History or Novel?: The first translation of Persiles y Sigismunda
It has to be said, and with regret, that Cervantes’ Persiles y Sigismunda is no longer a best-seller, even in Spain, and in the United Kingdom it is little known outside of university Spanish departments.
But it was not always like this. As is well known, Cervantes completed this, his last work, four days before his death in
1616. It was published in 1617. In 1618 it was translated into French by François de Rosset1 and later in the same year, Matthew Lownes went to the Stationers’ Hall in London and paid to have his interests registered
in an English translation. That translation appeared under the title, 'The Travels of Persiles and Sigismunda. A Northern
History.'2 From the title-page, we can see that the book was printed for Matthew Lownes by his brother, Humphrey, and was sold at their
shop in St Paul’s Churchyard in London. Together with Barry Ife, I have been interested in this translation for several years,
for purely practical reasons: there are very few English translations of the Persiles, and this first one is still, I think, the best, even though it has never been reprinted. So what follows are some thoughts
which might go towards an introduction to a modern edition, and they were prompted by a recent conference organised by Barry,
Jenny Mander and Nicholas Cronk in Oxford on The Rise of the European Novel.
The sub-title of the English Persiles, 'a northern history' was no more than a literal translation of Cervantes’ own sub-title,
historia setentrional. But language has shifted since 1619 and 'history' has taken on a more restricted meaning; today, historia is more likely to be translated as 'novel', a concept which did not exist in English culture in 1619. Now, the question of
what constitutes a novel occupied the scholars at Barry’s conference for four days of intense debate, and this is not the
time or place to rehearse the arguments: you will have to wait for the Actas to come out. For this evening’s purposes, I am taking a deliberately simplistic approach by working to the The Oxford English Dictionary's definition. According to OED, a novel is, ‘A fictitious prose narrative or tale of considerable length (now usually one long enough to fill one or more
volumes), in which characters and actions representative of real life of past or present times are portrayed in a plot of
more or less complexity.’ The earliest citation in OED is 1639, but OED is notoriously imprecise when it comes to dates of first usage. It records ‘novels’ in the plural, meaning short stories,
as being in use from the mid-sixteenth century, but the English Persiles, at 399 quarto pages, is on a much grander scale than this. Before the story even starts, then, translation leads us into
an ambiguity, and this is something to be kept in mind.3 Over the years, hispanists have been reluctant to call the Persiles a novel, preferring to label it a “romance”, the formulation adopted by Forcione for his seminal study of 1972 on Cervantes’ Christian Romance, or a “prose narrative” -- a term which I admit to having used myself. But I am now satisfied that the Persiles satisfies the dictionary’s definition of a novel, and indeed, in the wake of the Oxford conference, the eminent critic Diana
de Armas Wilson has indicated that this is the position which she has now adopted.
So in 2001, de Armas Wilson and Darby think that Cervantes’ last work was a novel. But to return to the ambiguous title, what
did the English readers in 1619 think that they were buying from the Lownes brothers’ bookshop? Pure fiction? A factual account?
Something which was a bit of both, which one might say was how much history of the period was written? Perhaps we should begin
by reminding ourselves of the main features of the narrative.
