@@nameistanya the modifier “being silent” is misplaced, pointing to the word “it” (which is pointing to the phrase “every word”). the way your comment is written, you are saying that every word with a gh is silent. “every word with ‘gh’ being silent in it” would be more correct
The only time I've ever come across IPA is through my best friend, who is an opera singer. She's fluent in IPA because it allows her to sing in practically any language.
@@maliziosoeperverso1697 Hahaha no. We usually write the IPA above the staff/below the text while learning the piece. Similarly, we write the translation so we know what it is we're actually singing too.
If everyone in the world started using the IPA, after enough time, both graphic and phonetic aspects of any symbol may be slightly altered by the needs of each individual culture, making it not universal.
I mean... this is kinda already done. The phonology of a language that is decently well studied tends to use a fair bit of simplifications in their phonemic transcription. For example, English /ɹ/ or occasionally /r/ is in most dialects actually much closer to [ɹ̠]. This phoneme has a fair bit of different phonetic realisations too. Furthermore, /h/ in a lot of european countries is realised as a voiceless version of the followinɡ vowel. In english /ʌ/ tends to be closer to [ɐ]. In swedish /a/ tends to be closer to [ä]. Also in swedish /ʉ/ is understood to mean somethinɡ like [ɪ̙ᵝ]. /t/ may also represent any plosive from dental, denti-alveolar, alveolar, and postalveolar. What's more is that these may also be realised as either laminal or apical. Phonemically most of these consonants are just writen as /t/, but [t] typically seems to represent the voiceless apical alveolar plosive. Another example from Swedish is that /ɾC/ is often realised as a retroflex version of /C/, where /C/ can be /t, d, s, n, l/. So /rt, rd, rs, rn, rl/ become [ʈ, ɖ, ʂ, ɳ, ɭ]. Here's just a few examples how ipa is used inconsistently among different languages. Phonemic transcriptions oftentimes try to break down the language into ad simple a set of phones as possible, and also choosing to use the simplest symbols avaible to represent these, unlike phonetics which tries to get more accurate transcription, instead losing generalisation among the language speakers, but gaining a more accurate transcription of this particular speaker or dialect.
That's how it currently is, and it would FINALLY stop being that way if everyone switched to IPA, because people would be encouraged to actually spell hoe they pronounce.
@@bacicinvatteneaca Well, the idea is that the way people speak will always disconnect from the orthography, regardless of how precise it is. Are you implying (as mentioned in the video) that the only solution is to then have every individual pronounce words as THEY individually say it? Then where's the 'universal' benefit of the IPA, and worse; the orthography reduces in effectiveness because of how inconsistent words will be even across individual people. Sure, maybe we do go ahead and standardize, idk, English in the UK (so everyone spells things the same way with the IPA). But, then the problem of a gradual disconnect of speech to orthography reappears, making it no better than the standard English Latin alphabet-based orthography we have now. I'm not saying that the IPA would be useless to think of as a substitute, but I believe in no way will it make the world better with universal adoption.
They could always do the Vietnamese thing. Write the tones if they’re critical to understanding. Obviously they’re going to be writing two to six times the total number of characters, but there’s a chance that someone without the time to memorize 12,000 characters (and come to grips with the 102 radicals) would still be able to read four words in a newspaper with a couple tries.
@@polyhistorphilomath It's not about tones but rather the phonotactics. Chinese has a lot of phonemes with the same tones but with varying meanings. What those words/sentences mean depend heavily on the context of adjacent logographs and their patterns. Correct me if I'm wrong, but I think that's the gist of it. It's easier to just understand the current writing system than the IPA. Edit: grammar
@@polyhistorphilomath There have been several attempts to romanize Chinese characters, but in the end it was mainly use for pronunciation purposes. To keep it short, Chinese have very rigid restrictions on what kind of sound can make a syllable and a one syllable one character rule. This leads to a huge number of homophones. I’ll give you an example with English. You can obviously tell the difference between to, too and two even though they sound the same, but would you ever want to write them all as “to”? Let’s start with the phrase “I want to”. Without context this single phrase is ambiguous, but with context the meaning become clear. I want to (two) chicken, I want to play to (too), I want to (to) go home. Now imagine every word also have 4 homophones. I have read texts in bopomofo when I was growing up and I can tell you it is a pain to read. If they are going to do it, I rather they do the Japanese thing where they add pronounciation (furigana) along side kanji.
@@reachforbluesea I think I see your point. If there are many homophones and relatively few possible syllables (when transliterating without tones) then many of the advantages of alphabetical writing are lost. Apparently the size of the syllable inventory in Vietnamese (as the obvious counter example-tonal, but written in Latin characters) is different. Feel free to correct me, but it must be large enough that the possible word shapes also have a wide variety. I assume this because the Vietnamese actually use their writing system, after all. I can’t imagine happily using the Latin alphabet for a language where word shapes were not helpful in reading quickly. Just to be clear, I mean word shape in the sense of the rough outline and/or other visual peculiarities [such as the location of denser patches of character strokes relative to the outline] in an alphabetically written word.
@@polyhistorphilomath reading into Vietnamese and Korean a bit more, both Hangul and romanized Vietnamese were imposed specifically so commoners didn't have to go through the very rigorous schooling required of the scholars. and, well, it worked.
IPA would be like if a musical scholar thought "our musical notation system only applies to the 12 tone scale. Other countries use 5 tone scales or even 22 tone scales. To account for this I've created a new way of writing music where every note is written as its frequency in hertz and the amount of time that note is played is written in milliseconds above it. Techniques are all represented by a unique symbol, for example pizzicato is represented by the letter д with an umlaut and rotated 135 degrees clockwise. Also, "pizz" is written below it in italics to avoid confusion. The musical score also comes with a golden record which when played in a projector will display a semiotic representation of a string being plucked as well as a spectrogram and a diagram of a hydrogen atom. This is just to show that we are an intelligent species to any alien species that may come across these musical manuscripts"
@@pedrosso0 I was trying to make an overly absurd and convoluted system as a joke however, our current notation system does have its limitations and having one that accounts for different cultures as well as modern music would be useful. Imagine if it was also designed with music software in mind. It could be a happy medium between programs like musescore and programs like abelton and pro tools.
Yeah, people saying that this is a good idea kinda forget that different styles use different temperaments. For example, in "standard" Baroque tuning, the A is at 415 Hz, and in French baroque tuning, the A can be as low as 392 Hz (that's a whole tone down!)
@@paranodrum9171 Aren't Chinese keyboards actually in latin and then through software different combinations become Chinese characters? That's at least how all the keyboards I've come across work.
@@masterkamen371 There is no one standard Chinese keyboard as there is for English, though I believe the ones you described are the most common, where you type how the character sounds in the latin alphabet and the software gives you a number of sugestions.
Remembers seeing some linguistic/orthography subreddit where whenever someone made an English spelling reform there'd always be someone saying "wAɪ̯ NɒT dƷʌSt JuːZ Aɪ̯ Pi Eɪ̯"
Argh. You would be making something like Arabic script but stuck in the opposite trap. You want 900+ distinct glyphs to be all swirly. Good luck. Arabic is hard enough to read. IMO. It’s like a post-reform abjad that only got to keep five distinct letterforms-leaving all the work of disambiguation to be done by sporadic dots. Okay so now all we get to read is swirls and dots. Congratulations, it is illegible.
@@polyhistorphilomath Standard written Arabic is OK, kufic is harder to read, nasta''''''''liq is illegible. People do cursive Vietnamese and it's even stupider than the standard form. Suffice to say if I had to read that language for any length of time I'd be typing this comment in Braille having gouged my own eyes out.
I actually had the same idea of “Why can’t we just use the IPA?” sometime within the last 2 weeks. Then this video came into my recommendations, and made a lot of good points.
Finnish is pronounced pretty much exactly as it's written. People speak in everyday situations a bit differently to how they write but it works just fine. Some of the critiques do still apply.
The IPA obviously makes no sense as a language since it’s made for translating between languages. But in making artificial languages, like Esperanto, I think phonetically consistent is something that would be extremely useful going forward, so there’s no ambiguity for the word.
It mostly boils down to most languages having incosistent or convoluted rules for spelling things. And we can't forget that asian languages which depend on context for the words to make sense would just make reading harder, since it's not rare for different words to have identical spelling. I can't trally think any language aside from Greek, that would benefit from this change
As someone who has to interact with phonetics fairly frequently, Vocaloid SAMPA, specifically whatever modified version of XSAMPA Hatsune Miku's English bank uses, is about 27 times worse than IPA. I hate XSAMPA so much. Being a vocal synth user is a punishment unlike any other I have learned many phonetic systems against my will. Vocaloid SAMPA, Delta (which is a modification of Vocaloid SAMPA's English transcriptions, meant to be easier for Japanese speakers to read), Arpabet, CMUDict for French (this SUCKED bc its only meant for European French and I speak Canadian French), CZSAMPA (a modified version of Vocaloid SAMPA based on written English and the GenAm accent), Erable French (I modified CMUdict for Canadian French, gods preserve us), the IPA I need for English and Japanese, voice synths are a special type of hell sometimes
@@jolkert_ Vocaloid SAMPA is even WORSE. It's SAMPA but modified and also phonetically inaccurate. With the example I gave of Hatsune Miku's English bank for Vocaloid 4, she sings in Japanese-accented American English. Vocaloid SAMPA was written solely for Recieved Pronunciation and did not add symbols for key phonemes in American English (the dialect most English Vocaloid VBs sing in) until the release of Cyber Songman for Vocaloid 4 (yes that's the real name of the voicebank) in 2015, and only as an optional extension to English voicebanks. I am so eternally grateful that SynthesizerV uses Arpabet for English, Hepburn for Japanese, and only pulls out the XSAMPA when it's dealing with Mandarin
The IPA is way too technical to be used on large scale. Sure, I once terrorized my friends on a groupchat using only the IPA for like a week, but imagine using it everywhere. The IPA is a good method to supliment language learning, but the actual writing system carries on history. I personally think English spelling is okay as it is, and this comes from a person that had to learn words one by one, like logographs since primary school (that is how we learn English in Romania). I cannot imagine English, French or Greek without the horrible historical spellings. Also, there are languages that didn’t have any written form, so linguists decided to write them in the IPA officially. So the IPA is actually used on a large scale to write some endangered languages.
But Ai fowr wun think Ingliś wuud be betur ouf with æt līst fonetic spelings (awlthoh it wuud make menī wurds hārdur tū distingwiś frowm īč uthur, laik “bee” & “be”), its curint spelings ār u mes. Awlsow rīnaym thu leturs acowrdinglī plīz.
The IPA isn't even used consistently within English. It depends on the context and how narrow and updated a convention is. It's also not narrow enough to transcribe a lot of phonetic differences between languages.
Yeah it is very common to transcribe the wrong vowel, forget about aspiration or mistranscribe a rare sandhi. Native speaker often don't notice these things in their own language.
That's the whole point of mandating its phonetic use and completely abandoning phonemic considerations: force people to be cognisant of how they use their mouths. No conventions, ever.
@@bacicinvatteneaca I'm sorry but that is the most ridiculously terrible and batshit crazy, or simply ignorant take on language I have ever heard. "conventions" aren't the problem(or the only problem). Allophones are. Dialects are. Ideolects are. It would be impossible to destroy any one of those things with ANY AMOUNT of draconian linguistic prescriptivism. Being "cognisant of how they use their mouths" is not something you can expect of people at least to the degree you are suggesting. Without even knowing you, I can guarantee that the sounds you make are not always the ones you think you are making phonetically. And it wouldn't even be "better". There is a significant amount of phonetic sandhi in English that would literally do nothing but needlessly complicate spelling by forcing a spelling distinction where it would be irrelevant. As a concrete example what is the benefit of transcribing the word "symphony" with an /ɱ/ and the word "more" with an /m/? There are no minimal pairs between them. 99.9% don't know that there is a different sound that exists there. Literally nobody would notice if you made the wrong sound there. In fact, most people if they were asked to say the word slowly would make the wrong sound there. Your suggestion is about equivalent to. "That's the whole point of mandating that people only ever die in coffins." Yeah it would be more convenient in some narrow way, but how exactly are you going to get it to happen? And how exactly would it be better? It certainly wouldn't make it easier to teach reading or writing.
Yep. Take, e. g., the [æ] sound in American English (but not in British). Words "back", "ban" and "bank" will be transcribed with the same [æ] symbol, but in fact they have three different vowels. Or the famous "dark L" in English, which usually does not have a separate transcription.
