Commitment and Degrees of Belief

I've recently got interested in degrees of belief. (Hang around Canberra long enough and it's bound to happen.) I've been particularly enjoying a paper by Lina Eriksson and Al Hajek called 'What Are Degrees of Belief' (forthcoming in a Studia Logica special issue on formal epistemology, edited by Branden Fitelson, to appear 2007). I won't go into the details of their paper here since it is not publicly available yet. But here's something a bit different that I started thinking about after reading their paper and chatting to Al about it.

(Caveat: I don't know much about the literature on this topic, so I'm making no claims to originality.)

It would be nice to have a unified way of explicating the notions of belief and degree of belief. My hunch is that the notion of commitment - which clearly has both on/off and gradable aspects - can help. Belief could be explicated as commitment in the on/off sense, degrees of belief as degrees of commitment in the gradable sense.

The tricky bit is going to be specifying the right notion of commitment. Here are some of the things it's not:
- the kind of commitment you take on by making a promise
- the kind of commitment you have to a proposition p in virtue of believing something which entails p (thanks to Al for this one)
- the kind of commitment you can have to a cause or a person.

Here are some things that might help pin down the right notion:
- it's a propositional attitude
- we aim to have this kind of commitment in the on/off way to a proposition p only if p is true.

Incidentally, some readers of this blog may be interested to know (if they don't already) that Studia Logica is currently calling for papers on vagueness. (I just spotted this while finding the link for the above reference.)

Life Matters

Daniel and I might be appearing on ABC National Radio on 21st December on a programme called 'Life Matters'. Discussing what, I hear you ask? The semantics of conditionals? The a priori and the Canberra Plan? Methodology and epistemology?

Amazingly, it's none of these they're interested in; it's the philosophy of flirting. Watch this space for confirmation ...

Book News

Some good news: my book Grounding Concepts will be appearing with Oxford University Press. The central aim of the book is to develop the idea of concept grounding as the basis of a new kind of epistemology for arithmetic (see my paper Knowledge of Arithmetic for a sketch). It also tries to show how some of my ideas about realism and knowledge are supposed to fit into a coherent whole together with the concept grounding story.

Now I just have to get down to writing up the final version. Any volunteers to read and criticize draft chapters would be welcome ...

(PS: Sorry about the delay with the BK workshop photos. I'm having some technical difficulties with my camera.)

AAP(NZ)

Pictures from the Basic Knowledge workshop will be appearing here soon ... I am currently away at the New Zealand AAP.

Ed Mares gave the opening presidential address on Sunday, arguing that validity was to be identified with information-preservation and not with truth-preservation. Since information-preservation was so defined as to yield the rules of a relevant logic, this allowed Ed to maintain that the connectives have meanings given by their classical truth-conditions without thereby committing himself to classical logic.
Amongst other interesting stuff, Ed argued (by way of arguing that we do not need Looney Tune, or 'that's all', facts) that whether a model is complete or partial is not a matter of whether the model contains a 'that's all' fact but a matter of our attitude to that model (i.e. whether we regard it as complete or partial). I think it would be preferable, to avoid raising realist eyebrows, to say simply that it is a fact about the model rather than a fact within the model, and leave us out of it.

New (Final?) Draft of My Boghossian Paper ...

... is now online.

More Photos

I've just posted some more photos from the grad conference.

Priest On The Logic Of Relativism

Today at the Arché grad conference Graham Priest entertained us with a fun talk on the logic of relativism.

Amoung other things, Graham considered the following problem that (some) relativists seem to face. Suppose you are the kind of relativist who thinks that as a relativist you should not make absolute claims (A), but only claims about what holds according to some perspective (SA - where S stands for 'syat', a Sanskrit term of art which apparently is used by Jains to mean something like 'from some perspective'). This gives rise to a regress: one should not assert SA absolutely but only the qualified SSA; not SSA but SSSA; and so on. It turns out nothing is assertible.

Graham proposed a system in which A and SA (and hence SSA, SSSA etc.) are all logically equivalent, and suggested that this would solve the problem, since whenever one asserts A one is thereby asserting (something equivalent to) SA, SSA and so on.

