Play language

Play is a fusional and polysynthetic language that reached its peak around the year 4200 in Memnumu and points nearby, when it was the language of more than one fourth of the human population. It was known for its simple phonology and extremely difficult grammar. Players often used long compound words that seemed to hardly resemble their parts. For example, from tatabūpu "coconut palm" and peep "to shake", one could say

Tatapaeikupupites?

Why are you shaking the coconut tree?

Such words corresponded to whole sentences in other languages, and were commonplace in everyday Play speech. This, among other features, made the Play language so impenetrable to outsiders that, during wars, Players were able to simply carry on their daily routines without fear of foreign spies. The history of the Play language relied largely on its extreme difficulty.

Scratchpad

Butterfly

16:25, 6 May 2022 (PDT)

One of the few Play words with /ss/ is žessušapa "butterfly", literally "beautiful animal having its own refuge".

Semantic emphasis

15:02, 2 May 2022 (PDT)

Classifiers are never phonetically stressed, and so cannot be used for semantically prominent information. Thus a phrase such as "all life in the sea" would not be translated by simply "all" (usually /nafata/) with the sea life classifier suffix /-pa/. Play has a separate word for the whole ocean, vaŋapa, which is like the titular words of Andanese in that the classifier suffix is redundant since the word does not occur outside this class. This can be seen as saying something like "the sea in the sea". Therefore, its own locative form requires an additional -m, and this occurs after the /-pa/, not before.

Paisi nafatam vaŋapam fayaaušeppa.

All life in the sea is frightened.

Note that the verb for fear here is reflexive. And note that it takes a /-pa/ suffix to indicate where it takes place, one of a few verbs to require this.

Tense and aspect

04:48, 28 April 2022 (PDT)

Tense

The Play past tense marker is -ib-, which could combine with the plural marker /-ub-/ to form -uy-. The present tense was unmarked, and the future tense was indicated through mood markers, so the past tense /-ib-/ is the only tense marker in the entire language.

The /b/ would be expected to disappear, reshaping this affix to /i/ and then in some cases to palatalization of the preceding consonant. Moreover, primitively it would have also invaded coarticulates like /tʷ/, creating /tiw/ > /tib/ > /tip/ instead of retaining its own /b/. But this might not happen.

Aspect

In most languages of the family, aspect is marked through a construction resembling serial verbs ("to start while reading", "to read while starting", etc), with different solutions as to which is the so-called head verb. In Play, this could suffice, or there could be a more elaborate solution.

Noun-verb coupling

02:28, 6 April 2022 (PDT)

One very difficult aspect of Play for outsiders was the idea that two open classes, nouns and verbs, could fuse together to create new words that seemed atomic because they did not have an audible morpheme boundary and because both often underwent stem changes in addition. For example, pupa means book and vāu means to read, but neither of these words is audible in the compound word

Pukūavesa.

The book of yours that you read to me.

Which functions as a noun just like any other, since it begins with a root and ends with a classifier suffix, just like /pupa/ itself.

This is the verbal embedding written of below. Although Play did use a certain small set of a few dozen verbs much more commonly than all others, the fact that a phonetic formula existed for the embedding meant that the class of embeddable verbs was in fact the class of all verbs, and therefore was open.

In most other languages, even highly complex ones, when fusion of two content words occurred, one of them belonged to a closed class.

Morpheme order

Note the difference in morpheme order between pukūavesa "the book of yours which you read to me" and pupas "I see a book". Here, /-pas/ is submerged /-ba-s/ and /-sa/ is submerged /-s-ba/; the same two morphemes but in opposite orders. The change from /b/ to /p/ is because of the word itself.

Using 3>4 to indicate 1>3

Note that e.g. muaeapippu can mean "the pine tree I planted for you" if there are no other actors in the sentence or in a previous sentence, but if it is replaced by muaeapipu, without the extra /p/, as described below it CANONICALLY means "the pine tree that [someone] planted for [someone else]", and not "the pine tree that I planted [for (NULL)]". There must be a way to specifically designate 1>3 to the exclusion of the 1>2 and 3>4 interpretations.

Essentially, this comes down to a question of whether the innermost person marker is BENEFICIARY or PATIENT.

It is possible that Play simply uses nulls in both slots when 1>3 is intended, even though the sentence seems to indicate 3>4. This would be explained by a lack of other nouns in the sentence, and would depend on this.

Perhaps this problem will be solved by taking my mind out of the "pseudo-person" mode and using the morphemes as they are: transitivity markers, with person provided solely by context.

Poetic animacy

08:37, 28 March 2022 (PDT)

Play could be the originating language for spontaneous words such as fipipu "tree", literally "I see a goal", because it is personless. The /-u/ here is not the evidential but rather a submerged form of the classifier for forests and fields.

In Gold, the equivalent word hʷiməči would simply mean "that which sees a goal" if used without a classifier, and by adding a classifier a new word for tree such as hʷiməčidu could be created, but the word would not closely parallel existing word structures.

Addressing inanimates

The Play idea is part and parcel with the idea that animates can be addressed: tes "you are a pine tree" (no classifier suffix). Structurally this requires an animate listener because all listeners must be; semantically therefore this requires a verb /te/ "to be like a pine", and that it be something humans can do. Metaphorical use is, nonetheless, understandable since there will never be a tree that is a human, and therefore it is clear when an inanimate is being addressed.

The reason there is no classifier suffix is because if there were, the resulting Play word *tepu would not mean a pine tree, but rather a tree that says "you are a pine tree" (or "I see a pine tree"; compare below).

Constructions like this also serve a second meaning: Tes "you are a pine tree" also means "I see a pine tree", even with no evidential suffix on. This is similar to zero-marked constructions in English: Play

Sufubas!

There's treasure here!

Can also just be translated as "Treasure!" given the right context.

The need to differentiate between the 2nd person and impersonal meanings of these words led to the increased use of evidentials, which eventually became mandatory in post-classical Play's "pusiba system".

Note also the subtle difference in morpheme order; here, the word ends in -ba-s, a classifier followed by the polysemic /-s/ suffix that can also mark possession. But the word about a book ends in -s-ba, surfacing as -sa, showing the same two morphemes in reverse order. This is the difference between addressing an inanimate (or mentioning it) and involving it in a wider sentence. (NOTE: I will try to explain this better later so that I dont get confused myself. It may be that it is better to analyze Play as having two /-s/'s with similar meanings but different semantic roles.)

The fibula

Despite the existence of poetic animacy, it may or may not be possible for inanimate objects to "talk" as in the Old Latin inscription manios med fhefhaked numasioi, where an item of clothing speaks and uses the pronoun "me". What matters most is whether or not this will conflict with the ability to label something. The part that means "___ made me" would be mikəči in archaic Play and mikip in classical Play. But the use of a passive verb implies that an active is possible, which inanimates cannot do.

Question particles and suffixes

11:40, 25 March 2022 (PDT)

Remember that the Play open question suffixes, tīs and tes, couple to noun classifiers and therefore most often drop their /t/ (but sometimes /t/ overwhelms the classifier instead). Thus for example, one can say

Pisisaes?

Where do you live?

When referring to an unspecified place, but

Pisišītes?

Where do you live?

When specifically asking about a building.

The mismatch in forms is because vowel-final classifiers retained their silent /ʕ/ (and /ʕt > d > Ø/) but consonant-final classifiers came to behave as though they had never had an /ʕ/.

These words were often used in just this state, but could also interplay with verbal embedding. For example,

Pukūaveepaes?

Which of my books did you read to me?

Animate nouns' question markers

Animate nouns cannot take classifier suffixes, but nonetheless the question markers tīs and tes simply attach to them just the same, being suffixed directly to the stem instead of staying behind a classifier.

This, however, means that there can never be a Play word paradigm like that of IE, where the question words "who? what? where? when? why?" are just five different case markers attached to a single particle. Rather, Play has the opposite morpheme order, with the question particle being an invariable morpheme attached to the stem. Thus one must say

Tates?

