1Over ninety years ago, on August 15, 1911, George Santayana, a well-known Harvard professor, philosopher, poet, and humanist, born in Madrid, gave a lecture at the University of California, Berkeley, on the topic The Genteel Tradition in American Philosophy. And these were his first words on that particular occasion, which I adopt, and adapt:

2"Ladies and gentlemen," he began, "the privilege of addressing you today is very welcome to me, not merely for the honour of it, which is great, not for the pleasures of travel, which are many, when it is Winnipeg and the University of Manitoba1 that one is visiting, but also because there is something I have long wanted to say which this occasion seems particularly favourable for saying.…"

3That "something I have long wanted to say" deals here with the many "blank spaces" still found in the history of translation. And no wonder that such gaps and blanks may still be found, when our discipline (or subdiscipline) is but a young branch in the large tree of translation studies. I do not think, however, as Alexander Gross does, that "the field of translation history may be in its infancy."2 Perhaps "in its adolescence," but definitely not in its infancy, because, as Bassnett and Lefevere wrote a few years ago (1998, 1), "History… is one of the things that happened to Translation Studies since the 1970s." I would even correct that statement, and say "since the mid-1960s," for it was in 1965 that Georges Mounin published in Italian his Teoria e Storia della Traduzione.

4Since those forty pages by Mounin, and over the last thirty years, a long chain of titles of general scope have followed: After Babel by George Steiner (1975), The True Interpreter by Louis G. Kelly (1979), Interpretatio by Frederick M. Rener (1989), and then a good number of titles by Michel Ballard, Hans J. Vermeer, André Lefevere, and Susan Bassnett, Henri van Hoof, Jean Delisle, Judith Woodsworth, and Francisco Lafarga, among others; and before, after, and around them a countless number of nation-based, author-oriented monographs, conference proceedings, chapters of books, monographic issues of academic journals, and so on. It may be true, as Alex Gross claimed in 1996, that "no single book in this field can [yet] be considered a model of clarity and accessibility";3 nevertheless, and again in Gross's words, "books like these are still of enormous value: they offer a rich harvest for those prepared to cut their way through their burgeoning undergrowth."

5There is no doubt that since the mid-sixties, or even the seventies, a good distance has been covered, and we are now quite far from where we were in those days. Things have changed very quickly, at least from a doctrinal point of view. Already in the early 1980's Susan Bassnett stated in quite clear terms: "No introduction to Translation Studies could be complete without consideration of the discipline in an historical perspective" (1980, 39); and again, a few pages later: "The history of Translation should therefore be seen as an essential field of study for the contemporary theorist" (1980, 75). On the same lines, Antoine Berman (1984, 12): "The construction of a history of translation is the first task of a modern theory of translation." And he underlines the word modern.

6However, when going over the long history of translation (or rather the miscellaneous and multifarious mosaic of partial historical studies, because a history as such is still to be written), the reader cannot help thinking he or she is being presented with a distorted image of the translation panorama throughout the centuries. And the only reason is, to be frank, that we are still far from having at our disposal a global and globalizing vision of what the translating activity has been throughout its approximately four thousand, five hundred years of history.

7It is true that there are parts of that history that are well charted, the many translations of the Bible for instance. However, it is also true that there still remain "vast unknown territories" in that universal history, territories which concern not only places and times but also whole fields of inquiry and research. If we think of the history of translation as a mosaic, there can be little doubt that there are still many small pieces or tesserae missing, as well as large empty spaces yet to be filled in. The full design is far from complete. Much is still unknown.

8I claim to uncover or reveal nothing new because, as professionals or amateurs of this discipline, I imagine that we are all also well aware of all those gaps along and across the coordinates of time and space. Let this paper therefore be nothing but a brief, cursory review of the several large uncultivated fields we can expect to plough in the near future.

Oral translation or interpretation

9Almost everybody would agree that one of the most notorious empty spaces in our field is the history of oral translation or interpretation. There are so many books of national and international scope on the history of translation, and so few on the history of interpretation. Only in very recent years has attention begun to be paid to the activity of interpreters from a historical point of view, although limited in most cases to the twentieth century, with few and sparse references to their work before our times or in the centuries before America was discovered.

10Most studies appear methodologically fragmentary and more like aggregations of "innumerable anecdotes found in secondary sources"4 than true histories of a centuries-old profession. Such miscellaneous collections of data and anecdotes are extremely useful, certainly, but are still far from the idea of a true history. Let us consider such works as Ruth A. Roland's little-known Translating World Affairs (1982), which, revised and updated by Professor Jean Delisle, was reprinted five years ago under the new title of Interpreters as Diplomats: A Diplomatic History of the Role of Interpreters in World Politics (Ottawa: University of Ottawa Press, 1999). Similarly, we could mention Karttunen's Between Worlds: Interpreters, Guides, and Survivors (New Brunswick, NJ: Rutgers Univ. Press 1994).

11However, the past is full of documents, both manuscript and printed. Take, for instance, the Anabasis. In the year 399 BC (the year in which Socrates was condemned), Xenophon led his troops, the "Ten Thousand," some 1,000 miles through the unknown and hostile territories of Kurdistan and Armenia; they reached the Greek city of Trabzon on the Black Sea early in 400 BC. The chronicle Xenophon wrote of that expedition, the Anabasis, brims with dozens of references to the mediation of professional and occasional interpreters from Greek into Persian, Armenian, Tracian, Carduchian, Macionese and Mossynaecian.

12But the Anabasis is just one among thousands of examples. The diary of Antonio Pigafetta's circumnavigation in 1519–1522; the several voyages of James Cook; the accounts of English, French, Spanish, Dutch, and German explorers, travellers, and chroniclers in North, Central, and South America; the explorations of Africa and Asia; the work of missionaries throughout the world — all of these are full of notices about interpreters. According to Bede in his famous Ecclesiastical History of the English Church and People, Saint Augustine landed on the island of Thanet in 597, bringing with him several "interpreters of the nation of the Franks."5 An Arab boy from Jerusalem, serving as interpreter, helped Henry M. Stanley to finally discover Dr. Livingstone in the African village of Ujiji. And Bombay and Nasib traveled as interpreters of the party, led by John Hanning Speke, that discovered the sources of the Nile.

