The study of Persian mystical poetry in the post-Mongol period reveals a complex literary landscape shaped by spiritual upheaval, institutional formation, and textual innovation. In this context, Sultan Bahaʾ al-Din Muhammad-i Walad, known as Sultan Walad (623–712 AH/1226–1312 CE), occupies a position of singular importance.Footnote 1 As the eldest son of Mawlana Jalal al-Din Muhammad-i Balkhi, universally known as Rumi, Sultan Walad inherited not merely a biological legacy but a spiritual and literary tradition that he would spend his lifetime preserving, interpreting, and transforming.
For much of the twentieth century, scholarly assessment of Sultan Walad remained overshadowed by his father’s towering presence in world literature. Although his historical role in establishing the Mevlevi order was widely acknowledged, his literary contributions received comparatively little critical attention. Abdolhossein Zarrinkoub, in his influential studies of Mawlana, treated Sultan Walad’s poetry as essentially secondary to his organizational achievements.Footnote 2 Badiozzaman Forouzanfar’s foundational monograph on Mawlana similarly afforded Sultan Walad’s literary output only peripheral consideration.Footnote 3 This tendency found perhaps its most explicit expression in the work of Abdulbaki Golpınarlı, who, although acknowledging Sultan Walad’s fidelity to his father’s teachings, characterized his verse as comparatively pallid.Footnote 4 E. J. W. Gibb’s assessment in his monumental history of Ottoman poetry proved equally dismissive, concluding that Sultan Walad was less a poet than a mystic teacher who merely employed verse as a pedagogical vehicle.Footnote 5 The introductions to modern editions of Sultan Walad’s masnavis (a Persian poetic form composed of rhyming couplets) by Jalaluddin Homaei and Ali Soltani Gerd-Faramarzi perpetuated similar assumptions, treating him primarily as a chronicler of the early Mevlevi community rather than a literary artist in his own right.Footnote 6
This scholarly consensus has begun to shift in recent decades. Lars Johanson’s Reference Johanson1993 study drew attention to Sultan Walad’s remarkable linguistic creativity, particularly his role in fashioning some of the earliest literary verses in Anatolian Turkish.Footnote 7 Franklin Lewis, in a significant reassessment, has argued directly against the assumption that Sultan Walad was merely a versifier, presenting him instead as someone deeply attuned to the musical and literary cultures of Anatolia, a poet, not simply a teacher who happened to write in verse.Footnote 8 Most recently, Michael Pifer has reframed our understanding of Sultan Walad’s relationship to his father’s legacy, demonstrating that the practice of mutābaʿat (emulation) in medieval Anatolian literary culture functioned not as a marker of derivative inadequacy but as a generative means of constructing poetic authority and extending spiritual genealogies. As Pifer observes, pre-Mevlevi sources exhibit no anxiety about the resemblance between father and son; such resemblance was precisely the point, and through emulation Sultan Walad shaped both his own legitimacy and the reception of his predecessors.Footnote 9
Yet as this scholarly reassessment has proceeded on conceptual and theoretical grounds, the manuscript tradition of Sultan Walad’s Divan has remained largely unexamined. We possess sophisticated arguments for why Sultan Walad should be taken seriously as a poet, but we lack direct textual evidence of how he actually practiced his craft: evidence that would ground these arguments in the material processes of composition and revision. This study addresses that gap.
The present study emerged from an attempt to produce a critical edition of Sultan Walad’s Divan that would meet contemporary scholarly standards. This project, involving the examination of manuscripts in libraries across Turkey, Iran, and other repositories, led to an unexpected discovery that fundamentally alters our understanding of the text’s composition and transmission. The identification of two distinct versions of the Divan, not merely variant manuscript traditions but deliberate authorial revisions, opens new perspectives on Sultan Walad’s poetic practice and provides the material evidence that recent conceptual reassessments have lacked.
Central to this discovery is a manuscript that has remained largely unknown to scholarship despite its extraordinary importance. This manuscript, designated as Manuscript A in our study, was scribed by Nizam al-Din Katib, identified as Sultan Walad’s brother-in-law, on 15 Jumādā al-Awwal 694 AH/25 April 1295 CE. What makes this manuscript unique is not merely its early date, written during Sultan Walad’s lifetime, but the extensive evidence it contains of authorial revision. The manuscript bears numerous corrections, deletions, marginal additions, and complete rewritings that appear to represent Sultan Walad’s own process of revising his collected poems.
This article presents a comprehensive analysis of these two versions, demonstrating through detailed textual criticism that both represent authentic stages in Sultan Walad’s poetic development. The revisions, far from being minor adjustments, reveal a poet engaged in systematic improvement of his work, enhancing metrical precision, deepening metaphorical resonance, and refining the musical qualities essential for the poems’ use in samāʿ sessions (the Sufi ceremony of whirling dervishes). This evidence supports and extends recent scholarly reassessments of Sultan Walad, moving from arguments about his potential as a poet to direct observation of his editorial practices.
The implications of this study extend beyond the specific case of Sultan Walad to broader questions in Persian literary studies. How did mystical poets of the classical period approach the revision and compilation of their collected works? What can manuscript evidence tell us about the relationship between oral performance and written text in Sufi contexts? How should modern editors approach texts that exist in multiple authorial versions? These questions, here focused through the lens of Sultan Walad’s Divan, speak to fundamental issues in the study of classical Persian literature.
Sultan Walad: Spiritual Lineage and Literary Practice
Spiritual Formation and Succession
Sultan Bahaʾ al-Din Muhammad-i Walad, eldest son of Mawlana Jalal al-Din Muhammad-i Balkhi, occupied a unique position in the formation of the Mevlevi tradition. Born in Laranda (modern Karaman) on Friday, 25 Rabīʿ al-Ākhir 623 AH/15 May 1226 CE, he received his earliest mystical instruction from Burhan al-Din Muhaqqiq Tirmidhi (d. 638 AH/1241 CE), a close disciple of his grandfather Sultan al-ʿUlamaʾ who arrived in Konya to complete Mawlana’s spiritual education after the latter’s death. For the young Sultan Walad, Burhan al-Din represented a living link to the family’s spiritual heritage.Footnote 10
The arrival of Shams al-Din Tabrizi in 642 AH/1244 CE transformed the household. When Shams disappeared in 643 AH/1245 CE, Mawlana chose Sultan Walad to lead the search party to Damascus, a selection reflecting both his spiritual connection to Shams, whose disciple he had become alongside his father, and his diplomatic capabilities. The month-long return journey, during which Sultan Walad walked while Shams rode, provided intensive spiritual instruction that Shams himself characterized as surpassing what could be attained through a lifetime of worship.Footnote 11 Following Shams’s final disappearance, Sultan Walad accepted the successive spiritual companions his father chose: Salah al-Din Zarkub, whose daughter Fatima Khatun he married, and Husam al-Din Chalabi, the inspiration for and scribe of Mawlana’s Masnavi. After Mawlana’s death in 672 AH/1273 CE, Sultan Walad ceded leadership to Husam al-Din, and following his death, approximately twelve years after Mawlana’s passing, to Karim al-Din ibn Biktumur, described in the Walad-nama as a hidden saint who led the community for seven years. Only after this period did Sultan Walad assume full leadership, spending the final three decades of his life systematizing Mevlevi ritual practice and composing his literary works.Footnote 12
The Divan: Composition and Chronology
Sultan Walad’s literary corpus includes three masnavis, the Walad-nama (also known as Ibtida-nama), Rabab-nama, and Intiha-nama, collectively containing over 20,000 verses, along with the prose Maʿarif. Most significant for this study is his Divan, containing approximately 13,000 verses. In the preface to his Walad-nama, Sultan Walad provides explicit testimony about its composition:
“His Excellency composed divans in various meters and rubais, so I followed his example and created a divan. Finally, friends requested that because I had followed Mawlana (may God sanctify his exalted secret) in making a divan, it was necessary to follow him in writing a masnavi. Therefore, to emulate His Excellency, I began this Masnavi on 1 Rabīʿ al-Awwal 690 AH/30 March 1291 CE.”Footnote 13
This establishes that the Divan was compiled before 690 AH/1291 CE, making it Sultan Walad’s earliest major work. However, the manuscript evidence complicates this chronology: a manuscript dated 694 AH/1295 CE containing extensive revisions indicates that Sultan Walad continued refining his Divan after beginning his masnavis.