The narrative of the Persiles is arranged into four books and begins in medias res, with a ritual of human sacrifice which Diana de Armas Wilson believes is heavily influenced by the literature of discovery,
and particularly the writings of the Inca Garcilaso. Another obvious influence on the structure of the first two books is
classical epic, particularly the Odyssey and the Aeneid. The first two books of the Persiles are a maritime journey through strange lands, where the Baltic and North Seas replace the Mediterranean; and the home for
which the travellers seek is a spiritual home, in Rome. There are witches and shape-changing werewolves, reminiscent of Circe;
an athletics contest, which mirrors the classical funeral games, and a king’s daughter who, like Dido, falls in love with
a stranger who comes from the sea and then deserts her. Like Homer’s Odysseus, Cervantes’ Periander (the English form of Periandro) tells much of the story in the first person, and again like Odysseus, he is an untrustworthy narrator. In chapter 15 of
book 2, for example, he relates an elaborate episode of landing on a wonderous island, which only in retrospect does he reveal
to have been a dream. He also, in chapter 20, explains how he tamed a wild horse by riding it over a cliff’s edge, a story
which he claims to be true. But unlike wily Odysseus, Periander has his critics, chiefly his companion Maurice, an elderly
astrologer who brings a note of irritable scepticism to Periander’s tales (I quote throughout from the 1619 translation and
using the English names): "This terrible skip of the horse seemed to Maurice a very hard matter, who would gladly at the least
he had broken three or four legs to give the more probability to the discourse of such an unmeasurable and outrageous leap."
This is not the first time that Maurice has raised objections to Periander’s story, which he thinks is too long-winded and
self-indulgent, and one effect of his role in is to make more credible Periander’s unlikely stories precisely by voicing the
reader’s own disbelief and occasional feeling of longeur.
The third and fourth books of the Persiles are clearly of a different tone and style. They take place on land: the travellers have reached Lisbon and continue to Rome
on foot, through familiar territory and familiar times. At the end of book 2, there have been references to Charles V's being
in a monastery in Yuste and then, in chapter 21, to some recent news: "there was no other talk but of the wars of Transylvania,
the stirs of the Turk, the death of Charles the Fifth, Emperor of Rome and King of Spain, who was a terror to the enemies
of the Church; and the sadness of Leopoldus King of Denmark ..." I will be coming back to King Leopoldus; but at this point,
we should note that Cervantes sets up quite precisely the events of books 3 and 4 to happen in the early years of the reign
of Philip II. This journey by land is not without incident, and perhaps the most striking example is a woman who falls from
a tower but is saved by her farthingale's acting as a parachute. There are other implausible occurences, such as newly-born
babies being handed to strangers and a Scottish countess wandering around the Dordogne with the skull of her murdered husband.
New characters are introduced, and each brings with them a story: Cervantes openly acknowledged that he owed his technique
to the Aethiopica of Heliodorus, with which this structure has obvious analogies. But the narrative springs in these later stories -- what
Hitchcock called the “Maguffin” -- tend to be the coincidences of everyday life, albeit rather colourful lives, rather than
the supernatural events of the two northern books.
English culture in 1619 had very little in the way of prose fiction on the scale of Persiles. English literature had been telling stories since the Anglo-Saxons composed Beowulf, but mostly in verse: one might instance Chaucer’s Troilus and Crisseyde, Sir Gawain and the Green Knight by an anonymous contemporary of Chaucer, and even Edmund Spenser’s Faerie Queene. In prose, there was Malory's enormous compendium of the Matter of Britain, the Morte d'Arthur, dating from the mid-fifteenth century; Sidney's Arcadia, Lyly's Euphues and Greene's Pandosto, all from the late sixteenth century. Urania, by Sidney’s niece Mary Wroth, had probably been written by 1619 but was not yet in print. Arguably, these all belong to
the romance tradition. In the 1590s and 1600s there was a vogue for 'rogue pamphlets.' These were short fictions which framed
themselves as warnings for the unwary but were really a vehicle for telling witty tales of the clever tricks which were perpetrated
in London, usually against foolish visitors up from the country: The Gull's Hornbook, by Thomas Dekker, is the best known of the genre.