13The chanceries of Europe are also full to overflowing with documents that tell of interpreters involved in embassies and legations (both secret and official), peace and trade treaties, settlements of frontiers, royal marriages. Another example deals with eighteenth-century ship's interpreters, a profession about which very little has been written, but which has existed (for at least five centuries) with its own statutes and regulations, including full details of their professional deontology and even what we would call today a "code of honour" (Santoyo 2003).

14How much longer shall we wait for a general history of interpretation, not even of the world, but of a continent or a country?

The daily practice of translation

15In the universal chronicle of translation the focus has always been on the book. And the more significant the original, the more studied any of its translations. There are thousands of pages on the translations of Shakespeare's poems and plays into this or that language, as there are thousands on the many translations of the book par excellence, the Bible, or the works of Marco Polo, Cervantes or Seneca, Ptolemy or Galen. These are no doubt the sort of translations which had the most influence on the development of culture and knowledge, at least in the West, and maybe they are the only ones worth recalling throughout the centuries. Remember the sentence of King Alfred, in the last years of the ninth century, when he admits in the prologue to his version of Cura Pastoralis, "It seems better to me… that we too should turn into the language that we all can understand certain books which are the most necessary for all men to know" (Swanton 1979, 31–32).

16In point of fact, such translated texts are the only ones that count in any history of this art and craft, and it is quite evident that only minor references to other types of texts are found in the partial histories of translation published so far. Everyday, common, unerudite, unscholarly translations have hardly ever attracted the attention of historians. So exclusively have the spotlights been directed at the book that any other sorts of texts have been left in the shadow of history.6 However, books, whatever their nature, classical or modern, technical, historical, literary, philosophical, or religious, are but a part of the total landscape of translation, and from a quantitative point of view not even the most important part. During four hundred years, from the seventh to the tenth centuries, "as far as we know, only one Western book was actually translated into Arabic" [Orosius's Adversus paganos historiarum libri septem] (Lewis 1982, 76, 141); still, the stream of translated documents between Christians and Muslims flowed uninterrupted, and many medieval chronicles brim with the names and performances of their translators.

17Think, for instance, of the relations between Europe and the Mongols during the thirteenth century. Not a single book seems to have been translated between Mongolian and any European language, Latin included, during this century of Mongolian expansion; notwithstanding, the chronicles of the mutual relations abound with "acts of translation," messages, letters and documents which went to and fro in the hands of successive emissaries (William of Rubruc, friar Giovanni di Pian del Carpine, and friar Ascelino of Cremona among them), translated from Mongolian into Latin, from Latin into Russian, Persian, or Mongolian, from Greek into Mongolian, from Latin into Arabic or Syriac, and so forth.

18Between the eighth and the fifteenth centuries, and much later on, translation covered spheres of interest and activity which clearly overflowed the narrow limits of the book: texts, most of them, of a pragmatic, matter-of-fact condition, which since the beginning of time have been present almost daily at school, at court, at church, in monasteries and chanceries, on routes of pilgrimage, at ports, harbours, and interstate frontiers. These practical, everyday forms of translation are not as well documented as those of a more "cultural" or "scientific" character, but even so they do not deserve the historiographical silence which has so far surrounded them.

19In this respect, the only sources one can usually come across are slight (or worse, trivial) commentaries which hardly go beyond two or three lines, scarcely a whole paragraph. Such is the case, for instance, with Henri van Hoof's Petite histoire de la traduction en Occident. A chapter containing a detailed account of the translational activity in France during the twelfth, thirteenth, and fourteenth centuries, centred above all on the translations of works by Gregory of Tours, Livy, Aristotle, Seneca and Cicero, Lucan, Virgil, and others like them, ends with this single sentence relative to "other" types of texts: "That translation has, moreover, played an administrative and diplomatic role, there can be no doubt" [our translation7]. And without any other explanation, the sentence is followed by a short anecdote about a certain letter which in 1447 was "translated from Saracenic into early French" (1986, 14; our translation). Full stop and new chapter.

20However, the numerous medieval and Renaissance testimonies of the daily practice of translation all tend to confirm that in many countries it was a major factor of personal and official communication, present in everyday life at least since the end of the seventh century and throughout the Middle Ages. This circumstance has hardly ever been taken into account, however, dazzled as we are by the brighter lights of "the book."

21No wonder, then, that in England at the beginning of the eighth century, Bede the Venerable translated from Latin several prayers and liturgical songs for the many illiterates who only knew their own language ["qui tantum propriae linguae notitiam habent"]; or that in the spring of the year 813 the Third Council of Tours decreed the translation of all sermons "in rusticam Romanam linguam aut thiotiscam," that is, into the vulgar French or German of the time.

22Five centuries later, in the first decades of the fourteenth century, an amazing procedure was the normal, everyday way of doing things at the courts of the County of Kent, in the south of England (Clanchy 1979, 161):

First of all the jurors were presented with the justices' questions… in writing in either Latin or French. They replied orally, probably in English, although their answers were written down as veredicta by an enrolling clerk in Latin. When the justices arrived in court, the chief clerk read out the enrolled presentments or veredicta in French, mentally translating them from Latin as he went along. On behalf of the jurors, their foreman or spokesman then presented the same answers at the bar in English. Once the presentments, in both French and English oral versions, were accepted by the court, they were recorded in the justices' plea rolls in Latin.

23A surprising exercise, no doubt, of linguistic and translational juggling, favoured by and derived from the situation of bilingualism and trilingualism in England at the beginning of that century. This was, by the way, very common throughout Europe of the Middle Ages, from Norman Sicily to Ireland and from the lands of Central Europe to the Iberian Peninsula, here with a variety of five languages plus Arabic and Hebrew.

24Most of those translations were not intercultural but intracultural performances; there was no transference, no transmutation from a source cultural polysystem to another target polysystem; not even their teleology had a cultural character: these translations were always made with a strictly local purpose and in order to make known strictly local contents. Quite far, therefore, from the translation of books, made to transcend "the moment and place of their production and reception" (Romano 1991–92, 222).

25If translation is to be thought of not only as a translinguistic, but above all as a transcultural phenomenon, such a point of view must be left aside when approaching medieval, Renaissance, even modern everyday translations. If books were translated in medieval times "because of the wealth of knowledge they were thought to provide" (Chang 1994, 19), in everyday translations it is the direct, immediate, local, utilitarian necessity of understanding contents that inspires the act of translating. There is no express desire for cultural transcendence. "Est latine, non legitur": as it is in Latin, nobody understands it, and therefore it is translated. Such could be the ultimate reason for so many everyday translations.

