Cosmopolitan Play languages

The Play party spoke a language they called Bābākiam and spread it to the territories they conquered. Soon, dozens of rival parties also spoke Bābākiam, including the Jokers, the Swamp Kids, the Flower Bees, the Corals, the Creamers, and the Dolls. At its peak, Bābākiam was the language of more than one third of the world's population. But the Players were the first people to spread it outside its original compact homeland in Paba, and every one of the other groups traced their association with the language to early contacts with the Players.

Bābākiam was the primary language of the Play party, so much so that both the Players and their rivals often referred to the language as Play. But the Players arose from a group of people with a long history of diglossia, meaning they spoke a second language, Late Andanese. Because of the nature of their society, both languages had survived side by side for more than 2,000 years, with little influence on each other despite the fact that many people spoke both. This situation is partly for cultural reasons and partly because the two languages were quite different. Only a few traits characterized both languages; for example, Bābākiam and Late Andanese both had very small phoneme inventories, and both lost their tone contrasts at about the same time.

Scratchpad

Possibly revive the voicing shift from Yeisu Kasu, Mevumep, or wherever it was. that language did not have /b/, so the shift made perfect sense there. here, there would be both an inherited /b/ and a new /b/.

Invasion of Tata

In 4138, the Play party invaded Dreamland and joined their conquest to the Play-held state of Tata. Their language was identical with Bābākiam. Later, the Swamp Kids invaded Tata a second time, and even though they opposed the Players, their languages were the same, and the invasion helped cement the role of Babakiam as the primary language of Tata and adjacent areas near (but not in) Baeba Swamp.

Note that the language of Tata arose from a blend of two different dialects; even though the invasions were only 70 years apart, they had originated from different areas. Even so, the differences were mainly in vocabulary, since the phonology of both dialects was exactly the same and the grammar of both dialects was flexible enough that speakers of both had no trouble listening to others.

The Tataans borrowed words from the indigenous Dreamlandic languages, which had by this time diverged widely even from each other. They borrowed relatively few words from the non-Dreamlandic Lenians, however.

Invasion of Amade

In 4140 (approximately), the Play party invaded Amade but did not completely subdue the locals. But the Firestones who invaded Amade in 4162 also spoke Babakiam, and they completely replaced the native languages. The Firestone branch may have had a greater survival of Late Andanese vocabulary than the Players and Swamp Kids in Tata. Very few of the words originating in Dreamland reached either of the groups in Amade.

This might be the "vuonah" language.

If, as stated on Tropical_Rim#Later_history, the Firestones conquered more than just Amade, their invasion woul have a great and indelible change on the demographics of the area, as they were very different in appearance from the dark-skinned natives, and reached 85% of the population within just two years. It is unlikely that the tribes around them would ever push the descendants of the Firestones back out.

The Firestones had light skin and typically dark hair, but some had brown hair or even shades of red and blonde. As above, they were a majority in their area, so even as they mingled with the dark-skinned tribes around them, they changed little over the next 4,500 years, and did not feel out of place in their tropical habitat. They were probably the shortest people in the area, however, as the short aboriginals of Kxesh, Atlam, and the Star Empire had been replaced by taller tribes that had immigrated in slow waves over the preceding years. This would have slowed the rate of intermarriage somewhat, although it may have conversely brought the Firestones closer to the few remaining true aboriginal tribes.

Baeban Play dialect continuum

The consonant inventory of Play was

Bilabials:    p   m   b   f  (w)
Alveolars:    t   n       s   
Palataloids:          ž   š  (y)
Velars:       k   ŋ

The vowels were /a i u ə ā ī ū/, but double vowel sequences were common and could contrast with single long vowels.

This branch could equally well be derived from the Swamp Kids' Bābākiam as spoken in 4206. The Swamp Kids settled further south initially, but they were pushed together and in fact the Swamp Kids ended up enslaving the descendants of the Players.

This language was spoken in both Tata and the Cupbearers' republic of Xalmana. At one point, Xalmana governed all of Tata, but Baeba Swamp governed all of Xalmana.

