Federman’s Languages.

Another intriguing passage from Alien Tongues (see this post), this time about Raymond Federman (footnote 56 on p. 198):

Raymond Federman, born in Paris, came to the United States at nineteen, after escaping from a deportation train and spending the rest of World War II working and hiding on a farm in southern France. Although he did not start learning English until he came to the United States, four years in the U.S. Army, including a stint in the 82nd Airborne and six months on the front lines in Korea, vastly accelerated his acquisition of “colloquial” English. Federman writes in both French and English, and his own observations about his bilingualism frequently support the generalizations made in this chapter. When he began writing his novel Amer Eldorado (circa 1970), he worked, alternately, on the French and English versions — one day on the French version, the next day on the English, and so on. He kept that up for about six months: “It drove me crazy. The two languages were pulling apart, pulling together, encouraging one another, defying one another, feeding one another, or perhaps I should say devouring one another. Eventually I dropped the English text and finished the French which became Amer Eldorado (published in 1974 by The Editions Stock). Then I went back to the English text and worked on it for three more years. That became Take It or Leave It. But no longer the same book. Amer Eldorado is about 200 pages long. A first-person narrative. Though Amer Eldorado is contained, in a manner of speaking, in Take It or Leave It (which is about 500 pages long), it is there not as a translation but as a loose adaptation. Moreover, Take It or Leave It uses two narrative voices (first and third person).” Federman always feels a sense of incompleteness when his work exists in only one language. When he writes poems, he immediately does a version in the other language (whichever), because he has the feeling that the original text is not finished until there is a version in the other language. He usually abandons self-translation of his novels, however, “for reasons of time, laziness, etc.,” but the result is that he feels his novels are never finished. Translations by other people do not do the trick. The only time Federman did an immediate translation of a prose text was for The Voice in the Closet / La voix dans le cabinet de débarras, where the two texts coexist in the same book, working from either end (the French text is rectangular, the English one, square). Federman says that his ambition is to write a book — admittedly, totally unreadable — in which the two languages would come together in the same sentence (there are a few such pages in Take It or Leave It). The cover would say “translated by the author,” but would not indicate from which language into which. (Federman did once publish a bilingual text [“D’une parenthèse à l’autre” / “From One Parenthesis to Another”] wherein the English version says “translated from the French by the author” and the French version says “translated from the English by the author.”) Behind the playfulness, there is a “need to abolish the ‘original.’ In fact, between the two texts translated by the author, there is no original, no possibility of origin” (personal communication).

I posted an example of one of his bilingual texts back in 2008.


  1. Federman says that his ambition is to write a book — admittedly, totally unreadable — in which the two languages would come together in the same sentence

    IIRC there are poems that are readable as Urdu or Persian, giving different meanings.

    In a more comic vein, there are those Franglais stories that Miles Kingston wrote… I wonder if they are as hilarious to Francophones as to Anglophones.

  2. I wish I could my previous comment — rekhta began as poems with alternative lines in Urdu and Persian but IIRC there are also poems in which the same line is readable as Urdu or Persian…

  3. Afanasii Nikitin’s travelogue of India had similar Russian-Turkic (Arabic/Persian) text.

    В Ындейской земли гости ся ставят по подворьем, а ести варят на гости господарыни, и постелю стелют на гости господарыни, и спят с гостми. [Сикиш илиресен ду шитель бересин, сикиш илимесь екъ житель берсен, достур аврат чектур, а сикиш муфут]; а любят белых людей.

    Ындея же [какъпа чектуръ а учюсьдерь: секишь илирсень ики жител; акичаны ила атарсын алты жетел берь; булара достур]. А [куль коравашь учюзь чяр фуна хуб, бем фуна хубе сиа; капъкара амьчюкь кичи хошь].

    The Turkic text is used for revealing description of sexual habits of Indian prostitutes while the Russian text simply says there are prostitutes in India, but doesn’t go into shocking detail.

  4. Oh you mention such racy details, but not his heartfelt prayers in Church Slavonic and Arabic!

    (I searched to find out where exactly he went.)

  5. Afanasii Nikitin’s travelogue of India had similar Russian-Turkic (Arabic/Persian) text.

    Thanks very much for that!

