What did you accomplish today? [2011–2019]

A forum for all topics related to constructed languages
User avatar
MrKrov
banned
Posts: 1929
Joined: 12 Aug 2010 02:47
Location: /ai/ > /a:/
Contact:

Re: What did you accomplish today?

Post by MrKrov »

Bluntly: self-aggrandizing is what.
User avatar
Lambuzhao
korean
korean
Posts: 5405
Joined: 13 May 2012 02:57

Re: What did you accomplish today?

Post by Lambuzhao »

Lao Kou wrote:
Egerius wrote:
Chagen wrote:I ended up somehow ballooning the verbal classes from 7 to 25.
...no wait, I forgot the syllabics. That means we get 30+ with ease.
Linguistic supernova blowing up in 3, 2, 1... [xD]
Major Tom saw it in 4, 3, 2, 1... Long live Major Tom
My momma said "To get things done
U better not conujugate Major Tom"
Vechnaya pamyat to the both of 'em [<3] [<3]

...Was just talking with my son about Space Oddity.
How odd it's mentioned herein. And now~ish.
:wat: [;)]
User avatar
Chagen
runic
runic
Posts: 3338
Joined: 03 Sep 2011 05:14
Location: Texas

Re: What did you accomplish today?

Post by Chagen »

elemtilas wrote:

So . . . what's the issue with 30 verbal classes? Surely the more the merrier? :mrgreen:
Ridiculous complexity in a language that is already super complex. And worse: complexity that adds no new grammatical or syntactic wrinkles, just more brain-burden.
gestaltist wrote: Are you reworking Polish or something? [xD]
Nah, lol. Verbs in Pazmat are arranged according to their base-vowel, and whatever extra stuff has an effect on what that verb's grades will look like. mat- "to speak" is an a-base, midh- "to ask" is an i-base, and śtars- "to move" is an ar-base, because it has different grades than an a-base like mat- (śtars/śtōs/śtaus vs. mat/mēt/meyt). The simplest verbs, such as mat- and midh-, which follow the "normal" ablaut patterns for their vowels, are grouped as "simple verbs".

Though, not all verbs with -ar- are ar-bases: scar- "to shout, scream" is a technically an a-base, as verbs are ranked according to how they appear when suffixed with a vowel, and verbs which end in -ar- (as opposed to -arC- like śtars-) act like normal a-bases this way: "I am shouting" is scērana akin to mētana "I am talking", compared to śtōsana "I am moving (s.thing else)". This is becaue almost every verbal suffix ends in a vowel. However, for the few times they don't (such as when compounding, or nominal suffixes such as -tū (but not when the suffix begins with /r/ itself), then verbs like scar- would behave like ar-bases, as in the word scōtū "loudspeaker". As such, they are sometimes considered not a-bases nor ar-bases, but a separate class: the mixed rhotic a-bases (confusingly enough, ar-bases and er-bases are called "rhotic verbs"!)

If that were not enough, Pazmat grammatical tradition often has redundant classes. Remember how I said the ar-bases and er-bases were considered "rhotic verbs"? Well, -irC-, -urC-, and -orC- (i.e ir-base, ur-base, and or-base) verbs are also part of the rhotic verbs. However, these verbs behave identically to their non-rhotic counterparts, making these classes honestly superfluous. Look at the identical grades between vus- "lay prone" and burst- "feel (emotionally)":

vus/voys/vūs
burst/boyrst/būrst

In Pazmat grammatical tradition, vus- is a u-base and burst- is a ur-base, even though they behave identically and nothing prevents burst- from being a simple u-base. It's classed as a ur-base solely so the rhotic verbs have five classes like the simple ones, to appear more "orderly" and "pleasing". The traditional even calls these three "rhotic" verbs "fake rhotics". This does inflate the amount of classes a little more. Of course, I could always change the ablaut patterns of the three fake rhotics to make them true rhotics, but that's a lot of extra complexity...

Also, remember how I said the simple verbs had five classes? They actually should have seven, including the ṛ-bases and ṇ-bases, but these are considered a separate subset called "syllabics" (there are no syllabic rhotics, because honestly how the hell do you pronounce -ṛrC- or -ṇrC-). Then we have the ṃ-bases, except those...became their own weird class as ṃ disappeared early. Then syllabics are split into "open" and "closed", depending if they have a consonsant post-the syllabic or not (which honestly changes almost nothing...I think).

Moving on from that there are what can be considered Pazmat's "true" irregular verbs: verbs whose roots end in -y- or -w-. These verbs are fine when suffixed with vowels (as nearly all verbal suffixes are...except the infinitive), it's when consonants are suffixed that they act weirdly (as of now, -y- becomes -ṣ- and -w-...uh, geminates verbs or messes with the vowels before it?*). This does mean that like 95% of the time they act like regular old simple verbs, but...gotta elaborate on everything, man!

