Translating to Music

Author(s): Sigmund Spaeth

Source: The Musical Quarterly, Vol. 1, No. 2 (Apr., 1915), pp. 291-298
Published by: Oxford University Press
Stable URL: http://www.jstor.org/stable/737851 .
Accessed: 31/10/2013 16:07
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp

.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact [email protected].

Oxford University Press is collaborating with JSTOR to digitize, preserve and extend access to The Musical
Quarterly.

http://www.jstor.org

This content downloaded from 143.106.201.145 on Thu, 31 Oct 2013 16:07:52 PM

All use subject to JSTOR Terms and Conditions

TRANSLATING TO MUSIC
By SIGMUND SPAETH
NY ONE who has ever devoted much time to translating

songs, operas or cantatas into English must frequently be

annoyed by such a remark as "How wonderful it is to be

a linguist!" As a matter of fact, being a linguist is the least
important qualification for the delicate task of translating to
music. With the aid of cosmopolitan friends, dictionaries, or
even hack-translators, a literal version of any prose or verse is
not difficult to obtain. It is the poetical insight, the understanding of the intentions of both the composer and the author
of the text, and the ability to imitate the salient features of the
original combination that must determine the skill of the musical
translator.
In the midst of the still raging controversy concerning
"opera in English," the question has been asked again and again,
"Why translate at all?" There can be only one answer. The
translation of words set to music is permissible in so far as it has
an educational value. If a song or an opera can broaden its sphere
of influence through an English version, then by all means let it
be translated. Better the half-knowledge that comes from even a
weak imitation than complete ignorance.
American music-publishers, recognizing this need, have
generally made it a rule to supply all foreign works with singable
English equivalents, not with the intention of encouraging the use
of translations, but that every composition may have a fair chance
to deliver its message. The amateur who has not yet mastered
the pronunciation of a foreign language may still acquire a certain
appreciation of a masterpiece through an adequate translation,
and thus be tempted to seek out the less accessible beauties of
the original.
From the educational point of view, therefore, English versions
of vocal music are desirable. Why cannot the hysterical propagandists of "opera in English" be satisfied with this verdict
instead of insisting upon their artistic value as well?
A well-known lecturer, speaking from the stage of the Century
Opera House in New York, recently compared grand opera with a
beautiful picture, veiled, however, by the unfamiliar language in
291

This content downloaded from 143.106.201.145 on Thu, 31 Oct 2013 16:07:52 PM

All use subject to JSTOR Terms and Conditions

The Musical Quarterly

which it was presented. "Why not remove this veil of language,"
said the ardent enthusiast, "so that all the beauties of line and
color may become clearly visible? "
The utter absurdity of this analogy scarcely requires comment.
If an opera is to be compared with a picture, then surely the language in which it is written represents not an external covering but
the very lines and colors of which the picture is composed. Carrying out the analogy to its logical conclusion, the English version
would have the same relation to the original opera that a photograph or a lithograph would have to an original painting. Nobody
denies the educational value of the photograph or the lithograph,
but who would argue that it was the same thing as the original,
and who would look at it if the original itself could be seen?
(It is assumed that the translation is at least a fair copy, not, as is
unfortunately so often the case, a ridiculous caricature.)
To argue that the English version of any vocal work is equal
to the original is a mere waste of time. Such a contention implies
that the setting fits at least two sets of words equally well, which
at once eliminates it as a true work of art. (This statement does
not attempt to undervalue the melodic beauty of such tunes as
those of the old folk-songs, which have been fitted with widely
differing sets of words. It merely contends that no setting of this
kind may truly be called a work of art, as it is entirely lacking in the
quality of "inevitableness." A fine melody is not necessarily an
artistic setting.)
There is a vast difference between setting music to English
words and adapting an English translation to music already
fitted to the words of another language. The discussion of English
as a fit medium for song really has nothing to do with the question.
Even supposing that English were a more singable language than
Italian or French or German (which, especially in the case of the
first two, would be extremely difficult to prove) the translation and
consequent changing of a work of art would not on that account
be excusable. For it may be taken for granted that the deliberate
changing of a work of art is to be condoned only on extraordinary
grounds. And the only valid ground for translating the text of
any vocal music is increased intelligibility, implying greater
educational possibilities.
It would be folly to argue that there is no real difference
between an original Italian, French or German text and an English
translation. No matter how careful and resourceful the translator,
the gulf between the two is still impassable. And when the words
of a Romanic language have to be reproduced in a Germanic, as in

