Against Contrastivism

A conversation last night with Yuri and Andy helped me to get clearer on the argument I was trying to press in my last post. Here's the much more succinct way to make the point. It's an argument against forms of contextualism that put relevant alternatives into the proposition expressed by knowledge attributions.

Suppose I'm in a nonskeptical conversation, talking about Henry, who is standing in front of a barn. I have no reason to suspect any funny business, so I say, sensibly enough:

(K) Henry knows that he is standing in front of a barn.

Here are are three pretty plausible claims:

(1) If there isn't any funny business going on, my utterance of K is true.

(2) If it turns out that (unbeknownst to me) Henry is in fake barn country, (looking at the only real barn) my utterance of K is false.

(3) My sentence (K) expresses the same proposition, whether or not it turns out that Henry is in fake barn country.

If you think all of these things, then you can't think that the proposition I express builds in the relevant alternatives. Either the possibility that <the thing Henry is standing in front of is a fake barn> is relevant, or it's not, but it's his environment, not my context, that makes it relevant.

So if you want a Schaffer-style extra-argument-place approach to knowledge, this provides a reason not to let that argument place be one for a set of relevant alternatives. You might instead be a function from subject's situations to sets of relevant alternatives.

Yesterday I also included a parallel argument relying on pragmatic encroachment sorts of cases. I think it's a good argument too, but this one proceeds on less contentious premises.

Are 'epistemic standards' contrast classes?

Suppose you agree with Jonathan Schaffer that 'knows' takes an extra argument place, and that variation in what fills this slot explains the context-sensitivity of 'S-knows-that-p' attributions. Knowledge relates, say, a subject, a proposition, and, let's call it, the 'epistemic standard'. Nothing yet is assumed about what sort of thing that is. Schaffer thinks the epistemic standard is a contrast class, or, equivalently, a question. So "George knows he has hands" can express either of these three-place knowledge relations, depending on the context:

(1) K (George, <George has hands>, {<George has hands>, <George is not the sort of creature who has hands>, <George has lost his hands>})

(2) K (George, <George has hands>, {<George has hands>, <George is deceived by an evil demon into thinking he has hands>})

We could disambiguate in English, saying, "George knows he has hands rather than stumps", to get something like (1), or "George knows he has hands rather than being deceived by an evil demon into thinking he has hands" to get (2).

Identifying the epistemic standard with contrast classes does much of the work that a contextualist might legitimately seek -- for instance, it disambiguates the sentence above into a modest one and one inconsistent with skeptical intuitions -- but I think there might be a few reasons to prefer something different. Here are two related challenges to contrastivism that are not, I think, general challenges to the shifty variable approach.

Gettier cases. Suppose Henry's in fake barn country, but you and I don't have any reason to think that's so. (Neither does he.) I say to you, "Henry knows that he's looking at a barn." My sentence is ambiguous; what's the shifty epistemic standard? We can get an anti-skeptical reading with a contrast class of, e.g., barns and silos. But the antiskeptical reading in this case is very counterintuitive. To get the skeptical reading, which is the standard one, we need the contrast class to include fake barns. But given our situation, it's quite mysterious how the fake barn possibility got to be part of the semantic content of my sentence. (Compare the possible world in which everything seems exactly the same to all three of us, but Henry is in real barn country.)

Low-Attributor, High-Subject Bank Cases. Hannah and Sarah are desperate to get to the bank before Monday, and they have good but not great reason to think they can go tomorrow. You and I don't know and don't care how important it is to them; I say to you: "Hannah knows the bank will be open tomorrow." Here are two representative possible contrast classes: {<bank open tomorrow>, <bank not open Saturdays>}; {<bank open tomorrow>, <bank recently changed its policy and will be closed tomorrow>}. The latter is needed for the skeptical reading, which is the intuitive one; but what about me and my context could make it the case that this skeptical possibility is relevant? Again, compare the corresponding low-subject stakes version, which seems just the same to me.

Now I'm a good semantic externalist. So don't take me to be arguing that in each case, the intrinsic similarity entails sameness of semantic content. The argument can't go that directly. Nevertheless, in the relevant cases, it does look to me pretty strange that fake barns should appear in my content only if Henry happens to be in fake barn country, or that the policy-changing case is part of my content only if Hannah and Sarah's stakes are high. Intuitively, these features to which I'm blind are relevant to the truth of the knowledge attributions, but they are not relevant to their truth conditions.

