Vote Up A Language! (A community language-construction project)

[The Anarresti]

Vote Up A Language!
a community language-construction project

(Current Voting Question at bottom of this post)

Folks, ever since J.R.R. Tolkien himself, the great-grandaddy of all worldbuilders, first published his works, invented languages have been at the heart of fantasy. We've all been there, (or at least, I've been there), that place were we want to make our own language, to claim some of that glory, but don't have the time or the know-how to do it. Quite honestly, it is a lot of work.

So, I, a bored linguistics major, present to you a completely unprecedented kind of community project: making a language from scratch. If you've ever wanted to have a hand in making something unique, flavorful, and powerfully expressive, now's the time.

Here we go:

The BasicsOk. The way it's going to work is that I am going to introduce a handful of linguistic terms or characteristics. There will then be a period for discussing and voting. Depending on what issue is currently up for debate, the discussion period may extend for a couple days before the voting period, or may be coterminous with the voting period. Note that I'll try to input as little of my own decisions into the language-making-process as possible: this is partly because I want to preserve the excitingly democratic nature of the project, and partly because I want a life outside of the forum and running this thing will take up enough time as it is.

The discussion of the language will proceed in ten phases: Typology, Phonology, Time, Space, Word Categories, Word Construction, Pronouns, Sentence Construction, and Funky Business. Please keep the discussion limited to the topic at hand; part of the fun is making a language unlike any other, and crossing topics increases the chance that one's own knowledge of a pre-existing language, be it Klingon or Cantonese, will influence what one says. Additionally, keeping the discussion centered on one topic at a time makes it that much more accessible to those who lack a lot of linguistic knowledge.

The lengths of each phase will vary wildly, as will the voting structure for each. Sometimes I will simply ask you to vote between several cut-and-dried options; other times I will float a topic, define some terms, and let everyone else hash out the specifics (whereupon I will take the leading specific ideas and put them to a vote). Hopefully, over time, we will move to more free-form discussion as people get comfortable with making a conlang.

Additionally, because decisions about the language we make early on will affect what decisions we make later, the actual choices that I lay out in the blurbs for each phase will become more specific later.

After all the specifics of the language itself have been ironed out, I will translate a short paragraph of English into the language, and post it the forum, along with a word-by-word gloss and a soundcloud file (or something similar) of me reading it aloud so that we all can know what it actually sounds like. Then we move to the tenth and final phase of discussion: Who Speaks It.

If this thing takes off the ground, I might have trouble managing it all, so anyone who has interest in being a co-curator of the thread, to run things when I can't, please PM me. You don't necessarily need linguistics knowledge to help out, just organizational skills and a can-do attitude.

The Phases of Discussion
Typology done: Linguists classify languages into four basic categories, based on how their grammar works: Fusional, Analytic, Agglutinative, or Polysynthetic.

Fusional languages may be the most familiar to playgrounders, as all Indo-European languages, to a greater or lesser degree, fall in this category. In such a language, changes in secondary meaning are expressed by changing the the form of the word itself.
For example, in English, in order to express the past tense form of the phrase, "I go," we exchange the word "go" for its past tense equivalent, and say I went.
Examples of Fusional Languages include Spanish, Latin, and Russian.

Analytic languages, on the other hand, tend to have short words, and if a speaker needs to change a meaning, he or she will simply add another word to the sentence. For example, a speaker of an analytic language may say: I go yesterday to demonstrate past tense.
Examples of Analytic Languages include Vietnamese, Mandarin Chinese, and Tibetan.

Agglutinative languages add affixes to express variation in meaning. An affix is something like a prefix, suffix, circumfix, or infix; it carries real meaning but must be bound to a word to exist grammatically. In contrast to a fusional language, which may have a different word for each separate case, such as third-person-past-feminine, or first-person-future-masculine, an agglutinative language just adds an affix to the same root to express each added meaning: in the case of a first-person future perfect verb, for example, a speaker of an agglutinative language may add 3 different suffixes to the same root word.
A speaker of an agglutinative language may say I goed to demonstrate past tense.
Examples of agglutinative languages include Turkish, Finnish, and Korean.

Polysynthetic languages are basically agglutinative languages amped up to eleven. The key feature of a polysynthetic language is that a speaker could, conceivably, express an entire clause with one word, such as the (in)famous Yupik word tuntussuqatarniksaitengqiggtuq, which means "He had not yet said again that he was going to hunt reindeer."
A speaker of an agglutinative language may say Goed-me to demonstrate that concept. The running example kinda breaks down here, as English is most certainly not polysynthetic.
Examples of polysynthetic languages include Greenlandic, Ainu, and Nahautl.
Our language will be Polysynthetic.

Phonology: Issues discussed in this phase will concern the phonology of the language, e.g. what sounds make it up, and how do those sounds combine to make words. (For simplicity's sake, I am unilaterally declaring that we are not inventing a signed language, because the phonology of signed languages, which consists of things such as handshape, movement, and location, is nearly impossible to represent with the Roman alphabet.)
We will decide how many consonants and vowels the language has, whether it uses clicks (very few do), and whether it is tonal, emphatic, uses vowel length, or none of the above. I will decide exactly which vowels and consonants are to be used.
I will pick only sounds that an English speaker could be reasonably expected to be able to distinguish and produce.
Time: This discussion will be determined a lot by what we decided on for the typology. Here, we will decide how our language discusses time and action: what tenses it has, if it is more important to express past, present, or future tense, or if an action is completed, continuous, or the like? Additionally, any other wacky things you guys come up with regarding time.
Space: Similar to the above. The bulk of the discussion will probably center around number: will the language use a 1-2-many counting system, or will it have the full set of numbers? Will our language have a plural form? A dual form? A separate "very plural" for enormous things? Will we make a distinction between uncountables (like water) and countables (like rocks?)
Also, if there's anything to be decided on prepositions, or other wacky things regarding how our language expresses spacial relationships.
Word Catagories: Some languages have seperate catagories for nouns and verbs, and a noun or verb that falls into one catagory is grammatically treated differently than a word in the other catagory. Grammatical gender, as this phenomenon is known in Indo-European languages, is a bit of a misnomer, as the gender of a word is not necessarily tied to the gender of the object it references.
We will decide what different classes to have, if any. Examples of such include masculine/feminine nouns in French, masculine/feminine/neuter in German, animate/inanimate nouns in most Algonquin languages, and strong/weak verbs in Old English.
Word Forms: Now that we've decided, grammatically, how words work, we need to decide exactly how they will be made. This entails inventing words, suffixes, particles, etc to put in place what we already decided. Plus, you know, actually generating a lexicon of sorts.
Pronouns: These can be weird. Here, we decide exactly what our pronouns are. At the very least, we need three (generic first, second, and third persons), but most languages have six: plural and singular for the three persons. If we have grammatical gender, we may need more pronouns for the different genders.
We can also do all kinds of funky things, like having a separate set of pronouns for referring to animals and children, separate formal and informal forms of "you," or a whole set of "he/she/they" pronouns strictly for talking about the dead.
Sentence Forms: Exactly how to make a sentence or phrase. Where does the predicate, subject, or object go in a sentence? Does an adjective come before or after the noun it modifies? How many different kinds of sentences can you make, and how, exactly, do you make them?
Also, here, we decide on how to treat verb "to be," including its absence, if any. (Many languages simply do without the verb "to be" sometimes, e.g. saying "he working" instead of "he is working.")
Funky Business: Anything we missed, or fresh ideas, or something else new. For example, this would be the stage where we decide that the language is sexually dimorphic, i.e. that men and women each speak a different version of it, or that there are fifteen different words for "snow."
-Reading Passage Interlude-

Who Speaks It? Who indeed? Here, we can place it to a pre-existing group of people in a made-up world, such as the Giantkin from the Tears of Blood community-made campaign setting, or make it Gnomish in The Giant's "Barbarian Setting," or we could start a new thread to build a world, people, and mythology around the language, a la Tolkien.
Please, please, please do not discuss this one until we have finished the language, because the people we think may speak it will strongly influence how we think it "should" work. (For example, if it's for the Tears of Blood human Phoedran culture, there would be an unconscious push to make it sound "greek-y."


The language construction kit (http://www.zompist.com/kit.html) is a fantastic resource, and I would certainly advise anyone who plans on becoming a regular on this thread to look it over. It contains pretty much everything you would need to know to approach this project with no prior linguistics knowledge.

Also, here (http://www.paulmeier.com/ipacharts/) is a complete, interactive International Phonetic Alphabet chart, with textual explanations of each of the sounds as well as a recording of someone pronouncing them.

A Reminder, and A Warning:

Firstly, I would like to remind everyone that this is an English-language forum. Years from now, when our fledgling Playground-tongue is finally complete, we may be tempted to use it to to chitchat amongst ourselves. Remember that is very rude to talk in a code in a public forum, so when that heavenly day of viable speech comes, please limit use of it to PMs or threads specifically made for such a purpose.
In the meantime, please gloss (e.g. provide an english translation for) any word, phrase, or sentence you make in our Playground language, unless you are specifically talking about that word or phrase, not using it to communicate, and the meaning of the word or phrase has already been established or could be readily deduced.

Light, glossed banter in the thread should be fine. We are, after all, only human :smallbiggrin:


Additionally, as the Warning, please do not discuss any real-world culture. We are here to discuss the mechanics of a language, nothing else. Bringing up a real-world language as an example of a trait of a language is fine, like what I did above, but real-world geopolitics or ethno-cultural theory has no place on the thread.



Current Questions

Discuss, Nominate & Second: What phonetic features would you like to see in our language? This can be as broad as "I think we should allow syllables to end in consonants" to "I think we should have the voiced uvular fricative somewhere in the language." (Unlimited nominations and seconds. OP, The Anarresti, reserves right-of-veto for phonetic characteristics deemed too difficult for English speakers to replicate)

Vote Format

You may express it as a one word answer, or as part of a sentence, but please made only the key word bold. E.g. "Fusional," or "I think a polysynthetic language sounds absolutely dandy."

If you feel like changing your vote, please go back to your original post, unbold and strikethrough your old vote, and write your new vote inside your old post, bolding the key word. When I am tallying up the votes, I don't want to have to backtrack.

Nominations for the Ballot:

This won't come up for the first phase of voting, but will be allowed for every round after. The way it'll work is that during the discussion period, anyone will be allowed to nominate a linguistic feature to be placed on the ballot, as long as that feature is germane to the discussion. If someone else seconds that nomination, then the feature will be placed on the ballot, most likely as a yes or no question.
If you do nominate a feature, please remember to
a. State clearly what the feature is. If it's something you vaguely remember, wikipedia is a surprisingly useful tool for explaining linguistics.
b. If you use any linguistics jargon, please define it.
c. As with the voting, please bold the key word or phrase.

For each discussion phase, I will make a list of questions, which need to be answered. The answers will then be put on the ballot, as options to vote from. For an extended example:

Phase: Sentence Forms

Discuss & Nominate: How will our language form questions?


Why don't we make it so that in our language, you make questions by putting the verb at the front of the sentence?


Hmm. That'd be a little odd. What about moving the predicate to the front? You know, the whole verb phrase, instead of just the individual word.


Yes, that's what I meant. I nominate that our language will form questions by moving the predicate to the front of the sentence.


Moving the predicate to the front for questions, seconded


Why not keep it simple, and retain the word order of a statement. To ask questions, I nominate that we use question words only.


Sure. Question words only, seconded

(later, on the ballot)

Vote: How will our language form questions? (choose one)
1. Move the predicate to the front of the sentence
2. Only use question words.

Once we move onto the voting phase, nominations for the ballot will no longer be accepted.
Polls are Closed
Nominations are Open
Polls close at N/A
Nominations close at tbd

[The Anarresti]

This post is reserved for decisions we make about the language.

Elaboration on Grammar: none yet

Running Lexicon (of invented Playground-Language words): nope

Glossary (of linguistics terms, or meta-terms we invent): There are some, but I can't be bothered to define them just now.

Voting Record: derp

Possible English Passages
The North Wind and the Sun (broad community support)
Something from Beowulf (suggested)
Clive Barker, "After a battle lasting many ages..." (suggested)

[Everyl]

Ooh, this looks like a fun project. I'm very interested in languages, but have no formal background in linguistics.

For now, I'll vote for polysynthetic. It should be a nice mental workout to figure out a grammar structure that alien to my native language (and to the second languages I've studied).

Edit: Typo fix

[Owlglass Moot]

Always a pleasure to see another linguistics major! We're like unicorns.

My vote's agglutinating.

P.S. Any chance of adding a vote for morphosyntactic alignment at some point? Or would that fall under "Sentence Forms", I guess? I'm a sucker for active-stative and tripartite languages.

[The Anarresti]

Owlglass M.- Hail and well met! The later voting sessions are going to be more free-form, with as many options, and even voting issues, as possible originating from the thread instead of the OP (c'est moi). When we get to Sentence Forms you're more than welcome to explain what a tripartite language is and why everyone should vote for it :smallbiggrin:
"Sentence Forms" should probably be renamed "Clause Forms," I guess.
With the possible exception of the first vote, I want to work bottom up, from smallest to biggest. I was going to vote on phonemes first, then on the morphological typography, then on word forming, etc. I switched up the first two because the typology is so much more straightforward and makes for an easier first vote, but the overall trend from tiniest pieces to bigger is there.

