Oyster language

See history for non-linguistic information.

Scratchpad

Standalone genitive

03:57, 7 May 2022 (PDT)

The Late Andanese genitive hi- is real, even if it turns out to not be cognate to the infix -əh-, which only becomes genitive in Play. Lava Bed languages would all inherit this, but as a verb, not a prefix or a particle. It is possible that in Oyster, it becomes the primary means of indicating a genitive relationship, just as in Late Andanese. Compounds with definite meaning would still omit the genitive, however, as they did in Late Andanese, Play, and nearly every other such language.

Because /hì/ is not /-ə̀h-/, it can itself take that infix, and the result would simply be /hì/ again because the schwa would both disappear and impart its tone on the remaining vowel, which was already high. Therefore the word is defective from the beginning. Presumably, the /-ək-/ infix still works, and the 3rd and 4th person suffixes still work, which would all be in patient mode most of the time. (Note: if Gold preserves it, it would be /his/.)

Oyster shifts /p b/ > /w/, so the suffixes for boy, woman, and teenager would all have this /w/, and it may even shift /w/ > /Ø/ through analogy with further suffixes containing labials. This would allow Oyster to evolve a system like Galà, only to lose it again as the grammar "breaks down". Still it is possible that Lava Bed morphology is preserved in Oyster in certain words such as this and is only discontinued for content words.

If Oyster preserves the suffixes, it would have words like hì-u "of the woman; not SAP", hì-a "of the boy; not SAP", etc., and the hiatus would remain until at least 1900. Later it probably just becomes /sŭ/, /să/, etc. But remember that it is required to mark BOTH participants on the verb, so it is not "of the boy" really but rather "of the boy; SAP/non-SAP/third-party" and so on.

Since this is a verb, it doesnt matter where in the sentence it appears. thus for example "book-X woman" and "book woman-X" both mean "the woman's book". But it is likely that one order becomes dominant in time.

Relation to other verbs

Remember that the order of morphemes was fixed as THIRD + FOURTH, but that each could be either agent or patient. It is most likely that when both morphemes appear, 3rd is most commonly patient and 4th is most commonly agent, since this puts the object closer to the root of the verb. Since the verb means "belong to; member (of)", the PATIENT is actually the object that is "bigger", that is, the possessor. This is the opposite situation as what appeared in Gold and Play.

Accordingly, the genitive particle is probably much like /š/ + GENDER VOWEL OF POSSESSOR + /b/ (if IDENTITY is animate) + GENDER OF IDENTITY (whole syllable, except vowel-only if animate). this can be repersented as

1. /š/
2. GENDER VOWEL OF POSSESSOR
3. /b/ (if identity is animate)
4. whole syllable of identity (but just a vowel if animate)

pep

Is there still a language anywhere that has CVC- classifiers? The Subumpamese family has been crushed out of existence, so it can't be the source of that.

Background

See Lenian_languages#Oyster.

A few FILTER people may have settled in baeba or southern Dreamland.

Some Oysters may have spoken Andanic_languages#Proto-Olati_.281300.29_to_Olati-A_.282672.29, from the South Andanic (Olati) branch. This family is very conservative, and shows many similarities with the languages of Atlam, even though Atlam is far away both geographically and genetically. West Andanic and Atla have lack of innovation in common, meaning that they resemble each other because they both still resemble their MRCA even after 2,000 years apart (long branch attraction). It is possible that the Oysters, so proud of their language, find kinship in Atlam even after they slaughtered thousands of soldiers in Lobexon and Kxesh (and perhaps even some in Atlam).

If the language is assumed to be identical to Olati-A (which it probably isnt), then the consonant inventory would be

Labials:     p   b   m   f   v       
Alveolars:   t   d   n   s       l
Palataloids: č   ǯ   ň   š       y
Velars:      k

And the vowel inventory would be /a e i o u/, on two tones, and a long series.

/ǯ/ may be a mistake, as there seem to be no paths to it in the sound changes. But perhaps it comes from /dʲ/, in which case it would be much rarer than /č/ but still phonemic.

The Oysters likely meet up with a group of settlers speaking a Southwestern Tip language. This group is the furthest west in the tropics, but also settled one area of land further north in the desert.

