Why epistemology, again

Just a reminder:

There are tons of things for which we have reasons. We have reasons to fear, reasons to hope, etc. These are reactions that are conceptually distinct from belief. So why think that the study of what reasons there are to believe things might help vindicate the study of what moral reasons we have to do things? Because unlike most of these areas, it's hard to pass off epistemic statements as being second rate from a cognitive point of view. If epistemic statements can't be true or false, then we would seem to be in a boatload of trouble. And what we really want is to distinguish between reasons to fear something and reasons to believe something, and say that reasons to act morally are more like the latter than the former.

At least, that's the idea. If reasons to believe are ultimately not true/false then all of our knowledge seems to be in trouble.

Anselm on Justice and Injustice

T: Remember also that we've already agreed that once justice has been lost, nothing but the obligation to justice remains in him beyond what he had before he received justice.
S: Right.
T: But before he had justice, he was not unjust and did not have injustice.
S: No.
T: Therefore, either there is no injustice in him and he is not unjust once justice has been lost, or else injustice and being unjust are nothing.

This is Anselm in "On the Fall of the Devil." I think that the point that he's making here is that you require principles of justice and morality in order to criticize one for failing to live up to those norms. When one completely lacks the norms of ethics, one can't be criticized for failing to live up to those norms.

I believe that this is related to the point I've made about things being epistemically optional. One cannot be criticized for failing to live up to the epistemic norms before one has accepted epistemic normativity in some way.

Another post on Enoch and Shechter

I don't think that I properly understood what Enoch and Shechter were trying to do until I started to think things through on my own. I was initially puzzled by their article(s). How could they hope to provide any sort of justification that itself goes beyond justification? OK, that sentence was incomprehensible, lemme try again. Here's what I thought that they were trying to do: provide a justification for a principle such as inference to the best explanation. What's wrong with that? Well, they also claim that inference to the best explanation (IBE) is basic, and isn't justified by anything else. You don't have to be that deep of a thinker to recognize a problem here: they want to justify something that itself is unjustifiable. What?!

In retrospect, I feel kind of stupid about this picture of their work. It's awfully simplistic. I always knew that this couldn't be quite what they were getting at, but I didn't know what they were aiming at either.

Here's what I think they're trying to do:

It's hopeless to try to justify the truly, truly basic beliefs. Some beliefs really are unjustified--for example, the first epistemic beliefs that you take on will be unjustifiable. After all, if all you have is a knowledge that epistemic realism is true (and so you are able to entertain the possibility that some given epistemic sentence is true) how are you able to evaluate the truth of some epistemic sentence? You have nothing available to justify it with, since you don't know of any other epistemic sentences that are true. On the other hand, this also means that you can't be rationally criticized for accepting this sentence as true, since the epistemic resources for challenging any belief-practice epistemically simply aren't available yet. So as far as the truly basic beliefs go--they're optional in a particular sense. Indeed, this is what I argue for in the first chapter. And so you can't expect this to help ethics at all, since ethics is not optional in this sense at all.

What about Enoch and Shechter? Are they simply trying to justify the unjustifiable basic beliefs? This would be silly. And we know that they are trying to provide justification ("How are Basic Belief-forming methods justified?" is the title of their article, after all). So the most plausible reading is that they're not actually trying to justify sentences that are unjustifiable (duh, right?).

So what does it mean to be a basic sentence? All it means is that it's a basic method. Here's an important quote: "We cannot justify our use of IBE by appealing to other belief-forming methods, since IBE is a basic rule. Thus, there is nothing in virtue of which we are justified in using IBE. Or so it may seem." I think that the idea in this sentence is that there are no methods--no other beliefs?--upon which IBE rests in order to gain justification. But justification is still possible, they argue, because pragmatic considerations can provide epistemic justifications. Then they give their account.

It remains unclear to me in virtue of what the pragmatic account doesn't deserve to be called a "belief-forming method" though. Because, if my understanding of them is correct, what they're saying is that a sentence like "a thinker who does not inquire about the world around him is intuitively doing something wrong" determines what has positive epistemic value. That is, what I think they're saying is that what is truly basic in epistemology is not the belief-forming methods, but rather certain projects that are "intrinsically indispensable"--certain projects have great, positive, epistemic value. Why not just say that there is a foundational belief-forming method: "You are justified in believing that which is indispensable to any of the following projects: explanatory, deliberative, etc." That seems just as much as a method as the other stuff. And if I've understood them correctly, this is much more clear then bringing in all these pragmatics.

You would still have to take this new, foundational principle for granted, but you're going to have to take something for granted (and you won't be able to be criticized for it--how could someone even begin to criticize you?)

I think that I still have worries about Enoch's argument that ethics gets in via this way of seeings things. But at least this makes way more sense to me than my first few readings.

Intrinsic indispensability of epistemology

I borrow the phrase "intrinsic indispensability" from Enoch, and I use it in a much more limited capacity than he does.

