These things are interesting, this sound is so common in Swedish so it feels like it would be common everywhere but I can't think of any English word where I use it. Like, chocolate has that sound in Swedish but not in English. It is kinda like when I learned that J and Y are pronounced differently in English, both are pronounced like Y in Swedish and the J sound doesn't exist.
I didn't even hear the difference between them when native English speakers talked for over a decade. It was only when I saw people making fun of a guy for not knowing the difference that I looked it up, and now the difference sounds so obvious!
YeGoblynQueenne 1597 days ago [-]
It's like how I grew up speaking French and never realised that it only has two genders. I'm Greek and Greek has three, including a neutral gender that French lacks. It seems I was always making the assumption in my head that, .e.g. "table" or "oiseau" were neutral ("une table" is female in French and "un oiseau" is male. In Greek they're both neutral).
I was a bit shocked when I realised, too. I mean, I studied French formally at some point and I had learned that it only has two genders but it never, you know, sunk in.
na85 1597 days ago [-]
In Canadian French there's an easy remedy if you can't recall what the gender of a noun is: just insert some profanity.
Québécois vulgarity tends to be inserted between the article and the verb, in a sort of semi-adjective form. So if you forgot if le porte or la porte was correct, you'd say l'ostie de porte which translates roughly to "the goddamn door" but lets you collapse le or la into l' and avoid the error altogether.
YeGoblynQueenne 1596 days ago [-]
Oh, but that's the funny thing. I don't struggle with the article. I use le/la or un/une etc correctly. It just seems I do it without ever thinking what it means.
So I'd map "la table" to "το τραπέζι" and "un oiseau" το "ένα πουλί" and never notice that the gender has changed from one language to the other.
Unless I think about it.
And the same with pronouns also.
squiggleblaz 1596 days ago [-]
My takeaway from your post is probably completely surprising to some, but:
A trapezium is a table? (Checks etymonline) And apparently indeed, trapezium comes from a Greek word that means "little table" from "trapeza" meaning table. The things you learn.
(For those who can't read Greek, the Greek writing in YeGoblynQueenne's post are "to trapezi" and "ena pouli". I can read Greek since my mother studied Christian theology and had the alphabet in prominent display to help her learn it, but I can't understand a word of it.)
YeGoblynQueenne 1596 days ago [-]
Hah. Yes, it's the same word. Or same word-root anyway. Interestingly, a "table" as in a matrix, is called a "pinakas" (πίνακας). Which translates as "board" to English (as in whiteboard, cheeseboard etc).
Have you ever visited Greece? It would be a strange experience, being able to read all the signs but not knowing what they mean. I know because I was in Bulgaria for a conference this summer and I could read everything... but understand not a word :)
That's because it's all in a Cyrillic alphabet which is very similar to the Greek alphabet (it's partly derived from it). Come to think of it, so is the Latin alphabet, but I'm used to seeing Latin signs everywhere, including in Greece.
nicoburns 1596 days ago [-]
As a native english speaker who also speaks some french and spanish, I had the opposite experience:
I had trouble deciphering the alphabet at speed, but once I had I could often understand the word based on it's similarity to spanish (e.g. Σάββατο / Sábado) or english (έξοδος / exit).
close04 1596 days ago [-]
> I can read Greek since my mother studied Christian theology
I can read (a reasonable amount of) Greek after studying a lot of physics/math :D. But I only understand the words that trickled into the other languages I speak or can recognize a root for.
YeGoblynQueenne 1596 days ago [-]
Yes, that's another interesting experience- how other people pronounce Greek letters and words that they've learned from science textbooks. Sometimes I forget myself and say something in a Greek way and people just stare at me :)
E.g. I was working on a paper with my thesis advisor and I used the plural of "lemma", "lemmata" (λήμμα, λήμματα). It's actually in the English dictionary, but he advised me to say "lemmas" instead because people reading the paper may not be native English speakers and it would just make the text harder to read. I always want to sneak in fancy words in my papers, but it's a bit tricky to do it right..
jhbadger 1596 days ago [-]
And a trapezoid is called that because it is like a table, with a flat part on top.
tasogare 1596 days ago [-]
Yes, native speakers don’t thinks about grammatical gender of words, it’s just coming up naturally. The worrying trend however, is the politicization of language, based on a Saphir-Whorf-like hypothesis (and I would say on Newspeak as well) which lead to the horror that is called écriture inclusive. The Académie condamned this new orthography but it’s still wildly in use in administrative documents.
squiggleblaz 1596 days ago [-]
"Condemned". The distinction between condemn and damn seems to go back to Latin, if not further.
I'm not sure it should matter what one particular authority decides, in terms of language use. People, including people writing administrative documents should be free to write in the manner that feels most comfortable to them. If this creates a political problem, the political solution is the ballot box and altering the education or hiring policies.
Customary criticism is fine, but it makes no sense to say "you shouldn't say foo because some organisation that shouldn't have authority says you shouldn't say foo"; instead, you should have some direct reason why foo is problematic.
Your criticism shouldn't be circular either, for instance "you shouldn't use 'singular they' because 'they' is plural" is completely, aside from being historically wrong, ineffective as criticism, since it is merely because of the incorrect description of the word as "plural" by some authority that you believe it to be plural, whereas most people use it regularly to refer to a non-specific person.
Bayart 1596 days ago [-]
>People, including people writing administrative documents should be free to write in the manner that feels most comfortable to them
They should also write in a way that conveyed information effectively. The so-called inclusive writing is extremely disruptive to the flow of reading and makes job offers, administrative documents and so on a pain to navigate (all for the sake of political positions very few people actually hold), defeating the point of language. I'm sure fellow French speakers will know what I'm talking about.
lordnacho 1597 days ago [-]
There's quite a lot of language stuff you don't realise if you grew up with it. Sometimes you learn a rule without knowing what on earth the rule actually it. I had this come up once in a Danish class. The choice was between "hans" and "sin" which is a somewhat subtle thing. The thing is I never got any of the answers wrong, but I still don't really understand the difference.
On the learning side, you also often get told rules that don't entirely make sense. For instance I'm told in a Mandarin class that there's a difference between 不 and 没 which is whether it's subjective or objective that something isn't true. But that's neither clear when saying something, nor clear in terms of examples I've been given. What's really weird is I speak another Chinese dialect that also has this distinction, and I also can't tell you the rule there.
yorwba 1597 days ago [-]
> I'm told in a Mandarin class that there's a difference between 不 and 没 which is whether it's subjective or objective that something isn't true.
I don't think that rule holds. 没 is usually used to negate either 有 "have" or something in the past. E.g. 我没告诉你。 "I didn't tell you." 不 is used for everything else, i.e. 我不告诉你。 "I'm not telling you." (The interesting thing is that "have" also has a past aspect in English, e.g. "I have told you." But Mandarin doesn't use 有 as a general past marker, only the negation 没)
> I speak another Chinese dialect that also has this distinction, and I also can't tell you the rule there.
Which one? Is the rule the same as in Mandarin, i.e. can you produce correct Mandarin sentences despite not knowing the rule, relying just on intuition? I think Cantonese uses 冇 to negate 有.
thaumasiotes 1597 days ago [-]
> 没 is usually used to negate either 有 "have" or something in the past.
More technically, 没 is the negative marker that applies to the verb 有, and 不 is the negative marker that applies to everything else. It isn't the 没 that marks a negative in the past as being past -- it is the (omitted) 有. That 有 is commonly not omitted in negative past sentences -- and it's also possible to use 有 to mark past time in positive-polarity sentences, though this is much rarer than the same marking in negative sentences.
