tense-marking by affixes on both non-predicate and predicate NPs and pronouns

[Matthew Dryer]

Claude Hagège refers to tense-marking on pronouns in Hausa and other Chadic 
languages.  I am not sure whether we are talking about the same thing, but 
if he is referring to words that code both pronominal features of the 
subject and tense features, such as /ya/ in the following example from 
Hausa (some diacritics not shown)

   mùtûm  ya              ginà   gida
   man    COMPL:3SG.MASC  build  house
   'the man built a house'  (Newman 2000: 719)

then I think it is a mistake to consider such words as pronouns that 
inflect for tense, or, for that matter, tense words that inflect for 
pronominal features of the subject.  Rather such words are like 
inflectional affixes on verbs that code both tense and pronominal features 
of the subject, except that they are separate words.  In that sense, they 
are not really pronouns.  Words of this sort are also found in other parts 
of the world, such as Niger-Congo and Austronesian.  But they are rather 
different from pronouns that inflect for tense.

Matthew Dryer

--On Saturday, January 24, 2004 1:59 PM +0100 claude-hagege 
<claude-hagege at WANADOO.FR> wrote:

>
> Dear LingTyp members,
>    Since Roland Hemmauer has already sent a summary, it might be to late
> to add something to this discussion, but I would like to point out that
> tense(not aspect, mood, etc.)-marking by affixes (not clitics) on both
> non-predicate and predicate NPs and pronouns is a fairly widespread
> phenomenon all over the world. The case of Guarani is well-described in
> several monographs and articles. The same phenomenon is found in many
> Amerindian languages, both in North- and South-America, besides Yukaghir,
> Samoyed and other languages already mentioned by Elena and Florian. I
> should add that in Hausa and other Chadic languages, tense-marking on
> pronouns is quite regular. Let me quote an illustration from Comox, in
> which, like in other Salishan languages, this a very frequent phenomenon:
> not only can non-predicate NPs be tensed, like tala "money" (from dollar,
> incidentally), in tala-elh (money-PAST) used as subject or object of a
> verb, but tensed non-verbal predicates are common, as in
> hégos-elh té'e (chief-PAST) this "this man was the chief".
>  More data in C. Hagège, Le comox lhaamen de Colombie britannique,
> présentation d'une langue amérindienne, Amérindia, Paris, Association
> d'Ethnolinguistique Amérindienne, 1981, and in "Du concept à la fonction
> en linguistique, ou la polarité vebo-nominale", La Linguistique,vol. 20,
> 2, 1984, 15-28.
> Best regards to all,
> Claude.
>
> Claude Hagège, Chaire de Théorie Linguistique, Collège de France, Paris.
>
>
>
>