Exploring Four Block-Printed Indic Script Mahāpratisarā Dhāraṇī (Chinese: 大随求陀罗尼) Amulets Discovered in China

1. Introduction

Dhāraṇī maṇḍala amulets, those circular or square ritual objects inscribed with Indic scripts, have long been marginalised in Buddhist textual studies. Regarded more as protective charms than textual witnesses, they have largely escaped serious scholarly attention—particularly in terms of their Sanskrit content. Perhaps because these amulets were not typically intended for reading or chanting, or because their makers in medieval China often lacked any Indic literacy, scholars have neglected their textual significance. Instead, their focus has rested on the canonical dhāraṇī texts preserved in palm-leaf or birch-bark manuscripts, leaving these printed artefacts to be discussed primarily in art-historical or devotional contexts. Yet, this study reveals that many of these xylographic amulets preserve some of the earliest and most valuable versions of Buddhist dhāraṇī texts. By comparing recensions preserved on these artefacts, we can track evolving textual variants, script usage, and ritual priorities that might otherwise remain invisible.

This research examines four Indic-script Mahāpratisarā Dhāraṇī amulets from late Tang (9^th–10^th centuries) to early Song (10^th–11^th centuries) China. While they have previously been studied collectively in terms of iconography or devotional practices, their Indic-script content—whether in Siddhaṃ, Nāgarī, or hybrid scripts—has not been analysed with linguistic rigour. This project treats these amulets not only as material culture but also as legitimate Sanskrit textual witnesses. It introduces transliterations, comparative analyses, and interpretations of dhāraṇī variants, contributing to the growing field of editorial studies of Buddhist incantation literature.

A key argument of this project is that block-printed dhāraṇī maṇḍala amulets, especially those produced in China with little to no Indic literacy, nonetheless preserve early versions of Buddhist texts now lost in manuscript form. Though these amulets were not created for reading or chanting, and were often carved by printers unfamiliar with Sanskrit, their fine workmanship and faithful reproduction of older materials make them unexpectedly valuable textual witnesses. Through detailed analysis of mantra structures, seed syllable arrangements, script use, and visual layout, this study repositions these artefacts at the centre of Buddhist textual transmission. Some reveal script styles, textual recensions, or mantric elements found nowhere else, pointing to lost lines of transmission distinct from canonical manuscripts in Nepal or Tibet.

In doing so, this study also contributes to our understanding of medieval Sino-Indian cultural exchange. The appearance of up-to-date Nāgarī script, previously thought absent from Chinese Buddhist materials, reveals that Chinese Buddhists had access to—or at least knowledge of—contemporary South Asian textual styles. This points to ongoing scribal exchanges and devotional interactions between India and China during the late first millennium CE.

By foregrounding the editorial potential of these amulets, this project reimagines their role within Buddhist textual history. They are not simply ritual accessories, but dynamic records of textual adaptation, visual creativity, and cross-cultural religious transmission.

The following chapters are structured as follows: Chapter 2 provides the necessary historical and religious background, including the development of dhāraṇī literature in Indic Mahāyāna and its transmission to China, as well as the evolution of printing and amulet production in the medieval period. Chapter 3 presents four case studies—amulets commissioned by Xu Yin, Li Zhi-Shun, the Hangzhou National Archives, and Ruiguang Si—offering in-depth textual and material analysis. This chapter forms the core of the research, showcasing how close philological study of these artefacts can illuminate broader networks of Buddhist textual transmission.

2. Background

2.1. Dhāraṇī in Indic Mahāyāna

Dhāraṇīs are texts that emerged within Indic Mahāyāna from the first century CE and were transmitted during the earliest stages of the spread of Buddhism to China. Dhāraṇī are variously interpreted as symbolic or codes to significant teachings or qualities of the Buddha or, in contrast, as nonsensical phrases to encapsulate the ineffability of reality.

Dhāraṇī is the name for a phrase or an element of text found in Mahāyāna Buddhist literature, believed to bestow some form of support on the path, access to broader teachings represented in the phrase, or of protection or power. The dhāraṇī may be relatively short, consisting of a number of syllables, or quite long, often equal to a section within a sūtra. It may stand on its own or be found within a larger text or, if extensive, form a complete text in its own right, the equivalent of several pages long (around a few hundred short sentences or phrases).

The body of a longer dhāraṇī text is often divided into padas (smaller sections) of short phrases, invocations, and mantras (spells).1 In brief, dhāraṇīs usually consist of a list of feminine vocatives embedded with some series of literally incomprehensible yet potent syllables.