Also, truth be told, I'm a phonetician at heart, so I got more bogged down in details there. :smalltongue:

[Grinner]

Chalk me down for analytic.


P.S. Any chance of adding a vote for morphosyntactic alignment at some point? Or would that fall under "Sentence Forms", I guess? I'm a sucker for active-stative and tripartite languages.

Well there's something you don't hear everyday... :smallsmile:

[The Anarresti]

Update: I added a time the polls close (8:00pm, EST, Friday), and an additional discussion question, mostly for fun (What would make a good English paragraph to eventually translate?)

By the way, Grinner, would you mind bolding your answer? It'll just make it easier on me when I tally up all the votes.:smallsmile:

[Owlglass Moot]

I always liked Aesop's Fables or Grimm's Fairy Tales for translation. "The North Wind and the Sun" is the one I've seen used most often for conlangs.

[The Anarresti]

I assume you mean this:

"The North Wind and the Sun were disputing which was the stronger, when a traveler came along wrapped in a warm cloak.
They agreed that the one who first succeeded in making the traveler take his cloak off should be considered stronger than the other.
Then the North Wind blew as hard as he could, but the more he blew the more closely did the traveler fold his cloak around him;
and at last the North Wind gave up the attempt. Then the Sun shined out warmly, and immediately the traveler took off his cloak.
And so the North Wind was obliged to confess that the Sun was the stronger of the two."

Not a bad choice. It has the feel of natural speech, and isn't inherently tied to a particular group or culture.

[Ozy]

I'm a bit taken with the Na Dene languages right now, so I'm gonna cast my vote for polysynthesis. Whatever we choose, though, could we have the verbs be incorporating? An isolating tongue with incorparating verbs might be kinda fun too.

Also: isn't there a difference between isolating and analytic languages, the latter being a subset of the former? Or am I in error?


As for what to translate, we'll almost certainly want something more prose-y than this, but I do think this poem by clive barker could be interesting to do:After a battle lasting many ages,
The Devil won,
And said to God
(who had been his Maker):
"Lord,
We are about to witness the unmaking of Creation
By my hand.
I would not wish you
to think me cruel,
So I beg you, take three things
From this world before I destroy it.
Three things, and then the rest will be
wiped away."

God thought for a little time.
And at last He said:
"No, there is nothing."
The Devil was surprised.
"Not even you, Lord?" he said.
And God said:
"No. Not even me."
(But I also like The North Wind and the Sun)

[The Anarresti]

To be perfectly frank, I'm just one linguistics sophomore with a heart full of Sindardin and a head full of Mandarin. I honestly think that, if you all put your heads together, you're more likely to come to an accurate conclusion that I would on my own.

Regarding the excerpt, that seems good too. I'll take a crack at stuff that seems vaguely poetic.


By the way, I only contrasted heart/head and Sindarin/Mandarin because it sounded cool. The reason I'm still studying Mandarin after four years is because I find the language itself so fascinating, and the literature so broad and deep; it's not like I dislike it.
And, when I say "full of," I mean "studied in high school and picking back up now," not, "fluent in." So, all in all, my list of credentials is pretty short.

[Balyano]

My heart says polysynthetic, but my brain say agglutinative.

[sktarq]

And to round out the early voting I'm going Fusional

[Jendekit]

Regarding the paragraph to translate, I enjoy the North Wind & Sun so even if that doesn't become officially translated I'll do one myself.

Regarding the language itself, my vote is polysynthetic for the basic category

[sktarq]

been thinking a bit about this.
I think we need at least some idea of the culture we are making a language FOR in order to do this properly eventually.
and a bit about the world it is in.
If it is for DnD/PF type place then articles or the like native to say demi-humans or intelligent non-humanoids may be called for. The degree of formality expected within a culture may also be important-Spanish tu vs Usted as an example or various Japanese suffixes for relationship based ones. Very long lived races may have different ideas of future and past tenses than short lived one. None of this needs to be picked yet but how well can we make a generic language that adds to that.
It laid out that it should be only discussed in step ten but the form of who speaks it does make structural differences.

[Everyl]

I don't know about the linguistics majors, but I'm fine with building the language first, then considering what it implies about its speakers. Use the language to create the people, rather than the other way around.

[Alexkubel]

I'm torn between Fusional and Polysynthetic for what to go for.
my preferred translating phrase is not appropriate to post up, not just because it is involving various impolite words but mainly because it involves making an entire Thesaurus.

[BornValyrian]

This is really cool, hopefully this catches. My vote goes to Agglutinative.

The North Wind and the Sun sounds cool.

And I'm with Everyl on the language->Culture deal.

[Silverbit]

I'm rather interested in this; I barely know anything about linguistics, but I'm still intrigued by fictional languages. My vote is for Analytic.

Maybe we should try some stanzas of Beowulf? It's quite a challenge to translate, I hear. I approve of the North Wind And The Sun as well though.

We should build the language first, I think. It's a nice change from how culture building is normally done.

[Ajadea]

I'm going to vote Polysynthetic. This project sounds really cool. Definitely keeping an eye on it.

[Ellye]

I'll vote for Agglutinative. I like how consistent they can be.

[Ziegander]

I had my doubts about how this project would work, but with the OP being a Linguistics major and at least one or two others contributing to the thread, the parameters set up for how the language will actually be voted up, this is all sounding pretty crazy-mazing. I love all the possibilities for little flourishes that could be added to change how the language is spoken/written...

Anyway, my vote this round is Agglutinative.

[The Anarresti]

Hey y'all! So, I just added procedures for nominating items to the ballot. It's a moot point this time around, but once we move on to phonology, from then on it's open season.

sktarq does have a point about culture influencing language. I would say that we are intentionally going about it backward, like Tolkien did. First, we make the language, then we let the language determine what culture to fit it too. Of course there will be fine-tuning involved once we place the language, but that bridge can be crossed when we come to it.

I'm looking forward to the flourishes, too, but I'd like to point out that we are tackling the spoken language first; that's hard enough as it is. We haven't even decided if it will have a writing system at all. Once we vote up the phonetic rules and inventory, I will invent an appropriate way to write it with the English alphabet, an informal orthography, but that will just be a tool for our purposes.

An invention of a script can come later, but that's not the focus of this thread.


Also, regarding the passage to be translated: After this discussion round is over, I want to table that conversation, but then maybe resurrect it every so often while I tally the votes. I'll keep track of all the nominations for the passage and then choose my favorite when we are finished. Don't complain, I'm the one who's doing the translating :smallwink:

[Grinner]

@sktarq:

I'm no linguist, but I have attempted to make a language. One of the resources I found, the Language Construction Kit (http://www.zompist.com/kit.html), made a similar point, observing that one's vocal apparatus would affect the range of sounds one is capable of.

However, as we're just a bunch of strangers, developing this in concert with a whole culture, assuming its an isolated one, and its environment by extension is a bit more than we're capable of.

[The Anarresti]

The Language Construction Kit is a pretty cool resource; I'll stick it in the original post for people to read if they wish.

@Alexkubel: do you want to link to it?

[Thunderfist12]

Agglutinative language.

[Amblehook]

I would like to place a vote for Analytic.

[sktarq]

I guess my problem is this. Making a language with the idea of matching it up to a at least partially premade culture seems to be a recipe for making as generic a language as possible. In trying to allow for greater choice of where to use it later interesting choices will be voted down in that kind of system. JRRT didn't so much match up elves to his new languages as create their culture from inspiration around the language. I think that may be a better goal for us too. Creating the language as a starting point for the culture.
for example. In a fair number of langagues the upper class and lower class speak rather differently. Different loan word sources etc. And while there can be a large number of reasons for this. (Invading army takes over and upper class is made of decendents who spoke a different language, nearby cultural influences being considered positive/negative, refugee or slave movement bring new languages, institute preserving archaic version for nobles) any of these is writing part of the history of the culture we are linguistically making this for. That limits who can really use the language and thus if it to be fitted to a group that may not have that trait latter it would be an issue...so it would be avoided and perhaps the language would be poorer for it. If we always pulled away from that kind of option it certainly would be.

[Balyano]

I think it would be cooler to make the language and afterward look at the result and see what kind of culture the language's quirks might imply, then build a culture from that starting point, then and only then try and figure out what fantasy race might fit it. It would give a culture that doesn't fit the exact stereotypes that tend to get overused while letting some core idea of a race still be expressed in it. So you may end up with a race still feeling like themselves but in a brand new way that is very distinct from normal.

[BornValyrian]

Here's a tally so far:

{table=head]Typology|Votes
Polysynthetic|5
Agglutinative|6
Analytic|3
Fusional|1[/table]

Unless I'm missing something.

[Fortuna]

Let's go with polysynthetic, to bring the votes even.

[Jakodee]

polysynthetic, defiantly

[BornValyrian]

Bold your selection, so it isn't missed.

[Balyano]

How will ties, if any, be handled? And since these things are more of a continuum than strict categories will there be differing degrees of based on votes? For a hypothetical instance if both fusional and agglutinative had been tied or nearly so, would we then do something along the lines of ''verbal morphology is very agglutinative but nouns very fusional''?

[BWR]

Polysynthetic.

I know little about them so I want to learn through creation. It would be a chance to go one up on my sister who has to varying degrees studied 12 languages (I'm stuck at 4). For fun, while doing her biology stuff.

[Xuc Xac]

I vote analytic.

[The Anarresti]

Alright Guys, the votes are in, and we have a Polysynthetic language to make!

Now, it is true that these are very fuzzy categories, and as to the question about how it will work in practice, exactly how polysynthetic it will be, that is up to y'all. But, at least we have something to work off of, and a straightforward vote to get the ball rolling.

Next up, we are going to vote on the phonology & phonetics of the language. As it stands, there are a couple questions that I will be certain to place on the ballot in any case:

Will the language be tonal, differenciate based on vowel length, or neither?
A tonal language differentiates between words not just by the actual sounds that are said, but also by the relative pitch at which they are spoken. For example, in Mandarin chinese, the syllable "ma" held at a constant, relatively high pitch means "mother," but if you say it with a pitch which falls and then rises again, an up-and-down pitch, it means "horse."
Other languages, such as Japanese, Arabic, and Old English, differentiate based on the actual length of a vowel as well as the quality. That is to say, that, for example, the invented word "dod," pronounced quickly in normal speech, has a different meaning than it would if it was stretched out, where you said the same actual vowel, but held it for a longer amount of time.
Will this language allow for closed syllables, or will it only have open syllables?
A closed syllable is one that ends in a consonant, e.g. "dad," "ad" An open syllable is one that ends in a vowel, and may begin in a consonant. All languages have open syllables; many languages allow phonetically allow for closed syllables.
Many languages have a limited system of closed syllables: the language allows for syllables to end in a nasal (m, n, ng), an approximate (r), or something similar. These are consonants that have many vowel-like properties.
Will the language have few vowels (3), an average number (5), or a lot (12).
The smallest recorded number of vowels in a language is 3. The universally preferred vowel system is the Spanish one, of five pure vowels: a, e, i, o, u. Italian also has this vowel system. 12 is an arbitrary number. For reference, English has 11 pure vowels: very high by the world's standards.
Will this language allow for diphthongs?
A diphthong is a sound made by combining two vowels into one syllable. The english word "eye", pronounced in the Standard American or Standard British manner, is an example of a diphthong.
Will this language have a small number of consonants (8), average (20), or a lot (35)?
These numbers are arbitrary. For reference, American English has ~24 consonants
Will the language have clicks? These are sounds that are made purely with the mouth, without any air from the lungs. They are not hard to make, but for some reason, unknown to modern linguistics, they only appear in Khoi-San (e.g. "bushman") languages, in South Africa.
Even more mysteriously, it appears upon closer investigation that the so-called Khoi-San languages are not genetically related; that is to say that these languages do not share a common ancestor.
The "tsk-tsk" sound and the "kissy" sound are both examples of clicks that English speakers often make.


The next period will be one without voting, where you all discuss what other phonetic/phonemic qualities you want in the language, and pose questions to me and/or the other linguists. Also, feel free to read through the materials provided in the first post, the language construction guide and the fully explained interactive IPA chart.

For those just joining us, please read the first post, or at least skim it.

[BWR]

I wonder if this project might work better with an actual goal in mind. Not just 'create a language', but 'create a language for a certain race and/or culture'. Tolkien and Barker both had specific races for their languages, though off-hand I can't remember if they designed the languages with the race in mind or if they needed someone to speak the languages they made.
I think it would help if everyone was on the same page with what they want to see rather than just making it by vote.
IMHO

1. Sound length and tone.
Will the concept of long consonants be addressed later? Anyway, sound length is familiar but tonal languages are so alien to me, even though Norwegian technically has some residual tonal elements (which I'm incapable of detecting or producing in speech), so it might be fun to do that. Tonal for my sake, then though it might be fun to try both length and tone.

