Sajem Tan Listserv Logs

Would anyone be interested in attempting the project suggested here: https://listserv.brown.edu/archives/cgi-bin/wa?A2=ind0905D&L=conlang&D=0&P=29735 ?

Each collaborator has a specific role.

Thunder is responsible for the phonology. Rain is responsible for the morphology. Spider is responsible for the syntax. Flower is responsible for the semantics. River is responsible for the pragmatics. Stone is responsible for the orthography. Bee is responsible for the corpus. Bear is responsible for pedagogy. Bird is responsible for supervision.

Any further roles will be democratically determined by the collective.

How exactly would morphology and syntax be worked out independently? (And there are other very tricky pairs)

Does having Semantics separately mean the morphology guy will only create affix forms and such, while what they actually do is another person's job? (The same goes for whoever creates actual lexical entries)

I loved Brett's proposal and would like to see something like it implemented.

I suspect we'll have less than nine active collaborators, so probably some people should have more than one role, or we should have fewer roles with more responsibilities each. Also, some of these roles would have more work to do in early stages, some more work in later stages; Stone, Rain and Spider can't do much until Thunder's work is more or less done, nor can Bee or Bear do much until Rain and Spider are done. If the number of collaborators grows over time (unfortunately past experience suggests the opposite is more likely) then we can divide up the roles more later on.

Furthermore, perhaps the people with those roles shouldn't have *sole* responsibility for those areas, but rather a dispute-settling authority, making a final decision when there is no clear consensus or a tie in voting.

There was a paper on Fiat Lingua a while back on the sociolinguistics of collaborative conlangs. It might be instructive to (re)read that before starting a new project, to see if we can avoid some common mistakes.

http://fiatlingua.org/tag/gary-j-shannon/

Yes, this seems better. Let's do that.

This is intentional. It manages the workflow by making the language in stages.

I shall explain it with a mythic history of Common Honey:

So very long ago, during gray and overcast days past, Thunder rumbled. From the shaken sky tumbled Thunder's noises and with them Rain. By the fallen Rain, Flower was nourished and River was filled. Upon its web, between Flower and Stone, Spider gathered Rain. When Rain ceased and Thunder silenced, Bee gathered Flower's pollen. Bee then flew away to its hive to make our Common Honey. Upon the hive came Bear who so loved our Common Honey that it shared with all who would sample. Witnessing all this, from atop Stone, Bird declared, "May this recur until all our tribe partakes of Common Honey."

I was a little bit confused with the zebra thing, but yes, I'd be interested in doing this!

Which role would you like? What taboos and norms should our tribe follow?

In my previous post, I made a guiding mythology about the language and why it is called Common Honey. This should help produce a shared identity which fosters language use.

I'll join!

I volunteer to be stone.

-Daniel

Hey! That's me! Lucky me, I just happened to read this.

I'd like to be Flower, please.

It seems to me that since morphology and syntax have to go together that that means Spider and Rain will have to be friends!

It would be very difficult not to mention controlling if I were to try to decide the meaning of every word, but what I could maybe do could be to help decide how the meanings should come to be, perhaps create a system to record them, things like that.

<3,
Mungojelly/Flower

One of our medium-term goals should be to translate this creation myth into Common Honey. That gives us some structure for the concepts we need to lexicalize early on, etc.


It seems that Flower need not create all the words, but should coordinate the creation of words so they all harmoniously divide up meanings among them in ways that are interestingly different from other languages. For instance, Flower might decide that mental state concepts are lexicalized as adverbs to be used with a "feel" verb -- or as intransitive verbs with an indirect object being what you feel that way about -- or as transitive verbs with direct object ditto, or whatever. Then various other collaborators can make up individual mental-state words within that framework. And I suppose Flower would be responsible for maintaining the lexicon.

I'll volunteer to be Rain and/or Bee. I have a little experience maintaining corpora and doing frequency analysis etc. on them.

I suggest that Bee's duties should include collecting all the casual utterances in the language, as well as more formal texts, from the mailing list or forum and putting them into the wiki, then periodically posting an updated frequency analysis of the corpus as it grows.

Now this makes a more reasonable amount of sense. May i apply for either Rain or Spider?

I can be thunder or possibly river. I am also rather new to linguistics and conlanging, just thought you should know.

You'll be in charge orthography, including Romanization. If you like, you can make a neography/original script.

