Project Garnet Scratchpad

A forum for all topics related to constructed languages
Post Reply
User avatar
Corphishy
greek
greek
Posts: 740
Joined: 18 May 2013 18:28
Location: Route 102, Route 117, Petalburg City

Project Garnet Scratchpad

Post by Corphishy »

Hello and welcome to what is essentially my attempt to nail down the Vuase language once and for all after 7 years of waffling. What I have created is a proto-proto-protolanguage and a sketch of a few thousand years of linguistic history. That’s what this thread is for—to nail down more concretely the grammar etc. of both Project Garnet and subsequently Vuase.

Project Garnet is the moniker I have given to the language macrofamily within my conworld which currently doesn’t have a real name; I have just been referring to it as Vuaseland as it’s the world where Vuase is spoken. Speaking of, Vuase is a member of this language family. The reason I have picked the name Garnet is somewhat arbitrary. I could have called it any number of mineral groups, but garnet is the one I have decided on for this particular family. I also decided to partake in the CBB tradition of naming one’s conlang project after a mineral.

The great-grandmother of all the languages in this family is the so-called Garnet itself. I have no good name for this language, which has prompted this whole Project Garnet naming scheme I plan on moving forward with. You may occasionally see me refer to Garnet as SHW or šʰawšʰaw. The first is shorthand based on the second, which I rarely use because as you can see it’s a little unwieldy to type. The name has nothing to do with the word for the language or the people who speak it, and in fact the reduplication is probably something I wish to either retool or get rid of, but it more or less means to answer from a reduplication of the word for to say. I probably won’t use it here, but if I were talking about it on discord that’s what I would call it.
To be clear, this is a thread designed to help me work this language out, not to curate an already figured out product, hence the name “scratchpad.” With that said, let us begin with something that is very well figured out by now: the phonology. Garnet has an honestly quite large phoneme inventory, especially with regards to consonants; at least, it’s larger than any consonant inventory I’ve ever made. Let’s just get right into it.

/m n ŋ/ m n ŋ
/p pʰ pˀ b ᵐb t tʰ tˀ d ⁿd k kʰ kˀ g ᵑg ʔ/ p pʰ p’ b ᵐb t tʰ t’ d ⁿd k kʰ k’ g ᵑg ʔ
/f v s̻ s̻ʰ s̻ˀ z̻ s̺ s̺ʰ s̺ˀ z̺ x ɣ h/ f v s sʰ s’ z š šʰ š’ ž x ɣ h
/w r l j/ w r l y

A few questions may arise, chief among which (and the ones I am going to cover in this post) are what does glottalization mean in the context of this language, and why aren’t there nonsibilant aspirated or glottalized fricatives? The answer to the first question is that it is purposefully ambiguous, but that it may have either been ejective or implosive in nature. The answer to the second question is there used to be /fʰ fˀ xʰ xˀ/, but they had merged into their equivalent stops before Garnet had branched off into its various daughters.
Anyway, beyond those questions, there are also some clarifications to be made on the nature of sibilants with regards to apical vs laminal. Being as conservative/lazy as possible, one could interpret it simply as that—an apical/laminal distinction. However, it is more accurately a dental or denti-alveolar vs a post-alveolar distinction. Relatedly, the aspirated and glottalized sibilants were often realized as affricates, especially in the branch which lead to Vuase.

Beyond consonants, Garnet has a very normal set of vowels. They are the standard /i u e o a/ plus a length distinction. Standard fair, but unlike the consonants I have no real justification to include more than that.

The maximum syllable is of the form CVC. The only consonants which cannot appear in coda position are glottals and glottalized consonants (aspirated consonants are allowed, however).

That is it for the phonology of Garnet. It’s pretty simple all things considered. Some ideas I have kicking around in my head are the inclusions of voiced glottalized consonants and perhaps even also voiced aspirated consonants.

It is very late/early in the morning so I am done for today. Sometime tomorrow I plan on posting about the various daughters of Garnet and/or some grammatical information or even in-universe background. Thank you and good night everyone!
Aszev wrote:A good conlang doesn't come from pursuing uniqueness. Uniqueness is usually an effect from creating a good conlang.
Project Garnet
(used to be Bulbichu22)
User avatar
Corphishy
greek
greek
Posts: 740
Joined: 18 May 2013 18:28
Location: Route 102, Route 117, Petalburg City

Re: Project Garnet Scratchpad

Post by Corphishy »

Edit: i give up trying to get that image of the map to load. I suppose it's not necessary but it's still very frustrating
I should really stop saying that I'll do things the next day. I had planned on doing so but pretty soon after I posted the first post, I immediately had an idea on how the phonology of Project Garnet could be improved. I will say that this will officially be the last time I make any major changes to the “Proto-Garnet” phonology. Stepping away from phonology and looking at this project as a whole, I intend it to be “finite,” in that at a certain point there will be an end to major grammatical and phonological retoolings as I am prone to tinker with those things endlessly, especially phonology. I will just share the new phonology with you now and then I intend to speak more about my thought processes and the reasons for things being as they are. Strap in, folks, because this post I intend to be very long and full of a lot of things and a lot of reasonings for those things as well. The first of which is, as I said, the finalized phonology. Here it is:

/m n ŋ/ m n ŋ
/p pʰ pˀ b bˀ t tʰ tˀ d dˀ k kʰ kˀ g gˀ q qʰ ɢ~ʁ ʔ/ p pʰ p' b b' t tʰ t' d d' k kʰ k' g g' q qʰ ɢ ʔ
/f v s̻ s̻ʰ s̻ˀ z̻ z̻ˀ s̺ s̺ʰ s̺ˀ z̺ z̺ˀ x ɣ χ h/ f v s sʰ s' z z' š šʰ š' ž ž' x ɣ χ h
/w r l j/ w r l y

/i iː u uː e eː o oː a aː/ i ī u ū e ē o ō a ā

So basically all of the supplemental information i stated previously is still correct. I'll actually get back to some of the things I said then but let's put a pin in it for now. You will notice some key additions and removals. The latest members of this family are the uvulars and the voiced glottalized consonants.

Now I don't think I've really ever mentioned Bozeto-ei in the CBB, but it was a language I made a few years ago as a speedlang challenge that has now become a member of this language project. I used gleb to generate the phonology and among many things Bozeto-ei contains a single enigmatic q which while not common in its very limited vocabulary of 100 words occurs in enough environments and frequently enough that I could not justify trying to figure out a way to generate uvulars from a language which didn't previously have them. Not to mention that I don't offhandedly have a good way of figuring out how to do that anyway, at least not in a way which would also mess with the development of Vuase which is the most important byproduct of this extremely overengineered device I call a macrofamily. You will notice that the uvulars have no glottalized forms, this is because they merged with the glottal stop in early Proto-Garnet.

The second addition is the voiced glottalized series. I initially added both the voiced glottalized and voiced aspirated series, but I had the intention of merging them anyway. The reason for the merger being that they both produce a breathy vocalization in the consonant. The reason why I decided it should merge in the direction of glottalized rather than the direction of aspirated is because of the fact that Proto-Ugrandite, the name of the branch which eventually created Bozeto-ei, had implosives which obviously source themselves from these voiced glottalized consonants. You will in fact find that Proto-Ugrandite is fairly conservative phonologically, at least compared to Proto-Pyralspite (which also has a real name created before this naming convention--Proto-Bwn).
Now if you remember that pin, I'd like to take it down now and discuss exactly why I have made the decisions that I have. The first one I would like to discuss is the non-plain non-sibilant fricative-non-plain stop merger (rolls right off the tongue right? just call it Keynes’s Law after Jackk who suggested it to me on the unofficial CBB Discord server). Keynes’s Law has a few reasons for existing and also some actual implications for the development of the language. The primary reason is not that I cannot pronounce them, but I'll get to that too. The main thing is that they don't need to exist. I do not intend to add anything excessive or that which wouldn't help to reconstruct Vuase and Bozeto-ei to the best approximation as when they were originally created. I have no use for them, so I just got rid of them. However, when I heard this suggestion it did give me a very interesting idea, which is to have some kind of aspiration or glottalization assimilation. Basically when two obstruents are next to each other, they assimilate in voicing, as well as whether they are plain, aspirated or glottalized (this happens from right to left). this means that plain non-sibilant fricatives in some environments are realized as stops in daughter languages, which I thought would add some fun spice to this language family.

