dan at daneverett.org
Thu Mar 31 10:02:20 UTC 2011
In fact, there is at least one very active group that works with evaluating claims of simplicity and complexity, Josh Tenenbaum's lab at MIT's BCS Department. Their approach is quite different than you might expect, though, testing the relative complexity of the grammars needed to describe a language. Ted Gibson and Amy Perfors have worked with Josh and others to produce some interesting studies in this vein. One paper that has emerged from this research is here: http://tedlab.mit.edu/tedlab_website/researchpapers/Perfors%20et%20al%20InPress%20LingReview.pdf
On 31 Mar 2011, at 05:48, Esa Itkonen wrote:
> Simplicity and complexity are conceptually interdependent: if, and only if, you can define one, you can define the other. Between 1957 and c. 1997 it was confidently predicted that a valid definition of simplicity (conceptualized as a "simplicity measure") was just around the corner. But, as we all know, nothing came of it. Nowadays much the same is being claimed about complexity. This seems illogical, however, for reasons just indicated. (Never mind that simplicity and complexity are mainly thought to apply to grammars and languages, respectively. It would surely be odd if the simplicity/complexity of grammars in no way reflected the simplicity/complexity of languages.) Why is all this so difficult? Some hints at an answer may or may not be gathered from my 2011 piece on 'Simplicity vs. complexity' (= click first 'Homepage' and then 'Selected writings available as full texts'). Some historical and conceptual background is provided by 'Philosophy of linguistics' (= 2011, to a
> ppear in the 'Oxford Handbook of the History of Linguistics'). You are also free to have a (second?) look at what I wrote about this topic back in 2009.
> Homepage: http://users.utu.fi/eitkonen
More information about the Funknet