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

[Dormouse559]

For a naturalistic conlang, is it unrealistic to have separate personal endings on verbs for active, middle, and passive voice?

Not at all. The reflexive-pronoun origin for middle voice makes a lot of sense.

[Creyeditor]

For a naturalistic conlang, is it unrealistic to have separate personal endings on verbs for active, middle, and passive voice?

Here is what I have for Denkurian at the moment:

Spoiler:

My idea is that the middle voice evolved from an old reflexive pronoun.

IIUC, you just have the active personal endings/suffixes plus <-eks> in the middle voice and -<ast> in the passive. That looks very naturalistic.

[eldin raigmore]

My idea is that the middle voice evolved from an old reflexive pronoun.

That last part is especially realistic & naturalistic.

[felipesnark]

For a naturalistic conlang, is it unrealistic to have separate personal endings on verbs for active, middle, and passive voice?

Not at all. The reflexive-pronoun origin for middle voice makes a lot of sense.

IIUC, you just have the active personal endings/suffixes plus <-eks> in the middle voice and -<ast> in the passive. That looks very naturalistic.

That last part is especially realistic & naturalistic.

You are correct about the analysis of affixes, Creyeditor. In a future version of that documentation I will just list the middle and passive voice affixes separately. Thanks for the feedback, guys!

[felipesnark]

So a related question: Where do you we think the passive personal endings in some IE languages arose from? Like in the case of Ancient Greek and Classical Latin?

[eldin raigmore]

So a related question: Where do you we think the passive personal endings in some IE languages arose from? Like in the case of Ancient Greek and Classical Latin?

Search for “the middle voice suzanne kemmer”.
According to her, middle-voice markers usually arise as reduced forms of reflexive-voice markers (or possibly incorporated reflexive-person pronouns), as a way to distinguish those situations which are/were naturally reflexive*, from those that just happened to be reflexive just this once (for which the full, unreduced reflexive marker, would persist).

*A situation is “naturally” reflexive, if the addressee would just assume it was reflexive unless the speaker indicates otherwise; and the speaker knows the addressee would so assume, and counts on that.

The reasons the addressee would assume reflexivity might be language-specific or culture-specific, or might be more universal.

If I say “I bathed” or “I shaved” or “I dressed”, you assume I was the object as well as the subject. In our culture those verbs are naturally reflexive.

Not so if I say “the octopus ate”; there you assume it ate a crab or something. If I want you to understand that it exhibits symptoms of the neurological disease called autophagy, which sometimes afflicts Octopoda, I need to explicitly say “the octopus ate itself”.

——————————

Kenner goes on to list 11 types of situations (in addition to “naturally reflexive”) that middle-voice is frequently expanded to cover.
Most situations marked as “introversive” in languages that mark version but not middle voice, get marked as “middle voice” in languages that mark middle voice but not version. For example, “grow”, in “the calf grew to be a bull” as opposed to “the farmer grows tomatoes”.
But another of Kemmer’s examples is the “naturally reciprocal”. In most cultures I know about, for instance, if I say “we hugged”, my addressees would just naturally assume I meant “we hugged each other”. Kemmer says that in languages with a middle voice, the middle-voice’s job is often expanded to include such naturally-reciprocal situations.

—————

In Classical I.-E. languages, middle voice meant “the agent or their interests are affected”.
Kemmer gives evidence that those markings, too, arose as mentioned above.

====================

Maybe this post relies too much on just one author. But I’m convinced, at least presumptively.

[yangfiretiger121]

Once upon a time, Aʻatun only had the alveolar, retroflex (/ɽ/ due to the flap's absence), and uvular trills. However, the voiceless bilabial, voiced labiodental, alveolar nasal, retroflex nasal, and uvular nasal trills were added through, for the lack of better terminology, "sound economy." Could a palatal fricative trill (/ʝ̝/) have arisen through the same process? For example, depending on the environment, [mr nr] became /r̃ ɽ̃ ʀ̃/.

[Lambuzhao]

So a related question: Where do you we think the passive personal endings in some IE languages arose from? Like in the case of Ancient Greek and Classical Latin?

