Green who?

« previous post | next post »

In "Sign of the Times,"  I discussed a sign in a New York shop window written in Russian but ostensibly addressed to Chinese.  Now we have a bilingual sign in Russian and English:

The Russian means "Green Beach"; with a Slavic pronunciation of English, that might come out the same as "Green Bitch," both being pronounced "Green Beech."  Russians seem to have a hard time with the distinction between [i] and [ɪ].

This immediately reminds me of Geoffrey Pullum's "The bitches of Hong Kong."   Apparently Russians aren't the only ones who have a hard time with the [i] – [ɪ] distinction.

As a matter of fact, when teaching English to speakers of various Chinese languages, I've discovered that one of the hardest things — in terms of pronunciation — for them to master is distinguishing among the fine gradations of vowel quality that make such a crucial difference for understanding others and expressing oneself clearly:  pin, pine, pan, pane / pain, pen, peen, pun, poon, pone, peon ('ŏn', pē'ən; also pyūn), paen….

[Hat tip to Eric Kinzel]


  1. Skullturf Q. Beavispants said,

    April 3, 2009 @ 4:59 pm

    Do the Chinese languages lack a distinction between voiced and unvoiced final consonants? More than once, I've seen crustaceans listed on Chinese menus not as "crab", but as "crap".

  2. Mike Aubrey said,

    April 3, 2009 @ 5:18 pm

    Its so true. Russian simply does not have the eleven phonemic vowels that English does. [i] – [ɪ] in Russian function together as allomorphs in a highly complex system of stress and vowel reduction.

    This is probably the best thing I've seen all day – all week!

  3. marie-lucie said,

    April 3, 2009 @ 5:27 pm

    The vowel distinction is also a difficult one for French, Spanish or Italian speakers, among others.

    The word for "beach" in Russian seems to be a borrowing from French "plage", no?

  4. Russian Phonology & English Translation « ΕΝ ΕΦΕΣΩ said,

    April 3, 2009 @ 5:29 pm

    […] Green who? […]

  5. Kaoru said,

    April 3, 2009 @ 5:39 pm


    My French Phonology professor tells me this is an issue for native French speakers too. When she first moved here she got funny looks for telling people how much she adored the "bitch," meaning "beach".

  6. Jim said,

    April 3, 2009 @ 5:50 pm

    "Do the Chinese languages lack a distinction between voiced and unvoiced final consonants?"

    Cantonese has only unvoiced stop finals, so Cantonese speakers typically have a hard time learning to distinguish them in English. Mandarin and Shanghai don't, and I think they just learn the finals as appended intials, and they come pretty close to the English distinctions.

    "pin, pine, pan, pane / pain, pen, peen, pun, poon, pone, peon (pē'ŏn', pē'ən; also pyūn), paen…."

    They can approximate these if they use the same strategy – append an -n to syllables with diphthongs. In Mandarin a syllable can have a final either in a diphthong or ending in a nasal, but no final has both. That's true for Shanghai, but I don't know the situation in the other languages. It's probably pretty much the same.

    Mandarin has all the diphthongs mentioned above, but 'pin' and 'peen' are going to sound the same.

  7. Rodrigo said,

    April 3, 2009 @ 5:52 pm

    The brazilians apparently have the same problem. In the brazilian band CSS's "Meeting Paris Hilton":

    I went to the beach
    the bitch was so hot
    she came to me and said
    do you like the beach, bitch?
    I said back
    I wanna take you home, bitch
    'Cos I wanna treat you good, bitch
    What do you think of it, bitch?
    The singer pronounces "bitch" and "beach" identically.

  8. Daniel von Brighoff said,

    April 3, 2009 @ 6:18 pm

    One of the Russian émigrés I knew in college used to ask his friends if they wanted to accompany him the "Wisconsin Street Bitch". Fortunately, he was able to look back on this and laugh rather than cringe with embarrassment.

