A Safavid Text Adjusted for Ottoman Sensibilities: The Istanbul Manuscript of the Tuḥfa-yi Sāmī

Theodore S. Beers

Oct. 2022


This paper is meant to offer a case study of the changes that might creep into an Ottoman-Safavid-Mughal-era Persian manuscript for political or ideological reasons. It will also be an occasion to give a brief overview of the authorship, early reception, and codicology of the work in question: an influential Persian biographical anthology of poets (taẕkira) known as the Tuḥfa-yi Sāmī (ca. 957/1550).1 The author of this taẕkira was Sām Mīrzā (923–75/1517–67), a prince of the Safavid dynasty—so it is only natural to consider potential ideological aspects of both the text and the ways in which it was disseminated and received.

The matter on which we will eventually focus is an early, important copy of the Tuḥfa-yi Sāmī, which was executed in Istanbul in Ramaḍān 972 (April 1565). It is also a complete copy.2 Just as significantly, Sām Mīrzā was still alive at the time of its production. He was living in confinement at the prison fortress of Qahqaha in northwestern Iran, owing to the political anxieties of his older half-brother, Shah Ṭahmāsb (r. 930–84/1524–76); and, as far as can be determined from extant sources, he had stopped working on his taẕkira more than a decade earlier. In any case, it is fascinating to consider the copying in the Ottoman capital of a text written by a contemporaneous member of the Safavid ruling family. The two dynasties had been in and out of conflict for decades (though, in fairness, the mid to late 1560s was a relatively easy time in Ottoman-Safavid relations).3

The significance of the Istanbul manuscript of the Tuḥfa-yi Sāmī, as an early and complete witness of the text, was recognized only recently. When the first few editions of the Tuḥfa were produced, in the 1930s and again in the 1960s, this manuscript was not used. It seems to have been unknown or unavailable to the scholars involved in those projects. Credit for demonstrating the value of the Istanbul copy should go to Aḥmad Mudaqqiq-i Yazdī, who used it as the basis for a new critical edition in 2009. While the ca. 1967 edition of Rukn al-Dīn Humāyūn Farrukh has thus far remained better-known, the Mudaqqiq-i Yazdī edition should become the default option for researchers looking to cite the Tuḥfa.4

The fact that we have two serviceable critical editions of the Tuḥfa, with the most fundamental difference between them being Mudaqqiq-i Yazdī’s reliance on the Istanbul copy, makes it fairly straightforward to look for points of divergence between this manuscript and others from the same general period. We can ask, then, whether the copying in Istanbul of a text that was decidedly Safavid in origin—and which contains pro-Shiʿi statements, effusive praise of the Safavid dynasty, and, on a few occasions, criticism of the Ottomans—required that certain adjustments be made. As we will see, the answer is yes, albeit to a rather minor degree. In fact, by the end of this discussion, it will be fair to ask which is more noteworthy: the modest changes that seem to have been made to the text of the Tuḥfa to avoid offense in the Ottoman context, or the fact that it was acceptable to copy this work at all in 972/1565, while the author—a brother of the Safavid king—was still living, and while many of the political figures mentioned in the Tuḥfa were fresh in people’s minds.

Before addressing these points, however, it will be helpful to cover some preliminary material regarding Sām Mīrzā, his authorship of the Tuḥfa, and the reputation that the book developed in the first few generations of its circulation (including with reference to codicology). It remains the case that fairly little scholarship about Sām Mīrzā’s life and work has been published in European languages.5

Sām Mīrzā Ṣafavī

Sām Mīrzā can be considered a major figure in the early history of the Safavid dynasty (qua independent polity), if only because he was one of the four sons of Shah Ismāʿīl (r. 907–30/1501–24) who survived to maturity. He was born on 21 Shaʿbān 923 (8 September 1517), apparently somewhere in ʿIrāq-i ʿAjam, to one of Ismāʿīl’s harem women.6 This places him third in the birth order of the four aforementioned sons. Before him came Ṭahmāsb (b. 919/1514) and Alqāṣ (b. 922/1516). And the last son, Bahrām, arrived just one week after Sām, on 28 Shaʿbān 923 (15 September 1517). It is worth noting that Sām and Alqāṣ were allegedly borne by the same concubine,7 whereas Ṭahmāsb and Bahrām were from Ismāʿīl’s favorite wife, Tājlū Khānum of the Mawṣillū tribe. After taking the throne, Ṭahmāsb would have a generally warmer relationship with his full brother than with his half-brothers (both of whom he eventually had put to death).

In 927/1521, while little more than a toddler, Sām Mīrzā was sent to Harāt to serve as titular governor. This was in keeping with the custom of princely education and training in the early Safavid period, whereby a young male in the ruling family would be posted to one of the provinces, under the care of a tutor (lala) from the Qizilbāsh military leadership. What became unusual in the case of Sām Mīrzā was that he would remain in Harāt, on and off, for the entirety of his upbringing—and under the influence of one powerful family from the Shāmlū tribe.

Shah Ismāʿīl died and was succeeded by Ṭahmāsb in 930/1524, when the latter was ten years old. The impossibility of Ṭahmāsb’s ruling independently for the next decade led to a violent and unstable political situation among the constituent tribes of the Safavid Qizilbāsh.8 Various factions fought to exert influence over the young king—and even toyed with the idea of engineering his downfall and having him replaced with one of his brothers. In the mean time, the Safavids faced serious external threats on both the western and the eastern frontier. The Abū al-Khayrid Uzbek confederation, under the military leadership of ʿUbayd Allāh Khān (d. 946/1539), attempted continually from the late 1520s through the mid 1530s to take Harāt and its environs from the Safavids.9 They were successful in occupying the city on multiple occasions. (When this took place, Sām Mīrzā and the Shāmlū with whom he was associated were compelled to flee—a point to which we will return shortly.) On the western frontier, there was, of course, the Ottoman empire, which had dealt a demoralizing defeat to the Safavids at Chāldirān in 920/1514. The early to mid 1530s would see another, more extensive incursion by the Ottomans into Safavid territory. This resulted in the loss of Baghdad to the Ottomans, effectively permanently, in 941/1534.10

