The Art of Language Invention, Episode 27: The Evolution of Tone

This video discusses two strategies for evolving tone in a naturalistic conlang.
EDIT: Uhhhh...it should obviously be "ruedo", not "rodo". Don't even know wtf I was thinking... lol Maybe think of that as Spanish from the 1300s...
LINKS:
(1) 1:30: Consonant Cluster and Coda Evolution: • The Art of Language In...
(2) 2:49: That should be "ruedo". So embarrassed... However, given the other sound changes I used, the correct form wouldn't result in a minimal tonal contrast, so let's just pretend it comes from a long ō in Latin for this...
(3) 6:30: On Opposite Tones in Related Languages: en.wikipedia.o...

[@ArturoStojanoff]

As a native Spanish speaker that Spanish thing was really cool. It really made tone make a lot more sense than it ever had from just reading about it.

[@dsp6373]

In colloquial Chilean Spanish, especially when spoken very quickly, vowels are assimilated to each other, and consonants are elided so much that much of the lexical and grammatical information is now carried by tones. Thus, in Chilean Spanish... De qué habla? = è ke abla Es que habla = é ke abla Es que hablá! = é ke abla’ De que hablada? = è ke ablâ’ ...are now all homophones, but Chileans are using tones to differentiate what their vowel assimilation and consonant elision has turned into homophones. What happened to the once non-tonal Chinese languages which became tonal by elision and assimilation, is actually happening in Chilean Spanish. Spoken colloquial Chilean Spanish is very different to what is the otherwise artificially articulated neutral pronunciation which Chileans might force themselves to speak when they need to be understood by other Spanish-speakers who are not familiar with tones. Non-Chileans find it extremely difficult to understand Chilean Spanish precisely because Chileans are speaking their variety of Spanish with tones now, to distinguish words that Chileans have turned into homophones. Non-Chileans can’t hear the tones because they don’t use tones themselves, so they’re left wondering “what the f*ck are these Chileans saying?”. The above four example sentences, now being homophones in Chilean Spanish, all simply sound like “e ke habla” to non-Chilean speakers of Spanish. The first example was “de qué habla?” (“What is he talking about?”). In non-Chilean Spanish this would be pronounced “ðe ke abla”. The “d” in the word “de” had already become “ð” in standard pronunciations of Spanish. But even this “ð” has now altogether disappeared in Chilean Spanish where, so the word “de” becomes simply “e” but distinguished to Chilean earns by tone. The second example was “es que habla” (“it’s that he’s talking”) which in non-Chilean Spanish is pronounced “es ke abla” but in Chilean Spanish is pronounced “é ke abla”. As you might know already, the “s” in the word “es” has already become aspirated in many Spanish dialects (sounds almost like an English “h” sound), but even this aspirated “h”-sounding “s” has been lost in Chilean Spanish, so now “es” is also simply pronounced “e” but yet again distinguished by a different tone. The third example was “es que hablá!” (“Just [you] speak!”). This is the imperative (command tense) in the voseo form (note, Chileans use verbal voseo more so than pronoun voseo, so it would be “es que [tú] hablá” rather than “es que [vos] hablá”). In any event, the pronunciation here is yet again the same, but the change is simply one of tone, becoming “é ke abla’” in Chilean Spanish. Finally, the fourth example was “de que hablada?” (“what talking to?”). In non-Chilean Spanish pronunciation it would be “ðe ke ablaða”. As mentioned before the “d” had already become pronounced as “ð”, but even this is now lost in some dialects of Spanish where it becomes “abla’a”, but in Chilean Spanish they take it further and assimilate the two vowels (a’a) as one vowel (“â”) and pronounce it as “ablâ’” with a rising and falling tone, so that it becomes “e’ ke ablâ’”.

[@dsp6373]

@Dani Rojas three years later and it’s funny that you replied today of all days because I was listening to the song “Hawái” by Maluma and The Weekend (click kzbin.info/www/bejne/b2LZdnaEndugg7s for the video) and while Maluma is Colombian, not Chilean, I did notice in the Spanish lyrics of the song Maluma pronounces “nada” (“nothing”) not as /‘na.ða/ (Standard Spanish), nor as /‘na.a/ (Caribbean Spanish, including Colombia’s Caribbean coast), but rather he pronounces it as /‘nâ/ with a rising and falling tone. This was the first time I’ve noticed tones being used in Spanish other than Chilean Spanish. I was surprised. Tones are creeping in as Spanish becomes more and more simplified to monosyllables with so many homophones. That’s how Chinese became tonal, the spoken language kept simplifying into monosyllables. Thus, for instances, in the sentence “en la nada” (“in nothing”) which would be pronounced /en la ‘na.ða/ in Standard Spanish, and as /na ‘na.a/ in Caribbean Spanish, would then become “na nâ” in whatever dialect Maluma is singing in, and you can get the meaning just by the tone though both words are homophones. As to the Chilean “po”, I agree.

[@amyoosh]

wow this is all so interesting!

