SciELO - Scientific Electronic Library Online

 número27Novas responsabilidades para o IEM e para a MedievalistaO rei que esmorece e a rainha sanhuda: a crise dinástica de 1383-1385 através das emoções nas crónicas de Fernão Lopes índice de autoresíndice de assuntospesquisa de artigos
Home Pagelista alfabética de periódicos  

Serviços Personalizados




Links relacionados

  • Não possue artigos similaresSimilares em SciELO


Medievalista online

versão On-line ISSN 1646-740X

Med_on  no.27 Lisboa jun. 2020 


The Old French Translation of William of Tyre and Templars

Peter Edbury*

*School of History, Archaeology and Religion; Cardiff University. Cardiff, CF10 3AT, England,


Peter Edbury [1]

As is well known, Archbishop William of Tyre’s history relates the story of the First Crusade and the Latin settlements in the Levant as far as 1184. He was at work on his narrative of these events between about 1170 and 1184, and he died shortly before the collapse of the kingdom of Jerusalem in 1187. Historians have long been aware that behind the welter of detail and the measured Latin cadences - for William clearly considered himself something of a Latin stylist - lies much that is tendentious. At first reading he may seem even-handed and judicious, but a careful examination reveals that he had various agendas of which putting a favourable gloss on the history of the Latin East; praising the royal dynasty of Jerusalem, and using his history as a platform for bewailing the inadequate treatment the ecclesiastical province of Tyre had received at the hands of successive popes are among the more notable. He is also famous as an early and strident critic of the Military Orders - the Templars and the Hospitallers - partly no doubt because he was suspicious of their growing military and political power, but more especially because he resented the privileges they had received from the papacy which placed them beyond the jurisdiction of the secular church of which he, as archbishop of Tyre, was a leading member. It is likely, though not provable, that William was involved in attempts to have their privileges reduced at the Third Lateran Council of 1179. The Templars especially are cast in an unfavourable light, but, where other evidence for particular episodes exists, the veracity of William’s hostile stance can often be called in question[2].

William wrote his history to be read by his fellow clergy, and the surviving manuscripts show that in due course copies of his work found their way into the libraries of prominent Benedictine and Cistercian monasteries in England and northern France. At some point in the 1220s someone, probably working in the Paris region, translated William’s history into French. Now the audience was to be the laity - in particular, members of western European noble families who were themselves participating in the crusades to the East. With this new audience in mind, the translator adapted the text, omitting material of purely ecclesiastical interest and from time to time introducing new snippets of information or new slants on the events described[3]. The French translation would appear to have been a huge success. No less than fifty-one complete or substantially complete manuscripts dating from before 1500 survive in public collections in Europe and North America, and most of them have continuations which would have brought the narrative closer to the date at which they were copied[4]. This paper, however, is concerned with the translation, not the additional materials, important and interesting though they are.

We are fortunate that, thanks to the labours of Professor R.B.C. Huygens, we have a superb modern critical edition of William’s Latin text[5]. What we lack is a modern edition of the French translation, and historians have to make use of one or other of the two nineteenth-century editions[6]. When a number of years ago I embarked on a research project to investigate the manuscript tradition of the French text in an attempt to establish a stemma, identify those manuscripts that preserved a text that stood close to the original, and see what else a systematic investigation might reveal about the text and its transmission, I quickly realised that neither of the nineteenth-century editions is satisfactory[7]. The editors chose their manuscripts more or less at random from whatever happened to be to hand in Paris at the time and made no attempt to establish a stemma; in an arbitrary and mechanical fashion they forced the chapter divisions in the French text into conformity with those in the Latin text, and, finding no rubrics in the earlier manuscripts at their disposal - the original translation lacked chapter headings - they supplied them from those to be found in later, mainly fourteenth-century, copies. But what they completely failed to appreciate was that the text itself was not static.