Getting back to my personal inability to pronounce glottalized and aspirated fricatives, let's also discuss why I chose to have the nonplain sibilants be realized as affricates. This actually has some history to it and is not just me being lazy. In early drafts of the language, I just had s z š ž with no explanation as to what the actual values of these consonants were (which is something I like to do--leave fuzzy spaces to block out with and then come back and retool and iron out). However, I had these consonants do some weird things, namely having them become θ ð s z in Proto-Pyralspite (and š ž became t d in coda position) while becoming s z ts dz in Proto-Ugrandite. So sibilants in this language have had a history of being in free variation with affricates.

Uvulars, the recent addition, have some already mentioned quirks, namely that they’re missing a few pieces in the otherwise completely symmetrical paradigms. Well, as I had stated previously Bozeto-ei only has q, so I felt no reason to include the glottalized equivalents which seemed so close to the glottis already they would easily have merged. The reason I included other uvulars besides q is just because I figured I could easily get rid of them later. You'll notice I am a lot more casual and handwavy in this section. This is because as this is a new feature I expect to come back later and fill in some details, probably when I go to reconstruct Bozeto-ei from scratch.

A bit of an aside, but I'd also like to explain why I use the term glottalized and the superscript glottal stops instead of specifically implosive or ejective etc. It is because I intend these consonants to be pulmonic, at least in this stage of the language. Glottalization applies a breathy or creaky modality to the consonants and (in the case of Proto-Pyralspite and some modern Ugrandite languages) the vowels directly thereafter. Only in Proto-Ugrandite did these consonants begin taking on a non-pulmonic quality.

Oh, and lastly I got rid of prenasal stops because they look ugly, are cluttersome and they only existed to solve one quandary, that being the reason why Proto-Pyralspite has wl̩m while Proto-Ugrandite has ʔləb (those aren’t even the current forms of those words but that's neither here nor there). I figured that I can come up with a better explanation later, one which doesn't require the existence of an entire series of consonants for one word.
And now that I have exhaustively and exhaustingly discussed the phonemic invententory of this language, I would now like to briefly discuss some background on the conworld wherein these languages are spoken. I know this is not the conworlds section, so I will be brief; I will also be brief because I do not have that much information to share period, just a rough sketch in my mind.

The time gap between Proto-Garnet and Modern Vuase (the ultimate goal of this project and the stated exemplary modern day example of a Garnetian language) is about 3 to 4 thousand years. It is a language family spoken on a conworld which I have decided to call Ynuh. The origin of this name comes from fifth grade, where I invented a world which existed beyond the horizon. I got the name from flipping around a symbol I drew which at different rotations resembled a Y, an N, a U and an H. There is no canonical pronunciation of this name so feel free to say it however you'd like. Think of it as a tetragrammaton for this world.

here is a direct link because CBB won't display the image for some reason.

Here is a map of the world. The red is the modern distribution of Garnetian languages, and the orange is where Vuase is located. This is by no means the final version of the distribution, but it is the final version of the map--at least, the final version of this sketch of the world; I will go back and add details like mountains and biomes and all those things at a later date. I would also like to credit Davush for helping me to make this map and showing me a free vector program, Affinity Designer, to be able to make it.

As you can see, this language family spans both continents. My intent was that a few thousand years ago there were a series of islands or something of that description inside what could I suppose be called the trans-garnetian ocean. These islands could still exist and are just so small they aren't on the map. Or they could have sunk to rising sea levels after the end of an ice age. The how of things for the moment is unimportant. However, you may think that the people went from the continent on the right to the one on the left, as if analogous to the migration of humans on earth to the Americas. I can see why you might think that, as this world does vaguely resemble Earth in some ways. It is however the opposite which is true. In fact, the Garnetians migrated west to the snake-shaped continent very early in this timeline, making Proto-Ugrandite carry a few features with it which had already been collapsed or changed before Proto-Garnet became Proto-Pyralspite; that is to say, Proto-Ugrandite derives from what I call Early Proto-Garnet whereas Proto-Pyralspite derives from Late Proto-Garnet.

That is about all that can be said about the in-universe context for this language family. There are obviously other language families here but that is for a different thread and a different time. You may also notice that this is actually not that large of a language family. That was not necessarily intentional but I found it to be an interesting quirk and something a bit different, as you would expect a passion project which is soon to be a year old (aka something I've been working on for 1/7th of Vuase’s entire life) would be much larger and much more important in-universe.
Now that I have that out of the way, I wish to discuss some grammar. Specifically for this segment I wish to discuss a bit about the syntax of Proto-Garnet. Most Garnetian languages are VO languages. More specifically, Proto-Garnet is a primarily SVO language, though it also alternates to VSO. I am not entirely sure yet what triggers this word order shift, if anything does at all. I would love to hear some suggestions if anyone has any. Some ideas I had were for asking questions and perhaps also for dependant clauses. Either way, it has some features standard of many SVO languages, but also some which are not.

For example, most adjectives go after the noun they modify, but like some asian languages like Chinese and Vietnamese (if this information is wrong I apologize--all of the information about SVO languages that I have comes from the wikipedia article on the subject. I do not speak any of these languages), they place numerals before the noun they modify as if the numeral were the head of a “numeral phrase;” the same occurs with other adjective types as well, namely those which fall under the currently vague category of substantive, which describes adjectives which typically find themselves in noun-like positions. This also includes typically predicative adjectives as well. Examples of the kinds of adjectives which typically find themselves in this category include color words, and physical appearance words like height and weight. If you couldn’t tell this is a very fuzzy category because I am thinking about this right as I am writing this paragraph. Perhaps this distinction is expanded upon and altered in daughter languages, or the distinction is completely flattened in one direction or the other.

Something which is different but not necessarily crazy about Proto-Garnet is that it uses postpositions rather than prepositions. It shares this with West African languages, and in fact I took some other syntactical quirks from West African languages because of this. The reason why I decided to give this language postpositions is because all versions of Vuase and also Bozeto-ei have postpositions, which in Vuase’s case has always been a quirk of its word order, it being a primarily head initial language. So, I decided that instead of “correcting” this fact, I would help justify it by bringing in other aspects of languages which share this feature in the real world; such as the inclusion of prenominal genitives, which actually occur in other SVO languages and language areas despite it being the nonstandard ordering. I will also take this paragraph to say that the order of modifiers to nouns goes [substantive] [genitive] HEAD [other adjectives] [postposition].

Getting away from the modifiers of nouns, I also have information about the modifiers of other parts of speech. Firstly, adverbs, which are not treated as different from adjectives, are postverbal just as adjectives are postnominal. But more importantly, so are auxiliaries. Now, this is a weird feature for an SVO language to have, but I cannot really find a way to justify it not being the case because it’s how it works in basically every language which I have deemed to be Garnetian, chief among which being Vuase. I have no in-universe explanation as to how this could have happened, though I suppose at a certain point you need to stop and realize that not everything needs a very in-depth explanation or reason for being the way that it is.

And finally, on the topic of relative clauses, and all other dependent clauses, they are placed after the clause they modify. Specifically for relative clauses, they can also be placed directly after the noun that they modify. The way that relative clauses are formed is pretty simple, Proto-Garnet uses a pronoun retention system. There is not a separate resumptive pronoun or pronoun series, though in some daughters perhaps I may develop some.
That about does it for the syntactical information I have about Proto-Garnet. I suppose it is a lot, but for a long time it was just a half-finished outline in a google doc. That’s kind of why this post is so long, because I have a lot of information but most of it is either in my head or scattered haphazardly in a document or note on my phone. So I’m taking this opportunity to just get it all out there. And by it all I literally mean every ounce of information that I have about this language, Proto-Garnet. Now if you put it in the perspective of literally all I have about this language, it doesn't seem like that much. I am, however, not done yet. I also have some information I would like to share about the inflectional morphology of this language. Firstly, I would like to discuss the case system I have in mind for this language. I do not have the morphology and suffixes etc. specified yet, but I have the information that the morphology will carry. There are 5 cases in this language for now: Nominative, Accusative, Genitive, Instrumental, and Locative. This next section will be a little dry and formulaic so bare with me for a bit as I explain the function of each case.