2. Closed

3. Vowels
I like a lot of vowels, but I'm more than willing to give up on this point, or possibly have a number of allophones that pop up in a consistent pattern.

4. Dipththongs and more. Tripthongs, perhaps even more.
Fun fact. certain dialects of Norwegian can produce sentences that consist entirely of vowels. This can be transliterated as "Æ e i Å æ å" or in proper spelling "Eg er i Å eg óg". Literal meaning "I am in Å I too" - "I am also in Å"

5. I'd like to say a lot of consonants, but right now I'm envisioning something very liquid and heavy on the vowels. Breezy and breathy, perhaps. So few consonants, no more than 10 or so, perhaps 7-8.

6. Clicks.
No. Doesn't work with a soft language.
Other non-standard sounds might be fun. Perhaps there's a difference between egressive and ingressive sounds.

[Xuc Xac]

"Will the language be tonal, differentiate based on vowel length, or neither?"

Why not both?

"Will this language allow for closed syllables, or will it only have open syllables?"

I vote for closed syllables being allowed but with a limited number of consonants that are allowed in the syllable final position (i.e. the language has several consonants but only a few can end a syllable).

Will the language have few vowels (3), an average number (5), or a lot (12).

Since I voted for long vowels, I will suggest 5 pure short vowels, 3 of which can be lengthened for a total of 8 vowels.

Will this language allow for diphthongs?

Yes. I would go so far as to suggest occasional triphthongs (with a limit on vowel length, maybe a maximum of 1 long vowel).

"Will this language have a small number of consonants (8), average (20), or a lot (35)?"

Average, with a small number allowed in the syllable final position

Will the language have clicks?

Only as interjections (as in English "tsk tsk")

[Balyano]

tone no
I vote against phonemic tone, but I think it would be a good idea to use intonation to distinguish the beginning of a word in speech.

long vowels yes

diphthongs yes

vowels medium-high
I vote for 5-10 basic vowels so we can avoid having to use too many symbols for them in writing. So something like i, e, a, o, u, with diphthongs and the remaining pure vowels represented with digraphs. For instance the pure vowel ''a'' sound in ''cat'' might be represented by the digraph ''ae'' where as the diphthong ''a'' in ''bake'' might be ''ei''. If y and w are not used for consonants then they could be used as vowel symbols. If the number of vowels needing a digraph is small enough adding h might be a quick way to wright them all. So the ''ae'' might be ''ah'' and the ''ei'' might be ''eh'' instead. A small number of languages such as Ubykh from the Caucasus and Upper Arrernte from Australia are sometimes analyzed as have as few as two phonemic vowels. But I wouldn't suggest that few.


closed syllables yes
I vote for closed syllables with a limited inventory of final consonants that assimilate to the following consonant, if any. I would suggest something like -ʔ, -s, -n, so hypothetical naʔ-pi-sas-ken-ben > nappisaxkemben, though it might just be spelled natpisaskenben.


consonants medium
I vote for about 20 consonants, give or take. I suggest we start with a simple inventory, then vote a few phonemes in or out.
In different series of consonants if there is a gap there tends to be a pattern. Typically out of p, t, k, if one is missing it will be p. With aspirated stops ph is even more likely to be missing, and with ejectives even more so. Out of b, d, g it's usually g that is missing. With implosives g (ɠ) is the most likely to be missing. ŋ is more likely to be missing than m or n.

For example we might start with something like
p t k
b d g
s
z
r
m n ŋ


and then decide that we don't want p, g, z, or ŋ, but decide we do want f, h, L, ɬ, and tɬ, resulting in
t k
b d
f s h
ɬ
tɬ
L
r
m n

It may also be a good idea to have certain phonemes be a little ''loose'', with a lot of leeway for allophony. Perhaps allowing θ in place of ɬ and tʃ in place of tɬ if the speaker has trouble with lateral fricatives.


clicks no

[Balyano]

I wonder if this project might work better with an actual goal in mind. Not just 'create a language', but 'create a language for a certain race and/or culture'. Tolkien and Barker both had specific races for their languages, though off-hand I can't remember if they designed the languages with the race in mind or if they needed someone to speak the languages they made.
I think it would help if everyone was on the same page with what they want to see rather than just making it by vote.
IMHO

I'm of the opposite opinion. I think it would be cooler to make the language with no preconceived ideas of what it should be like. Then take a step back and think ''what sort of culture seems to gel with this?''. And then say ''what fantasy race could fit with that culture?''

It might allow for a race to get a culture that differs from their typical portrayal, but still fits in a way.

Don't get me wrong, I do like creating languages I think fit a fantasy race. I have a dozen half finished (read that as unlikely to be finished) ones for goblins, orcs, elves, dwarves, ect. But with a group of people voting on things a bit more unknown and unexpected becomes possible.

[BornValyrian]

1: I think that we should skip tonality, with maybe a few exceptions. I like the idea of the length of the vowel making it a different word, though.

I think we should keep to the 5-10 vowels but include diphthongs.

I'm thinking closed syllables should be restricted to finishing with non-stop sounds (so syllables don't end with sounds like t,b,p,k, etc.)

As for consonants, I think a smaller amount, like 10-15.

I like BWR's idea of a "breezy" language, more flowy.

And I think, like Balyano, that we should not worry about the culture/race until we have the language (I'm pretty sure this is what Tolkein did, though don't quote me on that).

[lsfreak]

This could be interesting, if I remember to keep up on it.

Tone:
Yes, but limited. My vote would be a two- or maybe three-tone register system, where a given word (or set of morphemes) takes a high, medium, or low pitch. Or perhaps Japanese-style, where a given syllable drops or "resets" the whole pitch of the word. Definitely not a Chinese-style contour system where almost every syllable has its own distinct tone, that would rapidly turn into a cluster**** with polysynthesis.

I'm not sure you need to unite tone with vowel length, given the frequency of tone and vowel length being independent, though I'm ambivalent about vowel length.

Syllable type:
Closed allowed. Given polysynthesis, I think we need closed syllables. Open-only syllables would likely be unstable, with vowels dropping out in order to reduce words to more reasonable lengths. And aesthetically, I loves me some consonant clusters.

Vowel number:
Average. Let's not overdo ourselves, though I'm all for a nonstandard-but-believable vowel set like /i y e ɤ a/. Alternatively, large if we're doing vowel harmony, where a given word has either one set of vowels or another, but not both (taking an English example, one set of words might allow "ee ay oo oh ah" while a different set takes "ih eh uh aw a"; or if you speak German, a much more systematic version of "u o a au" versus "ü ö ä äu"). Actually that might be my vote, large with vowel harmony of some kind.

Side note that I'm pretty sure it's agreed upon that the smallest vowel system is just two /ə a/, but with significant allophony [i y e ø ə a u o]. Even most 3- and 4-vowel systems I'm aware of have quite a bit of allophony; actually, off the top of my head, the only language I know of without at least five major allophones is Wichita with /i e a/ plus [o], though it's not something I've really paid attention to.

Diphthongs:
Yes/no. Meaning that certain vowel clusters may be realized as diphthongs or triphthongs, but they are fundamentally different vowels. mu + a + i is three vowels that happen to be pronounced as a triphthong, but there's not an actual triphthong uai. Or certain consonants trigger diphthonization, such as siq becoming siaq and qis becoming qeis.

Consonant number:
I'll say average for now, but I can't really answer until we decide how they're distributed. Trying to cram aspirates, plain stops, breathy stops, and ejectives into an average number of consonants could rapidly turn wonky.

Clicks:
No. Just too complicated, to pronounce, to incorporate, to put into an easy orthography.

[sktarq]

Firstly I'm going to have to ask how we plan to use the language. From it's placement in a world building forum for RPG games I would imagine it is being devised for use by DM's wanting a resource at their gaming tables more than an author of a novel series. I bring it up because a language for DM usage will have certain qualities that will lend it success in that format. - -Sounding very different than known languages would be a good thing as this would help with the fantasy element-vs sounding like the ST plugged a phrase into Google translate and then mangled the pronunciation
-Being able to pronounce easily-Heck we are going for big long words with the language type already and while fun academically if the person does the translation and then can not easily pronounce their own work then they are not likely to go through the work of translating in game-even moderate effort on that point slows the pace of the game and focuses attention on something beyond the story - a fast quick flowing name in a bizzare tongue or a riddle that the entire party can say at least semi quickly aids the story and more likely to be used again.
-and when in game is the language likely to be used? I bring this up because playing to these ideas could give us a language that sounds at its best in fields of use and thus comes off as cooler (and more likely to be used again) Really the bits that come up to my mind would be Names-People, Places, spells, institutions, social groups, weapons/tools, etc. Also Riddles, Curses, and prophecies. .. . and Inscriptions on various monuments. Throw in command words and possibly a bit of poetry and I think that would be it for the MAIN uses....


Alright Guys, the votes are in, and we have a Polysynthetic language to make!
Will the language be tonal, differenciate based on vowel length, or neither?
A tonal language differentiates between words not just by the actual sounds that are said, but also by the relative pitch at which they are spoken. For example, in Mandarin chinese, the syllable "ma" held at a constant, relatively high pitch means "mother," but if you say it with a pitch which falls and then rises again, an up-and-down pitch, it means "horse."
Other languages, such as Japanese, Arabic, and Old English, differentiate based on the actual length of a vowel as well as the quality. That is to say, that, for example, the invented word "dod," pronounced quickly in normal speech, has a different meaning than it would if it was stretched out, where you said the same actual vowel, but held it for a longer amount of time.
I mentioned above it should be easy to tern the written form (which is how people will be exposed to this even if it just their own notes to a verbal form in order for the game to go smoothly. Tonal variation makes this difficult-While cool academically and could have fun magic sysmpathetic tie ins. Nine and Death having only tonal differences giving the culture a thing for nine days of mourning, nine stages of the underworld etc I don't think it would be helpful in this case. As for length I this could work but would need very clear separation and even then I recommend against it.



Will this language allow for closed syllables, or will it only have open syllables?
A closed syllable is one that ends in a consonant, e.g. "dad," "ad" An open syllable is one that ends in a vowel, and may begin in a consonant. All languages have open syllables; many languages allow phonetically allow for closed syllables.
Many languages have a limited system of closed syllables: the language allows for syllables to end in a nasal (m, n, ng), an approximate (r), or something similar. These are consonants that have many vowel-like properties.
Honestly I'd prefer open only or a very short list of closable consonants-The main reason is that with the style of language we have chosen there is a lot of word parts that will added together in various orders. Keeping a highly stable syllable ending will make it easier to avoid awkward transitions between word parts that hang up an unpracticed speaker. I think this would give it the kind of back and forth that would do well in riddles and name that I thing would be important. On the other hand going with ALL closed would be nifty too-Either end really but all open seems easier.



Will the language have few vowels (3), an average number (5), or a lot (12).
The smallest recorded number of vowels in a language is 3. The universally preferred vowel system is the Spanish one, of five pure vowels: a, e, i, o, u. Italian also has this vowel system. 12 is an arbitrary number. For reference, English has 11 pure vowels: very high by the world's standards.
Honestly I think vowels give a language so much of its character and sound-It is the one area I'm totally fine it pushing people to have new symbols to get right-So lets go largish-12 works for me.-But assuming an alphabet is 1 letter to symbol (instead of syllabry lets give each on its own letter)



Will this language allow for diphthongs?
A diphthong is a sound made by combining two vowels into one syllable. The english word "eye", pronounced in the Standard American or Standard British manner, is an example of a diphthong.
I'd say no for the same reason as wanting to keep things with open (or nearly so) syllables-with the highly compound word nature of the language awkward attachments and transitions are more likely with diphthongs. Otherwise we'll just promote the - okay did the translations and...crap three diphthongs in a row if using the command future tense of the verb "taking" when combined with a masculine object - how did we not see this? and how do I say that anyway?



Will this language have a small number of consonants (8), average (20), or a lot (35)?
These numbers are arbitrary. For reference, American English has ~24 consonants
Meh - high vowels medium consonants or high consonants and medium vowels both work for me but I prefer the former so lets say Medium.



Will the language have clicks? These are sounds that are made purely with the mouth, without any air from the lungs. They are not hard to make, but for some reason, unknown to modern linguistics, they only appear in Khoi-San (e.g. "bushman") languages, in South Africa.
Even more mysteriously, it appears upon closer investigation that the so-called Khoi-San languages are not genetically related; that is to say that these languages do not share a common ancestor.
The "tsk-tsk" sound and the "kissy" sound are both examples of clicks that English speakers often make.
[/LIST][/SPOILER]
Good gods yes....Pops, clicks, snaps, lip pops etc. all work for me. If not sure how to work them in as part of words directly they could be a part of a limited set of unique sounds used to augment a language-either used only in magic, or math, or even as a punctuation guide. a pop being the equivalent of a period or "over" in radio speak...maybe a snap as the verbal equivalent of a quotation mark which would help compound quotes in general.