I showed this to my speech pathologist mother, who said she had never seen "don't" written like it is in the thumbnail. She guessed you would have a British or Australian accent. Also she said she used to write letters to people in IPA just to see if they could figure it out.
Another part of the problem is the incompleteness of IPA. Some languages (especially East Asian languages) include pitch in the pronunciation of words. Two words with an identical IPA representation can mean different things simply because the pitch structure of the word is different (for example, in Japanese).
@J Boss Fair point--though I was meaning to say 'having everyone use IPA to make cross-lingual pronunciation easy' is probably a bad idea because it is incomplete with regard to that aim, if not for countless other reasons.
As the first guy already mentioned, though, although the IPA *isn't* the ITA, tone markers do exist in the IPA. They're, uh, a bit questionable, like everything else in the IPA, but they're technically there. If I were to complain about anything, its that there's no real (good) way to express rhythmic differences in speech. You could have different people who speak various languages with similar phonemic inventories all say the same sentence, and they'd *still* pronounce them differently, with no real way for someone reading the IPA transcriptions to know. Of course, as you've already said, this isn't particularly in the realm of the IPA, but it'd nevertheless be interesting to see an attempt made to accommodate this
@J Boss Mostly, it'd be interesting to see how they'd go about making a system that captures that level of detail. It'd take more than just a single subscript number, though; I'd imagine it being akin to extremely detailed and complex musical notation. I'm a programmer, too, and I can definitely say that if you think audio parsing *and* speech synthesis on the level of being able to produce the "exact same sound" is "easy", then you should definitely get to making that, because that's state-of-the-art technology right there that people've been trying to achieve for years. My man, you could likely make hella money from something like that, and, if you're truly willing, I'd kinda just wanna see what you come up with either way tbh, cuz, as I said, it'd be interesting
@J Boss Ignoring the (possible) hyperbole in the number of subscripts you'd need, combining all of the data into a single subscript seems like a really bad idea. That's like taking every IPA phoneme, along with every combination of diacritic for each, and making them their own sound. Velarised /k/ and /p/ no longer just have a superscript /j/ attached here; they'd be entirely new symbols. And thats ignoring the fact that you can just keep adding diacritics if you'd like. Even if it takes having 254k different subscripts, I'd much rather have that, and be able to add and subtract them as I like to get more or less detail in transcription, than to be forced to have the maximal level of detail described by having a character per every combination. But even then, the idea I'm getting from what you're saying is that you think that, just because you have the series of binary data that represents audio, that you can then make a program that can take that data and somehow manage to parse it into IPA so accurate that anyone reading could convey the exact same sound. That, alone, is extremely incredible technology, and most modern solutions to this rely heavily on machine learning algorithms trained on a metric fuck ton of audio samples only to get a relatively accurate transcription of whats being said. If this was as easy as you make it seem, then, for example, UA-cam's automatic subtitles wouldn't be so absolutely bad at transcribing text. That's not even getting into the claim that making a speech synthesis engine to play back the IPA transcription would *also* be easy. I've admittedly seen some pretty good speech synthesis engines, but you can usually tell when the speech you're hearing was synthesized. You're expecting technology on the level of highly specialised deepfakes that's as accessible as something like the Vocaloid speech engine. These technologies *are* getting to the point you say is easy, but the sheer level of accuracy required in this hypothetical program that can supposedly take in any audio sample and both output exact IPA *and* exact speech synthesis just isn't quite there. The accuracy you say is easy generally requires extremely large data sets of audio, usually from a specific person, to get a pretty decent approximation. You *could* generate an IPA transcription of a given audio sample, and you *could* make a speech synthesis engine that plays back any given IPA transcription, but without heavy specialisation and a ton of work, I simply cannot see how a program could provide "exact" output that you say is "easy" or "not remarkable"
related to the dialects thing, but even if everyone somehow spoke the same dialect how would you account for micro differences in speech “oh no i accidentally added a tiny bit more height to this vowel and now i have to add a different diacritic” of course simply calrifying you only need a phonemic transcription would help but yeah dialectal differences as you said
@@bacicinvatteneaca at a certain point, it would 1. probably hinder readability, or at least slow down reading speed, and 2. basically require people to have linguistic knowledge and phonetic awareness of their own speech enough to use the IPA accurately (using it inaccurately kinda defeats some of the purpose of it) now assuming its purely phonemic transcription, then that still kind of requires people to know which of their phones are just allophones of a main phoneme; and also point 1 still stands
Erasing written language would also erase cultures. Many writing systems (Chinese, Arabic, etc) have a rich art of calligraphy. And some cultures have deep pride in their writing systems, like Korea’s Hangul.
Even if we switched every language to IPA, all it does is tell us how to pronounce words in any language, it doesn’t encode any meaning so if we came across a post in another language, all we knew is how all the words sound, not the actual contents of meaning. Also, I enjoy IPA as both a guide in Conlang making *and* singing international songs for choir.
The thing about different dialects having to use different writing systems was actually an issue with my own (Rhenanian flat German) dialect which is why - up to the third Reich and its educational reforms pushing High German as the standard in both written and spoken language to push the idea of "one German people" instead of the hundreds that existed as semi independent entities untill 1871 - many people around here used to write in Dutch as it more closely resembles/d our way of speaking (nowadays hardly anyone can speak in pure dialect anymore and most speak in "Mundart" - High German with a regional accent with varying degrees of strength depending on how official the situation is and wether they are conversing intra- or intertribally)
@@xhawkenx633 eigene Sprache ist glaube ich zumindest umstritten also zumindest eine Grauzone und Dialekt und Mundart können sehr gut im Hochdeutschen das Gleiche bedeuten, in meiner Region zumindest werden sie aber unterschiedlich genutzt. Wie bereits geschrieben ist Mundart nach meinem Verständnis Hochdeutsch mit Dialekt Betonung (Plattdütsch hat eine teilweise andere Grammatik und Satzstruktur, daher ist Dialekt sprechen zumindest in meiner Gegend noch einmal eine andere Sache. Ich spreche z.bsp. Mundart im Alltag (t->d, g->j, ch->sch, und eine gute Portion an Lehnwörtern aus dem Dialekt) und kann Dialektsprecher problemlos verstehen, kann aber leider selbst nicht Dialekt sprechen
The solution to accounting for regional dialects is to simply create universal logograms to be used in all languages instead. This plan has no flaws. Chinese characters will work for English with enough brute force.
Imagine if Chinese characters were adopted in English, similar to how the Japanese did, by mixing phonemes with the characters. Of course this isn't a good idea but it would look cool and it's definitely possible.
I think the issue in IPA is with it being meant for language analysis. IPA lists 68 different vowel sounds however most of these are not used in most language and many sound very similar to up to 5 other vowels as well as almost always being mutually exclusive to these similar sounds in a language. For language analysis these sounds may have nuances that need to be separated into different categories but as most of these sounds are mutually intelligible and/or indistinguishable, you could reduce these sounds to 20 or less letters. Consonants tend to have less similarities though and these 54 sounds can only be reduced to around 40-45 (there are some sounds that seem the same to me but are not in any languages i speak so I'm unable to give a precise guess), but at most we could easily make a simplified IPA with around 65 characters and separated into families of sounds (represented by a common letter of the family understood by most languages) so that if a sound doesn't match what someone wants to portray they can say what family its part of and use placeholder diacritics (which should not normally be used in this font) to show manner and place of articulation.
As someone with a linguistics degree when you said you hated those letters I felt that on a spiritual level. Damn retroflexives still have use, but doesn't mean I have to like them.
come on man retroflexes are not all that bad, like would you rather have a t with a tail or with an under dot. at least the tail seems like a separate letter than just a diacritic. (i say this as an indian, where most of our languages have a clear distinction between dentals and retroflexes)
@J Boss your comment about german is so stupid, this would mean that japanese is a sino-tibetan language because it has chinese words in it that's literally not how it works, and your comment clearly shows you don't understand anything about linguistics since you label it as unscientific and emotional, as if there isn't a gene named the sonic hedgehog gene, and the planets haven't been renamed to something more scientific than the names of roman gods so how about you leave the linguistics to the linguists and go back to whatever sad angry corner you came from? :)
This is not even mentioning the fact that you would have a much harder time identifying any given language due to the lack of orthographic variety. French and Swedish passages could look very similar at times, despite not even being in the same language group. Furthermore, imagine *loan words* that are just tossed into a script. In English, you can identify a loan word or phrase by italicizing the phrase or using quotation marks (e.g. "en masse"), but I think a lot of the context behind whatever requires a loan would be lost if all lingual scripts were homogenized. Even in English, the language of borrowing, we can usually figure out what language a word is from based on features of the written word itself (for example, two Zs in a row that make a /ts/ sound signify an Italian word), but all of that context would be lost if we decided, "Yeah, sh and sch and ch make the same sound in shall, anschluss, and pastiche, so let's just disregard the entire origin and meaning behind the variation of spelling and just change it all to sh." A lot of people point out English's propensity to have many different spellings for sounds as a degeneracy, but I firmly believe it's a virtue, because the spellings can provide historical context to a word AND help you figure out the meaning of the word without having to perform rote memorizing (Latin roots, kids). So having only one writing system based solely on the sound of a word would just be a disaster in terms of finding meaning in words, and on a more superficial level, identifying the language in which a writer is even writing.
If you can't distinguish French from Swedish in speech, why should you distinguish them in writing? If English flattens the sound of loanword in speech, why shouldn't it do the same in writing?
“Timmy, you wrote your essay in Californian. You’ll have to rewrite it.” “But we live in California” “Sorry Timmy, you have to write in Londonian like everyone else.”
It doesn't though. It doesn't account for nasal assimilation- tenpo isn't written tempo. Basically every sound has hundreds of potential allophones depending on your l1 - English speakers don't write aspiration markers, or distinguish between e ɛ and ei. When I say the sentence "tenpo ali la sina jo e ma e wawa e pona" I say it 'tʰɛm.pʰoʊ a.li la 'si.na jo e ma e wawa e pʰoʊ.na Which is close but not identical.
why would anyone want to give up their culture and start using an uglier version of the latin alphabet? i (as a cyrilic alphabet user) would never give up on the cyrilic alphabet. it is so pretty and kinda unique
I misinterpreted this as saying that the cyrillic alphabet is an uglier version of the latin one and i was about to get very mad at anyone who thinks that because the cyrillic alphabet looks sick
your videos are great man keep it up the german one blew my mind. been learning for 5ish years, and I never noticed the thing with s, ss, ß. Like I knew when to vocalize my s's, but just from seeing the pattern so many times, not from knowing that there's a rule to the pattern
Croatian, Serbian, Bosnian and Montenegrian all have phonetical alphabet. And realy easy to read and understand. We have 30 phonems and they are represented by 30 letters.
@@cakeisyummy5755 ma lik je vjv nije ništa istražio za video uzeo je 10 največih jezika i misli da je rješio sve svjetske jezike. A mi za njih navjerojatnije ne postojimo.
Also, some languages such as Chinese and Japanese become really confusing when written phonetically, they have relatively few syllables and thus relatively many homophones making their system of pictographic characters very important in written communication.
@@bacicinvatteneaca if it was that simple then don't you think china would have simply thrown away hanzi instead of creating simplified chinese? there's a reason these writing systems are still around and that's because they work for the language they were created. i think what people forget when they propose these kinds of spelling reforms is that writing systems have history, and hanzi and kanji exist in their current state after thousands of years of use. there have been multiple opportunities in both japan and china's history for them to do a writing reform and completely throw out hanzi/kanji to raise literacy rates but they didn't. also, what's the point of creating a completely phonetic writing system? sure, spelling becomes easier, but when you see a new word how do you know what it means? shallow orthography loses out on the advantages of deep orthography, where you can understand meanings of the words you don't know just from the way they're written. you need to remember that japanese and chinese have evolved alongside their writing systems for thousands of years, which means that words have been created in a way that makes sense when you write them in hanzi/kanji. in japanese, I'm able to understand unfamiliar field-specific terminology (like medical terms) just from knowing what each kanji represents, whereas in english I would have to look up what each term means, if i'm not lucky enough to remember what each latin or greek-derived part of the word means (if it even is derived from either of those). keep in mind I only have middle school level education in japanese, with some low effort "self-studying". hanzi was also used as a written lingua franca for years because even though spoken chinese and japanese are not mutually intelligible at all, because they shared writing it allowed them to communicate. in fact even in modern times, my own mother was able to sort of communicate with the chinese grandmas at my elementary school because they would sit outside and write what they were trying to say, and growing up I would navigate chinese websites like tudou and bilibili to try and find videos that would get copyright strikes on youtube. you might say that it's SO MUCH to learn because in daily use, chinese people need around 3000 hanzi and japanese people around 2000 kanji! a phonetic system it would reduce that to only the phonemes each language has! but if you think about the fact that you put hanzi/kanji together to create thousands of new words.... and the amount of words any language has... that's several thousand definitions that become intuitive just because you already understand what each part of the word means. it's impossible to do that with a purely phonetic system. one of the purposes of written language is to efficiently convey meaning. hanzi and kanji do an amazing job at that, so why should we get rid of them simply because you can't sound them out just by looking at them? there's a lot more to language than what sound each thing makes.
