A rebuttal of the notion of a trigger language

Not meaning to rock the boat, but trigger systems, as they're explained here, don't actually exist in natural languages. I think they only exist as conlangs, actually (I have one too). The trigger systems of Austronesia don't actually seem to be anything more than languages with multiple passive formation and applicativization strategies. The "trigger" isn't actually unmarked, it's simply in the case that the subject of an intransitive verb is put into. The morphology doesn't mark the role of the verb, per se, but merely marks what role the new subject played in the "underived" sentence. It would be something like the following:

English:

• Direct Object Passive: I ate a hamburger. -> A hamburger eat-PASS1 by me.
• Indirect Object Passive: I gave you a flower. -> You give-PASS2 a flower by me.
• Prepositional Object Passive: I walked into a store. -> A store walk-PASS3 by me. (Prepositional information lost.)

As it so happens, the form of the passive is the same for all three in English. They could very well be different, to give the hearer more information about the role of the subject (since its case is invariant). If you add in applicatives, which English doesn't have, you have a wealth of verbal morphology that tells what the role of the subject is.

So, the "trigger" actually is the syntactic subject--just the way the raised patient of a passive is the subject of the sentence. And these languages do have passive morphology--extensive passive morphology. The notion of the "trigger" language, then, is something exclusive to conlangs.

For example, one could create a language like the following:

• Verb: maka "eat"
• Subject Trigger: makana
• Object Trigger: makasi
• Indirect Object Trigger: makalo
• Genitive Trigger: makava
• Adessive Trigger: makawe
• Allative Trigger: makatu
• Abessive Trigger: makaje
• Ablative Trigger: makazo
• Inessive Trigger: makapi
• Illative Trigger: makaha

Etc.

In other words, a language with a whole bunch of cases that are simply marked on the verb, and in order to use one of these cases, the case must be used with the subject of the verb. This kind of language doesn't exist in the wild, but, given the idea of "trigger languages", there's no reason why it shouldn't.

The point of this post is the following:

1. To suggest that trigger languages, as they're described, do not exist naturally.
2. To suggest that such trigger languages exist as conlangs only.
3. To suggest that this should be noted on this page.

I don't mean to suggest that the trigger conlang is a bad thing (as I said, I have one myself), only that it isn't necessarily a representation of something that's naturally occurring.

-David J. Peterson —Preceding undated comment added 18:36, 25 September 2006‎.

Authorship removed

I removed the signature and the introductory text ("Hello, I'm here to...") because wikibooks pages belong to everyone, not just the original contributor. Since wiki pages are meant to be modified, it won't be too long before the words aren't your own anyway. :) - Furrykef 11:20, 14 April 2007 (UTC)

Revamping

I hope to clearly explain trigger languages and Austronesian alignment and redistribute this information across Alignment and Voice. See Alignment#Rethink. --Pi zero (discuss • contribs) 14:04, 21 December 2015 (UTC)