Churra

General information
Churra is a mostly isolating language with some fusional elements, and a heavy dependence on word order to illustrate many grammatical concepts, including tense. It has an exceptionally large consonant inventory, with many fricatives, but very few vowels. It is spoken in a small coastal region of the planet Ysla, by the Chasrae, an alien species of "crab-taurs" (crustaceans with an upper torso similar to a human) with a tri-sexual reproductive system.

It is peculiar in its large phonemic consonant inventory (68) including some sounds not possible for humans to produce.

Consonants

 * 'Ventral' sounds are non-pulmonic consonants, formed by the articulation of a stop and the channeling of air through a disparate voice organ that is not available to humans - a secondary, ventral windpipe or 'wind chamber' that is a vestigial remnant of the specie's once-present gills. It gives a rasping, whistling quality to the affected sounds, and can be approximated in human speakers by use of nasals, ejectives, implosives, or clicks at the corresponding point of articulation, which are not available to the species, who have no nasal passage.

Palatalization and Labialization
Any consonant can be modified from C to Cʲ or Cʷ (with the exception that /w/ cannot be labialized, and /y/ cannot be palatalized, for obvious reasons; however either can undergo the opposite process). These are phonemic distinctions, effectively tripling Churra's consonant inventory. This is written in the romanized orthography as Ce and Co.

Alphabet
Churra is written in ideographs and sometimes in a syllabry. Both are under development. The romanized alphabet used is included in the phonemic inventories chart.

Note that in some instances long sounds may simply be written doubled, e.g. ĥ = hh, î = ii. This is not the preferred method since the syllabic consonants can be interspersed with their non-syllabic counterparts, as can be seen in the word xŝsŝ, 'snake,' which with the doubled-character system would need to be transcribed xsssss. Despite what a reader's intuitions may be, the sequence in 'snake' should be composed of two audibly seperate syllables, [xs:.ss:] not merely a very long-held s sound [xs:::].

Stress structure
Churra, unlike human languages, has a trisyllabic stress pattern with foot-initial stress, leading to the grouping of its syllables into triplet feet; kè-ke-ke-ké-ke-ke. (A tri-trochee?) However, since a large percentage of Churric words are only one or two syllables long (degenerate feet are obviously permitted), the pattern often appears to simply be word-initial stress.

The word-head is final, feet count from left to right, feet are iterative, extrametricality is not permitted, and it is quantity-sensitive; syllables with a long vowel or a vowel followed by a glide are considered heavy.

Syllable structure
The basic syllable stucture in Churra is CV. However, word-initial vowels and word-final consonants are allowed; this makes it possible to occasionally know word boundaries, where consonant clusters not permitted at the word level occur.

Hiatus is never permitted; if an affix would create hiatus, the first vowel assimilates to the second, resulting in one long vowel; if either vowel is already long, the hiatus will be interrupted by an appropriate glide. Most consonant clusters are not permitted, save some exceptions; C+glide (G; phonemicized as labalialized or palatalized Cs), phonemic affricates (/c/ and /p/), and the long fricatives (F:). The long fricatives may function as either a long consonant or a voiceless vowel. In the available syllable structure, any phonemic consonant (except F:) or any short vowel counts as one mora, while long fricatives and vowels count as two. Each syllable may contain a maximum of three mora, although this rule is ignored by word-final consonants. All legal syllables in Churra can therefore be accounted for in the following: When a structure such as CF:V is attempted, the syllable is broken into two, which results in a short syllabic fricative; *[.kssa] is blocked, but [ks.sa] is permitted.
 * when the initial vowel is /a:/ :
 * a:V: → a:V:
 * e.g. â+ûla → âwula
 * a:V → a:G#, GV elsewhere
 * e.g. tâ+u → tâw ; â+iva →yiva
 * aV: → GV:
 * e.g. na+ûdik → noûdik
 * elsewhere
 * VːVː → VGVː | VːV → VGV | VVː → GVː
 * (C)V
 * e.g. ku, kû, u, û
 * F:V
 * e.g. ssa
 * (C/V)F:
 * e.g. kss, iss, ss
 * (C)VC#
 * e.g. bat, bât, at, ât

Regular sound changes

 * /hʷ/ = [ɸ]
 * /l/ = [l] /_#
 * /l/ = [ɾ] /elsewhere
 * /p/ = [pɸ] OR [pf] (free var.)