There's a straightforward information-theoretic reason why 有 can be so commonly omitted -- it is the only thing that can ever follow 没, so actually realizing it doesn't add much. Everyone knows it's there.
thaumasiotes 1596 days ago [-]
Hm, I responded too quickly the first time. Some comments on the rest of your comment:
> E.g. 我没告诉你。 "I didn't tell you."
This is the same sentence as 我没有告诉你, also fully idiomatic. The past sense is coming from the 有.
> But Mandarin doesn't use 有 as a general past marker, only the negation 没
This is plainly incorrect. 有 is used as a general past marker. It just isn't obligatory in positive-polarity sentences. Here are some things Chinese people have said to me:
你在中国 有看见过 美国没有的水果吗？("Have you seen any fruits in China that America doesn't have?")
我有吗 ("Did I?")
yorwba 1596 days ago [-]
What I meant by "not a general past marker" is that you can't in general use positive 有 where you can use negative 没（有） to talk about the past. For example, you wouldn't say 我有看这本书。 to express you've read that book, but instead 我看了这本书。 Even in constructions like 有……过 the 有 is optional, so I'd say the past aspect is carried by 过.
On the other hand, maybe that is a shortening of an originally obligatory two-part expression similar to Jespersen's cycle, famously observed in the appearance of "pas" and ongoing disappearance of "ne" in French "ne ... pas". https://en.wikipedia.org/wiki/Jespersen%27s_Cycle
As you can see, there's more than one way to negate something. Well it's hard for me to know if the rule is the same, given that I don't really know what the rule is in either! As per my previous comment I can tell you what sounds right, but I can't tell you why.
yorwba 1597 days ago [-]
The explanation for Cantonese in that Wikipedia article could be used just as well for Mandarin by simply replacing 唔→不, 冇→没, 唔好→不要, 咪→别,咗→了 ... so the rules are pretty much the same, they're just instantiated with different vocabulary.
1596 days ago [-]
ksaj 1597 days ago [-]
Swedish has masculine and neuter, but not feminine, interestingly enough.
blompa 1596 days ago [-]
Masculine and feminine have merged into 'common'/'utrum', used mostly for animate objects as pointed out by a sibling comment, although not all animate objects (e.g. barn, lejon).
Neuter has stayed as is.
There are, however, remnants of the masculine/feminine distinction left: e.g. klockan is ofter referred to as feminine (hon), and the "-e" in "storebror" (as opposed to "-a" which is used for utrum today and feminine in the past).
einr 1596 days ago [-]
There are, however, remnants of the masculine/feminine distinction left: e.g. klockan is ofter referred to as feminine (hon), and the "-e" in "storebror" (as opposed to "-a" which is used for utrum today and feminine in the past).
The -e/-a ending is still in common use not only in constructions like "storebror" but in general in adjectives pointing to a female or male person ("den fule mannen", "den fula kvinnan") although this usage is falling out of favour a bit ("den fula mannen" would not generally sound wrong to most modern speakers)
Another interesting remnant of old Norse that still lingers in modern Swedish is the accusative case ending -s in expressions like "till fjälls", "till skogs", "till sjöss", "till sängs" etc.
DFHippie 1597 days ago [-]
In which case it's probably better to say it has an animate/inanimate distinction. I was told by someone interested in historical linguistics that Indo-European first developed an animate/inanimate noun class system and then the animate class developed a feminine subset.
einr 1596 days ago [-]
In which case it's probably better to say it has an animate/inanimate distinction.
No, this is not accurate, neither historically nor according to modern usage. Animate objects tend towards utrum but plenty of inanimate objects are utrum too and there are no hard rules or mnemonics you can use to divine which is which.
It's "ett tåg" ("a train") but "en bil" ("a car"), "ett äpple" ("an apple") but "en banan" ("a banana"), "ett barn" (a child) but "en vuxen" (an adult). This is entirely arbitrary and non-obvious to non-native speakers and if you're learning Swedish, you just straight up have to learn which is which for every word.
henrikschroder 1596 days ago [-]
The biggest mindfuck about utrum/neutrum is that it exists in Danish and Norwegian as well, but the categorization is not 100% the same, some nouns are utrum in one language and neutrum in the other. So there's different arbitrary rules, and if you go from one of the Scandinavian languages to the other, you just have to learn the ones that aren't the same.
For example, "a friendship" is "et venskab" in Danish, but "en vänskap" in Swedish.
1596 days ago [-]
hashmush 1597 days ago [-]
It's the same with <ch> and <sh> for me. It took way too long for me to realize that they are different sounds in English ((Standard) Swedish does not have any affricates), or that <z> is supposed to sound different from <s> (/z/ does not exist in Swedish).
If anyone pronounces <ch> as <sh>, <z> as <s> and <j> as <y> they're probably Swedish.
blompa 1596 days ago [-]
I usually use the analogy that a Swede pronouncing "cheap" as "sheep" is equivalent to English speakers pronouncing "skjuta" as "tjuta".
As a native English speaker, it took me a while to hear the difference between Swedish "tj" and "sj" (at least the way it's pronounced in Northern Swedish, like "schhh").
colde 1596 days ago [-]
Yeah, the <ch> and <sh> being the same is usually my Swedish detector. It is surprisingly common amongst swedes speaking english, and not something that danes or norwegians do.
henrikschroder 1596 days ago [-]
> (/z/ does not exist in Swedish)
It does, but only in the onomatopoetic word "bzzz", as in the sound of a flying insect. And if you were to drop that sounding s, if you said that a fly went "bsss" around you, people would be weirded out, thinking the fly pissed on you or something. So pretty much every Swede can make the sound [z], and can hear the difference between [z] and [s], but only for this single word. It's so weird!
tilt_error 1597 days ago [-]
I moved to Sweden from Norway at a time when writing a check really was a thing :) It baffled me that swedes would pronounce 'check' with an initial 'k'-sound (as in "käk" or "käck"), which is not at all like the pronunciation in Norway (which is similar to the English pronunciation). What a strange country :-)
henrikschroder 1596 days ago [-]
Similarly, Danish has kept the 'dj'-sound in English loan words like juice, where Swedish dropped the initial fricative. That tripped me up a couple of times while in Denmark...
simongray 1596 days ago [-]
Danish generally just adopts loanwords wholesale, never bothering to adapt the pronunciation. We have lots of words from English, French, and German that are spelled and pronounced basically like in those languages.
Now as for Swedish, what trips me up is that I've met so many Swedes that (otherwise) speak great English, but seem to pronounce J sounds in a distinctly Swedish way when speaking English. I've had fluent English-speaking Swedes in a university setting asking if I want to hear a "yoke" (joke) and or maybe telling me about their "yeans" (jeans). It's so weird. It's a bit like how many French people seem unable to pronounce an H when speaking English, but Swedes are damn good at English, unlike the average French person.
henrikschroder 1596 days ago [-]
Yeah, the "blindness" to sounds that don't exist in your native language is funny, and usually the best way to figure out where a non-native speaker is from. If you can't even hear the difference between "sheep" and "cheap", or "joke" and "yoke", you have no way of pronouncing it right in the first place!
Likewise, I can identify English with a Danish accent pretty quickly, but I can't for the love of me speak it, because I'm not good enough at Danish. But a dead giveaway is that Danish realizes 'R' as [ʁ], and many Danes pronounce it the same way when speaking English. :-P And there's something with the vowels that's just weird that Danish people do... I can't put my finger on it.
