Reforming the Ŋarâþ Crîþ verb and relational inflection system.
- use the Project Elaine framework for verb conjugation and make use of stem fusion
- allow derivation from verbs to nouns (i.e. provide all necessary stems)
- support at least the same forms as the old system
- be aesthetically pleasing
Relationals are included in the scope of this project because of the existence of finite and nominalized forms.
To subjunctive or not to subjunctive?
Things that could possibly use a subjunctive mood:
- future tense
- negative polarity – not happening.
- optative mood, probably with a particle
- with nominalized VPs acting as adjuncts (e.g. locative and semblative cases), distinguish between a realized action and an intended but unrealized action
In general, add marking for subjunctive in contexts where it isn’t already shown.
Short and long verbs
Proposal: separate verbs into short (1 full syllable in stem) and long (2+ full syllables in stem) verbs.
The third-person singular present imperfective stays the same for short verbs but is the stem fused with null for long verbs.
This, however, produces ambiguity with verbs such as ⟦crešarat⟧, whose 3sg would be ⟦crešar⟧, which is also the 2pl of ⟦crešit⟧. This verb could still be classified as short because it is a compound of ⟦crešit⟧ and ⟦sarat⟧, the latter of which is short. Also, verbs whose stems have only 1 syllable in the stem could be classified as long anyway if the final bridge is sufficiently complex (e.g. ⟦neftrit⟧).
Perhaps the condition should be whether fusion with null gives at least 2 syllables? Alternatively, perhaps long verbs should be rare even among 2+-syllable stems.
Stems: I, N (nonpast), P (past), R (nominative-rcase relative), Q (other-rcase relative), L (locative)
Theme: Θ (⟦a⟧ or ⟦i⟧), with transformed forms Θτ, Θφ, Θψ
|Person \ Number
|-Θτ · ηn
|-Θφ · ηcþ
|Person \ Number
|-Θτ · ηnis
|-Θφ · ηcis
Use the N stem for the present tense and the P stem for the past tense. If the aspect is perfective, then eclipse the form. For the past tense, suffix ⟦-þ⟧ for the direct 1sg, 3sg, and 3pl forms; otherwise, suffix ⟦-ta⟧ after the object affix or use the Pt stem (TODO: how to choose?).
The second-person dual direct suffix is ⟦-Θφns⟧ if there are no subsequent suffixes but ⟦-Θφn⟧ if there are.
Object affixes remain unchanged.
|Person \ Number
add a distinct ‘equative’ form? (used to assert that one action is tantamount to another)
- this is the bulkiest section
- genus 0 could use some simplification; can we also simplify 2?
- want to keep distinction between type I and type II genera
infinitive form is ⟦IΘt⟧
converb forms would probably use the I stem, but a different stem is a possibility for some of them
introduce distinct subjunctive forms for locative and semblative cases?
adnominal forms for last 4 cases not used that often; do we really want to keep them?
dative nominalized forms are used often, especially as objects of relationals, so probably want to shorten them to an inflected form
S stem of derived nouns derivable from N stem by modifying the final bridge
L stem exists separately and is definitely distinct from N (perhaps use it for locative, instrumental, & abessive nominalized forms?)
for calculus affixes, infix them in every stem
might want more irregular verbs beyond ⟦eþit⟧ and ⟦telit⟧
for the existing irregular verbs, 2du ends with a simple coda, but will this hold for future ones?