*:I've been debating changing Pazmat's historical phonotactics, and allow /j w/ to be appear post-vocally like how they could in Proto-Indo-European, forming...diphthongs, sort of, but now really? I'm thinking of having these resonants screw up the ablaut of their vowels, or mess with the consonants following them (though I'm loath to alter the consonantal portions of roots too drastically), or perhaps become i/ī and u/ū under some circumstances (for instance: *dááyt > *deyyt > deyīt or maybe the /a/ never changes to /e/ and the result is dayīt or dayat or SOMETHING...feels extremely strange to expand a monosyllabic root to a disyllabic one though .__.

See, this is emblematic of one of my problems: I have to elaborate on every single possible permutation. The result is I'm wracking my brain to come up with all of them and coming up with solutions which is just extremely taxing...

Anyway, Pazmat grammatical tradition, like the Arabic one, uses a set of made up verbs to describe every class. In Pazmat's case, it's mV(r)th- for consonantal/"closed" verbs and mV(r/y/w)- for the ones which end in special consonants or are vocalic. The results gives us the following classes; in the following chart, a * in a class's name means that it's redundant and identical to its simple counterpart in all respects, while a ** means it's identical to the simple counterpart except when a consonant is directly suffixed. Along with the three grades, three normal forms are given: an reduced ar-stem, a tū-stem, and an infinitive, showing the long and overlong grades, respectively, when a consonant is suffixed. The only reason an ar-stem and tū-stem are given when both are long grade is because rhotic simple a-bases and e-bases (that is, verbs which end directly in -ar or -er) act differently when -r- is suffixed compared to any other consonant. Yes, the grammarians go to that level of detail...except I realized that this does not show what happens when an r is suffixed to an overlong root. OH WELL. That's such an extremely rare occurance it doesn't matter!

Oh, and one final thing: three asterisks (***) is reserved for a few classes which don't even exist (and probably never did) but were made just to make things even more orderly. They are classes involving ṃ.
Spoiler:
(I decided to cut out the nominal examples since it was taking forever and I'm only focused on the amount of classes)
(using a made-up verb for every class feels kinda boring when I could use real verbs, but I'm too lazy to trawl through my dictionary, and the point is to show off the ablaut with as few extraneous bits as possible)

Image

(this is not set in stone and honestly I will probably change some of this stuff).
There are 28 classes total, but cutting out all the 17 redundant ones give us a mere...11. Wow. That is a SHITLOAD of redundant classes. Granted, I really want to make the redundant classes unique, but I don't want classes to appear too similar...I'm very limited with only five vowel qualities, and four diphthongs (though I want to add a few more).

You know, I probably should have posted this explanation in my Pazmat thread. Now I'll have to write it all over again later...I spent like an hour making this pointless chart and post.
Nūdenku waga honji ma naku honyasi ne ika-ika ichamase!
female-appearance=despite boy-voice=PAT hold boy-youth=TOP very be.cute-3PL
Honyasi zō honyasi ma naidasu.
boy-youth=AGT boy-youth=PAT love.romantically-3S
jimydog000
greek
greek
Posts: 583
Joined: 19 Mar 2016 04:14
Location: Australia

Re: What did you accomplish today?

Post by jimydog000 »

I made around 200 nonsense words in my unnamed conlang, with it's phonotactics.
A signature.
User avatar
elemtilas
runic
runic
Posts: 3021
Joined: 22 Nov 2014 04:48

Re: What did you accomplish today?

Post by elemtilas »

Chagen wrote:
elemtilas wrote:

So . . . what's the issue with 30 verbal classes? Surely the more the merrier? :mrgreen:
Ridiculous complexity in a language that is already super complex. And worse: complexity that adds no new grammatical or syntactic wrinkles, just more brain-burden.
I'm sure if anyone can uncover ever new ever wonderful syntactical or grammatic wrinkles it'ld be you.
There are 28 classes total, but cutting out all the 17 redundant ones give us a mere...11. Wow. That is a SHITLOAD of redundant classes. Granted, I really want to make the redundant classes unique, but I don't want classes to appear too similar...I'm very limited with only five vowel qualities, and four diphthongs (though I want to add a few more).

You know, I probably should have posted this explanation in my Pazmat thread. Now I'll have to write it all over again later...I spent like an hour making this pointless chart and post.
Just copy-n-paste. Anyway, looks like you're well on the way to approaching the complexity of, e.g., English for complexity!

Here's an interesting take on syntactic verb classes in English.
User avatar
Chagen
runic
runic
Posts: 3338
Joined: 03 Sep 2011 05:14
Location: Texas

Re: What did you accomplish today?

Post by Chagen »

elemtilas wrote:
Chagen wrote:
elemtilas wrote:

So . . . what's the issue with 30 verbal classes? Surely the more the merrier? :mrgreen:
Ridiculous complexity in a language that is already super complex. And worse: complexity that adds no new grammatical or syntactic wrinkles, just more brain-burden.
I'm sure if anyone can uncover ever new ever wonderful syntactical or grammatic wrinkles it'ld be you.
I'd love to but ablaut in Pazmat works like it does in PIE--it's, honestly, completely arbitrary. I like it because it adds variety to inflections.

