© Anthony Pym 1999
On Sunday 5 April 1422, while staying in a Toledan monastery, Don Luis
de Guzmán, Grand Master of the Order of Calatrava, commissioned
‘vna biblia en rromançe, glosada e ystoriada’ (a bible in Romance,
with explanations and illustrations) from rabbi Mose Arragel, who was at
that time in Maqueda, to the north-west of Toledo. Rabbi Mose initially
refused the request because, among other reasons, it was against his beliefs
to produce a sacred book with illustrations. Yet the Grand Master insisted,
the rabbi gave way, and a text was produced. That translation was then
scrutinized and ‘corrected’ at the Franciscan monastery in Toledo, with
the whole revision process being completed in June 1430, more than eight
years after the initial commission. The resulting manuscript is a truly
magnificent document, both in view of its superb illustrations and,
perhaps more important for the kind of history being constructed here,
because of the way the translation manages to express both Jewish and Christian
In this case the translation project involved negotiation in rather more than a metaphorical sense. Throughout, it was an asymmetric collaboration of Jews and Christians, not only in the disputed commissioning and the long revision process but perhaps also in the manuscript production itself. Keller (1992: 156) believes it is ‘highly probable’ that the manuscript was copied by Jews, since notes in Hebrew are sometimes found. Through a serious of diplomatic concessions, it seems, representatives of the two religions could agree well enough to put a bible together.
Much of how this worked is explained in the rabbi’s long explanatory introduction, and can indeed be inferred from many of the work’s 6,000 or so glosses, which make this a privileged document for the study of intercultural translation processes. Yet suggestions might also be gleaned from minor discrepancies such as we find, for example, in the full-page miniature (hopefully reproduced on the front cover of this book) which portrays the hierarchy of production. Rabbi Mose is at the bottom delivering the bible to the Grand Master, who is apparently both there and magnificently enthroned at the top, with the translated bible, because of its importance, clearly outsizing the belittled translator. The illustration (fol. 25v of the manuscript) appears just after rabbi Mose’s introductory text and directly opposite the beginning of the translation proper, at the first book of Genesis. What is strange here is that the bible shown in the miniature is open at the same text as the one it faces, the beginning of Genesis, allowing for a potential mise en abîme. The only problems are that, first, the bible in the miniature does not show any illustrations (i.e. the scene does not show itself), and second, the translation given in the illustration, In principio creauit deus celum et terram..., is obviously the Vulgate and cannot be confused with rabbi Mose’s Castilian translation, En el prinçipio crió el señor los çielos e la tierra.... Why should the illustration so obviously contradict the fact of the translation?
Nordström (1967: 30) suggests this was because the illustrators simply did not have access to the translation, which was perhaps undergoing revisions at the time. This may be so. But why believe it was an accident or a trivial case of disorganization? We might also see the contradiction as a motivated separation of translation and illustration. Just as the graphic artists were visibly not illustrating the work of a rabbi (they do not cite his translation, so they could claim not to have seen it), the work of the rabbi is not presented in the illustration. In fact, for the illustrators and perhaps for the commissioning ideologues, what they are working on is the bible as they best recognize it, in the form of the Vulgate used in their churches. Here, thanks to the double representation, the Vulgate and the Jew’s Castilian translation are projected onto exactly the same space, as if one were worth the other. Such might be a claim to divine equivalence.
Of course, there may also have been minor potential advantages on the Jewish side. Only through outwardly separate production processes could anyone meet the rabbi’s initial objection, namely that his faith prohibited him from producing an illustrated text. The translator ostensibly did not work with the illustrators; the illustrators ostensibly did not work on the translation; yet an illustrated translation was the result. And in this intercultural separation of tasks, all might have been more or less happy; or at least, no articles of faith were radically upset.
Here we shall focus on this process of bringing different ideologies into the production of a common text, and specifically on the way the rabbi’s translation strategies and glosses fed into such a process. But first, in order to appreciate how an intellectual rabbi might relate to the Grand Master of a warrior order, a few notes are necessary on the modes of patronage prevalent at the time.
Changes in Patronage Structures
We have jumped over a century or so. No, Hispanic translators did not
suddenly become inactive after the passing of Alfonso X. In fact, they
were remarkably productive in the fourteenth century. Since lamentably
little attention is given to that period (see Santoyo 1994), let us briefly
summarize the necessary background.
Repeatedly opposed by the aristocracy, Alfonso X renounced his imperial candidature in 1275 and was finally deposed by his son in 1282, leaving Castile-León in political turmoil (so much for his wisdom!). In the absence of strong crown sponsorship, the practice of translating from Arabic tapered off in the ensuing decades, albeit without entirely disappearing. As we noted earlier, in 1455 ‘Isa de Jabir, an alfaqui from Segovia, had started translating the Qur’an into Castilian for Juan de Segovia, who then translated it from Castilian into Latin, thus using the same relay method that had been developed in Toledo (see Cabanelas Rodríguez 1952; d’Alverny 1989: 202). But this translation was not carried out in Hispania. Juan de Segovia was in Ayton, in the Savoy, whence he had brought ‘Isa de Jabir from Segovia. Hispania was no longer the place where anyone went looking for knowledge from the Arabic. At best it could provide the occasional native informant.
An associated sea change concerned general patronage structures. We have seen the general sponsorship of translations move from church to crown in the course of the thirteenth century. Crown patronage, both direct and through the employment of translators as royal secretaries, certainly continued in fifteenth-century Aragon-Catalonia, particularly when Alfonso IV set up his court in Naples from 1443 to 1458 (the Crown of Aragon extended to Sicily, Naples, and momentarily as far as Athens). In Castile, however, the center of gravity shifted to the nobility. The centralized authority of the Castilian king had been weakened; noblemen could do virtually what they liked within their territory; they paid remarkably little heed to national purposes; they fought among themselves; they occasionally paid for the services of scholars and translators. Translations thus entered the struggles of virtual or real civil war.
