BETA
This is a BETA experience. You may opt-out by clicking here

More From Forbes

Edit Story

Failing the Science Test

This article is more than 10 years old.

Over at the Discovery Institute's Evolution News and Views, Casey Luskin has a post listing various ways in which intelligent design theory can be tested. (More than five years after the Dover trial, the Institute is still arguing that ID is science.)

For several reasons the post is unconvincing, but what struck me is how Luskin initially went after a scientist at Dartmouth, Mark McPeek, and tried to claim that a recent paper by the biologist and his colleagues amounts to indirect confirmation of intelligent design.

According to Luskin, the paper in question, MicroRNAs and metazoan macroevolution: insights into canalization, complexity, and the Cambrian explosion, published in BioEssays in July 2009, is a boost for intelligent design because the paper's authors frankly couldn't come up with a viable 'materialist' explanation for how some forms appeared in the fossil record.

If you read the paper, however, you are left scratching your head as to how Luskin came to that conclusion.

Since Luskin never evinced any interest in actually contacting the scientist for his thoughts on his paper's alleged shortcomings, I decided to.

Mark McPeek is the David T. Mclaughlin Distinguished Professor of Biology at Darmouth College. His specialty is evolutionary biology and ecology. And he was already aware of Luskin's blog post of his paper.

"No, this paper does not make an ID argument for the Cambrian Explosion," he told me by email.  "It makes the proposal that the new class of RNA molecules called microRNAs may have played a very significant role in the evolutionary diversification of the Metazoa (i.e., animals).  MicroRNAs are very diverse across the Metazoa, and the first ones seem to have arisen in the ancestor to sponges.  Our paper proposes a model for how microRNAs arise and then can influence the evolution of the lineage that has this new microRNA."

Luskin was simply reading into the paper what he wanted to see.

But what about ID making testable predictions? I asked. For example, in Table 3. of Luskin's post, he lists two points which he claims can be explained by ID ... but not by evolutionary theory.

(3) Similar parts are commonly found in widely different organisms. Many genes and functional parts not distributed in a manner predicted by ancestry, and are often found in clearly unrelated organisms. (Davison, 2005; Nelson & Wells, 2003; Lönnig, 2004; Sherman 2007)
(4) There have been numerous discoveries of functionality for "junk-DNA." Examples include recently discovered surprised functionality in some pseudogenes, microRNAs, introns, LINE and ALU elements. (Sternberg, 2002, Sternberg and Shapiro, 2005; McIntosh, 2009a)

Where does Luskin derive the idea that there are "genes and functional parts not distributed in a manner predicted by ancestry"?

"Point 3 is interpreting examples of parallel and convergent evolution as evidence that the designer solves the same problem the same way in different organisms," McPeek said. "I think the point there is that if organisms are random collections of traits, more closely related taxa should be more similar, and so therefore things that are far apart phylogenetically must be more different.  Thus, when they aren't, that is de facto proof of design."

But for McPeek, this is simply the selective ignoring of how natural selection works.

"This is exactly the issue of testability," he said. "On this specific point, ID and evolution make exactly the same predictions.  However, the mechanisms of evolution that produce this outcome are (and have been) imminently testable and tested, and so are legitimate mechanisms to postulate as an explanation."

But to test ID?

"I'd have to know what mechanisms the creator was using to produce that outcome and see if I can replicate those mechanisms." [italics mine]

This is where ID really loses scientists. (Not that scientists have ever been their targeted audience, but that's a topic for a separate post.) Luskin seems to insist that scientists are supposed to devise tests to determine if any prospective intelligent agent (God of the Bible or an alien) is responsible for guiding patterns in the history of life. Since people can discern easily when someone else of our species is responsible for intelligent designs or actions, it therefore should be a simple matter by this argument to do the same for the natural order.

But, as McPeek and hundreds of other scientists have pointed out, this argument fails.

"Any idea or hypothesis is fair game for analysis and testing," he said. "However, their evidence is invariably an analogy and not direct evidence.  I know things that are designed when I see them - they have order.  And so anything with "order", was clearly "designed". That's basically been the ID argument and evidence since Paley found a watch in the heath."

The problem with analogy as a scientific method, however,  is that correlation is not causation. "Any argument by analogy is simply a correlational argument," said McPeek. "Hence, all the arguments I have seen (and I've read a lot of the 'primary literature' of ID) from them are not 'scientific' tests."

"To me," he added, "the arguments for ID in specific cases all fall into one of two categories: (1) the argument by analogy (from above), or (2) science has no explanation for this, so it must have been designed. The first has the problems I stated, and as I always tell people - Scientists work on things we don't know anything about, so blaming us for not having a pat answer for everything is quite disingenuous."

Finally, for some bizarre reason, Luskin couldn't resist getting personal:

Dr. McPeek holds a prestigious position at an Ivy League school where he pursues research related to evolutionary biology. If Thomas Kuhn's ideas hold any merit, he's not likely to admit the veracity of a new, competing paradigm of biology. Also, his article makes it clear he's capitulated to the NOMA construct which pretends that, as he puts it, "science can only be mute on these issues, since we cannot empirically test the existence, actions or methods of God." While we might not be able to scientifically identify the designer as God, we can certainly find signs of intelligent action in nature.

Again, this must play well to the target audience, I suppose, but Luskin is on shaky ground. Hundreds of years before Darwin, Catholic scholars in Medieval Europe (you know, the 'dark' ages) were coming to the very same conclusion about how investigation of the natural order should be pursued.

As William of Conches put it in the 12th century:

"In studying nature we have not to inquire how God the Creator may, as He freely wills, use His creatures to work miracles and thereby show forth His power; we have rather to inquire what Nature with its immanent causes can naturally bring to pass."

Here's Adelard of Bath, one of William's contemporaries:

"The natural order does not exist confusedly and without rational arrangement, and human reason should be listened to concerning those things it treats of. But when it completely fails, then the matter should be referred to God. Therefore, since we have not yet completely lost the use of our minds, let us return to reason.”

You can just tell Adelard was a Brit. Almost Dawkinsian, that last line. But you get the idea. Leaving a Designer out of the methodology of science as a matter of principle  has a long, healthy tradition in Christian natural philosophy.

And that is why intelligent design is a failure.