This last point is important and deserves some explanation. What tended to happen in the middle ages was that copyists treated texts in Latin - the language of religion - with considerable respect. Admittedly mistakes were made in their transcriptions, but few scribes would take it upon themselves to emend what was in front of them. With vernacular texts it was quite otherwise. Copyists would, perhaps subliminally, modify the orthography and word order, thereby modernising the spelling and polishing the style. They would also introduce new paragraph breaks and run paragraphs together. They might supply chapter headings where none existed in their exemplar and revise the vocabulary: for example, some of the fifteenth-century manuscripts of the French William of Tyre use the word ‘pape’ for pope where previously we find the word ‘apostoille’. But more importantly, they would change things. Maybe it was a matter of covering up or correcting mistakes; maybe they thought they had additional relevant information; maybe they thought to explain things they found obscure. The point is that vernacular texts evolve. The copyists did not regard them as sacrosanct, to be preserved in their original form, and the French William of Tyre is no exception. What the nineteenth-century editors of the French William of Tyre left for posterity is a pastiche - an amalgam of readings culled uncritically from a variety of manuscripts.

Some account of my approach is called for. Making full use of Professor Jaroslav Folda’s list of all the surviving manuscripts, my first task was to acquire microfilm or microfiche of them all[8]. In the event I managed to obtain forty-nine out of the fifty-one - the two exceptions being a manuscript in the Bibliothèque nationale in Paris (ms. fr. 9086) which could not be microfilmed because the binding is too tight and a Turin manuscript damaged in a fire about a century ago and still not dealt with by their conservators. (Having visited the Bibliothèque nationale and consulted the Paris manuscript, I can understand why they are unable to photograph it; unfortunately, unlike the Turin manuscript, its text is rather important in this context). So, armed with the raw materials for my enquiry, I began by embarking on an analysis of the chapter divisions, taking the first six and last six words of each chapter, and setting them alongside the 1879 printed text, thereby identifying where the chapter divisions diverged and also where this small sample of the wording differed. The thinking behind this stage of my enquiry was rather crude, but, in the event, productive: I reckoned that the manuscripts which preserve a text closest to the original translation would also preserve the chapter divisions as they appear in the Latin text more faithfully. The point is that the date of the manuscript is no guide: a late manuscript could preserve an early version of the text, as indeed is true of the fifteenth-century Paris BN ms. fr. 2627 which presumably was itself copied from a manuscript close to the original. Sure enough, some of the manuscripts had chapter divisions that remained close to the Latin text. Others contain characteristic features of where particular chapters are divided or run together that place them in one of two broad but distinct categories: these can be identified as a western tradition and an eastern tradition. To take the eastern tradition first: Folda and other art historians have identified eight manuscripts as having been copied in the East, most likely in Acre, between the 1260s and 1291. These manuscripts have certain characteristics in common which they share with a small number of later manuscripts, two of which were copied in Italy and all of which would clearly seem to be derived ultimately from manuscripts from the East. Rather more manuscripts share a different set of characteristics that identify them as belonging in a different tradition, and it would seem that they were either the products of workshops operating in Paris in the thirteenth century or were derived from manuscripts originating from those workshops. So - and this is simplifying somewhat - on the basis of my analysis of the chapter divisions, the manuscripts can be sorted into three groups: those which contain an early version of the translation; an eastern tradition that can be linked to Acre; and a western tradition that can be linked to Paris[9].

Even so, within these broad categories there is a bewildering variety of chapter divisions. In trying to construct a stemma it soon became apparent that the divisions did not point to a simple lineal development; instead in both the Paris and the Acre groups it looks as if a strong measure of hybridisation had taken place. Let us suppose that someone wanted their own manuscript of the Old French William of Tyre; if he or she borrowed a copy from a friend and took it down to the workshop and gave instructions for a new copy to be made, what he or she would get would be something very similar to their friend’s, albeit with a few errors and arbitrary changes introduced by the scribe. If then at some later date someone wanted to use this copy for creating yet another one, this further copy would incorporate all the previous changes in that copy and add a few more. It would seem, however, that this is not what tended to happen. It looks instead as if a copyist would start copying one version and then switch to another and then to another again - and so a hybrid version would emerge that defies any attempt to fit it into a stemma. So we need to forget the scenario just described in which a client would commission a copy to made from an exemplar that he or she supplied. Instead we have to assume that the workshops in both Paris and Acre owned several manuscripts of their own and would use them to prepare copies that could then be bought ‘off the shelf’; the problem was they tended to get their own exemplars muddled, and the simplest explanation of how that would have happened is that they worked from unbound signatures - quires usually of eight folios. Copying a large bound manuscript is physically difficult; transcribing unbound signatures would have been much easier. So the workshops had copies ready for sale. So far as Acre was concerned, this made a lot of sense. A wealthy pilgrim from the West would not be in the East long enough to commission a copy to be made, illuminated and bound before returning home - yet those copies that survive do so precisely because they were taken away from the Latin East almost immediately and not destroyed along with so much else when Acre and the other cities fell to the Muslims late in the thirteenth century.

