The Cupbearer Coast (sometimes known by its Play name, Fayuva) is an area which in the 4100s came under the control of the speakers of the Play language, as several waves of mutually hostile invaders all spoke this language. Thus even when power changed hands several times in one century, the language of the population did not change. However, the Play speakers never fully drove out the languages of the indigenous inhabitants, such as Oyster, Middlesex, and North Dreamlandic.
The native population of Mipatatatai, also known as Tata, spoke the Baywatch language. But as the Dreamers, the Players, the Crystals, the Matrix, the Zenith, the Soap, the Tinkers, the Slopes, the Cold Men, the Mirror, the Phoenixes, the Dolls, the Cupbearers, and the Storms sent their armies into Tata to fight for control of the territory and adjacent Baeba Swamp, more languages appeared.
If both labialized consonants and prenasalized consonants are analyzed as clusters, the phonology would be
Bilabials: p m b Alveolars: t n s l r
And the vowels /a e i o u/ in both short and long forms. The geminates /pp ss tt/ still remained, and there were sound gaps of */te so bo lo ro/, except in a few rare words where contraction of long vowels before geminates had created new short vowels there.
The coronal stop /t/ was allophonically [k] before any /o u/. Unlike the neighboring Dolphin Rider language, however, it remained [t] before /a/, and therefore [t] is considered the primary allophone. It was not palatalized before /e i/.
Cosmopolitan Play languages
This is likely within the Baeban Play dialect continuum, despite the sharp political boundary between Baeba and Tata.
It is required that the Play speakers become the majority in Lypelpyp no later than 6800 AD, and the traditional writeup requires them to be the majority already by 5500 AD, though this timeline relied on settlement of Play speakers coming from Paba (where the birthrate was high for thousands of years) rather than from the Baeba region.
Try to find the string halpa in open74 dictionary, as it may refer to the original Cupbearer party name.
The Plup River (a place holder name) may be the route by which Play speakers left Baeba Swamp and moved northeastward into what had been an Oyster/Cappini patchwork, but since they had lost much of their power, they would not have been able to become the superstratum as they had along the coast. Any Play languages spoken in this area would have been fragmented and of no higher status than those of the aboriginals. Nonetheless, this is the most logical route to Lypelpyp, rather than relying on old roads.
Note that this is not the river that flows north into Olansele, though it is likely that the Play speakers would have control of both at least upstream. This river forms the southern border of Poise, but from there turns to the east instead of flowing into Poise's northern neighbor, Olansele. This means that any Play speakers living in Poise would be cut off from the coast by Moonshines.
It is possible that this chain of languages is derived from Blossom, which is a middle step in another chain, meaning that one chain intersects another and that the early dialect map is two-dimensional.
Daughter languages here may be given names related to alcohol. This is not world-internally canonical because the speakers referred to themselves with ordinary tribal names. Rather it is sort of a weak double pun .... these languages were spoken near Moonshine and the speakers were also involved in the bottling and distribution of wine from warmer climates into cold regions such as Moonshine; they also consumed alcohol themselves.
Wineapple speakers almost certainly settle the offshore islands as well, including "Wipep"; they may have also settled Sakʷalo and been later pushed back by Moonshine speakers. This was the language spoken in Olansele.
This branch may have actually begun diverging before 3370, but is given as such in the assumption that the sound changes in the first few centuries where Baywatch and Cappini had lost contact were minimal enough to be ignored.
THere may have been a marginal /u/ + vowel construction, left over from Dreamlandic ū + vowel.
Cappini (≈3370) to Wineapple (3958)
- The alveolars t n shifted to k ŋ before any of /o u/. This included the geminate forms.
- Any singleton s shifted to h unconditionally. The specific sequence /hu/ may have been pronounced as IPA [ɸu].
- Voiceless stops were lenited according to a two-step rule:
- When occurring before a singleton voiceless stop, the voiceless stops p t k shifted to f s h. (Thus, this is a "FiSH language".) This new /f/ would not have contrasted with the allophonic [ɸ] from the shift above, however.
- When occurring after a voiceless stop (whether single or double), the voiceless stops p t k again shifted to f s h. These changes left the language with a sparse distribution of singleton voiceless stops, so it could be said that any remaining /p t k/ geminated to /pp tt kk/ for ease of reckoning.
- The short o vowel shifted to a. /ō/ may have then shifted to /o/, but it was still spelled as a full /ō/.
- The geminates pp tt kk ss shifted to p t k s.