Structure and Multilingual Dimensions
The Divan’s organization follows the pattern of Mawlana’s Divan-i Kabir: poems arranged by meter (baḥr), and within each meter alphabetically by the final letter of the rhyme (ravī). This arrangement facilitated locating specific poems for recitation during samāʿ ceremonies. The critical edition on which this study is based, compiled from the most important manuscripts, contains 935 poems in various forms (ghazal, qasida, qiṭʿa [a classical Persian poetic form with rhymed second hemistiches], tarjīʿ [a poem made of stanzas linked by a repeated refrain couplet]) and 458 rubais.Footnote 14
The multilingual aspect deserves particular attention. Although the majority of poems are in Persian, the Divan includes Arabic, Turkish, and Greek verses, reflecting thirteenth-century Konya’s cosmopolitan environment. Sultan Walad composed at least 367 lines completely in Turkish across his works, including approximately seventy-five consecutive lines in the Walad-nama and about 160 lines in the Rabab-nama. This substantial corpus demonstrates sustained engagement with Turkish as a poetic medium rather than occasional experimentation, perhaps, as Lewis speculates, a means of establishing his own identity beyond his father’s primarily Persian oeuvre.Footnote 15
Poetic Philosophy: Craft in Service of Spirit
Sultan Walad’s understanding of poetry differed fundamentally from that of secular court poets. In numerous verses throughout the Divan, he articulates a philosophy of poetry that subordinates aesthetic concerns to spiritual expression:
شعر چه باشد بَرِ من تا که بلافم زان فن / هست مرا فنّ دگر غیر فنون شعرا
What is poetry to me, that I should boast of that craft?/I have another art, different from the poets’ arts.Footnote 16
This “another art” (fann-i digar) represents not a rejection of poetic craft but its transformation in service of mystical expression. Sultan Walad frequently characterizes his poetry as taʿbīr (interpretation) rather than shiʿr (poetry) in the conventional sense. This distinction has important implications for understanding his approach to revision, as documented in the manuscript evidence. In defending the spiritual legitimacy of his father’s poetry and his own, Sultan Walad argues that the poetry of saints functions as tafsīr (exegesis) and sharḥ (commentary) on the Qurʾan, because the saints have effaced themselves and act only through divine inspiration, moving across the page like pens held in God’s hands. By contrast, he dismisses the verse of professional poets as merely taf-i sīr (the scum of garlic), a vivid metaphor that plays on the phonetic similarity to tafsīr while emphasizing the fundamental difference between divinely inspired interpretation and the self-serving display of worldly versifiers.
The tension between spontaneous spiritual expression and crafted literary art appears throughout Sultan Walad’s poetry. Many ghazals contain interjections and direct addresses that suggest composition in states of mystical enthusiasm:
وقت سماع است ای ولد بگذر ز گفت بی عدد / بشنو رباب و دفّ و نی پایی بزن دستی فشان
It is the time of samāʿ, O Walad, pass beyond endless words,/Listen to the rabāb [bowed or plucked lute-like string instrument], daf [large frame drum], and nay [reed flute], stamp your foot, clap your hands.Footnote 17
Yet the manuscript evidence reveals extensive revision of these apparently spontaneous compositions, suggesting a complex relationship between mystical experience and literary expression.
Relationship to Mawlana’s Poetry
Throughout Sultan Walad’s Divan, it is exceptionally rare to encounter a poem that does not feature Mawlana either directly or indirectly. This pervasive presence operates on multiple levels, transforming Sultan Walad’s poetry into continuous dialogue with the master who shaped his vision. He frequently employs dual signatures, concluding poems with his father’s poetic signature “khāmūsh” (silence) alongside his own takhalluṣ (pen name) “Walad”:
وقت سماع است خَمُش
It is time for samāʿ, SilenceFootnote 18
بس کن ازین گفت ولد
Enough of this talk, WaladFootnote 19
خامُش ازین گفتِ زبان
Silence from this spoken word.Footnote 20
This dual signature system serves multiple sophisticated functions: it acknowledges the ultimate source of his poetic inspiration, maintains the connection to his father’s spiritual authority, and suggests that true mystical expression ultimately transcends words, a paradox particularly poignant for a poet who, like his father, produced thousands of verses in the attempt to articulate the ineffable.
The quality and authenticity of Sultan Walad’s imitation has led to a fascinating phenomenon in the manuscript tradition: several of his finest ghazals have been mistakenly incorporated into various recensions of Mawlana’s Divan-i Kabir. This misattribution, far from diminishing Sultan Walad’s achievement, testifies to his remarkable success in capturing the essential qualities of his father’s poetic voice while maintaining his own spiritual authenticity. It suggests that, at his best, Sultan Walad achieved such perfect attunement to his father’s spiritual and poetic wavelength that even experienced scribes and scholars could not distinguish between their works.
A striking example of this phenomenon can be seen in a ghazal by Sultan Walad that has been so consistently misattributed to Mawlana that Annemarie Schimmel named one of her books after its final verse, believing it to be Mawlana’s work.Footnote 21
عشق گوید راست می گویی ولیک از خود مبین / من چو بادم تو چو آتش من تو را انگیختم
Love says: “You speak truly, but do not see from yourself/I am wind, you are fire; I have stirred you.”Footnote 22
With this context established, we turn to the manuscript tradition that preserves Sultan Walad’s Divan and to the extraordinary evidence of authorial revision that has remained unrecognized in previous scholarship.
The Manuscript Tradition: Discovery and Analysis
Previous Editions and Their Limitations
Before examining the manuscript evidence that forms the basis of this study, it is necessary to assess the existing printed editions of Sultan Walad’s Divan. This assessment reveals not only the specific shortcomings of these editions but also the broader challenges facing Persian textual criticism in the twentieth century.
The first printed edition, published by Feridun Nafiz Uzluk in Istanbul in 1941, represented a pioneering effort to make Sultan Walad’s Divan available to modern readers. Uzluk, a Turkish physician and scholar with deep interest in Mevlevi literature, produced an edition that despite good intentions was undermined by careless execution and technical shortcomings. His introduction, written in Turkish, provides valuable historical context but reveals significant limitations in his approach to textual criticism. Uzluk’s limited facility with Persian poetry led to numerous misreadings, and his unfamiliarity with the conventions of classical Persian prosody resulted in defective verses being printed without comment.
The second edition, published in Tehran in 1959 by Asghar Rabbani (writing under the pen name Hamid), essentially reprints Uzluk’s text with a new foreword by Saeed Nafisi. Although Nafisi offers insights into Sultan Walad’s literary significance, the edition itself provides no improvement to the textual basis. Rabbani’s primary contribution was reordering the poems, but this intervention proves particularly problematic given the functional significance of the traditional meter-based arrangement. His reordering, done without reference to manuscript evidence, disrupts the organizational system that facilitated the poems’ use in samāʿ sessions, a practice for which Sultan Walad’s ghazals, with their long radīfs (the repeated word or phrase following the rhyme in each line) and explicit references to ritual recitation, were specifically designed.
