Yay or Nay? [2011–2018]

[zyma]

Uh, in any case, my answer is "yay" to making them phonemes, if that's still what's being asked.

Yes, though I realized that /nʶ/ is likely to merge with my uvular nasal, leaving only /mʶ/...

If /nʶ/ can only appear before three vowels, maybe you could say it's already merged with the uvular nasal before all the other vowels, unless the uvular nasal is restricted in a similar way when it comes to phonotactics. And I wouldn't think that /nʶ/ would be all that more likely to merge with the uvular nasal than /mʶ/ would be. You should be fine keeping the three distinct if you want to, especially if you're going to have the uvularized nasals phonetically realized in a somewhat notably different way compared to other uvularized and uvular phonemes.

[HoskhMatriarch]

OK, I have all the CALS features decided for Hoskh except comparative constructions. I kind of want to use the particle comparative, but it's really overused in conlangs and I probably already spent all my europoints. I really don't like the fixed case locational and exceed comparatives though, because saying something like "I love you exceed him" with the exceed comparative doesn't tell you if I love you more than I love him or I love you more than he loves you, and the same goes for the locational. However, with the conjoined one, it doesn't seem like there's a way to put it into relative clauses such as "this house, which is bigger than that house..." since it depends on it being two clauses, it causes ambiguity with "this big that small" when "that" might not actually be small, only less big ("dragons big giants small" and with other adjectives "cheetahs fast racehorses slow" etc.), and I can't even figure out how to say things like "this car's gas mileage is 5 MPG higher than that one's" with it (which I can with the two fixed case ones). So should I just go ahead and use the particle comparative since it seems to kind of be the best?

[zyma]

OK, I have all the CALS features decided for Hoskh except comparative constructions. I kind of want to use the particle comparative, but it's really overused in conlangs and I probably already spent all my europoints. I really don't like the fixed case locational and exceed comparatives though, because saying something like "I love you exceed him" with the exceed comparative doesn't tell you if I love you more than I love him or I love you more than he loves you, and the same goes for the locational. However, with the conjoined one, it doesn't seem like there's a way to put it into relative clauses such as "this house, which is bigger than that house..." since it depends on it being two clauses, it causes ambiguity with "this big that small" when "that" might not actually be small, only less big ("dragons big giants small" and with other adjectives "cheetahs fast racehorses slow" etc.), and I can't even figure out how to say things like "this car's gas mileage is 5 MPG higher than that one's" with it (which I can with the two fixed case ones). So should I just go ahead and use the particle comparative since it seems to kind of be the best?

Well, "I love you more than him" is technically just as ambiguous as hypothetical constructions like "I love you exceed him" or "I love you from him". I don't speak any languages that don't use particle comparatives, so I can't exactly tell you how languages with different comparative constructions disambiguate things if context doesn't do the job for them, but I'm almost entirely sure that they do have ways of doing so, since there are functioning natlangs that don't use particle comparatives. No way is truly better than any other way, but people who grow up speaking languages that use particle comparatives are more likely to have the easiest time understanding and using that kind of construction.

But, anyway, I would say use particle comparatives, unless you really don't want to. What makes you think particle comparatives are "overused", and why would it matter even if they objectively were? Also, why worry about "Europoints" if doing so makes it more difficult for you to make your language the way you want it to be? (I don't mean to sound like I'm trying to dismiss you; I really do want to know your thoughts on these questions. It's fine if you personally want your language to be very unlike European languages, but I don't want you or anyone else to feel pressured or obligated to make conlangs that are unlike SAE languages because of other people.)

[HoskhMatriarch]

OK, I have all the CALS features decided for Hoskh except comparative constructions. I kind of want to use the particle comparative, but it's really overused in conlangs and I probably already spent all my europoints. I really don't like the fixed case locational and exceed comparatives though, because saying something like "I love you exceed him" with the exceed comparative doesn't tell you if I love you more than I love him or I love you more than he loves you, and the same goes for the locational. However, with the conjoined one, it doesn't seem like there's a way to put it into relative clauses such as "this house, which is bigger than that house..." since it depends on it being two clauses, it causes ambiguity with "this big that small" when "that" might not actually be small, only less big ("dragons big giants small" and with other adjectives "cheetahs fast racehorses slow" etc.), and I can't even figure out how to say things like "this car's gas mileage is 5 MPG higher than that one's" with it (which I can with the two fixed case ones). So should I just go ahead and use the particle comparative since it seems to kind of be the best?