First is the nominative case. The nominative has the least unique uses of any case. It is just the subject of a clause. It is also the dictionary form, and is the unmarked form of a noun.

The accusative case is also fairly simple, but slightly more interesting. It has the function of marking the object of a clause, and much like several indo-european languages, it functions as an “accusative of duration of time” indicating how long an action was performed. It is also used as a vocative.

The genitive case is where things begin to get interesting as from this point on every case has myriad function. Obviously, as a genitive it is a marker of possession. Specifically it marks alienable possession: there is another system for inalienable possession which I will discuss later. It is also used for amounts, and parts of a whole. So, it is used in phrases such as “2 out of 3 cheeses,” “wheel of cheese,” and “some cheese.” It is used in conjunction with amount words like some, many, a few, etc. Note that you use it mandatorily with any amount words, unlike in English where you can omit it in some contexts. When used by itself, as in not modifying another noun, it has an ablative sense as in “from the cheese.” The ablative sense also applies to temporal motion toward the future (“I will eat the cheese in an hour from now”) Lastly, the ablative describes facing away from an object and also describes the impetus of an action (“I am sick from eating too much cheese.”)

The instrumental case also has a lot of uses. Firstly, there is no instrumental/comitative distinction in this language, so it makes no distinction between using the object to complete an action and completing an action with the company of the object. It is also used as a dative and a benefactive. You use this case, in the comitative sense, whether you and the object are doing the same action as each other. That may sound confusing as I'm not sure how I would describe this in a technical sense, but let's take 2 sentences “I ate dinner with you” and “I sang a song to you.” In the first sentence, we are both eating, but in the other I am singing while you are listening. However, I am saying that those would both use the instrumental case. I suppose that it has to do with the fact that while you are not the direct object, you are still either in direct receipt of the action or are directly participating in it. But this is getting into a level of depth that is starting to bend my brain into a pretzel so for now I will move on. But before I do, this case actually has one more sense. It is used as a semblative, and used for comparing two nouns, especially to point out their similarity (“this cheese is like a smiley face in your mouth” or “I love this cheese like my son”), and also to form “as X as Y” type constructions.

And finally, the locative case. This case is kind of a catch-all case for any spatial or temporal relationship. It describes motion towards, into, through, over, under, and around as well as proximity to, and position in, on, or near something. it also describes points in time such as well as motion forward in time (“I ate the cheese an hour ago” or as it would be translated more literally “I ate the cheese an hour to now” which is cromulent english but a little wonky, especially in that circumstance). The locative also conversely to the ablative senses of the genitive case is used to mark the ends of actions and also the deadlines for actions (“you must eat all this cheese in the next hour.”)

I'm sure that in that giant wall of text that I have missed something. I don't believe it matters at the moment because what is most important is that I have enough information to be able to feel my way through the things I don't have as sure a grasp on much better than if I just told you the names of the cases and just kept all the minutiae in my head. However, now that that is out of the way, I have a few more miscellaneous grammatical musings because we are now reaching the end of my knowledge of this language and I can scramble to try to fill in some of the blanks I've left here and move on to blanks I haven't even discussed, such as the fact that the verbal morphology currently doesn't exist because I cannot decide on anything. but that's neither here nor there, let us get back to the topic on hand.

Proto-Garnet has no singular or dominant strategy for marking plurality, which is partly by design and partly because I cannot decide on one. Some words use reduplication, some words have some kind of word affixed onto them, and still others do other things. The daughter languages reflect this, borrowing words with multiple fossilized number systems.

Lastly, Proto-Garnet has a fairly productive and versatile system for head marking the person of a heads’ primary dependant. Its uses include person marking on verbs and postpositions, and inalienable possession on nouns. It may also mark dummy nouns or dependants on these substantive adjectives (which definitely need a better name because that's not really what I mean), but at the moment I am unsure of what I want to do with them. I am also unsure if I want adjectives to agree with the number and/or case of a noun.

Now that you have read almost 4000 words of me spilling out my brain gumbo, I would like to know your thoughts. I am sure there must be a few questions to be had. Also, if you have a suggestion as to what I should call the class of adjectives which form adjectival phrases where they are the head as opposed to the noun being the head, please let me know. In fact, if you have any suggestions period on what I could add to my language I would love to hear it. I have been writing this post for about 3 days straight now and am dying to hear what people have to say about it. As always, stay tuned for the next time I post something; it probably won't be nearly as long as this was, but I doubt anything will be this long for a very long time.
User avatar
qwed117
mongolian
mongolian
Posts: 4094
Joined: 20 Nov 2014 02:27

Re: Project Garnet Scratchpad

Post by qwed117 »

is this the map you wanted? you posted it on the discord
Image
Spoiler:
My minicity is [http://zyphrazia.myminicity.com/xml]Zyphrazia and [http://novland.myminicity.com/xml]Novland.

Minicity has fallen :(
The SqwedgePad
User avatar
Corphishy
greek
greek
Posts: 740
Joined: 18 May 2013 18:28
Location: Route 102, Route 117, Petalburg City

Re: Project Garnet Scratchpad

Post by Corphishy »

qwed117 wrote: 22 Aug 2019 02:30is this the map you wanted? you posted it on the discord
Yes thank you. I was trying to get it to appear in the post itself but it wasn't working and neither was the link I provided so I had planned to give up on it and fix it later. But this works too. For all of you who have read the thread but didnt get to see the map, this is she.
Aszev wrote:A good conlang doesn't come from pursuing uniqueness. Uniqueness is usually an effect from creating a good conlang.
Project Garnet
(used to be Bulbichu22)
User avatar
Jackk
roman
roman
Posts: 1487
Joined: 04 Aug 2012 13:08
Location: Damborn, Istr Boral

Re: Project Garnet Scratchpad

Post by Jackk »

Corphishy wrote: 22 Aug 2019 00:43 /p pʰ pˀ b bˀ t tʰ tˀ d dˀ k kʰ kˀ g gˀ q qʰ ɢ~ʁ ʔ/ p pʰ p' b b' t tʰ t' d d' k kʰ k' g g' q qʰ ɢ ʔ
Five types of stops for each point of articulation - nice :D I've never been brave enough to do even three. Also thank goodness ɢ alternates with ʁ because the latter is like 5x easier for me than the former :P Or is there some environmental conditioning?
Corphishy wrote: 22 Aug 2019 00:43 Now I don't think I've really ever mentioned Bozeto-ei in the CBB, but it was a language I made a few years ago as a speedlang challenge that has now become a member of this language project.
This is an interesting point - did Bozeto-ei have any similarities to other Garnetian languages before you decided to incorporate it. It sounds like an interesting challenge to connect unconnected languages with an ancestor - how often do you find you have to change a descendent language to make the ancestor consistent?

