Loading ...
Sorry, an error occurred while loading the content.

Notes on Óre; or the perils of dictionary translation

Expand Messages
  • Carl F. Hostetter
    Students of both linguistics and computer science (and I was both) usually encounter in their undergraduate careers a story (probably apocryphal) that has it
    Message 1 of 4 , Jun 25, 2002
    • 0 Attachment
      Students of both linguistics and computer science (and I was both) usually
      encounter in their undergraduate careers a story (probably apocryphal) that
      has it that, in order to test an early attempt at machine translation, an
      English phrase was fed into a computer to be translated into Russian, and
      the Russian result then fed through to be translated back into English. The
      English input was: "The spirit is willing, but the flesh is weak." The
      English output was: "The vodka is strong, but the meat is rotten."

      The lesson the student is meant to derive from this latter-day parable is
      this: translation relies not just on a set of lexicons and a mapping or
      correspondence among entries, but requires semantics and context: in other
      words, an understanding of the full range of usages and connotations, in
      short of _meanings_, of not only individual words, but of the sentences in
      which they occur. Any instructor of elementary second-language courses will
      recognize this phenomenon: given a translation task, linguistically naïve
      and/or less than diligent students will simply perform a word-for-word
      translation based on dictionary head-words and principal meanings. More than
      one German student has loudly exclaimed "Ich bin heiss!", meaning to express
      "I am hot", because it is a word-for-word translation of the English phrase.
      (In German idiom, however, it in fact means "I'm horny". One should instead
      say "Es ist heiss", lit. "It is hot".)

      Unfortunately for the prevailing efforts to "speak Quenya" (a naïve
      shorthand for what in fact constitutes an almost exclusively _written_
      activity of translation by more or less mechanical selection from a
      secondary amalgamation and presentation of words and grammatical devices
      more or less arbitrarily chosen or rejected from many decades of Tolkien's
      ever-shifting and evolving linguistic creativity and coerced into an
      idiosyncratic system for which consistency is nonetheless claimed and
      artificially imposed), its principle mode of operation -- the compilation of
      wordlists by excision of words and translations from their contexts
      (semantic, literary, historical, and comparative), and the use of such
      secondary compilations _in place of_ (rather than in _aid_ of) consultation
      and consideration of the primary sources -- tends to concentrate this
      dictionary effect. (In fact, in a remarkable disavowal of Tolkien's own
      crucial and seminal realization that although "'legends' depend on the
      language to which they belong ... a living language depends equally on the
      'legends' which it conveys by tradition" (L:231), Helge Fauskanger, the
      pre-eminent proponent and practitioner of "spoken Quenya", and the most
      widely cited "source" in Internet discussions of Tolkien's languages,
      proudly proclaims that his "Quenya course . . . aims to present
      Quenya-as-Quenya, almost entirely divorced from Tolkien's fiction"
      (http://groups.yahoo.com/group/quenya/message/398).)

      In illustration of this "dictionary effect", consider the case of the Quenya
      word _óre_ and its prevailing usage in "spoken Quenya". We know from
      Tolkien's late (c. 1968) notes on this form (published in _Vinyar Tengwar_
      41, July 2000), that although translated as 'heart, inner mind' in _The Lord
      of the Rings_ (LR:1096), "'heart' is not suitable [in meaning as a
      translation], except in brevity, since _óre_ does not correspond in sense to
      any of the English confused uses of 'heart': memory, reflection; courage,
      good spirits; emotion, feelings, tender, kind or generous impulses
      (uncontrolled by, or opposed to the judgements of reason)" (VT41:11). From
      the discussions that follow, it emerges that Tolkien thought of _óre_ more
      nearly as an analogue of what in English might be described as an "inner
      voice" or conscience: that inner faculty of Incarnates (mortal and immortal
      alike, though in fallen Man open to the corruption of Melkor) that warns or
      advises as to what is right and moral and wise, beyond personal experience
      or wisdom, but arising from without, even from the Valar, and thus
      indirectly from Eru himself.

      But in surveying the usage of _óre_ in "spoken Quenya", one will search long
      and hard, if not entirely in vain, for any reflection of the moral and
      metaphysical distinctions that occupied Tolkien when considering his
      invention. That this is so is due to what Tolkien himself states in his
      _Notes_: "What the _óre_ was for Elvish thought and speech, and the nature
      of its counsels -- it says, and so advises, but is never represented as
      commanding -- requires a brief account of Eldarin thought on the matter." It
      is precisely this sort of "account of Eldarin thought on the matter" that is
      compressed out of wordlists and dictionaries of Tolkien's languages; that is
      lost when words are ripped from their context in Tolkien's etymologies,
      essays, and fiction; that is banned from discussion when it might offend
      non-Eldarin sensibilities; and that is not only ignored but _not even
      guessed at_ by the vast majority of the producers and consumers of "spoken
      Quenya", who read only what they can download for free, and who therefore
      eschew primary sources entirely in favor of such secondary references. Helge
      Fauskanger's "Quenya Corpus Wordlist" says of _óre_ only that it means
      "'heart' (inner mind), also name of tengwa 21. (Appendix E) Cf. the
      description of Galadriel in PM:337, that 'there dwelt in her the noble and
      generous spirit (órë) of the Vanyar'"; with the result that the army of
      "Quenya speakers" can and do use _óre_ whenever they "need" to translate
      'heart', 'mind', or 'spirit', without regard to nuance, connotation, or
      context (be it English or Elvish).