Contrary to contrastivism, the subject is not irrelevant; the subject's practical and environmental features do play some role in determining what alternatives are relevant. This, to me, suggests that we might not want to put the relevant alternatives themselves into the semantics of my sentence. Don't make epistemic standards contrast classes; we can let the shifty epistemic standard be something else. Here's a modest suggestion: epistemic standards are functions from situations to contrast classes. A given standard tells you, for any situation the subject might be in, which possibilities are relevant. The speaker's context fixes the standard; the standard and the subject's situation fix the relevant alternatives. So speaker and subject are both 'relevant'.

Notice, by the way, that you don't have to go along with the encroachment stuff  to prefer this treatment to Schaffer's. The Gettier case provides, I think, a less contentious way of motivating just the same point. If you're one of those contextualists who is motivated in part by denying pragmatic encroachment, then you should think that no standard will deliver different alternatives for situations that differ only in stakes. (So you'll explain away intuitions about bank cases.) But unless you're also willing to explain away intuitions about Gettier cases, you should still have the standards be sensitive to the subject's environmental situation.

Contextualism, Indexicalism, and Constrastivism

One of the questions I've been thinking about lately (unrelated to most of my recent blog posts) concerns the best linguistic implementation of contextualism about 'knows'. I've committed myself to contextualism in a couple of papers, but so far I have tried to avoid a commitment to any particular semantic treatment of 'knows'.

I take it the minimal commitment of contextualism starts with something like this: statements of the form 'S knows that p' can express different propositions in different conversational contexts. And to this, presumably, we add that the context-sensitivity of these sentences derives from the 'knows' -- it's not enough that various singular terms (the 'S's') or statements of propositions (the 'p's') are sometimes context-sensitive. But this leaves open some choices for what to do with 'knows'.

One choice is to treat 'knows' as an indexical, which refers to different relations in different conversational contexts. Call this "indexicalism." The other choice is not to treat 'knows' as an indexical, suggesting that something about it generates context-sensitivity in some other way. (It is unfortunate that 'non-indexical contextualism' has been used as a name for a different view, which is not a contextualist one by the standard of the previous paragraph.) Jonathan Schaffer has argued for a version of contextualism of this latter type. Schaffer's view, 'contrastivism,' is that 'knows' univocally picks out a three-place relation, relating a subject, a proposition, and a contrast class. The contrast class is often left implicit in sentences of the form 'S knows that p', and so it is filled in tacitly; since different contexts will suggest different ways of filling it in, 'S knows that p' ends up expressing different propositions in different contexts. (Schaffer uses his terms differently, too; he calls my indexicalism 'contextualism'.)

Schaffer's isn't the only way of being a contextualist who doesn't treat 'knows' as an indexical. Contrastivism is only one example of a view of this kind -- this kind of view needs a name! I really want to use 'non-indexical contextualist'... do y'all think I could reclaim that label? I think the view I'm describing is extremely well deserving of that name... Anyway, whatever you want to call it, there are lots of things besides contrast classes that might be arguments for 'knows'. I rather suspect that a view of this sort is correct, and one of my projects at the moment is to articulate such a view and explain why it might be preferable to any other form of contextualism.

So we have at least three different forms of contextualism on the table:

1. Indexicalism. 'Knows' is an indexical; which epistemic relation it expresses depends on the conversational context. As far as I can tell, no one has actually seriously defended this view, even though it's often taken to be the standard claim of contextualism.


2. Contrastivism. Jonathan Schaffer's view. 'Knows' univocally expresses a ternary relation Kspq, relating a speaker, a proposition, and a contrast class. This last is often filled in by the conversational context.


3. Non-contrastivist extra-argument-place views. (I guess we need a name for this one too.) 'Knows' univocally expresses a ternary relation Kspx, where x is something other than a contrast class, and is often filled in by the conversational context. Until we get more specific, we might think of x as standing generically for an 'epistemic standard'.