Trying to sort the manuscripts into groups on the basis of characteristic patterns of chapter divisions allowed for various hypotheses to be formulated - some of which I have just outlined for you - but to test these hypotheses what was needed was a detailed examination, line by line, word by word, of the text of some sample chapters making use of every manuscript. The two chapters I chose both relate to the Templars: XII.7 is William’s account of the origin of the Templars; XX.30 tells the story of the Templar murder of an Assassin envoy who was returning from negotiations with King Amaury, an incident which would appear to date to 1174. In each case I took the seven manuscripts which seem to me to preserve texts closest to the original translation and prepared a critical edition of them. They show some variants spelling and some copyists’ blunders, but otherwise there are no significant differences between them, and their homogeneity would seem to be a pointer to their proximity to the original version of the translation[10].

There then followed two separate exercises: an analysis of how the text of the translation as established by this process differs from the Latin, and, secondly, an analysis of the changes introduced into the other manuscripts which give later versions of the text. A few examples will suffice to show changes introduced by the translator. To take XII.7 first: at sentence 3 the translator rightly added that Hugh of Paiens came from Troyes; at sentence 7 he added that the brigands ‘were accustomed to do great evil’; and at sentence 9 he added that knights were among those donating clothing to the Order in the early days of its existence. But the real interest comes in closing sentences (18, 19): only the translator links the Templars’ new found wealth with their desire to free themselves from patriarchal authority; only the translator records that it was at this point the Templars turned to the pope to gain exemption from patriarchal authority, and only the translator who concludes the chapter by noting that the Templars are still even now persisting their aggressive and litigious behaviour. In short the translator heightens William’s criticisms. When we turn to XX.30, we find the same is true. There are various minor points of difference: for example, where in the first sentence William had recorded that the king ‘so it is said’ (ut dicitur) was prepared to refund the Order from his own resources, the translator was categorical: the king would recompense the Order, and as if to underline this assertion he added the money would be paid ‘from such a place whereby they would see themselves well paid’. But in this chapter the big changes again come near the end. It is the translator who introduced the idea in sentence 17 that the master of the Assassins would seek revenge by killing the king. William, in the Latin text, then says that if Amaury had recovered from his final illness he would have taken the matter up with ‘the kings and princes of the lands of the world’, but in the translation the whole of the second part of sentence 18 is new: the king would send messengers ‘to expose the great damage the Templars had done to the Christian faith and especially to the kingdom of Syria: thus it was reckoned that they would be so incensed against them that each would drive them from his dominions’. In short, the translator in both chapters was harsher in his condemnation of the Templars than William.

So what happened later? It is often difficult to be sure whether a subsequent change was deliberate or the result of careless copying. In XII.7 sentence 16 three manuscripts instead of the chevalerie of the Templars speak of the fraternité of the Templars; at sentence 2 four other manuscripts have the Templars renouncing prosperité nor proprieté. Some of the Acre manuscripts emphasise the difference between the Templar knights and sergeants at sentence 15. Several manuscripts clearly date to after the suppression of the Order in 1312, yet the contemporary references in sentences 16, 17 and 19 are retained unaltered. That is odd, especially as one fifteenth-century manuscript uniquely (and correctly) identified the abbot of Clairvaux mentioned in sentence 11 as St Bernard, and another fifteenth-century manuscript rightly supplied the detail that Troyes is in Champagne. Turning now to XX.30, in sentence 12 we are told the Gautier des Mesnil acted ‘with the agreement’ of other Templars, but a group of four manuscripts tells that he did so ‘on the advice’ of other Templars while another five go further and say he acted ‘at the command’ of the others. Two manuscripts, both datable to the 1290s and ascribed to a Parisian workshop, provide a chapter heading: ‘La grant desloiauté que li Templier fisent dont Diex les doit haïr et touz li siecles.’ (‘The great treachery that the Templars did for which God and the whole world should curse them.’) We cannot know how far back this particular rubric was composed, but its presence in two manuscripts copied in Paris about ten years before the arrest of the Templars is suggestive. The point is that between them the manuscripts contain a substantial number of minor variants which may or not be of any significance. But not one of the later manuscripts attempts to tone down the criticisms of the Order to be found in both William and the early versions of the translation. If anything, the criticism is heightened. Of the surviving manuscripts almost a third can be ascribed to northern France and dated to between the 1260s and the arrest of the Templars 1307: the stories they contained would have reflected, and would have helped shape, a climate of opinion in secular society that meant that when in 1307 the king of France set about destroying the Order, the Templars found they had few supporters.


