If you are not a linguist, a conlanger, or some other form of dedicated linguistic dilettante, you’ve probably never heard of ergativity. (If you are a conlanger, you may be sick of hearing about it, since it’s a perennial subject on the CONLANG list.) Ergativity is one of the weirdest linguistic phenomena that I know of, and because I’m that kind of nerd I’m going to share it with you.
Let’s take a simple transitive sentence:
The duck(A) bit him(P).
If you remember your grammar at all, you know that the duck is the subject of #1, while the object is him. In English, we have the bare remnants of a case system for our pronouns, which requires us to use the form He for the subject and him for the object. You’ll notice that I’ve marked these A and P, which stand for Agent and Patient. My reason for using these terms will become clear below.
Now let’s add an intransitive sentence:
Here the "subject" is he. I’ve marked He as S for Subject here.
In English, as in most languages spoken in Europe, the Agent of a transitive verb and the Subject of an intransitive verb take the same form. In fact, we are usually just call them both "subjects" and forget about it. In more formal terminology, the combined A and S case is the nominative, while the P case is the accusative.
But this is not necessarily the case in all languages. There are languages in which the Subject and Patient take the same form, and the Agent gets left out. This is called ergativity.
If English were an ergative language, we might express the sentences like this:
The duck(A) bit him(P).
Note that our A, P, and S labels haven’t changed. But now the S and the P forms are the same, while the A form is different. In such a language, we refer to the combined S and P form as the absolutive case, and the A form as the ergative case.
If, by chance, you know a language with a more robust case system than English, it may help to think about how this would work in those languages. Take your Russian, German, or Latin and recast it with the subjects of all of the intransitive verbs in the accusative case. This is how an ergative language works. The only difference is that in a nominative/accusative language we think of the nominative form as the base form and the accusative form as the marked form, while in an ergative/absolutive language the absolutive form is unmarked while the ergative form is marked.
There’s more to it than that, and as usual Wikipedia has a decent overview, though I prefer the one with Cthulhu. Some ergative languages you may have heard of include Basque, Hindi, Georgian, and Tibetan.