Conditioned sound changes (IPA)

 * i > e /j_, Cʲ_
 * i: > eɪ /j_, Cʲ_
 * u > o /w_, Cʷ_
 * u: > oʊ /w_, Cʷ_
 * u, u: > ʉ, ʉ: /palatal_
 * a, a: > ɑ, ɑ: /_uvular_, _h_
 * C > C [+voice] /V_V
 * C > C [-voice] /_#
 * ð > θ /F:_
 * h > x /_i
 * h: = x: /_i, _palatal

Pronouns & Grammatical Number
The pronouns which serve as the verbal arguments in Churra have three distinctions - class, person, and number. The only aspect of Churric nominal declension that the pronouns do not exhibit is gender (animacy).

The pronouns use a two-way class system (subject/object), encode three persons (first, second, and third), and capture the language's five-way number system.

Despite the fact that all other relevant case-marking elements of Churra make a three-way distinction for Subject, Direct Object, and Indirect Object, the pronouns only have S and O forms. Pronoun order is relied on to convey case: the order is IO-S-DO.
 * taras lu zu
 * kill.TR 1SS 2SO
 * 'I killed you'
 * *taras zu lu
 * kill.TR 2SO 1SS
 * '*I killed to you'
 * *taras lu cu
 * kill.TR 1SS 2SS
 * '*I you killed'
 * *taras cu lu
 * kill.TR 2SS 1SS
 * '*You killed I'
 * taras cu ilu
 * kill.TR 2SS 1SO
 * 'You killed me'
 * *taras ilu cu
 * kill.TR 1SO 2SS
 * '*You killed to me'

Grammatical Number
While the meaning behind the number distinctions are more transparent when applied to nouns (see 'the noun,' which also includes the nominal declensions for number), the five-way system allows for a variety of ideas to be communicated.

Zero/Negative
The use of Zero/Negative (Z) pronouns in a sentence can convey a similar message as the negative in other languages.

Note that in realworld terms, there is no observable difference between 'We didn't go' and 'None of us went' ; neither allows that any of the 'we' group went, and neither precludes that nobody else (none of you/them) went. In general, however, there is also a negative particle which allows for the exact number of non-goers to be stated: ixa.
 * ba ix
 * go.IT 1Z.SUBJ
 * 'I/we didn't go'
 * 'None of us went'
 * ba ixa lu
 * go.IT NEG 1SG.SUBJ
 * 'I didn't go'

Singular and Plural
The singular (SG or just S) and plural (PL or just P) numbers behave nearly identically in Churra as they do in English; they captures the concept of 'one' and 'more than one' (roughly: see 'paucal' below). Therefore the first person singular subject pronoun lu can be thought of as equivalent to English 'I,' and the plural counterpart is the same as English 'we.' When used to refer to a mass noun, the singular acts as a singulative.
 * ba lu 
 * go.IT 1SG.SUBJ
 * 'I went'
 * ba di
 * go.IT 1PL.SUBJ
 * 'we went'
 * yuf la gumssa. yulssu vu.
 * COP PART.SJ grain. fall.IT 3SG.SJ
 * there is some grain. one grain fell.

Paucal
The main difference between plural in English and plural in Churra is that the Churric form conveys, specifically, 'many,' or at least 'more than a few.' This is because it is the paucal (PC)'s job to encode 'a few.' When used in conjunction with mass nouns, the paucal means roughly 'a small percentage':
 * ba idi
 * go.IT 1PC.SUBJ
 * 'a few of us went'
 * yuf la gumssa. yulssu iwi.
 * COP PART.SJ grain. fall.IT 3PC.SJ
 * there is some grain. some of it fell.

Mass/Collective
Mass/collective (C) number is similar to 'noncount' nouns in English; it is used mostly with fluids (e.g. water, air), and large collections of miniscule objects (e.g. grain, rice, sand). However it can also apply to large groups of anything which are then considered a collective (similar to 'a herd of deer' vs 'several deer').