Well, there is a reconstructable set of verbal endings for the PIE Medio-Passive.
The primary endings mostly bore a telltale /r/ at the end.

The /r/ clearly survived into (and other IE descendants), but was dropped in .

preserves the 3SG, 1PL, 2PL, and 3PL desinences for the PRS.PASS .
preserves the 1SG, 3SG, 3PL desinences for the PRS.PASS.

The Proto */mai/ PRS.PASS.1SG and */sai/ PRS.PASS.2SG desinences were innovations.

The Old */ris/ PRS.PASS.2SG and */menei/ PRS.PASS.2PL desinences were innovations.

The original PRS.PASS.PTCP ending */mnos/ survived intact into Ancient . It did not remain as a productive morpheme in Latin verb conjugation, but there are a number of nouns and adjectives that clearly come from former Proto-Italic PRS.PASS.PTCPs, such as femina, alumnus, calumnia, and possibly *vortumnus, *volumna and *autumnus. It has also been suggested that the aforementioned innovated PRS.PASS.2PL ending in could be simply the PRS.PASS.PTCP.PL form.

qv-
https://en.wikipedia.org/wiki/Proto-Ind ... ve_endings

[KaiTheHomoSapien]

It has also been suggested that the aforementioned innovated PRS.PASS.2PL ending in could be simply the PRS.PASS.PTCP.PL form.

Sihler says about "-mini": "One of the enigmas of the classical scholarship, ranking in mystery with what name Achilles took when he hid among the women, is the explanation of Latin -mini...the form looks like nothing at all, or worse, bears a distracting resemblance to what are certainly unrelated elements in cognate languages."

I remember noticing that -mini resembled the Greek middle participle form, but why that would persist in the 2nd plural beyond me. Guess we'll never know on that one

[Lambuzhao]

Indeed. That's quite a distraction!

---

Besides , there are a number of /r/ passive cogeners among the IE brood:

The r-passive (mediopassive voice) was initially thought to be an innovation restricted to Italo-Celtic until it was found to be a retained archaism shared with Hittite, Tocharian, and possibly the Phrygian language.

https://en.wikipedia.org/wiki/Italo-Celtic


, Avestan/Old Persian and Sanskrit also have MedioPassive verb conjugations (!!), though it is obsolescent and non-productive in Armenian.

[Lambuzhao]

For a naturalistic conlang, is it unrealistic to have separate personal endings on verbs for active, middle, and passive voice?

Not at all. The reflexive-pronoun origin for middle voice makes a lot of sense.

That, and, by way of another example of Passive construction, Coptic Egyptian often just used the ACT.3PL form of the verb with the Objective Pronominal endings tacked on.

[k1234567890y]

For a naturalistic conlang, is it unrealistic to have separate personal endings on verbs for active, middle, and passive voice?

Here is what I have for Denkurian at the moment:

Spoiler:

My idea is that the middle voice evolved from an old reflexive pronoun.

I guess this is ok? PIE had the separation of active and mediopassive voice among its verbal-person conjugations

[Nachtuil]

In languages with a system of polypersonal agreement on the verb, is that marking ever lacking for possession verbs? Especially if the possession verb is irregular?

[k1234567890y]

In languages with a system of polypersonal agreement on the verb, is that marking ever lacking for possession verbs? Especially if the possession verb is irregular?

uncertain, maybe unlikely? I guess either they still exist on possession verbs or such verbs have become highly irregular...or maybe you can have this as the semantic development to have a "no-mark possessive verb"? demonstrative > "to be(copula)" > "to be(locational and existential), to have", but uncertain

btw I tend to use the existential verbs for possession verbs too, as it is not uncommon for natlangs not to have a specific verb for the meaning "to have"

[brblues]

Hi guys, first post! I wanted to ask here whether my phoneme inventory, and the respective romanizations, seem naturalistic / workable. I also posted this to reddit last night, just as full disclosure, let me know if such crossposting is not well liked here.

So far I got two romanizations with very slight differences, the phonemic transliteration and the simplified transliteration. The purpose of the latter is to make typing on an English-layout keyboard easier by cutting down somewhat on the diacritical marks. In cases where both are provided, it is always the phonemic transliteration that comes first, followed by the simplified transliteration after a semicolon.





