@@OkandUtlanning it's just pretending that different dialects have comparable phonemes. You could make a diaphoneme for every reflex of a PIE consonant and pretend like all of Europe has the same phonology.
Romanization systems certainly have their place, but broad IPA does too. The first half of this video just seems like pointing out things that are certainly true about the IPA, but I don't understand why they're issues. Your argument seems to be that because broad IPA doesn't provide the precise pronunciation of phonemes, it shouldn't share symbols with narrow IPA. But the problem with that is that broad IPA *does* indeed provide the pronunciation. [θɹ] is, indeed, a valid pronunciation of /θɹ/, same with [θɹ̪ʷ]. Sure, you would never hear a GA speaker say [sænd], but that doesn't matter. Broad IPA gives you the important information about phonemes so that you are understandable to a native speaker. Narrow IPA gives you the minute details about sounds so that you are natural sounding to a native speaker. Both have they're place, and I think it's strange and downright untrue to say that broad IPA doesn't give you the pronunciation. Even if broad IPA didn't give you anything even close to the correct pronunciation, that doesn't matter because narrow IPA exists, and we know if a transcription is phonetic or phonemic by those fancy little brackets or backslashes. If people are confused about which is which, then that's the people's fault, not the IPA's. I totally agree with you that IPA looks ugly and is hard to type; that's why we have romanization. In fact, many romanization schemes are semi-phonetic rather than phonemic. Nobody can agree on how many phonemes are in Mandarin, so we just use Pinyin, which is phonetic, not phonemic. IPA and romanizations both serve different purposes, and there's no reason to replace one with the other. I love your passion for phonemic/phonetics, but I don't think you're right about this.
Wow this is... Something. 1. Phonemes and Phones are irreconcilable? Why's that? If we ignore using the wrong symbol for a specific phone, and let's ignore vowels because it's like trying to say at what point does red become blue, [p] represents a phone that is created with the top and bottom lips and does not make use of the vocal cords. /p/ represents much the same, but when you could have variants of the sound, such as one where it's voiced or aspirated, but the language, whichever one it may be, does not care about these phonetic differences when trying to differentiate words... then what matters is /p/, not [p~b~pʰ] (although doing the latter does help you sound native). 2. "/rɨba/ is bad because we use IPA." Saying this doesn't make it true. Why is using /ɨ/ bad because it's IPA inherently? What matters is that /ɨ/ represents a vowel with formant values probably between 240-250F1 and 2400-595F2 (i.e., somewhere high in the middle of the mouth with unrounded lips). Of course, if 95% of Polish speakers say [ɪ] (at least as of the 21st century), then why would changing the symbol be a bad thing? 3. Speaking of... Let's go back to the first example, if I hear that a language has a /p/ phoneme, and I use [p] where it's actually not used, that's not the fault of the IPA for writing down phonemes, that's the fault of the resources I used to learn the language for not describing that it's [b] between vowels, especially if this allophony has been well recognized since . It should be retired because people may pronounce /p/ wrong in the wrong phonetic environment? Sounds like you need a better teacher. 4. Phonemes are arguably specific to *people*. My /i/, no matter how the hell that's actually pronounced, will *never* be the same as your /i/, as a woman's /i/, as a child's /i/. A trans person's /i/ isn't the same from before or after their transition. But yet, our brains still all process it as /i/. The only time we don't process it as /i/ is if the allophony is broken, or the wrong phoneme is used entirely. Part of what makes the IPA useful is it's _standard_ . When I see /i/, I know what that sounds like regardless of language or voice. 5. Why does it matter how Proto-Germanic is written? Side note, I wonder how hard it'd be to do historical linguistics if you didn't have phonemes or at least phonetic descriptions of languages. How would we know that English linguists from, I dunno, 1400 know how to pronounce as /feːs/ even though today most English speakers don't. It's much easier to say, "The vowel in is /eː/" than "The tongue is in the middle of the mouth and held for long and sounds like French's ". Today we do have the formant values but that doesn't work for consonants I think. Also PIE is romanized poorly, at least when it comes to syllabic consonants and I feel bad for those who didn't realize that. But that's not the IPA's fault, that's whoever decided to romanize PIE like that. We write historic languages in phonemes btw, when we're talking about the phonemic (and phonetic btw) realizations of such. I can go to wiktionary, look up 'cat', and find both the spelling and the IPA /ˈkɑt.tuz/. 6. What does it matter where English's /θ/ comes from in relation to Spanish's /θ/? This is a non-sequitur. I would bet money that if you made a Spanish person learn English, they would go; "Oh, they spell /θ/ as th, while we spell it as c/z". They would recognize that their /θ/, which doesn't exist in all dialects btw, is the same phone as English's /θ/, regardless of the fact that Spanish's derives from Latin's palatalized C and English's from PIE's spiriantized /t/. That's not to mention that I, as an English speaker, can understand Spanish's /θ/, I see it as, for the lack of a better term, the same phoneme as English's /θ/. Also, can you provide a source for how different, phonetically or not, Spanish /θ/ is to English's? 7. At 3:23... Yes. 8. Double brackets are cool. That's not an argument against IPA imo. Also, maybe I'm just looking in the wrong places but could you also provide a source for that? 9. Phonemic transcription is... To represent the phonemes accurately. Specifically _the phonemes_. What matters when talking to English speakers is /w/, not [o̯]. [o̯] is not the phoneme most people would use, I suspect, assuming you didn't make that up to make your point. I'm not saying it's a bad thing you did, just saying that even if true it still doesn't prove anything. 10. Does IPA look ugly? I mean yeah, kinda. The only way to fix that is to use, I dunno, Korean? That literally makes its characters built on how the phones are made! Write a bilabial stop like /=/ (not an actual equals sign, just two lines representing the lips). Oh wait but that's hard to type, isn't it? Having to make an entirely new writing system?? Besides I kinda like IPA's aesthetic or lack thereof. 11. Diacritic spam can be an issue I suppose, but one that's infrequent and can be made of actual languages imo. This argument is against diacritics, not IPA. 12. "Oh noes, I have to go to a webzsite to type some characters that are not easily accessible to me!! this is the worst writing system ever!!" Is not an argument. It's whining. What if there were a way to easily type IPA on a keyboard (like on phones for example. I downloaded an IPA keyboard on my phone but I don't use it often because I don't do conlanging and such on the phone). 13. X-Sampa looks worse than IPA and still has the same problems in theory it's just less condensed (I'd rather look at and go to a website to type /ʙ/ than ever type B\ because it looks stupid). It's IPA for computer and we've moved passed the need for it. Besides, IPA _is_ 'metric,' X-Sampa is defined by IPA kind of like US Customary units. h2o\ or whatever the X-Sampa was looks... horrifying. I'd rather write /ɦɔ/.