Which human?

Instead of "who". This situation mostly affects animates, because if the thing being asked about is inanimate, it is often the patient of a verb. Note that the /t/ is not dropped, because this word never had /ʕ/.

Specifics of meaning

05:51, 20 April 2022 (PDT)

The two Play suffixes tīs and tes have identical meanings, as the originally distinct /tīs/ has been reanalyzed as the essive form of /tes/. Using /tīs/ adds a slight amount of additional emphasis to the question because the syllable takes longer to say. This means that, as above, there is no division among Play question words comparable to English "who/what/when/where/why". Rather, there is only a "what". This situation is common in related languages as well, and was inherited from Gold.

Words for "why"

Assuming the question marker must be the outermost morpheme in a word, it is therefore impossible to build a construction such as *setifum "for what reason?" (Note the etymologically incorrect inflection of /tes/ > /tifu/, which had become generalized.) Instead the Players would use setes "what reason?" and then place an inflection with opposite meaning on the rest of the sentence. That is, the sentence would say something like "What made you do this?"

The existence of this construction implies that it could be done in a single word, where /se/ takes the place of the expected classifier suffix, or possibly without even using the word /se/. Remember, though, that Play's /-u-/ was still just a plural affix. If a word such as

Fūvanabubapsetes?

Why are you squeezing me?

Can really stand alone, it implies that the action was voluntary on the part of the patient. This may have been a relic of earlier times, however, if /-u-/ was already shifting from a plural to an imperative (with additional meanings bundled in) in classical Play speech. In either case, this phrase could probably drop its /u/:

Fūvanaapsetes?

And break up the words:

Fūvanaap setes?

And reorder them to put the question word at the beginning:

Setes fūvanaap?

All without changing any other inflections. However, if a form without se exists, the /-tes/ suffix must be at the end of the word as it cannot stand alone.

Combinations

The word fi is another morpheme for "reason, motive" that is essentially synonymous with /se/ but appears in different contexts. From tatabūpu "coconut palm" and peep "to shake", one can form

Tatapaeikupupites?

Why are you shaking the coconut palm?

Demonstratives

11:40, 25 March 2022 (PDT)

Poswa has no demonstratives, and must use verbs with person markers on. (That is, "by me", "by you", etc.) It is possible that Play was the same way; but Play has a very complex noun structure already and adding a demonstrative suffix to it all would not overburden the speakers since it would only appear where certain other suffixes could not appear. For example, it seems logical that the demonstrative suffixes would never appear on nouns that were also tagged with the question particle tīs ~ tes (see below for derivation); if the question particle is also part of the same series of affixes as the "belief" mood markers, then it stands to reason the demonstratives can not cooccur with them either.

However, it is perhaps more likely that Play will not have demonstrative suffixes, but instead will use verbal embedding, as below, indicating "the pineapple you show me", and so on. Another alternative is to use locatives; this could be affixed to either a verb or a noun.

This would mean that Play can combine the demonstratives with the question tags after all, unlike many related languages in which the two use the same slot (or separate word).

Alternatively, a particular demonstrative is mandatory on questioned words.

Demonstratives of place

For example, taibem "children" can become

taimipippa "the children I see in the water". Note that -pi- here is specifically a marker of distant sight; the object must be visible and yet far away from the speaker (not necessarily the listener), and that /pip/ is formed through analogy (earlier Play would have had /peči/).

taibem patapa "the children playing in the water" (though this would more commonly have an evidential /-ba/ at least in later stages of Play)

taimipippa patapa "the children I see playing in the water". Note that the locative must be repeated twice because verbal embedding of the type used here demands it, even for animates.

Less common are constructions like

taibem baapippa "the children in the water I see"; usually the word with the demonstrative is pushed to the front of the clause instead, which means changing the inflections around on the other words. Note that /baa/ here means "fresh water" specifically; /pa/, the classifier for water and sea life, is almost never used as a root.

Deictic demonstratives

The locatives above cannot be used for deixis because their meanings are bound to literal location and not abstract concepts. For this, Play can only use verbal embedding yet again, with verbs such as "talk about" and the like. There is no tense marking on this. From pupa "book", one can say pukuefipa "the book I talk(ed) about", which suffices to translate both "this book" and "that book", depending on context, whenever a locative meaning is not intended.

Comparison to inanimate verbal embedding

The word taimipippa above has a morpheme structure very similar to the classical verbal embedding used above. What's different is that the noun classifier suffix has reverted to its archaic literal sense, "in the water", rather than classifying the children as sea life. This is required when using animate nouns, because animate nouns cannot have classifiers. Thus, the demonstrative must always couple with a classifier suffix indicating a place, and there is no generic classifier that just means "place". However, the bare -a, representing the word for field, is used in any case where some other classifier does not apply.

Note that this also discourages the creation of words such as *bāātaiyaappa, "the captain I obey (in the water)", as if meaning "my sea captain", because, while grammatical, it uses a circumfix yaa-...-pa "to obey in the water", which is semantically disjoint, and moreover contains just a single -p- in between. Usually with this type of construction, the classifier suffix is semantically linked to the head noun, not to the embedded verb. Even a circumfix yaa-...-a "to obey in a field" would sound odd to the Players. Rather, a two-word solution would be used.

Verbal mood and associated morphemes

True mood markers

Since there are no person markers (see #noperson below), all of the verbal modifiers like "need to", "want to", and so on behave like mood markers, and this system would need to trace back to Gold. Since they express the viewpoint of the speaker, they are mood markers in the strictest sense, indicating the speaker's emotions only.

It is possible to combine the mood markers with the transitivity markers, but note that "I want [me] to hit you" does not reverse to *"you want me to hit you", nor to *"you want [you] to hit me", but rather to "I want you to hit me".

But Players cannot use mood markers to express similar concepts where the person doing the wishing is not the speaker. Thus "You want me to row" cannot be a one-word sentence with just a mood marker in Play, even though it has one less participant than "I want you to hit me", because there is no such thing as a 2nd person mood marker.

Play inherited many of its mood markers from Gold, but because they were grammatically identical to the little-used postverbal locatives, any word indicating a location could be used metaphorically in a once-off manner just like a mood. Thus the Play mood markers formed an open class. However, they could not be stacked; every verb had only one mood.

The core mood markers in Play were:

Expressing knowledge

1. -Ø, the indicative mood. This was not a morpheme that had gone silent, such as /ʕ/, but simply a convenient construction to form a pattern.
2. -pa, the interrogative mood. This is etymologically the same morpheme used below in the strong desiderative mood, but without the linking [M] morpheme. It is also used as a conjunction meaning "or (else)". Note, however, this interrogative marker is never paired with the question marker -tīs, nor can one substitute for the other. Rather, /-pa/ modifies a complete sentence and /-tīs/ modifies a single word. Note also that /-pa/ marks polar questions.
3. -ŋi ~ [M₂], the negative, expressing strong doubt.

   This cannot be compounded with the other mood markers, and thus behaves as a true mood. This means that there is no negative imperative, or any similar compound. Instead, to make a negative imperative, the part of the sentence that the speaker wishes to negate would take a negative morpheme. A compounded form exists, -[M₂]-u, but this has the same meaning as the bare form and appears only because the morpheme would otherwise reduce to just /i/ after passive verbs and would be ambiguous after verbs whose stem ends in /i/ because it would collide with the imperative. Note however that /-[M₂]-u/ is aimnu ~ imnu ~ emnu because the elided /d/ goes to /n/ instead of disappearing.

There is no Play mood corresponding to subjunctive, but the subjunctive grammatical category overlaps with some of the proper moods and also with the Play suffixes -p -s, which show relations between two verbs or verb-final clauses.

Likewise, Play has no clear match to the IE conditional mood.

Expressing desire

1. -m, the imperative mood. This is inherited from Gold's -ṅ.