I'd like to add one or two more diphthongs to Pazmat, but I'm not sure which ones. I really would love to have one spelled <iy> for that Arabic feel (how is that pronounced in Arabic exactly, by the way?), but that in Arabic only appears at the end of words it seems and so in the middle it does look kinda weird. I could also do /uj/ but that also sounds weird. There's only one Vu diphthong in Pazmat right now, /aw/ (/ew/ got turned into /o:/) but none of the others (/iw ow ew/) really fit well how I want the language to look. /aw/ is honestly the only Vu diphthong I think sounds good. /ew/ sounds goofy /ew/ sounds too close to /o/ because English is weird and doesn't have /o/, and /iw/ also sounds weird. I want Pazamat to only have the "classic" five vowels /a i u e o/ so that limits me quite a bit.

For the record, /ej/ has the same problem as /ow/ for me: since English has /ej ow/ instead of /e o/, the former sound almost identical as the latter to me. /ey/ is pretty much only there because it was there from the beginning.

All in all Pazmat's ablaut patterns are a little weird. They don't really have any good logical reasoning, and exhibit some weird patterns which...can't really be used in an interesting way. For instance, I've noticed a pattern amongst the five vowels: The mid and low /a e o/ have a long vowel for their long grades, and a diphthong for their overlong grades:
a/ē/ey
e/ī/ay
o/ā/oy (really wish I could come up with something better for this, having /oj/ as a form of /o/ just bugs me)
Meanwhile the high vowels /i u/ have diphthongs for their long grades and long vowels for their overlongs:
i/au/ū
u/oy/ū

...but nothing can really be done with this. Well, outside of changing the ablaut patterns which feels like sacrilege now that they're s

Truthfully, a large amount of my problems come from my own flaws and weird hang-ups. For instance, I feel the need to laboriously elaborate every single possibility. The result is that I tax my brain thinking of every single possibility instead of coming up with a decent framework and working on stuff as it appears. As such, I blow all my brainpower for the day endlessly going over charts and calculating forms until my head explodes from checking everything and I burn out for the day, and I keep doing it day after day after day until I just want to smash something.

Another thing I often do is come up with a cool new rule, and then realize it fucks some earlier thing up or has far-reaching implications. For instance, I recently decided that /w/ in Old Pazmat became a weakly fricated approximant /ʋ/, spelled <v> in the orthography (though distinct from the identically-spelled /v/, and viewed as a variant of /w/ by Pazmat speakers in certain situations: more specifically, post-consonantally and non-initially. This is a great way to explain otherwise inexplicable Cv clusters: tvīsena- "I am soaking/relaxing in hot water"* and madhaśva "we(incl.) lead" in Old Pazmat were twēsena and madhaśwa.

*: Apparently I do not actually have a single root with initial Cv in my dictionary. Funny. I made tves- "soak in hot water", and then made a counterpart yarch- "soak in cold water" because that's a quirky and unique set of verbs.

Sounds great...except this means that madhawa should be madhava. Also, any verb ending in -w- like tuw- "fly" would exhibit a -v- in almost every single conjugation (tuvū "it flies", toyvutha "it is flying", toyvuvyū "it flew", etc.) Granted this is easily fixable by removing the "non-initial" requirement. I think /w/ > /ʋ/ in between vowels is an interesting change, but...from what? When? See, this is what I mean, I end up stressing my brain to think of every single situation and I burnout. Maybe before /u/, but not initially (so wurfarā "boy" remains unchanged, though initial /wu/ will probably be pronounced [hʍ~ɦw] or summat, my voice settles on either that or a raised aspirated [w̝ʰ]...is that even a thing?

To be honest, I don't think any wV syllable other than /wa/ sounds nice for Pazmat. /wi wu we wo/ sound "weird" except initially. I might have all instances of them become /v/ medially, though they remain initially and when the /w/ is part of a root, I guess. Not like they appear that often anyway: -awa is about the only (V)wV suffix in Pazmat right now IIRC.

Also this doesn't explain what happens when /w/ is pre- vocal or final. Rare, yes, but let's say "bird" is tuwā, derived from tuw-. "a bird" is *toyw, "to a bird" is *toywyā. Now I've thought that /w/ > /ʋ/ before resonants (/r l m n ŋ w j/), so "to a bird" is toyvyā. But what happens when /w/ is final? Is there an epenthetic vowel? I have a few ideas: for instance, post dipthongs like in *toyw, it could vocalize to a /u/, breaking the diphthong: toyu "a bird", and in the case of -auw# you get -avu I guess: ḥavu "a gasp" (cf. hiwā "the gasp")...it feels incredibly weird having irregular nouns like this, even if they follow a logical rule.