[@virtem7686]

@@dsp6373 I'm also 3 years late, but I will give my opinion too. And I 100% agree that Chilean could become tonal or is in their way to be, everything makes too much sense now that you and this video slap it on my face... Btw in the last example isn't hablada, is hablaba, and in the voseo example have a extra pecularity cause had vulgar chilean own their voseo conjugation, which end on , so instead of hablas is hablai (came from and is some sort of merged singular-plural)

[@dsp6373]

@@virtem7686 I am convinced Chilean is now tonal, but it simply hasn’t been recognized as such. When I hear something in Chilean, and it doesn’t make sense literally from the phonemes I’m hearing, I try to hear if there’s a tone, and then I deconstruct the tone and I either insert the missing consonant or I double the vowel to resuscitate the lost assimilated vowel followed immediately by inserting the missing consonant that existed between the two vowels: I hear “na” and it doesn’t make sense by context, I listen if there’s a tone and sure enough there is a tone, it’s a rising-falling tone “nâ” then > double the vowel to ‘na.a > insert a missing weak consonant to ‘na.ða > replace the weak consonant to the strong counterpart ‘na.da = nada (“nothing”). That’s the only way I make sense of what is being said in Chilean. Eventually, a non-Chilean will get accustomed to their manner of speech if exposed enough (without the Chilean having to revert to full enunciation of Spanish which does not come naturally to them) and I think the tones become subconscious to understand to a non-Chilean but the non-Chilean himself will never employ the tones because it’s just too foreign a concept. It’s one thing to subconsciously register the tones after enough exposure, and it’s another thing altogether to employ it oneself.

[@bluetannery1527]

“Fricatives usually go through a period where they become h before they vanish” That’s exactly what’s happening in Spanish. The silent “h” at the onset of many many Spanish words used to be the unvoiced dental fricative, “f.” See cognates between Spanish “hacer” and French “faire.” Is this prophetic

[@xmvziron]

I think you mean voiceless biliabial fricative.

[@keegster7167]

XMV Ziron It’s neither. It’s labiodental. It’s possible that it was bilabial, maybe, though. I’m not sure. They’re very similar, bilabial and labial fricatives.

[@yeetyeet-jb6nc]

in some dialects all coda fricative go through this change (and the original h came form *x so spanish really likes this sound change i guess)

[@markmayonnaise1163]

Typically a prophecy is a prediction of something in the future, not hundreds of years in the past

[@יואביאייל]

Does this happen in all dialects? Or just some if them?

[@ToqTheWise]

I'm making a mermaid conlang based on an experiment someone on reddit did where they stuck their head in the bathtub to see what sounds they could distinguish from each other. The result is a very limited phonetic inventory. There are only three vowels which are æ, i, and u and the only consonants are p, t, and k. I borrowed Xhosa's clicks for some flavor and originally I had ejectives as well however I was getting lightheaded from trying to pronounce a ejective and a click in the same word. I also decided after almost swallowing my tongue that only one click or at least one kind of click was aloud to occur next to each other. Because pronouncing a word like "xaciqu" is exhausting. So I'm not very good at math but I figure I'm going to need all the lexicon building tricks I can use and tone seems like it would be useful. I'm thinking now I might go the long way and start with a terrestrial proto-lang which evolved into an aquatic...meso-lang after undergoing an extreme sound change and adapt tone from there. ...but that sounds complicated and my girlfriend was giving me crap about how I always get caught up in worldbuilding and never write anything. ...Damnit, I'm going to have to do this aren't I? Wait hold on is that even possible? Like...that's quite the evolution and I think entire words would go extinct. Yeah, I might not then. Well...I'll try a few sample words and see what it sounds like.

[@JayAlcala23]

A topic I'm really interested in and hope you cover is how to construct and evolve triconsonantal root languages like the semitic family. How does one arrive at those funky vowel filling templates? I honestly don't think I'll ever make one, but am REALLY curious as to how they come about.

[@rubbedibubb5017]

kzbin.info/www/bejne/e4GlqqKrZap8mJY here’s a video!

[@Dedalvs]

Short answer is this: Get a bunch of single consonant suffixes that allow for generic semantic extensions plus lots of syncope, epenthesis, and some vowel assimilation.

[@pauleugenio5914]

@@Dedalvs Translation: Put the lime in the coconut and mix it all up.

[@QuotePilgrim]

I checked your channel just a few hours before you uploaded this video and upon seeing the latest video was uploaded 5 months ago I thought you wouldn't upload anymore. It's such a crazy coincidence that the exact day I decide to take a look at the channel you upload a video. I'm glad you haven't stopped making videos.

[@RanmaruRei]

New video, hooray!!! Really, I glad to see new video here after 5 months of silence. They are very educational and quite entertaining.

[@Dedalvs]

I was not prepared for the class I taught at Berkeley this summer, and once I got in the crunch, all my free time was gone. After that, lot of travel, lot of work. But I'm still going to add videos when I can!

[@pmaha5603]

คือดีงามฮะ ผมนี่ไลก์เลย 6:30 hah, so true. Even in different rigional dialects, the tones are different.

