Khotanese as a Language of the Tarim Borderlands

Abstract

This paper examines the Khotanese tradition in the political and cultural history of the Tarim Basin in the first millennium CE by foregrounding its role as an active facilitator within a multicultural and continually transforming geopolitical environment. The paper approaches Khotanese as both a medium through which local forms of social organization were articulated and a mediator embedded in wider circuits of exchange linking the Tarim Basin with South, Central, and East Asia. Particular attention is given to the linguistic and textual evidence for interaction with other traditions in and around the Tarim Basin, and to cases in which adaptation, (re-)composition, and translation can be associated with identifiable historical settings, institutions, and actors. The paper argues from selected examples that the history of Khotanese illuminates how regional languages sustained local authority while remaining deeply entangled with transregional formations of knowledge, culture, and exchange, and how, in the contested spaces of imperial borderlands, local communities had to hone cultural prestige in and through their languages in order to maintain their standing.

1. Introduction

This study examines Khotanese as a case through which to explore the linguistic ecology of the Tarim Basin, located in present-day Xinjiang Uygur Autonomous Region in the People’s Republic of China, in the first millennium CE and, more generally, the ways in which regional languages operated within, and were shaped by, the interregional networks of interaction that modern scholarship has come to describe as the Silk Road(s).1

The Tarim Basin has long been regarded as one of the major cultural contact zones of Eurasia (see, e.g., Rong 2022); recent scholarship, however, has emphasized that its oasis centers were not only waystations along long-distance routes of exchange (Neelis 2011, pp. 289–309; Zürcher 1990, p. 181), but from an early stage onward core areas of their own (Chen and Hiebert 1995), shaped by ongoing processes of adaptation and transformation. The region was characterized by a high density of intersecting economic (Barisitz 2017), political (see, e.g., Baumer 2014), cultural (see, e.g., Rong 2022), and linguistic vectors (see, e.g., McRae and Nattier 2012). As part of the eastern Silk Road corridor, the Tarim Basin was linked to South, Central, and East Asia through an archipelago of oasis polities whose viability depended on their ability to accommodate and recalibrate competing economic, political, and cultural logics. Rather than a transit corridor between empires, it constituted an imperial borderland zone in which multiple spheres of power overlapped and where local polities actively shaped the conditions of connectivity. Tarim Basin polities exemplify the long-term persistence of local linguistic, cultural, and institutional diversity within transregional networks of exchange, rather than their absorption into homogenizing imperial structures. They can be conceptualized not merely as a zone “in between” larger powers, but as a historically active borderland in which local actors, languages, and institutions functioned as media and mediators shaping the circulation, transformation, and localization of cultural and political forms across Eurasia.2

Within this wider mosaic, the kingdom of Khotan emerged, from the first century CE onward, as an important node of Buddhist interaction (Cribb 2023; Brough 1962, p. 55). The consolidation of Kushan rule intensified the Tarim Basin’s links with Gandhara—the principal northwestern nexus of early Buddhist activity—facilitating the transfer of religious repertoires, textual and artistic forms, as well as associated material and devotional practices into local institutional and social contexts (see Malzahn and Fellner, forthcoming; Fellner and Malzahn, forthcoming; cf. also Malzahn and Zin 2025).

The available evidence suggests that Khotan (very much like other Tarim polities) was not a mere recipient of these incoming traditions but engaged in processes of selective appropriation and recontextualization. South Asian religious and intellectual forms were adapted to local institutional settings and combined with existing cultural practices, giving rise to a distinctive regional idiom. Its influence took shape through reciprocal encounters in which traditions developed within Tarim Basin polities—emerging from sustained contacts among neighboring urban centers—entered into dialogue with the institutional and ideological frameworks of surrounding imperial formations, leaving discernible effects on both sides of these interactions. To fully grasp this constellation, one must turn to the surviving material record—a comparatively vast yet fragmented corpus.3

Against this background, the paper treats Khotanese not as an instance of a repository of translated material from major literary languages but highlights its status as a locally embedded medium of cultural production and projection in its own right. From this perspective, Khotanese provides a historically situated point of entry into the question of how the languages of Tarim Basin polities were shaped by, and responded to, the economic, political, cultural, and intellectual interactions that linked the region to its wider trans-Eurasian environment.

Among the polities of the Tarim Basin in the first millennium, the kingdom of Khotan stands out for its political longevity, its sustained religious patronage, and its comparatively rich literary production in a local language—that is, a language shaped within the institutional and cultural environments of the eastern Tarim borderlands and not aligned with any of the dominant languages of the surrounding empires: Khotanese. The language’s comparatively long attestation permits an analysis of its dual function: as a medium of local political and cultural life and as one of the vehicles for participation in wider networks of Buddhist transmission and intellectual exchange across the eastern Silk Road.

In this perspective, the paper examines how Khotanese emerged and was sustained within the settings of a regional polity at the borderlands of the eastern Silk Road. It considers how the language engaged with external textual and cultural models through processes of adaptation, re-use, and re-composition that can be traced in the surviving sources. On this basis, the case of Khotanese is used to illuminate three interrelated dimensions of the linguistic ecology of the Tarim Basin:

(1)

the embedding of a locally rooted language—not belonging to any imperial center—in the political and cultural routines of a borderland polity;

(2)

the formation of regional literary and doctrinal styles that integrate autochthonous traditions with different Buddhist models through translation, paraphrase, and (re)composition; and

(3)

the role of such regional languages as intermediaries within wider circuits of textual, religious, and intellectual exchange.

The analysis is organized in five parts. Section 2 situates Khotan and the Saka languages within the geopolitical and cultural context of the Tarim Basin. Section 3 expands this frame to consider the broader linguistic ecology of the Tarim Basin, documenting the plurality of languages and scripts in use and the mechanisms that sustained this diversity. Section 4 constitutes the core of the paper and is divided into three subsections: Section 4.1 traces the linguistic influence on and of Khotanese; Section 4.2 analyzes the profile of the Khotanese corpus, focusing on its locally inflected character; and Section 4.3 addresses the role of Khotanese in translation and transmission processes, particularly in relation to Silk Road and Buddhist networks. Section 5 concludes by situating the Khotanese case within wider debates on the role of language in structuring transregional religious ecumenes and articulating local agency on the Silk Road.

Based on recent scholarship, the objectives of this study are thus threefold. First, to situate Khotanese within the multilingual environment of the Tarim Basin not as a marginal variety, but as a central language of the heyday of the Silk Road, sustained and continuously shaped by the dynamics of a highly plural region. Second, to examine the orientation of Khotanese as both a locally embedded medium and a language that participated in wider transregional interactions through translation, manuscript circulation, and the movement of people and ideas. Third, to assess, on the basis of selected examples, the contribution of Khotanese to the transmission, adaptation, and localization of Buddhism in the eastern Silk Road borderlands, thereby investigating how linguistic practice intersected with cultural production and political relations in this historical setting.

Rather than assuming explicit language policies that cannot be documented, the following discussion proceeds from the extant Khotanese material and from the roles that Khotanese agents played in cross-regional networks. It underscores the importance of viewing local Buddhist traditions not only as minor, but as vital direct evidence from hubs of the Silk Road and sites of historical and cultural agency—shaped by and shaping the Tarim region’s networks of interaction.

2. Historical and Geographic Context

2.1. The Tarim Basin and Its Strategic Role on the Silk Road

Situated at the interface of several major civilizational spheres, the Tarim Basin functioned as a pivotal hinge in the long-distance networks that traversed Eurasia during the first millennium CE (Barisitz 2017; Baumer 2014). To the north extended the zones of steppe polities, whose fluctuating configurations—from the Xiongnu to the Mongolian empire—exerted pressure through mobility, tributary diplomacy, and military assertion. To the east, the imperial formations of China—especially under the Han and Tang dynasties—sought to integrate the region into their sphere of influence, aimed at sustaining transregional corridors. To the west and south, the Indo-Irano-Hellenic and South Asian spheres—with key nodes in the Kushan Empire and the Buddhist centers of Gandhara and Mathura—acted as cultural vectors, transmitting Iranic patterns of prestige and institutions and Indic forms of doctrine and erudition. Within this multilayered frontier zone, the polities of the Tarim Basin, including the city-state kingdoms of Khotan, Kucha, and others, operated as mediating agents—geopolitically constrained yet culturally generative, situated at the juncture of distinct linguistic, political, and cultural ecologies.

