(Conlangs) Q&A Thread - Quick questions go here

A forum for all topics related to constructed languages
Salmoneus
MVP
MVP
Posts: 1636
Joined: 19 Sep 2011 19:37

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Salmoneus » 12 May 2019 15:59

Of course sound changes have exceptions. English is a fantastic case in point! A big part of why our spelling is so frustrating is the fact that English vowels in particular have been subjected to a whole bunch of irregular, uncompleted (and sometimes incompletely reversed!) sound shifts.

Take "father". Most (all?) modern dialects have undergone the tensing of the vowel in "father", but for many dialects it's only happened in that one word. In SSBE, however, it's spread, and (almost) everyone has tensing (now backing) in "rather". However, there's uncertainty over whether to have backing in "lather", and nobody at all has backing in "gather". Traditional RP even had a Mass/mass split, where the backing applied to one homophone but not the other - similarly, "crass" and "bass" (the fish) are unbacked, while "grass" and (usually but not always) "brass" are backed. The equivalent tensing processes have been equally irregular in several other dialects - eg, in American dialects with a post-father split, tensing has been resisted by "mad", "bad" and "glad", but not by "sad" or "lad" - whereas (some) English English used to have (and I think Australian still has?) tensing in "bad" and "sad", but not in "lad", and had a jam/jam split.

Similarly, further back, the foot-strut split was similarly irregular: hence 'pudding', but 'budding'. Notably, this included a put/putt split (the same word either undergoing or not undergoing the split depending on meaning). And this change followed the equally irregular shorting of /u/ - hence "blood"/"flood" (shortening before the foot-strut split), "foot"/"good" (shortening after the foot-strut split), and "food"/"boot" (no shortening).



However, I think aelfwine has it the wrong way around: commonly used words are more likely to adopt innovative pronounciations that then spread through the lesser-known word, rather than remaining more conservative (although VERY little-used words may easily be spelling-pronounced to match the common ones).

User avatar
Vlürch
sinic
sinic
Posts: 264
Joined: 09 Mar 2016 21:19
Location: Finland
Contact:

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Vlürch » 12 May 2019 20:01

Ælfwine wrote:
12 May 2019 02:57
The more I immerse myself in linguistics, the more it seems that certain dogma, such as "sound change is exceptionless" is false. (But what do I know.)
Every time I say that, I get shit... but yeah, logically sound changes aren't always 100% regular even in part just because people's speech isn't 100% regular. I'd argue that especially with all the languages of Eurasia, which have had such extensive contact with each other for millennia, exceptions really could be made for certain "rules".

The discrepancies and seemingly irregular sound correspondences between languages could be explained by there having been intermediary languages that the words passed through before entering the destination language, or they could be related but the languages that formed a "bridge" between them went extinct without leaving a trace so the relation became muddled as a huge chunk of the continuum is missing, but saying that is apparently the same as saying "aliens, bruh". [>_<]

Not trying to start a flamewar or anything, and I've come to accept that the Eurasiatic language family, if real (which I think is the case), includes Indo-European and that my past denial of that was based on nationalist pseudoscience even if I thought it was based on fact. I'm trying to make up for it by at least apologising whenever the issue comes up. Also, sorry that this is a somewhat long borderline off-topic post.

Porphyrogenitos
sinic
sinic
Posts: 249
Joined: 21 Jul 2012 08:01
Location: Buffalo, NY

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Porphyrogenitos » 13 May 2019 05:10

The ship sailed on Neogrammarian sound change decades ago (or at least on an absolutist conception of it). We know that sound change isn't perfectly exceptionless but it's necessary to pretend it is when, for example, embarking on the reconstruction of a language family. And that assumption works pretty well - it just might become necessary to edit in certain irregularities later.

