A-only agreement

David Gil gil at EVA.MPG.DE
Sun Sep 20 01:59:20 UTC 2009 

Dear Paul and everybody,

Of course I would analyze the English verb suffix -s as bearing 3rd
person features. However, the Classical Malay form di- differs from
English -s in the following relevant respects:

(a) first, di- is probably more appropriately analyzed as a clitic,
rather than an affix, a fact which renders it less characteristic of
agreement markers cross-linguistically (while, admittedly, not
precluding an agreement analysis).

(b) secondly, and more importantly, di- does not form a paradigm with
its putative 1st and 2nd person counterparts ku- and kau- (the way even
English -s does with all of the other zero forms in its paradigm).
Specifically, whereas in the di-V construction, an agent phrase may
freely occur after the verb; in the ku-V and kau-V constructions, ku-
and kau- ARE the agent denoting expressions -- it is either extremely
rare or impossible for another coreferential agent phrase to occur after
the verb (or anywhere else for that matter). (Also, I suspect that the
pre-verbal patient phrases of the two constructions differ with respect
to their respective arrays of subjecthood properties, though since I am
now in the field I can't consult the relevant literature to provide
precise references.)

For these reasons, I prefer the traditional analysis of di- as a
patient-oriented voice marker, as opposed to a 3rd person agreement marker.

I must confess, however, that, while I agree that each variety of a
language should in principle be described on its own terms, Paul is
correct in suspecting that my view of the situation in Classical Malay
is coloured by my familiarity with other kinds of Malay/Indonesian. But
in this particular case I think such colouration might be justified.
Classical Malay is a formal literary register that does not exist in
isolation: nobody spoke only Classical Malay. Rather, Classical Malay
lies at the top end of a lectal cline, at the other end of which are
various local colloquial varieties of Malay. And in most of these latter
varieties, the "3rd person only" constraint on the di- construction is
absent. (See the numerous prescriptive grammars inveighing against the
use of di- with 1st or 2nd person agents.) Accordingly, it is not
unreasonable to strive for a unified description of a diglossic
speaker's competence, which, I would suggest, might look as follows: (a)
Panglossic Malay has (i) one "passive"-like construction with di-,
unmarked for person; (ii) a second "passive"-like construction involving
the 1st and 2nd person proclitics ku- and kau-; and (b) the literary
register of Malay imposes an additional constraint to the effect that if
both "passive"-like constructions are possible, ie. in the case of 1st
and 2nd person, only the second, more specific, of the two constructions
may be used. Thus, in the literary register, di- ends up being used only
in 3rd person constructions without it being an agreement marker bearing
3rd person features.

Concerning the bibliographical references: since I am now in the field,
I am not in a position to check, but it is my recollection that there is
still an ongoing and very lively debate on the origins of di- in Malay
(cf. the recent paper by Sander Adelaar in a volume of Pacific
Linguistics). But of course, the origins of the form do not necessarily
tell us what the correct analysis of it is in each and every one of the
descendant dialects.

