For all the bad rep French gets about not pronouncing letters in words, at least the t in tsunami is clearly pronounced. As is the p in psychologie, pseudo and pneumonie.
While it’s true from an English speaker’s perspective, to a non-native whose alphabet contains the ‘ts’ sound as a separate letter: there’s no ‘t’ sound in our ‘ts’ sound (which is just the letter ‘c’ for us - boring details in the spoiler section), since it’s not a series of ‘t’ and ‘s’, but one consistent sound on its own, deserving its own letter.
So we would naturally say the ‘ts’ in tsunami doesn’t have a ‘t’ or an ‘s’ (or any silent letter) in it.
spoiler
The 3 c’s in ‘pacific ocean’ are all pronounced differently. We don’t have that madness. Spelling to pronunciation is consistent. The first ‘c’ is just an ‘s’ sound, so we just use our letter for the ‘s’ sound. Which is ‘sz’. It might look weird, but only because the letter ‘s’ is preserved for the ‘sh’ sound. Unusual, but consistent.
The middle ‘c’ is pronounced ‘k’. We just use ‘k’ when it sounds like ‘k’.
So what do we use ‘c’ for? The ‘ts’ sound. Consistently. ‘tsunami’ is spelled ‘cunami’ here.
This is why it sounded so weird when as a kid I watched an American film and they looked at a written word and went “How do you pronounce it?”.
WTF mate, it’s written right there!
Okay but even as a native hungarian i have to say “sz” is just insane. Polish is so much more logical with “s” making the non-shushing sound and “sz” making the shushing sound. Also “gy” is not very logical as it should be “dy” but when the italians first tried to write hungarian down they used g cause in italian g can sound like “gy” in many cases. And then lastly “ly” is just utterly useless, it used to represent a seperate sound with some words contrasting them “boj” vs “boly” was a thing from what i could find, but the dialect modern standard hungarian was based on did not have this distinction so they just used “j” everywhere but kept “ly” where there wasnt a word with “j” that was spelt the same otherwise. This means modern day hungarian speakers in for example slovakia will say all “ly” as the sound for “ly” and “j” will be a mix between the two. As for features they couldve written down but didnt, in both of the dialects i grew up with(family from szeged, but grew up in the north west) there are two “e” sounds, sometimes written as “e” and “ë” when the distinction is wanted. Szeged is an example itself with it being possible to write it as “Szëged” because in the dialect there its pronounced as “Szöged”. On the other hand where i grew up its “Szegæd” where “æ” is pronounced kind of like “a” in english “ash”. Point is hungarian spelling is very much not logical, there are even more examples like “lesz” actually being pronounced with long sz usually so “lessz” would be more accurate, “is” and “se” being clitics like “-e”, and there actually being a long and short “e” distinct from “é” in some words technically making them phonemes like in “ere”(his vein) pronounced “ere” vs erre(this way) pronounced as “eere” or more often written as “ēre”. So yeah many holes tho still a thousand times better than most european languages spelling.
Okay but even as a native hungarian i have to say “sz” is just insane. Polish is so much more logical with “s” making the non-shushing sound and “sz” making the shushing sound.
Totally agreed. Like, also, everything else that you mentioned, but especially this. I really don’t get where they got the idea from, since every (most?) language around has it the other way around, including Latin and German (and their derivatives like romance languages and such). Maybe Turkish and/or Romanian that I can think of, but AFAIK even they put diacritics on their ‘s’ when referring to the ‘sh’ sound (I don’t speak any of those, though, so I might be horribly wrong).
It’s especially annoying because most Western European speakers are more used to the Polish sz-s system (which makes more sense), completely butchering words like Szoboszlai (I’m playing E-Football and the commentator pronounces his name like three different wrong ways, about never right).
I had a relatively old teacher ~30 years ago that said ‘j’ and ‘ly’ are pronounced slightly differently, giving two examples, and I couldn’t hear any difference : ).
I used to go to uni in Szeged so it was quite eye-opening reading about the weird 'e’s. So if I got that right, that’s the reason they say stuff with ‘ö’ instead of ‘e’ in certain cases (because it’s not exactly ‘ö’, but that’s the closest regular letter we’ve got)?
And yeah, ‘bicikli’ is actually pronounced more like ‘bicigli’ (or not? Could be regional) and stuff, but what I meant by ‘consistent’ is that almost always if you see something spelled a certain way you can expect it to sound that certain way, without bullshit silent letters and completely different consonants or vowels. Not in terms of phonemes but actual letters.
