Why Some Languages Sound More Beautiful.

This piece by Bernd Brunner may not have any particular conclusion to offer (“In the end, beauty in language is just one of those things”), but it’s always enjoyable to think about ineffabilities like “why do so many people think German sounds awful?” — and the fact that the author is German (it’s translated by Lori Lantz) makes it interesting for those of us who are used to hearing it talked about from an English-language perspective, and this is a thoughtful paragraph:

Until three years ago, before I started to learn Turkish, I didn’t really feel strongly about the language one way or another. It certainly didn’t sound particularly beautiful to me. But then I began to distinguish sounds as words or components. What’s more, I understood that the ways Turkish combines these components to produce meaning are radically different from the ways Indo-European languages function. As speaking and understanding Turkish required me to perform some mental acrobatics, my perspective on the language shifted dramatically. My deeper appreciation of Turkish not only went along with a deeper understanding of the country’s culture and people, but I also began to realize why Turks who learn German speak the way they do. And, of course, pride in mastering a language, only if to a certain extent, colored my emotional attitude towards it. Simply by learning Turkish, I was inclined to find it more beautiful.

Thanks, Paul!

Comments

  1. David Eddyshaw says

    Personally I’ve always found (standard) German one of the more euphonious-sounding languages. On the other hand, despite having every personal reason to find Welsh pleasant to listen to, I’ve never been able to bring myself to Tolkien’s view that it’s particularly beautiful aurally.

    I had a Circassian colleague some years ago who spoke Turkish, Arabic and English (this last better than any other non-native speaker I’ve ever met.) The last of his family who actually spoke Circassian was his grandmother; according to him, it sounded like someone with a bad cough.

  2. Lucy Kemnitzer says

    I am convinced that association is a big part of how we perceive the beauty of languages. My first exposure to German was Ernst Busch singing Six Songs for Democracy, (youtube link to a list of many Ernst Busch recordings) and the language will always sound like optimism and love and determination to me.

    On the other hand, I can’t think of a language that sounds uniformly ugly to me, anyway.

  3. David Eddyshaw says

    If the business about Japanese people finding French and the Polynesian languages beautiful is right, the explanation that springs to mind would be that, like Japanese, they aren’t big on consonant clusters and aren’t stress-dominated. So that would go with the intrinsically all too plausible “unfamiliar = ugly” hypothesis well enough.

    It would also account for native English speakers *not* finding German ugly, always assuming i’m not the only one …
    In general, the sound system of German is a lot more like the English one than the French or Italian is (as witness the fact that Germans find it a lot easier to acquire a good English accent than French speakers do, by and large.)

    I suspect that anything like this is in practice going to be overwhelmed by prejudices about the speakers and their culture, though.

  4. Another person who finds German a lovely language. For many Americans of my age, I imagine the first impression of German speech was often the horrible tinny pseudo-Prussian whine of Colonel Klink on Hogan’s Heroes: first impressions do linger.

  5. I agree with David Eddyshaw that German (Hochdeutsch) is euphonious – more so than, say, Parisian French. Which is about as objective and universal as saying that peppermint tea tastes better than camomile tea. But I like peppermint tea.

    There’s a phenomenon in some unrelated Asian languages that often strikes Europeans as ugly: the tendency to nasalize and extend certain (usually final) syllables, sometimes apparently with the implication of teasing / disbelief / derision, but also just as an informality marker. This is very common in Cantonese (don’t know about other Chinese languages) and in Malay/Indonesian (don’t know about Indonesian regional languages). It also sounds present in Thai, but I’m just an aural spectator there and may be mistaken.

    There may also be something roughly similar in Quebecois, where – an impression only – some speakers heighten extended nasalization as their speech becomes more colloquial. I know it’s purely subjective to say this isn’t pretty, but…

  6. Russian sounds like a great, grand river to me, sometimes a cascading waterfall. A cultured St. Petersburg accent makes me giddy with delight.

    I am not fluent in Polish, but when I listen–its softer music calls to me alluringly. Polish is a friend of poets.

  7. I’ve introspected a bit about what makes me find languages beautiful (over time I’ve come up with a “perfectly euphonic” phonetic system for a conlang I will never actually construct) and it seems like there are two major factors:

    (1) a wide variety of syllable weights: Japanese sounds too staccato, German too ponderous, Turkish too undifferentiated.
    (2) a relatively small percentage of stop consonants, in favor of fricatives, nasals, liquids.

    Of familiar living languages, Portuguese wins on both counts — I enjoy both major varieties.

    It’s not unlikely that the familiarity thing is playing a role here w.r.t. Russian, although I’ve never perceived Russian to be particularly pleasant.

  8. J.P. says “Polish is a friend of poets.”

    More like a passionate but somewhat abusive lover. Beautiful language, but doesn’t comb well into poetic rhythms. The Poles seem to succeed at poetry despite their language, not because of it.

  9. Love generally requires both connection and an element of surprise. It’s probably the same with languages we love?

  10. Welsh sounds like nothing. I don’t know how it’s possible, but I watched the movie Hedd Wyn and it seemed like they were talking English except that they weren’t.

  11. I first heard Turkish some two decades ago on a visit to Istanbul. What a wonderful surprise! Especially so when contrasted with the rest of the Levant and its gatto-gargling-a-hairball gutturals.

  12. Another data point to set beside ‘guttural’ and ‘nasal’ scunners: although кукурyза is one of my favorite words in any language, I find the abundance of palatalized consonants in Russian unpleasant. I remember watching an interview with Olga Korbut and thinking she sounded like she had a mouthful of mush. Perhaps I was also reacting to the ‘nasty Asiatic vowel’.

  13. marie-lucie says

    English and French are extremely different from each other in sound and stress. Germans seem to be able to accommodate to both languages quite well. The German people I have heard speak either French or English were all doing a very good job.

    According to a Spanish speaker I know, Portuguese is particularly beautiful because tiene la sensualidad del francés y la claridad del español. I don’t have a particular opinion of the qualities of French – it sounds normal to me! As for Spanish clarity, it depends on the speakers. I have found Mexican and Chilean women very clear, but not Chilean men.

  14. The structure of Russian is more friendly to poets, I’d agree with “F” here; however, there is something about Polish that has facilitated the creation of absolutely amazing poetry, especially in the past 100 years.

    Recently, I heard some recordings of poems in Polish and was entranced.

    I am so accustomed to Russian, that I do not always hear its harshness.

  15. I suspect that anything like this is in practice going to be overwhelmed by prejudices about the speakers and their culture, though.

    But then why do I and many other foreigners react differently to the sound of Norwegian, Danish & Swedish? I didn’t know much about the cultures when I moved to Norway, but spoken Danish stood out like a sore thumb to me because it sounded ugly (and was also very hard to understand). I thought Swedish sounded the best; I liked the way certain words and syllables were emphasised. Though prejudice plays a part, as with anything confined to one particular culture, my opinion is that I react to spoken language in the same way as to music. Sometimes I’ll like a piece only after I’ve heard it many times. Nowadays, I love to hear the weirdnesses of spoken Danish – they sometimes remind me of dialects in England – and I’m tired of the quirks of Swedish.

  16. It’s funny how Russian palatalized consonants and Bantu African ones feel so different to me.

  17. “Portuguese is particularly beautiful because tiene la sensualidad del francés y la claridad del español.

    That’s true of Brazilian Portuguese – I would have said the exact opposite is true of the European variety, not that it doesn’t have it’s charms.

  18. Stefan Holm says

    Some quarter of a century ago I read a book by Swedish uralist Björn Collinder. For some reason he referred to a study of the alleged ‘beuaty’ of European languages. They had measured on an oscilloscope the smoothness and ‘pureness’ of sound waves in everyday speech (a prejudice per se). It meant that pure vowels scored higher than diphtongs, the vowel /a/ higher than /i/, fricatives lowered the score and so on (can’t remember all details).

    The result was that Italian ‘won’ among the Romance languages, Slovenian among the Slavic and Swedish among the Germanic (the latter probably the reason why Collinder mentioned it). Of course it’s a matter of taste but one can at least wonder why Italian is the language par preference for so many world famous opera arias and choirs.

    @AJP Crown: I think many of us Swedes find Danish (like English) much better suited for pop/rock music (sounds a little ridicolous in Swedish) while Swedish is far better for ‘chansons’, ballads and the like. So – beauty can vary with the purpose/context.