(I have the opposite problem, I can't make that 'R' sound at all, which is why my Danish sounds pretty funny to native Danes.)
simongray 1596 days ago [-]
The Danish accent is definitely instantly recognisable, but I also can't put my finger on what it is. It's definitely something about the vowels. Danish is a very vowel-heavy language, so I'm sure some of that bleeds over into English.
jacobush 1596 days ago [-]
I'm old enough that I have used a check a couple of times, but probably fewer than 10. I remember not being certain how to pronounce it "xeck" (käck) or "sheck" (sjeck) or even "tcheck", to improvise an English sort-of-phonetic spelling.
marcus_holmes 1597 days ago [-]
I'm learning German, and only found out last weekend that the reason Germans pronounce W's as V is because they don't hear the difference. I was trying to get my (German) gf to explain why some words start with a "V" sound and some with a "W" sound, and she couldn't tell the difference. It's all the same sound to them. I now know how to pronounce German W's properly (the sound exactly halfway between W and V). Of course, they pronounce V like F, so I still have to get used to that ;)
claudius 1596 days ago [-]
Well, the English "V" is nearly identical to German "W"; the German "V" is nearly identical to English and German "F" and the English "W" doesn't quite exist in German. Someone had to point out to me as well that I was mispronouncing "water" as "vater" instead of "uuater" about ten years after I was somewhat fluent in the language. I never heard the difference before and native English speakers typically don't go around correcting the pronunciation of strangers….
henrikschroder 1596 days ago [-]
It's probably English that's the outlier here, though. 'W' is a double-V in other European languages, only in English is it called a double-U, and treated like a vowel.
estomagordo 1596 days ago [-]
Treated like a vowel now?
henrikschroder 1596 days ago [-]
Other way around then? Is a consonant, but a defacto vowel? Whatever, English is weird.
OJFord 1596 days ago [-]
Why do you say it's a de facto vowel?
It isn't used like a vowel. There's no word 'XwY' for any string of letters X,Y - and you can't prounounce 'xwy' (for any single letters x,y) without adding vowel sound.
checkyoursudo 1596 days ago [-]
I suspect in the sense that 'w' acts like a long vowel or a diphthong in many cases in English when it is used in the middle or at the of a word.
E.g., (spelling vs pronunciation) 'town' vs 'toun', 'own' vs 'oan', 'bow' vs 'bo' or 'bau', etc.
It definitely seems like the most vowel-like consonant after 'y'.
> There's no word 'XwY' for any string of letters X,Y
Awe, ewe, owe, awl, owl, own, two...
But my point was that in all other European languages, the letter 'W' is clearly a consonant, usually realized as [v] or [ʋ], and seen as a variant of 'V'.
English is the odd one out, English sees it as a variant of 'U', and it's not clearly a consonant, it's a semi-vowel as you pointed out further down.
OJFord 1596 days ago [-]
Right, I meant consonants, not 'any letter'. (You could do examples like that for any consonant in place of 'w'.)
OJFord 1596 days ago [-]
Hindi (which I'm learning) is similar - व (usually transliterated 'va') is somewhere between an English 'w' & 'v' - not as much 'lip vibration' as the latter but more than the former.
It varies across the v <-> w spectrum depending on region, and also some words seem mostly nearer one or the other (but it's the same character, there isn't a second).
You sometimes hear that in native speaker's English, but more 'v's sounding nearer a 'w' than the other way around (as in native German speakers).
Fascinating how growing up with different languages teach us different sounds, and those we didn't use can be so difficult to produce, even if we can hear and understand the difference.
I struggle with two things in Hindi: aspiration (transliterated as 'ta' vs 'tha' for example, but that's not an English 'th', it's 'ta' with air forced out) which I usually either don't pronounce clearly, or it's way over-exaggerated; and retroflex consonants, which I can hear but often struggle to pronounce correctly quickly enough to flow with the rest of the word. And particularly 'd.a' which when correctly pronounced contains (to my ear) some 'r'.
marcus_holmes 1596 days ago [-]
The German "ch" sound is the thing I struggle with. Not a hard "ch" as in "church", but a soft sound like "shhh" (but in the back of the throat). Pretty hard for an English speaker to do, especially straight after/before rolling an "R"
1596 days ago [-]
lordnacho 1597 days ago [-]
The bible story about Shibboleths is quite an interesting example of this, and also where the term comes from.
ahh, killing 42,000 people because of a mispronunciation. Gotta love that pro-life attitude that religion has ;)
77pt77 1596 days ago [-]
I had completely forgotten about this story.
It was nice to read it once again...
ljm 1597 days ago [-]
I’m not a linguistic but there seems to be an amazing ambiguity around I, E and Y.
The English E covers е, и, (roughly) ы in Russian. The English y covers the same but also partially у and ё.
When you get to dialectic English, then all of those vowels have different vocalisations or they otherwise don’t exist. And in all cases Y is a vowel and the same as J.
I love languages and linguistics and this kind of thing.
What’s more, take this to Spain and South America and in Castellano your ‘ll’ is more like an English ‘y’, but in Latin American it’s more of a ‘dj’ or an English ‘j’.
jdmichal 1597 days ago [-]
What you're seeing is the fact that /j/ (English Y as in <you>) is a semivowel, which is paired with /i/ (English long-E as in <feet>). This means that the sounds are almost exactly the same -- the consonant version tends to be a bit more restricted.
EDIT: Also, the Spanish <ll> as /ʒ/ (S in English treasure) is not nearly as extent in South America as you might think. It's still mostly limited to Argentina and Uruguay:
EDIT EDIT: I misread the symbols a bit, and in other areas of South and Central America it's /ʝ/ instead of /ʒ/, which is not a very big difference at all. (Palatal vs postalveolar... Basically neighbors.)
DonaldPShimoda 1597 days ago [-]
> Also, the Spanish <ll> as /ʒ/ (S in English treasure) is not nearly as extent in South America as you might think. It's still mostly limited to Argentina and Uruguay:
In Puerto Rican Spanish (at least, my grandmother's family's accent), the <ll> is more of a /dʒj/ sort of thing. "llama" comes out to "djyamma", with the "dj" softened significantly. Not sure how to express that more precisely in IPA, though.
jdmichal 1597 days ago [-]
Maybe /ʝ/? That's the palatal version of /ʒ/, which explains your desire to add a /j/ in there.
The Wikipedia I linked is about the merging of <ll> and <y> to /ʝ/ if you want to read more. Interestingly, I don't hear this a lot in the northern Mexican Spanish I hear, even though all of Central America is marked as merged.
I didn't read it correctly the first time, and I was wrong to call it out in the first place. The difference between /ʝ/ and /ʒ/ is not so big for me to have said anything.
ljm 1597 days ago [-]
That was precisely what I was trying to allude to :) ‘ll’ is a soft ‘dj’ in Columbia also.
heavenlyblue 1597 days ago [-]
I find it fascinating how Greek ф became th in English and then was again reinvented as F. But English still pronounce all of the Greek ф’s as th.
jdmichal 1597 days ago [-]
I'm unsure what you mean by this... English <th> is inherited from Proto-Germanic, not Greek.
Also, did you mean Greek θ (theta)? θ started as aspirated /tʰ/, and it's a pretty common sound change for aspiration to become a fricative, in this case /θ/. φ also made a similar journey from /pʰ/ to /ɸ/ to /f/.
DonaldPShimoda 1597 days ago [-]
Phi is, I believe, pronounced as a voiceless bilabial fricative, which does not exist in English. Instead, we have the voiceless labiodental fricative /f/. If I remember my phonology correctly, Japanese uses the bilabial sound but in romanized text it's spelled with an f, which causes English speakers to use the labiodental sound.
Not sure why phi would be mapped to the voiceless interdental fricative ("th") though. I thought that came from theta. But I have not studied Greek, so I really don't know haha.
jdmichal 1597 days ago [-]
Modern Greek uses /f/ also. It was only /ɸ/ for a brief transition period. Spanish, however, uses the voiced version /β/ as an alternation for /b/, hence <Cuba> /kuβa/. (This is sometimes described as a mumble, since it's not a very common sound in Indo-European languages and people don't know how to interpret it.)