[Alexkubel]

Will the language be tonal, differenciate based on vowel length, or neither? I Have NO IDEA!
A tonal language differentiates between words not just by the actual sounds that are said, but also by the relative pitch at which they are spoken. For example, in Mandarin chinese, the syllable "ma" held at a constant, relatively high pitch means "mother," but if you say it with a pitch which falls and then rises again, an up-and-down pitch, it means "horse."
Other languages, such as Japanese, Arabic, and Old English, differentiate based on the actual length of a vowel as well as the quality. That is to say, that, for example, the invented word "dod," pronounced quickly in normal speech, has a different meaning than it would if it was stretched out, where you said the same actual vowel, but held it for a longer amount of time.
Will this language allow for closed syllables, or will it only have open syllables? open seems better.
A closed syllable is one that ends in a consonant, e.g. "dad," "ad" An open syllable is one that ends in a vowel, and may begin in a consonant. All languages have open syllables; many languages allow phonetically allow for closed syllables.
Many languages have a limited system of closed syllables: the language allows for syllables to end in a nasal (m, n, ng), an approximate (r), or something similar. These are consonants that have many vowel-like properties.
Will the language have few vowels (3), an average number (5), or a lot (12). how about 24?
The smallest recorded number of vowels in a language is 3. The universally preferred vowel system is the Spanish one, of five pure vowels: a, e, i, o, u. Italian also has this vowel system. 12 is an arbitrary number. For reference, English has 11 pure vowels: very high by the world's standards.
Will this language allow for diphthongs? I dunno
A diphthong is a sound made by combining two vowels into one syllable. The english word "eye", pronounced in the Standard American or Standard British manner, is an example of a diphthong.
Will this language have a small number of consonants (8), average (20), or a lot (35)?
These numbers are arbitrary. For reference, American English has ~24 consonants

Will the language have clicks? why not? These are sounds that are made purely with the mouth, without any air from the lungs. They are not hard to make, but for some reason, unknown to modern linguistics, they only appear in Khoi-San (e.g. "bushman") languages, in South Africa.
Even more mysteriously, it appears upon closer investigation that the so-called Khoi-San languages are not genetically related; that is to say that these languages do not share a common ancestor.
The "tsk-tsk" sound and the "kissy" sound are both examples of clicks that English speakers often make.


I have a question how do we write posts in it? (sample idea imminent)
I mean you *high whistle* thought that *click*d not write *tsk* forum in it?
(translation) I mean you really thought that we'd not write on the forum in it?
note: sample needs work

[BornValyrian]

I have a question how do we write posts in it? (sample idea imminent)
I mean you *high whistle* thought that *click*d not write *tsk* forum in it?
(translation) I mean you really thought that we'd not write on the forum in it?
note: sample needs work

If we use these (I don't want to, but that's me) we'll probably use non-letter characters. Using your imminent example: I mean you ^ thought that ''d not write * forum in it.

Of course some punctuation might not be used in this language.

[lsfreak]

I have a question how do we write posts in it? (sample idea imminent)
I mean you *high whistle* thought that *click*d not write *tsk* forum in it?
(translation) I mean you really thought that we'd not write on the forum in it?
note: sample needs work

Most (all?) of the South African Bantu languages use c q x, as they're unused in the orthography, with modifiers like xh (aspirated) or ngq (prenasalized breathy). The International Phonetic Alphabet uses ǀ ǁ ! for the same. That's one of the reasons I said to avoid them, they're a bit of a nightmare to Romanize well. The languages they occur in are also largely of CV shape; that is, they have no closed syllables, which as I said above I think is largely incompatible with polysynthesis (take a look at this (http://wals.info/combinations/22A_12A#2/26.7/153.0) map, where all the languages that have the most synthesis on the verb also have complex syllable structures).

[Silverbit]

Neither.

Allow closed syllables.

A lot of vowels.

Yes to diphthongs.

A lot of consonants.

No clicks.

[Fortuna]

I find it curious how many people are bold-voting, considering that our illustrious OP has said this period is going to be one without voting. :smallconfused:

In any case, I don't really feel all that strongly on a lot of those points, but I definitely want clicks in - clicks are cool. I wonder whether it might be viable to make a language whose only consonants were clicks? I mean, from an informational perspective there's no problem, so why not? Might be kind of interesting.

[Xuc Xac]

I find it curious how many people are bold-voting, considering that our illustrious OP has said this period is going to be one without voting. :smallconfused:


The OP also said we're voting and the questions are "on the ballot". It wasn't very clear.

[Everyl]

The OP also said we're voting and the questions are "on the ballot". It wasn't very clear.

Given the future tense used in the post in question, I was under the impression that we are discussing things that will be put to a vote on a ballot at some point in the future. Like, the options we pitch now will become the choices we vote for, since many of them have open-ended or nearly open-ended possibilities.

And now for my opinions/ideas...

Tone/vowel length:
If we're going to use tone, we should keep it simple. Tone is difficult to convey in Roman letters, and tricky for native speakers of non-tonal languages to detect and communicate; tone-deaf people can find it nearly impossible to master, even. Anything much more complicated to distinguish than rising, falling, and flat tones would make things too difficult, in my opinion.

Vowel length, on the other hand, doesn't seem very difficult to me. Maybe I've been living in Japan too long, but I find it to be one of the easier phonetic forms that doesn't exist in English to hear and produce. It's not hard to write, either, if you have a reasonable-sized pool of vowels. Short 'e' versus long 'ee' - same sound, said for different durations. It only gets tricky if you have larger numbers of vowels and/or diphthongs, like in English.

I don't have much of an opinion on open vs. closed syllables. I'd say go with whatever makes it easiest to make big, polysynthetic compound words pronounceable.

I'd say, keep the vowel count close to average. It makes the written form much easier, and avoids the need for overloading basic vowels when writing the language using Roman letters. Maybe I'm too concerned with avoiding using IPA characters, though.

Diphthongs and consonant count - whatever works well for building polysynthetic words. I don't know enough about the subject to have a strong opinion on that.

As for clicks, I agree with the ideas above that, if there are clicks, they should have special grammar functions. Spoken punctuation, emphasis, an audible separator between words that shouldn't be fused, something like that. I don't think that clicks would flow well in a polysynthetic compound words, but I think it could be cool to have them in the language somehow, if done well.

[Ozy]

Vowels:
I think orthography is an important thing to consider here. The roman alphabet only really has seven symbols which could be used for vowels without people doing a double take (a, e, i, o, u, y, and w), and I find digraphs to be a bit unwieldy, so I think it'd make the most sense to have just seven vowels. We could possibly have a length distinction if we want more phonemic vowels, or add some allophones for character. Having 12+ vowels could be cool, but I think it'd probably just be too hard to deal with

I'd argue no tones, for the same reasons as the people above.


Consonants:
I'd like a medium number of consonants, but a large number would also be okay. Sparse consonant inventories seem very meek.


General sound/a weird idea:
So I really like the look and feel of consonant clusters, but at least a few people wanted a more flowy, vowelcentric language. I came up with a compromise which is sincerely odd, but I thought I'd post it regardless.

Basically, I was wondering what would happen if we had sonorancy strictly increase (or increase with slight variation) over the sentence -- grinding, crushing, crumbling sounds steadily giving way to airier and breezier ones. The end of each sentence would be significantly lighter than the start of the next, which might sound a bit odd, so maybe there could be a set of syllables considered typical of each weight/sonorancy (like solfege syllables, kinda) that could act as downward steps from airier back down to darker.

For example, if we had five different heights of syllable (1 being the darkest sounding, 5 being the lightest sounding), and one sentence ended with a syllable at height 5, and the next sentence started with one at height 2, then the proto syllables 4 and 3 would occur between the sentences, kinda waterfalling down between levels of sonorancy.

Yeah so that was basically the idea. If you consider complex gender systems it's possible that one could evolve into something like this, but it's still a bit far out there.

Could be cool? But I dunno.


Specific phoneme suggestions:
v, w, l, dz, ts, dl, tl

And I'll give a tentative no on clicks - if they really make other people excited we should go for them, but they could be hard to deal with and might drive this language towards kitchen sinkiness

Maybe we could have some weird allophony like [v~l]. I can't think of any mechanism which would cause that to happen, but it'd be neat.

[BornValyrian]

[
General sound/a weird idea:
So I really like the look and feel of consonant clusters, but at least a few people wanted a more flowy, vowelcentric language. I came up with a compromise which is sincerely odd, but I thought I'd post it regardless.

Basically, I was wondering what would happen if we had sonorancy strictly increase (or increase with slight variation) over the sentence -- grinding, crushing, crumbling sounds steadily giving way to airier and breezier ones. The end of each sentence would be significantly lighter than the start of the next, which might sound a bit odd, so maybe there could be a set of syllables considered typical of each weight/sonorancy (like solfege syllables, kinda) that could act as downward steps from airier back down to darker.

For example, if we had five different heights of syllable (1 being the darkest sounding, 5 being the lightest sounding), and one sentence ended with a syllable at height 5, and the next sentence started with one at height 2, then the proto syllables 4 and 3 would occur between the sentences, kinda waterfalling down between levels of sonorancy.

Yeah so that was basically the idea. If you consider complex gender systems it's possible that one could evolve into something like this, but it's still a bit far out there.

Could be cool? But I dunno.


Specific phoneme suggestions:
v, w, l, dz, ts, dl, tl


This sounds complicated and awesome. Does this basically mean that as a sentence progresses it gets flowier, with less consonants and more vowels as you progress through the sentence?

[Jendekit]

General sound/a weird idea:
So I really like the look and feel of consonant clusters, but at least a few people wanted a more flowy, vowelcentric language. I came up with a compromise which is sincerely odd, but I thought I'd post it regardless.

Basically, I was wondering what would happen if we had sonorancy strictly increase (or increase with slight variation) over the sentence -- grinding, crushing, crumbling sounds steadily giving way to airier and breezier ones. The end of each sentence would be significantly lighter than the start of the next, which might sound a bit odd, so maybe there could be a set of syllables considered typical of each weight/sonorancy (like solfege syllables, kinda) that could act as downward steps from airier back down to darker.

For example, if we had five different heights of syllable (1 being the darkest sounding, 5 being the lightest sounding), and one sentence ended with a syllable at height 5, and the next sentence started with one at height 2, then the proto syllables 4 and 3 would occur between the sentences, kinda waterfalling down between levels of sonorancy.

Yeah so that was basically the idea. If you consider complex gender systems it's possible that one could evolve into something like this, but it's still a bit far out there.

Could be cool? But I dunno.


Yes, yes, yes! I have never heard (pun not intended) of something like this in a language before, and I find this to be far more interesting than clicks as part of the language. This gets my vote.

[lsfreak]

I find it curious how many people are bold-voting, considering that our illustrious OP has said this period is going to be one without voting. :smallconfused:
Because we're all idiots, clearly. (I skimmed and read right over the part where we're not voting now.)


I wonder whether it might be viable to make a language whose only consonants were clicks? I mean, from an informational perspective there's no problem, so why not? Might be kind of interesting.

Only having clicks pretty much demands CV-only syllable structure. They're just too hard to cluster together. They're also hard to produce, and from a realism point of view (if that's something you value, I do) it would be expected that they de-click into easier-to-pronounce consonants if there's no easier-to-pronounce consonants already to there to keep it from happening. Also, it's something that's hard for people to pick up, so if we choose to go a click-heavy route, we need to accept that it's not something many of the people who would like to use, can.

So here's some thoughts:
Open syllables only likely makes things harder to pronounce, not easier. Because as the word drags on longer and longer, unstressed vowels will naturally start falling out in order to shorten what you have to pronounce. Compare these two, which is easier to pronounce quickly with stress on the first syllable?
ka.mi.su.che.fi.ma.sa.yi
kam.such.fim.si
The big thing isn't disallowing clusters, but being careful that they don't stack up to inordinate length. But as I said, I like consontant clusters, so I am a bit biased.


Maybe we could have some weird allophony like [v~l]. I can't think of any mechanism which would cause that to happen, but it'd be neat.
L velarized next to back vowels, which turns into [w] and then hardens into [v]. :smallamused:
The big problem with weird allophony is that the more sounds you have, the less room there is - other consonants take up enough acoustic and articulatory space that they're unlikely to happen. Not to say it's impossible, but it's the very small inventories that have the weird allophones, because they have so much "open space" for them to fill.

(I had more thoughts, but must run to work now.)