@@Pashmimi "shallow orthography loses out on the advantages of deep orthography, where you can understand meanings of the words you don't know just from the way they're written" ...unless you don't know the logogram
I remember being told about the IPA when I was in school and this idea was my first reaction. Obviously, my six year old self was not a linguist. I'm glad that, after all these years, I finally have an answer to my then incredulous question. You really do learn something every day
It's even funnier when you realize that some sounds in your language go outside of the most known characters in IPA. We had this dilema in class with English native speaker and we tried to explain how to pronounce our names using IPA. Turns out that it's hard to explain a sound that lays between to sounds commonly used in English.
In my native tounge, Swedish, there are often variations in the spoken word that indicate how formal your speech is, or to make emphasis or for clarity. Beyond dialect that is, so we're talking about the same native swede in the same dialect pronouncing the same word differently depending on circumstances. So in informal spoken swedish most endings of the most common words are simply usually dropped, but if you put a stress or emphasis on a word then those dropped letters can come back again. You have this sentence for example: "Var vill du vara" = "Where do you want to be". But most of the time both "var" (where) and "vara" (be, exist) are pronounced exactly the same. As simple "va" with a long A, as in how a posh British person says "bath". And then you have that the word "vad" = "what" that is even more frequently pronounced as "va", that ending D is rarely heard even in more formal settings. So sometimes in this type of sentence the first "va" might often be pronounced in full with the D or R at the end just to make the distinction between if it's asking about "Where" or "What" clear. But if the context already makes it clear then we will immediately drop those pesky endings and just say: "Va vill du va" and let the situation and our intuitive sense of grammar make the distinction between the words "var", "vad" and "vara". As a side note, formality here is actually determined more in how you pronounce the word "vara" than in the other two words. Even people reading the news on TV might frequently drop those D's and R's in "vad" and "var". But they'll rarely contract "vara" to "va" while reading the news. And swedish is loaded with this kind of stuff, silent letters that aren't always silent, occasionally you have multiple choice in how a word is pronounced, and we have words that even in IPA would look identical but are instead distingiushed by stress and rythm and melody e.t.c. How could you possibly make a spelling reform to incorporate these kind of things?
My standard English textbook in schools in China has IPA for all words and I memorized them, I found it useful in communicating English sounds at least..
Wow, i think this is the first time that i'm fully convinced something would be a bad idea from being convinced that something is a great idea, and in less than 4 min, Bravo!!
@J Boss Heb je de video bekeken? ¿Viste el video? Avez-vous regardé la vidéo? Hast du dir das Video angesehen? Viu o vídeo? Hai guardato il video? আপনি কি ভিডিওটি দেখেছেন? あなたはビデオを見ましたか? Apakah Anda menonton videonya? 你看过视频吗？ Pinanood mo ba ang Video? क्या आपने वीडियो देखा है? Вы смотрели видео? Umewahi kutazama video? Ingabe wayibukela i - video? Did you watch the Video?
Fun thing about using a standardised dialect to spell things: If people are recognising sets of letters as words, but the sets of letters they're reading don't conform to the accent they're speaking in, they have two options: 1. Change their own accent. This is a fair amount of work, and good luck getting lots of people to do it. 2. Just learn to recognise that this clump of letters is that word, and speak it the way you always have. The latter means that now the same set of letters will now be spoken in different ways by different people, and that's exactly the problem we already had. So, uh, yeah.
@J Boss And if you did it for English you would have to convince the entire Anglosphere, around half of Africa, India, Pakistan, the Philippines, and ostensibly everyone involved in international trade to do the same. Easy.
@@joshs5577 And after finishing off English, we would then have to do the same with every dialect of every language currently in use on earth. It would be a monumental undertaking at best, and borderline impossible at worst.
Curious, has there ever been attempts to create a completely constructed script to replace the IPA that's more featural and less Euro-centric? Seems like something that an amateur linguist or Internet group might've attempted, after all the countless English spelling reforms proposals.
Presumably you'd need to derive it from another region as "standard" instead. While logically you'd do that with Africa, the fact remains that linguists specialising in Africa are much thinner on the ground. Eurocentrism in the IPA primarily comes from pragmatism rather than purely from bias.
I think he meant reinventing an alphabet completely, solely according to the qualitative diversity of human speeches (without any other bias)... Ot eventually that the only other bias would be the numbers of speakers.
@@Turagrong Even a total reinvention is going to be biased towards of the specialty linguistic sphere of whoever made it. "Zero bias" is impossible so long as anyone is interpreting anything.
@@Halophage If it's made by a team consisting of people "from around the world" who all also can speak some representative languages of other major groups to a good level... then not that much...
All of the "how to pronounce" searches I've done in my entire life do not amount to the time it would have taken me to memorize the stupid fucking IPA.
Whenever I consider spelling in IPA, I can't agree on the specific phonetics of that very sentence. So yeah, amazing as an auxiliary, less so in everyday writing. Maybe a less ambitious and more flexible option would be more coherence between phonemic approaches. Explaining Chinese "zh" and "r" to fellow French gets old very fast.
Ben tiens, tu me rappelles quand j'ai dû expliquer à mes amis que Ganyu ne se prononce pas "ganne-you" mais bien "ganne-u", ou bien que "zhongli" c'est pas "jongli" mais "tchongli" Le pire c'était "Kazuha", je me suis fait reprendre pour l'avoir prononcé "Kadzuha", alors que eux le prononçaient "Kazoua" x)
@@aronasmundurjonasson3175 Only 2020's problems: la prononciation dans Genshin. Ça illustre bien le problème. Entre les langues qui prononcent Z "z" et celles qui prononcent "dz"; les variations pour étendre le répertoire ZH/Ž "j" et les Chinois qui ne font rien comme les autres "dj", c'est tellement confus que tes potes ne sont qu'à moitié en tort.
"So, to type that sound, you have to hit control twice, then press alt + shift together for two seconds, release them, then press control + alt, release, then dance three times around your room to summon Satan and then you press the A key and you're done"
A thought I had on getting rid of C is that since you're already making a large scale reform, you go all the way with it. C becomes what is currently CH because it's unintuitive to non-native speakers for 2 consonants to combine and make an entirely different sound to its constituents. And the softened intervocalic S in your example is always written with a Z. Given the context of that you can probably imagine what I'd do do eliminate pairs like SH, PH, and NG. As you can probably tell by now I know nothing about linguistics and would like to know why I am dumb.
In German, we once had a system where sometimes ss is always pronounced "s", but sometimes the preceding vowel is pronounced long, sometimes short. It got confusing with words like "in Massen" (within limits) and "in Massen" (in masses). The spelling reform turned the one where the preceding vowel is pronounced long into an ß (15 years later or so also introduced its capital version as ẞ, I think they had forgotten about that lol), so now it's "in Maßen" and "in Massen". This reform was made in 1996, as I understand it, shortly before I was born, so in those days, but slowly less and less over the years, older people started using the new rules
I get why you didn't get into this, but there's also the eternal question of "should we do a phonetic or a phonemic transcription?" Like, do you write the American word "city" as /sɪtɪ/ or /sɪɾɪ/?
Phonemes are intresting, but I believe that as long as you're desinging that writing system for only one language, they would make the system harder without any benefits in almost all cases.
What kind of question is that? The whole point is to make it phonetic, so that one can signal differences in pronunciation between dialects, and so that people become aware of the sounds they make. There shouldn't be any conventional spelling ever.
The only good use of IPA, apart from learning a new language, is the promotional campaign of the Polish city of Szczecin. Foreigners cannot pronounce it, so there were billboards with the name written in IPA, I thought it was so smart and hilarious.
not forgetting the massive overuse of hyphens and h's, or the insistence on the writer's particular middle Yank or Melbourne accent being the only correct pronunciation
And I'm not a native English speaker. To me, that is just the best way to represent English phonetically, just like Pinyin for Chinese. IPA is designed to be international and work for all languages.
If it would lead to everyone understanding every language in the world it would be a good thing, but sadly it doesn't and so doesn't justify the work that had to be done to first convince every language representive in the world of this idea, then teach everyone this language and change every written word to fit the IPA norms. I learned french and english in school and sure the ipa stood in brackets behind the vocabulary lists we had to learn, but did I use them? Most of the time not because even though it is given you don't really learn how to read it and aren't that interested to change that because once you know how to spell a word the IPA is never really relevant again. Also if it would be really that intuitive why are people just refereing to the letters and how they are pronounced in wellknown english words when trying to tell someone how to pronounce something in another language like for example german? Because it is just easier for the human brain to connect the pronounciation with sounds they already know instead of using signs that should serve the purpose to remember that sounds but are so rarely used that they just don't. Thanks for sharing this insightful discussion! :)
Nah, tbh it could well be that it wouldn't hurt to learn at least parts of the IPA in like year 5, and then actually use it, to easier learn languages. But then again, especially these days, we can just listen to a word in an online dictionary. Besides, pronounciation isn't the absolutely most important part of language anyway, grammar and the vocabulary are probably way more important I don't though get why Mr Klein here had to dump on this girl like that, it's not a terrible idea at first glance, we do a pretty similar thing with the Latin alphabet for most European languages - i.e. most people on four continents write with a Latin alphabet.
I had this idea myself a few weeks ago too, but my immediate reaction to it was 'and that sounds like it has a dozen huge holes in it' so this video is exactly what I was looking for! Though, while it's still clearly a bad idea due to the other core problems, some of them could be solved by inventing a different IPA that's actually comprehensible as a script and not eurocentric. Doesn't fix things like the dialect problem and so it's still a bad idea, but it's interesting to imagine how much closer we could get to a decent option.
How is the dialect problem a problem when it's the whole point of the reform? To defeat centralising prescriptivism, and force people to be cognisant of their speech.
There's also the problem that, as any undergrad phonetics student realizes... it's arbitrary. The class is dozens of hours of analyzing speech and by the end of it any two prospective linguists are STILL going to disagree on a paragraph's worth of spoken word in many places. This isn't to say we don't have tools for analysis - you'll see more of Praat than you ever imagined - but it's not like there's one single quantified threshold that separates an /ɑ/ from an /ɔ/. And that's if you can even hear the difference.
A lot of symbols sound very similar, if not identical, and other sounds seem to be unrepresented. Sometimes symbols get used to represent two different sounds too, like when the schwa (COMMA vowel) gets used to represent the sound at the end of the word ‘comma’ and the French word ‘je’. The main unrepresented sound is the way that most people in Birmingham (the one in England) say words like ‘cut’ or ‘put’, it’s not normally the same as the way someone from the North of England would say them, it’s more like how someone with a very broad West Country accent, or some people in some parts of the Southern states of the U.S, as well as many African-Americans, would say them.
I never heard of this and at most reminds me of the time the French tried to make a Metric system for Time (it failed because they ended up keeping the other time measurement around causing confusion)
I thought of this when I first took a linguistics class, but thanks to my studies of Japanese, I very quickly disabused myself of that notion. Japanese would be COMPLETELY INCOMPREHENSIBLE using the IPA due to its poor sound variety and pitch accent. There’s so many potential readings that are used by multiple, even dozens of different kanji - how the heck do I know if this ‘shou’ is the one in front of ‘elementary school’ or the one in front of ‘young man?’ As much as I love the idea of a true international language, the IPA is not it, not even close.
Why doesn't humanity make a new alphabet based on the list of phonemes the IPA has, but instead of weird latin alphabet stuff, we start from scratch creating a BRAND NEW ALPHABET that takes into account all possible combos and creates the most complete yet hypersimple list of graphemes. There, solved.
@J Boss i like diacritics because they narrow so much the sound, but on their current form they’re so ugly. Maybe they can be integrated in letters, like in my native tongue catalan there is a difference between c and ç
@J Boss Ugh, for my part i would hate to have to write cz, as in, two letters for one sound. Surely an alphabet can be built which has a grapheme for every phoneme. At least up to a point, because if you count every letter + diacritic combination in the IPA, you end up with hundreds of letters.
