My project, Kaujasas. Questions, critiques solicited!

[matsu]

I've been working on this language, name settled on Kaujasas, for over a year. It is supposed to be what is most logically pleasing to me in a language, with all of the functions laid out as I'd like them. But as I participate in this forum more and more, I find more issues with grammar, word formation and so on that I'd not previously considered. So I'm looking for third party eyes to take a look at this. Here is a brief (compared to my work-in-progress book) summary of the grammatical functions:

Background

Kaujasas is an Indo-European language, at least in its preference for creating words; it prefers to borrow German words, and with Finnish as an "aesthetic" prestige language (mostly because the phonosyntactic rules are nearly identical between Finnish and Kaujasas). It's an SOV language, with a nominative-accusative alignment (considering changing to ergative-absolutive alignment), with a 23-letter Roman alphabet.

Orthography

Vowels


Consonants


There is no vowel harmony as Finnish has, but diphthongs only form when the otherwise non-useful vowel harmony would not be violated. Diphthongs don't form at morpheme boundaries.

Phonotactics

Native words are limited to the following structures: CV, CVC, CVV, CVVC, CVCC, VC, V, VV, VVC and VCC. Consonant clusters are not allowed; this doesn't count glides. Geminate vowels an consonants are phonemic. Most consonants can geminate; the only vowel that cannot is /ə/. Stress is penultimate.

Syntax and sentence structure

Kaujasas is strictly left-branching. Adjectives come before nouns, adverbs before adjectives and verbs, relative clauses before noun phrases, and so on. Postpositions are used instead of prepositions, and subordinating conjunctions come at the end of their clause. The language is a subject-prominent language.

There is no wh-movement; question particles are used instead.

Word classification

Words don't strictly fall into European categories. Nouns, verbs, adverbs etc. are generally the same, but adjectives blur the lines. They are divided into predicative and attributive adjectives; predicative adjectives act as verbs and eschew a copula. Instead of modal verbs (e.g. "could", "should", etc.), Kaujasas has modal particles which follow the verb. (Does this violate a strict left-branching approach?)

Nouns and pronouns

There are three inflections for number: singular, plural, and collective.

There are no cases per se; Kaujasas uses a system of postpositions, of which case markers get included. Those that are not purely genitive, dative or accusative use postpositions. The two can't be combined. Non-case postpositions are almost all divisible into two segments: direction of movement (or lack thereof) + location. For example, the postposition (e.g. "ellean") is divisible into the morphemes elle- (movement away) and -an "near".

Pronouns distinguish between exclusive and inclusive "we". Pronouns can take adjectives as well. (E.g. Hylë he ilke os vöhai on cytsere. clever-ADJ 2;SG self GEN book-PL ACC protect-HAB-PRS "Clever she protects her own books.")

Demonstratives

Often adverbs, demonstratives are mostly formed by adjectives and pronouns/adverbs. For example, wh-questions are mostly composed with vë (adjective form of ve, "what"), proximals with sitë "this" medials with satë "that", etc.

Verbs

Verbs do not directly reflect indicative or subjunctive moods; moods are reflected with modal particles instead, which follow the verb. Verbs are inflected for transitivity, aspect and tense; they are not inflected for person.

Verbs distinguish between transitivity, unaccusative intransitivity and unergative intransitivity. I'm considering lumping transitivity and unaccusative intransitivity into an ergative, and unergative intransitivity into an absolutive. I don't know much about this alignment though, so I'm uncomfortable making the change, but nominative-accusative isn't working for me.

Aspects include: neutral (bare), perfect, progressive, perfect + progressive, continuous, perfect + continuous, and habitual. The tenses are past, present and future.

Verbs also take suffixes for frequentative, momentane, continuative, intentive, extentive, performative, factive, causative, translative, and resultative aspects. This is a direct inspiration from Finnish.

With prefixes and inflections, a verb can look like this:
töhine -> elletöhine -> elletöhinie -> elletöhinehtie ->
elletöhinehtoie -> elletöhinehtoiäie ->
elletöhinehtoiäiesë ->
elletöhinehtoiäiesëjë ->
elletöhinehtoiäiesëjë so
to take -> to take off (undress) -> to undress (unerg.) -> to quickly undress ->
to quickly and purposefully undress -> to quickly and purposefully undress time after time ->
(I) am quickly and purposefully undressing time after time (prog.) ->
(I) have been quickly and purposefully undressing time after time ->
Were it that (I) have been quickly and purposefully undressing time after time...

Adjectives

Adjectives have a comparative and superlative. Attributive adjectives use a copula and do not decline; predicative adjectives do not use a copula and do decline, but the neutral aspect retains the -ë adjectival ending.

Adverbs

Adverbs, even adverbs of time, are always marked with -o. If an adverb is pluralised, the plural marker comes before the adverb ending.