It has been argued that Cervantes’ English contemporaries were uncomfortable with the concept of fiction. Robert Maslen, in
his 1997 study of English Fictions, says that, "Throughout the sixteenth century, prose fiction seems consistently to have been regarded, by its authors as
well as by its readers, as the most slippery of literary mediums. Its slipperiness lay partly in the difficulty of defining
what it was." Since prose fiction did not fit into the classical system of genres, one solution was to pretend that it was
factual: hence the rogue pamphlets.4 Foreign prose fiction was, if anything, even more suspicious, since it often came from Catholic countries, and Maslen quotes
the humanist Roger Ascham as saying that, 'Ten Sermons at Pauls Cross do not so much good for moving men to true doctrine,
as one of these books do harm with enticing men to ill living.' The closest English has to offer to the Spanish picaresque
novel is Thomas Nashe's The Unfortunate Traveller of 1594 and a series of stories about heroic apprentices and merchants by Thomas Deloney. English readers who wanted a sustained,
prose fiction -- what we would call a novel -- had to turn to translations, such as John Shelton's of Don Quixote, the first part of which was published in 1612.
The 1619 English translation of the Persiles is anonymous in every respect; neither the original author nor the translator is named. The printer and publisher are identified
by their initials: we know who they were only because printing in England was very tightly regulated by the Stationers’ Company,
whose record books have survived. The title page says that the original text was written in Spanish, "translated afterward
into French; and now, last into English." Although this is a slightly ambiguous formulation, it is evident that the translator
worked from the French intermediary, from which some errors were imported, such as the translation in the very last paragraph
of the novel of bisnietos as ‘nephews’.
The translator is coy about her,5 or his, identity. There is a preface to the reader, but it is not signed. From the fact that the translator did not wish
to be named, it can be assumed that he or she was not someone who earned a living by writing. There may, indeed, have been
social implications: this could be someone who did not want to be seen to be earning a living at all, although Warren Boutcher,
in his article last year in The Oxford Guide to Literature in English Translation, argues that in the early seventeenth-century the profession of translator did not exist; the nearest approximation would
be, he says, "a schoolmaster such as Philemon Holland, or hack scholar-translators employed by printer book-sellers to turn
out English copies of foreign books".6 Publishers in England, says Boutcher, would bring out an edition of a popular European work to benefit from the market,
and would use local translators to “English” it. This, presumably, is what happened with the Persiles, and the fact that it did appear in English is possibly an indicator of the text’s success in France and Spain. The attraction
was not the author, because Cervantes is not mentioned by name; the anonymous translator merely says in the epistle to the
Reader that, "The author is a Spaniard, whose style becomes him well in his own mouth; and his works of this kind have raised
his name, and approved his spirit, not alone in his own country, but in others." I would argue here that the very fact that
the translator was given a page for a dedicatory epistle suggests that this is not a common hack writer such as Boutcher describes,
but someone of standing, at least in the printer’s eyes.
The translation on the whole is faithful, in some cases remarkably so. In particular, Cervantes' references to Roman Catholicism
as the true Church are retained, and this was a bold step for a book published in the staunchly protestant capital of a protestant
country. The Gunpowder Plot, we must remember, had taken place only 14 years previously and religion was very much a political
issue. Indeed, one reason why no names were attached to the English publication of Persiles might be prudence -- a sense on the part of the translator and publisher that anonymity was the better part of valour. Such
differences as there are from the Spanish text can often be explained by examining the transmission of the text from Spanish,
through French and into English. One might assume that the English translator's taste was less towards copiousness than was
Cervantes'. Where Cervantes describes his character Transila defending herself from attack as, brava como una leona y airada como una tigre, the English translation conflates the two similes into one: "furious as a lioness". In other cases, though, one can see
a wish to avoid tautologies. For example, in the Spanish novel, when the nursemaid is dying she asks for the confirmation
that she died in the Catholic faith to be conveyed to mis padres aun fueren vivos, o alguno de mis parientes: "my parents, if they are still alive, or any of my relatives." de Rosset renders this into French as mon pére ou ma mére ... ou quelqu'un de mes parents. The English translator seems to have taken pére and mére as being the same as parents and used only “parents”.
Top of page
These are the sort of details of difference which one would expect to find in a literary translation, where the translator's
input is more than merely mechanical and artistic judgement is to be exercised. One of the changes between translation as
an art in the Renaissance and in modern times is in the mode of the translator's mediation between the reader and the text.