Since you were the first to mention the idea of taboos and mannerisms, what did you have in mind for this set of roles?
Off-topic question: Are you mungojelly of Lojbanistan? If so, you should make a new Lojban video.

Agreed. Also, feel free to rewrite the English version if you want it to flow better or if you want to elaborate on it. I'm no writer!

The role of Bee will most importantly be to encourage original literature and poetry (or to write it, if need be).

If Jim chooses Bee, then Rain is available; if he chooses Rain, then Spider is available.

Both are currently open, so you are free to choose. If nobody else joins the project, you can have both if you like. No experience is necessary; this will be a learning experience for us all.

I can do that too.
I'll take Bee, and if we don't have more volunteers soon, I'll take whichever of Rain or Spider that Guilherme doesn't want (seeing that Bee won't have much to do in the early stages when Rain and Spider will be busy).

This sounds really interesting! If there are any slots left open, I might be interested in filling the role. Rain sounds particularly interesting to me.

Scott

Guilherme has his choice of either Spider or Rain. Whichever remains, you may may have.

Should I be Bird? Or should I leave that to someone with more experience with this mailing list? In any case, as soon as all the roles are filled, we should begin the language.

A suggestion to the tribe of collaborators: one word should cover "language", "honey", "vomit", and "speak/talk". My reasoning is that our language is a collaborative effort like making honey and its results should be just as sweet, figuratively speaking. And since honey is bee vomit, it seems vaguely appropriate by metaphorical extension for them to be a single word.

That sounds like a really interesting and unique way to build a language. Unfortunately I really won't have any time to join in with it....still not sure how I'm going to manage time for word creation for Lexcember. Would love to see updates about how the creation goes though.

Considering it seems i really have to choose, i'll go with Spider. Bear in mind i wont have much free time for the next two weeks, but after that it's a huge blob of free time until mid-february (also known as holidays).

It is granted.

If I counted correctly, all roles have been filled. Proceed to discuss amongst yourselves and post your preliminary thoughts.

I don't see anyone having volunteered as Bear.

Preliminary Thoughts:

An alphabetic system with separate characters for common morphological components.
-ed = %
-ing = *
-s = $
the = @
"having walked the dogs" -> "have* walk% @ dog$"

I'm uncertain about the inclusion of "the" above. Naturally, whether or not this would work would be very dependent on the morphology.

If you think this wouldn't work, I'd be fine using this for something else and coming up with a different system.

What I got for the roles so far:

Thunder (phonology): Samantha
Rain (morphology): Scott
Spider (syntax): Guilherme
Flower (semantics): Brett
River (pragmatics): Samantha (possibly)
Stone (orthography): Daniel
Bee (corpus): Jim
Bear (pedagogy):
Bird (supervision): Shanoxilt (possibly)


Let me know if there is a mistake.

I think we can wait a while to recruit Bear, since their work can't start until several other people have done a lot of their work. Possibly Thunder, Rain, Spider or Stone could take over as Bear after the major part of their work is completed.

I like the idea.

I suppose you mean, whether the system should have separate glyphs for common grammatical particles as well as for morphological affixes? I don't see why not. On the other hand, with the equivalent of "the" specifically, I'd prefer to mark definiteness morphologically, or to not mark it at all and mark referentiality instead (or a system of orthogonally marking both definiteness and referentiality such as someone proposed here a few weeks ago).

More broadly, I'd suggest that this collaboration method is perhaps better suited to an abstract artlang than to a naturalistic one. The method is designed to grow a "conculture" as a real-world subculture, and it may not need a fictional conculture as well, with the complex irregularity and diachronic history that would generally go along with that. If it's unnaturally regular, we are more likely to learn and use it, and Bee is all in favor of using the language a lot. On the other hand, "unnaturally regular" need not mean "as simple as an auxlang"; it should have enough weirdness to make it interesting, but not so much as to make it too difficult to use.

Secondly, I'd suggest that our initial goal should be for Thunder, Rain, Stone and Flower to have enough groundwork laid in the next few weeks that those of us who are doing Lexember can coin words in Common Honey as well as our personal conlangs in the last few days of the month. If Spider has enough basic syntax in place for people to write example sentences for their new words, even if they are tentative and subject to revision, Bee would be delighted.

Some suggestions for Rain and Flower:

Let's go over a list of morphological categories and see which ones we want to mark obligatorily or optionally or not at all. I'll tentatively suggest that verbs be marked for mood, polarity, aspect, the speaker's attitude to the situation, and (for verbs of motion) path/direction. I'd prefer not to mark tense, but if so, let's distinguish recent past from remote past.