      This dictionary approach is not only not good linguistics, it is not even
      good translation. It is certainly no way to learn to "speak" a foreign
      language (even less so an invented language that never has been spoken), and
      no way to teach others to "speak" such. Which is why, ultimately, the study
      of Tolkien's languages (as opposed to the study of a secondary account or
      compilation of facts about Tolkien's languages) is entirely dependent upon
      and inseparable from both Tolkien's fiction and the methods of historical
      and comparative linguistics.

      "The _óre_ is strong, but the meat is rotten."


      |======================================================================|
      | Carl F. Hostetter Aelfwine@... http://www.elvish.org |
      | |
      | ho bios brachys, he de techne makre. |
      | Ars longa, vita brevis. |
      | The lyf so short, the craft so long to lerne. |
      | "I wish life was not so short," he thought. "Languages take |
      | such a time, and so do all the things one wants to know about." |
      |======================================================================|
    • gentlebeldin
      ... More precisely, it would be Mir ist heiss , lit. for me (it) is hot . The same impersonal construction with the logical subject in dative is often used
      Message 2 of 4 , Jun 26, 2002
      • 0 Attachment
        --- In lambengolmor@y..., "Carl F. Hostetter" <Aelfwine@e...> wrote:

        > More than one German student has loudly exclaimed "Ich bin heiss!",
        > meaning to express "I am hot", because it is a word-for-word translation
        > of the English phrase. (In German idiom, however, it in fact means "I'm
        > horny". One should instead say "Es ist heiss", lit. "It is hot".)

        More precisely, it would be "Mir ist heiss", lit. "for me (it) is
        hot". The same impersonal construction with the logical subject in
        dative is often used to describe personal feelings ("mir ist kalt" --
        I'm cold, "mir ist schlecht" -- I'm feeling unwell). It's even more
        common in Russian, and one can translate such sentences literally
        from German to Russian (see footnote below), but not to English.
        Literary translation is something many German students of English do,
        however, so don't be surprised when you hear a German asking for
        a "bloody steak" (instead of a blue or rare one). :-)

        The same impersonal construction seems to be intended in Quenya for
        some verbs, especially _oola-_ (dream) and _or(a)_ (urge, impel,
        move). The impersonal construction with dative for "dream" is known
        from Russian ("mne snilos'", lit. it dreamed to me), and the same
        construction "mir träumte" can still be found in 19th century German
        (it's out of use now, replaced by "ich träumte" in everyday speech).
        HF quotes these cases in his course, but he has to admit "We don't
        know very many Quenya verbs that invite such constructions, though."
        (http://www.uib.no/People/hnohf/less-d.rtf, lesson 18)

        Without having a *complete* list of those verbs or even a hint at the
        underlying principle, how can one hope to speak/write correct Quenya?
        The literal translation of the personal "I dreamed" is incorrect, as
        it would be in Russian, and the same applies to other verbs we simply
        don't know to be of the same type.

        [Misusing a word because there is no evidence of its more proper usage
        is one thing -- though as you note, at a Platonic level it does not bode
        well for "spoken Quenya" -- but it is quite another, and quite inexcusable,
        to misuse a word simply because one can't be bothered to check primary
        sources (or even to cite them). Carl]

        Hans

        footnote: It's interesting (but hopelessly off topic) that German
        semantics is often closer to that of Russian that it is to the
        semantics of a Germanic language like English.
      • Petri Tikka
        ... [...] ... In Finnish the I am hot construction is done by the same way as denoting possession: Minulla on kylmä. If translated literallyu, it would be
        Message 3 of 4 , Jun 30, 2002
        • 0 Attachment
          --- In lambengolmor@y..., "Carl F. Hostetter" <Aelfwine@e...> wrote:
          [...]
          > More than one German student has loudly exclaimed "Ich bin heiss!",
          > meaning to express "I am hot", because it is a word-for-word translation
          > of the English phrase. (In German idiom, however, it in fact means "I'm
          > horny". One should instead say "Es ist heiss", lit. "It is hot".)

          In Finnish the "I am hot" construction is done by the same way as
          denoting possession: "Minulla on kylmä." If translated literallyu, it
          would be "I have cold" or (even more poorly) "At I is cold". "I am hot"
          translated without human inteligence, "Minä olen kuuma", would still
          connote being hot, but in a sexual way (in a different way from "horny").

          (All this reminds me of a silly word play. _haista_ is the intransitive
          word for "to smell" in Finnish. _heissen_ is the word for "be named" in
          German. "Ich Heisse Petri" could be interpeted in Finnish (in "hoono
          soomi") as (approximately) "I smell like Petri".:)

          > Unfortunately for the prevailing efforts to "speak Quenya"

          I wonder: where did the idiom arise from? Hmmm...

