Re: USAGE: Circumfixes

From: David Peterson

> How about English?
>
> long > elongate (no *elong or *longate)
That's a case of two affixes, a prefix (ex-'from') and a suffix (-ate 'to
make something gain an attribute'); I wouldn't consider that a circumfix.

> That sole example aside, though, Georgian's got tons of 'em (though not
actually in the verbal paradigm, as
> some have said).  Like the Malay example that was given, these are also to
derive nouns.  Here's a > website:
>
> http://www.armazi.com/georgian/
The Georgian cases are true circumfixes, since the prefix part or the suffix
part do not stand on their own: kartuli 'Georgian' > Sakartvelo 'Georgia'.
That case also has internal Ablaut, where /u/ becomes /ve/.

(Georgian is normally classified as agglutinative, but it looks more
fusional to me.)

Circumfixes can also be found in Arabic in the case of present-tense verbs:

?aktubu 'I write'
taktubu 'you write' (m. sg.)
taktubi:na 'you write' (f. sg.)
yaktubu 'he writes'
taktubu 'she writes'

The -i:na in the 2nd feminine singular is not used for the 3rd feminine
singular, so it cannot be used alone to mark a feminine subject.

A real peculiarity of Arabic is the use of a combination prefix (?i-) and
infix (-t-) for verbal class VIII, a reflexive class:

kataba 'he wrote'
?iktataba 'it wrote itself', or 'it was registered'

But the prefix part of that is really of euphonic nature, an epenthetic
vowel like initial e in Spanish _espan~ol_. Arabic words cannot begin with
two or more consonants, unlike Hebrew and Aramaic.

(I decided not to use Unicode this time.)