I

This paper, in various versions but stubbornly refusing to allow itself to be published, has been around for a long time—for almost as long, in fact, as I have known the distinguished recipient of the present Festschrift. Nonetheless, unlike old wine, it may merit decanting into new bottles. Its starting point is that the utilisation of other languages implies communication, which is something of which David Morgan, in his examination of the polylinguistic cultural world of the Mongols, has always been mindful, observing recently that “[o]ne of the principal difficulties that historians of the Mongol Empire have to deal with is the number of languages in which their sources were written”.Footnote ³ The same problem, both for historians at the present day and for the rulers and diplomats who in past centuries had dealings with them, may be said to have existed to a lesser extent with regard to the Ottomans but, as the late George Steiner once observed, “between verbal languages, however remote in setting and habits of syntax, there is always the possibility of equivalence”. In a form of words which may serve as a Leitmotif for the present study, Steiner rightly points out that his observation is still valid, “even if actual translation can only obtain rough and approximate results”.Footnote ⁴

How rough, and how approximate, these results frequently turned out to be within the context of seventeenth-century Anglo-Ottoman diplomatic relations may become apparent in the following pages. The establishment of diplomatic relations between the Ottoman Empire and the English crown, more than four centuries ago, brought in its train not only diplomatic and economic advantages to one side or the other, but also created for the Tudor monarchy a number of what would nowadays be called communication problems, of a type which was novel to Elizabeth I and her ministers but, to take a post-Pirennian view, nowadays as unfashionable as it is pertinent, one which had long existed across the conflicted Mediterranean frontier between Christendom and Islam.Footnote ⁵

Thus, from the 1580s onwards, the English crown and its ministers entered into a diplomatic and mercantile relationship—‘circles of discourse’, we may term them—with a state, the language of government and diplomacy of which was, for all practical purposes, both unknown and unknowable in Whitehall. At the Porte there was not such a perceptual problem: communication between the ambassadors of Christian states and Ottoman officialdom was secured through the ancient institution of the dragomanate, which comprised the corps of dragomans, or interpreters, the indispensable if to a degree equivocal go-betweens: drawn largely, after the earliest period of Anglo-Ottoman relations, from the ranks of the non-Muslim (zimmî) Greek Orthodox, Latin and Armenian subjects of the Sultan. Thus, in Istanbul or Izmir, in Aleppo or in the port towns of the North African regencies, an ambassador could communicate with a Grand Vizier or his kaymakam, a consul with the local pasha or dey, through the linguistic and intra-personal skills of generations of Timones, Perones, Tarsias, Porfiritas and Ovaneses, while, thanks to both the linguistic abilities of the dragomans and the drafting skills of the Muslim kātib attached to the embassy or consulate, letters received from the English crown and destined for the Sultan or the Grand Vizier, or petitions (‘arzuhal) from the English ambassador to the Divān-i Hümāyūn, could be sent up to the Porte in proper documentary form and in good literary Osmanlıca.

How widespread was the knowledge of literary Ottoman; how wide were the contemporary ‘circles of discourse’, within and outside the living seventeenth-century milieu, the living Ottoman language, and what one may term the traditions of Levantine dragomannic polyglottism?Footnote ⁶ For example, one would greatly have wished to discover more about a Scotsman by the name of Watson (we do not know his first name), whom the English traveller George Wheler encountered in Istanbul in 1675.Footnote ⁷ Watson had lived in Turkey for four or five years and had learned the language. He had obviously learned it to a level beyond that of the colloquial, and was an habitué of the Turkish second-hand book market. To the great surprise of the avid antiquarian Wheler, Watson informed him that there was a “Bazar, or Exchange” in Istanbul where scientific manuscripts in Arabic, Persian and Turkish were dealt in, but that it was “dangerous for Christians” to frequent it; and that the Ottomans “keep annual Registers of all things that pass throughout the whole Extent of their Empire and of the wars they have with their neighbouring Countries”. Watson also informed Wheler that official annalists—“Historians and Writers who have a salary”—were employed in the Sultan's palace to record all important occurrences, and that Turkish-Arabic and Turkish-Persian dictionaries were to be had, together with grammars, books of history and poetry, and works on chiromancy, talismans and science.Footnote ⁸ But who was Watson? We appear not to know more about him than the scanty remarks offered by Wheler.