Particles

Borrowed from Japanese and Cantonese, there are question particles, particles that express surprise, excitement, etc.

[Kereb]

Kaujasas has modal particles which follow the verb. (Does this violate a strict left-branching approach?)

Not at all. You'd expect auxiliaries (or a particle that does similar) to head their verb phrases and so to occur at the very end.

[Kereb]

Native words are limited to the following structures: CV, CVC, CVV, CVVC, CVCC, VC, V, VV, VVC and VCC. Consonant clusters are not allowed; this doesn't count glides.

what is allowed for those CC up there then?

[kodé]

Cool stuff! Here's a few comments.

There is no vowel harmony as Finnish has, but diphthongs only form when the otherwise non-useful vowel harmony would not be violated. Diphthongs don't form at morpheme boundaries.

Are there any height restrictions on diphthongs? Are both rising and falling diphthongs allowed? What about same height diphthongs (e.g. [yi])?

Native words are limited to the following structures: CV, CVC, CVV, CVVC, CVCC, VC, V, VV, VVC and VCC. Consonant clusters are not allowed; this doesn't count glides. Geminate vowels an consonants are phonemic. Most consonants can geminate; the only vowel that cannot is /ɕ/. Stress is penultimate.

Do you mean words or roots? If it's only for roots, can affixes take all these shapes too, or are they more limited? If you want a simpler description of allowable monosyllables, you could just say that onsets can be null or simple, nuclei can be short or long, and codas can be null, simple, or complex; also, superheavy -VVCC rhymes are disallowed (which makes perfect sense). Also when you say consonant clusters are not allowed, do you mean inside a syllable or across syllables too (also Kereb's question).

There is no wh-movement; question particles are used instead.

This is a little weird: most languages I've seen that both are wh-in-situ and have question particles don't use both simultaneously.

Words don't strictly fall into European categories. Nouns, verbs, adverbs etc. are generally the same, but adjectives blur the lines. They are divided into predicative and attributive adjectives; predicative adjectives act as verbs and eschew a copula. Instead of modal verbs (e.g. "could", "should", etc.),

Couldn't you just say predicative adjectives are stative verbs? Also, is there a semantic determination of which properties are expressed as attributive adjectives and which as predicative adjectives?

Adverbs, even adverbs of time, are always marked with -o. If an adverb is pluralised, the plural marker comes before the adverb ending.

What do you mean by a pluralized adverb? Is it that the subject of the verb is plural? Or does it make adverbs distributive?

[matsu]

Cool stuff! Here's a few comments.

There is no vowel harmony as Finnish has, but diphthongs only form when the otherwise non-useful vowel harmony would not be violated. Diphthongs don't form at morpheme boundaries.

Are there any height restrictions on diphthongs? Are both rising and falling diphthongs allowed? What about same height diphthongs (e.g. [yi])?

Here is a chart on the diphthongs:
The diphthong chart is virtually the same as in Finnish. It's my understanding that Finnish diphthongs are mostly closing, so [yi] is permissible. Admittedly, I don't know much about height restrictions on diphthongs. I should look into that!

Native words are limited to the following structures: CV, CVC, CVV, CVVC, CVCC, VC, V, VV, VVC and VCC. Consonant clusters are not allowed; this doesn't count glides. Geminate vowels an consonants are phonemic. Most consonants can geminate; the only vowel that cannot is /ə/. Stress is penultimate.

Do you mean words or roots? If it's only for roots, can affixes take all these shapes too, or are they more limited? If you want a simpler description of allowable monosyllables, you could just say that onsets can be null or simple, nuclei can be short or long, and codas can be null, simple, or complex; also, superheavy -VVCC rhymes are disallowed (which makes perfect sense). Also when you say consonant clusters are not allowed, do you mean inside a syllable or across syllables too (also Kereb's question).

You'll have to forgive me. Phonology really wasn't my strong point in uni, so I can't exactly understand your first question. Could you show me examples of what you mean in living languages? When I say that clusters aren't allowed, I mean for onsets (so no /pr/, no /tc/ (except in borrowings) and codas within a syllable. So for example, a word like *rusk would be illegal because there is a cluster for a coda, and a word like *prät for the same violation in the onset. But a word like perusta "foundation" is fine because even though there is the cluster /st/, it is occurring across syllables and not within.

There is no wh-movement; question particles are used instead.

This is a little weird: most languages I've seen that both are wh-in-situ and have question particles don't use both simultaneously.

Maybe I phrased it wrong... it's supposed to reflect off of East Asian languages, namely Japanese, e.g.:

• 私は学生です。 "I am a student."
• 私は誰ですか？"Who am I?"
• Lisneer ese. "I am a student."
• Vë lyt ese ka ? "Who am I?"

