Nevertheless, prompted by Mark Liberman's LL post, I decided to read Quasiregularity and its discontents. I had no expectation at all that it would have anything to do with semantics, but I remember reading with excitement Rumelhart and McClelland’s 1986 chapter about a parallel distributed processing model learning the past tense, and I was very curious to know how things have developed in the fifteen years or so since I stopped reading much in that area.
And yet I did find various areas of relevance. The repeated references in the article to the effect of phonology and semantics suggested that I was right to feel that I need to learn more about that area and support the claims of various theorists that a theory of syntax has to deal with semantics.
The discussions of modeling also reminded me of a paper by Gelman & Shalizi I read this summer, which I got the general idea of, but the particulars of which often required statistical knowledge outside my ken.
Not a particularly coherent first post, but I've got to get back to reading.
Gelman, A., & Shalizi, C. R. (2013). Philosophy and the practice of Bayesian statistics. The British Journal of Mathematical and Statistical Psychology, 66(1), 8–38. doi:10.1111/j.2044-8317.2011.02037.x
Seidenberg, M. S., & Plaut, D. C. (2014). Quasiregularity and its discontents: The legacy of the past tense debate. Cognitive Science, 1–39. doi:10.1111/cogs.12147