This content downloaded from 143.106.201.145 on Thu, 31 Oct 2013 16:07:52 PM

All use subject to JSTOR Terms and Conditions

Translating to Music

293

translating from Italian or French into German or English, the

gulf is all the greater.
In representing either the linguistic subtleties of the Romanic
languages or the broad vowel sounds of the Teutonic, English
suffers from peculiar limitations. Our vocabulary, to be sure, is
very large, and seems to contain a word or phrase for every
possible thought. But a great part of these resources cannot be
utilized in a fixed musical setting because of the rigidity of the
English accent. In French, where there is rarely any marked
accentuation, and in Italian or German, where every letter and
every syllable is regularly given its full phonetic value, it is a
comparatively easy matter to achieve correct musical declamation,
i. e. the exact correspondence of the musical accent with that
of the spoken word. But in English a word rarely has more than
one syllable which may safely receive a decided musical accent.
And there are many words, particularly monosyllables, which are
utterly incapable of sustaining any musical accent whatever.
Proceeding from the mere rhythm to the general sound and
the mental effect of words, the limitations of the English language
are still found to be great. It is significant that we make a very
clear distinction between "commonplace" and "poetical" words.
Many a word is barred from English poetry as being too "prosaic."
In translating an operatic libretto, "poetical" words and phrases
have to be substituted for the everyday language of the original.
"But," objects the propagandist of "opera in English," "if
our literal translations of certain passages sound crude and banal,
does not the original have exactly the same effect on those who
are familiar with the language?"
The answer to this question is to be found in European
poetry. There the distinction between the poetical and the
commonplace is never clearly defined. The Latin imagination is
able to supply what the spoken or written word merely suggests, and
the crudest realism is easily invested with romance. The greatest
poems are often expressed in the simplest words. So the foreigner,
listening to operatic banalities which would make the literal AngloSaxon shudder, glorifies them by the power of his imagination.
For him there is no sudden leap from the sublime to the ridiculous.
But the Anglo-Saxon imagination, if it exists at all, is self-conscious and analytical. It finds it difficult to build up an illusion,
and even when it succeeds there is constant danger of a sudden
collapse. That is why it insists upon "poetical" terms in its
operatic translations, placing one more limitation upon the already
harassed English language.

This content downloaded from 143.106.201.145 on Thu, 31 Oct 2013 16:07:52 PM

All use subject to JSTOR Terms and Conditions

294

The Musical Quarterly

Finally must be considered the various effects of sound in the

foreign languages whose exact imitation is practically impossible.
Not only the many differences in the pronunciation of both
vowels and consonants, but also the varieties of rhyme, alliteration,
onomatopoeia, etc., must be taken into account. Wagner is
especially fond of onomatopoeia, frequently using German words
whose actual sound suggests their meaning. Alliteration is never
easy to imitate. As for rhyme, the European fondness for combinations of two or more syllables has been the bugbear of many a
translator, for, aside from the present participle and words with
such suffixes as -tion, there is a poverty of double rhymes in the
English language.
Admitting, then, that no English translation, however good,
can possibly hope to equal the original, the whole problem becomes
merely a question of how much artistic inferiority may be overlooked
for the sake of increased intelligibility. And this question, in all
fairness, must be applied to each vocal composition, separately;
its answer will depend largely on the skill of the translator. It
may be interesting, therefore, to examine in detail the problems
with which one is confronted in attempting to fit an English
version of a foreign text to a preconceived musical setting.
Correct declamation is the first duty of the translator. There
is no excuse for the slightest flaw in accents. That some great
song-writers have erred in this respect, even in their original
settings, has nothing to do with the case. The song-writer sets
his text according to his own conception of its metre and accent.
But the translator has before him a series of definite rhythmical
combinations, to which the natural accents of his English words
must be fitted exactly.
It is unnecessary to enter into the details of faulty accentuation that are to be found in published translations. Hundreds of
examples could be pointed out, merely by running through a few
collections of songs and half a dozen operatic scores. The mistakes
include not only the emphasizing of non-accented syllables, but
the placing of sustained and important notes upon such trifling
words as the definite and indefinite articles, monosyllabic conjunctions, prepositions, etc. The writer has yet to see or hear an
operatic English translation that is not full of false accents.
Almost as important as the purely mechanical problem of
declamation is that of dramatic emphasis. Here the insight of the
translator into the intentions of the composerisanimportantfactor.
He must constantly ask himself why certain words are peculiarly
suited to the musical expression that has been given them, and

