I think that something similar happens when – in translation of verbal texts from one language to another – one tries to produce the same effect with two different substances of expression.
In cases of rewording, like definition or paraphrase or inference, where the content is interpreted in a more detailed way, we can say that:
LS1/C1 → LS2/C1a where C1a > C1
A Linguistic Substance1 that conveys a Content1 is transformed into a Linguistic Substance2 that conveys a Content1a where Content1a is > than Content1 (and I use > not in a strictly logical sense but as shorthand in order to say ‘the same content but more detailed’, as when I defined cocaine as an alkaloid).
Substance in translation
In elementary processes of translation (for instance, by satisfactorily translating I am eating a piece of bread as Sto mangiando un pezzo di pane or Je suis en train de manger un morceau de pain) a Linguistic Substance1 that conveys a Content1 is transformed into a Linguistic Substance2 that aims at conveying the same Content1
LS1/C1 → LS2/C1
Usually we are not overly interested in the fact that a Substance1 is transformed into a Substance2 because in cases of elementary translation (as in rewording) we are mainly interested in grasping the content. But we have felt disturbed when we have tried to translate a few lines from Hamlet into definitions and paraphrases. This means that for certain texts there are many stylistic features such as, for instance, metre, or the sound of words, or the brevity of an exchange, that are independent from the structure of a given language (a hendecasyllable displays the same features both in English and in Italian) and which must be respected by a translation.
I have suggested that these texts are those that we consider as endowed with aesthetic qualities. However, let us take a step backwards. Is it certain that changes in substance are absolutely irrelevant as far as non-poetic, trivial and everyday expressions are concerned? We have seen in one of my previous essays that even when translating the French expression mon petit chou we feel that we should preserve certain qualities of the uttered substance of the expression, but this requirement holds even for the most utilitarian expressions.
We know that when an English text is translated into French or Italian or German it becomes a little longer, even if the translator is able to keep the same number of words, because English certainly has more monosyllabic words than, let us say, German. Italian words are very long but less so than many German ones, while German uses fewer words than Italian or French because it can resort to many compounds. Such differences can be qualified to such an extent that typographers and publishers are able to calculate how to compose parallel texts in different languages.
Let me consider the first paragraph of the second page of the User’s Guide to the musical instrument Casio CTK-671. The English text reads:
384 tones, including 1000 ‘Advanced Tones’.
A total of 238 standard tones, including piano, organ, brass, and other presets provide you with the sounds you need, while memory for 10 user tones lets you store your own original creations. 100 of the present tones are ‘Advanced Tones’, which are variations of standard tones created by programming in effects (DSP) and other settings.
Three versions, Italian, French and German follow. In this table I registered first the number of words and then the printed lines for every language:
It is evident that the English text is the shortest, that the German one uses fewer but longer words, and that the Italian text is a little longer than the French one. Considering all the lines of the whole of the second page, the results are:
– which means that the French, Italian and German texts shift to page 3. With unimportant variations, more or less the same ratio is respected in the following pages of the manual. If by chance the German text were sixty lines long, a typographer who did not know German might assume that it was not a translation but rather a paraphrase of the English one.
Probably nobody would protest if the Italian instructions for the Casio CTK-671 were three pages longer than the English. However, although it is admitted that in translation proper the substance of the expression changes – since we are shifting from the sounds of one language to the sounds of another – even in the most practical cases there is a sort of implicit stricture by which a certain ratio between substances must be respected.
This encourages me to say that every translation proper has an aesthetic or poetic aspect. If to interpret always means to respect the spirit (allow me this metaphor) of a text, to translate means to respect also its body.6
Naturally this duty of respect becomes mandatory in every text with an aesthetic aim. But we can say that, from the translation of the Casio manual to the translation of Shakespeare’s sonnets, there is a continuum of possibilities of respecting the substance.
Substance in translating poetry
Now, it is universally acknowledged that in translating poetry one should render as much as possible the effect produced by the sounds of the original text, even though in the change of language a lot of variations are unavoidable. One can miss the real body of a discourse, but try at least to preserve, let us say, rhythm and rhyme.
So in these kinds of translation we have a process of this kind:
LS1ES1 / C1 → LS1a ES1a / C1a
where not only a Linguistic Substance1 but also many Extra-Linguistic Substances1 conveying a Content1 are transformed into a Linguistic Substance1a and Extra-Linguistic Substances1a supposed to be aesthetically equivalent to the source ones, and conveying a Content1a aesthetically equivalent to the source one.
