Writer's block. OK, let's try this on the blog before trying to write it down again.

I have a draft due sometime before I go to sleep tonight, and I'm having trouble getting the ideas clear enough to write down. So here's a first draft of my first draft.

Moral realism is the view that morality is factual, sometimes true, and not fake (like, it's not mind-dependent, or it's not like we're just identifying morality with social norms and peer pressure). How can one argue for this view? One way is to try, somehow, to latch moral realism onto some other discourse that we feel much more comfortable about.

For example, here's a recap of Sturgeon's argument: if you're a scientific realist, you think that our scientific views are largely true. And so you probably think that inference to the best explanation, or something awfully close to it, is true. This tells us that unobservables like protons or electrons exist, and it might also tell us that numbers exist. Sturgeon argues that ethical beliefs are necessary for making our explanations better, and so inference to the best explanation should tell us that there are moral facts. Of course, one could just say "Feh, I don't like science either" but we're not taking up that argument. We're just trying to argue for the conditional, "If you accept science (then you accept inference to the best explanation, and if you accept inference to the best explanation) then you accept moral facts.

Let's say that fails. What other ways are there to approach an argument for moral realism?

In Enoch we saw another way. But what he's still doing, the rock of his argument, is scientific realism. Put it like this: inference to the best explanation is true, we know that from scientific realism. If you think that inference to the best explanation is true, then you believe that it is justified. Then Enoch provides a justification for IBE. Then he says that the justification he provided also justifies some new principle--inferring that what is necessary for deliberation is true.

I've spent time trying to argue that this approach is misguided. First, I think that it's easy to see how you could have conflicts between what's necessary for deliberation and what's necessary for explanation. What this comes down to is there being a problem with creating something like two epistemic realms--they're going to interact and cause problems (though maybe make things better too). Also, the argument gets off the ground by demanding that a basic belief, such as IBE, give itself a justification, but maybe that's an unfair requirement. Maybe we need to simply take some principles as primitive and unjustified, or maybe we need to be coherentists and not ask for a single justification for every belief. So Enoch's argument is sensitive to alternatives that seem more attractive than his account.

There's another account that came before Enoch's, but it's vague and not fully worked out (as far as I can tell). S-M suggests the following:

Still another reply, compatible with the first two but relying specifically on neither, shifts attention from science and from mathematics and logic, to epistemology itself. To think of any set of considerations that they justify some conclusion is to make a claim concerning the value (albeit the epistemic as opposed to moral value) of a conclusion. To hold of science, or mathematics, or logic, that there is a difference between good evidence or good arguments and bad ones is again to commit oneself evaluatively. This raises an obvious question: under what conditions, and why, are epistemic claims reasonably thought justified? Whatever answer one might begin to offer will immediately provide a model for an answer to the parallel question raised about moral judgments. There is no guarantee, of course, that our moral judgments will then end up being justified. The epistemic standards epistemology meets might well not be met by moral theory. But there is good reason to think the kinds of consideration that are appropriate to judging epistemic principles will be appropriate too when it comes to judging other normative principles, including those that we might recognize as moral. This means that any quick dismissal of moral theory as obviously not the sort of thing that could really be justified are almost surely too quick.

In other words, the epistemic realm is another normative realm. The question S-M is asking is what justifies our justifications? What's the second-order justification of justifications? Now, if we're just asking for more justifications, why do we think that we'll get anywhere new? Put another way, we're asking for normativity to provide the basis for normativity. I think the idea is that we'll get somewhere, and that since we're providing justifications of something normative, that will also provide a model of how to deal with a different normative realm, such as ethics. In his longer essay S-M suggests that it's conceivable that maybe there would even be something that justifies both our epistemic beliefs and our ethical ones. His example is a long shot, but it's possible, I guess. Enoch really seems to me to be operating in that model: trying to find what justifies our epistemology, and then reflecting that back on ethics.

There's another thing S-M wants to say, which is that if some argument would conclude that normative stuff isn't good in general it would be too strong. But there's a natural weakening of these arguments to give a prima facie reason not to believe in the existence of something. We can allow stuff into our ontology, even if it's problematic to do so.

OK, one thing out of the way: this probably gives a good argument against a specific interpretation of the argument from queerness. Specifically, if the argument from queerness gives us a reason not to believe in ethical stuff, it should also give us a reason not to believe in epistemic stuff. But queerness says "There's nothing else like it," but here we went and found something else like it.