It was remarkable enough that the teenaged Shah Ṭahmāsb was able to survive and remain on the throne through the first dozen years of his reign. Addressing the twin threats posed by the Ottomans and the Uzbeks required Ṭahmāsb and the Safavid army to move back and forth from one frontier to the other.11 And there were internal challenges to be navigated—including an apparent assassination attempt on Ṭahmāsb in 941/1534 by Ḥusayn Khān Shāmlū, who had served as Sām Mīrzā’s lala in Harāt and had a daughter married to the prince.12 Ṭahmāsb made it through all these difficulties. He avoided a pitched battle with the Ottomans of the kind that had been so damaging to Shah Ismāʿīl, and employed scorched-earth tactics to limit the enemy’s territorial gains. He managed to repel each Uzbek invasion in the northeast, and the Safavids bested the Uzbeks in a major battle at Jām in 935/1528. (This would, in fact, be remembered as the greatest military victory of Ṭahmāsb’s five-decade reign.)13 As of the mid 1530s, Ṭahmāsb had reached adulthood and was at last in a position of security and confidence. Part of taking control over the Safavid polity involved managing his younger brothers.

Sām Mīrzā had grown up under unusual circumstances in Harāt. In its ideal form, the convention of sending young princes to the provinces would have them moved occasionally, and placed in the company of different groups within the military.14 What Sām experienced was a period of more than a decade in which he was, with few exceptions, based in the same part of the Safavid realm, becoming ever more tightly affiliated with the same branch of the Shāmlū tribe. When that faction fell from grace, owing to the alleged assassination attempt by Ḥusayn Khān Shāmlū and certain other suspicious developments, Sām Mīrzā also lost Ṭahmāsb’s confidence. He was placed under a kind of house arrest in the imperial encampment from 943/1537, and he would not be permitted to re-enter public life until 956/1549.15 It is clear from both Safavid and external sources, in fact, that Sām Mīrzā’s period of greatest relevance was his youth, and that, after his first quasi-imprisonment, he would never again be in a position to draw wide attention.16

The end of Sām Mīrzā’s “house arrest” coincided with the resolution, in 956/1549, of the revolt of his brother, Alqāṣ Mīrzā, which had been a thorn in Ṭahmāsb’s side since 953/1546.17 Alqāṣ was finally taken into custody, to be executed several months later in the spring of 957/1550. (Bahrām Mīrzā also died in the fall of 956/1549, under somewhat mysterious circumstances—leaving Ṭahmāsb and Sām as the only two of Shāh Ismāʿīl’s sons who would reach the age of forty.) Sām Mīrzā took this opportunity to petition Ṭahmāsb for a change of his circumstances. He was then authorized to move to the city of Ardabīl, to serve there as local governor (if perhaps only in name) and custodian of the Safavid family shrine. It was after this transition that Sām, who had a lifelong passion for literature and the arts, was able to complete a project on which he had been working intermittently for some time: a biographical anthology of poets (taẕkira), which he titled Tuḥfa-yi Sāmī.18

The Tuḥfa seems to have been finished in 957/1550 or not long thereafter, judging from clues in the text. It should be emphasized that the completion date of the work does not imply much about its composition process. Taẕkira authors often spent decades collecting their material and drafting notices;19 and this could easily have been the case with Sām Mīrzā. As will be explained further below, many of the entries in the Tuḥfa relate to individuals that Sām met in the 1520s and ’30s in Harāt. The move to Ardabīl in 956–7/1549–50 simply placed Sām Mīrzā in a position in which he could gather additional and newer data, produce a coherent text, etc.

Before discussing the Tuḥfa in greater detail, we should briefly cover the remainder of Sām’s life. He managed to hold his position in Ardabīl for about a dozen years, until 969/1562. We do not know a great deal about his activities in this period, except that he tragically lost one of his sons, Rustam Mīrzā, in 961/1554, and that he was later permitted to travel to Mashhad to visit Rustam’s grave. According to the one source that provides an actual narrative of Sām Mīrzā’s later years—the chronicle Khulāṣat al-tavārīkh (999/1591), by Qāżī Aḥmad Qumī—the prince encountered ever-increasing problems with other leading families in Ardabīl.20 His situation deteriorated to the point that he came to be viewed again as a liability by Shah Ṭahmāsb and his inner circle. In the winter of 969/1562, Sām Mīrzā reportedly asked Ṭahmāsb if he could retire to Mashhad and live out his days in prayer. The king instead had him sent to the prison fortress of Qahqaha in the northwest, where a few Safavid princes—including Ṭahmāsb’s son Ismāʿīl Mīrzā, future Shah Ismāʿīl II (984–5/1576–7)—were already confined.

Sām spent about six years at Qahqaha. For co-inmates he had his two sons, of roughly adolescent age; the surviving sons of Alqāṣ Mīrzā; and Ismāʿīl Mīrzā. Communication among these imprisoned princes was both limited and monitored. According to the account given in the Khulāṣat al-tavārīkh, there was an occasion on which Sām Mīrzā sent a qaṣīda to Ismāʿīl Mīrzā, promising to support him in the event that he succeeded Ṭahmāsb. At some point after this, the papers in Ismāʿīl’s quarters were confiscated and sent back to the court for inspection. When Ṭahmāsb learned of the correspondence between Sām Mīrzā and Ismāʿīl Mīrzā, with its potentially conspiratorial undertones, he reacted negatively (as might be expected). In Jumādā II 975 (December 1567), the king had a group of men sent to Qahqaha to execute all of the princes except for Ismāʿīl. Sām Mīrzā, his sons, and the surviving sons of Alqāṣ Mīrzā were duly garotted.