The Cupbearers may be an important source of population for Cosmpolitan Nama, and therefore could be a substratum for the Poswobs who arrived 1,400 years later. It is even possible that the Cupbearers joined the aboriginals and came to be seen as part of the Repilian tribal assortment.

IT IS LIKELY THAT CUPBEARER, BOTTOM, BLOSSOM, TRUE SWAMP, AND ANYTHING SPOKEN POLEWARD OF TATA ARE ALL PART OF THE SAME DIALECT CHAIN EARLY ON, EVEN IF POLITICALLY HOSTILE, AND CAN HAVE A SHARED PROTOLANGUAGE WITH A BREAKUP TIME AS LATE AS 4800 AD.

A large group of Play-speaking Doll slaves broke free in the year 4217, and may have survived recapture.

These dialects are ordered from east to west.

Play (4138) to Next Step

This is the language of the slaves who switched sides and became abusers in Tata. A female military leader named Šasuasa, who also spoke Play but was born in Creamland, invaded Tata to rescue the slaves. Šasuasa made significant progress but was herself overthrown in 4221. Then, the Dolls who had been abusing the Matrixes in Tata fled to a safe spot at some distance from Tata. It is not clear where this is.

Play (4138) to Šasuasa's dialect

The Matrix state was so crippled by enemies on all sides that even with immense numerical superiority it is unlikely they would have been able to resist invasion. All that matters is whether the invaders who swept into Tata spoke Play or some other language.

Play (4138) to Quest

A possible fifth language for the descendants of the Dolls who lived in nature and considered themselves more physically and mentally hardy than the Cupbearers, Pacifists, and Bottoms.

This is the only Play language in Baeba Swamp whose territory is monoethnic, but it is only in fringe areas, mostly towards the north, with poor natural resources. Here they were safe mostly because they lived on the least desirable land. Meanwhile, other Quests moved around in search of a peaceful place to live. Thus there may be two or even more languages. On the other hand, the migratory Quests would likely have adopted the languages of the nations they settled in even if they preferred to remain culturally separate even there.

Play (4138) to Cupbearer (4800)

The political separation date of this dialect is 4206 AD. It is the northernmost dialect that is within Baeba Swamp proper.

The Cupbearers were willing slaves, who attempted to submit to the Swamp Kids, even though the Swamp Kids repeatedly slaughtered their Cupbearer slaves.

1. Long vowels in initial syllables became double: ā ī ū became aa ii uu. This shift did not happen if the long vowel was supported by another following vowel. Meanwhile double vowels in final syllables became long: aa ii uu shifted to ā ī ū. Medial syllables followed morpheme boundaries.
2. Any bʷ žʷ created by the previous shift changed to b. Likewise any bʲ žʲ changed to ž.
3. The vowels i ī shifted to ʲi ʲī, except after velars in native words. (Velars did not occur in Dreamlandic in this position, but some languages had a written "k" for /tš/.)
4. Any remaining prevocalic i u shifted to ʲ Ø. The allophones of labialized consonants thus came to occur in new environments.
5. The vowel sequences ə əi əu shifted to i ī ū. These were not palatalized.
6. The cluster mt (and loaned Dreamlandic /nt/) shifted to r.

   Consider making this shift extremely restricted, such as /mtu/ only before /i/, so as to preserve the structure of native words while allowing the borrowing of the foreign words. /žu/ is another possiblity.

7. Dreamlandic's word-initial geminates became integrated into the phonology of Cupbearer by first admitting an epenthetic schwa (which did not occur initially in native Play words) and then deleting that schwa.

   Remember, though, that Baywatch having these geminates in 3370 AD does not mean that forks of it still had them in 4600 AD etc.