@J Boss do you speak multiple languages? When i was 5 i could speak spanish catalan french dutch and english. There is so much nuance that is completely lost with the current alphabet. NO WAY the portuguese and french ã is an. It sounds completely different from that.
@J Boss But then you're rendering useless the whole idea of having an actualized IPA alphabet. The spelling you propose favors some languages over others. It's not universal.
3:16 That Malayalam text means something closer to "-of dogs". "Dogs" alone would be "പട്ടികൾ". Should be. It's one way of saying it, in this case it's using the same root word. Just a plural instead of both a plural and a possessive form (is that the right term?) I should probably mention I can't actually read or write that well in Malayalam since I didn't grow up there so I never picked up that skill. My parents however speak it natively and having to speak to them (and other family sometimes) in it is what kept me from forgetting what little of it i knew.
The problem with the International Phonetic Alphabet is the same problem with globalism, the fact that one-size-fits-all ideas look good on paper, and the people devising these schemes never consider all the thousands of myriads of details that billions of individuals across hundreds of different cultures and languages HAVE to agree upon just to make the scheme functional.
@@aronasmundurjonasson3175 What the devil is WRONG with you? I hope you tell a mob boss that. Then your filthy, disgusting mind will be sleeping with the anchovies!
@@rama7267 You misspelled "plausible" Rama7, let me make a video about this run-of-the-mill understandable mistake of yours, dissing you for some reason
This is already a very short video, but I feel like if anyone ever presents the idea of everyone using the IPA for everything, it can get shot down in one second by pointing out that pronouncing is not the same as understanding or speaking a language. Like, it's just fundamentally a proposal that maybe, in an ideal world, might possibly fix one problem (reading a word and not being sure how to pronounce it) at the cost of creating probably literally a hundred thousand others in the realm of, you know, actually using a language in the ways that language use actually occurs. Like, this isn't really mentioned in the video, but the distinctions between languages (and even dialects) carry a lot of information about what language is even being written, make it possible for people to parse it and its grammar, etc. No doubt some of that carries over to IPA use given the different "profiles" of each language in terms of sounds used, but you would lose huge amounts of contextual info. For each way it might make your life easier (no more memorizing thousands of kanji or hanzi, I guess!) it'd make your life more complicated in a hundred thousand ways. You can already get a near-perfect phonetic reading of Japanese if you just learn kana, read stuff with full furigana, and hide all the kanji (and somehow forget they're even there, as that would be an unfair grammatical hint). Good luck parsing anything more complicated than an illustrated children's book.
@@bacicinvatteneaca reading IPA means you can pronounce a word, not know what it means, and actually removes a lot of information about what language is being spoken, information which is necessary to understand what the sounds mean.
im currently learning japanese and i legit cant even imagine how itd work. like there are so many kanjis that are pronounced exactly the same but are just written differently
By relying on Japanese and Western words over ambiguous Chinese-derived ones. Actually, do you remember Aoharaido? Its name uses kun readings of ao and haru for seeshun. Such calquing may also be used.
@@АлександрАлександров-щ7к8т how limiting would it become for the language's vocab tho- and sorry i dont understand your example. im assuming youre talking about ao haru ride? i dont watch that anime/read that manga so i have no idea what youre saying :"D
It's extremely easy, actually: the same way as in speech. If homophones aren't a problem in speech, they're not a problem in writing, and if they ARE a problem in speech, make them stop being homophones,
Anyone have any arguments against removing Q from English? Also I am unreasonably mad at that dialect map for Ireland labeling everything as just "hiberno-english". Accents in Ireland are just as varied as in the UK, if not more. Each county has at least 1 dialect unique to itself and more populated ones have multiple
There aren't many arguments against changing the consonant letters in English spelling to more sensible ones. Vowels are all over the place in different dialects, but English speakers mostly agree on consonant sounds. Q and X especially don't need to exist or can be repurposed for other sounds.
Q is the ugliest Latin letter. I would not be sad to see it go, even though I would rue its absence in Scrabble. That dialect map shows highly inconsistent levels of detail. If there is one Hiberno-English dialect group then there is one southern England English dialect group for instance.
Q only exists in English because it borrowed words from French and Latin, and now there's not much to be done A bit like K and W being part of the French alphabet, even though you only find them in words of foreign origin
There could be analytical ways of choosing sounds based on frequency and allophones. That could help as a basis for many reforms. Nice video, nice criticism
I could see myself making a similar argument while also not necessarily believing it. I think that's just how some people learn from other people. It's kind of like shooting beams of particles at each other in an accelerator and seeing what kind of particles are produced by the collisions.
1. Nobody wants to memorize hundreds of characters that represent a single sound 2. Having it be Latin based would disrespect thousands of years of history for non-Latin users. 3. You'll still be understood even if your pronunciation isn't 100% correct
Learning another alphabet isn't that hard. At least when they work similar to the one(s) you already know and don't use too many characters. Also a people's script is a strong part of their identity, just like their language is. Any plan to replace that with one globally standardized system is gonna get A LOT of resistance all around the world! I mean, pretty much every country in the world has people complaining about their regional dialect being killed by standardizing the language on a national level already.
There's also the simple fact that spoken language changes a LOT faster than written language
he did address that at around 2 minutes when he was talking about each person with talking problems would wright by ear
@@nameistanya Not every. Rough, tough, ghoul. English spelling isn't just historical, it was bad from the beginning and simply got worse.
@@nameistanya the modifier “being silent” is misplaced, pointing to the word “it” (which is pointing to the phrase “every word”). the way your comment is written, you are saying that every word with a gh is silent. “every word with ‘gh’ being silent in it” would be more correct
Which is why we wouldn't need any spelling reforms
Them changing isn't really too much of a problem, as historic spelling is pretty common.
The only time I've ever come across IPA is through my best friend, who is an opera singer. She's fluent in IPA because it allows her to sing in practically any language.
So she needs sheet music that's especially made for her with IPA lyrics?
I’ve also heard theatre and movie actors sometimes read in IPA because it allows them to fully nail a regional accent they aren’t used to hearing.
that’s kinda lit
@@aphrog649 Dat be kina lit mane, ya feelz meh cuz? lie fo reelz niqqa.
@@maliziosoeperverso1697 Hahaha no. We usually write the IPA above the staff/below the text while learning the piece. Similarly, we write the translation so we know what it is we're actually singing too.
If everyone in the world started using the IPA, after enough time, both graphic and phonetic aspects of any symbol may be slightly altered by the needs of each individual culture, making it not universal.
I mean... this is kinda already done. The phonology of a language that is decently well studied tends to use a fair bit of simplifications in their phonemic transcription. For example, English /ɹ/ or occasionally /r/ is in most dialects actually much closer to [ɹ̠]. This phoneme has a fair bit of different phonetic realisations too. Furthermore, /h/ in a lot of european countries is realised as a voiceless version of the followinɡ vowel. In english /ʌ/ tends to be closer to [ɐ]. In swedish /a/ tends to be closer to [ä]. Also in swedish /ʉ/ is understood to mean somethinɡ like [ɪ̙ᵝ]. /t/ may also represent any plosive from dental, denti-alveolar, alveolar, and postalveolar. What's more is that these may also be realised as either laminal or apical. Phonemically most of these consonants are just writen as /t/, but [t] typically seems to represent the voiceless apical alveolar plosive. Another example from Swedish is that /ɾC/ is often realised as a retroflex version of /C/, where /C/ can be /t, d, s, n, l/. So /rt, rd, rs, rn, rl/ become [ʈ, ɖ, ʂ, ɳ, ɭ]. Here's just a few examples how ipa is used inconsistently among different languages. Phonemic transcriptions oftentimes try to break down the language into ad simple a set of phones as possible, and also choosing to use the simplest symbols avaible to represent these, unlike phonetics which tries to get more accurate transcription, instead losing generalisation among the language speakers, but gaining a more accurate transcription of this particular speaker or dialect.
That's how it currently is, and it would FINALLY stop being that way if everyone switched to IPA, because people would be encouraged to actually spell hoe they pronounce.
@@bacicinvatteneaca Well, the idea is that the way people speak will always disconnect from the orthography, regardless of how precise it is. Are you implying (as mentioned in the video) that the only solution is to then have every individual pronounce words as THEY individually say it? Then where's the 'universal' benefit of the IPA, and worse; the orthography reduces in effectiveness because of how inconsistent words will be even across individual people. Sure, maybe we do go ahead and standardize, idk, English in the UK (so everyone spells things the same way with the IPA). But, then the problem of a gradual disconnect of speech to orthography reappears, making it no better than the standard English Latin alphabet-based orthography we have now. I'm not saying that the IPA would be useless to think of as a substitute, but I believe in no way will it make the world better with universal adoption.
The real wasting of time - is to learn different languages and then translate them 😆🤦
Not really, you just don't know enough to do anything but guess
“Ja ja ja ja ja ja”
Either comes from an amused Spaniard or an enthusiastic German.
Or from a Polish person who wants to assert his dominance.
Or from a Frenchman who had a stroke
ʝa ʝa ʝa ʝa ʝa ʝa
or a guy with a stutter trying to say jam
@@kubacakagoomba can you explain 😂
Many languages, particular Chinese, would actually become ambiguous and harder to understand with IPA, not the other way around
They could always do the Vietnamese thing. Write the tones if they’re critical to understanding. Obviously they’re going to be writing two to six times the total number of characters, but there’s a chance that someone without the time to memorize 12,000 characters (and come to grips with the 102 radicals) would still be able to read four words in a newspaper with a couple tries.
@@polyhistorphilomath It's not about tones but rather the phonotactics. Chinese has a lot of phonemes with the same tones but with varying meanings. What those words/sentences mean depend heavily on the context of adjacent logographs and their patterns. Correct me if I'm wrong, but I think that's the gist of it. It's easier to just understand the current writing system than the IPA.
Edit: grammar
@@polyhistorphilomath There have been several attempts to romanize Chinese characters, but in the end it was mainly use for pronunciation purposes. To keep it short, Chinese have very rigid restrictions on what kind of sound can make a syllable and a one syllable one character rule. This leads to a huge number of homophones.
I’ll give you an example with English. You can obviously tell the difference between to, too and two even though they sound the same, but would you ever want to write them all as “to”? Let’s start with the phrase “I want to”. Without context this single phrase is ambiguous, but with context the meaning become clear. I want to (two) chicken, I want to play to (too), I want to (to) go home. Now imagine every word also have 4 homophones. I have read texts in bopomofo when I was growing up and I can tell you it is a pain to read. If they are going to do it, I rather they do the Japanese thing where they add pronounciation (furigana) along side kanji.
@@reachforbluesea I think I see your point. If there are many homophones and relatively few possible syllables (when transliterating without tones) then many of the advantages of alphabetical writing are lost.
Apparently the size of the syllable inventory in Vietnamese (as the obvious counter example-tonal, but written in Latin characters) is different. Feel free to correct me, but it must be large enough that the possible word shapes also have a wide variety. I assume this because the Vietnamese actually use their writing system, after all. I can’t imagine happily using the Latin alphabet for a language where word shapes were not helpful in reading quickly.
Just to be clear, I mean word shape in the sense of the rough outline and/or other visual peculiarities [such as the location of denser patches of character strokes relative to the outline] in an alphabetically written word.
@@polyhistorphilomath reading into Vietnamese and Korean a bit more, both Hangul and romanized Vietnamese were imposed specifically so commoners didn't have to go through the very rigorous schooling required of the scholars.
and, well, it worked.
IPA would be like if a musical scholar thought "our musical notation system only applies to the 12 tone scale. Other countries use 5 tone scales or even 22 tone scales. To account for this I've created a new way of writing music where every note is written as its frequency in hertz and the amount of time that note is played is written in milliseconds above it. Techniques are all represented by a unique symbol, for example pizzicato is represented by the letter д with an umlaut and rotated 135 degrees clockwise. Also, "pizz" is written below it in italics to avoid confusion. The musical score also comes with a golden record which when played in a projector will display a semiotic representation of a string being plucked as well as a spectrogram and a diagram of a hydrogen atom. This is just to show that we are an intelligent species to any alien species that may come across these musical manuscripts"
Ok the last parts are unecessary, but the first part... That doesn't seem that bad.
@@pedrosso0 yeah, he almost got some good ideas but chooce make a joke.
