(Conlangs) Q&A Thread - Quick questions go here [2010-2020]

[holbuzvala]

Any ideas how infixes arise?

I know analogised metathesis is one way, but I wanted to find a way for vowel infixing.

[Pabappa]

Any ideas how infixes arise?

I know analogised metathesis is one way, but I wanted to find a way for vowel infixing.

make your verb (or whatever youre modifying) into a circumfix while the element in the middle remains dynamic. e.g. proto-Pumpkin lapala "to speak" vs bu lapala "spoke" ----> Early Pumpkin ta lapala "to speak" and ta bu lapala "spoke".... with obligatory verb prefix ta- on both.... ----> Late PUmpkin talapala "speak" and tabulapala "spoke". Of course yourll probably want some sound changes in there, its not likely the words would stay absolytely the same. French, Portgueuese, etc have done similar things with their verbs.

The above is the easiest and probably the most common way to do it. I have many -VC- infixes in my conlangs, which IM planning to derive from analogized reduplication, but for this to work, the language really needs to be CV or very close to it. that is, CVCV gets reanalyzed as C(VC)V.

[Pabappa]

Any ideas how infixes arise?

I know analogised metathesis is one way, but I wanted to find a way for vowel infixing.

make your verb (or whatever youre modifying) into a circumfix while the element in the middle remains dynamic. e.g. proto-Pumpkin lapala "to speak" vs bu lapala "spoke" ----> Early Pumpkin ta lapala "to speak" and ta bu lapala "spoke".... with obligatory verb prefix ta- on both.... ----> Late PUmpkin talapala "speak" and tabulapala "spoke". Of course yourll probably want some sound changes in there, its not likely the words would stay absolytely the same. French, Portgueuese, etc have done similar things with their verbs.

The above is the easiest and probably the most common way to do it. I have many -VC- infixes in my conlangs, which IM planning to derive from analogized reduplication, but for this to work, the language really needs to be CV or very close to it. that is, CVCV gets reanalyzed as C(VC)V.

[eldin raigmore]

Any ideas how infixes arise?

When it’s too cold or wet or hot to work on your motorbike’s engine under the shade tree you usually use,
your SO suggests: “Honey, you look so uncomfortable doing that outside in the weather! Why don’t you bring it indoors and fix it here?”

This is much rarer than suffixing, which is when s/he tells you to work on it later.

It’s even rarer than prefixing, when s/he tells you you should have repaired it yesterday.

Then there’s ambifixing (s/he has to call the ambulance because things didn’t go right);
circumfixing (s/he has a lawn party outside in the yard and you have to walk around the guests and refreshments to make your repairs);
and suprafixing (not quite sure what that is).

—————

HTH!

[elemtilas]

When it’s too cold or wet or hot to work on your motorbike’s engine under the shade tree you usually use,
your SO suggests: “Honey, you look so uncomfortable doing that outside in the weather! Why don’t you bring it indoors and fix it here?”

This is much rarer than suffixing, which is when s/he tells you to work on it later.

It’s even rarer than prefixing, when s/he tells you you should have repaired it yesterday.

Then there’s ambifixing (s/he has to call the ambulance because things didn’t go right);
circumfixing (s/he has a lawn party outside in the yard and you have to walk around the guests and refreshments to make your repairs);
and suprafixing (not quite sure what that is).

[cedh]

Any ideas how infixes arise?

I know analogised metathesis is one way, but I wanted to find a way for vowel infixing.

For a vowel infix between two consonants, epenthesis based on phonotactic restrictions (e.g. breaking up difficult clusters) or prosodic considerations (e.g. to prevent two stressed syllables in a row) are realistic sources. Originally these vowels will probably only occur in words of a certain shape, but if these words are frequent enough, analogy can turn the epenthetic vowels into morphologized infixes.