1. I honestly have no idea how to take this. You tried to prove the point that broad and narrow transcription are reconcilable. The example you gave was "a narrow transcription represents a single specific phone," and "a broad transcription represents a single specific phoneme." Where in that did you connect the two? You even pointed out that the broad transcription doesn't care about subtle differences in allophony. Narrow transcription...does? You've tried to argue they are comparable because they use similar symbols, which is EXACTLY the misinformation i was trying to point out that broad transcription IPA spreads. 2. Because the IPA sets up exectations in narrow transcription, that broad transcription fails to meet (by virtue of WHAT IT IS). When I made the first video, some guy yelled at me in the comments because "The polish y vowel is [ɘ], not [ɪ]!!!!" There's obviously no way for a single phoneme symbol to reflect every realization. So what should it attempt to reflect? My argument is that it should attempt to reflect how the phoneme is perceived, which in this case would be /y/. Although I suppose there is an argument to be made that it should reflect the most common pronunciation. 3. If we're writing down all the ways that phoneme can be realized, what's the benefit of writing it using the "default" one? If people have, as you put it, a good teacher, they already understand it. So now that we've satisifed the linguists, maybe we should satisfy native speakers of the language, by making it something they can understand. 4. I'm convinced you've written one of my own talking points here, in the hopes of tricking me into arguing against myself. Yes, the phoneme /i/ is pronounced infinitely many ways. Yes, we still perceive it as /i/. My question is, why do we perceive it like that? I would argue that it's partially because we write it as /i/. That being the case, my argument would say that's a reason to use it as a broad transcription symbol, but that does technically supercede the IPA, which currently uses /iː/. 5. Because it's a language. Are you talking about "wikitionary uses a transcription for this one word I looked up" or "people constantly and consistently use an IPA-based broad transcription in addition to the * thing" 6. It matters because that's how phonemes are described. Right after Grimm's Law, could you say that the phoneme *t was realized as [θ] in Germanic, and [t] in the standard dialects? Yes, you could say that. That's how phonemes work, they were historically the same phone. And it doesn't matter if you perceive the DEFAULT pronunciations similar. Because they AREN'T ALWAYS PRONOUNCED THAT WAY. 7. Sí 8. They aren't currently in the IPA, so it's an indictment of the IPA that it doesn't already have that. And as far as I can tell, Wikipedia has MASSIVELY overblown the popularity of using them, so there might not BE any sources. 9. Guess what old buddy old pal? You're right! [o̯] is not the phoneme most people would use, because it's a phone, not a phoneme! And it's at least the realization I use. What I'm doing is TRYING to represent the phonemes accurately. And that entails NOT using the ipa, which is going to be inaccurate because it confuses phonemes with phones. What is accurate is writing broad transcription to match the ways people perceive the sounds. 10. This feels like a strawman, but this is fully subjective and not a point that need to be argued. 11. The diacritics are PART OF THE IPA! How am i meant to argue against them without arguing against the ipa. 12. Hey if we can improve it why not 13. Yes, X-sampa has the exact same problems as the IPA (since it's just a reskin). It's all arbitrary symbols, why not just use the one that's easier to type? Honestly, I really shouldn't have to respond to anything after #1. If you don't understand that phonemes and phones are different, you can't understand any of what i'm laying out.