Despite their limited military capacity, their exposed position between larger imperial and steppe powers, the oasis states of the Tarim Basin participated in the transregional circuits of trade, communication, religious transmission, and symbolic capital. Khotan and Kucha in particular demonstrate how peripheral polities could leverage their position to exert far-reaching influence: as centers of Buddhist patronage and transmission, they actively participated in shaping the intellectual and religious topography of the Silk Road and East Asia. Through the cultivation of multilingual elites, the sponsorship of textual production and transmission, and the establishment of diplomatic ties oriented towards multiple directions, these polities maintained a sphere of cultural authority which stood in contrast to their marginal geographical and political status. These dynamics underscore the extent to which the Tarim Basin as borderland was not merely a conduit but a site of articulation and transformation within the broader Silk Road sphere—where linguistic plurality, political pragmatism, and cultural creativity intersected in the service of local autonomy and transregional relevance (see Malzahn and Fellner, forthcoming; Fellner and Malzahn, forthcoming).

2.2. Khotanese and Tumshuqese

Khotanese and Tumshuqese constitute especially instructive cases for examining the multilingual landscape of the eastern Tarim Basin. They represent varieties that developed in close connection with the oasis polities of Khotan and Tumshuq, respectively. Among the Eastern Middle Iranic languages, they are more closely related to Sogdian than to Bactrian. A modern relative of Khotanese and Tumshuqese—though likely not a direct descendant—is Wakhi (Emmerick 2009).

Khotanese is attested in a substantial body of manuscript fragments written in an adapted form of the Indic Brahmi script (Sander 2005), ranging from around the 5th to the 11th centuries CE. Its literary corpus encompasses various genres, often revealing a multilingual textual environment in which Iranic, Indic, and occasionally Chinese, Tibetan, and Turkic elements intersect (see below). This corpus, though almost entirely preserved in overall Buddhist contexts, suggests that Khotanese was not only a vehicle of doctrinal transmission but also functioned as a language of administration and everyday life, integrated into the broader structures of elite literacy (Maggi 2009, 2020).

The distribution of Khotanese sources is conditioned by the fundamentally accidental nature of archeological survival and early twentieth-century retrieval under imperialist conditions. The extant corpus does not represent a systematic or comprehensive archive, but rather a series of chance preservations shaped by site function, depositional practices, and the contingent pathways of discovery and collection. Crucially, it derives from distinct archeological and institutional contexts that decisively shape its textual profiles. Manuscripts recovered from sites within the historic kingdom of Khotan—most notably Dandan Oilik, the Domoko oasis, Khadalik, and related locations such as Mazar Tagh and Balawaste—include a sizeable body of administrative and documentary material, especially concentrated in the late eighth and early ninth centuries. These texts document everyday practices of governance, taxation, labor organization, and military administration and are often relatively tightly datable. By contrast, the Khotanese material from Dunhuang, preserved primarily in the famous Cave Library, is overwhelmingly Buddhist and literary in nature, dominated by sūtras, ritual texts, verse compositions, and related works, largely dating to the tenth century. This contrast also reflects the differing functions of the sites themselves—settlement and administrative contexts in the Khotan region versus a monastic deposit context at Dunhuang. Thus, the surviving corpus is shaped by site-specific practices of production, use, and preservation (see below).

The three principal phases of the Khotanese language—Old, Middle, and Late—can be broadly assigned to the 5th–6th, 7th–8th, and 9th–10th centuries, respectively, although their exact boundaries and internal stratification remain debated in the literature. Old and Middle Khotanese are attested in manuscripts recovered from the core region of Khotan and extending eastward as far as Endere, east of Niya. Late Khotanese, by contrast, is the language of the manuscript corpus found at Dunhuang (Skjærvø 2003b, pp. lxx–lxxi).

Tumshuqese, by contrast, is attested in a small and heterogeneous corpus of documents, mainly secular with only a few fragmentary, but quite interesting Buddhist texts (Emmerick 2009, p. 379).

The textual and material evidence from Khotan and Tumshuq points to a long-term process in which these languages became embedded within urbanized environments. Their trajectories suggest that locally rooted languages, when embedded in literate institutions and connected to systems of religious authority and cultural prestige, could support both locally oriented cohesion and outward-facing forms of communication. In Khotan, this development was shaped to a particularly significant degree by the mechanisms of elite Buddhist patronage (R. Chen 2023).

2.3. Khotan and Buddhism

Khotan, strategically situated along the southern rim of the Tarim Basin, was one of the most enduring polities of the pre-Islamic Tarim Basin. Its political longevity—presumably, already from the second century BCE4 and, directly attested, from the first centuries CE (Cribb 2023) until the Islamic conquest in the early 11th century—coincided with its emergence as a pivotal node in trans-Eurasian networks of trade (Barisitz 2017), diplomacy (Wen 2023), and religious transmission (Meinert 2015b; Meinert and Sørensen 2020a; Kasai and Sørensen 2022; Doney et al. 2023). Despite its limited military reach and the persistent encroachment of larger political formations, Khotan maintained a remarkable degree of cultural autonomy through the cultivation of strategic alliances, the integration of transregional religious institutions, and the development of a distinct scriptural culture. Central to this formation was the articulation of Buddhist kingship, which provided the ideological infrastructure for both internal governance and external representation (R. Chen 2023).

The historical reconstruction of Khotan remains constrained by the fragmentary nature of the sources and their marked elite orientation. Surviving documents and inscriptions derive almost exclusively from courtly or monastic milieus, as well as from some local administrative contexts, even though these spheres are connected.

These are complemented by references in Chinese annals5 and Tibetan historiographic texts (see Emmerick 1967; Beckwith 1987), which—though invaluable—offer externally mediated views often inflected by the logics of larger political entities and their interests.

As a result, the available record privileges the perspectives of royal patrons, dignitaries, and high-ranking clerics, while rendering the wider social field of Khotanese society only partially visible. This imbalance is, of course, by no means unique to Khotan, but it must be borne in mind when generalizing from the surviving materials.

A further layer of complexity is introduced by the rich tradition of origin narratives associated with the kingdom, which reached their codified form between the 6th and 8th centuries and are preserved in Chinese and Tibetan transmission. These legends—which detail the mythic arrival of South Asian princes, miraculous relics, and divine interventions—were not historical accounts, but rather discursive instruments that naturalized the Buddhist polity and its institutions within a sacral-historical cosmopolitan framework. To treat these accounts either as transparent chronicles or as mere fiction would be methodologically naïve; they are better approached as cultural artifacts of a particular phase in the consolidation of royal and religious authority. Their internal coherence, chronological layering, and rhetorical density point to a mature narrative apparatus that was embedded in the performative and scriptural life of the court-monastery complex (R. Chen 2023).

The following is a brief chronological overview of the history of the kingdom of Khotan:6

2nd c. BCE to 3rd c. CE: Khotan was first brought to Chinese attention through the mission of Zhang Qian 张骞 (died c. 114 BCE), sent by Emperor Wudi 汉武帝 (r. 141–87 BCE) around 140 BCE to seek alliances against the Xiongnu. His report, preserved in the Hanshu 汉书 and Shiji 史记, mentions Khotan (Yutian 于寘) as a jade-producing region in the southern Tarim Basin. In the Eastern Han period, China launched campaigns to secure control over oasis polities along the Silk Road. Khotan became entangled in tensions with other Tarim Basin city-states, while coming under the influence of the Xiongnu and the expanding Kushan Empire.

Local documentation from this period emerges in the form of Sino-Kharosthi coins (from the 1st to the 2nd century CE), which bear bilingual inscriptions in Chinese and Middle Indic in Kharosthi script. Though possibly issued by Khotanese kings, the exact correlation with names found in Chinese sources remains unconfirmed. A Kharosthi document from Endere (CKD 661), possibly datable to the 3rd century CE, contains the name of a Khotanese king of kings, Vijitasiṃha, and other Iranic elements, indicating an Iranic-speaking ruling elite in the region.

There are early traces of Buddhism at this time. Textual evidence includes the retrieval by Zhu Shixing 朱士行 (203–282) of a Prajñāpāramitā manuscript around 260 CE, which became the source for T.221. Additionally, the so-called Gandhari Dharmapada from Kohmari Mazar near Khotan suggests early scriptural transmission. However, there is no archeological evidence for residential monasteries before the late 3rd century. Radiocarbon-dated temple remains from Domoko confirm links to Kushan-period Gandharan Buddhism (Wu 2012).

4th–early 5th c.: Little is known about the 4th century, though Chinese annals note tributary missions. Local dynasties in Khotan seem to have consolidated power. Sericulture seems to have been introduced at this time. Faxian’s 法顯 (337–422 CE) visit in 400 CE reveals a kingdom thoroughly committed to elite-sponsored Mahāyāna Buddhism. The royal monastery Gomati(ra) was then a prominent institution.

5th–7th c.: The period is marked by relative stability and growing political integration with Chinese polities. Chinese records note Khotanese envoys and monastic missions. Chinese pilgrims like Song Yun 宋云 (visit 519) and Xuanzang 玄奘 (602–664 CE; visit 644/645) describe a well-organized monastic society supported by the crown and centered around prominent monasteries. These institutions were tied to foundation legends and relic cults as seen in the Prophecy of the Li Country, a Khotanese text translated into Tibetan (see Emmerick 1967) and the Annals of the Li Country (Thomas 1935, vol. 1, pp. 303–32). This and other textual (see below) records together with some artistic evidence (see E. Forte 2020) point to a model of royal Buddhist kingship aligned with devotional frameworks.