[@HBMmaster]

nice

[@faheemsyed1674]

Conlang Critic You should critique his tonal Spanish language when it's done.

[@trafo60]

That's what I was gonna say

[@HBMmaster]

tom this isn't even my video can you not

[@JayAlcala23]

Spice

[@parthiancapitalist2733]

Itzchimalli Metziuhki, we are here after one month. Let's celebrate

[@dsp6373]

In colloquial Chilean Spanish, especially when spoken very quickly, vowels are assimilated to each other, and consonants are elided so much that much of the lexical and grammatical information is now carried by tones. What happened to the once non-tonal Old Chinese languages which became tonal Middle and Modern Chinese through elision and assimilation, is actually happening to Chilean Spanish. Thus, in Chilean Spanish... De qué habla? = è ke abla Es que habla = é ke abla Es que hablá! = é ke abla’ De que hablada? = è ke ablâ’ ...are now all homophones, but Chileans are using tones to differentiate what their vowel assimilation and consonant elision has turned into homophones. Spoken colloquial Chilean Spanish is very different to what is the otherwise artificially articulated neutral pronunciation which Chileans might force themselves to speak when they need to be understood by other Spanish-speakers who are not familiar with tones. Non-Chileans find it extremely difficult to understand Chilean Spanish precisely because Chileans are speaking their variety of Spanish with tones now, to distinguish words that Chileans have turned into homophones. Non-Chileans can’t hear the tones because they don’t use tones themselves, so they’re left wondering “what the f*ck are these Chileans saying?”. The above four example sentences, now being homophones in Chilean Spanish, all simply sound like “e ke habla” to non-Chilean speakers of Spanish. The first example was “de qué habla?” (“What is he talking about?”). In non-Chilean Spanish this would be pronounced “ðe ke abla”. The “d” in the word “de” had already become “ð” in standard pronunciations of Spanish. But even this “ð” has now altogether disappeared in Chilean Spanish where, so the word “de” becomes simply “e” but distinguished to Chilean earns by tone. The second example was “es que habla” (“it’s that he’s talking”) which in non-Chilean Spanish is pronounced “es ke abla” but in Chilean Spanish is pronounced “é ke abla”. As you might know already, the “s” in the word “es” has already become aspirated in many Spanish dialects (sounds almost like an English “h” sound), but even this aspirated “h”-sounding “s” has been lost in Chilean Spanish, so now “es” is also simply pronounced “e” but yet again distinguished by a different tone. The third example was “es que hablá!” (“Just [you] speak!”). This is the imperative (command tense) in the voseo form (note, Chileans use verbal voseo more so than pronoun voseo, so it would be “es que [tú] hablá” rather than “es que [vos] hablá”). In any event, the pronunciation here is yet again the same, but the change is simply one of tone, becoming “é ke abla’” in Chilean Spanish. Finally, the fourth example was “de que hablada?” (“what talking to?”). In non-Chilean Spanish pronunciation it would be “ðe ke ablaða”. As mentioned before the “d” had already become pronounced as “ð”, but even this is now lost in some dialects of Spanish where it becomes “abla’a”, but in Chilean Spanish they take it further and assimilate the two vowels (a’a) as one vowel (“â”) and pronounce it as “ablâ’” with a rising and falling tone, so that it becomes “e’ ke ablâ’”.

[@eheshzoumi7224]

Spanish evolution!!!! I NEED TO SEEE. please Mr. Peterson tell us what the show or name of the language is. I wish to learn.

[@eheshzoumi7224]

? Why did the yuck thing added? I did not add that and I definitely never use the heart emoji and I cant edit it out

[@parthiancapitalist2733]

I thought you said Mr. Postman since I was reading the comments quickly (It's a Beatles song)

[@Dedalvs]

It was _Into the Badlands_. All the dialogue is now up on my AO3: archiveofourown.org/users/Dedalvs/pseuds/Dedalvs/works?fandom_id=7214387

[@McCainenl]

Yay finally a new video! I was considering sending a tweet asking whether you meant to do new ones... It's so helpful for amateur conlangers!

[@jakepenny4366]

Very interesting! As someone who learnt Mandarin before starting Japanese it is interesting how loaning occurs back and forth, especially when I hear a Korean word with a shared root. I'm very interested to see what happens with regard to your tonal Spanish. It would be interesting to see how differences by dialect would affect the end result... This reminds me a lot of pitch accent in Japanese.

[@keegster7167]

Awesome! I have been waiting for an episode for a while.

[@QuotePilgrim]

I have one little nitpick, your phonetic transcription of "hablo" is incorrect, it should be more like [ˈaβ.lo]. I speak a language where [o] and [ɔ] are separate phonemes, and the one you articulated and Spanish speakers always articulate is definitely [o]. Spanish speakers don't hear the difference between the two sounds, though. EDIT: I just noticed you do the same thing with other Spanish words. Spanish speakers always articulate [o], [ɔ] doesn't really exist in Spanish.

[@citrus8788]

Which language do you speak that distinguishes those to sounds?