[Balyano]

I would like to see i, e, a, o, u as the vowels in their normal state with y and w from vowel harmony.
Have the first vowel in a word set its harmony.
Have i and e shift to w, have o, and u shift to y.
An a can go with either group with its front/back forms being allophonic.
A long vowel becomes a diphthong.
The second component of a diphthong sets the vowel harmony for the following syllables.
A sequence of two or more consonants blocks the harmony and resets it for the following syllables.
Affricates count as a single consonant for vowel harmony.

the result


basic form front harmony back harmony
i = i i ɤ
e = ɛ ɛ ɤ
a = ä a ɑ
o = ɔ ø ɔ
u = u ø u
ii = iː iː ui̯
ee = ɛː ɛː ɔi̯
aa = äː aː ɑː
oo = ɔː ɛu̯ ɔː
uu = uː iu̯ uː

written as
ø y
ɤ w
ä/a/ɑ a
iu̯ iu
ɛu̯ eo
ɔi̯ oe
ui̯ ui

allophone
ɬ θ before i or u
ʟ v before o or u
tɬ tʃ before i, e, or y

an example

tlek tlek
tɬɛk > tʃɛk

kutlek kutlwk
ku-tɬɛk > kutɬɤk

kutlekti kutlwkti
ku-tɬɛk-ti > kutɬɤkti kt blocks vowel harmony for i

kutlekti kutlwktimy
ku-tɬɛk-ti-mo > kutɬɤktimø first half of the word has back vowel harmony, second half has front vowel harmony due to kt sequence

kutlekiimo kutlwkuimy
ku-tɬɛk-iː-mo > kutɬɤkui̯mø i̯ component of diphthong resets the vowel harmony of later segments to front vowel harmony

[Owlglass Moot]

1. In terms of tones and vowel length, I'd prefer neither. Phonemic tone with polysynthesis sounds especially nightmarish to me, ack.

2. Closed syllables, for delicious consonant clusters! And what does everyone think of maybe allowing liquids or nasals to function as syllabic nuclei in some cases, like in Proto-Indo European?

3. Average number of vowels would be my preference. And yes to diphthongs!

4. Average number of consonants. I know we're going for reasonable English sounds, but I'd love to see the inclusion of some that are realized phonetically (but not phonemically) in English, like [ɬ] (the "l" in "please" in many dialects).

5. Clicks are fun, but if we're going to go balls to the wall we should do ejectives instead. :smallbiggrin:

Edit: Also, is there any love for vowel harmony?

[Ajadea]

I like the idea of a tonal language with mostly-open syllables, allowing only a short list of consonants at the end. However, as someone with experience with Chinese, I will freely admit my bias towards being able to understand and pronounce such things. I like the idea of producing a more flowy, almost melodic language.

An average number of consonants and vowels would help streamline this language a bit, keep it from getting too complicated. I don't know enough about languages to put in an opinion on diphthongs unfortunately. In a similar vein, clicks might be difficult to represent and pronounce.

[Recherché]

I'm not a linguist (i just live with one) so I'm not actually sure about all the options presented.

1: Tonal and vowel length seem like they'd be pains to type out and pronounce so no thank you

2:Closed syllables sound good though perhaps they should be limited to certain phonemes. Too many in a polysnthetic language sounds like it would be a nightmare

3: For some reason I like the idea of a high number of vowels and dipthongs

4: To keep from having too many phonemes this language should probably have a medium to low number of consonants

5: I like the idea of the occasional but rare click

And yeah mild vowel harmony sounds cool

[Xuc Xac]

1: Tonal and vowel length seem like they'd be pains to type out and pronounce so no thank you


Tones are easy to represent. You can use accent marks like Vietnamese (ma má mã mà mạ mả) or you can use tone numbers like Pinyin (ma1 ma2 ma3 ma4).

Vowel length is even easier to write. You can use a long vowel marker like Māori or just double the vowels like in Finnish.

[BWR]

I support using diacritics. Now having tone and length on the same vowel will be difficult - that would almost necessitate using double vowels to represent length unless someone knows how to stack diacritics on the screen.

[nelk114]

Hi. I just found this and, having an interest in linguistics, thought it would be interesting to take part in.

I like the idea of a polysynthetic language, given the (relative) conciseness it can yield.

As for the current part, my initial thoughts are along the lines of:
--At least one of tone and/or length: these are both interesting features and could give syllables in the polysynthetic language a fairly compact variety (a combination perhaps--short vs long falling syllable? Or perhaps a simpler pitch-accent or stress-based system?)
--A fairly large phoneme inventory, although perhaps weighted towards either vowels or consonants (probably consonants, given that vowels would already have distinctions as above). Not to large, though, to allow some kind of allophony and/or harmony...
--Closed syllables, though for ease of pronunciation perhaps with limited codas for ease of pronunciation. I actually quite like Ozy's idea with sentences (or perhaps more loosely clauses) getting continually more vowelly as they progress. This could allow, for clicks to be allowed, but only at or near the beginning of a sentence...
--Diphtongs I would tend to keep to a minimum, but allow others to be composed from individual vowels as in lsfreak's suggestion...

In any case, this could be interesting...:smallsmile:

[Owlglass Moot]

Stress patterns are really interesting, and I kind of like them as an alternative to tones and vowel length.

Basically a deviation from the normal stress of the word would have an effect on the meaning. Like in the English word "record". It can be record (verb; to record something) vs. record (noun; that which is recorded).

It could be more regular than in English, though. Like pluralizing a noun could be done by shifting the stress to the final syllable or something.

[nelk114]

I was looking back over the original post, and have noticed that we seem to be doing it slightly wrong...
At the moment, I think we are actually supposed to be discussing what to put in the next ballot, i.e. the options on which we will vote.
Looking over the previous posts, I have compiled some of the suggestions:
--Allophones
--Vowel harmony and by extension perhaps consonant harmony...?
--General sound i.e. vowelly/consonant-heavy...or perhaps Ozy's idea...
--Stress/pitch accent as alternatives to tone

If i've missed any, feel free to correct me, or, as I believe we are supposed to do, add some more:smallsmile:

EDIT: I seem to have missed Consonant gemination as a corollary to vowel length and Other kinds of consonants--aside from pulmonic consonants and clicks, there are things like pops, ejectives, etc...

Also, one more suggestion of my own:
--Different vowel/consonant distinctions e.g. nasalisation/breathy voice for vowels or voicing/palatalisation/aspiration for consonants

[Ozy]

so I compiled y'all's opinions on the matters at hand.
here's a big summary. It's broken into categories with their own summaries below, each of which also comes with all related quotes (unless I goofed up, which is probable)

SUPER SUMMARY
- Tone and vowel length are up in the air and should both be voted on.
- Closed consonants are a shoo-in and we should instead vote on whether or not to limit the coda.
- We should vote between a medium and large amount of vowels.
- Diphthongs seem to be a definite yes.
- A medium amount of consonants.
- We should vote both on whether or not we want clicks and whether or not they should be treated separately from other sounds.

CATEGORY SUMMARIES
TONE
- "Tonal for my sake"
- "Both"
- "Tonal"
- "If we're going to use tone, we should keep it simple."
- "Limited ... a two- or maybe three-tone register system, where a given word (or set of morphemes) takes a high, medium, or low pitch. Or perhaps Japanese-style, where a given syllable drops or "resets" the whole pitch of the word."
- "these are both interesting features and could give syllables in the polysynthetic language a fairly compact variety (a combination perhaps--short vs long falling syllable? Or perhaps a simpler pitch-accent or stress-based system?)"

- "I don't think [tones] would be helpful"
- "Neither"
- "No tones"
- "Neither"
- "No thank you"
- "Tone no"
- "Skip tonality"
Summary: 6 people for tonality, 7 against. Of those who were for, 3 wanted a limited system.

VOWEL LENGTH
- "it might be fun to try both length and tone."
- "Both"
- "Long vowels yes"
- "I like the idea of [vowel length]"
- "Vowel length ... doesn't seem very difficult to me." (pretty sure this was tacitly for length)

- "Neither"
- "Neither"
- "No thank you"
- "I think length could work but ... I recommend against it."Summary: 5 people for vowel length distinction, 4 against

CLOSED VS. OPEN SYLLABLES
- "Closed"
- "Closed syllables ... with a limited number of consonants that are allowed in the syllable final position"
- "Closed syllables with a limited inventory of final consonants that assimilate to the following consonant, if any."
- "Closed syllables ... restricted to finishing with non-stop sounds"
- "Closed allowed. Given polysynthesis, I think we need closed syllables. Open-only syllables would likely be unstable"
- "Allow [closed syllables]"
- "Whatever makes it easiest to make big, polysynthetic compound words pronounceable." (this would be closed syllables, as lsfreak pointed out)
- "Closed syllables, for delicious consonant clusters!"
- "Closed syllables sound good though perhaps they should be limited to certain phonemes"
- "Closed syllables, though for ease of pronunciation perhaps with limited codas for ease of pronunciation."

- "I'd prefer open only or a very short list of closable consonants ... on the other hand going with ALL closed would be nifty too"
- "Mostly open"Summary: 10 for closed syllables, 2 against. Of those who were for, 5 wanted limited codas.

VOWELS
- "I like a lot of vowels, but I'm more than willing to give up on this point, or possibly have a number of allophones that pop up in a consistent pattern."
- "Large with vowel harmony of some kind."
- "Lets go largish-12 works for me"
- "A lot"
- "A high number"

- "5 pure short vowels ... with [3 long vowels]"
- "5-10 basic vowels ... something like i, e, a, o, u, with diphthongs and the remaining pure vowels represented with digraphs."
- "5-10 vowels"
- "Close to average"
- "Just seven vowels"
- "Average"
- "Average"
Summary: 5 people want a large amount of vowels, 7 want an average amount.

DIPHTHONGS
- "Dipththongs and more. Tripthongs, perhaps even more."
- "Yes. I would go so far as to suggest occasional triphthongs (with a limit on vowel length, maybe a maximum of 1 long vowel)."
- "Yes"
- "Include diphtongs"
- "Yes"
- "Yes"
- "A high number"

- "No"
- "keep to a minimum"
- "Certain vowel clusters may be realized as diphthongs or triphthongs, but they are fundamentally different vowels. mu + a + i is three vowels that happen to be pronounced as a triphthong, but there's not an actual triphthong uai."Summary: 7 for, 3 against. 2 also were into triphthongs.

CONSONANTS
- "About 20 consonants, give or take"
- "A lot"

- "Medium or large"
- "Average"
- "Average"
- "Medium"
- "Average"
- "Average"
- "Medium to low"
- "A smaller amount, like 10-15.

- "Few consonants, no more than 10 or so, perhaps 7-8"Summary: 2 for a large amount, 8 for a medium amount, 1 for a small amount.

CLICKS
- "No"
- "No"
- "No. Just too complicated, to pronounce, to incorporate, to put into an easy orthography"
- "No"

- "Occasional but rare"
- "Only as interjections"
- "If there are clicks, they should have special grammar functions. Spoken punctuation, emphasis, an audible separator between words that shouldn't be fused, something like that."
- "Good gods yes ... possibly a limited set of unique sounds used to augment a language" (i.e. of special syntactic function)
- "Clicks are fun, but if we're going to go balls to the wall we should do ejectives instead"
- "I definitely want clicks in - clicks are cool."Summary: 4 Against, 6 for. 4 expressed interest in limiting clicks to a special syntactic function, and not letting them appear in normal words

And here's a quotedump of the miscellaneous proposed ideas. I don't have the time right now to go through it, but nelk has a pretty comprehensive list above and here's the raw data anyway- "I'm envisioning something very liquid and heavy on the vowels. Breezy and breathy, perhaps."
- "I vote against phonemic tone, but I think it would be a good idea to use intonation to distinguish the beginning of a word in speech."
- "It may also be a good idea to have certain phonemes be a little ''loose'', with a lot of leeway for allophony. Perhaps allowing θ in place of ɬ and tʃ in place of tɬ if the speaker has trouble with lateral fricatives."
- "I like BWR's idea of a "breezy" language, more flowy."
- "Maybe we could have some weird allophony like [v~l]"
- "v, w, l, dz, ts, dl, tl"
- "So I really like the look and feel of consonant clusters, but at least a few people wanted a more flowy, vowelcentric language. I came up with a compromise which is sincerely odd, but I thought I'd post it regardless.

Basically, I was wondering what would happen if we had sonorancy strictly increase (or increase with slight variation) over the sentence -- grinding, crushing, crumbling sounds steadily giving way to airier and breezier ones. The end of each sentence would be significantly lighter than the start of the next, which might sound a bit odd, so maybe there could be a set of syllables considered typical of each weight/sonorancy (like solfege syllables, kinda) that could act as downward steps from airier back down to darker.

For example, if we had five different heights of syllable (1 being the darkest sounding, 5 being the lightest sounding), and one sentence ended with a syllable at height 5, and the next sentence started with one at height 2, then the proto syllables 4 and 3 would occur between the sentences, kinda waterfalling down between levels of sonorancy."
- "This sounds complicated and awesome." (on above)
- "Yes, yes, yes!" (on above)
- "I would like to see i, e, a, o, u as the vowels in their normal state with y and w from vowel harmony.
Have the first vowel in a word set its harmony.
Have i and e shift to w, have o, and u shift to y.
An a can go with either group with its front/back forms being allophonic.
A long vowel becomes a diphthong.
The second component of a diphthong sets the vowel harmony for the following syllables.
A sequence of two or more consonants blocks the harmony and resets it for the following syllables.
Affricates count as a single consonant for vowel harmony."
- "maybe allowing liquids or nasals to function as syllabic nuclei in some cases"
- "I'd love to see the inclusion of some that are realized phonetically (but not phonemically) in English, like [ɬ]"
- "Any love for vowel harmony?"
- "I like the idea of producing a more flowy, almost melodic language."
- "mild vowel harmony sounds cool"
- "allow some kind of allophony and/or harmony..."
- "I actually quite like Ozy's idea with sentences (or perhaps more loosely clauses) getting continually more vowelly as they progress. This could allow, for clicks to be allowed, but only at or near the beginning of a sentence..."
- "Stress patterns are really interesting, and I kind of like them as an alternative to tones and vowel length."
- "Will the concept of long consonants be addressed later?"
- "Perhaps there's a difference between egressive and ingressive sounds."

