Vocabula

[Arnold Zwicky]

david bowie, to robert hartwell fiske:
 >I'm a generally honest person, and I don't believe much language
 >change is due to ignorance. So, unless we're defining "ignorance"
 >differently, I'd like to see your proof--particularly your proof of
 >(and definition of!) "much".

well, yes, a lot depends on what counts as "ignorance".  internal
change - excluding dialect and language contact - in lexicon,
morphology, syntax, and semantics is very much a matter of two
mechanisms: reinterpretation/reanalysis (including
grammaticalization), and regularization/generalization. (in the case
of the lexicon, creative extensions of meaning via metaphor and
metonymy play an important role as well.)  both types of change
involve a mismatch between the grammars of succeeding generations of
speakers.  people in general are of course deeply ignorant of the
details of other people's grammars; we devise our own grammars on
the basis of what other people say, in what circumstances, and have
no direct access to their minds.  for the purposes of getting along
in social life, it's enough that our grammars be close enough.  they
could never be entirely identical.

we all work from a position of ignorance, and we get huge numbers of
things (a bit) "wrong", from the point of view of other, especially
older, speakers.  in many cases, lots of people will independently
make the same "mistake" - reinterpreting an adjectival participle in
the predicate as a passive, say, or refashioning irregular past
participles to make them identical to past tense forms - because the
mistake is invited by the data of the language; in retrospect, such
changes seem almost inevitable.  then the innovations spread from
individual speakers through social networks.  and, of course, more
idiosyncratic innovations can spread that way too, much the way
fashions of all kinds spread.

so "fun" becomes an adjective (as well as a noun) because new
generations of speakers are ignorant of its exclusively nominal status
for older speakers, but have the evidence of sentences like "That was
(really) fun" that "fun" patterns like an adjective.

ignorance is the human condition here.  so *certainly* it plays a
huge role in language change.

but to apply "ignorance" as an epithet, implicating that somehow
people *should* be aware of the status of material in other people's
(the best people's) grammars - indeed that they should know who counts
as the best people in the first place - and that they should recognize
that the grammars of the best people are intrinsically superior itself
betrays a deep ignorance, of the nature of language, not to mention
displaying a patronizing meanness.  (i understand that acting superior
can provide an enormous amount of self-satisfaction, but it's ugly,
and most unbecoming in people who want to present themselves as
representing the best of human values.)

i have no problem with scholars who provide information about the
details of established formal written standard varieties (as, i keep
pointing out, MWDEU and the Cambridge Grammar of the English Language
do).  in fact, i celebrate them.  they provide an important social
service, to specific audiences.  notice, however, that they don't
traffic in ignorance-talk.

arnold (zwicky at csli.stanford.edu)