Gap Arguments and ID

Gil Dodgen at Dembski's blog has a post about ID and "god of the gap" arguments, drawing on quotes from Del Ratzsch in this interview. I think Dodgen's comments make little sense. In the interview, Ratzsch argues that god of the gaps arguments are not necessarily wrong. Dodgen quotes part of that, but leaves some out as well. I'll post the quote from Ratzsch in fuller context below the fold.

TGL: Design arguments are often associated with the idea of "gaps" in nature. How important are gaps to design arguments?

DR: For some design arguments, they are crucial. For instance, Dembski's 'explanatory filter' is completely gap-driven. I don't see that as necessarily a defect. We routinely employ gap arguments in all sorts of contexts. For instance, the SETI program is a gap-searching project - trying to find signals which nature alone couldn't or wouldn't produce, then constructing alien-civilizations-of-the-gap arguments. Further, it is nowhere written in stone that nature has no causal or explanatory gaps of the relevant sort. For all we know nature may contain gaps which can only be bridged by divine action. And that could be intentional - God might like running some things in nature hands-on, and might have created nature to allow for that in its normal operations. Anyway, gaps and gap arguments as such are unproblematic in principle.

Of course, there are serious questions concerning how we might identify supernaturally bridged gaps as such, especially scientifically. Such gaps might thus be inconvenient - or worse - for our investigative efforts and procedures, and we might have no choice but to assume continuity as a working strategy, but we can hardly demand that nature conform herself to our limitations - i.e., we cannot very appropriately issue inflexible ontological edicts propped up solely by reference to our epistemological limitations. Reality, it seems to me, is a bit more independent and robust than that. But many see gap-type arguments as having a troubled history and a troubling character.

But not all design arguments are gap arguments. In fact, during the heyday of natural theology - late 18th and early 19th centuries - non-gap arguments involving the interlocking structure of laws, cosmic order, the elegance and beauty built into nature, and the like, were very widely seen as preferable to and more powerful than gap arguments. And many contemporary ID advocates also embrace some non-gap arguments, including the 'frontloading' picture mentioned earlier. Front-loading types of design pictures go back at least to Augustine.

More specifically, gaps have to do with e.g. mechanical causal histories, whereas design has to do with intentional histories. Those are in many cases intimately related issues. Gaps can be important clues to design, since depending on the context an actual mechanical, causal gap could suggest agency as a causal factor, and it is a relative short step from there to design. But the issues are distinct, and the ritual allegation that design views are all God-of-the-gap theories is inaccurate philosophically, as well as historically and contemporarily. There are, again, many (especially lay) design-gap advocates, but the blanket universalization is straightforwardly mistaken.

It is also worth noting that if nature is designed and if it does contain causal or explanatory gaps, then any prohibition on gap theories will nearly guarantee that science - discarding one failed non-gap theory only by replacing it with another (not yet failed) gap theory - will not self-correct in the usual advertised way, and that science will never correctly understand the relevant phenomena.

Dodgen cited bits and pieces of that, but even taken as a whole I think it has some serious flaws. He is essentially arguing that we cannot rule out the possibility that there exists some causal gap which cannot be filled with a natural explanation and that would require a supernatural one, a point which I take to be self-evidently true. No, we cannot rule that out. The problem, of course, is that no one has yet come up with one and the innumerable attempts to argue based upon a lack of natural explanation in the past have always turned out to be false.

So while we cannot rule out the possibility that there might exist some gap that must be filled with a supernatural explanation, history would strongly suggest that we keep looking for such an explanation rather than giving in to the easy temptation of ascribing an unconfirmable, untestable explanation. History is replete with assumptions of supernatural explanations that turned out to be false, from the animism or shamanism of early man to their closely related cousins in modern religious doctrines.

It was once thought that thunderstorms, disease, lost wars and bad crops, among many other things, were signs of the anger of this god or that, while good health, bumper crops and military victories were proof that the locally popular deity was pleased with us. Now, of course, we know that weather is caused by the natural interaction of atmospheric forces, disease by genetic anamolies or microbes, good crops by fortuitous conditions and military victories by good planning and superior weaponry.

