January 16th, 2006

pedant, law and order


I've been reading Robert Wright's Non-Zero: The Logic of Human Destiny, recommended by a friend who's an "old high AI" person (she thinks, for example, that relational databases are a good model for thought and truth, which I find frustrating).

Anyway -- Non-Zero is a 300-page book with a one-paragraph premise: that human culture (and biological evolution as well) has a stochastic trend towards greater levels of organization, because non-zero-sum parts of the world require collaboration to harvest the (relatively) positive cells in the matrix. While it's an interesting premise, I think it would have gained substantial strength from distillation into a 30-page pamphlet instead, and it doesn't seem to present an argument that would convert those not already very close to the idea of collaboration and cooperation. The evidence presented is anecdotal and handwaving, and seems to thrive on three or four example tribes (the Shoshone of the scrub desert and their rabbit-catching insta-mini-government "rabbit bosses" get a little too much screen time, but nevertheless we learn no additional details about the form of their society).

Wright argues that this model justifies cultural evolutionism, which was rejected in the 20th century as a form of thinly-disguised racism and a justification for imperialism. I am uncomfortable with this argument, since Wright's own model bears little in common with the theories of cultural evolution that the cultural anthropologists of the 20th century so opposed. Wright then assumes that Margaret Mead and Franz Boas would be opposed to his proposal, but many of their objections to 19th and 20th century "cultural evolution theory" do not apply to his proposal. Nevertheless, Wright positions himself as the radical upstart, when it is not at all clear that he is upsetting any applecarts at all.

Wright's writing style has an occasionally off-putting contrast between form and content: while Wright clearly wants to be taken as seriously engaging with the questions of cultural evolution, he also occasionally drops into arguing with sarcastic analogy or jokey colloquialism. While the informality is occasionally welcome, it belies his attempt to be taken seriously -- argument by buddy-ness is not convincing.

Overall: (blackwingedboy take note) don't spend your paycheck on it. If you're interested in cultural evolution questions, skimming this as a survey isn't a bad idea. But it's not a serious introduction to the subject, and its overall argument can (as above) be summed up in a page or two. Despite Bill's interest, this is what libraries are for.
linguistics, asl

[lazyweb] Searching a social network for [tact]

caracola starts a discussion that includes the statement "ASL has no sign for tact". This statement struck me as suspiciously snowclone-ish, and I'm curious if anybody might know about such a sign in ASL.

Some of my friends here might have resources that have more information. Any ideas of where to look? Anybody know the sign?

To complete my jargon and topic list: social networks (how), sign language (what), and snowclones (why)!
amused, smiling

Artifact ID day

okay, I think the local Artifact ID Day is pretty cool.
Do you have something that you can't quite place? If it's from somewhere along the Pacific Rim or North America, one of the Burke's curators may be able to help you figure it out. Experts will be on hand to identify objects from the Northwest Coast, Pacific Islands, and Asia, in addition to specimens from throughout the natural world (including bones, fossils, or animal teeth).
I live in an apartment, so this isn't something I will bring something to, but it's the kind of thing that would be fun with kids.

[lazyweb] CAD programs for Linux?

well, not so lazy -- I've done a fair amount of googling, but there is much linux power and knowledge among my F-list.

I'm toying with CAD programs for Linux, partially to avoid other work and partially because I'm curious, and I'm looking to move into a new apartment. It would be pretty neat to be able to lay out the floor plan and have a digital model of it. (useful, maybe not. but neat!)

any suggestions? Inkscape seems too low-level -- extremely powerful, but not an architecture tool. QCaD seems promising, but I haven't tried it out (it also may be headed to closed-source). Am I missing any other major contenders?