Date: Wed, 22 Oct 1997 16:18:40 EDT
From: Larry Horn LHORN[AT SYMBOL GOES HERE]YALEVM.CIS.YALE.EDU
Subject: Re: whole nuther ballgame
Fritz Juengling writes
This kind of reanalysis is legion,
both in this direction (an ewt -- a newt) and especially in the opposite (an
orange, an umpire, an apron: all from stems with historical initial n-).
add 'nickname' from 'an ekename' to the list.
The SED also turned up 'nants' for 'ants' and 'nangnail' for 'hangnail.' The
progression must have been a hangnail an angnail (with usual loss of
h]) a nangnail.
I wonder. Since, as I think Dennis P. reminded us recently, 'hangnail' is
a folk-etymologized (or, as the OED puts it, an "accommodated") variant of the
earlier a(n)gnail (which was around for 8 centuries before the first 'hang-
nail' showed up, I don't think Fritz's chronology is that likely. In fact,
the history (as given under 'agnail' in the OED is quite interesting; when
'hangnail' first appeared, it was itself viewed as a dialect form, along with
the Scottish 'anger-nail' and 'nangnail'. As for the latter, I'm not sure this
was your standard metanalysis, as in the nuncle/nickname/newt set. Wright's
English Dialect Dictionary includes not only 'nangnail' and 'gnangnail' for
'corn', 'bunion', 'ingrown nail', etc., but also the verbs 'nang' and 'gnang',
with the sense 'gnaw', 'ache'. There seems to be a clear relation here, if
only one of an obvious folk-/motivated reanalysis. Both (g)nang and (g)nang-
nail were especially frequent in Yorkshire dialect, it appears.