For a vowel infix adjacent to another vowel, I would suggest a form of umlaut that does not result in wholesale fronting/rounding/whatever, but in diphthongization, followed by deletion of the triggering vowel or glide. For instance:
*kas > kas
*kas-i > *kais-i > kais
*kas-u > *kaus-u > kaus
*kas-ja > *kais-ja > kais-a

[yangfiretiger121]

Currently, Sjialɱa [çi̥ˈɔ̥ᵝŋ̊.m̪̊ɔ̥ᵝ] fixes stress to a word's penult mora. Can a pitch accent system coexist with that?

[Vlürch]

Currently, Sjialɱa [çi̥ˈɔ̥ᵝŋ̊.m̪̊ɔ̥ᵝ] fixes stress to a word's penult mora. Can a pitch accent system coexist with that?

I don't see why it couldn't, and AFAIK it does in (at least some of) the Yugoslavic languages?

[Omzinesý]

Currently, Sjialɱa [çi̥ˈɔ̥ᵝŋ̊.m̪̊ɔ̥ᵝ] fixes stress to a word's penult mora. Can a pitch accent system coexist with that?

I don't see why it couldn't, and AFAIK it does in (at least some of) the Yugoslavic languages?

Moraic languages can well have pitch accent. Take Japanese as an example.
What does "on the penult mora" mean in your lang? Are morae vowels? Does it appear on the moraic element or the syllable that includes the mora?

[yangfiretiger121]

Currently, Sjialɱa [çi̥ˈɔ̥ᵝŋ̊.m̪̊ɔ̥ᵝ] fixes stress to a word's penult mora. Can a pitch accent system coexist with that?

I don't see why it couldn't, and AFAIK it does in (at least some of) the Yugoslavic languages?

Moraic languages can well have pitch accent. Take Japanese as an example.
What does "on the penult mora" mean in your lang? Are morae vowels? Does it appear on the moraic element or the syllable that includes the mora?

I'm aware that Japanese has a pitch accent. I'm just trying to see if fixed stress can coexist with pitch accent and how so, if it can. The long vowels Romanized with acute accents (<á é í>) because they are already seen to have higher pitch. Thus, I'm answering my own question in a way. But, while I started out trying to see if something similar to Japanese with downsteps (ꜜ) or upsteps (ꜛ) can coexist with fixed stress, I'm, now, trying to see if a tonal system can coexist with fixed stress. Could the following system coexist with fixed stress: ˦ as tone one for long vowels and general stress, ˥ as tone two for stressed high vowels, and ˧˥ as tone three for high vowels before stressed high vowels?

The phones marked with an asterisk (*) below include numerous allophones.

The work-in-progress post refers to a mora as 1.) (C)V(n) with [p*, p͡ʔ, t*, t͡p, k*, q͡ʡ, s*, ç, ʎ̝̊, ʨ, j̊, ɰ̃̊, ɾ̥~ɬ, ʘ, ʘ̃] restricted to onsets; 2.) (n) being [m̥ m̪̊ n̥ n̥͡m ç̃ ŋ̊] and represented as L in broad moraic transcription; and long vowels [ɔ̥ᵝː ə̥ː ɨ̥ː] being two morae with the second represented as V in moraic transcription.

Examples (without up/downsteps)
Sjialɱa /çɨ̥.ɔ̥ᵝˈL.m̪̊ɔ̥ᵝ/ [çɨ̥.ɔ̥ᵝˈŋ̊.m̪̊ɔ̥ᵝ] [çi̥ˈɔ̥ᵝŋ̊.m̪̊ɔ̥ᵝ] (current)/Sjiaĺɱa /çɨ̥.ɔ̥ᵝˈL˦.m̪̊ɔ̥ᵝ/ [çɨ̥.ɔ̥ᵝˈŋ̊˦.m̪̊ɔ̥ᵝ] [çi̥ˈɔ̥ᵝŋ̊˦.m̪̊ɔ̥ᵝ] (proposed)
W̃áí /ʘ̃ɔ̥́ᵝ.Vˈɨ̥́.V/ [ʘ̃ɔ̥́ᵝːˈɨ̥́ː] (current)/W̃a᷄ï /ʘ̃ɔ̥˧˥ᵝ.Vˈɨ̥˥.V/ [ʘ̃ɔ̥˧˥ᵝːˈɨ̥˥ː] (proposed)

[Omzinesý]

Currently, Sjialɱa [çi̥ˈɔ̥ᵝŋ̊.m̪̊ɔ̥ᵝ] fixes stress to a word's penult mora. Can a pitch accent system coexist with that?

