June 21, 2009

This Post is Old!

The post you are reading is years old and may not represent my current views. I started blogging around the time I first began to study philosophy, age 17. In my view, the point of philosophy is to expose our beliefs to rational scrutiny so we can revise them and get better beliefs that are more likely to be true. That's what I've been up to all these years, and this blog has been part of that process. For my latest thoughts, please see the front page.

Intelligent Design and Scientific Instrumentalism

My apologies for the unscheduled hiatus. Over the past two weeks, I was working frantically to complete my term papers, and then taking some much needed rest. Incidentally, I expect to post a draft of a paper on Aristotle under the title "The Homonymy of Predicative Being" in the near future, and hopefully there will also be a draft of my Kant paper in the next month or two. Anyway, I should now be getting back in the swing of things. Currently, I am reading the last couple issues of Faith and Philosophy which had piled up during the school year, so that's what I'm writing about today.

John Beaudoin's recent paper "Sober on Intelligent Design Theory and the Intelligent Designer" contains the following fascinating remark in a footnote:

[William] Dembski has suggested that the designer referred to in ID theory need not be real: it could in principle be treated by design theorists as a mere useful fiction, if that should better fit with a particular design theorist's philosophy of science.

Beaudoin cites Dembski's No Free Lunch, p. 15, and The Design Revolution, p. 65.

I haven't bothered to read too much on the whole ID thing because it is not closely related to my main philosophical interests and from a theological/religious perspective seems like a mere distraction. Furthermore, most ID types seem to me to exaggerate the problems of 'orthodox' evolutionary biology. (Note that, contrary to popular belief, most of the original advocates of ID - as opposed to its politicizers - believe in macro-evolution, and this, at least, is a significant difference from most of the old 'creation science' establishment.) However, based on what I have read and heard, Dembski seems to be the most respectable of the bunch, and this remark reinforces that impression. By this assertion (which, granted, I am getting second-hand), Dembski is restricting ID to the claim that a designer (of undetermined nature) is an unobserved postulate of a well-supported scientific theory. That is, we have the same sort of evidence for a designer that we have for, say, quarks. By claiming that a scientific theory is well-supported, I mean to claim that we have good reason to suppose that it is correct in whatever sense of 'correct' is appropriate to scientific theories. So, according to Dembski, we can reasonably believe that the correct theory of biology postulates a designer.

The question raised is, does it follow from this that the designer actually exists? Most people would say yes: the correct theory of physics postulates quarks, so we know that quarks actually exist. However, some theories in the philosophy of science, most notably 'instrumentalism,' deny this. Berkeley was the first major proponent of instrumentalism, and it is defended today by Bas van Fraassen. On Berkeley's view, natural scientists are engaged in "fram[ing] general rules from the phenomena, and afterwards deriv[ing] the phenomena from those rules." People who do this, he says, "seem to consider signs rather than causes" (Principles 108). By these signs "we are instructed how to regulate our actions in order to attain those things that are necessary to the preservation and well-being of our bodies, as also to avoid whatever may be hurtful and destructive to them. It is by their informations that we are principally guided in all the transactions and concerns of life" (New Theory of Vision 147). Thus in a later work he says that the legitimacy of the concept of 'force' (as a technical term of physics) does not depend on some sort of metaphysical cognition, or any fact of fundamental ontology. Rather, 'force' is a legitimate concept because "by considering the doctrine of force, men arrive at the knowledge of many inventions in mechanics, and are taught to frame engines, by means of which things difficult and otherwise impossible may be performed" (Alciphron 7.7).

Instrumentalism is thus a metaphysically deflationary view of science: it holds that scientists are in the business of constructing useful generalizations and are entitled to whatever conceptual apparatus is useful - including the postulation of entities which do not, in fact exist. The correctness of a scientific theory is to be measured by the success of its empirical predictions, not by its correspondence to metaphysical reality.

Accordingly, if Dembski is to convince an instrumentalist that ID is natural science and not metaphysics, he must drop the claim that ID proves the existence of a designer, because the instrumentalist denies that natural science can prove the (metaphysical) existence of anything.

Now, trying to define 'science' is so problematic that many (staunchly anti-ID) philosophers of science have stopped trying. This difficulty is really the motivation for the differences between previous formulations of 'naturalism' and the approach Penelope Maddy calls 'second philosophy' in her excellent book by that title. The courts may say that ID is prohibited from schools because it is 'unscientific,' but today ID's philosophical opponents tend, it seems to me, to prefer to claim that the reason it shouldn't be taught in schools is that it is laughably bad science.

