realism and the limits of scientific explanation

Long time, no blog. I finally got back a few days ago from the last of my visits to schools for final job interviews. It was very interesting and instructive to observe non-Pittsburgh philosophers in their native habitats. I should know by the end of this week where I'll be next year.

In lieu of an actual post, I am putting up the handout I used at a couple of my job talks. As a result, it looks programmatic/ bullet-pointy; but I tried condensing this into a normal post, and it was just far too long. If you can make out what's going on, I would really appreciate any feedback/ comments/ eviscerations from readers.


The argument

(P1) Scientists do not accept explanations that explain only one (type of) already accepted fact.
(P2) Scientific realism, as it appears in the no-miracles argument, explains only one type of already accepted fact (namely, the empirical adequacy or instrumental success of mature scientific theories).
(P3) Naturalistic philosophers of science “should employ no methods other than those used by the scientists themselves” (Psillos 1999, 78).

Therefore, naturalistic philosophers of science should not accept scientific realism as it appears in the no-miracles argument.

Explanation and defense of (P1)

Explanations that explain only one type of already accepted fact
(i) generate no new predictive content, even when conjoined with all relevant available background information [‘already accepted fact’], and
(ii) do not unify facts previously considered unrelated [‘only one type’].

Evidence for (P1): Scientists reject
- Virtus dormativa-style explanations
- ‘Vital forces’/ entelechies as explanations of developmental regularities
- Kepler’s explanation of the number of planets, and the ratios of distances between them, via the five perfect geometrical solids
- ‘Just-so stories’ in evolutionary biology

The no-miracles argument for scientific realism

Abductive inference schema
(1) p
(2) q is the best explanation of p
Therefore, q

No-miracles argument for scientific realism
(1) Mature scientific theories are predictively successful.
(2) The (approximate) truth of mature scientific theories best explains their predictive success.
Therefore, Mature scientific theories are (approximately) true.

Proponents of the no-miracles argument (Putnam, Boyd, Psillos) accept (P3), appealing to naturalism to justify their abductive inference to scientific realism. Putnam claims that scientific realism is “the only scientific explanation of the success of science” (1975, 73).

The argument for (P2): Scientific realism (i.e., the claim that mature scientific theories are approximately true)
(i) generates no new predictions,
(ii) unifies no apparently disparate facts, and
(iii) explains only one previously accepted fact, viz., science’s predictive success.


Anonymous said...

Well, I came to the argument with a firmly entrenched bias. But while I've always thought that scientific realism doesn't actually do anything that we need to have done, I thought this was an unusually elegant way of getting to that conclusion.

Kenny said...

At first I was thinking that this would just show that scientific realism isn't supposed to be an explanation of anything, but on reading your statement of the 'no miracles' argument (I don't know the literature on this), it sounds like you really have removed this argument for scientific realism. However, on thinking about this a bit more, I wonder whether realism really does have the feature that you cite, namely that of explaining only "one (type of) already accepted fact". According to your justifications, it sounds like there's weight on "one (type of)" to show that the purported explanation does no predictive work, and "already accepted" to show that the purported explanation does no unificatory work. Now whether or not the claim that "mature scientific theories are predictively successful" is only "one (type of) already accepted fact" seems to be relevant only insofar as it means that scientific realism does no predictive or unificatory work.

The claim that "mature scientific theories are predictively successful" does look like just one fact. However, it seems that "T is predictively successful" might mean something like "there is an O such that T predicts O and O is true", where "O" is something like an observation statement (or whatever it is that science predicts). Thus, we get "for all mature scientific theories T, there is an O such that T predicts O and O is true". On a substantive notion of truth, this does in fact look like just one type of claim - for each theory we claim that there is an observation statement it predicts that has the property of truth. But on a disquotational account of truth, the claim "O is true" is just going to be equivalent to some statement phrased in the language of the relevant science. So the universal claim really is something like a conjunction including predictions from every science - some about observations of plants, some about observations of UV radiation from distant galaxies, some about observations of bubble chambers, so about observations of litmus paper, and so on. So there might be some sort of unification going on in an explanation of "mature scientific theories are predictively successful". (Of course, since the putative explanation itself uses the truth predicate, it might not end up being any more unified, but that seems to me to be a slightly different worry.)

As for the question of whether scientific realism predicts anything new, we can consider which claims of the form "T is a mature scientific theory that is predictively successful" are accepted. Certainly this is accepted where T ranges over things like quantum mechanics, relativity, neo-Darwinian genetics, and so on. However, it isn't entirely clear that every relevant instance is already accepted. For instance, one might suggest that string theory is a mature scientific theory, and yet some physicists seem to allege that it might not be predictively successful (because they claim it makes no predictions whatsoever). If it really is a mature scientific theory, then scientific realism would predict that string theory is approximately true, and thus that it actually would make successful predictions, and that it's opponents are just ignoring some predictions that it does make (which their theory already makes equally well). So scientific realism might have some predictive power.

I don't know if these objections are actually successful - they seem a bit convoluted to me.

Anyway, good luck!

Anonymous said...

I am no fan of the no miracles argument,and I'm sure I missing something obvious here, but how about the following defense of it: a poor explanation is better than no explanation at all, and the realist explanation is the only explanation on offer. Putnam, recall, paired his NMA with a negative argument for realism- the failure of non-realist explanations of success.

Greg Frost-Arnold said...

Protagoras -- Thanks for the vote of confidence!

Kenny -- I need to think about what you've said a bit more. I'm actually not completely sure I've understood you; I'll study your comments and let you know if anything occurs to me.

David -- Welcome, and thanks for the comment. It's a good point, and prima facie plausible; but I think the stuff under my heading "Evidence for (P1)" counts as (quasi-)empirical evidence for the claim that there are some explanations that are so bad, that (the vast majority of) scientists would rather leave the fact unexplained. And more generally, scientific explanations do come to an end somewhere, for a particular science at a particular time (e.g. with fundmental laws).

marco said...

Greg -
I enjoyed this - original stuff. Two objections:
(i) there are many examples of scientific posits explaining only one type of fact. E.g. we believe we see the trajectory of a particle because that explains the trail in the cloud chamber.
(ii) One way of running the NMA is theory by theory. So let's grant that scientists only like a theory if it explains a range of types of fact. Theory T explains F1, F2 and F3 -type facts. But explanation is factive: only true theories explain. So T is true.

Greg Frost-Arnold said...

Hi Marco --

Thanks for this. Re: (ii), I think you can now see why I asked, in the post your commented on previously, why I would want [local realism + global anti-realism] to be viable: my argument says that the case-by-case NMA is OK, since (good) individual theories do generate new predictions or unify previously disparate phenomena. But the global NMA doesn't. (I have not given an argument FOR anti-realism, just an argument against ONE justification for global realism. So [Global anti-realism + local realism] is not forced upon me, but I do wonder whether it's consistent.)

Re: (ii): I probably need to think about this a bit more, and you may be right. I hadn't really thought about SINGULAR explanations much; I think it's clear that positing the existence of electrons does generate new predictions and perhaps unifies various phenomena, but I hadn't really thought about positing THIS PARTICULAR electron's existence explaining THIS PARTICULAR track. But, for the NMA issue, I think that doesn't matter, since the NMA isn't supposed to offer a singular explanation.

Anonymous said...

David - "a poor explanation is better than no explanation at all"

Is it? That sounds more like a defence for religion than for a philosophy of science. What's wrong with suspending judgement? It's not as if any progress is being arrested without that explanation.