« May 2005 | Main | July 2005 »

June 16, 2005

St Louis

Here's something that I've been meaning to post for a while. It seems that Peter McBurney knows a lot more about the logic scene in St Louis than I do. He writes:

I keep meaning to ask you if you know that St. Louis (which I know you're currently away from) is a hotbed of argumentation theory? One of the founders of computational argumentation, Ron Loui, is in Computer Science at Wash U:

http://www.cse.wustl.edu/~loui/

and in the Philosophy Department at SLU is William Rehg:

http://www.slu.edu/colleges/AS/philos/frehg.html

I have co-authored a paper with Bill and my former PhD supervisor (Simon Parsons), on assessing the socio-political legitimacy of systems for computer-supported argument.

Ron Loui supervised the PhD of Guillermo Simari, who returned to his native Argentina and by force of personality established the National University of the South (UNS) in Bahia Blanca as a major world centre for computational argument. (The other locations are the Universities of Liverpool and Utrecht, IRIT in Toulouse, and the AI Lab at Cancer Research UK in London.)

Appropos nothing, really, except that St Louis is famous for more than TS Eliot and Miles Davis!


And the arch, of course. And Scott Joplin and Chuck Berry.

And that fabulous song lyric:

St Louie woman, with her diamond rings,
Drags that man around, by her apron strings.
If it wasn't for powder, and her store-bought hair,
that man I loved, wouldn't have gone nowhere.

It's quite a city, though I've yet to find the hair store. I was back there for a few hours yesterday, (I'm on the road a lot at the moment - I'm in London now and it'll be the southern hemisphere by the end of the month) and was happy to have a chance to catch up with my computer science colleague, Aaron Stump.

Thanks Peter.

Posted by logican at 4:52 PM | TrackBack

June 13, 2005

On the BBC

A link sent by (of all people) my dad. (This is noteworthy because my dad is, in general, rather suspicious of philosophy. I think it stems from when I was an undergraduate and tried to explain the problem of induction to him. Since then he's suspected that philosophy secretly wants to subvert science (he trained in chemistry - surely the most real of the traditional sciences. I bet if we took a survey of working scientists and asked them whether they thought their best theories were true or merely empirically adequate, the chemists would have the highest proportion of people answering "true." Maybe I'm biased when it comes to chemists though. I lived with a bunch of graduate chemists for years. You can see why I need a blog - I could go on like this forever...)

Anyway, Radio 4 is having a vote to see who the public considers to be the greatest philosopher of all time. You can also take their philosophy quiz, and, best of all, see what various celebrities said when asked who the greatest philosopher of all time is. In an ideal world, this question would have been put to the likes of Britney Spears, Posh and Becks, and the cast of Big Brother and you'd have to drink every time someone replied "my mum." (If I'd had to pick people to ask I think Moby, Julian Cope and Bob Geldof might have been high on the list. Andrew Marr and Stephen Fry could definitely stay on too.) But in this world they've asked the likes of Anne Widecombe and Mariella Frostrup. As far as I can tell, each respondent has only had to pick a name and perhaps a quote, and then someone from Radio 4 wrote up a few paragraphs about each philosopher, complete with links to your friend and mine, the Stanford Encyclopaedia of Philosophy.

Who is my favourite philosopher of all time? I don't really know. I might have said Tarski until recently. It's a hard call because you want it to be someone who's main message you agree with, but it also seems to matter than they worked on something important, and then it should be someone with whom you feel something of a connection. So here's a short list. It's very mixed, and people are on it for very mixed reasons: Berkeley, Tarski, Epicurus, Mill, Kripke, Wittgenstein, Marx, Russell. The thing is, half the people on this list I kind of hate too. I don't know how we ever manage to like whole people; they're so complicated.

Posted by logican at 3:07 PM | TrackBack

June 10, 2005

Kripke's Incompleteness Proof

Michael De, over at Lumpy Pea Coat, was wondering whether there is any literature on Kripke's version of Gödel's incompleteness proof. As he puts it:

Kripke gave a proof of the incompleteness theorem through ventriloquism in a lecture at Beijing University in 1984 using a very different (algebraic) approach, but I cannot find it anywhere. Does it exist? Has the lecture been transcribed? Anyone?