[@amanciojoao]

@@citrus8788 As a native portuguese speaker, they sound totally different to me.

[@dyld921]

@@citrus8788 Vietnamese distiguishes between [ɔ] and [o], as well as [ɛ] and [e]. We have 11 vowels.

[@asgerhougardmikkelsen8770]

@@citrus8788 danish does

[@MarcosKunBass]

That spanish you made is awesome, as a spanish speaker I like it and I may take it as an inspiration for my short spanish to make my highschool notes shorter and secret

[@willowpalecek7050]

This was a fantastic series and I'm sad it's over.

[@yeryomagabitov8878]

What's a good way for one to go about making a language family in conlang?

[@azhadial7396]

I think the best way is to do things chronologically. You start with your proto-family language. Then you decides that your proto-family language is spoken in different places (countries). Then you evolve your proto-family language differently depending on the places it is spoken leading to similar yet different languages. But you need to find the correct amount of evolution so the languages still look familiar without being almost the same and so the languages are all different without being unrecognizable compared to other languages of the same family. If you do dialects, you simply evolve your language a little bit depending on more precise places (regions of countries). Then you can repeat it again, and again so you end up with a really large language family with sub-families. You also need to consider history. If your language is spoken between 3 countries and 1 of them is conquerred by an other, most likely its language will disappear being replaced by the conquerer's one. Countries/Territories can also be split and lead to different languages. And a language of the family may be affected by another language of another family and will end up looking quite different from the other family members (like English which was highly influenced by French).

[@t.k.abrams4720]

Literally you can do it any way you want. I started by just making one language, then branched from there and the language I started with has ended up being somewhere in the middle of the language family tree.

[@dominicpayne3494]

I'm doing this iteratively (iterationally?), i start with a protolang, build up vocab with appropriate sample sentences. create two or more daughter languages, deciding sound changes for each, then semantic shifts, and any grammemes that emerged from sample sentences in the previous. make sure to keep the ancestor langs for when you find a concept you would need for them, then add them to the daughter's accordingly. the trick is variety, so David's videos on evolution are a good place to start

[@李白-f5u]

That explains all voiced initial consonants in middle Chinese turned into voiceless and the vowels behind them tents to be a lower tone. And some linguists reconstructed old Chinese end with -s as in a falling tone in middle Chinese

[@frederickkeith8054]

It's good both to see a video and to see you're rocking a Defiance shirt.

[@jonseilim4321]

Interesting, so is this how Chinese became a tonal languages? Can we recreate the "original" Chinese language?

[@ArturoStojanoff]

Yes, it's called Old Chinese: "Most scholars now believe that Old Chinese lacked the tones found in later stages of the language, but had optional post-codas *-ʔ and *-s, which developed into the Middle Chinese rising and departing tones respectively." (Wikipedia)

[@trafo60]

Yes. The reconstructed predecessors of Chinese didn't sound like modern Chinese at all, they were much more consonant-heavy. Basically what happened is that final stops were lost, resulting in rising tones; final fricatives lost resulted in falling tones; the remaining syllables had level tone. Later on, distinctions between voiced and voiceless consonants where lost, with the formerly voiced consonants resulting in rising tones. This resulted in another tone split, making for a total of six tones. Mandarin has merged these into "only" four; most other Chinese varieties, such as Cantonese, are more conservative.

[@dsp6373]

Jon Sei Lim, you are correct. Old Chinese was not a tonal language. It developed tones through consonant elision and vowel assimilation. In colloquial Chilean Spanish, especially when spoken very quickly, vowels are assimilated to each other, and consonants are elided so much that much of the lexical and grammatical information is now carried by tones. What happened to the once non-tonal Old Chinese languages which became tonal Middle and Modern Chinese through elision and assimilation, is actually happening to Chilean Spanish. Thus, in Chilean Spanish... De qué habla? = è ke abla Es que habla = é ke abla Es que hablá! = é ke abla’ De que hablada? = è ke ablâ’ ...are now all homophones, but Chileans are using tones to differentiate what their vowel assimilation and consonant elision has turned into homophones. Spoken colloquial Chilean Spanish is very different to what is the otherwise artificially articulated neutral pronunciation which Chileans might force themselves to speak when they need to be understood by other Spanish-speakers who are not familiar with tones. Non-Chileans find it extremely difficult to understand Chilean Spanish precisely because Chileans are speaking their variety of Spanish with tones now, to distinguish words that Chileans have turned into homophones. Non-Chileans can’t hear the tones because they don’t use tones themselves, so they’re left wondering “what the f*ck are these Chileans saying?”. The above four example sentences, now being homophones in Chilean Spanish, all simply sound like “e ke habla” to non-Chilean speakers of Spanish. The first example was “de qué habla?” (“What is he talking about?”). In non-Chilean Spanish this would be pronounced “ðe ke abla”. The “d” in the word “de” had already become “ð” in standard pronunciations of Spanish. But even this “ð” has now altogether disappeared in Chilean Spanish where, so the word “de” becomes simply “e” but distinguished to Chilean earns by tone. The second example was “es que habla” (“it’s that he’s talking”) which in non-Chilean Spanish is pronounced “es ke abla” but in Chilean Spanish is pronounced “é ke abla”. As you might know already, the “s” in the word “es” has already become aspirated in many Spanish dialects (sounds almost like an English “h” sound), but even this aspirated “h”-sounding “s” has been lost in Chilean Spanish, so now “es” is also simply pronounced “e” but yet again distinguished by a different tone. The third example was “es que hablá!” (“Just [you] speak!”). This is the imperative (command tense) in the voseo form (note, Chileans use verbal voseo more so than pronoun voseo, so it would be “es que [tú] hablá” rather than “es que [vos] hablá”). In any event, the pronunciation here is yet again the same, but the change is simply one of tone, becoming “é ke abla’” in Chilean Spanish. Finally, the fourth example was “de que hablada?” (“what talking to?”). In non-Chilean Spanish pronunciation it would be “ðe ke ablaða”. As mentioned before the “d” had already become pronounced as “ð”, but even this is now lost in some dialects of Spanish where it becomes “abla’a”, but in Chilean Spanish they take it further and assimilate the two vowels (a’a) as one vowel (“â”) and pronounce it as “ablâ’” with a rising and falling tone, so that it becomes “e’ ke ablâ’”.