Well, "I love you more than him" is technically just as ambiguous as hypothetical constructions like "I love you exceed him" or "I love you from him". I don't speak any languages that don't use particle comparatives, so I can't exactly tell you how languages with different comparative constructions disambiguate things if context doesn't do the job for them, but I'm almost entirely sure that they do have ways of doing so, since there are functioning natlangs that don't use particle comparatives. No way is truly better than any other way, but people who grow up speaking languages that use particle comparatives are more likely to have the easiest time understanding and using that kind of construction.

But, anyway, I would say use particle comparatives, unless you really don't want to. What makes you think particle comparatives are "overused", and why would it matter even if they objectively were? Also, why worry about "Europoints" if doing so makes it more difficult for you to make your language the way you want it to be? (I don't mean to sound like I'm trying to dismiss you; I really do want to know your thoughts on these questions. It's fine if you personally want your language to be very unlike European languages, but I don't want you or anyone else to feel pressured or obligated to make conlangs that are unlike SAE languages because of other people.)

Well, "I love you more than him" might be ambiguous in English, due to the fact that people seem to be switching all pronouns after "than" to accusative, but in languages like Latin and German it's normal to say something like "I love you more than he" in contrast to "I love you more than him" which have different meanings: "I love you more than he (loves you)" vs. "I love you more than (I love) him".

Here's how commonly used particle comparatives are:
http://cals.conlang.org/feature/121/

I don't feel obligated to make conlangs unlike SAE languages because of other people. I just think it's too weird to have a language with 11 inflectional categories per verb, polypersonal agreement, active-stative morphosyntactic alignment, pharyngeals and lateral obstruents, and then random euro features. I'm not going to make my languages less like SAE languages just because people might complain, but I'm not going to make them more like SAE languages just because people might complain either

[Creyeditor]

OK, I have all the CALS features decided for Hoskh except comparative constructions. I kind of want to use the particle comparative, but it's really overused in conlangs and I probably already spent all my europoints. I really don't like the fixed case locational and exceed comparatives though, because saying something like "I love you exceed him" with the exceed comparative doesn't tell you if I love you more than I love him or I love you more than he loves you, and the same goes for the locational. However, with the conjoined one, it doesn't seem like there's a way to put it into relative clauses such as "this house, which is bigger than that house..." since it depends on it being two clauses, it causes ambiguity with "this big that small" when "that" might not actually be small, only less big ("dragons big giants small" and with other adjectives "cheetahs fast racehorses slow" etc.), and I can't even figure out how to say things like "this car's gas mileage is 5 MPG higher than that one's" with it (which I can with the two fixed case ones). So should I just go ahead and use the particle comparative since it seems to kind of be the best?

Some Indonesian dialects use a locative comparative. To me, sentences like 'I am hungrier now than yesterday.' are very difficult. 'I am hungry from yesterday' is ungrammatical, so if I really wanted to use the locative comparative, I would have to say something like 'I am bigger from me yesterday' which is only slightly better. Native speakers usually use some kind of other of other paraphrasis: e.g. 'Yesterday I was hungry, but today I am very hungry'. What I'm trying to say is, that there might be one 'standard' strategy for comparison, but there might be also other strategies for peripharil uses.

[zyma]

Well, "I love you more than him" might be ambiguous in English, due to the fact that people seem to be switching all pronouns after "than" to accusative, but in languages like Latin and German it's normal to say something like "I love you more than he" in contrast to "I love you more than him" which have different meanings: "I love you more than he (loves you)" vs. "I love you more than (I love) him".

Exactly. I'm sure there are plenty of languages out there with different comparative constructions that have ways to largely avoid ambiguity too.

Here's how commonly used particle comparatives are:
http://cals.conlang.org/feature/121/

That data's only based on conlangs that have been entered into CALS, though. Well, anyway, moving on.