Both editions share fundamental methodological flaws: using manuscripts without systematic comparison, lack of critical apparatus documenting variant readings, no discussion of the manuscript tradition or stemmatic relationships, absence of indices and scholarly tools for accessing the text, and failure to recognize the existence of multiple versions of the text, which is the central discovery of this study.
The Discovery of Two Versions
Upon initiating the critical edition of this valuable Divan, two distinct groups of manuscripts emerged with differences too significant to be attributed merely to scribal errors or interpolations. This discovery pointed to the existence of two separate versions of the Divan, a finding that became central to this research.
In Persian codicological terminology, the term “version,” or taḥrīr, carries specific technical meaning. As Najib Mayil Haravi explains in his comprehensive study of manuscript traditions, taḥrīr refers to “different recordings of words or senses, as well as deletions and additions of content.”Footnote 23 This definition encompasses not merely variant readings but systematic patterns of revision that reflect deliberate authorial or editorial intervention.
The phenomenon of multiple versions is not uncommon in Persian literature, in which the manuscript tradition preserves evidence of authors revisiting and revising their works. Some versions were created by authors themselves for improvement, and others arose from political or religious motivations. For instance, Najm al-Din Daya Razi’s Mirsad al-ʿIbad, created in 618 AH/1221 CE, underwent author-driven revisions, including additions and stylistic enhancements, when he sought refuge with ʿAlaʾ al-Din Kayqubad.
Other versions emerged from the efforts of accomplished scribes or manuscript owners, who summarized or altered texts according to their interpretation for personal use. Unlike author-revised versions, these versions lack scholarly value and credibility in textual criticism.
Crucial to unraveling this textual puzzle was one particular manuscript, which will be referred to as Manuscript A in the detailed analysis to follow. This unique manuscript, the oldest chronologically, was scribed on 15 Jumādā al-Awwal 694 AH/25 April 1295 CE by Nizam al-Din Katib, Sultan Walad’s brother-in-law, and proved to be pivotal in our research. All questions regarding the quality and relationship of the two versions were resolved upon accessing this manuscript. Its exceptional value lies in the fact that it contains both versions of the Divan together, with revisions likely made by Sultan Walad himself. This characteristic makes Manuscript A an invaluable resource for understanding the evolution of the text and Sultan Walad’s own editorial process.
In approaching the critical edition of an author’s separate versions, a quantitative and qualitative analysis is essential. When an author has revised part or all of a work, producing a new version with lexical or structural changes, the final version intended for publication should form the main text. The earlier, uncorrected version should be included as marginalia for reference and further study.
It is crucial to note that a mere combination of different versions is neither scholarly nor acceptable. This practice can lead to a conflated text that represents neither the author’s original intent nor any specific historical version of the work. The approach adopted in this research allows for tracing the evolution of the text and understanding Sultan Walad’s motivations for alterations.
The Manuscript Survey: Methodology and Findings
This research involved an extensive examination of numerous manuscripts of Sultan Walad’s works. In total, over thirty manuscripts were studied, including copies of the Divan, Walad’s masnavis, poetic anthologies (jung), and works by Mawlana. From this extensive collection, seven manuscripts were selected as the primary basis for this edition due to their historical significance and textual integrity.
The examination of these manuscripts presented unique challenges that underscore the complexity of this scholarly endeavor. Many manuscripts suffered from physical damage, including missing pages. In several cases, pages had been rearranged and rebound incorrectly, disrupting the original sequence of the text. Some scribes added extensive marginal notes in extremely small and cramped handwriting, making these annotations particularly challenging to decipher. Many manuscripts exhibited inconsistent or unclear placement of diacritical marks, particularly the dots that distinguish between similar letters in the Persian alphabet. This inconsistency often made it difficult to accurately identify specific letters and, consequently, to interpret words correctly.
Particularly challenging were the Turkish poems within the manuscripts, in which scribes from different regions or time periods used various letter forms to represent the same sounds, reflecting dialectal variations or evolutionary changes in the Turkish language. This linguistic complexity necessitated a comprehensive understanding of historical Turkish phonology and orthography.
Despite these challenges, the wealth of material provided invaluable insights into the textual history and evolution of Sultan Walad’s Divan. The process of navigating these difficulties not only enriched the critical edition but also shed light on the transmission and reception of Sultan Walad’s poetry across different regions and periods.
Based on these criteria, seven manuscripts emerged as primary witnesses for establishing the text.
Description of Primary Manuscripts
The seven manuscripts selected as primary witnesses each contribute unique evidence for establishing the text and understanding its transmission history. Their individual characteristics and interrelationships illuminate both the circulation of Sultan Walad’s poetry in early Mevlevi circles and the broader patterns of manuscript production in medieval Anatolia.
Manuscript A
Manuscript A (Konya: Provincial Public Library, Feridun Nafiz Uzluk Collection. Manuscript 6654), dated 15 Jumādā al-Awwal 694 AH/25 April 1295 CE, stands as the most significant witness to Sultan Walad’s Divan. The manuscript’s colophon identifies its scribe as “Nizam al-Mawlawi”:
تمام شد دیوان ولدی الهادی الی الصراط السوی و الموصل الی عنایة الابدی و للسعادة السرمدی بعون الله و حسن
توقیعه در دست فقیر حقیر نظام المولوی ... پانزدهم ماه جمادیا لاول سنه اربع و تسعین و ستمایه
The Divan-i Waladi, the guide to the Straight Path and the connector to Eternal Grace and Everlasting Happiness has been completed, by the help of God and His gracious favor, by the hand of the humble and lowly Nizam al-Mawlawi ... on the 15th of the month of Jumādā al-Awwal, the year 694 AH.
He was among Mawlana’s disciples and married Hadiyya Khatun, daughter of Salah al-Din Zarkub, making him Sultan Walad’s brother-in-law, because Sultan Walad had married Hadiyya’s sister Fatima Khatun.Footnote 24 Nizam al-Din was celebrated as one of the great calligraphers of his age; Aflaki refers to him with such honorifics as “Sultan of scribes and master of masters; a second Ibn al-Bawwab.”Footnote 25
His identity is confirmed through comparison with another manuscript in his hand: a copy of Mawlana’s Masnavi (Istanbul: Murad Molla Library. Manuscript 1311) completed in 695 AH/1296 CE, just one year after Sultan Walad’s Divan. In that manuscript’s colophon, the scribe identifies himself more fully as Nizam al-Mawlawi al-Waladi al-ʿArifi, expanding upon the designation in Manuscript A with additional affiliations. The colophon also records Sultan Walad’s exact birth date, including the day and hour, information found in no other historical source and available only to someone within the immediate family circle.Footnote 26 Paleographic comparison of the two colophons reveals identical handwriting, definitively establishing that the same master calligrapher produced both manuscripts.
This manuscript comprises 714 pages with 19 lines per page in broken taʿlīq script characteristic of the Seljuk period. The codex measures 23.5 × 17 cm and contains extensive corrections, deletions, and marginal additions in what appears to be multiple hands. The phrase wa la-hū maddallāhu ẓillahu wa barakatahu (May God prolong his shadow and blessings) appears in red ink between ghazals, a formula typically used for living persons, confirming its production during Sultan Walad’s lifetime. The nature of these corrections (systematic improvements to meter and meaning, marginal notes containing copying instructions for future scribes, and their consistent adoption in all subsequent manuscripts) points to authorial rather than scribal intervention. Walad ChalabiFootnote 27, a later owner and descendant of Mawlana’s family, examined these features and concluded that the corrections were made by Sultan Walad himself, an assessment that accords with the textual evidence examined in detail below. The manuscript shows evidence of careful production: Turkish and Arabic poems are vocalized, the correct order of misplaced verses is marked with symbols, and prosodic meters are noted in margins when entering new meters.