Oh and
Does this basically mean that as a sentence progresses it gets flowier, with less consonants and more vowels as you progress through the sentence?Yep! :smallsmile:

[The Anarresti]

Hey guy! So sorry I haven't responded in a while, I kinda left you all hanging. Real life, donch-y'noe. On the other hand, I am so incredibly pleased that everything has been functioning smoothly in my absence!

A big thanks to Ozy for compiling everything in a handy-dandy summery: I'll set up an official ballot for the clear-cut issues right away, and start parsing through the miscellaneous proposals in order to present them in a voteable manner.

Quick straw poll: [B]Based on how quickly the average poster in this thread responds, how long should the polls be open for?[/B

Really, HUGE thanks to Ozy. And to everyone else who has taken a share of administrative tasks upon their own shoulders. I really appreciate it. It's amazing, seeing this thread come together in a true community-project fashion so quickly. *sniff*

[Balyano]

Based on how quickly the average poster in this thread responds, how long should the polls be open for?

Perhaps 4 days to discuss things followed by 3 days of voting? Rinse and Repeat.

[BWR]

Based on how quickly the average poster in this thread responds, how long should the polls be open for?

Perhaps 4 days to discuss things followed by 3 days of voting? Rinse and Repeat.

Sounds good. That should leave most of us enough time to read and think things through.

[Ozy]

That seems like a pretty good plan to me too

[Grinner]

And me as well.

It's important to keep the project from stagnating.

[Owlglass Moot]

Sounds fine to me as well.

[BornValyrian]

I'm fine with this too

[Balyano]

Perhaps we should have a vote soon to revitalize interest in this thread?

[Grinner]

Something, please.

I may not be much of a linguist, but I did want to follow this project through to completion, if only to get a better grasp of the process.

[Ozy]

So since the anarresti hasn't been here in a while (maybe more real life attacked?) Why not let's just go ahead and vote on the things which we obviously need to vote on.

BALLOT:
- Should the language be tonal or atonal?
- Should the language have a vowel length distinction or not?
- Closed syllables are allowed, but do you want there to be a limited or open selection of codas? (Do you want restrictions on how syllables can end? Also discuss what restrictions you might want)
- Do you want a medium or large amount of vowels?
- Do you want no clicks, clicks which act like any other sound, or clicks, but with special grammatical purpose


As always, please put jo votes in bold.
Voting to end on Friday.


If you want things to discuss:
- Allophony
- Vowel Harmony
- General sound. E.g. breezier or earthier or something crazy like the idea I posted earlier.
- Consonant gemination
- Liquids or nasals, rather than just vowels, as syllabic nuclei
- Stress patterns
- Uncommon consonant/vowel distinctions (egressive/ingressive, breathy, creaky, labialization, etc.)

And maybe suggest any specific sounds you'd want in the language?
https://en.wikipedia.org/wiki/IPA <- sound resource

(Is this okay? or do you think we should wait for OP more)

[Ozy]

I'll myself go with:
- atonal
- yes length distinction
- non-limited codas
- medium amount of vowels (ideally 7 before we consider length distinction)
- clicks with some special purpose

[Everyl]

Atonal
No length distinction
Limited codas, but I don't have time to research examples today
Medium vowels, I like languages without the overabundance of vowels of English.
Clicks, but only with special purpose

[BWR]

Tonal
Length distinction
non-limited codas
medium-plus to lots of vowels (11-12)
No clicks

[Balyano]

Atonal - but only in the sense that no root has an inherent tone, I'm all for having tone be used to show word boundaries.
Length Distinctions
Limited Codas - preferably something like t, s, r, l, n, with some degree of assimilations so that a final stop assimilates to a following one yielding a geminate stop (sot + ka > sokka vs so + ka > soka), similar rule with a nasal coda and a following consonant (pan + pit > pampit, kan + mir > kammir vs ka + mir > kamir)
Medium Vowels - preferably something we can write with i, e, a, o, u, y, w and use ii, ee, aa, oo, uu, yy, ww for long vowels
No Clicks - more than willing to concede this one if the clicks are a simple yes/no/i don't know vocal short hand like uh-uh and uh-huh and a few others that I have no idea how to type, or perhaps a form of emphasis that replaces the first consonant of an emphasized word with a particular click based on the consonant being replaced

[BornValyrian]

Atonal, Though if tone was used for other things, I wouldn't mind
Length Distinctions
Limited Codas: I'm thinking mainly cutting out stops like b,t,k, etc.
Medium number of Vowels like 7 to 10
No Clicks though if they make a minor appearance, that won't bother me too much.

As for the discussion topics, for the general feel I like the breezier language or Ozy's craziness.
If I understand vowel harmony, its that certain vowels are paired together in words. So if a word has an "i" in it, but the word starts with "a", the "i" would switch to "e" by merit of the "a"? If so, that sounds kinda cool.
Stress patterns: I think a simple one like the middle syllable of a word, with the last syllable in two syllable words

[sktarq]

Atonal
No Length Distinctions
Limited Codas
Medium Vowels
Clicks, for special grammer

[nelk114]

For my vote, I'll go:
--Tonal, but tending towards a more limited pitch-accent (http://en.wikipedia.org/wiki/Pitch_accent) system, i.e. a kind of middle ground between stress and tone.
--Vowel length distinction
--Limited codas
--Given tonality and length, Medium number of vowels
--Clicks, allowable in any word, but possibly heavily constrained

Discussionwise, I'm for vowel harmony, and it would be nice to see some kind of strange consonant/vowel distinctions - perhaps palatalisation/aspiration (or both as allophones) for consonants and nasal in addition to open vowels. I'm fine either way on gemination.
As for general sound, I've taken to like Ozy's 'craziness' - if we did this, I would be for having clicks (and while we're at it perhaps ejectives...) limited to the most consonant-heavy areas, perhaps even just the first few syllables of the first word (taking into account that our language is polysynthetic).
As for specific consonants...how about k͡p (https://en.wikipedia.org/wiki/Voiceless_labial-velar_stop), ç (https://en.wikipedia.org/wiki/Voiceless_palatal_fricative), ɰ (https://en.wikipedia.org/wiki/Velar_approximant), and/or ʙ (https://en.wikipedia.org/wiki/Bilabial_trill) (just a few interesting suggestions)?

[Xuc Xac]

- tonal
-a vowel length distinction
- Closed syllables, a limited selection of codas
- a medium amount of vowels
- no clicks

[Recherché]

-atonal
-no vowel length
-limited closed syllables
-large number of vowels
-Clicks as special markers

I'm definitely in for some mild vowel harmony though we probably should try not to make it too difficult to learn.

One thought I had would be to have clicks mark the ending of the one sentence-word/the beginning of another

[Ozy]

Hey guys - something came up a few days ago, so I won't be able to be super active in this thread for a bit. (It's nothing bad, but it has caused me to lose what free time I had)

Could someone else tally up the votes and think of what to work on next? Our fearless leader still seems to be AWOL.

[Balyano]

So far it looks something like this unless I miss counted.

3 tonal
6 atonal

6 length distinction
3 no length distinction

0 no codas
7 limited codas
2 unlimited codas

7 medium vowels
1 medium plus to lots of vowels
1 large number of vowels

4 no clicks
4 clicks with special purpose
1 clicks but heavily constrained

so it looks like
atonal
length distinctions
limited codas
medium vowels
and a tie between no clicks and clicks with a special purpose

considering this


No Clicks - more than willing to concede this one if the clicks are a simple yes/no/i don't know vocal short hand like uh-uh and uh-huh and a few others that I have no idea how to type, or perhaps a form of emphasis that replaces the first consonant of an emphasized word with a particular click based on the consonant being replaced


No Clicks though if they make a minor appearance, that won't bother me too much.

it appears that it will be clicks with a minor special purpose

[Balyano]

In the hopes of getting the talking going again I figured I would throw out a couple ideas. Hopefully people will respond with their own ideas. Describe some ideas of your own, talk about what you like or hate from other peoples ideas, modifications to those ideas to make them better, etc. Perhaps if you like an idea you can post an idea that works well with it. Or if you hate an idea give an alternative to it. Or if you like an idea but it doesn't work or could be better give some modifications or at least some insight on what's wrong with it.

So here are a couple ideas of my own. Each has some examples with the words made up randomly without taking each other into consideration.

This is an idea I mentioned previously about vowel harmony.

I would like to see i, e, a, o, u as the vowels in their normal state with y and w from vowel harmony.
Have the first vowel in a word set its harmony.
Have i and e shift to w, have o, and u shift to y.
An a can go with either group with its front/back forms being allophonic. A long vowel becomes a diphthong.
The second component of a diphthong sets the vowel harmony for the following syllables.
A sequence of two or more consonants blocks the harmony and resets it for the following syllables.
Affricates count as a single consonant for vowel harmony.


basic form front harmony back harmony
i = i i ɤ
e = ɛ ɛ ɤ
a = ä a ɑ
o = ɔ ø ɔ
u = u ø u
ii = iː iː ui̯
ee = ɛː ɛː ɔi̯
aa = äː aː ɑː
oo = ɔː ɛu̯ ɔː
uu = uː iu̯ uː

written as
ø y
ɤ w
ä/a/ɑ a
iu̯ iu
ɛu̯ eo
ɔi̯ oe
ui̯ ui

allophone
ɬ θ before i or u
ʟ v before o or u
tɬ tʃ before i, e, or y

Some examples


tlek tlek
tɬɛk > tʃɛk

kutlek kutlwk
ku-tɬɛk > kutɬɤk

kutlekti kutlwkti
ku-tɬɛk-ti > kutɬɤkti kt blocks vowel harmony for i

kutlekti kutlwktimy
ku-tɬɛk-ti-mo > kutɬɤktimø first half of the word has back vowel harmony, second half has front vowel harmony due to kt sequence

kutlekiimo kutlwkuimy
ku-tɬɛk-iː-mo > kutɬɤkui̯mø i̯ component of diphthong resets the vowel harmony of later segments to front vowel harmony



An idea I had about verbs.
I thought maybe have it so the verb has a set of affixes attached to it that represent the objects and subjects. They would act like pronouns, demonstratives, articles, etc. There would be quite a few of them and they would each have their own implications. One affix may suggest the subject is not quite up to the task at hand, another may praise the object, another may imply the subject is young, another may imply that they are feminine, etc. Some might even imply different things based on what the referent is. Using a pronoun that means ''dog'' to a canine probably has neutral implication, but applying it to a soldier may imply loyalty or if meant insultingly it might imply mindless obedience.


examples:
For instance let's say a king who's name means iron is receiving poor advice from someone who's name means oak.

maseetsone pekaetretakisune kitaywatomeke
maseetsone pekaetretakisune kitaywatomeke
mase-et-so-ne peka-et-reta-ki-su-ne kitay-wa-tome-ke
oak-name-agent-third person iron-name-king-patient-voluntary-third person he (inadequite)-advize-he (praiseworthy)-past tense

Mr. Oak gave poor advice to his royal highness King Iron, who willfully listened.


But what if Mr. Oak wasn't an idiot and actually knew what he was talking about.

maseetsone pekaetretakisune kitaywatomeke
maseetsone pekaetretakisune tomewatomeke
mase-et-so-ne peka-et-reta-ki-su-ne tome-wa-tome-ke
oak-name-agent-third person iron-name-king-patient-voluntary-third person he (praiseworthy)-advize-he (praiseworthy)-past tense

Mr. Oak gave wise council to his royal highness King Iron, who willfully listened.


Having them attached to the verb means you can completely omit the subject or object or both and still have a complete sentence.

if we already know Mr. Oak is the one giving advice then:

pekaetretakisune kitaywatomeke
pekaetretakisune kitaywatomeke
peka-et-reta-ki-su-ne kitay-wa-tome-ke
iron-name-king-patient-voluntary-third person he (inadequite)-advize-he (praiseworthy)-past tense

His royal highness King Iron willfully listened to poor advice.


If we instead only knew the king was listening:

maseetsone kitaywatomeke
maseetsone kitaywatomeke
mase-et-so-ne kitay-wa-tome-ke
oak-name-agent-third person he (inadequite)-advize-he (praiseworthy)-past tense

Mr. Oak gave poor advice to a willing listener, who was a far greater man.


Or for brevity:

kitaywatomeke
kitay-wa-tome-ke
he (inadequite)-advize-he (praiseworthy)-past tense
An idiot gave poor advice to a superior individual.

Please discuss, post your own ideas, etc and keep this project alive.