There is the impression, from surviving sources, that Shah Ṭahmāsb and his advisors endeavored to conceal the reality of this mass execution. Qāżī Aḥmad Qumī—writing a few decades later, around the beginning of the reign of Shah ʿAbbās—claims that Ṭahmāsb publicly attributed the deaths of his brother and nephews to an earthquake at Qahqaha. Indeed, the earthquake story is reported as fact in another chronicle of this period, the Takmilat al-akhbār (978/1570) of ʿAbdī Beg Shīrāzī—an author closely affiliated with Ṭahmāsb’s court.21 But we have further hints that Sām Mīrzā was executed. Yet another chronicler, Būdāq Munshī Qazvīnī, mentions the arrest of Sām Mīrzā’s killer in the aftermath of Ṭahmāsb’s death in 984/1576.22 Many of the best-known sources on early Safavid history, such as the Tārīkh-i ʿālam-ārā-yi ʿAbbāsī (1038/1629), are simply silent on the executions at Qahqaha in 975/1567.23 The topic of Shah Ṭahmāsb’s increasingly repressive treatment of other members of the dynasty throughout the second half of his reign is not well covered in the primary sources. One needs to dig a bit to reconstruct the sequence of events. In any case, the execution of Sām Mīrzā along with his sons and nephews represents an important “purge” within the Safavid dynasty. This event deserves greater attention, including in juxtaposition to the contemporaneous challenges in the dynastic politics of the Ottomans and Mughals.

Sām Mīrzā’s life can be divided overall into four periods: first, his upbringing, which took place largely in Khurāsān (everything up to 943/1537); second, the period in which he was under “house arrest” in the itinerant imperial encampment (943–56/1537–49); third, his tenure as titular governor of Ardabīl and custodian of the Safavid shrine (956–69/1549–62); and fourth, his imprisonment at Qahqaha (969–75/1562–7). Sām had an interest in literature and the arts from a young age, and he enjoyed the company of prominent men in these fields during his youth in Harāt.24 It is anyone’s guess when he began to assemble the materials that would eventually become his Tuḥfa. But it was around the time that he was allowed to settle in Ardabīl that he brought the work to completion, adding its latest entries in 957/1550 or shortly thereafter. In the long run, after a turbulent life dictated by political conflict and leading to a violent end, Sām Mīrzā would attain a permanent legacy through the success of his taẕkira. It has been one of the more enduringly popular works in that genre.

The Tuḥfa-yi Sāmī: A basic overview

The Tuḥfa-yi Sāmī is a biographical anthology (taẕkira) with a little over seven hundred notices, organized into seven chapters. (There is also a preface and a short concluding passage.) This anthology has several distinctive features. First, its entries tend to be quite brief. It is often the case that Sām Mīrzā mentions the name of a given individual, devotes a couple of sentences to a biographical description, and then quotes one or two lines of poetry attributed to that person. Seven hundred notices, then, add up to a work of relatively modest scale.

Second, the figures included in the Tuḥfa are either contemporary with Sām Mīrzā, or close to it. The oldest among them belong to the generation of ʿAbd al-Raḥmān Jāmī (d. 898/1492). Most, however, lived in the early to mid tenth/sixteenth century. Sām Mīrzā occasionally mentions an individual’s year of death in his biographical notice. Such cases become useful clues for the dating of the Tuḥfa itself: the latest deaths mentioned took place in 956/1549.25 (There are also two notices in which 957/1550 is referred to as the “current year.”26 It is on this basis that we can say that the Tuḥfa was completed in 957/1550 or not long thereafter.) At any rate, the years cited in the Tuḥfa, like the careers of the individuals profiled, generally fall within Sām Mīrzā’s lifetime. It was a noteworthy decision on Sām’s part to limit his taẕkira to discussion of recent and contemporary figures. Most Persian anthologists did not follow this approach, instead aiming to cover the whole sweep of literary history up to their time. Sām Mīrzā explains his deliberate restriction of scope in the preface of the Tuḥfa, where he notes that earlier taẕkiras—he cites the Bahāristān (892/1487) of Jāmī,27 the Majālis al-nafāʾis (896/1491) of ʿAlī Shīr Navāʾī (originally written in Turkic), and the Taẕkirat al-shuʿarāʾ (892/1487) of Dawlatshāh Samarqandī—have already ensured that the master poets of generations past have a secure legacy. Sām’s stated goal is to do the same for the poets of his era.28

Third, in contrast to the temporal restrictiveness of the Tuḥfa-yi Sāmī, it is unusually open in terms of the kinds of people who are included. We can refer to the Tuḥfa as a biographical anthology of poets, since poets indeed represent the central and largest category of individuals. The literary focus of this taẕkira is also made clear by the fact that every notice, by convention, includes at least a bit of poetry attributed to the person in question. Apart from this, however, the variety of figures discussed in the Tuḥfa is considerable.

Again, the body of the text is organized into seven chapters—and each of these represents a category in a descending social hierarchy. The first chapter focuses on members of ruling houses (including the Safavids) who were known to have some interest in poetry. The second chapter is divided into two sections, with the first devoted to sayyids, and the second to ʿulamāʾ (including several physicians). The third and fourth chapters deal with vuzarāʾ, men of the pen (arbāb-i qalam), and various other high-status individuals.29 It is only in the fifth chapter that Sām Mīrzā turns to poets (shuʿarāʾ) in the professional sense. This structure makes it clear that, from Sām’s perspective, being a poet is generally not comparable to, say, having a notable sayyid lineage. The fifth chapter, like the second (and fourth), is split into two sections: one for well-known poets, the other for more obscure figures. (Given that the typical reader of the Tuḥfa is probably most interested in notices on major poets of the early to mid tenth/sixteenth century, the first section of the fifth chapter can be considered the key passage in the book. Indeed, the first printed edition of the Tuḥfa—which was published in Patna in 1934—omitted the other chapters.)