It is worth noting here that the C marker for nouns (see 'the noun') will not always appear on nouns which behave collectively, such as 'water,' but these are nevertheless referred to with collective pronouns.
 * ba tilu
 * go.IT 1C.SUBJ
 * 'all of us went'

Reflexive Pronouns
Since reflexives must be used in conjunction with another pronoun, it is unimportant to know number on a reflexive - these follow their subject.


 * unut lu
 * hit.TR 1SG.SUBJ
 * 'I hit it'
 * unut lu fss
 * hit.TR 1SG.SUBJ 1RFX
 * 'I hit myself'
 * unut di fss
 * hit.TR 1PL.SUBJ 1RFX
 * 'We hit ourselves'
 * unut ix fss
 * hit.TR 1Z.SUBJ 1RFX
 * 'I/we didn't hit myself/ourselves'

The Verb
Verbs must show transitivity and tense, and can have modal, aspectual, and further valency information added by means of particles.

Transitivity, Valency & Voicing
Each verb has an inherent transitivity feature, and come in series of related verbs to allow for different transitive forms. There are five different types of verb; transitive, intransitive, ditransitive, 'dative,' and 'passive.' The 'endings' are often violated by exceptions. The transitivity feature was once an inflection on the verb, but for most verbs it has become fossilized and non-decompositional.

Inherent transitivity allows third-person-singular pronouns to be optional. If the verb is IT and has no visible subject, that subject is a third person; if the verb is DT with no pronouns attached, then all three persons are third persons. Other distinctions and combinations of persons may be made with valency-modifying prefixes (also known as voice markers):
 * sa- indicates passivization of the subject pronoun on TR or DT verbs.
 * sataras ilu
 * kill.TR 1SG.OBJ
 * 'he is killed by me'
 * fe(a)- indicates causation on DT verbs; with fe-, the IO causes the S to act on the DO.
 * featar ĥ ilu cu
 * kill.DT 1SG.OBJ 2SG.SUBJ
 * 'he made you kill me'
 * ju- is the counterpart to fe-: the DO causes S to act on IO.
 * jubalir ilu cu
 * speak.DT 1SG.OBJ 2SG.SUBJ
 * 'he made me speak to you'
 * ci- indicates 'dipassive,' or a verb with both a DO and IO.
 * cigufiz ki muzaqoi i mŝ a 
 * DP.pull.PA the.IO ground the.DO flower
 * 'the flower was pulled from the ground'

Tense
Churra has a two-tense system, future and non-future (which will be referred to as 'past' at times for sake of simplicity, despite inaccuracy). There is no morpheme, auxiliary verb, or verb form that indicates tense; instead this is acheived through word order.

In future-tense phrases, the arguments will appear before the verb; in the non-future, the arguments appear after the verb. This is true regardless of the argument form (pronoun vs noun), unless it is an entire embedded clause in a future-tense sentence, in which case the impersonal sub-clause pronouns appears before the verb, and the longer, imbedded clause appears afterwards.

Since the third-person pronouns are optional, many phrases will have no visible arguments or other methods of indicating tense-aspect or temporal information. In this case it is generally assumed that the phrase is in non-future tense, except in certain narrative conditions (e.g. following a future-tense phrase). While the non-future corresponds to English past and/or present for the most part, there are some instances where a phrase that would be present in English is future in Churra. This is especially true of phrases with an irrealis mood, or when a verb has consequences that may extend into the future.
 * mouga (run.IT) 'he runs/ran'
 * vu mouga (3SG.SUBJ) 'he will run'
 * mouga lu (1SG.SUBJ) 'I run/ran'
 * yaj lu mouga ss... (3SG.IO) 'I will run to...'

Aspect
Another important set of distinctions of the Churric verb are its aspects. These always appear directly before the verb (except for voice markers). The first six aspects must, in a way, agree with the phrasal tense. Adding su to a future phrase, for instance, would cause an ungrammatical and infelicitous utterance.

The remote-perfect/simple aspect, frequently shortened to 'remote,' indicates that an event occured a long time in the past or will occur far in the future. This is therefore one of the few ways to know for sure that a non-future phrase is past, and not present. It has been partially co-opted to merely indicate past tense for this reason.

The anterior-progressive is generally just called 'anterior.'