26This was to be the leitmotif (explicit) from the twelfth century onwards, as it had been (implicit) before. When in 1299 King Fernando IV of Spain confirmed the privileges of the town of Castrojeriz, he did it with a text translated from the original Latin into Castilian, "because the said privileges are in Latin, and the laymen cannot understand them" (Santoyo 1997, 169). When in 1378 Alfonso Pérez, a canon in the Spanish town of León, ordered a notary public to make a copy of four documents that were in Latin, he at the same time ordered their translation into Spanish, "because they were in Latin and therefore were obscure and difficult to understand" (ibid., 172–173).

27In such cases, it may be a surprise to many how the agents involved in the textual transfer express their will of exactness and radical fidelity to the letter of the document — quite far from the "liberty" one finds in other sorts of translations (literary, philosophical, scientific, and so forth). They had no interest in the form, nor in style: what really mattered to them was the identity of contents. In the translation of a document from San Miguel de Escalada (León) in the year 1380, the translator states that he has "turned the said Latin into vulgar Romance, everything word for word" [uerbo por uerbo] (ibid., 173). Two years earlier, when Alfonso Pérez, a Spanish canon, ordered the translation of several Latin documents into Spanish, three times he repeated that he wanted a translation "faithfully done," "faithfully translated from Latin into Romance," "faithfully translated and interpreted from Latin into Romance" (ibid., 172–173). This is a way of behaving far removed from what Lemarchard (1995, 30) believes to have been habitual among medieval translators of books. She says that "they felt perfectly authorized to modify a text in accordance with the audience it was meant for… ; a translator tackled his task feeling perfectly legitimized to graft any sort of comment onto the text, without even mentioning that he was deviating from the original text in order to add on something of his own invention."

28Translation was an everyday component of medieval life, no doubt, and much more frequent and quotidian than what today's lack of studies may suggest. This is a very important aspect of translating activity (and of the history of translation) which it is absolutely necessary to recover, and the sooner the better if we want to have a complete picture of what translating signified for the daily life of medieval Europe. An aspect which, of course, should be approached from conceptual premises somewhat different from those of today.

Pseudo-translations

29In spite of the cultural significance they have had throughout the last millenium, and in spite too of being "closely linked to genuine translation in terms of cultural position" (Toury 1995, 45), the field of pseudo-translations (or fictitious translations) is another empty chapter in our History.

30In 1721 Charles-Louis de Secondat (1689–1755), Baron of Montesquieu, published his first literary work: one hundred and fifty letters under the title Lettres persanes, two small volumes apparently printed in Amsterdam. In the very prologue the "anonymous" translator had written: "Thus my role is limited to that of translator" [our translation]. Actually, as if in fact it were a translation, and in the best style of a belle infidèle, the "translator" added immediately afterwards:

I took all pains to make the work correspond to our morals. I relieved the reader of the Asian language as much as I was able and saved him from innumerable rarefied expressions, which would have bored him to heaven.… But that is not all I did for him. I cut back on the long compliments, which the Orientals proffer no less lavishly than us, and I passed over an infinite number of minute details. (Montesquieu 1949, 131–132; our translation)

31Technically speaking, there was nothing new in the Lettres: as a collection of translated letters it had been preceded by, among others, the Lettres portugaises traduites en français, traditionally ascribed to Gabriel Joseph de Lavergne (1669), and above all by L'esploratore turco by Giovanni Paolo Marana (1684), which had already been translated into French with the title of L'Espion turc. This last work, for instance, is a collection of 531 letters supposedly written in Arabic by the Turk Mahmut and also supposedly translated into Italian by Marana.

32From their very first edition the Lettres persanes met with an "immense and immediate" success ("immense et immédiat"), as Roger Caillois has defined it. After the first edition in Amsterdam, 1721, the Lettres were reedited over thirty times during the author's life (always in French) in Amsterdam, Cologne, Leipzig, London, and Paris. It was, no doubt, one of the French editorial successes of the century.

33The most obvious and immediate consequence of that editorial fever was the long series of rewritings which throughout the eighteenth century were published in France as second parts, sequels, and imitations of the Lettres, all of them pseudo-translations written by foreigners of the most varied nations and cultures: Lettres d'une turque à Paris, Lettres d'une péruvienne, Lettres de Nédim Coggia, Lettres juives, Lettres cabalistiques, Lettres chinoises, Lettres siamoises, Lettres iroquoises, Lettres d'Osman, Lettres d'Amabed (by Voltaire), and so on.

34At the same time, a second wave of rewritings had begun to take shape, this time under the form of translations: into English (Persian Letters, translated by John Ozell in 1722), into German (Persianische Briefe, 1759), into Russian (Piersidsieia Pisma, 1789), and so on.

35As in France, the two English translations by Ozell (1722) and T. Floyd (1755) also were the direct (I would say the only) cause of the birth of a new literary genre in that literature: that of "letters written by foreigners." All were fake translations from this or that language, all rewritings, in one way or another, of Montesquieu's "Persian" letters. The list of titles is as long as that in France, even longer: Letters from a Moor at London, Athenian Letters, Sequel of the Letters Written by a Peruvian Princess, Letters from an Armenian in Ireland, A Letter from Xo-ho by Horace Walpole, Chinese Letters by Oliver Goldsmith, Letters of Clement XIV, Spanish Memoirs, Letters of a Hindoo Rajah by Elizabeth Hamilton, and so on. The complete eighteenth-century catalogue includes several dozen titles of major, middle, and minor literary value.

36But Montesquieu's Lettres persanes are just one instance among the several hundred pseudo-translations with which the cultural and literary history of the West has been interspersed. Fictitious translations include such literary milestones as the Historia regum Britanniae by Geoffrey of Monmouth in the first half of the twelfth century; Orlando innamorato by Boiardo in the fifteenth century; Don Quixote in 1605; Zadig by Voltaire; The Castle of Otranto by Horace Walpole in the eighteenth century; The Manuscript Found in Saragossa by Ian Potocki, and "Rappaccini's Daughter" by Nathaniel Hawthorne in the nineteenth; and in the twentieth century, The Immortal and Doctor Brodie's Report by Borges, The Council of Egypt by Leonardo Sciascia, Mrs. Caldwell Speaks to her Son by the Nobel Prize laureate Camilo José Cela, and The Duchess's Diary by Robin Chapman.