The model patron of this period was undoubtedly Íñigo López de Mendoza, marquis of Santillana (1398-1458). He was a particularly active sponsor of translations, receiving Latin versions directly from the Italian peninsula and having authors such as Virgil, Ovid, and Seneca rendered into Castilian. His translations were not limited to mere cultural capital. Since he also found time to defend the realm and then eventually lead the nobility against the king, we should not be surprised to find Santillana commanding a 1441 translation of something like L’arbre des batailles, a French handbook of military legislation. Nor should we be amazed to learn that Álvaro de Luna, Santillana’s arch political rival, commanded a translation of the same book in virtually the same years (Alvar and Gómez Moreno 1987). Certain translations might help win or justify wars. From the perspective of scholarly ethics, of course, certain other translations might also help educate or refine noblemen who could perhaps then win wars with a little more moral justification. Either way, the purpose of translations, in whatever genre, was generally considered in terms of fairly immediate utility.
Luis de Guzmán, Grand Master of Calatrava, was functionally part of this warring nobility. Founded in 1158, the Order of Calatrava was named after the frontier town, south of Toledo, from which it was called upon to defend the recent conquests from the Moors. Although the order initially included both monks and laymen, it then became a group of knights whose aim was to fight Islam. More important, as befits frontier society, these religious knights were not under the command of the monarchy. From 1187 they were a branch of the Cistercian Order and thus owed ultimate allegiance to the pope. Thanks to their relative political independence, the Order of Calatrava gained enormous economic and political power in the course of the centuries, exercising control over many Hispanic regions (it had been master of Calaceit in the thirteenth century) and, in 1404, successfully resisting attempts by king Enrique of Castile to have his cousin appointed as its Grand Master. The order itself chose Luis de Guzmán as its head; its decision was ratified by the papacy in 1414; Calatrava’s independence from the monarchy was made clear.
It was thus from a position of relative independence within a weak monarchy—Juan II had not turned two when he came to the throne of Castile in 1406—that Luis de Guzmán became a very active player in the military struggles of the fifteenth century. Much of his power was accrued from alignment with Álvaro de Luna, who virtually ruled Castile for the young Juan II before eventually being ousted and excecuted. When the Grand Master’s bible was completed in 1430, it seems he was away fighting alongside the marquis of Santillana, at that time defending Castile from invasion by Alfonso V of Aragon. So the magnificent presentation scene depicted in the miniature never actually took place. War was the first priority of the age.
Like a surprising number of his fellow patrons (perhaps from Raimundus Toledanus down), the Grand Master appears not to have been particularly enthralled by scholarship. Although he ordered a magnificent bible, the parties most interested in the manuscript’s contents rather than its magnificent appearance were undoubtedly the Grand Master’s two cousins: Arias de Ençinas, who was father superior of the Franciscan Monastery in Toledo, and Vasco de Guzmán, archdeacon of Toledo (Keller 1992: 148). So the actual translation, despite all the changes in patronage structures, was still largely controlled by church structures in Toledo. A warrior nobility was needed for the finance, but scholarship, especially of the religious kind, was not to be identified with the people actually paying. This set up a double patronage structure. On the one hand, the agent of relative action, the nobility, put up the money, remained relatively external to the translation process, and valued the outward appearance or symbolic value of the resulting books. On the other, the agent of ideological justification, the church, was intimately concerned with the translation process and with what the translator’s words might actually say. Working for the two, somewhere, was a Jewish translator.
This peculiar relation between scholar and patrons creates serious problems with respect to the ultimate loyalties of most fifteenth-century translators, be they Jewish or not. It seems plausible that the translators, like other intellectuals of the day, not only took their money where they could find it but were also interested in the long-term education of the nobility, broadly in keeping with the humanist models that were creeping into Hispania. A patron like the marquis of Santillana would openly declare his susceptibility to humanist aspirations, lamenting the limits of his education (particularly his difficulties with Latin), and commissioning translations in order to overcome those limits.
Yet the marquis was perhaps something of an exception to the rule. In a treatise by Alonso de Cartagena from this period (ed. Lawrance 1979), written for the first count of Haro, we find a pessimistic description of what the ‘men of action’ willingly read, and a call for structured reading programmes specifically for the medium genus, for people between the active and the contemplative life, which would include counts, marquis, and perhaps Grand Masters. This virtually amounted to a call for the development such an intermediary class, where people could both fight and think. Cartagena recommended that the reading programmes comprise books that were ‘healthy for the mind, morally uplifting, and adapted to the capacities of the reader,’ leaving heretical works to be read only by professional scholars (Lawrance 1979: 11-12). We shall return to Cartagena’s position in the next chapter. For the moment, what is of interest here is that rabbi Mose’s bible translation, which certainly incorporates ‘heretical’ glosses and is a serious work of detailed scholarship, was clearly not part of any such reading programme. It was primarily intended for professional readers, for churchmen, for relatively contemplative ideologues, who were not necessarily to be confused with the paying and fighting patron.