  19. Danish (like English) much better suited for pop/rock music (sounds a little ridicolous in Swedish) while Swedish is far better for ‘chansons’, ballads and the like.

    Yes. That seems perfect.

  20. A less travelled aspect of this discussion is a comparison of the sound of commercial voiceovers and documentary narration in different languages. Of course, this is not normal speech but an idealised form of speech that is supposedly found attractive by speakers of the language. (Similarly for radio broadcasts).

    I’ve never particularly liked the sound of American women’s speech, but the voice and language used in American commercials seems much more inviting than most women’s voices I’ve heard. Less strident and nasal, with more ‘mellow’ pronunciations. I also once heard a Frenchman comment (in a derogatory way) that Japanese female voices in voiceovers and commercials sound high-pitched because Japanese like their women to sound hysterical.

    On the other hand, I suspect that in many languages males have deeper voices and sound ‘authoritative’, although this possibly doesn’t apply to British English.

  21. spoken Danish stood out like a sore thumb to me because it sounded ugly (and was also very hard to understand).

    Well, come on, even the Danes admit that.

  22. It’s all about the potato, folks.

    Someone in a grumpily misogynistic mood said that there are two kinds of women when it comes to voices, bellowers and shriekers. I was careful (and fortunate) to marry a bellower.

  23. marie-lucie says

    Portuguese : The Spanish speaker was from South America, so he must have referred to Brazilian Portuguese.

    The first time I heard BP (having already taken a class in European Portuguese) was years ago when the songs of Carlos Jobim became popular in North America. I turned on the radio one day and heard a language which sounded like French, but I could not understand a word! I had the impression that someone had cut up a French tape into syllables and reassembled the pieces at random. After a few minutes of intent listening I heard what sounded like a potential French word: “courasson”: curação ‘heart’, of course!

  24. marie-lucie says

    Japanese

    The first time I heard Japanese (many years ago) was in a samurai-type movie, with French subtitles. The men barked rapid orders in gruff voices, sounding like brrr …. brrr …. brrr …., while the women twittered like birds in high-pitched voices. Most of the time it was impossible to guess that they were speaking the same language, or even any human language at all. .

  25. come on, even the Danes admit that.

    Yes, I know they do but do they really believe it or are they just being polite? Because my point is that I really love to hear Danish now I’ve gotten over the initial shock. It’s a similar experience to hearing some “difficult” pieces of modern music where you don’t always appreciate them the first time around but eventually come to like them better than simple pop tunes.

  26. …on another topic, I’ve only recently noticed that Swedish has a very odd S sound, one that is really hard to reproduce where all the air seems to come out of the sides of the speaker’s mouth rather than from the centre.

  27. Gale and I went out for her birthday dinner last night, and at the table next to us two women at the table closest to us were speaking animatedly but softly together. Gale asked me, as she often does, “What language are they speaking?” Atypically, I was completely unable to answer.

    The conversation was fluent and obviously that of native speakers. Both were evidently Europeans: one a typical Nordic in appearance, the other quite Central European looking except for the red hair. The first did most of the talking, rare code-switches into English; at least I think I heard heavily accented English fragments like “and she said to me”, but I could be mistaken. The other had a brief conversation with her daughter (sitting next to her) in English, but I could only hear what her daughter said, so it’s possible it was a bilingual conversation.

    Phonological and prosodic features and the absence of any words I could come close to recognizing (though they were clear enough, no potatoes here) pretty much ruled out any mainstream Germanic or Romance language. It was not any kind of East Slavic nor yet Polish; I can’t absolutely rule out other West or South Slavic languages, but my feeling was that they were improbable. It was not Basque, either. Ad ignorantiam, I was left with a Rhaeto-Romance or Daco-Romance language and some kind of German very different from Standard German as plausible candidates, but I could hear nothing that even vaguely suggested either. I suppose some kind of Roma or Sinti is also a possibility.

    I wound up saying “Why don’t you ask them?” but Gale recoiled in horror from the very suggestion of it. I continued to listen and ponder until they left the restaurant, but no luck.

  28. Alon Lischinsky says

    But then why do I and many other foreigners react differently to the sound of Norwegian, Danish & Swedish?

    My (entirely unscientific) opinion is that it’s simply the weirdness factor. Things like stød or the peculiar contrasts between voiced and unvoiced stops make the sounds of modern Danish hard to recognise for speakers of most Western languages, while Norwegian and Swedish have a more standard phonemic repertoire.

    English has a couple of unusual phonemes in its own inventory (e.g., /ð/), but I suppose that familiarity from film and TV has made them less noticeable across the globe.

  29. I wound up saying “Why don’t you ask them?” but Gale recoiled in horror from the very suggestion of it. I continued to listen and ponder until they left the restaurant, but no luck.

    Man, I would have been over at that table, abashed but determined, asking the question I have asked any number of times in my life, often on NYC subways. Sample answers: Portuguese (often hard to recognize even when you can read it!), Armenian, Albanian. Otherwise I can’t get to sleep at night.

  30. Man, I would have been over at that table

    Me too!

    abashed

    Nah!

  31. I think the sound of a language is where the folk art aspect of language is really obvious. The sound of a language tends to reflect the landscape of the homeland of that language. There’s a reason why Lushootseed or Nisga’a sound so jagged and lush – if douglas firs couldd speak this is how they would sound – why Thai and Cambodian sound pulpy – you can just hear the humidity and lush vegetation – and why Hungarian and Finnish have that galloping rhythm and why Finnish consonants in particular sound like solid, softly rounded cobbles that have been through four glaciations. Japanese has a staccato clarity that I associate with lots of rain and clean air.

    It’s a cliche, but heat and humidity have always been the excuse for the way Southerners speak English. And that’s an example of the amount of leeway speakers have to exercise this esthetic dimension of their speech. They could sound like New Englanders if they wanted to. There is nothing inherent in English phonology that makes either group go one way or the other.

    Alon, there are varieties of English that have plenty of stød, and that make account for some of the negative impression English speakers may have – those are not high-prestige varieties of English and that may stigmatize the feature.

    As for poetry, there is poetry and there is poetry, and it depends on your taste in poetry as to which language more suited to poetry. If you like Emily Dixon, or the similarly crystalline Tang jueju, then there is going to be no language that matches Chinese for terse, blunt grandeur. You can build sentence after meaty sentence without having to waste even one syllable on administrative morphological nonsense. But if you don’t mind, or even like, frills and syntactic grace notes and morphological padding, then SAE langauges will suit you fine. As for me, dificulty is an aspect of enjoying poetry. Irish poetry is beautiful, but after all, it’s in Irish; how hard can ithat be What impresses me is when English poetry manages to be beautiful. When someone can take the language of shopping lists, technical manuals and tabloids and make poetry of it, that impresses me.

  32. they left the restaurant

    Damn. Now we’ll never know. Wait! Come back!

  33. How about Samisk?

  34. It would be interesting to study this systematically: to control for bias, I would suggest that the best approach would be to have hearers rank various language samples in terms of how esthetically pleasing they find them, without identifying the languages, and indeed without the samples consisting of any real language: just write out each “sample” in IPA and have them all read by the same person (A student of phonetics, ideally). Now, while each sample would be in a made-up language, each sample should exhibit systematic features of phoneme inventory, phoneme realization and phonotactics. My own guess (no more than that!) is that a language with no phonemes unknown to the listeners’ L1 and CV phonotactics will be ranked as more beautiful than others (The story of Japanese speakers finding Polynesian languages pleasant to the ear suggests this).

    Having recently spent some time in Denmark I thought I would observe that Danish is to me neither ugly nor beautiful. Just utterly baffling. I could half-read newspaper headlines and signs and the like, but I had the sinking sensation that I could spend the rest of my life in Denmark and become a competent reader of Danish, but never come any closer to understanding this language, no matter how hard I might try. This stands in sharp contrast to Swedish and Norwegian, both of which I find euphonious and both of which I sense I could become reasonably competent in as an L2 speaker (with enough time and work, of course).

    John Cowan: Just a guess: Czech? In my experience that is the least-Slavic sounding Slavic language, and because of language purism over the history of its standardization there are few words which a non-Slavic-speaking foreigner can spontaneouskly recognize. Indeed Romanian sounds more “Slavic” to my ear than Czech does, and in any case its vocabulary is quite cosmopolitan.

  35. ‘Indeed Romanian sounds more “Slavic” to my ear than Czech does, ”

    Do you think it’s the palatalization or the voice quality, etinne? I have the same impression as you.