Thus, to attempt an answer to the question posed above, we are obliged to recreate the ‘circle of discourse’ with relevance to a specific area. This paper will concentrate on a topic accessible in practical terms mainly to English scholarship: the study and use of the Ottoman Turkish language in the context of Turkish documents sent to or from the English crown or relating to the affairs of the English Levant Company, principally in the late seventeenth century.

It has to be said that few of these documents exist in satisfactory editions; few have been published; and many (as it would seem) no longer exist. The exception to this last statement is that a virtually complete series of Ottoman Royal Letters (nâme-i hümâyûn) from this period, addressed to the Crown, are preserved, along in most cases with their contemporary translations, in the Public Record Office, London [now incorporated within The National Archives at Kew].Footnote ⁹ The volumes for the seventeenth century relating to England in the corresponding ‘Register of Imperial Letters’ (Nâme-i hümâyûn defterleri; NHD) series of registers in the Turkish state archives in Istanbul, in which the documents, or drafts of them, would have been entered, appear to no longer exist in the Turkish state archives (Başbakanlık Osmanlı Arşivleri; BOA) in Istanbul.Footnote ¹⁰ Notice should be taken, however, of what appears to be a stray nāme-i hümāyūn defteri from the very end of the seventeenth century, which is preserved in the British Library,Footnote ¹¹ and of two preceding volumes from the same series, which are preserved in the University Library of Göttingen and which contain some relevant documents.Footnote ¹²

Of the Turkish registers kept by the English embassy, in which copies and contemporary translations into Italian of almost all Turkish documents relating to English affairs were entered, only one, from the period of the embassy of Sir William Trumbull (1687-91) has survived to the present day.Footnote ¹³ There is also a surviving register of Turkish documents from Smyrna (Izmir) from the time of William Raye's consulship at the end of the seventeenth and into the early eighteenth century. Nonetheless, the modalities of English translation practice remain largely unstudied, at least for the seventeenth century.Footnote ¹⁴

A distinction may be made at the outset, however, between ‘official’ translations, done into English, in England, of Turkish official communications, and ‘unofficial’, ‘working translations’, done mainly into Italian, the lingua franca of the dragomannic institution, or English, or some other western language, within the territorial limits of the Ottoman Empire—including the “Regencies” of North Africa, where Turkish was employed as a chancery language until the nineteenth century. In the latter case we are also faced with the problem of deciding the language—either the Turkish of an original document or the Italian (or Spanish, or even English) of an original or a ‘working translation’—of the document from which the ‘official’ English translation of a particular Ottoman document was made.Footnote ¹⁵

It goes without saying that English archival practice, both in London and at the embassy in Istanbul, when it came to the orderly preservation of documents in unfamiliar scripts and often of unwieldy dimensions, was not necessarily haphazard or casual, but the almost complete loss by fire, earthquake or neglect of the English embassy and consulate registers of Turkish documents over the course of several centuries has resulted in very considerable lacunae in the surviving holding of the relevant documents, a topic too vast to be entered into here.

Despite these gaps in our sources, an attempt may be made to schematise the stages of documentary transmission, as follows:

1. I. An Ottoman document, written in Turkish, and relating to relations with England, is drafted by clerks in the relevant department of the central chancery.Footnote ¹⁶ A copy of the document is retained in the appropriate register of outgoing documents. (For illustrative purposes the notional document may be assumed to be a nāme-i hümāyūn (‘Royal Letter’) addressed to the English Crown.Footnote ¹⁷

2. II. The document was brought from, e.g., the Ottoman court at Edirne, to the English embassy at Istanbul, either by an embassy dragoman sent to Edirne for that purpose, or sent direct, making use of the Ottoman state courier network (ulaklık).Footnote ¹⁸

3. III. In the embassy chancellery the document was unsealed, read, and a ‘working translation’ into Italian was made by an embassy dragoman. The texts of the document and its translation were copied into the current embassy Turkish register-book.Footnote ¹⁹

4. IV. The document, together with its translation, was sent to England by the most convenient route. In times of war between the Ottomans and Austria, documents could be sent by sea; the land route, however, was still made use of, even in wartime. Footnote ²⁰

5. V. Alternatively, or in addition to the translations mentioned above, on its arrival in England the document might, on the orders of the relevant Secretary of State, be sent for translation. The Secretary would issue instructions to whoever was available: scholars in the University of Oxford; former chaplains to English ambassadors at the Porte, or if, no competent translator could be found in England, particularly in William III's reign, when the king would be on campaign in Flanders, the document might be sent to Holland for translation.

