Rubbio, rubeo, rossetto, rubeolo, rubescente, rubecchio, rossino, rubefacente, suggeriva Roberto. Nein, Nein, si irritava padre Caspar. E Roberto: come una fragola, un geranio, un lampone, una marasca, un ravanello . . .
Weaver’s translation reads:
Ruddy, ruby, rubescent, rubedinous, rubent, rubefacient, Roberto suggested. Nein, nein, Father Caspar became irritated. Roberto went on: like a strawberry, a geranium, a raspberry, a cherry, a radish . . .
When checking Weaver’s translation I did not realise that he was quoting only six English colour terms instead of my eight Italian terms. Probably he could not find more words for red and thought that six were enough to suggest the impatience of Roberto. Lozano too had to surrender because, as she says,3 in Spanish she only found six terms. Moreover in Spain in the seventeenth century a geranium was called a pico de cigüeña (crane bill) and – beyond the fact that this term is now obsolete and incomprehensible to modern readers – its use risked introducing an ‘animal’ element into a ‘vegetal’ series. Thus she substituted the geranium with another flower, equally red:
Rojo, rubro, rubicundo, rubio, rufo, rojeante, rosicler, sugería Roberto. Nein, nein, irritábase el padre Caspar. Y Roberto: como una fresa, una clavellina, una frambuesa, una guinda, un rabanillo . . .
Let me praise Schifano for having found eight terms: rouille, rougeâtre, rubis, rubicund, rougeaud, roussâtre, rocou, rubéfié.
To come back to coral, and featuring that in a given language there weren’t, let’s say, enough terms for yellow, I encouraged my translators to change colours freely when they ran out of words. In those seas, coral and fish come in all colours, and that a given coral or fish was red or yellow was not important; I repeat, what counted was that the same term would not be repeated in the same context and that the reader, like Roberto, experienced a great chromatic variety through a linguistic variety.
So when I said that si vedeva il fegato poroso color colchico di un grande animale (that is, in my translation, there was a sort of porous liver the colour of an autumn crocus) I left the colour undetermined, since crocus petals may be yellow, lilac or whatever. Weaver chose saffron, Kroeber lilarote.
When I speak of polipi soriani (where soriani evokes the striped coat of a tabby cat), Weaver speaks of cypress-polyps, Lozano of polipos sirios, and Schifano of polypes ocellés (ocellated). The German translator found a more literal solution by speaking of getigerte Polypen. But when, a little later, I speak of tuberi tigrati di ramature negricanti, and the translators used striped tubers, tubercules tigrés and raigones listado, Kroeber, having already called the polyps getigerte, was then obliged to change both form and colour for my tubers, speaking of gelblich geflammete Knollen schwärzlichen Astwerks, that is, more or less, of black branches, which show yellow protuberances.
In all these examples I invited the translators to disregard the literal sense of my text in order to preserve what I considered to be the ‘deep’ one, or the effect it had to produce.
Would we say that my translators (with my approval) have changed my text? We certainly would. In spite of this, these translations say exactly what I wanted to say, that is, that my three characters were sick of literature, and that the coral of the Southern Seas is incredibly marvellous – and a literal translation would have made these effects less perspicuous.
However, we must admit that these translations are referentially false.
Thus, to preserve the effect of the text, translators were entitled not only to make radical changes to the literal meaning of the original text, but also to its reference – since in Italian Diotallevi is said to have seen a hedge, Casaubon to have said non spirava un alito di vento, Roberto to have seen such and such colours and forms, while in other languages this is not the case. Can a translation preserve the sense of a text by changing its reference?
No one would allow a French translator of Hamlet to write, for instance, that Hamlet, instead of seeing the ghost of his father, saw the ghost of somebody else. Obviously, one could say that Shakespeare cannot be changed, not even to help his foreign readers to understand certain situations, while Eco’s translators can do what they want. But we have seen that it is not so wrong, in an Italian translation of Hamlet, to mention a topo (that is, a mouse) instead of a rat. Yet such a variation becomes highly relevant, as we have seen, in a translation of Camus’s La peste.