Manuscript B
Manuscript B (Tehran: National Library of Iran. Manuscript 3060) contains 266 folios with 15 lines per page in old Seljuk Naskh script. The manuscript begins with the Walad-nama, whose colophon reads: “The book of the Masnavi Waladi, the guide to the Straight Path, was completed on the first of Shaʿbān in the year six hundred and ninety[?] by the hand of the humble Sadr al-Waladi.” The scribe, Sadr al-Waladi, was Sultan Walad’s son. The final digit of the date is illegible due to damage or fading; the National Library of Iran records the date as 690 AH/1291 CE, whereas Saeed Nafisi read it as 698 AH/1299 CE. The precise year therefore remains uncertain, although it falls within the range of 690 to 699 AH/1291 to 1300 CE. The Divan follows on renumbered pages, with the heading wa la-hū maddallāhu ẓillahu wa barakatahu (May God prolong his shadow and blessings) in red for most ghazals. According to Nafisi, who used this manuscript for his research, it was formerly in Kabul before its transfer to Tehran. The manuscript follows the first version of the text.
Manuscript C
Manuscript C (Istanbul: Suleymaniye Library, Halet Efendi Collection. Manuscript 328), undated but likely produced between 690 and 694 AH/1291 and 1295 CE, preserves 307 folios with 17 lines per page. The manuscript lacks its beginning and end, with many pages disordered during rebinding. Significantly, it opens with the first lines of the Walad-nama before transitioning to the Divan, suggesting that these works circulated together. Comparison with Manuscript A suggests both were copied by the same scribe, Nizam al-Din, adding to its textual authority. The manuscript contains 783 poems in various forms and 112 rubais, with corrections that may reflect authorial revision, although these are less extensive than those in Manuscript A.
Manuscript D
Manuscript D (Konya: Mevlana Museum Library. Manuscript 2136), scribed by Salah al-Din Katib in Seljuk Naskh, comprises 175 folios with 15 lines per page. The manuscript is arranged alphabetically rather than by meter, containing only 425 poems and 31 rubais. As Golpınarlı notes in the museum catalog, this arrangement and limited content suggest production during Sultan Walad’s lifetime. The manuscript shows significant textual variations, with evidence of at least three different hands contributing to the text.
Manuscript E
Manuscript E (Konya: Mevlana Museum Library. Manuscript 4696) was completed on 22 Dhū al-Qaʿda 712 AH/6 March 1313 CE by Nur al-Bukhari, just four months after Sultan Walad’s death. Written in Seljuk Naskh with 313 folios and 19 lines per page in two columns, it contains 773 poems and 441 rubais. The temporal proximity to Sultan Walad’s death and its following of the second version make it a crucial witness for establishing the final form of the text.
Manuscript F
Manuscript F (Istanbul: Suleymaniye Library, Halet Efendi Collection. Manuscript 135), dated Rabīʿ al-Awwal 717 AH/May 1317 CE, represents a selection from Sultan Walad’s Divan occupying folios 110b–202b of a larger anthology. It contains 252 poems and 47 rubais in 19 lines per page. Despite numerous scribal errors that disrupt meter, its textual affiliations with Manuscript D suggest derivation from a common source.
Manuscript G
Manuscript G (Istanbul: Suleymaniye Library, Halet Efendi Collection. Manuscript 139), scribed by Hasan ibn ʿUthman al-Waladi on 11 Rajab 722 AH/31 July 1322 CE, served as Uzluk’s base text for the 1941 edition. This complete manuscript of 834 pages in two columns and 17 lines per page contains 918 poems and 442 rubais, arranged by meter with careful rubrication. Its provenance through the Mevlevi order, including ownership by Asrar Dede (d. 1211 AH/1796 CE), enhances its textual authority.
Collation of these manuscripts reveals two distinct textual groups. The first version is represented by Manuscripts B, C, D, F, and the base text of A before correction, whereas the second version is found in Manuscripts E, G, and A after correction, containing systematic revisions in vocabulary, imagery, and structure. Manuscript A’s unique preservation of both versions, with visible evidence of the transformation process, provides unprecedented insight into Sultan Walad’s methods of composition and revision.
It should be noted that none of the three principal witnesses to the second version is complete. Manuscript A has damaged and missing pages, as does E; Manuscript G, although the least damaged, still contains unreadable sections. The critical edition of the second version therefore requires all three witnesses working in conjunction, with each manuscript supplying readings where the others are deficient.
The Unique Status of Manuscript A
Manuscript A occupies a unique position as a witness to the transformation from the first to the second version. The manuscript presents a complex stratigraphy of textual layers:
• Base text: The original text as scribed by Nizam al-Din Katib, largely corresponding to the first version
• Corrections: Systematic corrections in black ink, often changing single words or phrases
• Deletions: Entire ghazals crossed out with diagonal lines
• Marginal additions: New ghazals or replacement verses written in margins
• Annotations: Notes about poem arrangement and copying instructions
The physical evidence suggests a sustained campaign of revision rather than sporadic correction. The systematic nature of the changes, their literary quality, and their adoption in subsequent manuscripts all point to authorial revision rather than scribal intervention.
The Two Versions: Analysis of Variations
Categorizing the Variations
The variations between the two versions of Sultan Walad’s Divan can be organized into several distinct categories, each revealing different aspects of the poet’s revisionary practice. This analysis draws on extensive collation of the manuscript witnesses, with Manuscript A providing crucial evidence of the revision process itself.
Substitution of Words for Poetic Improvement
Given the invocations in the ghazals, Sultan Walad composed most of these poems in a state of samāʿ just like his father, such that whatever arose from the heart found immediate expression on paper. However, upon rereading and deeper contemplation, he became aware of various shortcomings and resolved to leave behind a work of greater literary merit. Most corrections applied to the ghazals aim to improve meter and enhance poetic ornamentation. Unnecessary repetitions are eliminated, and greater attention is paid to the semantic power of words in creating more harmonious proportions and meaningful contrasts. The poet demonstrates meticulous care in selecting phrases and idiomatic expressions.
Example 1
Initial version:
هستم من مغز جهان زانم از جمله نهان / درگذر از قشر مهان مست شو از جام صفا
I am the essence of the world, thus hidden from all/Pass beyond the feeble shell, become intoxicated from the cup of purity
Final version:
هستم من مغز جهان زانم ازین جمله نهان / درگذر از قشر مهان مغز شو و نغز و کیا
I am the essence of the world, thus hidden from all these people/Pass beyond feeble shell, become essence, refined and noble.
The substitution of azīn (from this) for az (from) resolves the metrical caesura, creating a smoother rhythmic flow. Moreover, positioning maghz (essence) in direct opposition to qishr (shell) establishes a more powerful conceptual contrast, and the phonetic harmony between maghz and naghz (refined) adds an additional layer of literary beauty through internal rhyme.
Example 2
Initial version:
تو منی من توام دوی خود نیست / در چنین عشق نیک و هم بد نیست
نیک و بد در جهان آب و گِل است / نیک و بد در جهان بی حد نیست
You are me, I am you, there is no duality/In such love there is neither good nor bad
Good and bad exist in the world of water and clay/Good and bad do not exist in the limitless world
Final version:
تو منی من توام دوی خود نیست / در چنین عشق نیک و هم بد نیست
نیک و بد در جهان محدود است / این دو ضد در جهان بی حد نیست
You are me, I am you, there is no duality/In such love there is neither good nor bad
Good and bad exist in the limited world/These two opposites do not exist in the limitless world.