I don't feel obligated to make conlangs unlike SAE languages because of other people. I just think it's too weird to have a language with 11 inflectional categories per verb, polypersonal agreement, active-stative morphosyntactic alignment, pharyngeals and lateral obstruents, and then random euro features. I'm not going to make my languages less like SAE languages just because people might complain, but I'm not going to make them more like SAE languages just because people might complain either

Here is the comparative constructions map for WALS. There are only 167 entries, sadly, but hopefully it can still be helpful. Even though particle comparatives are the smallest group shown on the map, they exist in more than one area outside of Europe. I don't think there's anything inherently incompatible between particle comparatives and the features you listed, even though they don't typically happen to coexist on Earth. I find this especially true if your language isn't spoken on Earth.

But what I think and what I find aren't really relevant. If you don't like the way particle comparatives fit in with the rest of your language, feel free to ditch them! As you said, complainers shouldn't be dictating your creations.

Hopefully at least some of this was helpful. I wish I knew any languages with non-particle comparatives so I could give some advice about how natlangs work with those. (Thanks to Creyeditor for informing us about comparison in Indonesian!)

[Bristel]

A Germanic language derived from Old Saxon (Old Low German) but is heavily influenced by a local Celtic language?

A sister to that language but is influenced by Romance languages?

[zyma]

A Germanic language derived from Old Saxon (Old Low German) but is heavily influenced by a local Celtic language?

A sister to that language but is influenced by Romance languages?

Awesome concepts; "Yay" to both!

[Bristel]

Might not develop the Celtic conlang any more right now, but I could use its quite large lexicon for borrowings and influence on this Saxon language. (as for the Romance influenced conlang, I think it should be something around France (maybe far enough south for Italian dialects).

[protondonor]

I'm making a diachronic language and just realized that I haven't been keeping track of the etymology of my vocabulary. Should I start over and keep an etymological dictionary, go back and reconstruct the words I do have, or press forward and not worry about it? I don't have that many words yet (between 100 and 200 words at the moment).

Also I recently realized that a word I had used to create a copula actually has a really well-preserved meaning (to sit) in all the other languages in the family. Should I find another copula, or is it reasonable that in just one subfamily out of a large family, the word for "to sit" would have become an existential copula, whereas in almost all the other languages in the family, it preserved the original meaning?

[zyma]

Might not develop the Celtic conlang any more right now, but I could use its quite large lexicon for borrowings and influence on this Saxon language. (as for the Romance influenced conlang, I think it should be something around France (maybe far enough south for Italian dialects).

Sounds good to me.

I'm making a diachronic language and just realized that I haven't been keeping track of the etymology of my vocabulary. Should I start over and keep an etymological dictionary, go back and reconstruct the words I do have, or press forward and not worry about it? I don't have that many words yet (between 100 and 200 words at the moment).

I'm afraid I don't quite understand. If possible, could you show an example or two of what you've been doing so far?

Also I recently realized that a word I had used to create a copula actually has a really well-preserved meaning (to sit) in all the other languages in the family. Should I find another copula, or is it reasonable that in just one subfamily out of a large family, the word for "to sit" would have become an existential copula, whereas in almost all the other languages in the family, it preserved the original meaning?

I'd say keep "to sit" as the copula; languages can be weird, and outliers (for lack of a better word) like this do happen, in terms of semantics (like your situation), phonology, morphosyntax, etc.

[HoskhMatriarch]

I'm making a diachronic language and just realized that I haven't been keeping track of the etymology of my vocabulary. Should I start over and keep an etymological dictionary, go back and reconstruct the words I do have, or press forward and not worry about it? I don't have that many words yet (between 100 and 200 words at the moment).

Also I recently realized that a word I had used to create a copula actually has a really well-preserved meaning (to sit) in all the other languages in the family. Should I find another copula, or is it reasonable that in just one subfamily out of a large family, the word for "to sit" would have become an existential copula, whereas in almost all the other languages in the family, it preserved the original meaning?

I would reconstruct the words because it'd be fun. If you wouldn't find that fun, don't do it, though.

I would keep the "to sit" copula. Copulas develop from words with other meanings all the time, and there are similar examples all throughout Indo-European. https://en.wikipedia.org/wiki/Indo-European_copula

[loglorn]

I say to start keeping track of etymologies while there's time. When your lexicon gets bigger, working backwards to find out etymologies is going to be a hell lot of work, so it's better to do it now, because keeping track of etymology is quite useful.

---

Also, protondonor, how hardcorely are you working out semantic drift? How (if you actually are) are you calculating how much will change per millenium?