Both languages had simple syllables and fairly small phoneme inventories, but there were many differences between the two inventories. Southwestern Tip may have had only four vowels, likely /a e i u/, meaning it was unbalanced. Also, Oyster had three tones while Southwestern Tip probably had only two. The consonant inventories were quite different as well: Southwestern Tip had /m̄ n̄ ŋ̄ ŋ g h w/, all unmatched in Oyster; there was also a /b̲/, but it may have still been /mb/ at the time. Meanwhile Oyster had /p f v s č ǯ ň š/ all unmatched in Southwestern Tip. Most of Tip's coronals were palatalized before /e i/, however, so they could count as /č ǯ ň/ for transcription purposes.

If a unified script appears, they may consider /v/ and /w/ to be the same sound, as well as writing the palatals as coronal + /i/. Southwestern Tip apparently does not have /ke ki/, even at the phonetic level.

Note that Oyster is being compared with the Southwestern Tip proto-language because the Oysters were an earlier migration than the SW Tip colonists. It may be that Oyster is older even than the Tip proto-language and that the comparison will have to be still further back. Alternatively, Oysters do not meet up with the SW Tip colonists until several hundred years after the Oysters' colonies were founded.

Grammar

Though part of the Lava Bed clade, this language almost certainly "freezes out" the grammar and becomes much simpler. It is still more difficult than Late Andanese, but without the efficiency that comes with difficulty in related languages. For example, it may be that mandatory /-h-/ insertion exists, but that it no longer carries the information that it does in Lava Bed languages and therefore just takes up space.

One of these languages is the language of administration of the Anchor Empire (3958 - 4018), because the Dreamers could not learn Play and had little reach over Play territory at any rate. Therefore this language comes to be called Anchor. It may be identical to the Thunder language. The Moonshine scholars later mis-classify the whole Olati family as being part of Dreamlandic.

Because the speakers specifically took pride in their language, like the later Moonshines, they were embarrassed when their language was placed alongside "rival" languages such as those they had displaced in the tropics, which more often than not appeared be more efficient and more expressive than their own. Their experience was very similar to the later Moonshine diplomats, and this was no coincidence, as the Moonshines grew out of the Crystals who were closely tied to the Oysters, all of whom shared the same belief that language determines culture.

In the early days of Oyster, their language was acoustically similar to Play, because Play was still tonal and had consonant frequencies similar to Oyster, as its evolution towards the well-known baby-talk sound was a gradual process. Therefore the Oysters considered the developing Play language to be very similar to their own, and did not attempt to suppress it. The Oysters probably did not realize that Play's grammar (even then) was much more difficult than Oyster's, because they did not realize yet that their own language was so simple.

MRCA (~500) to High Oyster (~3141)

It is not clear if 2057 is the correct date for the break with the mainland, but it would likely mean little. The Islanders considered themselves to have their own language, but it was very similar to Olati-A.

This was originally written up with the whole of the list taking place before 2057.

Pre-Oyster changes

The Andanese/Gold dialect of Tapilula had the consonants

Rounded bilabials:                     hʷ  w
Spread bilabials:      p       m   b   f  (Ø)
Alveolars:             t       n   d       l
Rounded alveolars:     tʷ      nʷ  dʷ         
Velars:                k   ḳ   ŋ   ġ   h   g

1. The accent pattern involved in certain infixes with accented schwa switched to favor the following vowel. e.g. ăpo "field", apə̀ho "field (possessive) became /ăpo apəhò/.
2. The "labial" vowel ə disappeared, syllabified nearby consonants or turned to i if the nearby consonants were not possible to become syllabic. Note that it never occurred after labialized consonants. Sequences such as /pəh/ collapsed to form aspirated consonants, though these behaved as clusters.
3. The velar nasal ŋ changed to n in all positions.
4. The stops p b t became w w k (the /t/ shift was allophonically [th > tx > kx > kh]) except when occurring:

   After an accented or high-tone vowel (but not before);

   In a consonant cluster of any kind; or

   In a monosyllabic word. (This likely excludes CVC words created by the preceding shifts, however, as they were grammatically still alternating with CVCV forms.)

5. tʷ dʷ nʷ > kʷ ġʷ ŋʷ.

   This shift led to consonant changes in a small number of roots ending in -tu du nu which were commonly followed by vowels, usually /o/. In some of these words, the /o/ became fused to the root, and these consonants were shifted further to labials.

