General linguistic terminology (bound morph(eme))

[Laurence Horn]

At 10:10 AM -0400 5/27/04, Jesse Sheidlower wrote:
>I wonder if we can drift a little afield to discuss some
>general linguistic terminology.
>
>A correspondent recently asked if there is a name for
>linguistic forms that can only occur with an affix
>(except perhaps in jocular use[1]), as "emptive" (no
>"pre-"), "gruntle" (no "dis-"), and so forth. My
>first thought was "bound morpheme". I looked this up
>in the CGEL and didn't find anything relevant; the
>Oxford English Grammar gives "bound morph". I had
>never heard this before, and Google gives only a
>small number of examples, though from technical-ish
>sources. There are a lot more examples of "bound
>morpheme".
>
>Is "bound morph" a favored term in linguistics, and if
>so, what's the inspiration for the switch from "morpheme"?
>
>I was also concerned with the applicability: I usually
>see this in reference to affixes, such as -ly or -ed.
>Is it acceptable to use "bound morph(eme)" in relation
>to roots? If so, would they be called "bound root
>morph(eme)s"? Or is there another term?
>
>Thanks.
>
>Jesse Sheidlower
>
>[1] Interested readers are highly recommended to seek out Jack
>Winter's superb piece "How I Met My Wife," in The New Yorker,
>25 July 1994, which sustains for an entire page a stream of
>such and similar forms. Links to presumably illegal copies may
>be found on Google and won't be reproduced here.

I regularly give the Jack Winter piece to my students (I'm assuming
that's fair use, but I don't really know) as an exam question and ask
them to identify the process these forms illustrate, where the
intended answer is back-formation.   Granted, they represent a
subcategory of back-formation, since other cases of back-formation
don't involve an actual affix getting stripped away, but that is part
of what's going on here.

Larry