I don't see why it couldn't, and AFAIK it does in (at least some of) the Yugoslavic languages?

Moraic languages can well have pitch accent. Take Japanese as an example.
What does "on the penult mora" mean in your lang? Are morae vowels? Does it appear on the moraic element or the syllable that includes the mora?

I'm aware that Japanese has a pitch accent. I'm just trying to see if fixed stress can coexist with pitch accent and how so, if it can. The long vowels Romanized with acute accents (<á é í>) because they are already seen to have higher pitch. Thus, I'm answering my own question in a way. But, while I started out trying to see if something similar to Japanese with downsteps (ꜜ) or upsteps (ꜛ) can coexist with fixed stress, I'm, now, trying to see if a tonal system can coexist with fixed stress. Could the following system coexist with fixed stress: ˦ as tone one for long vowels and general stress, ˥ as tone two for stressed high vowels, and ˧˥ as tone three for high vowels before stressed high vowels?

The phones marked with an asterisk (*) below include numerous allophones.

The work-in-progress post refers to a mora as 1.) (C)V(n) with [p*, p͡ʔ, t*, t͡p, k*, q͡ʡ, s*, ç, ʎ̝̊, ʨ, j̊, ɰ̃̊, ɾ̥~ɬ, ʘ, ʘ̃] restricted to onsets; 2.) (n) being [m̥ m̪̊ n̥ n̥͡m ç̃ ŋ̊] and represented as L in broad moraic transcription; and long vowels [ɔ̥ᵝː ə̥ː ɨ̥ː] being two morae with the second represented as V in moraic transcription.

Examples (without up/downsteps)
Sjialɱa /çɨ̥.ɔ̥ᵝˈL.m̪̊ɔ̥ᵝ/ [çɨ̥.ɔ̥ᵝˈŋ̊.m̪̊ɔ̥ᵝ] [çi̥ˈɔ̥ᵝŋ̊.m̪̊ɔ̥ᵝ] (current)/Sjiaĺɱa /çɨ̥.ɔ̥ᵝˈL˦.m̪̊ɔ̥ᵝ/ [çɨ̥.ɔ̥ᵝˈŋ̊˦.m̪̊ɔ̥ᵝ] [çi̥ˈɔ̥ᵝŋ̊˦.m̪̊ɔ̥ᵝ] (proposed)
W̃áí /ʘ̃ɔ̥́ᵝ.Vˈɨ̥́.V/ [ʘ̃ɔ̥́ᵝːˈɨ̥́ː] (current)/W̃a᷄ï /ʘ̃ɔ̥˧˥ᵝ.Vˈɨ̥˥.V/ [ʘ̃ɔ̥˧˥ᵝːˈɨ̥˥ː] (proposed)

If you are asking, if fixed stress and pitch accent can coappear, absolutely yes. I have been, at last, learning Swedish pitch accents. It's a very simple example https://learningswedish.se/courses/1/pa ... ord-accent

Pitch is basically just one means the stress can realize, beside strength and length. It just is more fluid so that a high pitch can appear on several syllables simultaneously while stress always appears on one. If there are more peaks, you just have to define what "stress" means as a term to describe the language.

[Zekoslav]

If I understood this well, stress would be fixed on the penultimate mora, and pitch would be independent of it? I can imagine this, it exists in Latvian, which is an interesting mixture between pitch accent and tonal language: there's stress on the first syllable of the word, and a distinction of three tones on all syllables.