37Sometimes received by the reading public of their time not as originals but as true translations (as was the case of The Castle of Otranto by Horace Walpole in England and of Papa Hamlet by Arno Holz and Johannes Schlaf in Germany), pseudo-translations occupy a very special place in the history of literature. But it should not be forgotten that they also belong, in more ways than one, to the history of translation.

38It may be worth recalling here Toury's words on the subject (1995, 41): "Pseudotranslating has not always been so marginal as it may now seem to be.… At the same time, pseudo-translations are far from a mere curiosity, which is how they have been treated all too often in the literature. In fact, they often prove highly revealing for cultural studies, especially in their historical facet, including culture-oriented Translation Studies."

39By the way, it is quite surprising, and paradoxical too, that on the one hand, all throughout the centuries, "translators have been widely scorned at times and their work severely criticized"; they "have been distrusted, and even called turncoats and traitors" (Joly 1995, xiii); whereas, on the other hand, their work, their translations, have given rise to a narrative technique used by writers of no lesser stature than Voltaire, Cervantes, Montesquieu, Walpole, Goldsmith, Borges, Cela, and Hawthorne, among many others.

Self-translations

40Another vast territory without history is self-translation, defined thirty years ago by Anton Popovič as "the translation of an original work into another language by the author himself" (1976, 19). It is surprising how mistaken commonplaces can go on and on for decades. Self-translation, I have no doubt, is an area of translation studies that so far has been almost forgotten, perhaps because we all think that it is and always has been something absolutely marginal, a sort of cultural or literary oddity. Several testimonies, chosen at random among many others, may bear witness to a widespread opinion.

41Antoine Berman, in the first pages of his essay L'Épreuve de l'étranger: "For us, self-translations are exceptions" (1984). Grady Miller, in a paper read at the American Translators Association Annual Conference: "Historically, few authors have dared to translate their own works" (1999, 11). Professor Christian Balliu in a paper published three years ago in META, the Canadian quarterly: "It must be admitted that examples of self-translation… are extremely rare in literature and form the exception" (2001, 99; our translation). Four years ago John Benjamins published a volume, edited by Allison Beeby et al., under the title Investigating Translation. In one of the contributions, by Helena Tanqueiro, the reader is informed that "[throughout history] only a few [writers], very few indeed, actually translated their own work" (2000, 50). In his introduction to The English Writings of Rabindranath Tagore, Sisir Kumar Das, Professor of Bengali and Comparative Literature at the University of Delhi and Vice-president of the Comparative Literature Association of India, states in a fairly definite way: "Undoubtedly he [Tagore] is the only major writer in the literary history of any country who decided to translate his own works to reach a larger audience" (1994, 10). It is worth repeating (and remembering) his words: Tagore is the only major writer in the literary history of any country who decided to translate his own works.

42Exceptions, few, very few authors, rare enough, rarissimes. Even György C. Kálmán, of the Institute for Literary Studies of the Hungarian Academy of Sciences, rashly described self-translation in 1993 as "a borderline case" of both translation and translation studies. That seems to be today "the state of the art." And when so many people, so well qualified, and all through the last forty years agree on such a description and definition, no doubt again, they all must be right. Self-translation lacks any place in history; small wonder, then, that Professor Brian T. Fitch, University of Toronto, could write sixteen years ago in his book Beckett and Babel: "Direct discussion or even mention of self-translation is virtually non-existent in writings on theory of translation" (1988, 21). Small wonder, again, that only six years ago Shuttleworth and Cowie could also write in their Dictionary of Translation Studies: "Little work has been done on autotranslation" (1997, 13).

43Against the grain of all these judgements and opinions, the sheer truth is that self-translation has been present in the history of this art and craft at least since the times of the Jewish historian Flavius Josephus in the first century of our Common Era. He wrote his first work, the seven books of The Jewish War, in his mother tongue, Aramaic, for the benefit of the Jewish communities in the Diaspora. Some years later he himself reviewed and translated it into Greek, correcting at the same time some of the errors he had before run into. In the preface to his Greek version he clearly states: "I have proposed to myself, for the sake of such as live under the government of the Romans, to translate those books into the Greek tongue which I formerly composed in the language of our country.…"8

44Two thousand years of self-translations after Flavius Josephus are still waiting for inclusion in the pages of our History. Hundreds, perhaps thousands of texts have been written in one language and then transferred by their authors into a second tongue: in medieval Spain, in Renaissance Italy, in seventeenth-century England (think of John Donne), or in present day Belgium, Turkey, Canada, or Brazil.

45Self-translators have included the medieval philosopher Raimundus Lullius, the humanist Leonardo Bruni, Thomas More, Étienne Dolet, Du Bellay, Jean Bodin, John Calvin, Pietro Bembo, the poet Andrew Marvell, the philosopher Spinoza, the Italian playwright Carlo Goldoni, the French poet Mallarmé, James Joyce, and the Nobel Prize recipients Mistral, Tagore, Pirandello, Beckett, Singer, Brodsky, and Milosz. Plus Julien Green, Romain Gary, and Elsa Triolet in France, Karen Blixen in Denmark, Aitmatov in Russia; and in our own days Raymond Federman, Nancy Huston, and an endless list of authors, particularly within the world of literature.

46In Spain, over two hundred writers are nowadays translating their works from Catalan, Basque, or Galician into Castilian, even into French. In India, dozens of writers are also nowadays transferring their works from Urdu, Bengali, Hindi, Gujarati, Malayalam, or Tegulu, mainly into English. And the same phenomenon is taking place in South Africa, Puerto Rico, India, Ireland, the United States, and elsewhere.

47In Canada you may well know the case of Honoré Beaugrand (1848–1906), journalist, traveller, and also mayor of Montreal, who wrote "La chasse-galerie" in French and then translated it into English, publishing it in 1891 in French in the journal La Patrie and the next year in English in The Century Illustrated Monthly Magazine. He also translated another of his stories, "Macloune" (vide Grutman 1994), from French into English.

48Being, on the one hand a bilingual country and on the other both a land of asylum for many exiles and a nation where foreign authors thrive in academic circles, it is small wonder that Canada is home to an extensive list of authors who are now self-translating their originals into a second language: Nancy Huston, Patrice Desbiens, Guy Arsenault, Melvin Gallant, Moin Ashraf, Alfredo Tutino, Claude Hamelin, Jean Grondin, Nicole Brossard, Chava (or Chawa) Rosenfarb, Guy Maheux, and Daniel Gagnon, among others.