The use of incantation of mantra and dhāraṇī existed more broadly in Indic religion and became familiar within the Buddhist monastic community since at least the first century of the common era.2 This incantational function of dhāraṇī is a feature that continues from its Indic origin into East Asia. While similar in function to mantras, the latter are shorter and usually preserved in the formula that begins with ‘oṃ’ and sometimes ends with ‘svāhā,’ neither of which has a clear discursive meaning.3 Some dhāraṇī have known names. The one that concerns us most here, the Mahāpratisarā dhāraṇī, is found on amulets discovered in China, which also usually contain mantras written immediately after the dhāraṇī proper.

As noted above, dhāraṇī can also be the focus or even content of entire texts—dhāraṇī sūtras (incantation scriptures). These sūtras are sometimes named with a long title that encompasses one or more terms denoting power, such as dhāraṇī, mantra, and vidyā; other terms like kalpa, pratyaṅgirā, and sūtra are also applied.4

Given their mantric components and functions, dhāraṇī sūtras, such as the Mahāpratisarā dhāraṇī sūtra, are attributed to a new Buddhist soteriological path within the Mahāyāna tradition during the first half of the first millennium—the Mantranaya (Method of Mantras). This is an early stage of what we now call tantric Buddhism in general, which received its name for its unique employment of mantras and other potent linguistic phrases for soteriological and practical purposes.5 Although modern scholars usually refer to this school as Mantrayāna, it is a name received far later than its actual establishment. This study uses the name Mantranaya for chronological clarity, for it occurs earlier than the seventh-century appearance of the Vajrayāna (the Path of Diamond or the Diamond Way).6

2.2. Mahāpratisarā and Transmission (Textual History and Iconography)

The extant versions of the Mahāpratisarā dhāraṇī sūtra usually include two dhāraṇīs, four mantras, an introduction to the background and cause of the nidāna (the frame story), a frame story of the Buddha teaching the dhāraṇīs, two kalpas (ritual manuals) for each dhāraṇī, instructions for amulet-making and healing, general sections enumerating the anuśaṃsāḥ (various benefits), and nine narratives about how the dhāraṇīs were used in the past. Among these nine narratives, six seem to be original works of this scripture, while the second, third, and seventh can be found in earlier Buddhist literature.7

The Mahāpratisarā dhāraṇī sūtra is composed probably no later than the sixth century, instructing the creation of a protective and wish-fulfilling amulet through the writing down of the Mahāpratisarā dhāraṇī with provided rituals.8 This sūtra was incorporated into a collection of five dhāraṇī sūtras called the Pañcarakṣā (Five Protections) some centuries after its composition.

The first dhāraṇī and mantras from the Mahāpratisarā Dhāraṇī Sūtra are often inscribed on physical objects to empower the object and create ‘amulets.’ For the sake of convenience, I shall refer to this dhāraṇī as the Mahāpratisarā Dhāraṇī in this investigation. These amulets may be in the form of paper, silk, bracelets, armlets, necklaces, and bricks, for example.

Baosiwei (Chinese: 寶思惟, d. 721)9 was a Kashmiri monk who arrived in Luoyang in 693 CE and became the first person to translate the Mahāpratisarā Dhāraṇī Sūtra into Chinese.10 His translation represents an earlier phase of the text’s transmission into East Asia. It was almost forty years later that Vajrabodhi from South India discovered the ‘incompleteness’ of this edition.11 Later, in the mid-eighth century, Amoghavajra (Chinese: 不空, fl. 705–774), a key figure in the development of esoteric Buddhism in China, produced another translation.12 Amoghavajra was one of the most influential Buddhist monks of the Tang dynasty, known for his efforts in transmitting tantric Buddhist texts from South Asia and translating them into Chinese. His translations of dhāraṇī literature, including the Mahāpratisarā Dhāraṇī Sūtra, played a crucial role in shaping the esoteric Buddhist practices of the period. The widely applicable potency of this particular dhāraṇī informed its prestige and popularity around Asia and, in turn, has led to it being explored by scholars from different fields.

In most Mahāpratisarā Dhāraṇī Sūtra manuscripts dated after the sixth century, after introducing and telling the frame story, the Buddha describes the benefits of using the Mahāpratisarā Dhāraṇī.13 Then, the Buddha teaches the first kalpa (ritual), which at the center of it introduces the first dhāraṇī. Nine narratives are applied after this kalpa to demonstrate the efficacy of the dhāraṇī in various circumstances. Between the seventh and eighth narratives, there are the four mantras. Following this is a demonstration of how to make an amulet with this dhāraṇī, which ends the first part of the scripture. The second part follows the same order as the first, beginning with the enumeration of benefits when using the second dhāraṇī, then the second kalpa with dhāraṇī, further description of benefits regarding the narratives in the first part, and finally, an instruction on how to perform a healing ritual with the recitation of the second dhāraṇī.14