As for "Yugoslavic languages"... I don't think there's anything exactly like that. Macedonian has fixed stress on the antepenultimate syllable but no pitch, and some Kajkavian dialects near the Hungarian border have a limit on pitch accent position but within that limit it's free, like Ancient Greek (albeit it's two, rather than three mora). Everywhere, pitch is bound to stress.

[clawgrip]

I'm just trying to see if fixed stress can coexist with pitch accent and how so, if it can.

If you are asking, if fixed stress and pitch accent can coappear, absolutely yes. I have been, at last, learning Swedish pitch accents. It's a very simple example https://learningswedish.se/courses/1/pa ... ord-accent

Pitch is basically just one means the stress can realize, beside strength and length. It just is more fluid so that a high pitch can appear on several syllables simultaneously while stress always appears on one. If there are more peaks, you just have to define what "stress" means as a term to describe the language.

That's really the main point...stress does not really have a fixed definition. In English, stress is a combination of vowel length, vowel quality, loudness, and pitch. If you use pitch for a pitch accent system, you could still have other features that represent a non-pitch stress (much like secondary stress in English).

[Nloki]

Maybe I'm contradicting myself right now according to what I previously said, but the main problems don't concern grammar. I kind of know what I want my conlang to be like in terms of grammar, but I want it to be as a prioristic as possible

I can definitely understand wanting your conlangs to be "as a prioristic as possible", and that's why I've sometimes changed words or morphemes or bits of grammar in mine to ones I didn't like as much when I realised the ones I liked more were "too similar" to things in natlangs or other conlangs, but as hard as it is to force yourself not to do that after the realisation has hit you, why should that be the solution? I mean, where do you draw the line? It's the same in music: sometimes I come up with a great riff or some really cool lyrics, and then, sometimes only after finishing the song, I realise it's a subconscious rip-off of something. However, this is where you have to ask yourself: is anything truly original? If not, what counts as original enough?

The way I see it, while there can still be innovation in every creative field, it's becoming increasingly difficult to invent (or discover) new things or even combine things in an original way because so much has already been made and there are more and more people in every creative field... but more importantly, you're more and more likely to know those other people have already done what you thought was so original. I'll again use an analogy with music: when I started out making stuff that combines cybergrind/goregrind, synthcore/deathcore and deathstep/brostep, I didn't know of any other artists doing the exact same thing and it was even called unique and original by some people... but there were artists doing it years before me (literally before those genres became well-defined), and better than me. I could swear I tried my hardest to find artists who'd done it before, but I couldn't find any. Now it takes two seconds on Google to find them. Why? I have no idea. Maybe I just sucked at googling back then, or maybe Google's algorithms have gotten better. But that doesn't matter, it's good that I didn't know others had done it because if I had known, I may not have done it myself.

It's the same with conlanging: there are different types of languages you can mix the features of, but at the end of the day all of them have probably already been mixed one way or another long before you thought about it... but that doesn't mean you shouldn't do it or that your conlang won't be unique. You could even make the most run-of-the-mill Romlang but it'd still be unique, as it would be your take on what a run-of-the-mill Romlang is; it would be compared to the countless other run-of-the-mill Romlangs out there and people would probably say "it's too similar to Romlang X and Romlang Y", but there would be at least one feature that differed from all the other Romlangs. That's comparable to different artists/bands in the same genre.

Let's switch the analogy to black metal for it to be more accurate. Black metal has been done to death and back, but does that stop 666 new black metal albums from being released every single day? Of course not, because people who like black metal don't care that 99% of it is technically speaking the exact same thing; in fact, there are people who think that doing anything new in black metal is bad. And the argument makes sense: there's a working formula for black metal, so why do anything differently? Of course that doesn't stop people from making black metal that isn't like the other 99% of black metal, and I think that's a good thing. But sometimes I like to listen to the most run-of-the-mill black metal out there, and I'm glad people still make it because there's always something different when different artists/bands do it, no matter how minor the difference is.

