Definition of "ergative"

The most common situation is for nouns to inflect according to a nominative-ergative pattern, while pronouns at least superficially follow a nominative-accusative pattern. That is, nouns have a single case (nominative) marking intransitive subject and transitive object functions, and another case (ergative) for transitive subject function.

1972, R[obert] M[alcolm] W[ard] Dixon, “Australian Languages”, in The Dyirbal Language of North Queensland (Cambridge Studies in Linguistics; 9), Cambridge, Cambridgeshire: University Press, page 4

For Podopa, the actor of most transitive verbs (and of some intransitive verbs as well) may occur with or without the ergative case suffix, but with a semantic difference. The ergative suffix indicates that the actor is acting independently, is self-motivated, and exerts his personal control over the situation; while its lack indicates that the actor is performing according to his set social obligations, not according to his own independent will, and does not assert his personal control over the situation.

1986, William A[uguste] Foley, “Nominals”, in The Papuan Languages of New Guinea (Cambridge Language Surveys), Cambridge, Cambridgeshire, New York, N.Y.: Cambridge University Press, published 1999, page 108

The ergative case marks the agent of a transitive verb. The ergative suffix is -le/-re/-lle/-ʔille. The form of the ergative suffix is /-le/ for the indefinite and /-ʔille/ for the definite after the consonants /ʔ/, /k/, /t/, /p/, /b/, /ŋ/, /n/ and /m/.

1987, George van Driem, “Nominal Morphology”, in A Grammar of Limbu (Mouton Grammar Library; 4), Berlin, New York, N.Y.: Mouton de Gruyter, page 39

In Section 1 I will discuss the existence of a class of ergative adjectives in Dutch […]. It will be demonstrated that there are a number of arguments supporting the claim that the class of adjectives should be divided into ergative and unergative adjectives. A large number of adjectives that are unergative according to the tests provided in Section 2 appear to be ergative with respect to their argument structure.

2000, Hans Bennis, “Adjectives and Argument Structure”, in Peter Coopmans, Martin Everaert, Jane Grimshaw, editors, Lexical Specification and Insertion (Current Issues in Linguistic Theory; 197), Amsterdam, Philadelphia, Pa.: John Benjamins Publishing Company, pages 27–28

Another difference between C. Barwar and Kurdish is the fact that in C. Barwar the compound perfect construction is not ergative. […] In Kurdish, on the other hand, the corresponding compound construction, which appears to have been the model for the NENA [North-Eastern Neo-Aramaic] construction, is ergative in form when the verb is transitive. The loss of the ergative inflection in C. Barwar and most other NENA dialects is again a development internal to NENA. The original ergative type of construction has survived only in a few Jewish dialects on the eastern periphery.

2008, Geoffrey Khan, “Introduction”, in HdO: The Neo-Aramaic Dialect of Barwar (Handbook of Oriental Studies; Section 1 (The Near and Middle East); 96), volume 1 (Grammar), Leiden, Boston, Mass.: Brill, page 22

Samoan, for example, differs from the usual pattern displayed by split ergative languages in that the appearance of the ergative is grounded in sociolinguistic factors as well as syntactic ones. The more formal register of Samoan requires the ergative on all postverbal transitive subjects. The less formal register allows the ergative not to be expressed at all.

2006, Miriam Butt, “The Ergative Dragon”, in Theories of Case (Cambridge Textbooks in Linguistics), Cambridge, Cambridgeshire: Cambridge University Press, page 178

Unlike those with subjectivized ergatives, such locative clauses naturally do not allow for imperatives (*Contain the apples).

1971, John M[athieson] Anderson, “Locative”, in The Grammar of Case: Towards a Localistic Theory (Cambridge Studies in Linguistics; 4), Cambridge, Cambridgeshire: Cambridge University Press, published 1976, part III (Locative and Ablative), page 90

Woodbury (1975) does argue, however, that absolutives are more relativisable in Greenlandic than are ergatives, on the grounds that (1) RCs [Relative Clauses] formed on ergatives are somewhat more restricted in the distribution in matrix clauses (p. 21) than are those formed on absolutives, and (2) for certain verb classes ergatives cannot be relativised out of the active participle (p. 27).

1987, Edward L. Keenan, Bernard Comrie, “Noun Phrase Accessibility and Universal Grammar”, in Edward L. Keenan, Universal Grammar: 15 Essays (Croom Helm Linguistics Series), Beckenham, Kent, North Ryde, N.S.W.: Croom Helm, part 1 (Cross Language Variation), page 26

Ergatives share close similarities with agentless passives: Both are intransitive, both lack an agent, while the patient appears in the subject position. As the acquisition data show, learners seem to treat ergatives like passives.

1994, Virginia Yip, “Grammatical Consciousness-raising and Learnability”, in Terence Odlin, editor, Perspectives on Pedagogical Grammar (Cambridge Applied Linguistics), Cambridge, Cambridgeshire: Cambridge University Press, page 128

Combining two ergatives in one clause is not always ungrammatical in Agul; but one of the ergatives must be used in a non-agentive function, e.g. instrumental or temporal.

2012, Michael A. Daniel, Timur A. Maisak, Solmaz R. Merdanova, “Causatives in Agul”, in Pirkko Suihkonen, Bernard Comrie, Valery D. Solovyev, editors, Argument Structure and Grammatical Relations: A Crosslinguistic Typology (Studies in Language Companion Series; 126), Amsterdam, Philadelphia, Pa.: John Benjamins Publishing Company, page 66