    Even speakers who have mastered the phonemic contrasts between tense and lax vowels can still stumble on the subphonemic distinctions between short and long ones. I struggled for years to understand why it sounded to me like a German friend was unvoicing his final stops even though I could clearly distinguish a contrast between such minimal pairs such as bid and bit. Finally, I realised that the [ɪ] was of equal length in both cases, not phonetically lengthened before /d/ as is typical among native speakers. (It immediately made me wonder if I was making the opposite error in my German.)

  9. Morten Jonsson said,

    April 3, 2009 @ 6:52 pm

    Native English speakers merge those sounds too, of course. Witness breeches and britches, creeks and cricks. And the residents of Beech Creek, Pennsylvania, might spell the name of the town that way, but that's not how they say it.

  10. phleabo said,

    April 3, 2009 @ 8:00 pm

    Spasibo bolshoiye; that was mne smeshnoye. I read the Russian part and didn't think anything of it – took me a couple seconds to register the English for how it was spelled.

  11. Amy Stoller said,

    April 3, 2009 @ 8:20 pm

    As a dialect coach in the US, I've had to work on the i / ɪ distinction with clients whose first languages were French, Italian, Spanish, Japanese, and Korean, as well as Russian, though to my ear, it does seem to have a particular je ne sais quoi in Russian accents.

    A similar difficulty occurs with the distinction between ʊ and u. I haven't seen any good street or store signs to serve as illustrations, more's the pity.

  12. D.O. said,

    April 3, 2009 @ 8:30 pm

    Where this photo of the sign comes from? It might very well be a joke. Very funny nonetheless. It is strange that anyone so ignorant of the beach/bitch difference will spell the last word correctly. If a Russian would play it "by the ear" the spelling might be "bich". "Bitch" would be quite an educated guess. And if someone look it up in dictionary…

  13. chris said,

    April 3, 2009 @ 9:34 pm

    @ Daniel von Brighoff:

    I'm really confused by your comments about your German friend's pronunciation. Are you saying that he was pronouncing words like bid and bit correctly except that he made no distinction in the length of the vowel, and that this is different from the way native English speakers pronounce these words? Or are you saying that his problem with the voicing of the final consonants was due to the fact that there is no difference in the length of the vowel in English, and that he would have had an easier time of it if the vowel had been phonetically lengthened before /d/ as it always is for native German speakers??

    Either way the interpretation seems incorrect. In my dialect of English (non-American) there is absolutely no difference in the length of the vowel in bid and bit.

    [(myl) Where are you from? And is your opinion about your vowel lengths impressionistic, or have you measured? In either case, I'd be interested to see a recording of some appropriately controlled materials, because in all the varieties of English that I've ever seen, or seen reported, there's a reliable and quite large difference in vowel duration before voiced vs. voiceless consonants, especially for stressed syllables in word-final position. ]

    Vowel length has no phonemic significance in this pair. The words still have a clearly different sound due to the voicing of the final. In German, on the other hand, consonants are never voiced in final position; /d/ simply does not occur as a stop. Germans often misspell the word "standard" as "standart" for precisely this reason. You'll also see "lock on" instead of "log on". For them both sets of words would be pronounced identically. So I think your friend was just having difficulty voicing those final stops, as you originally assumed. Vowel length was almost certainly not the problem, since German does distinguish between [i] and [ɪ] .

  14. Bryn LaFollette said,

    April 3, 2009 @ 10:33 pm

    @chris: If you've spent any time studying the phonetics of English, Chris, you'd be aware that English has an allophonic change in vowel duration conditioned by a voiced versus voiceless syllable coda. You're right in that there is no phonemic distinction of vowel length in English here, but there are many English allophonic variations that are very difficult to train yourself to become aware of. Spend some time looking at spectrograms of words which have these conditions, though, and you'll see it right away. I believe Daniel von Brighoff was absolutely accurate in his explanation, too. The entire point he was making was that his natively German speaking friend had managed to do away with applying the German coda consonant neutralization to his English, but there was still something that didn't sounds quite right about his distinction between "bid" and "bit". And as for his fear about applying his English allophonic lengthening in German, I was told by my German speaking friends that I was in fact doing this very thing in my German!

