Skip to content

Comment on Chomsky and the Two Cultures of Statistical Learning (2011)parent

Comments

Read the article above. There is a link at the top of this submission to an essay by Peter Norvig, arguing (correctly, in retrospect) that Chomsky's approach to language modelling is mistaken.

Obviously I did read the article. And I know how the hn site works.

I have a passing familiarity with the debate over Chomsky's theories of universal grammar etc. I didn't notice anything in the article that would cause disgust, and so I wondered what I was failing to understand.

If you have read many books by Chomsky, it might make you angry that you have wasted so much time on what turned out to be a fundamentally mistaken theory.

The people who downvoted this apparently didn't read the article.

Isn't "theory turned out to be wrong" just the price of doing science? And its a good thing: something has been learned.

AboutSource Built by g1lg1l

Hackerly is an independent reader for Hacker News, built on the public HN API. Not affiliated with Y Combinator.