Help With Sound Changes

A forum for all topics related to constructed languages
Post Reply
User avatar
Üdj
rupestrian
rupestrian
Posts: 22
Joined: 07 Sep 2022 00:24

Help With Sound Changes

Post by Üdj »

Hello!

I was hoping to get some suggestions for sound changes in my new language. Sound change has always been the hardest part for me; I just never really have ideas.

Here's the phonology:

Stops: /p b t d k g q/ <p b t d k g q>
Nasals: /m n/ <m n> plus /ŋ ɴ/ as allophones of /n/ before velar and uvular consonants respectively
Fricatives: /f s ʃ x/ <f s sh ch> and the uvular fricative <kh>, which it won't let me type for some reason.
Liquids: /l j w/ <l y w>
Vowels: /i u e o a ɑ/ <i u e o ä a>
Syllable structure: CCVC
Illegal sequences: /wu/, /wo/, /uw/, /ji/

So far, I have some sound change ideas, but the words don't change very much, and I need help.
/ɸ/, /ʍ/, and /ɴ/ are great sounds, and nothing you say can convince me otherwise.

Native: :usa: Conversational (on a good day): :bra: Moderate: :epo:
User avatar
Omzinesý
runic
runic
Posts: 3545
Joined: 27 Aug 2010 08:17
Location: nowhere [naʊhɪɚ]

Re: Help With Sound Changes

Post by Omzinesý »

What do you want to have after the sound changes?


Ideas that came to my mind without thinking:

Geminated stops become ejectives
pp tt kk qq -> p' t' k' q'

Palatalization of alveolars before /i/
t -> c etc.
My meta-thread: viewtopic.php?f=6&t=5760
User avatar
qwed117
mongolian
mongolian
Posts: 3998
Joined: 20 Nov 2014 02:27

Re: Help With Sound Changes

Post by qwed117 »

Üdj wrote: 14 Sep 2022 00:09 Hello!

I was hoping to get some suggestions for sound changes in my new language. Sound change has always been the hardest part for me; I just never really have ideas.

Here's the phonology:

Stops: /p b t d k g q/ <p b t d k g q>
Nasals: /m n/ <m n> plus /ŋ ɴ/ as allophones of /n/ before velar and uvular consonants respectively
Fricatives: /f s ʃ x/ <f s sh ch> and the uvular fricative <kh>, which it won't let me type for some reason.
Liquids: /l j w/ <l y w>
Vowels: /i u e o a ɑ/ <i u e o ä a>
Syllable structure: CCVC
Illegal sequences: /wu/, /wo/, /uw/, /ji/

So far, I have some sound change ideas, but the words don't change very much, and I need help.
Can you give us a set of say like 20 example words? Also a lot of the change between languages is grammatical or derivational. There are quite a few words that are more or less identical between Spanish, Sardinian and Italian. If you need ideas you can hit up the Index Diachronica.
Spoiler:
My minicity is Zyphrazia and Novland
What is made of man will crumble away.

The SqwedgePad
User avatar
Üdj
rupestrian
rupestrian
Posts: 22
Joined: 07 Sep 2022 00:24

Re: Help With Sound Changes

Post by Üdj »

Here are some example proto-words:

äqste ['aq.ste]
tachwa ['tɑ.xwɑ]
manqeli [maɴ.'qe.li]
läkh [lax] *the x in this word is supposed to be the uvular fricative, but for some reason, the CBB forum's text boxes just turn that character into x.
eyeta [e.'je.tɑ]
geshalä [ge.'ʃɑ.la]
ngafu ['ŋgɑ.fu]
/ɸ/, /ʍ/, and /ɴ/ are great sounds, and nothing you say can convince me otherwise.

Native: :usa: Conversational (on a good day): :bra: Moderate: :epo:
User avatar
qwed117
mongolian
mongolian
Posts: 3998
Joined: 20 Nov 2014 02:27

Re: Help With Sound Changes

Post by qwed117 »

Üdj wrote: 14 Sep 2022 04:13 Here are some example proto-words:

äqste ['aq.ste]
tachwa ['tɑ.xwɑ]
manqeli [maɴ.'qe.li]
läkh [lax] *the x in this word is supposed to be the uvular fricative, but for some reason, the CBB forum's text boxes just turn that character into x.
eyeta [e.'je.tɑ]
geshalä [ge.'ʃɑ.la]
ngafu ['ŋgɑ.fu]
Known phpBB problem, you can make it a capital 'X' if it becomes a problem. Do you have an idea of the inventory and phonotactics you want to end up with? I'm noticing that your words primarily have penultimate stress, and are mostly disyllables. Do you have an idea of the derivational morphology of the protolanguage?