Let's figure out how we want to handle motion verbs. Do they primarly mark path of motion or manner of motion or the shape/consistency of the moving/moved thing? I'd suggest having the basic dictionary form of verbs mark manner of motion, and add a direction/path affix in actual use. E.g., "gather" (one of the words we'll need for the creation myth) could be a basic "to put/to set in place" verb plus a "from all directions" affix.

Are our names for animals and plants going to be cladistic, or based on a folk taxonomy of our own devising? We'll obviously need several words for animals and plants early on. I'd suggest that we translate the bee in the creation myth and the title Bee with a word for honeybee specifically, and have another word for bumblebees, orchid bees and others that don't produce honey.

On the contrary, I think that Bear can begin as soon as the others. Teaching it while it is still in flux will reveal what needs to be fixed.

Why not both?

I love this idea, deeply and truly. I will tuck it away for future use. I hope that the roles help keep the project rolling a little better than they usually do.

Any room for spectators/observers? I don't really have enough time for my own conlangs lately, but I do love to watch the collaborative process unfold.

Mia.

So, do we just begin discussion ideas? Or is there a more formal process to this?

I see the instruction

So I think the answer to your question is "the former".

I'm thunder (phonology), so that means sounds? And patterns. So, does anyone have any favorite sounds that they want in the language?

Yes.
I'm fond of semivowels and consonant+semivowel clusters.

I've always been fond of lateral fricatives. And the voiced post-alveolar fricative /ʒ/ is always a favorite of mine. But I'm not adamant on anything. :)

Sure! But don't hesitate to contribute if you like. We also need learners of Common Honey.

Just begin. Any formalities we need will develop organically.

I like /ks/ .

If we have that, let's have /ts/ and /ps/ as well (and maybe /qs/?).
Do we want them initially / finally or only medial?

Are /ʒ/, /ks/, /ts/, /ps/ and /qs/ IPA or something else?

I'd bet so. I like /θ/.

Considering i take care o syntax, would you mind if this was OVS? All my languages (excepts for my first, which was rather sad) are either SOV or VSO and i don't really feel like doing an SVO just now.

I think OVS would be interesting, I don't mind what it is

I like them initially, but I'll go with whatever popularity dictates.

That seems good to me.

Does anyone have any ideas about the roles beyond the language collaboration?

Yes. /ʒ/ is the consonant sound in "Asia". /q/ is an uvular plosive like in Arabic, like /k/ but further back in the throat.

This looks like a good resource for seeing the IPA symbols and hearing the sounds pronounced in mp3 files:

http://web.uvic.ca/ling/resources/ipa/charts/IPAlab/IPAlab.htm

Bee likes OVS well enough.

Thunder: OVS means object-verb-subject word order.

O Swift feathered one, Bee thinks it fitting that Bee should address the swift feathered one (and refer to them) by contrasting Bee's featherless wings and ambling flight pattern with the swift feathered one's. Bee apparently also talks about themselves in the third person, for less obvious reasons.

Rain: we will need derivational operations for deriving "feathered" from "feather" and "feathered one" from "feathered". The first could perhaps also derive "wet" from "water" or "liquid", "rich" from "money" or "wealth" and so forth?

Flower: what other senses might our word for "swift" have? What about "feather"?

A proposal:

Thunder, Rain, River, and Stone always speak in the first person.
Flowers always speaks in the second person.
Spider, Bee, Bear, and Bird always speak in the third person.

Question: Do first-person inanimate pronouns exist? If so, they would be perfect for Thunder, Rain, River, and Stone. Maybe we could have noun classes emerge from our roles.

Personal request: Could we skip the usual translations of the Pater Noster and the Tower of Babel?

Spider doesn't do Pater Noster and Babel much. Spider prefers The North Wind and the Sun. (we already have thunder and rain, so why not wind?)

Do Thunder and Rain have any ideas already? Spider'd like to build from that.

Bee approves, on the whole, but is unclear how Flower always speaking in the second person would work. Does Flower address the subject of Flower's every sentence as "you"? Or only Flower themselves?

So are we having animate-inanimate noun classes, or nine noun classes based on the nine roles? If the latter, then it seems that Thunder would use a first-person thunder pronoun, rain would use a first-person rain pronoun, etc., while Bee would et alia would use third-person pronouns of the appropriate noun classes.

