The Mystery of the Disappearing Diamond

Jonathan Kvanvig has an interesting paper on Fitch's knowability paradox in which he argues that the paradox is a puzzle for everyone (not just those who believe that if a proposition is true then it is knowable). The puzzle is one of modal collapse; Fitch's argument shows that:
1. p --> possibly Kp
(if p then it is possible that p is known by some being at some time)
implies:
2. p --> Kp
(if p then p is known by some being at some time).
But how come the possibility operator can just disappear like that? Everyone has work to do explaining this, according to Kvanvig.

Thinking about the way the Fitch argument works suggests the following explanation:
Nothing of the form (p and not Kp) is knowable. But given 1, if anything of that form is true, then it is knowable. That's why, if 1 is true, nothing of the form (p and not Kp) is true. And that's why if 1 is true then 2 is true too.

Kvanvig must think there's something unsatisfying about this simple explanation. But what? Do we need further explanation of one or more of the claims it draws upon? I don't think so, but even if we do it could surely be given. More plausible, I think, is the thought that the simple explanation doesn't really explain the disappearance of the possibility operator; it just explains why 2 follows from 1.

If the worry is something like this then it is interesting. What counts as a good explanation of a fact does plausibly depend on (among other things) the way the fact is presented. The thought here would be that presenting the implication of 2 by 1 as a case of surprising modal collapse makes the simple explanation offered above inadequate (even if it is a good-enough explanation of the same fact under a different description).

But I am left with two questions:
A. What exactly is wrong with the simple explanation, considered as an explanation of the modal collapse? What virtue would a satisfying explanation have that this one lacks?
B. If we can answer question A, do we have any reason to think that there will be a good explanation of the modal collapse so described (i.e. an explanation that has the virtue we've identified as lacking in the simple explanation)? Or might it be that the best we can do is explain why 1 implies 2, leaving the fact that 2 takes the form of 1 without the 'possibly' a sort of coincidence?

Holiday

I'm on a computer-free holiday until Sunday 31st, so won't be posting any more until then.

Vague Existence

In a forthcoming paper in Analysis (not yet available online) my colleague Elizabeth Barnes discusses some worries about ontic vagueness for sparse theories of properties (both theories of Universals and trope theories). One worry about sparse Universals is that issues of vague instantiation seem as relevant to them as to plentiful Universals. Another is that 'Aristotelians' about Universals (people who think Universals only exist instantiated) have some work to do if they are not to be committed to the vague existence of certain sparse Universals.

Rather than discussing the details of Elizabeth's interesting discussion, I just wanted to note an intuitive response I had to it. I often find myself tempted to believe in Universals and not to be Arisotelian about them, so I can regard Elizabeth's worries about vague existence as one more reason not to go Aristotelian. But the non-Aristotelian still faces the worry about vague instantiation. Instinctively, though, I am less worried by this that by the threat of vague existence.

But why? Vague instantiation is ontic vagueness - real vagueness out there in the world, not generated by our language - just as much as vague existence is. An initial thought is just this: I can understand (sort of) how two things, a Universal and a particular, could be vaguely related to each other, but not how one thing could exist vaguely.

A first attempt to say why is that if there is something of which it is true that it exists vaguely, then that thing exists. And we're not going to be comfortable with saying both that it exists and that it's a vague matter whether it exists or not. If vague instantiation is the worst thing we have to deal with, no similar problem arises - there is no comparable motivation to say both that some particular instantiates some Universal and that it is vague whether it instantiates it or not.

The obvious response is that the defender of vague existence is not committed to saying there is something which exists vaguely, merely that it is vague whether (to borrow Elizabeth's example) an Einsteinium Universal exists. As Katherine Hawley puts it in this paper, a 'modest' commitment to vague existence is a commitment to this claim:

it is indeterminate whether anything is F (for some F), this is not as a result of semantic indecision, and yet nothing is such that it is indeterminate whether it is F.