I'm curious as to whether these exhaust the options for the contextualist. They do if the only ways that 'S knows p' can be relevantly context-sensitive are for 'knows' to be an indexical, or to take an extra argument place supplied by context. Anybody see any more choices here?

The Knowledge Norm of Practical Reasoning

For reasons exactly like the ones outlined in the previous post, these two claims are importantly distinct:

(1) If S knows p, then S can appropriately rely on p in practical reasoning.

(2) If S knows p, then p is warranted enough to justify S in phi-ing, for any phi.

I argued a couple of weeks ago that (1) is not strong enough to establish pragmatic encroachment. I suggested then that this was a problem for Fantl and McGrath; the discussion with Jeremy in the comments thread is part of what helped me to see the distinction between (1) and (2) more clearly. If their argument proceeded from (2), rather than (1), my argument doesn't apply. However, in light of the important distinction between these two claims linking knowledge and action -- I won't speak for anybody else, but this is a distinction I certainly hadn't been thinking clearly about before recently -- we should, if relying on claims like (2), proceed carefully, distinguishing arguments for (2) from arguments for (1).

I gave in my most recent post, linked at the top, an argument against a strictly weaker principle than (2). If that argument was right, then (2) is false. Let phi be an action that, for accidental reasons concerning the background environment, is only justified if S has some super-knowledge access to p. The example from that post was, let phi be the assertion that p, and let the background be such that S has promised, in a morally weighty way, not to assert p unless S knows that she knows that she is absolutely certain that p; let p be known, but the higher condition not be met. Then S is not justified in phi-ing, under the circumstances, even though she knows p, and even though, were p better warranted, she would be.

But maybe that argument went a little too quick. For maybe it's question-begging to assume, as I did, that one can know p without being in the super-epistemic position, under the circumstances described. Maybe the act of promising collapses that distinction. If so, then my argument against (2) can be resisted. Indeed, it's sort of the point of (2) that the 'standards' for knowledge raise to as high a level as one might need in any given circumstance.

But -- and here's the main point of this post -- one can retain (1) without collapsing that distinction. That's another respect in which (1) is interestingly different from (2).

Assertability and Norms of Assertion

Here's a crazy thesis that nobody holds:

(1) If S knows that p, then S is permitted to assert that p.

There are boring counterexamples to (1). For instance, there are cases in which I am morally forbidden from asserting things that I know. This, of course, shows nothing interesting about the relationship between knowledge and assertion. So to capture the content of the interesting claim in the neighborhood, we move from (1) to something subtler and less crazy; maybe something like one of these:

(2) If S knows that p, then no epistemic shortcoming in S can explain why asserting that p would be inappropriate.

(3) If S knows that p, then S is in a strong enough epistemic position to assert that p.

(4) If S knows that p, then improving S's epistemic position won't put S in a position to assert p.

Maybe (2)-(4) are equivalent to one another; I'm not sure. They do a better job than (1) does at rendering certain kinds of cases of knowledge without assertability irrelevant. But they, too, are subject to conclusive refutation in a way orthogonal to the knowledge norm of assertion. My argument against (1) concerned cases in which I'm morally prohibited from asserting something, even though I know it -- maybe telling the Nazis where the Jews are hiding or something like that. But there are also cases in which I'm morally prohibited from asserting something, even though I know it, where a failure in my epistemic position plays a role in explaining the prohibition. There are boring cases like this. For instance, suppose that I've made a promise to assert that p only if I know that q. (Maybe promising by itself is insufficient for the relevant duties; stipulate that we're talking about a promise that carries serious moral weight.) Now suppose I know p, but don't know q. It's impermissible to assert that p, even though I know p -- and even though a deficiency in my epistemic position plays a role in explaining why p is unassertable. These cases, too, show that (2)-(4) don't get to the heart of the matter.

And it won't help to relativize the claims about strength of position to p, either. That is, these attempts at the knowledge norm face the same problem:

(5) If S knows that p, then S is in a strong enough epistemic position with respect to p to assert that p.

(6) If S knows that p, then improving S's epistemic position with respect to p won't put S in a position to assert p.

(7) If S knows that p, then p is, for S, warranted enough to justify S in asserting that p.