  36. Trond Engen says

    I asked that question on the London Underground once. The language sounded Finnish, but I couldn’t sort out a single word. Still I ended up asking if it was Estonian. Obviously in hindsight, it was Latvian. Lithuanian sounds very different to me, hard to discern from its Slavic neighbours. I like the sound of both.

    As for John’s language, yeah, maybe Czech, though I think he’d recognize the Slavic prefixing. So I’ll say Latvian.

  37. David Eddyshaw says

    Agree about the melodiousness of Turkish. I was quite struck by this when I first visited Turkey, despite the fact that I was an extremely philhellenic teenager at the time and everybody was grumpy all day because it was Ramadan.

  38. Étienne, English (and no doubt French) speakers who go and live in Denmark seem to learn it as fast and with as little difficulty as Norwegian is learnt here. It may just be a question of learning a few constantly enforced pronunciation rules, though I wouldn’t count on it. The many Norwegian dialects lead to all sorts of rules (e.g. pronoun gender) being almost worthless.

  39. David Eddyshaw says

    @MR:

    “Welsh sounds like nothing.”

    You surprise me. Are you from the UK? If you’re used to hearing Welsh people speaking English, it can be difficult to tell them apart if you aren’t close enough to make out the words, because the sound system of Welsh English is a good deal affected by Welsh, even for people who can’t actually speak it. Most people’s English sounds very unlike Welsh, though. They unaccountably lack voiceless laterals, velar fricatives or high unrounded back vowels, and seem to be able to manage single unvoiced stops between vowels without any trouble. Bizarre. And they can’t sing.

  40. Jim: the palatalization, definitely: indeed the best “definition” of Romanian I have ever heard is “Italian spoken with a thick Russian accent”.

    AJP Crown: Interesting observation. I wonder whether the dialect diversity of Norway, as an obstacle to learning Norwegian, might not be in practical terms the equivalent of the phonological complexity of Danish. And this phonological complexity is definitely real, and indeed was a common conversational topic with linguists in Denmark, Danes and non-Danes alike.

    David Eddyshaw: I’ve had a similar experience with Welsh English versus Welsh, which I found hard to tell apart when I couldn’t hear things clearly enough. Similar such phenomena are attested elsewhere: a Swedish linguist of my acquaintance told me the same regarding the language of the Swedish minority in Finland: from afar this linguist (whose L1 is Swedish, please note!) had trouble telling whether Finnish or Finland Swedish was being spoken.

  41. Stefan Holm says

    @AJP Crown: that ’very odd S-sound’ you think you hear in Swedish must simply be the IPA [x]. It’s close to the initial sounds in Spanish Juan and Russian khleb (bread) or the final in German Buch and Bach, Scottish loch – i.e. a velar fricative. I think it is a quite common consonant around the world although lacking in English.

    As for Danish I don’t believe the real obstacle to be the glottal stop, ‘stød’ Instead it’s a very far pushed reduction of both consonants and syllables. In the Norwegian and Swedish words for street: ‘gate’ and ‘gata’ respectively you will here all four letters clearly pronounced in normal speech (due to the tonal accent system) but the Danish variety ‘gade’ you will hear as just [gæ:]. This in turn is for a non native impossible to distinguish from their pronunciation of ‘gabe’ (yawn) or ‘gave’ (gift).

  42. Trond Engen says
  43. Ètienne, the practical way of dealing with the dialects in teaching Norwegian to foreigners must be to ignore them all but the one that’s local to where the class is being held. That at least is my uninformed guess, and I can’t imagine how else they could handle it. I don’t know what happens when a person from Trondheim wants to teach Norwegian to foreigners in Oslo; maybe they use Oslo dialect, but I know Norwegian people who would find that idea irritating.

    The voiceless rounded dorso-palato-alveo-velar fricative. – I can’t get it to play, unfortunately, but only an oaf would confuse it with the German ch in Bach or Buch.

  44. Étienne.

  45. Czech has fixed initial stress, though, and this language didn’t (unless I was completely mishearing the word boundaries). I didn’t think about the Baltic languages, though: that seems like a real possibility now.

    David Eddyshaw: Wuddaya mean, can’t sing? I’m an anglophone and I sing very well indeed.

    Etienne: I’m adding the line about Romanian to the next edition of “Essentialist Explanations”.

  46. David Eddyshaw: Wuddaya mean, can’t sing? I’m an anglophone and I sing very well indeed.

    I think he was talking about Brits. You know, like the Beatles song, “And Your Brits Can’t Sing.”

  47. marie-lucie says

    Jim: There’s a reason why Lushootseed or Nisga’a sound so jagged and lush – if douglas firs couldd speak this is how they would sound

    Having heard both, I don’t find them very similar.

    Welsh English

    As a student, not yet very fluent in English, I spent a couple of weeks visiting an English family in Herefordshire, close to the Welsh border. One day we went over the border and stopped at a store. I found Welsh English easier to understand than English English, as it sounded to me close to how French people would speak English.

    English English

    A few years ago my daughter and I were in France visiting my family. Once, we were in a restaurant and noticed four people at a neighbouring table: a middle-aged man with (most probably) his two sons and one of them’s girlfriend. From their looks and clothing, they were obviously European. They were close enough for us to hear what they were saying, but we could not identify what language they spoke. After quite a while my daughter (English-dominant French-English bilingual) exclaimed “Ils parlent anglais!” After that she understood most of what they were saying, but I did not.

  48. des von glanzende-overduidelijkheid says

    I think maybe it’s just one of those things?

    But also, I personally find the unreduced wovels of Italian always make it sound loud, bordering on obnoxious. But I don’t get that with Finnish, so it is possible this prejudice has – clutch your pearls, everyone! – other, non-linguistic roots.

  49. Stefan Holm says

    Trond & AJP: If you carefully read the Wiki article you link to you will find it critical to the idea of Swedish fricatives being abnormal in any way. It may be that quantum mechanics allows a particle to simultaneously be in two places – but that’s just until we actually observe it. The same goes for fricatives, they are articulated in one place at a time.

    The vast majority of Swedes uses a velar fricative, i.e. the place of articulation (position of the tongue) is essentially the same as for the consonants ‘k’, ‘g’ and ‘ng’. That makes it belong to the same family as the German ach-laut (with the Ladefoged/Maddieson comment that it ‘may be slightly further forward’).

    The problem is, that in a few dialects – and sociolects, most important the Stockholm elite one, a supradental fricative is common. This has lead to quarrels for as long as one can remember with the popular velar being called vulgar and the supradental niminy-piminy or theatrical. As a result a range of intermediates occurs – from just above the upper teeth via the palatum to the velum. And the degree of rounding is very much a question of colouring from adjacent vowels.

    But we all keep the articulation in one place at a time. This nine-fricative sentence is an enough tongue-twister even for natives: ‘Sju sköna sjuksköterskor skötte sju sjösjuka sjömän’ (Seven beautiful nurses nursed seven seasick sailors).

  50. It sounds as though there was a class war in Sweden over whether to transcribe “sj” as [x] or [ʃ] and so the IPA very generously added an extra letter [ɧ] to its alphabet to spare everybody’s blushes.

  51. Stefan Holm says

    Very much so, mollymooly. And I can imagine that Per Lindblad’s extensive X-raying our tongue positions some 35 years ago was a very Swedish way of making everyone happy and satisfied. (Also – until his sad passing away a couple of years ago the distinguished Swedish professor Gösta Bruce was President of the IPA).

  52. des von glanzende-overduidelijkheid says

    On the other hand, is there a more satisfying thing to say – in any language – than “hook-top heng”?

  53. I can articulate a simultaneous retroflex-velar fricative okay, or even an alveolar-velar fricative, as these involve the tongue tip. A true palatal articulation defeats me, however.

  54. Bathrobe says

    I recently accosted a young lady speaking an unidentified language with her friend on the subway. It was Tibetan.

  55. marie-lucie says

    JC: To prevent the tip of your tongue from interfering with your attempts at palatal articulation, stick it firmly behind your lower teeth. You will still be able to practice a variety of palatals, velars, uvulars, whether fricatives or stops. I am not sure what Swedish does, but German palatal “ich” and velar “ach” both fall in that non-tip category. But you don’t need to hold your tongue tip in that position once you are familiar with the feeling of what the rest of the tongue is doing.

  56. M-L, you are close enough to both of them to distinguish them, the way an Italian will immediately distinguish between Castilian and Portuguese.