6. VI. Once the ‘official’ translation was received, action could be taken. Replies were drawn up (in English or Latin), and were sent with covering instructions to the ambassador at the Porte, where they would be translated into Turkish and forwarded for presentation to the Sultan or the Grand Vizier to whom they were addressed.

7. VII. The incoming Turkish document and its translation(s) were filed away in the care of the custos rotulorum or Keeper of the Rolls (i.e., of State Papers), to survive, or not, as chance and accident might prescribe.

There is insufficient space to deal with translators and their translations made within Ottoman territory, either at Istanbul or elsewhere.Footnote ²¹ We have to do here with translations undertaken within the ‘active’ living Ottoman milieu, undertaken by translators in Istanbul, principally the so-called ‘English’ dragomanate, men of uncertain intellectual abilities and disparate origins, although universally of ẓhimmī status, who habitually utilised not English but Italian as their medium of translation. This was a world, effectively at some remove from academic scholarship, but one which possessed its own social and cultural imperatives.Footnote ²²

II

More significant, for the purposes of the present essay, were the translators of Turkish documents employed in England; in other words what we may term (loosely) the English Ottomanists of the later seventeenth century. Our predecessors (for we may think of them as such) constituted a body of men who were few in number, but all (or almost all) of them were in holy ordersFootnote ²³ and some — an even smaller number — held university posts. Amongst their number may be mentioned William Seaman, best known as the translator into Turkish of the New Testament, and himself active as a translator of incoming Ottoman documents.Footnote ²⁴ Also deserving of mention is the equally obscure figure of William Hayley, chaplain to Sir William Trumbull during his embassies to the court of Louis XIV (1685-87) and to the Porte (1687-91), and also active as a translator, in this case of letters from the Ottoman court to William III.Footnote ²⁵ As mentioned above, Hayley had been Trumbull's chaplain from the time of the latter's appointment to the Paris embassy in 1685, and it is clear that he must have acquired a sound knowledge of Ottoman Turkish during the four years that he was with Trumbull at the Porte, since it is while Hayley was once more in Oxford after 1691 that he undertook the translation of at least two Ottoman documents addressed to the English crown. In the following summer of 1692 his name crops up in correspondence between Trumbull and Thomas Coke, the English chargé d'affaires at the Porte, regarding the translation of certain letters from the Ottoman Sultan Ahmed II and his Grand Vizier to William III. The occasion for this correspondence was the sudden death at Belgrade, while en route to the Porte, of William Harbord, the equally ill-fated successor to Sir William Hussey as William III's envoy from London to the Ottoman court. The originals of the letters sent from Turkey on this occasion, together with the translations undertaken by Hayley, have survived amongst the Turkish documents in the National Archives; register copies of the Turkish originals can be found in the Ottoman nāme-i hümāyūn defteri, already referred to above, preserved in the British Library.Footnote ²⁶

Neither Seaman nor Hayley's work can be dealt with in detail; instead, the present paper will concentrate on the problems encountered in the translation of Turkish documents undertaken by Thomas Hyde, Bodley's Librarian and professor of Arabic and later of Hebrew in the University of Oxford, and Translator of Oriental Documents to the Crown.Footnote ²⁷

One of the reasons for Hyde's appointment, immediately after the Restoration, to the post of Translator of Oriental Documents to the Crown may be found in his relationship with two of his contemporaries at Queen's College, Oxford: Joseph Williamson, who held office as Under-Secretary of State from 1660 to 1674, and as secretary from thence until 1679, as well as holding the post of Keeper of the State Papers from 1661 to 1701; and Sir Leoline Jenkins, one of the Secretaries of State between 1680 and 1684.Footnote ²⁸ Hyde also appears to have been on good terms with at least one of their successors after the revolution of 1688, in the person of James Vernon, joint or sole Secretary of State for much of the period between 1697 and 1702, although he seems not to have corresponded with Vernon's noble predecessors in that office.Footnote ²⁹ A further reason may be found in an anecdote retailed by Hyde much later, in the reign of William III. According to Hyde, writing in the context of the receipt of an allegedly fraudulent translation of a letter from the then dey of Algiers to the king, written in Turkish, he recalled that “many years previously” an embassy from the sharīf of Morocco had arrived at the court of Charles II with its credentials written in Arabic, accompanied by a “pretended translation[which contained] not so much as 2 lines of the Arabick Letter; but quite contrary, making him [sc. the sharīf] threaten us with war, and to demand Tribute of us, etc., whereas he came with an embassy of peace”. Charles II, accordingly, Hyde continues in a satisfied tone, “finding the usefullness of a faithfull interpreter, would alwayes expressly order his letters to be sent to me, as Sir Lionel [sic] Jenkins told me”.Footnote ³⁰

