Date:  Tue, 12 Dec 1995 08:46:27 EST

From:  Larry Horn <LHORN[AT SYMBOL GOES HERE]YALEVM.CIS.YALE.EDU>

Subject: Re: can/can't



Don writes:



>When international students have asked about American pronunciations of

>'can' and 'can't' -- complaining that they don't hear a -t -- I point out

>that the vowel in 'can' lasts a little longer than the one in 'can't'.

>They find that explanation useful.   DMLance



Or you can refer them to Hans Marchand's "Remarks about English Negative

SEntences", American Studies 20 (1938): 198-204, who observes that post-

auxiliary negation tends to be signalled more by vowel quality, stress, and

rhythm than by the presence of a segmental element. In fact,Jespersen's "Nega-

tion in English" (1917, p. 11) also contains this remark:



   If we contrast an extremely common pronunciation of the two opposite

   statements "I can do it" and "I cannot [sic] do it", the negative

   notion will be found to be expressed by nothing else but a slight

   change of the vowel [ai kaen du: it | ai ka:n du: it].



Jespersen is describing a different dialect from mine, but the point is that

vowel quality and length and phrasal rhythm is more of a cue than the presence

of a [t]. Where the rhythmic distinction neutralizes is just when the modal is

contrastively stressed: he CAN come/he CAN'T come. (I discuss this on p. 458 of

my 1989 book _A Natural History of Negation_, where I note I've occasionally

heard the repair query "can-yes or can-T'?")





        Larry