6. The labial fricative f shifted to hʷ.
7. The labialized sounds kʷ ġʷ hʷ w changed to k ġ h g when they preceded a vowel followed by a labial consonant (including /w/ and other labiovelars).
8. ŋʷ> ŋ.
9. final /b/ > /w/. must, now, add to other list!
10. Tautosyllabic vowel sequences òi ài èi converged to ē. This did not affect syllable-straddling words like /tùya/. Likewise, èu àu òu in the same environment converged to ō.
11. Duplicate vowel sequences àa èe ìi òo ùu shifted to long vowels ā ē ī ō ū. But the same sequences with the opposite tone pattern did not shift.

Independent changes

1. The labialized consonants kʷ ḳʷ ġʷ hʷ w shifted to p p b f v unconditionally.
2. The aspirate clusters bh dh shift to p t. Then mh mf merged as mp. This shift did not affect /nh/. Any /ŋh/ would also be shifting to /mp/ here.
3. The velar nasal ŋ shifted to m except when before a vowel followed by a labial consonant, in which case it became n.

   This assumes that the Islanders retained an earlier dialectal pronunciation of /ŋʷ/; this is why the shift is conditioned in a way reminiscent of the earlier shift.

4. The velars k g ġ h shifted to č y y š unconditionally.

   Some /k/, particularly in root-medial position, will defy the shift due to a grammatical alternation similar to those described elsewhere. In effect, /č/ will become the genitive form of /k/ just as /š/ becomes the genitive form of /y/.

5. The uvular stop ḳ became k.
6. The sequences lh shifted to s.
7. Remaining aspirated clusters deaspirate.
8. Before a vowel, the sequences ay ey oy shift to ē. iy uy shift to ī. For grammatical reasons, the sequences /aš eš iš oš uš/ sporadically shift as well.
9. Long vowels before hiatus became short, but retained their stress. (Thus, /ia/ is really /ia/, not /ja/).

   At this point, it's possible that some of the original vowels that merged will return since they had survived in the genitives.

10. The voiced stop d shifted to r.
11. Final high vowels i u disappeared to Ø at least post-tonically. CVCVC word stems take over as the underlying forms of many words, but inflections can still cause the old vowels to reappear.

    Between two labials (e.g. /piob piep/ "pine sap"), the vowels were preserved as /u/, so long as the first was not /m/. Between certain other clusters, it was preserved as /i/. It is possible these two vowels were mere allophones of each other in this position, however.

Thus the consonant inventory was

Labials:     p   m   f   v   b    
Alveolars:   t   n   s   r   l
Palataloids: č       š       y
Velars:      k

The inventory here differs slightly from the mainland dialect.

Medial /t/ and /k/ both gradate into /č/.

Analogy might repeat the old vowel shifts. For example, when /ayu/ > /ēu/, it could trigger a repeat of the earlier /eu/ > /ō/ shift because that old shift would have still been operating as a surface filter when words changed from one classifier to another. Thus /payùta/ >> /pōta/. Words that do not do this will likely soon undergo a different shift as the prevocalic /e i/ turn into /j/ and create palatalized consonants.

The sound changes above suggest that primordial hì followed by /b/ and then a vowel would by this point have become just š followed by the original vowel but now on a low tone.

Substratum influence

Coronal consonants were weakly palatalized before any /i/. The velar stop /k/ was not palatalized because it instead pushed surrounding /i/ into a high central vowel.

All consonants were weakly labialized when bordering a /u/ in either direction. Neither this nor the shift above were phonemic, however, and when the high vowels were lost, they did not leave behind labialized and palatalized consonants.

Morphological adaptations

Classifiers

The classifier inventory is very different than that of related languages. It is likely that only a two-way distinction was made in prefixes' vowels: /a/ vs all other vowels, which rotated according to vowel harmony, though possibly not completely so. For example, /u/ might precede /i/ in some words.

Other shifts

Since /kʷ/ > /p/ happened twice, the second shift could be analogized to the first shift, and the requirement for stress shifts removed. In final syllables, /pu tu ku po/ can all theoretically interchange, though it is likely that one member of each set will only play with one other member. The less common /du bu bo/ could also interchange, though in this case there is no way to form two pairs.