@@zzineohp 1. I don't understand your point. Saying things doesn't make them true and this video doesn't prove the initial point. 2. Let me spell this out for you... If you *know* that Broad Transcription gives you phonemes, and that Narrow Transcription gives you allophones, regardless if you have Narrow Transcription, you should be able to tell how a language is generally pronounced. It is INTENTIONAL. Metaphorically speaking, you're don't like hammers because they can hammer in nails into wood. That's the bloody point! Broad transcription tells you phonemes, it uses IPA characters BECAUSE BROAD TRANSCRIPTION IS A WAY TO USE IPA. Do you want there to be TWO phonetic alphabets based on phonemes and phones!? 3. The teacher is Narrow transcription, or at least an actual lecturer telling you such. Sometimes _it's better to read IPA than having to intuit systems you don't know._ 4. If that's one of your points then you mangled it. If I heard a Spanish speaker say sí, and I heard an English speaker say see, then I would still realize both and as /i/. I don't know why it works like that, ask Chomsky or someone in the neuroscience-y aspect of linguistics. 5. Proto-Germanic, as it is currently understood and presented, is technically not a language. It's a hypothetical reconstruction of the most recent ancestor of all Germanic languages. It barely has any concrete reality, unlike say Latin. That's not to say that PG isn't real but it's a reconstruction, we don't have many (if any) records of Proto-Germanic. We have Gothic and Old Norse, but not Germanic. Besides, we still use IPA to show how languages evolved, Latin or Latin't. 6. BROAD TRANSCRIPTION IS MEANT TO BE BROAD. Accent or otherwise if all that matters between English and Spanish's dental fricatives is that you have the tip of the tongue at the teeth and you push the air out of your lungs without using your voicebox, THAT is the sound that matters. THAT is /θ/! Using this pseudo-Etymological fallacy is irrelevant. What does it matter that English says father where Latin says pater? At the point of Proto-Germanic, at least in theory, the speakers _probably_ didn't recognize /b, d, f/ as /bʱ, b, p/, and we definitely don't now! Besides, historical spelling is worse, "Long and short I" being/aj/ and /ɪ/??? That's NOT EVEN LINGUISTIC THAT'S JUST WRONG! That yelling wasn't at you btw. 7. Okay let's skip seven from now on. 8. You literally showed a Wikipedia page in the video... 9. Your personal phonetic realization of /w/ as [o̯] is not universal and that's why, even in other people's speech, it's not generally written as [o̯] because that corresponds to /w/ (and at least for me, IS [w]). Besides, here again broad transcription is doing its job. If you encounter a word with /w/ and you know how to pronounce /w/ even if you say [o̯], then what's the problem? We're not catering to YOUR specific idiolect? 10. It isn't a strawman but all I'm gonna say is literally: "We should not use X because it looks ugly," is not a valid reason, REGARDLESS if there are other reasons. 11. My point being that languages use diacritics and you'd have to zoom in on those too. Ça va, sí, hündchen, etc. Even Proto-Germanic does it, and worse with the (over)long nasal vowels! 12. idk what to respond to this, I'll just re-iterate the point that "We can't type with it" isn't very convincing because, at least as an American, we know that if we wanna type with other scripts we gotta download a keyboard for that anyways! 13. See point 12: if we hypothetically developed a keyboard (even if you had to download it) to use IPA, this would be BEYOND irrelevant. I should fucking know that phones and phonemes are different, I've told you the definition multiple times if I remember correctly. At the very least I am a conlanger who uses the IPA a LOT. I don't see the issue in using phonetic symbols to represent the phonemes of a language, how the fuck else are you supposed to do it? Apologies for vulgar language.
you've convinced me, I've had so many headaches realizing and finding the ipa equivalent phonemes that I pronounce or want in my short conlangs, it's horrilbe to try and type out, and I arrived at a homebrewed Xsampa (sorta) on my own to skip the ipa keyboard step in the end too
Wrote out a whole comment about my allegiance to the IPA and deleted it because I choose happiness. anyway you should make a video about ancient greek and the laryngeals
Two things 1. They should really make an IME that converts Xsampa to IPA when typing, like Chinese IME that allows you to use a qwerty keyboard to type pinyin and it'll show you the available hanzi you can choose and convert it. 2. I wonder if you can make an alphabet that to specifically write phones, the actual acoustic sound, instead of phonemes, regardless of intent or priming. Mostly because I want to be able to transcribe the Yanny-Laurel and ua-cam.com/video/7Yt6EfOCBwQ/v-deo.htmlsi=g8oB_FVTGOFbRvvZ
I think the contradictions in your channel are hilarious, dont let anyone get you down. if you ever find a girlfriend just dont show her your youtube channel until 6 months in
I really hope my comment (the third one at 2:15) didn’t read the wrong way. I wanted to express doubt. At no point did I think “he used that symbol wrong; I should tell him”, and I hate thinking someone might’ve thought I thought that way
Yeah, i always feel bad about using people's comments, and I definitely appreciate all the times where I legitimately get something wrong, I just thought it was funny how many people there were on this instance
it's still a helpful tool dog, retiring it all together and using phonetic alphabets for every different language / group of closely related languages just to avoid these issues is ridiculous, the cons are such small nitpicks, especially in comparison to the pros of having a universal phonetic alphabet.
It's good for Narrow Transcription, and my whole argument is about how that makes it bad for broad transcription. And we don't have to make up a new phonetic alphabet, you already have a different one for every language. But we keep pretending they're all part of the same system
@@zzineohp “We”? Some do, but linguists understand the difference between phonemes and phones, that’s why we write /t/ and [t] differently. You are annoyed at a misconception that perhaps was influenced by this dual purpose, but I dunno, I don’t think it’s as common as you think? But what would I know, I live in a hole and don’t have friends.
@@zzineohp The whole point of broad transcription is to give a general idea, and not high accuracy. Even narrow transcription isn't perfect, because every individual has small differences in pronunciation of the same sound for any language. And IPA is useful for talking about sounds in writing without the need for audio samples and also without the need for learning how to read every single language (which is 7,139 languages officially). This type of videos of yours are really seems like just a rage bait for engagement anyways.
@@tovarishchfeixiao no I understand that is currently the point of broad transcription. My point with this video is that a) it doesn't do that job particularly well and b) there are other things it can do instead of that. If you wanna debate me on that, fine, but don't take it for granted that giving a general idea is the only important job.
Sure extremely narrow transcription or transcription of incredibly rare phones is a hassle and there are good reasons to use XSAMPA in certain situations (I have used XSAMPA and I csn testify it can be incredibly useful) but as you've clearly demonstrated, it's infinitely more bothersome to read the more it deviates from the unmodified letters. If I'm reading or writing a text about Danish phonology as someone who has good enough knowledge of the language, I of course already know that the Danish /ð/ is very different from the English /ð/ but the point of using the IPA for the transcription of phonemes isn't to give an exact pronunciation guide but to represent this abstract concept with a reasonable enough symbol. Any other symbol to represent that Danish phoneme would completely miss the mark - how do I know that for sure? Because if you know a little bit about Danish you already know which phoneme I'm talking about. I'm just never gonna write [D_-_o_e] unless I'm absolutely forced to
Traditional Danish phonetic transcription is an impressionistic mess loosely based on IPA at best. I've only found one book by Ruben Schachtenhaufen that's legible if your native language isn't Danish and you already know IPA, where the phonemes you're talking about is transcribed as /ɤ/, which while still isn't perfect, at least it shows that it's a semivowel, not a consonant.