8. The clusters pn mn sn all merged as nn. Then pm sm shifted to mm, and pŋ sŋ mŋ became ŋŋ.
9. The clusters ps pš shifted to tt before any palatalized vowel. Then pt shifted to tt unconditionally.
10. The clusters ms mš shifted to nt (not /r/) before any palatalized vowel.
11. The cluster pk shifted to pp before any palatalized vowel, and otherwise to tt. Then mk shifted to nt.
12. The voiceless fricative f shifted to h.

The language at this stage had three vowels: /a i u/, and unlike Blossom to the south, did not have labialization. Long forms /ā ī ū/ of all three vowels were common, as were diphthongs.

The consonant inventory at this time was

Bilabials:          p   m       w   b
Spread labials:     pʲ  mʲ
Alveolars:          t   n   s  (l   r)
Postalveolars:      tʲ  nʲ  sʲ (lʲ) ž
Palatals:           kʲ  ŋʲ  hʲ  y    
Velars:             k   ŋ   h

Names of daughter languages could be along the lines of "Hungry Plants", etc.

Play (4138) to Blossom

The language of the United Pacifist League. If the dialects are smeared out, this language could actually be identical to Cupbearer, and function as a political name only.

This language is spoken in the lowlands of Baeba Swamp, immediately south of the Cupbearer territory.

This group eventually ended up speaking Poswa, but the language came from Paba whereas the political party arose thousands of miles away in Baeba Swamp.

This language may borrow many words from Playwatch.

1. Long vowels in initial syllables became double: ā ī ū became aa ii uu. This shift did not happen if the long vowel was supported by another following vowel. Meanwhile double vowels in final syllables became long: aa ii uu shifted to ā ī ū. Medial syllables followed morpheme boundaries.
2. Any bʷ žʷ created by the previous shift changed to b. Likewise any bʲ žʲ changed to ž.
3. The vowels i ī shifted to ʲi ʲī, except after velars in native words. (Velars did not occur in Dreamlandic in this position, but some languages had a written "k" for /tš/.) Then u ū became ʷu ʷū.
4. The vowel sequences ə əi əu shifted to ɜ i ʉ. This created a new /ʲʉ/, but /wʉ/ instead became /ʷu/.

   This was originally written with the outcomes of /ə/ and /əi/ swapped, even though it was counterintuitive. This is because the shifts did not happen at the same time.

5. The sequences f fʷ shifted to hʷ. Then sʲ š šʲ merged as sʲ.
6. The voiced bilabial stop b came to be pronounced bʷ.
7. The sequence fʲ shifted to hʲ. Thus /f/ was removed from the phonology.
8. Foreign loans with /l r/ were integrated as in surrounding dialects by matching them with uncommon native sounds and then adapting to the proper pronunciation.

Note that the language at this stage retained the original four phonemic vowels, /a ɜ i ʉ/, all of which could be preceded by /ʲ/ or by /ʷ/.

The consonant inventory at this time was:

Bilabials:        p   m   bʷ     (w)
Alveolars:        t   n   r   s   l
Palataloids:              ž   šʷ (y)
Velars:           k   ŋ       h*

Consonant coarticulations exist for all stops and nasals, and for the voiceless fricatives /s h/. The outlier /šʷ/ was rare. There was no plain /h/.

An expanded consonant table can be presented:

Rounded labials:        pʷ  mʷ  bʷ  hʷ  w
Plain labials:          p   m    
Palatalized labials:    pʲ  mʲ
Close alveolars:        tʷ  nʷ      sʷ
Alveolars:              t   n   r   s   l
Postalveolars:          tʲ  nʲ  ž   š   lʲ
Retroflexes:                        šʷ
Palatals:               kʲ  ŋʲ      hʲ  y
Velars:                 k   ŋ
Labiovelars:            kʷ  ŋʷ      

Blossom has a richer phonology than its neighbors to the east and west; the dialects with five vowels have more restrictions on those vowels than Blossom with its four free vowels.

Blossom Reunion changes

Note: these sound changes may be out of order, and this is important, because they involve chain shifts such as pk mŋ > pt mn > tʷ nʷ > kʷ ŋʷ, all unconditional.