   In Poswa, the imperative can take all three person markers. A Nenets grammar linked from Wikipedia describes these as hortative, imperative, and optative. Since these are just person-marked forms of a single mood, only one Play morpheme is required, but it could perhaps be better called the jussive mood.

   The imperative mood is also the form used after a particle like "if", meaning it can function as the "IF" mood. (This could be called conditional or subjunctive, but different languages use this mood differenty.)

2. -[M]-na, the common desiderative mood. Expresses the speakers' desire for the situation to come true; often used as a gentle imperative. This is formed by affixing /-na/ to the 1st person "dirty feet" morpheme below, meaning it appears variably as -amna ~ īmna ~ ūmna depending on the final vowel in the stem of the verb, and appears as -ana when the verb ends in one of /p s/, as in the case of reflexive and reciprocal verbs.
3. -[M]-pa, the strong desiderative mood. Expresses the speakers' immediate needs.

Note that the desideratives and the imperative can be marked for all three persons. The common desiderative often corresponds to an English word like "should" when used with the 2nd or 3rd person.

Expressing belief

Morphemes in this category may team up with, absorb, or lose out to the evidentials /bu bi ba/ that came to dominate in early Poswa. (The rise of /bu bi ba/ in Poswa was due to the convenience of beginning with /b/, and therefore, they could have been common in pre-Play, lost ground by the time of classical Play, and then gained it all back when they triggered an important sound change.)

1. -tau, the potential mood; this is a catch-all for statements where the speaker does not know how likely the statement is to be true, is not requesting an answer, and shows no emotional investment in the situation. It is similar to Play's habilitative and permissive verb markers ("the cat can run away"; "the boy is allowed to speak"), but applies also to inanimates ("the leaf might fall") and even to sentences without a verb ("it might rain"). Thus it does not require volition, and can be used in combination with volition ("The boy might be able to run away").

   This is from /tə̀la/, not */tə̀ndu/. It is very distantly cognate to the question marker tīs, but not even scholars would see the connection because /tīs/ lost the forms in which its presemblance was most evident. Nonetheless, note the morphemes expressing categories of belief below.

2. -tata, expressing strong belief.
3. -tiu, expressing weak belief.
4. -tetu, the dubitative, expressing doubt.

Despite the different vowels above, the first morpheme in all of them is /tə-/. Play also has a habilitative morpheme taus; this is not a mood, but rather a verb morpheme that patterns like the aspects of IE languages. This is not cognate to the mood markers (remember, /tau/ is from earlier /tə̀la/).

Evidentials

These are referenced in the #Pusiba_system section below, which is a later stage of the language when the transitivity markers -p -s -Ø had combined with evidentials to form new person markers. The combined forms already existed in classical Play, but were fluid and each evidential could combine with any of the three transitivity markers. This could be called color-flavor locking, using a physics analogy just as the 2x2 Play system is likened to gluons.

Pusiba system (post-classical Play)

• bu to feel; to know intuitively; unshareable knowledge (later, a 1st person marker)
• bi to see; shareable knowledge (later, a 2nd person marker)
• ba to know; to have seen; presently unseen knowledge (later, a 3rd person marker)

Hypothetical earlier system

If the evidentials pattern like the "belief" morphemes above, then -bu is the atomic morpheme and the other two come from this morpheme followed by /Ci Ca/ where C is a consonant that dropped out during the evolution of the language.

An alternative is that the three were independent morphemes all along, and originally began with different consonants, but were analogized to /b/ due to sound rules like /pn pb/ > /p p/ where any disappearing consonant could be later assumed to have been a /b/. This could only have happened in the post-classical Play period, however, because in classical Play the morphemes were not yet tied to the person markers. In this setup, the morpheme ba must truly be /ba/ and not with some other consonant because in both classical Play and the later maturation period it was used after a vowel and therefore no analogical substitution was possible.

VARIOUS MORPHEMES USED IN MOODS

• ta, from Tapilula /tà/.
• te, from Tapilula /tə̀/, but most likely appearing as /ta/ most often. /tə/ + /ta/ is /tata/ and is a legal compound. The question particle tīs has a variant form tes, both of which appear to be related to /te/, since a simple infix could create both forms of the question particle. Indeed, they are ultimately related to each other, but at a much older layer of the language.
• u, from Tapilula /ndù/. There may be a second use of /u/, originally not a separate morpheme, that is seen in such as /tau/ above. This is just from Gold /təl/, with /a/ instead of /ə/ for reasons given above.
• va, the /la/ needed to create the above. Likely not used in Play even as a fossil.
• ŋi, used in at least two morphemes.
• vi, used in at least one.
• tu, used in at least one.
• na, a second /na/ that is not related to the first and comes from pre-Tapilula /nat/.
• [M], which is from Tapilula /ŋà/ rather than the 1st person dirty feet suffix.

There was once a compound, -paim (for expected [M]-paim), a negative desiderative, but it fell out of use as the rule against compound moods became solid. The ones with /-t-/ are not seen as compounds and the dubitative is not seen as a negative mood, but only a weakly positive one. Similar constructions nonetheless existed to cast doubt on just a single part of a sentence ("the team's unlikely success") without using a mood marker. These could be described as irrealis NOUNS.

Future tense constructions

Like its descendants, Play has no proper future tense. However, the setup is very different: Poswa's grammaticalized -u- inflection, which marks the imperative, is actually the Play plural marker, having semantically shifted through the meanings of voluntary and then involuntary cooperation. This morpheme fills the same slot as the past tense marker -i-, and therefore the three-way Poswa setup of i/Ø/u (past/present/imperative) is naturally intuitive and convenient. In Play, the morphemes could still be stacked, so plural past tense was expressed easily enough, as -uy- right alongside -ub- for plural and -ib- for past. (The /b/ insertions are normal and are not duplicated when the morphemes are stacked; thus /-ubib-/ was wrong. The expected analogical -e-, which could have arisen from the /ui/ > /ə/ sound rule, did not occur either.)

This however meant that Play had the ability to create a past tense imperative.

Use with nouns

At least some of the evidentials could be attached to nouns as well as to verbs; the question at the top of the page (04:32, 28 April 2022 (PDT)~), Tatapaeikupupites?, is an example of this, since tatapaeikupu is actually a noun, not a verb. That said, the linking morpheme -p- is important in this constriction, and so it could be considered that it is a verb after all even though it is not etymologically the same /p/ that marks "self".

better words for cardinal directions

cardinal directions

These words will all be inherited from Gold, but some may be padded or replaced. Gold in turn inherited the words from Tapilula, which was a maritime society, and therefore needed words for navigation. Gold is less likely to have replaced the inherited words, but may have used padding.

Play words here are nouns, meaning "northern area" etc and not verbs or abstract nouns such as "northern-ness" etc.

If pi "boat" survives, it could mean "south across a sea", etc.

the basic cardinal roots for the four directions are very repetitive: šavafa "north", šatua "south", sata "east", and šasuša "west"; these are mingling with the more specific terms that mean "north across the water" and the like.

In Play, initial s- š- p- can all shift to Ø after a preceding word in the (primitive) locative case; Leaper cannot do this. Thus Play can create words such as Pubumafuata "east of Pubumaus", Pubumafuaašatua "south of Pubumaus by sea", and so on, even using proper nouns. These words cannot be used for "eastern Pubumaus" and so on; they always indicate a place outside the root word placename. Another formula uses separate words and assumes a preexisting genitive case instead of the locative, in which case the consonants are not deleted. Yet another formula uses freestanding words such as piaipsata "east of the river by boat", with three morphemes (pi / paip / sata) stacked together, following the name of a river, and so on. The word pi "river" is ultimately not a duplicate of the p- "boat" morpheme, and did not gain its /p/ for thousands of years after the formula was created, and so would not seem redundant.

other directional words

the word piša can still mean "across the water" without specifying a particular direction. it would just be /pi/ "boat" + /ša~fa/ "directional location", with the middle morpheme removed. It may be better to make the word paipa, however, using the inherited /paip/ morpheme from Gold /paiḳ/.

new locative case

Play's locative case is marked by -m; inherited constructions survive, but tend to be bound to their morphemes. After this /-m/, any š shifts to f, but other consonants remain intact.