The social-class hierarchy continues (downward?) in the sixth chapter, with poets of Turkic background. But the seventh chapter is most surprising of all: it describes “other miscellaneous commoners” (sāʾir-i ʿavāmm) and their efforts, often unsuccessful, to compose poetry. While the fifth chapter of the Tuḥfa has the greatest relevance for traditional literary history, it is the seventh that makes this work stand out among taẕkiras of the period.30 Sām Mīrzā shares stories of ordinary people—a wrestler (pahlavān), a butcher, a locksmith, a kalla-pācha vendor—who were participating, in their own ways, in the culture of Persian poetry in cities across the Safavid realm. In several cases he mocks them openly, and it becomes clear that the basis for their inclusion in the Tuḥfa is that Sām Mīrzā finds their attempts at verse amusing or somehow unusual.31

The seventh chapter forces the reader to confront the question of what kind of text, exactly, the Tuḥfa-yi Sāmī is supposed to be. We tend to think of taẕkiras in terms of documenting the careers and reception of select poets, or of tracing an outline of the development of Persian poetry—defining the canon, arguing for the importance of certain stylistic trends, etc. Sām Mīrzā does carry out some of those traditional functions in the Tuḥfa. He provides notices on key poets of his lifetime, such as Bābā Faghānī (d. 925/1519) and Ahlī Shīrāzī (d. 942/1535).32 What is potentially more surprising in this taẕkira is the preponderance of material that seems to have been added for other reasons. Yes, Sām Mīrzā comments on the lives and works of the most important Persian poets of the early to mid tenth/sixteenth century. But that is the focus only in the first part of the fifth chapter. The earlier chapters in the Tuḥfa often read more like a register of dignitaries. The second part of the fifth chapter consists mainly of very short notices on obscure poets—nearly three hundred of them. And the seventh chapter goes in an entirely different direction, cataloguing misfit would-be poets from the urban lower classes. If one reads the whole book, one may be more inclined to view it as an adab anthology, organized as a collection of short biographies—with poetry as a unifying theme, but perhaps only on a formal or conventional level.

The Persian taẕkira would, over the course of the early modern period, come to be a rather diverse genre, in which an author could choose their own approach along numerous axes. If the Tuḥfa had been written in the twelfth/eighteenth century, its structure and contents probably would not stand out as idiosyncratic. But Sām Mīrzā was working in the mid tenth/sixteenth century, and relative to his predecessors—Jāmī, Navāʾī, Dawlatshāh, Fakhrī Haravī (d. after 974/1566)—he produced something strikingly original.33 Nowhere is this clearer than in the seventh chapter.

Ultimately, the Tuḥfa can be different things for different readers. It is, in part, a taẕkira of poets that provides valuable coverage of literary developments in Iran in the early Safavid period. On another wavelength, it offers a variety of amusing anecdotes from Sām Mīrzā’s own experiences and what he had heard from fellow literati. And just as importantly, the Tuḥfa serves as a kind of microcosm of society as it appeared to a member of the Safavid ruling family. Sām Mīrzā has gathered information about the careers and poetry of men34 from all walks of life—kings, paupers, and anything in between—and he has set them in a hierarchy of his own design. This taẕkira is a multifaceted work that can be approached from both literary and (for lack of a better term) sociological perspectives. However one chooses to read the Tuḥfa, it should be acknowledged as one of the most influential taẕkiras of its era. As will be described below, there is no lack of extant manuscripts of the Tuḥfa-yi Sāmī, including from the period immediately following its composition; and the text has been edited for publication (in whole or in part) four times since the early twentieth century. Few works in the genre have received such attention.

Codicology and early reception history of the Tuḥfa

As has been noted above, there is a relative abundance of extant manuscripts of the Tuḥfa-yi Sāmī. This is one of the most securely surviving taẕkiras of its era—which probably should not come as a surprise, given the prominence of the author. To attempt a comprehensive review of the codicology of the Tuḥfa would take us far afield, but it may be helpful to mention a handful of early manuscripts that are now held at libraries across the Near East and Europe, simply to give a sense of the interest in copying this work from the initial decades of its textual life (including manuscripts that date from the final years of Sām Mīrzā’s life, when he was imprisoned at Qahqaha).

There are, for example, three manuscripts of the Tuḥfa at the British Library. One of them (Add. 24,362) is an incomplete copy, but it is dated 969/1561–2, which is exceptionally early. Again, that was the year in which Sām Mīrzā was sent to Qahqaha. Another of the British Library manuscripts (Or. 3490) is dated 976/1569—barely more than a year after Sām was killed. The third copy (Add. 7,670) is undated but appears to be from the same general period. Charles Rieu assigned it to “about the close of the 16th century.”35

There are two manuscripts at the Staatsbibliothek zu Berlin, one of which (Petermann 352) Wilhelm Pertsch judged to be relatively old.36 In Paris, Silvestre de Sacy had access to a copy of the Tuḥfa that had been in the personal collection of Antoine Galland. And de Sacy describes this codex and the text itself at length in one of the volumes of Notices et extraits des manuscrits de la Bibliothèque nationale.37 That copy is dated 1001/1593. Another manuscript with a noteworthy history is Persian MS 317 at the John Rylands Library at the University of Manchester.38 This was the personal copy of Nathaniel Bland, one of the first orientalists to study Persian taẕkiras in a systematic manner.39 The manuscript is reportedly complete, and it is dated 977/1570—only five years later than our Istanbul copy.

We could easily go on in surveying early manuscripts. There are, for example, multiple copies that are still held in former Ottoman libraries in Turkey.40 Apart from what is referred to in this paper as the “Istanbul manuscript”—which is now held at the Iranian National Library in Tehran—one of the older copies in Iran is at the Āstān-i Quds-i Rażavī in Mashhad. This manuscript is missing some leaves and lacks a colophon date, but it, too, is thought to be from the late tenth/sixteenth century.41 If one wished to carry out a critical edition of the Tuḥfa-yi Sāmī, using only manuscripts that were produced up to a few decades after Sām Mīrzā’s death, it would be entirely feasible. This is a work that was copied widely as soon as it became available. Indeed, one of the Safavid chroniclers of this period, the aforementioned Qāżī Aḥmad Qumī, reported that, by the time that he was writing (around 1590 CE), the Tuḥfa had spread to the corners of the world.42 This is hyperbole, but not to the degree that might be imagined.

