Thursday, 2 June 2011

I can has чизбургер?

When our cruise reached St Petersburg, eight years had passed since my previous most recent visit to Russia. One of the differences that struck me on this occasion was the increased number of public signs in English. When I say “in English”, I do not mean that most English-speaking people would have recognized them as being in English — because they were English words written in Cyrillic.

I can’t speak Russian, but I can read Cyrillic. I don’t know exactly what the Russian for ‘café’ is*, but I’m pretty sure it isn’t кофe шоп. OK, кофe kofe is ‘coffee’, but шоп shop? This is the English shop written in Russian letters.

I saw several signs advertising БИГ МАК (big mak). which you may recognize as a brand of hamburger.

And there was a fast food joint offering пицца (pitstsa), which I suppose is Italian via English.

The treatment of the English word house was interesting. Sometimes it appeared with relative phonetic accuracy as хаус (khaus), but at other times less accurately as хауз (khauz). (The first word in the lefthand image is steyk.)

Here is ‘second-hand’, nicely demonstrating Russian inattention to the English e-æ distinction. (Actually, this makes me ask why the second part of Big Mac doesn’t come out as МЭК or МЕК (mek) rather than МАК. Perhaps it was an executive decision by the American owners of the franchise.) As for ‘grill dog’, I have never come across this menu item in an English-speaking country. I wonder if it is a Russian invention. It seems to be a frankfurter in a panini. The sign below also mentions Хот дог (khot dog), Френч дог (french dog), Гамбургер (gamburger), Бургер ролл (burger roll) and finally Чизбургер (chizburger).
*(PS: I’ve now looked up the Russian for ‘café’, and my sources give it as кафе kafe or буфет bufet.)


  1. да, сингз ар геттинг аут аф ханд, феллоу комрад.

  2. хауз for house isn't so bad, considering that final consonants are devoiced in Russian. It might even be more appropriate, since Russian spelling is morphophonemic.

  3. In Serbia, I suppose, you'd see all these signs in English (although we seem to prefer the more Italian 'caffe' to 'café'). But outside official signs, we would indeed say 'чизбургер' and 'хот дог' and 'биг мек' (never 'биг мак').

  4. I have seen (in photos) signs for Пицца Хат. If the logo had not been so easy to recognize, "Хат" would probably have given me some trouble, as [a] and [ʌ] (in "Хат" vs. "hut") are farther apart in our vowel space than they are for, say, someone from the UK. The double "ц" is also surprising. If it is pronounced as [t:s], then apparently the word is treated as a borrowing from Italian rather than from English. But I would expect it to be pronounced simply as [ts].

  5. Oops. Omitted information: "our vowel space"=that of Americans.

  6. МЭК would look very unnatural: Э is normally used only initially, even in borrowed words.

    The use of Г in Гамбургер shows that it is moderately old: the newer style is to use Х as in Хот дог. In the novel A Torrent of Faces (1967) by James Blish and Norman L. Knight, which is set in the year 2794, there is a city in Missouri named Gitler. It was founded by Ukrainians (or perhaps Ukrainian Canadians) and named after one of the liberators of Ukraine; apparently the full truth had been distorted or forgotten.

  7. Here is ‘second-hand’, nicely demonstrating Russian inattention to the English e-æ distinction.

    I don't speak Russian, but according to the Wikipedia Russian phonology page, there are only five vowel phonemes: /i, e, a, o, u/.

    Let's assume that /i/ is used for KIT-FLEECE, /e/ for FACE-DRESS, /a/ for STRUT-PALM, /o/ for GOAT, and /u/ for GOOSE-FOOT. Russians would then have somehow to squeeze in TRAP, LOT, and THOUGHT. It seems a little unfair to criticize them for squeezing TRAP into /e/ rather than /a/, when presumably either one would be equally appropriate.

  8. It's funny how English words with /æ/ come into Polish with either /a/ or /ɛ/, but I can't find any reason why it's one or the other.

    Some words of English origin with /æ/ always realized in Polish as /a/: badminton, compact, establishment, fan, fax, hamburger, iPad, Mac, scanner, snack, scrabble, tablet.

    Some words of English origin with /æ/ always realized in Polish as /ɛ/: backhand, businessman, flash, hand, happening, happy, manager, match, stand.

    This seems to be consistent with your Russian examples.

  9. @Michal:

    It looks as if spelling pronunciations and trademarks have /a/, while words learned through speech have /E/.

  10. Or when they were borrowed? Given the [a]-ward drift of TRAP in RP-type British English, that would make some sense.

    As a Northerner, I still find TRAP [ɛ] and STRUT [a] strange (and liable to cause confusion); it makes me think people are learning from South Africans or something.

  11. I think we've discussed the TRAP conundrum in English borrowings into Russian and Polish on here earlier.

    There doesn't seem to be a good, clear rule for Polish.

    Here's my take: Polish people tend to believe that TRAP = Polish /ɛ/. As a result, that's what they'll use when "more aware" that a given item is English (probably with less assimilated items, or when trying to accentuate the "Englishness" of the item).