This period marked the maturation of Buddhism in Khotan. Khotan began translating and rendering Buddhist texts into Khotanese, adopting and adapting the Indic Brahmi script (Sander 2005)—Tarim Brahmi (Fellner et al. 2019, pp. 61–64)—using Chinese-style locally produced paper (Helman-Ważny 2021), in material formats ranging from pothī-style books (Baums 2021)—bundles of oblong, loose leaves modeled on Indian palm-leaf manuscripts—to Chinese-style scrolls composed of multiple joined folios extending to several meters in length (Sander 1988). The Book of Zambasta, based on but not directly translated from Indic sources and the Bhaiṣajyaguruvaiḍūryaprabhasūtra, probably translated from Chinese (Loukota 2019) demonstrate an already broad and expanding scriptural knowledge and literary production and awareness, also evidenced by numerous Sanskrit works found in the realm of Khotan.

7th–8th c.: Following the Tang conquest of the Western Turks, Khotan became a Tang vassal. The Anxi Protectorate oversaw administration, and the post of Vice Military Governor was often held by Khotanese royalty. Chinese state monasteries were established and often co-administered with Khotanese kings. These sites integrated Khotanese and Chinese Buddhist traditions.

798–c. 840: After 798, Tibetans took control of Khotan. Khotanese Kings ruled under the oversight of Tibetan prefects at Mazar Tagh. Despite the subordination, local administration and dynastic continuity persisted. Buddhist institutions remained intact. Monks such as Śīladharma translated Chinese and Sanskrit texts into Tibetan.

9th—10th c.: After the fall of the Tibetan empire, Khotan reasserted its independence. Alliances with Dunhuang through marriage and diplomacy flourished. Khotanese Kings sent envoys and married into the Cao family. Royal patronage continued in Dunhuang, as seen in cave murals and literary dedications. Khotanese monks in Dunhuang honored their king during the pravāraṇa (end of the rainy season) ceremony, as evidenced by a 10th-century Khotanese panegyric (Bailey 1965, pp. 102–8). The text affirms the king’s ritual sovereignty and generosity towards the Buddhist community. It concludes with a retelling of King Kaniṣka’s endowment of a stupa on Aśvaghoṣa’s counsel, illustrating how the Kushan model of dharmic kingship was still alive in late Khotan.

1006/7—12th c.: The conquest by the Kara-Khanid Khanate around 1006 led to the dissolution of the Buddhist kingdom. Refugee elites fled to Dunhuang, Amdo, and northern China. Only a few Khotanese are recorded after 1006, but those known—such as Aligu in Amdo and Wang Yuande 王元德 in Taiyuan—remained devoted to Buddhism and upheld royal traditions in Tibetan and Chinese contexts.

The Khotanese language was shaped by historical institutional frameworks while also serving as one of the means through which they were maintained and enacted. The extensive Buddhist corpus in Khotanese was not a spontaneous expression of a vernacular Buddhist tradition but the outcome of systematic activity embedded within elite political and religious structures. At the same time, the language served as a medium for translating, reframing, and localizing different traditions within the discursive frames and cultural expectations of a Khotanese-speaking polity (see below). The emergence of Khotanese as a literary and liturgical language reflects a broader process by which it was mobilized to construct, legitimate, and reproduce a distinctive configuration of power, prestige, and piety in the context of the Eastern Silk Road:

“The interplay between religion and secular power, as it played out in the Buddhist centres located along the Silk Road during the 6th to 14th century, often followed a model where mutual benefit played a crucial role. The Buddhist religion was dependent upon benevolent rulers, who extended their graces and economic muscle to sustain religious establishments, sponsor rituals and the production of holy scriptures, and additionally promote religious leaders, while Buddhism on its part lent its prestige, salvific promises (including divine protection), and its role as a shared, stabilising factor to the glory of the various rulers. This was particularly important in the multi-cultural setting of the area we are dealing with, where Buddhism’s role as a unifying factor also involved identity politics and negotiating among competing ethnic and/or religious groups.”.

(Meinert and Sørensen 2020b, pp. 2–3)

3. The Linguistic Ecology of the Tarim Basin

The linguistic ecology of the Tarim Basin during the first millennium CE presents a case of exceptional multilingual density and stratified language use, shaped by sustained interaction between local polities, transregional networks, and multicultural connectivities. Far from being a marginal periphery, the Tarim region functioned as an integrative space where linguistic plurality was not merely tolerated but seems to have been institutionalized across religious, scholarly, administrative, and diplomatic domains—developments underpinned by the region’s strategic role in Silk Road trade, which provided the material and infrastructural foundation for such sustained cultural and intellectual exchange. Royal courts, monastic centers, and translation workshops served as hubs of cultural production and exchange—sites that not only preserved this linguistic richness but also brought to light a cultural complexity that has, in modern times, reshaped entire fields of inquiry, from Buddhist studies and historical linguistics to manuscript culture and the history of translation. This diversity seems not incidental but structurally embedded in the Tarim Basin’s geopolitical function as a hinge between Central, East, West, and South Asia.

At least twenty-one languages7 and seventeen scripts8 have been identified across the Tarim Basin and its broader cultural orbit, reflecting an extraordinarily intricate and multilingual environment. These languages encompassed a range of language families, including Sino-Tibetan (Trans-Himalayan), Indo-European, Turkic, Mongolic, and Semitic. Among them were languages of major global agents—such as Chinese, Indic, Persian, Syriac—as well as regionally embedded varieties like Khotanese and Tocharian (a separate branch of Indo-European). Also attested were important languages such as Bactrian, Sogdian, Parthian, Old Uyghur, Khitan, Tangut, and various Middle Indic varieties. Each of these languages was tied to specific historical communities, cultural trajectories, and institutional settings.

This linguistic diversity was matched by an equally rich array of scripts. These scripts were often adapted for use across multiple languages, reflecting both the flexibility of scribal traditions and the overlapping religious, administrative, and diplomatic functions they served. For instance, Brahmi was not only employed for Indic languages but was also adapted for Tocharian, Khotanese, Tumshuqese (and other Iranic varieties), and Old Uyghur (see Maue 2019; Sander 1986). The script-language matrix thus reveals a high degree of fluidity and pragmatic adaptation, further confirming the region’s role as a laboratory of cultural interaction and innovation.

The locally attested languages with the most substantial corpora closely linked to Buddhist institutions—Khotanese and Tocharian—reflect long-standing traditions of translation, commentary, and original composition. Both Khotanese and Tocharian developed as literary languages within structured religious settings. Both of their respective corpora include canonical Buddhist texts, autochthonous embellished and enriched literature of various genres, and administrative documents, all reflecting efforts to establish idioms adapted to local environments and needs (Maggi 2009; Pinault 2016).

While Khotanese is rightly emphasized as a major Buddhist literary language of the southern Tarim Basin, its role as an administrative language—especially within the historic kingdom of Khotan—deserves equal attention in this context. A substantial portion of the earliest and most securely localized Khotanese material, particularly from sites such as the Domoko area, Khadalik, and related locations within Khotan proper, consists of administrative and documentary texts. These attest to a well-established bureaucratic use of Khotanese in local administration and organization. The prominence of Buddhist literary texts in the surviving corpus should therefore not obscure the fact that Khotanese functioned over a long period as a language of state and administration, embedded in everyday practices of authority and regulation. The relative visibility of these functions in the extant material is itself shaped by patterns of preservation and recovery, but taken together, the evidence points to a broader functional spectrum than a purely monastic or literary profile would suggest.

The South Asian linguistic influence—especially from Sanskrit and Middle Indic Varieties, particularly Gandhari—was pervasive throughout the region. Sanskrit and Gandhari, although rarely used for original literary composition in the Tarim Basin, retained canonical authority and served as a source language for translation and adaptation. Indic elements were integrated into local Buddhist literary styles, often resulting in hybridized textual forms that blended regional languages with calquing and borrowing of Indic lexical material and phraseology. The spread and regional adaptation of the Brahmi script, Indic literary forms, and erudition attests to the region’s connection with South Asian cultural paradigms.

The material foundations of this literary and religious exchange were equally shaped by transregional influences. While paper and ink were often locally produced, their manufacture followed Chinese techniques, reflecting the technological prestige and practical dominance of Chinese models in the region (Helman-Ważny 2021). In times of scarcity or disruption, Chinese paper was imported, repurposed, and reused—sometimes bearing traces of earlier texts—highlighting both the interconnectedness of the Tarim Basin with broader networks of material circulation and the pragmatic strategies employed to sustain textual production in a resource-constrained environment. Studies of Tocharian (Mascheroni and Musitz, forthcoming; Pan and Chen 2021) and Dunhuang (Wen 2023, pp. 182–85) corpora suggest that such practices were widespread, although more work remains to be done on regional variation.