FWIW, some (most? all? the one's I've heard, at any rate) Spanish dialects pronounce the letter 'f' as a voiceless bilabial fricative (I almost wrote 'phricative'). E.g. fotografia (sorry about the missing accent).
The letters 'b' and the 'v' are pronounced in Spanish as voiced bilabial fricatives after a vowel, and as stops after a nasal consonant (which in that position will be pronounced as [m]), and generally in utterance-initial position.
Jverse 1596 days ago [-]
One thing I've noticed a lot of Swedes don't think about is that 'y' is pronounced the same as 'i' in English, as in the word 'gym'. They instead use the Swedish 'y' sound (like the German 'ü').
boomlinde 1596 days ago [-]
Another Swede-ism I've noticed is the lack of distinction between e.g. "chip" and "ship" or "choose" and "shoes". Probably because so many similar spellings are bunched up under /ɧ/
balboah 1596 days ago [-]
I’m Swedish, didn’t quite get the part about J and Y being pronounced the same? Sounds different to me
einr 1596 days ago [-]
I think he's talking about Y when used as a consonant at the start of English words. Y exists in Swedish as a distinct vowel sound but is not used as a consonant like in English.
Swedes tend to pronounce English "yes" and "yoke" as well as "juice" and "joke" all with the same Swedish J sound.
bklyn11201 1597 days ago [-]
Google Translate can provide a basic sample of the sj sound. Ask for a Swedish translation of "seventy-seven hospitals near the lake" (sjuttiosju sjukhus nära sjön):
I lived in Sweden for a little over a year. This sound wasn't the hardest to figure out, but it's one of the harder ones to remember to use since the natural instinct is to make the "syuh" sound.
To make the sound with your mouth, make a really guttural 'H' sound like a golf swing.
russellbeattie 1597 days ago [-]
I have to assume IPA is basically gibberish to 99.9% of the world, if not more. Is there a digitally synthesized version? Whenever I'm reading Wikipedia/Wiktionary and see the phonetic version of a word, I wish I could click on it and hear the sounds (without having to have someone upload a sound file). The pronunciation key (https://en.m.wikipedia.org/wiki/Help:IPA/English) isn't particularly helpful.
Yes, reading IPA is very hard, but it's hard for a good reason. It's the best way we have of describing human-made sounds unambiguously, because our regular alphabet doesn't work cross-language.
I agree it would be cool if there was a good IPA speech synthesizer, that would certainly help! There's some discussion on the issues and problems with it here: https://news.ycombinator.com/item?id=13514258
usr1106 1596 days ago [-]
I studied English and French 2 decades before affordable computers and speech synthesizers. Reading IPA for a single language is not all hard. You just get used to the symbols needed in that language.
If you take a random language and believe that you can pronounce it correctly by reading IPA, well that won't work. Or at least be very hard.
DagAgren 1596 days ago [-]
Unambiguously, except for the symbol this article is about, which is ambiguous.
VectorLock 1597 days ago [-]
If the annual Wikipedia top-banner donation ask simply said "Donate $5 and we'll make the IPA keys produce an actual sound" I'd donate.
crazygringo 1597 days ago [-]
Nope, just gibberish to 99.9% of native speakers. BUT...
Because English is spelled non-phonetically, English learners across the world generally learn IPA together with English, to learn pronunciation.
Granted it's not the entire set of IPA sounds, and some students are better than others... but seeing as there many more speakers of English as a second language than native speakers (according to some sources, though this can be disputed), there must be at least hundreds of millions of people with a rudimentary knowledge of IPA, enough to sound out words in English.
apersom 1596 days ago [-]
The IPA doesn't actually describe a sound. It describes what you do with your mouth to produce that sound. Where you place your tongue, whether your lips touch or not etc.
The actual sound that gets produced can vary by context, what sound you made before that and what will come after etc. (And who is saying it off course).
It seems limited to IPA symbols that also appear in English. (Understandable, given that it uses an English TTS synthesizer under the hood.) It didn't produce any sound for ʈ͡ʂʐ̩
funnybeam 1596 days ago [-]
Interestingly all four of those pronunciations for Greenwich are valid.
As a local I hear (and use) all of them interchangeably and am only vaguely aware of the difference if I consciously think about it
girzel 1597 days ago [-]
Living in China, I was curious to discover that many people there were actually pretty familiar with IPA, I assume it's taught as part of advanced language learning, to cope with all the very different sounds in other languages. On several occasions people expressed great puzzlement that I had no idea how most IPA symbols were pronounced.
yorwba 1597 days ago [-]
I'm pretty sure IPA is the de facto standard for conveying English pronunciation in textbooks around the world, not just in China.
What do foreign language textbooks in English-speaking countries use to teach pronunciation?
Geimfari 1597 days ago [-]
Almost all textbooks and dictionaries use their own transcription key, I haven't come across plain old IPA in language learning textbooks, only in linguistics textbooks.
usr1106 1596 days ago [-]
IPA was in use in every German high school book for English or French in the 1980s. I have no reason to believe it has changed, but I haven't used any since then.
Using own systems I have seen in very naive tourist guides and in text books by English speaking publishers.
brummm 1597 days ago [-]
I mean, one could say the same about musical notation. It might not be 99.9%, but probably close to 75%.
Tomte 1596 days ago [-]
The bane of my learning Swedish.
I do understand this sound on a cognitive-analytic level (some phonetics courses at university help when learning sounds). I can pronounce it, at least passably.
There are big regional variations to pronouncing sj and so on. Fine, I can live with that, just pick one and use it consistently, preferably that of your teacher.
But when it comes to the sound the article is about, I hear something else.
At some point I cut together many instances of words with that sound from Swedish pop songs and told my teacher which sound (IPA) I clearly hear in all those instances. She told me flatly that I'm wrong. Those singers clearly sing another IPA sound.
It must be similar to when Russians try to show me the difference between hard and soft consonants. They tell me they really over-pronunciate now and they sound starkly different, I'm not even sure I hear a difference.
bjoli 1596 days ago [-]
As a native swedish speaker I like to tease people with word stress (accute vs grave accent). Words like tomten, regel and slutet mean different things depending on accent, which is very confusing for people who don't have swedish as a native tongue.
Even English has words that are pronounced differently depending on context.
For example "I read to my son" could be pronounced two different ways depending on tense. "I read", vs. "I red".
There are other examples too, off the top of my head "I polish my shoes", vs. "I have a polish friend.". Or "the bow of a boat", vs. "at the end of the play the actors take a bow".
bjoli 1596 days ago [-]
I understand what you mean, but those are not grave or accute accent differences. In swedish T<o>mten and T<o>mt<e>n (for lack of a better way to show stress off the top of my head) the o and e both sound the same, but have intonation differences.
henrikschroder 1596 days ago [-]
The vowel sound is different between past tense and present tense "read", even though they're spelled the same.
But that's not what's going on in the Swedish example above. The vowels are exactly the same between "anden" (the spirit) and "anden" (the duck), but the pitch accent is different, the "song" of the words is different.
(And if you can't hear the difference in that video, don't worry too much, it's Swedish that's weird, not you. :) )
goodcanadian 1596 days ago [-]
Polish is fine, but Polish should be capitalized. ;-)
I am wondering about your accent, though, because, to me, "bow" is pronounced the same in both of those examples. Robin hood shooting an arrow from his bow, however, has a different pronunciation which is the same as the bow tied on my present.
bjoli 1596 days ago [-]
He/she might mean the bow of a ship? If I am not completely wrong about my English pronunciation, that is.
goodcanadian 1596 days ago [-]
Yes, the bow of a ship is pronounced the same as taking a bow. At least, it is in my accent.