    @ Morten Jonsson: Native English speakers merge those sounds too, of course. Witness breeches and britches, creeks and cricks.

    I'm not sure I follow what you're talking about. These are minimal pairs for me, clearly distinctive. Are you saying there are English dialects that collapse the distinction of the vowels in these examples?

  15. Simon Cauchi said,

    April 3, 2009 @ 10:40 pm

    I'm having trouble imagining a green beach. Beaches can be sand or shingle or even rock, and the colours vary — brown or grey or almost white or even black — but green?

  16. Ellen K. said,

    April 3, 2009 @ 10:44 pm

    I've never noticed it before, but saying them to myself now, bid is indeed longer than bit.

  17. Sparadokos said,

    April 3, 2009 @ 11:24 pm

    @ B.O.: why would you suppose this is a joke, but "Crap Your Hands Elmo" is not? Your point about the origin of the mistake is a good one. I wondered that myself. The person who wrote this was obviously familiar with English, and just as obviously neglected to consult a dictionary before sending the text down to the sign shop for production. I think either: a) they confused French orthography with English for this word (e.g. Tchaikovsky), or more likely, b) they are familiar with both Beach and Bitch and confused them since they pronounce them as allophones.

    @Simon Cauchi: I know of Green River, Green Bay, Green Lake, and yes, Green Beach, Hawaii (obviously not the one pictured).

    The photo is from Tsentralny Park Kultury I Otdykha (Central Park of Culture and Recreation) in Saint Petersburg.

  18. Sparadokos said,

    April 3, 2009 @ 11:39 pm

    I meant to say "homophones", not "allophones" ("…they pronounce them as homophones").

  19. Daniel von Brighoff said,

    April 4, 2009 @ 12:06 am


    Bryn has dealt quite accurately and effectively with your questions, so I'll content myself with a bit of mopping up:

    1. You seem to be confusing phonemic and phonetic notation when you say "in final position; /d/ simply does not occur as a stop". It's [d] which doesn't appear in final position (in normative pronunciation, that is; regional accents are a different matter). If you claim that /d/ is absent here as well, then that makes it difficult to explain alternations like ['ʁa:t], pl. ['ʁɛ:tə] "councillor(s)" vs. ['ʁa:t], pl. ['ʁɛ:dɐ] "wheel(s)", which are quite easily handled by means of a syllable-final devoicing rule.

    2. The contrast between [i] and [ɪ] is one of tenseness, not length.In Standard German, tense vowels in stressed syllables are also long, but there still exist minimal pairs based on tenseness alone (albeit not for all speakers), e.g. ['ʁɛ:dɐ] "wheels" vs. ['ʁe:dɐ] "shipowner".

  20. Steve said,

    April 4, 2009 @ 2:52 am

    Same problem for Greek learners of English, with the pron problem affecting spelling. My students were writing about the problem of rubbish at the seaside: 'we must put beans on the biches'.

  21. Pekka said,

    April 4, 2009 @ 2:56 am

    There's also a photo of a "Baltic Bitch Hotel" that has been making the rounds on the humorous pics sites for some time. It's easy to find the image with a web search, but I don't know anything about where the photo was taken. Here's a link.

  22. Bill Walderman said,

    April 4, 2009 @ 8:48 am

    'The word for "beach" in Russian seems to be a borrowing from French "plage", no?'

    Marie-Lucie: Yes. Russian is full of borrowings from French, among other languages.

    Wasn't there a recent post about the difficulty of non-native speakers learning certain minimal pairs recently, focusing specifically on the tense/lax distinction in English and some Indonesian phonemes? I haven't been able to find it–could someone point me to it?

  23. Doc Rock said,

    April 4, 2009 @ 9:07 am

    One peach of a pitch!

  24. Ray Girvan said,

    April 4, 2009 @ 9:18 am

    'The word for "beach" in Russian seems to be a borrowing from French "plage", no?'