/si/ > /s/

05:06, 8 March 2022 (PST)

Note that the sound change of si to s before a vowel is usually incorrect. This is seen with, for example, the final suffix /sa/ replacing /-s ya/ in words like tasa "drink" that make use of classifier suffixes. It is still legal when the suffix is /-ba/ since /sb/ > /s/ in all Play words, but words like *vapasas "war" are incorrect and could at best have been created through analogy. The correct form would be vapasias, and in fact vapias was the most common word, without the middle morpheme.

The sound change IS legitimate when the /s/ derives from Trout's /f/ phoneme, since Gold did /fj/ > /f/, and then shifted all /f/ to /s/. (It may actually have been [f fj] > [fʲ] > [s].)

Also note that Play can use š here, since it can also generalize the pre-shifted Gold forms with /h/, and /hi/ > /š/ in all Play words before a vowel. But Play did not usually use these pre-shifted forms.

Particles

Poswa is analyzed as being entirely without particles, since the particles and conjunctions can all be analyzed as fronted verbs. This was probably true in Play as well, although perhaps less remarkable since the verbs have a zero morph form.

In Late Andanese, ka/ki/pa/pi mean "and/if/or/but", with /pa/ also serving as a question marker. In Play, the direct cognates of these would be ka/či/pa/pi if there were no semantic shifts, no reanalysis, and no analogy based on sound changes. However, it is unlikely that Play would preserve these single CV particles given the complexity of its grammar and that because such words were unstressed, they could be mistaken for the final syllables of words whereas this was much less likely in Andanese.

Andanese influence might explain the surprising conservatism of the Play particle inventory, where otherwise so many short words were lost or replaced with fusional inflections.

/ka/

There might be ka .... ka for "both ___ and ____" with the first /ka/ optional. These are verbs, so they inflect for 1st and 2nd person (presumably as kap and kas), but note that these are context-sensitive like the other person markers and that /-p -s/ really mean "self" and "nonself". This implies that it might need to precede the noun to really mean 1st person. Alternatively, they could be conjugated as full verbs, meaning that they would need repeated vowels, etc.

Play also had a second "and" word, ču, which survived into Poswa. The two may have coexisted even in Play. In some contexts, nonetheless, /ka/ might be replaced by pa if the word-initial /k/~/p/ alternation that affected content words spread analogically to function words.

/ki/

This would appear as /či/ in Play, and is not related to the later morpheme /pis/ that appears in Poswa, which comes from Play pes and tīs, the latter being a question marker (distinct from /pa/ below). However, pes might have meant "if" in Play. Note that these were CONJUGATED like ordinary verbs at least in pre-Play, and that therefore any word could add an /s/ or /p/.

Like /ka/ above, this could appear as pi in some contexts, because /č/ would also become /p/ in the alternation mentioned above. (č > ǯʷ > b > p) This still is not the ancestor of Poswa/Pabappa /pis/ but could have helped influence it, especially in Pabappa where i > i is possible.

/pa/

Preservation of pa as the question particle is likely. Poswa reflects it as /pu/, which is regular, but from Play /pū/ rather than /pa/. (The artificial /ču aa/ etymology is to protect /p/ from gradation in later stages of Poswa.) Play would not have used this, and therefore the Play question particle should be /pa/. Then Play could use pipa (/pi/ + /pa/) as its word for "or; but alternatively". This is a compound of the two words above and implies that /pi/ was probably not used alone but carried a meaning similar to "but".

The Spanish-like structure pa .... , pa ..... for "either, or" is likely to be valid in Play, however, and particles may in general be more protected from compounding when occurring in absolute initial position since they would not be mistaken for final unstressed syllables. It might be pa ... pipa instead, or both forms could be valid.

/pi/

This word means "but". It likely disappears in Poswa, and it may be that even in Play, it only occurred as the first element of a compound. Since Play already has pupi "polar opposite", it may be that this was the second element of that, and that further compounding was possible. This word is listed in the dictionary as representing pup ži, however.

Other particles

It's not clear if Play would have atomic negative particles corresponding to "not", "neither", "nor", etc or if they are all just compounds of other particles with a prefixed (or perhaps suffixed) element carrying the negative meaning. Pu could be the catch-all negative morpheme here, but note that pi already means "avoid".

Better numerals

06:21, 6 February 2022 (PST)

Play society inherited long words for numerals just as the societies around them did. The Andanese had long numerals too; the tradition of monosyllables comes from using their syllabary, whose first ten members were a la i ha ka u ma ga na li and thus represented the digits 0 to 9. Play simply borrowed these rather than using the first ten members of their own syllabary, because the digits in the script were also borrowed. The Play speakers (at the time, usually known as the Pubu people) did not achieve Andanese-like numeracy even when they developed a strong school system, because they typically did not work in trades requiring strong math skills. Only after they absorbed the Andanese and took over commercial trades did they begin to use mathematics in daily life. Therefore the numeral system contained only a few basic roots, with even some low numbers being visibly derived from compounds of other numbers, or of transparent word roots.

Other important ideas

Play also had a different word for ten that was used in certain contexts and could have provided the basis for inflected forms like "tenth", "set of ten", and so on.

The correlation between pās "four" and fuppās "eight" is partly a coincidence, as the extra morpheme in the latter was originally an infix, not a prefix. There is no standalone numeral fup.

In the original scheme, the words for numbers were always expressed in the essive case, meaning that they did not become the heads of their noun phrases.

Ordinals, both the fractional type and the "place in line" type, were expressed using the locative case. The analogy here is that a half of something is the second part of it, and a third of something is its third part, and so on.

Thus the bare roots did not appear in daily usage. This cannot be preserved in the current setup, although something superficially similar will be in Play because most of the inherited numeral words end in -s, which although not historically a genitive, could be analyzed as one. Both Play and Leaper will have opportunities to lose this final /-s/, separately in each language, leading to the final forms being more divergent than one might expect.

In the Gold language, numerals were also long, since this is the ancestor of Play and Play mostly kept the syllable counts of inherited Gold words intact. In Leaper they became shorter, largely through analogy but also through ordinary sound change. The Gold numerals have a distinct Indo-European look about them since they mostly end in /-s/.

Table

Numerals

	Play	Late Andanese	Gold	Leaper	proto-Dreamlandic
0
1					(pi-)huanči
2	nūs	nuka	nūs	LEAPER	anupia
3	pama	hama	gʷàma	LEAPER	amia
4	pās	hika	gʷās	LEAPER	ipia
5	putas	hia	gʷutas	LEAPER	alupia
6	putamas	alukimaka	gʷutamas	LEAPER	arumia
7	batanās	apunuka	gʷatanās	LEAPER	kuana
8	fuppās	kulika	hʷukpas	LEAPER	ihu
9	butapās	kukuluka	gʷutapās	LEAPER	kuhuapa
10	mabumās	mukuka	magʷumās	LEAPER	muahu

The Gold words above are those chosen to best match the Play descendants, and were not the only forms in use. For example, the initial /gʷ/ in some words could revert to its original /h/, matching Andanese, in some constructions, and this was part of an alternation still alive at the time. Essentially it distinguishes whether the words were seen as extended plural markers or as nouns in their own right. Thus Play appears more conservative than it is.

The Leaper words will be almost identical to those in Gold, but as above, may descend from different forms because they were variable at the time.