From the late first millennium onward, Turkic languages—especially Old Uyghur—entered the linguistic landscape with increasing prominence. This shift was not merely demographic but tied to the incorporation of the Tarim Basin into the political and cultural orbit of Turkic-speaking polities, particularly the Uyghur Khaganate. Multilingual documents begin to appear, illustrating not only language shift but also continuity and accommodation, as older cultural traditions remained in use alongside new ones. This process of linguistic and cultural layering did not occur in a vacuum but drew upon earlier traditions that had already established models for textual production and religious expression. Notably, Tocharian—once a dominant literary language in the northern Tarim Basin—provided important precedents in script usage, translation practices, and genre conventions. Even as Tocharian gradually receded from active use, its legacy persisted in scribal habits, manuscript formats, and the transmission of Buddhist texts. The continued presence of Tocharian alongside languages like Old Uyghur reflects a pattern of adaptation rather than replacement, in which incoming traditions built upon the institutional and intellectual infrastructures of their predecessors (Kasai 2023).

Religious institutions sustained by elite patronage—particularly Buddhist monasteries—played a central role in the production and preservation of this multilingual corpus. The predominance of Buddhist texts in the surviving manuscript corpus is less a direct reflection of the overall textual culture of the Tarim Basin than a consequence of patterns of preservation and recovery. What endures is primarily the output of literate institutions—above all monasteries—whose textual practices were embedded in doctrinal, ritual, and pedagogical routines that favored material durability and reproduction. This evidentiary imbalance necessarily skews our reconstruction of the region’s linguistic and cultural dynamics toward elite and religious contexts. Yet it also highlights the central role played by religious, particularly Buddhist institutions, in shaping the conditions under which local languages were standardized, codified, and inscribed into broader transregional formations.

In this context, interpreters and translators—frequently mentioned in various sources—served as institutional mediators and indispensable actors in diplomatic and cultural exchanges (see Lung 2011). Chinese and Tibetan records, for instance, refer to interpreters, while Khotanese itself mainly attests to multilingual practice through loan strata and the participation of Khotanese monks in translation teams (see below). Their presence points to the lack of any single spoken language that prevailed across all domains of interaction and underscores how communicative competence in the Tarim Basin depended on context-specific forms of multilingualism. These individuals were more than transducers; they acted as intermediaries between different political, cultural, and religious spheres, facilitating communication and mutual understanding across linguistic and conceptual boundaries. Their activity seems to point to the broader social embeddedness of multilingualism. The linguistic and scriptural heterogeneity of the Tarim Basin thus emerges as more than a reflection of commercial connectivity and geographic convergence. It was a historic configuration, shaped by strategies of prestige, investments into textual production, and the ongoing interplay between transregional integration and local initiative. The result was a deeply rooted multilingual ecology generated through prolonged cultural interaction. This linguistic plurality played a key role in the formation of political authority, the consolidation of religious institutions, and the negotiation of cultural identities across diverse communities, allowing distinct yet interconnected traditions to coexist within a landscape defined by overlapping spheres of influence.

4. The Borderland Character of the Khotanese

4.1. Linguistics

The linguistic profile of Khotanese is inseparable from the historical and cultural processes that shaped the Tarim Basin as one of the most dynamic areas of the Silk Road. While rooted in the Eastern Iranic linguistic continuum, Khotanese emerged as a regionally grounded yet transregional integrative medium, structured by the demands of Buddhist textual transmission, shaped by prolonged exchange with Indic models, and embedded in an environment of intensive multilingual interaction. Its development exemplifies how local languages on the eastern Silk Road did not merely absorb external influences but actively contributed to the formation of a shared communicative infrastructure across and beyond the Tarim Basin.

A key indicator of this transregional role is the stratified nature of Indic influence on the Khotanese lexicon. Khotanese Buddhist texts preserve a layered inventory of borrowings from Indic languages, reflecting both the chronology and geography of Buddhist transmission. The earliest wave of Indic influence appears in the form of Gandhari loanwords, which entered the region alongside the introduction of Buddhism during the early centuries CE. This first phase reflects a period when the dominant idiom of Buddhist textuality in the northwest was still Gandhari, and when Khotanese translators were working primarily from Middle Indic sources, possibly oral as well as written (Loukota 2023).

Already in the Book of Zambasta and related texts, a substantial layer of learned Sanskrit vocabulary is present. In the later stages of Khotanese textual production, particularly during the 7th to 9th centuries, a marked Sanskritization of the lexicon can be observed. This second wave of borrowing introduced a range of more technical terms drawn from the Mahāyāna scholastic tradition. These borrowings from Sanskrit testify to the growing influence of South Asian scholastic centers and to intensified contact between Chinese and Tibetan Buddhism, where Sanskrit had by then become the primary language of scriptural authority.

The resulting bilingual lexicon preserved earlier Gandhari while integrating Sanskrit forms (see

Table 1. Examples of loanwords across different Indic layers (source of the Khotanese word in bold).

	Khotanese	Gandhari	Sanskrit	cf. Toch. B
Toponyms	Jīyavana-	Jedavana-	Jetavana-	Jetavaṃ
	Vaiśśāli-	Veśali-	Vaiśāli-	Vaiśāli
Theonyms	Kālśava-	Kaśava-	Kāśyapa-	Kāśyape
	Ekaśrṛṅga-		Ekaśṛṅga-	Ekaśriṅke
Cosmology	vaśära-	vayira-	vajra-	waśir
	Kaläyugga-		Kaliyuga-	kaliyuk
Monasticism	päṇḍävāta-	piṃḍavada-	piṇḍapāta-	pintwāt
	vajrropama-		vajropama-	vajropame
Doctrine	vimūha-	vimoha-	vimokṣa-	wimokṣ
	kleśa-	kileśa-	kleśa-	kleś
Text titles	avidharma-		abhidharma-	abhidārm

), resulting in a layered, diachronically structured vocabulary that reveals the changing axes of cultural and religious prestige. A similar phenomenon in a different Buddhist tradition can be observed regarding Tocharian (Pinault 1995).

However, Indic influence in Khotanese was not merely a matter of straightforward lexical loans. The process of translation entailed significant conceptual adaptation. Many borrowed concepts were also calqued using Iranic roots, producing words that were simultaneously intelligible to local audiences and doctrinally faithful to Buddhist concepts (e.g., khot. dāta-: skt. dhárma-).

Khotanese scribes and translators thus developed a locally anchored Buddhist terminology that combined Indic and Iranic elements into a coherent idiom of religious instruction. This is particularly evident in texts such as the Book of Zambasta, where metrical structure, lexical choices, and thematic composition all reflect the integration of Buddhist content into a local literary form (see below).

Moreover, Khotanese functioned not only as a recipient of lexical and conceptual material but also as a source language in its own right, illustrating the multidirectional nature of linguistic exchange in the Tarim Basin. Khotanese (and its close relative Tumshuqese) contributed a range of Buddhist and administrative and material-cultural terms to Tocharian (see

Table 2. Examples of Khotanese loanwords in Tocharian.

	Tocharian	Khotanese	Gloss
Buddhism	TB pātro TA pātär	acc.sg. OKh. pātru	alms-bowl
	TA pissaṅk	LKh. bi’saṃga- (OKh. bälsaṃga-)	bhikṣusaṃgha-
Administration	TB orśa TA oräś	OKh. aurāśśa-	councilor
	TB kāmarto TA kākmart	acc. sg. *kamardu (OKh. kamala-)	chief
Medicine	TB tvāṅkaro	acc. sg. OKh. *tvaṃgarau (LKh. ttuṃgara-)	ginger
	TB ṣpakīye	LKh. ṣvakā-	suppository

* indicates linguistically reconstructed form.

). These borrowings include both lexical items and calques, many of which were transmitted in monastic or scholastic contexts (Dragoni 2023).

Loan words like these demonstrate the permeability of linguistic boundaries in the region and highlight the role of Khotanese as an intermediary in the transregional circulation of knowledge. The directionality and specificity of these borrowings suggest more than casual contact. They point to sustained institutional interaction between Khotanese- and Tocharian-speaking communities, most likely within shared or adjacent monastic networks. Khotanese thus served not only as a liturgical and literary language within the kingdom of Khotan but also as a bridge language in a wider religious and intellectual system. Its influence on neighboring traditions reflects its functional centrality in the localization and transmission of Buddhist doctrine along the routes of the eastern Silk Road.