Modern translators aim to be "a sheet of glass";7 their work should be transparent and the reader should see straight through it to the original text. To cite Warren Boutcher
again, the role of the Renaissance translator was to intercede between the reader and the text, and he gives the examples
of John Harrington and Philemon Holland, both of whom were, he says, "engraved on the title-pages of their translations".8 Cervantes himself is fascinated with the recension of text through translation. Famously, Don Quixote is supposed to be a translation from an original by an Arab historian, Cide Hamete Benengeli. The same device is used for
the Persiles, and Cervantes allows himself the liberty of criticising the author of the original edition: It seemeth that the author of this history was a more expert lover than good historiographer, because he bestoweth well-nigh
the first chapter in the beginning of his second book to define the jealousy which Auristela had conceived [...]. But in this
translation we will omit those definitions, as being over-long and elsewhere handled, and come to the truth of the matter.
At times, Cervantes as the authorial voice intervenes between the reader and Cervantes as the translator, as when Maurice
is becoming fractious about the length of Periander’s narrative at the beginning of book 2, chapter 21: "I know not if I can
assure whether Maurice and some others of the company were glad that Periander had ended his discourse: for the longest are
oftentimes the most irksome, be they of never so great importance." Identifying the "I" of this statement requires some thought,
to penetrate Cervantes' -- triple, is it? -- layers of narrative. Again, what does the English reader have in front of herself?
A translation from the French, of a translation from the Spanish, of a translation from: what?
As far as I am aware, there is no contemporary, documentary evidence of the reception of the Persiles by English readers, with one exception. The playwright John Fletcher took a plot from one of the inset narratives and turned
it into a play, The Custom of the Country (1621). The story, in book1 chapter 11 of the Persiles, dealt with a non-Christian marriage-ritual, in which the bride was raped by the groom's male relatives. I have dealt with
this in detail elsewhere;9 for the present discussion, the point to note is that Fletcher takes Cervantes' story, which is set in an island off the
British mainland, and translates the action to Italy and Lisbon; and my argument is that the islands between Ireland and Scotland
were, for an English theatre audience, relatively familiar territory, whereas the story required an exotic location where
disbelief could willingly be suspended, which the south of Europe offered.
As I have indicated, Cervantes is careful to locate Persiles y Sigismunda specifically in the identifiable and recent past. We know that the setting for the novel is not contemporary, because Sigismunda
has lived to see her great-grandchildren; so it must have happened at least fifty or sixty years before 1616, about the middle
of the sixteenth century. Sure enough, this timeframe can be followed, with a little allowance for poetic licence. A character
discussing the contemplative life draws an analogy with "Charles the Fifth in a monastery" (2.9): Charles V withdrew to the
monastery at Yuste in 1555. We have noted a reference to his death, in 1556. One of the protagonists, a generation older than
the others, began his adventures when he left Spain for "the War which Charles the Fifth made against certain Princes of Aleman"
(1.5): perhaps a reference to Charles's troubles with the Schmalkaldic League. Madrid -- still a relatively new capital city
in Cervantes' day -- is helpfully marked as "where Philip the Third later kept his court" (3.5). This is a case of telling
history in the future tense, since the character is speaking as a middle-aged man in the 1560s and Philip III came to the
throne in 1598, but it adds to the topicality of the work. One of the chief characters, an elderly scholar and astrologer
from the Western Isles who is called Mauricio, is identified by Diana de Armas Wilson with the FitzMaurice clan in Ireland,
who had dealings with Spain during the reign of Elizabeth I.10 There is one anomaly, a character named Rosamunda whose role is to make sexual advances to one of the young men in the party
and come to a bad end. An English reader will inevitably identify this character with Rosamund Clifford, a twelfth-century
beauty who was the mistress of Henry II of England and Anjou; their love affair, and the jealousy of Eleanor of Acquitaine,
was the subject of a ballad by Thomas Deloney and there are frequent references to it in the literature of the period. How
well-known the legend was to a Spanish readership, I cannot judge. I would suggest that the world of Persiles y Sigismunda was one with which the Spanish reader could identify; it was a real world and had nothing of the mythic in it. As we have
seen, much that seems fabulous in the first two books may have had its origins in the Spanish literature of discovery.11 There may have been some odd stories of customs in the far north, the Spanish reader may have thought, but that did not
mean that they were not true: one had only to read the new accounts of the Americas to know that much of the world outside
Spain was still not Christian, and anything could be expected to be happening in the barbarian lands which Spain had not conquered.