          > coerced into an idiosyncratic system for which consistency is
          > nonetheless claimed and artificially imposed),

          Consistency is impossible to reach from scarce material of internal
          inconsistency. How could one consider it possible?

          > the use of such secondary compilations _in place of_ (rather than in
          > _aid_ of) consultation and consideration of the primary sources

          I personally only use wordlists that cite the sources, and check the
          sources, if available to me, before using the word. The sad thing is, I
          don't yet have any VT-issues.

          [I'm mailing _VT_ 44 off to everyone tomorrow, along with all outstanding
          back-issue orders. You'll have your packet of _VT_s, and lots of
          interesting reading, within a week. Enjoy! Carl]

          Adapting a saying from the Shire: "A dictionary is as dictionary does".
          With this I mean, IMHO, that a good dictionary could be made to serve as
          a way to learn languages, if sufficently complete. It would, in itself,
          contain explenations using "the methods of historical and comparative
          linguistics". I think that at least one is made similarly to that:
          http://www.jrrvf.com/~hisweloke/sindar/main.html .

          [I quite agree that it is _possible_ to create a dictionary of
          Tolkien's languages that will be _useful_ to scholar and student alike,
          through full and thorough source citation, cross-reference among entries,
          and by very careful consideration in writing the entries. But the fact is
          that the dictionaries I've seen thus far fall short of the mark. Didier's
          dictionary (what I've seen of it -- I _do_ wish Didier would get the
          Tolkien Estate's blessing/permission and publish it) does look like the
          best effort I've seen to date; but even he implies that it has
          shortcomings along the lines that I've mentioned (see:
          http://groups.yahoo.com/group/elfling/message/15989). In any event, no one
          has ever learned to _speak_ a dead language -- and Tolkien's languages are
          most assuredly that, and what's more, they have never been anything but --
          by reading a dictionary. Carl]

          PS: What do you think about my "Kalevala translations" at
          http://www.geocities.com/petristikka/elvish/ ?

          Petri Tikka
        • Petri Tikka
          ... Yes, I agree. Almost all of them are poorly and quickly done, without citations and without any cross-references. One must say in defence of the ones Helge
          Message 4 of 4 , Jul 1, 2002
          • 0 Attachment
            --- In lambengolmor@y..., "Carl F. Hostetter" <Aelfwine@e...> wrote:
            > [I quite agree that it is _possible_ to create a dictionary of
            > Tolkien's languages that will be _useful_ to scholar and student alike,
            > through full and thorough source citation, cross-reference among entries,
            > and by very careful consideration in writing the entries. But the fact is
            > that the dictionaries I've seen thus far fall short of the mark.

            Yes, I agree. Almost all of them are poorly and quickly done, without
            citations and without any cross-references. One must say in defence
            of the ones Helge Fauskanger has done, that they indeed do have
            source citations (most don't), but of course, only from publications
            available to the maker at the time of the creation.

            [Part of the purpose of Web publishing is that documents can be kept up to
            date with ease. I would have thought that Helge could have updated his
            word-list some time in the past two years. Carl]

            And so, the instance of the misinterpretation of _óre_ in "the Quenya
            Corpus Wordlist" is not, in fact, the wholy best example, for VT41
            had not been published during the making of the Wordlist. So the
            discussions in VT41:11 weren't available for citations. The case is
            different in his newer private "Quettaparma Quenyallo", since it
            does refer to this issue.

            [Helge has refused me a copy of his dictionary, so I wouldn't know. I do
            wish he'd get the Tolkien Estate's blessing/permission and publish his
            dictionary properly. As for _óre_, I never said it was the _best_ example;
            but it is a good one. Further, Helge's is not a "misinterpretation", but
            it is certainly an _incomplete_ interpretation, presenting as it does
            only a small part of the available information about this word, and its
            significance across the decades of Tolkien's creativity. Carl]

            But even that wordlist, indeed, falls "short of the mark", not being
            carefully made at all points. But it is the best Quenya-dictionary
            I'm aware of. A much better one could be made, of that I am sure.

            > Didier's dictionary (what I've seen of it -- I _do_ wish Didier would
            > get the Tolkien Estate's blessing/permission and publish it) does
            > look like the best effort I've seen to date; but even he implies
            > that it has shortcomings along the lines that I've mentioned (see:
            > http://groups.yahoo.com/group/elfling/message/15989).

            Yes, I am aware of them. But at least it is attempting to be accurate.

            > In any event, no one has ever learned to _speak_ a dead language

            No, not to fully speak. That is impossible for the simple
            fact that to learn to fully _speak_ one must hear people
            who actually _speak_ the language. If no speakers are
            alive, the atempt to do so is doomed from the start.

            > -- and Tolkien's languages are most assuredly that, and
            > what's more, they have never been anything but --
            > by reading a dictionary. Carl]

            A dictionary can be a wonderful helper, but its contents
            are dead without context.

            Petri Tikka
          Your message has been successfully submitted and would be delivered to recipients shortly.