This content downloaded from 143.106.201.145 on Thu, 31 Oct 2013 16:07:52 PM

All use subject to JSTOR Terms and Conditions

Translating to Music

295

he must try to produce similar if not identical combinations in

his translation.
An illustration will make this point clear. In Schumann's
"Widmung," the second line of the poem is "Du meine Wonn',
o du mein Schmerz." The composer has beautifully expressed the
meaning of the line by rising to a joyous note on the word " Wonn',"
and then descending in regular progression to a contrasting low
note on "Schmerz." The meaning of the two sustained tones in
the musical phrase (coming on "Wonn'" and "Schmerz") is
unmistakable. The same effect can be procured in translation
only by giving the first note a word expressing joy and the second
a slXarply contrasting word indicating grief, pain, or sorrow.
Yet at least two translators have rendered the line, "Thou all my
joy and sorrow art." The flat, meaningless verb "art," in other
words, is expected to fit the poignant note which Schumann has
given to the dramatic German word "Schmerz"!
Another example of lack of insight on the part of a translator
is to be found in Schumann's "Die Beiden Grenadiere." The
line "Sie liessen die K6pfe hangen," is dramatically expressed
by a series of descending tones, clearly suggesting the drooping
heads of the dispirited grenadiers. But this effect is entirely lost
in the version "Their hearts were depressed and aching." That
the line is in no sense a literal translation makes little difference.
(Literalness is one of the last things to be considered in translating
to music.) But to miss the entire dramatic effect of a line, to give
no hint of the hanging of heads, when both the German text and the
music definitely express this thought, is inexcusable.
In one of the first lines of Schubert's "Wanderer" are the
words "Es braust das Meer," whose meaning is dramatically
suggested by the music. One translator gives as his version of the
phrase, "The ocean foam"! Here is no hint of the force of
"brausen," a very expressive German verb.
In his famous setting of the "Erlkonig," Schubert makes use
of a number of dramatic devices which the careful translator
must imitate if possible. The terror-stricken cry of the child,
for instance, is intensified in the phrase "Siehst, Vater, du den
Erlkinig nicht?" by the position and sound of the word "nicht."
But this phrase has been translated, "Seest, father, thou the
Erlking, alas!" A pitiful subterfuge, even supposing that an
actual child ever could have used such an exclamation as "alas!"
When the music, as is often the case, gives a direct imitation
of the sense of the words, then the translator's version must be
even more carefully chosen. Two examples will suffice. Debussy,

This content downloaded from 143.106.201.145 on Thu, 31 Oct 2013 16:07:52 PM

All use subject to JSTOR Terms and Conditions

296

The Musical Quarterly

in the song "C'est l'extase langoureuse,"gives a direct imitation

of the rolling sound of pebbles under water, in setting the phrase
"Le roulis sourd des cailloux." The translation, "Where waves
on stormy shoresdie," scarcelymakes the point clear.
For a second example we may turn again to Schumann's
"Grenadiers." The line "Viel' Schwerter klirren und blitzen"
contains two strongly onomatopoetic words ("klirren" and
"blitzen") which the composerhas set to music directly imitating
the clashing and flashingof swords. One translatorgives the line
as "And swords clash and muskets rattle," a version which commits nearly every possible sin against musical translation. The
declamationis bad, with the important word "clash" actually on
an unaccented note, while the second syllable of "muskets" is
awkwardlydivided, and the unvocalic "le" of "rattle" comes on a
fairly sustained tone. The translationis not only far from literal
but introduces the entirely extraneous idea of "muskets." But
worstof all, it absolutelymissesthe dramaticimitation of the music