I'm not a linguist but it seems that most phonemes exist as areas (or multiple discontinuous areas as seen in TFA) in phonetic space while IPA attempts to describe only points in phonetic space[0]. Wouldn't it be better to instead just define X = {(F1min,F2min,...),(F1max,F2max,...)} for every dialect instead of making a universal X = (F1,F2,...) and then using it describe any area in phonetic space that happens to contain it?
This is effectively the difference between phonetics (the sounds we make) and phonology (how we decide which sounds to make)
Qwertystop 1597 days ago [-]
IPA seems to describe areas as far as I can tell; each consonant is how-do-you-make-the-sound and where-is-the-sound-made, not specific frequencies, charted on a grid. The vowel chart is "where is your tongue in your mouth".
knolax 1597 days ago [-]
But it seems that the "where the sound is made" aspect is being described as discrete whereas the positioning approaches more of a spectrum. Even if it were discrete, the fact that most phonemes can be produced with multiple positions would require at least a list of points.
DonaldPShimoda 1597 days ago [-]
IPA is about phones, not phonemes. Phones are described precisely and are (mostly [1]) constant across all languages; phonemes are broader and are language-specific.
Multiple phones can belong to a single phoneme within a given language. For example, English groups the aspirated stops with their non-aspirated counterparts (e.g., [pʰ] and [p] both belong to /p/ because they are non-contrastive, despite being different sounds phonetically).
This is why it is the International Phonetic Alphabet, and not the International Phonemic Alphabet.
[1] Worth noting that there can be some variance among different speakers when it comes to the articulation of specific phones, but the IPA essentially is the result of determining whether languages draw any meaningful distinctions among these. If there are separate symbols in IPA, the phones are noticeably distinct. It is very uncommon for multiple sounds to get mapped to a single phone in IPA by rule, and this usually happens based on some dispute among linguists. A good example might be the "tensed" consonants of Korean, which have their own diacritic applied but are not fully understood (meaning linguists cannot precisely identify what, if anything, separates them from their un-tensed counterparts).
knolax 1597 days ago [-]
Ok so does that mean phones are atomic and also consistent across languages? If so that makes a lot more sense. Thanks for the clarification!
henrikschroder 1596 days ago [-]
Phones are pretty consistent, but I think the biggest piece of the puzzle is that different languages don't care about all the differences. A phone exists in IPA if a language cares about the difference.
For example, if you pronounce "sju sjösjuka sjömän" with [ʃ] instead of [ɧ], I'll understand you just fine, but I'll instantly know that you aren't a native Swedish speaker. (Or, you are, and you're speaking a Swedish dialect that has replaced [ɧ] with [ʃ], because of course that's also a thing)
But if you mis-pronounce "skön" (nice) as [ʃøːn] instead of [ɧøːn], you're getting awfully close to [ɕøːn], which is how you pronounce "kön" in Swedish, which means gender. So as a Swedish speaker, the phoneme has changed, I no longer know if you're saying "skön" or "kön", or maybe "schön" in German? Swedish "cares" about this difference. But you, as an English speaker, might not be able to hear the difference.
And if there was a hypothetical language that had even further subdivisions of these sounds, speakers of that language would be upset that neither you nor me could tell the difference between two sounds that speakers of that language care about.
So the mapping of phones to phonemes is highly language (and dialect) dependent, and you should think about it in terms of continuous ranges or tolerances, instead of a discrete 1:n mapping.
DagAgren 1596 days ago [-]
Except for the symbol the article is about.
tom_mellior 1597 days ago [-]
Not to poop on this article, but isn't the point of IPA that every phoneme has "its own dedicated IPA symbol"?
crazygringo 1597 days ago [-]
Actually no -- otherwise there would be thousands of them, because languages often have sounds that are very close to a sound in another language but not identical.
The point of IPA is to provide a phonetic alphabet that can provide enough symbols, each covering a range of similar sounds across languages, such that for each language, every sound within that language has a unique symbol to differentiate it from other sounds within that language.
E.g. "bête" in French and "bet" in English are identical in IPA: /bɛt/. But all three sounds are definitely slightly different, which is how you can tell whether the speaker is, say, American or French.
The process of coming up with IPA and deciding what sounds in different languages deserve the same symbol or not involved just as much art as science, since there are tons of gray areas, especially when dialects are involved.
hoseja 1596 days ago [-]
Huh. I thought IPA was universal. Bummer.
vilhelm_s 1597 days ago [-]
No, it's not a phonemic alphabet. The idea is that it should have a symbol for each phone, i.e. each sound you can make with your mouth. Different languages will then cluster those phones into different equivalence classes (phonemes), but the standard practice is to pick a representative sound from the equivalence class, and use that as the IPA letter for the phoneme. Allocating a symbol to a phoneme is is odd.
zeckalpha 1597 days ago [-]
Yes, but most phonemes are in multiple languages.
gpvos 1596 days ago [-]
The link to Caramelldansen inside the article does not actually contain the swedish pronunciation, because it has some English mishmash text there. Here is the Swedish original, with the linked text at 0:58: https://youtu.be/PDJLvF1dUek
lukego 1596 days ago [-]
This is the same sound as in "cool whip" a la Family Guy, right? Like the way a southern gentleman would pronounce "wh."
Wait, what? Would you pronounce those two words with a different sound? I can't think of any Swedish dialect where that is true?
gfaure 1596 days ago [-]
The whole point of this article is that sj-ljudet has a bunch of different realisations, some closer to the initial in "shell" and some closer to the initial in "whip".
croon 1596 days ago [-]
Also Swedish, and disagree. GP is close(r).
Are you perhaps from Norrland? Then I see what you're saying.
scandinavegan 1596 days ago [-]
Pronouncing the sound is one thing, spelling it is another.
Here's a (probably non-exhaustive) list of 65 different variants. The article in Swedish, but it should be possible to understand the list:
Edit: I don't know of any rules for when to use which spelling. When I was a kid, I just learned how the different words were spelled individually, even though the sound is the same. Same thing when I'm teaching my kids: "Sju (seven) is spelled like this, schack (chess) is spelled like this, charmig (charming) is spelled like this", and so on.
boomlinde 1596 days ago [-]
IMO the list confuses a number of different dialectical oddities and makes it seem more of a labyrinth than it really is.
It makes some rather sveamål-centric assumptions, even then for some subset of sveamål, which is rather notorious for joining long stretches of consonants ending with an "sj" phoneme into just "sj". For example any of the variants that start in "r" only happen in dialects where the "r" is barely pronounced. In most of Sweden, it is distinct; guttural or rolling. Others only occur when consonants that are distinct in other dialects are dropped, and definitely not only the ones marked as "sloppy pronunciation".
Others occur only when crossing words in compound words with distinct "sj" sounds (again, in some dialects), where pronouncing them as a single phoneme again may happen in some sveamål. Some of them seem really forced.
henrikschroder 1596 days ago [-]
Yeah, the list has examples where someone in some dialect might pronounce a word with the sound.
I would say that a word like "duschschampo" has two different sounds for me. I'm from Stockholm so I would do the "schsch" part like /ʃ ɧ/, I have a friend from Kalmar who would say it like /ɧ ɧ/, and probably, somewhere, someone would merge them into a single /ɧ/, but I have no idea where you'd find that person.
neilwilson 1596 days ago [-]
It's not uncommon for languages to have a sound with a dedicate phoneme.
There are still some slots free in the sounds matrix that no language as yet uses.