Khotanese also exhibits a discernible layer of Chinese and Tibetan loanwords, which attest to the kingdom’s sustained integration into the broader political and cultural dynamics of the Tarim Basin. These lexical borrowings reflect structured and multi-layered contact, encompassing economic exchange, administrative subordination, and religious interaction. Chinese-derived terms—especially from the 8th to 10th centuries—frequently pertain to Tang dynasty bureaucratic titles, fiscal instruments, and units of measurement, indicating the extent to which Khotan was embedded within the Chinese political order. Similarly, the Tibetan stratum, though less systematically documented, is indicative of the period of Tibetan hegemony and the concomitant administrative and ecclesiastical entanglements (Yoshida 2015; Emmerick 1985b). In sum, the Chinese and Tibetan loanword strata underscore the intensity and institutional depth of Khotan’s transregional interactions; representative examples are presented in

Table 3. Examples of Chinese and Tibetan loanwords in Khotanese.

	Khotanese		Gloss
Economy	kiṇa	斤 jīn	measure of weight
	khara	tib. khal	measure of grain
Administration	tcerthūśi	节度使 jiédùshǐ	military governor
	bulāni	tib. blon	minister
Cultural	thaiśi	大师 dàshī	great teacher
	stānaḍa	tib. *ston-bla	teacher superior

* indicates linguistically reconstructed form.

.

Khotanese translation practice represents one of the most revealing sites where Silk Road Buddhism can be observed in action—not as the passive transfer of fixed meanings, but as a process of semantic negotiation within a multilingual environment. From the outset, Khotanese translators worked in a layered Indic–Iranic contact zone in which many Buddhist terms had already entered the language as loans (saṃtsāra-, satva-, bodhi-, dharma-), while others were rendered through Iranic-based calques such as uysnaura- “breath-bearer” for Skt. sattva-. This created a stratified lexicon in which technical meaning could be distributed across inherited Iranic vocabulary, Indic loans, and newly coined compounds.

Unlike the Tibetan tradition, which later enforced one-to-one correspondences in Buddhist text transmission, Khotanese never stabilized a fixed terminological grid. Instead, translation operated through semantic clustering. In the Saṅghāṭasūtra (Canevascini 1993), for example, the single Khotanese noun balysūsti- is used to render bodhi- “awakening,” sambodhi- “perfect awakening,” buddhatva- “Buddhahood,” and buddha-jñāna- “Buddha-knowledge.” Conversely, buddha-kṣetra- “Buddha-field” appears variously as balysānā-tcārami- “Buddha-land,” balysāna-mäṣṣa- “Buddha-field,” and as the direct loan buddhakṣetra-—sometimes within the same text. These shifts do not signal inconsistency so much as an assumption that meaning resides in a semantic field rather than in a single lexical item. Doctrinally dense terms were frequently rendered through paraphrase grounded in Buddhist commentary traditions. Thus tathāgata- is translated in the Saṅghāṭasūtra as kye hutsutu pando tsute “he who has gone along the right path,” while the Vajracchedikā (Konow 1916) has ttāharā-tsūka- “the perfectly going one” and ttāhirau-hvāñāka- “the perfectly speaking one.” Khotanese translation thus incorporates doctrinal interpretation directly into lexical form (Maggi 2009, p. 344).

A characteristic feature is the use of dyadic and triadic renderings, pairing near-synonyms to convey one Sanskrit term. Sanskrit priya- “dear” may become bria-ajsera- “dear and inviolable,” while Sanskrit niryātita- “requited” is rendered as hüḍaimä närṣätaimä “I gave and presented.” In many cases, an Indic loan is coupled with an Iranic word, anchoring imported doctrine in the native semantic system. Sanskrit asatya- “untrue, false” is rendered as asäda adātia “evil and unlawful”. Conversely, long Sanskrit synonym chains—especially in the Suvarṇabhāsottamasūtra’s (Skjærvø 2004) formulas like ārakṣayā paritrāṇam parigraham paripālanam (śāntim svastyayanam) kṛ- “protecting and bringing security to the land”—are sometimes reduced to a single Khotanese verb such as rakṣo yanāmä “to protect,” showing that translators adjusted rhetorical weight rather than copying surface form (Skjærvø 2012, p. 130).

Variation itself was a principle, possibly a sign of Iranic poetic tradition (Skjærvø 2012, p. 127). In the Suvarṇabhāsottamasūtra, Sanskrit sādhukāram adāt “he gave approval” is rendered not only as usāvañau hvate se śuru śuru “he spoke approval: good, good,” but once expanded as usāvañau hvate ysūṣṭe nä śuru śuru “he spoke approval—he was pleased: good, good,” drawing on the formulaic ending ysūṣṭe nä “they were pleased.” In the Śūraṅgamasamādhisūtra (Emmerick 1970), the same Sanskrit phrase becomes buljäte se śuru śuru “he praised: good, good,” because the translator associated approval with verbs used elsewhere for praise. Such shifts show how formulae migrated across semantic networks within the Khotanese literature (Skjærvø 2012, p. 131).

Chronologically, translation strategies evolve. Old Khotanese texts favor interpretative renderings and Iranic religious vocabulary: in the Suvarṇabhāsottama, the Indian goddess Śrī is translated as śśandrāmatā, a name derived from Zoroastrian Spəṇtā Ārmaiti (Maggi 2009, p. 345). Late Khotanese texts, by contrast, rely more heavily on Sanskrit loans, reflecting centuries of multilingual Buddhist literacy.

Finally, poets explicitly theorized their own practice of rendering and translation. The Book of Zambasta insists that the Buddha does not speak meaningless words and that translation must therefore preserve semantic truth (Z 23.7-8; Emmerick 1968, p. 345); the preface to the Khotanese Siddhasāra (Bailey 1945, pp. 1–104) prays to translate “in accordance with the true meaning,” not merely the words (Si A1; Emmerick 1983b, p. 21). Translation in Khotan was thus conceived as a hermeneutic act embedded in Buddhist doctrine itself.

Together, these examples show that Khotanese translation was neither mechanical nor inconsistent. It was a flexible, commentary-driven, semantically layered philology—one that allowed Khotan to participate actively in Buddhist textual culture while preserving a distinctive Iranic linguistic and literary horizon within the eastern Silk Road.

In sum, the borderland character of the Khotanese language is manifested at multiple levels: in its layered Indic vocabulary, in its adaptive translation practices, in its role as a donor language to neighboring traditions, and in its adaptability to express Buddhist concepts in idioms rooted in Iranic linguistic heritage. Far from being a passive conduit of transmission, Khotanese emerges as an active participant in the formation of a shared yet differentiated Buddhist ecumene across the Tarim Basin. Its stylistic features are not only the residue of contact but the outcome of deliberate choices shaped by doctrinal needs, literary conventions, and social institutions of translation. As such, Khotanese represents a model case of how languages in the Silk Road context mediated between supra-regional Buddhist norms and regional communicative needs, textual fidelity and local resonance, authority and intelligibility.

4.2. Written Corpus

The literary production in the Khotanese language provides critical insight into the function of writing of local languages in the Tarim polities along the eastern Silk Road—not merely as media of translation, but as sites of creative engagement, recomposition, and local self-fashioning within transregional frameworks. While—at least based on the currently extant material—mostly rooted in the Buddhist textual tradition, Khotanese literature exhibits a distinct profile shaped by its position at the intersection of Iranic literary heritage, Buddhist doctrinal authority, and the multilingual textual culture of the Tarim Basin.9

Khotanese Buddhist literature is not derivative in the narrow sense of mechanical one-to-one translation from Sanskrit or Chinese sources. Rather, it constitutes a corpus of adapted, restructured, and often innovatively composed texts that reflect both the specificities of the Khotanese cultural context and the broader religious and literary dynamics of the eastern Silk Road networks. Many of the surviving texts are not direct renderings of canonical works but vernacular compositions that integrate doctrinal content with local narrative strategies, metric conventions, and rhetorical devices.

The original Khotanese doctrinal compendia constitute one of the clearest expressions of local intellectual production within the Tarim Basin (next to the Tocharian Buddhist epic theater tradition of works like the Maitreyasamitināṭaka; Pinault 2020; see also Kumamoto 2002). Works such as the Book of Zambasta (Emmerick 1968), the Mañjuśrīnairātmyāvatārasūtra (Bailey 1981, pp. 113–35), and the Book of Vimalakīrti (Skjærvø 2003b, pp. 489–99) are not translations in any strict sense but synthetic compositions that reorganize Buddhist doctrine into forms adapted to the pedagogical, ritual, and institutional needs of a Khotanese-speaking Buddhist community. Their function is therefore not merely exegetical but constitutive: they establish a locally authoritative register of Buddhist discourse capable of supporting Buddhist education and courtly patronage. In this domain, Khotanese operates as a language of doctrinal articulation and institutional legitimacy, enabling the Khotan polity to frame its own version of Buddhist learning while remaining intelligible within wider transregional traditions.