For take the special case of the sort of example given above where q is a proposition about S's epistemic position with respect to p. For instance, suppose I've promised not to assert p unless this condition is met: I know that I know that I'm absolutely certain that p. Suppose also that I know p, but don't know that I know that I'm absolutely certain that p. Under the circumstances, p is unassertable, because my epistemic position with respect to p isn't strong enough. But that doesn't show us anything interesting about the knowledge norm of assertion, if the latter is meant to be understood as showing something interesting distinctively about assertion.

What the knowledge norm of assertion suggests is that there's a special way that assertions can fail qua assertions. It says that if S knows p, then an assertion that p doesn't fail in this particular way. It doesn't provide any sufficient conditions for not failing in some other way, even when you build in all of these conditions about S's epistemic position.

Therefore, none of (1)-(7) are closely related to:

(Norm) Knowledge is the constitutive norm of assertion.

This post has gone on long enough for now, but I'll close by just asserting that many attempts to argue against the knowledge norm of assertion really look like arguments against some or all of (1)-(7); this is a mistake. I've argued that (1)-(7) are definitely false, for boring reasons that don't have anything to do with assertion in particular. If all you have is a case along with intuitions about what is known and what is unassertable, and why it's unassertable or under what circumstances that unassertability would be rectified, then you don't have anything strong enough to speak directly to (Norm). To evaluate (Norm) via the method of cases, you'd need to have intuitions about whether the assertion suffers from a particular kind of failure qua assertion. These, I think, we rarely have at any kind of pretheoretic level.

Knowledge Norms and Pragmatic Encroachment

I'm thinking a bit more today about the point I made in a post yesterday about the use of intuitions about cases to evaluate knowledge norms. That point was basically that facts about whether S knows p and whether S is well-enough situated epistemically in order appropriately to X don't by themselves say anything about the knowledge norm of practical reasoning; S may know p without X's being appropriate just by virtue of p's not being a good enough reason to X. Yesterday I used this observation to rebut a certain kind of argument against knowledge norms.

I now think that, in addition to that use, this observation undercuts certain implications sometimes drawn from knowledge norms. In particular, I think it points to a lacuna in one of the central arguments of Fantl and McGrath's defense of pragmatic encroachment. I think this is a pretty fair reconstruction of their §4.3:

1. KJ: If you know that p, then p is warranted enough to justify you in phi-ing, for any phi.


2. Consider some low-stakes action, X, for which LOW has q as a sufficient reason; LOW can appropriately perform X. (Their example: Matt knows the train is a local; this justifies his boarding it.)


3. There is a possible counterpart of LOW in the same epistemic position but with higher stakes, HIGH, such that HIGH cannot appropriately perform X. (Their example: Jeremy really needs to get off at a local stop; the stakes are too high for him to risk boarding.)


4. So q is not well-enough warranted to justify X in phi-ing.


5. So HIGH does not know q.


6. So two subjects in the same epistemic position can differ with respect to knowledge of q.


7. So purism is false.

(For maximum faithfulness to F&M's broader intentions, we should understand this entire argument, including its conclusion, as being offered under the conditional assumption that fallibilism is true -- that it's possible to know that p, even though there is some epistemic possibility that not-p.)

I'm concerned with the move from (3) to (4). It has pretty much the same form as the arguments sketched above, except that it's holding different bits fixed. Jessica Brown argued from what she saw as intuitive verdicts about knowledge and appropriate action against the knowledge norm; Fantl and McGrath argue from intuitive verdicts about appropriate action and the principle of the knowledge norm against the knowledge verdict Brown found intuitive. My argument shows the flaw in both of these arguments. (In a sense, they are instances of the same argument, run in different directions.)

The move from (3) to (4) relies, like Brown's surgeon case argument, on the assumption that the knowledge of q, the non-actionability of X, and the knowledge norm are incompatible. But as I've shown, they're not. They're incompatible only on the assumption that q is, if possessed as a reason, a good enough reason to X. And there just aren't obvious intuitive verdicts about facts like these; neither are there clear theories that dictate which claims like these to accept. It's obvious enough whether S knows q, and whether S would be justified in Xing -- but whether q itself is well-enough justified to be among the reasons S has for Xing is an esoteric question on which naive intuitions are silent.

