Play substratum languages

See User:Soap/history.

The Play substratum languages are a branch of the Lenian languages, also known as the Dreamlandic languages, spoken by people who sailed in the opposite direction and ended up with a smaller but more bountiful piece of land to live on as compared to the Dreamers who won control of a large span of land that had abundant seacoast and good fishing but was mostly too dry for agriculture.

It is possible that the "PDP" languages will be added somewhere in here, as their sound changes are no longer appropriate for their location in Dreamland now that the history of Dreamlandic has been shifted back 800 years. PDP could even be the new Thaoa.

They can be called the Pàndu languages, but this is simply a mnemonic term of convenience that happens to begin with the same letter as Play and Pabappa. The native cognate of this name is fianku and neither this name nor their full name begins with or contains the letter p. Likewise while the people may have used the name Puňanupi for themselves, this is not a tribal name and is here only because it's reminiscent of the sound of the Play language which they eventually ended up speaking.

Although Late Andanese speakers lived only where Play speakers lived, and although they were a minority who were eventually absorbed, the Players always considered them to be equals, and not as a substrate to stand on.

Early evolution

Shape of the proto-language

The phonology could be analyzed as

                       CONSONANTS                 VOWELS

Rounded bilabials:     pʷ  mʷ      w              a   u  
Plain bilabials:       p   m      (Ø)             a i u              
Alveolars:             t   n   s   l   r          a i u               (/ta ci cu/)
Postalveolars:         č   ň   š   λ   ř          a i   ə             (/ča či čə/)
(Palatals:             ć   ń   ś   y)             a i   ə
Velars:                k   ŋ   h                  a i u
Labiovelars:           kʷ  ŋʷ  hʷ                 a   u

Though it is possible that PSL had followed Nuclear Dreamlandic in shifting the schwa vowel /ə/ to merge with /a/. The list up above showing it occurring after labialized consonants is probably a mistake, and in fact it only occurs after palatalized consonants. (If it is not a mistake, it is talking about the extremely rare /ʷe/ of the proto-language.)

A minimal phonology incorporating the early loss of /f h/ is

p  m   
t  n  s
k  ŋ   
r     l

a  i  u  ə

Naming

This overrules anything I have written elsewhere (cant find it now).

A likely cover term for all of these people is Yaři, the name of a river running through the western area of the territory. But the tribes within the group had colloquial names for themselves and for each other, mostly beginning with labial consonants exemplifying the sound of their language. This was because they were surrounded by people who spoke languages without such a sound. The Yaři considered this a positive trait and were proud of it, but Yaři remained the official name of the whole group of tribes and they did not seek to replace this with an "iconic" name such as Puňanupi above.

Proto-PSL (500 AD) to Slime Forest (~2500)

Alternate names: New Thaoa This is the replacement language for Thaoa. There will be one standard language and a crop of dialects spoken further north and east. The maturation date here is a guess based on the fact that this list was originally about the same length as one that ended around 2500 AD.

Despite the very short list, this is accurate for 2,000 years because this branch happened to be very conservative. Its daughter language was also very conservative. There may have been divergent dialects even before 2500 AD however (remember Tuq, etc when there was Thaoa).

The voiceless non-sibilant fricatives f fʲ h hʷ shifted to Ø Ø Ø w.
The palatalized labials pʲ mʲ shifted to p m.
Before any of /a ə u/, the sequences ky ŋy ly shifted to t n l. (That is, every vowel but /i/.) It also happened, even before /i/, when an [u] (including /au/) preceded.
Any remaining ky ŋy ly then shifted to č n̆ l.
cu > tu.
kw ŋw w > p m Ø .
ř š > r s. A bare /y/ probably still existed but could be analyzed as an allophone of /i/.
ku ŋu > Ø (or maybe /t n/), but sometimes turns the following consonant into a dorsal.
Consider using schwa instead, even though it would leave a hole less likely to be repaired by a later shift.
Long vowels before hiatus became short. This likely requires phonemicization of /y w/ as in Shining Wave.
ču ňu > ku ŋu.

Vowels are /a i u ə/ and consonants are

 Bilabials:       p   m               
 Alveolars:       t   n   s   l   r 
 Postalveolars:   č   n̆   y
 Velars:          k   ŋ

Slime Forest (2500) to Andano-Thaoa (4185)

Alternate names: New Palli

This branch is extremely conservative. The maturation date is accurate even if the start date is moved back by a few hundred years. See below for an alternative; it is unlikely that both languages exist.

The voiceless fricative s shifted to h unconditionally. There was no /f/, so there was no staircase shift.
Geminates became simple. Sequences of nasal + stop became simple stops as well, except that a preceding w r y h shifted to m n n ŋ. (Note: Andanese did NOT do this the same way.) It may have also lengthened the preceding vowel, but vowel length soon disappeared anyway except possibly for /aa/.
The postalveolars č n̆ y shifted to s n Ø unconditionally. ("The snow shift.")
The flap r shifted to l.
The schwa vowel ə shifted to i.
The vowels i u before a hiatus shifted to y v, but this was no longer phonemic.

