on a tangent, has anyone heard of J. G. Wolfe? I have about 5 papers from
him written in the late seventies or early eighties. He develops a
computer program which learns the grammer. It was usually fed one text
over and over until the structures in memory stabalized, and all
irregularities (i.e. learning bought over buyed) were acounted for. I
found it fascinating reading--it fits very well with empiracal data taken
from children learning language--and if a) no-one else knows who I am
talking about, and b) someone desires it, I will go back over them and
present a more accurate summary.
blegh.
tom.