Chain A (up to 4800 AD)

1. The sequences pk mŋ shifted to pt mn.
2. The sequences ptʲ mnʲ shifted to kʲ ŋʲ .
3. The sequences pt mn shifted to tʷ nʷ.
4. Then šš šʷ merged as ʂ. Similarly, ss sʷ merged into ṣ.
5. The postalveolar fricatives š ž shifted to h g.
6. The labialized alveolars tʷ nʷ became kʷ ŋʷ.
7. The geminates pp mm shifted to pʷ mʷ.

Chain B (after 4800 AD)

1. Palatalized labials depalatalized.

Petals

It is possible that whole separate languages could appear for guilds, such as the soap guild, the bookbinders' guild, etc so long as the guild arises from a specific area with a specific natural resource.

Play (4138) to White Pants (c. 5200)

Note, this language was originally named Cupbearer South because it belongs to the Cupbearer tribe, but it is not descended from the main branch of Cupbearer languages.

The language of Kēk. It may be that the Cupbearers who fled into Kēk remained stranded there, and came to speak a different language. The ending date may be too late.

1. Long vowels in initial syllables became double: ā ī ū became aa ii uu. This shift did not happen if the long vowel was supported by another following vowel. Meanwhile double vowels in final syllables became long: aa ii uu shifted to ā ī ū. Medial syllables followed morpheme boundaries.
2. Any bʷ žʷ created by the previous shift changed to b. Likewise any bʲ žʲ changed to ž.
3. Word-final p s shifted to the vowel length marker ː. Word-final m shifted to n, pronounced as a nasal vowel offglide.
4. The sequences ss sš sž shifted to ss šš žž; these are geminates functioning as single consonants. /žž/ may have been used to absorb loaned /l/ and possibly /r/, before the speakers later adopted the proper pronunciations.
5. Any remaining s in the syllable coda shifted to the vowel length marker ꞉.
6. The vowel sequences ai au əi əu shifted to e o ē ō.

   Check to make sure this is not a mistake; note that it's the opposite of the usual order. The bare schwa is also not yet handled.

7. The clusters pn mn all merged as nn. Then pm shifted to mm, and pŋ mŋ became ŋŋ.

   This may allow word-final /p/ to be rescued from merging with /s/, since it would remain distinct before a word beginning in a nasal, except in situations where it would also be in conflict with original word-final /m/. Note the front-loading of syllables took place early and that these word sequences would have been pronounced as if they had prefixes instead.

8. The clusters pk mk shifted to tt nt.
9. Any remaining codas assimilated to a following consonant, although the syllables were front-loaded as in Dreamlandic languages.
10. The velar stop k shifted to tš (as if from /pš/) before any front vowel or glide.
11. Through grammatical replacement, the velar stop k often shifted to t in word-initial position (from codas), to p after primordial stressed syllables in transparent compounds, and to Ø after a long vowel. Remaining k was then deleted.

The consonant inventory at this stage was

                  PLAIN                        CLUSTERS

Bilabials:        p   m   b   f   w            pp  mm
Alveolars:        t   n   r   s   l   c        tt  nn  ss  nt  ns
Postalveolars:            ž   š   y   č                šš
Velars:               ŋ                            ŋŋ

The vowel inventory was probably /a e i o u ā ē ī ō ū/.

Coarticulations were not significant in this dialect, and /si š/ could still form a minimal pair. The geminate consonants parallel the distribution of labialized consonants in eastward dialects.

1. Before a vowel, the sequences pu mu tu nu su šu ŋu shifted to pp mm tt nn ss šš ŋŋ. These were considered as single consonants.
2. The geminates pp mm tt ŋŋ shifted to pʷ mʷ kʷ ŋʷ. The other geminates remained as such.
3. Before a vowel, the sequence si shifted to š.

Play (4138) to Cupbearer West

The Cupbearers who fled Kēk soon found themselves in trouble again and attempted to exit Baeba Swamp altogether. They were captured a third time, and they were held under such tight control that they became the largest group and were no longer found only in the west.