    Stefan,
    As for Danish I don’t believe the real obstacle to be the glottal stop, ‘stød’ Instead it’s a very far pushed reduction of both consonants and syllables.”

    Maybe, but not necessarily. I think that affects Swedish or Norwegian speakers most. My Swedish is quite rusty but it is still my reference pooint for Scandinavian laguages. I remember very distinctly a creepy experience I had on an SAS flight. The pilot came on and did his liitle introduction, first in Danish and then in English. I had an immediate visual of a stroke victim piloting the plane, and let me tell you, it was not pleasant. And it is exactly that mumbling reduction you speak of that was responsible.

    On the other hand it is exactly that reduction that makes Shanghainese and all those Wu dialects sound “soft” compared to Mandarin. It is not only a reduction in consonants but a thorough-going simplification of the vowel and tone phonology too. Shanghai sounds like Mandarin with the bones taken out, and it’s a beautiful effect.

  57. m-l: I meant that a true palatal-velar simultaneous fricative, as [ɧ] is supposed to be, defeats me. That is because it would be simultaneously lamino-alveolar and lamino-velar, which I can’t do in any way that would make the double frication distinct. But pronouncing an apical fricative and a laminar fricative at the same time is possible for me. The further apart the points of articulation are, the easier: a simultaneous (apico-)interdental and (lamino-)velar fricative, for example, is not all that hard.

    When I try the stop equivalent of [ɧ], coarticulating [k] and [c], it just sounds like [c].

  58. @David Eddyshaw

    No, American. Hedd Wynn is about the only exposure I’ve ever had to Welsh speakers. So I just checked out some videos on YouTube. I can hear the individual elements that are different from AmE or BrE, but it still sounds like nothing. Like, it doesn’t have a foreign flavor.

  59. marie-lucie says

    JC, sorry for the misunderstanding. I am not familiar with those doubly-articulated fricatives.

  60. Etienne said In my experience {Czech} is the least-Slavic sounding Slavic language,

    How are you defining “Slavic”? To my ears, Czech just sounds like an odd dialect of Polish and is quite obviously Slavic. I would think Croatian would sound the least “Slavic” to non-Slavic speakers – some dialects are phonologically very close to Italian.

  61. Vanya: I guess what I mean is that, in my (quite limited, compared to other hatter!) experience of hearing Slavic languages spoken, Czech seems to me to be the “odd man out”, and strikes me acoustically as a kind of generic “vanilla” European language. Israeli Hebrew often gives me a similar impression, and indeed this is another candidate for the mystery language John Cowan heard.

  62. “m-l: I meant that a true palatal-velar simultaneous fricative, as [ɧ] is supposed to be, defeats me.”

    John, the friction is between your barely parted teeth and there is some kind of raisng of the sides of the tongue. English speakers often hear it as a voicless -w, except that there’s no rounding. It’s not a hard sound to learn if you hear it directly from a speaker of Swedish.

  63. Hmm, seems like my comment got lost. In short, I ruled out Hebrew because in the Hebrew conversations I have heard in NYC, the listening party is saying “ken … ken … ken …” (Hebrew for ‘yes, right, correct’) more or less throughout. Here the listener (most of the time the redhead) was pretty silent.

  64. MR,
    “I can hear the individual elements that are different from AmE or BrE, but it still sounds like nothing. Like, it doesn’t have a foreign flavor.”

    Considering that something like Welsh was probably the substratum of a lot of British dialects that gave rise to Modern English, especially the southwestern dialects that already sound so American, that shouldn’t be surprising.

  65. Stefan Holm says

    Jim, John, Marie-Lucie et. al.: If a native may attempt to sort things out there are in addition to the unproblematic ’s’ and ’f’ three voiceless fricatives in Swedish:

    [ʃ]: A supradental/ palatal ‘s’, pretty close to the ‘sh’ in English she, shoe, shall (maybe slightly more back/palatal) and used in the spelling combination ‘rs’ (like Sean Connery in ‘of course’ etc.). It corresponds to the ‘thick’, i.e. supradental, pronunciation of ‘rd’, ‘rl’, ‘rn’ and ‘rt’, heard even in dialects of English.

    [ç]: A palatal, non-sibilant, unrounded fricative – or easier: the initial sound in English ‘chat’, ‘cheese’ without the initial ‘t’. It’s used word initially with the spellings ‘kj-‘ and ‘tj’ (kjol=skirt, tjock=thick) and ‘k-‘ preceding a front vowel (e, i, y, ä, ö: kemi=chemistry, Kina=China, kyss=kiss, källare=cellar, köl=keel). You can think of it approximately as a voiceless [j] in year, you, yell.

    [x]: A velar (and the controversial one). This has been very productive in Swedish and can be spelled in +40 different ways. The inherited ones are the word initial ‘sj-‘ (själv=self, sjuk=sick), ‘stj-‘ (stjärna=star), ‘skj-‘ (skjuta=shoot) and ‘sk-‘ preceding a front vowel (skepp=ship, skina=shine, sky=sky, skämmas=be ashamed, sköld=shield). It has also attracted loanwords, particularly from French, like ‘-ti-‘, ’-si-’, ‘-ssi-’, ‘g-’, ‘j-’, ‘ch-’ (station, pension, mission, genant=embarassing, journalist, charm) and also from other languages (schäfer=German shepherd, Shanghai, Charkov).

    The problem, as I’ve mentioned earlier, is that in some dia- and sociolects the [x] is replaced by the [ʃ] or by something in between. After all, since we are not clones, we all pronounce individually, otherwise it wouldn’t be so easy for us to recognize people solely by their voice. But this has lead to the idea of the ‘co-articulated’ consonant [ɧ], which I simply don’t think exists. Even the IPA adds the comment ‘disputed’ to it. The very term ‘co-articulated’ seems redundant to me. It’s also sometimes used about English ‘w’ and the vowels/diphtongs in French nuit (night) and Swedish ny (new). For what purpose?

  66. [ʃ]: – What you describe is a retroflex sibilant. e.g. “fors” eller “färsk”.

    ” But this has lead to the idea of the ‘co-articulated’ consonant [ɧ], which I simply don’t think exists. Even the IPA adds the comment ‘disputed’ to it. The very term ‘co-articulated’ seems redundant to me. It’s also sometimes used about English ‘w’ and the vowels/diphtongs in French nuit (night) and Swedish ny (new). For what purpose?”

    I agree with you on every one of those examples. The idea that “w” is co-articulated sounds like it is a non-Engish speaker who is just having difficulty with the sound, and the same is probably true of “y” in Swedish. English “r” on the other hand really is co-articulated.

  67. /w/ is definitely coarticulated: it’s a velar approximant with labialization. The tongue position is almost that of [ç], just pulled down a little to be an approximant rather than a fricative, and the labial component is obvious. The native speaker has spoken.

  68. Stefan Holm says

    Am I missing something, John, in the very definition of the word coarticulated? Doesn’t just every human sound include at least some state of your (1) vocal chords (2) tongue (3) lips and (4) whether or not breath is allowed to pass through your nose? In that sense all sounds are coarticulated. My objection concerned the suggested ‘simultaneous’ pronunciation of [ʃ] and [x], where the difference (at least in Swedish) is only about the position of the tongue, which can’t be in two states simultaneously. And what would be so special about ‘w’, essentially a very short [u]?

    And this applies, Jim, even to your comment upon the English ‘r’. It’s roughly a voiced variety of [ʃ]. Compare ‘row’ and ‘show’: The tongue position is basically the same (slightly more back in ‘row’) In ‘row’ you activate your vocal chords and in ‘show’ you allow a little less air to pass through the space between the tip of your tongue and the area above your upper teeth. (And again: the final ‘w’ in both words is realized as a very short [u]).

    So ‘coarticulated’ – I still don’t get it.

  69. David Marjanović says

    Personally I’ve always found (standard) German one of the more euphonious-sounding languages.

    Which kind? 🙂 Except for the existence of a stage pronunciation, Standard German pronunciation has never been standardized. It’s more uniform now than 100 years ago, unsurprisingly – but there are still differences in just the kinds of things people tend to notice as “harsh”, “strong”, “guttural”, “melodious”, “dark” and the like: presence/absence of aspiration, voiced plosives, syllable-final fortition, (moderately) long consonants, [ɑ], [χ], [ɵ]… not to mention Swiss intonation, where the stressed syllable of each word gets a lower pitch by default, the next syllable has a high pitch, and the one after may have a falling one.