(Also the gemstone/mineral naming convention is neato - you shouldn't run out of names for a good while)
Corphishy wrote: 22 Aug 2019 00:43 I just had s z š ž with no explanation as to what the actual values of these consonants were (which is something I like to do--leave fuzzy spaces to block out with and then come back and retool and iron out).
Jonathan Keynes approves of this practice. [B)] Your justification for the sound variation here and with the uvulars makes lots of sense. :D
Corphishy wrote: 22 Aug 2019 00:43 There is no canonical pronunciation of this name so feel free to say it however you'd like.
*pronounces it /"mo:tSaj/* [:P]
Corphishy wrote: 22 Aug 2019 00:43 As you can see, this language family spans both continents. My intent was that a few thousand years ago there were a series of islands or something of that description inside what could I suppose be called the trans-garnetian ocean. These islands could still exist and are just so small they aren't on the map. Or they could have sunk to rising sea levels after the end of an ice age.
Ooh nice, island peoples! I have a certain fondness for this sort of history. Do the people speaking Garnetian languages retain any particular naval prowess? In particular, did they ever lose the ability to cross the trans-Garnetian ocean, and for how long?
Corphishy wrote: 22 Aug 2019 00:43 Proto-Garnet is a primarily SVO language, though it also alternates to VSO. I am not entirely sure yet what triggers this word order shift, if anything does at all. I would love to hear some suggestions if anyone has any. Some ideas I had were for asking questions and perhaps also for dependant clauses.
I second the dependent clauses idea, and also suggest relative clauses. Questions also works, and consider other moods - maybe hortatives/optatives or other subjunctive-ish sentences like "May the Queen reign for many more years!".
Corphishy wrote: 22 Aug 2019 00:43 I will also take this paragraph to say that the order of modifiers to nouns goes [substantive] [genitive] HEAD [other adjectives] [postposition].
Postpositions, excellent [}:D]. This whole paragraph seems well-reasoned - i especially like that the particular adjectives that precede the noun look like a rather different set than what I'm used to from Romance languages. :D
Corphishy wrote: 22 Aug 2019 00:43 When used by itself, as in not modifying another noun, it has an ablative sense as in “from the cheese.”
Surely I've seen genitive=ablative before, hmm [:D] I was initially a bit surprised but after some thought I really like this detail and it makes semantic sense to me. Case shenanigans like this are the main reason i really do want to do a case-heavy language at some point.
Corphishy wrote: 22 Aug 2019 00:43 Proto-Garnet has no singular or dominant strategy for marking plurality, which is partly by design and partly because I cannot decide on one. Some words use reduplication, some words have some kind of word affixed onto them, and still others do other things. The daughter languages reflect this, borrowing words with multiple fossilized number systems.
Again excellent :D yay for inconsistency. I might suggest that in PG itself, number marking is optional for most nouns (except those which are reduplicated, perhaps) and so when people started to want to mark all nouns for number different strategies arose in different places and for different semantic classes of nouns.

This is indeed a lot of information but it's very well organised. [:D] It's certainly more than I've written down for Duban or Boral grammar :) Glad to have read it!
terram impūram incolāmus
hamteu un mont sug
let us live in a dirty world
User avatar
Corphishy
greek
greek
Posts: 740
Joined: 18 May 2013 18:28
Location: Route 102, Route 117, Petalburg City

Re: Project Garnet Scratchpad

Post by Corphishy »

Jackk wrote: 22 Aug 2019 08:42thank goodness ɢ alternates with ʁ because the latter is like 5x easier for me than the former :P Or is there some environmental conditioning?
I forgot to mention this little tidbit in the post (as I said, given with as much information as I put, I was bound to have overlooked at least something). Basically from what I have read ɢ is pretty universally hard to pronounce, and I would like to be able to pronounce this language, at least to an extent. That, coupled with the fact that it doesn't technically need to exist but I decided to include it anyway because it would make the whole chart just a little bit more symmetrical and not like it was lopsided with only one random uvular, I figured the least I could do for myself is to be able to pronounce every phoneme. So, no, to answer your question there is no environmental conditions, they are just in free variation with each other. Perhaps at one point they were distinct but merged due to not having any or enough important minimal pairs to warrant it.
Jackk wrote: 22 Aug 2019 08:42did Bozeto-ei have any similarities to other Garnetian languages before you decided to incorporate it. It sounds like an interesting challenge to connect unconnected languages with an ancestor - how often do you find you have to change a descendent language to make the ancestor consistent?
Its similarities, as well as Uzi's similarities to the more traditional Granetian languages I had been making (Vuase and its relatives on the dragon-head continent) were pretty minimal to be honest. There were, however, a few words which looked vaguely similar. The primary reason for including them in this language project is that vocabulary generation is probably the hardest part of language invention for me, so if I can have as many languages share the same pool of vocabulary, that's much easier on me. As for changing descendants to fit the ancestor; besides vocabulary changes, I don't tend to change much at all of the descendant. For me it's mostly about the challenge of reconciling these two disparate groups of languages with a common language, and changing the descendants to fit the ancestor would be too easy to do.
Jackk wrote: 22 Aug 2019 08:42Do the people speaking Garnetian languages retain any particular naval prowess? In particular, did they ever lose the ability to cross the trans-Garnetian ocean, and for how long?
I never really thought about it, but since you mention it, perhaps at some point they were a much larger naval culture than they are now. I don't think I want there to have been any cross-polination of Ugrandite and Pyralspite languages after the colonization of the snake-head continent prior to the early industrialization and globalization which takes place in the modern history of the world. So, there could have definitely been a loss of technology, for what reasons I am not entirely sure of because again I have never thought about that before. Thank you for making me think about it though!
Jackk wrote: 22 Aug 2019 08:42I second the dependent clauses idea, and also suggest relative clauses. Questions also works, and consider other moods - maybe hortatives/optatives or other subjunctive-ish sentences like "May the Queen reign for many more years!".
Since you mention hortative, perhaps this alternation is used for both asking and answering questions, as hortative has to do with the encouragement or discouragement of actions. And perhaps expanding upon that, alternation has other imperative functions such as as you said optative and maybe jussive--expressions of commands and suggestions, as well as questions and answers. Again, thank you for the suggestions!
Jackk wrote: 22 Aug 2019 08:42Surely I've seen genitive=ablative before, hmm I was initially a bit surprised but after some thought I really like this detail and it makes semantic sense to me.
I was partially influenced by romance languages' usage of the preposition "de," which has both possessive but also partitive and in some circumstances ablative senses (such as describing where someone is *from*). I'm also trying to break away from my usual problem of just including a case and not really using it in any interesting or well-thought-out ways other than exactly as SAE uses them, which I have then internalized as the "normal" way in which cases are used. Specifically for this case I typically only use the genitive to mark possession, which may be because of English influence as well.
Jackk wrote: 22 Aug 2019 08:42 I might suggest that in PG itself, number marking is optional for most nouns (except those which are reduplicated, perhaps) and so when people started to want to mark all nouns for number different strategies arose in different places and for different semantic classes of nouns.
I had this idea as well, at least the part about different noun classes being developed. A creation of several different plurality systems which become codified as marking some certain "kind" of noun i think would be very interesting, and in certain branches could lead to the development of a proper noun class/gender system. In the case of Vuase, by the time of its modernity it has mostly flattened out this wacky system with a fairly regular one, any irregularities nonwithstanding.
Jackk wrote: 22 Aug 2019 08:42 This is indeed a lot of information but it's very well organised. It's certainly more than I've written down for Duban or Boral grammar :) Glad to have read it!
Thank you and thank you so much for your feedback. You've given me quite a bit of brain candy to chew on here for my next post.
Aszev wrote:A good conlang doesn't come from pursuing uniqueness. Uniqueness is usually an effect from creating a good conlang.
Project Garnet
(used to be Bulbichu22)
User avatar
Corphishy
greek
greek
Posts: 740
Joined: 18 May 2013 18:28
Location: Route 102, Route 117, Petalburg City

Re: Project Garnet Scratchpad

Post by Corphishy »

So instead of stockpiling information to then spew into this thread in another megapost, I figure I should instead post the information that I have when I have it. Since the last post, I have mostly been thinking about verbal morphology, mostly because it is the least defined area of this language’s grammar. Namely, I have been trying to work out how the modality of this language works. What I have worked out so far is that Proto-Garnet has four moods which are marked on the verb (though there will most likely be more which are created periphrastically). These four moods are: indicative, subjunctive, jussive and potential. This is another section where I will just list and explain the uses of these things.

The indicative mood is the most common mood, the one which is used for information which is known, either from first-hand or general, gnomic knowledge. This aspect is used regardless of whether the statement is true or false, it is only used to refer to the certainty of the information. It can also be used for predictions and statements of the future, which specifically are certain to happen.