And yeah, ‘bicikli’ is actually pronounced more like ‘bicigli’
(I’m still salty about ‘bicikli’ keeping the ‘k’ but the short for ‘lakodalom’ is ‘lagzi’ and not ‘lakzi’)
Yeah it works the same in every single neighbouring language, s is always the normal s(sz for us) and s+something is sh(s for us), i actually dont know where this convention comes from as the y marking for soft sounds is actually logical except for gy but thats a completely exceptable way to write it.
As for the szeged ö the sound is actually the exact same as other the other ös, but yes in the other dialects they are extra sounds that dont have their own way of writing. Also in standard hungarian this is not present so you wont hear it in budapest, northeast and from younger speakers in larger cities.
I usually pronounce bicikli with with a pretty solid k but i can easily imagine a lot of people dont, ive never actually payed attention to that specifically. The reason its bicikli but not lakzi is because of certain phonological effects having to do with “l” being much closer to a vowel than “z”. For bicikli/bicigli the voicing is just at a different point while for lakzi you would need to articulate a k, then start voicing, and then articulate a z and then transition to the i which is much harder to do. But dont worry there are thousands of examples like this that very much have stupid rules for spelling.
Phonetically though, the sound in question is composed of a t and an s sound. You can recognise that regardless of your own language.
Ts is an affricate, composed of a stop and fricative with the same place of articulation. The analysis is similar to analysing diphthongs as being composed of two vowel sounds.
Not exactly, a few languages will pronounce them differently, treat them differently for syllable weight, geminate them and even contrast them. A few examples being, in languages that need this que, “ts” is pronounced longer than “c”. Syllable weight and syllable structure often treats them as one, with some languages being CV for example where a syllable is always composed of one consonant and one vowel yet they will allow affricates which means they are one consonant(this is alsk the case for prenasalization). In languages that have geminates its often a rule that if theres a geminate in a syllable, its always the coda(so after the vowel) and there can be no other consonant in the coda if its geminate yet it will allow affricates, this is the case for hungarian and sami languages for example from what i know. And lastly in polish for example they even contrast directly with the wikipedia example of “dżem” vs “drzem” where dż is a single retroflex affricate segment and drz is a retroflex d and z after eachother but are not an affricate. They should probably do research on how this looks in your throat and do those sonic analyses they did with prenasalized stops to determine they are actually pronounced differently and arent just quirks in the syllable structure.
“means they are one consonant” is the perspective of the language’s speakers, not of the actual sounds they are making. The former is a subjective matter, the latter is essentially objective. Whether speakers believe they are saying one consonant or two, a “ts” affricate always consists of stopping the air with the the tongue behind the teeth (the t) and then releasing it with a lot of turbulence with the tongue still behind the teeth (the s). Those sounds are physically distinct, even if speakers think of them as comprising a single logical unit.
This is also consistent with there being a slightly differently pronounced sequence comprising a t and an s sound which is not analysed as a single consonant! Both are two physical sounds.
In my language most native speakers don’t realise that diphthongs are made of two vowel sounds, but they still are :)
I mean you have to draw the line somewhere, many consonants are composed of multiple articulotary parts that you have to perform seperately. Simple stops, or better called plosives, can themselves be unreleased for example which means a /t/ for example is already a closure /t̚/ and then the release of /t/(idk if theres a diacritic to mark a release cause you already need a closure to be able to release something) so a t can be analyzed as /t̚t/ which might sound stupid but with english some of these interactions are actually noticable with depending on stress and the speakers accent a final t can be many things like unreleased(/t̚/), released(/t/) or even ejective(/t’/). Of course its absurd to assume things like these are valid analyses in any but a few select languages but that once again means that how you analyze things matters a lot on the language being analyzed. Who am i to say a language doesnt contrast /t̚/ and /t/ word finally, if you try it out you can hear the difference. So yeah to wrap it up a plosive is made out of a stop and a release and so is an affricate except that the release is of a different manner, but crucially, at the same place of articulation. This doesnt mean youre wrong cause i still think its a huge problem in linguistics that we just slap some random ass discription on things without elaborating and assuming everyone follows the same conventions. I have the same problem with narrow transcriptions, they should be very accurate but ive seen some stuff narrowly transcribed in a way that wouldnt pass as a proper broad transcription. In some languages affricates function completely as a single consonant and in some languages it just doesnt me sense to make that distinction.