LATER CHANGES

After about 4800, it may be that Baeba's growing influence causes the Play languages to develop tones and a CV structure, and that these changes proceed in waves originating from Baeba Swamp, with eastern Play languages showing weaker influence. It is even possible that the languages fuse into a dialect chain and then split apart again along different lines, meaning that some of the languages have more than one parent.

Also note the Dreamlandic influence in White Pants, not present even just a few miles eastward; this is because, while the dialects were in contact, they crossed an important political boundary.

1. Long vowels came to have a new tone, spelled with the circumflex tone, as with â.
2. In the coda, ɜm shifted to um.
3. The alveolar flap r merged into l.
4. Word-final s p disappeared, and changed the previous vowel to a high tone. One dialect (probably Cupbearer) had already given up its word-final codas, meaning that there were two possible reflexes. But note that this tone occurred ONLY in final position.
5. Late Andanese loanwords came to have a mid tone on their final syllable, with all other syllables becoming low tone.
6. Native Play words came to have a mid tone on the first syllable of their root, except where recent changes caused shifts, and all other syllables became low tone.

Play (4162) to Firestone (6843)

The maturation date is of no linguistic significance, but rather of political significance. Late Andanese contributes much vocabulary.

The consonant inventory of Play was

Bilabials:    p   m   b   f  (v)
Alveolars:    t   n       s   
Palataloids:          ž   š  (y)
Velars:       k   ŋ

The vowels were /a i u ə ā ī ū/.

1. Long vowels in initial syllables became double: ā ī ū became aa ii uu. This shift did not happen if the long vowel was supported by another following vowel. Meanwhile double vowels in final syllables became long: aa ii uu shifted to ā ī ū. Medial syllables followed morpheme boundaries.
2. The double vowels ii uu became ʲi ʷu in all positions.
3. Any bʷ žʷ created by the previous shift changed to b. Likewise any bʲ žʲ changed to ž.
4. Between two consonants in a single syllable, the diphthongs au əu əi changed to ō u e. iu ui ii uu > ə ə i u. The change was bypassed whenever a consonant cluster was frontloaded onto the next syllable, however.

   NOTE, this shift is also in Poswa and Pabappa's list, but does not seem to have happened, except when there was a third component to the vowel nucleus. That is, /iui/ > /ʲə/, etc, rather than just /ui/ > /ə/.

5. ā aa changed to aba in all positions.
6. The sequences hi hih shifted to ś, except when the whole of the syllable was stressed.
7. The fricative sequences f sf changed to h h.
8. The fricative sequences ž sž shifted to l s.

   Note, this was recently edited after a long break.

9. The sequences mn mt pt shifted to nn nt tt.
10. All coda -p generated a high tone on the preceding vowel and then disappeared. Note that this reflex might become synonymous with gemination.
11. Before a vowel, the vowel sequences iy uw (however spelt) shifted to ī ū. These did not occur in Pabappa.
12. Before a vowel, the sequences ti si ni li shifted to č š ň λ. Note that this only occurred in open syllables. The lambda is the same as /ł/.
13. Before a vowel, the sequences tu su šu nu lu shifted to tʷ sʷ šʷ nʷ b. Note that this only occurred in open syllables.
14. Before a vowel, the sequences ki kih ŋi shifted to ć ć ń . They may have shifted to a dental almost immediately when before any vowel other than /i/.

    Note that there were still minimal pairs like kʲiw "plain" vs kʲu "mountain range", so this shift is important.

15. Before a vowel, the sequences ku ŋu hu shifted to kʷ ŋʷ hʷ. Note that /hʷ/ was not IPA /f/ and that the nasal was nasal all the way through.
16. This triggered a further shift of kuh huh to kʷ hʷ. Also, kuś huś (as from Late Andanese /kuhihii/ "sword") shifted to ćʷ śʷ.
17. All native Play words with no heavy syllables took the tone pattern MLL, and Andanese loans took the tone pattern LLM (there were no heavy syllables in Andanese).
18. The stress shifted to the highest mora in the word. All remaining closed syllables were high, and the first mora of a long vowel was high.
19. Any short i was backed to ɨ after any velar or labial consonant. Long /ī/ did not shift.