But I'm still uneasy. As good Quineans, anything that we quantify over is something that we should be happy to say exists. Therefore, you might think, no Einsteinium Universal can be among the range of our quantifiers, since we are not happy to say that an Einsteinium Universal exists (that would be in tension with the claim that it is indeterminate whether an Einsteinium Universal exists). But if no Einsteinium Universal is among the range of our quantifiers, we can truly say that nothing is an Einsteinium Universal. And that looks very much like a claim that no Einsteinium Universal exists, which is again in tension with the claim that it is indeterminate whether an Einsteinium Universal exists.

Maybe there is room to try and block this argument by claiming that it's indeterminate whether an Einsteinium Universal is among the range of our quantifiers. After all, our quantifiers range over everything that exists, and it's indeterminate whether an Einsteinium Universal exists.

But if our quantifying over something goes along with it's being the case that we should be happy to say that that thing exists, this proposal ought to give the result that it is indeterminate whether or not we should be happy to say that an Einsteinium Universal exists. Yet the defender of ontic vagueness will actually be unhappy about saying this, and will presumably think she is right to take that attitude, since saying this would amount to taking a stand where the world itself does not.

Putting the point another way, if it's a vague matter whether we're quantifying over an Einsteinium Universal, it's a vague matter whether we're committed to one or not. But the defender of ontic vagueness would surely deny that she is committed to the existence of an Einsteinium Universal (I am assuming that the view is meant to be neutral as to whether an Einsteinium Universal exists or not - i.e. not committed either way). And this sits uncomfortably with the thought that it's a vague matter whether she is or isn't so committed.

Dispositions

Lewis (1997) proposed that:

Something x is disposed at time t to give response r to stimulus s if and only if, for some intrinsic property B that x has at t, for some time t’ after t, if x were to undergo stimulus s at time t and retain property B until t’, s and x’s having of B would jointly be an x-complete cause of x’s giving response r.

There are lots of things to say about this, but one of them is that we sometimes seem to have dispositions based on extrinsic properties. For instance, I am disposed to get upset when someone is rotten to Muriel, and the underlying basis of this disposition is that I like her. Liking Muriel is, however, surely an extrinsic property of mine rather than an intrinsic one.

The restriction to intrinsic bases is introduced to resolve certain problem cases which occur if we leave it out. A sorcerer watches over a fragile glass, prepared to make it cease to be fragile if it is ever struck. Why doesn't that mean the glass has a disposition to lose its fragility when struck? According to Lewis, it's because the property which served as the basis of the proposed 'disposition' would not be intrinsic to the glass.

But maybe Lewis puts the intrinsicness requirement in the wrong place. Maybe what matters is that s and x's having of B cause changes in the intrinsic properties of x which are sufficient to bring it about that x gives response r to the stimulus s.

If that's right, the real reason the glass does not have the disposition to lose its fragility when struck is that the striking and the property of being watched over by a sorcerer do not cause a change in the glass's intrinsic properties that suffices to bring about the loss of fragility. The way this property and this stimulus bring about that response is through causing changes in the sorcerer and his relation to the glass.

By contrast, the cases where we seem to have dispositions with extrinsic bases are ones where the relevant extrinisic property, together with the stimulus, brings about the response just by causing some change in the intrinsic properties of the bearer of the disposition. For instance, I get upset when someone is rotten to Muriel because of the way that the rottenness and my liking for Muriel cause changes in my intrinsic properties.

In Virtue Of

I've just finished reading a very interesting paper on metaphysical dependence (the kind of fact-fact relation that philosophers often use 'in virtue of' to express) by Gideon Rosen, which he presented at the last Arche modality workshop. I won't comment on the contents of the paper as it's a work in progress, but I did want to post some thoughts about a question I was inspired to think about by reading it.

