Last Thoughts on Naturalism

by Brian on August 1, 2003

The discussion threads on naturalism have been lots of fun, but I’m going to have to leave them behind to head off to my favourite little philosophical conference. Unless the thread lasts another week (an eternity in blogtime!) it will be done before I return. So I thought I’d close with a point of agreement.

Matt Evans discusses the “impossibility of criticizing anyone else’s Founding Moral Principle, no matter how outrageous it might seem to me.” This is a little strong – it’s possible to criticise anything. (Just listen to me for a while if you don’t believe this.) What isn’t possible is to make criticisms that the other party will find moving. There’s something to this. It’s why moral arguments are usually coherence arguments, I think. But I don’t think there’s much distinctive about morality here. I think something similar is true in every area of human study. A quick (and very compressed) tour through 20th century philosophy of science might explain why I think that.

The 20th century opens with Duhem pointing out that data never conclusively refutes a theory. One can always find enough auxiliary hypotheses to save the theory. There’s a hard question then about when theory change in response to data is appropriate. For two generations, one of the central questions in philosophy of science was when theory change is rational. The eventual conclusion was that if you’re looking at something as broad as all of science, you can’t say much more than “theory change is rational when smart scientists are moved to change their theory by the data.” If you want to know anything more specific, you have to find out what smart scientists are like. And hence we see the (quite productive) movement in late 20th century philosophy of science to philosophers learning more about particular sciences to try and say something much more specific.

The problem here is that if someone wants to hold onto a core proposition (as Lakatos put it) it’s going to be very hard to conclusively show that they are wrong. Theories change not because their core claims are refuted beyond all doubt, but because in the judgment of enough scientists, it looks preferable to adopt an alternative theory.

(Some people, myself included, think that this model generalises pretty well to all of philosophy. Philosophical argumentation can show you the costs of rival theories, but eventually it’s an act of judgment nor argumentation to determine which is the cheapest theory.)

So I think Matt’s right – someone whose judgment is so bad that their fundamental theories are completely wrong-headed cannot be talked out of this by argumentation. But this isn’t just about morality. Someone who really wants to hold on to a Ptolemaic theory of astronomy can find enough reasons to preserve their theory in the face of any countervailing data. (Remember all those epicycles!)

This is why I’ve been a little surprised by some of the responses to my claims that there are valid arguments from descriptive claims to normative claims. One common thread in the responses is that these arguments wouldn’t move a determined opponent of the conclusion. True enough. But some arguments from the data to astronomical conclusions wouldn’t move a determined opponent either, and that doesn’t show that our astronomical beliefs rest on faith, or cannot be incorporated into a naturalist picture, or don’t follow from perfectly ordinary descriptive claims.

None of this should be taken to be a claim that moral argument is never effective. We can have quite productive disagreements with people who share enough core premises with us. But if we try comparing views with radical opponents, we’ll find that many disagreements rest on quite different judgments about how costly various moves are. Those disputes tend not to be so productive. And this, very roughly, is why I much prefer moral disputes with people with similar political views, because we’ll often be trying to spell out the consequences of our shared premises, than with radical opponents, because then we’ll be in the (rather unpleasant) game of trying to convince the other guy that we’ve misvalued something. Sometimes this can be worthwhile (sometimes after all we do misvalue things and maybe this can be revealed somehow) but often it is less than productive.