Nevertheless, the instrumentalist does have a strong distinction between the 'scientific' and the 'metaphysical' - correct metaphysical claims are literally true and correct scientific claims aren't (necessarily)! It does seem right to me (perhaps because I have instrumentalist leanings myself) to say that if the claim that ID is 'science' is to stick, it will have to be possible to consistently hold ID and instrumentalism, and thus to consistently hold both that ID is a component of the correct theory of biology and that there is no designer. Of course, since this requires some exotic philosophical theories, it is unlikely to assuage any ID opponents.

Posted by Kenny at June 21, 2009 9:08 PM
TrackBack URL for this entry: http://blog.kennypearce.net/admin/mt-tb.cgi/493


Interestingly, I think Daniel Dennett might technically qualify as an instrumentalist IDer, as per his 'design stance' and use of 'Mother Nature'.

I think it's pretty much impossible to be an instrumentalist about the designer if one's acceptance of the designer is based on a causal inference to the designer, which is what most IDers are aiming at. But if one accepted the designer for theory-improving reasons (if one thought it improved the theory's simplicity, elegance, scope, whatever), than one could do so -- all one would be committed to saying is that a theory with a designer works better as a theory than the available alternatives.

Posted by: Brandon at June 22, 2009 10:00 AM

Yes, but those who accept full-blown instrumentalism (like Berkeley and van Fraassen) go through 'causal' reasoning to establish that scientific theories should posit certain entities all the time, and nevertheless deny that these entities 'exist' in the strict sense. This is why, in the passage quoted, Berkeley denies that 'causal' reasoning in the natural sciences is genuinely causal.

However, it seems to me that you are talking not about full-blown instrumentalism that reads all of science this way, but simply the use of 'useful fictions' in science. It does indeed seem that most writers in evolutionary biology, including virulent anti-ID types like Dennett, use design language as a 'useful fiction.' However, in the above quote the phrase "if that should better fit with a particular design theorist's philosophy of science" suggests to me at least that Beaudoin takes Dembski to be talking about a full-blown instrumentalism which holds that all unobserved postulates of scientific theories to be merely useful fictions. This is why I picked up the quark example.

In van Fraassen's case, there is skepticism at the 'epistemological level,' so for all we know scientific theories might accidentally line up with metaphysical reality, but we can't know this and it has nothing to do with the correctness of the theory. This is opposed to Berkeley who thinks he knows that there are things postulated by correct scientific theories (corpuscles, for instance) which nevertheless don't exist. (That this is Berkeley's view is undeniable if one takes a unitarian approach to the early works and Siris; if one thinks that the early metaphysics has been abandoned by the time of Siris, which few scholars do any more, then it is harder to say whether the early Berkeley thinks science should be 'purified' of non-existent entities.)

So even if we use causal reasoning to get to the designer, an instrumentalist can still deny that the designer exists.

Posted by: Kenny at June 22, 2009 10:37 AM

'Useful fiction' is not quite right, because it's really instrumental requirement that's the point: i.e., that the instrument, the theory itself, requires it to do its job (or do its job well, or to do it fully). This is a much stronger and narrower category than 'useful fiction', and, in fact, doesn't require (although is consistent with) taking what's posited to be false. But I take your point about the difference between the local and global instrumentalism in this context.

Posted by: Brandon at June 22, 2009 6:07 PM

Yes, that's what I mean. I didn't mean to imply that instrumentalism holds that scientific theories must be literally false; rather, instrumentalism holds that whether they are literally true doesn't matter to their correctness. At any rate, the local vs. global distinction is indeed what I had in mind.

Posted by: Kenny at June 23, 2009 12:01 AM

For the sake of argument, let’s assume Bob believes the following:
#1. Bob is a Christian who is staunchly an instrumentalist with respect to historical science, and he thinks realism with respect to historical science is logically impossible.
#2. Bob thinks evolutionary development (Evo Devo) is currently the conceptual apparatus that helps scientists make the most sense of observations.
#3. Bob thinks it is logically possible that young age creationism will become far more progressive and it will be able to explain far more observations than Evo Devo. After all, according to Bob, Evo Devo is merely a conceptual apparatus that helps scientists construct useful generalizations and scientists are entitled to use whatever conceptual apparatus that is useful to them at any given time.

It seems to me that it would be really hard to be both an instrumentalist and think young age creationism might be true. In other words, it seems like young age creationism and realism should go hand-in-hand.

