Thursday, August 23, 2012

Voice and applicatives in Hra'anh

I'm really having fun using applicatives in Hra'anh. It currently uses three, though two of them work more like oblique particles.

First off, I need to start by explaining myself. A lot. Let's start with some terminology. First off, you should have learned in school the difference between a transitive verb and an intransitive verb. If you didn't, here's the difference. "I ran" is intransitive; it has only one argument, called the subject. "I ate the sandwich" is transitive because it has two arguments, the subject (I), and the object (the sandwich).

Let's go now to a little thing called voice. This is a subject that is near and dear to my heart. A lot of that is probably because voice is the underdog of English class. Every teacher I've ever had has said, "Don't use the passive voice. It is weak and useless." This, of course, is hogwash. English has two voices, though I would argue that certain constructions have what's called the middle voice (e.g. "How do you feel?" and "I feel fine.") But that's a different post for a different time. We use the active voice ("the horse kicked the man") and passive voice ("the man was kicked by the horse"). Hra'anh adds the antipassive voice, which English doesn't really have. I'll explain why voice is important later.

Now let's cover what an applicative is. Some linguists argue that the applicative is in fact a voice. However, I would be inclined to agree with conlanger William Annis in that applicatives are really valence-changing operators. Valency is a quality of a verb that indicates how many arguments it takes. In English, some take zero arguments ("it rains"); some take one ("rain fell"); some take two ("rain hit John"); some take three ("I threw John the football"). In some engineered languages, verbs can take five, possibly more (!) arguments.


 The obliques

I'm not talking about those dastardly oblique abdominal muscles that I can never seem to work. I'm talking about the applicatives ke [kʼɛ] and doke [do.kʼɛ]. These are the passive and antipassive oblique applicatives respectively.


How voice works in English

Why do we have voice? It's all about keeping the most important (salient) things in the foreground of the discourse. Language is biased toward human beings. That's why we have constructions like "I was hit by a car." The whole "by a car" thing is what's called an oblique argument. It can be taken out of the sentence, and it still makes perfect sense. "I was hit." The car, which is the agent, isn't as important as the fact that I was hit. In fact, if you use the active-voice version, "a car hit me," it almost seems ungrammatical. 


How voice works in Hra'anh

Voice is ever more important in Hra'anh because Hra'anh is unlike English. Verbs in English are happy to switch valency, e.g. "I ate" vs. "I ate the sandwich." Verbs in Hra'anh can migrate downward in valency, but NOT upward, at least not without special marking (enter the applicatives!). Many verbs that are in English transitive are in Hra'anh fixed semantically at the intransitive level, and voice shuffling must be used to introduce a new argument.

The passive voice in Hra'anh works much the same way it does in English. "E khoite pep topoka" (the book hit him) goes to "Topoka pep ke e khoite" (he was hit by the book). Notice the use of the passive applicative "ke". The passive sentence could be said "Topoka pep" with perfect clarity. What hit him? Nobody cares. But if you must ask, the book hit him: "ke e khoite."

Enter the antipassive voice. It's hard to explain in English, but here's how it works: the passive voice demotes the agent (the book in the previous example); whereas, the antipassive voice does the opposite and demotes the patient or object. Again, we don't have this voice in English, so let's use a verb that's happy with or without an object, "to eat". If I were to say, "I ate," Hra'anh would be perfectly happy translating that directly: "Ier amira." Because this verb, kiamir, is one of the semantically intransitive verbs, in order to include an object, an applicative must be used: "I ate the sandwich" is "Ier amira doke e sandoesh." If you were to say "Iok e sandoesh amira," you'd be wrong because that sentence is badly formed. It may be perfectly correct syntactically, but it would be like saying "I fell the cliff" instead of "I fell off the cliff."

Notice that both the passive and antipassive voices in Hra'anh use word order shuffling to indicate which voice is used. The default word order is SOV (subject-object-verb). In passive and antipassive clauses, the order is shuffled to SVO, which is evident with the antipassive's omission of the object, SV(O). Passive clauses omit the subject, which then further fudges the word order to VO(S). Hra'anh has a highly configurational syntax, despite it being an agglutinating language (i.e. it uses bound morphemes to mark case).