[nelk114]

A rather lengthy response to the vowel harmony idea:
I suppose the vowel harmony system could be modified depending on the exact details of the eventual phoneme inventory? For example, say we ended up with geminated consonants (since we haven't voted on such features yet). I suppose these would reset harmony as well? Similar considerations would apply should we have stress, and perhaps with certain special consonants/consonant combinations. And how would it cooperate (hypothetically) with Ozy's system? Perhaps vowels get more independent and thus less allophonic/harmonious as the sentence continues...
Thinking about it further, it would be possible to have 2 identical sounds representing different sounds distinguished by their harmonic context; perhaps a little complex to learn, but interesting to think about...
Additionally, how about other types of vowel? Would nasal vowels affect all subsequent vowels (until a harmony reset)?
Now that I'm typing this, I've also come up with yet another idea: different 'levels' of harmony/reset... e.g. a double consonant or affricate would perhaps just reset position (i.e. front/back) but it would take a triple or geminated double consonant to reset nasalisation...
Also, perhaps an artificial reset could be used for some kind of emphasis?
Overthinking can be fun...!

As for the verbs idea, it seems interesting, and I'd probably go with it, but it's probably more relevant once we get onto morphology of some kind... Still, it could go well with your click-based 'form of emphasis'! Nice idea to think about.

I can't think of anything phonology-based of my own for the moment... I'm wondering, though, when are we going to deal with orthography, i.e. a (hopefully standard) written representation of the language? Even if we decide on a provisional one for the sake of convenience at the keyboard, it would surely be useful to have a sure way of knowing how to read/write any propositions we may have...

Food for thought :smallsmile:

[Balyano]

In response to nelk114's musings. I think geminates would count as two consonants for the purpose of harmony resets, but for nasals how about instead of two or three consonants resetting nasal harmony have it be voiceless consonants like p, t, k, s etc. reset nasalization, whereas a voiced consonant doesn't. So sanmaru would have a nasal ma and ru, but sanmaku would not nasalize the u from ku because the voiceless consonant blocked it. pambedbu would be all nasal where as pambetbu would not nasalize bu. An artificial reset for emphasis seems very natural and useful, and if harmony is shown in the orthography it would be like using italics or underlining a word.

[nelk114]

So you're suggesting, effectively, two independent harmonic systems coexisting?
For example, take the following hypothetical cases (N.b. this is just an example; we may or may not have these phonemes):
Pũka-->Pũkɑ
Pũga-->Pũgɑ̃
Pũkta-->Pũkta, and
Pũgda-->Pũgdã
Is that what you had in mind, or am I misunderstanding? Interesting in any case.
I suppose if we had palatalised consonants, for example, they would have their own effects (for instance fronting the harmony).

[Balyano]

So you're suggesting, effectively, two independent harmonic systems coexisting?
For example, take the following hypothetical cases:
Pũka-->Pũkɑ
Pũga-->Pũgɑ̃
Pũkta-->Pũkta, and
Pũgda-->Pũgdã
Is that what you had in mind, or am I misunderstanding? Interesting in any case.
I suppose if we had palatalised consonants, for example, they would have their own effects (for instance fronting the harmony).

That is exactly what I had in mind.

And lets not forget about non-stop consonants.
Pũla-->Pũlɑ̃ and so forth

Pũgi-->Pũgɯ~ (how did you add the tilde?)
or using the idea for the harmony i posted
Pũgi-->Pũgɤ~ (again how do I add the tilde to the letter?)

[Ozy]

I hear tell of a language with rhotic vowel harmony, which could be pretty fun to do


The Yurok language is well known among linguists for several unusual features. One is phonological: Yurok is among a very few languages of the world to have the rhotic vowel sound /ɚ/ among its phonemes (another such language is American English). In addition, Yurok has a rhotic vowel harmony process by which underlying non-high vowels /a/, /e/, and /o/ may become /ɚ/ in a word that has /ɚ/; for example, the root /nahks-/ 'three' becomes [nɚhks-] in the word [nɚhksɚʔɚjɬ] 'three (animals or birds)'.

but yeah in general some sort of harmony would add a lot to the language. I like the idea of two harmonies coexisting, too -- maybe a fronting/backing harmony and a rhotic/null harmony? (I personally don't really like the aesthetic of nasal vowels)


we might as well start thinking on a vowel inventory? Sounds I like: ɚ ɛ e u

[Balyano]

With the rhotic harmony we could possibly have it where a sequence of two vowels or a diphthong could break it, but in a different way. A diphthong may just not assimilate and block further harmony, but a sequence of two vowels may harmonize the first vowel then turn it into an approximate and block the harmony for the rest of the word.


pa - tɚn - ate
patɚnɚtɚ

pa - tɚn - e - ate
patɚnɻate

pa - tɚn - ajte
patɚnajte

or perhaps the diphthong becomes rhotic
pa - tɚn - ajte
patɚnaɹte

[nelk114]

The rhotic harmony seems like it could work. How would they interact (if at all) with rhotic consonants (i.e.'r's)? Perhaps a reset?
I still think it'd be interesting to have an alternative distinction between consonants besides voicing, for example aspiration or palatalisation (I also just remembered prenasalisation, which could be interesting...); these could perhaps interact with vowel harmony in their own ways.
In the presence of two independent vowel harmonies, I won't bother suggesting throwing consonant harmony into the mix...:smallsmile:

N.b. the tilde is in the 'combining diacritical marks' section of unicode; Times New Roman/Arial/Courier New should have it and some other fonts may as well. They can be accessed through Character map if need be. The tilde beside the letter is still comprehensible though.

[BWR]

This thread is doing what I wanted it to do: force me to look up a whole bunch of stuff to keep up with the discussion (my OE philology isn't that heavy on actual phonology).

The two sets of vowel harmony seem fun, but I wonder if we are going about this the wrong way. Should we build a language from intricate sound systems and add meaning later, or should we try to make a proper vocabulary and grammar and add flavor later? Because, correct me if I'm wrong, these proposed systems are flavor, not meaningful, right?

[Balyano]

Should we build a language from intricate sound systems and add meaning later, or should we try to make a proper vocabulary and grammar and add flavor later? Because, correct me if I'm wrong, these proposed systems are flavor, not meaningful, right?

I figure do some phonology so we know what to make the roots and affixes look like. Then some basic grammar to know what kinds of affixes and function words we will need. Then come up with a list of basic vocabulary that the language will need, roots and affixes and such. Then make words for the vocabulary list that fits the phonological rules we made. Then give it a tune up and make changes where necessary. So we probably shouldn't get too intricate at first. But we still haven't determined the phonology yet.

Then look for some gaps in vocabulary to fill, possibly take a subject like smithing, figure out a bunch of terminology that it needs, come up with words for those terms, possibly basing new roots off old ones to mimic the look of a natural language. For instance the root for anvil may look suspiciously like the root for boulder. The root for forge may look suspiciously like someone took the roots for fire and hole and stuck them together and deleted a sound or two. ect. Could even have multiple words derived from the same compound by simplifying it in different ways, simulating the compound being made and simplified at different times in the languages development or even that they were the same word developed two different directions by different dialects and were then borrowed between them.

example of the above idea lets pretend

doku = boulder
dokwa = anvil

tsil = fire
nefi = hole
tsilnej = forge looks like tsilnefi with the f deleted and the ei sequence turned into a diphthong
tsillef = brick oven formed from the same compound by instead assimilating the n to l making a geminate l, and deleting i

[BWR]

Ok, that makes sense. So what we really need to do first off is determine what to work towards. Did we ever reach a conclusion about what sort of basic sound we wanted? I suggested, and at least one other person liked, the idea of a very airy language filled with vowels and soft consonants. Are there any other suggestions? Because that might be a good thing for the next vote.

[Ozy]

For general sound, it looks like the only ideas were your breathier, breezier, melodic tongue, and my idea, which I've quoted in full below.


I really like the look and feel of consonant clusters, but at least a few people wanted a more flowy, vowelcentric language. I came up with a compromise which is sincerely odd, but I thought I'd post it regardless.

Basically, I was wondering what would happen if we had sonorancy strictly increase (or increase with slight variation) over the sentence -- grinding, crushing, crumbling sounds steadily giving way to airier and breezier ones. The end of each sentence would be significantly lighter than the start of the next, which might sound a bit odd, so maybe there could be a set of syllables considered typical of each weight/sonorancy (like solfege syllables, kinda) that could act as downward steps from airier back down to darker.

For example, if we had five different heights of syllable (1 being the darkest sounding, 5 being the lightest sounding), and one sentence ended with a syllable at height 5, and the next sentence started with one at height 2, then the proto syllables 4 and 3 would occur between the sentences, kinda waterfalling down between levels of sonorancy.

Could be cool? But I dunno.

Are there any other ideas? Should we discuss the pros/cons of each of those?

[BWR]

That is an interesting idea. Anyone else have suggestions before we start discussing things?

[Balyano]

I'll be honest I don't know what constitutes ''breezy'' ''airy'' ''dark'' or ''light''. Is breezy and airy something with lots of fricatives? or just voiceless fricatives? Is dark voiced and nasalized sounds and light voiceless ones? Are these music terms that everyone knows but me? Also how do I get superscript on this thing?

For Ozy's idea, if I understand it correctly, perhaps the ancestor of the language had multiple degrees of stress, i'm aware of real languages with three degrees of stress so why not. The roots contain aspirated, tenius and prenasalized stops, and voiceless fricatives. When they are suffixed onto another root the difference in stress, or a difference in stress in an ancestral version of the language, causes the aspirated stops to become tenius, the tenius to become voiced, and the prenasalized to become nasals, and the fricatives to become voiced fricatives. Further suffixes are even less stressed and what would have been tenius becomes voiced, voiced to voiced fricatives, nasals to nasal vowels separated by approximants, and the voiced fricatives to approximants. And have the vowels under go reduction with the progression until they become schwas.

Not sure how to do superscript in here, tried pasting but it de-superscripted it, so just pretend that ''ph'' means aspirated p and ''mb'' means prenasalized b.

for instance with labials
primary stress > secondary stress > unstressed
f > v > w
ph > p > b
p > b > v
mb > m > Vnw (Vn here means nasalize the preceding vowel)
perhaps

tɬ > ɬ > l
ts > s > h
or
ts > s > z

lets pretend we have

kat- phe- ŋgep-
katpɪnɰəv (the n there is supposed to be superscripted)

te- te- te-
tedɪðə

[Ozy]

I'll be honest I don't know what constitutes ''breezy'' ''airy'' ''dark'' or ''light''. Is breezy and airy something with lots of fricatives? or just voiceless fricatives? Is dark voiced and nasalized sounds and light voiceless ones?

Here's what comes to mind for me, at least, when I hear the terms light or dark.

Darkest:
- consonant clusters
- voiced sounds
- nasals and stops
- place of articulation in the back of the mouth
- consonantal nuclei (l, r , m, n)

Lightest:
- (C)V structure
- unvoiced sounds
- approximates
- articulated closer to the front of the mouth

I'd be curious how BWR's definitions compare, because 'light' and 'dark' aren't exactly clearly defined.


Also how do I get superscript on this thing?This converter (http://aveneca.com/jepax.html) is a pretty useful tool. If you don't know x-sampa, just remember that to superscript something, put an underscore before it.

I'm low on time now, so I can't say much on the bulk of your post, but it looked pretty cool and well thought out from what I saw. If we do decide consonant clusters are something central to a dark sound, those would probably make up the roots and the different stress levels would cause certain sounds to elide. Or something.

[BWR]

I'd be curious how BWR's definitions compare, because 'light' and 'dark' aren't exactly clearly defined.


I was thinking lots of vowels and vowel clusters, avoiding too many consonants breaking things up, a low number of 'hard' consonants like stops, affricates, plosives, etc. Most consonants would be 'soft' like approximants.
The idea I had, hard as it is to realize (in the sense 'make real'), was something that sounded sort of like a gentle breeze that picks up and dies down.

[BornValyrian]

This was basically what I thought with breezy or airy^

[Balyano]

perhaps discussion or a vote on

lots of consonant clusters
vs
occasional consonant clusters
vs
few if any consonant clusters

also perhaps discussion on what types of cluster
type example
CAVX - twas
CCVX - ktor
CCAVX - gvle
CCCVX - spta
SCVX - stum
CSVX - psel
CNVX - kni
CSCSCSCSACSCSSCSSCASCCCACCCCSSSAAVX - tfpskfdvrcxtsztfhklsbd☢jdgkqfsfrlep
etc.
C = any initial consonant
A = approximant or liquid
V = vowel or syllabic consonant
X = coda or lack there of
S = fricative
N = nasal


maybe a vote along the lines of

1: very breezy with lots of fricatives and approximants and few clusters
2: very choppy with lots of stops and clusters
3: something based on Ozy's idea with one style giving way to the other
4: something else entirely

what do you think we should vote on? What should the ballot look like?

are we counting the results of the last vote so that we have an language that is
atonal
vowels with length distinction
limited set of codas
medium number of vowels
clicks, but only with special purpose

or is this contested?

[BWR]

We probably should work within the constraints of the last vote, or else there would be little point in voting about it.


1: very breezy with lots of fricatives and approximants and few clusters
2: very choppy with lots of stops and clusters
3: something based on Ozy's idea with one style giving way to the other
4: something else entirely

This seems like a good thing to vote on. It should give us an idea of what sort of sounds we have to work with and what to aim for. Does anyone else have any ideas or comments they want to make before we get to the act of voting?