The question is: can it happen that a single fact P obtains wholly in virtue of one fact Q and also wholly in virtue of a distinct fact R? I'm not particularly wedded to any answer to this question, but here are some prima facie reasons for answering 'yes'. The fact that my poppies are red obtains in virtue of the fact that they are scarlet. (Actually all my poppies are disappointingly pink this year - but let's pretend they're not.) And the fact that these poppies are red obtains in virtue of their surfaces being microphysically structured in such a way as to look red under normal conditions. A little more controversially, the fact that Jeff has decided to go somewhere sunny for his holiday obtains in virtue of the fact that he has decided to go to Spain for his holiday, and it also obtains in virtue of the fact that Jeff's brain is (or has been) in a state constitutive of his having decided to go somewhere sunny for his holiday.

But one could try to resist this by saying that in these cases the 'two facts' mentioned are really the same fact. On the face of it, that seems crazy: a poppy's being microphysically structured in such a way as to look red under normal conditions doesn't even entail that it is scarlet. So how can these two facts be the same? One response might be to argue that there is no real (worldly) fact corresponding to anything as general as the 'fact' of the poppy's being scarlet or that of its being microphysically structured in such a way as to look red under normal conditions; to argue, that is, that the only real facts are very detailed facts about the poppy's particular microphysical structure. We could then say that the two sentences cited when saying what it is in virtue of which the poppy is red are (regardless of what we happen to think about the matter - and regardless of how much we know about the worldly - i.e. particular - facts) just two ways of picking out the same particular fact or facts about the poppy.

An objection to this view is that it makes 'in virtue of' relations into relations of fact-identity (or perhaps at best fact-inclusion). For presumably the worldly fact(s) corresponding to my poppy's being red are the same as those corresponding to its being scarlet and its surface having a certain microphysical structure. But I can imagine this bullet being bitten; maybe IVO claims really are made true by relations of fact-identity or fact-inclusion. Maybe they are interesting, non-trivial and assymmetric only for epistemic reasons, having to do with the ways the facts are picked out. This line has its advantages if you are inclined to think that worldly metaphysical grounding is a weird sort of notion that we are well shot of.

A less radical way of taking this kind of approach, however, would be to argue that although the less particular facts are also 'real', they are not the sorts of things that facts can obtain 'in virtue of'; only the very particular facts can play the metaphysical grounding role. When we appear to cite a less particular fact as the metaphysical ground of some other fact, we are actually just gesturing towards the kind of real (particular) fact that is doing the work.

There are problem cases for either version of this sort of strategy, however. It seems that my mother is a parent in virtue of having given birth to me, and she is also a parent in virtue of having given birth to each of my brothers. But however particular you take them to be, these facts are surely not the same fact.

In response to this kind of point, perhaps we might deny that she is a parent wholly in virtue of having given birth to me. Although she could have been a parent wholly in virtue of having given birth to just one of us, and indeed was so at one point, maybe we just need to be careful not to confuse that claim with any claim about the actual metaphysical ground of the present tense fact that she is a parent.

One way to motivate that line could be to argue, in sympathy with the more radical of the two options considered above, that, as things currently stand, the sentence 'My mother is a parent' must pick out a bunch of particular worldly facts which involve (the microphysical parts of) me and all my brothers, and so could not obtain wholly in virtue of her having given birth to me. It must pick out this bunch of facts, we might think, because there's no unspecific wordly fact (her being a parent) that it could pick out, because unspecific facts aren't 'real', and to say it picks out specific facts about (the microphysical parts of) her and only one of her children seems to be privileging that child in an unprincipled way. (And we don't want to encourage sibling rivalry.)

However, we don't need to say anything as strong as this in order to appreciate that there is some intuitive motivation for the claim that my mother's being a parent does not obtain wholly in virtue of the fact that she has given birth to me (or to just one of my brothers). Or so it seems to me.

3Dism and 4Dism

Aha - something I've suspected for a while but been rebuked for thinking has been defended in print by someone else. In a recent paper in Erkenntnis (the link requires a subscription), Kristie Miller argues that three-dimensionalist and four-dimensionalist views about persistance are intertranslatable, equally simple, and equally explanatory. Though I'm not sure I'd go so far as to say they're 'metaphysically equivalent' as she does (I guess I hear that as conveying a bit more than she intends by it, as captured by her four clauses on p. 92), I have often wondered whether the two theories are really just two ways of describing the same underlying facts, and if so how much really hangs on which of them we choose.