I don't like the fact that Xsampa is based on ASCII which in turn is based upon the american keyboards and it's ugly AF (waaaay uglier than IPA). If encoding standards had been invented in Europe the initial set of symbols encoded would have been different and probably if something like XSAMPA was developpe it would have been way closer to IPA in its look *sigh* well this didn't happen. But my point is xsampa comes out of an English centric original historical development and I think we should seek to emancipate from it instead of bringing it back. I've been typing IPA on macs and this is globally a way better experience than on windows. WIndows keyboards are archaic like they use almost no alt key's and the special characters interface is just impossible to use (not saying mac keyboards are perfect but still), so to me it seems a good chunk of the issue is tech based rather than alphabet based. Spanish /θ/ is very different to English /θ/ as they have very different realisations and they are part of different system and as phoneme are relativistic concepts this make them indeed tottally different. My question would be: then what's the alternative ? Would it be writing the English phoneme as /T/ or /þ/ but then what about the spanish one ? if we use Xsampa it would be /T/ too so it doesn't fix the issue. We could use /ç/ but then what do we use to write phonemes realised as [ç] in other languages ? And what about germanic languages that have a phoneme that can sometimes sound like english /θ/ ? are they allowed to use /þ/ ? Using the same alphabet for both the broad and narrow transcriptions just seems more convenient to me.
I'm still unclear what your point is, especially with the first half of the video- that having IPA symbols be shared with symbols used for writing natural languages makes it misleading and insufficiently international? XSampa sounds like it would only make things worse. You know what, if you know so much about transcription, how about you make a Broad Transcription system you actually do consider decent.
No, that was the opposite of my point, I said that having IPA symbols be shared with narrow symbols makes it misleading and insufficiently language-specific. X-Sampa doesn't really change that, it's still an international alphabet. Using regular latin symbols would be fine, and i'm saying better because it wouldn't be connected to the misleading narrow symbols.
Ah, but the "You already know it" argument applies to ~100% of the relevant population, unlike "people who use measurements"! I completely agree with the "lack of error bars" argument, as I understand it. Which is, ironically, why phonemes are useful.
@@zzineohp That's bad for describing phonemes when we have a system for describing phones in general. Phonemes *are* phones, they're just the contrastive phones within a given language or dialect.
Isn't the point of IPA that it distinguishes phonemes that are... distinctive? A shorthand you can use once you already know how the language works. Also, when you describe English to someone who already knows English you can just use /b/ and /p/ while to someone who speaks a true voice language you can clarify that by /b/ you mean [p] and by /p/ you mean [pʰ].
yes exactly that is my point. The broad transcription symbols rarely, if ever, match up with the narrow transcription symbols. So we should stop using the same alphabet for both of them.
@@zzineohp So that you'd have to learn new symbols for every language instead of just using the same ones with maybe the odd exception that's used slightly differently? Right now you can just click random on Wiktionary and pronounce any word in any language well enough that you'd be understood. You can't describe every dialect and accent of a language with a single transcription, no matter what symbols you use.
@@matt92hun you already have to learn new symbols, except it's worse than that because you have to learn new meaning onto the same symbols. You're ignoring the differences between the "real" pronunciation and the phoneme, but those subtle differences are what phonology is all about.
@@zzineohp I typed out a really long comment, then I realised that maybe I just didn't get your point. Are you saying that pronunciation with differences that are indistinguishable to a native speaker's ears should be marked in broad transcription? Or that you should only use narrow transcription? Would a cashier at your corner store hear the difference between [b]en and [p]en?
@@matt92hun i don't understand how you can disagree with me so hard we're both arguing that Phonemes and Phones are wildly different, and should use different symbols.
Three reasons why I dislike the IPA; A) it looks ugly. B) it is based on European alphabets making it very patronising outside the Western world. C) broad transcription is shit.
excited for next weeks quantum mechanics video
Ditto.
big day for India Pale Ale haters! (what is a linguistics)
"But what about diaphonemes?" I'm actually just coming for you. Turn around, I'm behind you right now.
But what about diaphonemes?
@@OkandUtlanning it's just pretending that different dialects have comparable phonemes. You could make a diaphoneme for every reflex of a PIE consonant and pretend like all of Europe has the same phonology.
you're WHAT???? 🤯🤯🤯
I was watching your video with interest until you mentionned that thorn is cool. I hate thorn.
@@ayrtonpavot3096 why though
Romanization systems certainly have their place, but broad IPA does too. The first half of this video just seems like pointing out things that are certainly true about the IPA, but I don't understand why they're issues. Your argument seems to be that because broad IPA doesn't provide the precise pronunciation of phonemes, it shouldn't share symbols with narrow IPA. But the problem with that is that broad IPA *does* indeed provide the pronunciation. [θɹ] is, indeed, a valid pronunciation of /θɹ/, same with [θɹ̪ʷ]. Sure, you would never hear a GA speaker say [sænd], but that doesn't matter. Broad IPA gives you the important information about phonemes so that you are understandable to a native speaker. Narrow IPA gives you the minute details about sounds so that you are natural sounding to a native speaker. Both have they're place, and I think it's strange and downright untrue to say that broad IPA doesn't give you the pronunciation. Even if broad IPA didn't give you anything even close to the correct pronunciation, that doesn't matter because narrow IPA exists, and we know if a transcription is phonetic or phonemic by those fancy little brackets or backslashes. If people are confused about which is which, then that's the people's fault, not the IPA's.