The replacement of āb u gil (water and clay) with maḥdūd (limited) corrects the prosodic irregularity caused by two short syllables where one long syllable is required. Furthermore, the juxtaposition of jahān-i maḥdūd (limited world) with jahān-i bī-ḥad (limitless world) creates a more philosophically resonant contrast. The substitution of īn du żidd (these two opposites) for the repetitive nīk u bad (good and bad) not only improves the poem’s fluency but also evokes the broader concept of duality in the phenomenal world.
Complete Recomposition of Ghazals
In certain instances, the poet begins correction only to conclude that the required changes exceed simple emendations. Recognizing that the poem requires fundamental restructuring, he crosses out the entire ghazal and rewrites it in the margins, often preserving the core theme while transforming its expression.
Example
Initial version (crossed out):
مست شود بی خمری هر که لبانت بمزد / هیچ ورا عیب مکن خدّ تو را گر بگزد
عشق تو آتشکده است تابش در سینه زده ست / گرچه دلم خام بود از لهبش زود پزد
Intoxicated without wine becomes whoever tastes your lips/Do not fault him at all if he bites your cheek
Your love is a fire-temple, its radiance has struck the heart/Though my heart was raw, it quickly cooked from its flame
Final version (written in marginalia):
مست شود بی می و جام آنکه لبانت بمزد / هیچ مکن عیب ورا گر دو رخت را بگزد
بر سر چرخ دل و جان عیش عجب دارد آن / کو چو مه از نور خورت بی دهنی شیر مزد
Intoxicated without wine or cup becomes whoever tastes your lips/Do not fault him at all if he bites your two cheeks
Upon the sky of heart and soul, wondrous pleasure has/He who like the moon, from your sun’s light, drinks milk without a mouth.
While maintaining the opening motif of mystical intoxication, the revision elevates the imagery from conventional fire metaphors to cosmic symbolism. The substitution of earthly passion with celestial mysticism demonstrates Sultan Walad’s evolving poetic sophistication and deepening spiritual perspective.
Revisions of Radīf and Their Significance
The radīf holds unique importance in Persian poetry, serving both as a structural device and semantic anchor. In Sufi samāʿ ceremonies, the radīf assumes even greater significance as participants often chant these radīfs, transforming them into a form of dhikr (remembrance) that intensifies spiritual experience.
The musical quality of Sultan Walad’s poems closely follows that of Mawlana’s Divan-i Kabir, with many ghazals employing identical meters and rhyme schemes. As Mohammad-Reza Shafiei Kadkani observes in Musiqi-yi Shiʿr: “What distinguishes Divan-i Kabir from all other Persian divans is the nature of its radīfs … Mawlana employs radīfs in their longest possible form… . These special radīfs often carry no semantic weight, their repetition creating pure auditory pleasure and enhancing the poem’s lyrical quality. In reading, they can be bypassed semantically: we might say these poems contain no meaning beyond their musical concepts.”Footnote 28
Given his imitation of his father’s ghazals and their performance in samāʿ ceremonies requiring communal repetition, Sultan Walad paid careful attention to these long radīfs, revising them with acute awareness of their effect in group recitation.
Example
Initial version with radīf “man khāna namī-dānam” (I do not know the way home):
ای کرده تو مهمانم در پیش درآ جانم / زان روی که حیرانم من خانه نمی دانم
ای گشته تو را جویا پیر و بچه و برنا / پنهان ز ره بی جا من خانه نمی دانم
O you who have made me your guest, O my soul, come forth/For I am so bewildered that I do not know the way home
O you whom old and young and youth all seek/Hidden from the path of the placeless realm, I do not know the way home
Final version with radīf “man juz tu namī-dānam” (I know nothing but You):
ای کرده تو مهمانم در پیش درآ جانم / زان روی که حیرانم من جز تو نمی دانم
ای گشته تو را جویا پیر و بچه و برنا / پنهان ز ره بی جا من جز تو نمی دانم
O you who have made me your guest, O my soul, come forth/For I am so bewildered that I know nothing but you
O you whom old and young and youth all seek/Hidden from the path of the placeless realm, I know nothing but you.
This revision transforms a radīf expressing spatial disorientation into one declaring exclusive divine knowledge, creating a more powerful vehicle for mystical practice, where each repetition becomes an affirmation of tawhid (divine unity).
Changes in Rhyme
When revising individual words affects their final letters, comprehensive rhyme revision becomes inevitable. What may begin as correction of a single rhyme word necessitates rewriting all rhymes throughout the ghazal. This meticulous attention to rhyme consistency demonstrates that Sultan Walad, inheriting his father’s mastery of poetic music, understood the crucial significance of rhyme position in achieving sophisticated verse and seamless flow.
Example 1
Initial version (rhyme: -āb):
جمع رنودیم همه درده امروز شراب / کاسه بخور کوزه بکش تا همه گردیم خراب
سیمبرا خمر چو زر درده هر شام و سحر / بهر مرا هر نفسی می زن مطرب تو رباب
We rinds [free-spirited figures in Persian mystical poetry who defy conventional piety] have all gathered, give out the wine today!/Drink from the cup, draw from the jug until we all become intoxicated
O silver-bodied one, give out the gold-like wine every eve and dawn,/For my sake, O minstrel, play the rabāb with every breath
Final version (rhyme: -ā):
جمع رنودیم مها درده امروز صلا / کاسه بخور کوزه بکش تا همه گردیم فنا
سیمبرا خمر چو زر ده به همه شام و سحر / بهر من خسته جگر مطرب بنواز نوا
We rinds have gathered, O moon, give out today’s proclamation!/Drink from the cup, draw from the jug until we all become annihilated
O silver-bodied one, give golden wine to all every eve and dawn,/For me with wounded heart, O minstrel, play the melody.
Example 2
Initial version (rhyme: -āst):
توبهِ نو کرد دلم از بد و نیکی که مراست / نعرهِ جانی ز درون بیل ب و کامی برخاست
رسته ز کبریم و ز کین جسته ز کفریم و ز دین / هستی ما نیستی است رفتن ما بی سر و پاست
My heart made new repentance from the bad and good that is mine,/A soul’s cry arose from within, without a lip or palate
Freed from pride and grudge, escaped from blasphemy and faith/Our existence is non-existence, our going is without head or foot
Final version (rhyme: -ā):
توبۀ نو کرد دلم از بد و از نیک شها / نعرۀ جانی ز درون بی دهنی شد پیدا
رسته ز کبریم و ز کین جسته ز کفریم و ز دین / برتر ازینم و ازین نیست شده هستی ما
My heart made new repentance from bad and from good, O King,/A soul’s cry from within became manifest without a mouth
Freed from pride and grudge, escaped from blasphemy and faith/I am beyond this and from this, our existence has become naught.
Example 3
Initial version (rhyme: -ūd):
از سوی روزن شه ما آمد خود را بنمود / گفت بیا گفت بیا سوی من ای بنده تو زود
وصل مرا ترک مکن در چَهِ هجران بمرد / چونکه ز بالا بده ای میل مکن سوی فرود
From the window our King came and showed himself/Said “Come, come to me, O servant, quickly”
Do not abandon union with me, you will die in the well of separation/Because you are originally from above, do not incline toward descent
Final version (rhyme: -ān):
از سوی روزن شه ما آمد و بنمود رخان / گفت بیا گفت بیا سوی من ای بنده روان
ترک مکن وصل مرا از چَه هجران به درا / چونکه ز بالا بده ای خیز درین پست ممان
From the window our King came and showed his face/Said “Come, come to me flowing, O servant”
Do not abandon union with me, emerge from the well of separation/Because you are originally from above, rise up, do not remain in this lowness.