(Also you can DM me a larger list of proto-words. A lot of changes only become evident when you see close words diverge, and distant words converge)
Spoiler:
My minicity is Zyphrazia and Novland
What is made of man will crumble away.

The SqwedgePad
User avatar
Üdj
rupestrian
rupestrian
Posts: 22
Joined: 07 Sep 2022 00:24

Re: Help With Sound Changes

Post by Üdj »

Some ideas I had for the new phonology:
- Irregular stress is very fun; contrastive stress is even better.
- It might be nice to turn [ŋ] and [ɴ] from allophones of /n/ into true phonemes.
- I definitely want a fun new vowel system; if I could trigger length somehow, I could raise long vowels and get ɛ and ɔ.
/ɸ/, /ʍ/, and /ɴ/ are great sounds, and nothing you say can convince me otherwise.

Native: :usa: Conversational (on a good day): :bra: Moderate: :epo:
User avatar
eldin raigmore
korean
korean
Posts: 6094
Joined: 14 Aug 2010 19:38
Location: SouthEast Michigan

Re: Help With Sound Changes

Post by eldin raigmore »

Üdj wrote: 14 Sep 2022 13:50 - It might be nice to turn [ŋ] and [ɴ] from allophones of /n/ into true phonemes.
I cannot pronounce [ɴ] .
I cannot imagine both [ŋ] and [ɴ] co-occurring in the same phoneme inventory together except as allophones of each other.
I assume it’s attested in some natlang; is it? Does anyone have some examples?

But I can easily imagine neither one being an allophone of /n/.

To me your ideas seem feasible, as long as I’m not the one trying to “feas” them.
What would you hope to accomplish by adding two or so new consonant phonemes and two or so new vowel phonemes?
Technically I suppose “I just want to” is entirely acceptable as an explanation; and technically whatever your explanation, you don’t have to wait on anyone else to accept it.

But usually a conlanger like you seem to be, does have something in mind. Or so I think.

If you don’t, though, you’re in good company, along with e.g. Martin Luther’s preface to his translation of Romans.
User avatar
VaptuantaDoi
greek
greek
Posts: 733
Joined: 18 Nov 2019 07:35

Re: Help With Sound Changes

Post by VaptuantaDoi »

eldin raigmore wrote: 14 Sep 2022 21:52
Üdj wrote: 14 Sep 2022 13:50 - It might be nice to turn [ŋ] and [ɴ] from allophones of /n/ into true phonemes.
I cannot pronounce [ɴ] .
I cannot imagine both [ŋ] and [ɴ] co-occurring in the same phoneme inventory together except as allophones of each other.
I assume it’s attested in some natlang; is it? Does anyone have some examples?
Greenlandic
Salmoneus
MVP
MVP
Posts: 2674
Joined: 19 Sep 2011 19:37

Re: Help With Sound Changes

Post by Salmoneus »

Üdj wrote: 14 Sep 2022 04:13 Here are some example proto-words:

äqste ['aq.ste]
tachwa ['tɑ.xwɑ]
manqeli [maɴ.'qe.li]
läkh [lax] *the x in this word is supposed to be the uvular fricative, but for some reason, the CBB forum's text boxes just turn that character into x.
eyeta [e.'je.tɑ]
geshalä [ge.'ʃɑ.la]
ngafu ['ŋgɑ.fu]
The first thing is to reiterate that words don't have to all look totally different. Changes in grammar, lexical replacement (and semantic shift), and small but systematic sound shifts can yield languages with very different feels. In practice, lots of words can continue to look very similar across thousands of years of evolution!

Famously, there are sentences that are identical in spelling (though slightly different in pronunciation) in English and Afrikans, despite nearly two thousand years of divergent evolution between them:
My pen was in my hand
My hand was in warm water

If we compare the words to the original Proto-Germanic:
my < PGmc mi:naz
hand < handuz
was < was
in < in
warm < warmaz
water < wato:r
pen < Latin penna

As you can see, these words have barely changed! Particularly when you realise the -az ending was probably entirely lost by West Germanic (the latest ancestory of English and Afrikaans) nearly two thousand years ago. [the -uz had lost its -z, though the -u stayed around a little longer, at least in English]

Even more extreme is a sentence like: In mare irato, in subita procella, invoco Te, nostra benigna Stella. This sentence is exactly the same in Latin and in Italian.

[indeed, even when sound changes or replacement do alter individual words, languages can retain very similar phonologies overall. That's how we end up with sentences like I vitelli dei romani sono belli, which have valid translations in both Italian and Latin despite the words being mostly unrelated historically (in Italian it means "the calves of the Romans are beautiful", whereas in Latin it means "Go, Vitellius, to the sound of the Roman war god!").]