    In general, the sound system of German is a lot more like the English one than the French or Italian is (as witness the fact that Germans find it a lot easier to acquire a good English accent than French speakers do, by and large.)

    I’d say the sound systems of French and English are about as far apart as is possible in Europe, and German is in the middle. I find French a bit easier to pronounce than English.

    (Part of that is cheating, though – my dialect happens to have nasal vowels, even though not the same set as French.)

    More like a passionate but somewhat abusive lover. Beautiful language, but doesn’t comb well into poetic rhythms. The Poles seem to succeed at poetry despite their language, not because of it.

    Maybe what’s going on is the absence of vowel length in Polish? While it’s not phonemic in Russian either, stressed vowels are (all else being equal) distinctly longer than unstressed ones in Russian, while in Polish there’s no difference.

    Perhaps similarly, hexameter has never caught on in English or German, but is said to be common in Hungarian poetry because vowel length is phonemic in Hungarian while stress placement is not.

    It’s funny how Russian palatalized consonants and Bantu African ones feel so different to me.

    They’re simply not the same thing. Russian has palatalized labial, palatalized alveolar and palatalized velar consonants; Bantu languages have palatal consonants that occupy a separate place of primary articulation. (…At minimum, that’s the case with the [ɲ] in matswenyeho in the South African anthem.)

    In Europe, palatal plosives ([c ɟ]) occur in Hungarian (ty, gy) and Latvian (ķ, ģ).

    solid, softly rounded cobbles that have been through four glaciations

    More like seventeen…

    Czech seems to me to be the “odd man out”, and strikes me acoustically as a kind of generic “vanilla” European language

    Most kinds of Czech take their vowel length extremely seriously. It’s phonemic and has no relation to stress whatsoever.

    The idea that “w” is co-articulated sounds like it is a non-Engish speaker who is just having difficulty with the sound

    [w] = [ɰʷ], except that the latter transcription makes a claim about which articulation is primary and which secondary, which doesn’t make sense for an approximant. I guess you could also transcribe it as [βʷˠ] (bilabial approximant, rounded, velarized), making the opposite and equally useless claim.

    Reportedly, the Japanese w is a rounded bilabial approximant without a velar component (that would be [βʷ] then, I guess). This seems to agree with what little Japanese I’ve heard, but I haven’t heard enough to tell for sure.

    the position of the tongue, which can’t be in two states simultaneously

    The same part of the tongue can’t be in two places at the same time. Different parts of the tongue are usually under no such restriction. I have no trouble articulating [x] and [ʂ] at the same time, and the resulting sound is easy to distinguish from both – the question is whether it actually exists in any sound system.

    And this applies, Jim, even to your comment upon the English ‘r’.

    No. The English /r/ is quite strongly rounded: [ɹʷ ~ ɻʷ]. There’s a reason for “be vewy, vewy quiet – I’m hunting wabbits”.

    The English /ʃ/, as it happens, is also rounded, but much less so; a bit more than the French one, less than the German one*, and much less than the actual [ʃʷ] that appears to exist in Setswana and various Caucasian languages. To find a completely unrounded [ʃ], which I found surprisingly different to articulate at first, you have to go all the way to Navajo.

    * The most strongly lip-rounded sound in the entire language, at least where I come from.

    And again: the final ‘w’ in both words is realized as a very short [u]

    There is no /w/ in row or show; both end in a diphthong that ends in [ʊ].

  70. marie-lucie says

    My parents, having spent the war years under German occupation, associated the sounds of German from soldiers barking orders at the French population. If French people at the time learned one German word, it was “raouss” (Raus!, ‘[Come] out!’), a dreaded word which could presage arrestation and worse. Years later, my father and I had the opportunity to go Berlin for a week with a small group, of which only one person (not me) was reasonably fluent in German. My father quickly learned the numbers and how to order a beer. After this experience of hearing German in its normal, everyday, peaceful environment, he had the opportunity to take German courses, and later he and my mother took several trips to Germany and Austria (to the Salzburg festival among other things). My mother had started to learn German too and was pleasantly surprised that the language could sound gentle. She was charmed by the sounds of ich liebe dich.

  71. Am I missing something, John, in the very definition of the word coarticulated? Doesn’t just every human sound include at least some state of your (1) vocal chords (2) tongue (3) lips and (4) whether or not breath is allowed to pass through your nose?

    Yes and yes, I think. We call a sound coarticulated when two articulation positions are necessary to producing it. I can articulate /k/ in English with almost any tongue-tip position and not be misheard, but /w/ requires both a velar and a labial gesture to get right.

  72. More on Swedish fricatives here.

  73. David Marjanović says

    …Wow, I wrote “different” instead of “difficult”.

    Anyway, I just watched the first 7 minutes of Hedd Wyn. I find the intonation completely unremarkable. Practically everything else, though, does not sound like English (outside of Scotland); it’s quite unlike the video of a rant in Irish-in-the-English-sound-system I once heard.

  74. gwenllian says

    How are you defining “Slavic”? To my ears, Czech just sounds like an odd dialect of Polish and is quite obviously Slavic. I would think Croatian would sound the least “Slavic” to non-Slavic speakers – some dialects are phonologically very close to Italian.

    Yes, I think the mystery language could be a South Slavic variety. Much to my surprise, I’ve had English-speaking Americans mistakenly guess what I’d been speaking as Italian, and on another occasion I was mistaken for Greek. I haven’t experienced that from English speakers from the UK, though. The couple of times people in the UK made an assumption about the language it was to ask me if I’d prefer to speak to a Polish-speaking employee.

    Regarding Welsh, I’ve always found it striking how much unlike English it sounds, given the absolutely overwhelming influence English exerts on it. So much of modern Welsh consists of one calque after another, but the sound doesn’t seem to have been influenced at all. As David Eddyshaw notes, the Welsh sound somewhat foreign in English as well (though not so much in the Scouse-influenced NE). It seems to remind other Brits of Indian English.

    I first heard Turkish some two decades ago on a visit to Istanbul. What a wonderful surprise! Especially so when contrasted with the rest of the Levant and its gatto-gargling-a-hairball gutturals.

    It’s the opposite for me. I find the sound of Hebrew and Arabic (though I haven’t had extensive exposure to most of its varieties) very pleasant. Turkish not so much.

  75. My mother had started to learn German too and was pleasantly surprised that the language could sound gentle.

    Moi aussi!

  76. GeorgeW says

    “I find the sound of Hebrew and Arabic (though I haven’t had extensive exposure to most of its varieties) very pleasant.”

    While I am interested in these languages, I don’t find them paticularly pleasant. But, not unpleasant as well. I have had a lot of exposure to Arabic, particularly in the Gulf and Egypt. I do like the intonation of Egyptian Arabic. It seems more rhythmic to me than other varieties. I also prefer Lebanese Arabic to that of the Gulf.

  77. David Marjanović says

    I’m thinking about the Czech influence on Viennese (massive immigration and very quick assimilation in the 19th century). The intonation is the same; when the Czech speak, I feel at home even if I don’t understand a word. The Viennese /ʃ/ is identical to the Czech one, less rounded and more fronted than elsewhere in German; and in and around the 12th district (Meidling), /l/ is outright pharyngealized, which clearly has to be blamed on the velarized Czech one. All other influence has been on vocabulary and a bit of grammar. Czech is full of voiced obstruents, but has word-final fortition; Viennese lacks both. Czech has phonemic vowel length; Viennese has phonemic consonant length instead. Czech distinguishes the usual 5 vowel qualities; Viennese has 4 rounded front vowels alone, but lacks [ɔ] despite having something quite similar.

  78. Is there an alternation between single vowels with double consonants and double vowels with single consonants, as in Standard Italian?

  79. George Gibbard says

    The English /r/ is quite strongly rounded: [ɹʷ ~ ɻʷ]. There’s a reason for “be vewy, vewy quiet – I’m hunting wabbits”.

    Not all English varieties are created alike. I (Michigan) am a ‘buncher’, meaning I articulate [ɹ] with the ‘front’ of the tongue (meaning the front of the dorsum of the tongue — behind the blade of the tongue — not the tip of the tongue) raised behind the alveolar ridge, and I don’t round it at all, whether in a syllable onset or a coda. I gee up with the impression that rounding of this sound seems, well, infantile.