Gnomic aspect is used to state that an event (or state) is always true, or that a statement is a general (usually well-known) truth, e.g. sa juma tîla 'birds fly.' It is sometimes used as a means of trying to make a sentence seem more truthful and honest.

The stative is used to describe a situation which is ongoing but unchanging.

The semantic difference between the future and non-future of the stative and gnomic is minor at best, which is probably why there is only one form for each.
 * wira lu 'I run/ran'
 * su wira lu 'I ran a long time ago'
 * *li wira lu 'I ran a long time from now'
 * lu li wira 'I will run a long time from now'
 * bu wira lu 'I am/was running'
 * lu mi wira 'I will be running'
 * ku wira lu 'I had been running'
 * lu xi wira 'I will have been running'
 * sa wira lu 'It is true that I run'
 * lu sa wira 'It is true that I will run'
 * na wira lu 'I (have) always run'
 * lu na wira 'I will always run'

Aspects are only allowed to stack if one has future/non-future distinction and the other does not; in this case, the gnomic or stative particle appears before the other. In this method the stative and progressive may be combined to create the pluperfect.
 * sa bu sirif zu lu
 * ST PR.P wait.DA 2SG.OBJ 1SG.SUBJ
 * 'I have been waiting for you'

Mood
Most modal sentences, especially ones that use irrealis moods, that would take the present in English are in future tense in Churra.

Determiners & Case
There are three types of determiners in Churra: articles, demonstratives, and possessives. Articles communicate case and definiteness, while demonstratives encode spacial relation to the speaker or subject. It is possible to use both partitive and/or tool articles and demonstratives together.

Articles
The 'proper noun' articles are only required when the sentence's structure leaves case ambiguous.
 * a mŝa 'the flower (sj)'
 * at alois 'the sun (sj)'
 * mŝa 'a flower (sj)'
 * lat ara 'some still water (sj)'
 * i alois 'the sun (oj)'
 * ki alois 'to/at/etc the sun'
 * tin alois 'with the sun'
 * vujaw utuga icibirilĥk ma Tum
 * laugh.DA monkey PAUC-trick-PSS.IO the.SUBJ Tom
 * 'Tom laughed at the monkey's tricks'

Demonstratives
There are 36 demonstratives in Churra, which encode three axes of information: case, proximity, and vertical direction. What is immediately apparently is that there is a sound-meaning correspondence between the vowels and the height distinctions (above, equal/eye-level, below). This sound symbolism is present in other parts of the language as well.

The four levels of proximity are proximal, medial, distal, and surrounding, which correspond roughly to English this X, that X, yonder X (cf. Sp. aquell@), and the surrounding X.
 * lu nisit mŝa qapul
 * 1SG.SUBJ DEM flower want.TR
 * 'I want that flower up there'
 * bu wudiwi tal ara lu
 * PR.NF swim.DA DEM still.water 1SG.SUBJ
 * 'I am swimming in still water'

Possessive Case
There are two ways to form the possessive in Churra - by using a possessive pronoun (like English my, your, her) or by using the possessive declension. (When using possessive pronouns, but the pronoun and the case marker must be used.) This is the only form of declined case in Churra. Unlike in English, these endings are not applied to the possessor (Mary's kitten) but to the possessed. The possessor is left without any article.
 * Mali nanîm 'Mary's kitten (sj)'
 * wazat a kavoa Mali naniŝd
 * chase.TR the.SUBJ dog Mary kitten-PSS.DO
 * 'the dog chased Mary's kitten'
 * wazat a kavoa vim naniŝd
 * chase.TR the.SUBJ dog 3SG.PSS kitten-PSS.DO
 * 'the dog chased her kitten'

Adpositions
These fall under the category of determiners in Churra since they may replace articles and demonstratives in the phrase, standing in place of case or directional marking. They are generally found before the noun they modify.

The Noun
The noun in Churra has little morphology. However, it does show number, and something akin to animacy.

Number
Churra has a five-way number distinction. The behaviour of these numbers is not entirely predictable to an English speaker (see also 'pronouns and grammatical number,' above).

