Such traces of a later text are present in a particular earlier text by virtue of both texts’ participation in a textual system and by way of our reading. But there is one way in which these traces are sometimes made more materially present, and that is when reading becomes (literally) writing, that is, in a translation. Translations themselves are, of course, of necessity later than the work they translate, and it is to this work that they have the strongest intertextual (or, in Gerard Genette’s terms, hypertextual) relation. There is no reversal here. But translators in their belatedness may read and therefore write a work as referring to still other texts, some of them not yet written at the time of the source text. In the world of translation, not only may Proust be found (or made) to quote Shakespeare in his title, rendered by Scott Moncriev as Remembrance of Things Past; Homer can rephrase a line from The Godfather, Aeschylus can echo Edgar Allen Poe, and Euripides can quote an African-American spiritual.

As with instances of intertextuality in general, anachronistic quotations will only play a role in interpretation where the reader knows the quoted text. Many readers will presumably take Robert Fagles’s Odyssey to be alluding to The Godfather when Hephaestus says (8.401 in the translation, 8.358 in the original): ‘Now there’s an offer I really can’t refuse!’ Fewer will realize that the opening words of Stanley Lombardo’s Odyssey, ‘Speak, Memory’, are borrowed from the title of Nabokov’s memoir, and fewer still that Lombardo is quoting Thomas Nashe in a simile at Iliad 16.306–9 (16.297–300 in the orginal). Sometimes a translator supplies the necessary information. In her preface to Hecuba, Marilyn Nelson tells us she has added two allusions in the chorus’s concluding lines (1852–68), one to Robert Hayden’s poem ‘The Middle Passage’, and one to the spiritual, ‘Nobody knows the trouble I’ve seen’; she also tells us why she has done so.

. . .

It may also be a combination of text and context that marks what is said as quotation. Is Gilbert Murray’s Aeschylus quoting Poe’s ‘The Raven’ in the second stasimon of the Agamemnon (which describes Menelaus’ longing for Helen) when we find a repeated ‘nevermore’ for the Greek (‘not afterwards’) in lines 420–26?

. . .

It can of course be argued that this sort of translation by quotation is not radically different in its slightly surreal anachronism from translation itself; after all, if Homer can speak twentieth-century English, why can’t he quote Nabokov? Or—supposing we accept that the shift of language per se is by convention to be ignored—if the Greek tragedians can, in Murray’s versions, write choruses in the style of Swinburne, then why can’t they allude to Poe? A less perverse way to ask this question might be: if we can use all the linguistic and stylistic resources of our time to express what an earlier author says, why should the words of an author known to our time be excluded? But in fact such anachronistic quotations or echoes seem to lack the transparency (or what some would call the illusion of transparency) we otherwise grant to language and poetic idiom.

It therefore seems more natural to find such quotations in works that otherwise play freely with anachronism—such as many twentieth-century translations of comedy—or that declare themselves to be something other than translation. Mahon’s play, for example, is entitled The Bacchae: after Euripides, and we find in it not only other quotations from twentieth-century sources (‘It’s still the same old story a fight for love and glory’) but twentieth-century place-names (‘Saudi deserts’), terminology (‘clinical psychosis’), and sound effects (‘Pow!’). Nelson’s Hecuba, described as ‘rendered freely into verse’, evokes in a number of places the parallels between the condition of the Trojan women and chattel slavery in the Americas. The presence or absence of anachronistic quotation has sometimes been taken as an index of the sort of version we are dealing with.

. . .

The effects of such intertextuality are varied. In translations that freely engage in anachronism—such as many versions of comedy in the past few decades—the quotations, like other modes of anachronism, seem to serve primarily though not exclusively to bring the work into the translator’s own time, to make its preoccupations and its sense of humour accessible to contemporaries by not only speaking their language but also citing what they have read. Other translations—such as Nelson’s Hecuba and Mahon’s Bacchae—use anachronistic quotation to suggest a connection between the events of the original text and a later cultural context not necessarily contemporaneous with the translation (America in slavery times, the Second World War). Since something long ago is thereby identified with something recent, the anachronism serves in a sense both to mark and to collapse (for the reader) the difference in time.

. . .

Anachronistic quotation, then, may ignore chronological sequence, joke about it, or collapse it in one of several ways. The conceptual and functional distinctiveness of these modes should perhaps make us wary of supposing that we have here a single literary practice, but in every instance the quotation entails some sort of play with time (whether overtly thematized or not), and in most it tends to close up the temporal distance between ancient writer and modern readers.

. . .

On the face of it, archaism seems almost the reverse of anachronism. Where anachronistic quotation suggests either a disruption or an erasure of chronological sequence, the use of archaism underscores the antiquity of the source text relative to its translation, the distance of the original from its modern readers. It is the expression of this distance—or rather of some facsimile of it—that occupies many translators in the Victorian period, the heyday of archaizing translation.

. . .

We may, however, see archaism in translation as akin to, or as almost another species of, anachronistic quotation. Both, after all, entail a further element of intertextuality beyond the translation’s relationship to its original. The literariness of archaizing language is inescapable: if you are to translate a text into an earlier version of your own language, you can only have acquired that earlier version from other texts.

. . .

These two types of translation—archaizing proper, which seeks to translate the old as old, and archaizing appropriately, which seeks to translate the old into a suitable idiom which is itself old—may of course coexist in the work of a single translator. They may also both be made to serve ancillary purposes as well. Two examples will suffice, one from each of the theorists I mentioned earlier. As Steiner notes, there are times when archaizing translation serves to assist a reappropriation of an earlier period in the target language. Emile Littre, for example, translated Homer into medieval French not just because the French of that period seemed to him to suit Homer, or in order to suggest the antiquity of the Homeric poems, but because he wanted to demonstrate that medieval French was itself capable of representing the greatest of the ancients and was thus worthy of a respect that had been denied to it.

. . .

For some translators, then, particularly in the Victorian period, the Classics are always (though in a variety of senses) old, and should be so rendered; but to others, both before and after, they are (for various different reasons) new, and our translations must keep them so. We may also find translators who have tried, as it were, to deal with the problem of the Classics—being by definition somehow both old and new—by making use of what we might call a peculiarly privileged archaic intertext, the King James Bible.