Given that every single god of the gaps argument ever offered has eventually been proven false by the slow and steady accumulation of knowledge and understand using the scientific method, it is understandable that scientists would be immediately skeptical of new arguments taking the same form - "science can't (yet) explain X, therefore God must have caused X." So while we cannot in principle rule out the possibility that such an argument may yet end up being true, scientists are entirely justified in continuing to do the hard work of providing explanations with firm assurance that what has happened in the past will continue to happen in the future - that science will continue to provide valid explanations for what we see around us.

Ratzsch seems to understand this, which is why he later says:

But all that said, I think that one should be very wary of identifying something in nature as indicating a gap. Creationists and some (especially lay) ID advocates have been very - overly - ready to see gaps, and some opponents of ID have made the complete absence of any gaps in the entire cosmos a matter of principle. Neither position, it seems to me, fosters a suitable readiness to see what nature might have to say on the issue.

And as Ratzsch also notes, most of the major ID arguments rely upon precisely this type of god of the gaps argument. He specifically points to Dembski's explanatory filter, which he has been critical of in the past, as one such argumen. But in reality, all of the major ID arguments require the failure of science to explain a given phenomenon as their first premise. Dembski's explanatory filter requires the failure of "chance and necessity" (i.e. of natural explanations) as the first premise.

But so does Behe's "irreducible complexity" (IC). And all of Wells' arguments are purely negative, aimed at showing that science has not (yet) explained this or that. And I think Ratzsch is mistaken when he says that frontloading is not a gap argument. He says:

Among the problems with this shortcut dismissal are that although many design theories are indeed gap theories, some are not (e.g., what ID advocates sometimes refer to as "frontloading," according to which God built all the relevant potentials and capacities and directions into the original creation, subsequent developments being the designed, natural, unbroken, and continuous unfolding of those primordial potentials).

But frontloading cannot possibly be justified without making a gap argument; if the natural interaction of natural forces is sufficent to explain biochemical complexity, there is no way to justify the need for any sort of frontloading. More importantly, I would argue that frontloading scenarios do not make sense in light of the evidence that we have. If it was true that God programmed in all the relevant potential for later developments, then wouldn't we expect to see that bacteria, the first cellular life on earth (at least that we have evidence for) and the most primitive organisms still in existence, have DNA coding for later developments like feathers and lungs and eyes? They do not, of course.

It should also be noted that the frontloading hypothesis requires more than just that initial programming, it would also require a God that continually protects that initial genome from mutation. Unexpressed portions of the genome mutate rapidly and uncontrollably because natural selection cannot weed out harmful mutations; that's why so much junk DNA develops over time in those areas of the genome which do not code for actual proteins. So in fact, the frontloading hypothesis not only is a god of the gaps argument, but it explicitly requires a continually intervening supernatural force - precisely what ID advocates like to claim a design inference does not necessarily require. Frontloading simply does not save ID from the standard arguments against such inferences.

It should also be pointed out that analogies with SETI or with human-designed objects just don't mean much in this discussion. One simply cannot analogize natural and supernatural causation. We can infer human design in archaeology when looking at an object, but we can only do so because we know a great deal about the limitations and abilities of human designers (and about the limitations and abilities of natural forces to produce inanimate objects). We have no such knowledge of supernatural design, and that is precisely one of the reasons why such explanations are not taken seriously in science.

Once one invokes supernatural interference, all bets are, quite literally, off. Because we know nothing of how a supernatural entity might operate, what limitations if any such an entity might operate under, or how they interact with the natural world at all, we simply can't say anything meaningful or make any meaningful predictions based upon such a hypothesis. That is exactly why ID explanations cannot be falsified, because no matter what the data might say, you cannot rule out a possible supernatural explanation.

If the fossil record or the molecular sequence data showed one pattern, then obviously the supernatural designer simply chose to do it that way; if it shows the opposite pattern, then the supernatural designer chose to do it that way. There isn't even a coherent way to distinguish between a supernatural designer intervening in nature and violating natural laws to produce a given phenomenon or object, and a designer working through natural laws to do the same thing.

In short, there is no set of data that could possibly be inconsistent with a supernatural entity choosing to make it that way. And theories that can explain anything are scientifically sterile; explaining everything actually means explaining nothing. It's like blaming the rain on the action of leprauchans. When it rains, you see evidence of the leprauchans in action. When it doesn't rain, then the leprauchans have chosen not to make it rain. Both X and ~X are equally explained by such a hypothesis. And in science, that means one hasn't actually explained anything at all.