I am not sure of any but some english dialects might use i to mark word stress which could make it phonemic? Im reaching a bit here but i know i read some analysis of some english dialect doing this with ejective versus modal voiceless stops but i cant find it lol and its hard to find these studies and writeups, i usually just read them and then can never find them again. Another good candidate would be some asian, especially south and east asian language that often have required unreleased stops at the ends of words which sre required for your pronounciation to sound correct, so thats halfway there, now there just has to be one that actually contrasts them. But overall in most cases that i know of the release is simply replaced with something else, like in affricates but in some languages there are such things as oral(the normal release), fricative(those make the affricates), nasal, lateral and glottal releases. Dont ask me about glottal releases, i have no idea how those work, im not good with the glottal stuff but for lateral and nasal releases you can imagine what english does(depending on your dialect) with clusters where there is a t and an l or an n right after(idk if there are ones with n). You actually dont release those like in the word “atlantic” and yes we got em, wiktionary transcribes it as [ætˈlæɾ̃.ɪk], which once again doesnt even matter cause im sure a lot of people pronounce it that way but [æt̚ˈlæɾ̃.ɪk] would also be correct. But once again linguistics is a huge circle jerk, so never take anything seriously related to it. Theres the swedish sj sound or /ɧ/ which is a completely brainfuck but im not gonna get into it but basically they have a unique ipa symbol for a swedish sound that has like 20 ways of getting pronounced based on dialect and the sorrounding sounds, has something like 50 ways of being written(im not kidding here) and some of the most excepted ways of describing it with more narrow ipa transcriptions are humanly impossible to pronounce 💀
True, I might have phrased it wrong. The way we (I?) pronounce that ‘c’ the tongue position is slightly different from ‘t’ (less of the very tip of it, a tad bit further to the back, close to the tip), hence I thought it deserves the status of a separate sound instead of a transition between the two. Mind you I’m not a linguist and it may be completely bonkers from an actual expert’s perspective.
It’s the difference between a purely phonetic viewpoint and one based in the language - you’re not really wrong to see it one way, just that there’s another as well.
But you certainly can say “let’s” and not just “le’'s” right? We don’t have it word-initial in a standalone context, but it exists. Certainly, it should be pronounced then, right?
No because in careful speech for example “let’s” is pronounced as “l-e-t-s” while in languages that do have the “c” sound, even with careful, slow pronounciation its “l-e-c”. English, like every language, has rules about clusters and its basically a rule that “s” always has to be farther away from the syllable nucleus(the vowel basically) than the stop(t in this case). This results in “step” being valid but “tsep” not being valid. Theres also the extra layer of brits being more european so they often pronounce it as “cunami” while americans pronounce it as “sunami” from what i gather. Also another thing is that english doesnt have any base words that distinguish the “c” sound, theyre always loanwords and you can pronounce them as “s”
For all the bad rep French gets about not pronouncing letters in words, at least the t in tsunami is clearly pronounced. As is the p in psychologie, pseudo and pneumonie.
While it’s true from an English speaker’s perspective, to a non-native whose alphabet contains the ‘ts’ sound as a separate letter: there’s no ‘t’ sound in our ‘ts’ sound (which is just the letter ‘c’ for us - boring details in the spoiler section), since it’s not a series of ‘t’ and ‘s’, but one consistent sound on its own, deserving its own letter.
So we would naturally say the ‘ts’ in tsunami doesn’t have a ‘t’ or an ‘s’ (or any silent letter) in it.
spoiler
The 3 c’s in ‘pacific ocean’ are all pronounced differently. We don’t have that madness. Spelling to pronunciation is consistent. The first ‘c’ is just an ‘s’ sound, so we just use our letter for the ‘s’ sound. Which is ‘sz’. It might look weird, but only because the letter ‘s’ is preserved for the ‘sh’ sound. Unusual, but consistent. The middle ‘c’ is pronounced ‘k’. We just use ‘k’ when it sounds like ‘k’. So what do we use ‘c’ for? The ‘ts’ sound. Consistently. ‘tsunami’ is spelled ‘cunami’ here.
This is why it sounded so weird when as a kid I watched an American film and they looked at a written word and went “How do you pronounce it?”. WTF mate, it’s written right there!
Okay but even as a native hungarian i have to say “sz” is just insane. Polish is so much more logical with “s” making the non-shushing sound and “sz” making the shushing sound. Also “gy” is not very logical as it should be “dy” but when the italians first tried to write hungarian down they used g cause in italian g can sound like “gy” in many cases. And then lastly “ly” is just utterly useless, it used to represent a seperate sound with some words contrasting them “boj” vs “boly” was a thing from what i could find, but the dialect modern standard hungarian was based on did not have this distinction so they just used “j” everywhere but kept “ly” where there wasnt a word with “j” that was spelt the same otherwise. This means modern day hungarian speakers in for example slovakia will say all “ly” as the sound for “ly” and “j” will be a mix between the two. As for features they couldve written down but didnt, in both of the dialects i grew up with(family from szeged, but grew up in the north west) there are two “e” sounds, sometimes written as “e” and “ë” when the distinction is wanted. Szeged is an example itself with it being possible to write it as “Szëged” because in the dialect there its pronounced as “Szöged”. On the other hand where i grew up its “Szegæd” where “æ” is pronounced kind of like “a” in english “ash”. Point is hungarian spelling is very much not logical, there are even more examples like “lesz” actually being pronounced with long sz usually so “lessz” would be more accurate, “is” and “se” being clitics like “-e”, and there actually being a long and short “e” distinct from “é” in some words technically making them phonemes like in “ere”(his vein) pronounced “ere” vs erre(this way) pronounced as “eere” or more often written as “ēre”. So yeah many holes tho still a thousand times better than most european languages spelling.