Still on the fence about the yellow consonants, and would probably only include them as allophones. I don't have any systematic formulation for phonotactics yet, though I know what it's supposed to look like and got vocab; I'm just extremely new at phonology and find it hard to generalize the rules.

[zyma]

Welcome to the board!

I don't think there should be an issue with "crossposting". It's never come up before, to the best of my knowledge, but I don't think that many of us are also on the conlang Reddit page.

What do you mean by "workable", exactly?

The tables are a little hard to read because they're so small (and clicking on them takes me to a website that I would apparently have to sign up for), but things seem good for the most part.

A few things that stand out, if you're going for naturalism. These are just some thoughts; you don't have to change anything you don't want to.

• /p͡f/ is rare, especially as a phoneme, but German has it, so I don't think that should be a problem.
• Having /ʀ/ and /ʁ/ contrast is also rare, but, at least according to Wikipedia, they're apparently separate phonemes in Luxembourgish.
• It seems strange to have /ɭ/ and /ʟ/ without /l/. Does anyone know of any natural languages that do something like this?
• I would add /s/, personally.
• It sticks out to me that, not counting /ʒ/ because it's highlighted in yellow, /v/ and /ʁ/ are the only voiced fricatives, and that /ʁ/ doesn't have a voiceless counterpart (unless that's supposed to be /x/ or /h/, despite those being at different POAs), while /ʃ/ doesn't have a voiced counterpart (again, not counting /ʒ/). Does anyone know of an inventory like this in a natural language?

Also, does your language have front vs. back vowel harmony? I can't quite tell from your description.

[brblues]

Thanks so much for your post, shimobaatar!

You have indeed uncovered a few mistakes and oversights on my part. By "workable" I just meant whether it would be possible to work with, but I guess that's a question only I myself can answer, so let's just stick with (at least some semblance of) naturalism for now!

First of all, sorry about the size of the pictures. I somewhat naively assumed I could easily import tables from OneNote or Excel into a board, and when that proved impossible (if somebody knows a way, please let me know!), I just went with a not-too-great screenshot and the first image hosting page I could find.

/p͡f/ is rare, especially as a phoneme, but German has it, so I don't think that should be a problem.

I'm German, so really no problem there - and I may be biased, but still surprised that it seems to be a rare phoneme!

Having /ʀ/ and /ʁ/ contrast is also rare, but, at least according to Wikipedia, they're apparently separate phonemes in Luxembourgish.

That was a mistake, I actually wanted to have "ɹ" (instead of ʁ) . That should be better?

It seems strange to have /ɭ/ and /ʟ/ without /l/.

It is indeed strange and another mistake, I wanted /l/ instead of /ɭ/. So basically a phonemic contrast between dark and light l.

I would add /s/, personally.

Definitely, and I must have deleted that before, weird. Right before posting I noticed that I had accidentally deleted the alveolar plosives and nasal, and even checked again to see if nothing else had gone amiss, but I did miss that one. So thanks for uncovering all these mistakes!

It sticks out to me that, not counting /ʒ/ because it's highlighted in yellow, /v/ and /ʁ/ are the only voiced fricatives, and that /ʁ/ doesn't have a voiceless counterpart (unless that's supposed to be /x/ or /h/, despite those being at different POAs), while /ʃ/ doesn't have a voiced counterpart (again, not counting /ʒ/). Does anyone know of an inventory like this in a natural language?

I noticed there are quite a couple of isolated ones, which is why I thought to include /ʒ/ despite not being a fan of it. So I'd also be happy to hear whether it is somewhat naturalistic to have an inventory like that.

And yeah, the language does have a (somewhat odd?) form of front vs back vowel harmony. I deliberately left this out as it's a bit tough to explain what I got in mind, but I hope this makes some sense: Some morphology vowel is based on the relationship between the vowel phoneme in the primary (stress-bearing) syllable and secondary syllable (following the primary syllable) of the two-syllable root. Vowel phonemes, with the exception of ə <e>, which is regarded as neutral, are divided into four front /i y e ø/ and three back vowels /u o a/. The "melody" of a noun/verb can be described as:

- Neutral: both primary and secondary syllable either front (verbs) or back (nouns)
- Falling: primary syllable front, secondary syllable back
- Rising: primary syllable back, secondary syllable front

There is then some complications when some things are combined, but I'm gonna spare you that for now!

