Conlang Random Thread

[missals]

It's a kinda weird but if it's just a gap in patterning of CC onsets it's not unbelievable or anything. What would be unbelievable if a language allowed, say, CCC onsets but not CC ones.

[Nortaneous]

It's a kinda weird but if it's just a gap in patterning of CC onsets it's not unbelievable or anything. What would be unbelievable if a language allowed, say, CCC onsets but not CC ones.

The existence of CCC clusters implies the existence of CC clusters, but it's possible for a language to allow a specific C₁C₂C₃- cluster but not C₁C₂-, or C₂C₃-, or C₁C₃-.

No C₁C₂-: Erzya gʒn-, *gʒ- [in a Russian loanword with ʒn- due to hypercorrection; the related language Moksha had kʃ- > ʃ- 'in a front-vocalic environment]

No C₂C₃-: Ungarinjin brɻ-, *rɻ [contrasting with br- and bɻ-]

No C₁C₃-: English str-, *sr- [unless you have it in "sriracha", but that's a spelling pronunciation]

(source)

[dɮ the phoneme]

Is it plausible for all roots in a language to begin with a voiced consonant? This has arisen in one of my conlangs due to a pervasively prefixing morphology, together with a lack of root-initial clusters in the proto-language, and subsequent intervocalic voicing of obstruents.

[Xwtek]

Is it realistic that I have /tst tɬt tɕt > st ɬt ɕt/?

[Zaarin]

Is it plausible for all roots in a language to begin with a voiced consonant? This has arisen in one of my conlangs due to a pervasively prefixing morphology, together with a lack of root-initial clusters in the proto-language, and subsequent intervocalic voicing of obstruents.

Unusual but probably not impossible. It should certainly give your language a distinctive flavor.

Is it realistic that I have /tst tɬt tɕt > st ɬt ɕt/?

Perfectly. A similar change (to the first) happened in Greek and Celtic.

[cenysor]

David Peterson is working on Denis Villeneuve's Dune:

https://io9.gizmodo.com/game-of-thrones ... 1833921344

[malloc]

The abstract mechanics of my conscript have developed nicely, but I have no idea what the script should actually look like. Despite my best efforts, I cannot find any inspiration. My attempts at doodling conscript ideas thus far keep looking like existing scripts and don't feel much like the language they are meant to record.

[bradrn]

The abstract mechanics of my conscript have developed nicely, but I have no idea what the script should actually look like. Despite my best efforts, I cannot find any inspiration.

Have a look at Ian James's scripts; they're exceptionally good. Mattias Persson has made a lot of scripts as well. Omniglot's conscript section is also very good.

[Qwynegold]

As I was about to go to bed I started to think about the word for "then" in one of my conlangs. I thought I had two words for "then", but couldn't remember what exactly the distinction was. (Now that I looked at the grammar, it turns out I only had "now" and "then" which I had classified as "proximal" and "distal temporal adverbs". ) Anyway, I did come up with three ideas about how one could have two "then".

1. Then1 is used about events that refer to yourself and then2 about events that refer to a 2nd or 3rd person.
2. Then1 refers to yourself. Then2 refers to an event someone else has told about. But in instructions or suggestions, then1 is used if it came from you, regardless of who should perform the action. E.g. "Can you tell me how to get to the library?" "Turn left at the first crossing, then1 turn right." Here then1 is used because the instruction comes from the 1st person, even though it's the 2nd person who should turn right.
3. Then1 is used in past tense and then2 in future.

To me it seems like idea #3 could easily happen in a natlang, but I don't know of any languages that have several words for "then". Do you?

EDIT: I just realized that Swedish has "sen" and "då". >_< But I can't explain the difference.

In English, then is not just a distal temporal adverb. It's also sequence conjunction. Many languages distinguish between distal temporal adverb and conjunction with sequence meaning.

In Indonesian, we have 3 words for "then":

1. Lalu, kemudian: this is used for then with meaning the next action in sequence
2. Saat itu: Used for "that time" meaning of then. (i.e. distal temporal adverb meaning of then)
3. Maka: Used in conditional construction. (Jika saya tidak pergi, maka saya tidak bisa bekerja = If I didn't go, I wouldn't be able to work then. Using kemudian instead of maka, the sentence would be read as If I didn't go and I couldn't work)

Aah, thank you! Swedish sen is a sequence conjunction!

[Qwynegold]

Is it realistic to have a diphthong like this:

/æɪ ɑʊ ɪæ ʊɑ/

Finnish has /æi ɑu iæ/ plus many more. /ʊɑ/ (or /uɑ/) is more iffy. It does occur as a common mispronunciation of /uo/, and it could appear in foreign words, e.g. Kuala Lumpur.

[Qwynegold]

Is it plausible for all roots in a language to begin with a voiced consonant? This has arisen in one of my conlangs due to a pervasively prefixing morphology, together with a lack of root-initial clusters in the proto-language, and subsequent intervocalic voicing of obstruents.

You have already come up with an explanation for this oddity, so go for it.