I have some other ideas, such as -oywu- > -oyuu- > oyū, such that "I am flying" is toyūna, not toywuna. Another way to go about it would be -oywu- > -owwu- > -ōwu-, thus tōwuna "I am flying". I like this second one because it preserves the root's consonantal structure (tVw) and I generally don't like fucking with that around too much. This would apply in in the case of -aywu- and -eywu-, but these are extremely rare compared to -oywu- (which will appear in the Imperfect and Perfect of any w-final u-base root like tuw- No wait, they would appear in w-final e- and a-bases conjugated to the Stative: maw- "float" > meywubbū > mēwubbū "it begins to float"; nyew- "squeeze" > nyaywubbū > nyāwubbū "he begins to squeeze".

But...as much as I like this, the part of my brain that hates messy irregularity is screaming at me for it. We've got a situation where a root appears differently in one specific ablaut grade when one specific vowel follows it. Nothing else in Pazmat is like this. It just feels wrong. In the case of w-final a-bases like maw-, the Overlong grade is identical to the Long one in one specific conjugation, whereas in w-final e-bases like nyew- the Overlong grade looks like an o-base's Long grade.

But isn't this a good thing? After all, no language is 100% regular, and irregular stuff like this is common. Yet, it just...I don't know. I just don't.

After writing this, I realized that I could just analogize this, so that, say, -oywV- becomes -ōwV- in all cases. It sounds strange that one irregularity could force analogy, but -oywu- happens to be in the most common conjugation for w-final u-bases, so it makes sense, and makes the irregularity now fit Pazmat's tendencies. Single forms have been the analogy pivot before in natlangs anyway...this does not help for the other w-final roots, however, where -aywu- and -eywu- aren't nearly as common...but then again, that's one of the few times the Overlong grade ever appears at all! (I really need to use it more)

...I haven't even talked about when obstruents are suffixed to w-final roots. I thought about geminating the consonant. But that messes up the consonantal structure of the root and I hate that (even though that happens in y-finals, where the <y> becomes <ṣ>). Christ. I need to get over my weird hang-ups.

This doesn't even affect grammar! You know, the stuff I really would like to work on again. Ugh. In any case I think I'm gonna add a Cessative and think about making the Stative a derived stem (that is, you form a new verb meaning "to begin to [VERB]" with the stative which is derived like a normal verb), along with the Cessative I guess. But if I go that far, might as well make the Desiderative a derived stem.

You know, I think I should make Desiderative, Stative, and Cessative stems for participles. Pazmat uses participles for relative clauses and this means awkward participle and infinitive combinations (i.e "want to X" in a relative clause requires an infinitive with a participle of ray-, which means "to desire". At the same time, I like having periphrastic combinations. Having the verb inflect with everything feels like a...cop-out? I don't know. I really don't. God dammit. I could have the infinitives merge with the participles to form single words, but that also feels lazy.

I also played myself, as the Desiderative is based around -ar-, roughly, except -ar- is also the present active participle suffix (though I should probably change that..maybe to placing the base vowel in its place? That does sound nice), so I dunno what Desi. stem that isn't near-identical to the present active participle stem would even look like. Also, an almost completely unrelated note, I like the idea of the -ara- Desi.pres. ending becoming aCC in casual speech, so, say, milarana! "I want to lie down!" becomes milanna!. Except that looks a little...undistinct? What the fuck Chagen, it is pretty distinct, stop being an idiot. Maybe Pazmat needs a set of irrealis person markers. Or maybe that's just me coming up with an extremely overcomplicated solution to a simple problem again. Yeah.


Dear god look at this super long post. I'm sorry guys, I just find this ranting and describing my various processes and work to be therapeutic in a way. That's why they get so long, I just keep adding stuff because it's entertaining to write these posts even as they get super long.
Nūdenku waga honji ma naku honyasi ne ika-ika ichamase!
female-appearance=despite boy-voice=PAT hold boy-youth=TOP very be.cute-3PL
Honyasi zō honyasi ma naidasu.
boy-youth=AGT boy-youth=PAT love.romantically-3S
User avatar
KaiTheHomoSapien
greek
greek
Posts: 641
Joined: 15 Feb 2016 06:10
Location: Northern California

Re: What did you accomplish today?

Post by KaiTheHomoSapien »

^One thing I've liked about studying PIE ablaut is the extent to which it often seems arbitrary. The same ablaut patterns occur throughout the language, but they don't have the same functions when they do. For some things, it seems to be redundant. Why change /e/ to zero in the root in the genitive case; isn't the genitive ending enough? I've used this same kind of ablaut in my conlang, and I don't mind it at all. I love the variety that it adds.
Image
User avatar
Parlox
greek
greek
Posts: 495
Joined: 10 Feb 2017 20:28
Location: Ehh

Re: What did you accomplish today?

Post by Parlox »

I have finished (I think) a new writing system bases off of insular script.
:con: Gândölansch (Gondolan)Feongkrwe (Feongrkean)Tamhanddön (Tamanthon)Θανηλοξαμαψⱶ (Thanelotic)Yônjcerth (Yaponese)Ba̧supan (Basupan)Mùthoķán (Mothaucian) :con:
User avatar
elemtilas
runic
runic
Posts: 3021
Joined: 22 Nov 2014 04:48

Re: What did you accomplish today?