What does aesthetically equivalent mean?
In his beautiful book Le Ton beau de Marot7 Douglas Hofstadter considers different translations from Dante Alighieri’s Divine Comedy, starting from the principle that the most important stylistic and metrical feature of that poem is the invention of the tercet – that is, strophes of three verses each, all in hendecasyllables, and with the rhyme-scheme ABA, BCB, CDC and so on. Hofstadter, to stress the quasi-musical structure of a series of tercets, provides this graphic representation:
It’s evident that such a form is independent of any natural language and can be embodied in different linguistic substances. Hofstadter takes the opening tercets of the Canto Terzo:
PER ME SI VA NE LA CITTÀ DOLENTE,
PER ME SI VA NE L’ETTERNO DOLORE,
PER ME SI VA TRA LA PERDUTA GENTE.
GIUSTIZIA MOSSE IL MIO ALTO FATTORE:
FECEMI LA DIVINA PODESTATE,
LA SOMMA SAPIENZA E ‘L PRIMO AMORE.
DINANZI A ME NON FUOR COSE CREATE
SE NON ETTERNE, E IO ETTERNO DURO.
LASCIATE OGNE SPERANZA, VOI CH’INTRATE.
Queste parole di colore oscuro
vid’io scritte al sommo d’una porta;
per ch’io: «Maestro, il senso lor m’è duro»
and starts examining some English translations of these verses. For instance Robert Pinsky8 renders them:
THROUGH ME YOU ENTER INTO THE CITY OF WOES,
THROUGH ME YOU ENTER INTO ETERNAL PAIN,
THROUGH ME YOU ENTER THE POPULATION OF LOSS.
JUSTICE MOVED MY HIGHER MAKER, IN POWER DIVINE,
WISDOM SUPREME, LOVE PRIMAL. NO THINGS WERE
BEFORE ME NOT ETERNAL; ETERNAL I REMAIN.
ABANDON ALL HOPE, YE WHO ENTER HERE.
These words I saw inscribed in some dark color
Over a portal. ‘Master’, I said, ‘make clear
Their meaning, which I find too hard to gather.’
Then he, as one who understands: ‘All fear
Must be left here, and cowardice die. Together . . .
Hofstadter comments: ‘It pains me to break things off so abruptly here, especially to leave unclosed the pair of quotation marks opened near the bottom, but what can I do? After all, Pinsky chose to redistribute Dante’s semantics quite radically across tercets. Indeed, in this canto Dante has forty-five tercets, Pinsky has but thirty-seven. The aesthetics of this decision boggles my mind’ (p. 533).
In a collection of translations edited by Daniel Halpern9 Hofstadter picks up the translation by the great Irish poet Seamus Heaney:
THROUGH ME IT LEADS TO THE CITY SORROWFUL.
THROUGH ME IT LEADS TO THE ETERNAL PAIN.
THROUGH ME IT LEADS AMONG THE LOST PEOPLE.
JUSTICE INSPIRED MY MAKER ABOVE.
IT WAS DIVINE POWER THAT FORMED ME,
SUPREME WISDOM AND ORIGINAL LOVE.
BEFORE ME NO THING WAS CREATED EXCEPT THINGS
EVERLASTING. AND I AM EVERLASTING.
LEAVE EVERY HOPE BEHIND YOU, YOU WHO ENTER.
I saw these words inscribed above a gate
in obscure characters, and so I said,
‘Master, I find their sense hard to interpret.’
Hofstadter complains first of all that Heaney does not preserve either the metre or the rhyme, but is very severe also on many other counts:
In favour of Heaney, I can say that he renders a tercet by a tercet. It’s good to see nine lines of gate inscription, instead of just seven. But, I regret to say, there’s not much more praise that I can offer. Look at the first line: ‘Through me it leads to the city sorrowful.’ ‘It’?! What is this ‘it’? And ‘city sorrowful’ is pretty sorrowful. If this line had been written by a high-school student, I would have struck it out in bright red ink and said to start again from scratch. To my ear, the sentence doesn’t even sound like it was written by a native speaker! (p. 535).
Then Hofstadter considers a translation by Mark Musa,10 who did not use rhymes, claiming: ‘My main reason