Again, the same applies to conlangs. Again, there are countless Romlangs, some of them practically indistinguishable from one another, but each of them still has something that differentiates them from all the others. If that can be accomplished with Romlangs, you really shouldn't have to worry about your a priori conlang being too similar to other languages. Nothing is ever 100% original, but literally everything you do will always have your own "touch" to it.

I don't know how to manage to coin basic roots or stems. Seriously, not even derivation yet. Coining stems. And that's supposed to be somehow the easy part, but it's not (at least not for me).

That's actually one of my biggest problems (or sources of procrastination) too, so you're not alone. Have you tried just making a long list of meaningless mono- and bisyllabic words and then assigning them meanings from a randomly generated long list of words? Some may call it cheating, but in metal that's also what some people say about drum programming, inhaled high screams and tons of other things. Point is, who cares as long as it works? Anyone who cares isn't someone you should take seriously on that issue if you disagree with what they're saying.

And I think that's what I lack: phonoaesthetic taste! (or however it could be called like).

I doubt you lack it. Isn't it more likely that you just have more than one taste? That's perfectly fine (and normal), anyway. Who really only likes Tolkienesque fantasylangs and whatnot? I'm sure they exist, just like people who only listen to the most formulaic black metal exist (and, actually, I'm pretty sure those two groups have some overlap), but most of us are into more than one thing even if we may have a "passion" for one thing above all others. That's why sticking to just one conlang (or one group of conlangs that share any kind of connection) may actually reduce your dedication to it and your creativity, since you're more likely to get indecisive over what features you want to include or exclude based on your current mood.

Phonoaesthetic taste isn't a monolith, just like musical taste. On some days you might love /d͡ɮˤ/ even if on most days you'd go "ew wtf get this shit outta my conlang!", just like on some days you might love Katy Perry's Firework even if you'd go "ew wtf get this shit outta my head!" if you randomly got it as an earworm.

I cannot discern among lot's of different roots that could be assigned to an only meaning and I just use logical strategies to choose whether a word should be included or not. For example; nēr for "man". But then I realise Quenya already uses nér (ner-) for "man"! So I start trying to coin some monosyllabic words that could fit the purpose and I find out - I don't like any of them! (Happens always the same way) so I give up for a while just to try again and achieve doing nothing.

I think that's normal if you think about conlanging in too "strategic" a way. Natlangs are often not logical and often have words that just sound wrong referring to the things they actually mean, etc. and even conlangs that strive to be logical and neat often end up having illogical with inconsistent sound symbolism, etc. Maybe try to assign words the most unfitting meanings and see if that helps you shuffle them around in a way that you like more? That's assuming you come up with meaningless roots first and then assign meanings, not the other way around. If you don't do it that way, maybe try that?

I also don't want to be recognized in any way for conlanging (since it's one of my hobbies which I suck the most at)

You mean you don't want people to be able to connect your accounts on conlanging forums to your accounts elsewhere or in real life or whatever?

It's always the same way; the more I learn or acknowledge, the less I success on something. It's sort of upside down but it's what most of my hobbies end up being: an endless hell.

That's how I am with music in some ways, as in the better I get the longer it takes to finish anything and I feel like I can't just do some weird shit and not really care about the quality. Unfortunately I don't know what an actual solution would be, but maybe it could be to just go back to doing things the way you used to do them even if it's hard? I've been trying that lately but can't yet say whether it works or not... but anyway, you implied you also have other hobbies besides conlanging. Maybe you could try to mix conlanging into your hobbies in some way, would that make it easier? I don't mean like randomly starting to yell in a conlang to your friends with a dead ass serious face if you're playing football or whatever, that'd be awkward as hell, but you know what I mean.

Thank you again for your extensive post Vlürch. I actually didn't have much time to browse on the CBB because, despite everything I typed before... I've been able to conlang again!
In fact, in the couple last days I've been able to surpass my previous works of months! (Or at least that's what I think). Almost literally as if a light bulb turned on in my head!
Metaphors apart; I certainly appreciate your post, and much of others’ posts answering the matter. I'll see if I can post Nlokian 5 today. Thank you so much!!