Is there another way to go? I want to throw out the possibility that maybe the game's over once we have some normative beliefs, specifically epistemological beliefs. Here's one argument: being justified in something, we'll assume, has some aspect of normativity to it. And so here's one possibility. Let's say that we analyze justification as, "X is justified in believing Y" as "X ought to believe Y." And so epistemology is full of ought-statements about belief. Then the only thing that distinguishes ethics from epistemology is that ethics deals with a lot more actions that belief (assumption: belief is an action). But it's not just justification that's normative, because there are some actions that involve being justified. You know something only if you're justified in believing it.

Another way for the game to be over is if we showed that ethics was doing nothing more than justifying beliefs. "You ought to do X" could be reinterpreted as "You ought to believe that you ought to do X." And what would be wrong with that? Plenty. It's actually a remarkably bad argument. Moving on...

A third option is that ethics and epistemological norms are just quite different things. As different as ethics and etiquette. Well, then what? Then I guess we're back to Enoch and S-M's efforts. Which isn't necessarily a bad thing. It's just that Enoch is problematic (and I think the problems come from pushing epistemology too hard and by carving up our knowledge) and S-M doesn't really offer anything. So if there's anywhere to go, maybe it's by just saying "belief is an action, so why not?" but that's a pretty rotten argument too, because you could just as easily say "picking up your fork is an action, so why not?" Oy, still not getting anywhere.

Enoch on S-M

1.A.8 Sayre-McCord
In the last section of his (1988a) Sayre McCord presents – in just four pages – what is by far the most detailed and careful attempt at the second strategy I know of. Despite its lack of crucial details, this section clearly anticipates my line of argument for Robust Realism.
Sayre-McCord starts by noticing that the explanatory project – indeed, the explanatory requirement itself – is normative through and through, for it requires, at the very least, that we evaluate explanations, and choose the best one. He notes that there is little to recommend a view that is realist about some normative facts but not about moral ones, and so he concludes that our engagement in the explanatory project already commits us to evaluative, and so to moral, facts. There is just no way of engaging in explanation without relying on normative facts.
Sayre-McCord goes even further than that. He argues that the respectability of normative (and in particular moral) facts and properties does not depend on their indispensability to the explanatory project:
The legitimacy of moral theory does not require any special link between explanatory and moral justification. (280)

Instead, what guarantees the respectability of moral and other normative facts are their justificatory, not their explanatory, role:
Just as we take the explanatory role of certain hypotheses as grounds for believing the hypotheses, we must, I suggest, take the justificatory role of certain evaluative principles as grounds for believing the principles. (278)

This is, I take it, an explicit rejection of the first strategy of Harman’s Challenge, a rejection of the explanatory requirement, and the beginning of an argument for normative realism from a different kind of indispensability. Though Sayre-McCord does not use the term “indispensability argument”, he does often say that evaluative facts are indispensable (e.g., 279). And he suggests that we talk in this context, instead of an inference to the best explanation, of an inference to the best justification (ibid.).
Now, Sayre-McCord does not give some crucial details here (details which I try to give in this chapter and in the rest of this essay): What exactly does indispensability amount to? Why does indispensability to the explanatory and the justificatory projects justify ontological commitment? Why is there a justification-related need to invoke irreducibly evaluative facts and not just, say, psychological ones about one’s brute desires or preferences? Furthermore, in some important respects the line he seems to suggest is different from mine: For one thing, I cannot see how anything like inference to the best justification can be made to work . More generally, it seems to me the justificatory work normative facts do matters to us because of the deliberative indispensability of justification. What is intrinsically indispensable, in other words, is the deliberative project, not the justificatory one. The latter only matters because the former does. This is why I think the argument for Robust Realism is better put in terms of deliberative rather than justificatory indispensability.
Despite the lack of details and these differences, and despite Sayre-McCord’s commitment to Metaphysical Naturalism, it is clear, I think, that his suggestions anticipate – in broad outline, at least – my indispensability argument for Robust Realism.