    The English /ʃ/, as it happens, is also rounded, but much less so; a bit more than the French one, less than the German one*, and much less than the actual [ʃʷ] that appears to exist in Setswana and various Caucasian languages.

    I’m certain I don’t round English [ʃ] either.

    There is no /w/ in row or show; both end in a diphthong that ends in [ʊ].

    Put me down for [ɹo] and [ʃo] (also normal n nearby Canada; but I do hear even [ɐʊ] from the younger folks).

  80. George Gibbard says

    that was supposed to be “grew up”, I don’t know how “gee up” happened.

  81. George Gibbard says

    So why do I say ‘infantile’? I guess “bunched” American [ɹ], like Czech [r̝], is one of those sounds it takes children an especially long time to master. I think I frequently encounter children who are not able to say [ɹ] like adults, and who try to use lip rounding to achieve the low F3 of the sound, but who presumably later learn to say it like adults. Elmer Fudd is a character who never acquired adult pronunciation.

  82. George Gibbard says

    Sorry, I mean Bugs Bunny.

  83. How is that? Elmer Fudd is the one who hunted wabbits.

    On the topic of English [ʃ], my impression is that increased rounding is a subtle marker of posh speech on both sides of the Atlantic.

  84. I’m thinking about the Czech influence on Viennese (massive immigration and very quick assimilation in the 19th century).

    Yes indeed, when I first got to Vienna I assumed some native Viennese I met must be Czech or Slovak immigrants who just spoke very fluent German. Then a few years later it took me a while to realize the owner of our Stammlokal actually was indeed a Czech immigrant with very good German, for a long time I assumed she was Viennese.

  85. Fascinating!

  86. Lip-rounding in English /r/ is most perceptible word-initially, as a result of the merger of older English /wr/ and /r/. This is usually thought to be /wr/ > /r/, but it’s more like /r/ > /wr/.

  87. George Gibbard says

    I accept that labialization of English /r/ is widespread, but I contend that (1) it is not the norm in Michigan, and if I stand in the mirror and say “Rhapsody” or “federal regulations” I don’t observe any; (2) it does, however, occur in child speech, even where it is not common in adult speech, because it is sort of acoustically similar: lip rounding lowers F2 as well as F3, while /r/ for me has lowered F3 without lowered F2. (whereas /l/ has high F3, while all other sounds in American English can be identified without paying attention to F3.)

  88. David Marjanović says

    Is there an alternation between single vowels with double consonants and double vowels with single consonants, as in Standard Italian?

    There is in Austrian Standard German, except that the long fricatives* happily occur behind long vowels and diphthongs as well, diphthongs occur in front of consonant clusters, and various loans retain a short stressed vowel in front of a short consonant (words with bb, dd, gg**, but also Kanada for example); those always sound “overly hasty” to me.

    In the Central Bavarian = Danube Bavarian dialects, including Viennese (and mine), vowels are longer when they’re stressed and when all else is equal, which it never is; I do think long consonants are a factor, but so are emotion/intonation and the fact that German, as a whole, is stress-timed like English – which also has a noticeable effect on vowel length in the standard language.***

    Eastern Austrian dialects, again including Viennese, have further monophthongized the opening and closing diphthongs (leaving the centering ones); that makes it even easier to pronounce them as fully short.

    * By and large from the High German consonant shift, which turned short plosives into long fricatives.
    ** /bː dː gː/ exist only across morpheme boundaries; the High German consonant shift got rid of them otherwise.
    *** Once I went home by train through the Czech Republic. Trains from north of Vienna loop around the city and end at what was at the time the south station. The convention in Europe is that the names of stations are never translated, so the Czech conductor announced Wien Südbahnhof as the destination. All vowels in this are phonemically long, so the conductor pronounced it [ˌviːnˈziːdbäːnˌɦɔːf]. Now, that’s of course a complete adaptation to the Czech sound system, plus the [z] which is foreign to Austria but of course used by most native speakers of German; that’s not remarkable. What really struck me is how long it took to say. Between the stress-timing and the fact that fully unstressed vowels have to be fully short in German, a native pronunciation would be more like [ˌviːnˈsyˑd̥b̥anˌhoˑf].

    I’m certain I don’t round English [ʃ] either.

    There’s a completely unrounded [ʃ] in one of the soundfiles on this page. It’s unlike anything I’ve encountered in English or even French. What do you think?

    I think I frequently encounter children who are not able to say [ɹ] like adults, and who try to use lip rounding to achieve the low F3 of the sound, but who presumably later learn to say it like adults. Elmer Fudd is a character who never acquired adult pronunciation.

    Yes. I was trying to talk about why such children end up at /w/ rather than, say, /l/ (…which, incidentally, must be where Molly, Sally and Hal come from; that took me many years to notice).

    That said, I have no idea about Michigan. And unfortunately I have no experience with trying to identify formants either.

    if I stand in the mirror and say “Rhapsody” or “federal regulations” I don’t observe any

    Rounding isn’t necessarily visible. The most visibly rounded sound in my kinds of German is /ʃ/, despite the existence of a whole list of phonemically rounded vowels.

  89. George Gibbard says

    Actually the Navajo unrounded [ʃ] strikes me as unremarkable. I think in the southern United States you will find a more palatalized [ʃ].

    Elmer Fudd may of course be imitating a dialect where adults round [ɹ], but he could also be going for low F3 in a dialect where they don’t (whereas as I say, /l/ has high F3 which must be due to “velarization”, often the only reflex of /l/ in my pronunciation, which I understand we are now supposed to write as [ʶ]). My impressions of this come from when I was fooling around in the UCSD phonetics lab, recording myself and looking at spectrograms in PRAAT. I no longer have these recordings, or a microphone on my computer, but I would think that other people could do the same thing. I do recall that in my informant’s recordings of Persian, /r/ and /l/ had nothing that I could see in common with my English consonants I transcribe the same way, i.e., no discernible influence on formants. [which was just autocorrected to “from ants”]

    what is rounding if it is not visible? the main effect of rounding is lowering F2, which is the same as the acoustic effect of backness (although rounding is supposed to have a somewhat different effect on F3). Similarly, the main effect of vowel height is lowering F1, which is the same as the acoustic effect of advanced tongue root. And I believe a vowel distinction that has originated as rounding may be realized as backness, or original ATR may be realized as height, without anyone much noticing.

  90. George Gibbard says

    Downolaod PRAAT for free: http://www.fon.hum.uva.nl/praat/

  91. George Gibbard says

    As for Molly, Sally I suppose it’s possible that these are so old that what’s going on is that children couldn’t produce IPA [ɾ] (the alveolar tap) and produced IPA [l] instead.

  92. David Marjanović says

    Thanks for the link to PRAAT, I’ll look into it! I have a microphone 🙂

    have further monophthongized the opening and closing diphthongs

    …except there weren’t any opening ones to begin with. Oops. I mean the cognates of Standard German ei (both), au (both) and eu, all of them closing.

    “velarization”, often the only reflex of /l/ in my pronunciation, which I understand we are now supposed to write as [ʶ]

    That would be uvularization; but I have heard a native speaker say midline in a way that I heard [ʀ] instead of [l]. (It may well have been an approximant rather than a one-contact trill.)

    what is rounding if it is not visible?

    A lot can be outsourced from the lips to the tongue and perhaps other mouthparts.

    I think rounding is generally visible on my lips – in that I simply open my mouth less far than for unrounded vowels, without making any effort to bring the corners of my mouth closer together…

  93. GeorgeW says

    Isn’t “rounding” by definition a rounding of the lips?

  94. Yes, that confuses me too.

  95. Christsophe says

    @Stefan Holm

    ‘beuaty’?

    Beware of the Vowelkyries!

  96. J. W. Brewer says

    For some reason this evokes memories of my 9th-grade German teacher barking “Runden Sie die Lippen!”* at the class as we tried to figure out how to utter ü, which was not in our Anglophone set of vowels. I suppose it was arguably redundant because there was nothing else relevant we could have rounded, but we were sufficiently lacking in phonological sophistication it probably didn’t hurt to make it explicit.

    *She used “du” with us one-on-one because we were kids, but not being a native speaker I think she may not have been sufficiently comfortable in the familiar second-person plural pronouns to use them without stopping to think about it so she defaulted to the polite forms.

  97. *She used “du” with us one-on-one because we were kids, but not being a native speaker I think she may not have been sufficiently comfortable in the familiar second-person plural pronouns to use them without stopping to think about it so she defaulted to the polite forms.