The question of loyalties becomes slightly more complicated when we consider the translator’s position as a Jew. In 1391, only some 30 years prior to the commissioning of the translation, Jews had been massacred in Seville, Cordoba, Montoro, Jaen, Toledo, Madrid, Segovia, Valencia, Barcelona, Palma, and Girona. According to some accounts, about half the Jews in Hispania converted to Christianity. (We might also note that, in July 1390, the son of the Great Rabbi of Burgos was baptized as Pablo García de Santa María, who became bishop of Cartagena in 1401, bishop of Burgos in 1435, and wrote as Alonso de Cartagena, author of treatises on reading programmes, translations, and much else as well.) In 1410, only twelve years prior to the bible commission, Martí I closed the Jewish quarter (call) in Barcelona and took all the land, buildings and goods for himself (Miró Montoliu 1996: 19). In 1412 Catherine of Navarra proclaimed that all Jews were to live in separate quarters. From February 1413 to November 1414 there was the public disputation of Tortosa, held before the antipope Benedictus XIII and in the presence of about 1000 Christians, opposing collected rabbis from the Crown of Aragon. The disputation took 22 months to decide whether or not the Messiah had already come, famously resulting in fourteen rabbis converting to Christianity (Marcó i Dachs 1977: 295-305). In 1415 Benedictus XIII issued Et si doctores Gentium, a series of anti-Jewish measures that probibited, among much else, the reading of the Talmud, the circulation of Hebrew books that contradicted Christian doctrine, and Jews pronouncing the names of Jesus, Mary, and so on. The bull also closed many synagogues and severely restricted the professions available to Jews. Note that this is just seven years prior to rabbi Mose’s commission. Given the general climate, it is not too surprising that there were an estimated 250,000 conversions to Christianity between 1391 and 1415 (Marcó i Dachs 1977: 300). By 1449 there were widespread anti-Jewish riots; in the 1460s a generalized expulsion was being considered by Alonso de Espina (Schonfield 1992: 10). In 1465 Joan II of Aragon tried to get rid of non-baptized Jews by sending them to Mallorca (Miró Montoliu 1996: 19). In 1477 the mistrust of Conversos was such that the Holy Inquisition was established. By the time it was abolished in 1834, the Inquisition would have burned alive some 31,000 people.
As rabbi Mose set down to work on his translation in 1422, Hispanic anti-Semitism might have gone underground for a while yet it could never have been far from the social surface; it would of course surge up in the definitive expulsion in 1492. Although Lazar describes the fifteenth century as a period when ‘Jewry was raising itself out of the period of turmoil to new heights of activity in literary, scientific, philosophical and kabbalistic studies’ (1992: 160), one should not pretend it was an entirely comfortable time for Hispanic Jews. Then again, at least in the frontiers of the world, one should not expect intellectual productivity and social comfort to go hand-in-hand.
Perhaps more important, any renewal of Jewish activity might have been favored not just by an occasional Grand Master with religious cousins, but also by kings, especially those monarchs who had not entirely forgotten the tradition of protecting the ‘three religions’ of frontier society. Both Juan II of Castile and Alfons V of Aragon (IV of Catalonia) abolished the anti-Jewish laws set in place by the previous generation. Both sought to restore the integrity of Jewish communities despite widespread religious opposition. When Alfons V ordered the judges of Girona to protect the Jews during the Easter of 1418, his order indicated that they needed serious protection. The general strategy to be seen here is that of the ‘king’s Jews’, the figure of Jews as professional intermediaries benefiting from the direct or indirect protection of the Crown, in the face of attacks from the Church and the nobility. It is a figure that had survived at least from 1215, when the kings of Castile and Aragon refused to apply the fourth Latran Council’s declaration that Jews be excluded from civic life; the ‘king’s Jews’ were very much the translators working at the Alfonsine court in the thirteenth century; they might also have included the Converso Cartagena, who was in the employ of Juan II from 1420 to 1440, the years in which he will be most active in our stories. The figure of the ‘kings Jews’ would nominally reappear in 1481, when king Fernando (‘the Catholic’) would move to protect non-converted Jews, his direct ‘vassals,’ from the inquisitions being set up to denounce false Conversos.
A certain leftist history might try to associate this crown-jewry allegiance with the similar loyalties manifested by the peasantry at the time. As the warring nobility drove the land to economic ruin, popular movements did indeed declare their loyalty directly to the monarch, jumping over the strata of feudal society (cf. Pérez 1995). One might then romantically see an intellectual caste, largely comprising Jews and Conversos, remaining secretly loyal to monarch and people, and for that reason attempting to instill humanist values in the warring nobility so that they, the nobility, would learn to fall in line with monarch and people. That is a nice story. Whether or not it has much basis is largely a question for our next chapter, where the strength and meaning of those humanist values will be assessed. For the moment, let us merely stress the significant risks faced by Jewish intellectuals. Despite long periods of cooperation between king and Jewry, when the crunch came the Hispanic kings had and would privilege other alliances. Alfonso X turned against his main Jewish administrator; and as for Fernando defending the Jews in 1481, just two years later he and his queen Isabel would order 4,000 Jewish families to convert or leave; in 1486 he would order the expulsion of Jews from Zaragoza and Albarracín; the rest is well known. Although there may have been significant cooperation under the figure of the ‘king’s Jews,’ there is little reason to believe there was deep-seated long-term trust.
Of course, when one’s patron was the head of a Christian militia, acting in the interests of churchmen in Toledo and with relative independence from the crown, there was surely even less reason for trust.
If we now think in terms of the real or virtual disputations that framed translations like the Latin Qur’an, we can perhaps measure some of the uneasiness with which a rabbi might receive a command from a Christian patron. Was this translation really meant to educate a Christian reader? Had the Grand Master and his cousins really turned to a Jew simply because there were no good Christian scholars around? Or was part of the patrons’ purpose—shared by both church and nobility—to gain information, the inner secrets, that could then be used against Jews in disputations, in the hunting out of false Conversos? The strategies of the Qur’an translation were still pertinent, except that now we have the translator on the other side, ideologically aligned with the source culture. It is important to understand the terms on which he agreed to undertake the translation.