Post by elemtilas »

Chagen wrote:I'd love to but ablaut in Pazmat works like it does in PIE--it's, honestly, completely arbitrary. I like it because it adds variety to inflections.
Not entirely arbitrary. Of course, Pazmat ablaut doesn't háve to be so arbitrary!
I'd like to add one or two more diphthongs to Pazmat, but I'm not sure which ones. I really would love to have one spelled <iy> for that Arabic feel (how is that pronounced in Arabic exactly, by the way?), but that in Arabic only appears at the end of words it seems and so in the middle it does look kinda weird. I could also do /uj/ but that also sounds weird. There's only one Vu diphthong in Pazmat right now, /aw/ (/ew/ got turned into /o:/) but none of the others (/iw ow ew/) really fit well how I want the language to look. /aw/ is honestly the only Vu diphthong I think sounds good. /ew/ sounds goofy /ew/ sounds too close to /o/ because English is weird and doesn't have /o/, and /iw/ also sounds weird. I want Pazamat to only have the "classic" five vowels /a i u e o/ so that limits me quite a bit.
Mine has /o/. Like /bot/. Anyhoo. English defs weirds, and ymmv, etr.
For the record, /ej/ has the same problem as /ow/ for me: since English has /ej ow/ instead of /e o/, the former sound almost identical as the latter to me. /ey/ is pretty much only there because it was there from the beginning.
/plet/
All in all Pazmat's ablaut patterns are a little weird. They don't really have any good logical reasoning, and exhibit some weird patterns which...can't really be used in an interesting way. For instance, I've noticed a pattern amongst the five vowels: The mid and low /a e o/ have a long vowel for their long grades, and a diphthong for their overlong grades:
a/ē/ey
e/ī/ay
o/ā/oy (really wish I could come up with something better for this, having /oj/ as a form of /o/ just bugs me)
Meanwhile the high vowels /i u/ have diphthongs for their long grades and long vowels for their overlongs:
i/au/ū
u/oy/ū

...but nothing can really be done with this. Well, outside of changing the ablaut patterns which feels like sacrilege now that they're set.
Why not? You could combine ablaut gradation with accent shift. Perhaps build some verbal nouns or some other interesting structure on one or other of the grades. A heavier grade accented form could sap the strength of a neighbouring vowel and reduce it or delete it.
Truthfully, a large amount of my problems come from my own flaws and weird hang-ups. For instance, I feel the need to laboriously elaborate every single possibility. The result is that I tax my brain thinking of every single possibility instead of coming up with a decent framework and working on stuff as it appears. As such, I blow all my brainpower for the day endlessly going over charts and calculating forms until my head explodes from checking everything and I burn out for the day, and I keep doing it day after day after day until I just want to smash something.
Sounds about right.
Another thing I often do is come up with a cool new rule, and then realize it fucks some earlier thing up or has far-reaching implications. For instance, I recently decided that /w/ in Old Pazmat became a weakly fricated approximant /ʋ/, spelled <v> in the orthography (though distinct from the identically-spelled /v/, and viewed as a variant of /w/ by Pazmat speakers in certain situations: more specifically, post-consonantally and non-initially. This is a great way to explain otherwise inexplicable Cv clusters: tvīsena- "I am soaking/relaxing in hot water"* and madhaśva "we(incl.) lead" in Old Pazmat were twēsena and madhaśwa.
Excellent! Complexity breeds complexity. I'd say keep it up!
*: Apparently I do not actually have a single root with initial Cv in my dictionary. Funny. I made tves- "soak in hot water", and then made a counterpart yarch- "soak in cold water" because that's a quirky and unique set of verbs.
I love sets like that. Queranaran is chock full of the things. But that's just how Daine are: if they have a verb for something, they also find they have to have a verb for a range of oppositional points. (.i. "soak in hot water" would be opposed by not only "soak in cold water" but also "dunk in hot water" & "dunk in cold water" and "soak / dunk in warm / cool water"). Probably also verbs for "dipping one's left foot into cool / warm water" and "dipping one's right foot into warm / cool water".
Sounds great...except this means that madhawa should be madhava. Also, any verb ending in -w- like tuw- "fly" would exhibit a -v- in almost every single conjugation (tuvū "it flies", toyvutha "it is flying", toyvuvyū "it flew", etc.) Granted this is easily fixable by removing the "non-initial" requirement. I think /w/ > /ʋ/ in between vowels is an interesting change, but...from what? When? See, this is what I mean, I end up stressing my brain to think of every single situation and I burnout. Maybe before /u/, but not initially (so wurfarā "boy" remains unchanged, though initial /wu/ will probably be pronounced [hʍ~ɦw] or summat, my voice settles on either that or a raised aspirated [w̝ʰ]...is that even a thing?
Salright. You don't have to create Pazmat in a fortnight, you know. If you're halfway done with it by the time you're 60, I'd be frankly surprised.
...But...as much as I like this, the part of my brain that hates messy irregularity is screaming at me for it. We've got a situation where a root appears differently in one specific ablaut grade when one specific vowel follows it. Nothing else in Pazmat is like this. It just feels wrong. In the case of w-final a-bases like maw-, the Overlong grade is identical to the Long one in one specific conjugation, whereas in w-final e-bases like nyew- the Overlong grade looks like an o-base's Long grade.
But this is what is becoming very cool about Pazmat. It's actually accreting irregularity, spawning seemingly senseless divergence and, in short, looking for all the world like an actual language.