And as for definite articles and other grammatical particles, if they are to behave in an anomalous manner, they are going to anomalously participate in a sound change rather than anomalously resist it. Articles are some of the most frequent words in the language and are going to be highly prone to reduction - "practice makes imperfect". Take the English article the, which, along with a group of other grammatical particles that began with /θ/, spontaneously became phonemically voiced, even though voicing of /θ/ was regularly triggered only by an intervocalic environment (as in bathe). Their use was so frequent that speakers just began leniting their initial consonant.

So if there's a vowel shift from o: > u:, I don't think an article would be likely to resist that.

User avatar
CarsonDaConlanger
sinic
sinic
Posts: 250
Joined: 02 Nov 2017 20:55

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by CarsonDaConlanger » 13 May 2019 16:05

Reyzadren wrote:
11 May 2019 00:27
CarsonDaConlanger wrote:
10 May 2019 16:23
Speaking of which, how do langs with no tense and only aspect convey a generic past tense? Do they use perfective aspect? Wouldn't that just make it a tense anyways? For example how would they say "John fought the snake?"
Langs without tenses don't convey tenses because they don't need to. Most conlangers place too much value on tense and assume that tenseless langs need to somehow express them whenever a tense language does, especially with regards to tense agreement, which is false. The strategy and usage depends on the specific language really.

Relevant example from that other natlang that I speak: There are several things that one could use which would cause a resemblance of past tense. Using the word "already" (which I assume is what you mean by "perfective"), using a generic word that means "I am a past tense", add an additional clause such as "and this happened in the past", 0 (ie, do nothing). Which do you think is the most common method to translate "John fought a snake"?

The answer is 0. Ie, "John fight snake". This might seem strange to you, but nobody speaking this natlang ever spams "I am a past tense" (or the other ways) in every sentence, even when it is obvious from the translation that there was a past tense involved. The natlang simply doesn't care. In fact, my conlang does this too.
Wait, so how do you know that the action happened in the past?

User avatar
Creyeditor
mongolian
mongolian
Posts: 4490
Joined: 14 Aug 2012 19:32

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Creyeditor » 13 May 2019 17:35

From the context and world knowledge. Or you don't know at all, if it does not matter.
Creyeditor
"Thoughts are free."
Produce, Analyze, Manipulate
1 :deu: 2 :eng: 3 :idn: 4 :fra: 4 :esp:
:con: Ook & Omlűt & Nautli languages & Sperenjas
[<3] Papuan languages, Morphophonology, Lexical Semantics [<3]

User avatar
Reyzadren
greek
greek
Posts: 458
Joined: 14 May 2017 10:39
Contact:

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Reyzadren » 14 May 2019 00:47

CarsonDaConlanger wrote:
13 May 2019 16:05
Reyzadren wrote:
11 May 2019 00:27
The answer is 0. Ie, "John fight snake". This might seem strange to you, but nobody speaking this natlang ever spams "I am a past tense" (or the other ways) in every sentence, even when it is obvious from the translation that there was a past tense involved. The natlang simply doesn't care. In fact, my conlang does this too.
Wait, so how do you know that the action happened in the past?
I don't, but I don't need to because the speakers of that natlang don't care about whether it happened in the past or not. Much like we don't need to know about where it happened, why it happened or how it happened.
Image Soundcloud Profile | Image griuskant conlang

User avatar
eldin raigmore
korean
korean
Posts: 6369
Joined: 14 Aug 2010 19:38
Location: SouthEast Michigan

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by eldin raigmore » 14 May 2019 06:19

Reyzadren wrote:
14 May 2019 00:47
I don't, but I don't need to because the speakers of that natlang don't care about whether it happened in the past or not. Much like we don't need to know about where it happened, why it happened or how it happened.
More than one language with evidentiality, doesn’t require and typically doesn’t say when something happened/happens/will happen; but absolutely mandates that the speaker say how they know.