Como citar este artigo

Referência electrónica:

EDBURY, Peter - “The Old French Translation of William of Tyre and Templars”. Medievalista 27 (Janeiro - Junho 2020). [Em linha] [Consultado]. Disponível em .


Data recepção do artigo / Received for publication: 7 de Julho de 2019



[1] A slightly revised version of a paper originally read in 2008. Since then Philip Handyside has considered a number of the themes touched on here in far greater detail. HANDYSIDE, Philip − The Old French William of Tyre. Leiden: Brill, 2015. See especially pp. 102-5.

[2] NICHOLSON, Helen - “Before William of Tyre: European Reports on the Military Orders’ Deeds in the East, 1150-1185”. in NICHOLSON, Helen (ed.) - The Military Orders 2: Welfare and Warfare. Aldershot, U.K.: Ashgate, 1998, pp. 111-18.

[3] HANDYSIDE, Philip − Old French William of Tyre,… pp. 27-107 passim Leiden: Brill, 2015.; PRYOR, John H. - “The Eracles and William of Tyre: an Interim Report”. in KEDAR, Benjamin Z. (ed.) − The Horns of Hattin. Jerusalem: Yad Izhak Ben-Zvi; London: Variorum, 1992, pp. 270-93; HAMILTON, Bernard − ‘The Old French Translation of William of Tyre as an Historical Source”. in EDBURY Peter W.; PHILLIPS, Jonathan (ed.) − The Experience of Crusading, Vol.II, Defining the Crusader Kingdom. Cambridge: Cambridge University Press, 2003, pp. 93-112.

[4] FOLDA, Jaroslav − “Manuscripts of the History of Outremer by William of Tyre: a Handlist”. Scriptorium 27 (1973), pp. 90-95.

[5] HUYGENS, Robert B.C. (ed.) − Willelmi Tyrensis Archiepiscopi Chronicon, 2 vols., CCCM 63. Turnhout: Brepols, 1986.

[6]L’estoire de Eracles empereur et la conqueste de la terre d’Outremer”. Recueil des historiens des Croisades. Historiens occidentaux 1 (1844); WILLIAM, of Tyre Archbishop of Tyre; PARIS Paulin − Guillaume de Tyr ses continuateurs: texte français du XIIIe siècle, 2 vols. Paris: Firmin Didot, 1879-80.

See HANDYSIDE, Philip − Old French William of Tyre,…, pp. 108-113.

[7] For what follows, see EDBURY, Peter W. - “The French Translation of William of Tyre’s Historia: the manuscript tradition”. Crusades 6 (2007), pp. 69-105; EDBURY, Peter W. - “The Old French William of Tyre and the Origins of the Templars”. in HOUSELY, Norman (ed.) − Knighthoods of Christ: Essays on the History of the Crusades and the Knights Templar presented to Malcolm Barber. London: Routledge, 2007, pp. 151-64; EDBURY, Peter W. - “The Old French William of Tyre, the Templars and the Assassin Envoy”. in BORCHARDT, Karl; JASPERT, Nikolas; NICHOLSON, Helen (eds.) − The Hospitallers, the Mediterranean and Europe: Festschrift for Anthony Luttrell. London: Routledge, 2007, pp. 25-37.

[8] Above note 4.

[9] See now HANDYSIDE, Philip − Old French William of Tyre…, pp. 132-6.

[10] EDBURY, Peter W. − “The Old French William of Tyre and the Origins of the Templars”…, pp. 154-6; EDBURY, Peter W. - “The Old French William of Tyre, the Templars and the Assassin Envoy”…, pp. 29-31.

Creative Commons License Todo o conteúdo deste periódico, exceto onde está identificado, está licenciado sob uma Licença Creative Commons