So conlangers shouldn't worry too much about totally changing all or most of the words in a language - at least unless they're attempting a very-long-timedepth language family!


That said, a few small changes CAN radically change the appearance of words and the feel of a language quite quickly. In that spirit, here's your wordlist in two daughter languages:

(I'm not suggesting these are particularly great sound changes or interesting daughter languages, just plausible pathways to quite different locations)


Idea one:
Spoiler:
eje>e:
umlaut before *i
unstressed nasals drop, lengthening preceding vowels
/x/ and /X/ drop in codas, lenghtening preceding vowels
breaking before low vowels
intervocalic voicing
loss of final vowels when this doesn't produce a final cluster
etc
aksɛ
tɑ:w
ma:qew
la:
ɛ:ad
gʒɑw
ŋɔf

Idea two:
Spoiler:
chain shift: i > ja, u>i, o>u, ɑ>o
o/e > u/i in unstressed final syllables
vowels develop falling tone before a uvular
uvulars merge with velars
eje > e: > é
vowels develop rising tone after voiced consonants
voicing lost
x > h
final h lost
hw > f
nasal+stop clusters become voiced stops
àsti
tòfu
màgelja

étu
kéala
gofi
User avatar
Üdj
rupestrian
rupestrian
Posts: 22
Joined: 07 Sep 2022 00:24

Re: Help With Sound Changes

Post by Üdj »

Thanks for everyone's help! This is such good advice and will be so useful. (Also, that Italian - Latin thing is very funny.)
/ɸ/, /ʍ/, and /ɴ/ are great sounds, and nothing you say can convince me otherwise.

Native: :usa: Conversational (on a good day): :bra: Moderate: :epo:
User avatar
sangi39
moderator
moderator
Posts: 2888
Joined: 12 Aug 2010 01:53
Location: North Yorkshire, UK

Re: Help With Sound Changes

Post by sangi39 »

/q/ and uvular /x/ could be a nice source of changes in the vowels, e.g. /i u e o a ɑ/ > [ɪ ʊ̙ ɛ ɔ ɑ ɑ] when adjacent to the uvulars within the same syllable:

['aq.ste] > ['ɑq.ste]
[maɴ.'qe.li] > [mɑɴ.'qɛ.li]
[lax] > [lɑx]

This could then be followed by two changes, 1) vowels lengthen before coda /x/ (both the uvular and the velar one), and 2) all of coda /q x x/ (uvular a velar) are dropped, giving:

['aq.ste] > ['ɑ.ste]
[maɴ.'qe.li] > [mɑɴ.'qɛ.li]
[lax] > [lɑ:] (something like <läch>, with the velar fricative, would give [la:] instead

That gives you a phonemic distinction between long as short /i u e o a ɪ ʊ̙ ɛ ɔ ɑ/. After that, you could collapse the vowel system down, e.g.:

/i: u: e: o: a: ɪ: ʊ̙: ɛ: ɔ: ɑ:/ > /i: u: e: o: a: e: o: ɛ: ɔ: ɑ:/
/i u e o a ɪ ʊ̙ ɛ ɔ ɑ/ > /i u e o ɛ e o ɛ a a/

['ɑ.ste] > ['a.ste]
[mɑɴ.'qɛ.li] > [maɴ.'qɛ.li]
[lɑ:] > [lɑ:] ([la:] > [la:])

The majority of your words will probably look the same after this, but it does give you the the option to play around with some fun morphology, like, say, a dative case that appears as a suffix -/q/ that, due to these changes, instead surfaces as a vowel alternation in the daughter languages ['aq.ste] > ['a.ste] vs. ['aq.steq] > ['a.stɛ], and you get a handful of vowel mergers in there as well, which can always be fun [:D]
You can tell the same lie a thousand times,
But it never gets any more true,
So close your eyes once more and once more believe
That they all still believe in you.
Just one time.
User avatar
qwed117
mongolian
mongolian
Posts: 3998
Joined: 20 Nov 2014 02:27

Re: Help With Sound Changes

Post by qwed117 »