[@WoangLuh]

I talked about this on Quora a few years ago (revised)L www.quora.com/Where-did-Chinese-tones-come-from/answer/Robert-Matthews-%E9%A6%AC%E5%AD%B8%E9%80%B2

[@violet_broregarde]

@@dsp6373 this is fascinating, thank you for sharing :D

[@otakufreak40]

So what did you do with words like creer and leer? Also, [ɔ]?

[@eruyommo]

otakufreak40 Well, you already can say that leer and creer have already a tonal distinction in marked pronunciation. Actually, most of the things he 'invented' are already present in some variants of Spanish. So he was not so original.

[@Dedalvs]

Actually the full set of sound changes is pretty crazy. These were just a couple. _Lot_ of extra stuff happens.

[@OmegaTaishu]

Thanks for this vid, I am a bit interested in tones, and had no idea of how to start evolving those.

[@joshsellers527]

Thanks for this video! It allowed me to create a protolang for my protolang haha! (Bc my current protolang uses level tones and now I can construct backwards from that) That means even more diverse daughter languages!! I'm pretty pumped about how expansive my conworld is gonna be now

[@NaVVtiLuSPS3]

"I rolled" is actually said "ruedo" :v

[@Dedalvs]

LOL OMG! Why am I even like this?!?! I feel like redoing the whole video... SO DUMB!

[@solautumni]

Actually "I roll" would be "ruedo". "I rolled" should be translated as "rodé" :)

[@NaVVtiLuSPS3]

True, actually xD I was thinking about another thing while writing

[@parthiancapitalist2733]

Ruedo my r's

[@Zwerggoldhamster]

Creeper Pro: You roll your arse? XD

[@messenger3478]

Thank you sooo much! This will definitely help me develop a pitch accent Proto Elvish language I've been wanting to make!

[@PhantomKING113]

Minor correction, not important xd: The first person singular present indicative form of the verb "rodar" is not "rodo", but rather "ruedo" (this happens to a bunch of verbs with "o" as their tonic syllable, although it isn't a regular pattern I think, you get things like "tocar" (to touch something or to play music), which goes to "toco"). Stressed o going to ue, as I'm sure you are aware, was a pretty prevalent change in old Spanish, in many words, specially nouns. Also, e sometimes went to ie, I think (this is much more noticeable in Asturian and Galician than in Spanish though). Which brings me to... idk if the vowels would fuse back together or just continue split appart (like ao going to ≈au or something), but what you are doing is still great, as always.

[@t.k.abrams4720]

You should do an episode about signed conlangs and the unique possibilities you have with signed conlangs

[@dustov]

Just moved to Hong Kong so this is interesting on both that personal level and the linguist/conlanger level. Great video.

[@sailorjerry3720]

Stumbled across this whilst seaching the following question: "Do all languages evolve to become tonal?" Now wondering your thoughts. Thanks for your thorough analysis. Very informative. Enjoyed.

[@trafo60]

I've been waiting for this!

[@martinkullberg6718]

This is intresting , I am constructing a rom-lang with a bit of a tone in it, it's called Daoqiänghua cendat . It is a language spoken in a fictional valley in the Himalayas, forming a group of romance languages which Daoqiänghua cendat is the standardised form of. The speakers are descendents of a roman legion send as a gift to a chinese emperor but they ended up in that valley after mixing up with chinese.

[@trickvro]

"Don't take the evolution of that tone with me, young man!" ... I dunno, that just popped into my head.

[@pampelius1267]

Hi David! Thank you so much for your incredibly interesting and useful videos! Can you make one about noun incorporation or how to make a polysynthetic conlang? I can find very little about polysynthesis in other conlang related sources. Thank you!

