Yipta

General information
Yipta, meaning roughly "correct speech" or "well-formed speech" is a language isolate spoken by a number of paleolithic hunter-gatherer bands living in a region of gently-rolling hills and oak woodlands. Its speakers have traditionally relied heavily on acorns as a staple food, but population pressure and increasing competition between bands over productive oak groves has led many to rely more heavily on smaller seeds, including wild varieties of wheat, barley, and peas. These people are unknowingly poised on the verge of the neolithic revolution.

Phonology
The phonemic inventory of Yipta is quite small, including 11 contrasting consonant phonemes and 4 vowel qualities, with no phonemic contrast in, length, stress, tone, or phonation.

Allophony
Major sound rules in Yipta include the following:

/s/ is pronounced [s] before or after /a/ or /o/ and in clusters with /p/, /t/, and /l/. It is pronounced as [ʃ] before or after /i/ or /u/ and in clusters with /q/, /k/, /j/ and /w/. However, when multiple sibilants occur in the same word, they assimilate to the place of articulation of the final sibilant in the word, regardless of their immediate environment

There are a number of phonological processes that affect the stop phonemes /p, t, k, q/ in consonant clusters. Sequences of a glottal stop and a non-glottal stop are realized as the ejective stops [p', t', k', q'], while sequences of two identical stop consonants are realized as a single aspirated stop [pʰ, tʰ, kʰ, qʰ], except in the case of a sequence of two glottal stops, where the cluster simplifies to a single glottal stop.

Non-stop consonants behave similarly, becoming the glottalized allophones [s', ʃ', ˀm, ˀn, ˀl, ˀj, ˀw] in clusters with glottal stops and simplifying to clusters of the form [Ch] when geminated. I.e. /ll/ is pronounced [lh], /mm/ is pronounced [mh], etc. [h] in these clusters is a voiceless glottal transition with little to no friction at the place of articulation.

Between a vowel and another nonidentical, pre-glottal stop consonant, /p, t, k, q/ are realized as the non-sibilant fricatives [ɸ, θ̠, x, χ]. In intervocalic clusters of two stop consonants, certain consonants are "stronger" and more resistant to spirantization. The hierarchy, from "strongest" to "weakest" is as follows: t, p, k, q. So, between vowels, the cluster /pt/ would be pronounced [ɸt], while the cluster /pq/ would be pronounced [pχ]. A cluster like /yqt/, on the other hand, would be pronounced [jqθ̠], because /t/ is between a stop and a vowel, while /q/ is not.

Vowels are lowered and retracted before and after /q/: /i/ surfaces as [e], /u/ as [o], /a/ as [a̠] or [ɑ], and /o/ as the "over-rounded" back-open vowel [ɒ̹]. Vowels also display a degree of lowering and/or retraction in proximity to /k/, but how noticeable this is depends on the speaker.

Phonotactics
In general, the syllable structure of Yipta can be represented as (C)(C)V(C)(C), with some restrictions. A syllable must always have an onset or a coda; it may or may not have both, but no Yipta syllable consists of only a vowel.

Consonant clusters within syllables may consist of two dissimilar stops and/or nasals, a stop or nasal and an approximant, or a stop or nasal and /s/. Clusters involving approximants always obey the sonority hierarchy, such that one finds the syllable /tya/ but not /yta/ and the cluster /tna/ but not /nta/. /s/ patters more freely with stops in syllable-internal clusters, such that the syllable /spa/ contrasts with /psa/.