Üdj wrote: 14 Sep 2022 13:50 Some ideas I had for the new phonology:
- Irregular stress is very fun; contrastive stress is even better.
- It might be nice to turn [ŋ] and [ɴ] from allophones of /n/ into true phonemes.
- I definitely want a fun new vowel system; if I could trigger length somehow, I could raise long vowels and get ɛ and ɔ.
Here's the couple of things that I've done
Spoiler:
t palatizes to tʃ in front of the front vowels e, i
The uvular consonants 'color' the vowels preceding them, causing the backing of the front vowels, and raising the low back vowel to a mid back vowel, ie. a ɑ e i o u > ɑ o ɤ ɯ o u
Similarly they 'color' and lengthen the vowels that follow them, ie. a ɑ e i o u > ɑ: o: ɤ: ɯ: o: u:.
The vowels o, u trigger ɑ in a previous syllable to raise to o.
The vowels i, u trigger e in a previous syllabe to lengthen to e:.
(This doesn't affect the set of words you posted but coda labial consonants trigger an offglide of the previous vowel)
Coda plosives and uvular fricatives both become a glottal stop. The glottal stop is metathesized to the onset of a syllable that doesn't have an onset, and lengthens syllables where it cannot metathesize.
o, u lengthen in stressed syllables, ɤ ɯ round in all positions.
Vowels are lengthened in stressed syllables
Short a brightens ɛ, except when followed by a nasal
The short vowels lax, ie. e, i, o, u > ɛ, ɪ, ɔ, ʊ, and contrastive length is lost
Voiced consonants become unvoiced in stressed syllables.
The back low vowels merge.
d and x (the velar fricative) becomes r
Consonants following a nasal are lost.
Lax vowels are lost word finally, if they're not preceded by a cluster. Else, they are tensed

'aq.ste → 'ʔa.stʃe
'tɑ.xwɑ → 'ta.Xwa
maɴ.'qe.li → ma.'ɴel
'laX → 'la
e.'je.tɑ → ɛ.'jɛ.ta
ge.'ʃɑ.la → gɛ.'ʃal
'ŋgɑ.fu → 'ŋof
'te.Xan → 'tʃɔ.ran
'po.di → 'por
po.'di.ʃo → po.'tɪ.ʃ
Now, we've achieved a lot of your goals. Stress is a mix of penult and final stress, albeit predictably (and thus not contrastive); /ŋ ɴ/ are both phonemes that can't be easily explained, except as an underlying /q X/. You have a contrastive tense-lax distinction
Admittedly not too much has changed, but with a similar level of lexical and grammatical change, the language would likely be unintelligible to the speakers of the proto-language
Spoiler:
My minicity is Zyphrazia and Novland
What is made of man will crumble away.

The SqwedgePad
User avatar
eldin raigmore
korean
korean
Posts: 6094
Joined: 14 Aug 2010 19:38
Location: SouthEast Michigan

Re: Help With Sound Changes

Post by eldin raigmore »

I cannot imagine both [ŋ] and [ɴ] co-occurring in the same phoneme inventory together except as allophones of each other.
I assume it’s attested in some natlang; is it? Does anyone have some examples?
Greenlandic
I’d rather do without the uvular nasal but include the palatal nasal.
Just because I can pronounce bilabial & alveolar & palatal & velar nasals, but not uvular nasals.
But it looks like Greenlandic does have the nasal phonemes Üdj has been proposing!
Thanks, Vaptuandoi!
Last edited by eldin raigmore on 16 Sep 2022 21:38, edited 2 times in total.
User avatar
Omzinesý
runic
runic
Posts: 3545
Joined: 27 Aug 2010 08:17
Location: nowhere [naʊhɪɚ]

Re: Help With Sound Changes

Post by Omzinesý »

I don't see a problem in /N/.
If /nd/ -> /n/, /ng/ -> //, the old coda /n/ could be pushed to /N/ when not followed by a stop.
My meta-thread: viewtopic.php?f=6&t=5760
Nortaneous
greek
greek
Posts: 663
Joined: 14 Aug 2010 13:28

Re: Help With Sound Changes

Post by Nortaneous »

eldin raigmore wrote: 14 Sep 2022 21:52 I cannot pronounce [ɴ] .
I cannot imagine both [ŋ] and [ɴ] co-occurring in the same phoneme inventory together except as allophones of each other.
I assume it’s attested in some natlang; is it? Does anyone have some examples?
Mapos Buang, some dialects of Bai (PDF), and it's alleged for some Sino-Tibetan varieties spoken in/around Tibet but that's all preliminary fieldwork so who knows.
User avatar
Ahzoh
mongolian
mongolian
Posts: 4134
Joined: 20 Oct 2013 02:57
Location: Canada

Re: Help With Sound Changes

Post by Ahzoh »

non-syllabic /u/ or /w/ become /p/ or /k/ when found in the coda of a syllable. non-syllabic /i/ or /j/ become /k/ (if u>p) or /c/ (if u>k)

ajta > /akta/ or /acta/
sausu > /sapsu/ or /saksu/
Image Śād Warḫālali (Vrkhazhian) [ WIKI | CWS ]
Post Reply