The subjunctive is used for a few things, all of which have to do with the uncertainty or unrealness of a statement. It describes statements which are reported or assumed/presupposed to be correct or incorrect, or actions which one wishes to have happen (this is different from a polite command or suggestion; compare “I want her to leave” which in one instance means “I wish she would leave” and is just a statement of one's desire, but in another instance means that you are attempting to tell someone in an indirect, admittedly passive aggressive way, that you would like them to be removed from the premises. In the first instance, you would use the subjunctive, but in the second you would use the jussive. I bring this specific example up in particular because as I explained this system to the people of the discord server, I got this question). It is also used as a quotative, related to its reportative meaning of “he said that X” (where X would get the subjunctive). In certain daughters, Vuase among them, this sense has spread for use in all relative clauses, and it is possible that this analogy was present even in late Proto-Garnet.

Anyway, speaking of the jussive, it is the mood concerned with orders and requests to do things. Obviously, it has imperative uses, but it is used for all persons (hence why I have chosen to call it jussive). Besides giving orders, either in the form of commands or suggestions, it also gets used for answering polar questions (an extension of the hortative sense of approval/encouragement, it also came to mean affirmation or agreement, as if to say “i must agree” or “i must not agree”). It alos has a permissive sense, occupying the half of the semantic space of the English word “can” which means “I am allowed to do.” Besides that, it also has an interesting tensepectual meaning, that obligative sense of “i must do” coming to be used as a future (this is by no means the primary future tense marking, but it is a way to say things about the future which are to occur not because of general knowledge but because of obligation or necessity).

Lastly, the potential is the mood which has to do with statements that are predicated on other statements being true or false. As such, it is used as a conditional (and as a counterfactual), and is used in “if then” statements, where it occupies the “if” slot (the “then” slot is usually indicative but it doesn't have to be). You can also use it to say things like “I would go, but I have homework” or “I would go, if I didn’t have homework” (in this instance the first half is indicative and the second half is potential). However, this mood doesn't need to qualify another statement, it can also be used on its own. When it is, it describes actions which may happen--i.e. ones which have “potential” to happen. It also occupies the other half of the semantic space of the word “can.” I don’t know what the real word is, but I have been calling it abilitative, or “I am able to do.”

Any semantic space that I have no doubt forgotten about can be fleshed out later. This kind of thing is one which needs ample example sentences to translate until I get a really solid grasp on it. However, as you can see, these moods only exist as grammar and currently much like the case system has no morphology associated with it. That's perhaps something I need to work on even more, because without morphology this is really just a theoretical language with all these aspects which occur in theory but which cannot be exemplified at all.
Aszev wrote:A good conlang doesn't come from pursuing uniqueness. Uniqueness is usually an effect from creating a good conlang.
Project Garnet
(used to be Bulbichu22)
User avatar
Jackk
roman
roman
Posts: 1487
Joined: 04 Aug 2012 13:08
Location: Damborn, Istr Boral

Re: Project Garnet Scratchpad

Post by Jackk »

Corphishy wrote: 25 Aug 2019 22:52 In certain daughters, Vuase among them, this sense has spread for use in all relative clauses, and it is possible that this analogy was present even in late Proto-Garnet.
Do you suppose it spread to all complement clauses first (as in "I know that ...", "I am happy that ...", "I fear that ...", "I told her that ...") where the clause is a complete clause on its own, and only then to relative clauses (as in "the woman that I saw", "two cats who were playing together") where the clause ("I saw", "were playing") isn't complete on its own?
Corphishy wrote: 25 Aug 2019 22:52 [The jussive] also gets used for answering polar questions (an extension of the hortative sense of approval/encouragement, it also came to mean affirmation or agreement, as if to say “i must agree” or “i must not agree”).
I like this! :D I wouldn't have thought of it myself but it is very aesthetic.
terram impūram incolāmus
hamteu un mont sug
let us live in a dirty world
User avatar
Corphishy
greek
greek
Posts: 740
Joined: 18 May 2013 18:28
Location: Route 102, Route 117, Petalburg City

Re: Project Garnet Scratchpad

Post by Corphishy »

Jackk wrote: 26 Aug 2019 18:35Do you suppose it spread to all complement clauses first (as in "I know that ...", "I am happy that ...", "I fear that ...", "I told her that ...") where the clause is a complete clause on its own, and only then to relative clauses (as in "the woman that I saw", "two cats who were playing together") where the clause ("I saw", "were playing") isn't complete on its own?
Yes that makes the most sense, but also there is not much of a difference in Proto-Garnet, because Proto-Garnet uses a resumptive pronoun in its relative clauses, meaning that "the woman that I saw" would become "the woman that I saw her," in which case "I saw her" is a complete sentence on its own.
Aszev wrote:A good conlang doesn't come from pursuing uniqueness. Uniqueness is usually an effect from creating a good conlang.
Project Garnet
(used to be Bulbichu22)
User avatar
Corphishy
greek
greek
Posts: 740
Joined: 18 May 2013 18:28
Location: Route 102, Route 117, Petalburg City

Re: Project Garnet Scratchpad

Post by Corphishy »

So, it's been a while. I assure you though, this hiatus was spent wisely: I've not stopped working on Proto-Garnet. In fact, I am writing this because it is time to start wrapping up production on Proto-Garnet. I will still be adding little snippets of grammar and also more vocabulary, but at that point is when I consider a language to be "finished" (something that has been a very controversial subject matter between me and fellow conlangers: whether a language can ever be considered "finished")--but that's besides the point.

Let us first take a moment to flesh out some of the concepts which I had left fuzzy in the last megapost. One such topic, which will tie into today's topics, is the trigger for word order shifting in Proto-Garnet. If you will remember, PG is SVO, but sometimes switches to VSO. It does this for two major reasons: to mark dependent clauses, and to ask questions and answer questions negatively (these two things are tightly linked in Garnetian languages, many Ugrandite languages developing an interrogative mood which comes from a negative particle, and is related to the prevalence of negatively framed questions as a feature of Proto-Garnet).

In some daughters, this use was also expanded for marking irrealis moods, and in Spessartine, the branch which eventually became Vuase, it expanded even further to be the primary word order (perhaps due to a politeness register which overrode the old, now rude, way of speaking).
The other, more complicated in my opinion, fuzzy category is the substantive. In all of the months since I came up with this idea, I haven't found a better name for them, and the more I use it the more it justifies itself. The reason why I chose the name is because the term “substantive adjective” means “An adjective used alone in the absence of the noun that it modifies” (from wiktionary). This is a pretty okay description of what a substantive in Garnetian does. A more accurate description is that only some adjectives are allowed to be used without a noun. This class of adjectives which can also have special properties, and are treated much more nounishly than regular adjectives are.

The kinds of adjectives which are substantives follow a comparable, but not similar, set of adjectives as with romlangs and BAGS adjectives (note: this system was NOT inspired by romance at all, but rather by the way in which PG puts things like numerals and genitives before the noun they modify). I have created this classificatory system as opposed to a finite list because the part of speech known as substantives in Garnetian grammar is a very malleable one, and what does and does not belong is different from daughter to daughter (and in many Pyralspite languages, namely Vuase, the distinction was entirely leveled). Often times when a substantive word or phrase is replaced with an adjective in a language's development, it can be "promoted" to a substantive.

In Proto-Garnet, the kinds of words which were substantives include:

Measure words: this includes numbers, lengths and weights, and age

Weather: this includes things such as temperature, humidity, and condition (windy, rainy, etc). Despite the name, in many cases this category expands to descriptors of objects with regards to temperature as well. Note also that words such as "windy" and "rainy" are able to be used to mean "wind" and "rain"

Senses: this includes emotions, physical perception, textures, and colors

And lastly, participles are treated as substantives as well, and so do oblique forms of nouns (particularly the genitive. I don't know if there is a word for “a phrase made up of two nouns, one of which is a special case which describes it a relationship between the other noun”)

The two primary features of substantives is that they appear before the nouns they modify, and they are able to be used predicatively and without a head noun. Now, adjectives are able to do this as well, but they need extra morphology to be able to. Specifically, they need the help of the head-marking suffixes. I briefly touched on them, but their functions are myriad, and the same set of suffixes are used for multiple parts of speech, each with a slightly different purpose. For adjectives, they allow them to appear without an explicit head.