Counterpossible Conditionals

I've been thinking on and off for a long time (and without making too much progress) about counterfactuals with impossible antecedents. I have generally tended towards the view that they are trivially true: there are no possible worlds where P, hence every world where P is a world where Q. But this looks troubling when we want to distinguish between pairs like:

1. If the square root of 2 were rational, it could be expressed as n/m, with n, m integers.
2. If the square root of 2 were rational, it couldn't be so expressed.

Intuitively, you might think, 1 looks true and 2 looks false. Why, if they are both trivially true?

The answer that tempts me is that counterfactuals (including counterpossibles) are often used for, and often naturally heard as, expressing claims which are distinct from their literal contents. When we evaluate 1 as true and 2 as false, we are actually evaluating the sort of claim which is often conveyed by sentences like 1 and 2, although it is not what they literally mean. (In the case of 1, it might be something like the claim that any number which is rational can be written as n/m with n, m integers, so in particular this is true of the square root of 2).

A worry that a colleague put to me this morning concerning this view is that we have an intuition that it is wrong even to think things like 2, but stories which focus on the pragmatic communication in conversation of claims distinct from literal content will not capture that intuition.

I wonder whether one might respond that accepting in thought that 2 is true could be a mistake if it relied upon (or amounted to) mistakenly assuming that the non-literal content often conveyed by claims like 2 was true. If one accepted 2 in thought without realizing that it was only true because its antecedent was impossible, one might (e.g.) be making the mistaken assumption that no number which is rational can be expressed as n/m with n, m integers, so in particular this is true of the square root of 2.

It wouldn't be misleading to think 2 was true if one's reason for doing so was that one knew its antecedent to be impossible. But in this case I don't get the intuition that there is anything wrong or mistaken about thinking 2 is true.

Ebert vs Boghossian on Epistemic Analyticity

In an interesting forthcoming paper, my colleague Philip Ebert discusses an argument, due to Paul Boghossian, to the effect that we can have a priori knowledge of logical principles through implicit definition, on the basis of the following kind of 'template' argument (slightly revised by Philip from its original form):

1. If ‘and’ is to mean what it does, then ‘P and Q --> P’ has to be valid (by the nature of implicit definition)
2. ‘and’ means what it does
3. ‘P and Q --> P’ is valid
4. P and Q --> P

Philip objects that the essential ‘disquotational step’ from 3 to 4 cannot be made unless one already understands ‘and’. (This claim is supported by a trivial principle about disquotation). This understanding of ‘and’ in turn requires (by a Context Principle) that one understands sentences like 4. But by Boghossian’s lights, 4 is epistemically analytic, so that understanding it suffices for being warranted in believing it. Hence anyone who can use the disquotational step already has a warrant for believing the conclusion of the argument. This, Philip says, means that the argument fails to transmit a warrant for its conclusion.

I think one should reply on Boghossian's behalf that the epistemic analyticity of 4 only means that a warrant for believing 4 is available to anyone who understands 4, not that understanding implies possession of a warrant. But no worry about transmission of warrant failure arises unless understanding 4 implies possession of a warrant.

Philip comments on this response in his paper, though he does not put my mind at rest. I won't post my comments on his comments here however (though I will happily share them with parties interested enough to email me for them). Rather, I want to suggest that there is a different (though not entirely dissimilar) problem with the template argument as an account of our knowledge of P and Q --> P, namely that it doesn't tell us very much until we are told how premise 2 is known.

One reason to think knowledge of 2 is no trivial matter is that it implies that 'and' has a meaning . But if 'and' is implicitly defined as whatever makes ‘P and Q --> P’ (and some other rules) valid, knowing 2 requires knowing that something makes ‘P and Q --> P’ (and the relevant other rules) valid. How is this fact known? Plausibly, the way we know it is by knowing that P and Q --> P, but that can't be Boghossian's answer.