Richard Zach came through with the most important source, Putnam's "Nonstandard Models and Kripke's Proof of the Gödel Theorem" Notre Dame J. Formal Logic 41, no. 1 (2000), 53–58.

But I could also remember a talk that David Lewis gave at the University of Melbourne on Kripke and a version of the incompleteness proof, and Allen Hazen, who was in attendance, had turned out to have some rare notes that were relevant to the topic. Unfortunately it's not a lost Kripke manuscript, or even notes on a Kripke lecture, but, as Allen writes:

Notes Graham Solomon (Philosophy Dept, Wilfried Laurier University, deceased) sent me, coming from Bill Demopoulos (Philosophy Department, U of W Ontario), of -- not Kripke's original lecture, but -- a lecture by ... you guessed it ... Hilary Putnam.

Allen went on, in subsequent email:

Kripke, when he actually does write, gives marvelously followable expositions, and the proof is interestingly different from G's. More different than is George Boolos's proof, which I assume you are familiar with (ch. 26 of "Logic, Logic and Logic").

(In my opinion, it's worth listening when Allen says "I assume you are familiar with..." All through my Australian breaks from grad school he completed that sentence with the names of brilliant and wonderful things I'd never come across.)

He has also kindly given me permission to make the Putnam/Kripke/Solomon/Demopoulos notes available here, for anyone who's interested:

Posted by logican at 1:43 AM | TrackBack

June 8, 2005

logicnazi moves out

I don't know whether any of you guys are interested in, you know, logic, or philosophy, (or even math(s)), but should you be, you'll be happy to learn that frequent (and appreciated) commentor logicnazi has started a new blog called "Computational Truth." He writes:

The name computational truth refers to my two majors areas of interest. My mathematical interest in computability/recursion theory and my philosophical interests in truth and representations thereof. Well really my philosophical interests may be closer to the theory of mind but that is only because I think answers to the puzzles of truth (like the liar) and representation lie in the theory of mind.

To which I say, to the first part: cool interests, and to the second part: huh?

The idea that the answer to the Liar lies in the philosophy of mind reminds me (unfairly, I'm sure, since logicnazi hasn't elaborated) of trying to explain the Liar to someone in a club once (you think I'd learn...) and getting the response: but that doesn't seem hard to figure out; lots of people can just tell when someone is lying. (I suppose the idea was that we utter the sentence in conversation with one of them and they will be able to assign the appropriate truth-value based on our shifty demeanour or lack of such.) But, other problems aside, this is to be distracted by inessential features of the paradox (like the fact that it's called "the Liar" and that one traditional way of setting it up involves a Cretan claiming that all Cretans are liars) for the paradox itself. We can recreate the paradox without any mention of anyone lying, using only classical logic and our favourite unrestricted disquotational T-schema, and so the paradox seems squarely located in logic and the philosophy of language. Something has to give, either in our logic, or in our theory of truth. Since we have the paradox without commitment to any particular principle in the philosophy of mind, how could getting principles in the philosophy of mind right help us here?

Well, probably I'm not using enough imagination. I suppose work in mind might give us an error theory of our acceptance of one of the principles or argument forms (notice how I managed not to write "inferences"? I really am trying to be consistent about the inference/implication thing) that land us in the paradox, in the sense that it might explain how we came to accept them, even though they aren't true or valid. That would be a non-trivial role for philosophy of mind (or at least, for psychology) in the solution to the paradox, but still, there would have to be a solution in logic or the theory of truth for it to support.

But enough confused speculation. logicnazi's new blog promises to be very intriguing...

Posted by logican at 11:15 PM | TrackBack

Men of substance, men of style

Hey there, Philosophy Man. Tired of plain, classic shirts? According to this article

Plain, classic shirts just wouldn’t do for the adventurous Philosophy male.

Never fear, Philosophy Men to the rescue.

Posted by logican at 2:32 AM | TrackBack

Via Arts and Letters Daily

Alicia Shepard reports her experiences of student complaints about grades. I was surprised by the same phenomenon last semester. I've never yet had the "but my parents pay your salary" complaint, but I have had the following given by students as grounds for a grade increase:

And just in case you're reading this, guys, none of these have ever worked. (To have even a ghost of a chance you need to go for miscalculation, death or serious illness.)