I mentioned predicate adjectives because in Proto-Garnet, you cannot say something like “the cheese is good” because despite the thing that good is describing obviously being “cheese,” for whatever reason Proto-Garnet does not like it when an adjective doesn’t have a head, and they’re unable to be detached like this as they are in English. I’m sure this isn’t too crazy of a grammatical feature, but I honestly don’t know. I know that languages often treat predicative and attributive adjectives differently. I suppose one could say that Proto-Garnetian disprefers to use predicative adjectives at all.

But, that's enough waffling on about my funny adjectives. It’s time to get into the meat of this post: verbs.

The first of two major topics will be on aspect in Proto-Garnet. This has been a major thorn in my side, as describing the function of various aspects is pretty complicated and hard to do. I’ll try to make sense of it as much as I can.

Firstly, the unmarked form of the verb is called the continuous aspect. Its primary function is to describe ongoing actions with internal structure. It is the most basic of the aspects.

This is in contrast with the perfective, which is used to describe actions as a whole, or as a single point or unit. The perfective is also used to describe “background information,” such as what you were doing before or while something happened, or info about the setting etc.

The third aspect is the habitual aspect, which describes, well, habitual events and gnomic statements. In some cases, it can be used in conjunction with time words like “tomorrow” or “soon” to form a future tense. Verbs which are stative conflate the habitual and continuous, and are unmarked for both aspects. Also, the past habitual in most circumstances is really only used to form “used to” type phrases. When you want to do a “gnomic past,” it is more common to use the past perfective.

I am sure these descriptions are very bare bones, and I don't know how many example sentences I’d have to make before I “got” when to use what.
Edit: I totally didn't realize that I had already posted about mood in Proto-Garnet. It's been far too long I suppose


Thankfully this post is much shorter than 4000 words (only about 1100). Honestly, this kind of wraps up all of the knowledge on the basics of Proto-Garnet grammar that I have. If you have any questions about how certain things work, I can definitely provide example sentences. I also intend at some point to add a little post which shows the actual morphology of these things to give your brain something more digestible than an entire description of a grammatical feature’s usage.

Anyway, stay tuned for the next segment of Project Garnet: Project Pyralspite.
Aszev wrote:A good conlang doesn't come from pursuing uniqueness. Uniqueness is usually an effect from creating a good conlang.
Project Garnet
(used to be Bulbichu22)
User avatar
Jackk
roman
roman
Posts: 1487
Joined: 04 Aug 2012 13:08
Location: Damborn, Istr Boral

Re: Project Garnet Scratchpad

Post by Jackk »

Great to see a new update! [:D] You have the opposite strategy to me (long effortposts every so often instead of bitesize things every day); it's probably less annoying to the audience [xD]

Ok, let's have a look...
Corphishy wrote: 21 Jan 2020 23:49 PG is SVO, but sometimes switches to VSO. It does this for two major reasons: to mark dependent clauses, and to ask questions and answer questions negatively.
I've tried before to have languages mixing SVO and VSO but never quite got it to work in a way I like (Chudihr is like this, for example), so I like to see it in other langs. Associating negatives and questions like this is a cool idea!
Corphishy wrote: 21 Jan 2020 23:49 it expanded even further to be the primary word order (perhaps due to a politeness register which overrode the old, now rude, way of speaking).
Do you have any specific ideas about what this originally-polite structure looked like?
Corphishy wrote: 21 Jan 2020 23:49 Measure words
Weather
Senses
[P]articiples ... [and] oblique forms of nouns
Nice collection of concepts! These feel like a very natural class to me.
Corphishy wrote: 21 Jan 2020 23:49 I don't know if there is a word for “a phrase made up of two nouns, one of which is a special case which describes it a relationship between the other noun"
Something like "genitive phrase" maybe, lol. Can PG do things like "The woman on the mat has black hair" (or some sentence using a similar locative phrase)?
Corphishy wrote: 21 Jan 2020 23:49 in Proto-Garnet, you cannot say something like “the cheese is good”
So... do you say like "There is the good cheese" or like "The cheese is a good thing"; how would a PG speaker most naturally express the concept?

Corphishy wrote: 21 Jan 2020 23:49 [aspects]
Interesting division of the space into three aspects! Do you mind if I ask for some examples: which aspects would be used in the following sentences?
  • I will work every day.
  • What happened last night?
  • Have you ever been to Prague?
  • The house is being built.
terram impūram incolāmus
hamteu un mont sug
let us live in a dirty world
User avatar
Corphishy
greek
greek
Posts: 740
Joined: 18 May 2013 18:28
Location: Route 102, Route 117, Petalburg City

Re: Project Garnet Scratchpad

Post by Corphishy »

Jackk wrote: 22 Jan 2020 16:52 Do you have any specific ideas about what this originally-polite structure looked like?
The idea is that it evolved from som irrealis form, but using the indicative to indicate some other nuance (such as to mute the intended meaning, or to be able to expand to more kinds of phrases). It perhaps evolved from the jussive, which is used for requests, but used the indicative to soften it (not directly stating the request, only implying it with word order). Then, it expanded to phrases which weren't requests at all. This is just an off the cuff idea after having slept on it.
Jackk wrote: 22 Jan 2020 16:52 Something like "genitive phrase" maybe, lol. Can PG do things like "The woman on the mat has black hair" (or some sentence using a similar locative phrase)?
Yes, you could say that, hence why I didn't want to just say "genitive" when I meant any oblique. By the way, that particular sentence would be translated as

rēt'iz mište mijoŋ ib'izo onzē
cloth-LOC woman have-HAB black hair-ACC
"The woman on the mat has black hair"
Jackk wrote: 22 Jan 2020 16:52 So... do you say like "There is the good cheese" or like "The cheese is a good thing"; how would a PG speaker most naturally express the concept?
You would say basically "the cheese is good-it" or to say in Proto-Garnet:

qēʔelzatuj oŋ ožeɣ
cheese is good-3S
"The cheese is good"

I suppose you could also say the first sentence, and in fact I may use such a construction as the "more proper" way of saying it in Garnetian languages. Thank you for the suggestion.
Jackk wrote: 22 Jan 2020 16:52 I will work every day.
What happened last night?
Have you ever been to Prague?
The house is being built.
1. eŋ ab'arma wije en mejzi.
1S work after all day-LOC
"I will work every day."

I decided that the typical way to form the "normal" future tense is wije, which is also a postposition and a particle meaning "after, in front of, beyond" etc.

2. Impalar ib'izoz qu pʰonši?
happen-PST night-LOC before what
"What happened last night?"

Notice the word order shift to VSO, because this is a question. Also notice that the verb is both A) in the indicative and B) unmarked for aspect. This is because "to happen" is a stative verb, and as such it doesn't make as many distinctions as active verbs.

3. Tob'īmīr la enzi Piragzi tin?
go-PRV-PST NEG any Prague-LOC 2S
"Have you ever been to Prague?

Notice that since "to go" is an active verb in PG, the perfective was used. This is the "default" past tense form for questions. Also notice the negative particle: since this is a polar question, it is customarily negatively framed. Just as a fun aside, I just realized this is actually a redundant marker for interogatives. I have no problem with that, redundancy is always nice.

4. ošī pʰeran oŋ.
house build-PASS.PTCP be
"The house is being built."

This is a fun one because it's a passive construction, which I had forgotten to talk about. However, very simply they use the passive participle and copula. I might should elaborate more intricate tenspects/auxiliaries+nonfinite verb constructions, but for now I'll table that as A Thing I Can Use to Develop Daughterlangs instead.

Those translations had me racking my brain a bit, hence why this took a few days. I'm not very practiced with translating things into my conlangs, so thank you for that!
Aszev wrote:A good conlang doesn't come from pursuing uniqueness. Uniqueness is usually an effect from creating a good conlang.
Project Garnet
(used to be Bulbichu22)
User avatar
Corphishy
greek
greek
Posts: 740
Joined: 18 May 2013 18:28
Location: Route 102, Route 117, Petalburg City

Re: Project Garnet Scratchpad

Post by Corphishy »

So, long time no see. I've been piecemealing around, fiddling with outer language projects--I do that sometimes so I don't get burned out on this one, which despite how fun it is, can often become too much for me. I tend to lose myself, and then when I return, what usually follows are some sweeping reforms to the system. This time is no different.