@@vincentwhite938 He did. I would actually like to see a generalized system of writing down notes and styles
@@pedrosso0 I was trying to make an overly absurd and convoluted system as a joke however, our current notation system does have its limitations and having one that accounts for different cultures as well as modern music would be useful. Imagine if it was also designed with music software in mind. It could be a happy medium between programs like musescore and programs like abelton and pro tools.
Yeah, people saying that this is a good idea kinda forget that different styles use different temperaments. For example, in "standard" Baroque tuning, the A is at 415 Hz, and in French baroque tuning, the A can be as low as 392 Hz (that's a whole tone down!)
You missed the best reason: how large a keyboard would one need?
The key-cap industry would become very lucrative
Might I interest you in Chinese keyboards Mister Sir?
@@paranodrum9171 Aren't Chinese keyboards actually in latin and then through software different combinations become Chinese characters? That's at least how all the keyboards I've come across work.
@@masterkamen371 There is no one standard Chinese keyboard as there is for English, though I believe the ones you described are the most common, where you type how the character sounds in the latin alphabet and the software gives you a number of sugestions.
@Rtu Barmen I assume that's how cunnilingus go about it. Now tell me with a straight face it's not cumbersome :)
Remembers seeing some linguistic/orthography subreddit where whenever someone made an English spelling reform there'd always be someone saying "wAɪ̯ NɒT dƷʌSt JuːZ Aɪ̯ Pi Eɪ̯"
Lol I can read that, I like IPA
@@midloran me too
@@midloran That's not IPA if uppercase letters are used, though...
I read
”Why not just use IPA”
Waɪ not?
the thing about the IPA being ugly kinda makes me wanna try and make """cursive IPA"" now
u mean cursed right?
Argh. You would be making something like Arabic script but stuck in the opposite trap. You want 900+ distinct glyphs to be all swirly. Good luck. Arabic is hard enough to read. IMO. It’s like a post-reform abjad that only got to keep five distinct letterforms-leaving all the work of disambiguation to be done by sporadic dots. Okay so now all we get to read is swirls and dots. Congratulations, it is illegible.
@@Smeiksmeiksmeik cursive IPA? More like cursed IPA
It does exists en.m.wikipedia.org/wiki/Cursive_forms_of_the_International_Phonetic_Alphabet
@@polyhistorphilomath Standard written Arabic is OK, kufic is harder to read, nasta''''''''liq is illegible.
People do cursive Vietnamese and it's even stupider than the standard form. Suffice to say if I had to read that language for any length of time I'd be typing this comment in Braille having gouged my own eyes out.
I actually had the same idea of “Why can’t we just use the IPA?” sometime within the last 2 weeks. Then this video came into my recommendations, and made a lot of good points.
səʊ wɒt ɪz ɪn ðə weɪ ɒv əˈdɒpʃ(ə)n?
Finnish is pronounced pretty much exactly as it's written. People speak in everyday situations a bit differently to how they write but it works just fine. Some of the critiques do still apply.
The IPA obviously makes no sense as a language since it’s made for translating between languages. But in making artificial languages, like Esperanto, I think phonetically consistent is something that would be extremely useful going forward, so there’s no ambiguity for the word.
It mostly boils down to most languages having incosistent or convoluted rules for spelling things. And we can't forget that asian languages which depend on context for the words to make sense would just make reading harder, since it's not rare for different words to have identical spelling. I can't trally think any language aside from Greek, that would benefit from this change
@@Youcanatme
...
"So what is in the way of... adaption?"
west african languages: i'm gonna pretend i didn't hear what i clearly just heard
Click plop click click
@@fluffynator6222 thats more southern african, west african would be like ngbwã bha
literally ipa
@@mahatmaniggandhi2898 lol how did you get that middle name?
@@fluffynator6222 I just imagine those words between those IPA hieroglyphs
As someone who has to interact with phonetics fairly frequently, Vocaloid SAMPA, specifically whatever modified version of XSAMPA Hatsune Miku's English bank uses, is about 27 times worse than IPA. I hate XSAMPA so much.
Being a vocal synth user is a punishment unlike any other I have learned many phonetic systems against my will. Vocaloid SAMPA, Delta (which is a modification of Vocaloid SAMPA's English transcriptions, meant to be easier for Japanese speakers to read), Arpabet, CMUDict for French (this SUCKED bc its only meant for European French and I speak Canadian French), CZSAMPA (a modified version of Vocaloid SAMPA based on written English and the GenAm accent), Erable French (I modified CMUdict for Canadian French, gods preserve us), the IPA I need for English and Japanese, voice synths are a special type of hell sometimes
Well I mean that's mostly because SAMPA is literally just. "help I need to use IPA but I only have ASCII" lol
@@jolkert_ Vocaloid SAMPA is even WORSE. It's SAMPA but modified and also phonetically inaccurate. With the example I gave of Hatsune Miku's English bank for Vocaloid 4, she sings in Japanese-accented American English. Vocaloid SAMPA was written solely for Recieved Pronunciation and did not add symbols for key phonemes in American English (the dialect most English Vocaloid VBs sing in) until the release of Cyber Songman for Vocaloid 4 (yes that's the real name of the voicebank) in 2015, and only as an optional extension to English voicebanks.
I am so eternally grateful that SynthesizerV uses Arpabet for English, Hepburn for Japanese, and only pulls out the XSAMPA when it's dealing with Mandarin
vocal synths are the world's worst gateway to linguistics and i will die on this hill
we found the vocaloid user
The IPA is way too technical to be used on large scale. Sure, I once terrorized my friends on a groupchat using only the IPA for like a week, but imagine using it everywhere. The IPA is a good method to supliment language learning, but the actual writing system carries on history. I personally think English spelling is okay as it is, and this comes from a person that had to learn words one by one, like logographs since primary school (that is how we learn English in Romania). I cannot imagine English, French or Greek without the horrible historical spellings.
Also, there are languages that didn’t have any written form, so linguists decided to write them in the IPA officially. So the IPA is actually used on a large scale to write some endangered languages.
But Ai fowr wun think Ingliś wuud be betur ouf with æt līst fonetic spelings (awlthoh it wuud make menī wurds hārdur tū distingwiś frowm īč uthur, laik “bee” & “be”), its curint spelings ār u mes. Awlsow rīnaym thu leturs acowrdinglī plīz.
@@badpiggies988 That spelling reform is cursed. I also tried one ever worse, let me switch to it. Ay̆ ëgrĩ ðæt yt yz hād țë lēn fë mëŭst p̌ĩpĺ, bât Ýñglyš, Frênč, Grīk ë Mǽndëryn wũdń’t bī Ýñglyš, Frênč, Grīk ë Mǽndëryn wyðáŭt ðë hốrëbĺ hysțốrykĺ spếlyñgz. May̆ néy̆tyv lǽñgwyǧ, Rëŭméy̆nyën, ķūd õlsëŭ jūz sâm spếlyñ ryfõmz, bât yt wūdń’t fīl ðë sey̆m. Yc ǧâst may̆ ëp̌ýnyën.
Fīl frī țë lāf æt mī fë ðys țếrëbĺ, bât fënếtykli ǽkjurët ryfõm.
@@badpiggies988 Бро̄ контэкст кӓн дисти́нгўиш сэд ўӑрдз. Дьі о̄нлі різӑн дье'р кӑнси́дӑрд дифрӑнт ўӑрдз из біка́з ов аур шиті ортьо́грӑфі.
@@topazbutterfly1853 Aj lajk haw jū hæv dhė old Tėgálėg skript, Bajbajin, in jr displej nejm
@@tompeled6193 I feel intelligent being able to read this.
The IPA isn't even used consistently within English. It depends on the context and how narrow and updated a convention is. It's also not narrow enough to transcribe a lot of phonetic differences between languages.
Yeah it is very common to transcribe the wrong vowel, forget about aspiration or mistranscribe a rare sandhi. Native speaker often don't notice these things in their own language.
That's the whole point of mandating its phonetic use and completely abandoning phonemic considerations: force people to be cognisant of how they use their mouths. No conventions, ever.
@@bacicinvatteneaca I'm sorry but that is the most ridiculously terrible and batshit crazy, or simply ignorant take on language I have ever heard. "conventions" aren't the problem(or the only problem).
Allophones are. Dialects are. Ideolects are. It would be impossible to destroy any one of those things with ANY AMOUNT of draconian linguistic prescriptivism.
Being "cognisant of how they use their mouths" is not something you can expect of people at least to the degree you are suggesting. Without even knowing you, I can guarantee that the sounds you make are not always the ones you think you are making phonetically.
And it wouldn't even be "better". There is a significant amount of phonetic sandhi in English that would literally do nothing but needlessly complicate spelling by forcing a spelling distinction where it would be irrelevant. As a concrete example what is the benefit of transcribing the word "symphony" with an /ɱ/ and the word "more" with an /m/? There are no minimal pairs between them. 99.9% don't know that there is a different sound that exists there. Literally nobody would notice if you made the wrong sound there. In fact, most people if they were asked to say the word slowly would make the wrong sound there.
Your suggestion is about equivalent to. "That's the whole point of mandating that people only ever die in coffins." Yeah it would be more convenient in some narrow way, but how exactly are you going to get it to happen? And how exactly would it be better? It certainly wouldn't make it easier to teach reading or writing.
@@bacicinvatteneaca [bliv mi wɛn aɪ tʰɛɫ juː, ju dõ wɑ̃ˀ ɛvɹibəɾi ɹaɪɾɪŋ ə kʰəmpʰliːˀli ækjəɹɪˀ tʰɹænskɹɪpʃn̩ ɔv wʌˀ ðɛɚ ækʃuəli seŋ. moʊs pʰipɫ̩ wʊn̩t bi eɪbɫ̩ tʰə ɹid θɪŋz ðeɪd hæv noʊ tɹʌbɫ̩ ʌnɾɚstændɪŋ vɝbɫ̩i. ðɪs izn̩ˀ ivn̩ ən əbskjɝ daɪəlɛkˀ, d͡ʒʌs maɪ neɪɾəv vəɹaɪəɾi ɔv kælɪfɔɹnjə ɪŋɡlɪʃ wɪt͡ʃ juv pɹɑbli hɝd ə mɪljɪn taɪmz ɔn tiːviː, bʌɾ ɪf ju ɚn̩ˀ ɪkstɹimli fəmɪlɚ wɪθ boʊθ aɪ pʰi eɪ æn ðə pɪkjuliɛɹɪɾiz sʌmtaɪmz pɹɛzn̩ˀ in əmɛɹɪkn̩ daɪəlɛks ðɛn ju pɹɑbli wõˀ ʌnɾɚstænˀ mʌt͡ʃ]
Is that what you really want? You expect kids to be able to learn to write that way?
Yep. Take, e. g., the [æ] sound in American English (but not in British). Words "back", "ban" and "bank" will be transcribed with the same [æ] symbol, but in fact they have three different vowels. Or the famous "dark L" in English, which usually does not have a separate transcription.
I showed this to my speech pathologist mother, who said she had never seen "don't" written like it is in the thumbnail. She guessed you would have a British or Australian accent.
Also she said she used to write letters to people in IPA just to see if they could figure it out.
Another part of the problem is the incompleteness of IPA. Some languages (especially East Asian languages) include pitch in the pronunciation of words. Two words with an identical IPA representation can mean different things simply because the pitch structure of the word is different (for example, in Japanese).
I think there are diacritics for pitch in IPA but yeah, still not good.
@J Boss Fair point--though I was meaning to say 'having everyone use IPA to make cross-lingual pronunciation easy' is probably a bad idea because it is incomplete with regard to that aim, if not for countless other reasons.
As the first guy already mentioned, though, although the IPA *isn't* the ITA, tone markers do exist in the IPA. They're, uh, a bit questionable, like everything else in the IPA, but they're technically there.
If I were to complain about anything, its that there's no real (good) way to express rhythmic differences in speech. You could have different people who speak various languages with similar phonemic inventories all say the same sentence, and they'd *still* pronounce them differently, with no real way for someone reading the IPA transcriptions to know. Of course, as you've already said, this isn't particularly in the realm of the IPA, but it'd nevertheless be interesting to see an attempt made to accommodate this
@J Boss Mostly, it'd be interesting to see how they'd go about making a system that captures that level of detail. It'd take more than just a single subscript number, though; I'd imagine it being akin to extremely detailed and complex musical notation.