I totally agree with you that IPA looks ugly and is hard to type; that's why we have romanization. In fact, many romanization schemes are semi-phonetic rather than phonemic. Nobody can agree on how many phonemes are in Mandarin, so we just use Pinyin, which is phonetic, not phonemic. IPA and romanizations both serve different purposes, and there's no reason to replace one with the other. I love your passion for phonemic/phonetics, but I don't think you're right about this.
Wow this is... Something.
1. Phonemes and Phones are irreconcilable? Why's that? If we ignore using the wrong symbol for a specific phone, and let's ignore vowels because it's like trying to say at what point does red become blue, [p] represents a phone that is created with the top and bottom lips and does not make use of the vocal cords. /p/ represents much the same, but when you could have variants of the sound, such as one where it's voiced or aspirated, but the language, whichever one it may be, does not care about these phonetic differences when trying to differentiate words... then what matters is /p/, not [p~b~pʰ] (although doing the latter does help you sound native).
2. "/rɨba/ is bad because we use IPA." Saying this doesn't make it true. Why is using /ɨ/ bad because it's IPA inherently? What matters is that /ɨ/ represents a vowel with formant values probably between 240-250F1 and 2400-595F2 (i.e., somewhere high in the middle of the mouth with unrounded lips). Of course, if 95% of Polish speakers say [ɪ] (at least as of the 21st century), then why would changing the symbol be a bad thing?
3. Speaking of... Let's go back to the first example, if I hear that a language has a /p/ phoneme, and I use [p] where it's actually not used, that's not the fault of the IPA for writing down phonemes, that's the fault of the resources I used to learn the language for not describing that it's [b] between vowels, especially if this allophony has been well recognized since . It should be retired because people may pronounce /p/ wrong in the wrong phonetic environment? Sounds like you need a better teacher.
4. Phonemes are arguably specific to *people*. My /i/, no matter how the hell that's actually pronounced, will *never* be the same as your /i/, as a woman's /i/, as a child's /i/. A trans person's /i/ isn't the same from before or after their transition. But yet, our brains still all process it as /i/. The only time we don't process it as /i/ is if the allophony is broken, or the wrong phoneme is used entirely. Part of what makes the IPA useful is it's _standard_ . When I see /i/, I know what that sounds like regardless of language or voice.
5. Why does it matter how Proto-Germanic is written? Side note, I wonder how hard it'd be to do historical linguistics if you didn't have phonemes or at least phonetic descriptions of languages. How would we know that English linguists from, I dunno, 1400 know how to pronounce as /feːs/ even though today most English speakers don't. It's much easier to say, "The vowel in is /eː/" than "The tongue is in the middle of the mouth and held for long and sounds like French's ". Today we do have the formant values but that doesn't work for consonants I think. Also PIE is romanized poorly, at least when it comes to syllabic consonants and I feel bad for those who didn't realize that. But that's not the IPA's fault, that's whoever decided to romanize PIE like that. We write historic languages in phonemes btw, when we're talking about the phonemic (and phonetic btw) realizations of such. I can go to wiktionary, look up 'cat', and find both the spelling and the IPA /ˈkɑt.tuz/.
6. What does it matter where English's /θ/ comes from in relation to Spanish's /θ/? This is a non-sequitur. I would bet money that if you made a Spanish person learn English, they would go; "Oh, they spell /θ/ as th, while we spell it as c/z". They would recognize that their /θ/, which doesn't exist in all dialects btw, is the same phone as English's /θ/, regardless of the fact that Spanish's derives from Latin's palatalized C and English's from PIE's spiriantized /t/. That's not to mention that I, as an English speaker, can understand Spanish's /θ/, I see it as, for the lack of a better term, the same phoneme as English's /θ/. Also, can you provide a source for how different, phonetically or not, Spanish /θ/ is to English's?
7. At 3:23... Yes.
8. Double brackets are cool. That's not an argument against IPA imo. Also, maybe I'm just looking in the wrong places but could you also provide a source for that?
9. Phonemic transcription is... To represent the phonemes accurately. Specifically _the phonemes_. What matters when talking to English speakers is /w/, not [o̯]. [o̯] is not the phoneme most people would use, I suspect, assuming you didn't make that up to make your point. I'm not saying it's a bad thing you did, just saying that even if true it still doesn't prove anything.
10. Does IPA look ugly? I mean yeah, kinda. The only way to fix that is to use, I dunno, Korean? That literally makes its characters built on how the phones are made! Write a bilabial stop like /=/ (not an actual equals sign, just two lines representing the lips). Oh wait but that's hard to type, isn't it? Having to make an entirely new writing system?? Besides I kinda like IPA's aesthetic or lack thereof.
11. Diacritic spam can be an issue I suppose, but one that's infrequent and can be made of actual languages imo. This argument is against diacritics, not IPA.
12. "Oh noes, I have to go to a webzsite to type some characters that are not easily accessible to me!! this is the worst writing system ever!!" Is not an argument. It's whining. What if there were a way to easily type IPA on a keyboard (like on phones for example. I downloaded an IPA keyboard on my phone but I don't use it often because I don't do conlanging and such on the phone).
13. X-Sampa looks worse than IPA and still has the same problems in theory it's just less condensed (I'd rather look at and go to a website to type /ʙ/ than ever type B\ because it looks stupid). It's IPA for computer and we've moved passed the need for it. Besides, IPA _is_ 'metric,' X-Sampa is defined by IPA kind of like US Customary units. h2o\ or whatever the X-Sampa was looks... horrifying. I'd rather write /ɦɔ/.