The syllables /čo ču/ may be selected against, even though they are the regular outcomes of the fairly common /ko ku/ of the parent language. Possible means for this are:

1. Retention of the primordial /e~o~ə/ alternation, such that /čo/ is replaced by /če/ whenever possible;
2. In final syllables, retention of /k/ due to grammatical analogy; this could even happen before the other three vowels. On the other hand, /ku/ may interplay with /p/ and other consonants.

Single-consonant classifier prefixes arise from the loss of /g/ in both Oyster and Middlesex; these could be padded with traditional CV classifiers to become the CVC- classifiers that appeared in an early writeup of Xap II, although their use would be sporadic at best since they can only appear before vowel-initial roots, and none of the speakers would create consonant clusters to accomodate the CVC prefix format.

In verbs, /v/ and /p/ may also interchange, and in the later history of the language this /v/ shifts to /b/, perhaps interweaving itself with the other adaptation. This is because /p/ in accented syllables shifted to /w/, and then to /v/ (never to /j/, because this happened in final syllables only).

Some words with medial /-š-/ could change this to /-y-/.

A very old alternation between ya and wo (in any context, but primarily over morpheme boundaries) may have survived here. Likewise, when the sequence ai appears over a morpheme boundary, the result is o, not */e/.

THe sequence kiy- before a vowel could become š-. This is an extremely old sound change, but because of its intuitive result, it could survive. This covers inherited /ḳu-g-/. Alternatively this could become /pi-/.

Primordial final consonants in low-tone words may be retained, padded with vowels, and the stress shifted to the penultimate syllable. The consonants may not necessarily be reflected very well, due to massive analogy; what matters is that the reflex is a trisyllabic morpheme where the surrounding languages have disyllables.

This means that most, and possibly all, noun roots end in a CVCV sequence with a high tone on the penultimate syllable, making tone irrelevant in noun roots. The language still retains tone shifts of semantic value.

Prefixes

Note that fi- fu- ši- šu- f- š- are all merely different forms of the same reflexive verb prefix, and may come to be parsed as bimorphemic. In theory, m- mi- mu- could also exist.

The accusative of all prefixes beginning with p- could be in b-, extending an alternation that might have been present in just a few prefixes. This is etymologically sound when /p/ reflects true /p/ but not when it reflects /kʷ/. However, it may well be that there was no surviving example of such alternation in the entire language.

Daughters

The Lantern Empire was run by the Crystals and the Thunderers, who spoke Middlesex and Oyster. It is likely that the languages were slightly offset from the tribes, however .... in the far north there were some Crystals whose inherited language was Oyster because they were the descendants of converts, and it is possible that Crystals in the far south also spoke languages derived from Oyser.

Proto-Oyster (~2050) to Birch (~3310)

This sound change list is very short because it actually covers only about 200 years. It is possible that under Crystal influence, all three of /f v b/ merge into /b/.

Labials:     p   m   f   v   b    
Alveolars:   t   n   s   r   l
Palataloids: č       š       y
Velars:      k

1. The postalveolars č š shifted to c s.

   This may allow for some grammatical shifts of /s/ > /l/, reflecting inherited /lh~h/ and /l/. A preceding shift had already shifted some of this analogically to /y/.

2. The labial fricative f shifted to h.

   A likely followup shift, perhaps even simultaneous, is /v/ > /b/, even though this would leave the language with just /p b m/ for labials.

3. The vowel sequences èa èe èi èo èu shifted to ʲā ē ē ʲō ʲō.
4. The vowel sequences ìa ìe ìi ìo ìu shifted to ʲā ʲē ī ʲū ʲū.
5. The true palatals kʲ hʲ shifted to č š.
6. The consonants t n s r l h became palatalized to č ň š ř λ š before any /i/. Labials and /k/ did not shift.
7. Palatalized labials depalatalize. However, it is possible that the reflexes of at least pʲ mʲ are t d rather than /p b/.

At this stage, the consonant inventory was

Labials:         p   m       v   b
Alveolars:       t   n   s   r   l
Postalveolars:   č   ň   š   ř   λ
Velars:          k       h

Proto-Oyster (~2050) to Plume (3431)

This may be the same language as Birch. It may borrow many loans from Baywatch.

Plume (3431) to High Thunder (4149)

When the Play party takes over the Anchor Empire, they note that there are still people speaking the Thunder language living in the northern states.

Plume (3431) to Defector (3948)

Exists only if the speakers did not simply switch to speaking Baywatch.