[yangfiretiger121]

Thanks for the help, guys.

At one time, Sjialfa had voiceless rhotic, lateral, palatal, and palatal lateral flaps [ɾ̥, ɺ̥, c̆, ʎ̮̊]. Are the changes to the rhotic, lateral, and palatal lateral ones stated below plausible?

{ɾ̥ ɺ̥ → ɬ → ʎ̝̊ ()}
[ʎ̮̊ → ʎ̝̊ () → ʟ̝ ()]

[Ælfwine]

Thanks for the help, guys.

At one time, Sjialfa had voiceless rhotic, lateral, palatal, and palatal lateral flaps [ɾ̥, ɺ̥, c̆, ʎ̮̊]. Are the changes to the rhotic, lateral, and palatal lateral ones stated below plausible?

{ɾ̥ ɺ̥ → ɬ → ʎ̝̊ ()}
[ʎ̮̊ → ʎ̝̊ () → ʟ̝ ()]

The first is fairly plausible. The second I find less plausible, but only because [ʟ̝] is sort of rare. Also, I assume it retains voicelessness or no?

---

I have a fairly typical set of plosives, /p b t d k (g)/. Only /g/ is rare but becoming more common due to loanwords.

In the initial position, voiced plosives are devoiced, but a phonemic contrast still exists due to aspiration. In medial position, voiced plosives are fully voiced, and aspiration is usually retained on voiceless plosives. In final position, voiced plosives are devoiced again, and voiceless plosives are unaspirated, neutralizing the distinction between the two.

So, would my plosive consonants in the absolute final position be phonemically "voiceless" /p t k/ or "voiced" /b d g/? This development is similar to Gothic's devoicing of fricatives word-finally, but except for /ɣ/ they are usually analyzed as "voiceless."

[yangfiretiger121]

Thanks for the help, guys.

At one time, Sjialfa had voiceless rhotic, lateral, palatal, and palatal lateral flaps [ɾ̥, ɺ̥, c̆, ʎ̮̊]. Are the changes to the rhotic, lateral, and palatal lateral ones stated below plausible?

{ɾ̥ ɺ̥ → ɬ → ʎ̝̊ ()}
[ʎ̮̊ → ʎ̝̊ () → ʟ̝ ()]

The first is fairly plausible. The second I find less plausible, but only because [ʟ̝] is sort of rare. Also, I assume it retains voicelessness or no?

Got it. Yes, the palatal and velar lateral fricatives are/would be voiceless.

If I make the changes, would retention of <r> or switch to <l> be more likely?

[eldin raigmore]

... the palatial and velar lateral fricatives ...

Palatial sounds are made with the tongue pressed against the palace.

I assume autocorrect “corrected” palatal to palatial?

[yangfiretiger121]

... the palatial and velar lateral fricatives ...

Palatial sounds are made with the tongue pressed against the palace.

I assume autocorrect “corrected” palatal to palatial?

Yep. Damn phone!

[yangfiretiger121]

Thanks for the help, guys.

At one time, Sjialfa had voiceless rhotic, lateral, palatal, and palatal lateral flaps [ɾ̥, ɺ̥, c̆, ʎ̮̊]. Are the changes to the rhotic, lateral, and palatal lateral ones stated below plausible?

{ɾ̥ ɺ̥ → ɬ → ʎ̝̊ ()}
[ʎ̮̊ → ʎ̝̊ () → ʟ̝ ()]

The first is fairly plausible. The second I find less plausible, but only because [ʟ̝] is sort of rare. Also, I assume it retains voicelessness or no?

Got it. Yes, the palatal and velar lateral fricatives are/would be voiceless.

If I make the changes, would retention of <r> or switch to <l> be more likely?

For the record, I'm only aware of the IPA extensions for the palatal and velar lateral fricatives [ʎ̝̊ (), ʎ̝ (̬), ʟ̝̊ (), ʟ̝ (̬)] being in the Gentium, Charis, and Doulos fonts at the moment.