I've been sitting on this for a little bit. I've been completely recombobulating this language's dictionary, and in the process also its phonology and grammar. Yeah, remember when I said this language was basically done? Well, now it's not again! However, I have greatly simplified many things which did not need to be as complicated as they were (note "complicated" and not "complex").

Let's begin this post with an updated phonology. It has mostly been a case of downsizing rather than much moving or shaking. This is she:

/m n ɲ/ m n ñ
/pʰ p ɓ tʰ t ɗ kʲʰ kʲ ɠʲ kʰ k ɠ qʰ q ʔ/ p b b' t d d' c j j' k g g' q ğ '
/f θ s z ʃ ʒ x/ f þ s z š ž x
/w r l j/ w r l y

I feel the need to sprinkle in some justifications or reasons for the various changes I am introducing into this project. So, starting from the top, I'll explain why I got rid of the velar nasal/replaced it with a palatal nasal. Primarily the reason is because in sketches of Proto-Bozeto-ei, its ŋ originated from an earlier ɲ, and in Vuase's history, *ɲ eventually merged with *j. On top of this, Both Bozeto-ei and Uzi have or had a palatal series, which is why I replaced the earlier laminal/apical distinction with an alveolar/palatal one. In Pyralspite languages, this doesn't end up mattering much, but in Ugrandite languages it does figure. This is alo why I made them palatalized velars in PG instead of full palatals. In Pyralspite, they simply merge with plain velars.

Another change is that I removed the voicing distinction of nonsibilant fricatives. This is because of Bozeto-ei and Uyi, which have s z š ž, but no v or ɣ etc. I also got rid of the uvular fricative, because it isn't represented in any daughters (BOZ, the only documented daughter retaining uvulars, only has q, due to the loss of aspirated/unaspirated distinction). I also added *θ, primarily to decomplicate the process by which Pyralspite languages have or had θ but Ugrandite languages did not. It, along with the getting-rid of voicing on stops and the orthographical change thereof, was also to better help me better reconcile the differences between the proto-form and the daughter languages. Before, I would have a bunch of plain p t k becoming b d g in Vuase and quite frankly it sounded and looked ugly to me. However, having that same change occuring, but not changing anything orthographically (*b > b is the same as */p/ > /b/) made it much more palatable to me.

As for the vowels, the main difference is the removal of the long vowels, as of the 3 documented Garnetian languages (Vuase, Bozeto-ei and Uzi), only one of them (Uzi) has phonemic vowel length. Instead, I reconstructed Proto-Garnet as having an 8-vowel system. This is she:

/i ɨ u e ə o æ ɑ/ i ï u e ë o ä a

Also, the overall syllable complexity of Proto-Garnet has increased from CVC to CCVCC. The specifics of what exactly is allowed as an onset and coda cluster are as follows:

ONSET
- OL (plain obstruent + liquid)
- ON¹ (plain obstruent + nasal)
- FO (fricative + obstruent (does not have to be plain, but usually is))

CODA
- LO (liquid + plain obstruent)
- NO (nasal + plain obstruent (does not have to be, but usually is homoorganic)
- FO (fricative + plain obstruent)

1. only heteroorganic clusters of this type are allowed. For example, *dm is allowed, but *dn is not.

There is still a restriction on glottalic consonants in coda position, and an additional restriction is on aspirated stops in coda position. However, the progressive assimilation of aspiration/glottalization with stops still occurs intervocalically, however fricatives no longer affected by this process. This does sadly mean the death of Keyne's Law in Proto-Garnet. My condolences Jackk.

Going back to what I had said earlier about making the phonological changes more palatable by reducing the orthographical changes, That is also why I got rid of vowel length and increased the syllable complexity (along with the fact that every Pyralspite language has a syllable structure like that, and it is only Ugrandite languages have a simple syllable structure, though in reality it is only Uzi with a truly "simple" syllable structure). It was also very garish to me seeing all of these *i's and *u's running around which in the end were reduced to schwa and often deleted in Vuase. By just introducing central vowels, which both BOZ and VUA share, I was able to greatlly diversify the distribution of vowels in the language.

Moving on from phonological changes, let's discuss the (now VASTLY simplified) grammar. Nouns have not changed much, they are still marked for case, though I split the genitive case into two cases, the possessive and ablative. I don't want to spend a year describing the minutae of each case's functions again, but I will say that not much has changed. This split between possessive and ablative does however highlight another new change, which is that Proto-Garnet now has an animacy system, as infantile as it may be in this stage. There are three primary things which this affects grammatically:

1. animate nouns can sometimes be marked for number, but inanimate nouns never can. The primary way in which this is done is via reduplication of the final rhyme of a word.
2. inanimate nouns are unmarked in the nominative and the accusative, however animate nouns are specially marked in the accusative (differencial object marking)
3. possession on animate nouns is marked by using the ablative case and a possessive postposition (this is how alienable possession is marked); inanimate nouns, however, can either be marked this way, or by using the possessive case to indicate inalienable possession, depending on the noun.

As for verbs, their morphology has been greatly lobotomized. Verbs now only distinguish between past and nonpast via a past-tense suffix -är. To compensate however, Proto-Garnet has a set of particles as well as auxiliaries which convey more complex meanings. These both go after the verb.

And finally, on the topic of adjectives, I have gotten rid of the concept of the substantive for the most part. It was one of those things which I was saying were complicated rather than complex. Anyhow, the concept still survives as a new distinction between so-called "quantitative" and "qualitative' adjectives (a concept which persists in all three documented Garnetian languages, and is in fact one of the main reasons I decided to combine all these languages. Of course, the real meta reason these all share this concept is because I just think it's neat). Quantitative adjectives, which include numerals as well as words like "many" "few" and even measure words like "tall" or "wide," go before the noun, while qualitative adjectives (the rest of them) go afterwords.

That is all I have for now. I will maybe return with a complete list of common verbal particles and auxiliaries, and then some more grammar notes when I have time to reconcile the grammars of these three disparate languages. Despite the extra free time afforded by this global pandemic, reading through the notes of me 4-5 years ago is still an undertaking. Anyway, see y'all next time!
Aszev wrote:A good conlang doesn't come from pursuing uniqueness. Uniqueness is usually an effect from creating a good conlang.
Project Garnet
(used to be Bulbichu22)
User avatar
Corphishy
greek
greek
Posts: 740
Joined: 18 May 2013 18:28
Location: Route 102, Route 117, Petalburg City

Re: Project Garnet Scratchpad

Post by Corphishy »

Small update: I have gotten rid of the voicing distinction on sibilants, and will instead derive them with sound changes in Ugrandite languages. Also, I have gotten rid of the phonemic distinction between *q and *qʰ, they will instead be one phoneme written <q> whose exact pronunciation depends on environment
Aszev wrote:A good conlang doesn't come from pursuing uniqueness. Uniqueness is usually an effect from creating a good conlang.
Project Garnet
(used to be Bulbichu22)
User avatar
Jackk
roman
roman
Posts: 1487
Joined: 04 Aug 2012 13:08
Location: Damborn, Istr Boral

Re: Project Garnet Scratchpad

Post by Jackk »

Woo, new update! [:D]
Corphishy wrote: 20 Mar 2020 18:09 I tend to lose myself, and then when I return, what usually follows are some sweeping reforms to the system. This time is no different.
big mood Hopefully this iterative process will converge on something lol.
Corphishy wrote: 20 Mar 2020 18:09 However, I have greatly simplified many things which did not need to be as complicated as they were
Good to hear! Hopefully this will make it easier to develop the descendants.
Corphishy wrote: 20 Mar 2020 18:09 Primarily the reason is because in sketches of Proto-Bozeto-ei, its ŋ originated from an earlier ɲ, and in Vuase's history, *ɲ eventually merged with *j. On top of this, Both Bozeto-ei and Uzi have or had a palatal series, which is why I replaced the earlier laminal/apical distinction with an alveolar/palatal one. In Pyralspite languages, this doesn't end up mattering much, but in Ugrandite languages it does figure. This is alo why I made them palatalized velars in PG instead of full palatals. In Pyralspite, they simply merge with plain velars.
All very sensible. [+1]
Corphishy wrote: 20 Mar 2020 18:09 Before, I would have a bunch of plain p t k becoming b d g in Vuase and quite frankly it sounded and looked ugly to me. However, having that same change occuring, but not changing anything orthographically (*b > b is the same as */p/ > /b/) made it much more palatable to me.
Nice, haha [:D] Yeah, I find having a lot of <b d g> [:D] an overwhelm an aesthetic.
Corphishy wrote: 20 Mar 2020 18:09 ONSET
- OL (plain obstruent + liquid)
- ON¹ (plain obstruent + nasal)
- FO (fricative + obstruent (does not have to be plain, but usually is))