Everybody who accepts (2) and (3) has to think there's some important difference that derives from a change in the stakes that bears on actionability. But you can think this without giving up on KJ, fallibilism, or purism, if you want to. You can say that what propositions are good enough reasons to X depends in part on the stakes. That is: whether p is known is stakes-independent; so too is whether p is warranted enough to be a reason for phi-ing, for any phi. What varies by stakes is whether p, supposing that it is a reason, is by itself a good enough reason to phi. When the stakes are high, p, though still genuinely a reason, isn't a good enough reason. In lower stakes, p is a good enough reason. Insensitive knowledge; insensitive reasons for action; sensitive needs of actions for reasons.

This does look to me like a significant and substantive gap in F&M's argument. I'm generally pretty sympathetic to F&M-style views -- I'm a different sort of contextualist than those who are motivated by 'intellectualism' -- but this does look to me to be a potentially promising avenue for resisting pragmatic encroachment from F&M-style arguments.

It will not obviously help, however, with the knowledge intuitions about bank cases and the like. Lots of pragmatic encroachment people seem to be turning their emphasis away from these cases recently, in favor of broader theoretical arguments. Certainly, this seems to be one of F&M's aims. I rather suspect, though, that the pragmatic encroachment theorist may end up needing to rely on judgements about cases more than they think. (I don't know that that's a problem.)

Knowledge Norms and Intuitions about Cases

Here's a boring thought experiment that doesn't demonstrate anything.

Smith burgled the house last night; Detective Stanley is investigating the crime scene. He acquires evidence sufficient for knowledge that the burglar came in through the window, but finds very little evidence about whether it was Smith or someone else who committed the crime.

Here are two intuitive verdicts that aren't in any tension at all:

1. Stanley knows that the burglar came in through the window.


2. Stanley would need to have more evidence in order for it to be appropriate for him to arrest Smith.

Everybody can accept these obvious claims. In particular, these obvious claims are in no tension with the knowledge norm of practical reasoning, which claims that p can be an appropriate reason for action for S if and only if S knows that p. It would be an anemic objection to the knowledge norm to point out that Stanley knows that the burglar used the window, but needs more evidence in order for it to be appropriate to arrest Smith. That the burglar used the window just isn't a strong enough reason to arrest Smith, so this case doesn't tell us anything about what is and is not a reason. So it doesn't tell us anything about knowledge norms.

The moral of the story is that claims about who knows what, and about what actions are inappropriate, are in general insufficient to refute the knowledge norm of practical reasoning. (So, mutatis mutandis, for the knowledge norm of assertion.)

When you look at the case given above, this moral is really obvious. But sometimes, I think, it's neglected. Jessica Brown, for instance, argues against the knowledge norm of practical reasoning by citing this case:

A student is spending the day shadowing a surgeon. In the morning he observes her in clinic examining patient A who has a diseased left kidney. The decision is taken to remove it that afternoon. Later, the student observes the surgeon in theatre where patient A is lying anaesthetised on the operating table. The operation hasn’t started as the surgeon is consulting the patient’s notes. The student is puzzled and asks one of the nurses what’s going on:

Student: I don’t understand. Why is she looking at the patient’s records? She was in clinic with the patient this morning. Doesn’t she even know which kidney it is? Nurse: Of course, she knows which kidney it is. But, imagine what it would be like if she removed the wrong kidney. She shouldn't operate before checking the patient’s records.

We have, as before, a pair of intuitive verdicts: one attributing knowledge, and another denying appropriateness of action. Brown considers this to be a counterexample to the knowledge norm of practical reasoning, but the case of the burglar shows that this cannot not enough. Just as the burglar argument was transparently invalid, because the burglar's use of the window wouldn't be sufficient reason for arresting Smith, Brown's argument is valid only on the assumption that the disease in the left kidney would be a sufficient reason for operating without checking the charts. But Brown has given us no reason to think that is so. It's entirely open to the defender of the knowledge norm to argue that knowledge of p is sufficient for p to be a reason, but that in this case, p isn't a good enough reason for action.

This strategy is always available. I think this shows that trading in intuitions about who knows what, and who ought to do what, is not a helpful strategy for evaluating knowledge norms.