    Interesting that’s actually the opposite of what many Germans do – while the official standard is formal pronoun in singular and plural in formal situations, many Germans use the formal Sie in the singular and the informal Ihr in the plural. I’ve never encountered informal du for the singular and formal Sie in the plural, like your teacher did.

  98. J. W. Brewer says

    Well, as I mentioned, she was not a native speaker, and I really don’t know how well she would have fared in conversation with a native speaker who didn’t go into full-dealing-with-foreigners mode (speaking slowly, pronouncing with exaggerated clarity, avoiding idioms and slang, etc.). But since the public schools I went to growing up lamentably no longer offer instruction in German at all, I don’t want to quibble about how what they offered 35 years ago might have been imperfect. I do think there may have been a self-reinforcing loop — I think we students learned the ihr-forms in the context of the memorized “tables” of conjugations/declensions but used them less than the du-forms or Sie-forms in actual sentences or bits of stilted dialogue.

  99. “full-dealing-with-foreigners mode (speaking slowly, pronouncing with exaggerated clarity, avoiding idioms and slang, etc.).”

    And louder.

  100. David Marjanović says

    The term “endolabial rounding” seems to have disappeared from Wikipedia except for a link to Wiktionary.

    many Germans use the formal Sie in the singular and the informal Ihr in the plural.

    I haven’t noticed. I have noticed university professors doing this (in Vienna) – apparently to make clear that they meant the plural, because the polite form doesn’t distinguish singular and plural.

  101. the polite form doesn’t distinguish singular and plural.

    What form of the verb does it take?

  102. 3pl

  103. Danish copied the system of using the (original) 3pl pronoun as formal 2sg and 2pl, but had the foresight to first make verb declension independent of number and person.

  104. I haven’t noticed. I have noticed university professors doing this (in Vienna) – apparently to make clear that they meant the plural, because the polite form doesn’t distinguish singular and plural.
    Interesting. That’s not the sociological environment where I’m used to hear this in Northern / Western Germany; here, it’s mostly among people outside of academia – e.g. colleagues in an office who would use Sie with each other individually, but use Ihr speaking in a group, or people adressing an unknown group (so you’d expect Sie) at a pub or restaurant table or on the street with Ihr. The reason may be the same, that the singular and plural are not distinguished for the polite form; it also may be a remainder of previous use of the 2nd plural as a polite form.

  105. David Marjanović says

    people adressing an unknown group (so you’d expect Sie) at a pub or restaurant table or on the street with Ihr

    Inconceivable! 🙂

    Unless of course if everyone is under… 35 maybe, and not dressed too formal. The politeness horizon isn’t done receding yet.

  106. Unless of course if everyone is under… 35 maybe, and not dressed too formal. The politeness horizon isn’t done receding yet.
    Well, yes, I should have foreseen objections like this, and I should have made it more clear that I’m talking about situations where, at a restaurant or pub, the waiters address each single guest they talk to with Sie, but uses Ihr when talking to those same guests as a group. That’s different to the general receding of formal Sie and its reduced use among young people.

  107. David Marjanović says

    situations where, at a restaurant or pub, the waiters address each single guest they talk to with Sie, but uses Ihr when talking to those same guests as a group

    …That matches the university situation I mentioned, but a restaurant or pub? That’s rather stunning.

  108. marie-lucie says

    A few years ago the King of Spain addressed an audience of writers (perhaps prize winners) using “vosotros” rather than “Ustedes” (perhaps I read about it in a Spanish newspaper or magazine, which quoted part of the King’s speech). I doubt that he would have addressed individual writers with “tu” rather than “Usted”, but I could be wrong, as I remember reading that the Spanish upper class is fond of “tu”, leaving “Usted” to lower mortals. I wonder if this case is similar to the German one about how waiters address customers individually with “Sie” but as a group with “Ihr”.

  109. As far as I know, the King says tu to everyone, including (famously) his fellow heads of state. See “¿Por qué no te callas?”. Granted, that could have been the T of insult.

  110. As far as I know, the King says tu to everyone

    As would I, if I were a king. What’s the point of being king if you can’t say tu to everyone?

  111. I would say rather that it is one of the few consolations afforded to someone who winds up having to be a king.

  112. Well, that could be regarded as just a rephrasing of what I said.

  113. You tend to concentrate on the sound of any language only if, and as long as, you cannot speak it. Once you start to learn it and its words become meaningful, you stop “hearing” the way it sounds. During a trip to South Africa, I happened to overhear a conversation in Zulu. I found the sound of it as melodious as Italian. I’ve met several foreigners who, on hearing me talk thought I spoke Spanish, I guess this is a good thing, I like the “clear” sound of Spanish myself. Not that it matters what others say, my language is my language and a part of myself, how could it be ever compared to any other?

  114. marie-lucie says

    Ariadne: You tend to concentrate on the sound of any language only if, and as long as, you cannot speak it. Once you start to learn it and its words become meaningful, you stop “hearing” the way it sounds.

    Exactly!

  115. Alessandro says

    Italian as spoken in Western Lombardy and Milan is the most beautiful language. This is the Italian you hear on advertising spots in TV or in american movies overdubbed in Italian.
    Here are examples on how to speak the beautiful version of Italian:

    https://www.youtube.com/watch?v=5ePx-fe2_jE

    https://www.youtube.com/watch?v=H_81jUy_Rd4

    https://www.youtube.com/watch?v=lYZAmq3MED0

    https://www.youtube.com/watch?v=P-6Eul-6dBo

  116. David Marjanović says

    You’re aware that different people have different tastes, right? I find nothing particularly beautiful in any of these, and on top of that I don’t like Patrizia Mottola’s voice. 😐

  117. Stephen C. Carlson says

    Aww, I was really hoping that Arnold would be dubbed into Italian with an Austrian accent.

  118. What Uralic languages sound like:

    https://www.youtube.com/watch?v=SjKfijHMWyg

  119. John Cowan says

    I’ve met several foreigners who, on hearing me talk thought I spoke Spanish, I guess this is a good thing, I like the “clear” sound of Spanish myself.

    There are objective similarities: five simple vowels without length distinctions, stress accent on one of the last three syllables, [θ] and [ð] (though the latter is phonemic in Greek and allophonic in Spanish), lenition of voiced stops, final consonants /s/, /r/, /n/ (and also /l/, /d/ in Spanish). Albanian belongs here too, of course.

    And of course there’s the vocabulary:

    Armonia neurotica en el microcosmο de la metropoli cultura narcisista en una monarquia dogmatica simfonia cacofonica, pandemonium en atmosfera melodia simbolo, melodrama y tragedia. Orgasmo ideologico del barbarismο a la teoria politico dislexico en parodia onirica tirania fantasma, dilema megalomano de un metabolismο retorico sin tesis ni antitesis. Este mi ultimo tango en Atenas tango lloron, que corre por mis venas. Patriota heroicο, tragicο, sistematico hipocrecia paranoica sin dialogo esotericο teatro ironicο,sindicato plasticο y epicentro de la epidemia, quimera, utopia. Energia hyperbole, antidotο democraticο Laberinto critico sin entusiasmo, sin rima musica epidermica en un pentagrama masoquista y la simetria toxica de un epilogo necrologico. Este mi ultimo tango en Atenas tango lloron, que corre por mis venas. Hay oasis aromatico, paralelo, fisiologico profeta enigmatico, fenomeno cronico y ortodoxo sin racismos ni extremismos, sin tabues etnicos en lirica extasis sus praxis es el melodico y fantastico antropo. —Daniel Armanto

  120. What Uralic languages sound like

    Some languages sound remarkably Russian, others not at all. Some Sami languages sound quite similar to Finnish, others barely so.

    I haven’t listened to all of them.

  121. I just found this recording of Tahitian. It’s a voiceover, read from a prepared translation of a Mormon sermon. Not my cup of tea, but you can follow the along with the written text. Very noticeable is the distinct Tahitian metathesis CVʔ > CʔV, so you hear lots of t’- and r’- clusters. Not quite ejectives yet, but maybe they’ll get there some day if the language survives.

  122. I haven’t listened to all of them.

    I didn’t know that Mari has borrowed the -rak morpheme to form comparatives.

  123. D J EDDYSHAW says

    Some languages sound remarkably Russian

    I was struck by that too. But I suspect that would have been much less the case if I actually knew Russian: what do our Russians think?

