Three Dollar Quill

In this post:

• Just a giant spew of made-up grammar facts.

---

I'm making a judgment call, and stopping the daily sentence quota. I'm just slipping a lot, and it's hard to deal with.

Anyway, let's see how this works.

There are several categories of word, which each have different requirements for the associated information:

• verbs mark person, then number, then aspect
• nouns and adjectives mark number, then case
• names and pronouns mark person, then number, then case
• conjunctions between nouns mark case

Persons are:

• first
• second
• proximate
• obviative

The obviative is used to mark a distinct third person from the proximate, which is the default, except in certain sentences.

Numbers are:

• singular
• dual
• plural

The plural number is used for mass nouns and some abstract concepts.

Aspects are:

• imperfective: sets up the verb as a background action
• perfective: sets up the verb as an event, possibly in the context of a background action
• inchoative: indicates that the verb is beginning
• cessative: indicates that the verb is ending

The latter two can be used to distinguish meaning where some languages would use different verbs.

So far, there have been two basic sentence structures.

A typical sentence starts with a verb, and follows it with noun phrases with an associated case, which generally go in order:

• nominative
• accusative
• dative
• intrative, comparative, privative, adessive, instrumental (these have never appeared in a sentence together)
• inessive

There is also a genitive case, which doesn't appear on its own at the noun phrase level, but does appear within certain constructions.

There was also, until about five minutes ago as I write this, a "reflexive" case in the lexicon. It functioned as a combination of nominative and accusative.

Looking over this, I'm probably going to get rid of the privative case. I think it can be replaced with something like "[ala (adj)].number.genitive [thing].number.genitive'[wan (adj)]".

Anyway, the other major sentence structure just puts the accusative noun phrase before the verb, and makes the nominative noun phrase optional; it's sort of a passive construction.

Notes on specific cases:

• Dative can often mean "for", as in a benefactive, or possibly a causative. It also signifies the indirect object. And, in some contexts, the destination. ... I just realized I used it wrong for all sentences with [pana (v)]. Aw, they're now less bonkers. Although, I did have to introduce a nonce indirect object into one, for reasons.
• Inessive is used with time periods to indicate tense, purely lexically.
• The comparative is used a lot to translate adverbs modifying verbs. It's possible that I'll change out the "like a fast thing" constructions with stuff like "like a cheetah". Taking about the manner of a verb with nouns rather than adjectives.

The genitive case is used to form possessives, superlatives, and comitatives, and can probably do other things.

To use the genitive:

• Start with a noun phrase in genitive case.

• Move the head from the beginning to the end.

• Convert the end of the head to a contraction.

• Suffix it with an adjective that agrees with the noun before the genitive phrase.

  • If the adjective is [jo (adj)], then this expresses that the referent of the modified noun phrase is owned by the referent of the genitive phrase.
  • If the adjective is [wan (adj)], then this expresses that the referent of the modified noun phrase is accompanied by the referent of the genitive phrase.
  • For most other adjectives thus far, the referent of the modified noun phrase is exemplary within the context of the genitive phrase, in terms of the adjective.

Adjectives normally follow the noun they modify. I currently have determiners under the "adjective" umbrella.

It's possible for a sentence to have multiple verbs, and the mechanics of this were deliberately chosen to confuse:

There is no infinitive form. Each verb has a subject. The subjects MAY be equivalent noun phrases. If they are, the subjects after the first (... I think) can be omitted. If they are not, the verbs after the first agree with the first verb, and not their subject, in terms of person and number. The verbs are not required to agree in terms of aspect, because trying aspect agreement took it from "pretty confusing" to "way too confusing".

Usually, the first verb has a special meaning when used like this:

• [sin (v)] again
• [tu (v)] twice
• [mute (v)] often, usually
• [wile (v)] want, hope, should
• [o (v)] imperative
• [jo (v)] to have something as a result of
• [sama (v)] probably
• [ken (v)] can
• [toki (v)] converts the rest of the sentence to a yes or no question

When I combined multiple verbs with the passive form, I concluded that the first verb has to include a nonce indirect object, if it doesn't already have a direct object or later case.

So far as verbs of motion, the language is mostly "verb-framed", and encodes the manner of motion using the instrumental case.

There is no specific equivalent of "to be". So far, I've used [lon (v)] to talk about location, and [sama (v)] to talk about appearance.

I haven't explored this in the example sentences, but I want the "prestige dialect" to favor "name-like" noun phrases over pronouns. Basically, the way this would work is that there are certain "name-like" nouns that encode person in addition to number and case, and these would be used both to conclude each part of a name, but to construct "name-like" versions of phrases along the lines of "humble servant" or "most merciful".

One thing that I haven't really thought about hard with these sort-of glosses I'm doing is the "indefinite deteminer", [sin (adj)], the equivalent of "a(n)". It exists to introduce new ideas, and there's no definite counterpart; things are assumed definite by default.

I think that's everything I've worked out so far.

I think what I want to do is take a break, then come back to this and try to use Describing Morphosyntax on it.

---

Next time, I wrap this up for now.