Posted by logican at 1:38 AM | TrackBack

June 4, 2005

Beamer Update

Beamer turned out to be very easy to use with TeXShop; I didn't run into the problems logicnazi encountered when using it with a LaTeX application installed via fink. I had to get a few extra packages from c-tan to get everything to run smoothly though: pgf.sty, xcolor.zip, xxcolor.sty and pgfshade.sty.

As a first project I converted my slides from the Formal Epistemology Workshop, and now they have colours, boxes, and "overlays" galore. My approach to the conversion was basically just to get it all more or less working and then trawl the Beamer manual for cool stuff to add. The result, as you might expect, would make Josh Parsons blanche, (honestly, check out that link,) but I think I'm in a position to produce something more, ... streamlined ... for the AAP in Sydney.

Posted by logican at 9:14 PM | TrackBack

June 1, 2005

Beamer

One more thought post-FEW. I really must download Beamer and use it for making presentations in future. The Beamer presentations looked very cool.

Posted by logican at 3:59 PM | TrackBack

Formal Epistemology

Now that I have a chance to draw breath, I'd like to say that the Formal Epistemology Workshop was fantastic, and I recommend next year's conference to anyone who doesn't come out in a rash at the sight of symbols. I think the mixture of sessions - papers from (exceptionally talented) graduate students, papers from established scholars, and tutorials in interesting topics - is a very good idea, and I learned a great deal, talked to many interesting and clever people, and came away with great comments on my paper - especially from my commentator, Peter Vranas (he didn't mention anything about the slide striptease trick) - and ideas for new work. Thank you very much to Sahotra Sarkar and Branden Fitelson for organising it all.

Like Jonah, I found that there was plenty of material that was over my head, but when I admitted this to Brian Skyrms he suggested that some of what one learns at a conference like this is new stuff from papers that one understands, but one also learns that there are these other subjects and tools out there which one might want to investigate later.

One highlight for me was Cristina Bicchieri's tutorial on experimental game theory. Two of the simple styles of game were Ultimatum games and Dictactor games. In an Ultimatum game, the first of two agents is given $10 by the experimenter and has to offer to divide it between themselves and agent 2. Agent 2 then chooses to accept or reject that offer. If they accept, the money is split as offered, if they reject, both agents get nothing. In the Dictator Game version, player 1 simply gets to decide how to split the money. What makes this experimental game theory is that instead of running the games with idealised agents, we run them with real people, in these cases as anonymous one-shot games. There are all kinds of variations on the games in which, say, agent 2 isn't told everything that agent 1 is told, and in which the two agents have to solve problems before the game and one is told that they did better than the other, or the agents are rechristened things like "buyer" and "seller." And the results are just interesting.

And sort of shocking. And in some cases kind of funny. In the simplest version of the Ultimatum game, one might expect the 1st agent to offer the lowest amount possible, and the responder to accept it. But (and this is kind of funny) it turns out that it's only small children who play the game this way. In general, offers below 20% are simply rejected half of the time and it's common to offer around or slightly less than half the money to the other agent. The experimenters thought that giving people more money to play with might make them behave "more rationally", but people still rejected offers below 20%.

Professor Bicchieri then considered some explanations for this - altruism, preference for fairness, social norms etc. - and used experimental games to test them, with a lot of surprising results. One distressing general trend was that people seemed to be more responsive to what they thought others would count as acceptable behaviour than they were to any independent judgements of acceptable behaviour (and so for example, even in one-shot games, whether or not their partner will find out how they choose makes a big difference to way they choose.)

I imagine this work will be of interest to ethicists like Gil Harman and John Doris, whose research draws on empirical results, and - though it isn't obvious exactly what follows from it for ethics - it does seem plausible to me that surprising empirical research like this could support some interesting work in ethics and political theory, especially in areas where the conventional wisdom on people's reasons for action is really dodgy (e.g. immigration.)

(Note - If you've been following the Hume's law discussion, that might - for a second - seem like an odd thing for me to say, but even if you can't get an ought from an is, lots of us believe all kinds of normative claims (e.g. torturing the innocent is wrong, polluting the earth is bad) and these could give us surprising normative results when teamed with surprising empirical data.)

Posted by logican at 10:13 AM | TrackBack