Status Updates From LOT 2: The Language of Thou...

“In their deepest heart of hearts, the great majority of psychologists think that talk of “truth” is unscientific (not to say vulgar)… namely it’s got to turn out to be some sort of association, either among ideas or between ideas and the world… and it’s not hard to imagine a machine that can form associations. Imagining such machines is what connectionists do for a living.”

“The ways in which different kinds of things are similar to one another aren’t, in general, similar to one another.”

“Tokening M(X) pretty generally doesn’t cause want to think its associates; a good thing to since if it did one thinking would be forever, getting derailed by one’s associations…
your cognitive lexicon isn’t network of associations. It’s a system of files… associationism took the thinking out of thinking…”

Yes and that’s why word2vec models are NOT models of natural language semantics

“THERE IS, AS A MATTER OF PRINCIPLE, NO SUCH THING AS A PSYCHOLOGICAL THEORY OF MEANING” (capitalization is from Fodor himself, delightful)

Pretty great so far