1. I honestly have no idea how to take this. You tried to prove the point that broad and narrow transcription are reconcilable. The example you gave was "a narrow transcription represents a single specific phone," and "a broad transcription represents a single specific phoneme." Where in that did you connect the two? You even pointed out that the broad transcription doesn't care about subtle differences in allophony. Narrow transcription...does? You've tried to argue they are comparable because they use similar symbols, which is EXACTLY the misinformation i was trying to point out that broad transcription IPA spreads.
2. Because the IPA sets up exectations in narrow transcription, that broad transcription fails to meet (by virtue of WHAT IT IS). When I made the first video, some guy yelled at me in the comments because "The polish y vowel is [ɘ], not [ɪ]!!!!" There's obviously no way for a single phoneme symbol to reflect every realization. So what should it attempt to reflect? My argument is that it should attempt to reflect how the phoneme is perceived, which in this case would be /y/. Although I suppose there is an argument to be made that it should reflect the most common pronunciation.
3. If we're writing down all the ways that phoneme can be realized, what's the benefit of writing it using the "default" one? If people have, as you put it, a good teacher, they already understand it. So now that we've satisifed the linguists, maybe we should satisfy native speakers of the language, by making it something they can understand.
4. I'm convinced you've written one of my own talking points here, in the hopes of tricking me into arguing against myself. Yes, the phoneme /i/ is pronounced infinitely many ways. Yes, we still perceive it as /i/. My question is, why do we perceive it like that? I would argue that it's partially because we write it as /i/. That being the case, my argument would say that's a reason to use it as a broad transcription symbol, but that does technically supercede the IPA, which currently uses /iː/.
5. Because it's a language. Are you talking about "wikitionary uses a transcription for this one word I looked up" or "people constantly and consistently use an IPA-based broad transcription in addition to the * thing"
6. It matters because that's how phonemes are described. Right after Grimm's Law, could you say that the phoneme *t was realized as [θ] in Germanic, and [t] in the standard dialects? Yes, you could say that. That's how phonemes work, they were historically the same phone. And it doesn't matter if you perceive the DEFAULT pronunciations similar. Because they AREN'T ALWAYS PRONOUNCED THAT WAY.
7. Sí
8. They aren't currently in the IPA, so it's an indictment of the IPA that it doesn't already have that. And as far as I can tell, Wikipedia has MASSIVELY overblown the popularity of using them, so there might not BE any sources.
9. Guess what old buddy old pal? You're right! [o̯] is not the phoneme most people would use, because it's a phone, not a phoneme! And it's at least the realization I use. What I'm doing is TRYING to represent the phonemes accurately. And that entails NOT using the ipa, which is going to be inaccurate because it confuses phonemes with phones. What is accurate is writing broad transcription to match the ways people perceive the sounds.
10. This feels like a strawman, but this is fully subjective and not a point that need to be argued.
11. The diacritics are PART OF THE IPA! How am i meant to argue against them without arguing against the ipa.
12. Hey if we can improve it why not
13. Yes, X-sampa has the exact same problems as the IPA (since it's just a reskin). It's all arbitrary symbols, why not just use the one that's easier to type?
Honestly, I really shouldn't have to respond to anything after #1. If you don't understand that phonemes and phones are different, you can't understand any of what i'm laying out.
@@zzineohp 1. I don't understand your point. Saying things doesn't make them true and this video doesn't prove the initial point.
2. Let me spell this out for you... If you *know* that Broad Transcription gives you phonemes, and that Narrow Transcription gives you allophones, regardless if you have Narrow Transcription, you should be able to tell how a language is generally pronounced. It is INTENTIONAL. Metaphorically speaking, you're don't like hammers because they can hammer in nails into wood. That's the bloody point! Broad transcription tells you phonemes, it uses IPA characters BECAUSE BROAD TRANSCRIPTION IS A WAY TO USE IPA. Do you want there to be TWO phonetic alphabets based on phonemes and phones!?
3. The teacher is Narrow transcription, or at least an actual lecturer telling you such. Sometimes _it's better to read IPA than having to intuit systems you don't know._
4. If that's one of your points then you mangled it. If I heard a Spanish speaker say sí, and I heard an English speaker say see, then I would still realize both and as /i/. I don't know why it works like that, ask Chomsky or someone in the neuroscience-y aspect of linguistics.
5. Proto-Germanic, as it is currently understood and presented, is technically not a language. It's a hypothetical reconstruction of the most recent ancestor of all Germanic languages. It barely has any concrete reality, unlike say Latin. That's not to say that PG isn't real but it's a reconstruction, we don't have many (if any) records of Proto-Germanic. We have Gothic and Old Norse, but not Germanic. Besides, we still use IPA to show how languages evolved, Latin or Latin't.
6. BROAD TRANSCRIPTION IS MEANT TO BE BROAD. Accent or otherwise if all that matters between English and Spanish's dental fricatives is that you have the tip of the tongue at the teeth and you push the air out of your lungs without using your voicebox, THAT is the sound that matters. THAT is /θ/! Using this pseudo-Etymological fallacy is irrelevant. What does it matter that English says father where Latin says pater? At the point of Proto-Germanic, at least in theory, the speakers _probably_ didn't recognize /b, d, f/ as /bʱ, b, p/, and we definitely don't now! Besides, historical spelling is worse, "Long and short I" being/aj/ and /ɪ/??? That's NOT EVEN LINGUISTIC THAT'S JUST WRONG! That yelling wasn't at you btw.
7. Okay let's skip seven from now on.
8. You literally showed a Wikipedia page in the video...
9. Your personal phonetic realization of /w/ as [o̯] is not universal and that's why, even in other people's speech, it's not generally written as [o̯] because that corresponds to /w/ (and at least for me, IS [w]). Besides, here again broad transcription is doing its job. If you encounter a word with /w/ and you know how to pronounce /w/ even if you say [o̯], then what's the problem? We're not catering to YOUR specific idiolect?