Published editions of the Tuḥfa

The Tuḥfa-yi Sāmī has been edited for publication on four occasions since the early twentieth century—perhaps more than any other Persian taẕkira. Three of these are complete editions, based on progressively superior sets of manuscripts that were available to the scholars involved. The first was produced by Vaḥīd Dastgirdī in 1936; the second, by Rukn al-Dīn Humāyūn Farrukh in the mid to late 1960s; and the third, by Aḥmad Mudaqqiq-i Yazdī in 2009. Beyond these three full editions of the text, there is a partial edition, covering the fifth chapter of the Tuḥfa—that is, the chapter devoted to “real poets”—which was carried out by Mawlavī Iqbāl Ḥusayn in 1934.43 (He used two early manuscripts which were, and presumably still are, held at the Khudā Bakhsh Oriental Public Library in Patna.)

Among Persianists of our time who have read the Tuḥfa-yi Sāmī—and many have at least glanced at the text—the most commonly used edition, by far, has been that of Humāyūn Farrukh. Few scholars in this field seem to be aware of the existence of the Mudaqqiq-i Yazdī edition, let alone the remarkable manuscript on which it is based. It is worth noting here that Mudaqqiq-i Yazdī’s work is not flawless. He made a few questionable decisions, which would be a topic for another paper.44 There has also been a problem across all editions of the Tuḥfa, including this most recent one, which is that untangling the biography of Sām Mīrzā is difficult. Both Humāyūn Farrukh and Mudaqqiq-i Yazdī were evidently unaware of when and how Sām Mīrzā died. This, of course, influenced their opinions on questions such as the completion date of the Tuḥfa-yi Sāmī. Whatever imperfections there may be in the Mudaqqiq-i Yazdī edition, however, it is clearly the best available at this point.45

The Istanbul manuscript

The manuscript copied in Istanbul in 972/1565 by ʿAhdī b. Shamsī al-Baghdādī is now held in Tehran, at the National Library and Archives of Iran. The title page bears the library stamp of Nāṣir al-Dīn Shāh Qājār, so the codex must have been taken to Iran by that point. Fascinatingly, one of the flyleaves at the end contains a note in Russian—in an elegant Cyrillic cursive—dated 1805. It appears to be a snippet of poetry, on the theme of the arrival of spring, likely translated from Persian into Russian. Two names are written at the bottom of the note: Kolom’ P. Khamdalov (Коломь П. Хамдалов) and Mirza Mahomet Khoja (Мирза Магомет Ходжа).46 (Perhaps the latter is supposed to be the name of the author of the original Persian poem?) In any case, we can say that this manuscript moved around to some extent. It was copied in Istanbul; later entered the custody of a Russian-speaking individual; and finally remained in Qajar Iran.

A digitized copy of the manuscript can be accessed through the web portal of the National Library.47 The images are black-and-white and are provided in a fairly low resolution, but the text is legible. The contents take up 142 folios, with up to seventeen lines of text per page. Again, the Tuḥfa is not a large taẕkira by volume. More than seven hundred entries in fewer than three hundred pages implies short entries indeed—and a few pages at either end of the book are taken up by Sām Mīrzā’s preface and conclusion.

In the images that are available, it can be difficult to identify where one notice ends and the next begins. Even chapter transitions are only faintly recognizable. Red ink appears to have been used for some headings—i.e., they come out lighter than the main text in this black-and-white reproduction—but there is no other clear separation between passages.48 What visual framing exists in the manuscript is for the purpose of setting a divider between the first and second hemistichs of lines of poetry.49 The handwriting looks to be nastaʿlīq of reasonable quality. This does not come across as a finely produced copy of the Tuḥfa; nor does it seem to have been written in haste. Most importantly, there are no missing portions.

ʿAhdī b. Shamsī al-Baghdādī

A fair bit is known about ʿAhdī b. Shamsī al-Baghdādī.50 He was a Turkic speaker from Baghdad. He was also, it should go without saying, fluent in Persian. And he is best remembered as the author of the Gülşen-i şuara, the third work in the traditional sequence of Ottoman Turkish tezkires—after those of Sehi of Edirne (d. 955/1548) and Kastamonulu Laṭīfī (d. 990/1582).51 While the date of ʿAhdī’s birth and the circumstances of his childhood are unclear, we know that he came from a literary family in Baghdad. Shamsī was his father’s pen name (takhalluṣ), with the actual name being Shams al-Dīn. And one of ʿAhdī’s cousins, who went by the sobriquet Rindī, is given a short notice in the Tuḥfa-yi Sāmī.52 There is no indication that ʿAhdī and Sām Mīrzā knew each other, but only one or two degrees of separation stand between them.53

ʿAhdī spent over a decade, including most of the 1550s and the first half of the 1560s, in the central Ottoman lands. He was, for a time, in the entourage of Şehzade Selīm (soon to be Sultan Selīm II), including when the latter was based in Konya.54 ʿAhdī can be viewed as a representative of the part of Baghdādī society that profited from their new integration into the Ottoman world, in the initial decades when it was becoming clear that Arab Iraq would continue to fall within that sphere of power.

The Gülşen-i şuara has usually been dated to 971/1563–4, partly on the basis that the abjad value of the title is supposed to be 971. But some of the events described in the tezkire continue into 972/1564–5, which has led to uncertainty surrounding the proper dating of the work. One point that seems not to have been commented on in scholarship is the possibility that Gülşen-i şuara, with the final hamza of şuara included and counted, would have an abjad value of 972. This might solve the dating problem. At any rate, the authorship of this tezkire is taken to coincide roughly with the point at which ʿAhdī ended his sojourn in Anatolia and returned to Baghdad, where he would remain (for the most part) until his passing in 1002/1593–4. (That is, at least, the most commonly accepted death year for ʿAhdī.) The Gülşen is an important tezkire in its own right, about which a great deal could be written.55 In general, as far as I have understood, this work is distinctive in providing documentation of the careers of Turkic poets from the eastern parts of the (recently expanded) Ottoman empire. ʿAhdī had an uncommon perspective as a Baghdādī visitor to the literary scene in Istanbul and its environs.