But I wondered how this proposed solution would work. The thought motivating the problem was not that the relativist should say SA as well as A, but that the relativist should say SA instead of A: that she was wrong to assert the unqualified A. If A and SA, SSA etc. turn out to be equivalent, the wrongness of asserting the unqualified A has not (for all that's been said so far) been removed. So for all that's been said so far, none of SA, SSA etc. are assertible because they are all equivalent to the unacceptable unqualified A.

Grad Conference Photos

Here are some of my photos from the Arché grad conference. More will follow ...

Tip

Look out for more coverage of next weekend's Basic Knowledge workshop over at Plurality of Words. "Extensive blogging" of the event is "promised"! (Can't get out of it now, Andreas!)

Stanley on Shared Content

Neglecting Hegel on the a priori yet again, Jason Stanley gave an exciting talk about contextual sensitivity and shared content yesterday.

The (alleged) problem of shared content for those who believe in rampant contextual sensitivity is that is that to the extent that the proposition expressed varies with context, it is harder to explain why people find it so easy to communicate (i.e. grasp what propositions others are expressing) .

Richard Heck has suggested that grasping the exact proposition expressed is not important - you just have to grasp one that is similar enough.

Jason proposed instead that one should respond by saying that it is not so hard as you might think to grasp the exact same proposition that one's interlocutor expresses. Adpoting a Russellian view of propositions enabled him to argue (if I got him right) that my grasping the exact proposition you express requires only a kind of de re understanding. It only requires that I know, de re of the things you were talking about and properties you ascribed, that you said that those things have those properties. Since the Russellian proposition expresed is just a construct out of those things and properties, not a Fregean sense, mode of presentation doesn't matter: it doesn't matter what descriptions of these objects and properties I have available, nor whether I can distinguish them from close relatives, as long as I end up with the appropriate piece of de re knowledge about them.

One question this raised in my mind is whether this move really addresses the spirit (as opposed to the letter) of the shared content problem. If we go Russellian about propositions, then arguably grasping the proposition expressed is not all that's important for communication. Modes of presentation ought to matter too.

Addendum: Thanks to Jason for correcting my spelling of 'Russellian'. I should have mentioned that the term 'problem of shared content' is due to Herman Cappelen.

Arche Trip

I'm back at Arché for a couple of weeks, where I'll be attending the Arché graduate conference and organizing the Basic Knowledge Workshop. I will be posting some event reports here as I go along.

Tonight I'm off to see Jason Stanley open the grad conference. The title is TBA, so it remains to be seen whether he will address the important yet neglected topic of Hegel on the a priori, as requested by Paula and me.

Nolan Flirts With Fame

The Aussie press has now picked up Daniel's response to my paper The Philosophy of Flirting - see the fourth item down on this page of The Australian.

The reporter offers some sage advice inspired by the work of Nolan: "If you want to flirt, send text messages, not testicles."

PGR and Other News

The long-awaited Philosophical Gourmet Report is now up. I note with interest that the ANU is ranked 15th in the world, along with my alma mater Cambridge. Nottingham comes in at joint 51st in the world.

In other news, I'm very pleased to have been offered Associate Fellowship of two research centres in the last week: the Centre for Metaphysics and Mind in Leeds and my old haunt Arché in St Andrews. Both are places where there's lots of good philosophy going on, so this is really exciting (and it raises my number of institutional affiliations to four for the rest of this academic year)!

Davies Paper

A draft of another of the papers for the Arché Basic Knowledge Workshop is online: 'Two Purposes of Arguing and Two Epistemic Projects' by Martin Davies.

Boghossian on Epistemic Analyticity

I'm posting the nearly-final draft of my paper on Boghossian on Epistemic Analyticity. Comments welcome. The paper upholds the objection that knowledge of meaning through implicit definition cannot be a source of a priori knowledge, since in order to use implicit definitions, one must already know the propositions knowledge of which we are trying to account for. This claim itself is not new, but I aim to do four new things: firstly, drawing on some recent work by my former colleague Philip Ebert, I put a new structure on the objection, showing how it works on either of two possible readings of one of Boghossian's premises. Secondly, I argue that Boghossian's recent attempts to answer this sort of objection are unsuccessful. Thirdly, I offer some new side-objections to Boghossian. Finally, I resist Ebert's reasons for thinking what's wrong with Boghossian's argument is that it fails to transmit warrant (and also explain how the objection I defend is different).