49Self-translations are not at all exceptions, nor are they rare enough, nor few, very few indeed; we cannot keep saying that they are "not very common in the field of creative writing," or that "few authors have dared to translate their own works," or that they are borderline cases. Research in the history of this particular area will show, is in fact beginning to show, that as Christopher Whyte wrote two years ago, "self-translation is a much more widespread phenomenon than one might think" (2002, 64). And worthy, therefore, of receiving much more attention than it has so far received.

Forgotten texts

50Another field awaiting the work of historians is the rescue and recovery of many forgotten texts. Nowhere have I seen mentioned, for instance, one of the very first reflections of a translator on his own work, to be found in the prologue of Ecclesiasticus, one of the books of the Bible that is non-canonical for Jews and Protestants but canonical for Roman Catholics. Written in Hebrew at the beginning of the second century BCE, Ecclesiasticus, or the book of Sirach, was translated into Greek around the year 132 BCE by the anonymous grandson of the author. In the prologue to his translation he writes (mind the year, 132 BCE):

You therefore are now invited to read it [this book] in a spirit of attentive good will, with indulgence for any apparent failure on our part, despite earnest efforts, in the interpretation of particular passages; for the words spoken originally in Hebrew are not as effective when they are translated into another language. That is true not only of this book but of the Law itself, the prophets, and the rest of the books [of the Bible], which differ no little when they are read in the original.…

51And in a new paragraph:

I therefore considered myself in duty bound to devote some diligence and industry to the translation of this book. Many sleepless hours of close application have I devoted in the interval to finishing the book for publication, for the benefit of those living abroad who wish to acquire wisdom and are disposed to live their lives according to the standards of the Law.9

52Everybody quotes Cicero's famous sentence, so often repeated: "In quibus non verbum pro verbo necesse habui reddere." Very few (if any) remember that, over a hundred years before Cicero, it was Terence who first used the expression "verbum de verbo" (word for word) when speaking of his translation ofMenander's The Brothers [Adelphi], a comedy played in Rome in the year 160 BCE: ". . . eum hic locum sumpsit sibi in Adelphos, verbum de verbo expressum extulit."

53We badly need to recover, for instance, the entire tradition in Arabic, which spans from the eighth century to our own days, and which when cited, if cited at all, is often dismissed in less than one page. A quick review of several translation theory anthologies, from Paul A. Horguelin to Douglas Robinson, makes quite clear the absence of non-European thinkers and writers, either Arabs or Chinese, Indians or simply Latin Americans.

54We in the West are quite unfamiliar, for instance, with many of the dicta in the Arab tradition of translational thought and criticism, among them the opinions of Hunayn ibn Ishaq, the great ninth-century expert in translation from Greek into Syriac and Arabic. We are also quite unacquainted with the opinions of another contemporary of his, Abû Utmân al-Jâhiz (830 approximately), and with his surprising and pioneering statements on the difficulty, even impossibility, of translating poetry (al-Jâhiz 1938, 75–79):

If translated, the very essence of poetry is destroyed. Poetry is only enjoyed by the people to whom it belongs. As a literary manifestation, it is untransferable; it never is universal, because it is always tied up to, and trapped by the language in which it was written.… Poetry cannot be translated, should not be translated, because, when translated, its music, rhythm and poetic structure disappear, its whole beauty fades away, and nothing really worth admiring is left; in point of fact, translated poetry ends up by becoming plain prose.…

55And again, with regard to the "musts" of a translator:

A translator must measure up to the subject matter he is translating, must have the same knowledge as the author he is translating. He must be well versed in the language he translates, as much as in the language into which he translates, so that he be equal in both.… The more difficult a discipline and the fewer experts in it, the more important are the difficulties a translator finds, and the more exposed he is to making mistakes.…

56All this, and much more, in the year 830.

57At the dawn of the tenth century, Ahmad ibn Yûsuf wrote in his Epistle on Proportion and Proportionality: "In addition to having reached a more than remarkable knowledge of the languages from which and into which he translates, it is also necessary that the translator have a very good grasp of the subject matter he is translating" (Gil 1985, 44).

58Mention can also be made of Abû 'Alî ibn al-Samh, at the end of the tenth century, and the extreme care with which he collated several translations of the same text. Or the name and opinions of the translator of Aristotle, al-Hasan ibn Suwâr al-Hammar, around the year 1000: "In order to reproduce the meaning, the translator needs to understand it by means of the language from which he is translating, so that he perceives it as fully as the original author expressed it; he must know the use of the source language, as well as that of the language into which he is translating" (Hugonnard-Roche 2001, 27). Or, just as a last example, the name and opinions of Salâh al-Dîn al-Safadî, halfway through the fourteenth century, with details of the two methods used by translators in his own time (Badawi 1968, 33):

There are translators who examine every single Greek word and its meaning, select then an equivalent word in Arabic, and just write it down; then they pass on to the next word, and so they proceed along till the translation is finished. But this is a wrong method, and that for two main reasons: first, because in Arabic there are no equivalents for every Greek word, and as a consequence we see that many Greek terms are left untranslated; secondly, because the syntax and sentence structure in one language does not always correspond to that of another language.… The second method of translating is much better: it consists in reading and fully understanding the sentence, and then translating it by means of an equivalent sentence; it does not matter whether or not the words themselves are of the same value.…

Translated texts as survivors of lost originals

59In spite of their cultural significance, no history of translation has taken into consideration the role of translated texts as survivors of lost originals. In a recent paper Joâo Ferreira Duarte asked (2003, 16): "Do they [translations] have a function in history other than being the afterlife of the original, as Walter Benjamin put it?" Of course they have, particularly in the case of lost originals, which have come down to us only in translation: they now function in history as true originals, because the text from which they derived has disappeared, and the translated text has assumed the function of the original. However, that function, as far as I know, has hardly been chronicled, studied, or estimated.