Dreamlandic diverged earlier than the others, but because the numbers are old, it can be considered to have branched off at the exact same time. The only way in which it appears as an outlier is that the numbers are treated differently in the grammar. For example, only the numbers 4 and 5 attained the -pia postclitic that evolved to -s in Play and to -ka in Andanese (in the number for 2 it was part of the root even primordially). They may also be the heads of their noun phrase, whereas in Gold they were not and in Andanese the distinction wasn't meaningful. All numbers from 2 on upward are prefixed with u- ~ w-, which is why the bare vowels did not fall off. This changes for classifiers. The numeral for 1 also takes classifiers.

Possible extensions

These are mostly etymologically unsound, though some are based on cognates.

• buta(s)- "plus five", by analogy between the numerals for 9 and 4. This is helped by the similarity to the numeral for five and by Players' familiarity with opaque alternations of initial consonants between related words. But Play does not use prefixes of any form, so the speakers would need to also assume that numeral words belong to a special class. Note that the insertion of /s/ here is because /sp/ > /p/ in all Play words.
• fup- "double", by analogy between the numerals for 8 and 4. See above about the lack of prefixes in Play, and note that a dual suffix -bup already exists. It would thus have to be that again numerals are a special class that go even after suffixes and yet can stand alone. See also below for a reason why fup may fit well up front after all.
• -ma(s) "plus one", the only etymologically sound derivation, though one that would be more likely if there were at least some lingering use of /pa/ to mean a pair of something. Moreover, since /sm/ > /s/ in all Play words, it would have to be that the final /s/ in /putas/ was already being analogized as a genitive even though other /s/'s were not.
• Pabappa uses the word tam to mean "cardinal numbers", but it was probably a noun, and not a morpheme that actually appeared on the number words. This word was added at a time when the words for the numbers were much shorter.

Other assorted number words

Those in Play

• -(t)eu "to agree"; used to form expressions like "both", "all three", and so on. It is possible that this is actually a prefix. Note that this might be redundant with the "all" below, but also that the form below is construed as an ordinal. If /teu/ is a prefix, the redundancy might be tolerated.
• -šafu- forms fractions; attaches to locative, which can by itself also signify fractions. Play uses the same word for "fifth (part)" and "fifth (in line)", but can disambiguate when needed. /šafu/ is really šaus and may appear as such if it is not itself padded with a second locative suffix.
• tap "to clap the hands"; used for multiplication by ten. Possible replacement for inherited mabu ~ mapup, which is transparently derived from the numeral for ten.
• fa, another word for ten. This is simply a doublet of /mabu/, deriving from a variant form of the word 4,000 years earlier in which the analogical medial /g/ ("to keep the rhythm") was not retained. That is, mʷŏgu ---> /mabu/ and mʷŏu ---> /fa/.

• This is the same /fa/ that means "cloud", "to hide", etc. and thus also appears in the words for zero and for "each, every".

• -bup, the dual ending. Appears irregular but is actually regular (c.f. /bip ~ pip/ "sun" for a similar alterntion).
• -bu, the "separate" plural ending. Originally a doublet of -bup (hence the comment above about it appearing irregular), but evolves to an independent morpheme by the time of Play. The oblique stem of both is pu-. The meaning of "separate" plural may have evolved from a construction like -bem-bu.
• -bem, the "working together" plural ending. May be also a collective. The oblique stem is mi- and survives as such into Poswa and likely Pabappa.
• -nem, meaning one, and may behave differently than other numerals.
• nasu ~ nafa, meaning "all, whole, entire". At least in its /nasu/ form, it changes meaning to "expand, make larger" in Poswa. May have a variant form /namu/. Note that Poswa gets its word for "each, every" from /fau vis/, where the /fau/ is yet another variant of /mabu/ (and probably should be replaced in the lexicon by either /mabu/ or /fa/). See below, as it is possible that the /fa/ may be separable.
• ta, used in the words for zero (pume fata) and "each, every" (nafata). At least in the case of /nafata/, it is construed as an ordinal ("the all'th place; all of them") rather than a cardinal.
• pume, a Play-specific word for zero, but cannot be used alone. Cognate to pupi "polar opposite" and therefore likely found only inside Play.
• na, the palm of the hand. Used to plump up the word /fata/ to mean "each, every". Thus /nafata/ is a very succinct way of saying "that which is hidden in the palm of the hand, (now) shown". It can only be formed inside Gold because of the monosyllabic word for palm. It is possible that this is perceived as standing in for a numeral, and that when speakers say "all seven" and so on, the /na/ is dropped so that the ordinary number word can stand in. See also the /teu/ prefix above.

Gold and others

The very old Play morpheme /fa/ appears in proto-Dreamlandic as mua. The expected variant reflex /mpua/ does not appear at least in the numerals up to 10, but may appear elsewhere as the morpheme was flexible even in the MRCA.

• duʕ is the Gold word for zero. It means "in the /dù/" where dù is a word for something sacrificed or cut off, and also functions as a negative in some other constructions.

Morpheme order in higher numbers

In Play, the word for "times ten" would almost certainly be placed in initial position, unlike IE. Rather than having two roots for ten, Play had a root word meaning ten and a root word meaning "a group of ten", rather as English has "twelve" and "dozen". Put another way, Players would see 60 as six tens, not ten sixes, and thus the word for 60 would not be analyzable as "six [times ten]" as in English. Rather, it would be something like fa(p) or tap followed by the ordinary word for six. It may or may not have a mandatory classifier suffix. ša(p) is also possible, but less likely.

Possible better nominal possession marker

WHY THERE ARE NO PRONOUNS AND NO PERSON MARKERS

10:31, 11 February 2022 (PST)

Play nouns can take up to FOUR person markers: possessor, agent, patient, and beneficiary. No other language, even Poswa, has all four of these inflections marked on nouns. For example, from fapasa "pillow", one can say

Fapašavaapās "the pillow which you bought for me which (got) you"

which combines four person markers on one word. The 1st person is marked twice, once for possessor and once for beneficiary, but in English it would be uncommon to repeat it twice. Since there are only three persons in Play, any noun with four person markers on will have at least one pair of markers that is the same. The verb in the sentence is not present here, but implies that the listener was affected, as for example if the pillow had fallen on their head. Nonetheless fapašavaapās is a noun, and can be the subject of a sentence.

Note that it is still possible to analyze Play as having no person markers because all the morphemes that here mark 1st person really just match the possessor marker, and since that is the same marker, it can be claimed that there is only a "self:non-self" distinction except for the patient markers.

The possessor also serves for "i am a doctor", etc, since by context the speakers assume a state of being instead of a relation of ownership. Therefore the possession marker has nothing to do with the gluons, unlike in Poswa where the counterpart of gluons is considered an augmented possessive. This means that the "verbalizer" is the same as the possession marker except that the possession marrker is padded by a classifier suffix.

In Play, -p -s -Ø can mean both "I am a ____" and "I have a ____" depending on the classifier suffix (if present), and that in both cases, they are attached to the oblique form, which is the same as the Poswa B-stem, and was analyzed as a verb. Therefore there is no need for "another" verb to serve as the nucleus of the sentence.

Even if the paradigm in Gold was originally -k -s -ḳ, there would have been an unmarked form. Thus Play has -p -s -Ø because it gets its third person from the unmarked form, in essence substituting "a book" for "his book", etc.

The outermost markers are in fact etymologically derived from true person markers, but because the 3p form is Ø, Play teachers analyze them as being like the others and distinguishing just self vs nonself (with the speaker as "self").

Gluons

08:34, 29 January 2022 (PST)

It can be that the Gold accusative markers were -k -s -ḳ for 1st, 2nd, and 3rd person, meaning that they were inherently marked for person. These could have been possession markers too, although their basic meaning would have still been verbal, .... that is, "I have a ___". Genitive -s is older than Gold, though.

Vowel-initial roots

If a verb /a/ existed, its infixed forms /aka aha/ etc would merged to /a/ when suffixed, meaning that /a/ could be a verb.

The four gluons

08:18, 23 January 2022

Gold was able to indicate the person of both the agent and patient with just one morpheme, and that one morpheme took only four values: -k -Ø -s -ʕ for 1:1 1:2 2:2 2:1.