Complete Deletion of Ghazals
Certain ghazals appear crossed out without replacement, including panegyric poems. Manuscripts derived from the corrected version of A show inconsistent treatment of these deletions: some scribes include a limited selection of the crossed-out ghazals, others incorporate all of them, while still others exercise selective judgment. This variation suggests either that not all erasures originated with Sultan Walad or that subsequent scribes questioned whether to honor these deletions.
Examples of non-panegyric ghazals completely crossed out:
Ghazal beginning—
هرچه کند با تو خدا هیچ مگو چون و چرا / زهر ورا قهر ورا می خور همچون حلوا
Whatever God does to you, say not “how” or “why”/His poison, His wrath, drink it like sweetmeat.
Ghazal beginning—
امشب من با بت خود باده خورم تا به سحر / با صنم آمیزم خوش همچو که در شیر شکر
Tonight with my idol I drink wine until dawn/Mingling sweetly with the beloved like sugar in milk.
Ghazal beginning—
بردی دل و جان و دین ای قاضی کاب ما / زین پس نشوی غایب در یقظه و خواب ما
You’ve taken heart and soul and faith, O Judge of our city Kaab/Henceforth you won’t be absent from our waking or our dreams.
Panegyric and elegiac ghazals that were deleted:
Elegy for ʿAlam al-Din Qaysar—
فغان فغان که برفت آنچنان کریم امیر / فغان فغان که شد آن مایۀ غنی و فقیر
Alas, alas that such a generous prince has departed/Alas, alas that the source of rich and poor has gone.
In praise of Majd al-Din ʿAli ibn Muhammad—
منم منم که تو را خاصبنده فرمانم / بقای عمر تو را از خدای خواهانم
It is I, I who am your devoted servant in command/The continuation of your life I seek from God.
In praise of Sultan Ghiyath al-Din Masʿud—
جهان به فرّ شهنشاه نو گرفت نظام / زهی طراوت ملک و سعادت اسلام
The world took order through the glory of the new King/What freshness of dominion and felicity of Islam!
In praise of the city of Aksaray—
ای مردم آقسرا بدانید / هر یک ز شما دو صد جهانید
O people of Aksaray, know/that each one of you is two hundred worlds.
Patterns in the Revision Process
Analysis of the complete corpus of revisions reveals several consistent patterns illuminating Sultan Walad’s evolving poetic consciousness. Most prominently, initial spontaneous expressions frequently yield to more carefully formulated theological statements about divine nature and mystical experience. This movement toward greater doctrinal precision suggests Sultan Walad’s increasing awareness of his dual role as both ecstatic poet and responsible spiritual teacher, charged with transmitting accurate mystical doctrine through verse.
The revisions consistently enhance the poems’ musical qualities (meter), improving their suitability for samāʿ performance through smoother prosody, more resonant radīfs, and strategic internal rhymes that create additional rhythmic layers. This attention to acoustic properties reflects the practical function of these ghazals in Mevlevi ritual, in which poetic rhythm guides both physical movement and spiritual states. The transformation from merely competent verse to musically sophisticated poetry demonstrates Sultan Walad’s growing understanding of how sound and meaning interweave in mystical expression.
A marked philosophical deepening characterizes many revisions, with concrete imagery frequently evolving into abstract theological discourse. Simple devotional expressions transform into sophisticated mystical philosophy, suggesting Sultan Walad’s own spiritual maturation between the composition of the two versions. This pattern parallels the theological development visible in his masnavis, where later works display increasing doctrinal complexity and philosophical subtlety.
Many revisions bring poems into closer alignment with Mawlana’s distinctive poetic practice, particularly in the deployment of paradox, cosmic imagery, and ecstatic language. This intertextual refinement reveals Sultan Walad’s conscious modeling of his father’s style while maintaining his individual voice, achieving that delicate balance between faithful imitation and creative innovation that characterizes his entire literary output.
Sultan Walad’s practice of systematically revising his poetry was not an innovation but an inheritance from his father. Scholarship on the manuscript tradition of Mawlana’s Masnavi has demonstrated that Mawlana himself engaged in extensive revision of his compositions. According to Aflaki’s account in Manaqib al-ʿArifin, after Husam al-Din Chalabi transcribed the Masnavi, he would read the entire text aloud to Mawlana, who “elucidated the obscurities of the text and, in all likelihood, introduced corrections and alterations.” Aflaki reports that Husam al-Din “read the entirety of the volumes of the Masnavi aloud to His Holiness the Lord seven times, performing spiritual audition while he listened, becoming acquainted with the mysteries and hidden treasures of its secrets. He explicated every niche of the Masnavi, letter by letter and word by word, recording them with golden clarity and adding vocalization marks.”Footnote 29
The colophon of the famous Konya manuscript of the Masnavi dated 677 AH/1278 CE confirms that Mawlana introduced revisions into his early compositions. Copies were made from these revised exemplars, creating what Nicholson, in his critical edition, identified as two distinct groups of manuscripts; one metrically and stylistically more refined than the other, representing the final form of the Masnavi.Footnote 30 Sultan Walad himself participated in this process: variant readings recorded in his hand appear in manuscripts of the sixth book of the Masnavi.Footnote 31 Therefore when Sultan Walad subjected his own Divan to systematic revision, he was following a practice he had witnessed firsthand in his father’s workshop; a practice in which he had directly participated as a corrector of the Masnavi itself.
This familial tradition of revision takes on particular significance in light of Aflaki’s account of scribal interference during Mawlana’s lifetime. Shaykh Fakhr al-Din Sivasi, an early companion entrusted with transcribing Mawlana’s poetry, would “at times interfere with the words of the Lord, correcting with his pen and distorting words without permission,” behavior for which Mawlana severely rebuked him.Footnote 32 Such unauthorized alterations made the preservation of authentic, author-supervised texts a matter of urgent concern within the Mevlevi community. It is in the light of this background of vigilance against textual corruption that we should understand Manuscript A of Sultan Walad’s Divan: scribed by his brother-in-law Nizam al-Din, produced under close supervision, and bearing evidence of the author’s own revisions.
The Question of Authenticity
The attribution of both versions to Sultan Walad himself rests on the convergence of textual, codicological, and historical evidence. Although no single piece of evidence is decisive in isolation, the cumulative weight establishes the authorial nature of these revisions with a high degree of probability.
The most compelling evidence lies in the character of the annotations themselves. First, the revisions consistently improve the poems: correcting metrical irregularities, enhancing semantic precision, and refining the radīfs for more effective performance in samāʿ ceremonies. Such changes require both technical mastery of Persian prosody and intimate familiarity with the poems’ liturgical function: expertise that a later scribe or owner would be unlikely to possess. Second, the margins contain copying instructions directing future scribes on proper arrangement of the text. For example, one marginal note in the rubais section reads:
پنج رباعی از حرف شین زیر حرف ط و ظ نوشته شده است. در حالت نبشتن احتیاط کنند.
Five rubais from the letter shīn have been written under the letters ṭā and ẓā. When copying, be careful [to place them correctly].
A scribe copying for personal use or a later reader annotating the manuscript would have no occasion to issue such directives; only someone supervising the text’s production and transmission would do so. Third, the systematic scope of the revisions (encompassing vocabulary, rhyme, radīf, and complete recomposition of entire ghazals) indicates a sustained campaign of revision rather than sporadic correction.