No. Generally saying I'd say it is pronounced more to the front of the mouth, without involving the throat.
xxpor 1597 days ago [-]
That's just regular [x]
sarbaz 1597 days ago [-]
Only in some dialects.
jefftk 1596 days ago [-]
This reminds me of /ꜧ/ which is used as a hypothetical phoneme that sounds [h] at the beginning of a syllable and [ŋ] (as in "ing") at the end of one. Since a given context can only ever have either [h] or [ŋ], there are interesting philosophy of science questions over why we shouldn't analyze it as /ꜧ/. (Though I don't think anyone actually thinks /ꜧ/ is right.)
taejo 1596 days ago [-]
For some people "being it" [biŋɪt] and "be hit" [bihɪt] are minimal pairs. I can't think of any vowels that can both come at the ends of words and before [ŋ] in my own speech, though.
jefftk 1596 days ago [-]
English has juncture, so "the grey tape" and "the great ape" are also distinct. We'd write your two as [biŋ+ɪt] and [bi+hɪt].
headcanon 1597 days ago [-]
Sorry but I'm not clear on what IPA means, I assume we're not talking about the beer?
"Stjärn" is not the Swedish word for star, that's "stjärna".
viceroyalbean 1597 days ago [-]
The author seems to be from Finland, so I wonder if it's a difference between "normal" Swedish and Finnish Swedish (finlandssvenska)
nils-m-holm 1597 days ago [-]
It's "stjärna" in both, it is just pronounced a bit differently.
tilt_error 1597 days ago [-]
The Finnish form of Swedish resembles how Swedish was like a hundred years a go.
"Skjorta" (a shirt) uses the pronunciation described in the article, but was earlier pronounced as "Skiorta" which I believe is the form used in Finland. Same thing with "Stjärna" (a star) which was formerly pronounced "Stierna".
On the other hand, I was born a Norwegian so I could be all wrong :) Norwegians have a special relationship with "kj"-sounds -- which the younger generation is avoiding. Youngsters are actually replacing the hard "kj" with the soft sound described in the article.
jacobush 1596 days ago [-]
Stierna is closer to proto-indo-european. :)
blissofbeing 1597 days ago [-]
I'm more curious with this question
> "A better question, however, might be why aren’t other similar cases around the world treated in the same fashion…? "
It would seem you can't really divorce science from politics, ever.
knolax 1597 days ago [-]
TBF linguistics as a field doesn't seem to apply empiricism as strictly as the hard sciences. It seems that it was only relatively recently that descriptivism became the norm.
ksaj 1597 days ago [-]
When I was a kid, they taught us to say words like "white" with that sound for the 'wh'. I thought it was rather odd, since nobody outside of English class ever said it that way.
blompa 1596 days ago [-]
The 'wh' sound /ʍ/ isn't the same as the Swedish 'sj' sound /ɧ/, but depending on the realisation (like how it's pronounced Stockholm) it can sound fairly similar. The 'sj' sound is more fricative.
I didn't even hear the difference between them when native English speakers talked for over a decade. It was only when I saw people making fun of a guy for not knowing the difference that I looked it up, and now the difference sounds so obvious!
I was a bit shocked when I realised, too. I mean, I studied French formally at some point and I had learned that it only has two genders but it never, you know, sunk in.
Québécois vulgarity tends to be inserted between the article and the verb, in a sort of semi-adjective form. So if you forgot if le porte or la porte was correct, you'd say l'ostie de porte which translates roughly to "the goddamn door" but lets you collapse le or la into l' and avoid the error altogether.
So I'd map "la table" to "το τραπέζι" and "un oiseau" το "ένα πουλί" and never notice that the gender has changed from one language to the other.
Unless I think about it.
And the same with pronouns also.
A trapezium is a table? (Checks etymonline) And apparently indeed, trapezium comes from a Greek word that means "little table" from "trapeza" meaning table. The things you learn.
(For those who can't read Greek, the Greek writing in YeGoblynQueenne's post are "to trapezi" and "ena pouli". I can read Greek since my mother studied Christian theology and had the alphabet in prominent display to help her learn it, but I can't understand a word of it.)
Have you ever visited Greece? It would be a strange experience, being able to read all the signs but not knowing what they mean. I know because I was in Bulgaria for a conference this summer and I could read everything... but understand not a word :)
That's because it's all in a Cyrillic alphabet which is very similar to the Greek alphabet (it's partly derived from it). Come to think of it, so is the Latin alphabet, but I'm used to seeing Latin signs everywhere, including in Greece.
I had trouble deciphering the alphabet at speed, but once I had I could often understand the word based on it's similarity to spanish (e.g. Σάββατο / Sábado) or english (έξοδος / exit).
I can read (a reasonable amount of) Greek after studying a lot of physics/math :D. But I only understand the words that trickled into the other languages I speak or can recognize a root for.
E.g. I was working on a paper with my thesis advisor and I used the plural of "lemma", "lemmata" (λήμμα, λήμματα). It's actually in the English dictionary, but he advised me to say "lemmas" instead because people reading the paper may not be native English speakers and it would just make the text harder to read. I always want to sneak in fancy words in my papers, but it's a bit tricky to do it right..
I'm not sure it should matter what one particular authority decides, in terms of language use. People, including people writing administrative documents should be free to write in the manner that feels most comfortable to them. If this creates a political problem, the political solution is the ballot box and altering the education or hiring policies.
Customary criticism is fine, but it makes no sense to say "you shouldn't say foo because some organisation that shouldn't have authority says you shouldn't say foo"; instead, you should have some direct reason why foo is problematic.
Your criticism shouldn't be circular either, for instance "you shouldn't use 'singular they' because 'they' is plural" is completely, aside from being historically wrong, ineffective as criticism, since it is merely because of the incorrect description of the word as "plural" by some authority that you believe it to be plural, whereas most people use it regularly to refer to a non-specific person.
They should also write in a way that conveyed information effectively. The so-called inclusive writing is extremely disruptive to the flow of reading and makes job offers, administrative documents and so on a pain to navigate (all for the sake of political positions very few people actually hold), defeating the point of language. I'm sure fellow French speakers will know what I'm talking about.
On the learning side, you also often get told rules that don't entirely make sense. For instance I'm told in a Mandarin class that there's a difference between 不 and 没 which is whether it's subjective or objective that something isn't true. But that's neither clear when saying something, nor clear in terms of examples I've been given. What's really weird is I speak another Chinese dialect that also has this distinction, and I also can't tell you the rule there.
I don't think that rule holds. 没 is usually used to negate either 有 "have" or something in the past. E.g. 我没告诉你。 "I didn't tell you." 不 is used for everything else, i.e. 我不告诉你。 "I'm not telling you." (The interesting thing is that "have" also has a past aspect in English, e.g. "I have told you." But Mandarin doesn't use 有 as a general past marker, only the negation 没)
> I speak another Chinese dialect that also has this distinction, and I also can't tell you the rule there.
Which one? Is the rule the same as in Mandarin, i.e. can you produce correct Mandarin sentences despite not knowing the rule, relying just on intuition? I think Cantonese uses 冇 to negate 有.
More technically, 没 is the negative marker that applies to the verb 有, and 不 is the negative marker that applies to everything else. It isn't the 没 that marks a negative in the past as being past -- it is the (omitted) 有. That 有 is commonly not omitted in negative past sentences -- and it's also possible to use 有 to mark past time in positive-polarity sentences, though this is much rarer than the same marking in negative sentences.
There's a straightforward information-theoretic reason why 有 can be so commonly omitted -- it is the only thing that can ever follow 没, so actually realizing it doesn't add much. Everyone knows it's there.
> E.g. 我没告诉你。 "I didn't tell you."
This is the same sentence as 我没有告诉你, also fully idiomatic. The past sense is coming from the 有.
> But Mandarin doesn't use 有 as a general past marker, only the negation 没
This is plainly incorrect. 有 is used as a general past marker. It just isn't obligatory in positive-polarity sentences. Here are some things Chinese people have said to me:
你在中国 有看见过 美国没有的水果吗？("Have you seen any fruits in China that America doesn't have?")
我有吗 ("Did I?")