The Book of Zambasta, the most extensive and formally ambitious of these works, exemplifies this mode of textual production. Although deeply informed by Mahāyāna thought, it is neither a translation nor a commentary but a literary reconfiguration of Buddhist cosmology, ethics, and soteriology in a style shaped by local poetics and conventions. Its metrical organization already marks this status, as Khotanese verse neither straightforwardly reproduces Indic nor Iranic patterns.10 At the level of content, these compendia display dense intertextuality and thematic recombination. Earlier materials are reused, reframed, and expanded; narrative motifs are redeployed in new doctrinal settings; technical vocabulary is glossed and paraphrased in ways that presuppose an active hermeneutic engagement with various traditions. Transmission here is not conservational but generative, taking place within an evolving canon that could be adapted to shifting pedagogical, ritual, and political contexts—an approach characteristic of eastern Silk Road literary cultures more broadly, where texts circulated in modular and reconfigurable forms across linguistic and regional boundaries.

At the same time, these compendia integrate religious authority with political self-representation. The Book of Zambasta and related texts explicitly invoke royal patronage, the legitimacy of local monastic institutions, and the court’s role as protector of the Dharma. While such motifs draw on widely shared Buddhist models of kingship, their articulation in Khotanese gives them a distinctly local inflection. The righteous ruler appears not only as a doctrinal ideal but as a figure embedded in the institutional life of Khotan itself. Through a locally inflected mode of literary production, political authority is thus aligned with transregional Buddhist norms, allowing the Khotanese polity to present itself simultaneously as a locus of Buddhist erudition and learning and as an active participant in the “Buddhist globalization” (Meinert 2015a, p. 1).

It is precisely at this juncture that the Avadāna and narrative literature assumes its particular importance. Texts such as the Aśokāvadāna (Dragoni 2013), Kaniṣkāvadāna (Bailey 1954, pp. 107–8, 1965), and Sudhanāvadāna (de Chiara 2013, 2014) provide the narrative counterpart to the doctrinal construction of Buddhist kingship, translating models of righteous rule, karmic causality, and kingly piety into forms accessible to Khotanese audiences. Their transmission in Khotanese often stands in parallel to Indic, Chinese, and Central Asian versions, and comparison across these traditions shows that translation functioned as narrative calibration rather than simple linguistic transfer. The Khotanese Aśokāvadāna, for example, is not a rendering of any known Sanskrit, Tibetan, or Chinese text but a selective recomposition of the Aśoka–Kunāla narrative complex, rearranging episodes, importing motifs from other branches of the tradition, and introducing elements shared only with certain Chinese witnesses (Dragoni 2013). Through such recombinatory strategies—shifting the role of religious instructors, reassigning narrative actions, and reconfiguring causal sequences—the Khotanese version adapts the imperial Buddhist memory of Aśoka to a local horizon that likely intersected with Khotan’s own foundation legends (see above). Translation here thus operates as a medium of selective appropriation, allowing Khotan to inscribe itself into the transregional memory of Buddhist ecumene while articulating its own place within that narrative order.

The Mahāyāna sūtra corpus in Khotanese marks a different layer of textual practice from the original compendia and narrative recompositions: it is the most direct interface between local religious institutions and the transregional canon, and it makes visible—at the level of manuscript transmission, translation technique, and genre selection—how Khotan positioned itself within the wider context of Buddhist text transmission. In the Prajñāpāramitā tradition—the earliest stratum of Mahāyāna sūtra literature—the Khotanese record is both selective and revealing. While the Prajñāpāramitā corpus ranges in India from vast compilations of up to 100,000 verses to extreme condensations of a single syllable, only short Late Khotanese texts from Dunhuang are extant: the Vajracchedikā (Konow 1916), the Hṛdayasūtra (Skjærvø 1988), and the Adhyardhaśatikā Prajñāpāramitā (Emmerick and Vorob’ёva-Desjatovskaja 1995, pp. 24–34). This concentration on highly condensed formats points to a phase of reception in which Prajñāpāramitā doctrine—centered on the realization of emptiness as the path to liberation and Buddhahood—was mediated in forms suited to memorization, recitation, and ritual deployment rather than encyclopedic exposition.

At the same time, these short texts are anything but mechanically reproduced. The Khotanese Vajracchedikā, for example, augments the Sanskrit dialog between the Buddha and Subhūti with a metrical introduction and extensive paraphrastic expansions that draw on established commentarial traditions. Similarly, the Hṛdayasūtra circulates both in translation and in an independent Khotanese commentary, while the Adhyardhaśatikā is transmitted in a bilingual format that juxtaposes Sanskrit esoteric sections with Khotanese exoteric framing. In all cases, translation operates as doctrinal interpretation and pedagogical adaptation, situating Prajñāpāramitā thought within the concrete liturgical and instructional practices of late Khotanese Buddhism rather than reproducing it as a neutral canonical artifact.

Across the wider Mahāyāna repertoire—the Ratnakūṭa (Skjærvø 2003a), Sukhāvatīvyūha (Skjærvø 2003b, p. 176), Saṅghāṭa (Canevascini 1993), Bhaiṣajyaguruvaiḍūryaprabharāja (Emmerick 1985a), Vimalakīrtinirdeśa (Skjærvø 2003a), Śūraṅgamasamādhi (Emmerick 1970), and the Suvarṇabhāsottamasūtra (Skjærvø 2004) among others—the Khotanese materials repeatedly occupy positions that are neither fully identical with known Sanskrit, Tibetan, or Chinese versions nor simply idiosyncratic. Instead, they often preserve intermediate or early recensional states, and in some cases appear to reflect Central Asian textual lineages that cut across the later canonical stabilizations familiar from Nepalese Sanskrit or Tibetan transmission. The Suvarṇabhāsottama is paradigmatic in this respect: the Khotanese witnesses span multiple stages of the text’s development and at points seem to antedate, or at least correct, later recension in other branches of the tradition. Such evidence places Khotan not at the periphery but within the workshop of Mahāyāna textual history, where translation, copying, and recension were themselves modes of participation in canon formation.

Finally, the devotional and protective economy of Mahāyāna is strongly foregrounded: deśanā texts—often written in the first person singular and oriented toward confession, vow-making, protection, and merit—, hymns, namo lists, and dhāraṇī sūtras show how canonical materials were operationalized within liturgical practice and lay religiosity. Taken together, the Mahāyāna sūtras in Khotanese document not only what Khotan read, but how it read: through adaptation as exegesis and through genre selection as an index of societal needs and strategies of religious prestige.

Alongside sūtras and narrative translations, the corpus includes a small śāstra-like stratum of doctrinal prose, above all the unparalleled Old Khotanese Bodhisattva Compendium (Bailey 1963, pp. 91–102) on Bodhisattva duties, whose manuscripts even show later “updating” through interlinear Late Khotanese overlays—an instructive trace of continued use and refunctionalization. This doctrinal register shades into a larger penumbra of fragmentary, often unidentified Mahāyāna texts, e.g., Ratnadvīpa (Skjærvø 2003b, pp. 368–70).

Vajrayāna in Khotan is attested by a small cluster of Late Khotanese texts from tenth-century Dunhuang: a short rosary treatise, probably mediated through Tibetan models, and two dated deśanā compositions, the Vajrayāna verses and the Vajrayāna verses of Cā Kīmä-śani (971 CE) (Skjærvø 2003b, pp. 551–56). The same register recurs in royal panegyric, in the Jātakastava colophon, and in the Sanskrit–Khotanese conversation manual, indicating a socially embedded Vajrayāna strand within late Khotanese Buddhism.

Besides the Buddhist corpus, the Khotanese record preserves a non-doctrinal stratum that is analytically indispensable for the paper’s central claim: Khotanese was not only a vehicle for imported religious material, but a locally embedded medium of social practice—capable of articulating intimacy, mobility, courtly authority, practical knowledge, and multilingual competence within a shifting borderland environment. This material is heterogeneous (poetry, letters, panegyric, itinerary, medicine, bilingual aids) and often difficult to interpret precisely because it rarely offers secure parallels and because much of field’s lexical knowledge is still calibrated on religious usage. Yet precisely this “unparallel” texture is what showcases Khotanese as a language across institutional and everyday domains, not merely a monastic or translational register.