If we are to have nine noun classes, it will probably make learning the language much easier if the classes are phonologically based, and do not require memorization of each noun's class. E.g., words ending in a high vowel are of River class, words ending in a liquid are Rain, words ending in a voiced plosive are Stone, etc.?

Bee does not promise to never translate them, but will not make them a high priority. Bee prefers original writing to translations as a general thing.

I imagine it would be like reading a recipe or an instruction manual. I'm not sure what the second question is asking.

I'll leave that to Flower and Rain to decide.

Another proposal:

Our tribe/language should have a lots of symbols representing it. They would include, but not be limited to, a flag, a seal, an emblem, a motto, and an anthem.

The motto could be based on this passage from Jonathan Swift's _The Battle of the Books_:

"...[W]e have rather chosen to till our hives with honey and wax; thus furnishing mankind with the two noblest of things, which are sweetness and light."

Possibly some polysemy or homophony could give us a motto that means both "Honey and wax" and "Sweetness and light".

So, using the example I had before, "walked" would be "walk%", but what about irregular conjugations? Should "went" be "go%", "went", or something slightly more complex ("went%"?)

Perhaps Flower would address themself as "you", the rest of us as "y'all" (or some equivalent), and individuals by name/title.

If we have irregular conjugations (which Bee would prefer to avoid) then Bee thinks writing "go%" to be read as "went" would be the most logical way to apply the orthography you described. A morphogram would signify a value for a grammatical category, or a derivational operation, whatever allomorph it's expressed by phonologically in a given context.

But Bee would prefer to avoid irregularity, unless it grows into the language and we don't notice it's irregular until we (or at least some of us) have already learned the irregular bits. Then the morphograms would signify a morpheme (or cluster of frequently collocated morphemes) with an invariant phonological form, or a form that varies predictably with its phonological envornment (sandhi, elision, etc.).

This makes sense to Bee.

On that note, Bird wonders how many valencies the average Common Honey verb should have. In Lojban, most words have two or three arguments in the predicate.

Bee thinks that at most two arguments should be denoted by word order, one before the verb and one after, and an open-ended number of other arguments should be marked by adpositions or case affixes. OVS word order suggests postpositions. Bee also suggests that most or all arguments should be droppable when clear from context.

Traffic in this thread has been low for a few days. Are Thunder, Rain and Spider busily at work behind the scenes? Or busy with Real Life?

Spider likes postpositions, and is OK with having only two arguments marked by word order. Spider is also a huge fan of applicatives of all sorts and is kindly suggesting that to Rain. Also, Spider loves pro-dropping (or mostly-everything-dropping, like Japanese).

Spider has been very busy for the last month, although the amount of work has been exponentially decreasing, culminating at having absolutely nothing school-related to do after the 13th. When there is time Spider will work more heavily in this project (although not as heavily as Spider's main project). Spider'd also like some input from Rain, which will be of great help.

I have been thinking of agglutinating and head marking for basic morphology. I like applicatives too, so I'm all for Spider's suggestion of that.

Spider, how variable do you want the word order in general? If we are going highly non-configurational, then we'd probably have a lot of morphology to support it. Maybe we'd go more dependent marking there....

Do we want noun classes? That would factor in a lot into the morphology.

Bee is pleased at the idea of having many applicative operations.

Bird tentatively suggested having noun classes, but Bee is not enthusiastic about it. If we do have noun classes, Bee would like the noun class of a word to be predictable from its phonological form (e.g., ending in a consonant = inanimate, ending in a vowel = animate?).

Spider is somewhat of an extremist when it comes to noun class, optimally liking either no classes, or a whole ton of classes, while finding 2-5 class systems somewhat damp.

One there's been in Spider's mind for a while, is that it'd e much appreciated if there was conditioned word-order variation to SVO and/or VOS, circumstances still undecided. What are your thoughts on the matter, Rain?

I have been busy with real life and behind the scenes. I will post a progress update in the next few days.

What do you all think about clicks? Does anyone have any objections to having clicks in our language?

No objections from Bee.

Bee suggests that Rain (and other collaborators) take a look at

http://conlang.wikia.com/wiki/List_of_derivation_methods

Now that's interesting. Favorited.

Which clicks did you have in mind?

Tangent: what sorts of pronouns should we have? If we decide to have gendered pronouns, we should include queer-inclusive and/or epicene pronouns.