On the other hand, maybe that is a shortening of an originally obligatory two-part expression similar to Jespersen's cycle, famously observed in the appearance of "pas" and ongoing disappearance of "ne" in French "ne ... pas". https://en.wikipedia.org/wiki/Jespersen%27s_Cycle
As you can see, there's more than one way to negate something. Well it's hard for me to know if the rule is the same, given that I don't really know what the rule is in either! As per my previous comment I can tell you what sounds right, but I can't tell you why.
Neuter has stayed as is.
There are, however, remnants of the masculine/feminine distinction left: e.g. klockan is ofter referred to as feminine (hon), and the "-e" in "storebror" (as opposed to "-a" which is used for utrum today and feminine in the past).
The -e/-a ending is still in common use not only in constructions like "storebror" but in general in adjectives pointing to a female or male person ("den fule mannen", "den fula kvinnan") although this usage is falling out of favour a bit ("den fula mannen" would not generally sound wrong to most modern speakers)
Another interesting remnant of old Norse that still lingers in modern Swedish is the accusative case ending -s in expressions like "till fjälls", "till skogs", "till sjöss", "till sängs" etc.
No, this is not accurate, neither historically nor according to modern usage. Animate objects tend towards utrum but plenty of inanimate objects are utrum too and there are no hard rules or mnemonics you can use to divine which is which.
It's "ett tåg" ("a train") but "en bil" ("a car"), "ett äpple" ("an apple") but "en banan" ("a banana"), "ett barn" (a child) but "en vuxen" (an adult). This is entirely arbitrary and non-obvious to non-native speakers and if you're learning Swedish, you just straight up have to learn which is which for every word.
For example, "a friendship" is "et venskab" in Danish, but "en vänskap" in Swedish.
If anyone pronounces <ch> as <sh>, <z> as <s> and <j> as <y> they're probably Swedish.
As a native English speaker, it took me a while to hear the difference between Swedish "tj" and "sj" (at least the way it's pronounced in Northern Swedish, like "schhh").
It does, but only in the onomatopoetic word "bzzz", as in the sound of a flying insect. And if you were to drop that sounding s, if you said that a fly went "bsss" around you, people would be weirded out, thinking the fly pissed on you or something. So pretty much every Swede can make the sound [z], and can hear the difference between [z] and [s], but only for this single word. It's so weird!
Now as for Swedish, what trips me up is that I've met so many Swedes that (otherwise) speak great English, but seem to pronounce J sounds in a distinctly Swedish way when speaking English. I've had fluent English-speaking Swedes in a university setting asking if I want to hear a "yoke" (joke) and or maybe telling me about their "yeans" (jeans). It's so weird. It's a bit like how many French people seem unable to pronounce an H when speaking English, but Swedes are damn good at English, unlike the average French person.
Likewise, I can identify English with a Danish accent pretty quickly, but I can't for the love of me speak it, because I'm not good enough at Danish. But a dead giveaway is that Danish realizes 'R' as [ʁ], and many Danes pronounce it the same way when speaking English. :-P And there's something with the vowels that's just weird that Danish people do... I can't put my finger on it.
(I have the opposite problem, I can't make that 'R' sound at all, which is why my Danish sounds pretty funny to native Danes.)
It isn't used like a vowel. There's no word 'XwY' for any string of letters X,Y - and you can't prounounce 'xwy' (for any single letters x,y) without adding vowel sound.
E.g., (spelling vs pronunciation) 'town' vs 'toun', 'own' vs 'oan', 'bow' vs 'bo' or 'bau', etc.
It definitely seems like the most vowel-like consonant after 'y'.
https://en.wikipedia.org/wiki/Semivowel
Awe, ewe, owe, awl, owl, own, two...
But my point was that in all other European languages, the letter 'W' is clearly a consonant, usually realized as [v] or [ʋ], and seen as a variant of 'V'.
English is the odd one out, English sees it as a variant of 'U', and it's not clearly a consonant, it's a semi-vowel as you pointed out further down.
It varies across the v <-> w spectrum depending on region, and also some words seem mostly nearer one or the other (but it's the same character, there isn't a second).
You sometimes hear that in native speaker's English, but more 'v's sounding nearer a 'w' than the other way around (as in native German speakers).
Fascinating how growing up with different languages teach us different sounds, and those we didn't use can be so difficult to produce, even if we can hear and understand the difference.
I struggle with two things in Hindi: aspiration (transliterated as 'ta' vs 'tha' for example, but that's not an English 'th', it's 'ta' with air forced out) which I usually either don't pronounce clearly, or it's way over-exaggerated; and retroflex consonants, which I can hear but often struggle to pronounce correctly quickly enough to flow with the rest of the word. And particularly 'd.a' which when correctly pronounced contains (to my ear) some 'r'.
For those that don't know.
It was nice to read it once again...
The English E covers е, и, (roughly) ы in Russian. The English y covers the same but also partially у and ё.
When you get to dialectic English, then all of those vowels have different vocalisations or they otherwise don’t exist. And in all cases Y is a vowel and the same as J.
I love languages and linguistics and this kind of thing.
What’s more, take this to Spain and South America and in Castellano your ‘ll’ is more like an English ‘y’, but in Latin American it’s more of a ‘dj’ or an English ‘j’.
https://en.wikipedia.org/wiki/Semivowel
EDIT: Also, the Spanish <ll> as /ʒ/ (S in English treasure) is not nearly as extent in South America as you might think. It's still mostly limited to Argentina and Uruguay:
https://en.wikipedia.org/wiki/Ye%C3%ADsmo#She%C3%ADsmo
EDIT EDIT: I misread the symbols a bit, and in other areas of South and Central America it's /ʝ/ instead of /ʒ/, which is not a very big difference at all. (Palatal vs postalveolar... Basically neighbors.)
In Puerto Rican Spanish (at least, my grandmother's family's accent), the <ll> is more of a /dʒj/ sort of thing. "llama" comes out to "djyamma", with the "dj" softened significantly. Not sure how to express that more precisely in IPA, though.
The Wikipedia I linked is about the merging of <ll> and <y> to /ʝ/ if you want to read more. Interestingly, I don't hear this a lot in the northern Mexican Spanish I hear, even though all of Central America is marked as merged.
I didn't read it correctly the first time, and I was wrong to call it out in the first place. The difference between /ʝ/ and /ʒ/ is not so big for me to have said anything.
Also, did you mean Greek θ (theta)? θ started as aspirated /tʰ/, and it's a pretty common sound change for aspiration to become a fricative, in this case /θ/. φ also made a similar journey from /pʰ/ to /ɸ/ to /f/.
Not sure why phi would be mapped to the voiceless interdental fricative ("th") though. I thought that came from theta. But I have not studied Greek, so I really don't know haha.
https://en.wikipedia.org/wiki/Koine_Greek_phonology
(Click through the "stop consonant" sections.)
The letters 'b' and the 'v' are pronounced in Spanish as voiced bilabial fricatives after a vowel, and as stops after a nasal consonant (which in that position will be pronounced as [m]), and generally in utterance-initial position.
Swedes tend to pronounce English "yes" and "yoke" as well as "juice" and "joke" all with the same Swedish J sound.
https://translate.google.com/#view=home&op=translate&sl=en&t...
https://translate.google.com/#view=home&op=translate&sl=sv&t....
https://en.wikipedia.org/wiki/Sj-sound
To make the sound with your mouth, make a really guttural 'H' sound like a golf swing.
Honestly, we're all basically just hoping whoever wrote it knows what the symbols mean. An example: Greenwich (https://en.m.wikipedia.org/wiki/Greenwich).
I agree it would be cool if there was a good IPA speech synthesizer, that would certainly help! There's some discussion on the issues and problems with it here: https://news.ycombinator.com/item?id=13514258
If you take a random language and believe that you can pronounce it correctly by reading IPA, well that won't work. Or at least be very hard.