    It is not clear if this is distinct from /ə/ or not. Note that this shift is what keeps /ki kii/ from merging the way /hi hii/ had.

20. When occurring as a FALLING diphthong, the sequence iu shifted to ʷɨ̄. This could only occur in an open syllable after a consonant plus a glide.
21. The labialized palatals ćʷ ńʷ shifted to dentals ṭ ṇ. This did not affect postalveolars, nor did it affect the palatal fricative.
22. Syllable-final m came to be pronounced as vowel nasalization, but was spelled as if it were final n.
23. When on a low tone, the sequences sa si su ha hi hu collapsed to s š sʷ h ś hʷ even before consonants now. Some grammatical affixes were reinforced by using a longer form containing a second vowel. (This had happened even in Late Andanese.) Then, hʷɨ joined the shift as the postpalatal fricative ɦ; any śʷ likely also shifted to ɦ. These shifts did not become phonemic in final syllables (which only came into play in Play words), but in casual speech before another word, these vowels were dropped without affecting the phonemic shape of the words.
24. A series of rapid, interdependent sound changes involving contraction of CVC sequences into clusters:

    The sequences pk mŋ shifted to pp mm.

    The sequences pɨs pɨš pɨt pɨl bɨl shifted to ps pš pt pl bl.

    The sequences pus puš put pul bul shifted to psʷ pšʷ ppʷ ppʷ bbʷ.

    The geminates may attract stress to the preceding syllable since they are likely to be analyzed as sequences of a high tone and a singleton consonant. This may require a shift back to singletons for environments where this is not possible.

    The sequences mɨl mɨn mɨŋ shifted to ml mn mŋ. Then mul mun muŋ became mmʷ mnʷ mŋʷ.

    The sequences tɨl kɨl shifted to tl kl. Then tul kul became ttʷ kkʷ.

Consonant clusters were mostly front-loaded onto the trailing syllable. Unlike Play languages of the north and east, however, nasals could not be front-loaded; indeed the only truly phonemic coda in the language is /n/, and a word cannot begin with a nasal-stop cluster.

Note also the prosodic distinction between fortis nasals /mm nn ŋŋ/, which pattern like ordinary consonants but can only follow high tones, and the newly created geminates /mm nn ŋŋ/ which pattern like clusters and can occur in any tone environment. The former may be spelt as single m n ŋ because the high tone mark will be sufficient to distinguish them and because there were no simple lenis nasals after a high tone.

Play (4108) to Tadpole (~5547)

Because the Tadpoles were native to their territory, they may not have shared the sound changes such as /bʷ žʷ/ > /b/ that characterized all of the other branches including the branch that stayed behind in Paba. Even if the voiced consonant changes are included, it is still possible to omit the vowel changes, meaning that the full Play vowel inventory could remain.

The consonant inventory of Play was

Bilabials:    p   m   b   f   w
Alveolars:    t   n       s   
Palataloids:          ž   š   y 
Velars:       k   ŋ

The vowels were /a i u ə ā ī ū/.

Unlike all other dialects of the language, Tadpoles still had the sequences bʷ žʷ bʲ žʲ, which had elsewhere shifted to /b b ž ž/.

Early changes

1. A schwa ə in a word in which the following syllable had /a/ changed also to a.
2. The labiovelar glide v came to be spelled w.
3. The original stress accent, in which verbs were stressed on the final syllable of the root, was restored. (This is an entry of convenience; in reality, it had never changed in this dialect.)
4. The sequence žʷ shifted to gʷ.

   In theory, the shift in the other branches could have been žʷ > gʷ > bʷ > b, since it incorporates all preexisting /bʷ/ aswell, but this would imply that it was not simultaneous with the shift of /bʲ/ > /ž/ in those branches, which could not possibly have gone through a stage with /l/.