CODA
- LO (liquid + plain obstruent)
- NO (nasal + plain obstruent (does not have to be, but usually is homoorganic)
- FO (fricative + plain obstruent)
Cool phonotactics, especially the FO asymmetry. [<3]
Corphishy wrote: 20 Mar 2020 18:09 This does sadly mean the death of Keyne's Law in Proto-Garnet. My condolences Jackk.
[:'(]
Corphishy wrote: 20 Mar 2020 18:09 1. animate nouns can sometimes be marked for number, but inanimate nouns never can. The primary way in which this is done is via reduplication of the final rhyme of a word.
2. inanimate nouns are unmarked in the nominative and the accusative, however animate nouns are specially marked in the accusative (differencial object marking)
3. possession on animate nouns is marked by using the ablative case and a possessive postposition (this is how alienable possession is marked); inanimate nouns, however, can either be marked this way, or by using the possessive case to indicate inalienable possession, depending on the noun.
All very cool! I enjoy seeing how noun class interacts with the rest of the grammar.
Corphishy wrote: 20 Mar 2020 18:09 As for verbs, their morphology has been greatly lobotomized.
[}:D] As a fan of more analytic strategies for marking stuff in general, I can't disapprove haha
Corphishy wrote: 20 Mar 2020 18:09 Of course, the real meta reason these all share this concept is because I just think [the adjective distinction is] neat
Best reason for anything, ofc. :p


Keep it up! [:D]
terram impūram incolāmus
hamteu un mont sug
let us live in a dirty world
User avatar
Corphishy
greek
greek
Posts: 740
Joined: 18 May 2013 18:28
Location: Route 102, Route 117, Petalburg City

Re: Project Garnet Scratchpad

Post by Corphishy »

Jackk wrote: 22 Mar 2020 15:59 big mood Hopefully this iterative process will converge on something lol.
Yeah, as much as I love this process of endlessly tinkering with this language like a middle aged dad tinkering with his car, I think I will soon be done with this part of the project and can move on to derive daughters.
Jackk wrote: 22 Mar 2020 15:59 Cool phonotactics, especially the FO asymmetry.
The fact that you called it an asymmetry gave me the idea that perhaps at one stage there was a metathesis of onset OF to FO. Since you gave me the idea, it only seems fair to dub this sound change the new Keyne's Law. I could even make it productive in PG, turning a word like *fudšay "wild" or *kubþe "wick" into *fušday and *kuþbe in certain daughters, or perhaps also with compound words or affixes (turning a word like, say, *nogþe "ending" (from *nog "to end") into *noþge)
Aszev wrote:A good conlang doesn't come from pursuing uniqueness. Uniqueness is usually an effect from creating a good conlang.
Project Garnet
(used to be Bulbichu22)
User avatar
Corphishy
greek
greek
Posts: 740
Joined: 18 May 2013 18:28
Location: Route 102, Route 117, Petalburg City

Re: Project Garnet Scratchpad

Post by Corphishy »

Okay, so I've made yet another update to the Proto-Garnet phonology. Please understand that I am attempting to reconcile the phonology of 3 very different languages phonologically speaking. As such, I have made a few decisions. I have reinstated the voicing distinction in sibilants. I don't know why I thought it would be a good idea not to have it. Also, I have changed the aspiration distinction to a voicing distinction (you'll find I think that with every update I am making Proto-Garnet more Normal™).

On top of this, I have simplified the legal consonant clusters. An onset cluster can ONLY be of the type CL, where C is ANY consonant, and L is any liquid. This means that I have gotten rid of ON and FO clusters entirely. Instead, this is an innovation of Pyralspite. Specifically, via these sound changes

PG *Nw *Ny *Nr *Nl > PAS1 kʷN cN tN2? tN
PG *sw *sy *sr *sl > PAS skʷw scy skʷr3 stl

For example, in Proto-Garnet you have the word *sra'int "sword," which in Vuase ultimately becomes skraint, though the <ai> is often reduced in that environment to i or ɛ, depending on one's dialect, and the final <nt> is often reduced to just n, written mostly to keep track of it when it shows up in things like the plural skrainten /ˈskɾ̥ɛntɜn/. In Bozeto-ei, this word becomes instead something like rha-ït /r̥aˈʔɨt/.

More minorly, I have edited the kinds of cluster which can occur in the coda. Primarily, I have removed the FO clusters from occurring in the coda. I have also gotten rid of nasal+fricative clusters, mostly because they look ugly to me and even all those years ago when I added them to Vuase I didn't like them. Vuase doesn't even really have NC clusters (at least, not word finally): the only reason I am keeping them is because I do like the idea of NC clusters creating ghost segments like with skren/skrenten or om/omben (ear/ears). However, due to the fact that Vuase has its own written form, I will tend to write in the less phonemic, but more accurate to the in-universe orthography writing system (skraint and omb instead of skren/skrin and om). In Ugrandite, these NC clusters cause nasalization and lowering (as in the above example *-int > -ït).

I have kept the LO clusters. In Ugrandite languages, these are very simply reduced like VLO > VVO. In Bozeto-ei, this is the origin of diphthongs (similar to Vuase/Pyralspite langs but to an even greater extent: *erd against > PAS *erd, BOZ eït), and in Uzi it is the origin of vowel length (UVA4 ēl).

Finally, I have decided to get rid of *θ entirely. The reason for this is that it does not appear in any DDs5, and in both Vuase and Bozeto-ei, it is gotten rid of in the exact same way (by becoming t). However, in PAS, *z becomes *ð̞, which merges with r in Spessartine languages like Vuase. Also, š ž become velar in PAS, either by moving toward palatal/velar, or by being labialized similar to how *r was. Either way, š ž > x ɣ, and then x (as well as f) become h and eventually are deleted in Vuase, while ɣ merges with g. So, you end up with cognates like VUA agog vs BOZ ajoš from PG *ajož brave.

I make no pretense that this will be the final phonology update. However, it is starting to approach something I am happy with. Also, I have decided to include more references to PG's daughters, specifically the Documented Daughters, in order to shift focus in my mind away from endlessly tinkering with a protolang to trying to finish a tool I can then use to create daughterlangs. I will at some point speak about some of the grammar of PG: specifically I wish to nail down the common verbal particles and auxiliaries so that I can actually translate things without needing to coin a particle just for the specific tensepectual concept I need to convey at the time. See you then!

1. PAS is the abbreviation I use for Pyralspite: It comes from the etymology of the word pyralspite, which is a combination of Pyrope, Almandine and Spessartine. Similarly, Ugrandite is abbreviated as UGA: Uvarovite, Grossular and Andradite
2. I don't know what to make this, because sr > skʷr
3. kʷ happens because of labialization of postalveolars
4. UVA = Uvarovite, the branch which includes Uži and Lelendzi

5. Documented Daughters; Vuase, Bozeto-ei and Uži
Aszev wrote:A good conlang doesn't come from pursuing uniqueness. Uniqueness is usually an effect from creating a good conlang.
Project Garnet
(used to be Bulbichu22)
User avatar
Corphishy
greek
greek
Posts: 740
Joined: 18 May 2013 18:28
Location: Route 102, Route 117, Petalburg City

Re: Project Garnet Scratchpad

Post by Corphishy »

This Project is on permanent hiatus
Aszev wrote:A good conlang doesn't come from pursuing uniqueness. Uniqueness is usually an effect from creating a good conlang.
Project Garnet
(used to be Bulbichu22)
Post Reply