As Cuneo points out, there is no non-question-begging way to argue against epistemic nihlism (and let's pretend middle positions between nihlism and realism don't exist for this post, to make my life slightly easier) and for epistemic realism. Why? Because if epistemic nihlism is true then there are no arguments, evidence or reasons to believe anything. Hence, any argument against epistemic nihlism would beg the question against epistemic nihlism, simply by presupposing that there is anything such as an argument, or anything such as categorical reasons to believe anything.

So one can't argue against epistemic nihlism. Rather, one simply has to reject it for no epistemic reason. Your rejection of epistemic nihlism has to be because you want to, not because you have a good reason to. I think that this is a fairly secure conclusion.

Now, if you are embracing epistemic realism (because I'm pretending nihlism and realism are the only options) what exactly are you embracing without justification? Presumably, that epistemology can be true/false, some epistemic statements are true, and that they're true in a non-reductive way. But if this is all we embrace, we're still going to be lost, because we don't know which epistemic statements are true or false. This is more than a problem of epistemic access, but it's a built in problem; how can we justify belief in statements concerning what counts as good evidence or justification before we've determined the truth of ANY statements about what counts as good evidence or justification? There seems to be no good way to get started in the epistemic enterprise. So if one is going to reject epistemic nihlism and embrace epistemic realism, it's not enough to just embrace some kind of abstract realism, but one must also take certain epistemic beliefs for granted. (Unless you want to advocate the view that epistemic realism is true, and we just don't know anything about epistemology. This seems bad.)

I wonder if this is part of the answer to the challenge of epistemic access that exists parallel to epistemology in ethics. This is how epistemology gets started, and there's no corresponding story about ethics to be told.

Being more precise

We might even agree with Cuneo that disagreements such as "It is rationally permissible to cling to religious beliefs in the face of contrary evidence provided by our best science" is a first-order epistemic sentence. But note that it is a general, as opposed to a particular one. If we were to say, instead, "We're justified in believing that evolution is false because the Bible says so" the sentence is no longer plausibly viewed as an epistemic sentence. Rather, it's a scientific or religious sentence, but if this is an epistemic sentence than ALL sentences are epistemic, since from any plain old descriptive sentence "There's a tree over there" we can get a normative claim such as "I know that there's a tree over there" or "I'm justified in the belief that there's a tree over there." The point is that when it comes down to specifics and particulars it's hard to see how you could have a first-order epistemic sentence.

My futher claim is that the kind of disagreement that we should find interesting is first-order particular disagreements. This is the kind of lasting disagreement that we find striking in ethics.

This is still wrong, because I'm abusing the particular/general distinction, but I think that there is something here.

Summing up the previous post in a few sentences

What I began expressing in the previous post was the following picture:

There are some statements that are first-order ethical statements. There are also second-order statements about ethical concepts. There are also some second-order statements about epistemic concepts. But I'm tempted to think that there are no first-order epistemic statements. This is because the first-order sentences about what to believe don't seem to me to really be about epistemology, but the purely epistemic sentences seem to me to be second-order.

My reason for thinking in this way is from Cuneo's example of religious disagreement. It seems to me that if that's an instance of epistemic disagreement, then it's because what these people are really arguing about is the second-order issue itself, directly. Otherwise, it's not really an epistemic disagreement, because if that was considered an epistemic disagreement then ANY disagreement would be an epistemic disagreement, since every disagreement is epistemic in the way that religious disagreement is. Why? Because all disagreements can be translated into epistemic terms, since all disagreements are concerned with what is rational to believe, or what one ought to believe. In other words, they're all arguments.

Inching towards this argument

I keep on thinking I have an argument here, then thinking that I don't. I'm going to try and get this down on here, and maybe it'll make it easier for me to get it on the actual page later tonight.

Let's begin with moral disagreement. There's an argument from disagreement against moral realism. In short, it says that the kind of disagreement that we find in ethics is unable to be resolved rationally. However, the argument goes, when disagreement has this quality it's indicative that there is no fact of the matter that the two parties are disagreeing about.

Cuneo argues that epistemic realism faces the same challenge. What's the evidence? Consider religious disagreement. The disagreement there is deep in the same way that ethical disagreement is deep. That is, there seems to be no way to resolve this disagreement rationally. Now, Cuneo suggests that this disagreement about religious issues is really itself actually about epistemology, not religion:

"since the predicate 'epistemic rationality' in [the context of religious debate] is probably best understood as one that stands for a constellation of thicker epistemic concepts such as wisdom, understanding, judiciousness, epistemic humility, epistemic honesty, and objectivity, the disagreement in question is plausibly viewed as being concerned with whether religious beliefs exhibit these thicker epistemic merits."

Cuneo's argument is that the religious disagreement is actually epistemic disagreement. He explains that the disagreement is between those who believe that one ought to believe in the authority of holy books over all other evidence, or that "rationality fundamentally consists in obedience to God and religious authorities" and those who maintain "non-theistic epistemological pieties [who] reject all this. According to these views, obedience of disobedience to God plays no role in explaining why we succeed or fail to understand the world aright."