Not saying you should calculate, you can carry on without doing so. I'm just curious.

[protondonor]

I'm making a diachronic language and just realized that I haven't been keeping track of the etymology of my vocabulary. Should I start over and keep an etymological dictionary, go back and reconstruct the words I do have, or press forward and not worry about it? I don't have that many words yet (between 100 and 200 words at the moment).

I'm afraid I don't quite understand. If possible, could you show an example or two of what you've been doing so far?

Sure, my lexicon pretty much looks like this:
understand v. trans. kaste
urine (n.), urinate (v. intr.) guske
vein n. þūne
wade v. kjalja
wagon, cart n. raþa (< Proto-Indo-Iranian)

So I have annotations for when a word came from a different language family (so far there are only Proto-Indo-Iranian loans, there may be a few Proto-Balto-Slavic loans as well), but no annotations for what reconstructed Proto-Uralic word (this is a Uralic project) I derived the word from. So when I want to coin a new word, there's nothing I can reference for whether I've used a given P-U root before and if so, for what. Although for some of those it's pretty obvious, some of them have pretty idiosyncratic derivations.

Also, protondonor, how hardcorely are you working out semantic drift? How (if you actually are) are you calculating how much will change per millenium?

Not saying you should calculate, you can carry on without doing so. I'm just curious.

I don't have any particular formula for semantic drift, nor do I really have any idea what that would be. I've mostly been massaging the meanings of individual lexemes as I see fit.

Looks like the consensus is keeping sit as a copula is a-okay. I'm happy with that, because I really liked that derivation. I'm probably going to go back and reconstruct at least a few of the words I have, if only to make sure I did the derivations right the first time. And cause I've gotten quite a thrill out of looking at long Proto-Uralic wordlists.

EDIT: I have to put my lexicon in a better format anyway so it's going to become an etymological dictionary.

[zyma]

Sure, my lexicon pretty much looks like this:
understand v. trans. kaste
urine (n.), urinate (v. intr.) guske
vein n. þūne
wade v. kjalja
wagon, cart n. raþa (< Proto-Indo-Iranian)

So I have annotations for when a word came from a different language family (so far there are only Proto-Indo-Iranian loans, there may be a few Proto-Balto-Slavic loans as well), but no annotations for what reconstructed Proto-Uralic word (this is a Uralic project) I derived the word from. So when I want to coin a new word, there's nothing I can reference for whether I've used a given P-U root before and if so, for what. Although for some of those it's pretty obvious, some of them have pretty idiosyncratic derivations.

Ah, thanks!

What I personally would recommend doing is to stop making new words until you've reconstructed the etymologies of the hundred or so words you have now and recorded them with your lexicon. Once that's all done, keep adding new words while simultaneously recording etymologies so you don't have to reconstruct any more in the future.

Also, your project sounds so cool!

[protondonor]

What I personally would recommend doing is to stop making new words until you've reconstructed the etymologies of the hundred or so words you have now and recorded them with your lexicon. Once that's all done, keep adding new words while simultaneously recording etymologies so you don't have to reconstruct any more in the future.

Also, your project sounds so cool!

Thanks! That's exactly what I'm doing right now! Glad you like it, I hope I can eventually work it into something I can present here. It's over a year in the making, but I tend to conlang in unfortunately brief spurts.

[zyma]

Thanks! That's exactly what I'm doing right now! Glad you like it, I hope I can eventually work it into something I can present here. It's over a year in the making, but I tend to conlang in unfortunately brief spurts.

I hope we get to hear more about it on the board in the future as well! (But I can completely empathize with working in brief spurts of conlang productivity.)

[HoskhMatriarch]

I just realized that my language Hoskh has an affricate (only voiceless though) corresponding to every place of articulation it has a stop for (as well as aspirated vesions for some of them):

/t/ - /t͡s/, /t͡ʃ/
/k/ - /k͡x/
/q/ - /q͡χ/

Should I just add in a /ʡ͡ʜ/ and /p͡f/ so I can say it has affricates for every stop? I feel really tempted to but I also feel like there's probably some reason I shouldn't.

[Creyeditor]

I would say, adding those would be somehow too much.

[Adarain]

I kinda have to agree. Maybe add only p͡f, but it’s a pretty rare one overall. But not as bad as the one with the epiglottal… trill?