Play then changed /ʕ/ to /m/ by analogy since it had also changed a noun case ending in -ʕ to -m. Thus, these markers have nothing to do with the "odrty feet" markers that came from swallowed noun classifier prefixes, although by happenstance there is an /s/ in 2nd person in both and the /ʕ/ that gets replaced by /m/ is a passive just like /m/.

This setup still leaves -l -n -ḳ free to mark other things on verbs in Gold, so it could be that while Gold could indeed get by with just four person markers (including null), they used more even so.

The 1:1 1:2 2:2 2:1 probably correspond to 3:1 1:3 3:2 2:3 since the ohlny difference is the presence of an additional argument in the sentence. Then, there would be 3:4 and 4:3 for /Ø ʕ/ and reflexive and distributive for /k s/.

Evolution into Play

Even in Gold, the /ʕ/ phoneme had already mostly merged with silence. It may have hung on as a verbal marker indirectly by affecting tone. But Play lost its tones too, and therefore it was no longer possible to distinguish /ʕ/ from silence. Thus, Play retained the Gold verbal markers but they wre no longe sufficient to stand on the ir own. This is why Play ended up sticking evidential morphemes on, which were unrelated to the Gold verbal markjers. They were all third person verbs, maninfg that their marker was silence all along. Thus, there was no need to "inflect the inflections" in Play.

This means Play loses the irregular tense markers as well. Also, it means that Play starts out by using the reflexive verb markers for 1st and 2nd person, but the transitive verb marker for 3rd person. It may be that Play has a confusing system in which old, new, and ephemeral features are all mixed together. Try to avoid using evidentials as object person markers, however, since they need to evolve into subject person markers in Poswa and Pabappa.

Possibility of verbal embedding

If these verbs, which are both tenseless and seen as personless, are embedded within noun phrases, monstrously complex inflection paradigms could appear, mostly ending with /s/ plus a vowel. Essentially, rather than behaving as content morphemes, these incorporated verbs behave as inflections. Though very difficult, the system is maintained because Play's small syllable inventory ensures that there can only be so many different inflectional paradigms to memorize. Thus simplicity preserves complexity.

For example, from fausa "door", one could derive faufuufupusa "the door you opened for me", all without any markers for person, number, or tense. If such words were commonplace, it would help explain why Play was so notoriously difficult to learn. The internal meanings of these morphemes might take a few steps in their own direction, for example /s/ could be a reciprocal/distributive instead of being tied to just the 2:2.

Moreover, this would mean that Play has a feature which Poswa does not, and therefore that the evolution of Play into Poswa was not simply a linear progression in which more and more features were gained while none were lost.

Further examples:

1. pupumuba "toy blocks" + pata "play" =

   pupumuatasa "the toy blocks that I play with". The elision of /p/ is optional here because etymologically it was dropped when it came from /p/ but retained when it came from /kʷ/.

   pupumuataasa "the toy blocks that you play with"

   pupumuatašasa "the toy blocks we're playing with together"

   pupumuatakasa "the toy blocks we're playing with individually"

And so on. Note however that there is no way to distinguish "we" from "I" in the fourth example, because it is a true reflexive, whereas in the third example, it is a reciprocal and therefore must be plural.

Use with third person and unmarked arguments

Note that it is still possible to say, for example, sapupaipa "swimming fish", which is /sapu/ + /pai/ + /pa/. Importantly, the word for fish is in the nominative and not the oblique stem.

This looks like it can also mean "the fish I swim", but this is just because /sp/ > /p/ and the forms overlap. When the classifier suffix is /-ba/, this ambiguity is resolved, but note that handheld objects typically only take intransitive verbs. It is likewise for other classes. Thus for example pūpu "palm tree" becomes pūinupu "tall palm tree"; here, howver, a passive is possible: pūinuppu "palm tree taller than me".

Private verbs

Some constructions can be made with private verbs, which are monosyllables inherited from the ancient Tapilula language, mostly obsolete already in Gold. Play preserves the system, but sound change has worn down most of the consonants so that the private verbs today are mostly just single vowels. This means that their meanings are extremely broad and must be understood from context. For example, a has over a dozen meanings, including the bare possessive "to have" (although it is possible that Play had a zero-marked possessive construction, as if there were a silent private verb, that was replaced in Poswa and Pabappa by /a/).

Nonetheless, just as English has verbs such as "get", "do", "run", and so on, so too does Play have formations such as these:

1. sapūaapa "the fish you bought for me", showing that they do not always end in /s/ + vowel.
2. namaŋašasa "the chair you're sitting on".

The private verb va typically means "buy" when used with the handheld object classifier -ba, even though this /b/ is always swallowed by the infix /s/, and therefore not transparent. Even here, though, it can also mean "throw" or "clean".

The only consonants permitted in the private verbs are /m n ŋ p/ and the semivowels /v y/, which Play teachers consider to be allophones of the vowels. The /p/ comes only from /gʷ/ and not from primordial /p/, though some conflation of meaning has taken place since there are freestanding words /pa pi pu/ that can serve as verb stems.

Possible null construction

05:23, 25 January 2022 (PST)

There may be a null construction as well, for example, from fabu "to share one's war strategy" there can come fapupuse "the war strategy you shared with me". Note that it uses /pu/ instead of */u/; the "cards" here are /a i pu/, not /a i u/. The final -e represents the classifier suffix /-be/ "speech, song".

It could be said that the null construction above does not have a null verb, but rather a null stem. That is, the construction is Ø + PRIVATE VERB rather than NOUN + Ø.

Strictly speaking, however, it seems that a silent /ʕ/ should precede the construction, even though such a phoneme could never have stood alone. Even if it were analogically created, this would lead to a great many vowel-initial words, which occur nowhere else in Play, so it is likely that the construction does not exist, and that Play must use simple verbal embedding instead. The passives below such as "you took from me" would become actives.

Examples:

1. vumaaba "the handheld object you took from me" (/vuma/ "to take away" with /-aa-/ replacing /a/, thus as if representing a single "card" infix /-a-/).
2. tamavumaaba "the leash you took from me", as above, assuming that the null construction can be analyzed as Ø + PRIVATE VERB and therefore take an additional morpheme to fill in the gap.

Consonant clusters in etymologies

/st/ and other clusters

Remember that while /st/ appears frequently in the etymologies of words given in Play, this cluster did not occur in Play itself. Rather, it appeared in Poswa and Pabappa after those languages independently underwent vowel syncope, creating the opportunity for new compound words to be coined from Play words containing the cluster /st/, which previously would have been automatically reduced to /t/. This also applies to various other clusters, such as /pf sf sp ss/ and perhaps any cluster with a fricative in either position. Note that /pp pt mm mn/ were etymologically sound, but that /pk mŋ/ were not, so /pk mŋ/ would only appear in late coinages and therefore can be replaced by /pp/ arbitrarily in etmyologies. (The intent here is that /pk/ would only ever arise from the sequence /kʷk/, which would have become /kʷ:/ and thus /pp/.) There was also a rare change /kʷd/ > /kʷdʷ/ > /ptu/.

Later /st/ can either reflect Gold s + t or Gold s + d; these were pronounced /tʰ/ and /dʰ/ respectively, but since all voiceless stops were aspirated, this is just /t/ vs /dh/. Then, Play turned these into /t/ and /Ø/ while in Leaper they merged as /t/.

Nouns

See Pabappa/scratchpad#Scratchpad_ordered_by_date.

Dirty feet

It is likely that Play and Pabappa both lack pronouns just as Poswa does. Instead, 1P and 2P patient is marked on the agent of the sentence, using captured prefixes from the Gold/Trout era. The Gold 1P patient morpheme was -ŋa and the 2P patient morpheme was -hə. The "feet" of the nouns thus get "dirty" with the otherwise lost prefix of what had once been the following word. (Although the metaphor could just as easily be soap, so long as it's something that's sticky and has weight.)