Update: The link now points to a new draft, changes to which have been based in part on the discussion in the comments on this post.

Merely Verbal Disputes

I'm planning to join the metaphilosophical fray over what counts as a merely verbal dispute. My first thought is that what counts as a merely verbal dispute is likely to depend to a fair extent on what the parties to the dispute are interested in. One obvious way it looks like this can happen is that various prima facie substantive-sounding disputes can turn out to be merely terminological if the parties are self-consciously interested in settling a point of terminology. But this isn't the only way.

For instance, consider disputants A (an externalist) and B (an internalist), who argue as follows:
A: "We both know that there is an external world."
B: "No-one knows that there is an external world."

Suppose A and B agree that they each stand in relation E to the proposition There exists an external world, where E is the relation that A, being an externalist, takes to be sufficient for knowledge. And they also agree that nobody stands in relation I to that proposition, where I is the relation that B, being an internalist, takes to be necessary for knowledge.
Now suppose that when they realize they agree in these ways, they find that this enables them to resolve all the points they were interested in. Then we might be tempted to say that their original dispute was merely a verbal dispute about 'knows'. But in fact things aren't quite that simple - we want to allow that two people could resolve a substantive dispute about knowledge by thinking about the E and I facts, even where the original dispute was a substantive dispute about the knowledge facts (as opposed to the E and I facts) and not merely a verbal dispute about 'knows'. For instance, it could be that thinking about the E and I facts helps one of them to notice the facts about the knowledge relevant for resolving their dispute. Instead, I think that being resolvable in this way is a symptom of a merely verbal dispute. It is a symptom because it (fallibly) indicates that they never really diagreed about anything they were interested in, which I think is crucial for whether their dispute was merely verbal. (That's not supposed to be a criterion by itself either, but it's closer.)

But if noticing these points of agreement does not resolve their dispute (and the same goes for any other agreements of a similar kind that they may have), then it's tempting to say their dispute is not merely verbal but concerns a substantive point about knowledge. And that's going to be because this is evidence that what they're interested in is the facts about knowledge, and not just the E and I facts (or whatever).

In fact, however, there's something appealing about the thought that their dispute might still be merely verbal even if they are interested in knowledge for its own sake. For if there are no knowledge facts as distinct from the facts about E and I, then there is nothing substantive for their dispute to be about. So their dispute - if it is to have any point at all - must boil down to a disgreement about whether 'knowledge' tracks (and/or should track) E or I. If this is right, the interests of the disputants don't always settle whether the dispute is merely verbal. Still, they surely have an impact in many cases.

Pritchard's Basic Knowledge Workshop Paper

A draft of Duncan's paper for the Arche Basic Knowledge Workshop, 'Knowledge and Value', is now available. I'm looking forward to getting a chance to read it this weekend ...

A Priori Paper

I'm posting a draft of my new paper on the a priori. Comments very welcome, as always.

(Actually p) iff p

Here's an interesting schema from the point of view of thinking about the relationship between conceptual truth and necessary truth:

A: (Acutally p) iff p

For true but contingent propositions p, this is possibly false: worlds where not-p are still worlds where actually p. Do people reckon it's conceivable that A is false for such p?

Even if it *is* conceivably false, you might think it's the sort of thing you can tell is true just by thinking about the concepts it involves. This might lead you to suspect there are two grades of conceptual truth:

1. Things you can tell are true just by thinking about the concepts involved,
2. Things such that it is not conceivable that they are false.

The second grade looks strictly stronger. (Although I'd be interested to hear if people think they have counterexamples to this claim.)

(Thanks to Daniel for making me think about this!)

More Fame

Since my last post I have been interviewed on ABC Radio! (Although before anyone gets too excited, I should mention that it was ABC Radio Hobart). I'm about to be interviewed again by RTR fm in Perth. It seems the Aussies are interested in flirting.

What I want to know is why nobody takes this kind of interest in my *fascintating* papers on the a priori. :)