The Unseen wrote:
It sounds like you simply have an inflectional morpheme that marks an animate as a patient and an inanimate as an agent. So, it seems accusative. With an interesting twist, but still accusative.
That said, intransitives have some sort of split-s (aka Active-Stative) system going on.
I guess that's correct, but if intransitives are active-stative, isn't the entire language called active-stative? We may just be bantering around with terminology here, but is there a substantive difference that I'm missing? If it means anything, I'm not expecting that I've come upon a new type of alignment or anything - I'm pretty sure it's overall active-stative. It's just the case marking is different from anything I've heard of.
I really don't know. I was thinking that the active-stative distinction wouldn't carry over into the transitive (where it would be perceived as role reversal, not merely a subtext of meaning), but I've realized it's up to you. Does the marker not mark normally in transitive sentences when the alleged subject is more like a patient (or less like an agent, at the least)? Or does it mark syntactically like the patient?
Basically, what happens when you make an originally intransitive sentence transitive? There are a variety of responses, though, so I'm not sure what it would all mean...
There's an idea, have a language mark syntactically like an accusative language, but then take ergative or active-stative or phillipine inflectional morphemes. That's also called insanity.
I haven't really thought it through, but at first blush I would say that the marker depends all on the animacy, and even if the subject is patient-like it would probably be treated as an agent. Actually, typing that out makes me think that that might not be the case for a few verbs (like the verb "undergo" in English), but perhaps I could make the system perfectly regular and make the subject of a transitive verb have to be an agent.
I'm guessing that when an intransitive sentence becomes transitive, the original S moves into the role that it equals (either A or P) and a new noun is introduced for the other role. I've actually thought of one example like this before:
"The teeth are white"
Kó raχ llaeq
man white tooth-PL
"The man is whitening the teeth"
The verb raχ
is a stative verb when used intransitively, as its S is treated like P. However, when it used transitively, the S actually becomes P, and a new A is introduced. The P doesn't gain the inverse marker here because it is staying in its inherent inanimate role of P. If I use an animate noun here, it would take the suffix in both cases, just as the inanimate doesn't take the suffix in both cases, because of role fulfillment. Ex:
"The man is white"
Kó raχ kóll
man white man-INV
"The man is whitening the man"
In these situations, intransitive becoming transitive is a causative construction. I don't have a causative morpheme here - the verb is contextually understood to be transitive, and I may think of a causative morpheme in the future, but for now I'm not using one.
In English, an intransitive becoming transitive via causation moves the S to P and introduces a new A. Ex: "I went" and "God made me go." In this conlang..............it's the same thing. I was wrong when I said the role for S would be kept the same as before. Because if I had an active intransitive verb like "to go," and I made it causative, the S would move to P anyways. Ex:
Tloso phosá nahetl
God go-PERF I-INV
"God made me go"
I don't think this is a sympton of accusativity, though. According to this paper (http://assets.cambridge.org/052166/0394/sample/0521660394wsc00.pdf
, pg. 21), it's "rather rare" for any language to use the original S as A in causation, and the example they give the new A is actually in a peripheral argument.
So I guess that answer was a bit long, but it terms of causation intransitive > transitive is fairly basic. The only other instance of intransitive > transitive I can think of is an ambitransitive. In this situation, S can equal P or A when the verb is optionally used without an object. In the same paper listed above, there's an example of an active-stative lang that has ambitransitivity go both ways depending on the verb (S equaling A or P). S = A when the P "is irrelevant or unimportant in the transitive," and S = O (they use O for P) when "describing a spontaneous process or event in the intransitive." I don't have examples for my conlang yet, and I may or may not use those types of ambitransitivity, but there's an example to go off of.