    With older borrowings, now thouroughly assimilated, this was extended to their spelling (i.e. they use "e"). But newer borrowings tend to be graphemic; and if they're felt "sufficiently Polish", they're pronounced accordingly, with /a/.

    My favourite example: English flash 'lamp on your camera', an older borrowing, is rendered as flesz, and pronounced, accordingly, with /ɛ/. In the sense of 'computer graphics technology', it's newer, spelled as in English, but the pronunciation varies between /ɛ/ and /a/ -- seemingly it's on its way to being "Polished", but not there quite yet. But scan, an older one, has been assimilated as skan /skan/.

    However, there's still quite a lot of variation. For the record, Big Mac is /big mak/.

  12. It's funny how English words with /æ/ come into Polish with either /a/ or /ɛ/, but I can't find any reason why it's one or the other.
    Same in Italian, though I think /a/ is limited to very recent loans. For example, I say cracker as /'krɛker/ if I mean a type of food and /'kraker/ if I mean a computer criminal.

    BTW, according to Wikipedia, Russian /a/ is [æ] between soft consonants, so what would be wrong with Мякь for Mac (cf Irish meaits /mʲatʲʃ/ ← match)?

  13. Yes Army, what you say about Italian and wjarek says about Polish has long been happening on a vast scale in Japanese, with far more respellings and rerespellings than that, and corresponding proliferation of lexemes (but sometimes just of age-determined free allomorphs).

    And what you say of Russian: Japanese kyappu 'cap' and kappu 'cup'. And this principle seems to override considerations of disambiguation and English cred: kyasshu 'cache' is still doing service for kyasshu 'cash' as well. One really wonders why the Russians didn't go for it!

  14. I wonder how Г persisted for so long as the Russianisation of the English H - you see Gamlet, Ganover, Gitler and my personal favorite, the American president Gerbert Goover. But it seems like it's on its way out for new borrowings, and thank goodness - the X is so much closer. Using it entirely in place of H would give you a Russian accent, but would make you understood. If a Russian asks an Englishman if he likes Gamlet, the Englishman is pretty likely to be completely confused.

    Anyone know where H = Г came from?

  15. Army1987,

    convention. Just as we write kasha for Russian porridge, not kusher.

  16. Well the correspondence between Russian g and h in other Slavic languages goes way back.

  17. In fact, considerable parts of the Russian dialects have a fricative even today. (And, just to avoid misunderstandings, I'm not counting Ukrainian and Belarusian among those.)

  18. Мякь doesn't work because /k/ cannot be palatalised. To achieve the right vowel you'd thus need to change the final consonant to something else, but that'd sound less similar to the English original.

  19. @Thomas W

    /k/ and /g/ can be and are palatalised all the time (cf. кость — кисть, and горе — гиря), although I agree that when it comes to adopting English sounds preference usually is given to the consonants at the expense of vowels (and while /æ/ is present as an allophone of /a/ in Russian, it is not used, thus preserving the English consonants' being non-palatalised).

  20. @Thomas W
    «To achieve the right vowel you'd thus need to change the final consonant to something else, but that'd sound less similar to the English original.»

    That doesn't deter the Japanese. The Japanese examples I have given above show the advantages of not attempting any such similarity, but rather maintaining distinctions in the English originals.

  21. @Lane

    I don't know the answer to your question, but it's obviously not about English (and you've mentioned a German example yourself), there are a plenty of, say, Greek words that have this contrast as well (hero — герой, Herodotus — Геродот, Herostratus — Герострат).

    One way to explain it is euphony. Not only /g/ sounds more natural to a Russian ear (evidence anecdotal), it also helps to avoid unnecessary obscene terms (as you would get a хер-берт for Herbert).

  22. Really, old g turned h/ɣ/ɦ in many Slavic languages, including large parts of Russian. So, when people wrote г, they didn't imply g in the first place.

  23. «To achieve the right vowel you'd thus need to change the final consonant to something else, but that'd sound less similar to the English original.» (Thomas W)

    And come to think of it you can't say either к cannot be palatized or that it is all the time. It's OK, emperor, I know what you mean, but the velars are a bit of a morphophonological oddity. What you can say, and what I suspect Thomas means, is that кь is even more of an oddity, i.e. that к doesn't get _marked_ for palatization, but that doesn't mean it can't be, the obvious example being кью, for anyone who doesn’t think кю is satisfactory for Q.

    So Thomas, don't pick on Army's Мякь. It would be the obvious analogy for anyone who doesn’t think Мяк is similar enough to the English original.

  24. Lipman

    And did the g>ɣ in the genitives survive with the acoustic substitution v because of their relative preponderance?

  25. That's a complex issue, probably not.

  26. Lipman: In these parts we speak of /kɑʃə/, so how else would we spell it but kasha?

  27. Just one example, for the (most widespread?) pronunciation ˈkæʃə - the English spelling kasha wasn't chosen because it may be interpreted both ways anyway.