III

A useful starting point for any examination of Hyde's activities as Interpreter in Oriental Languages is provided by a handlist that he drew up, sometime after 1689, giving details of the translations which he had undertaken in the preceding two decades. This handlist has been published by the Hull-born eighteenth-century scholar Gregory Sharpe in the Prolegomena to his edition of Hyde's miscellaneous works.Footnote ³¹ Hyde did not limit himself to documents in Turkish: the list includes documents in Arabic and Persian, as well as in Turkish, i.e. in all of the elsene-i selāse, the “three [learned] tongues” of Ottoman (and not only Ottoman) literary usage, deriving from Mughal, Safavid, Ottoman, North African and Sharifian chanceries. The first items in the list are “five or six” documents in Persian, written at Surat, and translated into English by Hyde as early as 1660.Footnote ³² From this evidence it is clear that Hyde's activities as interpreter in oriental languages to the Crown must have commenced immediately after the Restoration. Other early translations undertaken by Hyde include Safavid documents dealing with the contentious disputes over customs duties levied on East India Company merchants trading at Hormuz,Footnote ³³ and Sharpe's no. 5, the first of a number of letters from successive deys of Algiers, described by Hyde as being “linguâ Arabicâ Afrorum scriptâ”; in other words, Arabic documents written in the customary Maghribī script.Footnote ³⁴

The first Turkish documents to appear in Hyde's list are a pair, numbered ‘6’ and ‘7’ which are described as, respectively, “a letter from the Prince or dey of Tunis, written in Turkish”, and “a letter in the same language and from a certain personage in the kingdom, whom they call Aga”.Footnote ³⁵ Hyde gives no indication concerning the date of these documents, or the names of the senders, but his description can be fitted to two documents which survive in the Public Record Office. One is a letter from Mehmed, dey of Tunis, to Charles II, dated 21 Jumādā I, 1093 (“12 June 1682”Footnote ³⁶ ); the other, wrtten or dated four days later, is from ‘Alī, ağa of the Janissaries of Tunis, and is also addressed to Charles II.Footnote ³⁷

How and when did Hyde learn Turkish? It was certainly the case that in Restoration England, Turkish, unlike Arabic, lacked the stimulus of official support and academic instruction. The reasons are fairly obvious: a knowledge of Turkish had nothing to offer for the prosecution of Biblical studies, nor did any Turkish sources, insofar as any are known, have anything to contribute towards currently fashionable attempts at the refutation of Islam. As a result, the great interest in seventeenth-century England in the Ottoman state as a political and military entity was not reflected in any corresponding concern for the prosecution of Turkish language studies, let alone the use of Turkish as a language of diplomacy.Footnote ³⁸ The subject produced no Pococke, and Hyde, who succeeded Pococke in the chair of Arabic in 1691, appeared to have been overshadowed by his predecessor's preeminence and longevity.

It is probable, therefore, that it was not until late in the reign of Charles II that Hyde was in a position to undertake the translation of Turkish, as opposed to Arabic and Persian documents, and there is a certain amount of subjective evidence that documents in Turkish caused him more trouble than those in the two other learned languages.Footnote ³⁹ What is certain is that for the remainder of the reign of Charles II, through the short reign of James II and down until the later years of William III, Hyde continued to provide translations of Turkish documents for the Crown. The Glorious Revolution of 1688, or more precisely the accession of William and Mary as joint sovereigns, provoked the issuing of a further number of what Hyde describes in his handlist as “diversas epistolas Arabicè et Turcicè scriptas à Principibus Orientalibus ad regem nostrum Gulielmum IIIm missas”.Footnote ⁴⁰ The extensive diplomatic correspondence which ensued between William III and the Ottoman court, to be discussed below, did not fall to Hyde for translation, but it is possible to identify letters from Tripoli and Algiers, written in 1101/1689 in reply to letters to those Regencies from William III announcing his accession to the English throne.Footnote ⁴¹

