Typically it is in the transitive sentence in the case ergative , while in the intransitive sentence it has the case absolute . In an accusative language like German, on the other hand, the same case is used in both cases ( nominative ; example: the man walks - the man sees the dog). Consistent ergativity is rare (example: Basque ); more common is split ergativity . Ergativity raises questions about the generality of the term “subject”; However, on closer analysis, the languages concerned are grammatically inconsistent, so that effects of a category “subject” can be found to varying degrees in such languages.
The term ergative language is ambiguous. In a broader sense, it describes any language that knows the phenomenon of ergativity. These include Greenlandic and other Eskimo languages , Basque , Georgian , Sumerian , Zazaese , Kurmanji , Pashto , Burushaski , Hindi / Urdu , Tibetan and Dyirbal (an Australian language ).
In the narrower sense of the relational language typology, he only means some of these languages that are not assigned to any other type. In WALS, Greenlandic and Burushaski are classified as ergative languages, Basque and Georgian as active languages and Hindi as three-part ( ergative-accusative language ); More below .
Ergative languages use the same grammatical case for the subject of an intransitive verb and the patient of a transitive verb, which is called the absolute and corresponds to the nominative in accusative languages, as it usually remains unmarked. Another case is used for the agent , i.e. the acting subject, of transitive verbs, namely the ergative.
An example of ergativity in Basque :
- Emakume-ak gizon-a ikusi you.
- Woman def. erg .sg man- def.abs.sg see- prf.3sg
- "The woman saw the man."
One could also illustrate this system by inventing a variant of German in which there is an ending -u for the absolute on a noun and an ending -o for the ergative. Sentences in such "Ergative German" would look like this:
- Child-u fell down
- Mrs.-o saw child-u.
Ergative scheme in relation to semantic roles
|bi valent - transitive||Ergative||Absolutely|
|monovalent ( intransitive )||Absolutely||(Absolute)|
Ergative construction and nominative accusative construction in comparison:
|Nominative accusative scheme||Nominative||accusative||Nominative|
Some languages that have morphological ergativity also have syntactic ergativity.
In ergative languages without syntactic ergativity - just as in accusative languages - the subject is always the agent of transitive verbs and the only argument of intransitive verbs. In languages with syntactic ergativity, on the other hand, the “subject” is the argument that is in the absolute, that is, the patient's transitive verb and the only argument of intransitive verbs. Most of the time, however, syntactic ergativity only occurs in some constructions, in the others the language behaves accusatively.
Syntactic ergativity becomes visible, for example, in the sentence combination of the sub-clauses: Ergative languages infer a missing argument in the absolute. In German we interpret the sentence “The student saw the teacher and went away” as “The student saw the teacher and the student went away”. In contrast, in ergative languages, the absolute argument of the first part of the sentence would be accepted as the subject of the second, intransitive verb. "The student (ERG) saw the teacher (ABS) and went away" is interpreted as "The student (ERG) saw the teacher (ABS) and the teacher (ABS) went away". A sentence of this kind can be used as a test by native speakers of a particular language to find out whether the language in question is syntactically an ergative or an accusative language.
Ergative languages generally have no passive forms . But they have an anti-passive , whereby the direct object is deleted and the subject changes from the ergative to the absolute, that is, the verb becomes intransitive .
Split ergativity occurs in languages that use another alignment pattern in addition to the ergative pattern .
This can either depend on the verb form, such as in Georgian or Zazaic (here certain tenses require ergative, other accusative marking), or it depends on the arguments of the verb as in Dyirbal , where pronouns show an accusative and nouns show an ergative ( see also animation hierarchy ), or in Inuktitut and other Eskimo languages , where a distinction is made according to definiteness . Even the Sumerian is an example of split ergativity.
- Example Georgian:
- Example Inuktitut:
- ᐊᖑᑦ ᓇᓄᕐᒥᒃ ᑕᑯᔪᖅ
- angut nanur-mik taku-juq
- Man [ abs . sg ] polar bear into . sg see- 3 sg
- 'The man sees a polar bear'
Some Indo-Iranian languages such as Pashto (in Afghanistan ), Hindi / Urdu , Kurmanji- Kurdish also use accusative constructions in the present group, whereas ergative constructions are used in the perfect group. This is due to the fact that the past participle has a passive meaning in transitive verbs, and of course active in intransitive verbs. We know that from German: “seen” is understood passively, as opposed to “gone”. This is how the perfect tense is formed in Hindi and in German as “Anita has gone” ( Anītā gaī hai , literally Anita has gone ). If one were to say “Anita is seen”, this would be understood passively in both languages (Hindi: Anītā dekhī hai ). That is why one says in German "Anita has seen a house". Hindi does not use a construction with any other auxiliary verb, but instead “ a house is seen through Anita”, in Hindi: Anītā ne ghar dekh ā hai . The ā of dekhā is a masculine ending because dekhā (seen) refers to ghar (house), not anita.
But if the patient (object) has been determined, it turns out that Hindi and Urdu also use an accusative (the verb ending -ā is then impersonal):
- Miriam Butt : Theories of Case. Cambridge University Press, 2006
- RMW Dixon : Ergativity. Cambridge University Press, Cambridge 1994, ISBN 0-521-44446-2 .
- A derivation of the Greek ἔργον 'Tat, Werk' is often assumed, but the first mention of the expression is related to a place-case. See Butt (2006), chap. 6th
- Marcus Kracht: Languages of the World . (PDF) Bielefeld University, lecture from the winter semester 2009/10, January 27, 2011, p. 38
- WALS : Chapter 98: Alignment of case marking of full noun phrases , Example (6)