[Lexicog] Circumfixes

[Mike Maxwell]

Andy Black wrote:
> I'm wondering what the standard ways are to show circumfixes in lexical 
> entries.
> 
> For example, does one create only one entry with the circumfix or does 
> one create, say, three entries:
> 
>    1. The left member (presumably a prefix)
>    2. The right member (presumably a suffix)
>    3. A combination of the two members.
> 
> In the entry for the circumfix, how does one typically indicate its form?
> 
> What does one do if either or both members of the circumfix have allomorphy?

A few thoughts on this thread.

1) Affixes in general are often clumsy in dictionaries, and you have to 
be a pretty sophisticated user to know enough to look them up among the 
words.  I suspect (but don't know) that most English dictionaries have 
few if any inflectional affixes listed--I wouldn't imagine -ed, ing, -en 
etc. are there; and few  derivational affixes, with the possible 
exception of the most common ones (un- and re-, for example).

A perhaps better way of handling affixes in dictionaries is to include 
them in a small grammar sketch.

IIRC, Bartholemew and Schoenhal's book on dictionary making talks about 
this question of putting affixes in the alphabetic part of the dictionary.

2) Some theoretically-minded linguists have claimed that there are no 
circumfixes (in Albert Bickford's strict sense; discontinuous morphology 
is well attested in the Semitic family, of course).  Anderson has used 
the existence of circumfixes to argue for his "amorphous" morphology, 
while others have use the extreme rarity of circumfixation to argue 
against his theory.

Circumfixes are briefly discussed on pg. 62 and 129 of The Handbook of 
Morphology, and of course in lots of other places.  The comment by 
Robert Beard on pg. 62 is worth repeating here:

   Circumfixation...is merely extended exponence involving
   a prefix and a suffix simultaneously.

This is a theory-based point, to some extent; Beard works in a version 
of Realizational Morphology.  Nevertheless, his point pertains to many 
of the putative examples of circumfixes, including some brought up in 
the course of this thread.  (And other theories of morphology, like 
Distributed Morphology, would I believe also re-analyze circumfixation 
as the accidental coincidence of meaning of two affixes.)

Rick Nivens brought this issue up in his discussion of infix/suffix 
pairs in West Tarangan, where he noted that while you might consider -i- 
"3s" and -na "3s.animate" to be such a pair, there is in fact a 
difference in what the two affixes mean: the suffix, but not the infix, 
signals animacy.  For that reason, the infix may appear where the suffix 
does not, making it look very much like two affixes that happen to 
partially share meaning, rather than a (in-)circumfix.

Many other examples of circumfixes can be analyzed in the same way, as 
being the accidental coincidence of a prefix and a suffix with 
overlapping meanings.

There is a good illustration of extended exponence (if not a discussion 
about circumfixes), and motivation for an early version of Realizational 
Morphology, in

    Matthews, P.H. 1972. Huave Verb Morphology: Some Comments
    from a Non-Tagmemic Viewpoint. International Journal of
    American Linguistics 38:96-118.

3) WRT their canonical representation, Gary Simons and Larry Versaw's 
manual for the old IT program talked about how to represent them in 
interlinear text.  I don't believe they talked about representing them 
in dictionaries.  Unfortunately, I just (reluctantly) threw out my copy 
of the IT manual, in preparation for a move...

4) Perhaps the best-known example of a (putative) circumfix is the 
German ge...en, as in 'gefunden', the past participle of 'finden' (root 
'find').  There's lots to be said about this from a theoretical 
standpoint, though, and it too falls into the category of suspect cases 
for being a genuine circumfix.

5) It may be that Semitic-style discontinuous morphology (and cases like 
German umlauting in plurals, mentioned by Albert) is more common than 
circumfixation consisting of a prefix and a suffix having identical 
meanings for reasons having to do with the historical origins of these 
patterns.  So from a practical standpoint--how (or whether) to represent 
discontinuous morphology in dictionaries--the issue remains, regardless 
of the theoretical status of "true" circumfixes.
-- 
	Mike Maxwell
	Linguistic Data Consortium
	maxwell at ldc.upenn.edu

	"When I get a little money I buy books;
           and if any is left I buy food and clothes."
	--Erasmus


 
Yahoo! Groups Links

<*> To visit your group on the web, go to:
    http://groups.yahoo.com/group/lexicographylist/

<*> To unsubscribe from this group, send an email to:
    lexicographylist-unsubscribe at yahoogroups.com

<*> Your use of Yahoo! Groups is subject to:
    http://docs.yahoo.com/info/terms/