    Marie-Lucie: Yes. Russian is full of borrowings from French, among other languages.

    I've seen the situation summarised, unkindly but not wildly innaccurately, as "in Russian, words for anything above the level of cutting cabbages are borrowed from French". A classic example is ресторан.

  25. Stephen Jones said,

    April 4, 2009 @ 11:12 am

    When a Spanish friend of mine was working as an au pair in the UK the young girls in the house asked her what she was doing with the bedlinen.

    "I'm ironing the shit' she replied. Taking cleanliness a step too far I'd have thought.

  26. Terry Collmann said,

    April 4, 2009 @ 3:52 pm

    @Bryn "@ Morten Jonsson: Native English speakers merge those sounds too, of course. Witness breeches and britches, creeks and cricks.

    I'm not sure I follow what you're talking about. These are minimal pairs for me, clearly distinctive. Are you saying there are English dialects that collapse the distinction of the vowels in these examples?"

    No, Morten means these are the same words (breeches = britches, creek = crick) spelt according to two different vowel pronunciation regimes.

  27. Achim said,

    April 4, 2009 @ 4:32 pm

    I have with some degree of success taught myself to not devoice voiced final stops (and fricatives, but that's another story) in English, and noticing that the preceding vowels are longer than their counterparts preceding a voiceless stop has indeed helped.

    But when I started to think about French, I was puzzled. Vowel length does not seem phonemic in French, so I did not bother, and still I get words like raide right. All French vowels always "long"?

  28. talpianna said,

    April 4, 2009 @ 9:29 pm

    One has only to look at what English speakers have done with place names in the USA bestowed by speakers of other languages, such as Colorado's Purgatoire River, known locally as the Picketwire.

  29. marie-lucie said,

    April 4, 2009 @ 10:45 pm

    @Achim: Vowel length does not seem phonemic in French

    It depends on who is speaking. It is not phonemic in Southern French, which has a more restricted set of vowels than Standard French. I hesitate to generalize, as my pronunciation is now archaic compared to that of young Parisians, but in traditional Standard and Western French (including Canadian French) there are definitely minimal pairs differing in length: to me il 'he' and île 'island' are not the same, neither are faite 'done, made (fem)' and faîte 'highest point (esp. of a roof)', or tache 'stain, spot' and tâche 'task' (the last pair with a distinction of both length and quality) even though for some people those distinctions are now obsolete. For me (and still many others) the distinction between long and short vowels is quite obvious when the words are at the end of a "breath-group", although the length difference tends to cancel in favour of shortness when the word is in another position within the group.

    I get words like raide right.: I am not sure I understand your point. Do you mean that you are lengthening the vowel in this word, as in English red? This word is not part of a potential minimal pair like the ones above, so slight lengthening (less than for a French long vowel) would barely be noticed.

  30. felix said,

    April 5, 2009 @ 5:35 am

    Do you mean that you are lengthening the vowel in this word, as in English red?

    In fact, vowels are shortened before voiceless consonants, not lengthened before voiced consonants. Or at least, that's how it seems to me, but "boot” is almost short enough to be a short vowel, whereas “hood” could never be mistaken for a long one. Which brings me to another point:

    some posters have made gross generalisations like "length is not phonemic in English&rdquo, or "tense vowels aren't long in English”. These statements are probably true of American English, but many forms of English have long tense vowels.

    my particular variant of english has phonemic long and short vowels, so hut/heart, can (aux.)/can (n.) = sell/Sal, shed/shared, bid/beard are length distinctions. Tense vowels no longer form a single grouping, but are either long vowels (e.g. palm, thought, goose) or diphthongs (e.g. face, goat, debatably fleece).

    In addition to that, it has the non-phonemic voiceless shortening still happens. So "cut, cud, cart, card" all have the same vowel but with increasing lengths. My impression is that the distinction between "cud" and "cart" is much longer than between "cut" and "cud" or "cart" and "card”, but it's possible this is just an illusion caused by the relative importance of the distinction.

