use(d) to

[Arnold Zwicky]

peter mcgraw:
 >In speech, how could you tell whether someone is saying "didn't use
 >to" or "didn't used to", since both are pronounced identically as
 >"didn't useta"?  I suspect this is the reason for the uncertainty in
 >spelling.

agreed.  CGEL says essentially this.

 >Grammatically, "didn't use to" seems perfectly straightforward to
 >me, since one would also say, "I didn't use the washrag", whereas
 >"didn't used to" seems problematical, since one would not say *"I
 >didn't used the washrag."

this, however, is problematic.  it assumes that speakers connect
the verb in [just@] (conveying tense/aspect semantics) with the
ordinary verb [juz] 'employ', which seems pretty dubious to me.
(maybe some do, but it's a pretty big leap.)

the alternative is to assume that tense/aspect [just@] is always
spelled <used to>, regardless of the syntactic context.

arnold (zwicky at csli.stanford.edu)