Plume (3431) to Creamland Intruder (4149)

At one point, Sapeepa invaded all of the south, trying to conquer Aboan nations like Thaoa. They poisoned water and claimed that they would make the land unlivable. (This is why Lenians still lived in Subumpam in later ages despite losing the war against the Stars; they were not merely immigrants from Paba.)

They by this time had some slaves from Baeba. Baeba had invaded and conquered them, but when they revolted, many Baebans were trapped in their territory and became slaves.

This language likely differs little from Thunder.

Proto-Oyster (~2050) to Paleo-Nunabetari (~3350)

Within months of their defeat in Repilia, the Lenians declared war on the dark-skinned aboriginals of Baeba Swamp. The war ended in a stalemate, but a third army soon arrived and conquered Baeba; they were more favorable to the Lenians than to the aboriginals.

This language only exists if some of the Lenians remained in Baeba after the war.

Proto-Oyster (~2050) to Poise (~3700)

Only if Poise has its own language. If this exists, it likely has Dreamer influence, but from an isolated branch of the family, not from Baywatch, Soap, or any other one of the major Dreamer languages. Minor Lenian languages states that Poise's Dreamers spoke a language practically identical to Baywatch, but this may be wrong, as they immigrated to the area at least a few hundred years earlier than the Baywatchers' contact time. Poise's Dreamers by 3700 AD spoke a language with no geminates and very little /s/. It may also have phonemicized /k ŋ/ but they would be confined to occurring before /o u/ except in loans. (Note the name of the state was Posisse early on.)

1. All s was deleted.

Proto-Oyster to RCV

This language was spoken by the RCV party in Nōslina.

Proto-Oyster to DOZ

This language was spoken by the DOZ party in Gàqa.

Proto-Oyster to Olansele

This language was spoken by the XIG party and some others in Olansele. It was just one of several languages spoken there.

Proto-Oyster to TOT

This language was spoken by the TOT party in Sapeepa. It was just one of several languages spoken there. There was no significant influence from Baywatch or Birch, just from Khulls. The contact began just as Khulls was shifting /d/ > /r/. At first the TOT speakers were dominant, but soon Khulls became the language of administration.

1. The palatalized labials pʲ bʲ mʲ shifted to p b m .
2. Coda stops disappeared, changing the preceding vowels to high tones unless the stops were voiced.

Proto-Oyster to Teacher

This language was spoken by in Hōkī.

Proto-Oyster to HOL

This language was spoken by the HOL party in Šelpa (Sàkwalo).

Proto-Oyster (2057) to Rusted Pearl (4143)

It is unlikely that the Rusted Pearls remained in place for over 2,000 years as their homeland changed hands from one outside power to another. It is not clear that they even continued to speak their ancestral language, because although their tribal identification was indeed passed down for 2,100 years their language may have changed several times as they moved. Note that the writeup simply gives their name in Late Andanese. If they did not remain in place, it is possible that they are part of the #Creamland Intruder branch, perhaps the westernmost part, and therefore their language would be identical to Plume until at least the year 3431.

At any rate, the tribe identifying itself as the Rusted Pearls lived in eastern Subumpam, near but not quite in the same area from which the original Oysters set sail.

Proto-Oyster (2057) to Soap (4205)

The Soap Bubbles considered themselves descendants of the Lazy Palms tribe, not the Oysters, but the Lazy Palms adopted the Oyster language fairly early in their history, and therefore the Soap language can be traced back to the same starting point.

I have always pictured the Soap Bubbles as having spent a long period of time in the southern deserts ("Dahawu"), in the same territory that the Blue Oysters moved to. (Eastward of "Sun Country".) However, they were politically a branch of the Thunderers, having split from them only in 3844, which means that their language is also identical to Plume up until 3431 and identical with Thunder until 3844. Thus the Soapies spent at the very most 350 years in the desert, and are unlikely to have absorbed any significant linguistic influence from the people living there, be they Crystals, aboriginals, or some other group.

Other groups

There may have been Oysters in the original Oyster Empire as well, despite them losing their war, because it is not clear that the whole of the population was pushed out, and because it seems that the Oysters regained control of at least some of their original territory sometime around the year 3040. This is separate from the Thunder invasion.

Contact with other languages

Some Gold-speaking Pabaps were moved to Lobexon in the year 2057.