Best,

David


> Dear David,
>
> I'm not sure what you're getting at here. I guess I don't understand your
> distinction. You seem to be admitting that prefixal di- "is used only with
> third person agents" yet you say it does not "bear third person features".
> Would it be possible for an affix to "be used only with third person
> agents" and yet NOT "bear third person features"? Does the English verb
> suffix -s "bear third person features"?
>
> For the record: That in Classical Malay, transitive clauses having the
> verbal di- prefix also have a third person agent is not "my" analysis,
> it's the canonical one. See, e.g., recently, van den Berg 2004. The uses
> of this prefix in other kinds of Malay/Indonesian aren't at issue here.
>
> - Paul
>
>
> R. van den Berg, "Some notes on the origin of Malay di-" Bijdragen tot de
> Taal-, Land- en Volkenkunde 160 (2004), no: 4, Leiden, 532-554)
>
>
>
> On Sat, September 19, 2009 10:14, David Gil wrote:
>   
>> Dear all,
>>
>>
>> I think it's only fair to point out that Paul's analysis of the Malay
>> sentences below as involving verbal agreement with a transitive agent is
>> dependent on the assumption that the prefix di- bears 3rd person features
>> -- an assumption that many have questioned (even in those
>> somewhat idealized varieties of Malay in which the prefix di- is used only
>> with 3rd person agents).
>>
>> David
>>
>>
>>     
>>> Dear Peter,
>>>
>>>
>>> I have argued (references below) for "discourse split ergativity" (John
>>>  Verhaar's term) in traditional Malay narrative such that precisely the
>>>  distribution that you describe occurs in certain narrative
>>> environments. Examples from texts:
>>>
>>>
>>> di-ambil  oleh pawang sedikit batang pisang
>>>       
>    3:ERG-take ERG magician some stalk  banana "The elephant-magician took
> a piece of banana stalk"
>   
>>> di-ambil-nya surat itu, di-renong-nya, kemudian di-letokkan-nya
>>> 3:ERG-take-3AGT letter the, 3:ERG-stare-3AGT, then 3ERG-put-3AGT
>>> "He took the letter, stared at it, and then put it down."
>>>
>>>
>>> ku-bilang gajah itu
>>>       
>    I:ERG-count elephant the
>   
>>> "I counted the elephants"
>>>
>>>
>>> datang-lah se-orang orang muda come-EVENT a (classf) man young "a young
>>> man came"
>>>
>>> So the transitive agent has the preposition oleh if a noun, or a
>>> special verbal clitic if a pronoun, while absolutives (intransitive
>>> subjects and objects) have no marker on the NP, and have independent
>>> forms (aku "I", ia/dia "3rd sg.") if pronouns.
>>>
>>> The verb has the prefix di- if the transitive agent is 3rd person,
>>> whether noun or pronoun. Transitive agents that are pronouns have
>>> special forms that are clitic to the verb (1/2 are proclitic, 3 is
>>> enclitic). S and O show no agreement with the verb.
>>>
>>> Paul
>>>
>>>
>>> ----------------------------
>>>
>>>
>>> Paul J. Hopper, 1987 “Stability and change in VN/NV Alternating
>>> Languages:
>>> A study in pragmatics and linguistic typology.” In M. Bertuccelli Papi
>>> and J.Verscheuren, eds., The Pragmatic Perspective, 455-476. Amsterdam:
>>> John
>>> Benjamins.
>>>
>>>
>>> Paul J. Hopper, 1983 “Ergative, passive, and active in Malay narrative
>>> discourse.” In F. Klein-Andreu, ed., Discourse Perspectives on Syntax,
>>> 64-87. New York: Academic Press.
>>>
>>>
>>>
>>> On Sat, September 19, 2009 07:30, peterarkadiev wrote:
>>>
>>>
>>>       
>>>> Dear typologists,
>>>>
>>>>
>>>>
>>>> do you know of languages where the verb would agree exclusively with
>>>> the transitive agent (A), but neither with P nor with S? It seems to
>>>> me that I have come across such systems, but I cannot recall any
>>>> particular details. By the way, if per chance someone knows of
>>>> agreement systems where both transitive A and P are cross-referenced
>>>> but S is not, this example would be highly appreciated, too.
>>>>
>>>> Many thanks in advance!
>>>>
>>>>
>>>>
>>>> Peter Arkadiev
>>>>
>>>>
>>>>
>>>> --
>>>> Эмоциональна� почта находит�� зде�ь:
>>>> http://mail.yandex.ru/promo/new/emotions
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>         
>>>       
>> --
>> David Gil
>>
>>
>> Department of Linguistics
>> Max Planck Institute for Evolutionary Anthropology
>> Deutscher Platz 6, D-04103 Leipzig, Germany
>>
>>
>> Telephone: 49-341-3550321 Fax: 49-341-3550119
>> Email: gil at eva.mpg.de
>> Webpage:  http://www.eva.mpg.de/~gil/
>>
>>
>>
>>
>>
>>     
>
>
>   


-- 
David Gil

Department of Linguistics
Max Planck Institute for Evolutionary Anthropology
Deutscher Platz 6, D-04103 Leipzig, Germany

Telephone: 49-341-3550321 Fax: 49-341-3550119
Email: gil at eva.mpg.de
Webpage:  http://www.eva.mpg.de/~gil/

More information about the Lingtyp mailing list