Negotiating the Brief
The details of the commissioning process are given in the long introduction
to the translation, where we actually find copies of the letters involved.
The Grand Master’s initial letter, dated 5 April 1422 (fol. 2r-a), states
that he wants the bible for two reasons: ‘first, the vernacular bibles
that we have are flawed [sson fallados] and their language is corrupt [el
su rromançe es muy corrupto]; second, those that we have are in
great need of commentaries [la glosa] for the obscure passages’.
The would-be patron also goes to some pains to stress that he would much
rather listen to the bible than go off hunting or play chess (well, he
is supposed to be head of a holy order), that he has two Toledan theologians
to ‘help’ with the work, and that the rabbi would be well rewarded for
his efforts, albeit in terms that are rather less than specific: ‘we shall
grant you many goods and privileges every year.’ The next letter is the
rabbi’s reply to the Grand Master, staunchly refusing the commission and
giving many reasons for the refusal, including his objection to the illustrations.
Then we have a rather more abrupt second letter from the Grand Master,
asking rabbi Mose to come to Toledo to meet Arias de Ençinas, who
will assist him with the translation. Clearly, Guzmán was not about
to take the rabbi’s refusal seriously. There follows a letter directly
from Arias de Ençinas to the rabbi, inviting him to Toledo and explaining
certain concessions with respect to the brief: the rabbi would only have
to make summaries of the main rabbinical positions on various passages;
the Grand Master genuinely desired to learn about those positions (thus
suspiciously repeating the claim that pure learning was the aim, although
why the Grand Master might have this desire is not clearly stated); the
Christian views to be included would be given by reverend Arias himself
(so the rabbi should not really be worried about them); and as for the
illustrations, well, the translator only had to leave a blank space after
each chapter, and the artists would find their models in an illustrated
bible in the cathedral (which explains away a few apparent contradictions
in the illustrations). The collection of letters finishes with a note saying
that rabbi Mose went to Toledo three days after receiving Arias’s conditions.
It was apparently an offer he could not refuse.
Why did the rabbi refuse once but not twice? To improve his bargaining position? If so, the ploy did not concern questions of payment, since no fixed recompense is mentioned in the first letter and there was thus no initial position that could be improved upon. One suspects that the initial refusal, if a bargaining strategy, served to buy respect for the rabbi’s position as a non-Christian. In effect, it allowed his objections to be negotiated and understandings to be attained in all the key areas: the rabbi would do his work, the Christians revisers would do theirs, and the rabbi and the illustrators would be kept diplomatically apart.
Let us note just one further detail. Toward the end of his letter of refusal, rabbi Mose explicitly makes an obeisance to Juan II, an act clearly illustrated in the manuscript (fol. 11r) where the translator is shown kneeling before the king and about to kiss his foot, with the caption: ‘The noble king Juan, son of the noble king Enrique the Second / His servant rabbi Mose’ (miniature reproduced and commented upon in Nordström 1967: 14). This is a strange act of reverence, particularly considering the mention of the very king, Enrique, who had fought against Guzmán’s appointment as Grand Master of Calatrava. Perhaps here, in expressed allegiance to a monarch who otherwise had nothing to do with this translation, we find part of a strategy whereby the translator (and the illustrators?) could mark his distance from his patron, gaining leverage and thus creating room to manoeuvre. The king may have been a useful figure in the negotiating strategies. However, once in the text itself, rabbi Mose begins with lengthy praise of God and then brief mention of his patron (fol. 1v-a), with not a word for any king in between.
Rabbi Mose’s declared strategy was to opine nothing but to report everything
(including, as we have seen, the letters that formed the context of his
reporting). In his introduction, written after completion of the translation,
he claims he has not criticized the beliefs of the Christians, he has simply
shown what Christians and Jews believe; and although he clearly stands
on one side rather than the other, readers will be able to choose whatever
interpretations they want: ‘Since I have done no more than relate or record
[memorar], everyone is left free to believe, dispute [disputar] and defend
their law as much as they can’ (fol. 15r-a). The translator thus strives
to remove his first person from any situation that might be seen in terms
of a disputation as such. He would have his commentaries be known by as
neutral a name as possible: ‘La Memoratiua,’ the Memorandum (fol. 15r-a),
as if he were on the sidelines, looking in on a disputation and merely
taking note of the points raised by both parties.
Rabbi Mose lists four further general principles by which be proposes to write his glosses. Since those principles form a fairly explicit regime, they are worth repeating (here we work from Lazar 1992: 165, who actually lists the ‘memorandum’ concept as a fifth idea; the names of the principles are ours):
1. Principle of default compatibility: When there are no major differences between Christian and Jewish interpretations, the gloss should be acceptable to both religions.
2. Principle of non-involvement: If Christian readers find an interpretation unacceptable, they should regard it as Jewish. And if Jewish readers should have the same problem, they should regard the interpretation as Christian. (This virtually amounts to the claim to be a ‘non-disputation;’ in effect, it means that the readers decide everything, the active third is excluded, and the rabbi can thus never be wrong.)
3. Principle of parallel manifestation: Wherever strongly divergent interpretations exist, they are clearly noted as such in the gloss. (This would also be the principle behind the long introduction, and indeed behind the decision to announce the principles themselves.)