Thus the dirty forms of the nouns are, for 1st person, -a -i -u -e > -am -īm -ūm -am, and for 2nd person, -a -i -u -e > -ās -īs -ūs -ās. Note the asymmetry and that the final -s can be lost in all four examples.

Accusatives could also be padded .... e.g. -p changing to -pu when the agent is plural (and also perhaps -ptu for /du/, which could indicate a single boy). -pi would be a 3rd person singular female agent (possibly also plural all-feminine) and either -pa or -pte for a 3rd person singular male. In theory the 1st and 2nd person agent could also be marked here, probably with -a and -e respectively (that is, the same vowels as are used for the patient forms), but since person is already marked on the verb, these markers would either be redundant or would be confined to peculiar verbless constructions. (Poswa does not have this problem because the verbal person markers are identical to the padding on objects in verbless sentences.)

There is also potentially a -ŋu when the patient is 3rd person but a child, and -a ~ -e if the patient is 3rd person masculine singular. Lastly, the feminine would be -i. These are etymologically sound, but it may be that the 3rd person patients were never marked, since nearly all such sentences would have the word for the patient explicitly present anyway. But consider that if gender were retained for patients (even if not for agents), these could be used to allow the speaker to avoid repeating words. This system is most likely to be retained if Play had a means of marking the gender of the agent on the verb.

These endings are similar to what later evolved in Poswa .... where inanimate (and 3rd person) agents are padded by person markers for the patient, and must take passive verbs, and in which accusatives are also padded in some constructions with the person markers for the agent (because there are no pronouns). It could be that Poswa's system arose as the Play system began to break down through sound change. Unlike Poswa, however, there is no tense marking.

Note that this is SLIGHTLY similar to proto-Semitic, but not quite the same.

Possession on nouns

The means by which Poswa and Pabappa mark possession on nouns evolves from the Play oblique case, which likely no longer stood alone with any meaning even in Play. It is possible, even so, that the 1st and 2nd person possessed forms could simply be derived from the oblique followed by -p and -s respectively, if assuming that these would not be confused with case-marked free forms.

Alternatively, the OBLIQUE + -a- + verbal person marker construction (whether with /-p -s -Ø/ or with longer forms) existed even in Play. But note that this construction exists in Poswa and Pabappa primarily because it participated in a sound change which made it much shorter; it would have been rather inconvenient in Play as it could have made words two or even three syllables longer just to mark possession (e.g. /tap/ > /takaakas/).

It is possible, also, that instead of /-i/ > /-ip/, it would be /-iši/, which respects the true etymology, which means that it would collide with the reflexive past.

Verbs

Remember the -V-š-V reflexive past, which might be effectively a past participle. e.g. "he scratched me" etc. It corresponds etymologically to a merger of the past tense forms of /p/ and /s/, so it might come to mark tense alone.

The verb person markers are -p -s -Ø for first, second, and third person, despite the fact that these same markers also mean reflexive/reciprocal/direct.

This system cannot cleanly evolve into that of Poswa. Poswa gets its person markers from Play's evidentials, but the tense markers go inside the evidentials instead of the original suffixes. Strictly speaking this is bad syntax, but the speakers must have gradually lost their knowledge of the original meaning of the morphemes. (But perhaps it is not so bad, since Poswa ends up doing the same thing with its remaining evidentials: they always stack onto present tense verbs, regardless of when the action occurred.)

An infix of -ay- or -[V]y- to mark passive voice is possible, assuming that all passive sentences would have third person subjects and that person would therefore no longer need to be marked. (Though hypotheticals like -k[V]y- would be understood by the speakers as ad-hoc formations.) This -y- survives into Poswa, meaning it must have been inherited from Play, and could have been the present tense counterpart of the V-š-V morpheme above, assuming there were no person markers.

In fact, an early infix of -əy- is required to have existed for Lava Bed languages to do what they do. It was originally just a simple /i/, but schwa was added at an early date. The development to -ay- in Play is because of the /ə > e > a/ chain.

Pusiba system

It is likely that the person markers -p -s -Ø belong to one conjugation, and -pu -si -ba to an extension of that conjugation, where they have been padded with evidentials. This system may not have been fully solidified at the time of the unified Play language, meaning that there could still have been -pa, -su, etc, but the context-dependent grammaticalized sound change of *pb *sb > p s had taken place long before this time and was therefore inherited rather than innovated separately in the daughters. (Poswa discards the interior person markers entirely, and adopts /-bu -bi -ba/ for its person markers, meaning that the evidentials must have persisted as independent morphemes, capable of being added to 3rd person verbs, until at least 5500 AD.)

The V-š-V morpheme above could easily be reinterpreted as simply reflecting the genitive, turning the genitive ending -s into a past passive participle marker. This would lead to endings -su -si -š-V for the 1st, 2nd, and 3rd person past passive participles, although these would be best suited for daughter languages where /-si/ was not also the ordinary 2nd person present tense ending. Thus, this development might happen in side branches of Pabappa but is less likely to happen in the branches that begin diverging before the split of Poswa from Pabappa.

Otherwise, past tense could be infixed into the evidentials, making the forms for the past -pe -sei -bī. This corresponds to the /-be -bei -bī/ that gave rise to the system used in Poswa.

Other information

Diachronics

See Babakiam/Sound_changes for a convenience link to edit the diachronics.

Scope

Although the Swamp Kids ruled from Săla instead of Paba, it's likely that the dialects were interchangeable, since the entire Empire underwent massive internal migrations. Therefore, Bābākiam is also the parent language of the Swamp Kids' outposts in Amade and Pipatia.

Gold (1900) to Play (4100)

The Play language evolved from the Soft Hands dialect of Gold, also known as Wolf in Wool, Broken Shields, and perhaps at least one other name. It drove out the Lazy Palms language and took relatively few loanwords. There were also several other languages spoken in this territory, including one language spoken by Star immigrants, probably a branch of Amade.

Wolf in Wool had not yet evolved its characteristic sound, so the relative scarcity of loanwords was not due to the acoustics of the language, but rather a cultural identification with the new language being imported from overseas. Any loans that were taken in had /e o/ shifting to /ə/ for the entire time period of this language, though /ē ō/ may have been borrowed as /əi əu/ or /ai au/ or either.

1. At the end of a syllable, the pharyngeal fricative ʕ disappeared and changed the previous vowel to a high tone. It also voiced the following consonant.
2. Syllable-final k ḳ ŋ changed to kʷ ḳʷ ŋʷ.
3. Feeding on the above change, in compounds, if the final consonant was one of /kʷ ḳʷ/ and the first consonant of the next morpheme was one of the velars k ḳ h ŋ, it also became labiovelar. Thus for example /kk/ > /kʷkʷ/ or /kʷ:/. It did not happen for other consonants. Prenasals did not shift; later, the cluster /ŋʷk/ becomes /mk/, which is pronounced as spelled but later becomes [ŋk], [mpt], etc depending on dialect.
4. In initial position, the labialized coronals tʷ dʷ nʷ shifted to t d n. Elsewhere, even in clusters, they decoupled to the sequences tu du nu.
5. The bilabial approximant w changed to v (in internal reconstructions, also spelled "β") before a vowel.
6. Then l lʷ both became w (not */v/) in all positions although it retained a rhotic allophone. The distinction between this new /w/ sound and the one that had just changed to /v/ is important later on, as it keeps sequences like /ʕl/ from being corrupted to /ʕʷ~gʷ/ and then on to /v/, /b/, and /p/. Rather, /l/ stays as /w/.

   Notably, the sequence sl (which was pronounced as IPA [hl] or for some speakers [ɬ]) shifted here to sw, and did not become */hʷ/ or */f/. That is, it behaved as the sequence that it was morphologically, instead of sliding with the phonetics into a new single consonant.

   NOTE ON POLITICS: Proto-Highland Poswa breaks off here.

7. The labiovelar consonants kʷ ḳʷ hʷ gʷ became p ṗ f v unconditionally. This includes sequences like /kʷl/, despite the precedent set by /sl/ above, because in this case, /kʷl/ was already [kʷ] at the surface level in the proto-language.
8. Sequences of two vowels in which the first vowel was i or u became rising diphthongs. Then all clusters of a consonant followed by a semivowel came to be pronounced as coarticulated single consonants. Thus pua became pʷa, pia became pʲa, and so on.
9. Stressed syllabic nasals were opened to sequences containing a schwa.
10. The voiced fricative g assimilated to a neighboring glide /j/ or /w/, thus creating sequences of /jj/ and /ww/. The shift thus was gj jg gw wg > jj jj ww ww. This includes g after /ī/ and /ū/.
11. The voiced fricatives d dh g became silent between vowels and occasionally in initial position (due to compounding).

    When I wrote this, there was no /ž/ in the language at this stage, and so it is possible that ž also shifts to Ø.

    NOTE ON POLITICS: This time period is around 3100 AD, near the beginning of the "Time of Happiness" (Yeisu Kasu: 3138 - 3302 AD). The branches of the language that fork off from mainline Bābākiam in 3138 all die out, and therefore all of their names in the history are written in Babakiam, but they could be revived as minor local languages, and there would be quite a lot of them.

12. A voiced consonant in a cluster after /p/ or /s/ changed briefly to ʕ and then disappeared.

    This shift is responsible for important consequences in verb morphology in Poswa more than 5000 years later. Note that the inherited clusters gh hg had been merged as h already in Gold; /hg/ was morphologically equivalent to /sg/, which explains why /sg/ shows up in Play as š instead of s like the others. Lastly, this shift explains why the Play toponym Fanašasa corresponds to Leaper Xʷanaxanta.

13. The voiced fricatives v z ž g changed to b d ǯ ġ before a high tone. Unlike other languages, Play considered the long vowels to be high tones here.

    This is how Play does /g/ > /k/ even though /g/ was a fricative. Note however that in hypothetical words like /vuau/, where a /d/ dropped out, the initial /v/ was part of a separate syllable, not stressed, and so did not shift to /b/.

14. The post-velar fricative consonants ħ ʕ, which had been developing labial compression, changed unconditionally to f v.
15. The velar fricatives h g were fronted to š ž unconditionally. šʲ žʲ became š ž. This includes the /čʲ/ sequence, which had long ago become [šʲ] but was maintained in spelling because of its importantly distinct grammatical behavior.

    Importantly, this shift included conditions in hiatus ("holes" in Play terminology), so that čiva became čua.

16. The labialized voiced stops bʷ dʷ ǯʷ ġʷ changed to b.
17. The palatalized voiced stops bʲ dʲ ǯʲ ġʲ changed to ǯ.
18. Any remaining voiced stops b d ǯ ġ changed unconditionally to p t č k (except when in clusters).
19. The voiced fricative žʷ changed to v.
20. Tones were eliminated. However the stress accent (nouns on the penultimate syllable, verbs on the ultimate) remained and became regularized.
21. The voiced stops d ǯ ġ (now found only in clusters) changed to n nʲ ŋ unconditionally.
22. Remaining v changed to b.
23. Remaining z changed to s.
24. Newly created vowel sequences beginning with i or u collapsed into rising diphthongs, thus creating a new series of palatalized and labialized consonants.

    This same shift happened twice but many words missed by the first change were captured by this change. Note, however, that the reflex of /buya/ is still /buya/; it did not become /bʷia/ and then /bia/.

25. The labialized consonants bʷ žʷ changed to b unconditionally. (Despite the fact that a nearly identical sound change had occurred only shortly before this one, this rule was very common in verb forms that were created by the shift of /bua/ > /bʷa/ > /ba/, and likewise for other vowels.)
26. The palatalized consonants bʲ žʲ changed to ž unconditionally. (The above shift also applies here; many verbs underwent a shift of /bia/ > /bʲa/ > /ža/.) This shift did not apply to words such as bivu, from earlier /buivu/, because the /i/ in this word was not [ʲ] but still a true /i/.
27. A schwa ə in a word in which the following syllable had /a/ changed also to a. Note that this is the only vowel change in the entire history of the language going back 3500 years, even before the Gold language, except for a few diphthongizations such as /ua/ > /wa/. However, the vowel system became very unstable in the succeeding period as the language developed into Poswa and Pabappa.
28. The stress was shifted to the first syllable in all words.

Phonology

Babakiam is the parent language of Poswa and Pabappa and thus shares with these languages many characteristics.

Vowels

There are four vowels, /a i u ə/, spelled a i u e. The first three vowels can also be long. The schwa is the rarest of the four vowels, and words with schwa are usually cognate to words with clusters or syllable-final consonants in closely related languages such as Leaper and Proto-Moonshine, which are separated from Babakiam by about 2700 years of divergence.

Play is notable for allowing unrestricted vowel sequences, dominated by /a/. All vowels can stack, but /i u/ contract into /j w/ (spelled y v) when bordering other vowels, and the schwa vowel e is rare, so the effect is most noticeable with /a/.

Because of grammatical embedding, it is trivial to create words with five vowels in a row: two for the root, two for a grammatical infix, and one more for an inflection (always the same as the preceding vowel). Additionally, a classifier suffix might also begin with a vowel, and sometimes the root word would have more than two vowels in a row, at least in its oblique form. An example with seven vowels in a row is taaeaaaa "the boy he borrowed".

Three-vowel chains were common in everyday speech, and four were occasionally heard, but substitutions were typically used when five or more vowels would appear in a row. For example, in the word above, the infix -aa-, here "borrow", is more commonly expressed by the synonymous -muna- when the root word already ends in /a/ or /e/. The result would be taaemunaaa, with two three-vowel sequences.

Consonants

The consonant inventory is very simple: /p b m f t n s š ž k ŋ/, but note that /w j/ are considered allophones of the vowels. It is unusual in that it lacks liquid phonemes entirely when all the languages around it have /l/ and most also have an /r/-like sound. Thus Babakiam sounds like stereotypical toddlers' speech. Bilabial consonants are by far the most common, as in Pabappa, Poswa, and the Outer Poswob languages. However, Play is not as extreme as its descendants, which are almost entirely free of dorsal consonants.

Phonotactics and sandhi

Most words end in vowels, but can also end in the grammatically feminine consonants /p m s/.

Sound changes

Babakiam stands apart from other descendants of the Gold language by its labial-friendly phonology. But it also stands apart by being extremely conservative with sound changes applying to vowels, having only one change in 2200 years, and that a rarely seen polyconditional one that shifted accented ə to /a/ when followed by another /a/ in the next syllable. Thus Nəma, the name of a large empire, became Nama in Babakiam. (This schwa vowel is normally written e in Romanized Babakiam, so the old name would have been Romanized Nema.)

Comparison of words

4200 Babakiam peskavu sabayiuŋaus

6000 Babakiam pyskary šalergos

8700 Poswa pwaršalios

8700 Pabappa pospalerba "soap bubble wand"

Orthography

see history for older informaiton

scratchpad

Andanese /la li lu ha hi hu/ = Play /ba bi bu ža ži žu fa fi fu ša ši šu/. The matching of the syllables may take advantage of Play's late sound shift of /bʷ žʷ bʲ žʲ/ > /b b ž ž/, such that Andanese syllables could be matched to Play consonants. For example perhaps /li lu/ could spell /ž b/, though this leaves open the question of how to read a plain /l/. Remember that Play did still have /žī bū/ etc.

since nearly all /ž/ in Play is from CVC collapse, it is overwhelmingly likely that the L glyph will "lean" to /b/.

Daughters

See Cosmopolitan Play languages, Cupbearer Coast, and possibly more.

History and culture

See Play language/history.