Here is Bob’s dilemma as I see it:
A) If Bob believes #1, then he can’t admit realism might be true without having a severe case of cognitive dissonance.
B)If Bob accepts #1 and #3 and rejects the possibility that realism with respect to historical science might be true, then he would be admitting it is possible that the best way to explain how the earth’s topography came to be is to postulate the existence of a useful fiction (a specific understanding of a God from a specific tradition within a specific religion that is either epistemically inaccessible or ontologically inaccessible or both) which just seems frivolous. On top of that, in this case he would have a severe case of cognitive dissonance as well because as a Christian he would not think his God is a useful fiction, but when he thinks about young age models he would be thinking the same God is a useful fiction.

Is my assessment of this thought experiment correct?

Posted by: Durrell Chamorro at November 7, 2010 8:45 PM

Hi Durrell,

We need to be careful, I think, about what we are adopting instrumentalism about, and about what instrumentalism actually entails. Instrumentalism doesn't entail that the 'fictions' are false. It is just the view that usefulness, rather than truth, is what we're after. Consider an analogy to a novel: a novel, as a work of fiction, is aimed at entertainment, artistic value, and perhaps also making some kind of social/political/etc. point. Whether anything in the novel is true is irrelevant to the purpose. If, as happens in fantasy novels, nearly everything said in the novel is false, that's fine. But if, as in historical novels, many things are true (though others are false), that's fine too. Truth is beside the point. An instrumentalist analysis of science says that the truth of scientific theories is beside the point, as long as the empirical predictions of the theories are true.

Presumably Bob, in your story, is not an instrumentalist about religious/theological discourse, just about scientific discourse. So if he has independent religious/theological backing for his young earth view, then the science could go either way without effecting him.

Posted by: Kenny at November 8, 2010 11:45 AM

Thanks Kenny!

When you say, "if Bob has independent religious/theological reasons his young earth view, then the science could go either way without effecting him," I take it you are saying, according to Bob, Bob can actually know how old the universe is due to his insider information that he has through his religious/theological backing. Thus, if Bob agreed that all the evidence in the universe pointed to an old universe, then all that would mean for Bob is it is merely useful for scientists to assume of the universe as old.

The problem I was thinking of in my thought experiment was Bob doesn't have any such religious/theological backing for a young age view. In fact, Bob thinks 1. it is logically impossible to actually know the age of the universe, but 2. Bob thinks it is logically possible that young age creationism is true.

The young age leadership thinks they can know a very accurate age for how old the universe actually is, but Bob thinks such knowledge is epistemically impossible for humans. Thus, when Bob says he thinks young age creationism might be more progressive, Bob is saying some day it might be more useful for scientists to think of the universe as being about 6,000 years old, and whether the universe is actually 6,000 years old is beside the point because nobody can really know for sure.

I'm saying, it seems highly unlikely that anyone would conduct research in any of the historical sciences with the assumption that the universe is about 6,000 years old unless they had religious/theological reasons for thinking that is in fact about how old the universe is. But according to Bob, humans can’t know that information so any religious/theological reasons for thinking humans can know that information must be false.

Therefore, when Bob says it is logically possible that young age creationism might be more progressive than evolutionary models, he is saying it is possible that secular science might find it more useful to assume the universe is about 6,000 years old.

Given the way all the present young age models rely so heavily on God to break all the regularities we currently observe in nature, I find it highly improbable that any secular research program would posit such an entity that just happened to intervene just right for the universe to be about 6,000 years old.

I take it you are saying if secular science came to the conclusion above, it would make sense for Bob to think his God was the entity breaking up all the regularities, but then given Bob’s instrumentalism with respect to historical science, Bob would have to think it is logically possible that his God broke up the regularities and it is possible that his God didn’t. Bob would be committed to thinking he can’t know the truth until He asks God in heaven.


Posted by: Durrell Chamorro at November 13, 2010 10:29 AM

What I was thinking was just this: according to instrumentalism, the fact that something is a conclusion of our best science, is only a reason for thinking that it is useful for making predictions, not a reason for believing it is actually true. (I'm assuming that the scientific instrumentalism is not coupled with some broader form of anti-realism.) As a result, if you weren't claiming that your creationism was part of science, then you could be a realist about your creationist claims, and not care whether they were contradicted by our best science or not.

Now, it is true that 'creation science' and 'intelligent design' are both characterized by claims to 'scientificness'. So on these views if you were a scientific instrumentalist then you would not think that the fact that science (according to you) needs to posit God is a reason for thinking that God exists. That, really, is the point.

Posted by: Kenny at November 13, 2010 10:38 AM

Right. You cleared things up for me. I was justing thinking of a related point.

Posted by: Durrell at November 13, 2010 3:02 PM

Post a comment

Return to blog.kennypearce.net