4. Localization disclaimer: The commentaries are the best the rabbi
could find in the rabbinical and Christian texts made available to him.
He is thus not responsible for not citing texts he has not seen.
Together, these principles may be viewed as a self-protection program, designed to keep the translator well out of the ring of any disputation and to absolve him of any complaints that might ensue. The way they work in the glosses is fairly clear, although questions of authorship sometimes arise. Two examples should suffice.
The first is a detail concerning how many members there were in Jacob’s family in Egypt (Exodus 1:5). The translation reads: ‘E fueron todas las almas que salieron de la ancha de Jacob setenta e çinco almas con Joseph que estaua en Egipto’ (And all the souls that came from the loins of Jacob were seventy-five souls with Joseph who was in Egypt). The commentary reads:
Seventy souls. That is the number in the Hebrew text, not more; but in the translation of Saint Hieronymus it is written ‘seventy-five’; and is similarly found in the Acts of the Apostles, chapter seven; and Nicolaus de Lyra says that the Law mentions the number twice, once saying seventy, and next seventy-five; but the Hebrew text mentions only seventy in both instances.
We thus see a range of variants mentioned; Hieronymus, Luke, and Nicolaus
de Lyra are all allowed their say (the number 75 paradoxically comes from
the Septuagint, in both the instances cited), but, insists the rabbi, the
Hebrew says 70, not 75, so there. Few could argue with the rabbi’s authority
on this point. And yet, in the actual translation, we find the words
e çinco (and five) have been added above the line.... The Christian
correctors allowed the rabbi to say what he liked, just as he allowed them
to say what they liked, but they ultimately directed the stylus that wrote
the translative number. So the translation and the gloss are left in glaring
contradiction; the regime for commentary seems not to have extended into
the translation proper.
Our second example concerns the use of the plural verb in Genesis 1:26: ‘dixo el Señor: fagamos omne a nuestra ymagen e nuestra semeiança’ - And the Lord said, let us make man in our image and our likeness’:
We have here a secret of faith, namely: one, calling for a plural, is not appropriately expressed in writing by saying: ‘let us make;’ even less should one believe that the Lord God wanted to take advice from angels, or spheres, or elements, which are all subjugated to God, and it does not stand to reason that he should have held counsel with his inferiors. But what stands to reason is as follows: the Father said to the Son: ‘Let us make man in our image and our likeness,’ and this likeness is meant for the Trinity [...]
And the explanation of the Trinity then runs on for three rather convoluted
sentences. Here it is easy enough to locate rabbi Mose’s input as the resolution
of the problem as a ‘secret of faith;’ and we can be fairly sure that the
passage about the Trinity, beginning ‘But what stands to reason is...,’
gives the words of the rabbi’s Christian revisers. However, in between
these two interpretations, exactly who is arguing against the idea that
God’s plural incorporated the angels and spheres (which had been created
before humans)? And why is that idea not named explicitly and ascribed
to one side or the other? This is surely far more than a mere
reporting of opinions. Presumably because the rejection of the ‘angel hypothesis’
was common to both the rabbinical and the Christian views, this intermediary
position could indeed be presented as a detached argument, in accordance
with the ‘principle of default compatibility’ announced above. And that,
of course, is why the compatibility principle should hold a higher position
in the regime than do the principles of non-involvement or manifest separation,
which in this this case are clearly overridden.
We thus have a set of principles that seem to work, more or less, for the production of commentaries. But how did they apply to the actual translating? What happened when, in the obligation to select one term or another, diplomatic accommodation was not always possible?
Even before we consider translation as an obligation to choose between
alternatives (which it need not always be), it should be clear that the
careful equilibrium underscoring rabbi Mose’s ‘commentary regime’ could
not really be applied in translation practice. This is because the translation
situation was profoundly asymmetrical, and in several ways.
First, although we have seen the rabbi envisaging a possible Jewish readership (‘everyone is left free...’), it is not at all clear how a bible produced for churchmen in Toledo was going to reach Jewish readers. Surely the short-term reception was all one side rather than the other, and the expressed regime depended on no more than a fiction of symmetry? Second, since there appears to have been no real negotiation about what ‘the bible’ was, the translation initially follows the order of books given in the Massoretic texts, and not that of the Vulgate. The latter imbalance, now favoring the Jewish side, is also clear enough in the thousands of glosses, which often refer back to the Hebrew text and draw on a wider range of Jewish scholarship than Christian commentaries (Rashi, Nahmanides, David Kimhi, Gersonides, as against Nicolaus de Lyra for the Christian readings). In fact, Lazar has called the translation ‘a Spanish-Jewish rabbinical bible draped in some Christian garments’ (1992: 162), the garments mainly being the 300 or so miniatures that might make the text look Christian, at least upon superficial inspection (perhaps the kind of inspection it was going to get from its patron). The third kind of asymmetry lies in the fact that, as we have seen, even when Hebrew scholarship was unequivocal and authoritative in its own terms (e.g. ‘seventy’ instead of ‘seventy-five’) the Christian theologians would still ignore that authority and adhere to their own traditions in the most foregrounded parts, namely in the translation rather than the commentary.
In view of these situational factors, quite apart from the linguistic constraints involved, the rabbi’s hermeneutics of translation could scarcely aspire to the symmetries he claimed for his commentaries. What we in fact find, in terms of both hermeneutics and politics, is an operative distinction between textual foreground (the translation proper) and textual background (the introduction, commentaries and glosses). Although much of the plurality of the translation process could find a place in the background, where it was indeed fairly well regulated by the rabbi’s explicit principles, this did not solve the more vital struggles over what words would be selected for the foreground, for the translation itself. No doubt aware of this problem, the rabbi does all he can to displace or postpone it, insisting that the plurality is in fact at all levels.