    Or vwawtker, if you like. Or names, eg Evun.

  28. 1. My wife tells me that кафе kafe and буфет bufet are not interchangeable. A bufet is likely to be a self-service counter; a more commodious outfit is more likely to be called a kafe.

    2. Хаус and хауз represent identical pronunciations.

    3. The muddle of representations of English orthographic a, u, and h is nicely illustrated in Elena's biographical dictionary with entries for:

    ГЕ́КСЛИ, Томас Генри (1925-95)
    = geksli, tomas genri
    i.e. Thomas Henry Huxley

    Immediately above is a mention of

    ГЕ́КСЛИ, Олдос
    =geksli oldos
    i.e. Aldous Huxley

    with a redirection to ХА́КСЛИ (=khaksli).

    Nineteenth century individuals are invariably (I think) transliterated with Г for Roman H. By the end of the twentieth century this had changed invariably (I think) to Х. I have a hunch that the trend to the new transliteration was quicker to start with English names than German names — but this could be illusory. Aldous Huxley was writing before Hitler was widely known outside Germany, but perhaps the Soviet Union began writing of Huxley somewhat later.

    I've never met a Russian who saw Cyrillisation as transcription. They transliterate and then pronounce the words with a spelling pronunciation — unless, of course, they can produce an authentic foreign-native-speaker-like pronunciation. But that would be despite the Cyrillic spelling.

  29. I knew I'd written about the biographical (actually encylopedic) dictionary some time some where. I find that it was a posting of this blog — with a link to some copies of the Huxley entries, plus mention of Mark Twain's Adventures of Geckleberry Finn click here.

  30. I remember reading that when Isaac Asimov's books were published in the Soviet Union, his name was printed as Айзак Асимов rather than his original Russian name of Исаак Юдович Озимов, partly to hide his Jewish origins. The use of s in his surname was an artifact of German-Yiddish romanization conventions; he always pronounced it /z/, thus:

    When Isaac's at a nudist camp,
    He promptly joins the fun;
    For When in Rome's his favorite quote
    As he tells everyone.

    So when order comes around
    "All clothing you must doff",
    Without a moment's hesitation
    Isaac Asimov.

  31. Presumably Озимов had penultimate stress, causing the pre-tonic О to be pronounced a?

  32. @John Cowan:

    Is that supposed to be "Isaac (h)as (th)em off"?

    Doesnt work for non-weak-vowel-mergers like me, I'm afraid.

  33. 'BTW, according to Wikipedia, Russian /a/ is [æ] between soft consonants, so what would be wrong with Мякь for Mac'

    - the 'a' vowel might be [æ] indeed in certain contexts, but we tend to have a [i]-transition vowel between a palatalized consonant and the succeeding vowel, so it wouldn't be just /m'æk(')/, but /m'iæk'/, and that sounds really far from the original.
    Moreover, from my teaching experience, I'd say that Russian students when studying British phonetics are very likely to take /æ/ for 'a', not for any kind of e-sound, so substituting English 'mac' for 'мак' seems only natural to me.

  34. @Emperor_spock and @Thomas W

    while /k/ and /g/ are indeed palatalized, /k'/ and /g'/ normally do not occur in word final position in Russian; thus /м'як'/ is quite impossible. This /k'/ and /g'/ distribution limitation has always been a headache for Russian phonologists :)

  35. btw Ha translates to Га in 95%. So, for example, Harry Potter translates (not transliterates) to Гарри Поттер.... and, Кофе-хауз is a name, not a translation ) cafe is transliteration of french word café. There's reasoning to use Га as pseudotransliteration of Ha, that goes from old, pre-revolutional grammar rules of russian.

  36. > Nineteenth century individuals are invariably (I think) transliterated with Г for Roman H. By the end of the twentieth century this had changed invariably (I think) to Х.

    It's a bit more complex: they are usually trahsliterated with Г if they were transliterated a long time ago, according to old tradition (there were no formal rules of transliteration). Since 1980s (I'm not sure in the date) they are usually transliterated with Х. The change of tradition started probably in 1960s. In the other words, Г is applicable only for persons which were known for a long time, and if a person is old but there is no tradition to translate it with Г, it should be translated with Х.

    Harry has a very old, traditional, and common transliterated form Гарри (maybe you know Garri Kasparov? That's his real name, and wide use of English names in Armenia and Azerbaijan is another long story), and it sounds very unwonted when transliterated as Хэрри, that's why Хэрри or Харри is rare.

  37. As late as the last years of the Soviet Union, a friend of mine called Howard sent off for a visitor's visa in the name of Howard and what he received said "Говард". (I won't write his surname here, but a TRAP vowel in it came back as "э".)

  38. I'd be prepared to believe that the reason that the recent examples (original post) use "г" in гамбургер but "х" in the other examples (хаус / хауз and хенд) is that гамбургер is an earlier borrowing - the first McDonalds opened in 1990 and no doubt Гамбург also existed as a much older borrowing and would have had an influence.