[Qwynegold]

The abstract mechanics of my conscript have developed nicely, but I have no idea what the script should actually look like. Despite my best efforts, I cannot find any inspiration. My attempts at doodling conscript ideas thus far keep looking like existing scripts and don't feel much like the language they are meant to record.

Have you thought about what writing materials they would use? Because that will influence how the script looks, so when you have that figured out you might get some inspiration.

[Raphael]

Newbie question: if lenition, or "softening", is one of the most common sound changes, then how do languages ever become "hard" in the first place? (Yes, this could have gone in the Linguistic Miscellany Thread, but I'm asking it here because the reason why i'm asking it is related to my attempts at conlanging.)

[dhok]

Newbie question: if lenition, or "softening", is one of the most common sound changes, then how do languages ever become "hard" in the first place? (Yes, this could have gone in the Linguistic Miscellany Thread, but I'm asking it here because the reason why i'm asking it is related to my attempts at conlanging.)

Fortition is a thing, even between vowels, and is not as rare as you might suppose. Off the top of my head: Spanish voiced fricatives (other than the allophones of the stops) devoiced unconditionally; some Salish languages had *j *w > tʃ kʷ if not after a consonant; sometimes voiced stops just unconditionally devoice (Tocharian, Germanic, Anatolian, Chinese).

[Raphael]

Thank you!

[Salmoneus]

While fortition does happen, it's true that (in the middle of words) it happens less than lenition does.

But the difference can be made up through cluster simplification and through compounding.

Say /t/ > /d/ intervocalically. But say that, first, unstressed vowels drop; and after, clusters simplify. So "kapa" > "kaba", but "katapa" > "katpa" > "kappa" > "kapa". And then there is compounding, so "kapa+katapa" > "kabakapa".

Fortition can also happen in specific pro-fortition environments, then then disappear. So, for example, maybe lenition doesn't happen to the onsets of stressed syllables. Then, "kápa" > "kaba", but "kapá" > "kapa".

[Raphael]

Thank you, too!

[Raholeun]

In the Sataw language, there exists an AUGMENTATIVE morpheme -ʔ. It gets suffixed onto a root to add a sense of size, intensity or briskness. This suffix is quite productive in Modern Sataw, as new words can be formed from older stems:

wawta [ˈwɒʊ̯ðɐ] "fish" > wawtaʔ [wɒʊ̯ˈtæˀ] "fish.AUG"

Likewise, loanwords may be modified:

denwa [ˈðɪnwɐ] "telephone" > denwaʔ [ðənˈwæˀ] "telephone.AUG"

It is not uncommon for the -ʔ to not only augment the original meaning of the word, but to cause a semantic shift. This shift in meaning is historical in nature. Sometimes only the from with a fossilized suffix has been preserved, while the original root has been lost from the language:

kanitaʔ [kənəˈtæˀ] "evil spirit, demon" < Proto-Austronesian *qaniCu "spirit" [1]

This derivational process seems mechanically regular at first sight. However, this morpheme (among others) may trigger a morphophonemic change; not end the end of the word where the suffix is added, but in the beginning of the word. This process is known as consonant ablaut, or alternatively as Sataw initial permutations. Spoiler alert: I explain these modern intial permutations ʔ~k~h~Ø by positing an earlier threesome of postvelars, algebraically notated as *q1, *q2 and *q3.

ʔasi "to fall, flow, spill" < Proto-Austronesian *qaCi ‘ebb’
asi-ʔ " to splash, overflow, cascade"

ʔutay "a (small) berry" < Proto-Austronesian *qutaɬ ‘small plant’
huti-ʔ "a large berry"

ʔamis "north" < Proto-Austronesian ‘north’
kamii-ʔ "far north"

My question is are there ANADEW examples of this happening? Would anyone know of some sources explaining the diachronic processes at work?


[1] Compare Bunun qanitu, Tsou hicu and Itbayaten anitu "ghost, spirit".
[2] Sataw derivational morphology is particularly promiscuous in nature, as morphemes may pair with both nouns and verbs, as well as other parts of speech.

[Raholeun]

Edit
It seems like I misspoke when stating qanitaʔ has no counterpart lacking the final, AUG -ʔ. There is in fact a lemma kanita in the Sataw dictionary: ‘heirloom, an object with spiritual connotations that is passed on through generations’ (n). *qanita ‘supernatural being’, very well attested in Formosan (Yamada: 2002).

[Pabappa]

Im a little confused. you dont specify that Sataw is a conlang, but I assume it is because it's what youre asking about .... it looks like a very specific and well thought-out conlang, given how closely its vocabulary tracks those of the other Austronesian languages, and how it has a word for telephone that;s loaned from Japanese as if it came to them during the runup to WW2. Very well done. But Im still not clear how to answer the question ... is this k~h~ʔ~Ø something you came up with on your own and dont have a good historical explanation of, or are there similar phenomena in Austronesian languages that youre trying to figure out? also is there a /q/? you type the word /kanitaʔ/ once and /qanitaʔ/ later. typo, i assume?