60It is not a case of one or two isolated instances. We know many ancient, medieval, and even Renaissance writings thanks to their translations, the originals having been lost for good and all. This is true in the case of a large portion of paleochristian literature, which only survives in translation and includes nothing less than the Gospel according to Matthew, which we now have in Greek. Both Papias and Jerome firmly believed, and insistently repeated, that Matthew had written his Gospel "in the Hebrew language" (Rajak 1984, 232) [Hebraeis litteris uerbisque]. In his Commentary on Matthew, Origen wrote: "I have learned by tradition that the Gospel according to Matthew… was written first, and that he composed it in the Hebrew tongue and published it for the converts from Judaism.…" And again, Jerome in his De viris illustribus: "Matthew… first composed the Gospel of Christ in Judaea, in Hebrew letters and words [Hebraeis litteris uerbisque]… ; but we do not know for sure who later translated it into Greek.… This Gospel in Hebrew is still held today in the library of Caesarea [Maritima], which the martyr Pamphilus carefully put together. I also was able to make a copy from the Nazarenes, who use this volume in Bresoa, a city of Syria.…"

61Many of the Greek works of Evagrius Ponticus (such as the Commentaries on the Cherubim and Seraphim) are only extant in Syriac and Armenian translations. The same could be said of Eusebius of Caesarea's Chronici Canones, which but for a few fragments survive in an Armenian version and in Jerome's Latin translation. And the same again is true of the greater part of Origen's Peri Arxon (On the principles), of which only a translated text made by Rufinus in the year 398 has come down to us. The Apology for Origen which the martyr Pamphylus wrote in Greek has also disappeared, and only its first book survives in Rufinus's Latin translation.

62The apocryphal account of St. Paul's voyage, so popular in medieval times, was originally written in Greek, but we only have its Latin version, known as the Visio Pauli (Blake 1972, 30).

63In the year 1264, Bonaventura da Siena, notary of King Alfonso the Wise, translated from Spanish into French and into Latin the Mira'j or Book of the Scale of Muhammad, which had previously been translated from Arabic into Castilian by Abraham of Toledo. Both the original in Arabic and the text in Spanish have disappeared, and only their two translations, into Latin and French, have survived (Liure de Leschiele Mahomet).

64Throughout the thirteenth century and the first decades of the fourteenth, Raimundus Lullius wrote many of his treatises in Arabic, in Latin, or in Catalan. Though some of his first originals are now lost, they are preserved in his own translations, such as the book On the Contemplation of God, which Lullius wrote in Arabic but which remains only in Catalan and in Latin.

65Let us skip over the fourteenth and fifteenth centuries, because there are too many cases of lost originals preserved only in translation to be mentioned here, and limit ourselves to a well-known example of the seventeenth century.

66In 1660 the Dutch philosopher Benedict (Baruch) Spinoza wrote in Latin his Short Treatise on God, Man, and His Well-Being. A few months later, at the request of some friends "whose Latin was less than fluent" (Nadler 1999, 186), he translated his text into Dutch. "Although originally written in Latin, like all of Spinoza's other writings, the work has only come down to us in a Dutch translation, and it was not published or generally known until its discovery in the nineteenth century" (Allison 1975, 23).

Mistakes

67Gaps, holes, blank spaces,… and mistakes, too, which must absolutely be amended — little pieces in the mosaic which definitely do not belong to it. In fact, one of the most important tasks of today's historians is to denounce, correct, and eradicate the serious mistakes that have slipped into a good number of present-day texts. Let us stop to consider two of these errors because, besides serving as examples, they concern basic points of the history of translation in Europe, and particularly in the Iberian Peninsula.

68Example number one: It is high time that we stop speaking of the socalled Toledo School of Translators. Never was there such a school in Toledo, neither in the twelfth century with Bishop Raymond nor in the thirteenth century with King Alfonso X the Wise. However, this school has proved to be one of the most lasting myths in the history of contemporary culture, present everywhere, from the Encyclopaedia Britannica to thousands of pages on the Internet.

69It was Amable Jourdain who in 1819 first made the mistake of speaking of a "collège de traducteurs" in that Spanish town, when in fact the translators he ascribed to the "Toledo School" carried out their tasks in many other parts of the country, places sometimes quite far from Toledo. The error, however, soon spread all over Europe, and after Jourdain many other nineteenth-century scholars used to refer to the same "school." One of them was Valentin Rose, who in 1874, in his article "Ptolomaeus und die Schule von Toledo," spoke of "eine förmliche Schule arabisch-lateinischer Buch- und Wissenschaftsübertragung" (Hermes, 8:3, 327–349, 327).

70No, there was no such School of Translators at Toledo, if "school" is to be understood in any of the normal senses of the word. Already in 1942 Angel González Palencia (1942, 118) had to admit that "the very few documents which so far have been found do not provide a basis to assert the existence of such a centre of translations." Half a century later, in 1998, Clara Foz published in Ottawa a book under the title Le traducteur, l'Église et le roi (soon translated into Spanish: El traductor, la Iglesia y el rey, Barcelona: Gedisa 2000), dealing precisely with translation in the Iberian Peninsula during the twelfth and thirteenth centuries. Her assertions leave no place for further speculation. She writes (1998, 105–107):

An examination of the data with regard to translators of that era and their travels shows that in fact, Toledo was only one of the places certain twelfth-century men of letters went to.… Nothing suggests, however, that a space devoted to translation activities existed in the Castilian capital and that the work was directed.… The precise framework under which their work was carried out remains open to conjecture.… Indeed, no specific description of the exact places where these activities took place exists. [our translation]

71Four years ago Anthony Pym published the volume Negotiating the Frontier: Translators and Intercultures in Hispanic History, whose first four chapters deal precisely with this "school." Three quotations gleaned from Pym's book (2000, 34, 35, 56): 1. "The references to Toledo are not strict in any geographical sense, since twelfth-century translations from Arabic were carried out in many parts of Hispania, often in the north"; 2. "No medieval author would seem to have referred to Toledo as a 'school,' much less as a school of translators"; 3. "The notion of a 'School of Toledo' (in very capital-letter senses) has been mythologized in such a way as to make its historical coverage as broad as possible."

72More or less the same can be said of the famous ninth-century Baghdad "school of translators." Myriam Salama-Carr (1990, 31; Pym 2000, 36) defines it as merely "a team or group of translators." And then there is Dimitri Gutas, Yale University (1998, 59), on the Baghdad "House of Wisdom" (bayt al-hikma) "school":

It was certainly not a centre for the translation of Greek works into Arabic; the Graeco-Arabic translation movement was completely unrelated to any of the activities of the bayt al-hikma. Among the dozens of reports about the translation of Greek works into Arabic that we have, there is not even a single one that mentions the bayt alhikma.… The first-hand report about the translation movement by the great Hunayn himself does not mention it.…

73Crystal clear, or as clear as daylight. The myth of the "Toledo School of Translators," however, has been so often repeated that it seems almost impossible to eradicate it. Few lines contain a larger accumulation of inaccuracies than the following short paragraph by Henri van Hoof in his Petite histoire de la traduction en Occident:

Starting in 1135, Archbishop Raymond of Toledo… founded a School of Translators, which was literally a school where courses were given and where, for more than a century and a half, Italians, Frenchmen, Englishmen, Jews, and Flemings earned renown, alongside Spaniards, in a vast translation undertaking sponsored by the Church (van Hoof 1986, 10; our translation).