Boghossian (in the appendix to his 1997 paper 'Analyticity', included in this anthology) addresses this sort of worry by saying that 'we are a priori entitled to believe that our basic logical constants are meaningful because we cannot coherently doubt that they are'. There are many things to say about this claim. One of them is that, in the presence of Boghossian's views on implicit definition, it seems to imply that one of the things we rely on to secure a priori knowledge that P and Q --> P is the fact that we cannot doubt that there is some meaning for 'and', even when it is defined as that which makes ‘P and Q --> P’ (and other relevant rules) valid. One wonders exactly how far this is from simply saying that we cannot doubt, of the rule expressed by 'P and Q --> P' on its intended interpretation, that it is a valid rule. But to say that would hardly be to provide a satisfying acccount of our a priori knowledge of P and Q --> P.

The Sensitive Kind

This is a more developed rambling about a topic I posted on recently over at Certain Doubts.

Consider the following situation-schema (instances of which are familiar from various areas, including theory of knowledge, ethics, gradable adjectives, epistemic modals, disputes of taste ...):

Two people give (apparently) different verdicts on two cases which prima facie you might think they should have given the same verdict on. (In the extreme, the two cases might be the same case. But it could also just be that the properties on which you would expect the verdict to supervene are held constant between the cases.) Yet we have an intuition that (we are tempted to express by saying that) neither person is wrong. Furthermore, we think we are dealing with a truth-apt discourse.

Responses to this situation can be grouped together into (at least) three families:
1. Responses which deny that we are dealing with a truth-apt discourse.
2. Responses which maintain that one person's claim was false but try to make this less counter-intuitive (e.g. by talking about assertability and/or the pragmatic communication of some truth through the utterance of a literal falsehood).
3. Responses which take seriously both truth-aptness and the intuition that (there is some sense in which) neither person is wrong, by arguing for some sort of hitherto unexpected sensitivity to some parameter.

I am interested in attempting to map the various options within the third family. A bewildering array of terminology (and terminological disputes) lurks here, so I'd like to propose a systematic approach.

The idea that defines the third family is that the truth of an utterance can be in some way sensitive to the value of some parameter or other. But there are two dimensions in which views within the family could vary: along a source axis and along a type axis.

The source of the mooted variation in truth-value may be something about the subject-matter, something about the utterer or something about the assessor of the utterance's truth-value. The type may be either variation in content (which could in turn amount to variation in sense or variation in reference) or variation in truth-value without variation in content.

This gives six views:

SSC, Subject-sensitivity of content
SST, Subject-sensitivity of truth-value
USC, Utterer-sensitivity of content
UST, Utterer-sensitivity of truth-value
ASC, Assessor-sensitivity of content
AST, Assessor-sensitivity of truth-value

A few comments on these positions:

i. 'Sensitivity of truth-value' is short for 'sensitivity of truth-value without sensitivity of content'.
ii. ‘Sensitivity’ strikes me as a better word to use than (say) ‘variability’ because we want to capture the thought that change in the source is what explains the change in content and/or truth-value, not just that the two co-vary. We want the idea to be that content and/or truth-value are sensitive - responsive - to changes the parameter we're interested in.
iii. ‘Subject’ means ‘subject-matter’ (we don’t want to restrict ourselves to cases where there is ‘a subject’ – e.g. a person).
iv. US_ and AS_ are usually assumed to be talking about sensitivity to utterer’s or assessor’s context. SS_ might be talking about sensitivity to a subject’s context but other interpretations are also plausible.
v. If we don’t intend to talk about subject’s context, though, SS_ risks becoming trivial unless we are careful to restrict attention to unexpected kinds of subject-sensitivity.
vi. SS_ must be understood as sensitivity to how/what the subject-matter is, not to how it is thought of, or else it’ll be a form of US_ or AS_. (You might think that SS_ could give rise to US_, though: if we accept SS_ and think that subject matter contributes to fixing utterance context, won't we end up with US_? I don't think so - because US_ is meant to be the view that it is facts about the utterer('s context) which explain the pattern of variation. Whereas in the envisaged scenario what's really doing the explanatory work is facts about the subject-matter.)
vii. As far as I can tell, the term ‘contextualism’ might with some precedent be used to refer to various of the six positions, provided the sensitivity in question is to someone or other’s context. I take the standard contemporary view to be that contextualism in epsitemology amounts to a form of USC.
viii. 'Relativism' likewise has many different uses, but seems currently to be frequently used to refer to forms of AS_ (and perhaps more usually AST).
ix. ‘Invariantism’ could refer to a denial of sensitivity of any of the various kinds, so will also have many uses (content-invariantism vs truth-invariantism; invariantism wrt assessor, invariantism wrt utterer, etc.).