5. The sequence bʲ shifted to a lateral l.

   Note that this means that the Tadpole dialect has a handful of native words with /l/, corresponding to /ž/ in all other branches.

6. The sequences fʷ šʷ merged as hʷ.
7. The diphthongs ai au əi əu shifted to ē ō e o. Plain schwa remained as such.

The consonant inventory of Tadpole was

Bilabials:    p   m   b   f   w
Alveolars:    t   n       s   l
Palataloids:          ž   š   y 
Velars:       k   ŋ       hʷ  gʷ

The vowels were /a e i o u/, which could be either short or long, plus a high mid vowel /ə/. The only codas were /p m s/.

Superheavy syllables in words like šaippa "mermaid" were fairly common; essentially long vowels behaved the same as short vowels. Likewise, long sequences of vowels could occur as in kaāatapa "coral reef".

Andanese loans were relatively few, but higher than in Pabappa.

This stage of the language may be around 4800 AD.

Free changes

These are the shifts that were not in contact with other dialects.

1. The clusters pt mt mn shifted to tt nt nn.
2. The clusters pm pn pŋ shifted to mm nn mʷ.
3. The clusters mŋ pk shifted to mʷ pʷ.
4. When occurring as a falling diphthong, the sequences iu ui īu ūi merged as ə̄. It did not affect the rising forms. Note that the distribution was similar to that of the other early Play languages, in that /ivi/ was [ʲuj], and so on. Thus, the new /ə̄/ vowel could only occur after an offglide, either /j/ or /w/, and it was colored by that offglide.

Culture

Play, including both Babakiam and Late Andanese, may have been the dominant language of the common people as far south as 20°N, where the old empires of Lobexon, Taryte, and AlphaLeap came together. This was "upper Taryte"; though still at a low elevation, it was on the north side of the river and therefore had come fully under Oyster control for a long period of time.

Alternatively, Tadpole settlement may have only reached as far south as 23°N, if it can be assumed that the sparsely settled desert area to the south of this had passed into the hands of some other power. This small detail is important because it may mean that there was no common border between the Tadpole and Firestone territories, even if the Firestones gained control of adjacent lands long after their initial war.

AlphaLeap would have considered this to be a part of Subumpam, and the border with AlphaLeap would have been primarily linguistic. To the south would live slaves firmly under Leaper control who spoke only the Leaper language, and to the north would be historically free people who spoke their own languages.

It is not clear how and why the Play languages replaced the Oyster languages in this area, however, as Paba could not have invaded them.

This area may in the long run come to be considered part of Nama, once Nama becomes fully integrated into the Poswob Empire.

The maturation date is of no linguistic significance, but rather of political significance. Late Andanese contributes much vocabulary.

Play in Creamland

In 4151, Creamland seceded from the Anchor Empire, and soon came to identify its people as a wing of the Play party. In eastern Creamland, Andanic languages and others may have persisted from ancient times, but the only language spoken in the capital was Play. It is possible that, after 4268, Creamland remains the only functioning nation outside Baeba's control and that its language splits slowly along geographic lines (not quickly, because the capital was still in control). Thus the Pabappa/Poswa branch would be just one of many branches, albeit by far the dominant one.

Alternatively, early Pabappa covers all of Creamland, and only begins branching after Poswa has, so although later Creamland has more than one language, those languages are more closely related to Pabappa than Poswa is. A third alternative would be to combine these situations, meaning that Creamland would have languages both closer and further away from Pabappa than Poswa is, and Poswa would therefore be just one of many typical branches of the Pabappa family, notable only because of its vastly greater territory compared to all others.

Play (4151) to Mr Òóple (c. 5000)

This is the first-branching form of the language. Sound changes should be identical or nearly identical to those of Pabappa, but greater retention of Andanese vocabulary is also a possibility. It is not clear if the Andanese here includes Late Andanese or if they were relic branches only.

This territory was still capable of invading neighboring nations as late as 4268, and was locally dominant even afterwards.

1. Long vowels in initial syllables became double: ā ī ū became aa ii uu. This shift did not happen if the long vowel was supported by another following vowel. Meanwhile double vowels in final syllables became long: aa ii uu shifted to ā ī ū. Medial syllables followed morpheme boundaries.
2. The double vowels ii uu became ʲi ʷu in all positions.
3. Any bʷ žʷ created by the previous shift changed to b. Likewise any bʲ žʲ changed to ž.
4. Between two consonants in a single syllable, the diphthong əu changed to u . In more restricted environments, also iu ui ii uu > y y i u. The change was bypassed whenever a consonant cluster was frontloaded onto the next syllable, however.
5. ā aa changed to aba in all positions.
6. At the beginning of a syllable and after /p m/, the semivowels w j shifted to r l.

   Though not marked in the history of Pabappa, it is likely that w j also shift to r l at the END of a syllable.

7. The medial clusters pt mt mn shifted to tt nt nn. Then pk mk mŋ became pt mpt mn . ms mš became mps mpš.

   The shift of /pt mt mn/ > /tt nt nn/ might need to be back-dated because it affects the way vowels were compressed. For example, Play /pk/ front-loaded onto a following syllable, but Play /pt/ did not. This would make more sense if the contrast was either /pk/ vs /tt/ or /pt/ vs /tt/.

8. The cluster sf changed to ff.

   In most words where an /sf/ cluster might be expected, only a single /f/ was found due to a much earlier shift that occurred in the Gold language. This had been maintained through analogy. Likewise, where one might expect /pf/, there was often just a /p/. The words where these clusters did occur were newly coined compounds.

9. The labialized alveolar stop tʷ shifted to pʷ in word-initial position or after one of /r l s/, and to pt between vowels.
10. The labialized consonants šʷ sʷ nʷ shifted to pš ps bʷ. There was also a rare word-initial /bʷ/, from earlier /bū/ > /bu/ > /bʷ/.
11. The consonant clusters mr ml shifted to br bl unconditionally.

    NOTE ON POLITICS: This branch will have very few loans from Leaper.

12. In initial position before a vowel, the voiceless labial fricatives f fʷ changed to w. It also happened often to a word-internal -f- preceded by a /w/ of any origin, but note that the sequence /fVf/ only appeared in words that were originally compounds. This shift did not affect fʲ.
13. The voiceless fricative f shifted to v between vowels. This still did not affect /fʲ/.
14. After a labialized consonant (except /w/), the schwa vowel ə changed to u.
15. The clusters pm sm pn sm shifted to mm mm nn nn.

At this point, the consonant inventory of the language was

Rounded bilabials:          pʷ  bʷ      fʷ  w
Bilabials:                  p   b   m   f   v
Labiodentals:                           fʲ
Alveolars:                  t       n   s   l   
Postalveolars:              tʲ      nʲ  š   ž  (sʲ)
Palatals:                   kʲ      ŋʲ      
Velars:                     k       ŋ       r
Labiovelars:                kʷ      ŋʷ

The vowel inventory is simply /a i u ə/ due to different analysis from Pabappa.

It is not clear which, if any, of the original /nʲ nʷ ŋʲ ŋʷ/ should shift to voiced stops. It is most likely that none of them shift, because in both Poswa and Pabappa, new /nʲ nʷ/ are created by a subsequent shift, and merge unconditionally with the primordials.

Play in Xema

The first Swamp Kids who moved to Xema were Andanese speakers who were cut off even from the other Swamp Kids and had nowhere else to flee to. Some later Swamp Kids moved to Xema, and these may have spoken Bābākiam, but it is unlikely their language would have survived for very long.

If this language exists, it is under strict control by Icecap Moonshine from a fairly early date.

Play along the Strand

It is possible that Play languages survive in the areas poleward of Tata, and may stretch as far as Moonshine territory. This is assuming that the Ghosts, Leapers, or other groups that speak Leaper do not take them over. If so, however, this should be subsumed under the Baeba dialect chain.