It should also be noted that we have other extant manuscripts copied by ʿAhdī. One of those is a manuscript of a work of Persian poetry titled Manẓar al-abrār, which has sometimes been attributed to ʿAhdī but was in fact written by his father, Shamsī. This copy, however, which is held at the Köprülü Library in Istanbul, was executed by ʿAhdī.56 The resulting confusion regarding the authorship of the work is understandable. I have also seen mention of a manuscript of the Qānūn al-adab, a sixth/twelfth century lexicographical work by Ḥubaysh Tiflīsī, with ʿAhdī identified as the copyist.57 That manuscript is now held at the Sipahsālār Library in Tehran.58 One may find it difficult to believe that the Istanbul manuscript of the Tuḥfa was actually copied by the same ʿAhdī b. Shamsī al-Baghdādī who wrote the Gülşen-i şuara. To have a manuscript of one well-known taẕkira in the hand of the author of another well-known tezkire seems too good to be true. So it is helpful to understand that there are other surviving manuscripts executed by ʿAhdī.

Atypical features of the Istanbul manuscript

At last we come to the object itself. The starting point for this paper was that I had noticed a few small, curious discrepancies between the edition of the Tuḥfa-yi Sāmī by Mudaqqiq-i Yazdī, which uses the Istanbul manuscript, and the edition of Humāyūn Farrukh, which is based on manuscripts in Iran, including the one at the Āstān-i Quds. The discrepancies in question occur in the first chapter of the Tuḥfa, which, again, is devoted to members of ruling families who were known to have composed poetry.

For example, in the text as it is given by Humāyūn Farrukh, Sām Mīrzā includes sharply critical remarks about the Ottoman Sultan Selīm I (r. 918–26/1512–20), who had crushed Shah Ismāʿīl at the Battle of Chāldirān in 920/1514.59 In the version of Mudaqqiq-i Yazdī, however, this condemnation is dampened.60 And it turns out that the difference has been caused entirely by the Istanbul copy. This is evident from Mudaqqiq-i Yazdī’s footnotes, which indicate that the text in the Istanbul manuscript—i.e., his base manuscript for the edition—differ in this passage from all of the other manuscripts and editions that he consulted.61 Generally speaking (and with circumspection), we can take the Mudaqqiq-i Yazdī edition as a reflection of the Istanbul copy, and the Humāyūn Farrukh edition as a reflection of the mainstream of the manuscript tradition of the Tuḥfa-yi Sāmī—with Mudaqqiq-i Yazdī’s critical apparatus clarifying the relation between the two. This is the approach that we will take in what follows.

Returning to Sām Mīrzā’s treatment of Selīm, it is perhaps noteworthy that, in ʿAhdī’s manuscript, the heading of this notice refers to Sulṭān Salīm Khān, whereas other copies have simply Sulṭān Salīm. And the same pattern is repeated for the following notice on Sultan Süleymān: again, Khān is added by ʿAhdī.62 This is, it seems, a small mark of respect for the Ottoman rulers. Near the beginning of the entry on Selīm, Sām Mīrzā notes that the sultan was extraordinary for his sharpness of perception (diqqat-i rāy), abundant resoluteness (kas̱rat-i ḥazm), and lack of mercy (qillat-i raḥm) in his thirst for world-conquering. These descriptions are kept in the Istanbul manuscript. Selīm was, after all, known as Yavuz (“the Grim”). But Sām Mīrzā (in most manuscripts) goes on to say that Selīm was famous for his blood-letting (safk-i dimāʾ) and harshness (tund-khūʾī). This is omitted in the Istanbul copy.63

The discussion of Selīm continues with an anecdote about his order to have his brothers killed upon his accession.64 In the Humāyūn Farrukh edition, the text charges Selīm with the deaths of his brothers and his innocent father (pidar-i bī-gunāh), and states that he did not think about the evil consequences (ʿāqibat-i vakhīm) or obeying the devil (muṭāvaʿat-i dīv-i rajīm).65 The Istanbul manuscript abbreviates this considerably, preserving only a simple mention that Selīm had his brothers killed (ḥukm bih qatl-i barādarān numūda)—which was a less controversial issue than the death of his father, Sultan Bayezid.66 The remainder of this entry, which is brief, notes that Selīm conquered much of the Arab world (aks̱ar-i mamālik-i ʿarabistān), and that he died in 926/1520 after a reign of eight years and nine months. Interestingly, however, the Istanbul manuscript then excerpts three lines of Selīm’s poetry, while other copies provide only one line. It may be that ʿAhdī decided to place a bit of extra focus on Selīm’s literary acumen.

We find more of the same phenomenon in the next entry, on Sultan Süleymān. ʿAhdī adds the honorific Khān to the heading. He also opens the notice with the prayer madda ẓilluhu, “may his shadow be extended.” We should keep in mind that this manuscript was copied in 972/1565, when Süleymān was still alive (albeit not for long). And ʿAhdī was indebted to the sultan and his family. The larger differences in content, however, occur toward the end of this notice. Again, the Istanbul manuscript quotes three lines of poetry from Süleymān,67 against the one line usually given in other manuscripts. More significantly, there is a passage missing in the Istanbul copy. Sām Mīrzā quips that Sultan Süleymān, due to his “opium abuse and fancies almost at the limit of insanity” (mudāvamat-i afyūn va khiyālāt-i qarīb bih sarḥadd-i junūn), invaded Iran on a couple of occasions.68 And thanks to God, and to the greatness of the Safavid king, Süleymān was forced to return home empty-handed (bī-sar va sāmān murājaʿat numūda). Sām Mīrzā goes on to praise Shah Ṭahmāsb for his Shiʿi faith that gave him resolution in fighting back the Ottomans.

Perhaps unsurprisingly, ʿAhdī has left out these statements. And he seems to have added his own commentary in its place. Süleymān is referred to as an emperor renowned for justice (bi-ʿadl va dād mashhūr) and filled with good character (bi-ḥusn-i khulq maʿmūr).69 In all versions of the Tuḥfa that I have seen, before quoting from the poetry of Süleymān, Sām Mīrzā acknowledges that he has some literary talent (ṭabʿ-i mawzūnī dārad). In the Istanbul manuscript, however, this is adjusted to state that it is the “august essence” (ẕāt-i humāyūn-ṣifāt) of Süleymān that possesses poetic acumen.