[@MsLaBajo]

Absolutely fascinating. I’m not a linguist but I have been attempting to speak my heritage language which was tonal but is now extinct. As you can imagine it is a mess figuring out tone because they were not documented for my language but it is tonal. It’s Mangue and it’s northern counterpart is Chiapanec. What was languages went extinct in the 19th century and while they have extensive documentation for lexicon we have nothing for the tonal structure. I think I have to learn everything I possibly can about tonal languages so that I can apply it to my language and reveal the tone of each and every single word in our dictionaries.

[@scptime1188]

Very helpful!

[@widmawod]

Thanks I really needed this video!!

[@mac5565]

The tonogenesis process I've been toying with roughly goes as follows: Syllables with /h/ and /ʔ/ become voiceless and creaky respectively; otherwise unstressed syllables are high and stressed syllables low. Later, voiceless vowels become super-high (presumably through breathy-voice: "breathy > high" is a principle I'm getting from Punjabi) and creaky-voiced vowels super-low. Then, through tone sandhi, loss of final vowels and a few simplifications, we eventually have a six-tone system (high, low, strong and weak falling, strong and weak rising). e.g. [uhˈke] > [u̥ké] > [űké] > [űgé] > [űé̯gĕ] > [weː˦˨g] How justifiable would something like this be?

[@lecreeperweeb4813]

As if Spanish conjugation wasn't enough complicated

[@KaaSerpent]

Where did you go? I just discovered your channel and got to the end. I crave more. :)

[@blacky6552]

3:02 the diacritic for rising tone is an arrow pointing down and for falling tone is an arrow pointing up... I get why, but it still hurts

[@piggyzig]

what do you mean

[@virtem7686]

I agree, I think an would had been better looks good and is more keyboard friendly

[@pradyutdas7358]

2:05 Mandalorian theme starts playing in my brain

[@humanmusic6409]

Time to go make a future tonal English! Now I must decide whether /ð/ mergers with /d/ before or after the development of tone... and would a cluster like /gz/ make a falling or rising tone? Also, could you do an episode on making Pidgins/Creoles?

[@niku..]

Zaylon Ballard voicing is always (or at least very often) associated with a low tone. For /gz/ it depends on what you want to do with it but it would probably create some sort of low (contour) tone.

[@humanmusic6409]

Niku I was thinking more because it's one part stop and one part fricative. Using the rules in the video, /dog/ would become /tǒ/, but would /dogz/ become /tǒ/ or /tò/?

[@niku..]

Zaylon Ballard I don't think there would be a distinction between /dog/ and /dogz/ since the /z/ is just an affix. But you could put a vowel in between g and z so you get two vowels with tone which can lead to tone sandhi. You could make /dog/ have a low rising tone and /dogz/ have a low flat tone. This could also create another distinction: dog > dó > to (imagine this o having a caron) dogz > dogiz > dogi > tòkì (k wouldn't be part of the actual affix which would now be -i)

[@calypso_lazuli]

What about vowels that are creaky voiced/vocal fry, breath, stiff, ect... I find information on these types of vowels very hard to find and I especially am interested in creaky voice. Murmured voice also sounds interesting. Is there any in depth information out there?

[@wtc5198]

Creaky voice and breathy voice can evolve from coda glottal stops and coda [h], respectively

[@Zwerggoldhamster]

Why are rising tones transcribed with an arrow down and falling tones with an arrow pointing up? xD

[@jo7801]

Could you please do an extra video about the Duolingo course for Valyrian?? :)

[@deepsolar169]

Can vowel liaison create different contour tones? Like a high tone next to a low tone, over time the two vowels merge and the tone becomes a high to low contour tone?

[@wtc5198]

yes

[@parthiancapitalist2733]

Also another question: how do consonantal roots arise? I have created some language families, but now I want to make a big one. I am taking inspiration for the concept of Proto-Indo-Uralic (two major language families share a common ancestor), and I want one branch to be have consonantal roots. Should I start with consonantal roots and make one branch agglutinative, or the other way around? Do you think deriving 50 languages is too much? Because if you look at all the languages (not just modern ones) it's like 70

[@BlaherTiger]

Have you got any academical work on that topic?

[@Ofordgabings]

David, are there any auxlangs you like? Also, do you know of good critiques on Esperanto?

[@trafo60]

Conlang Critic

[@Dedalvs]

No, there are no auxlangs I'm a fan as auxlangs. Separate from that requirement, I think Afrihili is pretty cool. This is the usual go to critique of Esperanto: web.archive.org/web/20170718033809/jbr.me.uk/ranto/ Really, though, it should be enough to say that Esperanto fancies itself a universal language drawn entirely from Western European languages.

[@Grakkagrunk]

@@Dedalvs That link was no short critique xD haha

[@t.k.abrams4720]

I have a theory that English is gonna develop more contractions in the next 100 years. A lot of people tend to shorten is and am to 's and 'm in a lot more places

[@asloii_1749]

What about /l/ or /r/ at codas? Which kind of tones would emerge from that?