  6. In some places, prayers and racy descriptions are hopelessly mixed and the languages too.

    А съезжаются все наги, только на гузне плат; а жонки все наги, толко на гузне фота, а иные ф фотах, да на шеях жемчюгу много, да яхонтов, да на руках обручи да перстьни златы. Олло оакь!

    And everyone comes here naked, save for a loincloth; and women are all naked, wearing only a veil around hips, while some have a veil with lots of pearls and rubies and hand bracelets and golden rings. Allah Akbar!

  7. The Allah Akbar here is used emphatically, to show his shock at sight of a crowd of hopelessly naked Indian ladies.

    Probably should be translated as “Oh My God!”

  8. капъкара амьчюкь кичи хошь

    And this little phrase embarrassed Soviet translators so much that they opted to omit translation of the key word from all editions of Nikitin’s travelogue.

    I can translate if anyone is interested, but it’s still very offensive…

  9. Please do, and explain any difficult words! We’re not ханжи here.

  10. Almost identical conversation happened on this site: http://amalgrad.ru/viewtopic.php?id=753

    Comments are funny:
    The first guy says: Согласитесь, немного странный перевод. Кто понял, тот понял. Переводить дословно не просите – все-таки на форуме есть и дамы.

    The second guy says: Ну всё, дамы побежали к тюркам за переводом!

  11. капъкара амьчюкь кичи хошь
    jet black little pussy I like

    Six centuries passed and the phrase still manages to be deeply offensive on several levels.

    It’s racist, misogynistic and probably in violation of laws against child pornography.

  12. another no no word for Soviet translators:

    секишь илирсень ики жител


    a fuck will cost you 2 shekels(?)

  13. Since we’re not ханжи here…


    To me, it seemed like “tight” vs “little” is a better translation – I don’t think the size has to do with age so much. Thoughts?

  14. В Ындейской земли гости ся ставят по подворьем, а ести варят на гости господарыни, и постелю стелют на гости господарыни, и спят с гостми. Сикиш илиресен ду шитель бересин, сикиш илимесь екъ житель берсен, достур аврат чектур, а сикиш муфут; а любят белых людей.

    In India, merchants (the word гость now means guest) are hosted in hostels, hostesses prepare food for merchants and make beds for merchants and sleep with merchants [if you wish to fuck her, the cost is two shekels, if you don’t want to fuck her, then it’s only one shekel, wife-friends are many and fucking them is free]; and they love white men.

    interestingly, while the non-Russian text in brackets is Turkish, the numerals for some reason are Persian (du, ek)

    Dostur avrat (literally wife-friends) probably refers to common Persian practice of temporary marriage.

  15. but mostly Nikitin uses Oriental languages for religious purposes to astonishing effect:

    Милостию божиею преидох же три моря. Дигерь худо доно, олло перводигерь дано. Аминь! Смилна рахмам рагим. Олло акьбирь, акши худо, илелло акшь ходо. Иса рух оало, ааликъ солом. Олло акьберь. Аилягаиля илелло. Олло перводигерь. Ахамду лилло, шукур худо афатад. Бисмилнаги рахмам ррагим. Хуво могу лези, ляляса ильлягу яалимуль гяпби ва шагадити. Хуя рахману рагиму, хубо могу лязи. Ляиляга иль ляхуя. Альмелику, алакудосу, асалому, альмумину, альмугамину, альазизу, алчебару, альмутаканъбиру, алхалику, альбариюу, альмусавирю, алькафару, алькалъхару, альвазаху, альрязаку, альфатагу, альалиму, алькабизу, альбасуту, альхафизу, алльрравию, алмавизу, алмузилю, альсемилю, албасирю, альакаму, альадюлю, алятуфу.