Bealer on the A Priori

George Bealer ('A Theory of the A Priori' p.5, and elsewhere) gives the following as a 'principle of holism':

A theory is justified (acceptable, more reasonable than its competitors, legitimate, warranted) for a person if and only if it is, or belongs to, the simplest comprehensive theory that explains all, or most, of the person's evidence.

Surely a non-holist could accept the bare biconditional, however? What's needed to characterize holism, as I've always understood it, is some claim to the effect that a theory's being justified in some sense depends upon its belonging to the best justified (most explanatory, if you like) comprehensive theory.

Less nit-pickingly, I take issue with Bealer's characterization of 'moderate rationalism' on p. 7 of this paper as a commitment to the view that:

A person's phenomenal experiences and intuitions comprise the person's basic evidence.

To have an 'intuition' in Bealer's sense that p is for it to seem to you as if p in an a priori way (p. 3). So Bealer is effectively characterizing all a priorism as rationalism. No fair - there are plenty of honest, hardworking, non-Quinean empiricists out there who acknowledge and try to account for a priori knowledge. But presumably Bealer does not intend that any empiricist should count as a 'moderate rationalist'.

Another quibble: I don't see why we should think that all 'truth-based' (as opposed to pragmatic, coherentist, etc.) epistemological stories are 'reliabilist' (p. 7). I think a form of explanationism is preferable to reliabilism (even 'modal reliabilism' of the kind discussed on p. 9), and it looks equally 'truth-based' to me.

A more substantial point arising from this quibble: with an explanationist account in place, the mere existence of a modal tie between the deliverances of modal intuition and the truth looks inadequate for a genuine evidential relationship. In fact, such a modal tie looks inadequate anyway: we need to know why the tie holds in order to know whether it is evidential. (Bealer seems to be sensitive to this, as he does try to offer an explanation of the tie.) God could have so configured our psychology that, unbeknownst to us, there is a modal tie of the kind Bealer wants between our being willing to guess and the guess being true. But (by the same token which leads Bealer to reject contingently reliable guesswork as evidence on p. 9), such modally stable good luck at guessing wouldn't make our guesses count as evidence.

Armstrong on Logical Necessity

I'm puzzled by this comment by David Armstrong (A Theory of Universals, p. 42):

If logically equivalent predicates which are not logically empty apply in virtue of the very same universals, and logically equivalent propositions which are not themselves logically necessary are true in virtue of the very same state of affairs, then some de dicto account of logical necessity must be correct. The logical necessity of propositions must, in some way, derive from the words or concepts in which the propositions are expressed. (See also p. 168.)

Why? Wouldn't one way of reacting to Armstrong's claims be to conclude that, in the relevant cases, logical necessity boils down to the (de re) necessity of self-identity between universals states of affairs? (Moreover there are many logical necessities to which Armstrong's comments here seem irrelevant, such as '((A-->B)&A)-->B'. But let's ignore them for now.)

Maybe he'd want to resist this because there are some cases where the identity of the universals in virtue of which two predicates apply does not give rise to a corresponding logical necessity. There are cases of 'contingent identification of properties', according to Armstrong.

