Khulls verbs are broadly similar to Poswa's.
- 1 Structure of verbs
- 2 Tense and person markers
- 2.1 Gender marking on verbs
- 2.2 Semantic associations of noun genders
- 2.3 Breadth of gender categories
- 2.4 True person markers
- 2.5 Use of gender markers
- 3 Transitivity
- 4 Voice
- 5 Aspect
- 6 Verbal particles
- 7 Other modifiers
- 8 Notes
Structure of verbs
- NOTE, THIS PAGE CONTAINS ONLY INTRANSITIVE VERBS FOR THE TIME BEING BECAUSE THE TRANSITIVES ARE *NOT* SIMILAR TO POSWA. POSWA'S TRANSITIVE VERBS ARE ACTUALLY DERIVED FROM A VERBAL AUXILIARY THAT PUMMELED SO DEEP INTO THE VERB THAT IT CAME TO BE SEEN AS PART OF IT.
Strucuture of verbal stems
Most verbs end in vowels. This is because in the Tapilula language, all words ended in vowels and verbs were formed by moving the stress to the final syllable. In the Gold language, some unstressed vowels were dropped, but since all verb roots were stressed on their final syllable, no verb would ever come to end in a consonant. Khulls, however, has exceptions to this, because some verbs have come to end in consonants. These are of two types:
- Verbs that formerly ended in a stressed schwa vowel /ə/. In Khulls, stress drifted away from the schwa early on, and later the schwa was dropped entirely, usually leaving behind a residue of labialization on the preceding consonant. An example of this is ḳʷahʷ "to sleep in a bed".
- Verbs that formerly ended in a stressed syllabic consonant, which was considered a vowel at the time. Khulls does not consider these to be vowels and they cannot carry stress except in a word containing no vowels. This category contains only words ending with syllabic nasals, as the other syllabic consonants did not exist in the Gold language.
Because of the process above, verbs cannot end in high-tone long vowels; that is, the ā and á tones never appear at the end of a verbal stem.
STRUCTURE OF VERBAL SUFFIXES
There are many elements here, even more than Poswa. Gender, number, person of agent, person of patient, and tense are all marked by morphemes that fuse with each other, and there are 7 persons instead of 3. There are also 7 genders. However, the genders of both agents are not completely marked; there is syncretism here. Number is sing/dual/plur(sep)/plur(coop), ther same as Poswa, although Poswa marks the nouns instead of the verbs. Number is marked only for agent(?). Tense does not fuse with aspect.
/k/ (not /ḳ/) marks 1st person. /x/ marks 2nd & 3rd, which are distinguish by noun class. These are the poswob noun recip/co-op mEmbers
It's impossible these actually mark trans & intransitive verbs, though that corresponds better to poswa.
gender prevents inanimate from being able to do transitive verbs ... either it collides with 2p or with a noun
ELEMENTS THAT NEED TO BE MARKED:
- Person of agent. Probably not marked for gender, therefore 1st person and 2nd person can be treated as pseudo-genders.
- Gender of agent. See above; the concept of gender and person might be unified outside of the 3rd person. Possibly make all compound persons epicene ... e.g. "you and I" is automatically epicene even if both are male.
- Number of agent. Here, too, there will be syncretism, which is why this is fusional rather than agglutinative.
- IF* gender is marked on person (as a relic of an older system in which it denoted possessed objects perhaps), it is irregular. given below are the consos correspondign to each gender, in bare form, then 1st person, then 2nd. (3rd would be same as bare form: )
♀ m m m ♁ p p p ⚳ s m m ☿ n nʷ ŋ ☼ d dʷ y ⚲ — w y ♂ t tʷ h
Note that /h/ merges with neuter since it always attaches to an aspirated consonant,. nʷ may be replaced with /m/ in some words, and then later extended to all weords.
Khulls may add an /l/ to all 1st person forms, on analogy with the words for men and children, which would be reflected as /nl/, /l/, and /l/ otherwise. this would likely cause /ml/ > /bl/ > /b/ (except in PMS). possibly also /nl/ > /dl/ > /d/.
- Person of patient. Probably marked in a relative sense, with categories for "self", "each other", and "other person", and only in the third case is an additional person marker required.
- Gender of patient. Only in 3rd person.
- Number of patient. Probably not marked on the verbn at all.
- "Declension": a dummy variable to prevent the 3 verb declensions from merging with each other. Grouped with patient rather than agent because agent is the more outward morpheme.
- Tense. Probably not merged with aspect or with mood.