[@pietertalens1256]

Awesome video. I wish I saw this a year ago haha. (obviously it wasn't up)

[@bluetannery1527]

As a huge lover of Spanish, this is fascinating

[@YaboiFoon]

It would be awesome if you could talk about noun cases.

[@deathbatforever4792]

Hi, David. This last year I've been working on a medieval fantasy story which I would like to write and publish, that lead me to creating one of the most prominent populations for my fictional continent and its language and extensive mythology. If I publish or not a story, I will continue working on the language by myself, but here's the concept behind it: According to the history (mostly known through mythology) of the people that speak it, it's the "language of the conqueror", as these people used to live in different clans, but having in common many customs, before merging all together under the rule of a more powerful clan. My story would take place thousands of years later and the language, now divided in a common form used mostly by high and middle-low classes in formal situations and by foreigners, and a few dialectal variations of it. I think I'm going to use the language as much as I can, seeing as important it is in that world, and here come my questions: about the Romanisation of it to translate the writing system of it (probably runes) into a book. The language (sorry: I still have to decide between a few names for it) ideally features Finnish-like vowel harmony as well as the consonants Þ and Ð. How can I simplify the concept of vowel harmony for possible readers? What's the best way to mark long and short vowels? Should I just give up vowel harmony? I'm guessing Þ and Ð can easily be simplified with "th" and "dh". If you have any suggestions, they're welcome. I'll wait. P.S. If I ever publish a book, I swear there will be a guide to my languages.

[@Dedalvs]

I have an entire video on romanization systems. Also, you shouldn't have to simplify a vowel harmony system for your reader, as I can't imagine there would ever arise a situation where they would need to learn it.

[@deathbatforever4792]

David Peterson Thank you!

[@omiumn.7829]

Really interesting idea, turning Spanish into a tonal language. Is this a personal project or for something? Also, from what I remember, I thought you didn't like making tonal languages because their too confusing for people who want to learn them.

[@oz_jones]

He said it was for a show.

[@jasondubose8160]

So then, since in my dialect of English, final unvoiced stops are glottalized and pretty much all the same thing, we are on our way to achieving tones??? For my dialect at least

[@incredulity]

Jason DuBose maybe

[@swagmund_freud6669]

No. Final unvoiced stops becoming glottalized happens all the time in languages. It's why Hawai'ian and Japanese have such restrictive phonologies. It's why T in French at the end of a word is silent. There are other things that need to happen for a tone system to appear that David brings up in the video (not saying these other steps couldn't occur some time in the future though).

[@יואביאייל]

Hard to know, it would be really cool if it will happen, though

[@sofijeffrey9797]

I have a suggestion for an upcoming video. Please do one on conscripts and how to make them. This is where I have been stuck with my conlang Ngma for a while now

[@stjacquesremi]

I wonder if in a hypothetical timeline, the Latin language (that has phonemic length distinction, like short, long, and in possibly rarer cases, over long), could developpe tones, and I wonder how it might look like after tonigenesis

[@Salsmachev]

So, if a language has a set of ejectives and a set of implosives and the implosives become ejectives, should the high tone on the following vowel go on the syllables that were originally ejectives or the ones that were previously implosives?

[@Mr.Nichan]

I'm currently woking on sound changes for a tonal conlang and I'm wondering whether some of my rules make any sense: I have three final fricatives: /s/, /x/, and /h/. They change in the following ways (essentially): Vh > V̤ (falling/low tone) Vx > Vh (falling/low tone) Vs > Vh (rising/high tone) My original thought was that it made since because [s] is overall higher pitched than [x].

[@fronk850]

I think it has a lot more to do with where your voice box is, which is affected by how tense or lax the vowels are, which might be affected by the consonants. This means that the position of the tongue and voice box matters more than the actual sound produced. This probably isn't helpful though, being a year after, lol.

[@Mr.Nichan]

@@fronk850 I still haven't gone much further with that particular conlang, though.

[@mayueno]

NICE

[@sheldonebbeler6125]

When a language loses tonal contrasts, do the reverse trends {ever/generally} occur? For example, could loss of a tonal system result in additional contrasts between Cs (e.g, voicing contrasts)?

[@rubbedibubb5017]

Does it work the same way for pitch accent languages?

[@milanschouten6533]

Why did you stop??

[@יואביאייל]

Doesn't the first way leads to Pitch Accent? I thought that languages with Pitch Accent like Swedish don't count as tonal language.

[@leonfricker8096]

if a lexeme becomes grammaticalised can it still also be a functional lexeme in the language? for example if i evolve an agentive suffix from a word meaning man lani = work, pato = man lanipau = worker, can pato still be an independent lexeme? would it also be phonologically reduced?

[@rredd7777]

Nice vid. So, is tone limited to these instances of consonant loss and distinction loss, or, do languages ever start using tone as a variable where there was no consonant/distinction loss? Say a language becomes tonal from one of these mechanisms, might it ever begin having tone in instances unrelated to the above losses?

[@niku..]