Salmoneus
MVP
MVP
Posts: 1636
Joined: 19 Sep 2011 19:37

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Salmoneus » 14 May 2019 23:26

Tense usually isn't that important. On the rare occasions when it is important, it's usually clear from context - because if tense is important, it's because something is urgent, and urgency is usually made very clear from context.

If someone tells you "house burn down", it's easy to tell if they mean in the past or in the present - if they mean in the past, they look quietly sad, while if they mean in the present, they're screaming and pointing. If the house is burning down in the present and they AREN'T screaming and pointing, but only looking quietly sad, this probably means you either can't, or shouldn't, do anything about it anyway, so it doesn't really matter that it's not (yet) in the past.

User avatar
elemtilas
runic
runic
Posts: 3512
Joined: 22 Nov 2014 04:48

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by elemtilas » 14 May 2019 23:46

Salmoneus wrote:
14 May 2019 23:26
Tense usually isn't that important. On the rare occasions when it is important, it's usually clear from context - because if tense is important, it's because something is urgent, and urgency is usually made very clear from context.

If someone tells you "house burn down", it's easy to tell if they mean in the past or in the present - if they mean in the past, they look quietly sad, while if they mean in the present, they're screaming and pointing. If the house is burning down in the present and they AREN'T screaming and pointing, but only looking quietly sad, this probably means you either can't, or shouldn't, do anything about it anyway, so it doesn't really matter that it's not (yet) in the past.
True, but wouldn't you think such context would depend rather a lot on whose house burn down. Someone from down the road or a few blocks over might look quietly sad while the fire is in progress. The homeowner might scream and point regardless of when the event took place, though I concur they will almost certainly do so while the fire is in progress! If anyone looks quietly sad on account of can't do anything about it, they're expressing something other than past tense.

yangfiretiger121
sinic
sinic
Posts: 233
Joined: 17 Jun 2018 03:04

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by yangfiretiger121 » 15 May 2019 19:44

Are the following arrangements plausible for my setting's current language:

1. Compound words, such as Ιφαυδύ [ʊ.ǃǃɑ.uˈȡʑuː] ("Star Navy"; formerly spelt Υεφαυδύ; βάϙ ['bɑːn] implied from word order), forming similarly to Latin while the Empire's or Emperor's possessions, such as Ϳό βάϙ Νάκέθε [joː 'nɑː.kɛ̥ː.ʇɛ̥] ("The Emperor's Military"; "emperor" formerly spelt Ϳάυ], form similarly to Japanese
2. Coalescence affecting VV sequences in simple words, such as ἀ [ḁ] ("great"; formerly spelt, εα) but not word-final vowel-into-word-initial vowel sequences, as seen in Star Navy's above translation
Last edited by yangfiretiger121 on 15 May 2019 20:49, edited 1 time in total.
Alien conlangs (Font may be needed for Vai symbols)

shimobaatar
korean
korean
Posts: 11580
Joined: 12 Jul 2013 23:09
Location: PA → IN

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by shimobaatar » 15 May 2019 20:08

yangfiretiger121 wrote:
15 May 2019 19:44
1. Compound words, such as Ιφαυδύ [ʊ.ǃǃɑ.uˈȡʑuː] ("Star Navy"; formerly spelt Υεφαυδύ; βάϙ ['bɑːn] implied from word order), forming similarly to Latin while the Empire's or Emperor's possessions, such as Ϳό βάϙ Νάκέθε [joː 'nɑː.kɛ̥ː.ʇɛ̥] ("The Emperor's Military"; "emperor" formerly spelt Ϳάυ], form similarly to Japanese
If you're asking whether or not it's alright for a language to use both compounding and possessive constructions, then yes.
yangfiretiger121 wrote:
15 May 2019 19:44
2. Coalescence affecting VV sequences in simple words, such as ἀ [ḁ] ("great"; formerly spelt, εα) but not word-final vowel-into-word-initial vowel sequences, as seen Star Navy's above translation
So, you're asking if it's OK to have vowel coalescence occur within morphemes, but not across morpheme boundaries, such as in compound words? I think that should be fine, especially if the elements of a compound can still appear as independent words.