Words don't strictly fall into European categories. Nouns, verbs, adverbs etc. are generally the same, but adjectives blur the lines. They are divided into predicative and attributive adjectives; predicative adjectives act as verbs and eschew a copula. Instead of modal verbs (e.g. "could", "should", etc.),

Couldn't you just say predicative adjectives are stative verbs? Also, is there a semantic determination of which properties are expressed as attributive adjectives and which as predicative adjectives?

Yes, I could call them that, but I call them predicative adjectives so I can lump them in with the other adjectives and provide a place to explain their differences. Generally, a predicative adjective will be dynamic, and an attributive adjective will be static, unchanging, or immutable. Certain things could be semantically one or the other; I don't imagine "clad-iron" being lumped in with the predicative stative verbs, and "insane" can be either a temporary status or a permanent affliction, and therefore could be both. There's something of a connection to the dual "ser/estar" relationship of adjectives in Spanish, but I prefer to compare it to Mandarin 很+adj (predicative) v. 是+adj+的 (attributive). The distinction isn't always perfect in Mandarin, and I could suggest that the distinction isn't perfect in Kaujasas and carries exceptions, instead of completely acknowledging a logical error on my part.

Adverbs, even adverbs of time, are always marked with -o. If an adverb is pluralised, the plural marker comes before the adverb ending.

What do you mean by a pluralized adverb? Is it that the subject of the verb is plural? Or does it make adverbs distributive?

When I say plural adverbs, I mean specifically temporal adverbs. Finnish has a distributive temporal adverb system as well, where it uses the suffix -sin, e.g. arkisin "weekends". Kaujasas simply adds a plural to the noun root before the adverbial ending, so "a weekend" would be vohë, "on the weekend" voho; more than one weekend is vohëen/vohi, and "on the weekends" is vohëeno/vohio.

[matsu]

Native words are limited to the following structures: CV, CVC, CVV, CVVC, CVCC, VC, V, VV, VVC and VCC. Consonant clusters are not allowed; this doesn't count glides.

what is allowed for those CC up there then?

I might need to revise that to say native "syllables". I would imagine CC using something like /h/, so "maht.ta" (not defined) would be permissable. Kodé made a suggestion for redefining in simpler terms syllabic constraints that I will look into.

[Linguist Wannabe]

Looks a lot like there has been some areal contact with Japanese, Korean or the Altaic languages (if you believe that they form a valid language family). Is this intentional?

Words don't strictly fall into European categories. Nouns, verbs, adverbs etc. are generally the same, but adjectives blur the lines. They are divided into predicative and attributive adjectives; predicative adjectives act as verbs and eschew a copula. Instead of modal verbs (e.g. "could", "should", etc.), Kaujasas has modal particles which follow the verb. (Does this violate a strict left-branching approach?)

Japanese is a prototypical strict left-branching language, and it forms its equivalent to the English "should" construction by placing stuff after the verb e.g.
taberu - eat
tabetara ii - should eat.
taberu kamo shirenai - might eat.

[matsu]

Looks a lot like there has been some areal contact with Japanese, Korean or the Altaic languages (if you believe that they form a valid language family). Is this intentional?

I don't think they form a family, but that aside, the contact isn't areal. The background of my language is an alt. history/sci-fi sort of thing, and Kaujasas serves as a fantasy auxlang. The features are intentional though; I studied Japanese for three years, so I'm very comfortable borrowing grammatical notions from it now.

Words don't strictly fall into European categories. Nouns, verbs, adverbs etc. are generally the same, but adjectives blur the lines. They are divided into predicative and attributive adjectives; predicative adjectives act as verbs and eschew a copula. Instead of modal verbs (e.g. "could", "should", etc.), Kaujasas has modal particles which follow the verb. (Does this violate a strict left-branching approach?)

Japanese is a prototypical strict left-branching language, and it forms its equivalent to the English "should" construction by placing stuff after the verb e.g.
taberu - eat
tabetara ii - should eat.
taberu kamo shirenai - might eat.

Right -- that's why I do it too. But I was never sure if that was actually a violation of left-branching or not.

My equivalents, by the way, are:

esse - eat
esse sajo (deliberative)/esse ino (exhortative) - should eat
esse kamo - might eat

[Linguist Wannabe]

Japanese is a prototypical strict left-branching language, and it forms its equivalent to the English "should" construction by placing stuff after the verb e.g.
taberu - eat
tabetara ii - should eat.
taberu kamo shirenai - might eat.

Right -- that's why I do it too. But I was never sure if that was actually a violation of left-branching or not.

My equivalents, by the way, are:

esse - eat
esse sajo (deliberative)/esse ino (exhortative) - should eat
esse kamo - might eat

If we consider the auxiliary verb to be the head of the phrase, then strict-left branching will always place it last.

Look at Greenberg's universal #16 (http://en.wikipedia.org/wiki/Greenberg' ... universals). "In languages with dominant order VSO, an inflected auxiliary always precedes the main verb. In languages with dominant order SOV, an inflected auxiliary always follows the main verb."