On the whole, then, this was a trustworthy narrative; and as Barry Ife has demonstrated,12 Cervantes' narrative sleight of hand is so skillful that even the woman saved by her farthingale-parachute is made to seem
entirely credible, no matter how much common sense and the laws of physics may suggest to the reader that the event is impossible.
For the English reader, however, the experience would have been different, and even more so for a Scot.13 I return to the point that, for the British readers, Scandinavia and the North Sea islands held different resonances and
connotations than they did for the Spanish. In the late Elizabethan and Jacobean periods, there was much scholarly activity
in uncovering and recovering the pre-Norman history of the British Isles.14 The Scottish monarchs had intermarried with Scandinavian royalty as much as they had with the French, the English had traded
with the Baltic states for centuries and had established the Muscovy Company to trade to Russia and Eric of Sweden was one
of the first suitors for the hand of Elizabeth I when she came to the throne in 1558.15 Northeastern Europe, to the Scots and the English, certainly was not the stuff of legend
Any major inaccuracy would have undermined the careful illusion of veracity which Cervantes had created. Introducing Henry
II's mistress Fair Rosamond into a novel which was purportedly contemporary with the early years of Elizabeth I's reign would
have seemed strange, but perhaps justifiable. Rosamond is the personification of lechery. She has been exiled from England
because of her sexual habits and in a particularly memorable scene, she pursues a young barbarian man across the snow and
ice, only to be repulsed by him. She dies of a mixture of unfulfilled lust and remorse for her past life, and is buried at
sea, as if the ocean could cool her ardour. She is therefore a counterfoil to the lovers who are the novel's protagonists
and who are, like the original models in the Aethiopica, ostentatiously chaste. The inclusion of Rosamond is, therefore, explicable in thematic terms, and would not necessarily
have seemed out of place to an English readership of the period -- if the work was a fiction.16
Much more puzzling would have been the presence in the novel of Arnaldo, Prince of Denmark and son of King Leopoldus. Arnaldo
is an important character who is present in much of the novel and who is driving the action when he is not physically there;
but, unlike Charles V, for the English or Scottish reader, he is a figment of the imagination. Denmark was a Lutheran state
and thus a natural ally of England and Scotland. The geopolitical importance of Renaissance Denmark is neatly summarised by
Leeds Barroll, in a recent cultural biography of James VI and I's consort: Denmark's king was sovereign not only over Denmark itself -- this consisting of the Jutland Peninsula, the islands of Fyn
and Sjaelland and their island groups -- but also a large collection of other Scandinavian and German territories. Danish
rule extended to the north over all of Norway; to the east over the provinces of the Scandinavian peninsula in present-day
Sweden, and the Baltic islands of Bornholm and Gotland; to the northwest, over Iceland, Greenland, the Faeroe Islands, and
the Orkneys; and to the south over the German duchies of Schleswig and Holstein [...] Further, because of its geographical
position, Denmark in the late sixteenth century could effectively block all three entrances from the Atlantic to the Baltic.17
Denmark was thus important, and of interest, to England at any time, but particularly so at the time of the publication of
Persiles and Sigismunda. From James's accession in 1603 until he was widowed in 1619, the Queen Consort of England was Anna
of Denmark, daughter of one Danish king and sister of another. The heir to the thrones of England, Scotland and Ireland, the
prince Charles who was to become Charles I, was actually half Scots, half Danish. Londoners visiting the booksellers in Paul's
Churchyard may well have known that recent kings of Denmark had all been Fredericks or Christians, and that none had been
called Leopoldus or Arnaldo. The very existence of Arnaldo signals this text in English as fiction.