Jack 08.01.03 at 2:20 pm

Are you saying that ought/is is a red herring in this case?
It seems to me that Matt would only be happy if everything had E-mark or FDA approved morality care labels on them or if you ears went red when you did something immoral.
Clearly these are absent from the world, at least in my experience. On the other hand a theist might claim to have a few such indicators and therefore be better off.
I would argue however that that is an illusion. First of all Different people repeatedly draw different conclusions from the same text so things aren’t that straightforward. Even the most explicit statements, “thou shalt not kill” for example, are open to interpretation and exceptions. Even supposing that there is a correct lead to be taken from the source there will be situations where that are not explicitly covered by the existing rules I like divorce by text message as an example.
That being the case the foundation of morality includes some process of argument or decision not directly supported by whatever super natural source the non-naturalist relies on.
It seems to me that someone who maintained that this extra stuff was all that was necessary might reasonably be called a naturalist and would not be easily dismissed on is/ought grounds. The attack would be that there were no foundations so you could never be sure that an argument was getting somewhere. To that the defence would be
i) I expect that I can always answer your why? questions.
ii) You (anti-naturalist) can never be sure if one of your arguments is resting securely on your foundations.
iii) I have no reason to believe in your foundations beyond bold assertion.
iv) Some of your foundations are offensive.
v) The foundations keep shifting anyway (should priests be celibate? How many wives can a man have?)
vi) What happens to Galileo? (Alternatively humans are fallible)


Matt Evans 08.01.03 at 3:10 pm

Thanks Brian for spending so much time explaining your perspective on this issue.

One point, however. Your summary of my argument, “someone whose judgment is so bad that their fundamental theories are completely wrong-headed cannot be talked out of this by argumentation” understates my claim. This problem doesn’t exist solely among those with bad judgment (actually I claimed that there is no rational basis for saying one fundamental moral premise is better than another). But even if you believe we can determine who has bad judgment, they are much greater in number than you suggest.

Take these two common moral premises: (1) all human beings, and only human beings, have equal moral worth, and (2) all sentient beings, and only sentient beings, have equal moral worth.

These two premises are both commonly held. (Though many who accept premise 2 omit the reference to equality). The allegation that one of them is simply captured by bad judgment doesn’t stick.

And even though these initial premises are very similar, their allegiance to different classes, humans or sentient beings, leads to vastly different results. Every argument over animal rights, unborn human beings, the comatose, and the severely handicapped will eventually be reduced to these core premises, at which point the debaters will realize they have nothing to say to one another. (This is assuming that these premises are fundamental, but I suspect they’re not, meaning there would be at least one more step of desconstruction before they reached bottom.)

Anyway, thanks for the posts and enjoy your conference!

– Matt Evans


Matt Evans 08.01.03 at 4:27 pm

One other thought.

A counter-example to your “Matt’s claim isn’t limited to morality” argument is that people view facts and values differently.

Many factual disputes can be resolved by more data. Though there are some skeptics who won’t be convinced no matter the evidence, most people accept facts.

There is very little consensus on values, however. Is sex outside of marriage wrong? Is eating animals wrong? It’s hard to imagine what facts could possibly resolve these questions.


Eric Olson 08.03.03 at 4:43 am

I’m coming very late to this discussion, but it occurs to me that the problems with naturalism and ethics stem from the fact that naturalsits and theists disaggree over the definition of “ought”. As a theist, I would state that “One ought to do X” is equivalent to “Doing X has positive consequenses in the long term for one’s spiritual well-being in the afterlife.” (Other theists may give different definitions, but I would say that this is realy what Christian morality at least boils down to.) When a naturalist says, “One ought to do X,” I suspect (correct me if I’m wrong) that he means “As a result of the evolution of human society, most people feel a strong urge to do X that is totally unconnected to the immediate pleasure or tangible gains resulting from X”. The Christian conception of metaethics requires the existance of a God, or at least of an afterlife. Doing X is right if X will help you to be saved, even if X contradicts your evolutionary programming (e.g. celibacy and martyrdom). The naturalist conception of morality is completely unrelated to any afterlife and completely dependant on widely held moral intuitions. Doing X is right if biological and societal evolution cause people with your combination of genes and environment to want to do X, even if doing so would cause you to go to hell (e.g. Ivan Karamazov rejecting his ticket to heaven). The real debate is not whether ethics are compatible with naturalism but what exactly ethics are.

(yes, I know that Ivan Karamazov is a straw man caracature of a naturalist. I used him because most real naturalists don’t even bother with the question of whether their actions would result in their damnation, for obvious reasons.)

Comments on this entry are closed.