This is not a Bad Thing at all.
But isn't this a good thing? After all, no language is 100% regular, and irregular stuff like this is common. Yet, it just...I don't know. I just don't.
It's a very good thing indeed. It's a sign a) that you're skilled in the Art and b) that Pazmat is a / the language you need to focus on long term. I don't mean "if I'm still working on it next Saturday, that'll be like forèver!" kind of long term. I mean this is clearly your heart language and ought to be lovingly worked on until can't work on it any more, be that fifty or sixty or seventy years hence.

(snip)
Dear god look at this super long post. I'm sorry guys, I just find this ranting and describing my various processes and work to be therapeutic in a way. That's why they get so long, I just keep adding stuff because it's entertaining to write these posts even as they get super long.
Please don't apologise for a long meaty post of this sort. I really don't mean to come off sounding insulting, but most posts we see on language work (the half-a-sentence-that-sums-up-a-month's-work kind of posts) are like this:

Image

Your post is the kind I (and I'm sure many others here) really enjoy! It's like this:

Image
User avatar
Inkcube-Revolver
cuneiform
cuneiform
Posts: 126
Joined: 05 Nov 2015 23:20
Location: Miami, FL

Re: What did you accomplish today?

Post by Inkcube-Revolver »

Firstly, very hungry for that plate of food there. Secondly, I agree with Elemtilas, the irregularity looks great and the way Pazmat works around the irregularities is very interesting. It's easy to say to just get over your weird hang-ups, but your only real options are to either continue working around them or just ignore them. I do a lot of both, and I've been swamped in phonotactics and phonology for one conlang for eight months.

EIGHT.

MONTHS.

This was not the time length I anticipated for working on this single facet of the language, and some days working on actual grammar seems like a pipe dream.
My goal was originally to have the phonological changes be consistent from one protolang to another, but with gradual rooms for error, and to fit in line with an original list of words I made back in September that fit the scheme. If they do, I pat my back until I find one that doesn't, rinse and repeat until my hands have long gone pruney and shriveled, and sounds are the only things I can see.
I like my languages how I like my women: grammatically complex with various moods and tenses, a thin line between nouns and verbs, and dozens upon dozens of possible conjugations for every single verb.
User avatar
Chagen
runic
runic
Posts: 3338
Joined: 03 Sep 2011 05:14
Location: Texas

Re: What did you accomplish today?

Post by Chagen »

I've run into yet another problem with Pazmat ablaut. So, <a> and <e> normally have the grades a > ē > ey and e > ī > ay. However, arC and erC have completely different grades (arV and erV do not, they behave as normal):
vat-"look at" > vētana"I am looking at" > veytubbī"I begin to look at"
gard-"replace" > gōdana"I am replacing" > gaudubbī"I begin to replace"
geg-"believe" > gīgena"I am believing" > gaygubbī"I begin to believe"
wers-"defend" > wūsena"I am defending" > wōsubbī"I begin to defend"

However, other VrC verbs act identically to their VC counterparts, e.g:
cidh-"boil" > caudhitha"it is boiling" > cūdhubbū"it begins to boil"
kirt-"approach" > kaurtitha"he is approaching" > kūrtubbū"he begins to approach"

I have no real reason for this. There simply isn't a logical explanation for why -arC and -erC verbs would ablaut differently. And these ablauts are from the very first work I've done on this language. I just can't throw 'em away no matter how much I try. I could make the irC-, urC-, and -orC verbs irregular too, but that would add a bunch of unnecessary complexity.

The long grade of burst-"feel" is boyrst-(boyrstuna"I am feeling"). I'm honestly not sure whether this is super ugly or not. At the same time I have no idea what I'd want -oy- to become if I really didn't want it there. I do have a new rule in place which simplifies diphthongs in certain positions, namely before a geminate semivowel, and I could apply it here, but eeeeehhhh

And that doesn't even work because it doesn't make sense. If I choose to simplify diphthongs before an -rC- sequence, why would -ārC and -ērC ablaut differently? They would become -ērC and -īrC: no diphthongs to simplify!

(originally, the idea was: -ārC > -auC > ōC and -ērC > -euC > ūC, but this doesn't make sense because <au> normally stays all the way to Modern Pazmat and <eu> became <ō> post-Middle Pazmat. I GUESS I could say that, say, -ārC became -āuC and āu > ō, likewise with -ērC becoming -ēuC and ēu > ū. Deriving a set of long diphthongs and doing wacky things with them might be fun...)

