floating clitics

[Jan Terje Faarlund]

At 17:18 01.12.99 -0500, Matthew S Dryer wrote:
>I am interested in finding examples of what I call "floating clitics",
>clitics which must attach phonologically to some other word, but where the
>host phrase or word cannot be defined in terms of some specific position
>in the clause (like second position or final position) or in terms of some
>specific host (like subject or verb).  My impression is that this sort of
>clitic is not uncommon, but does not seemed to be discussed much in the
>literature on clitics that I have looked at.
>
This type of clitics can be found in Zoque (Mixe-Zoquean, Mexico), where
cliticization in fact seems to be the predominant morphological
process/type. This language has a number of morphemes that always attach to
a host, but the host can belong to various categories and be in different
positions. One such clitic is _na'ahk_, which marks past tense. This may
attach to most lexical categories, such as a verb (a), a noun (b), a
numeral (c), a determiner/pronoun (d), a quantifier (e), an adverb (f), or
a participle (g).

(a)	Ihtungna'ahk tumnaka tumø pøt 
	be-PAST one-time one man 
	'Once upon a time there was a man'

(b)	ngyomohindina'ahk ihtpabø,
	his-wife-COMITATIVE-only-PAST live-DURATIVE-RELATIVE 
	'Who lived alone with his wife'

(c)	Myehchkøyihindi'ungna'ahk ihtyahpa.
	two-together-COMITATIVE-QUOT-PAST live-PL-DUR 
	'The two lived together'

(d)	Te'se'ungna'ahk nøhayahpa 
	that-like-QUOT-PAST they/him-call-APPL-PL-DUR 
	'That is what he was called'

(e)	Te ngyomo'isñe ni'isungna'ahk hi myuhse ñøyi.
	the his-wife-POSS nobody-ERG-QUOT-PAST NEG knew name 	
	'Nobody knew his wife's name'

(f)	Hene'ungdena'ahk tømahpabø wi'na'a te' kama,
	very-FOC-PAST yield-DUR before the cornplant	
	'The cornplant was very productive before'

(g)	te pøt ñøyibø'isna'ahk Ngo'daka
	the man call-PRTC-PAST Ngodaka 
	'he man called Ngodaka'

Eaxamples (c), (d) and (e) also show examples of another clitic with
similar properties, the quotative _ung_, which always precedes _na'ahk_
when they both occur with the same host. I should add that clitics are easy
to distinguish from free morphemes (words), since word boundaries are
always clearly marked, both prosodically (penulitmate stress) and by (the
lack of) various segmental processes, such as metathesis and palatalization.

Less exotic (from my point of view at least) is Norwegian, where in most
dialects the unstressed third person pronoun has a clitic form, e.g. masc.
'n (full form 'han') and feminine 'a (full form 'ho' (nom.) and 'henne'
(acc.)). Except for a few well defined cases, the unstressed pronoun
occupies the same position as a full NP, and thus attaches to whatever word
that precedes it. Norwegian being a regular V2 language, the host may often
be a finite verb (a), but it may also be a complementizer (b), a non-finite
verb (c), a preposition (d), a noun (e), or a pronoun (f); the host may
even have another preceding clitic (g), etc.

(a)	Da kom'n = then came-he
(b)	at'a gjekk = that-she left
(c)	No har eg sett'a = now have I seen-her
(d)	Vi tenkte på'a = we thought of-her
(e)	Da møtte Jon'a = then met John-her
(f)	Når møtte du'n? = when met you-him (=when did you meet him?)
(g)	Når møtte'n'a? = when met-he-her (=when did he meet her?)


********************************************
Professor Jan Terje Faarlund
Universitetet i Oslo
Institutt for nordistikk og litteraturvitskap
Postboks 1013 Blindern
N-0315 Oslo (Norway)

Tel. (+47) 22 85 69 49 (office)
     (+47) 22 12 39 66 (home)
Fax  (+47) 22 85 71 00