When rabbi Mose talks about his actual principles of translation, he is astute enough to cite a major Christian authority, no less than Hieronymus. He does this when making explicit reference to the principle we have earlier called alie translationes (see chapter two above), basically involving awareness that no one translation is definitive and that scholarly work thus has to deal with a plurality of versions. The rabbi finds this principle at the heart of the Christian tradition:
Sant Jeronimo is said to have done this translation [of the bible] three times, and only approved one of them, and none of us know which one. And in Madrid and in Cuellar in our Castile there are two Latin bibles that are much more in agreement with the Hebrew than the one normally used in Church today; and in the property of the very reverend master Arias de Enzinas my eyes have seen the Psalter with three Latin translations of each psalm, and very diverse in their Latin, and one of them much in agreement with the Hebrew.’ (fol. 14r-a)
In thus locating the plurality of translations within the context of
Christian authority and usage, rabbi Mose logically protects himself from
direct complaints about differences between his version and the Vulgate,
which would appear to have been his main concern. The success of his argument
is noted in the epilogue, where Arias Ençinas, writing prior to
the final correction process, concedes that ‘even though in the Romance
the Hebrew might deviate from Sant Jeronimo, it may be acceptable if it
does not offend our Holy Roman faith’ (fol. 24v-b). Faith will decide
the correctness of the translation! This remarkable position would become
of extreme importance in future years (it would soon correspond to the
secular appeals to ‘reason’ such as we will find them in Cartagena in the
next chapter; and of course ‘allein durch den Glauben,’ ‘by faith alone’,
would be Luther’s protestant cry in a century’s time ). Here it is simply
mentioned in passing, as a lower-order variant of ‘default compatibility,’
lying in wait for future regimes.
Having established the principle that there is no one true translation, rabbi Mose Mose then has to find way to justify his mission to produce a translation in one way rather than another. This is not an easy task; plurality itself does not offer firm guidelines. Once again, his strategy is to cite the authority of Hieronymus, from within the Christian tradition:
Sant Jeronimo, showing such great nobility, did three things in his translation: first, in his prologue to the bible he said that he was adventuring a great deal [que se auenturaua a mucho]; second, he left many parts in Hebrew with no Romance and then gave them interpretation with a, b, and c; third, he said in the prologue to his translation that Jewish scholars should be consulted about doubts in the translation, since they were the ones who most deeply knew the language. (fol. 13v-b)
This is all good protective stuff for a Jewish scholar working with
Christians. But could it actually help with translation decisions? The
first idea, that translators should visibly confess their difficulties,
is amply embodied in the rabbi’s long introduction and many glosses. The
second, that the Hebrew terms could be left untranslated and then have
their various possible meanings explained, virtually amounts to the writing
of commentary; it does not always help one decide what target-language
word to put in the translation proper, unless the whole text is to be left
in Hebrew. And the third idea, that Jewish scholars should be consulted,
is little less than claiming ultimate authority for the rabbi himself.
This means he can use a Christian justification for refusing Christian
translations of terms. But now that he has this authority—cleverly won,
to be sure, and thanks to a hermeneutics that here relies on no falsely
neutral position—how does he apply it?
Part of the translator’s ideal solution lies on the border of translation and commentary: rabbi Mose supplies a glossary of several hundred key terms that are thus invested with both Jewish and Christian meanings prior to the translation proper. Three examples (from Lazar 1992: 167):
Adonay: In the Hebrew language it means Señor [Lord], and this name of Señor or Adonay Dios was appropriate because he is the beginning, middle and end of all lordship [señorio] in angels, archangels, and the whole angelic nature and in quintessence and in kings and emperors. (fol. 15v-b)
Egleja [church]. Often it is said in this work: as commands the holy mother eglesia, and these formal words here mean in the way that [quieren aqui tanto dezir commo lo que] the sage scholars of the law commanded, and when those words are said, Jews can and may take it for their own sages, and Christians for theirs, since egleja means: ayuntamiento [assembly, congregation]. (fol. 16v-b)
Martir. Martires are those who die for the law and faith of God at the hands of the unfaithful, and when in a gloss I shall speak of the good and glory of the martires, Christians should take it for those who die for the faith of Jesus Christ, and Jews should take it for those who die for the love of Adonay their God, and in defense of their Mosaic law. (fol. 17v-b)
With numerous terms thus loaded with double meaning, the translator is presumably free to use them without risk of being misunderstood. The glossary becomes a mechanism for setting up two parallel readings for the one text, in keeping with the stated regime for commentaries and, in part, applying Hieronymus’s principle of listing the divergent readings. But the solution is not quite so neat. For example, although Adonay is explained in a way that would enable Christians to understand the term as a name for God, no explanation is forthcoming about the reasons why rabbi Mose uses this term in some places and Señor or Dios in others. And if the terms are really so interchangeable, why insist on Adonay in the first place? Further, although in this particular case we find a Hebrew term being explained in accordance with the advice drawn from Hieronymus, the general strategy is to select a Romance word rather than retain any Hebrew (with exceptions, as we shall soon see), and in cases of doubt to choose the most Latinate Romance (e.g. lux, dux, vespera), which tended to be not only closer to the Vulgate but also in keeping with the directions of fifteenth-century Castilian. So the glossary merely enables Christian readers to get what they are expecting, at least as long as they remain on the most foregrounded levels; its forked tongue, like most of the commentary, is a background labyrinth at best. Of course, the ultimate problem with the double-meaning strategy is that the real translation difficulties tend to be unpredictable; they often concern small terms with relatively incompatible meanings, as we shall soon see.