If we happen to have clicks, let's keep them the most symmetrical with the plosive system as possible.

Spider's not a fan of big pronoun inventories, the protolang of my main project having only 3. Spider also thinks queer-inclusive pronouns are something of a hassle.

Another point i just thought up, how naturalistic are we getting? Spider has a strong primal instinct of hardcore naturalism, so you better warn Spider.

Bee would like to have obviative/fourth-person pronouns of some kind. Third-person pronouns at least, and possibly others, should be in line with our noun class system (if any); and Bee would ideally prefer no noun classes, but if we must have them, then let's have a system which does not align with psychological gender or biological sex but something else (animate/inanimate or thunder/rain/spider/etc.)

Yes, please.

Bee also would prefer a fairly basic pronoun system, except maybe for an obviative. We can pluralize pronouns with the same strategy as nouns if need be.

Bee prefers a more abstract approach (since our goal is to learn and use this language a lot) but can be persuaded to more naturalism (which will make it harder ot learn) if the other collaborators are in favor of it.

Bird is a fan of extreme regularity. Irregularities will emerge anyway so there is no need to deliberately include them. Once we begin noticing the irregularities, it will give Bee and Bear something to do.

I think we should start with 365 root words (one for each day of the year), then add an arbitrarily agreed-upon amount each month or so. It allows the language to unfold in neat stages for both the learners and the creators.

Please do! It would be so valuable to Lojbanistan.

Should we make this a new role or should Bird cover this? If we decide on a new role, then I propose it be called "Rainbow" or "Wind".

Bee agrees in general terms, without committing to specific numbers as yet. The lexicon database table should have "date coined" as one of its fields, and "bundle" or "level" as a way of grouping words into clusters. A level might contain an entire year's worth of coinings, but if one is tackling a text written on 2016-09-03, one could search the lexicon for all words coined prior to that date and keep such a list handy, even though that might include some words from bundle 1 and some from bundle 2.

Since it requires fluency in multiple other languages, Bee thinks it should be multiple roles, one for each community we delegate and ambassador to. Wind might be the ambassador to Esperantujo, Rainbow the ambassador to Lojbanistan, etc. If one person knows multiple languages they might serve in multiple roles but one person is unlikely to know all the languages needed. (Bee is fluent in Esperanto and fairly so in Toki Pona, and could bridge to those communities as needed, once we have enough language and corpus to tell people about.)

Another aspect might be ambassador(s) to different English- and natlang-conlanging communities -- IRC #conlang, ZBB, CBB, Idiolengua, the various Facebook groups, etc.

Does anyone want to tackle the role of Spider?

I think I'm leaving the project. Primarily to focus on my main project, and secondarily because this is getting too far from anything I'd construct myself.

I could take Spider as well, since Bee won't have work to do on the corpus until we have some syntax to make sentences with. -- Unless we have a new collaborator who wants to join in?

Good luck with your main project!

It looks like we are heading rather strongly in the non-naturalistic and englang directions. As Rain, I'm cool with that (I'm a nice cool rain... wait, I'm getting off track...).

So in terms of morphology, I'd like to go with a self-segmenting morphology. I'd like to hear what thoughts Thunder has on this.

Reading some of the other suggestions, it sounds like dependent marking might work better, especially with the idea that oblique roles are would be marked as cases or adpositions. I favor cases, myself, and I'm all for a lot of them ;)

As for morphological operations, I'm always fond of some legerdemain here - not just adding affixes, but having things like stem changes and vowel/consonant grades. Have some non-concatinative morphology in there. Any thoughts on this?

No noun classes, it sounds like. However, I may go with declensions and/or conjugation classes named after the various roles in the language construction (I like the idea of a Spider declension, a Bee declension, etc.) I'd make the declension classes based off of phonological form, so one would not have to memorize declension with the word.

Bee likes the idea.

Bee tentatively agrees, but would like more details. If some words mark case (for instance) by stem changes instead of affixation, they should be high-frequency words.

Bee would prefer far fewer than nine declensions/conjugations -- preferably just one.

Hail Bee! Hail Spider! Bird approves.

So, do we have anything definitive yet? Has Thunder any phonology for us to view? Does Flower has any idea for lexemes?

If the others of the tribe wish for naturalism, Bird will concede. Bird just wants a successful project.

Thank you for your thorough response. :) Bird awaits consensus.