10. It isn't a strawman but all I'm gonna say is literally: "We should not use X because it looks ugly," is not a valid reason, REGARDLESS if there are other reasons.
11. My point being that languages use diacritics and you'd have to zoom in on those too. Ça va, sí, hündchen, etc. Even Proto-Germanic does it, and worse with the (over)long nasal vowels!
12. idk what to respond to this, I'll just re-iterate the point that "We can't type with it" isn't very convincing because, at least as an American, we know that if we wanna type with other scripts we gotta download a keyboard for that anyways!
13. See point 12: if we hypothetically developed a keyboard (even if you had to download it) to use IPA, this would be BEYOND irrelevant.
I should fucking know that phones and phonemes are different, I've told you the definition multiple times if I remember correctly. At the very least I am a conlanger who uses the IPA a LOT. I don't see the issue in using phonetic symbols to represent the phonemes of a language, how the fuck else are you supposed to do it?
Apologies for vulgar language.
@@lulujuice1
1. Alrighty then, guess there's no point trying to explain it to you. Have a nice day
@@zzineohp 1. isn't a problem if the IPA is used correctly, so I don't see the issue.
I think that English fails at the IPA, not the IPA at English 😂
Well it's not like any other language handles it better
you've convinced me, I've had so many headaches realizing and finding the ipa equivalent phonemes that I pronounce or want in my short conlangs, it's horrilbe to try and type out, and I arrived at a homebrewed Xsampa (sorta) on my own to skip the ipa keyboard step in the end too
Wrote out a whole comment about my allegiance to the IPA and deleted it because I choose happiness. anyway you should make a video about ancient greek and the laryngeals
Two things
1. They should really make an IME that converts Xsampa to IPA when typing, like Chinese IME that allows you to use a qwerty keyboard to type pinyin and it'll show you the available hanzi you can choose and convert it.
2. I wonder if you can make an alphabet that to specifically write phones, the actual acoustic sound, instead of phonemes, regardless of intent or priming. Mostly because I want to be able to transcribe the Yanny-Laurel and ua-cam.com/video/7Yt6EfOCBwQ/v-deo.htmlsi=g8oB_FVTGOFbRvvZ
re 2.: at that point we should probably just use spectograms lol, maybe just weighted towards the formants and sibilants possible to be made by humans
I think the contradictions in your channel are hilarious, dont let anyone get you down. if you ever find a girlfriend just dont show her your youtube channel until 6 months in
Or don't show your girlfriend at all.
Show her immediately as a litmus test
Or find a girlfriend who loves linguistics rants
Me when the broad transcription is broad:
I really hope my comment (the third one at 2:15) didn’t read the wrong way. I wanted to express doubt. At no point did I think “he used that symbol wrong; I should tell him”, and I hate thinking someone might’ve thought I thought that way
Yeah, i always feel bad about using people's comments, and I definitely appreciate all the times where I legitimately get something wrong, I just thought it was funny how many people there were on this instance
it's still a helpful tool dog, retiring it all together and using phonetic alphabets for every different language / group of closely related languages just to avoid these issues is ridiculous, the cons are such small nitpicks, especially in comparison to the pros of having a universal phonetic alphabet.
It's good for Narrow Transcription, and my whole argument is about how that makes it bad for broad transcription. And we don't have to make up a new phonetic alphabet, you already have a different one for every language. But we keep pretending they're all part of the same system
@@zzineohp “We”? Some do, but linguists understand the difference between phonemes and phones, that’s why we write /t/ and [t] differently. You are annoyed at a misconception that perhaps was influenced by this dual purpose, but I dunno, I don’t think it’s as common as you think? But what would I know, I live in a hole and don’t have friends.
@@terdragontra8900 well this is more about non-linguists, the linguists already have everything figured out.
@@zzineohp The whole point of broad transcription is to give a general idea, and not high accuracy. Even narrow transcription isn't perfect, because every individual has small differences in pronunciation of the same sound for any language. And IPA is useful for talking about sounds in writing without the need for audio samples and also without the need for learning how to read every single language (which is 7,139 languages officially).
This type of videos of yours are really seems like just a rage bait for engagement anyways.
@@tovarishchfeixiao no I understand that is currently the point of broad transcription. My point with this video is that a) it doesn't do that job particularly well and b) there are other things it can do instead of that. If you wanna debate me on that, fine, but don't take it for granted that giving a general idea is the only important job.
i disagree with this video
I'm deleting your comment bro
4:32 I thought I was about to hear about the quirks and features of a forgotten 80s car.
Yeah sure but im not gonna transcribe things as /12 b5 8d*ㅎঙ/ and pronounce it [ba li lamut]
This video was very well thought out and I agree with most of your points. Reported for terrorism
That's fair, I was technically high on drugs for most of the editing process
Sure extremely narrow transcription or transcription of incredibly rare phones is a hassle and there are good reasons to use XSAMPA in certain situations (I have used XSAMPA and I csn testify it can be incredibly useful) but as you've clearly demonstrated, it's infinitely more bothersome to read the more it deviates from the unmodified letters.
If I'm reading or writing a text about Danish phonology as someone who has good enough knowledge of the language, I of course already know that the Danish /ð/ is very different from the English /ð/ but the point of using the IPA for the transcription of phonemes isn't to give an exact pronunciation guide but to represent this abstract concept with a reasonable enough symbol. Any other symbol to represent that Danish phoneme would completely miss the mark - how do I know that for sure? Because if you know a little bit about Danish you already know which phoneme I'm talking about. I'm just never gonna write [D_-_o_e] unless I'm absolutely forced to
Traditional Danish phonetic transcription is an impressionistic mess loosely based on IPA at best. I've only found one book by Ruben Schachtenhaufen that's legible if your native language isn't Danish and you already know IPA, where the phonemes you're talking about is transcribed as /ɤ/, which while still isn't perfect, at least it shows that it's a semivowel, not a consonant.
