What the fact-value dichotomy is not

by Dave Maier

416328B1BJLA few posts ago I distinguished between philosophical and scientific/practical questions about the objectivity of science, and urged that we not get them mixed up. There’s a lot more to say about that, so here’s another chapter in our continuing story.

Philosophical questions about objectivity are metaphysical questions, and of course we invite confusion right away if we insist that as scientists we don’t do that metaphysics stuff (as if one could somehow avoid metaphysical commitments simply by saying so). A closely related question (or a different aspect of the same one) is that of the relation between fact and value. Whether they affirm it or deny it, all sides seem okay with calling this the “fact-value dichotomy,” so that’s what we’ll do too.

This dichotomy is also called the “is-ought” question. It’s pretty obvious that there’s a literal difference between asking how things are and whether they should be that way, but that doesn’t entail that the former questions are objective and the latter not (and of course this is where our question morphs into our earlier question about objectivity anyway). The natural context for this question (although not, as we shall see, the only one) is that of the objectivity of morality; and here too we see an obvious (if not conclusive) difference between scientific and moral questions. As Gilbert Harman points out, moral questions are not subject to experimental confirmation. If we want to know whether murder is wrong, we can’t just murder a number of people (under proper test conditions), crunch the numbers, and see. That doesn’t make sense.

As always, though, the problem with dichotomies is that they make it seem that if we’re not on one side of the fence then we’re on the other, and that’s all there is to it. (It doesn’t help matters that there are plenty of cases in which this is perfectly true; philosophy tends not to be one of them though.) Just because we can’t establish the truth or falsity of moral judgments experimentally doesn’t mean they can’t be true or objective or whatever you want.

But even so, how does this work? Not surprisingly, there are better and worse ways to think about this. Here’s a hopefully instructive look at one of the latter.

One sort of conversation I learned to avoid early on in life was one which pits Science vs. Religion. [Full disclosure: I was a card-carrying “skeptic” and subscribed for several teenage years to Skeptical Inquirer magazine, each issue of which features multiple insufferably condescending “debunkings” of this or that superstitious nonsense, whether this be the doctrine of transubstantiation or that Bigfoot is retired and living in Mexico (okay, I made that one up), so when I say I learned this “early on,” I don’t mean (*cough*) immediately.] I mean the sort of conversation in which participants may deem it significant that, for example, Isaac Newton (or some other certified Smart Science Guy) was a religious believer or that at one point the Bible seems to indicate that pi = 3. That sort of thing.

There are many reasons to avoid such conversations; one is that the fact-value dichotomy or its negation is often, as are many ideas in this context, used as a bludgeon.

I once saw in a volume about creationism an article called something like “Overcoming the fact-value dichotomy,” which makes it sound like something Hilary Putnam might say, and this was surprising to me given that it was coming from the creationist side. What it turned out that the authors meant by this, though, was something else entirely. Those who subscribe to this dichotomy, they said, are those who reject the objectivity (facticity) of morality, believing instead that moral values are subjective and thus that value judgments cannot be factual. Conversely, overcoming this dichotomy means seeing that they can be.

Well, all right, that may be a bit crude, one might say, but at least it rejects a facile subjectivism – that’s worth something, anyway. But listen to the rest of it and see if it doesn’t make your head spin. First, as I noted, this is precisely the opposite of “overcoming the fact-value dichotomy"; instead, it keeps the conceptual dichotomy firmly in place and simply puts value judgments (that is, judgments about value) into the one category rather than the other. Objective facts remain as conceptually removed from subjective value as before.

In Putnam’s sensitive discussion, on the other hand, overcoming this or anything worth calling a conceptual dichotomy means overcoming our natural tendency to put things into separate boxes once and for all, and instead seeing everything in the relevant domain as having aspects of both opposed ideas in various senses and proportions. To do this we must reject not simply a crude subjectivism about value but an equally crude objectivism about facts. This means extending our discussion beyond the purely moral.

So far, so good. But I promised I would make your head spin. Okay, check this out. Remember that these are creationists speaking; not surprisingly, they see their opponents as saying something like that science is the only reliable road to objective knowledge, and that revealed “truths” about moral law were merely speculative hearsay and unacceptably culturally specific to boot. Naturally they find this unacceptable. (Note: since I don’t remember exactly who the authors are or what exactly they said, you should probably take the following as a composite. People do say this, even if possibly not these same people.) However, not content simply to defend the objectivity of morality qua value, they turn the tables and go on the offensive. Last time I objected when “science critics” conflated a philosophical rejection of the subject-object dualism with substantive criticism of scientific practice. Here, on the other hand, we have a full-on metaphysical affirmation of that dualism, explicitly labeled, oddly, as an overcoming of another, turned into a wholesale rejection of science.