  124. I once heard an Irish speaker who absolutely sounded like someone speaking irish with a thick Russian accent. But I do not know what part of the island she was from:( Obviously, Irish does have many similarities to Russian (broad and slender consonants) but realizations vary.

  125. PlasticPaddy says

    @drasvi
    https://m.youtube.com/watch?v=OV375ffOyMw
    To me the vowels and the l are similar to Slavic (possibly more like Czech than Polish or Russian), but the r is not. Also the intonation is slow and careful. There is another recording with the same poem read in Munster Irish. You can compare, but for me the Munster accent is less rough and more musical sounding.

  126. What Uralic languages sound like

    Finally watched this; my reward for sticking with it was the startling finale of Прауральский язык (Proto-Uralic)! I wonder what that was from?

  127. Did you notice that it was the Lord’s Prayer recited in Votic?

    Izä meпee kumpa õlet taivaiza, pühätü õlkoo nimes sinun, litši tulkoo sinu valtas,
    õlkoo sinun tahtoos nii maaza kui taivaiza; anna meiпee tänänne õkapäivädnee
    leipäni ja anna meile meппee võlkani anteehsi kui i müö anteehsi annamma meппee
    võlgaliizile ja älä saata meitä manituhseesee a päässä meitä pahassa ku sinun on
    valta i võima i slaava igää kõikkee. (Ahlqvist 1874: 20)
    Our Father in heaven, hallowed be your name. Your kingdom come, your will be done,
    on earth, as it is in heaven; give us this day our daily bread, and forgive us our debts,
    as we also have forgiven our debtors and lead us not into temptation, but deliver us
    from evil. For the kingdom, the power, and the glory are yours forever.

    https://www.researchgate.net/publication/301346135_From_bourgeois_to_national_romanticism_Attempts_at_creating_Votic_and_Ingrian_written_standards_in_the_20th_and_21st_century

  128. Some languages sound remarkably Russian

    I was struck by that too. But I suspect that would have been much less the case if I actually knew Russian: what do our Russians think?
    For many of the languages spoken in Russia, it looks like there is a strong influence from Russian prosody. I assume many of these speakers speak mostly Russian in their everyday lifes; for the smaller languages, some may even not be native speakers but revivalists. There’s also some code switching to Russian in individual clips, and loan terminology in what seem to be excerpts from radio or TV news / reports.

  129. Yeah, I kept hearing “Ну, вот…” and the like.

  130. SFReader says

    https://youtu.be/9P77vYgnh-M

    This Moksha song sounds 120% Russian for Russians even though they don’t understand a word.

  131. Yep.
    I don’t know whether anyone has looked into this, but is the prosodic influence all one way (the predominance of Russian in the last century or so made the prosody of Finno-Ugric languages more like Russian) or does the specific prosody of Russian also owe something to substrate influence from assimilated Finno-Ugric languages?

  132. This Moksha song sounds 120% Russian

    It sure does; I kept having the feeling that if I listened a few more times I’d understand it. (I kept hearing хуя…)

  133. @Hans: “does the specific prosody of Russian also owe something to substrate influence from assimilated Finno-Ugric languages?”

    Some Russian historians and ethnographers believed the answer was yes and applied to more than just prosody or accent. But that was a long time ago – pre-1917 – and I doubt there’s evidence to support that view today in its entirety. E.g., Klyuchevsky believed (wrongly) that the ending -ogo pronounced as -ovo was a Finno-Ugric trait.

    On the other hand, the Komi-Permyak lady, while mixing Russian and Komi, speaks Russian in a peculiar way, phonetically and prosodically.

  134. (I kept hearing хуя…) in 90s it was a recognized genre among pop-singers here, writing songs based on misheard foreign lyrics. Cf. “Мальчик хочет в Тамбов” (Bate forte o tambor) etc

  135. Hatten är din! (One of my very first internet memes; Wikipedia.)

  136. It’s not just the prosody. Some vowels sound diphthongized after the Russian fashion, and there’s some palatalized stops, too, I think.

  137. är din! That I frist read as “ad-din”. I failed to misread “[Hatt]en”:(

  138. palatalized stops

    That’s an echt feature of many Uralic languages.

  139. “Мальчик хочет в Тамбов”

    Водки найду!

  140. That’s an echt feature of many Uralic languages.

    I guess I’d had some idea that was true. Still, did palatalization become more prominent somehow in some languages under Russian influence?

  141. Была б баба Люба, она б дала!

    (I do not know the source. Tutti Frutti surely has ‘была б баба Люба” part but the author says nothing about “она б дала”:(. I certainly heard “…она б дала” variant (something like whop /t’ba/ ) but it could be Russian rock’n’roll performers)

  142. “The last of his family who actually spoke Circassian was his grandmother; according to him, it sounded like someone with a bad cough” (DE)

    Could be a number of things from Ubykh (“84 distinct consonants (four of which, however, appear only in loan words).”) to any of these, e.g. this or this one. You likely know that this langauge family is famous for its phonology, but just in case;)

  143. “Мальчик хочет в Тамбов”

    Водки найду!
    I remember people singing А я рыба, я рыба in the 90s.

  144. did palatalization become more prominent somehow in some languages under Russian influence?

    Hard to tell. There are parallel texts in Finno-Ugric languages at the end of Volume 1, both in original scripts and transcription:

    http://languagehat.com/saras-family/#comment-3780868

  145. did palatalization become more prominent somehow in some languages under Russian influence?

    Strongly suspected to be the case for Ludian–Veps and a bit less strongly suspected to be the case for Mordvinic (Erzya, Moksha), which are the only Uralic languages that today have automatic palatalization of ~all consonants that precede front vowels. Native Uralic palatalized consonants tend to be exclusively coronals and/or triggered by front vowels (the latter history is prominent esp. in Eastern Samic, Eastern Finnish+Karelian, Estonian, Livonian). Then again Nenets apparently developed Russian-esque general palatalization independently, followed further by the loss of front/back distinctions in vowels.

  146. Mordvinic {…} Moksha

    I had a look at a picture of a relative who passed away last October and a picture of Mikhail Devyataev and was struck at how much they looked alike. But then, my mother’s village and Torbeyevo are less than 100 km apart as the An-2 flies.

  147. George Gibbard says

    There are a number of problems with my comments here.

    First of all, however an adult speaks is speaking like an adult, whether or not they have the average accent.

    Second, I was ignoring the speech of black speakers in Michigan, who have their own norm or norms of pronunciation.

    Third, I was not even right for white speakers in Michigan, most of whom have some labialization of r at the beginning of a word, though it can be subtle so that I used not to notice it.

    One real difference between white Michigan speech and say, New York, is that r between vowels tends to be treated as a coda in Michigan and lack labialization, where New Yorkers would treat it as an onset and labialize it.

  148. See now The Economist‘s Johnson (Jan 11th 2024), “What is the world’s loveliest language?“:

    Linguists tend to say that all languages are valuable, expressive and complex. They usually attribute negative attitudes to prejudice and politics. That is probably why no one has carefully studied the touchy question of which ones are seen as beautiful or ugly.

    That is until three scholars—Andrey Anikin, Nikolay Aseyev and Niklas Erben Johansson—published their study of 228 languages last year* [Do some languages sound more beautiful than others?]. They hit upon the idea of using an online film about the life of Jesus, which its promoters have recorded in hundreds of languages. Crucially, most recordings had at least five different speakers, as the film has both exposition and dialogue. The team recruited 820 people from three different language groups—Chinese, English and Semitic (Arabic, Hebrew and Maltese speakers)—to listen to clips and rate the languages’ attractiveness.

    What they found was that nearly all the 228 languages were rated strikingly similarly—when certain factors were controlled for (of which more below). On a scale of 1 to 100, all fell between 37 and 43, and most in a bulge between 39 and 42 (see chart 1). The highest-rated? Despite the supposed allure (at least among Anglophones) of French and Italian, it was Tok Pisin, an English creole spoken in Papua New Guinea. The lowest? Chechen. The three language groups broadly coincided in their preferences. But the differences between the best and worst-rated languages were so slight—and the variation among individual raters so great—that no one should be tempted to crown Tok Pisin the world’s prettiest language with any authority.

    Details at the link. Thanks, cuchuflete!

  149. Trond Engen says

    I would think the point of the study would be to understand the concept of beauty as applied to languages rather than any intrinsic beauty of the languages themselves. Might there be a tendency for certain preferences, e.g. for assumed simplicity in reproducing the phonological patterns (in unknown languages)?

Speak Your Mind

*