LINGUIST List 4.447

Thu 10 Jun 1993

Sum: Object agreement

Editor for this issue: <>


  1. Dan Everett, Object agreement-Summary

Message 1: Object agreement-Summary

Date: Wed, 9 Jun 93 09:06:04 -04Object agreement-Summary
From: Dan Everett <>
Subject: Object agreement-Summary

I posted the following message to Linguist several days ago. Responses
I have received to date are appended to my original query. Thanks to
everyone who replied.


I am looking for nonergative languages in which verbs agree with
the direct object but not the subject. (The intransitive subject should
also not trigger agreement on the verb, else the language would count
as ergative.)

I know of only one language that operates this way (in New Guinea,
described by Foley, I believe). Anyone know of any others?

Individual responses should be sent to me directly. I will post a summary
of any responses I receive.

Dan Everett

Although this is probably not what you are looking for, you might want
to know that Icelandic verbs taking non-nominative ("quirky") subjects
and nominative objects agree with the nom. object in number (not in
person). Icelandic has various types of non-nom. subjects (cf. Yip,
Maling & Jackendoff in Language 1987, Zaenen, Maling & Thr'ainsson in
NLLT 1985, Sigurdhsson in NLLT 1991) but it is basically only the ones
that take dat. subjects that have nom. objects. In such cases the
finite verb agrees w. the nom. object in number:

(1) M'er hafa alltaf l'ikadh thessar baekur
 me(D) have(3pl) always liked these books(Nplf)

(2) Honum hafa aldrei leidhst m'alv'isindi
 him(Dsg) have(3pl) never bored linguistics(Npln)


(Hoskuldur Thrainsson)

Your query doesn't seem to exclude languages in which some verbs agree
with the subject and others with the object. In Sora, a Munda language
of Koraput and Ganjam Districts, Orissa, India, there are transitive
verbs that can agree with either, e.g.

 amAn gij-t-ay
 you see-pres-1subj
 I see you

 nyen gij-t-Am
 I see-pres-2obj
 I see you

and there many verbs that take no subject and agree only with the

 be ill-pres-2obj
 You are ill (lit. ills you)

But intransitive verbs, marked with an -en, do agree with their subjects:

 nyen dimAd-t-en-ay
 I sleep-pres-intr-1subj
 I'm sleeping

Nominal objects can be incorporated, and then they become intransitive,
and agree with their subjects:

 nyen gAd-bOng-t-en-ay
 I cut-buffalo-pres-intr-1subj
 I'm sacrificing a buffalo

though, derivationally, these can also be used transitively:

 nyen BudO-n gAd-bOng-t-ay
 I BudO-art cut-buffalo-pres-intr-1subj
 I'm buffalo-cutting BudO, i.e. beheading him

However I don't think that anyone has analyzed Sora as ergative. If
this is anything like what you're looking for, you may want to ask my
colleague Stan Starosta about that. I do not think he subscribes to
LINGUIST, but you could email him at
He also knows Sora, and wrote his thesis on it in the 60s, and knows a
lot about ergativity, and he's a syntactician, unlike me.

It occurs to me though that in the 2nd "I see you" example, there is
no way to form this with 3rd person objects, but only with me/thee and
us/you. Therefore it might be claimed that this is incorporation of a
pronominal object rather than agreement. Although it is put after the
tense marker, not before as an incorporated noun would be, this
interpretation is supported by the fact that this construction doesn't
allow a syntactic espression of the object pronoun: *nyen amAn
gij-t-Am `I see you'.

David Stampe <>, <stampeuhunix.bitnet>
Dept. of Linguistics, Univ. of Hawaii/Manoa, Honolulu HI 96822

My database (cf. Dryer 1992 in Language) lists the following languages as
having pronominal affixes on verbs for objects but not for subjects. Not
all of these are agreement affixes (since some do not normally co-occur
with noun objects) and which of them count as agreement partly depends on
one's theoretical assumptions. Some of these have subject agreement, not
on the verb, but phonologically separate either adjacent to the verb or
elsewhere in the clause. The names in caps are the names of the genetic
groups (genera), with the name of languages in parentheses.

IJOID (Kolokuma Ijo)
GUR (Bimoba)
NAX (Ingush)
MUNDA (Mundari)
NICOBARESE (Car Nicobarese)
CENTRAL-EASTERN MALAYO-POLYNESIAN (Kiribatese, Ponapean, Woleaian)
PIMIC (Papago)
WARAO (Warao)

Please contact me if you want references.

Matthew Dryer


Maybe you know that French may present such an agreement in specific
cases; i.e. when the direct object is placed (for any reason) in front
of the verb when the tense is the "passe compose". La voiture ? Je
l'ai donne'e a` Pierre. The car ? I gave it to Peter; La voiture que
j'ai donne'e a`Pierre e'tait rouge. The car I gave to Peter was red.
But J'ai donne' la voiture a`Pierre. I gave the car to Peter.


Laurent Romary
Batiment Loria
B.P. 239, 54506 Vandoeuvre Les Nancy
tel : 83 59 20 37
fax : 83 41 30 79



American Sign Language has several verbs that agree with the direct object
but not the subject. ASL also has a large number of verbs that agree with
both the subject and the object, but I can't think of any verbs that agree
only with the subject. ASL is not an ergative language. Hope this helps,
and let me know if you need references.

Karen Emmorey
Laboratory for Cognitive Neuroscience
The Salk Institute


Again, thanks to all who replied.
Mail to author|Respond to list|Read more issues|LINGUIST home page|Top of issue