I'm a programmer, too, and I can definitely say that if you think audio parsing *and* speech synthesis on the level of being able to produce the "exact same sound" is "easy", then you should definitely get to making that, because that's state-of-the-art technology right there that people've been trying to achieve for years. My man, you could likely make hella money from something like that, and, if you're truly willing, I'd kinda just wanna see what you come up with either way tbh, cuz, as I said, it'd be interesting
@J Boss Ignoring the (possible) hyperbole in the number of subscripts you'd need, combining all of the data into a single subscript seems like a really bad idea. That's like taking every IPA phoneme, along with every combination of diacritic for each, and making them their own sound. Velarised /k/ and /p/ no longer just have a superscript /j/ attached here; they'd be entirely new symbols. And thats ignoring the fact that you can just keep adding diacritics if you'd like. Even if it takes having 254k different subscripts, I'd much rather have that, and be able to add and subtract them as I like to get more or less detail in transcription, than to be forced to have the maximal level of detail described by having a character per every combination.
But even then, the idea I'm getting from what you're saying is that you think that, just because you have the series of binary data that represents audio, that you can then make a program that can take that data and somehow manage to parse it into IPA so accurate that anyone reading could convey the exact same sound. That, alone, is extremely incredible technology, and most modern solutions to this rely heavily on machine learning algorithms trained on a metric fuck ton of audio samples only to get a relatively accurate transcription of whats being said. If this was as easy as you make it seem, then, for example, UA-cam's automatic subtitles wouldn't be so absolutely bad at transcribing text.
That's not even getting into the claim that making a speech synthesis engine to play back the IPA transcription would *also* be easy. I've admittedly seen some pretty good speech synthesis engines, but you can usually tell when the speech you're hearing was synthesized. You're expecting technology on the level of highly specialised deepfakes that's as accessible as something like the Vocaloid speech engine.
These technologies *are* getting to the point you say is easy, but the sheer level of accuracy required in this hypothetical program that can supposedly take in any audio sample and both output exact IPA *and* exact speech synthesis just isn't quite there. The accuracy you say is easy generally requires extremely large data sets of audio, usually from a specific person, to get a pretty decent approximation. You *could* generate an IPA transcription of a given audio sample, and you *could* make a speech synthesis engine that plays back any given IPA transcription, but without heavy specialisation and a ton of work, I simply cannot see how a program could provide "exact" output that you say is "easy" or "not remarkable"
related to the dialects thing, but even if everyone somehow spoke the same dialect
how would you account for micro differences in speech
“oh no i accidentally added a tiny bit more height to this vowel and now i have to add a different diacritic”
of course simply calrifying you only need a phonemic transcription would help
but yeah dialectal differences as you said
Why exactly are dialectal differences a problem? The obligatiom to spell them out isn't a bug, it's a feature.
@@bacicinvatteneaca at a certain point, it would 1. probably hinder readability, or at least slow down reading speed, and 2. basically require people to have linguistic knowledge and phonetic awareness of their own speech enough to use the IPA accurately (using it inaccurately kinda defeats some of the purpose of it)
now assuming its purely phonemic transcription, then that still kind of requires people to know which of their phones are just allophones of a main phoneme; and also point 1 still stands
Erasing written language would also erase cultures. Many writing systems (Chinese, Arabic, etc) have a rich art of calligraphy. And some cultures have deep pride in their writing systems, like Korea’s Hangul.
I love Arabic and Japanese calligraphy they look beautiful
Even if we switched every language to IPA, all it does is tell us how to pronounce words in any language, it doesn’t encode any meaning so if we came across a post in another language, all we knew is how all the words sound, not the actual contents of meaning.
Also, I enjoy IPA as both a guide in Conlang making *and* singing international songs for choir.
Yes. And? That's the same as in speech, with the added benefit of spaces. So?
Best would be logograms for words and IPA for proper names like cities or surnames.
The thing about different dialects having to use different writing systems was actually an issue with my own (Rhenanian flat German) dialect which is why - up to the third Reich and its educational reforms pushing High German as the standard in both written and spoken language to push the idea of "one German people" instead of the hundreds that existed as semi independent entities untill 1871 - many people around here used to write in Dutch as it more closely resembles/d our way of speaking (nowadays hardly anyone can speak in pure dialect anymore and most speak in "Mundart" - High German with a regional accent with varying degrees of strength depending on how official the situation is and wether they are conversing intra- or intertribally)
ah, the wild Plüütdäääätsch speaker
Ist Platdeutsch nicht eine eigene Sprache? Und sind Dialect und Mundart nicht das selbe?
@@xhawkenx633 eigene Sprache ist glaube ich zumindest umstritten also zumindest eine Grauzone und Dialekt und Mundart können sehr gut im Hochdeutschen das Gleiche bedeuten, in meiner Region zumindest werden sie aber unterschiedlich genutzt. Wie bereits geschrieben ist Mundart nach meinem Verständnis Hochdeutsch mit Dialekt Betonung (Plattdütsch hat eine teilweise andere Grammatik und Satzstruktur, daher ist Dialekt sprechen zumindest in meiner Gegend noch einmal eine andere Sache.
Ich spreche z.bsp. Mundart im Alltag (t->d, g->j, ch->sch, und eine gute Portion an Lehnwörtern aus dem Dialekt) und kann Dialektsprecher problemlos verstehen, kann aber leider selbst nicht Dialekt sprechen
Same in France. There's no one people in France, France always was made up of different peoples and cultures with their own dialects and languages.
@@hmvollbanane1259 hm... Das was du als Mundart bezeichnest kenne ich unter dem Wort Akzent
The solution to accounting for regional dialects is to simply create universal logograms to be used in all languages instead. This plan has no flaws. Chinese characters will work for English with enough brute force.
If brute force isn't working, you aren't using enough.
I'm a native Chinese speaker who has learned some Japanese and is terrorized by kun-yomi. This sounds like a nightmare.
Forget the alphabet, let's just start writing every language in logograms. This will not go wrong at all.
Eventually we will invariable adapt Ithkuil to all languages.
Imagine if Chinese characters were adopted in English, similar to how the Japanese did, by mixing phonemes with the characters. Of course this isn't a good idea but it would look cool and it's definitely possible.
I think the issue in IPA is with it being meant for language analysis. IPA lists 68 different vowel sounds however most of these are not used in most language and many sound very similar to up to 5 other vowels as well as almost always being mutually exclusive to these similar sounds in a language. For language analysis these sounds may have nuances that need to be separated into different categories but as most of these sounds are mutually intelligible and/or indistinguishable, you could reduce these sounds to 20 or less letters. Consonants tend to have less similarities though and these 54 sounds can only be reduced to around 40-45 (there are some sounds that seem the same to me but are not in any languages i speak so I'm unable to give a precise guess), but at most we could easily make a simplified IPA with around 65 characters and separated into families of sounds (represented by a common letter of the family understood by most languages) so that if a sound doesn't match what someone wants to portray they can say what family its part of and use placeholder diacritics (which should not normally be used in this font) to show manner and place of articulation.
As someone with a linguistics degree when you said you hated those letters I felt that on a spiritual level. Damn retroflexives still have use, but doesn't mean I have to like them.
@J Boss This isn't a helpful take and I don't know why you thought it was.
come on man retroflexes are not all that bad, like would you rather have a t with a tail or with an under dot. at least the tail seems like a separate letter than just a diacritic. (i say this as an indian, where most of our languages have a clear distinction between dentals and retroflexes)
@J Boss your comment about german is so stupid, this would mean that japanese is a sino-tibetan language because it has chinese words in it
that's literally not how it works, and your comment clearly shows you don't understand anything about linguistics since you label it as unscientific and emotional, as if there isn't a gene named the sonic hedgehog gene, and the planets haven't been renamed to something more scientific than the names of roman gods
so how about you leave the linguistics to the linguists and go back to whatever sad angry corner you came from? :)
This is not even mentioning the fact that you would have a much harder time identifying any given language due to the lack of orthographic variety. French and Swedish passages could look very similar at times, despite not even being in the same language group. Furthermore, imagine *loan words* that are just tossed into a script. In English, you can identify a loan word or phrase by italicizing the phrase or using quotation marks (e.g. "en masse"), but I think a lot of the context behind whatever requires a loan would be lost if all lingual scripts were homogenized. Even in English, the language of borrowing, we can usually figure out what language a word is from based on features of the written word itself (for example, two Zs in a row that make a /ts/ sound signify an Italian word), but all of that context would be lost if we decided, "Yeah, sh and sch and ch make the same sound in shall, anschluss, and pastiche, so let's just disregard the entire origin and meaning behind the variation of spelling and just change it all to sh." A lot of people point out English's propensity to have many different spellings for sounds as a degeneracy, but I firmly believe it's a virtue, because the spellings can provide historical context to a word AND help you figure out the meaning of the word without having to perform rote memorizing (Latin roots, kids). So having only one writing system based solely on the sound of a word would just be a disaster in terms of finding meaning in words, and on a more superficial level, identifying the language in which a writer is even writing.
If you can't distinguish French from Swedish in speech, why should you distinguish them in writing? If English flattens the sound of loanword in speech, why shouldn't it do the same in writing?
“Timmy, you wrote your essay in Californian. You’ll have to rewrite it.”
“But we live in California”
“Sorry Timmy, you have to write in Londonian like everyone else.”
You sir,
You are filling the linguistics shaped hole in my heart after Xidnaf left, I salute you.
No one ever should use the IPA for any orthography is a true statement aside from 1 language
toki pona
It doesn't though.
It doesn't account for nasal assimilation- tenpo isn't written tempo.
Basically every sound has hundreds of potential allophones depending on your l1 - English speakers don't write aspiration markers, or distinguish between e ɛ and ei.
When I say the sentence "tenpo ali la sina jo e ma e wawa e pona" I say it 'tʰɛm.pʰoʊ a.li la 'si.na jo e ma e wawa e pʰoʊ.na
Which is close but not identical.
@@chair547 broad transcription
Based
@@chair547 why not say it like the former, or spell it like the latter? Or both. Both could be fun.
@@bacicinvatteneaca you're right that would be fun
why would anyone want to give up their culture and start using an uglier version of the latin alphabet? i (as a cyrilic alphabet user) would never give up on the cyrilic alphabet. it is so pretty and kinda unique
I misinterpreted this as saying that the cyrillic alphabet is an uglier version of the latin one and i was about to get very mad at anyone who thinks that because the cyrillic alphabet looks sick
Я would become ja which would effectively change jarule to яруль
I would never even change the Latin alphabet for the ipa
@@lpnp9477 jarul'*
Imagine wanting to change the Georgian writing system, if that was lost this world would try be beyond saving
your videos are great man keep it up
the german one blew my mind. been learning for 5ish years, and I never noticed the thing with s, ss, ß. Like I knew when to vocalize my s's, but just from seeing the pattern so many times, not from knowing that there's a rule to the pattern
Croatian, Serbian, Bosnian and Montenegrian all have phonetical alphabet. And realy easy to read and understand. We have 30 phonems and they are represented by 30 letters.
Svi se ponašaju kao da mi ne postojimo.
Je li ti zbog naše Politike, ili je to jer Balkan jednostavno ne postoji za njih?
@@cakeisyummy5755 ma lik je vjv nije ništa istražio za video uzeo je 10 največih jezika i misli da je rješio sve svjetske jezike. A mi za njih navjerojatnije ne postojimo.
@@antoniobobek8845 Točno brate!
Also, some languages such as Chinese and Japanese become really confusing when written phonetically, they have relatively few syllables and thus relatively many homophones making their system of pictographic characters very important in written communication.
@JM Coulon no, it didn't. korean has that exact problem because all the homophones are written the same way, so they become context dependent...
If you can handle them spoken, you should be able to handle them written. If you can't handle them spoken, the languages should probably change.