Finally, there is the significance of the novel's geography to consider. Diana de Armas Wilson has traced the voyage of the
pilgrims in books one and two18 and she has demonstrated that the journey is geographically correct. The English translator, however, literally remaps it,
in chapter 11 of book 1. The pilgrims have escaped from the Barbaric Isle and are drifting southwards. They make landfall.
"They saw two men on the shore, to whom Transilla called, asking of what country they were, and what religion they professed.
They answered in a language which she understood, that this isle was called Holland [...] although the place were in a manner
desert, because the inhabitants were so few that all of them had but one house." The name “Holland” in Renaissance English,
as today, refers both to a province of the Netherlands and an area of East Anglia, and the two could have been confused;
but neither Holland could reasonably have been described as an island possessing only one house. One must turn back to the
process of translation.
The Spanish word here is Golandia, which de Rosset translates into French as Groenland. Possibly the English translator thought that this did not look right, checked the Spanish original and mistranslated it
with a more familiar word without considering too closely the geography. Conversely, perhaps he translated Golandia first as Gothland, the meaning cited by Diana de Armas Wilson,19 but confused it with the English Gothland, which is an area of the North Riding of Yorkshire on the moors between Whitby and Scarborough. Perhaps he just did not know
where it was.20 Cervantes’ geography was, like the novel form itself, ‘slippery’ to the English reader, and the translator did his best
to make sense of it. The congruence between Holland and Golandia, however, does suggest that the translator had access to the Spanish original as well as the French intermediary from which
he mainly worked.
Top of page
To take another case: on the first occurrence of Hibernia, or Hybernie in the French, he translates it correctly as Ireland. But in chapter 21 of Book 1, there is the sentence, y todos rogaron muchas veces a los cosarios que los llevasen de una vez a Irlanda o a Hibernia (‘and they all often asked the pirates to take them to Ireland or to Ireland’). This appears in the 1619 English translation
as, ‘All of them oftentimes entreated the pirates to carry them into Ireland or Scotland,’ a sensible emendation but one which
again changes the context of the story for what was, in fact, a local readership. For the Spanish reader, the pirates were
being asked to take the pilgrims to a far-off country; for the English or Scottish reader, they were being asked to bring
the pilgrims home.
Returning to the question with which we started: what sort of history is this
historia? Cervantes presents two books set in a territory which for the Spanish readership is terra incognita, but gives familiar signposts along the way, which lead to the final two books in a realistic setting. Yes, there are werewolves
and withces in the far north, but it happened just about the time that Charles V died and the story also takes place in identifiable
places such as Lisbon. Charles was a real King Emperor, Lisbon is a city in the Spanish Empire so, the logic goes, this is
a real story in which the protagonists happen to encounter magical events which, after all, are no more strange than travellers
have come across in the New World. For the English reader, it is a different story. The narrative is the same, but the shape
has shifted through the different cultural perceptions which geography imposes. The English reader's familiarity with the
Baltic makes the Persiles a narrative which cannot be trusted; it is revealed as such from the moment that the unhistorical Arnaldo Prince of Denmark
is first named in chapter 2 of book 1. Cervantes' meticulous framework of convincing facts, which makes so plausible the witches
and the werewolves and farthingale parachutes, simply does not work in the Northern History. Without it, the perspective changes, and the magic is foregrounded to take precedence over the realism. A faithful translation
it may be, and a novel; and a very good novel at that; but The Travels of Persiles and Sigismunda is a different novel from Los trabajos de Persiles y Sigismunda.
Top of page