eggcorn candidate: "undermind"

[Laurence Horn]

>Laurence Horn wrote:
>
>>
>>"I don't want to undermind her"
>>
>>This struck me as more of an eggcorn than a slip or "performance
>>error", especially since I could imagine a context in which
>>reanalysis might have taken place, e.g.:
>>
>>"I don't want to undermine Dad" ==> "I don't want to undermind Dad"
>>
>
>Larry, these are interesting examples.  But wouldn't it be simpler to
>assume that the proximity of 'e' and 'd' on the qwerty keyboard,

I'm as fond of the null hypothesis and Occam's razor and
hoofbeats-mean-horses-not-zebras as the next analyst, but I don't see
the force of this objection.  'e' is as close to 'w', 'r' and 's' on
the keyboard as it is to 'd', but I don't imagine we'd get the same
576 ghits for "want to underminw", "want to underminr", or "want to
undermins" that we get for "want to undermind".   In fact what we get
is a grand total of 0 hits for all three of these, which makes it
vanishingly unlikely that "want to undermind" is in fact a typo.

>  as well as
>their proximity in many lexical forms ( -ed words) argues for mere
>typographical error?

Not sure what you mean by proximity in lexical forms.  I don't think
you get people saying (or writing) "I mind some gold last week".  I
checked "mind some gold", which by your hypothesis ought to bring in
typos for either "mine some gold" or "mined some gold", and got
nothing relevant.  (I'm not counting the likes of "I wouldn't mind
some gold coins and diamonds".)

In any case, how would qwerty keyboards or typographical errors
affect the speech substitution?

>
>(I don't know what you mean by 'performance error'...)
>
I'm alluding to Noam Chomsky's distinction between competence and
performance.  Linguistic productions that violate the rules of the
competence grammar have occasionally be dismissed as performance
errors, especially when fatigue, inebrity, etc. are factors.  Slips
of the tongue are an example.  I'm claiming that's not involved here,
that speakers say and write "undermind" and write because that's the
form (or one of the forms) they have for what the rest of us say and
write as "undermine", and that a reanalysis is responsible.

Larry