    So one doesn't even need to look at languages as “exotic” as Finnish or Estonian to find multiple length distinctions: English, that most banal of languages, manages to find space for it too.

  31. marie-lucie said,

    April 5, 2009 @ 8:36 am

    @felix: multiple length distinctions: the point is whether those distinctions are significant (contrastive) or not. Even if the length of English vowels varies according to the specific vowel, the exact length of long vowels is irrelevant provided that each of them is longer than its short counterpart. So there are only long and short vowels in English, a two-way distinction, not four or five significant vowel lengths. It is my understanding that Finnish and Estonian can have three significant vowel lengths, unlike English. But that does not mean that English is "banal".

  32. language hat said,

    April 5, 2009 @ 10:00 am

    I'm having trouble imagining a green beach.

    Beaches come in many colors (though green is uncommon).

  33. Ellen K. said,

    April 5, 2009 @ 12:41 pm

    Felix: I'm curious where you are from. For me (live in Kansas City area, grew up in St. Louis area, parents from Chicago), each of those pairs you mention have a distinct sound difference, except for can/can, where there is no phonemic difference. Though, saying them in a sentence ("she can put the can…"), it seems the auxilary can be schwad, but that again is a sound difference.

  34. Arnold Zwicky said,

    April 5, 2009 @ 12:43 pm

    Morten Jonsson: "Native English speakers merge those sounds too, of course. Witness breeches and britches, creeks and cricks."

    No, the sounds are not merged in these pairs. Native English speakers have a phonemic distinction between the lax and tense vowels in question, in a large number of pairs, like lick vs. leak and pit vs. Pete. All that's going on is that some speakers have the lax vowel in certain words where other speakers have the tense vowel. It's an entirely lexical matter, like which vowel phoneme you have in either, tomato, radiator, apricot, etc.

  35. Ben Teague said,

    April 5, 2009 @ 2:52 pm

    Ray Girvan wrote, "I've seen the situation summarised, unkindly but not wildly innaccurately, as 'in Russian, words for anything above the level of cutting cabbages are borrowed from French'."

    It just depends. Find a reference book that names the parts of a sailing ship. The names are mostly from Dutch, as many are in English. Computer terms tend to originate in English (komp'yuter), printing terms in German (shrift).

  36. Simon Cauchi said,

    April 5, 2009 @ 3:34 pm

    May I recommend comparison of Matthew Arnold's "Dover Beach" with Anthony Hecht's parody, "The Dover Bitch", included in more than one anthology of light verse (e.g. New Oxford BoLV, Penguin BoLV).

  37. felix said,

    April 6, 2009 @ 1:39 pm

    Sorry about the delayed reply.

    Marie-lucie, you are right that there are only two phonemes. This is relevant for coming up with an IPA scheme, but I don't speak in phonemes, but rather in phonetics. So when I need to understand if someone's said “bit” or “bid”, I automatically consider the length of the vowel, because the voicing of the consonant is not always sufficiently distinct on its own. As other posters have mentioned, when Germans and other non-native speakers don't do clipping, you can easily misidentify “bit” as “bid”.

    On top of that, no significant distinction was identified for Finnish. “kiiitos” (three i's) would be the same word as “kiitos” (two i's), only sarcastic. But getting the length wrong in “bit”, even though it's redundant, causes misidentification of the word as “bid”. All I know about Estonian is that the Wikipedia article on it says:

    Of these, simple and long are segmentally phonemic, and the third length level is suprasegmentally phonemic and aided by a distinctive tonal contour.

    I don't know what that means. In particular, I don't understand the terms “segmentally phonemic” and “suprasegmentally phonemic”, although I do understand “segment” and “phonemic”, and I'm pretty sure I understand “suprasegmental”. “A distinctive tonal contour” is, of course, child's play. But it sounds hedged enough that the four-way English distinction is not far away in significance from the three-way Estonian distinction, and definitely more important than the three-way Finnish distinction. In practice, for word identification in running speech, I mean, not in theory, for coming up with an IPA scheme.