Particularism versus Methodism

Sosa distinguishes between particularism and methodism. Methodism starts by outlining what the correct epistemological methods are, and then checking to see how much knowledge we have given those methods. Most people don't really do that, I don't think. Much more common is particularism in epistemology. This is a methodological starting point; you start with knowledge, and then you attempt to justify and understand what norms and principles give us that knowledge. It's not just an attempt to describe our knowledge, but that description is in fact normative in this case. If it turns out from our study of knowledge that people require observation in order to know stuff, then observation is normative, in the sense that you only ought to believe something if it's been observed. That's the methodology for a lot of epistemology.

You might ask, then, why not do the same thing for ethics? Say that we're assuming that there is ethical knowledge, and that we need to understand how that's possible. Then how we gain that ethical knowledge is normative. Given that we do the same thing for epistemology, how can we distinguish the two?

The obvious difference is that whether there is ethical knowledge is debated. But I suppose if one was quite sure, on a first-person level, that there was ethical knowledge, there's nothing to stop him for proceeding as we do in epistemology. But it certainly won't convince anybody. But it's not aiming to. Just as epistemology often says that it has no need to answer the absolute skeptic, a moral epistemologist could say the same thing. Of course, it won't convince the skeptic, but maybe one doesn't want to. This only works if you are REALLY sure that there is ethical knowledge, but given that someone is this certain then it's hard to think of a way to criticize their methods given that there's an exact parallel methodology in scientific/general epistemology.

Metaethics and Philosophy of Religion

Another fruitful parallel? I like when philosophy does this. Here's Plantinga:

If a man believes that the star Sirius has a planetary system containing a planet with mountains over 40,000 feet tall, then if his belief is to be rational or reasonable he must have some reason or evidence for it. Similarly, it may be said, with the existence of God: the theist must be able to answer teh question "How do you know or why do you believe?" if his belief is to be rational; or at any rate there must be a good answer to this question. He needs evidence of some sort or other; he needs some reason for believing. Obviously this raises many questions. What is evidence? What relation holds between a person and a proposition when the person has evidence for the proposition? Must a rational person have evidence or reasons for all of his beliefs? Presumably not. But then what properties must a belief have for a person to be justified in accepting it without evidence? Is a person justified in believing a proposition only if it can be inferred inductively or deductively from incorrigible sensory beliefs? Or propositions that are obvious to common sense and accepted by everyone?

Math, religion, ethics all face the same challenge: if I can't sense it, can I really know it? The answer might seem to be no, but the idea that we only believe what we can sense fails horribly. So we try to latch on math, religion or ethics to those principles that guide us when we leave the realm of observation. In math there is the claim that this is enough for numbers (though Parsons raises the objection that is quite parallel to that raised by Enoch towards the naturalist Cornell realists about losing the special nature of math/ethics by resorting to the naturalistic argument). If ethics and religion fails this test, does that mean that they're out? Plantinga says "no" in the case of religion.

A philosophy of math thought

Almost without fail, when I talk to my math friends about whether mathematical statements are true or false I'm told:

"Well, it depends what you mean by true or false. Mathematics doesn't make any substantive claims, we just make conditionals. You start with some axioms, and then deduce from there. So all mathematical knowledge is conditional: if the axioms are true, then this deduction is true."

Why isn't this quite right?

First, this avoids answering basic questions about really basic math. Is 2+2=4 true or false? Can we only talk about it being true in Peano's world, or true in ZFC's world, but can't talk about it being absolutely true or false? What explains our choice to study the worlds where 2+2=4 is true instead of false? Is it just by whim? Can we then imagine living in a world where 2+2=5? Or maybe we pick the system because it matches up with our world. But doesn't that mean that we believe that 2+2=4 is really true? In that case we are committed to the existence of actual numbers. And if we want to maintain that there are no such things as number, we then have to explain why we study 2+2=4, and we have to explain why that statement seems to incredibly true about our world.

Second, it evades answering questions about the axioms. How do you choose which axioms to study? Why are we picking some axioms instead of others? If we're picking axioms that properly describe the geometry of earth, does that mean that we're studying something factual when we study that geometry? When it comes to set theory, and the foundations of things like arithmetic, we might be less willing to be wishy washy about which axioms are true. This argument is somewhat dependent on the one right above.

Third, Hilbert tried to be a formalist, but most folks think that he failed because there are tremendous tensions in trying to believe that we just work with a bunch of formal systems that are consistent when these systems can't prove their own consistency.

Explanation and Justification

I'm reading Peter Lipton's book on IBE right now, and it's a good read with nice chapters on induction and explanation.

