Infinitives in Konehian: internally verbs, externally nouns.
Hello folks. Today I put some effort into learning syntax trees*, because I got stuck with my grammar*, although I haven't gone through the more recent theories yet, so the way I illustrated these trees might be wrong or even outdated in fashion. I provided some glosses to help you out with the morphology. Now you might ask, why this specific look? I think my conlang is meant to be an a priori, but inspired from Romanian, Ancient Greek and with a funny twist, some active-stative alignment.
To be more specific, we are talking about an agentive-default fluid-s typology. I believe this means that the agentive case is the default unmarked one, whereas the patientive case is marked. In most cases, Konehian behaves like a Nominative-Accusative language I'd say. Verbs may have the active voice, the passive voice. But also, an antipassive, or at least, that is my goal. I would say that the S argument, the subject of an intransitive clause, can be either agentive or patientive, depending on volition (Sa and Sp). If A is the subject of the transitive clause, also known as an agent, and if P is the object of the transitive clause, the patient, I might just say that Sa = A and Sp = P, when it comes to case morphology. For these reasons, I dropped entirely labels such as NOM, ACC, ERG, ABS and just went with A (agentive case) and P (patientive case).
I believe the antipassive voice is used in Konehian to express a lack of volition, for a transitive clause. I would think that:
1SG.A pour-PRS.ACT.1SG water-DEF-P
"I pour the water."
would turn into:
1SG.P pour-PRET.ANTIP.1SG water-INS
"I spilled with water."
I am not sure 100% if this is how an antipassive may work, but that is what I concluded from my discussions with fellow conlangers. Word order is indeed SVO mostly. Konehian is not afraid of obliques. You can see what happens to "water" above. Also, the active voice is yet again, unmarked I'd say.
Now onto infinitives. I am not sure yet how bare infinitives work in Konehian, but "articular" infinitives are surely funky. They introduce an infinitive clause. In the main clause, these may function as nouns, they can be agents, patients, maybe even oblique. But within their own clause, they function as verbs. They decline for tense, which is really just aspect, voice but not number nor person! How are articular infinitives made? You just add the DEF.ABST article? -or, after the stem. Konehian definiteness works like in Romanian most likely, you do not have a separate determiner or article that precedes the noun, you just glue it to the end of the stem. Metasti (people) becomes metastia (the people). Mozu (zombie) becomes mozur (the zombie). Ruz (war) becomes ruzor (the war). Ditas (station) becomes ditaste (the station). You get it. Different articles, I suppose depending on the gender. Konehian has three genders: abstractive, animate and inanimate. You can see some of these featured in the given glosses. So I suppose that the articular infinitive uses an abstractive definiteness marker. Much like other abstract nouns. This makes sense, after all, if ubrèq means "to liberate", ubrèqor means "the deed of liberating", this deed being an abstract concept.
Funny thing, infinitive clauses, whose verb is the articular infinitive itself, accept both a non-overt subject and an object. The case marker on the articular infinitive is actually the case function of the NP of the infinitive clause, as governed by the matrix verb. In gloss 2.1.1, you can see that Ubrèqor Eyzenlomon functions as the NP, and it is the subject of the main clause. But in 2.1.2 example, ubrèqoron Eyzenlomon is the object of the main clause, as such, the infinitive shows that. In both instances, Eyzenlom is the direct object within the infinitive clause. Hence the -on ending.
In 2.2.1 example, I tested how a reflexive aorist agentive articular infinitive may work (weird naming convention, I know). By reflexive, I think I mostly mean the middle voice, although there is also a mediopassive voice when I tried to use the reflexive as an intransitive (but this happened in indicative verbs). See Romanian "Se moare mult." as such an example. The verb is reflexive and it accepts an adverb, but otherwise it has no object I'd say. Mediopassive in Konehian also happens only for 3SG. Anyways, since the subject is implied in infinitive clauses, and since the reflexive voice implies that the agent is also the patient of the clause, I tend to believe that the object of the clause is simply dropped, as it is equated to the implied subject. The focus is not on Thirinda (see the gloss for 2.2.1.) but on the deed of self-liberation itself. This also might imply that Thirindae (genitive of Thirinda) is not part of the infinitive clause. You could switch the articular reflexive aorist infinitive ubrèdàtor for something like english "cat". You can totally say "Thirinda's cat encouraged people further.", according to Konehian grammar, which could further be simplified to "The cat encouraged people further."
I would say that the aorist form of the infinitive is the least marked of all forms. I am really curious to see how this will work for the antipassive voice. An antipassive infinitive? That sounds fun. I could imagine the object of the infinitive clause to be put into an oblique case, or perhaps entirely dropped, but not so sure what happens to the implied subject. Overall, I am not sure if my reasoning for all of this is proper, nor how naturalistic any of this may seem, but I tried my best, it isn't perfect and there is plenty of work to do, but I was wondering what you think about this, overall. If you got some advice, it is welcomed.