    Describing English as “banal” I was of course being sarcastic. But I was trying to get across the idea that people look at English phonologically as if it's a solved problem and simple, but they try to find complexities in other languages which would be ignored as a problem of phonetics in English.

    @ellen k, I'm from Melbourne in Australia. different length distinctions are also made in New Zealand and parts of England.

  38. Ellen K. said,

    April 6, 2009 @ 9:57 pm

    Seems to me that just because vowel length helps us to understand the difference between bit and bid doesn't make it phonemic. Different allophones of one phoneme do help us identify another neighboring phoneme, sometimes.

    I think most of us consciously hear and say phonemes. But, of course, we do on some level hear and speak the phonetics.

  39. felix said,

    April 7, 2009 @ 3:11 am

    Seems to me that just because vowel length helps us to understand the difference between bit and bid doesn't make it phonemic.

    I wasn't saying it did. What I was saying is that it makes it significant. I don't really hear the vowel length difference between “bid” and “bit”. It's not like “bid” and “beard” where I hear different vowels. I hear the same vowel. In fact, you can solely manipulate the vowel length and it can still change the consonant I hear. But no description of English is complete unless it mentions the phenomenon, and it's especially relevant to non-native speakers because they need to make sure they perform clipping if they want to be reliably properly understood. In that regard, it's more important than the distinction between the two kinds of “th”, which even though unpredictable (i.e. phonemic), mixing them up won't cause misunderstandings.

    There's two phonemic lengths in Australian English. These can't be predicted and messing with them changes the vowel that is perceived. There's two (or more?) pairs of phonetic lengths. These can be predicted and messing with them either has no influence on perception, or creates an illusion that changes the consonant that is perceived. Those which fall in the latter category, are significant and clearly more important than “kiitos” vs “kiiitos” which was felt worthy of a post on Language Log.

  40. Victor Mair said,

    April 7, 2009 @ 10:38 am

    This discussion reminded my friend Bill Hannas of a mnemonic for how he initially learned the Russian word for "description":

    описание [Oh, piss an ya.]

  41. Hat tip to … & Stephen Fry « [ʃplɔk] said,

    April 9, 2009 @ 9:04 am

    […] Für ein Foto: “Hat tip to Eric Kinzel” (Language Log) […]

  42. Merri said,

    April 10, 2009 @ 8:43 am

    I had always thought that quality wasn't 100% necessary for a foreigner to distinguish between English 'i' and 'ea'. Quantity would do. Saying 'bitch' with short, forward 'i' (French or Italian style) would make it heard as 'bitch' , not 'beach', wouldn't it ? Granted, it would sound non-English, but even so …
    So the error is a bit more than just a side effect of Russian vocalism ?

    And BTW doesn't 'green' simply mean 'ecologically correct' ?

  43. Yuriy Zhilovets said,

    September 15, 2009 @ 2:25 am

    "Green" here just means "covered with grass and trees".

  44. Yuriy Zhilovets said,

    September 15, 2009 @ 2:31 am

    "in Russian, words for anything above the level of cutting cabbages are borrowed from French"

    I guess, this sentence was said by some Frenchman with very big language ego. French borrowings in Russian usually apply to niceties that were borrowed from France in the 19th century – magazine, restaraunt, knitted wear, scarf and all that stuff.

  45. Portia said,

    October 20, 2009 @ 1:35 am

    To reply to Amy, I definitely was amused by my French colleagues' talk of "pooting" things here or there. It was not a sound that had occurred to me might be difficult. Also heard "g-wild" for "grilled," and, of course, "bitch" for "beach."

    And yes, to another poster, Russian borrowed their term from the French "plage," which makes it one of my favorite Russian words!

  46. The Stupidest and Certainly Most Judgmental Tree-Identification Sign | said,

    May 20, 2011 @ 9:25 am

    […] language log This entry was posted in stupid mistranslations, stupid signs and tagged nature, trees. Bookmark the permalink. ← The Stupidest and Certainly Quite Off-Putting Exercise Class Sign […]

RSS feed for comments on this post