Translating as Decision-Making
Now armed with double meanings for several hundred terms, the ideal
reader may at last embark on the translation. Let us check a few contentious
passages, places where the Hebrew text potentially conflicts with christological
readings (once again following the transcriptions and explanations given
in Lazar 1992).
Our first example actually shows how the ‘double-meaning’ strategy can creep into the translation proper, albeit not without difficulties:
E la tierra era vana e vazia, e tenebra sobre fazes del abismo, e el spiritu del Señor era rretraydo sobre fazes de las aguas. (Genesis 1:2).
[And the land was barren and empty, and darkness over the face of the abyss, and the spirit of the Lord was drawn over the face of the waters.]
The point of possible contention here is the term spiritu (spirit),
used to translated the Hebrew ru’ah, which can mean either ‘spirit’ or
‘wind’ (Lazar 1992: 175). The Talmudic interpretation as ‘wind’ observes
that the biblical text has named three of the four basic elements (earth,
water and fire ) and so should logically name the fourth (wind). But a
Christian interpretation wants this wind to be ‘spirit,’ so that it might
prefigure the Trinity. Rabbi Mose thus had to choose, and his choice apparently
went the side of the Christians. Elsewhere he shows he is quite aware of
the reading as ‘air’ (in a later commentary he returns to the spiritu del
Señor and adds esto es el ayre, ‘which is air’), but there is no
sign of that interpretation here, neither in the translation nor in its
immediate gloss. So did the Christians win? Perhaps not so clearly. Lazar
notes ‘a long established Jewish tradition culminating in Maimonides’ according
to which there was no real choice to be made between these terms, since
the movement of wind is always ascribed to God (as in Numbers 11:31; Exodus
10:19; 15:10) and it must thus be his spirit. As long as this compatibilist
tradition was available, there was perhaps no profound choice to make and
the rabbi might as well have conformed to the desires of his Christian
correctors (in general accordance with the ‘principle of default compatibility’
Our second example shows a slightly different way of avoiding polemics:
Non se tirara el sçeptro de Juda nin el dux de la su ancha fasta que venga [en Sylo, added in margin] el que tien de seer enbiado... (Genesis 49:10; fol. 57r-a).
(The scepter shall not be taken from Juda nor the staff from his loins until the coming [in Sylo, added in margin] of he who has to be sent...
The first point of interest here is the place-name ‘Sylo’ for the Hebrew
shiloh, interpreted in many Christian versions as the name of the person
to come. Lazar (1992: 196) notes that the place-name was a popular reading
among Jewish commentators in the Middle Ages because it ‘voided the prophecy
from its messianic message [...] and was reduced to a merely historical
context.’ Indeed, in simply leaving out ‘Sylo’ (as rabbi Mose no doubt
originally did) and translating the person as ‘he who has to be sent,’
the rabbi was following the relatively neutral language of the Vulgate
(‘donec veniat qui mittendus est’). Yet this diplomatic translation was
perhaps not quite prophetic enough. In the glosses there are no less than
three further translations of the line, with the person named rather more
clearly in all of them: ‘fasta que venga el Mexias,’ ‘until the coming
of the Messiah.’ Other doubts are also modified: dux, which the rabbi had
taken from the Vulgate, disappears and is replaced with escriuano; and
de la su ancha (from his loins or thighs) becomes dentre sus pies (from
between his legs). Here the strategy would appear to have been to choose
the most neutral and non-polemical reading for the translation proper (in
this case partly supported by the Vulgate) and to leave the more polemical
variants for re-translations given within the glosses.
Our third example is a classic litmus test for any bible translation:
Por tanto dara el Señor el a uos signa; ahe que la alma conçebira a parira fijo, e llamara su nombre Emanuel. (Isaiah 7:14, fol. 267v-a).
(Thus the Lord will give you a sign; the alma will conceive and give birth to a son, and his name will be Emanuel.)
The rabbi knows the Hebrew ‘almah means ‘young girl’ and should be rendered as moça (as do some copies of the Ferrara Bible, cf. Lazar 1992: 193); his Christian correctors know the young girl is supposed to be a virgin and want her rendered as a virgen (as had been the case since the Septuagint). The solution here is to apply a principle gleaned from Hieronymus: to transcribe the Hebrew term—although without shocking people unnecessarily, since Castilian morphology has been adopted, potentially inviting confusion with the alma that means ‘soul’ in Castilian—, and then to explain the two alternative translations (moça and virgen) in the gloss. Yet this solution, despite its elegance, was not easily won. Lazar comments that a word, probably virgen, has been erased from the manuscript and replaced with alma. It was only after a debate of some kind that the ‘loan-word’ solution reached the foreground. But here, for once, the rabbi would seem to have gained neutral ground.
The Rabbi’s Regime
If we now go back to the four principles that rabbi Mose claimed to have used in his writing of the commentaries, we find that a few modifications and extensions are necessary if they are to account for his translating as well:
1. Default compatibility must be assumed and manifested: When there are no major differences between Christian and Jewish interpretations, the translation should be the most readily acceptable to both religions in terms of both word and concept, with priority going to the Latinate Castilian terms most familiar to the Church.
2. Functional ambiguity is legitimate: The one term can be invested with two related meanings, so that its use in the translation may then mean one thing for Christian readers and another for Jewish readers. This is in fact a linguistic application of default compatibility, relying on the fiction of a symmetrical readership.