    By Grace of God, I crossed the three seas. [The rest only the God knows, Allah only knows.] Amen! [In the name of God, the Merciful, the Compassionate. Allah Akbar, the God is good, the Lord is good. Jesus is the Spirit of Allah, peace be upon him. Allah Akbar. There is no god, but Allah. The God is Protecting. Praise be to Allah, thanks to the all-victorious God. In the name of God, the Merciful, the Compassionate. He is the God, there is no God, but him, he knows everything secret and hidden, He is the Merciful, the Compassionate, He has no equals.There is no God, but Allah. The King, the Holy, the Peace, the Granter of Security, the Controller, the Almighty, the Irresistible, the Supreme, the Creator, the Evolver, the Shaper, the Repeatedly Forgiving, the All Compelling Subduer, the Bestower, the Ever Providing, the Opener, the All Knowing,the Restrainer, the Expander, the Abaser, the Exalter, the Giver of Honour, the Giver of Dishonour, the All Hearing, the All Seeing, the Judge, the Gentle.]

    text in brackets is in a strange mix of Arabic and Persian.

    While Russia had no Inquisition, translating such a text into Russian probably was not a good idea…

  16. That piece about Jesus being Spirit of God…

    People have been burned at the stake for less!

  17. Thanks, extremely interesting stuff!

  18. Note that in this thoroughly Islamic text, one thing is conspicuous in its absence – profession that Muhammad is the prophet of Allah.

    With tricks like that Nikitin managed to spend four years in Muslim lands, going to their mosques, making namaz and saying prayers in Arabic, but never making the final, crucial step which would make him a Muslim and apostate.

    He confesses elsewhere in his travelogue that he has sinned, that he wasn’t a good Christian (but saying that it was impossible to be a Christian in Muslim land), nevertheless he maintains that he did not betray Christian faith.

    It’s a rather dangerous behaviour. As the Quran says:

    14. And when they come across those who believe, they say, “We believe”; but when they are alone with their devils, they say, “We are with you; we were only ridiculing.”

    15. It is God who ridicules them, and leaves them bewildered in their transgression.

    16. Those are they who have bartered error for guidance; but their trade does not profit them, and they are not guided.

  19. Yes, very interesting, thank you SFReader. I found a translation in the book Visions of Mughal India, which has some of the passages you cite, but without marking the different languages or including the Allahu Akbar and with much less colloquial diction. Presumably they are toning it down, rather than you spicing it up…

    What is ханжи? Google Translate gives Khanji, which doesn’t help. (I must admit I have always found it a little jarring that и is a vowel, consonants having different pronunciations is fine but crossing the consonant/vowel barrier is a bit too much.)

  20. ханжА
    = goody, hypocrite, pharisee, …

  21. https://www.youtube.com/watch?v=-wn6LvzH7kw

    I was interested to hear his voice.

  22. interestingly, while the non-Russian text in brackets is Turkish, the numerals for some reason are Persian (du, ek)
    I’ve seen Uzbeks (who otherwise would speak Uzbek among themselves) using Tajik / Persian numbers for counting when playing cards. So perhaps it was also usual for Turkic speakers in Nikitin’s times to use Persian numbers, at least for certain purposes like prices of goods as in your quotes.

  23. –Presumably they are toning it down, rather than you spicing it up…

    I think they were translating from the Soviet edition with translation into modern Russian which, I believe, was extremely politically correct – to the point of outright falsifying original text.

    At least that’s the most common complaint of Russian readers when they finally discover the original text.

  24. What is ханжи?
    The plural of ханжа (stress on the final syllable), for which juha gives a good set of definitions. (Not sure what you mean about и [= i]; it’s a vowel, with nothing consonantal about it.)

  25. The 1958 film Russian film on Afanasy Nikitin featured Padmini (fully dressed).

  26. marie-lucie says

    fisheyed: (I must admit I have always found it a little jarring that и is a vowel, consonants having different pronunciations is fine but crossing the consonant/vowel barrier is a bit too much.)

    “Crossing the consonant/vowel barrier” is nothing unusual. The “high” vowels are so called because they are pronounced with the tongue almost touching the palate. Bring up the tongue just a little bit higher and you get
    the ‘semivowels’ (also called ‘semiconsonants’). Depending on the language, the semi’s can be associated with a vowel, creating a diphthong (after the vowel, as in English “house” [haws]), and/or function as a consonant (as in English “wall”). In French the letter “i” can be pronounced as vowel, as in Marie /mari/ or as a semivowel before another vowel, as in Marianne /marjan/ (two syllables).

    Perhaps the problem you see is that the Russian ханжи ends in the vowel и alone, while a lot of Russian words end in ии, where only the first letter is the vowel [i] while a mark over the second one (sorry I can’t do it here) indicates a semivowel or semiconsonant pronunciation.