Because English is spelled non-phonetically, English learners across the world generally learn IPA together with English, to learn pronunciation.
Granted it's not the entire set of IPA sounds, and some students are better than others... but seeing as there many more speakers of English as a second language than native speakers (according to some sources, though this can be disputed), there must be at least hundreds of millions of people with a rudimentary knowledge of IPA, enough to sound out words in English.
The actual sound that gets produced can vary by context, what sound you made before that and what will come after etc. (And who is saying it off course).
See this diagram: https://i.pinimg.com/originals/f8/1e/ea/f81eeaa4921133c3b67b...
As a local I hear (and use) all of them interchangeably and am only vaguely aware of the difference if I consciously think about it
What do foreign language textbooks in English-speaking countries use to teach pronunciation?
Using own systems I have seen in very naive tourist guides and in text books by English speaking publishers.
I do understand this sound on a cognitive-analytic level (some phonetics courses at university help when learning sounds). I can pronounce it, at least passably.
There are big regional variations to pronouncing sj and so on. Fine, I can live with that, just pick one and use it consistently, preferably that of your teacher.
But when it comes to the sound the article is about, I hear something else.
At some point I cut together many instances of words with that sound from Swedish pop songs and told my teacher which sound (IPA) I clearly hear in all those instances. She told me flatly that I'm wrong. Those singers clearly sing another IPA sound.
It must be similar to when Russians try to show me the difference between hard and soft consonants. They tell me they really over-pronunciate now and they sound starkly different, I'm not even sure I hear a difference.
An example video: https://youtu.be/lXp7_Sjgm34?t=273
For example "I read to my son" could be pronounced two different ways depending on tense. "I read", vs. "I red".
There are other examples too, off the top of my head "I polish my shoes", vs. "I have a polish friend.". Or "the bow of a boat", vs. "at the end of the play the actors take a bow".
But that's not what's going on in the Swedish example above. The vowels are exactly the same between "anden" (the spirit) and "anden" (the duck), but the pitch accent is different, the "song" of the words is different.
Here's a video showing the difference:
https://www.youtube.com/watch?v=lXp7_Sjgm34
(And if you can't hear the difference in that video, don't worry too much, it's Swedish that's weird, not you. :) )
I am wondering about your accent, though, because, to me, "bow" is pronounced the same in both of those examples. Robin hood shooting an arrow from his bow, however, has a different pronunciation which is the same as the bow tied on my present.
https://www.dictionary.com/browse/bow agrees indicating IPA baʊ for both.
I just assumed they meant bow as in bow and arrow. The other example I did not know. "Read before you think before you write".
Ever-relevant poem: http://ncf.idallen.com/english.html
English is kind of a hot mess.
[0] https://sail.usc.edu/~lgoldste/General_Phonetics/Source_Filt...
Multiple phones can belong to a single phoneme within a given language. For example, English groups the aspirated stops with their non-aspirated counterparts (e.g., [pʰ] and [p] both belong to /p/ because they are non-contrastive, despite being different sounds phonetically).
This is why it is the International Phonetic Alphabet, and not the International Phonemic Alphabet.
[1] Worth noting that there can be some variance among different speakers when it comes to the articulation of specific phones, but the IPA essentially is the result of determining whether languages draw any meaningful distinctions among these. If there are separate symbols in IPA, the phones are noticeably distinct. It is very uncommon for multiple sounds to get mapped to a single phone in IPA by rule, and this usually happens based on some dispute among linguists. A good example might be the "tensed" consonants of Korean, which have their own diacritic applied but are not fully understood (meaning linguists cannot precisely identify what, if anything, separates them from their un-tensed counterparts).
For example, if you pronounce "sju sjösjuka sjömän" with [ʃ] instead of [ɧ], I'll understand you just fine, but I'll instantly know that you aren't a native Swedish speaker. (Or, you are, and you're speaking a Swedish dialect that has replaced [ɧ] with [ʃ], because of course that's also a thing)
But if you mis-pronounce "skön" (nice) as [ʃøːn] instead of [ɧøːn], you're getting awfully close to [ɕøːn], which is how you pronounce "kön" in Swedish, which means gender. So as a Swedish speaker, the phoneme has changed, I no longer know if you're saying "skön" or "kön", or maybe "schön" in German? Swedish "cares" about this difference. But you, as an English speaker, might not be able to hear the difference.
And if there was a hypothetical language that had even further subdivisions of these sounds, speakers of that language would be upset that neither you nor me could tell the difference between two sounds that speakers of that language care about.
So the mapping of phones to phonemes is highly language (and dialect) dependent, and you should think about it in terms of continuous ranges or tolerances, instead of a discrete 1:n mapping.
The point of IPA is to provide a phonetic alphabet that can provide enough symbols, each covering a range of similar sounds across languages, such that for each language, every sound within that language has a unique symbol to differentiate it from other sounds within that language.
E.g. "bête" in French and "bet" in English are identical in IPA: /bɛt/. But all three sounds are definitely slightly different, which is how you can tell whether the speaker is, say, American or French.
The process of coming up with IPA and deciding what sounds in different languages deserve the same symbol or not involved just as much art as science, since there are tons of gray areas, especially when dialects are involved.
https://www.youtube.com/watch?v=7ZmqJQ-nc_s
EDIT: Clear as mud now, https://www.youtube.com/watch?v=9x7C6qeh-SQ.
Are you perhaps from Norrland? Then I see what you're saying.
Here's a (probably non-exhaustive) list of 65 different variants. The article in Swedish, but it should be possible to understand the list:
https://sverigesradio.se/sida/artikel.aspx?programid=411&art...
Edit: I don't know of any rules for when to use which spelling. When I was a kid, I just learned how the different words were spelled individually, even though the sound is the same. Same thing when I'm teaching my kids: "Sju (seven) is spelled like this, schack (chess) is spelled like this, charmig (charming) is spelled like this", and so on.
It makes some rather sveamål-centric assumptions, even then for some subset of sveamål, which is rather notorious for joining long stretches of consonants ending with an "sj" phoneme into just "sj". For example any of the variants that start in "r" only happen in dialects where the "r" is barely pronounced. In most of Sweden, it is distinct; guttural or rolling. Others only occur when consonants that are distinct in other dialects are dropped, and definitely not only the ones marked as "sloppy pronunciation".
Others occur only when crossing words in compound words with distinct "sj" sounds (again, in some dialects), where pronouncing them as a single phoneme again may happen in some sveamål. Some of them seem really forced.
I would say that a word like "duschschampo" has two different sounds for me. I'm from Stockholm so I would do the "schsch" part like /ʃ ɧ/, I have a friend from Kalmar who would say it like /ɧ ɧ/, and probably, somewhere, someone would merge them into a single /ɧ/, but I have no idea where you'd find that person.
There are still some slots free in the sounds matrix that no language as yet uses.
https://www.youtube.com/watch?v=9uZam0ubq-Y
"Skjorta" (a shirt) uses the pronunciation described in the article, but was earlier pronounced as "Skiorta" which I believe is the form used in Finland. Same thing with "Stjärna" (a star) which was formerly pronounced "Stierna".
On the other hand, I was born a Norwegian so I could be all wrong :) Norwegians have a special relationship with "kj"-sounds -- which the younger generation is avoiding. Youngsters are actually replacing the hard "kj" with the soft sound described in the article.
> "A better question, however, might be why aren’t other similar cases around the world treated in the same fashion…? "
It would seem you can't really divorce science from politics, ever.
Also, some English dialects (such as those in Ireland and Scotland) definitely still pronounce 'wh' that way, I refer to the following article for an overview: https://en.wikipedia.org/wiki/Pronunciation_of_English_%E2%9...
The guy has some really good videos BTW!