But that doesn't seem particularly relevant to a discussion of the source of logical necessity (what it 'derives' from). Why not just say (for instance) that we get logical necessity in cases where the two predicates wear the identity of their corresponding universals on their faces, but we don't get it when this identity is not so obvious?

Presumably Armstrong will say that this means the difference between logically necessary propositions and the rest has something to do with the predicates used (i.e. whether they wear the identity of their corresponding universals on their faces or not) rather than with anything more worldly. That would be right, but it wouldn't establish anything about the source of logical necessity. Armstrong's claim would seem to be that, because we can get a logical truth using one set of predicates then use different predicates corresponding to the same universal and get something which isn't a logical truth, the predicates must be doing (at least some of) the work in generating the logical necessity we had in the first case. But that conclusion doesn't seem to follow.

By analogy, we wouldn't say that any part of the truth of a proposition was 'derived from' the words used to express it, just because if we had used different but co-referential words we wouldn't have got a truth. 'Lois believes that Superman flies' is true but 'Lois believes that Clark Kent flies' isn't - but the truth of the former is not in any way 'derived from' the words we used. It is 'derived' wholly from a fact about Lois's mental state, albeit one we need to use the right words to express.

One thing that might be going on is that 'logical necessities' are simply a priori truths. If that's it, then the distinction between them and cases of 'contingent' (a posteriori) identification of properties is something to do with methods of epistemic access, and of course that is something that could be 'derived from the words or concepts in which propositions are expressed'. On the other hand if we want to admit metaphysical necessity, why not say that this is present in all cases where we have identity of universals, and its source is always the same (the identity of the universals concerned), although in some cases we have a priori access to this identity and in others we don't?

(I am assuming that 'logical necessity' is not a syntactic notion, since if it were it would be trivial that logical necessity is 'in some way derived from the words or concepts in which the propositions are expressed'.)

Whatever floats your boat?

One of the comments made by an audience member when I gave my paper at the Joint Session was that Crispin Wright's defence of epistemic entitlement sounded reminiscent of Neurath's idea that, in epistemic enquiry, our position is like that of a ship at sea - we can replace various planks as we go along but cannot throw out the whole system and start again.

Wright states that in empirical enquiry we cannot but rely on 'cornerstones' unsupported by further evidence, such as the proposition that we are not brains in vats. Since he also thinks that a 'proper concept' of warrant will not make impossible demands upon us, he takes it that a proper concept of warrant will not require that we have evidence for these cornerstones in order to count as warranted in accepting them and relying on them in our enquiries.

It sounds as if Wright thinks since we cannot but trust in the truth of the cornerstones, it cannot undermine our rationality if we do so. This sounds akin to the coherentist thought that, since we are ships at sea and cannot build our belief systems from scratch as foundationalism seems to require, it must be that starting from scratch is not required in order for our beliefs to be justified.

Another audience member (I think it may have been Daniel Whiting?) pointed out that Wright seems to think the ship we are afloat on has particular planks that can't be removed without letting the water in, which isn't a standard part of the ship-at-sea metaphor. In this respect, I think, Wright's view seems to be closer to that of Lewis, who holds that certain ('Moorean') aspects of common sense are the planks of the epistemic boat which cannot (all) be replaced on pain of irrationality (as was recently brought to my attention by reading Daniel Nolan's helpful book on Lewis). A particular commonsense plank might be ditched, in order to best respect the rest of our commonsense views, but there is no rational way to ditch all the members of this subset of our beliefs simultaneously. I wonder, then whether Lewis thinks, as Wright does, that the very impossibility of rationally ditching this set of beliefs is what makes it rational to hold onto them. Or is it the other way around: that the fact we are rational to hold onto them makes it irrational to ditch them? The former (Wright-like) interpretation is suggested by the following passage:

'[I]t's not that common sense speaks with the voice of some infallible faculty of "intuition". It's just that theoretical conservativism is the only sensible policy for theorists of limited powers, who are duly modest about what they can accomplish after a fresh start'.