One issue we often see in this context is whether science is dogmatic in its naturalism. Defenders typically distinguish between metaphysical and methodological naturalism. The former is an a priori (and thus, one assumes, “unscientific”) metaphysical commitment to the idea that everything that exists is a physical object or can at least (say in the case of numbers, which are sets, which are … I don’t know, this view doesn’t make sense to me either) be explained in physical terms. This does seem dogmatic, seeing as science is supposed to be a method rather than a metaphysics. Thus methodological naturalism, the idea that although science is scrupulously agnostic about the existence of non-physical entities, its characteristic method requires that physical entities are the only things it is capable of dealing with (because observable, measurable, etc).

This of course leaves no more room for angels and whatnot than does the dogmatic commitment to metaphysical naturalism, and it is sometimes on precisely these grounds that the Science team declares “creation science” to be a pseudoscientific misnomer. However, this seems just as dogmatic a dismissal as was the (explicitly) metaphysical version. First, the very restriction of our commitment to the abstract methodological level (requiring, again, that the objects of science be observable, etc.) still leaves perfectly open the metaphysical question of what exactly the nature of such entities could be, making the limitation to physical objects gratuitous. Why couldn’t we observe the Virgin if she appeared to us? We’re a long way from Antony Flew’s essentially unobservable invisible pink unicorn or whatever.

Not only that, if the creationist “what good is half an eye” argument worked, it seems we wouldn’t have any reason to dismiss it on methodological grounds (especially since it’s purely negative and thus more defensible than a positive claim, which is one reason creationism evolved (heh) into “creation science” a.k.a “intelligent design” in the first place). The best reasons to think it doesn’t work aren’t methodological ones; they’re substantively scientific ones, showing for example that (and, importantly, how) the eye has evolved on several distinct occasions in the history of life. Briefly, the problem with “creation science” isn’t that it’s not science; it’s that it’s bad science. This verdict of course allows that it may improve in the future. But that’s okay with pragmatists like Putnam or Peirce, who famously urge that theoretical roadblocks not be placed in the path of inquiry.

So far, so good. But let’s return to our own creationists. Rather than challenging a dogmatic commitment of science to naturalism, they insist on it, identifying, as do naturalists themselves, science’s characteristic method with a limitation to the physical world. But that doesn’t mean, they claim, that science is objective. Empirical science is, of necessity, inquiry into appearances. But the realm of appearance is essentially less real than, well, the Reality which underlies and sustains it. However scrupulous science may be, it simply cannot overcome its essential limitation to merely human capabilities. Science’s vaunted methodological skepticism and tentative, fallible, revisable conclusions (as in Popper’s influential view) now look flimsy and uncertain next to the divinely guaranteed certainty of revelation.

Again, it’s not just that human reason pales before the divine kind. It’s that here we see the Platonistic inheritance of Christianity in its most virulent form. Scientists identify (too quickly, in my view) the methodological objectivity of science with the objectivity, in the metaphysical sense, of its characteristic domain (i.e, the natural realm). But on a Platonistic view, the merely physical realm is insufficiently infused with logos to count as the object of knowledge properly so called. Knowledge is of the intelligible realm, a realm of pure ideas, not mere matter. Remember that our context here is an explicitly moral one. At first it looked like affirming the objectivity of moral judgment brought it (back) up to the same status (qua objective) as factual judgment, presumably that of science. But now we see that it is really only the moral realm which counts as (metaphysically) objective.

Indeed, from this particular premodern perspective (one not committed, I should note, to “creationism” in our modern sense), metaphysics and morality are pretty much the same thing, concerning as they do the ultimate intelligible (qua ideal) structure of reality. I think it was conservative columnist Ross Douthat who opined that philosophy departments nowadays (he went to Harvard) had nothing to say to those of his persuasion because they no longer contained any “metaphysicians or moralists.” Missing his point completely, famous philosophers responded scornfully that of course there were metaphysicians and moralists there, pointing to analytic studies of necessity or personal identity (that’s metaphysics) or of consequentialism or deontology (that’s moral theory).

Still, this brand of creationism, whatever else might be said for it, makes utter conceptual hash of our original question about the dualisms of fact and value on the one hand and subject and object on the other, especially in the context (chosen, again, by our creationist authors) of modern empirical science. We should resist the misdirection of their claim to overcome, rather than affirm, the fact-value dichotomy, given their industrial-strength metaphysical realism. The issue is difficult enough without ideological salvos in the Science vs. Religion debacle masquerading as philosophical progress.

Next time: what is the “given”?

Like what you're reading? Don't keep it to yourself!
Share on Facebook
Tweet about this on Twitter
Share on Reddit
Share on LinkedIn
Email this to someone