Tesazo is the language of a fictional land of mine which I've actually never done serious conlanging on. Figure I'd speedlang a little of that tonight. Trying to use a more colloquial style of presentation; I hope it works well!
/t' k' kʷ'/<t' k' q'u>
/pʰ tʰ kʰ kʷʰ/<p t k qu>
/f v s z ɣ h/<f v s z g h>
/m n ŋ ŋʷ/<m n ng ngu>
/l ɾ w/<l r w>
The voiced fricatives are very obviously derived from voiced stops. In a lot of dialects they are prenasalized, particularly intervocallically; in others they induce nasalization on preceding vowels derived from this tendency. Similarly, the velar nasals are largely loanwords, mainly occurring natively as the result of assimilation or avoidance speech. I make no apologies for making <u> go with <q> unless before <u>, <qi qe qa> etc. are not the phonaesthetic vibe.
/i i: ɛ ɛ: æ æ: ɑ ɑ: ɞ ɞ: ɤ ɤ: u u: ɨ ɨ:/<i ii e ee ä ää a aa õ õõ o oo u uu y yy>
The vowel values are still a little uncertain, but I like the weirdness of having the central vowel rounded while the back value is unrounded (/ɤ/ is probably [ɤ̞] in practice). Long vowels are only distinguished in stressed syllables and are barely phonemic (deriving from now-lost coda glottal stops).
Stress is either on the initial or second syllable, marked with acutes if it varies in derived forms. If stress is on the second syllable prettonic syllables undergo horrifying "vowel crushing."
/i/ /ɛ/ /æ/--> [e] <e>
/ɑ / /ɤ/ /ɨ/ --> [ə] <o>
/ɞ/ /u/ --> [u.] <u>
I'm cribbing from Bulgarian here. What's not so Bulgarian is that this happens in unstressed /Vɾ/ sequences when /ɾ/ is the coda; the vowels are reduced but also coalesce with /ɾ/ to form rhotacized vowels [e˞] [ə˞] [u˞]. (I might remove the vowel reduction here.)
It's not predictable where the stress falls except second syllable stress is illegal if it's the last syllable; if more syllables are added on the underlying stress pattern can resurface.
kälquir --> kelquíri
['kʰæl.kʷʰe˞] -> [kʰel'kʷʰi.ɾi]
"fly" --> "be thrown, be fired"
The syllable structure is permissive of pretty heavy initial clusters, but less friendly to coda clusters/codas in general; another slavicism. The syllable structure can be simplistically described as (C)(C)(S)V(C2), where C is any consonant, S is any sonorant, and C2 is any consonant except /f s w/. Additionally the coda clusters /nt' ŋk' ŋkʷ'/ are permitted word finally; i.e. ['mɛ.nink'] "woman"
Onset clusters have slightly complex rules:
- Like obstruents must go with like obstruents; thus *pʰt' *k'z *vtʰ are illegal. The first obstruent assimilates to the MOA of the second or the closest thing; thus these would be repaired as /pt'/ /ɦz/ /pʰtʰ/. This might cross morpheme boundaries, idk.
- Voiceless fricatives cannot begin a cluster unless it precedes /l r w/; thus *fm *st *sf are all illegal, only occurring at morpheme boundaries when they can be plausibly assigned to neighboring syllables.
-Liquids and /w/ cannot begin clusters at all, and clusters of *rl *lr aren't legal even if preceded by another consonant.
Allophony is boring aside from what's already been mentioned. Nasal+aspirated stop clusters see the stops get voiced and have breathy voice and the same happens intervocalically; nasals assimilate to following stops.
Tesazo
-
Creyeditor
- MVP
- Posts: 5294
- Joined: 14 Aug 2012 19:32
- Contact:
Re: Tesazo
I really like this phonology. The vowel system looks kind of strange in a Germanic kind of way but the vowel crushing makes it look like there is some hidden pattern somewhere. The consonant system looks nicely Anatolic but also generically intetesting. I also like the constraint based description of allophony and stress. I would be interested in seeing a list of legal consonant clusters (or anything on morphosyntax really).
Creyeditor
https://sites.google.com/site/creyeditor/
Produce, Analyze, Manipulate
1
2 
3 
4 
4 
Omlűt & Kobardon & Fredauon Fun Facts & AMA on Indonesian
![[<3]](./images/smilies/heartic.png "<3")
Papuan languages, Morphophonology, Lexical Semantics
https://sites.google.com/site/creyeditor/
Produce, Analyze, Manipulate
1
2 3 4 4 Omlűt & Kobardon & Fredauon Fun Facts & AMA on Indonesian Papuan languages, Morphophonology, Lexical Semantics Re: Tesazo
Since I got Janko'd, here's some notes on Tesazo numerals.
Tesazo uses a vigesimal system. The basic numerals between 1 and 20 are:
1. äälo
2. k'en
3. tatpal
4. suq'uõ
5. pswõ
6. er
7. hopt'äl
8. mpy
9. urt'eva
10. t'ev
11. kmu
12. rõvwi
13. vzläk'
14. fõmõ
15. za
16. levun
18. opurit'ä
19. engk'
20. wava
(lmk if I messed up the phonotactics at all; I was using a word generator and threw out those that didn't fit, but might have missed a few.)