Ooh, I can finally actually understand one of my favorite Zzineohp videos well!
(I loved the original, but I too was tripped up on some of this.)
Me on my way to invent new phonemes, requiring them to expand the IPA and reattach my tounge
3:30 What unit does that scale measure, hectograms?
smh every place I speak about linguistics on has a bot that converts XSAMPA to IPA /lh
yooo what is that flag in your pfp?
when i need to write ipa symbols i just type 'i' into my searchbar and hit enter and my browser knows what I mean
it’s fine it’s fine it’s good enough it’s fine
I don't like the fact that Xsampa is based on ASCII which in turn is based upon the american keyboards and it's ugly AF (waaaay uglier than IPA). If encoding standards had been invented in Europe the initial set of symbols encoded would have been different and probably if something like XSAMPA was developpe it would have been way closer to IPA in its look *sigh* well this didn't happen. But my point is xsampa comes out of an English centric original historical development and I think we should seek to emancipate from it instead of bringing it back.
I've been typing IPA on macs and this is globally a way better experience than on windows. WIndows keyboards are archaic like they use almost no alt key's and the special characters interface is just impossible to use (not saying mac keyboards are perfect but still), so to me it seems a good chunk of the issue is tech based rather than alphabet based.
Spanish /θ/ is very different to English /θ/ as they have very different realisations and they are part of different system and as phoneme are relativistic concepts this make them indeed tottally different. My question would be: then what's the alternative ? Would it be writing the English phoneme as /T/ or /þ/ but then what about the spanish one ? if we use Xsampa it would be /T/ too so it doesn't fix the issue. We could use /ç/ but then what do we use to write phonemes realised as [ç] in other languages ? And what about germanic languages that have a phoneme that can sometimes sound like english /θ/ ? are they allowed to use /þ/ ? Using the same alphabet for both the broad and narrow transcriptions just seems more convenient to me.
English /þ/ and Spanish /z/.
@@zzineohp English /z/.
your videos are great, but could you please double the audio volume?
It already is doubled, to double it again I have to export it twice, which ends up ruining the video. It's an ongoing problem.
Can’t believe I wasn’t subscribed after Prosian 2
Interesting video
keep up the good work zzine!
I'm still unclear what your point is, especially with the first half of the video- that having IPA symbols be shared with symbols used for writing natural languages makes it misleading and insufficiently international? XSampa sounds like it would only make things worse.
You know what, if you know so much about transcription, how about you make a Broad Transcription system you actually do consider decent.
No, that was the opposite of my point, I said that having IPA symbols be shared with narrow symbols makes it misleading and insufficiently language-specific. X-Sampa doesn't really change that, it's still an international alphabet. Using regular latin symbols would be fine, and i'm saying better because it wouldn't be connected to the misleading narrow symbols.
when’s the zoo vlog coming out vro
“I shouldn’t have to zoom in”, get a better screen or make the text bigger?
That is called zooming in my friend
Next video should be about Prosian.
Ah, but the "You already know it" argument applies to ~100% of the relevant population, unlike "people who use measurements"!
I completely agree with the "lack of error bars" argument, as I understand it. Which is, ironically, why phonemes are useful.
Why are you even judging IPA at how good it is at broad transcription? It wasn't meant for that. It's a phonetic alphabet, not a phonemic one.
...which is why...we shouldn't be using it...for broad transcription. I think you understand my argument.
@@zzineohp What else are we supposed to use? An entirely separate alphabet? What do you have against broad transcription??
@@lulujuice1 just use the Latin Alphabet, or whatever alphabet the language uses.
@@zzineohp That's bad for describing phonemes when we have a system for describing phones in general. Phonemes *are* phones, they're just the contrastive phones within a given language or dialect.
@@lulujuice1 no they're not, phonemes and phones are different, they just coincide.
why did you drink 3 energy drinks for this
Because I hate sleep
Isn't the point of IPA that it distinguishes phonemes that are... distinctive? A shorthand you can use once you already know how the language works.
Also, when you describe English to someone who already knows English you can just use /b/ and /p/ while to someone who speaks a true voice language you can clarify that by /b/ you mean [p] and by /p/ you mean [pʰ].
yes exactly that is my point. The broad transcription symbols rarely, if ever, match up with the narrow transcription symbols. So we should stop using the same alphabet for both of them.
@@zzineohp So that you'd have to learn new symbols for every language instead of just using the same ones with maybe the odd exception that's used slightly differently? Right now you can just click random on Wiktionary and pronounce any word in any language well enough that you'd be understood. You can't describe every dialect and accent of a language with a single transcription, no matter what symbols you use.
@@matt92hun you already have to learn new symbols, except it's worse than that because you have to learn new meaning onto the same symbols. You're ignoring the differences between the "real" pronunciation and the phoneme, but those subtle differences are what phonology is all about.
@@zzineohp I typed out a really long comment, then I realised that maybe I just didn't get your point. Are you saying that pronunciation with differences that are indistinguishable to a native speaker's ears should be marked in broad transcription? Or that you should only use narrow transcription? Would a cashier at your corner store hear the difference between [b]en and [p]en?
@@matt92hun i don't understand how you can disagree with me so hard we're both arguing that Phonemes and Phones are wildly different, and should use different symbols.
Three reasons why I dislike the IPA;
A) it looks ugly.
B) it is based on European alphabets making it very patronising outside the Western world.
C) broad transcription is shit.
2:21 I think the real question is: why ⟨ǣ⟩ for the PRICE vowel (especially when you're using ⟨æ⟩ for the TRAP vowel and ⟨◌̄⟩ for vowel length)?