Ugh. I think I need to just let it go and keep this as an ugly but unfixable relic of Pazmat's older days. -arC and -erC have irregular ablaut but -irC, -urC, and -orC don't I guess. Either that or I will make the other -VrC roots kinda irregular but in a way that doesn't really mesh with the original two. Also, this would probably result in those other -VrC being irregular in in only one grade (whichever would normally have a diphthong), which is...weird.

Sorry this post is so long and somewhat confusing, I make these mostly as a vent and to collect my musings since I come up with stuff better when venting to others for inexplicable reasons. That's why they tend to presume at least minor knowledge of Pazmat and go kinda fast.

I should start work on a new post in my Pazmat topic. I think I work better when explaining things like a teacher than coldly and dispassionately working with charts and forms.
Nūdenku waga honji ma naku honyasi ne ika-ika ichamase!
female-appearance=despite boy-voice=PAT hold boy-youth=TOP very be.cute-3PL
Honyasi zō honyasi ma naidasu.
boy-youth=AGT boy-youth=PAT love.romantically-3S
User avatar
Chagen
runic
runic
Posts: 3338
Joined: 03 Sep 2011 05:14
Location: Texas

Re: What did you accomplish today?

Post by Chagen »

Double post, but I just came up with this and I love it:

savurayyaśamāmi...
Since (he) will want to visit (soon)...

And if you need to make it even longer...

savurayyaśamallītāmi..
Given that (he) (soon) will want to visit (us) so strongly...

Actually I think I can stretch these participles even further:

būrstayurayyṛtallītāva
(That time) when I so badly had wanted to make (her) feel
Nūdenku waga honji ma naku honyasi ne ika-ika ichamase!
female-appearance=despite boy-voice=PAT hold boy-youth=TOP very be.cute-3PL
Honyasi zō honyasi ma naidasu.
boy-youth=AGT boy-youth=PAT love.romantically-3S
User avatar
qwed117
mongolian
mongolian
Posts: 4094
Joined: 20 Nov 2014 02:27

Re: What did you accomplish today?

Post by qwed117 »

Chagen wrote:Double post, but I just came up with this and I love it:

savurayyaśamāmi...
Since (he) will want to visit (soon)...

And if you need to make it even longer...

savurayyaśamallītāmi..
Given that (he) (soon) will want to visit (us) so strongly...

Actually I think I can stretch these participles even further:

būrstayurayyṛtallītāva
(That time) when I so badly had wanted to make (her) feel
I love the aesthetic of Pazmat. [<3] [<3] [<3] [<3] [<3]
Spoiler:
My minicity is [http://zyphrazia.myminicity.com/xml]Zyphrazia and [http://novland.myminicity.com/xml]Novland.

Minicity has fallen :(
The SqwedgePad
User avatar
loglorn
mayan
mayan
Posts: 1728
Joined: 17 Mar 2014 03:22

Re: What did you accomplish today?

Post by loglorn »

qwed117 wrote:
Chagen wrote:Double post, but I just came up with this and I love it:

savurayyaśamāmi...
Since (he) will want to visit (soon)...

And if you need to make it even longer...

savurayyaśamallītāmi..
Given that (he) (soon) will want to visit (us) so strongly...

Actually I think I can stretch these participles even further:

būrstayurayyṛtallītāva
(That time) when I so badly had wanted to make (her) feel
I love the aesthetic of Pazmat. [<3] [<3] [<3] [<3] [<3]
damn ninjas...
Diachronic Conlanging is the path to happiness, given time. [;)]

Gigxkpoyan Languages: CHÍFJAEŚÍ RETLA TLAPTHUV DÄLDLEN CJUŚËKNJU ṢATT

Other langs: Søsøzatli Kamëzet
User avatar
cedh
MVP
MVP
Posts: 386
Joined: 07 Sep 2011 22:25
Location: Tübingen, Germany
Contact:

Re: What did you accomplish today?

Post by cedh »

Chagen wrote:I've run into yet another problem with Pazmat ablaut. [...]
I would tackle this problem diachronically. Suppose all Vr ablaut variants were regular at one point:
CarC CaˑrC CaːrC
CerC CeˑrC CeːrC
CirC CiˑrC CiːrC
CorC CoˑrC CoːrC
CurC CuˑrC CuːrC

Then, *r is deleted under certain circumstances, affecting the preceding vowel in some way. Currently it looks like *r > u̯ /{a,e}{ˑ,ː}_C only (as you said yourself), followed by monophthongization of long diphthongs. If you have a clash with other, pre-existing diphthongs with u-offglides, what about having the first step of r-lenition be a different approximant or rhotic that doesn't exist in either Pazmat or Proto-Pazmat? I don't know which sounds would fall under this condition, but some possibilities would be [ɾ], [ɹ], [ð̞], [ɰ], [ʁ̞], or [ɦ]. Then you can lenite *r in all V{ˑ,ː}_C situations and have the resulting combinations of a vowel followed by this approximant develop in a distinct way compared to other vowel-approximant combinations.
User avatar
Inkcube-Revolver
cuneiform
cuneiform
Posts: 126
Joined: 05 Nov 2015 23:20
Location: Miami, FL

Re: What did you accomplish today?

