« Via Richard Zach |
Main
| Some Animadversions on Hume's Law »
May 23, 2005
FEW
Like Jonah and Kenny, I'm off to FEW tomorrow. On Thursday I'll be presenting a paper that I wrote with Greg Restall, "Barriers to Inference," which is about how to formulate and prove inference barrier theses such as these:
Russell's Law: you can't get a universal claim from particular claims
Hume's Law: you can't get a normative claim from a descriptive claim (no 'ought' from an 'is')
Kant's Law: you can't get necessity-style claims from contingent claims
Hume's 2nd Law: you can't get claims about the future from claims about the past and present.
The slides for the talk are available as a pdf here. Since I don't really believe in reading out slides as a presentation method, there isn't a lot of explanation on the slides themselves. If you're interested in a more explicit exposition, you might like to take a look at my paper "In Defence of Hume's Law" and Greg's and my "Barriers" paper, both available here (and both well under 20 pages!)
FEW promises to be unusually technologically advanced for a philosophy conference (ok, CS-readers, stop smirking) and, according to the schedule:
There will be wireless access in the room. You are encouraged to bring laptops with wireless capability.
So perhaps I'll be able to post some news from the conference. (If I'd been reporting on the AAP last year, I could have told you about Karen Bennett rescuing Kit Fine from sharks...)
Posted by logican at May 23, 2005 02:32 AM
Trackback Pings
The trackback address for this entry is:
http://www.logicandlanguage.net/trakbak.cgi/67
Listed below are links to weblogs that reference FEW:
» FEW from Antimeta
I'm heading off to the second annual Formal Epistemology Workshop tomorrow, so I probably won't be posting here for a week. (I'm not coming back from Austin until the 31st.) Last year I commented on a paper by Brian Weatherson... [Read More]
Tracked on May 23, 2005 05:55 PM
Comments
See you there!
Posted by: Kenny Easwaran at May 23, 2005 03:23 AM
A question about Hume's first law:
Suppose ought implies can. So if I ought to do it, I can. Presumably what I can do is descriptive: it follows from my personal abilities, human nature, social influences etc.
Now if ought implies can, then if I cannot do it, I ought not to. Modus Tollens.
But now if what I can do is descriptive then what I ought to do is determined by what I can do. In which case(here is the question), are we not making out normative claims based on descriptive claims and Hume is wrong?
Posted by: Brendan at May 24, 2005 03:22 PM
Alas, I won't be at FEW, but I'm looking forward to reading what you and Greg Restall say about Hume's 1st Law (about how 'is' and 'ought' are totally different "copulations of propositions", as Hume so memorably put it....).
I'm weakly inclined to think that the last word on the Is-Ought issue is probably in Gerhard Schurz's monumental work of deontic logic, The Is-Ought Problem: An Investigation in Philosophical Logic (Dordrecht: Kluwer, 1997).
Schurz gives an answer to Brendan's objection, by the way, if I remember aright. Roughly, 'Ought' implies 'can' is what Schurz calls an "analytic bridge law", and he argues that the only plausible examples of analytic bridge laws are in a fairly precise sense neutral on all substantive normative controversies.
Posted by: Ralph Wedgwood at May 30, 2005 09:47 PM
If 'ought' implies 'can', 'cannot' implies 'not-ought', not 'ought-not'. If 'O' is a deontic necessity operator and 'M' a possibility operator, 'Op' entails 'Mp', so 'not-Mp' entails 'not-Op'. (But 'not-Mp' does not entail 'O(not-p)'.) So, from the claim that Sam cannot stop eating meat, it follows that it is not the case that Sam ought to stop eating meat. But it doesn't follow that Sam ought not stop eating meat.
'Can', like some other English modals, behaves strangely with negation; negation nearly always takes wide scope over 'can', which is why the way to formalize 'cannot p' is 'not-Mp' rather than 'M(not-p)'. (If you iterate negations you can get the other reading: "I cannot not do my homework". Here the first 'not' takes wide scope, the second scopes under 'can'. There are probably other ways to get the reading as well.)
Nevertheless, perhaps your point stands, for from a descriptive claim about one's inablities one can get a normative claim, so long as one counts claims about what is not obligated to do 'normative'. But maybe that doesn't conflict with Hume's claim?
Posted by: Dilip at May 31, 2005 08:21 AM