Linguistic commentary from a guy who takes things too literally

Anti-Passive! It’s Like a Passive for Ergative Languages!

Posted by Neal on March 4, 2013

This ain't no pasta!

Over the weekend, I speculated on how English might work as an ergative language. Today, on National Grammar Day, I’m taking it a step further into the reversed grammar of ergative languages, to show what might happen if you tried to use the passive voice in ergative English. What would that even look like, when our ergative English already has its transitive verbs agreeing with their patients?

In you’ve forgotten what that looks like, here’s the suite of sample sentences we arrived at in our morphologically and syntactically ergative version of English. The pronouns in red are in the ergative case; they denote agents who do things to others. The pronouns in blue are in the absolutive case; they denote either “subjects” (in this context, someone or something that performs an action which doesn’t directly affect someone or something else), or patients who are affected by an action.

    1. She kiss me.
    2. I kisses her.
    1. Her smiles.
    2. Me smile.

Before I go further, I need to comment on the vocabulary. First of all, the terms agent and patient have to be understood as referring not only to very obviously agentlike roles such as “hitter,” “writer,” and “creator,” and very patientlike roles such as “struck,” “written,” and “created,” but also to pairs such as “seer” and “seen,” or “one who loves” and “one who is loved.” This is true for ordinary English as well as our imaginary ergative variety.

Second, we saw earlier that the term subject has a specialized meaning when we’re talking about whether a language is ergative or not (or more concisely, its morphosyntactic alignment). It’s not just any subject; it’s the subject of an intransitive verb. So what term do we use when we want to talk about the subject of a transitive verb? In the last post, we called them agents, in keeping with the subject-agent-patient terminology of morphosyntactic alignment. But now, the agents aren’t going to be the … subjects? … anymore. In Language Universals and Linguistic Typology, Bernard Comrie’s attitude is “Tough noogies, welcome to the real world!” That is, it’s no simple matter crosslinguistically to say what’s a subject. Is it the thing that the verb agrees with? Maybe, but in some languages, transitive verbs agree with both their agents and their patients. So before reading further, say goodbye to your old notions of what’s a subject. The closest we’ll come is when we note which noun(s) a verb is agreeing with.

So now, onward to passives and antipassives. Ordinary English, as well as many other so-called “nominative-accusative languages” gives you two options for expressing a transitive verb. There’s the more straightforward option of the active voice, with agent as the noun that the verb agrees with (She kisses me; I kiss her); and the more complex passive voice, in which the patient is promoted to the place where verb agreement goes on, and the agent disappears or is expressed in a by phrase (I am kissed (by her); she is kissed (by me)).

In ergative English, the more straightforward option has the patient as the noun that the verb agrees with: She kiss me; I kisses her. The more complex option, analogous to the passive in ordinary English, is the antipassive. To show antipassive, I’ll use the passive morphology from ordinary English, to give you the maximum effect of how things are turned around here. In the antipassive, the agent gets promoted to agree with the noun, and the patient gets moved to the background:

  1. Me am kissed (by her). [Think of it as “I do some kissing (on her).”]
  2. Her is kissed (by me). [Think of it as “She does some kissing (on me).”]

That’s crazy!

Now actually, ergative English probably wouldn’t use an antipassive to express these thoughts. Just as ordinary English tends to use passive in situations where the agent is unknown, unimportant, or just less important than the patient; ergative languages tend to use antipassive when the patient is unknown, unimportant, or less important than the agent. (This is according to Ann Cooreman in “A Functional Typology of Antipassives”, in Voice: Form and Function, 1994, edited by Barbara Fox and Paul J. Hopper.) Functionally, they’re like detransitivized English verbs, such as eat, teach, write, etc. When you say “I’m eating,” or “I’ve taught for years,” or “Write every day,” the patient is assumed to be something obvious: food, courses or students, stuff you write.

Thinking about passives and antipassives, I thought about the disapproval of passives that speakers use to avoid placing (or accepting) blame. “Don’t say ‘Mistakes were made’! Admit your responsibility and just say, I made a mistake’!” If English were an ergative language, would English teachers tell their students, “Don’t say ‘Me am eaten!’ Take responsibility for your actions and say what you ate! Don’t try to gloss over it!”

Well, no, they wouldn’t, because all that was phrased in ordinary English. They’d say … let’s see … “Let no one say ‘Me am eaten’…” Ah, forget it! It was tricky enough to get my ergative examples straight as it was. Right now I am absolutely ergatived out!


6 Responses to “Anti-Passive! It’s Like a Passive for Ergative Languages!”

  1. the ridger said

    So …. is the so-called middle sort of ergative? The book read quickly, the bread cuts easily, she doesn’t frighten easily (as opposed to She doesn’t frighten me)? How about, what’s it called, locative focus (the garden is swarming with bees)?

    • Neal said

      No, these wouldn’t be ergative. Unfortunately, the term “ergative” has been used, without any good reason, for verbs that are just the opposite of ergative, such as intransitive “boil”. If English were an ergative language, the subjects of these verbs would be in the absolutive case, not the ergative.

      I prefer the term “unaccusative” for these verbs whose subjects fill a patient role: Their subject is something that you would expect to be accusative. Likewise, I like the term “unergative” for intransitive verbs whose subject has an agent-like role, but doesn’t actually affect another party. Prototypical unergative verbs include “walk” and “dance”.

  2. Ran said

    That is a really good explanation of antipassives. I’d understood them in theory before (passive:patient:nominative::antipassive:agent:absolutive), but now I have a sort of intuitive grasp of how they’re actually used. Maybe. For the next several minutes, at least. 😛

  3. EP said

    That’s a good point about passives and how they are used to avoid placing or accepting blame. I used to work for the military, for instance, and many a document was formulated in that form. It didn’t necessarily have to do with blame, but this passive-speak was the preferred mode of writing, militaryese, so-to-speak. Maybe it still is.

    • The Ridger said

      ARGH. I hate that accusation on the passive. You can avoid blame very neatly with the active (the bus blew up, someone lost the files), if that’s your goal, and the passive can make the blame even bigger and better (the doctor murdered his patient vs the patient was murdered by his own doctor!). Just once I’d like to read someone attack *avoiding responsibility* and not the passive.

      • Neal said

        Great examples of how passive voice is neither necessary nor sufficient for blame avoidance. However, it is *one* technique that makes it *easier* to avoid naming the agent.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: