"thing/think" [was: on the eggcorn beat]

[Arnold M. Zwicky]

On Apr 30, 2008, at 12:19 PM, Joel Berson (noting a difference in
judgments between jon lighter and benjamin barrett) wrote:

> Do I hear two generational divides within one lifetime?

there might be some association between age and choice of variant, but
i'd expect it to be weak.  there's probably some large-scale
geographical variation (between, say, American, British, and
Australian English) in the frequency of use for the two variants.  i'd
expect most other sociolinguistic factors to play a very weak role.

i make these predictions because i suspect think/thing is an "ice
plant variable", spreading primarily according to which variant you
happen to have noticed first.  whether you have "ice plant" as a count
noun or a mass noun (assuming you have it at all) probably depends on
which variant you encountered first.  so the spread of the two
variants has a significant amount of randomness in it.  there will
probably be a tendency for families to make clusters, but the items in
question are infrequent enough that it's unlikely the variants will
get associated with socially relevant factors.

but i could be wrong.

arnold

------------------------------------------------------------
The American Dialect Society - http://www.americandialect.org