#tonogenesis

[caesar-blogi]

The Emergence of Tone in Korean

This one is going to be a long post but hopefully it's still approachable even if you don't know about linguistics. I often make posts like this in my free time so let me know if you want me to post them to Tumblr more

Context: Other (Asian) Tonal Languages

One of the most well known features of Eastern and South-Eastern Asian languages is tone, the use of pitch to distinguish different words. In this map, white languages have no tones, pink have simple tones, and red have complex tones

Still, just in case you somehow don't know what tone is or just need a refresher, here are some examples from actual languages:

Mandarin Chinese 1. Flat 媽 mā ("mother") 2. Rising 麻 má ("numb") 3. Dipping 馬 mǎ ("horse") 4. Falling 罵 mà ("to scold") 5. Neutral 嗎 ma ("question marker")

Vietnamese 1. Flat ba ("three") 2. Low Falling bà ("grandmother") 3. Dipping bả ("bait") 4. Broken Rising bã ("residue") 5. Rising bá ("to embrace") 6. Mid Falling bạ ("to strengthen")

Japanese Even Japanese can be described as having tones. In English, stressed syllables are louder and longer than the surrounding syllables. In Japanese, stressed syllables have a higher pitch than other ones. "Pitch accent" can also be described as a simple tone system. While a controversial idea to some linguists, I think this makes perfect sense. In essence, Japanese only has a High and Low tone

Korean "Stop" Series

There's a class of consonants known as plosives or stops. They're called that because you completely stop airflow when you say them. Compare /s/ to /p/ (as in "sit" versus "pit"). When saying /p/, you build up air before releasing it

One more important thing to know, consonants can be said with an extra puff of air. We call these consonants aspirated. Hold your hand out flat, with your fingers together. Touch the tips of your fingers to your nose. If you say "spit" and "pit" back and forth you'll feel the difference. With "pit" there is a very strong puff of air that hits your palm. This is the aspirated sound [pʰ]. With "spit" it's a normal [p]. English doesn't distinguish these sounds, but Korean does

Korean has three sets of stops. Aspirated ㅍ /pʰ/, Plain ㅂ /p/, and Tense ㅃ /p͈/. This last one is fascinating. Natives I've asked have describe it as being pronounced "forcefully", calling it "narrow" or "strong". However, this is very different from reality. In truth the plain series is also aspirated, just less so. In speech, the series of sounds are actually Very Aspirated ㅍ [pʰʰ], Aspirated ㅂ [pʰ], and Plain ㅃ [p]

Gyeonggi Dialect

Around Seoul, there has been a merger of Aspirated and Plain stops. They are both said with the same amount of air. Words like "reward" 포상 /pʰosʰaŋ/ and "compensation" 보상 /posʰaŋ/ are now both pronounced [pʰo̞sʰaŋ]

However, a new contrast has emerged in this dialect. Historic Aspirates leave behind a high tone while historic Plain stops leave behind a mid or low tone. So they can still be distinguished by this feature. We have 포상 /pʰosʰaŋ/ [pʰo̞˥sʰaŋ] versus 보상 /posʰaŋ/ [pʰo̞˧sʰaŋ]

Conclusion

There isn't much more to say about this topic. In general, this is how tones develop. Speech is governed by the principle of minimal effort. People will pronounce things in the easiest way possible. Because the "ee" sound /i/ is said with the tongue very far forward, consonants will be drawn forward when near an /i/. This is how soft C came to be. When near /i/ there was a change /k/ > /kʲ ~ c/ > /ts/ > /s/. In the same way, certain consonants cause syllables to be pronounced with different pitches. If those consonants are lost, only the pitches remain and tone becomes contrastive. It's no longer a background feature speakers produce out of convenience, it's as basic as any consonant or vowel for distinguishing words

[dedalvs]

Hi I've been working on a conlang that underwent tonogenesis by losing coda non elective stops and becoming high tone (coda elective stops then become plain voiceless, giving a fun way for my language to be divided into different initials and finals) and low tone coming from the coda fricatives [s] [ɕ] and [ʑ], with mid tone happening in open syllables or with coda sonorants (though 2 of those sonorants from the proto lang [r] and [r̥] are now fricatives [ʁ] and [χ]). I'm also debating having intervocalic /s/ debuccalize to [h] and then delete and get contours from those new diphthongs like this: [ˈʊ˩.sə˧.nʊ˩] > [ˈʊ˩.ə˧.nʊ˩] > [ˈʊə˧˥.nʊ˩] > [wə˧˥.nʊ˩] but I'm still not feeling satisfied with this but then I remembered reading yours and Jessie Peterson's posts on instagram about Ts'ítsàsh some months ago where you guys had tonogenesis by consonant loss and also stress and I was curious about how that worked in Ts'ítsàsh because I think that could be what I'm missing, especially because stress in the proto language was based on morae and was therefore not fixed and I want my conlang's tone to be more dynamic. Sorry for the long introduction and thank you for your time.

What is an elective stop...? I've never heard that term.

Essentially, stress is another variable you can play with. You can ignore, as your system here has, or you can add it back in to potentially create even more tonal distinctions in your language. Stress is associated with a number of elements, depending on the language: volume, vowel length, pitch level, or some combination of these. The goal is to somehow give a particular syllable more prominence than other syllables. If stress is associated with pitch, well, tone is also associated with pitch. You can imagine, then, you have two syllables, however this works (i.e. whatever needs to happen to get you these two proto-syllables):

*pat

*ˈpat

Now you've got options. You already know that a loss of something like [t] in coda position results in high tone on the previous syllable. Throw in stress, and you have some options. Here are some examples. First, there's option A, which is what you have above:

*pat > pa˥

*ˈpat > pa˥

That is, stress plays no role, and it's simply the loss of the consonant that gives the syllable tone. Let's call this option B:

*pat > pa˦

*ˈpat > pa˥

This kind of additive tone. The loss of the coda [t] pushes the pitch one point higher, while older stress pushes the pitch one point higher. This results in a high tone and a super high tone, which is now a regular distinction your language makes.

Now here's an option C:

*pat > pa˧˥

*ˈpat > pa˥

Some tone languages don't allow contour tones on short syllables, but if yours does, this is another option. With the loss of [t] and stress you get a solid high tone. The the loss of [t] and no stress, however, the tone is dragged down a bit, resulting in a rising tone.

This is how an older stress system can play into creating a more complex tone system down the line. You have to be meticulous in writing down the rules, and, both Jessie and I can tell you from experience, it's easy to forget them, but it can lead to some more grist, if that's what you're after.

Again, though, this is not something you have to do in creating a tone language. Often stress plays no role in the development of tone.

[viaducdamarube]

My proto-form of my conlang is done! In its future form, it will be modelled after Vietnamese and Mandarin, but for now it doesn't have any tones, as those will have to be evolved later - I might go through how to do tonogenesis on here when I actually do that. BUT, like Mandarin and Vietnamese, it is very isolating. It obviously might not stay that way forever but for now it is. I've done three-ish pages of grammar (in a physical notebook because I'm archaic) as well as a page of introduction. If anyone is interested, I'll post a translation of, I don't know, the Lord's prayer? Or perhaps some short Buddhist tale, since I've found a couple I like. It's called dagna dai phot bos at the moment and I may post about it more later!

[forgottenbones]

The Punjabi language is fascinating!

[datasoong47]