In case anyone hasn’t yet taken the NIP journal survey, please click through and help the Northern Institute of Philosophy out with a very brief market research exercise.

Speaking of the NIP, I’ve been intending to post a link here to Walter Pedriali’s detailed report on the recent Entitlement Workshop.  Photos are included; they are (representatively) dark due to a lighting malfunction, but the illumination provided by our discussions naturally compensated.  See also the full photo gallery for the workshop.

… are on Facebook.  Be there or be mereologically variable (inclusive ‘or’).

See also: this page, with all the Original Monads tunes and the new tracks from the 21st CMs!

UPDATE: New song out now! The G. E. Moore Shift

This post concerns the strategy of postulating contextual variability to resolve/dissolve an apparent philosophical dispute.  Sometimes, it seems to me, philosophers feel methodologically aggrieved  when they encounter instances of this strategy.  I’m going to list some of the criticisms I’ve heard from (or thought up on behalf of) the aggrieved parties, and explain why I think they’re bad criticisms of the contextualist strategy.

(1) ‘It’s too easy.’

Why assume that good philosophy should be hard? Assuming we’re interested in getting at the truth, rather than proving that we can do difficult things, easiness is irrelevant. To put it another way, if difficulty is a theoretical virtue at all, I know of no reason to think it is a virtue of the kind that is relevant to rational belief. 

Also, it’s far from clear that it is easy.  There are, for example, many serious and important objections to epistemic contextualism which the epistemic contexualist needs to engage with in order to sustain her position.

(2) ‘It’s too broadly applicable.’

Why assume that philosophical strategies should not be broadly applicable?  In some arenas, features such as generality and breadth of explantory power are regarded as theoretical virtues, and moreover virtues of the kind relevant to rational belief.

Perhaps the concern is supposed to be that, given the breadth of applicability, it is difficult or impossible to place constraints on when the strategy is appropriate.  But I see no reason to suppose that breadth of applicability is directly relevant to the issue of how to find and justify such constraints.  Any strategy that is (or even might be) appropriate in some but not all of the debates in which it is applicable raises the methodological question of what constraints to place on its appropriate application.

(3) ‘It’s lazy.’  

It could be argued that reaching for the contextualist strategy, like reaching for one’s comfort blanket, enables philosophers to avoid doing the hard work of evaluating other strategies and making a proper assessment of their merits relative to those of the contextualist strategy.

But this is not a criticism of the strategy; it’s a criticism of a particular way of abusing the ready availability of the strategy.  Of course no defender of widespread contextual variability should forget to consider other options and assess their pros and cons.  But the same goes for every philosopher who has a view to defend.

Daniel Nolan and I are working on a draft paper looking at unmanifestable dispositions: dispositions, that is, to PHI in circumstances C where at least one of PHI and C is impossible (logically, metaphysically and/or nomically).  We think many things have such dispositions, and this isn’t trivial.  (A trivialist would say that everything has every unmanifestable disposition.)  And we think it matters.  It turns out not to be easy for standard kinds of theories of dispositions to accommodate the unmanifestable sort.

More details on the forthcoming Nottingham A Priori Workshop (9th October) are now available, including the programme and information on how to register.

Attendance is free (though we do require advance registration), lunch, tea and coffee are provided, and the speakers are Anthony Eagle (Oxford), Jessica Brown (St Andrews), myself (Nottingham) and Michael Devitt (CUNY/Nottingham Special Professor).

For enjoyers of geeky philosophy humour, I wrote a song called A Priori (link to an MP3 file) especially for you. Here are its lyrics.

I’ve heard a few philosophers mention recently that it seems to be more difficult than usual to get in touch with people at OUP at the moment.  This is also how it’s seemed to me over the last few months.

I wonder whether anyone else has noticed this, and particularly whether anyone knows of an explanation.

I was recently interviewed for the University of Nottingham Podcast series on the subject of how Twitter can be of use to philosophers.

The result is now available!

Michael Devitt is visiting Nottingham in October, and to mark the visit there will be a workshop on the a priori here on October 9th, at which he and I are both presenting papers. Also on the bill are Jessica Brown and the Ant Eagle. Let me or Greg Currie know if you’d like to be there.

For the occasion, I’ve been writing a paper engaging with some Devettian criticisms of the a priori, which I’ll present at the workshop.  I’ve just finished a draft that’s suitable for public consumption. Comments are welcome, especially if you spot anywhere I’m being unfair to Devitt!

Jeffrey Roland has an article (a subscription to Philosophia is required to view the full text) arguing that the prospects aren’t good for extending my 2005/2008 account of arithmetical knowledge to set theory.  The purpose of this post is to sketch a reply to his main argument.  (I’ll also then mention a couple of quibbles, because I can’t resist.)

Roland’s main challenge is that the concept set cannot be grounded in sensory input.  He says that ‘sensory input is too coarse-grained to make the distinction between the concepts of set and class’, since ‘these concepts are coextensive in the (finite) world of our observations’.  This, I think, is closely related to the kinds of underdetermination worry I discuss in my book (pp. 227-30).

Roland correctly anticipates that I will not accept that the kind of ‘coextensiveness’ he gestures to makes grounding for either or both of these concepts impossible.  For one thing, as he notes, I think that any concept that appears essentially in our best scientific theories is one we have reason to believe is grounded.  If all of mathematics is part of best science, then the concept of set is looking pretty good by these lights. I’m pretty confident that at least enough of mathematics is part of best science to make that concept indispensable to best science.

But a more illuminating way, I think, of replying to this kind of worry is to point out that empirical concept grounding is not supposed to be construed in a simplistic way, whereby some particular identifiable batch of experiences grounds each particular concept.  If that was the view, maybe it would be a problem that the same sets and classes look the same, sound the same, and so on.  But what I have in mind is that the concepts of set and class may prove their worth (individually or – more likely - in tandem with each other and with other concepts) by enabling us to make best sense of our experience of the world, and hence of the world.  I want to understand this process in a holistic (or at least molecularistic) way.  By analogy, triangular figures and trilateral figures look the same, sound the same, etc.; but that doesn’t make trouble for the view that the two concepts trilateral and triangular are both really useful for making best sense of all of our experience.

If concepts like set and class are useful in the way envisaged here, then in my opinion the best explanation of their usefulness is that there is something about world that both of those concepts (or perhaps their ultimate constituents, if they are compounds) are mapping onto.  And our responsivness to that is what makes those concepts grounded.

A related point worth making in response to underdetermination worries is that empirical concept grounding – like the empirical grounding of theories - should be construed as involving virtues besides empirical adequacy.  Conceptual virtues (simplicity, fruitfulness, appearance/capacity to appear in virtuous theories …) might be drawn upon to help explain why a concept is grounded despite the apparent inadequacy of the experience which grounds it. 

Contra Roland’s suggestion on my behalf, however, I certainly would not propose that we use such virtues as a basis on which to choose between the concepts set and class.  Rather, I should say that it is the possession of such virtues that makes each of those concepts a useful, and hence grounded, one.

Finally, two quibbles:

(1) I don’t propose an ‘analysis of knowledge’, just a necessary and sufficient condition which I hope is illuminating. 

(2) Roland asks: ‘[W]hy think that our concepts of such (transfinite) numbers are even apt for acquisition in response to sensory input, let alone actually grounded?’  I don’t actually require acquisition in response to sensory input. Even innate concepts can be empirically grounded on my view.