Tense and person markers
Gender marking on verbs
Gold nouns have complex gender inflections. See also Khulls verbs.
|Gender||Epicene ♁||Fem+ ♀||Fem- ⚳||Young Fem ☿||Unisex ☼||Neuter ⚲||Masc ♂|
|4 Greater Feminine ♀||♁||♀||♀||♀||♀||♀||♁|
|3 Lesser Feminine ⚳||♁||♀||⚳||⚳||⚳||⚳||♂|
|3 Young Feminine ☿||♁||♀||⚳||☿||☿||☿||♂|
|1 Unisex ☼||♁||♀||⚳||☿||☼||☼||♂|
|0 Neuter ⚲||♁||♀||⚳||☿||☼||⚲||♂|
|4 Epicene ♁||♁||♀||⚳||☿||♁||♁||♂|
|4 Masculine ♂||♁||♁||♁||♁||♂||♂||♂|
Semantic associations of noun genders
Most words for objects that are non-human, but living, belong to one of the animate genders. The assignment of gender largely follows semantic boundaries, but the choice of which gender to use has little in common with an object's physical characteristics. For example, all words for sea life belong to the greater feminine gender, even if they are words for male animals. Celestial objects also belong to the greater feminine gender. All words for birds, meanwhile, belong to the "maiden" gender, also known as the young feminine. (NOTE: This is true in some daughterl ags, not gold itself)
Greater feminine gender
The greater feminine gender is marked primarily by -m- and -s-. Words in the greater feminine gender usually belong to one of the following semantic categories:
- Adult human females.
- Edible objects, particularly processed foods rather than those which are edible in their natural form.
Lesser feminine gender
The lesser feminine gender is also marked primarily by -m- and -s-. Words in the lesser feminine gender usually belong to one of the following semantic categories:
- Adult human females.
- Celestial objects; fire.
- Snakes and worms.
- Abstract concepts such as love and beauty.
- Names of rivers and nations.
- Soft objects.
- Women's clothing and feminine hygiene products.
- All sea life, including penguins.
- Females of certain large, domesticated mammals.
- Soap and mixed potions.
- Money as an abstract.
Young feminine gender
The young feminine gender is marked primarily by -n-. Words in the young feminine gender usually belong to one of the following semantic categories:
- Young girls and unmarried women.
- Most fruits that can be eaten in one sitting.
- Birds other than penguins.
- Sharp objects.
- Most placenames other than those of rivers and nations.
- Money in the form of coins.
Manmade handheld objects are often found in this gender, from association with Mumba rae "hand", whose locative case yielded yaʕ in Gold, even though this word itself has not survived. (The word for coin may simply be derived from this very same morpheme, if it can be traced further back.)
The masculine gender is marked primarily by -t- and -d-. Words in the masculne gender usually belong to one of the following semantic categories:
- Men and boys.
- Males of some large domesticated animals.
- Some words for fruits. (More common in the descendants than in Gold itself.)
The epicene gender is marked primarily by -p- and -d-. It is always plural. Words in the epicene gender are generally words for groups of humans, or groups that include humans, but there are exceptions such as
- Water and other fluids whose name is derived from the word for water.
- Possibly round objects (from Mumba puarna).
The epicene has no singular form. It generally refers to groups of people of mixed gender, and therefore is never singular either as a subject or an object. It often corresponds to English "they/them". Epicenes can in fact refer to a single person, but only when of an entity whose size is unknown (e.g. "those who passed the test", even if only 1 student passes). Also, many words for mass nouns are epicene. For example, water.
The epicene therefore cannot be 1st person singular or 2nd person singular, either as a subj or an obj. Additionally, it never changes (in most langs) when serving as a patient for an agent of a different gender.
- BETTER IDEA... 1ST AND 2ND PERSON *CAN* TAKE GENDER INFLECTIONS, BUT IT MEANS "ME AND A WOMAN", "YOU AND A MAN", ETC.
The unisex gender is marked primarily by -d-. Words in this gender are reassigned to the masculine gender in many daughter languages. Words in the unisex gender usually belong to one of the following semantic categories:
- Babies whose gender is not known or not expressed.
- Reptiles, amphibians, and some small mammals of either sex.
- Grass, flowers, and small plants; apples and pears.
- Some words for diminutives.
The neuter gender is the only true inanimate gender. It does not have a thematic consonant. Neuters in bare form can never be the subject of a verb, and they have no distinct accusative case because the nominative serves also as the accusative. Words in the neuter gender take on the gender of their possessor, however, when the possessor is animate, which allows them to become the agent of transitive verbs. Words in the neuter gender tend to be words for
- Inanimate objects of all types not included in any of the above metaphorically animate categories such as celestial objects.