The instrumental

Now we have the all-powerful instrumental applicative, felo. While there is no formal animacy hierarchy, Hra'anh is highly resistant to low-animacy and inanimate subjects in transitive clauses. For instance, if I wanted to say, "The knife cut the bread," I would have to switch to passive voice and use an instrumental applicative: "Sefa e alom felo e epun." To say "e epunek e alom sefa" would be ungrammatical.

Also note that the syntax for the instrumental applicative varies based on voice. In the active voice, it is postpositional; in the passive voice, it is prepositional.


Strange musics

One notable verb that throws every applicative rule out the window is fola, "to sing". It is the only musical verb in Hra'anh (you don't play an instrument; you sing with an instrument). Not only that, but it is also semantically intransitive.

This is how music works:
  • I sing. "Ier kifola."
    • This is the indicative form, morphologically identical to the infinitive form.
  • I sang a song. "Ier folaa doke iah loime." (1SG sing-PST ANTP a music)
    • Perfectly normal, right? Well, for Hra'anh, at least.
  • I played a guitar. "Ier folaa doke iah kitara." (1SG sing-PST ANTP a guitar)
    • Wait, what? This literally translates to "I sang a guitar". Shouldn't it use the instrumental applicative? This doesn't share the same semantic map as the English equivalent.
  • I played a song on the guitar. "Ier kitara felo folaa doke iah loime." (1SG guitar INST sing-PST ANTP a music)
    • There's that instrumental. Literal translation: "I, with guitar, sang a song."
  • I played the guitar while I sang. "Ier kitara felo folaa."
    • There it is again. As you can see, using the instrumental applicative demonstrates concurrent action. It's an edge-case example and a fairly strange construction.
  • I sang a song while I played the guitar. "Ier folanh felo folaa doke iah loime momo folaa doke kitara." (I.ABS voice INST sing-PST ANTP a music while sing-PST ANTP guitar)
    • Beware! This is a very strange construction. While it is technically grammatical, it is not pragmatic, and using it will get you labelled iah faila, a foreigner. Don't go into that much detail; just say, "Ier folaa doke iah kitara." Nevertheless, it demonstrates the use of an adverbial phrase.


 Other uses for the applicatives


The applicative doke must be used to elevate postpositional phrases when motion is involved. Some good examples are found in my fandub of the "eagle running scene" from Spirit: Stallion of the Cimarron. I had to translate the following sentences: "I ran with the eagle. I danced in the wind." 

The first sentence I translated, "Ier chtena doke e iekol ahrem." I could have said, "Ier e iekol ahrem chtena," but that would have a different semantic map: "I ran in the same place as the eagle." In this case, proper semantics can be achieved only by using the applicative to elevate the postpositional phrase.

In the second sentence, the semantics are nearly identical with or without the applicative. "Ier selema doke e khush ail" (which I chose) and "Ier e khush ail selema" both mean "I danced in the wind." I chose what I chose for poetic reasons. The example with the applicative means that Spirit danced in concord with the wind, whereas the example without the applicative means that Spirit danced while the wind swirled around him, as if the wind was a container.

This also appears in the last sentence, "Feri fetekhuza za doke zuthire zek ahrem," which is translated, "But responsibility came with this honor."



The applicative ke is used to relativize verb phrases in some passive constructions. This is also seen in my Spirit dub. The first sentence is translated back to English as, "So I grew up." The sentence: "Thele chreka ier ke lelvakh za." A word for word translation would be, "Then I was made to be an adult." This is because Hra'anh has no verb for "become". (This will be delved into more when I talk about reflexives in Hra'anh.) The verb ze agrees with the verb kichrek in its tense inflection. The applicative elevates "lelvakh za", describing what Spirit was made to be. The applicative is not an oblique when it comes to this verb! In fact, when using "kichrek", which means "to do/make", in the passive voice, the subject is forbidden. I would argue that this usage is not in fact the passive, but rather the middle voice, yet another voice that English doesn't really have.

This applicative also appears in the next-to-last sentence: "Esi siro ioz bun ier shasa, chreka ier ke e shana taia horositoz za," translated, "Like my father who came before me, I became the king of all horses."

No comments:

Post a Comment