The science writer Carl Zimmer posted an invited reply on his blog from Joseph Thornton of the University of Oregon to my recent comments about Thornton’s work. This is the first of several posts addressing it. References will appear in the last post.
I must say, it never ceases to amaze me how otherwise-very-smart folks like Zimmer and Thornton fail to grasp pretty simple points when it comes to problems for Darwinian mechanisms. Let me start slowly with a petty complaint in Carl Zimmer’s intro to the post. Zimmer is annoyed that I think Thornton’s latest work is “great,” yet I thought his previous work published a few years ago was “piddling.” “Why the change of heart?” wonders Zimmer.
It’s really not that hard to understand. Here’s a little analogy to illustrate. Suppose some company claimed they could build a super-crane (tip of the hat to Daniel Dennett) which could hoist a whole mountain using a novel technology. Though untested, the great majority of the relevant engineering community was serenely confident it would work as advertised. In a carefully-devised, initial, “proof-of-principle” experiment, a laboratory at the University of Oregon demonstrated that the crane-technology could lift a smooth pebble. The work was published in Science, accompanied by a breathless editorial and a story in the New York Times. In a subsequent careful study published several years later in Nature, however, the same lab unexpectedly showed that if a pebble were even somewhat rough, the crane-technology would not lift it. Since mountains tend to be rough, too, if a super-crane wouldn’t move a rough pebble, then it certainly wouldn’t lift a mountain.
Of course, the initial work, although technically well-done, can fairly be called “piddling” compared to the promised capacity of the crane. The subsequent work, again technically well-done, was “great” because it demonstrated formidable difficulties for the technology at a very basic level that no one — not even (ahem…) the few skeptics — had expected. (I hate to be so pedantic, but unfortunately it seems necessary on this topic.)