Totally agreed. Like, also, everything else that you mentioned, but especially this. I really don’t get where they got the idea from, since every (most?) language around has it the other way around, including Latin and German (and their derivatives like romance languages and such). Maybe Turkish and/or Romanian that I can think of, but AFAIK even they put diacritics on their ‘s’ when referring to the ‘sh’ sound (I don’t speak any of those, though, so I might be horribly wrong).
It’s especially annoying because most Western European speakers are more used to the Polish sz-s system (which makes more sense), completely butchering words like Szoboszlai (I’m playing E-Football and the commentator pronounces his name like three different wrong ways, about never right).
I had a relatively old teacher ~30 years ago that said ‘j’ and ‘ly’ are pronounced slightly differently, giving two examples, and I couldn’t hear any difference : ).
I used to go to uni in Szeged so it was quite eye-opening reading about the weird 'e’s. So if I got that right, that’s the reason they say stuff with ‘ö’ instead of ‘e’ in certain cases (because it’s not exactly ‘ö’, but that’s the closest regular letter we’ve got)?
And yeah, ‘bicikli’ is actually pronounced more like ‘bicigli’ (or not? Could be regional) and stuff, but what I meant by ‘consistent’ is that almost always if you see something spelled a certain way you can expect it to sound that certain way, without bullshit silent letters and completely different consonants or vowels. Not in terms of phonemes but actual letters.
Yeah it works the same in every single neighbouring language, s is always the normal s(sz for us) and s+something is sh(s for us), i actually dont know where this convention comes from as the y marking for soft sounds is actually logical except for gy but thats a completely exceptable way to write it.
As for the szeged ö the sound is actually the exact same as other the other ös, but yes in the other dialects they are extra sounds that dont have their own way of writing. Also in standard hungarian this is not present so you wont hear it in budapest, northeast and from younger speakers in larger cities.
I usually pronounce bicikli with with a pretty solid k but i can easily imagine a lot of people dont, ive never actually payed attention to that specifically. The reason its bicikli but not lakzi is because of certain phonological effects having to do with “l” being much closer to a vowel than “z”. For bicikli/bicigli the voicing is just at a different point while for lakzi you would need to articulate a k, then start voicing, and then articulate a z and then transition to the i which is much harder to do. But dont worry there are thousands of examples like this that very much have stupid rules for spelling.
Phonetically though, the sound in question is composed of a t and an s sound. You can recognise that regardless of your own language.
Ts is an affricate, composed of a stop and fricative with the same place of articulation. The analysis is similar to analysing diphthongs as being composed of two vowel sounds.
Not exactly, a few languages will pronounce them differently, treat them differently for syllable weight, geminate them and even contrast them. A few examples being, in languages that need this que, “ts” is pronounced longer than “c”. Syllable weight and syllable structure often treats them as one, with some languages being CV for example where a syllable is always composed of one consonant and one vowel yet they will allow affricates which means they are one consonant(this is alsk the case for prenasalization). In languages that have geminates its often a rule that if theres a geminate in a syllable, its always the coda(so after the vowel) and there can be no other consonant in the coda if its geminate yet it will allow affricates, this is the case for hungarian and sami languages for example from what i know. And lastly in polish for example they even contrast directly with the wikipedia example of “dżem” vs “drzem” where dż is a single retroflex affricate segment and drz is a retroflex d and z after eachother but are not an affricate. They should probably do research on how this looks in your throat and do those sonic analyses they did with prenasalized stops to determine they are actually pronounced differently and arent just quirks in the syllable structure.
“means they are one consonant” is the perspective of the language’s speakers, not of the actual sounds they are making. The former is a subjective matter, the latter is essentially objective. Whether speakers believe they are saying one consonant or two, a “ts” affricate always consists of stopping the air with the the tongue behind the teeth (the t) and then releasing it with a lot of turbulence with the tongue still behind the teeth (the s). Those sounds are physically distinct, even if speakers think of them as comprising a single logical unit.