yangfiretiger121
sinic
sinic
Posts: 233
Joined: 17 Jun 2018 03:04

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by yangfiretiger121 » 15 May 2019 21:02

Okay. Thanks.
Alien conlangs (Font may be needed for Vai symbols)

User avatar
Davush
greek
greek
Posts: 507
Joined: 10 Jan 2015 14:10

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Davush » 16 May 2019 20:23

Hello,

I am hoping to develop a descendent of Classical Qutrussan which develops pitch accent or register tone. I don't really want full contour-tones à la Chinese, but I am struggling to come up with ways to develop pitch accent or a register tone system. Does anyone know of any papers that describe how a non-pitch accented language might have gained pitch/register tones?

I know loss of codas is often a factor, but the outcomes seem very different depending on language? I.e. final /s/ might lead to a higher or lower pitch? Do I pretty much have free rein in deciding the pitch outcome?

I have read that coda > pitch/tone usually occurs before voicing of stops influencing pitch/tone, but I'm quite unsure how to proceed. Any advice would be appreciated! Thanks!

User avatar
Creyeditor
mongolian
mongolian
Posts: 4490
Joined: 14 Aug 2012 19:32

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Creyeditor » 16 May 2019 20:26

Davush wrote:
16 May 2019 20:23
Hello,

I am hoping to develop a descendent of Classical Qutrussan which develops pitch accent or register tone. I don't really want full contour-tones à la Chinese, but I am struggling to come up with ways to develop pitch accent or a register tone system. Does anyone know of any papers that describe how a non-pitch accented language might have gained pitch/register tones?

I know loss of codas is often a factor, but the outcomes seem very different depending on language? I.e. final /s/ might lead to a higher or lower pitch? Do I pretty much have free rein in deciding the pitch outcome?

I have read that coda > pitch/tone usually occurs before voicing of stops influencing pitch/tone, but I'm quite unsure how to proceed. Any advice would be appreciated! Thanks!
I think there are several papers on how certain Franconian dialects in Germany developed pitch accent. When I google "Franconian pitch accent" the first page of the google hits gives me at least three .pdf-files about their history.
Creyeditor
"Thoughts are free."
Produce, Analyze, Manipulate
1 :deu: 2 :eng: 3 :idn: 4 :fra: 4 :esp:
:con: Ook & Omlűt & Nautli languages & Sperenjas
[<3] Papuan languages, Morphophonology, Lexical Semantics [<3]

User avatar
Davush
greek
greek
Posts: 507
Joined: 10 Jan 2015 14:10

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Davush » 16 May 2019 21:04

Thanks, Creyeditor! I didn't know Franconian dialects have developed pitch accent - how interesting. I will check them out.

Salmoneus
MVP
MVP
Posts: 1636
Joined: 19 Sep 2011 19:37

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Salmoneus » 16 May 2019 23:14

I'm no expert, but as a general guide, iirc I've encountered six different sources of tone:

- loss or neutralisation of coda consonants. Generally fricatives lead to low tones, sfaiaa. However, tones can change around once they're created, so... anyway, glottal stops being lost can result in either high or low - iirc, if they drop directly they tend to give high tone, while if they glottalise the vowel first that tends to give low tone. [I wonder if some fricatives giving low high tone might come via debuccalisation to /h/?] Voicing tends to lead to low tone.

- neutralisation of onset consonants. Voicing tends to lead to low tone. But again, tones shift - so a voiced initial might easily create a following 'rising' tone that then turns into an outright 'high' tone over time.

- neutralisation of vowel phonation. Phonation and voice can be closely connected.