What we find is that Sām Mīrzā’s notices on Selīm and Süleymān, which include sharply negative comments in most manuscripts—and probably in the “original text,” insofar as such a thing can be imagined—have been lightly sanitized by ʿAhdī al-Baghdādī. The entry on Selīm has gone from disparaging to something closer to neutral. And the notice on Süleymān, who was still on the throne, gains a positive valence.

If we look at the overall layout of the first chapter, it also seems that the ordering of notices has been altered in a significant manner. In the Humāyūn Farrukh edition, this chapter opens with a notice on Shah Ismāʿīl, who is followed by two Safavid princes: Bahrām Mīrzā (d. 956/1549) and Sulṭān Muḥammad Mīrzā (later Shah Muḥammad “Khudābanda,” r. 985–95/1578–87). Next comes Sultan Ḥasan (d. 944/1538) of the Kār Kiyā dynasty of Gīlān—which was still, at the time of the composition of the Tuḥfa-yi Sāmī, a semi-independent vassal kingdom to the Safavids. Sām Mīrzā follows this with notices on several of the Timurids, starting with Sulṭān Ḥusayn Bāyqarā (d. 911/1506). Next come three Mughals: Bābur (d. 937/1530), Humāyūn (d. 963/1556), and ʿAskarī Mīrzā (d. 964/1557). Then there is a notice on the Āqquyūnlū Sultan Yaʿqūb (d. 896/1490). Only after this does Sām Mīrzā turn to the Ottomans, Selīm and Süleymān.

In fact, it would be fair to say that the Ottomans are situated notably late in the first chapter. Could this be meant to imply disrespect? The idea never occurred to me until I saw the edition of Mudaqqiq-i Yazdī—and the Istanbul manuscript that underlies it—in which the notices on Selīm and Süleymān occur much earlier. In this version, they come just after Sultan Ḥasan of the Kār Kiyā. After the Ottomans, we find the entries on the Timurids, and the remainder of the chapter continues according to the same program followed in other manuscripts. We should at least consider the possibility that ʿAhdī al-Baghdādī changed the order of notices in this chapter so that it presents the Ottomans more prominently. In that case, it is interesting that he allowed the Safavids, along with their Kār Kiyā vassal, to maintain pride of place. Sām Mīrzā’s perspective is not completely negated—only his decision to set the Ottomans near the end of the list, even after the Āqquyūnlū.

These are the largest potentially ideology-related discrepancies that I have found between the Istanbul manuscript and the “mainstream” of the Tuḥfa-yi Sāmī textual tradition. I have examined several other passages in the text at which it is conceivable that there could be some incentive for ʿAhdī to make changes owing to political or religious sensitivities. For example, as has been mentioned above, the first entry in the Tuḥfa is given to Shah Ismāʿīl—and Sām Mīrzā praises him effusively. This is kept, for all intents and purposes, completely intact in ʿAhdī’s manuscript. One very small difference is that, when Sām Mīrzā is listing Ismāʿīl’s purported sayyid lineage, he reaches the end with Imam Mūsā al-Kāẓim and adds, ʿalayhi al-salām (“may peace be upon him”). This expression of reverence is omitted in the Istanbul copy.70

But there are many other points in the Tuḥfa at which Sām Mīrzā speaks favorably of Shiʿism, without the text having been altered by ʿAhdī. A few poets included in the Tuḥfa are described as specialists in the manqabat genre, and there appear not to be meaningful discrepancies in those notices.71 In fact, one might expect to find more “ideological adjustment” in this manuscript than is actually present. The impression is that ʿAhdī feels compelled to make changes only where Sām Mīrzā has been disparaging or disrespectful toward the Ottomans. Praise of the Safavid dynasty or of the Imāmī cause is evidently not a serious problem, in and of itself. It should be acknowledged that there are likely numerous details in the text of the Tuḥfa that ʿAhdī has added, removed, or modified. As has been noted above, ʿAhdī even inserted a notice on himself. But the main focus in this paper is the alteration of Sām Mīrzā’s work for political or ideological reasons.

Although the changes reviewed here are relatively subtle, there is enough in the organization of the first chapter and the notices on Selīm and Süleymān to suggest that a politically conscious intervention has taken place. In the edition of Mudaqqiq-i Yazdī, these are cases in which the editor notes that his reference manuscript differs from all other textual witnesses to which he has referred. Mudaqqiq-i Yazdī gives no indication, however, that he has investigated the question of why this might be the case.


In the end, the Istanbul manuscript stands as one of the curiosities surrounding the codicology and reception history of the Tuḥfa-yi Sāmī. It is undoubtedly striking that we have a manuscript of the Tuḥfa in the hand of a prominent Ottoman tezkire author. This fact seems not to have been commented on in prior scholarship—not even by Mudaqqiq-i Yazdī, who mentions the copyist’s name but does not delve into his biography.

A final question that ought to be posed, at least rhetorically, is whether we should find it surprising that a book written by a still-living member of the Safavid dynasty was being copied in the Ottoman capital so soon after its appearance. Or is this par for the course? Are there other documented examples of the rapid transmission of texts (especially those surrounded with political sensitivities) from the Safavid to the Ottoman realm in this period? Within the taẕkira genre, there is nothing quite like this case—in large part because it is already so unusual to have such a text authored by a member of a ruling dynasty—but it may be that there are other contexts in which it would be easier to find bases for comparison. I consider this an open question for the time being.


Primary sources

ʿAbdī Beg Shīrāzī. Takmilat al-akhbār. Edited by ʿAbd al-Ḥusayn Navāʾī. Tehran: Nashr-i Nay, 1990.

Būdāq Munshī Qazvīnī. Javāhir al-akhbār: bakhsh-i tārīkh-i Īrān az Qarāqūyūnlū tā sāl-i 984 H.Q. Edited by Muḥsin Bahrām-nizhād. Tehran: Āyina-yi Mīrās̱, 2000.