Lyric poetry provides the most immediate glimpse into this wider communicative ecology. Small stanza clusters—such as those appended to the Staël-Holstein roll from Dunhuang—cultivate an idiom of desire, presence and absence, and bodily affect that is unmistakably literary rather than documentary, yet it remains permeable to Buddhist framing. The larger Lyrical poem (Bailey 1956, pp. 34–48; Kumamoto 2000)—preserved in several overlapping manuscripts with a non-homogeneous tradition and shifting stanza order—is particularly revealing: it develops a spring-and-flower poetics with bards, birds, and lovers, only to close by turning that very esthetic of erotic plenitude into a moral lesson, mobilizing Indic mythic exempla of ruin through passion. The genre thus stages, within Khotanese itself, a characteristic borderland synthesis: Iranic-inflected lyric diction and formal play coexist with an explicitly Buddhist didactic horizon, suggesting not a secular enclave but a shared cultural register in which the monastery, the court, and literate lay life were not cleanly separated.11

The verse letters (Kumamoto 1982, 1991, 1993, 1996; Skjærvø 2003b, pp. 45–52, 508–9, 539–40) extend this logic into the social reality of mobility. Although they present themselves as communications to family, teachers, and friends “back home”, their status as unfinished drafts and their elaborate rhetorical texture indicate that they functioned as literary compositions rather than practical correspondence. They develop the theme of separation—often in the voice of travelers and monks on the road—into a high-mode poetics of longing, self-reproach, karmic reflection, and the material logistics of travel (clothing, supplies, exchange), thereby linking an inside perspective to the infrastructures of eastern Silk Road travel. These texts therefore show how a local language could articulate diasporic Khotanese identities within transregional space: Khotanese here is a medium of affective continuity under conditions of displacement.

At the opposite tonal pole stands the small burlesque fragment attributed to a Kīma-śanä/Jinshan figure (Kumamoto 1995, pp. 243–45), whose humorous numeration of age and meals is trivial only in appearance. It reveals a culture comfortable with playful self-mockery, and it also reminds us that scribal and literate milieus were not exclusively devoted to piety or high rhetoric. Such fragments, precisely because they sit at the margins of traditional genres, sharpen the picture of Khotanese as a living literary language.

Panegyric literature, finally, makes explicit the courtly and ideological stakes. The three eulogies on Viśa’-dynasty kings adopt an elevated rhetorical register reminiscent of Sanskrit praśasti, yet they remain anchored in concrete historical occasions—religious endowments, embassies, monastic foundations, and (in the case of Viśa’ Dharma) diplomatic engagement with Dunhuang and the Chinese world (Skjærvø 2003b, pp. 45–47, 285, 295, 522–24; Bailey 1965). Their diction is not merely ornamental: panegyric is the genre in which royal authority, Buddhist legitimacy, and the management of external relations are rhetorically fused. This is also where the broader argument about borderland soft power becomes most visible: Khotanese participates in transregional models of Buddhist kingship, but it does so through a locally controlled medium that can project legitimacy outward while organizing loyalty and institutional standing inward.

The same pragmatic embeddedness is visible in prose genres. The Itinerary (Skjærvø 2003b, pp. 524–26) maps a southbound route through Gilgit and Kashmir with a density of place-names, distances, and institutional markers that reads as travel knowledge made textual.

Medical writing extends this practical horizon into the domain of expertise. Rooted in the Indian āyurvedic tradition that spread with Buddhism, they attest to the practical institutionalization of medical knowledge in a local Tarim language. Alongside fragmentary materials (see Luzzietti 2023), two substantial Late Khotanese prose translations survive: the Jīvakapustaka (Konow 1941)12 and the Siddhasāra (Bailey 1945, pp. 1–104).

The Jīvakapustaka, preserved in a single tenth-century Dunhuang manuscript, is transmitted in a bilingual Sanskrit–Khotanese format. Framed as instruction given to the physician Jīvaka, it compiles ninety-three prescriptions drawn from multiple sources and organizes them systematically by preparation (antidotes, butter-based, oil-based, and powdered medicines). Each recipe combines instructions for preparation with indications for use, underscoring the text’s practical orientation. Although the Khotanese translation occasionally reflects difficulties posed by the error-prone Sanskrit text, it also stabilizes and clarifies technical content, with many prescriptions identifiable in Indian medical literature. The bilingual format thus documents not a simple source–target hierarchy, but Khotanese functioning as a working medium for the organization and application of therapeutic knowledge.

The Khotanese Siddhasāra, by contrast, was translated via Tibetan mediation, though the translator consulted the Sanskrit original and corrected errors. Unlike the Jīvakapustaka, it integrates extensive theoretical discussion—humoral theory, dietetics, pharmacology, and disease classification—alongside practical treatment. Its verse introduction makes explicit why translation was necessary: reliance on collections of prescriptions without theoretical understanding had led to misdiagnosis and death. Translation is here framed as a corrective intervention aimed at restoring conceptual clarity and ensuring effective care.

Together, these texts exemplify translation as applied knowledge transfer rather than participation in the prestige economy of sūtra translation. They show Khotanese operating as a pragmatic language of expertise and institutional responsibility, through which medical knowledge was domesticated, regulated, and made operational within the local community—further reinforcing the paper’s argument that Khotanese functioned as a fully embedded medium of social practice in the Tarim Basin.

A final cluster of bilingual texts makes the multilingual ecology of late Khotanese communities visible in miniature. An Old Turkish–Khotanese (Emmerick and Róna-Tas 1992) word list targets bodily terms and the technical lexicon of archery and horse equipment, plausibly tied to military instruction; Chinese phrases written in Brāhmī with Khotanese renderings suggest everyday language acquisition in a Dunhuang setting;13 and a Sanskrit–Khotanese conversation manual (Kumamoto 1988)—structured as practical dialog, including the question of whether one owns sūtra, abhidharma, vinaya, or vajrayāna books—shows language learning embedded directly in religious and institutional life. Taken together, these materials confirm the broader contention of the paper: in the Tarim borderlands, Khotanese operated as a medium and modality of social organization, cultural prestige, and pragmatic coordination, continuously negotiating between local needs and transregional formations of knowledge, practice, and power.

Taken together, these non-doctrinal genres show that Khotanese was already functioning as a language of courtly representation, practical expertise, transregional mobility, and everyday communication. From this, it follows that Buddhism was embedded in a literate system capable of recording, coordinating, and regulating social life. It is therefore no coincidence that the same linguistic and scribal infrastructure that sustained poetry, panegyric, medicine, and bilingual exchange also underpinned a substantial body of administrative and legal documentation. These texts record land sales, slave purchases, tax orders, military rosters, petitions, and emergency evacuations, showing that Khotanese was routinely used to manage people, property, and security across the kingdom.

Recent work on these materials has demonstrated that they are not random survivals but parts of coherent administrative archives, often linked to identifiable officials, towns, and years. The reconstruction of six such archives, ranging from the mid-eighth to the early ninth century, shows that Khotan operated a stratified bureaucracy extending from the royal court down to prefectures, townships, and villages, with Khotanese, Chinese, and Tibetan titles coexisting in a single institutional ecology. Orders issued by the king, relayed by prefects, and implemented by local officials and patrols document a regime of taxation, conscription, grain storage, and military preparedness that continued largely unchanged even after Tibetan overlordship replaced Tang authority (Zhang 2018).

Seen in this light, the literary and documentary materials belong to a continuum of textual economy. The same language that served to articulate Buddhist doctrine, royal eulogy, lyric expression, and medical knowledge also recorded land transfers, tax obligations, labor duties, and military organization. Khotanese was therefore not confined to religious or cultural communication but functioned as an administrative and legal medium embedded in the everyday operation of the Khotan polity. This convergence of literary and bureaucratic uses is central to the argument of this paper: Khotanese operated as a borderland language, mediating between Buddhist transregional networks, imperial power structures, and local social practice.

Khotanese literature is therefore not a peripheral byproduct of Silk Road transmission, but a primary body of evidence for how the eastern Silk Road functioned as a space of sustained cultural and textual production. Its defining features are formal hybridity, recombinatory composition, and close integration with political, ritual, and administrative institutions. It constitutes a local literary formation shaped within and shaped by Silk Road Buddhism—one that both participated in transregional knowledge circulation and anchored Khotan’s own cultural and institutional identity within that wider field. The result is a corpus that not only reflects religious devotion but also attests to the epistemic agency of the Khotanese people in cultivating and sustaining a distinctive regional identity within the broader context of the Tarim interaction zone.

The multilingual conditions of the Tarim Basin are inscribed directly in the manuscripts. Interlinear glosses, occasional code-switching, and bilingual formats show that Khotanese writing was produced in continuous contact with textual traditions of other languages. Translation and transmission in this setting were a routine mode of textual production, through which materials of various provenance were adapted to shifting audiences and settings. Such textual practices positioned Khotanese translators and monks as both custodians of doctrinal normativity and agents of cultural innovation, mediating between imported canonical materials, local traditions, and transregional environs.

4.3. Transmission

The transmission of Buddhist texts along the Silk Road was never merely a question of transduction; it was a structured social practice involving institutions, networks, and identifiable agents. Khotanese, as a language of Buddhist transmission, emerged not only as a product of external influences but as an active conduit through which texts, rituals, and doctrines were reconfigured and disseminated across Central Asia. Central to this process were Khotanese-speaking translators—some of whose names and affiliations are preserved in colophons, Chinese records, and Tibetan hagiographies—who played a key role in shaping the textual and doctrinal landscape of the Silk Road.