- neutralisation of vowel length. [long vowels tend to be higher-pitched, so if they lose length they can be just high tone]

- neutralisation of stress. [stressed vowels tend to be higher-pitched, so if stress moves or is lost, old stressed vowels can be high tone]

- mora loss or gain [words can have typical contours in a language, and attempting to keep the old contour when a word lengthens or shortens can produce a contrast]

User avatar
Aszev
admin
admin
Posts: 1530
Joined: 11 May 2010 05:46
Location: Upp.
Contact:

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Aszev » 17 May 2019 10:45

Here are two articles about ongoing tonogenesis in Afrikaans and Seoul Korean. You could also check out Punjabi (tone) and Serbo-Croatian (pitch accent).

In Mainland Scandinavian, pitch accent arose primarily from the contrast between mono- and polysyllabic words.

Osc. axl > Sw. ˈaxel (shoulder)
OSc. axull > Sw. ²axel (axis)

OSc. komr > Sw. ˈkommer (comes)
OSc. kennir > Sw. ²känner (knows) (> ˈkänner due to analogy)

OSc. endr > ˈänder (ducks)
Osc. vinir > ²vänner (friends)

OSc. (hund inn >) hundinn > ˈhunden (the dog)
OSc. etinn > ²äten (eaten)
Sound change works in mysterious ways.

Image CE

User avatar
Zekoslav
sinic
sinic
Posts: 222
Joined: 07 Oct 2017 16:54

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Zekoslav » 17 May 2019 12:21

I was toying with the idea to develop a pitch accent by distinguishing older and newer long vowels by pitch contour, but I didn't know it was actually attested. Hello Franconian!

I'm most familiar with tonogenesis by stress retraction, since that's the type of tonogenesis which happens most commonly in Western South Slavic, including my native dialect.

Originally, my dialect had free stress, but all stressed vowels had a non-distinctive falling contour. So words could be distinguished by stress like /ˈpeːta/ "fifth" and /peːˈta/ "heel". The first word had a non-distinctive HLL contour and the scond had a non-distinctive LLH contour.

Then, stress was retracted from short word-final vowels, because in Western South Slavic languages stress is horribly afraid of falling from the edge of the word and there's not many dialects which still keep it in that position.

So /ˈpeːta/ [ˈpêːtà] "fifth" remained the same, but /peːˈta/ [pèːˈtá] "heel" changed to [ˈpêːtá], and the words are now distinguished solely by pitch contour, HLL vs. HLH (well, it's a bit more complicated since the final syllable isn't completely de-stressed yet, but it seems that the contour is phonemic).
Languages:
:hrv: [:D], :bih: :srb: [;)], :eng: [:D], :fra: [:|], :lat: [:(], :deu: [:'(]

A linguistics enthusiast who would like to make a conlang, but can't decide what to call what.

- Tewanian languages
- Guide to Slavic accentuation

yangfiretiger121
sinic
sinic
Posts: 233
Joined: 17 Jun 2018 03:04

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by yangfiretiger121 » 20 May 2019 10:27

How likely is a derivation structure in which a word's first click denasalizes (if nasal) [ʞ̃ > ʞ], voices (if tenuis) [ʞ > ʞ̬], and de-clicks (if voiced) [ʞ̬ > g] for non-dental, non-lateral clicks with tenuis and voiced dental/lateral clicks becoming linguo-pulmonic stops [ʇ > ʇ͡q]?
Alien conlangs (Font may be needed for Vai symbols)

User avatar
Creyeditor
mongolian
mongolian
Posts: 4490
Joined: 14 Aug 2012 19:32

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Creyeditor » 20 May 2019 23:05

I don't think there is a conclusive answer based on natlangs. Not enough data to generalize.
Creyeditor
"Thoughts are free."
Produce, Analyze, Manipulate
1 :deu: 2 :eng: 3 :idn: 4 :fra: 4 :esp:
:con: Ook & Omlűt & Nautli languages & Sperenjas
[<3] Papuan languages, Morphophonology, Lexical Semantics [<3]

Post Reply