Thanks again for helping. I will post a corrected version of the inventory as soon as I get to it.

Thank you as well for the welcome - I hope to become a more or less active member here and look forward to it :)

[zyma]

First of all, sorry about the size of the pictures. I somewhat naively assumed I could easily import tables from OneNote or Excel into a board, and when that proved impossible (if somebody knows a way, please let me know!), I just went with a not-too-great screenshot and the first image hosting page I could find.

No worries! I wish I could help, but I'm not so great with that kind of thing myself.

/p͡f/ is rare, especially as a phoneme, but German has it, so I don't think that should be a problem.

I'm German, so really no problem there - and I may be biased, but still surprised that it seems to be a rare phoneme!

According to Wikipedia, at least, the only language variety that has it as a phoneme, apart from Standard German and close relatives, is a dialect of Tsonga, in southern Africa. The German phoneme is especially unusual, actually, because the stop component is supposedly bilabial, while the fricative component is labiodental.

That isn't to say you shouldn't have it in your language, though!

That was a mistake, I actually wanted to have "ɹ". That should be better?

That would definitely contrast more with whichever uvular sound you're keeping.

It is indeed strange and another mistake, I wanted /l/ instead of /ɭ/. So basically a phonemic contrast between dark and light l.

Ah, that makes sense. In many languages, though, especially in Europe, the "dark L" is actually the velarized /ɫ/ rather than the truly velar /ʟ/. There are some languages that have /ʟ/, though, so that shouldn't be a problem.

I noticed there are quite a couple of isolated ones, which is why I thought to include /ʒ/ despite not being a fan of it. So I'd also be happy to hear whether it is somewhat naturalistic to have an inventory like that.

If you're not a fan of /ʒ/, then I'd recommend leaving it out.

And yeah, the language does have a (somewhat odd?) form of front vs back vowel harmony. I deliberately left this out as it's a bit tough to explain what I got in mind, but I hope this makes some sense: Some morphology vowel is based on the relationship between the vowel phoneme in the primary (stress-bearing) syllable and secondary syllable (following the primary syllable) of the two-syllable root. Vowel phonemes, with the exception of ə <e>, which is regarded as neutral, are divided into four front /i y e ø/ and three back vowels /u o a/. The "melody" of a noun/verb can be described as:

- Neutral: both primary and secondary syllable either front (verbs) or back (nouns)
- Falling: primary syllable front, secondary syllable back
- Rising: primary syllable back, secondary syllable front

Oh, interesting. So it's not exactly like the systems of vowel harmony in, for example, Finnish or Turkish?

Thanks again for helping. I will post a corrected version of the inventory as soon as I get to it.

Thank you as well for the welcome - I hope to become a more or less active member here and look forward to it :)

No problem at all! I hope you enjoy the CBB!

[brblues]

After implementing the corrections, I now hopefully haven't screwed up the chart even more - this is the revised consonant grid (hidden under spoiler due to size):

Spoiler:

Oh, interesting. So it's not exactly like the systems of vowel harmony in, for example, Finnish or Turkish?

No not really, although I do like Turkish in particular, and the inventory of vowel phonemes is extremely similar too, except I have no dotless Iı /ɯ/. I wanted to do something different though. So while words with what I currentyl call the "neutral melody" do observe front-back harmony as they would in Turkish for example (I'm aware that there's some exceptions!), as soon as they get e.g. declined from absolutive to ergative case (nouns) or from the indicative mood into a different mood (verbs), they violate harmony. And it's exactly the nature of this "melody" (violated harmony) that conveys the declension / conjugation. Verbs aren't conjugated according to tense, aspect or person though, and nouns don't have number either, so quite a small number of permutations suffices.

All this may read quite confusing and convoluted without examples, and maybe it will eventually - or rather soon - also turn out that it's not actually feasible at all to work with such a language, but I'm trying!

[Pabappa]

Right clicking / thumbpressing the images should work ... I think he just didn't need to put the URL tags since the img tags automatically render.