@@bacicinvatteneaca if it was that simple then don't you think china would have simply thrown away hanzi instead of creating simplified chinese? there's a reason these writing systems are still around and that's because they work for the language they were created. i think what people forget when they propose these kinds of spelling reforms is that writing systems have history, and hanzi and kanji exist in their current state after thousands of years of use. there have been multiple opportunities in both japan and china's history for them to do a writing reform and completely throw out hanzi/kanji to raise literacy rates but they didn't.
also, what's the point of creating a completely phonetic writing system? sure, spelling becomes easier, but when you see a new word how do you know what it means? shallow orthography loses out on the advantages of deep orthography, where you can understand meanings of the words you don't know just from the way they're written. you need to remember that japanese and chinese have evolved alongside their writing systems for thousands of years, which means that words have been created in a way that makes sense when you write them in hanzi/kanji. in japanese, I'm able to understand unfamiliar field-specific terminology (like medical terms) just from knowing what each kanji represents, whereas in english I would have to look up what each term means, if i'm not lucky enough to remember what each latin or greek-derived part of the word means (if it even is derived from either of those). keep in mind I only have middle school level education in japanese, with some low effort "self-studying". hanzi was also used as a written lingua franca for years because even though spoken chinese and japanese are not mutually intelligible at all, because they shared writing it allowed them to communicate. in fact even in modern times, my own mother was able to sort of communicate with the chinese grandmas at my elementary school because they would sit outside and write what they were trying to say, and growing up I would navigate chinese websites like tudou and bilibili to try and find videos that would get copyright strikes on youtube.
you might say that it's SO MUCH to learn because in daily use, chinese people need around 3000 hanzi and japanese people around 2000 kanji! a phonetic system it would reduce that to only the phonemes each language has! but if you think about the fact that you put hanzi/kanji together to create thousands of new words.... and the amount of words any language has... that's several thousand definitions that become intuitive just because you already understand what each part of the word means. it's impossible to do that with a purely phonetic system. one of the purposes of written language is to efficiently convey meaning. hanzi and kanji do an amazing job at that, so why should we get rid of them simply because you can't sound them out just by looking at them? there's a lot more to language than what sound each thing makes.
I'll never understand why this isn't considered a problem in the spoken language
@@Pashmimi "shallow orthography loses out on the advantages of deep orthography, where you can understand meanings of the words you don't know just from the way they're written" ...unless you don't know the logogram
I always tell people that come up with these ideas that if it is so easy/simple to do something, then we'd be doing it already.
how does this not have hundreds of thousands of views lmao
because this gentleman just started
were early anon
Sure it isn't just a 17-year-old dissing his classmates for "being so staggeringly stupid" about a very nieche topic?
Having 200k+ Views is pretty Good
I remember being told about the IPA when I was in school and this idea was my first reaction. Obviously, my six year old self was not a linguist. I'm glad that, after all these years, I finally have an answer to my then incredulous question. You really do learn something every day
It's even funnier when you realize that some sounds in your language go outside of the most known characters in IPA. We had this dilema in class with English native speaker and we tried to explain how to pronounce our names using IPA. Turns out that it's hard to explain a sound that lays between to sounds commonly used in English.
0:27
Woah! The IPA made the IPA
In my native tounge, Swedish, there are often variations in the spoken word that indicate how formal your speech is, or to make emphasis or for clarity. Beyond dialect that is, so we're talking about the same native swede in the same dialect pronouncing the same word differently depending on circumstances. So in informal spoken swedish most endings of the most common words are simply usually dropped, but if you put a stress or emphasis on a word then those dropped letters can come back again. You have this sentence for example: "Var vill du vara" = "Where do you want to be". But most of the time both "var" (where) and "vara" (be, exist) are pronounced exactly the same. As simple "va" with a long A, as in how a posh British person says "bath". And then you have that the word "vad" = "what" that is even more frequently pronounced as "va", that ending D is rarely heard even in more formal settings.
So sometimes in this type of sentence the first "va" might often be pronounced in full with the D or R at the end just to make the distinction between if it's asking about "Where" or "What" clear. But if the context already makes it clear then we will immediately drop those pesky endings and just say: "Va vill du va" and let the situation and our intuitive sense of grammar make the distinction between the words "var", "vad" and "vara". As a side note, formality here is actually determined more in how you pronounce the word "vara" than in the other two words. Even people reading the news on TV might frequently drop those D's and R's in "vad" and "var". But they'll rarely contract "vara" to "va" while reading the news. And swedish is loaded with this kind of stuff, silent letters that aren't always silent, occasionally you have multiple choice in how a word is pronounced, and we have words that even in IPA would look identical but are instead distingiushed by stress and rythm and melody e.t.c. How could you possibly make a spelling reform to incorporate these kind of things?
broad transcription vs narrow transcription
but let's admit the "broad transcription" taken to a logical extreme, is just standard spelling.
My standard English textbook in schools in China has IPA for all words and I memorized them, I found it useful in communicating English sounds at least..
The IPA is basically Tekken inputs in scholar form
Wow, i think this is the first time that i'm fully convinced something would be a bad idea from being convinced that something is a great idea, and in less than 4 min, Bravo!!
@J Boss
Heb je de video bekeken?
¿Viste el video?
Avez-vous regardé la vidéo?
Hast du dir das Video angesehen?
Viu o vídeo?
Hai guardato il video?
আপনি কি ভিডিওটি দেখেছেন?
あなたはビデオを見ましたか?
Apakah Anda menonton videonya?
你看过视频吗？
Pinanood mo ba ang Video?
क्या आपने वीडियो देखा है?
Вы смотрели видео?
Umewahi kutazama video?
Ingabe wayibukela i - video?
Did you watch the Video?
How so? The arguments were awful...
Fun thing about using a standardised dialect to spell things:
If people are recognising sets of letters as words, but the sets of letters they're reading don't conform to the accent they're speaking in, they have two options:
1. Change their own accent. This is a fair amount of work, and good luck getting lots of people to do it.
2. Just learn to recognise that this clump of letters is that word, and speak it the way you always have.
The latter means that now the same set of letters will now be spoken in different ways by different people, and that's exactly the problem we already had. So, uh, yeah.
"Use the IPA for all writing"
"Wot? The Oy Pee Eh?"
@J Boss And if you did it for English you would have to convince the entire Anglosphere, around half of Africa, India, Pakistan, the Philippines, and ostensibly everyone involved in international trade to do the same. Easy.
@@joshs5577 And after finishing off English, we would then have to do the same with every dialect of every language currently in use on earth. It would be a monumental undertaking at best, and borderline impossible at worst.
What is a linguist’s favorite alcoholic drink?
IPA
randomly finding your channel on my home page was a blessing. i love your videos
oh wow love the captions
Curious, has there ever been attempts to create a completely constructed script to replace the IPA that's more featural and less Euro-centric? Seems like something that an amateur linguist or Internet group might've attempted, after all the countless English spelling reforms proposals.
Presumably you'd need to derive it from another region as "standard" instead. While logically you'd do that with Africa, the fact remains that linguists specialising in Africa are much thinner on the ground. Eurocentrism in the IPA primarily comes from pragmatism rather than purely from bias.
We don't need that, thanks. Every language should just use their own writing system. The IPA is a tool.
I think he meant reinventing an alphabet completely, solely according to the qualitative diversity of human speeches (without any other bias)... Ot eventually that the only other bias would be the numbers of speakers.
@@Turagrong Even a total reinvention is going to be biased towards of the specialty linguistic sphere of whoever made it. "Zero bias" is impossible so long as anyone is interpreting anything.
@@Halophage If it's made by a team consisting of people "from around the world" who all also can speak some representative languages of other major groups to a good level... then not that much...
0:20 you made a mistake here. You accidentally used the Australian world map
This sounds like a great idea for a dystopion sci fi short story.
All of the "how to pronounce" searches I've done in my entire life do not amount to the time it would have taken me to memorize the stupid fucking IPA.
And how much time does it take to memorize the stewpyd phukkin English orthography?
toki pona: Allow us to introduce ourselves.
Polish: Źdźbło
Me: ʑʥbwo? Impossible! No one could have that as a word, let's wait for the IPA.
The IPA: ʑʥbwɔ
Me: Cancel the Polish language!
Cancel the Danish language too!
The hangman game would be so fun with IPA
Whenever I consider spelling in IPA, I can't agree on the specific phonetics of that very sentence. So yeah, amazing as an auxiliary, less so in everyday writing.
Maybe a less ambitious and more flexible option would be more coherence between phonemic approaches. Explaining Chinese "zh" and "r" to fellow French gets old very fast.
Ben tiens, tu me rappelles quand j'ai dû expliquer à mes amis que Ganyu ne se prononce pas "ganne-you" mais bien "ganne-u", ou bien que "zhongli" c'est pas "jongli" mais "tchongli"
Le pire c'était "Kazuha", je me suis fait reprendre pour l'avoir prononcé "Kadzuha", alors que eux le prononçaient "Kazoua" x)
@@aronasmundurjonasson3175 Only 2020's problems: la prononciation dans Genshin.
Ça illustre bien le problème. Entre les langues qui prononcent Z "z" et celles qui prononcent "dz"; les variations pour étendre le répertoire ZH/Ž "j" et les Chinois qui ne font rien comme les autres "dj", c'est tellement confus que tes potes ne sont qu'à moitié en tort.
"...any language anywhere"
toki pona: 👁👄👁
I have a better idea: everyone should just speak Saami and forget other languages. That would eliminate confusion everywhere.
I thought it's called Sámi in English
@@suakeli English keyboards i believe don't have á
Dude I really love what you’re doing here, keep it up!!
Just imagine IPA keyboards...
"So, to type that sound, you have to hit control twice, then press alt + shift together for two seconds, release them, then press control + alt, release, then dance three times around your room to summon Satan and then you press the A key and you're done"
Just stumbled upon this video and I love the passion this guy has for linguistics
A thought I had on getting rid of C is that since you're already making a large scale reform, you go all the way with it. C becomes what is currently CH because it's unintuitive to non-native speakers for 2 consonants to combine and make an entirely different sound to its constituents. And the softened intervocalic S in your example is always written with a Z. Given the context of that you can probably imagine what I'd do do eliminate pairs like SH, PH, and NG.
As you can probably tell by now I know nothing about linguistics and would like to know why I am dumb.
In German, we once had a system where sometimes ss is always pronounced "s", but sometimes the preceding vowel is pronounced long, sometimes short. It got confusing with words like "in Massen" (within limits) and "in Massen" (in masses). The spelling reform turned the one where the preceding vowel is pronounced long into an ß (15 years later or so also introduced its capital version as ẞ, I think they had forgotten about that lol), so now it's "in Maßen" and "in Massen". This reform was made in 1996, as I understand it, shortly before I was born, so in those days, but slowly less and less over the years, older people started using the new rules
We should keep C but use it for the [t͡ʃ] sound. So words like child would be cild.
@@adio5071 there are south east asian languages that do that already actually xD
Well you could just spell Ch’s with Tsh which is actually better.
@@anonymousigggsoo3664 why adding more unnecessary letters to the consonant clusters?
So you hear one dumb idea from a classmate and now we have to deal with your wrath?
I get why you didn't get into this, but there's also the eternal question of "should we do a phonetic or a phonemic transcription?" Like, do you write the American word "city" as /sɪtɪ/ or /sɪɾɪ/?
ヘイ シリ プレイ ああああああああああああああああ
Phonemes are intresting, but I believe that as long as you're desinging that writing system for only one language, they would make the system harder without any benefits in almost all cases.
@@mihneaserban2746 so you mean phonetics? Because surely phonemic transcriptions would actually work to some degree
What kind of question is that? The whole point is to make it phonetic, so that one can signal differences in pronunciation between dialects, and so that people become aware of the sounds they make. There shouldn't be any conventional spelling ever.
Also, I'd write it as /sıdı/ because that's how I hear it
I love you videos so much. They have a very Bill Wats feel to them and your voice is very calming!
2:42 I have dysgraphia, reading a text of mine would be equivalent to learning Cyrillic and Greek simultaneously
When you think of a perfect comeback two weeks later in the shower
The only good use of IPA, apart from learning a new language, is the promotional campaign of the Polish city of Szczecin. Foreigners cannot pronounce it, so there were billboards with the name written in IPA, I thought it was so smart and hilarious.
"Stretching" with a bogan accent
Can we just appreciate the irony that the p in IPA is not the phonetic letter for the word it stands for?
As a person who understands IPA i find it harder to read pronunciation spellings like “AA-muh-kraan for omicron or -pree mee-uhm for premium
Same, it’s messy as hell
not forgetting the massive overuse of hyphens and h's, or the insistence on the writer's particular middle Yank or Melbourne accent being the only correct pronunciation
@@gurrrn1102 sometimes I wonder why we don’t teach first graders IPA and they can use it later on to clarify spelling as they learn new words
Respelling is a lot easier than IPA to me 🤔🤔🤔
And I'm not a native English speaker. To me, that is just the best way to represent English phonetically, just like Pinyin for Chinese.
IPA is designed to be international and work for all languages.
If it would lead to everyone understanding every language in the world it would be a good thing, but sadly it doesn't and so doesn't justify the work that had to be done to first convince every language representive in the world of this idea, then teach everyone this language and change every written word to fit the IPA norms.