The 'zero' number behaves almost identically to a negation particle: Zero nouns act as non-definite nouns, taking no article. One may specify that the X did Y by using another grammatical number, an article, and the negation marker ixa. Singular and plural are nearly-identical to English. The distinction between paucal and plural is best summed up as 'a few' vs 'many/ more than few,' ; the exact number being described depends on context, what the object is, etc. However, the number range for paucal is generally anything between or including two and six.
 * yuf feimŝa
 * COP Z-flower
 * 'there are/is no flower(s)'
 * yuf ixa boamŝa
 * COP NEG PL-flower
 * 'there are no flowers'

Many nouns considered collective (AKA 'mass' or 'noncount') in English will be unmarked, and therefore, apparently singular (e.g ara, 'still water'). However, they still behave collectively (taking collective pronouns and partitive articles). Nouns that appear with the ta- prefix are usually count nouns that have been made into a group, and now act as one entity.
 * wudiwi keu ara lu. yuf uqura vu.
 * swim.DA PART.IO still.water 1SS. COP coldly 3SS.
 * I swam in still water. It was cold.
 * Note that the 3SS pronoun is usually omitted, it is only shown here to illustrate that it is singular.
 * alisaj lu taras la takavoa koa tîla
 * see.TR-subc 1SS kill.TR PART.S C-dog PART.DO C-bird
 * I saw (that) a pack of dogs kill(ed) a flock of birds

Animacy/ Word Class
There are 19 common Churric noun endings. These are fossilized remnants from a once-productive animacy system, with five major classes of animacy; this was once decompositionable, but it is no more. While the animacy of the noun itself is no longer particularly important grammatically in Churra, there is a correspondence between semi-productive adjective animacy and the nominal form. Note that adjectives must agree with the animacy of their noun, and not the specific noun class. The specific classes are now completely defunct.

Animacy distinctions might not be completely intuitive to English speakers; for instance both ara 'still water' and fŝoa 'turbulent water' have fluid animacy, while narag 'the substance of water' is inanimate.

Modifiers
Churric adjectives are fairly simple, agreeing with their noun's animacy and having no other morphology. The adverbs are slightly more complex, dividing into temporal adverbs (which must agree with tense) and non-temporal adverbs, which behave identically to adjectives.

Distribution and morphology
These two word classes are only differentiated by their distribution; they are therefore considered the same word class in Churra, and will be referred to as adjectives henceforth for the sake of simplicity.

Churric adjectives follow the noun or verb they modify, and must agree with what they modify, on a six-tier system: verb (adverbs), and the five animacy sets. Therefore any one English adjective has up to six corresponding words in Churra (and if a form doesn't exist, it can be invented.) There are a few exceptions to the N-A order. For one, when a N+A construct is a compound (considered one noun), the order is reversed: The other exception is colour adjectives, which always preceed their noun, and which have only one form, regardless of animacy.
 * sujira 'clear (fl)' ; e.g. jiqoris sujira 'cloudless sky'
 * sujiqoi 'clear (in)' ; e.g. leit sujiqoi 'cloudless day'
 * sujir 'clearly/safely' ; e.g. ixa sujir lu 'I slept soundly'
 * kamût hanut 'large rock'
 * hanut kamût 'boulder'
 * zumal kamût 'blue rock'
 * *kamût zumal '*rock blue'

Adjectival Meaning
While adjective agreement is mostly just that - agreement - the fact that the classes are differentiated semantically allows different modes of the same adjective to pick out slightly different nuances of meaning, which do not correspond 100% to the English meaning.

Temporal adverbs
Temporal adverbs are inflected as either future or non-future and must also take specific places in the phrase depending on their agreement. In this way they can be considered somewhat similar to aspect or tense markers, since they are time-anchoring agents. The root forms of all temporal adverbs must begin with a vowel to allow for this process.

Some temporal adverbs may also function as adjectives (including non-temporal adverbs) but in this case they usually convey spatial information.
 * yimîn 'soon'
 * teimîn 'just now/ recently'
 * imîn 'near'

Vocabulary
Swadesh list entries are based on best available semantic counterparts; for example, the animal referred to by kavoa 'dog' is not actually the mammal we know on Earth, but simply a domesticated carnivorous pack animal kept for hunting purposes (which also happens to be an amphibious crustacean).