3. Divergent readings should not be foregrounded: Wherever strongly divergent interpretations exist, they may be noted in the background text (introduction, glossaries, commentaries) but not in the foreground text (the translation proper). In this way the plurality of possible translations is made compatible with the singularity of this particular translation.
4. Background readings may be contradictory: Interpretations listed in the background text may be contradictory, since they are reported discourse and not necessarily the translator’s opinions. Further, the background and foreground interpretations may contradict each other, for the same reason. One of the main purposes of the translation as a whole is to show precisely these contradictions.
5. Christians choose the foreground: Ultimate selection of the variant to be foregrounded is made by the Christian revisers, who may or may not accept the arguments of the translator (since this principle has less priority than the one allowing manifest contradictions).
6. The source text may be foregrounded: When decisions cannot be made as to which alternative is to be foregrounded, it is legitimate to leave the source-text term (e.g. alma) and explain its possible readings in the background text.
As in previous chapters, these principles have been arranged so that the first ones are presumed to be those on which the parties would most agree, then going down to those involving most signs of tension (namely the erasing and addition of terms in the manuscript translation). Could there be a seventh principle? We might posit, for example, that in cases of doubt about language matters the Hebrew scholar is to be believed, since this is indeed one of the principles that rabbi Mose borrows from Hieronymus. Yet when we find a concrete situation, as in the case of seventy or seventy-five, it is clear that the translator’s linguistic knowledge carries relatively little authority in the field of the foreground text, where it must compete with the similarly latent ‘correctness by faith’ principle. The substantiation of that seventh principle would require a slightly more humanist age.
The title of this chapter calls the translation a ‘Christian’s bible.’
This would seem to contradict Lazar’s assessment, noted above, that the
same text is basically ‘a Spanish-Jewish rabbinical Bible draped in some
Christian garments’ (1992: 162). So why should we stress the possessive
adjective rather than the intellectual content?
The main reason, one might suppose, is that the draping garments, the external appearance, the illustrations, and what we have been calling the ‘foreground’ translation, are extremely important for the reception of the text, and thus for its historical function as a translation. The historian or editor may read all the details, but that does not mean that all historical readers found the same details.
A second reason is that, if we try to trace the actual historical readers of this translation, we find remarkably few of them, and virtually all of them belonged to the Christian church or the nobility. Although commissioned by the Grand Master of Calatrava, that knighly sect was incorporated into the crown in 1487 (technically there were no more Moors to fight, so the order had no reason to exist; politically the crown was mopping up resistance). The bible translation was then held by the Holy Office of the Inquisition for some two hundred years, ‘sequestered,’ says Lazar (1992: 157). This is Hispanic history; the Inquisition was not a metaphor. There nevertheless seem to have been a few readers of the translation: a note of 1622 indicates that a Jesuit priest had borrowed it to consult the meaning of obscure words in the Shir ha-Shirim or Song of Songs, with the permission of the Grand Inquisitor. The same note requests that the bible be kept in a safe place ‘because both the friars from San Pablo de Valladolid and those from the San Lorenzo del Escorial monastery wanted to possess it’ (details from Paz y Mélia, cit. Lazar 1992: 158). In 1624 it was donated to Don Gaspar de Guzmán, Count of Olivares, in recognition of the ‘generous amount’ his ancestor had spent commissioning the translation. In the eighteenth century the manuscript was inherited by the House of Alba, in whose possession it now remains. It may have been a Jewish bible in content, but it has long been held in Christian hands, and held in a particularly restrictive way.
This restriction is by no means exceptional. Indeed, it might even be expected, to the extent that any plurality (like that of the public disputation) must eventually challenge accepted hierarchies of authority. In 1233 Jacme I had had a Council in Tarragona order all vernacular bible translations to be handed in to bishops to be burned within eight days (Enciso 1944; Lazar 1994: 357); a similar ban would occur in 1497, and we will see the issue being debated again in the mid sixteenth century. The church did not really want its Vulgate discussed in public. Yet the first prohibition seems not to have applied to Jews, who were apparently free to write and possess biblical texts in as much Romance as they wanted. And they did so: the rabbi Mose’s translation is only one of several such texts from the same and earlier periods. On 25 September 1492, after the expulsion of the Jews, some twenty Jewish bibles in Romance were burned in Seville alone (Lazar 1992: 196-7).
Now, for as long as Christians could not engage in the potential heresies of variant bible translations, powerful Christians could employ Jews to do that work for them, then socially restrict the result as much as possible. Similarly, for as long as Jewish scholars sought intellectual and financial exchange with the powers of Christian Hispania, they could avoid the heresies of illustrated bibles and the like by simply allowing others to do that work, politically restricting their own labor as much as possible. In a sense, this is the profound symbiosis manifested not only in an illustrated Jewish bible but also in Jewish intellectual content being long held, like a prize of battle, in very selected Christian hands. That long historical conquest, that active sequestering, should adequately justify the text being described as a Christian trophy.
A limited edition of 200 copies of the Bibla de Alba was published in Madrid in the early 1920s, transcribed by Antonio Paz y Mélia, who accompanied the edition with disparaging remarks about rabbi Mose’s work (1920-22: xvi), doing what he could to maintain the text within Christian ownership. In 1992 a limited and incredibly expensive facsimile edition of the work was produced by financially astute people who had something to do with a Fundación Amigos de Sefarad; the facsimile was accompanied by the volume of scholarly commentaries on which we have been relying throughout. The text, at least in this luxurious edition, appears to have returned to the Jewish side of the equation, in exchange for money. It might perhaps symbolize the exchanges and cooperation made possible in the multicultural Hispania of the past. But the social restrictions on its status and movement—now financial rather than overtly political—have not all been removed.