It's hard to buy this claim. It would seem that if this is what it takes to be an epistemic disagreement, every cognitive disagreement at all could be translated into an epistemic one. After all, isn't a disagreement between two scientists who (as they did in the past) wondered whether the evidence supports the existence of black holes be a disagreement over two detailed epistemic accounts of how to deal with empirical evidence? Of course, Cuneo might respond that this just strengthens the point: epistemic disagreement is everywhere.

I think that there's a distinction that should be made at this point: some confusions are philosophical, and others are not. If you are reading this...hell, no one is reading this. But suppose that someone was reading this. That person might, at this point have all sorts of alarms going off: "Oh no," this hypothetical reader says, "he's going to try to distinguish between philosophy and the rest of knowledge. This is going to be a mess." Yes, it is going to be a mess. But here we go anyway. Some disagreement or confusion is philosophical in nature. This disagreement has all the qualities that we claim ethical disagreement to have--it seems to be rationally unresolvable, as evidenced by a long history of open problems where little progress has been made.

(And of course there has been philosophical progress, but there are still many problems where progress hasn't been made, and it's not like that's a controversial statement. What might be annoying to some is that I call this philosophical disagreement, since I might besmirch some issues that have had success. Fine, I'll pick another thing to call it later. For now, philosophical disagreement.)

It's not that shocking that there is this sort of disagreement with these sorts of philosophical problems. These specifically are the hardest, deepest, most fundamental problems that we know of. That's part of the reason why we study them. So it's no great shocker that we disagree in philosophy, and it's not at all like disagreement about which flavor of ice cream is best. The latter has unresolvable conflict because there's no objective matter of fact about which ice cream is the best. In philosophy the reason why there is this sort of disagreement is because philosophical problems are specifically the hardest, most fundamental problems that seem to push at the limits of human knowledge.

Now, I think that religious disagreement of the sort that Cuneo talks about is properly philosophical. This is because the disagreements he is talking about are about the very nature of knowledge and justification: what does it mean to be justified? What does it mean to be rational? I don't need to give an account of the "philosophical" in order to claim that these are paridigmatically philosophical questions.

But this, I think, isn't the case when it comes to ethics. When it comes to ethics I think that there are plenty of non-philosophical questions that can be asked that are properly ethical. "Abortion is evil" Obviously, the meaning of a sentence depends to an extent on the meaning of the words in it, so what it means to be evil will matter when we determine the meaning of this sentence. But the sentence isn't ABOUT good and evil. The sentence is about abortion. OK, OK, OK, hypothetical reader. Yes, that's the sort of stupid thing you expect a really stupid undergrad to make. But is there a more precise way to put this weird intuition that I have? Is there anything here at all?

Let me try again. We know that there are certain typically philosophical questions, and among these we're used to the fact that there is disagreement. Now, I'm going to use an argument here that I'm just ripping off of Justin Clarke-Doane. Consider the disagreement about abortion. Now, the two people disagreeing might be deluding themselves by thinking that they can disagree about abortion with each other without the disagreement coming down to a disagreement about what is good/evil, and that in order to really take this argument down to it's rational roots they would need to advocate a theory of morality. That's fine. But that's not what's in their heads. What's in their heads is that they're disagreeing about abortion and whether you can do it or not. I take this to be a non-philosophical ethical disagreement. Again, they're not having a fight about the theory of good.

This, I think, is a difference between ethics and epistemology. People don't have non-philosophical disagreements about epistemology, but they do have non-philosophical disagreements about ethics. But it's the non-philosophical disagreements that indicate that something weird is going on, because, as I noted, we expect deep and difficult disagreement in philosophy. What's weird about ethics is that there's disagreement that has nothing to do with the theory of good, and this indicates that maybe people are just talking about something completely non-objective.

To elaborate a bit: "Believing what the Torah says is justified." "No, it's not." What is this sentence about? Is it about the Torah? No, I think it's about the nature of justification. Note also that belief in the nature of justification, the second-order stuff, that's not normative even though it's epistemic in a sense. Statements about the nature of justification aren't normative even though justification is a normative concept. I think that's another way of putting the point. They're not having a normative disagreement when they're having a philosophical one; it's not even really an argument about epistemology. It's an argument about philosophy. So maybe I should define philosophy as the second-order claims about epistemology, though maybe I should just dump "philosophical disagreements" for "second-order."

If this is so, then I think that this is a reason why the argument from disagreement applies to ethics in a way that it doesn't apply to epistemology. I'm going to sit with this point for a bit now, and see how I feel about it. Maybe I can make it stronger.

Note that if Cuneo were to read this, he's likely say "this person is massively misguided." Partly because in his version of the argument from disagreement the only role that first-order disagreement has is to indicate the presence of second-order dsiagreement. But I find this confusing: it doesn't seem to me part of any of the other versions of the argument from disagreement that I've read, and in fact it seems odd. Second-order disagreement about concepts is normal, to an extent. What's at issue in the argument from disagreement is what the best explanation of unresolvable conflict is, and the claim is that the best explanation of this is that there is no factual disagreement at all--and this seems true even if there is complete second-order agreement (we might all agree what it means when we talk about best flavor but still have unresolvable conflict here, no?). So I remain a bit puzzled with tihs, and I'm not sure what to do with it.