on those splendid words "klirren" and "blitzen," and therefore

completely frustrates the genius of the composer. Another

version, "While swords with clash are descending,"offerslittle in
the way of improvement,the final word,in particular,being utterly
unworthy of the original.
Thus far there have been mentionedonly the demandswhich
may fairly be made of the translatorby the composer. But the
singer must also be considered,at least in so far as he was consideredin the originalsetting. Translatorsshould make it a rule
to reproduceas far as possible the important vowel sounds of the
originaltext. For the convenience of the singer is generally regarded by the conscientiouscomposer who writes with the final
effect of the song in mind. A soft high note, for example, may be
set to an ee sound, making a smooth head-tone an easy matter.
On the other hand, if a full chest-toneis desiredthe average singer
would much preferan ah or an ay. In general the rule for the
translatormay be summed up in the words "Imitate the sounds
of the originaltext as far as possible, so that the translationmay
'sing' like the originalsong."
But the poet also has his rights, and the most conscientious
translators try as a rule to imitate also the original schemes of
rhyme and metre. This is often so difficultthat it can be accomplishedonly by the sacrificeof sense, accent, or dramaticemphasis.
If the translatorhas to choosebetweenthese evils, it is far better for
him to sacrificethe original verse-formthan any of the features
directly affecting the musical setting. Often a metrical scheme

This content downloaded from 143.106.201.145 on Thu, 31 Oct 2013 16:07:52 PM

All use subject to JSTOR Terms and Conditions

Translating to Music

297

cannot be exactly reproduced,owing to the liberties taken by the

composerand the differencebetween English and foreign accents.
When rhymes are emphasizedby the music, the translator can
hardly afford to omit them. But usually a modification is
permissible,as, forinstance,in the four-linestanzaswith alternating
rhymes(very common in French songs) where a rhyming of the
secondand fourthlines is quite sufficientin translation. The ideal
translator, of course, would be expected to make his version not
only "sing" like the original song but also "read" like the
original poem.
Last of all comes the question of the actual meaning of the
text. The average observer, to whom a translator is primarily a
linguist, may argue that this question should stand first instead of
last. To be sure, it is taken for granted that every translation will
reproduce in some fashion the meaning of the original. But in
respect to the accuracy of details, translating to music is a very
different matter from translating in general. The distinction
becomesclear when one reflects on the obvious differencebetween
a "readingtranslation,"for programpurposesonly, and a version
that is intended to be sung. In the former,every detail of idiom,
metaphor,poetical word-play,etc., should be reproduced,while in
the latter the demands of the musical setting must be supremein
importance.
In general it is safe to say that a literal, word-for-word
translationis never possibleor even desirable. The real meaningof
a French or Germanphrase may often be entirely obscured by a
literal translation. The translator should read the original poem
until he is thoroughlyimbuedwith its spiritandgeneralsignificance,
after which he may rewrite the entire text in his own words,
concentrating his mind upon the more important matters noted
above.
This has generally been the most satisfactory method of
procedure. Beginning with a clear conception of the general
thought to be expressed,the careful workmanfirst makes sure of
the vital points of accent and dramatic emphasis, gradually introducing also the chief vowel-soundsand the leading features of
the original rhyme-scheme. If he is able to include so much, he
is extremely fortunate and may well be satisfied even if not a
single word or phrase has been translated literally. The latter
possibility depends entirely upon the nature of the original text
and the language in which it is written. It is mucheasierto make
an English translationfromthe closely related German,forexample,
than from the French or Italian languages.

This content downloaded from 143.106.201.145 on Thu, 31 Oct 2013 16:07:52 PM

All use subject to JSTOR Terms and Conditions

298

The Musical Quarterly

The ideal musical translator is not merely a linguist, but a

poet and a musician as well. If all these qualities are not to be
found in one man, it would be better for severalindividuals to
collaborate. This, in fact, is now very often done. The late
Henry Grafton Chapman, whose translations are remarkablefor
their poetic touch and their thorough sympathy with the spirit
of the author,never hesitatedto consult with singersand composers
in order to make his work acceptable in its musical as well as its
literary quality.
If a poet could find the inspiration for a sonnet in reading
Chapman's translation of Homer, why should not the modern
Anglicizer of vocal music be ambitious for the future verdict
"He was a great interpreter"? Perhaps if the thoughtless propagandistsof "singing in English" and the carelesscritics of translations in general realizedthe stupendous difficultiesfaced by the
translatorof even the simplest song, they would be more inclined
to treat with respect and consideration a work worthy of the
most inspired poet-musician, instead of relegating it to the
underworldof grubbinghacks and soulless versifiers.

This content downloaded from 143.106.201.145 on Thu, 31 Oct 2013 16:07:52 PM

All use subject to JSTOR Terms and Conditions