Brad Lewis it could happen because of loan words from tonal languages. But this wouldn't affect native vocabulary

[@Dedalvs]

I don't know enough to say "no" for certain, but of all the cases I know of tone languages all seem to come from toneless ones at some point. I think it's certainly a possibility the language could just start out that way.

[@dominicpayne3494]

do you have any advice on producing the alveolar trill? the many many videos out there havent been very helpful, and it's starting to seem like sorcery. ive produced it on accident but can never replicate it since. and more to do with conlanging and perhaps a more loaded question, how might trills behave as a language evolves? im working on a language family, and the mother language has the bi-labial and linguo-labial trills. as you might figure, the Index Diachronica doesnt have a lot on these sounds. how might these consonants change in languages that lose them?

[@jake-jk8rj]

Linguolabial trill... *Laughs* Maybe have it so the daughter languages lost it because it sounded weird to foreigners and they didn't want to be laughed at.

[@dominicpayne3494]

First Last oh no no, well, maybe for a few daughters. I included it with the intent of a silly phonology. i guess a linguo-labial fricative can replace it in a few daughters

[@calypso_lazuli]

The best way to learn it is to take a word your familiar with like "water" or "ladder", you already know colloquially this is an "r" sound (I.e., warer, larer) and then take that "r" sound and now try to trill/roll it, warrrrrrrrer, larrrrrrrer.. Start incorporating it in Spanish words like perrrrrro, rrrrrrrrosa... Eventually you'll get it!!

[@contrapunctusmammalia3993]

What exactly is the perfective aspect in the present tense for verbs that aren't stative?

[@jacksonp2397]

My latest conlang uses tone to diffferentiate objects from subjects considering they are right next to each other in a sentence. This is rare to occur in a language, however, in my conlang many sentences have the subject or object implied since native speakers are quicker to point then to speak. The subject and object both occur at the beginning of the sentence. Also, case is not applied to pronouns, so the sentence "I kill" is ambiguous of whether or not it is stating I kill or commanding Kill me. Over time citizens began to use tone to differentiate between object and subject.

[@Dedalvs]

But _not_ in pronouns...?

[@jacksonp2397]

@@Dedalvs I'm just now looking at this comment 2 years later, and it's crazy to think how much my skills and preferences in conlanging have changed. It's been quite the journey; a journey no where near ending, thanks primarily to the conlanging community. Thanks David!

[@copyplanter]

Rô could have evolved from roo, a possible first person present singular form of roer (to gnaw). But I like your solution in the info box :)

[@Dedalvs]

I did not know that word. _Almost_ perfect (it's derives from Latin *ō with the missing *d I was looking for), but its past 3sg is royó. Cop out! lol

[@DTux5249]

Also love the vid

[@nasimakhtar8850]

Can you suggest an article on this topic?

[@Dedalvs]

There is an article on using tone in a conlang by Aidan Aannestad on Fiat Lingua here: fiatlingua.org/2018/04/

[@DTux5249]

Maybe ways to make a part tense

[@DarkrarLetsPlay]

Will you ever do more videos?

[@forcedcomicrelief3012]

Would it be possible for a language to lose its tones?

[@apta9931]

How would losing a schwa affect tone?

[@t.k.abrams4720]

Season 5 of The 100 I bet

[@VictorSantos-gy4uj]

He uses the word "tone" incorrectly at times. What he means is "stress", not tone.

[@filipinojalapeno1527]

how far did you get in the video

[@wolfgangalphamale1268]

As spanish native speaker, I find your evolution a little weird. The loss of the intervocalic voiced fricatives is very normal, especially in some dialects. However the evolution of the dipthongs/hiatus ao/au and eo/eu to o, in those positions is akward. I think they would probably go to au, which is a very stable dipthong, and eu, in tonic position, and maybe o in non-tonic. Also the loss of intervocalic /x/ is odd, because is a very strong sound, it is more uvular than velar.

[@Dedalvs]

Sounds changes can always happen or not happen. There's a _lot_ of time that passes here, if I'm working with the full set of sound changes I implemented. (Also, I never said *eo/*eu went to o.)

[@daeHknuJ]

I'm assuming you're a Latin American Spanish speaker, in which case you likely pronounce ⟨j⟩ as the glottal fricative [h], whereas a peninsular Spanish speaker would pronounce it as the velar [x]. This, perhaps, supports the plausibility of the intervocalic loss in the example. It is to be assumed that Latin American Spanish would exhibit different changes.

[@wolfgangalphamale1268]

daeHknuJ I'm a spaniard, so we pronounce the /x/ very strong, except in the south; but I agree, maybe it would be lost in Latin American spanish. I made a mistake with the eo/eu, which was ae/ai. In my opinion, ae would evolve either to ai or e, and ai would stay the same, because is a very strong dipthong

[@masicbemester]

ngl kinda want to make Tonal French

[@amtc4ihs2005]

Are there any languages out there with 32 tones?

[@rikkiegieler5638]

How did grammatical gender come about?

[@parthiancapitalist2733]

Rikkie Gieler he has a video on it