This same modesty is what motivates Wright's views. If Lewis's line is as similar to Wright's as I've suggested, certain worries I have about Wright's proposal would apply here too - in particular, the apparent step from 'nothing else is a sensible policy' to 'this one remaining policy is sensible'.

Explanation: is there anything it can't do?

Causal theories of meaning have trouble dealing with certain areas of discourse. My words 'five' and 'the empty set', for instance, don't seem to get their meanings from my having certain causal relations to the number five or the empty set.

So what would happen if we stopped thinking in causal terms we thought instead about explanation? Some theory along the following lines would presumably be the result:

S's term T means M iff there is the right kind of explanatory dependence between M (or Ms) and S's use of T.

Of course, the devil's in the detail of trying to say what the 'right kind' of explanatory dependence is. But explanation has often seemed to me to offer at least prima facie promising ways out of trouble in places where talking about causes, counterfactuals or necessitation looks initially plausible but eventually runs into difficulties. Theories of knowledge are one place where this idea has an application, as I argue in my CJP paper. I also think Simon Blackburn is right to move away from talk of what "guarantees" success towards talk of explanations of success, in trying to develop a more promising form of success semantics.

I'm glad to see that Simon's success semantics paper is now available online; I remember with pleasure an exceptionally productive afternoon seminar in Cambridge on this topic, where Simon, Anandi Hattiangadi, Steve Butterfill, and I thrashed out the issues. (Aside: many congratulations are due to Anandi on her new continuing job in Oxford!)

Joint Session 2005

The Joint Session of the Mind Association and the Aristotelian Society took place at the University of Manchester from July 5-8. Apart from being stiflingly hot in most of the sessions, it was a lot of fun. (Not so the journey home: an eight-hour train journey instead of the predicted four and a half. Wrong kinds of trees, falling power cables, broken down trains and sunlight on the lines, apparently. Luckily I had a philosophical travelling companion and a draft of a colleague's PhD thesis to read.)

Highlights included Simon Blackburn's talk, 'Paradise Regained'. During the question session I took issue with his claim (pp. 13-14 of the ASSV) that, even if we ditch the 'space of reasons' metaphor and any associated disjunctivism, we can still address Cartesian scepticism by maintaining that veridical perception and the like require successful immersion in the physical world of a kind brains in vats can't have. Surely dreamers have the right kind of immersion (unless this condition is so stringent as to make the view disjunctivist again)? Simon replied that he wasn't addressing that kind of Cartesian scepticism (wrong kind of Cartesian scepticism on the line). But even apart from the thought that dreaming scepticism and BIV scepticism have such a lot in common that it's unclear how satisfying an approach to the one can be if it doesn't tell us anything about the other, surely this means that quite a potent form of scepticism is still hanging around ready to motivate disjunctivism (at least, insofar as any such motivation for that kind of view was ever plausible). It at least enforces some restriction on the claim that 'scepticism [is] only a natural bogey once the spatial metaphors take hold'.

Other highlights included a great session by Jennifer Hornsby, Jason Stanley and Ian Rumfitt, and lots of high-quality open and graduate sessions. Unfortunately though the bar didn't have a late license; the resultant impromptu late-night drinking session on the lawn was enjoyable however (and very well attended). I gave a paper in the open sessions on entitlement and epistemic rationality, and got a lot of helpful feedback from the audience there.

I'm all in favour of the new format of the open sessions, though two small things would have considerably improved the experience for me. One concerns the arrangement of the rooms. Finding enough rooms close together must be an administrative nightmare, but it was very difficult to get to and from papers in the chapel, and some of the papers given there didn't get the audiences they deserved. The other would be some minimal refereeing, to ensure that all the papers given are broadly competent. (I appreciate that there is a Sorites series from obviously terrible papers to ones which are obviously OK, and I approve of the open ethos driving the current policy, but I wasn't the only one who found it frustrating sitting through papers that a referee could have ruled out in a few minutes, and I can't imagine the presence of such papers gives a great impression of British philosophy to overseas visitors.)