9/30/2009

Russell and Cubism

In my upper-level History of 20th C Philosophy course, we're currently reading Our Knowledge of the External World. Readers of OKEW will recall Russell's logical construction of "thing" (at an instant) in Lecture III: the set of aspects that would normally be said to be aspects of that thing. Aspects are sense-data. (These aspects include not only those actually perceived, but also those sense-data that would be perceived if a perceiver were there.) So, in other words, a thing (at an instant) is defined as the set of all the ways the thing would look (and smell, and feel, etc.) at that instant.

What struck me was how similar this is to the essential aim of cubist painting, which aims to capture multiple perspectives or aspects of an object simultaneously, on a single canvas. How far can this comparison between Cubist objects and Russellian ones be pushed? There's at least one difference: Cubist paintings (so far as I know) do not attempt to capture every perspective, just multiple perspectives -- whereas Russellian things exhaust all perspectives.

Also, someone else must have thought of this comparison before. Any references?

9/09/2009

Moving, plus evidence and truth-value

Very long time, no blog. The main reason is a cross-country move and a new job. For those who don't already know, I'm now at Hobart & William Smith Colleges, in upstate New York. But I also spent a very nice 2+ weeks in Australia, seeing old grad school friends as well as meeting new family members.

A discussion with a student today brought up an interesting question (which probably has been extensively explored in a literature of which I'm completely ignorant). We were talking about what sorts of linguistic expressions were truth-valued, and which were not. She suggested that there might be some sort of connection between being the kind of thing that has a truth-value, and being the kind of thing that can have evidence count for or against it. But what's the connection between these two conditions? Necessary? Sufficient? Neither? Both? -- and is there already a thorough treatment of this question in the literature?

Note: As I understand it, this is not a variant of verificationism, since one can imagine evidence that is unavailable to us weak, frail humans, with our very limited epistemic powers.

Note 2: If you take the position that the sentence 'Planet Vulcan is between Mercury and the Sun' is truth-valueless, instead of false, then that sentence is an example where we have evidence against a claim, but it nonetheless lacks a truth-value. But maybe the other direction of the connection would still hold?