Surface and deep stories
If, in order to preserve the deep sense of a text, a translation can change its references, to what extent are these changes possible? We have to reconsider the distinction between story and plot I mentioned in my first essay, as well as the nature of the reading process, by which every sentence or sequence of sentences conveying a story can be summarised (or interpreted) by a micro-proposition and several micro-propositions can be summarised by a more comprehensive macro-proposition.
Which were the ‘real’ stories told by the pages of my novels? The fact that Diotallevi saw a hedge or the fact that he was a sort of culture snob, able to perceive nature only if filtered through poetry? The fact that Roberto saw striped tubers instead of black branches with yellow protuberances, or the fact that he was amazed by an inexhaustible quantity of colours and forms and realised he was (as the title of the chapter suggests) in a Garden of Delights?
In order to make the ‘deep’ story of a chapter or of an entire novel detectable, translators are entitled to change several ‘surface’ stories.
For instance, the page on coral from The Island of the Day Before can be summarised as:
- Roberto is swimming on the coral reef.
He admires an immense variety of forms and colours.
These micro-propositions can be embedded in larger macro-propositions and the whole chapter could be summarised as: - Thinking that his brother is on the island, Roberto decides to learn how to swim.
In the course of his efforts, day by day, he discovers the marvels of the coral reef.
Among the coral he finds a sort of mineral form that he picks up as thinking it is the skull of Father Caspar.
The whole novel could be summarised by a hyper-macro-proposition that reads: - Roberto is shipwrecked on an abandoned boat near an island.
He is west of the 180° meridian, the island is east.
He will never reach it, and thus he fails to live in the day before.
Given that stories are embedded in this way, to what extent are translators entitled to change a surface story in order to preserve a deep one?
It is clear that every single text permits a different and individual solution. Common sense suggests that translators can change Roberto saw a striped polyp into Roberto saw an ocellated polyp but they certainly are not allowed to change the global macro-proposition (3) by writing, let us say, that Roberto reaches the island.
A first hypothesis is that one can change the literal meaning of single sentences in order to preserve the sense of the corresponding micro-propositions, but cannot alter the sense of major macro-propositions. But what about the many intermediate ‘shallow’ stories (between the literal meaning of single sentences and the global sense of an entire novel)? What if a translator decides to change a joke, or a play on words impossible to translate, by assuming that the sense of the story is not that a character told that joke but that he or she told a joke?
It is on the basis of interpretative decisions of this kind that translators play the game of faithfulness.
Radical rewriting
In my last essay I considered cases in which a partial rewriting (like the substitution of Baroque verses in Lozano’s translation) aimed at keeping the general textual effect. Let me now consider cases in which, in order to compensate for a loss, the operation of rewriting substantially alters the reference of the original text – which is profoundly transformed, disregarding the matter of semantic equivalence, in order to play the same game with the target text that the author played with the source text.
A typical case of radical rewriting is the Italian translation of James Joyce’s ‘Anna Livia Plurabelle’, part of Finnegans Wake. This translation originally appeared under the names of Nino Frank and Ettore Settanni (who certainly contributed to the work) but must be considered as the work of Joyce himself,4 as was the French translation, which was made by Joyce in collaboration with Beckett, Soupault, and others.5
Finnegans Wake is written in ‘Finneganian’, which could be defined as an invented language. As a matter of fact it is a multilingual text, but a multilingual text written by an English-speaker. Thus Joyce’s decision was based on the idea that the Italian and the French versions had to be multilingual texts written by an Italian or a French speaker.
We shall see in my next essay that Humboldt suggests that translating means not only leading the reader to understand the language and culture of the original but also enriching one’s own language. We can thus say that for Joyce the Italian and French translations of Finnegans Wake were conceived in order to lead the French and Italians to express what they were unable to express before (just as Joyce did with English).
Here is an example of rewriting that pushes the limits of the original creation:
Tell us in franca lingua. And call a spate a spate. Did they