- The voiced bilabial stop b shifted to v. This did not include the rare sequence /mb/.
- The clusters mp nt ns nk shifted to mb nd z ŋġ. Then, the prenasals mb nd ŋġ shifted to the simple stops b d ġ in word-initial position. Note that there was a very small amount of preexisting /mb/ that was not from /mp/.
- The sequences ia ua came to be pronounced ye wo.
- Before a vowel, the sequences fu hu merged as f.
- Before a vowel, the sequences si hi merged as š, while ni ŋi merged into ň. Likewise ti di zi became či ǯi ži.
- In word-initial position, facing a voiceless sound across a short vowel, the voiced stops b d ǯ ġ devoiced to p t č k.
- Between vowels, except post-tonically, the voiceless fricative h disappeared to Ø. Any preceding obstruents became voiceless (this may mean that distinctions like mb/mp became phonemic again).
- Any remaining h now shifted to š if bordering a front vowel in either direction, to f if bordering a rounded vowel in either direction, and otherwise (the sole remaining environment being /a-_-a/) disappeared to Ø, creating a long vowel /ā/ from the resulting hiatus.
The consonant inventory at this stage was
Labials: p b m f v w Alveolars: t d n s z l r Postalveolars: č ǯ ň š ž y Velars: k ġ ŋ
It may be that all consonants were palatalized before /i/, and that the postalveolar row can be reanalyzed as palatalized alveolars.
The vowel system may have been /a e i u ā ē ī ō ū/.
Even in 3958, the speakers were already in contact with Moonshine, and Moonshine may have influenced the evolution of the Wineapple languages. Wineapple could not influence Moonshine, however, because the standard dialect of Moonshine was that of the interior, whose high population growth continually refreshed and buried the outlying dialects of the border regions.
Possible paths for vowel evolution are:
- /e/ > /ə/, /ē ō/ > /ai au/, thus imitating Play, though there would be no */əi əu/.
- Conditional /i u/ > /ʲ Ø/, followed by conditional /e ē ī ō ū/ > /i i i u u/, probably followed by loss of at least one intervocalic consonant. This also imitates Play, but would make more sense if Play were peripheral to the area and that Wineapple was really being influenced only indirectly.
Cultural evolution of Olansele
The Play speakers became the dominant culture in the region after about 4220 AD, but they were likely not strong enough to create a Play-speaking state and therefore Olansele was at least bilingual and possibly trilingual (Play, Wineapple, Oyster).
On the other hand, it is also possible that the Play speakers seceded, taking the best land for themselves, and that the inland areas of Olansele formed a much poorer state which was economically dependent on the Play-dominated coast.
A third possibility is that the Play speakers never truly established themselves along the coast of Olansele, and though they may still have dominated sea trade, they did not control the land and therefore there was no daughter of Play that was spoken in Olansele, meaning that Wineapple was dominant after all.
Wineapple (3958) to Craft Brew (4800)
The consonant inventory of Wineapple was
Labials: p b m f v w Alveolars: t d n s z l r Postalveolars: č ǯ ň š ž y Velars: k ġ ŋ
- In posttonic position (nouns were stressed on the penultimate mora, while verbs were variable), the vowels e i u shifted to Ø ʲ ʷ. Long vowels could not be posttonic, so after this shift, the only vowel allowed in posttonic position was /a/.
- In immediate pretonic position, the vowels e i u usually collapsed to Ø ʲ ʷ in a manner precisely identical to the previous shift.
- Note that many of the relevant words were compounds in which this shift was automatically applied because the first element of the compound had a free form in which the involved syllable was posttonic. However, the shift failed to occur if the pretonic syllable stood alone. It also did not occur when the first consonant of the tonic syllable was a voiced stop, as these were still pronounced prenasally. It may also have not occurred before a voiceless stop, since these were historically geminates. This would require a preexisting shift, however.
- It may be that new pseudo-suffixes of /e i u/ were generated by this shift, since e.g. some roots would have shapes like /bep/ as nouns and /bepi/ as verbs.
- Likewise, because of the noun classifier prefixes, this shift allowed the creation of monosyllabic verbs beginning with clusters, corresponding to nouns that were either CVC[a] or CVC.
- Palatalized labials passed their coarticulation to a following consonant if there was one, and then depalatalized. This likely also occurred for labialized consonants.
- It is possible that unstressed /o/ shifted to /a/.
- The high vowel u shifted to o if an adjacent tonic syllable contained a mid vowel.
