we started the topic ergative languages. I am very confused how I could determine whether a language is an accusative or an ergative language. Is there some sort of analysis I could use to get evidence if a language is ergative or accusative?

  • 4
    Are you sure you understood what "ergative" and "accusative" means? Because their definition is the very way to determine whether a language is accusative or ergative. Is it the definition you don't understand or something else (if so, what exactly)? Commented Jan 3, 2017 at 14:45
  • I guess the OP has mixed the terminology, and they meant the contrast between the Ergative-absolutive languages versus the Nominative-accusative ones. Commented Jan 3, 2017 at 16:10
  • 4
    @bytebuster "ergative" and "accusative" are just frequently used shorthand for "ergative-absolutive" and "nominative-accusative" respectively, so I don't think that's the problem. Commented Jan 3, 2017 at 16:41

2 Answers 2


Simplifying things a bit:

  1. Take a sample of simple sentences where the verb is transitive, i. e. you can recognize a subject and an object. Note how the language marks which is the subject and which is the object (with case endings, with relative position, by person-number agreement with the verb, etc.).
  2. Now take a sample of sentences where the verb is clearly intransitive, i. e. it only has a subject. Again note how this subject is marked.
  3. Compare the marking.


  • If the subject of the transitive verb is consistently marked in the same way as the subject of the intransitive verb, and the "odd one out" is the transitive object, then the language is nominative-accusative.
  • If the object of the transitive verb is marked in the same way as the intransitive subject, then the language is ergative-absolutive.

In reality this is not nearly as neat, as there might be different alignments according to which tense-aspect-mood is in play (Georgian, Hindi) or there might be some intransitive verbs that take ergative subjects (Basque).


The first step is finding some sort of case marking. In this language, do "I like him" and "he likes me" use the same words, just in a different order? Or are any of the words themselves changed, like in English (I → me, him → he)?

Once you've identified case marking, note what indicates the subject, and what indicates the object. In the English example, "he" is subject-only, and "him" is object-only.

Now look at an intransitive sentence, like "he left". When there's only one noun, is it marked like a subject, or like an object?

If it's marked like a subject, you have a nominative-accusative language.
If it's marked like an object, you have an ergative-absolutive language.

In English, "he" is the subject form. So English is nominative-accusative.

Caveat: most languages with an ergative-absolutive distinction use it only in particular circumstances. For instance, Urdu is ergative-absolutive in some tenses, and nominative-accusative in others. So you may need to try several examples.

  • Note that even English John loves Mary has a "case marking" in this sense. It is the position of the argument towards the verb that defines the case. You can imagine ergative English as having sentences like: John walks and House builds Mary. In the latter example, the semantic patient of the transitive verb ("house") is marked the same way (preverbal position) as subject in the intransitive construction, thus it is technically in the absolutive position/case, while the semantic agent ("Mary") is in ergative position/case.
    – Eleshar
    Commented Jan 3, 2017 at 21:01
  • @Eleshar Definitely true. I didn't mention "syntactic ergativity" here since I've been told it's never been observed without "morphological ergativity" being present as well.
    – Draconis
    Commented Jan 3, 2017 at 21:04

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.