Walad Chalabi, one of Manuscript A’s later owners, examined the manuscript and reached a conclusion consistent with this evidence. Writing in blue ink on the manuscript’s first page, he observed:
درین دیوان شریف همۇ سرلوحۇ غزلیات )وله مدّ الله ظلّه و برکته) نوشته شده است. دلالت می کند که این دیوان شریف در
عهد شریف حضرت سلطان ولد نوشته شده است. نموده میآ ید که این دیوان مسوده شده است و به قلم حضرت سلطان ولد
تصحیح شده و بعضی را بر سر خط ابطال کشیده است. ولد چلبی.
In this noble Divan, all the headings of the ghazals are written as “wa la-hū maddallāhu ẓillahu wa barakatahu” (May God prolong his shadow and blessings). This indicates that this noble Divan was written during the blessed era of Sultan Walad. It appears that this Divan was drafted [musawwada] and corrected by the blessed hand of Sultan Walad, and some parts were crossed out. Walad Chalabi.
Walad Chalabi’s assessment, made some six centuries after Sultan Walad’s death, cannot by itself establish authorship. His hope that the manuscript preserves “the blessed hand” of Sultan Walad reflects the reverence with which the Mevlevi community regarded their founder’s texts. Yet his observations (that the corrections are contemporaneous with the manuscript’s production, that they exhibit a consistent hand resembling the marginal additions, and that the manuscript functioned as a musawwada from which other copies derived) accord precisely with what the textual and codicological evidence independently suggests. The value of his testimony lies not in establishing what we could not otherwise know, but in confirming that a knowledgeable member of the Mevlevi tradition, examining the physical manuscript, drew the same conclusions that the textual evidence supports.
The manuscript’s margins contain several different hands, all written in the Seljuk style (except for Walad Chalabi’s later annotations), indicating they date from the time of the manuscript’s creation or shortly thereafter. Although we possess a copy of the sixth book of Mawlana’s Masnavi as an exemplar of Sultan Walad’s handwriting, definitive paleographic attribution remains challenging, as one sample represents formal copying while the marginal annotations appear more hastily written. Even Walad Chalabi himself moderates his initial certainty in the rubais section, noting:
این خط شریف به علاوه های حواشی می ماند. ممکن است که این نسخه به نسخه های سائره مسوده شده باشد. این سطر
شریف خط مبارک جناب سلطان ولد بودن امید است. الله اعلم.
This blessed script resembles the additions in the margins. It is possible that this manuscript served as the draft [musawwada] from which other manuscripts were copied. This blessed line is hoped to be the handwriting of the blessed Sultan Walad. God knows best.
The manuscript’s provenance provides compelling evidence for authenticity. Ownership notes on Manuscript A’s opening pages record the names of successive owners, and Feridun Nafiz Uzluk, in his introduction to the 1941 edition, documents the chain of transmission:
For a period of time, it was in Konya under the ownership of ʿAbd al-Qadir Chalabi, who was the sar-ṭarīq (head of the Sufi order). It then passed to Walad Chalabi Efendi, and he graciously bestowed it upon this humble servant. I, too, will send it to the Mevlana Museum.
Uzluk himself was a descendant of Mawlana’s family, meaning that from its creation by Nizam al-Din Katib, Sultan Walad’s brother-in-law, to its eventual deposit in the Mevlana Museum, this manuscript never left the custody of the Mevlevi community.Footnote 33 This unbroken chain of familial transmission is significant, as the Mevlevi tradition regarded their founders’ works as sacred texts, preserving them without unauthorized alterations. The careful vocalization throughout the manuscript further attests to this reverence, as the Mawlana family strove to preserve the precise pronunciation of poems as their great master had recited them.
As the oldest extant copy of the Divan, written during Sultan Walad’s lifetime, Manuscript A reveals a complex revision history. Even before the visible corrections and deletions, the base text scribed by Nizam al-Din does not completely represent Sultan Walad’s first draft. This suggests that Sultan Walad revised his Divan multiple times, both before and after 694 AH/1295 CE when Nizam al-Din completed his copy.
The claim that Sultan Walad revised his Divan prior to Nizam al-Din’s transcription rests on a specific pattern in the manuscript evidence. If the base text of Manuscript A (before the visible corrections) represented Sultan Walad’s original draft, we would expect the other manuscripts of the first version (B, C, D, F) to match this base text closely, with variations attributable to normal scribal transmission. This is not what we find. Manuscripts B, C, D, and F, although clearly belonging to the first version and sharing its general textual character, contain readings that diverge from Manuscript A’s uncorrected base text in ways that cannot be explained by scribal error alone. These manuscripts sometimes preserve readings that are neither the base text of A nor its corrected state, but something else entirely.
This pattern suggests a three-stage revision process. First, an initial draft (no longer extant) from which Manuscripts B, C, D, and F ultimately derive. Second, a revised version of this draft, which Nizam al-Din transcribed as the base text of Manuscript A in 694 AH/1295 CE. Third, the further revisions visible in Manuscript A’s margins and corrections, which produced the second version followed by Manuscripts E and G. The base text of Manuscript A therefore represents not Sultan Walad’s first thoughts but an intermediate stage: already revised from an earlier draft, yet subject to further revision thereafter. It should be noted, however, that the divergences between the earliest recoverable stage and Manuscript A’s base text are far less extensive than the revisions visible in Manuscript A itself. The earlier revision appears to have been relatively modest, whereas the campaign of correction documented in Manuscript A’s margins represents a thoroughgoing transformation of the text.
A comprehensive reconstruction of this earliest stage lies beyond the scope of the present study, which focuses on the two versions preserved in Manuscript A. However, the evidence for its existence strengthens rather than weakens the central argument: Sultan Walad approached his Divan not as a fixed collection but as a work in continuous development, subject to repeated refinement over the final decades of his life.
The manuscript tradition provides additional confirmation of authorial revision. Later manuscripts dated after 694 AH/1295 CE follow the corrected text almost entirely, incorporating the changes and deletions found in Manuscript A. Manuscript E, begun during Sultan Walad’s lifetime and completed four months after his death, and Manuscript G, written ten years after Sultan Walad’s passing, both follow the revised version. This pattern indicates that Manuscript A served as the authoritative exemplar for subsequent copies, with its revisions accepted as authentic by the contemporary Mevlevi community. Manuscripts E and G contain all leading errors present in the corrected state of Manuscript A, plus additional separable errors unique to each.Footnote 34 These are not merely polygenic variants (common mistakes any scribe might independently make) but errors with clear directionality that establish the stemmatic relationship. Most decisively, where Manuscript A contains an error, E and G reproduce that exact error, proving they derive from A rather than from an independent source. At the same time, E and G each contain unique errors absent from the other, demonstrating that they copied independently from A rather than from one another. Combined with the Seljuk-period script of all marginal hands (excepting Walad Chalabi’s later annotations), this evidence confirms that the revisions were made close to the manuscript’s production, during Sultan Walad’s lifetime.
Internal evidence strongly supports authorial revision. The corrections consistently improve meter, enhance meaning, and increase poetic beauty, changes requiring both technical expertise and aesthetic judgment that only Sultan Walad would possess. The systematic nature of these revisions suggests they were undertaken during a period of reflection when the poet could review and unify his Divan. Such practice aligns with the literary culture of the time, as revision was common among accomplished poets, including Sultan Walad’s father, Mawlana Jalal al-Din Rumi.