- In unstressed position, the mid vowel e shifted to i if an adjacent tonic syllable contained /i/ (but not /u/).
- All consonants other than labials became palatalized before any /i/.
- All short e, even in stressed syllables, shifted to i if any other syllable contained a high vowel. Note, however, that this shift was quite rare, as all of the /i/ in classifiers prefixes and most other unstressed syllables had been lost by this time.
- The sequence rʷ shifted to w.
- The labialized nasals mʷ nʷ ŋʷ (and any others) merged as mm between vowels and to a single m otherwise.
Because of the vowel reductions, all classifiers except those containing /i/ contracted into single consonants. These lost their function as classifiers, as they could no longer carry the rhythm of the original CV classifiers, and therefore became inseparable prefixes on certain words. For example, Baywatch pensele "shirt" became fzel, with the f- reflecting the clothing classifier prefix but no longer seen as such. Instead there simply came to be a lot of words for clothes that began with /f/.
In the interior region, perhaps in Poise, there may be a shift of p b f v > f p v b, although the /p/>/f/ shift would take place AFTER /b/ > /p/, and be highly conditional, so it would require /pʰ/ as a middle step. It could even be that /p/>/f/ happened at the same time that Wineapple was also shifting /p/>/f/, a shift which was similarly conditional. (On the other hand, consider that Wineapple may have done its shift under Oyster influence.) Later, the main dialect of this Oyster language would shift /v/ > /b/, meaning that apart from a few /f/'s, the shift would actually be p b f v > p p b b, which would make it resemble Play.
It is not clear which branch of the family is even the ancestor of the language of Lypelpyp, but it should evolve into a consonant inventory based on /p b m n ŋ l r/ by 5200 AD or so, with all other consonants marginal.
It is likely that Lypelpyp was settled through the nation of Yīspʷilinâ, whose entry on Crystals indicates that it is likely a duplicate name of some other nation, but none of the other nations listed fit the description well. The nation of Poise is also nearby and if Yīspʷilinâ did not have its own language it may have had a similar profile to Poise.
Note that the first Thunder government (3884 - 3915) ruled all Crystals out of the Empire, but nations like Poise and Yīspʷilinâ may have seceded from the occupied Empire even before 3884 due to previous conflicts, and therefore never fallen under Thunder control.
Baywatch (3370) to Tata-A (4800)
This is cladistically part of the Minor Lenian languages and was even in contact with some of them.
This should probably begin around 4100 AD instead of 3370.
Note that the substratum language had likely already lost its /f/. see Oyster_language#Proto-Oyster (~2050) to Birch (~3310). Note also that the four-vowel inventory /a i u ə/ was not found in this area at the time, so there is no great pressure on this language to follow its shift of /o/ > /a/ with a shift of /e/ > /ə/.
Birch (which is probably intelligible with Plume) had lost its inherited freestanding /š č/, and although new palatals had been created from consonant + /j/, these were fairly rare, so there is no great pressure on Tata's Dreamer dialects to develop palatalization either.
It is also possible that the mainline Baywatch dialects influence this one.
- The alveolar stop t shifted to k before any /o u/.
- The vowel o shifted to a.
- In word-initial position, the sequences pp mp tt nt shifted to p b t d.
- The sequences s ss shifted to h s. It is likely that /hu/ was pronounced much like [ɸu], and perhaps other allophones existed before the other vowels; because these allophones arose after the shift of /t/ > [k], though, there was no contrast before historical /a/ vs /o/.
- The sequences mp nt ns shifted to mb nd z.
- The mid front vowels e ē shifted to ya yā. It is possible that the shift was just to /a/, but note that Middlesex was spoken nearby and did a conditional shift without its /e/ previously being iotated.
- The sequences ia ua (from pre-Baywatch deletion of /w/) came to be pronounced ya wa. There may also have been a iu, which could shift to yu. Also, it may be that these new sounds were pronounced much like IPA [je wo], as they had come almost entirely from similar sequences.
At this point, the Play army invades Tata and drives out the competing adstrates, meaning that influences from Crystal and Oyster cease to be significant.
- The long vowel ō shifted to ū. This was with Play influence. It is possible that it broke to /au/ instead, as Play was shifting /əu/ rather than /au/.
- Labialization was defeated. This left a new true /o/ vowel.
- All vowels became palatalized before /i/. This may have created a new phonemic /e/. However, Play's dialect continuum terminated in a three-vowel language, so it is possible that even these new vowels soon disappear.
- probably with [fs]