Khotan was uniquely positioned to mediate Buddhist transmission between South, Central Asia, and China. With institutional support from the royal court and a flourishing monastic culture, it developed into a major center for translation activity from the fifth century onward. While early translation efforts were oriented toward adapting Indic texts into Khotanese for use within local monastic and lay communities, the role of Khotanese speakers soon expanded into outward-facing networks of transmission, especially toward China and Tibet. The evidence shows that Khotanese translators were not only recipients of texts but also agents in their further dissemination and reinterpretation.

One of the most significant examples of this dynamic is the Buddhāvataṃsaka Sūtra, a text closely associated with the cult of Mañjuśrī. The compilation of the Buddhāvataṃsaka Sūtra itself may have had strong roots in the Tarim Basin. Indic śāstra literature often refers to sections of it as independent texts, without indicating their inclusion in a unified scripture. The title Buddhāvataṃsaka Sūtra appears in Sanskrit colophons, but the content of these early versions likely differed from the extant Chinese or Tibetan editions—suggesting that multiple localized recensions may have circulated, possibly including ones compiled or preserved in Khotan.

That Khotan played a central role in this tradition is confirmed by the historical episode of Zhi Faling 支法領, who, around the early 5th century, traveled to Khotan to obtain a copy of the Buddhāvataṃsaka Sūtra. The text was so revered that it was not allowed to be taken out of the kingdom, but after persistent requests, the king gifted him a version comprising 36,000 ślokas. Zhi Faling entrusted this manuscript to the monk Buddhabhadra 佛陀跋陀罗, who between 418 and 420 CE produced the first Chinese translation in 60 fascicles—the earliest full Chinese version of this foundational Mahāyāna text (Hamar 2020, pp. 352–56).

Later, during the reign of Empress Wu Zetian 武则天 (624–705), the Khotanese translator Devendraprajñā 提雲般若 rose to prominence. Arriving in Luoyang around 688, shortly after the death of the great Indian monk Divākara (a.k.a. Rizhao 日照; 613–687), Devendraprajñā was entrusted with key translation projects at the Weiguo Dongsi 魏国东寺, the imperial translation bureau. He first translated two short scriptures—the Section on the Inconceivable Buddha-realm (T 300.10) and the Section on the Cultivation of Loving Kindness (T 306.10)—both affiliated with the Buddhāvataṃsaka Sūtra tradition but not included in the canonical eighty-fascicle version. These texts, preserved in both Chinese and recently discovered Khotanese versions, attest to Khotan’s local canon and the courtly support it received (A. Forte 1979, pp. 296–97).

Devendraprajñā’s influence extended beyond translation. The Prophecy of the Li Country records that the Khotanese king Viśya Väkrraṃ, who had accompanied his father Viśya Saṃgrrāmä to China in 674 and was enthroned in Khotan in 692, built a monastery in his honor. This patronage underscores the interconnectedness of Buddhist scholarship, court politics, and cultural memory between Luoyang and Khotan (Emmerick 1967, pp. 1–75).

Empress Wu Zetian’s interest in the Buddhāvataṃsaka Sūtra culminated in the commissioning of a new translation project, which brought another Khotanese monk, Śikṣānanda 实叉难陀 (652–710), to the imperial court. The more complete Sanskrit manuscript he brought—possibly with official sanction from the Khotanese royal family—formed the basis of the 80-fascicle version of the Chinese Buddhāvataṃsaka Sūtra, completed in 699 at the Foshouji 佛授记 monastery with a preface by the empress herself. The final text was significantly longer than Buddhabhadra’s earlier 60-fascicle version (Hamar 2021).

Wu Zetian’s strong support of these Central Asian translators—especially Śikṣānanda—reveals not only the political pragmatism of her religious policies but also her ideological affinity with the Mahāyāna currents flowing through Khotan (Siu 2024, pp. 43–47).

In sum, Khotanese translators played a vital role in the transformation of Buddhist doctrine along the Silk Road. Their activity—anchored in royal patronage, manuscript circulation, and interregional pilgrimage—established Khotan as both a transmitter and a creative center of Mahāyāna thought. Their legacy is embedded not just in Chinese translations, but in the very structures of transregional Buddhist scholasticism that spanned from South Asia to China, Tibet, and beyond.

Taken together, the evidence suggests that Khotanese-speaking translators and scribes operated at the center of a multilingual infrastructure that sustained the flow of Buddhist texts across Central Asia. Their role was not passive or peripheral, but central to the mechanisms by which textual authority was transmitted and localized. The prominence of Khotanese monks in Chinese and Tibetan translation teams, the presence of Khotanese Buddhist vocabulary in Tocharian, and the wide geographical distribution of Khotanese manuscripts all testify to the active engagement of Khotanese agents along the Silk Road.

In this context, the borderland character of the Khotanese language lies in the historical agency of its speakers. As translators, scribes, and doctrinal specialists, Khotanese Buddhists helped forge the communicative circuits through which Indic Buddhist texts were rendered intelligible across linguistic and cultural spheres. Their efforts made Khotanese into more than a regional variety: it became a functional node in the transmission of Buddhist knowledge, capable of mediating between South Asian origin, Central Asian form, and East Asian reception.

5. Conclusions

The preceding analysis has framed Khotanese as a paradigmatic language of the Tarim borderlands, operating within a regional field in which influence and authority were exercised through cultural prestige. Embedded in courtly sponsorship, monastic institutions, and transregional connectivities, Khotanese emerged as both a medium and a modality of cultural exchange. Its role within the Tarim Basin’s linguistic ecology underscores the active participation of local polities in shaping the communicative infrastructure of the eastern Silk Road cultural transmission.

Three interrelated dimensions have guided this reassessment. First, the historical embeddedness of Khotanese within the complex multilingual environment of the eastern Silk Road reveals patterns of institutional support for its use as a medium of local and transregional agency. Far from being an epiphenomenon of imperial diffusion, the emergence of Khotanese as a scriptural and liturgical language reflects the recurrent involvement of the Khotanese court and its monastic circles in appropriating and localizing South Asian and Chinese models.

Second, the literary corpus—centered on but not reducible to Buddhist themes—demonstrates a high degree of compositional autonomy, marked by innovative forms, intertextual recombination, and a vernacular inflection of Buddhist discourse. Khotanese literature is thus best understood not as derivative transmission but as a site of cultural production in its own right that integrated parts of the local poetical tradition and tastes with Buddhist doctrines in a manner tailored to regional epistemic frameworks.

Third, the function of Khotanese as a language of translation and mediation attests to its embeddedness within broader networks of knowledge circulation. Khotanese monks and translators were agents in the movement of texts, concepts, and cultural repertoires across the Buddhist ecumene of the eastern Silk Road.

The linguistic and textual record of the Tarim Basin—remarkable in both depth and breadth—has long remained only partially accessible, and in a wider interdisciplinary scholarly context, the internal perspectives of these eastern Silk Road borderlands during their heyday are still only beginning to be comprehensively tapped. Until quite recently, direct knowledge of the languages that emerged from this region at the height of its connectivity in the first millennium CE was highly restricted and largely confined to a small circle of specialists. This situation is gradually changing through new accessible critical editions14, the deployment of digital tools15 and repositories16, increasingly sustained cross-disciplinary and interdisciplinary collaboration (see, e.g., Malzahn and Zin 2025), and the growing integration of research infrastructures (see, e.g., Fellner and Radisch 2025). As access widens, the Tarim region increasingly reveals itself as a unique setting for examining the long-term dynamics of imperial borderlands: spaces characterized by the relative persistence of local diversities and identities, and by their role not merely as intermediaries, but as media and mediators in their own right.

Within this broader perspective, the case of Khotanese is particularly instructive. It exemplifies the productive tension that defines imperial margins: the interplay between locally grounded spheres of meaning and participation in wider, shared cultural imaginaries circulating across Eurasian networks. Taken together, these dimensions compel a revaluation of Khotanese as a language that both indexed and enacted the conditions of its historical moment. Its trajectory illustrates how, under specific institutional and cultural configurations, a language could function simultaneously as a medium of regional identity and as a vector of translocal networks.

As research on Khotanese and related Tarim languages becomes more accessible and integrated into comparative inquiry, their relevance extends far beyond linguistics, philology, Buddhist studies, or the regional history of the eastern Silk Road. They offer critical insight into how languages in imperial borderlands operate as media of cultural articulation, institutional negotiation, and political self-assertion within globalizing historical processes. The gradual broadening of engagement with this multilingual heritage therefore promises to benefit a range of disciplines while significantly deepening our understanding of the linguistic ecology and historical development of the Silk Road’s borderland worlds.