IV

The study and knowledge of Ottoman Turkish and the practice of its translation in England was not without its problems. In the case of Thomas Hyde, there is considerable evidence to support the view that the translation of Turkish documents, partly for palaeographic reasons and partly because of the problems which he faced in dealing with Turkish syntax, caused Hyde more difficulties than he was accustomed to encounter in documents written in Arabic or Persian. He occasionally gave vent to his irritation at what he perceived to be sins of omission in this regard by the English consuls who had the responsibility of forwarding letters from the courts of the North African Regencies to the English Crown. Towards the end of his career, in 1697, Hyde had occasion to return to Vernon a letter from Algiers which had obviously caused him a great deal of trouble in the translating. Its subject-matter was the contentious business of a local woman found in flagrante in the house of a luckless Englishman by the name of Butler, but its text failed in part to correspond to the local translation forwarded with it by the English consul at Algiers. Concerning these errors Hyde wrote snappishly to Vernon:

Hyde continues prescriptively in similar vein: “the Consuls abroad should be advised to be more exact, and not to add what they please”.

Not long before this, in 1696, an earlier letter from the dey of Algiers to William III had caused Hyde great difficulty in completing its translation. He had hoped that the locally-produced translation, when it arrived, might assist him in his task, but it failed to do so. Hyde's letter of self-exculpation to Vernon merits quoting in full:

Hyde's problems with Turkish were palaeographic as well as linguistic. The two early (or possibly even the first) documents in Turkish translated by Hyde, the letter dating from 1682 sent by the dey and ağa of the Janissaries of Tunis to Charles II already noticed, had caused him some trouble. Hyde wrote to Leoline Jenkins on 16 September 1682, returning what he describes as “the two letters which by your order I lately received from Dr Wynne”. Hyde continues: “they are in the Turkish language, and what is worse, in the Divan hand which is very hard to read; otherwise I had not kept them so long in my hands”. He adds by way of self-defence, “Of the Dey's letter I have given you a verbatim translation, but the Aga's letter being worse written, there were some few words which I could not read: and therefore I could not venture to give an exact translation of it verbatim; but I have given you the sense and substance of it, which I dare warrant”.Footnote ⁴⁴

Hyde's self-taught skills in Turkish palaeography and diplomatic appear not to have improved with the years. With reference to the letter from the dey of Algiers to William III received in 1697, to which reference has already been made, Hyde further pressed Vernon to instruct the English consul at Algiers that he should ensure the Algerian authorities should “take care that the Turkish letters be written in another plainer hand; their Divan hand, in which this [document] is written, being very troublesome to read”. On the earlier occasion Hyde had ventured a comparison between Turkish documents, and those written in Arabic or Persian. Writing to Owen Wynne, to whom he could perhaps reveal more than to Williamson the problem he faced, he put the matter bluntly:

A year previously Hyde's irritation with the difficulties caused by the dīvānī script pushed him to make a further set of rather lame excuses to Vernon after being defeated temporarily by an earlier letter from Algiers:

In the light of Hyde's frequent complaints over the difficulties involved in translating documents written in Turkish, a number of questions must be asked, although the answering of them must regretfully be remitted in large part to another occasion. In the first place, how competent was Hyde as a translator; secondly, how do his translations compare with those made by other English scholar-translators of the period; and, thirdly, how do translations made in England, into English, compare with translations made locally, either in the North African Regencies or at the Ottoman court or in the English embassy at the Porte or its dependent consulates, in Italian or any other Mediterranean lingua franca? Remaining to be considered is the difficult comparison of translation practice both into and out of Ottoman Turkish, either directly to or from English or via (or into or out of) Italian. How far may ‘dragomanic’ translations of Ottoman documents be compared with those produced by Anglophone scholars, themselves working either with or without experience of the Ottoman linguistic and cultural milieu? Certainly, both the virtues and the defects of Hyde as a translator of Turkish may be seen in some of his extant translations mentioned above. With an attempt to answer these questions we may begin to complete the Anglo-Ottoman ‘circle of discourse’.Footnote ⁴⁷

We should perhaps also begin to view Ottoman documents and the angst and frustration which they produced in their hapless academic translators such as Thomas Hyde, in a completely new light. Mutual comprehension or incomprehension may not have been merely a function of linguistic or cultural divergences, and the circle of discourse may not always have remained unbroken.