Higher numbers continue to be formed vigesimally: k'en wava "forty", za wava õlevun "three hundred and sixteen" (the õ- prefix comes from õz "and;" it's õz- before a vowel. It's often dropped in colloquial language). 400 is het'.
Tesazo numbers have ordinal forms, which are regularly formed with the suffix -yz, e.g. t'evyz "tenth." If the numeral ends in a vowel, that vowel is deleted; pswyz "fifth." For a few numerals with two syllables, this results in a return of the stress to an underlying second position, resulting in a superficially irregular form; i.e. tatpal --> totpalyz. The numeral for one, äälo, has two ordinal forms; the regular elyz, and the irregular i.
There are also distributive forms, made by reduplication; levun-levun "sixteen each." Again, this causes vowel crushing, since the reduplicated forms operate as a phonological unit: fumõ-fõmõ "fourteen each."
Numerals precede nouns in the NP. They also must go with numeral classifiers:
k'en wäwäz tot'
two animal dog
"two dogs"
I'll develop these a bit more later.
Tesazo uses a vigesimal system. The basic numerals between 1 and 20 are:
1. äälo
2. k'en
3. tatpal
4. suq'uõ
5. pswõ
6. er
7. hopt'äl
8. mpy
9. urt'eva
10. t'ev
11. kmu
12. rõvwi
13. vzläk'
14. fõmõ
15. za
16. levun
18. opurit'ä
19. engk'
20. wava
(lmk if I messed up the phonotactics at all; I was using a word generator and threw out those that didn't fit, but might have missed a few.)
Higher numbers continue to be formed vigesimally: k'en wava "forty", za wava õlevun "three hundred and sixteen" (the õ- prefix comes from õz "and;" it's õz- before a vowel. It's often dropped in colloquial language). 400 is het'.
Tesazo numbers have ordinal forms, which are regularly formed with the suffix -yz, e.g. t'evyz "tenth." If the numeral ends in a vowel, that vowel is deleted; pswyz "fifth." For a few numerals with two syllables, this results in a return of the stress to an underlying second position, resulting in a superficially irregular form; i.e. tatpal --> totpalyz. The numeral for one, äälo, has two ordinal forms; the regular elyz, and the irregular i.
There are also distributive forms, made by reduplication; levun-levun "sixteen each." Again, this causes vowel crushing, since the reduplicated forms operate as a phonological unit: fumõ-fõmõ "fourteen each."
Numerals precede nouns in the NP. They also must go with numeral classifiers:
k'en wäwäz tot'
two animal dog
"two dogs"
I'll develop these a bit more later.
Re: Tesazo
Nouns in Tesazo are quite simple: they, technically, have no morphological marking at all.
However, the noun phrase more generally is fairly complex, so worth some discussion. The Tesazo noun phrase is largely left-branching, following the order of Numeral-Adjective-Noun-Demonstrative.
Numerals have been discussed already.
Adjectives are relatively simple in Tesazo. Like nouns themselves, they are uninflected and simply precede their following noun.
Demonstratives, unusually, follow nouns in Tesazo -- a regional feature. Tesazo demonstratives, of which there are three, cliticize to the noun as well.
igi -- proximal
nuki -- medial
naki -- distal
Possession is indicated with the particles le and er. The distinction between these is fairly nuanced. Le is canonically used with human or high animacy referents, while er is canonically used with inanimate objects and lower animacy creatures. However, each can be used, in a somewhat marked way, to indicate alienability. When used with human referents, for example, er implies a certain distancing that is specifically alienable -- this is especially dramatic with kin terms (which are often used with non-family affectionately), where it forces a reading of "we are not related." Meanwhile, le can be used with inanimate objects to express a kind of deep, existential connection that implies animacy -- i.e. someone who wears glasses would use le rather than er.
It's worth noting that in their canonical use, le and er do not have an inherent implication as to alienability, except when explicitly contrasted. Thus ktyna le amako could mean "the shepherd's mother," or "the shepherd's other female mentor or family friend" but ktyna er amako would force the latter meaning.
However, the noun phrase more generally is fairly complex, so worth some discussion. The Tesazo noun phrase is largely left-branching, following the order of Numeral-Adjective-Noun-Demonstrative.
Numerals have been discussed already.
Adjectives are relatively simple in Tesazo. Like nouns themselves, they are uninflected and simply precede their following noun.
Demonstratives, unusually, follow nouns in Tesazo -- a regional feature. Tesazo demonstratives, of which there are three, cliticize to the noun as well.
igi -- proximal
nuki -- medial
naki -- distal
Possession is indicated with the particles le and er. The distinction between these is fairly nuanced. Le is canonically used with human or high animacy referents, while er is canonically used with inanimate objects and lower animacy creatures. However, each can be used, in a somewhat marked way, to indicate alienability. When used with human referents, for example, er implies a certain distancing that is specifically alienable -- this is especially dramatic with kin terms (which are often used with non-family affectionately), where it forces a reading of "we are not related." Meanwhile, le can be used with inanimate objects to express a kind of deep, existential connection that implies animacy -- i.e. someone who wears glasses would use le rather than er.
It's worth noting that in their canonical use, le and er do not have an inherent implication as to alienability, except when explicitly contrasted. Thus ktyna le amako could mean "the shepherd's mother," or "the shepherd's other female mentor or family friend" but ktyna er amako would force the latter meaning.