  27. (Not sure what you mean about и [= i]; it’s a vowel, with nothing consonantal about it.)

    I feel really foolish typing this but in the interests of clarification, since it looks like a backwards N, the fact it is a vowel feels …weird to me. I apologize for the sudden intrusion of silliness into this thread.

    So macaronic texts, do all the manipravalam commentaries count? Or are they considered a dialect of their own?

  28. Good heavens, no need to apologize — silliness is always welcome in an LH thread! It’s just that I’m so familiar with Cyrillic that the “backwards N” thing didn’t even occur to me.

  29. If I understand Wikipedia article correctly, n comes from Greek ν and и from Greek η. u and n also look pretty similar and might be confused in sloppy handwriting, but I’ve never heard people complaining about it being between vowel and consonant.

  30. Google Translate is very strange. It translates ханжА as prude, ханжa as hypocrite, and ханжи as hypocrite. I realize I had the “khanji” issue yesterday because “detect language” rather than “Russian” was selected for input language — and Google detected Mongolian.

  31. I wonder why Google Translate detects it as Mongolian.

    There is no such word in Mongolian.

    The closest I can think of is ханжин (also spelled as ханз), which means something like http://en.wikipedia.org/wiki/Kang_bed-stove

  32. The detect-language algorithm is intentionally much simpler and cheaper than full translation. It looks for superficial things that might be a reasonable indication, given the set of languages it knows something about, of what language this one might be. Not too different from someone like the Hat or me, who can identify many languages when spoken even if we don’t understand a word of the conversation.

    Also, GT in very recent days seems to be falling back on Wiktionary or something very like it when the input is a single word. This is probably best, as no language detector can work on such short inputs: what language is chat, English or French?

  33. what language is chat, English or French?

    Urdu. (Looking at that website is make me faint with hunger).

    But the point here is that the word doesn’t exist in Mongolian, hence the English transliteration as a result, so it is strange to select Mongolian as the detected language.

  34. It looks vaguely Mongolian to the algorithm. If I showed you the word éburation and asked you to guess the language with no reference to any dictionaries, wouldn’t you guess French? Yet it is not a French word (well, I found one reference, but I suspect it is a typo or deliberate misspelling).

  35. Well, I would have guessed that the algorithm would resort to a dictionary (and so did you when you mentioned Wiktionary, no?), word frequencies and language size. Even if a word exists in both Mongolian and Russian, I would expect that detection algorithm to prefer a frequent word in a larger language like Russian to a rare word in a small language like Mongolian. But it doesn’t do this, obviously, which is what I find strange. The fact that capitalizing the last letter gives a different definition is also strange.

  36. I very much doubt that the algorithm consults any dictionary. It has to be very fast to serve its purposes, which are by no means limited to translation on demand. For example, there are situations where we want to present a text if it is in the user’s language and offer translation if it is not. For this purpose, we need to be able to identify the user’s language from the text without slowing down its delivery. (“We” is generic; I speak only for myself.)

  37. I know no details about Google’s language detection, but the algorithms I’m familiar with don’t use words at all. Often they are based on frequencies of digraphs or trigraphs in the text. I’d imagine some might use a handful of extremely common words for identifying languages. Looking up every word from the text in every language available would be far too slow and resource-intensive.

  38. The text was easier to read than I imagined: sikmek even exists in a Turkey-Turkish dictionary.

  39. I would think that GT first determines the writing system. That’s fast and very often determines the language. For example, looking only at a few characters serves to distinguish, say, Danish, German, French and English. This approach won’t work every time, but it surely speeds things up.

  40. But none of the characters in ханжи are unique to Mongolian, in fact, they exist in all languages using Cyrillic.

    And I don’t think it’s possible to detect Mongolian accurately by letters used.

    Kazakh Cyrillic, for example, uses every single letter of Mongolian Cyrillic plus seven additional ones.

  41. I think GT fell for the string хан, If there’s a Khan in it, it must be Mongolian. 😉


  1. […] Hat talks about the language use of writer Raymond Federman and tries to find a story with an unusual method […]

Speak Your Mind