Fakhrī Haravī, Sulṭān Muḥammad. Taẕkira-yi Rawżat al-salāṭīn, va Javāhir al-ʿajāʾib, maʿa dīvān-i Fakhrī Haravī. Edited by Sayyid Ḥusām al-Dīn Rāshidī. Hyderabad: Sindhī Adabī Būrd, 1968.

Ḥayātī Tabrīzī, Qāsim Beg. A Chronicle of the Early Safavids and the Reign of Shah Ismāʿīl (907–930/1501–1524). Edited by Kioumars Ghereghlou. New Haven, CT: American Oriental Society, 2018.

Iskandar Beg Turkmān Munshī. The History of Shah ‘Abbas the Great. 3 vols. in 2. Edited and translated by Roger M. Savory. Boulder, CO: Westview, 1978–86.

———. Tārīkh-i ʿĀlam-ārā-yi ʿAbbāsī. 2 vols. Edited by Īraj Afshār. Tehran: Amīr Kabīr, 1971.

Membré, Michele. Mission to the Lord Sophy of Persia (1539–1542). Translated by A. H. Morton. London: School of Oriental and African Studies, University of London, 1993.

Navāʾī, ʿAlī Shīr, Fakhrī Haravī, and Ḥakīm Shāh Muḥammad Qazvīnī. Taẕkira-yi Majālis al-nafāʾis. Edited by ʿAlī Aṣghar Ḥikmat. Tehran: Kitāb-furūshī-i Manūchihrī, 1984.

Qāżī Aḥmad Qumī. Khulāṣat al-tavārīkh. 2 vols. Edited by Iḥsān Ishrāqī. Tehran: Dānishgāh-i Tihrān, 1980–.

Sām Mīrzā Ṣafavī. Kitāb-i Tuḥfa-yi Sāmī (ṣaḥīfa-yi panjum). Edited by Mawlavī Iqbāl Ḥusayn. Patna: printed at Shanti Press, Allahabad for Patna University, 1934.

———. Taẕkira-yi Tuḥfa-yi Sāmī. Edited by Aḥmad Mudaqqiq Yazdī. Yazd: Sāmī, 2009.

———. Taẕkira-yi Tuḥfa-yi Sāmī. Edited by Rukn al-Dīn Humāyūn Farrukh. Tehran: ʿIlmī, n.d. (ca. 1967).

———. Tuḥfa-yi Sāmī. Copied by ʿAhdī b. Shamsī al-Baghdādī, 972/1565. MS 896 at the National Library and Archives of Iran, Tehran.

———. Tuḥfa-yi Sāmī. Edited by Vaḥīd Dastgirdī. Tehran: Armaghān, 1936.

Secondary literature

Akün, Ömer Faruk. “Ahdî.” İslâm Ansiklopedisi. https://islamansiklopedisi.org.tr/ahdi.

Beers, Theodore S. “The Biography of Vahshi Bāfqi (d. 991/1583) and the Tazkera Tradition.” Journal of Persianate Studies 8, no. 2 (2015): 195–222.

———. “The Lives of Sām Mīrzā (923–75/1517–67): Dynastic Strife and Literary World-Building in Early Safavid Iran.” PhD diss., University of Chicago, 2020.

Bland, Nathaniel. “On the earliest Persian Biography of Poets, by Muhammad Aúfi, and on some other Works of the class called Tazkirat ul Shuârá.” Journal of the Royal Asiatic Society 9 (1847): 111–76.

Çetin, Muazzez. “16. yüzyıl tezkirelerinden Gülşen-i şu’arâ’da eseri bakımından şâir.” MA thesis, Pamukkale University, 2019.

Çıpa, H. Erdem. The Making of Selim: Succession, Legitimacy, and Memory in the Early Modern Ottoman World. Bloomington: Indiana University Press, 2017.

Dickson, Martin B. “Sháh Ṭahmásb and the Úzbeks: The Duel for Khurásán with ʿUbayd Khán, 930–946/1524–1540.” PhD diss., Princeton University, 1958.

Dirāyatī, Muṣṭafā. Fihrist-vāra-yi dast-nivishta-hā-yi Īrān. 12 vols. Tehran: Kitāb-khāna, Mūza, va Markaz-i Asnād-i Majlis-i Shūrā-yi Islāmī, 2010.

Fleischer, Cornell H. “Alqās Mirzā.” Encyclopædia Iranica Online. https://dx.doi.org/10.1163/2330-4804_eiro_com_5253.

Mitchell, Colin. “Custodial Politics and Princely Governance in Sixteenth-Century Safavid Iran.” In The Safavid World, edited by Rudi Matthee, 79–110. London: Routledge, 2021.

Newman, Andrew J. Safavid Iran: Rebirth of a Persian Empire. London: I.B.Tauris, 2006.

Notices et extraits des manuscrits de la Bibliothèque nationale, vol. 4. Paris: Imprimerie de la République, year 7 of the French Republican calendar (1798–9 CE).

Pertsch, Wilhelm. Die Handschriften-Verzeichnisse der Königlichen Bibliothek zu Berlin, vol. 4, Verzeichniss der persischen Handschriften. Berlin: A. Asher & Co., 1888.

Posch, Walter. Osmanisch-safavidische Beziehungen (1545–1550): der Fall Alḳâs Mîrzâ. 2 vols. Vienna: Verlag der Österreichischen Akademie der Wissenschaften, 2013.

Reinert, Benedikt. “Sām Mīrzā.” Encyclopædia of Islam, Second Edition. https://dx.doi.org/10.1163/1573-3912_islam_COM_0990.

Rieu, Charles. Catalogue of the Persian Manuscripts in the British Museum. 3 vols. London: British Museum, 1879–83.

———. Supplement to the Catalogue of the Persian Manuscripts in the British Museum. London: British Museum, 1895.

Stewart Robinson, J. “Tadhkira 3. In Turkish literature.” Encyclopædia of Islam, Second Edition. https://dx.doi.org/10.1163/1573-3912_islam_COM_1140.

Uzun, Adem. “İslam ve bilim tarihinde sözlükçülük: Hubeyş Tiflîsî’nin Kânûnu’l-edeb adlı Arapça-Farsça sözlüğü.” Erdem 77 (2019): 151–78.