I learned french and english in school and sure the ipa stood in brackets behind the vocabulary lists we had to learn, but did I use them? Most of the time not because even though it is given you don't really learn how to read it and aren't that interested to change that because once you know how to spell a word the IPA is never really relevant again. Also if it would be really that intuitive why are people just refereing to the letters and how they are pronounced in wellknown english words when trying to tell someone how to pronounce something in another language like for example german? Because it is just easier for the human brain to connect the pronounciation with sounds they already know instead of using signs that should serve the purpose to remember that sounds but are so rarely used that they just don't.
Thanks for sharing this insightful discussion! :)
Nah, tbh it could well be that it wouldn't hurt to learn at least parts of the IPA in like year 5, and then actually use it, to easier learn languages. But then again, especially these days, we can just listen to a word in an online dictionary. Besides, pronounciation isn't the absolutely most important part of language anyway, grammar and the vocabulary are probably way more important
I don't though get why Mr Klein here had to dump on this girl like that, it's not a terrible idea at first glance, we do a pretty similar thing with the Latin alphabet for most European languages - i.e. most people on four continents write with a Latin alphabet.
You would be able to READ it. Maybe. If you're capable of making those sounds. But you wouldn't understand it.
Yeah, I doubt people will learn this mess of a script. And good luck trying to replace their perhaps 1000s or more years old script with this.
3:00 ugly as Ludwig van Beethoven? 😦
You just gotta embrace the chaotic nature of spoken and written language. If we fixed the problem, what would we make fun of?
Of how stupid it sounds, because it would reflected and therefore observable in spelling.
I had this idea myself a few weeks ago too, but my immediate reaction to it was 'and that sounds like it has a dozen huge holes in it' so this video is exactly what I was looking for!
Though, while it's still clearly a bad idea due to the other core problems, some of them could be solved by inventing a different IPA that's actually comprehensible as a script and not eurocentric. Doesn't fix things like the dialect problem and so it's still a bad idea, but it's interesting to imagine how much closer we could get to a decent option.
How is the dialect problem a problem when it's the whole point of the reform? To defeat centralising prescriptivism, and force people to be cognisant of their speech.
It's just a more complicated way of saying everyone else should speak MY language
There's also the problem that, as any undergrad phonetics student realizes... it's arbitrary. The class is dozens of hours of analyzing speech and by the end of it any two prospective linguists are STILL going to disagree on a paragraph's worth of spoken word in many places. This isn't to say we don't have tools for analysis - you'll see more of Praat than you ever imagined - but it's not like there's one single quantified threshold that separates an /ɑ/ from an /ɔ/. And that's if you can even hear the difference.
A lot of symbols sound very similar, if not identical, and other sounds seem to be unrepresented. Sometimes symbols get used to represent two different sounds too, like when the schwa (COMMA vowel) gets used to represent the sound at the end of the word ‘comma’ and the French word ‘je’. The main unrepresented sound is the way that most people in Birmingham (the one in England) say words like ‘cut’ or ‘put’, it’s not normally the same as the way someone from the North of England would say them, it’s more like how someone with a very broad West Country accent, or some people in some parts of the Southern states of the U.S, as well as many African-Americans, would say them.
I am down for that sequel
I never heard of this and at most reminds me of the time the French tried to make a Metric system for Time (it failed because they ended up keeping the other time measurement around causing confusion)
/ðɪs ˈvɪdioʊ meɪks mi sæd/
stop.
@@nequipaismaior3033 /ˈnəʊ̯/
jōr kɔment mēd mī sǣd😖
jes, ænd aj æm sæd tu. aj ɔlwejz sed ðæt ɪt kʊd bi mʌtʃ moɹ sɪmpəl ɪf wi kʊd now haw tu spel evɹiθɪŋ dajɹektli fɹʌm ðə pɹənʌnsiejʃən.
I am quickly falling in love with your vids
Imagine being unable to *read* Scottish English either.
That would be the point, yes.
There is no point, vocabulary would still be a problem. You would only be able to know how the work is spoken, not what it means.
I thought of this when I first took a linguistics class, but thanks to my studies of Japanese, I very quickly disabused myself of that notion. Japanese would be COMPLETELY INCOMPREHENSIBLE using the IPA due to its poor sound variety and pitch accent. There’s so many potential readings that are used by multiple, even dozens of different kanji - how the heck do I know if this ‘shou’ is the one in front of ‘elementary school’ or the one in front of ‘young man?’ As much as I love the idea of a true international language, the IPA is not it, not even close.
Dude you have such a great channel!
Weird I didn't know about it before
Why doesn't humanity make a new alphabet based on the list of phonemes the IPA has, but instead of weird latin alphabet stuff, we start from scratch creating a BRAND NEW ALPHABET that takes into account all possible combos and creates the most complete yet hypersimple list of graphemes. There, solved.
@J Boss i like diacritics because they narrow so much the sound, but on their current form they’re so ugly. Maybe they can be integrated in letters, like in my native tongue catalan there is a difference between c and ç
@J Boss Ugh, for my part i would hate to have to write cz, as in, two letters for one sound. Surely an alphabet can be built which has a grapheme for every phoneme. At least up to a point, because if you count every letter + diacritic combination in the IPA, you end up with hundreds of letters.
@J Boss do you speak multiple languages? When i was 5 i could speak spanish catalan french dutch and english. There is so much nuance that is completely lost with the current alphabet. NO WAY the portuguese and french ã is an. It sounds completely different from that.
@J Boss But then you're rendering useless the whole idea of having an actualized IPA alphabet. The spelling you propose favors some languages over others. It's not universal.
@J Boss you say nobody likes them but many people do like them. Are you from the usa btw?
This was very interesting, I think I just realized how cool I find linguistics lmao
aw that's so nice :D
It's a shame that this video had to be so mean-spirited to your classmate, because it was otherwise very good.
I don't even say Caribbean thensame every time. I switch it up between KARE-ah-Bee-an and KAH-rib-ee-an.
3:16
That Malayalam text means something closer to "-of dogs". "Dogs" alone would be "പട്ടികൾ". Should be. It's one way of saying it, in this case it's using the same root word. Just a plural instead of both a plural and a possessive form (is that the right term?)
I should probably mention I can't actually read or write that well in Malayalam since I didn't grow up there so I never picked up that skill. My parents however speak it natively and having to speak to them (and other family sometimes) in it is what kept me from forgetting what little of it i knew.
I really want to see that Controversial sequel
The problem with the International Phonetic Alphabet is the same problem with globalism, the fact that one-size-fits-all ideas look good on paper, and the people devising these schemes never consider all the thousands of myriads of details that billions of individuals across hundreds of different cultures and languages HAVE to agree upon just to make the scheme functional.
Yeah, you could tell them to first make two Redditors from different subreddits agree, and then come back with these clever ideas
@@aronasmundurjonasson3175 In the entire history of Italian cooking, no two people have ever agreed on pizza toppings. GARFIELD
@@ladymacbethofmtensk896 Why that? Doesn't everyone know that pineapple is the best topping on pizzas?
_run before I get mamma mia'd out of existence_
@@aronasmundurjonasson3175 What the devil is WRONG with you? I hope you tell a mob boss that. Then your filthy, disgusting mind will be sleeping with the anchovies!
I didnt give a shit about linguistics before watching this video. I still don’t but i was entertained. Good job.
Dude just tell us she rejected you, no need to dump on her for an idea that seems absolutely plausible at first glance
@@rama7267 You misspelled "plausible" Rama7, let me make a video about this run-of-the-mill understandable mistake of yours, dissing you for some reason
Both of you need to shut up
@@rama7267 Oh great, mix the mysoginstic sexism with anti-intersex sexism, wow
I'd say theres a need to dump on this idea
Its...bad
As clearly explained in the video
@@rama7267 gender ∉ R ; G ( set of all genders) ⊆ i
The cameos by a bottle of IPA are great because this is _precisely_ the kind of idea a linguist might have after one too many.
This is already a very short video, but I feel like if anyone ever presents the idea of everyone using the IPA for everything, it can get shot down in one second by pointing out that pronouncing is not the same as understanding or speaking a language. Like, it's just fundamentally a proposal that maybe, in an ideal world, might possibly fix one problem (reading a word and not being sure how to pronounce it) at the cost of creating probably literally a hundred thousand others in the realm of, you know, actually using a language in the ways that language use actually occurs.
Like, this isn't really mentioned in the video, but the distinctions between languages (and even dialects) carry a lot of information about what language is even being written, make it possible for people to parse it and its grammar, etc. No doubt some of that carries over to IPA use given the different "profiles" of each language in terms of sounds used, but you would lose huge amounts of contextual info. For each way it might make your life easier (no more memorizing thousands of kanji or hanzi, I guess!) it'd make your life more complicated in a hundred thousand ways. You can already get a near-perfect phonetic reading of Japanese if you just learn kana, read stuff with full furigana, and hide all the kanji (and somehow forget they're even there, as that would be an unfair grammatical hint). Good luck parsing anything more complicated than an illustrated children's book.
Tip: learning Hanzi is not as difficult as it seems. You can learn 500-800 in 3 months if you study correctly.
I really don't understand your point.
@@bacicinvatteneaca reading IPA means you can pronounce a word, not know what it means, and actually removes a lot of information about what language is being spoken, information which is necessary to understand what the sounds mean.
im currently learning japanese and i legit cant even imagine how itd work. like there are so many kanjis that are pronounced exactly the same but are just written differently
By relying on Japanese and Western words over ambiguous Chinese-derived ones. Actually, do you remember Aoharaido? Its name uses kun readings of ao and haru for seeshun. Such calquing may also be used.
@@АлександрАлександров-щ7к8т how limiting would it become for the language's vocab tho-
and sorry i dont understand your example. im assuming youre talking about ao haru ride? i dont watch that anime/read that manga so i have no idea what youre saying :"D
It's extremely easy, actually: the same way as in speech. If homophones aren't a problem in speech, they're not a problem in writing, and if they ARE a problem in speech, make them stop being homophones,
But in the spoken language it's suddenly ok?!
@@bacicinvatteneaca I sometimes have problems with homonymes even in English but what can I do
Anyone have any arguments against removing Q from English?
Also I am unreasonably mad at that dialect map for Ireland labeling everything as just "hiberno-english". Accents in Ireland are just as varied as in the UK, if not more. Each county has at least 1 dialect unique to itself and more populated ones have multiple
There aren't many arguments against changing the consonant letters in English spelling to more sensible ones. Vowels are all over the place in different dialects, but English speakers mostly agree on consonant sounds. Q and X especially don't need to exist or can be repurposed for other sounds.
I like the way Q/q looks on the page. Shallow argument but still an argument
Nah Q looks aesthetic asf 😎
Q is the ugliest Latin letter. I would not be sad to see it go, even though I would rue its absence in Scrabble.
That dialect map shows highly inconsistent levels of detail. If there is one Hiberno-English dialect group then there is one southern England English dialect group for instance.
Q only exists in English because it borrowed words from French and Latin, and now there's not much to be done
A bit like K and W being part of the French alphabet, even though you only find them in words of foreign origin
The only English spelling reform we need is an agreement to spell new loan words from other alphabets as plainly as possible.
There could be analytical ways of choosing sounds based on frequency and allophones. That could help as a basis for many reforms. Nice video, nice criticism
I could see myself making a similar argument while also not necessarily believing it. I think that's just how some people learn from other people. It's kind of like shooting beams of particles at each other in an accelerator and seeing what kind of particles are produced by the collisions.
1. Nobody wants to memorize hundreds of characters that represent a single sound
2. Having it be Latin based would disrespect thousands of years of history for non-Latin users.
3. You'll still be understood even if your pronunciation isn't 100% correct
1.4 billion Chinese and 400 million Arabs: We don’t fuck with that.
While I do think writing in a unified script is good in some ways, there’s a lot of problems such as those mentioned.
A unified script is very dumb. Way too many people already unnecessarily use latin.
@@iroga9764 a unified script can be helpful, especially in a hyper-united future. A script is as important to the language as the language is itself.
@@stegotyranno4206 A united future is a dark future. Death to the new world order and liberalism.
An example for how a unified script could work is Chinese. One written language represents several spoken languages.
Learning another alphabet isn't that hard. At least when they work similar to the one(s) you already know and don't use too many characters.
Also a people's script is a strong part of their identity, just like their language is. Any plan to replace that with one globally standardized system is gonna get A LOT of resistance all around the world!
I mean, pretty much every country in the world has people complaining about their regional dialect being killed by standardizing the language on a national level already.
Writing with IPA for some reason reminds me of one of those obscure West African languages and the heavily modified Latin alphabets they use.