The manuscript’s codicological features provide additional authentication. The careful notation system, extensive vocalization revealing thirteenth-century Anatolian pronunciation, nearly complete pointing (nuqṭa-guẕārī) of the letters, and minimal scribal errors all indicate meticulous production under close supervision. The vocalization of potentially ambiguous words particularly suggests an effort to preserve authorial intention for future generations.
Although the cumulative evidence strongly indicates that both versions belong to Sultan Walad, scholarly caution demands acknowledgment of the impossibility of absolute certainty in matters of medieval authorship. The convergence of paleographic, codicological, textual, and historical evidence creates a compelling case for authorial revision, yet definitive proof remains elusive. What can be stated with confidence is that the Mevlevi community of the early fourteenth century accepted these revisions as authentic, incorporating them into the manuscript tradition that would define Sultan Walad’s poetic legacy for subsequent generations. This contemporary acceptance, combined with the manuscript’s exceptional provenance and the literary quality of the revisions, establishes a probability approaching certainty that we possess in Manuscript A a unique witness to a major Persian poet’s creative process.
Conclusion
What does it mean to observe a medieval poet revising his work? The question may seem straightforward, yet the evidence presented in this study compels us to reconsider fundamental assumptions about authorship, textual identity, and the transmission of mystical knowledge in classical Persian literature.
The prevailing model of Persian textual criticism has long treated the author as a point of origin from which manuscripts descend, accumulating corruption as they pass through successive copyists’ hands. The editor’s task, in this model, is archaeological: to strip away accretions and recover an original, stable text presumed to exist behind the surviving witnesses. Manuscript A of Sultan Walad’s Divan disrupts this paradigm. Here we encounter not corruption but creation, not the degradation of an original but the transformation of one authentic state into another. The crossed-out verses are not scribal errors to be corrected but authorial decisions to be understood. The marginal additions are not interpolations to be excised but revisions to be honored. The text itself exists in motion.
This has implications for how we conceptualize the classical Persian divan as a literary form. Scholarship has generally treated the divan as a collection: poems gathered and arranged but essentially complete at the moment of compilation. Sultan Walad’s evidence suggests something different: the divan as a living document, subject to ongoing refinement as the poet’s understanding deepens and his craft matures. The temporal gap between the two versions, composed before 690 AH/1291 CE and revised after 694 AH/1295 CE, coincides precisely with the period during which Sultan Walad was composing his masnavis. We may be witnessing not merely technical improvement but the retroactive reshaping of earlier work in light of later spiritual development. The poet who revised these ghazals was not the same poet who composed them.
The nature of the revisions themselves demands theoretical attention. Sultan Walad’s corrections consistently move in a particular direction: toward greater musical sophistication, toward more precise theological formulation, toward closer alignment with his father’s distinctive poetic practice. This directionality challenges the assumption, implicit in much scholarship on mutābaʿat (emulation), that imitation represents a static relationship between model and follower. What the manuscript evidence reveals is imitation as process, an ongoing, dynamic engagement with inherited tradition that intensifies rather than diminishes over time. Sultan Walad did not imitate his father once and move on; he returned to his imitations, deepening them, refining them, bringing them into ever closer alignment with the model he sought to honor. Emulation, in this light, appears not as creative limitation but as creative discipline, a chosen constraint that generates rather than forecloses artistic possibility.
The revisions to radīfs prove particularly significant for understanding the relationship between written text and oral performance in Sufi contexts. When Sultan Walad transforms “I do not know the way home” into “I know nothing but You,” he is not merely improving a poem but reshaping a vehicle for collective spiritual practice. The radīf, chanted repeatedly during samāʿ, functions as a form of dhikr, and the choice of what words participants will repeat constitutes a form of spiritual instruction. The manuscript evidence therefore documents not simply poetic revision but liturgical composition: the deliberate crafting of texts designed to induce specific spiritual states through the embodied practice of recitation. This functional dimension of Sultan Walad’s poetry, its role as technology of transformation rather than object of aesthetic contemplation, must inform any adequate assessment of his literary achievement.
The materiality of transmission deserves equal attention. Manuscript A was not produced by a professional scribe working for hire but by Nizam al-Din, Sultan Walad’s brother-in-law, a master calligrapher embedded within the innermost circle of the Mevlevi community. The manuscript never left that community’s custody, passing from ʿAbd al-Qadir Chalabi to Walad Chalabi to Feridun Nafiz Uzluk, each a descendant of Mawlana’s family, before its deposit in the Mevlana Museum. This chain of transmission is not merely provenance but practice: the careful guarding of authoritative texts against the corruption that Mawlana himself had experienced when Shaykh Fakhr al-Din Sivasi altered his verses without permission. To produce a manuscript within this tradition was to accept responsibility for preserving authentic transmission; to revise such a manuscript was to exercise the authority that only the author himself possessed.
Walad Chalabi’s identification of Manuscript A as a musawwada, an author’s working draft that served as the exemplar for subsequent copies, illuminates why this physical evidence survives. Unlike a presentation copy, whose purpose is completeness and beauty, a musawwada is a working document: it accumulates corrections, bears traces of second thoughts, preserves the archaeology of composition. That the Mevlevi community retained this document rather than replacing it with a fair copy suggests they understood its value precisely as a record of process. The later manuscripts that descend from it carry forward Sultan Walad’s final intentions while the musawwada itself remains as witness to the transformation between versions, a material trace of creative labor that the finished product necessarily conceals.
The question of which version constitutes “the” Divan admits no simple answer, and this indeterminacy is itself significant. Modern editorial practice, shaped by print culture’s assumption of textual fixity, struggles with works that exist in multiple authorial states. Do we privilege the earlier version as closer to original inspiration, or the later as representing the author’s mature judgment? The evidence suggests that Sultan Walad himself would not have recognized this dilemma. His practice of revision was not replacement but accretion; the later version does not cancel the earlier but builds upon it, as commentary builds upon text. A truly adequate edition of Sultan Walad’s Divan might need to present both versions not as main text and variants but as equally authoritative witnesses to different moments in a continuous creative process.
The misattribution of Sultan Walad’s finest ghazals to his father, a phenomenon documented across multiple manuscript traditions and perpetuated even by experienced scholars, assumes new significance in light of this evidence. Such misattribution has traditionally been read as testimony to Sultan Walad’s success in capturing his father’s voice. But the revision evidence suggests a more complex interpretation: Sultan Walad did not simply capture his father’s voice but actively constructed it, bringing his own poetry into ever closer alignment with a model that he was simultaneously helping to define. The boundary between original and imitation, between Mawlana’s poetry and Sultan Walad’s, proves less stable than literary history has assumed, not because Sultan Walad lacked originality but because originality itself operates differently within traditions of sacred transmission than within modern conceptions of individual authorship.
This study has examined one poet, one manuscript, one tradition. Yet the questions it raises extend far beyond Sultan Walad to the foundations of Persian literary history. How many other classical poets revised their work in ways that surviving manuscripts might document? How many textual “corruptions” are in fact authorial alternatives? How should we understand the relationship between the mystical poetry composed in states of spiritual intensity and the same poetry as it existed after reflection and revision? The answers to these questions await future scholarship. What this study establishes is the necessity of asking them, and the extraordinary evidence that careful attention to manuscript witnesses can provide.
Acknowledgements
None.
Competing interests
The author declares no competing interests.
Financial support
This research received no specific grant funding from any funding agency, commercial or not-for-profit sectors.
Mina Raminsabet is the Farzaneh Family Lecturer of Persian Language and Literature at the University of Oklahoma. She holds a PhD in Persian language and literature from the University of Tehran. Her research focuses on Persian poetry, mystical literature, and textual criticism, with published books, articles, and translations in the field. She also serves as a contributing editor for World Literature Today.
Open access