This is also consistent with there being a slightly differently pronounced sequence comprising a t and an s sound which is not analysed as a single consonant! Both are two physical sounds.
In my language most native speakers don’t realise that diphthongs are made of two vowel sounds, but they still are :)
I mean you have to draw the line somewhere, many consonants are composed of multiple articulotary parts that you have to perform seperately. Simple stops, or better called plosives, can themselves be unreleased for example which means a /t/ for example is already a closure /t̚/ and then the release of /t/(idk if theres a diacritic to mark a release cause you already need a closure to be able to release something) so a t can be analyzed as /t̚t/ which might sound stupid but with english some of these interactions are actually noticable with depending on stress and the speakers accent a final t can be many things like unreleased(/t̚/), released(/t/) or even ejective(/t’/). Of course its absurd to assume things like these are valid analyses in any but a few select languages but that once again means that how you analyze things matters a lot on the language being analyzed. Who am i to say a language doesnt contrast /t̚/ and /t/ word finally, if you try it out you can hear the difference. So yeah to wrap it up a plosive is made out of a stop and a release and so is an affricate except that the release is of a different manner, but crucially, at the same place of articulation. This doesnt mean youre wrong cause i still think its a huge problem in linguistics that we just slap some random ass discription on things without elaborating and assuming everyone follows the same conventions. I have the same problem with narrow transcriptions, they should be very accurate but ive seen some stuff narrowly transcribed in a way that wouldnt pass as a proper broad transcription. In some languages affricates function completely as a single consonant and in some languages it just doesnt me sense to make that distinction.
Thanks for this perspective :) Do you know of any languages which contrast released and unreleased stops?
I guess a difference is that you can’t really have a stop release on its own…
I am not sure of any but some english dialects might use i to mark word stress which could make it phonemic? Im reaching a bit here but i know i read some analysis of some english dialect doing this with ejective versus modal voiceless stops but i cant find it lol and its hard to find these studies and writeups, i usually just read them and then can never find them again. Another good candidate would be some asian, especially south and east asian language that often have required unreleased stops at the ends of words which sre required for your pronounciation to sound correct, so thats halfway there, now there just has to be one that actually contrasts them. But overall in most cases that i know of the release is simply replaced with something else, like in affricates but in some languages there are such things as oral(the normal release), fricative(those make the affricates), nasal, lateral and glottal releases. Dont ask me about glottal releases, i have no idea how those work, im not good with the glottal stuff but for lateral and nasal releases you can imagine what english does(depending on your dialect) with clusters where there is a t and an l or an n right after(idk if there are ones with n). You actually dont release those like in the word “atlantic” and yes we got em, wiktionary transcribes it as [ætˈlæɾ̃.ɪk], which once again doesnt even matter cause im sure a lot of people pronounce it that way but [æt̚ˈlæɾ̃.ɪk] would also be correct. But once again linguistics is a huge circle jerk, so never take anything seriously related to it. Theres the swedish sj sound or /ɧ/ which is a completely brainfuck but im not gonna get into it but basically they have a unique ipa symbol for a swedish sound that has like 20 ways of getting pronounced based on dialect and the sorrounding sounds, has something like 50 ways of being written(im not kidding here) and some of the most excepted ways of describing it with more narrow ipa transcriptions are humanly impossible to pronounce 💀
True, I might have phrased it wrong. The way we (I?) pronounce that ‘c’ the tongue position is slightly different from ‘t’ (less of the very tip of it, a tad bit further to the back, close to the tip), hence I thought it deserves the status of a separate sound instead of a transition between the two. Mind you I’m not a linguist and it may be completely bonkers from an actual expert’s perspective.
It’s the difference between a purely phonetic viewpoint and one based in the language - you’re not really wrong to see it one way, just that there’s another as well.
But you certainly can say “let’s” and not just “le’'s” right? We don’t have it word-initial in a standalone context, but it exists. Certainly, it should be pronounced then, right?
No because in careful speech for example “let’s” is pronounced as “l-e-t-s” while in languages that do have the “c” sound, even with careful, slow pronounciation its “l-e-c”. English, like every language, has rules about clusters and its basically a rule that “s” always has to be farther away from the syllable nucleus(the vowel basically) than the stop(t in this case). This results in “step” being valid but “tsep” not being valid. Theres also the extra layer of brits being more european so they often pronounce it as “cunami” while americans pronounce it as “sunami” from what i gather. Also another thing is that english doesnt have any base words that distinguish the “c” sound, theyre always loanwords and you can pronounce them as “s”