This language thus shared its phonology with Late Andanese. Its grammar, which had originally been like that of Dreamlandic, also evolved towards the Late Andanese model, and though they did not borrow particles directly, the overall layout of the grammar was similar enough that speakers of both languages considered New Palli to be essentially Andanese with a different vocabulary.

Slime Forest (2500) to Play-Thaoa (4185)

Alternate names: New Palli

This is the same as the above, but the language evolves to the phonology of Play instead of Late Andanese. This is perhaps more likely than the above; it is unlikely that both languages exist.

The clusters mp nt ňč ŋk shifted to b d ǯ ġ at the beginning of a word and to mb nd ňǯ ŋġ word-internally. This is the same shift that Shining Wave had earlier done, and this may have been due to Shining Wave influence, but that language was rapidly being pushed out by Play.
The voiced stops d ǯ ġ disappeared between vowels. This may have only come up in loanwords, and by this time the language was already being influenced by Play instead of Shining Wave. The voiced stop b remained.

Proto-PSL (500 AD) to Shining Wave (1900 AD)

This is the dialect of the capital city. It is monolithic and with little dialectal variation for at least the first 1,400 years of its existence because the central government is very strong. After 1,400 years, the central government is the first to switch to speaking Gold, thereby freeing the outlying areas to develop dialects. But most of these areas soon also switch to Gold and therefore the language becomes the substratum of the developing Play language.

The voiceless non-sibilant fricatives f fʲ h hʷ shifted to Ø Ø Ø w. Even this /w/ disappears later on.
The palatalized labials pʲ mʲ shifted to p m.
Before any of /a ə u/, the sequences ć ń li shifted to ṭ ṇ ḷ. (That is, every vowel but /i/.) It also happened, even before /i/, when an [u] (including /au/) preceded.
NOTE that the shift is given differently in Slime Forest for the sake of convenience; Slime Forest had dentals too, but they unconditionally shifted later on to alveolars whereas in this language there was a conditional shift.
Also, prenasals like mp shifted to mb and the like even when there was no second voiced consonant. (This is one way in which this language differs frm Pabappa.)
Prenasals in absolute word-initial position wore down to plain voiced stops.
When padded by at least one additional voiceless sound, the voiceless stops p ṭ t k (but not /s/) strengthened to the geminates pp ṭṭ tt kk. Note that /s/ can trigger this shift, but did not participate in it.
Between vowels, the singleton voiceless sounds p ṭ t s š k shifted to b ḍ d z ž ġ. This includes coarticulates. The voiceless affricate /č/ was not affected by this shift, as it was now behaving as a cluster /tš/.
All geminate stops (but not nasals) became singletons. It is possible that the fricative /ss/ also shifted to a plain /s/, and was replaced by a shift from the affricates.
The dentals ṭ ṇ ḍ ḷ shifted to t m b b when not facing another labial, and otherwise to t n d l. Labiovelars did not count as labials for the purpose of this shift.
The sequences ći ńi ġyi shifted to ši ňi ži.
The affricates c cc merged as s, which in initial position or after a closed syllable shifted to s.
The affricates č čč merged into š, patterning as above.
Before any "hard" consonant, the syllables pu bu mu shifted to t t n, which assimilated to the following consonant. Then, voiced geminates like bb dd ġġ became voiceless.
The sequences kw ŋw ġw w shifted to p m b Ø.
λ > y.
ə > u.
iw uw > ū; iy uy > ī.
Long vowels before hiatus became short. In the dictionary the /i u/ are still spelled as "iy, uw" for clarity, but there is no /w/ in the language.

At this point, reached around 1900 AD, AlphaLeap invaded and introduced the Gold language, which began to drive out Shining Wave since Gold was the language of the sea. (NOTE: AlphaLeap was at this time cooperating with Nama, and Nama was probably the stronger power at the time. AlphaLeap was not really empowered to do land invasions, and so they may have merely patrolled the sea, while perhaps Nama sent some soldiers in to occupy the land. Nama had agreed, in any case, to use AlphaLeap's trade language, which was Gold.)

Inventory

The vowels were /a i u/. The consonants were

 Bilabials:      p   m   b    
 Alveolars:      t   n   d   l   r   s   z
 Palataloids:        ň       y   ř   š   ž
 Velars:         k   ŋ   ġ

All syllables could end in a vowel, /t/, or /n/, and that vowel could be long even if there was a coda. There may need to be a separate /w/, just like /y/, because vowel sequences like /ia ua/ can occur. It may also be necessary to distinguish /ii/ from /ī/ and so on, but it is unlikely that /aa/ and /ā/ are distinct.