[Ozy]

Maybe instead of just voting for one of the ideas or another, we could do a ranked vote for this one? Since there's more than just two options, and since I'd be happy with a few of them, I think that could be a better way to decide.


I also think that we should be voting on:
-lots of consonant clusters
-occasional consonant clusters
-few if any consonant clusters

as well as a vote of:
-very breezy with lots of fricatives and approximants
-very choppy with lots of stops
-something based on Ozy's idea with one style giving way to the other
-something else entirely

(which is a slightly modified version of what you wanted). The reason for splitting them up like this is that we could still have some very consonanty clusters which still sound breezy (shwlsl), and no clusters but still be choppy (katakataka), and in general, I think that phonotactics and phonetic inventory aren't similar enought to combine into one vote.

Is that good logic? I'm also game for the vote to start whenever, to keep this party rolling.

[nelk114]

Seems like a good idea.
Perhaps we could discuss the kinds of clusteers after we have decided how frequent they will be and how it sounds.

[BWR]

Maybe instead of just voting for one of the ideas or another, we could do a ranked vote for this one? Since there's more than just two options, and since I'd be happy with a few of them, I think that could be a better way to decide.


Works for me. So, with the larger number being the preferred outcome:


-lots of consonant clusters 1
-occasional consonant clusters 2
-few if any consonant clusters 3


-very breezy with lots of fricatives and approximants 3
-very choppy with lots of stops 1
-something based on Ozy's idea with one style giving way to the other 2
-something else entirely 1

[Balyano]

-lots of consonant clusters 2
-occasional consonant clusters 3
-few if any consonant clusters 1


-very breezy with lots of fricatives and approximants 2
-very choppy with lots of stops 2
-something based on Ozy's idea with one style giving way to the other 2
-something else entirely 1

I ranked breeziness and choppiness the same because I like a mixture of the two. I would have given Ozy's idea a 3, but I fear that it could make orthography and/or learning difficult. It would obviously work for a real language and it's a cool idea that I like, but it might make the learning curve too steep.

[Ozy]

-lots of consonant clusters 2
-occasional consonant clusters 3
-few if any consonant clusters 1


-very breezy with lots of fricatives and approximants 2
-very choppy with lots of stops 1
-something based on Ozy's idea with one style giving way to the other 3
-something else entirely 2

[nelk114]

-lots of consonant clusters 1
-occasional consonant clusters 2
-few if any consonant clusters 1

-very breezy with lots of fricatives and approximants 2
-very choppy with lots of stops 1
-something based on Ozy's idea with one style giving way to the other 3
-something else entirely 2

Just out of interest (although it doesn't seem to be getting much in the way of votes), what would 'something esle entirely' entail?

[Balyano]

Just out of interest (although it doesn't seem to be getting much in the way of votes), what would 'something esle entirely' entail?

Could be a few different things. For instance an even mix of stops and fricatives so that the language has a mix of chop and flow that doesn't neatly break down as one becoming the other. Or perhaps a slightly complicated relationship between them with lenition and fortition brought on by different circumstances like stress, emphasis, nearby words, adjacent phonemes from affixes, etc. Sandhi effects could be another way to shift between chop and flow. And there are other options as well.

[nelk114]

Interesting, if potentially slightly complicated--perhaps a mild version of some kind of sandhi or lenition/fortition with an overall manner based on Ozy's idea would be interesting. Combining this with double vowel harmony and polysynthesis...:smallsmile:

[Ozy]

haha that would indeed be something. we seem to have a mind for complicated but awesome things :)

what are y'all thinking with regards to the vote? should we wait a bit more or just call it done?

[Balyano]

haha that would indeed be something. we seem to have a mind for complicated but awesome things :)

what are y'all thinking with regards to the vote? should we wait a bit more or just call it done?

Well the last vote had 9 people vote and this one has had 4 people so far. I say wait until Monday, perhaps being the weekend we will get some more traffic.

That said.

ATTENTION PEOPLE PLEASE VOTE!!!

[Balyano]

Well, now that the forums are back up who out there is ready to conlang? And shall we call the last vote we were on finished and move on to the next discussion or should we see if anyone else is going to vote?

[Jendekit]

Why don't we see just how many votes were made and for what so we can judge if the poll needs some more votes or not.

[Jendekit]

You know how sometimes when you're going through old files you stumble across an old project that you never finished and forgot about? I just found the beginnings of a conlang on my old flashdrive.The vocabulary is less than a page and a quarter, and the grammar just has sentence structure, some tenses, and a few other things of the like. I was using it for a Pathfinder game and also have some sample sentences at the end of the two page document.

If anyone is interested :smallconfused: I could upload part of it up here or in a PM.

[Balyano]

You know how sometimes when you're going through old files you stumble across an old project that you never finished and forgot about? I just found the beginnings of a conlang on my old flashdrive.The vocabulary is less than a page and a quarter, and the grammar just has sentence structure, some tenses, and a few other things of the like. I was using it for a Pathfinder game and also have some sample sentences at the end of the two page document.

If anyone is interested :smallconfused: I could upload part of it up here or in a PM.

I'm interested, but I think it should be its own page or pm, i don't think it should be here, could get confused for part of this one.

[lsfreak]

Well, this seems to have lost a lot of steam, but I've vaguely been entertaining Ozy's idea. What I take as the most important thing for going towards more "breathy" is a movement from having consonant clusters to having fairly strict CV syllables. This is very roughly one idea for how to get it to work from a semi-naturalistic point of view

What I came up with was a verb-initial language. Verb-first languages bias rather heavily towards prefixing, and as this is a polysynthetic language, affixes are likely to abound. Polysynethsis likely means, at minimum, two agreement affixes. As an example


ka- še- ko- li
2S- 1S- PRES- see
I see you

Very likely, more. For the sake of this thought experiment, as a rather extreme example, the verb agrees with five persons (subject, object, recipient [to whom], benefactive [for whom], and direct causative [X makes Y, X forces Y to do]) and can add three other arguments (indirect causative [X has Y do, X convinces Y to do], instrumental [with what], and comitative [with who]). The order is the verbal complex, then the arguments, then a locational prepositional phrase that agrees with its object, and a purpose clause formed with a participle. Top row is the type of morpheme, second is the word, third is gloss, fourth is clarifying to what the morpheme refers (too lazy for official glossing abbreviations, hopefully they're not too hard).


INDCAUS- RECIP- COM- DCAUS- INST- BEN- OBJ- SUBJ- tense- root SUBJ OBJ INST RECIP COM OBJ- PP OBJ PTCP- tense- root
tis- ar- šek- pja- k'us- ni- k'atš- vi- tsa- saskem tif šah nuq pape tsima k'atš- ri mahšim q'up- sek- talit
INDCAUS 3S COM- 1P INST 1P 3S 3S PAST take dog bone mouth father brother 3S in tent PTCP- FUT- burn
them [father] [brother]- we [mouth] us [bone] [dog] [tent]

We made them have the dog, with [its] brother, take the bones to father in the tent with [its] mouth for us, for burning [i.e. ritual]
Note that the root verb is TAKE in this sentence; the "made" and "have" in English are supplied by the direct and indirect causatives.
Also note that the order of affixes on the verb is arbitrary, except for putting object/subject/tense right next to the root, as I don't have any offhand knowledge as to what's likely.

Tisaršekpjak'usnik'atšvitsa[B]saskem tif šah nuq pape tsima k'atšri mahšim q'upsektalis

If you'll notice, there's no consonant clusters except across syllables, with the exception of pja. And in roots (in bold) there's a harder restriction - there's no clusters involving two stops. So how does this work? Well, say the root is stressed heavily - and then most of the morphemes before the root had their vowel deleted in the prehistory of the language. Compare:

Tisaršekpjak'usnik'atšvitsasaskem tif šah nuq pape tsima k'atšri mahšim q'upsektalis
Tsarškpik'snk'tšvitssaskem tif šah nuq pape tsima k'tšri mahšim qp'esktalit

And bam, the natural order of the sentence means that the huge conglomeration of prefixes has tons of consonant clusters, while once the verb is over, the nouns, adverbs, and adjectives are largely CV with a few limited CVC clusters. Prepositional phrases and participles, both taking limited verbal inflection, sprinkle in a bit of extra clustering later in the sentence, but nothing like the first verb.

Of course, such a long chain of prefixes isn't going to crop up all the time, and many sentences will be a lot shorter [for example, I see the dog's bone might be k'tšeškli šah tif] but I think it gets the point across as one method for making a sentence grow towards being more "windy." While it could be possible to do this with a different sentence order, I think verb-initial works best because of how it affects the rest of the sentence structure. For example, if it was verb-final, as Japanese (among many many others), you'd end up with all the nouns at the beginning of the sentence. As I presented it, that would mean everything gets more and more rough as the sentence goes on, and switching it up for clustery roots and sonorous affixes just seems far less likely to me than clustery affixes and sonorous roots (though I don't have any solid proof, just intuition).


On a COMPLETELY different note, a few things that are also likely to be perceptually "dark" would be r-colored/rhotic/retroflexed vowels (as has been mentioned as possible harmony), pharyngealization and retracted tongue root (r-colored vowels tend to involve this is well), low tones, creakiness on the vowel (BUT creakiness can often have a high tone), and low vowels (ah/aw vowels versus high ee/oo vowels). While "bright/light" would be advanced tongue root, high tones, breathy voice (which correlates with both of the previous), and high vowels (ee/oo). To draw a connection to what Ozy already said, voiced sounds correlate with low tone and voiceless sounds with high tone. To a lesser extent, rounded vowels are darker than unrounded vowels - i/ü, e/ö, or Turkish ı or Korean eu versus u.

Personally I don't think "back" consonants are necessarily dark: I'd say palatals like sh, ch are "brightest," then dentals/alveolars like t, s, z, th, then labials and velars like p k kh/x. Much darker than those are uvulars/pharyngeals/retroflex like q, French R, American L/R, and the Arabic Sounds™. The palatals have a tendency to pull vowels front and high the uvular/pharyngeal/retroflex pull vowels back and low, building off what Ozy said about front/back vowels and I said about high/low.
[Due to the ambiguity of "windy" and "sonorous" and "dark" and the other words we're using, I'll also throw out that sounds like English p t k I find much less sonorous, but brighter, than the likes of mb nd ng, which are more sonorous and darker, and don't find either more or less windy.]

[nelk114]

We appear to be in a period of stasis once again...
I wonder whether our OP is likely to resurface?

Anyway, I thought I might as well post how I originally imagined Ozy's idea, given that both Balyano and lsfreak have made suggestions:
Essentially, there would be a root structure with a few abstract 'proto-consonants', and perhaps 3-5 registers which determine how the proto-consonants are realised, each of which is less or more consonant-heavy, cluster-heavy, choppy, etc. depending on exactly how we wish it to progress; there would be some rules to determine when there is a shift down, or perhaps occasionally up, a level.
This, as distinct from lsfreaks' suggestion with different phonotactics for affixes and stems, would be compatible with a (potentially completely) free word order, which a polysyntetic language would likely be able to allow anyway. Unlike Balyano's, it is more likely to be compatible with the long words a polysynthetic language would probably result in, since it would not simply go down a level every syllable (not necessarily, anyway).
Any thoughts as to which is the more likely, or any alternative ideas?

Also, with regard to the latest, rank-based vote, how do you propose we evaluate the choice?

[lsfreak]

Essentially, there would be a root structure with a few abstract 'proto-consonants', and perhaps 3-5 registers which determine how the proto-consonants are realised, each of which is less or more consonant-heavy, cluster-heavy, choppy, etc. depending on exactly how we wish it to progress; there would be some rules to determine when there is a shift down, or perhaps occasionally up, a level.
Could you elaborate a bit on how this might actually work out? Like, with a few samples? I'm thinking of a language with free word order (since you mentioned it) where the focus of the sentence ends up fronted and clustery while the less important/meaty parts of the sentence are airy and vowel-rich. If that's the case I'd end up questioning how this came to be; it seems like if you're going to shift the word order based on what's more or less important to the speaker, then it's already filling the role of the different "registers" without needing the complexity of having multiple forms of the same word. And the reverse; if the root itself changes, repositioning the word in the sentence for emphasizing or deemphasizing isn't needed and word order ends up fixed. (If this is what you mean though, inflection of words based on pragmatics rather than, or along with, semantic role is rather intriguing, though I don't know what to do with it).


This, as distinct from lsfreaks' suggestion with different phonotactics for affixes and stems, would be compatible with a (potentially completely) free word order, which a polysyntetic language would likely be able to allow anyway.
I'd be reluctant to make generalizations about word order. Australia and the Rift Valley are both hotspots of free word order despite not being polysynthetic, and (completely) free word orders are all but unheard of in the Caucasus, New Guinea, east India, and the southeastern US, despite being hotspots of polysynthesis. Now, we can probably assume it'll be freer than English, as synthetic languages generally are, but I don't think there's as much reason to assume a completely free (i.e. unidentifiable) word order until we decide on it.