- Very primitive animals seen as unable to act on their own behalf.
Since neuters cannot be agents of verbs(except a few irregulars), it may make sense to have 1st & 2nd person pronouns behave as if they were neuters. Essentially, a neuter 2nd person agent marking on a verb is equivalent to using the pronoun "you" in a pronoun-using language. In some ways, this will not "feel" like a neuter since the 1st & 2nd person args will be marked with consos, like the animate genders.
Thus, in Gold, plants actually rank higher on the animacy hierarchy than some animals.
Breadth of gender categories
The assignment of animals to particular categories is based on their habitat rather than their taxonomic ancestry. Thus, for example, penguins are grouped with fish in the "sea life" category, rather than with other birds. Bats, however, are found in the unisex gender, alongside other small mammals.
This tendency became much stronger in two of the daughter languages of Gold, Babakiam (ancestor of Pabappa and Poswa) and Khulls. This is because these languages dumped the classifier system but retained the categories to some extent, and forged new classifier-like morphemes that attached to the end of a word rather than the beginning. In both branches, the locative case became a zero morph in unstressed syllables, meaning that, for example, -pa "in the water" (for Babakiam) and -e "in the water" (for Khulls) came to appear on many nouns.
Gender is marked for both agent and patient, but there is massive syncretism. Thius charty should be read not to sat that the different elements on reach row are the same ,but that the different elements in each column are. Thus, e.g. if the obj is female, all of the "female" entries in the list imply the same marker ON THE OBJECY *(not the verb). BUT in the female ROQW, all of the entires thart are the same are th different ON THE OBJKECT>
The verbal endings are always unstressed and the stress migrates to the last syllable of the stem. The gender marker is inserted before the vowel.
- NOTE, these are actually 3 declensions,not 3 persons. Person is marked only by consos.
True person markers
The personal infixes -an- and -aŋ- may be retained from Tapilula. The first means 1st person subject, the other means 1st person object. In origin, however, they both just mean "human" and the second infix may in fact be related to the circumstantial noun case.
Use of gender markers
Khulls retains the use of the otherwise mostly obsolete consonantal gender markers inherited from the Gold language. For example, with the verb kʷî "to dream", one can say
- Šŭpe kʷînī.
- Suphoi (a girl's name) dreamt.
- Baṭà kʷîrī.
- Batak (a boy's name) dreamt.
Use of gender markers with verb stems ending in consonants
They seomtimes mingle with the consonants at the end of verbs, with each consonant affecting the other. For example, with the verb lixʷ "to talk", one can form sentences like
- Ŏma limpʷī.
- The woman talked.
- Lăxi likʷī.
- The man talked.
Note that the Roman orthography the verbs above implies that the verbs are accented on its ending rather than the stem; this is not true, and results from a convention of not adding accents to roots which are spelled without them in bare form.
Six infixes mark the object of the verb.
-aʕ- himself; themselves (each acting on himself) -al- himself and someone else (each acting on himself) -ey- some other people (acting on each one individually) -am- themselves; each other -an- themselves and some other people -im- some other people
- note, this might not be true. '-ey-' is cog to Poswa's passive former, which means its meaning here is nearly the opposite of what it should be.
The passive voice is marked by -ik-, and the mediopassive is marked by -ak-.
Inanimate nouns cannot be the subjects of transitive verbs, and therefore the voice markers are unnecessary as all verbs are assumed to be passive.
- note, do inanimates have genders in Khulls?
Imperfect aspect is marked by an infixed syllabic -ṁ-. Frequentative is marked by an infixed syllabic -ṅ-.
The particle pair ši ... ka indicates "if ... then".
Khulls has a diminutive. However, this is not a diminutive for nouns, but for verbs; it gives a verb the meaning "...like a baby" or "...like a child". It is an incorporated subject historically descended from the word ì "baby; child; something little". However, in most verbs, the subject has become silent and instead voices (and sometimes lenites) the initial consonant of the next morpheme in the verb. This process works the same regardless of whether the next morpheme in the verb is the verb stem itself, an incorporated object, or something else.
- From Mumba neyumpuṭ. Here, ne- is the classifier prefix.
- From Tapilula nə- "claw", which is cognate to Late Andanese gi- with a similar meaning to the Gold prefix.
- Possibly wrong; appears to be based on Andanese ni-, which lost an initial schwa.