Unlike most languages, /b d ġ/ are true voiced stops, but like Play with its sole voiced stop /b/, these voiced stops pattern like approximants (or fricatives), and therefore /ž/ could be placed in the same column for the sake of tidiness.

Notes on possible further sound changes

The proper name Duřinuunma shows that this language will either have a contrast between long vowels and double short vowels (like Play), or, if /uu/ > /ū/, it will have superheavy syllables (also like Play).

Earlier there were clusters /ss šš/, but because singleton /s š/ were so rare, I shifted the geminates to singletons as well. This could revert back if /z ž/ are never generated.

Note that the Purses opposed Thaoa (that is, Slime Forest) on linguistic grounds and therefore their language cannot be subsumed by Slime Forest. However also note that there may be a separate language for the "Wings" who were pro-Slime Forest politically but were ethnically closer to the Purses and perhaps identical to the Purses. Therefore, the Wings either spoke Purse or had a language of their own that was similar to it.

It is likely that the early sound changes here will be the same as those of Wings, and so they could be combined into a sort of parent language even if they did not think of themselves as sharing a language.

Neither group was much influenced by Play, the language of education, because they had little contact with the people near the capital for the first 700 years, and somewhat hostile relations afterward (for different reasons than Slime Forest, however). Play influence may have crept in towards the very end however.

It is likely that both languages will be quite conservative, and this may be an areal feature with Slime Forest.

Shining Wave (1900) to Purso-Yařabic (2980)

The starting consonant inventory was

Bilabials:      p   m   b    
Alveolars:      t   n   d   l   r   s   z
Palataloids:        ň       y   ř   š   ž
Velars:         k   ŋ   ġ

And the vowels were /a i u/. Long vowels were rare.

All coda nasals lengthened the preceding vowel (a i u to ā ī ū) and then disappeared.
All remaining geminate consonants collapsed to singletons. Thus the language became fully CV.
The double vowel sequences aa ii uu shifted to ā ī ū.
If the long vowels produced above had been nasalized on their second mora, that would now also spread to these new long vowels and to those that had been inherited from the parent language. It is likely that a new sound shift will delete some consonants and allow for the creation of new vowel sequences that will contrast with the long vowels. At this point, the language has many word-initial vowels but few internal vowel sequences. The inherited ones were mostly /ai au/ because /i u/ had both been worn down before vowels.
In word-initial position, the voiced stops b d ġ shifted to p t k. Likewise z ž in initial position became s š.
Consonants were deleted. Remember that prenasals had created long vowels, so v̄v was common but vv̄ was rare.
1. Before a vowel, the sequences ib ub shifted to w.
2. All remaining b ž ġ shifted to ɣ. (This shift is very similar to Play's, but Play did this around 3100 AD, so it is probably just a coincidence.)
3. The sequences aɣ āɣ shifted to a ā .
4. The sequences iɣ uɣ shifted to y w.
5. Remaining ɣ was deleted.
6. The sequences īi ūu shifted to ī ū (except possibly in absolute final position).
Any consonant occurring after the vowel sequences /ai āi ī ūi/ became palatalized. It could be said that hiatus was also palatalized, since /īa/ and the like were phonetically [īya] as in most languages of the region.
Any consonant occurring after the vowel sequences /au āu īu ū/ became labialized. The above comment about hiatus also applies here.
The sequences ai āi au āu all merged as ā. Thus the new labialized consonants only occurred after long vowels, and some of the palatalized consonants were also bound there.
All remaining d dʲ shifted to r ř.
The vowel sequences īu ūi shifted to ī ū except at the end of a word.
The palatalized labials pʲ mʲ shifted to p m.
The palatalized coronals tʲ nʲ lʲ rʲ sʲ shifted to č ň λ ř š . Thus two new consonants were created.
The palatalized velars kʲ ŋʲ shifted to č ň.
The labialized consonants kʷ ŋʷ shifted to pʷ mʷ.
The labialized consonants tʷ nʷ lʷ rʷ sʷ shifted to pʷ mʷ w b f.
The labialized consonants ňʷ yʷ řʷ šʷ shifted to mʷ w b f. There was probably no /λʷ/ or /čʷ/.

The consonant inventory at this time was

Rounded bilabials:    pʷ  mʷ             (w)    
Spread bilabials:     p   m   f       b
Alveolars:            t   n   s   l   r        
Postalveolars:        č   n̂   š   λ   ř  (y)       
Velars:               k   ŋ

And the vowels were still /a i u/, but now there was a contrast between long vowels and sequences. Palatalization (CʲV) was distinct from the sequence CiV, which was in turn distinct from CʲiV. And these two were themselves distinct from CīV and CʲīV, though these more complex sequences were uncommon. This manifold distinction is unlike Play. However, most of the contrasting environments are ruled out by sound changes even though they are phonetically valid.

