[R-lang] Re: Grammaticality judgments

[lngmyers@ccu.edu.tw]

Sorry to be pedantic, but there is no such thing as a "grammaticality
judgment". It's an "acceptability judgment". This terminological laziness
causes no end of confusion. Acceptability is observed; grammaticality is
hypothesized. Acceptability is evidence for or against a grammatical
hypothesis. This distinction goes back to Chomsky (1965) at least, as does
the claim that both acceptability AND grammaticality may be gradient, but
according to different scales.

So... if acceptability is gradient, this may be because grammar is
gradient, or it may be because grammar is categorical and something else
is making the acceptability gradient. The same caveat holds if
acceptability is close to categorical under certain circumstances - maybe
grammar is still gradient, but getting clumped up by something else.

Fortunately with clever experimentation and statistics these various
possibilities will eventually get sorted out. Unless people still insist
on using the same term for two totally different concepts.

James Myers
http://www.ccunix.ccu.edu.tw/~lngmyers/
Lngmyers@ccu.edu.tw