74There was no foundation, and consequently nothing was founded, by any archbishop; there was no collège, no véritable école, no courses whatsoever; nor, finally, was it the Church that, for over a century and a half, sponsored the work of those translators.

75This last remark could be true of the first period, in the twelfth century, but it is rather difficult to think in such terms of the second period, in the thirteenth century, unless van Hoof considers King Alfonso the Wise to have been a member of the ecclesiastical hierarchy.

76This has been an ill-fated appellation, inherited from an initial collège, translated into German as Schule, Übersetzerschule, and then into English as School of Translators, Scuola di Traduttori in Italian (Mounin 1965), Escuela de Traductores in Spanish, and so forth.

77Another quotation from Henri van Hoof's Petite histoire: "The French monk Peter the Venerable… translated the Koran in 1139.… Robert of Chester provided a new translation of the Koran in 1141–43, with Hermann of Carinthia" (van Hoof 1986, 10-11; our translation).

78Commentary and correction: Pierre de Montboissier, known as Peter the Venerable, Abbot of Cluny, never translated the Koran, neither in that year nor on any other date. Yes, he entrusted Robert and Hermann with the task, and this was the first time it was translated into any European language. But neither Robert nor Hermann could make a new translation, because the one by Peter the Venerable belongs to the realm of imagination.

79Van Hoof again (ibid.):

It was not until around 1200 that copies of Greek originals began to arrive in Toledo and that people began to recognize the value of making direct translations of them, without passing through a third language. From that moment on, the School of Translators translated… from Greek to Latin.

80Commentary and correction: there is no evidence whatsoever that around the year 1200 Greek originals began arriving in Toledo, nor that direct translations from Greek into Latin were being made there.

81Example number two: Four years ago J. F. Ruiz Casanova published in Spain (Madrid: Cátedra, 2000) a thick volume, over five hundred pages, entitled Aproximación a una historia de la traducción en España (Approximation to a History of Translation in Spain). Three short quotations from the very first pages: 1. "If we limit ourselves to the Iberian Peninsula… , we know that… in the fourth century bishop Ulfilas translated from Greek into Gothic all the Bible except the Books of Kings" (23–24); 2. "However, this Gothic Bible was not the first sample of translation in the Iberian Peninsula after the fall of the Roman Empire" (46); 3. "Apart from the translations made in the Iberian Peninsula from the 4th to the 6th centuries, done by Ulfilas, St. Martin and St. Pascasius…" (54).

82The three quotations repeat the same idea: in the Iberian Peninsula, Ulfilas translated from Greek to Gothic all the Bible, except the overly warlike Books of Kings. I imagine that many a Germanist has already gone through the roof: Ulfilas never was in the Iberian Peninsula, not even near it, and, consequently, he couldn't in any way have made his famous translation there. He was the Arian bishop of Lower Moesia, not far from today's town of Trnovo, in Bulgaria, and it was there, and in Constantinople, that he spent forty years translating the Bible. Exactly the same mistake has been repeated by Alberto Ballestero in his Diccionario de Traducción: "Translation in Spain starts officially at the beginning of the twelfth century… , although already in the fourth century bishop Ulfilas, who, besides Gothic also knew Greek and Latin, had translated the Bible into Gothic" (Ballestero 1998, 19; our translation).

83Errors like these are quite common in many pages of history, and, what is worse, in some histories of translation. As historians, it is our duty to remove, erase and rub them out completely.

The role played by translation in History

84Not only is translation the object of historical study, but it has also played a leading role in History (this time with a capital letter). Translation itself has made History, a particular which has seldom been taken into account. Think, for instance, of what has become known as the anti-Spanish Black Legend, something still alive and kicking in many pages of History, which keep on ascribing the beginning and success of the legend to a work by the Dominican friar Bartolomé de las Casas, the Short Account of the Destruction of the Indies (Breuissima relacion de la destruycion de las Yndias). However, this general opinion needs to be carefully nuanced. The Short Account was first published in Seville in 1552. A year passed, and ten, and twenty… , and dust piled up on the shelves on which the Short Account was lying, forgotten by everybody, without even a second edition for as long as twenty-five years.

85But in those twenty-five years many things had happened in Europe. The Low Countries were now up in arms against the Spanish government, England and France were also sworn enemies, and as in any war, propaganda became a powerful weapon. And the forgotten Short Account was a perfect propaganda tool against Spain, first because it was written by a Spaniard, and on top of that a man of religion. Secondly, because spreading abroad its contents, laying it on thick, helped display the intolerance and perversity of the Spaniards throughout their territories.

86In only five years, 1578 to 1583, the Account was translated into Dutch, French, and English, with five different editions in Antwerp, Paris, and London. This was followed year after year, insistently, throughout a whole century, by no fewer than thirty new editions and translations into French, Dutch, German, Latin, and Italian. Obviously, translating in that sociopolitical atmosphere was equal to manipulating very consciously — manipulating everything, from the very title to the colophon. Little wonder, then, that the first translation of the Short Account into French in 1579 bears the title of Tyrannies et cruautés des espagnols, perpétrées aux Indes Occidentales; or that the 1656 English edition appears under the title of The Tears of the Indians: Being an Historical and True Account of the Cruel Massacres and Slaughters of above Twenty Millions of Innocent People.

87However, it was the Latin edition published in Frankfurt in 1598 (Narratio regionvm Indicarvm per Hispanos qvosdam deuastatarum verissima) which caused a major impact on the public. Indeed, in a way, it became "canonical," particularly because of the sensational engravings of tortures illustrating a text which, manipulated over and over again to unimaginable extremes, was by now quite far removed from the original. Friar Bartolomé, for instance, had written (Pereña 1989, 216; our translation):

The care they took of them was to send the men to the mines to dig out for gold, which is an unbearable kind of work; and the women were sent to farms, in order to dig up the earth and cultivate the fields, which is a job for very strong and sturdy men.