A possible further shift is

The labialized consonants pʷ mʷ shifted to p m. Note that /pʷ/ was confined to occurring after long vowels, so /mʷ/ was the only "free" labialized consonant.

Shining Wave to Wings

Many of the sound changes here are shared with those of Purse. The most likely date for the split is around 2980 AD. This is not a politically significant time; rather, it is about 300 years after the Oyster War, meaning that after their homeland was destroyed, they still spoke mutually intelligible languages for about 300 years despite not having close contacts with each other. Only after this period of time did the language barrier become noticeable.

Proto-PSL (633 AD) to Proto-Western PSL

This branch is more diverse than the others, fracturing early. It can be compared to Slavic languages if Capital PSL is Russian and Eastern PSL is Siberian.

In the west there are three languages, only diverging around 1700 AD, and therefore a single language is sufficient to cover the territory up until 1900 AD. This also includes several more languages that were spoken inside the Shining Wave territory, which also did not diverge until around 1700 AD. These "accessory" languages survived primarily because the Shining Wave language that had dominated the territory up until then was squished by Gold, which then evolved into Play and became a marker of identity. WPSL speakers were those who did not adopt the Play-speaker identity (still not the same as the Play political party, but rather what was once called "Pubu").

These languages might delete final short vowels, or at least final short -i, making them look more conservative than their own parent language because they then restore the final consonants on all of their verbs, and possibly also nouns, which previously had only appeared when they were tailed with a suffix -(C)i. An alternative is that it is the much rarer schwa that gets deleted, but that this language somehow replaces the verbal suffix -(C)i with -(C)ə before the deletion.

Contact languages

The Gold language pushed out both PSL and the Fern languages. See also Tarwas and Tropical Rim.

Nik

The Nik language is a rare and perhaps unique instance of migrants not coming from the "Mumba" area of the larger island. These people may thus have been linguistically separated from the rest for 50,000 years or longer. Their language therefore may be effectively unrelated to all others. But it is possible that they learned a new language as they migrated, or that the migrants belonged to a class who had already learned to speak MRCA before they left their island.

Rasula

There were also dark-skinned settlers in Rasula. They may even have been there before the Lenians were, though this is perhaps unlikely, as it would be odd for them to bypass the much better land around the aboriginals' main settlements to fish along the coast. If they did arrive early, they could have a third branch of MRCA/DRM all to themselves. Otherwise, they will either be part of the Lava Bed clade, thus being Tropical Rim (but not within any other branch), or else early colonists who left AlphaLeap, which under the new writeup was an original settlement and not founded only in 1088. The latter plan would make their language part of the Tropical Rim/Extension.

The aspirated velar stop k became č before the vowel /i/. If another vowel followed, the /i/ disappeared. This happened even if the /i/ was accented.
When a "velaroid" consonant (/k ḳ ŋ h g l/) followed an accented high tone vowel, the vowel metathesized, leaving a closed syllable. Thus, for example, /àli/ > /ail/. These closed syllables were all high-toned, and are thus written without tone marks.
A schwa before another vowel in any syllable disappeared. Thus əa əe əi əo əu əə shifted to a e i o u ə. This happened in both open and closed syllables.
The double-vowel sequences aa ee ii oo uu əə shifted to the single vowels a e i o u ə in closed syllables only.
Andanese influence could shift tʷ dʷ nʷ > k ġ ŋ, and perhaps also ḳ > q. Thus the language sounds quite guttural, and the Play substratum speakers dislike the sound while the Rasulans think of the pre-Play population as babyish.
An early shift of ki > č(i), shared with Gold, might in this language also be followed by more palatalizations, but the ejective was immune as were all labialized consonants (this is why they became plain velars later).
i ə > ʲi i.
e o > ʲɜ ɜ.
stressed ɜ > a if another /a/ is in the word.
pʲ bʲ mʲ > f v m. /f v/ probably shift to /s d/ later on, as in Gold, implying they might have retained a palatal articulation.

Orthography

The PSL proto-language had the same simple alphabetic (only 14 glyphs) as did proto-Dreamlandic, but when the Andanese arrived, they adopted the Andanese syllabary in a series of reforms. First, they used their own native glyphs for single letters and the Andanese syllables for the CV syllables, with the native glyphs also used for silent letters that appeared in the grammar. That is, the word for man ended in a "silent /l/" that only appeared in the grammar and not in the bare form of the word. This was comparable to French liaison but it did not participate in cross-word liaison, only appearing in in-word inflections. Later, the native glyphs were^probably dropped because even the Andanese syllabary had glyphs for single consonants. (It is possible however that even then, the native glyphs were retained as grammatical markers of some kind, since they were simpler.)

History and cultural conflicts