If you think about it inference to the best explanation is a really interesting principle, if it's used for justification of inferences. Meaning, consider the following: I see water coming out of my wall, the best explanation of this is that a pipe has burst. Therefore I am justified in concluding that a pipe has burst. But what justifies this conclusion? IBE. So inference to the best explanation plays a justifying role. We then ask, "What justifies IBE?" Well, that is what I've been trying to do. Enoch and Shechter give an account of epistemic justification that allows for pragmatic considerations (of a very specific type) to play a role in justifying our basic principles, such as IBE. I think that this runs into problems that are pretty big, but it is a subtle argument with intuitive appeal.

Another attempt, which occurred to me, is to simply take certain beliefs as primitive. IBE is unjustified, or justified by things that it justifies. This suggests a coherentist picture of some sort. Enoch is decidedly working in a foundationalist epistemology, one that thinks that justification is linear and asymmetric.

What else is there? You might think that the notions of explanation and justification are simply linked--maybe the reason why inference to the best explanation is justified is because what you're really doing is inferring to the best justification, so that IBE is never leading you to believe something that isn't justified. But this doesn't work, since explanation is actually quite different from justification. This comes out well in Lipton. One way to see this is that we have a much easier time believing that that certain beliefs explain themselves or dont' require explanations, while it's much harder to come across beliesf that seem to justify themselves, or require no justification. Also, explanation seems to FAR outstrip knowledge. In fact, we only begin to seek an explanation of a fact after we are secure in our knowledge that the fact is true! So it's quite a substantial claim to say that explanation guides justification, as IBE seems to do (I think). Obviously, explanation cannot be all there is to justification. Is explanation a specific type of justification? Er, I'm confused.

Nearing the half-way point

It's about half way through this project. So how am I feeling about it? I'm having fun, I'm learning a lot, but I'm finding it really hard. Now, I don't mind finding it hard. If philosophy were easy no one would be doing it. The difficulty and seeming insurmountably of a question is almost definitional of a philosophical question. Easy questions don't get asked in philosophy. But that's all a consolation prize--I really would like to have gotten a much better or deeper understanding of something, and I don't think that I've really gotten there yet. If I keep on plugging away will I gain some new insight on the issues that interest me? Maybe yes, maybe no. There's no way to know.

Here's a disappointment: for now my focus is turning away from the intersection of math and ethics and focusing squarely on the intersection of philosophy of science, epistemology and ethics. We'll see how far that goes. But for now it seems as if philosophy of math has been asked to sit this one out. In retrospect, this was a long time coming, since the power of the indispensability argument is its claim to just be a plain-old boring scientific and inductive argument. What makes it so appealing in philosophy of math is that it seems to be completely acceptable to the scientist.

I'm going to keep on reading about philosophy of math on the side, wondering if maybe there isn't some way in which the mathematical platonist goes beyond inference to the best explanation in order to reach his conclusions. And maybe if ethical realism fails, ultimately we can tie mathematical realism to that sinking ship. Anti-realism tends to be far less interesting to me than realism, though. My deeper convictions are that math and ethics are clearly different than plain old empirical knowledge in some important respects, but that they're completely respectable arenas in which disagreements can be rationally voiced and where certain claims are true and others are false. But I'm going to be open here, and if I find an argument that pushes me towards anti-realism, I'll have to deal with that.

In the mean time, here's where the project currently stands. I don't feel too hopeful about where it is right now. I don't believe that I'm on the verge of finding something. But anyway, here it is: inference to the best explanation is a respectable principle. But what justifies inference to the best explanation? Do you need to fall back onto something normative in order to get the whole thing going? Is that any different from just saying that justification is a normative concept through-and-through? And if some sort of normativity is needed in order to justify scientific claims, how much normative theory creeps in? The question is analogous to one in math after the indispensability argument: so math is necessary for science, but how much math? In the same way, maybe some normative theory is necessary for science (in the sense that in order be justified we need normative theory), but how much normative theory? Does moral theory creep in?

I sincerely think that the answer is "no," but this is likely to be the topic of my half-way point paper that I'm writing this week. To be honest, I keep on hoping that I smell a trail that will send me in a slightly different direction. I think that this direction is absolutely fascinating, but I don't think that it's a very promising one for ethics. But here we go, anyway!