Post by Inkcube-Revolver »

Chagen, this may have been something you've considered before, but you can also work back diachronically and see how those -ar- and -er- forms came about, like maybe the <r> was an infix in an older stage of Pazmat, and stuck itself to the verbs of a developing or already present ablaut system that changed the way the morphology works. If you go through this route, you can even have a related language where that didn't happen and see the consequences/effects of morphology in that instance.
It could have been more regular in Pazmat's early history, and with the addition of the Vr- forms also triggering ablaut as one unit, you can play with those results.
Here's an example just to get the idea across:

*wat- "look at" > vat-
*wētana? *wātana > vētana
*wētubbī (broken up as diphthong) > veytubbī

*wart- "stare, gaze at" > **vart-
*wartana > **vōtana
**wartubbī> *wōtubbī > **vautubbī

*gad- "return (something)"
*gādana > **gēdana
*gētubbī > **geytubbī

gard- "replace"
*gardana > gōdana
**gardubbī > *gōdubbī > gaudubbī

So the <r> could have acted as an intensifier in the early stages, and would have originally blocked lengthening of the preceding vowel or would even be deleted after certain vowels, lengthening them in the process, which would later be discarded with verbs like burst-, which may have looked differently (not sure how the full forms in Pazmat look). Or verbs like burst- indicate how the other grades should reflect the ones without rhotics, explaining why ir-, or-, and ur- may be the way they are.

I don't know how much this helps, or, again, if you've already considered this option.
I like my languages how I like my women: grammatically complex with various moods and tenses, a thin line between nouns and verbs, and dozens upon dozens of possible conjugations for every single verb.
User avatar
alynnidalar
greek
greek
Posts: 700
Joined: 17 Aug 2014 03:22
Location: Michigan, USA

Re: What did you accomplish today?

Post by alynnidalar »

Been having fun today! I was complaining to myself that I'd like to get back in the habit of adding new words to Tirina regularly, as well as more "families" of words descended from Old Tirina roots. I'm not a huge fan of compound nouns in Tirina (Azen is more comfortable with them, but Tirina just... isn't), but it occurred to me that noun-adjective compounds fit a lot better. (I already had a few in Modern Tirina, in fact!)

So, I started coming up with some meaningful compounds in Old Tirina that might end up in Modern Tirina. I was doing them by hand, then ended up copying my Old Tirina adjectives and nouns into Excel and using it to randomly stick two together. Of course not all of the combinations turn out, but I've come up with a decent-sized list of compound that I quite like! As an additional bonus, it's helping shape some of my ideas about the culture of the Old Tirina speakers--something I don't know all that much about yet.

I doubt all of these compounds will "stick", but here's a few I've created descendants of for Modern Tirina:

grān krami lit. "empty heart", meaning "someone who's not very smart" -> kanami "a flake, an airhead"
emgaos fetar lit. "rigid bone", meaning the bones of the legs and arms -> inosedarn "femur"
tael inog lit. "totally freckled" -> taelinin "light-skinned person"
User avatar
Inkcube-Revolver
cuneiform
cuneiform
Posts: 126
Joined: 05 Nov 2015 23:20
Location: Miami, FL

Re: What did you accomplish today?

Post by Inkcube-Revolver »

@alynnidalar,

What are the sound changes going on that produce those results? They look very interesting, especially the first set of compounds
I like my languages how I like my women: grammatically complex with various moods and tenses, a thin line between nouns and verbs, and dozens upon dozens of possible conjugations for every single verb.
User avatar
Axiem
sinic
sinic
Posts: 316
Joined: 10 Sep 2016 06:56

Re: What did you accomplish today?

Post by Axiem »

I reproduced a bunch of algebra work to figure out where I got my star's mass size from given the length of Mto's year...

I really ought to blog that one, just so I can find it again without working through the math a third time.
Conworld: Mto
:con: : Kuvian
User avatar
alynnidalar
greek
greek
Posts: 700
Joined: 17 Aug 2014 03:22
Location: Michigan, USA

Re: What did you accomplish today?

Post by alynnidalar »

Inkcube-Revolver wrote:@alynnidalar,

What are the sound changes going on that produce those results? They look very interesting, especially the first set of compounds
Ahh, my one weakness!

Honestly, the Tirina sound changes are not necessarily very "sensible", and they're a bit of a mess. But it is what it is.

For grān krami > kanami, the major things that are going on are the loss of /r/ after /k g/, the loss of vowel length, /g/ > /k/, and /ng/ > /ŋ/ > /n/.

/g/ totally disappears by the time of Modern Tirina, but it leaves some interesting stuff behind.
Locked