88This same paragraph, at the foot of a horrifying engraving of tortures, became in the Frankfurt edition as follows (ibid.):

Those who worked in the mines, or elsewhere, and who were not diligent enough, were also treated in a most pitiful way; for not only were they tied up to poles, but they were also flogged with whips soaked in pitch, so that they were left as dead. And what is even worse, after having flogged them in such a cruel manner… the Spaniards also dropped burning grease on their wounds.

89The origin of the Black Legend, therefore, does not lie in the Short Account, which per se did little to create it, as was more than evident in those first twenty-five years after its publication, 1552 to 1577. The true origin must be looked for in the manipulated translations which uninterruptedly fuelled anti-Hispanic prejudices starting from 1578, translations into Dutch, French, German, English, Latin, and Italian. They were the true and only vehicle of diffusion of this legend all over Europe, at the service of very specific political and religious interests. They mark, as Eric Griffin has recently written, "a pivotal moment in one of the most successful propaganda campaigns ever carried out: as the Brevissima relacion is translated, printed, and reprinted in contexts far removed from that of its initial publication, the acts it recorded gave rise to the Hispanophobic typology."10

90This is just one example, but it could easily be multiplied ad infinitum because more or less the same could be said, in those same years, and along the same lines, of the Apology of William of Orange against Philip II, translated in scarcely twelve months (1581–82) into French, German and Latin; and the same could be said, again, of the Sanctae Inquisitionis Hispanicae artes aliquot detectae, published in Latin in Heidelberg in 1567 and immediately translated into English, French, and Dutch. The English edition, appearing in 1568, bore the title of A Discovery and Playne Declaration of Sundry Subtill Practises of the Holy Inquisition of Spayne.

91In our own days, a wrong translation might have been the ultimate reason for the dropping of the first two atomic bombs. On July 28, 1945, two days after the Potsdam Declaration, the Japanese Prime Minister Kantaro Suzuki declared at a press conference that the Potsdam Declaration was "a thing of no great value," and added: "We will simply mokusatsu it." In Japanese mokusatsu is a rather ambiguous word, rather untranslatable into English, and "certainly not by a single word, for there is no English equivalent."

92"Regrettably," it has been written,11 "the junior State Department official in charge of translating the Japanese reply lacked the necessary linguistic sophistication, and missed the subtle subtext of the reply altogether. Instead, he reached for a Japanese-English dictionary and translated mokusatsu by the closest single-word English equivalent, which happens to be 'ignore': 'We ignore the Potsdam Declaration.'" Which was also interpreted by the press in the United States and the United Kingdom as "We reject it."

93The result? The Allies took the statement as rejection of the Potsdam Declaration, and two atomic bombs were dropped on Hiroshima and Nagasaki, on the 6th and 9th of August.

94Se non è vero, è ben trovato, because this is how History is made, how it has been made throughout the centuries, threading its way through the silent protagonism of translated texts — so silent, indeed, that we are not fully conscious of how many strings translations may have pulled all along the centuries.

95Of course, there are many other gaps and "blank spaces" in the history of translation that could have been mentioned here: areas so far very little attended to, even clearly neglected and overlooked along the main axes of time and space:

• interesting debates and controversies, such as the thirty-year debate on translation between Jerome and Augustine, Rufinus, Pammachius, Oceanus, and several others, during the last decades of the fourth century and first decades of the fifth century;
• the incorporation of minority and/or minorized languages into translation history research;
• the impact of translations on many literary processes, the provision of new models for national literatures, and the influence of translations on the birth and development of new genres in some target polysystems (genres such as western novels, detective stories, science fiction, fantasy novels, even comics);
• the introduction through translation of new cultural, aesthetic, or political values;
• and above all, maybe before anything else, the urgent task of de-Westernizing the history of translation.

96Translation in Latin America, for instance, has hardly attracted the historian's attention: it is an uncultivated land still waiting to be ploughed. Five years ago Georganne Weller, Co-director of the Center of Applied Linguistics, Mexico City, remarked that "in general, and with a few notable exceptions, little attention has been paid to the history of translation and interpretation in Mexico by related professions, academic institutions, and professional associations." The same, or very similar, words could be said of Chile, Brazil, Argentina, Venezuela, or Peru.

97It may be my fault, but the truth is that I have never come across any mention, for instance, of the first printed translation from English into Spanish, which was made in Peru in the year 1594. An English ship, the Dainty, had been captured, and its captain, Richard Hawkins, arrested and charged with piracy. That was on the 22nd of June, after Hawkins had attacked Valparaíso. During the first days of August, Hawkins wrote a letter in English to his father, Sir John Hawkins, to inform him of what had happened. The Spaniards hastened to learn the contents of the letter, translated it, and printed it at once in Lima. It is only a short pamphlet of fourteen pages, lacking even a title page; it just starts with a modest heading, of this tenor: TRASLADO / DE VNA CARTA DE / RICARDO HAVQVINES, ESCRITA / en el puerto de Perico, en seys de Agosto, / de. 1594. años para embiar a su padre / Juan Hauqvines a Londres, tradu / zida de lengua Inglesa en lengua Castellana. [Translation of a letter by Richard Hawkins, written at the harbour of Perico on the 6th of August, 1594, to be sent to his father, John Hawkins, in London, translated from English into Spanish]. As far as I know, a single copy of this letter has survived; it is now in the British Library.

98The history of interpretation, the daily practice of translation, lost originals which only survive in translation, old theoretical and critical traditions of the East and Middle East which must be added to the already known tradition of the West, pseudo- and self-translations, translations as agents of History… gaps and blank spaces in a chronicle of four thousand, five hundred years, fragments of an unfinished picture, tesserae of a mosaic still waiting to be filled in. The task involved is immense, and there is much doubt that historians of other disciplines or cultural areas would be inclined to take it over (they haven't so far). Neither is it going to be the task of one generation, not even two, because, to echo a familiar saying, "The harvest truly is great, but the labourers are few."

99"Sunt autem et alia multa… , quae si scribantur per singula, nec ipsum arbitror mundum capere posse eos, qui scribendi sunt, libros." These are the very last words of the Gospel according to John in the Latin translation by Eusebius Jerome. Certainly, it is quite probable that when all these and many other blank spaces have finally been filled up, we historians of translation will also be able to repeat them, at least in the more understandable King James version: "There are also many other things [in the history of translation]… , the which, if they should be written every one, I suppose that even the world itself could not contain the books that should be written."