In addition to this general structuring of the content, one should notice a significant change and expansion to the scripture at the end of the seventh and start of the eighth centuries. Hidas points out that the scripture probably had only one kalpa, and therefore one dhāraṇī, when first compiled. One piece of evidence is Baosiwei’s 693 CE Chinese translation, which includes only the first dhāraṇī and kalpa. This translation further suggests that there had been refinements of the sūtra in the late seventh century to integrate it with Mantranaya, for its geographical location in the nidāna is still the historical Gṛdhrakūṭa instead of the later Mantranaya setting’s supernatural mountain, the Great Vajrameru, which appears to be in Amoghavajra’s translation.15

The Mahāpratisarā Dhāraṇī Sūtra was also translated into Tibetan, Uigurian,16 and Mongolian from the eighth century onwards.17 Most recently, Hidas published an annotated English translation of the entire sūtra in the same book where he puts the critical Sanskrit edition. This article primarily consults Hidas’ English translation.18

2.3. Dhāraṇī in China and the Development of Printing

Regardless of their origins, dhāraṇī were ascribed protective powers and became integral to various aspects of Chinese Buddhist practice. In China, they were used for longevity, virtue, power, wealth, warding off danger, ensuring a favourable afterlife, and aiding spiritual progress. To harness these benefits, dhāraṇī were presented in visual form as amulets, which were inscribed on various materials such as paper, silk, and bricks. By the late Tang, the widespread use of printing enabled the mass production of dhāraṇī in printed form, leading to the dhāraṇī amulets examined in this exploration.

Beyond their amuletic function, dhāraṇī, in their written form, were also regarded as relics of the Buddha or his teachings. Copies of dhāraṇī have been discovered in small stūpas (pagodas), a practice still observed in Buddhist communities today. This belief can be traced to South Asia, as Xuanzang (Chinese: 玄奘, fl. 602-664 CE) reported that people placed scripture fragments, referred to as ‘dharma relics,’ into stūpas in India.19 In China, this belief extended to the practice of writing dhāraṇī on banners.20

One of the most significant developments in the dissemination of dhāraṇī in China was their connection to the invention and spread of printing. The second year of Chángshòu (693 CE)21 was an eventful one: in the ninth lunar month, Empress Wu officially adopted the Buddhist title of Cakravartin (Chinese: 金輪聖神皇帝, "Divine Empress of the Golden Wheel").22 That same year, the Kashmiri monk Baosiwei arrived in Zhou and became the first person to translate the Sanskrit Mahāpratisarā Dhāraṇī Sūtra into Chinese in Luoyang,23 the political centre of Zhou.24

While the amulets explored here date from the late Tang, some two hundred years later, they emerged as a direct result of these earlier developments. On the one hand, xylographic amulets were identified as containing the Mahāpratisarā Dhāraṇī first translated by Baosiwei. On the other hand, they were shaped by the emergence of printing under Empress Wu and the belief in the power of sacred phrases. These beliefs included the idea that such phrases encapsulated the qualities of the Buddhas, were relics of the Buddha, ensured meritorious reproduction, and granted longevity and protection.

Some of the earliest printed works in existence are dhāraṇī, such as the Hyakumantō, produced during the mid-seventh century under Empress Shōtoku in Japan.25 Tim Barrett has suggested that the development of printing was influenced by various factors during the rule of Empress Wu Zhao (Chinese: 武曌, fl. 624-705 CE), including the belief in Buddhist sūtras as relics of the Buddha and the merit of reproducing texts for protection. Wu Zhao, the only female emperor26 in Chinese history, used Buddhist legitimacy to consolidate her rule, claiming to be both the reincarnation of Maitreya and a Cakravartin ruler.27 To emulate the Indian Cakravartin Aśoka, who was believed to have distributed 84,000 Buddhist relics, she initiated the mass copying of Buddhist scriptures.

Attracted by dhāraṇī sūtras for their claim to function 100,000 times more effectively than other scriptures, she promoted their translation and dissemination during her reign (690-705 CE).28 While printing would not experience a full resurgence until later in the Tang dynasty, Wu Zhao’s initiatives set a precedent for Buddhist printing, inspiring rulers such as Empress Shōtoku and influencing the production of dhāraṇī amulets.

In most studies of Mahāpratisarā Dhāraṇī amulets in China, these amulets are classified into three types based on their method of production: (1) hand-written, (2) partially hand-written and partially printed, and (3) fully printed.29 Table 1 at the end of this chapter details the discovery sites, colophons, devotee names, and sources of information where applicable. This classification provides insight into the historical development of amulet production alongside the evolution of printing technology in China.

While the potency of Buddhist phrases as relics and sources of power contributed to printing’s spread, all the amulets discussed here date no earlier than the eighth century, when printing had become widely established. Over time, a trend of increasing amulet size emerged towards the end of the tenth century. Examples range from an amulet associated with Madame Wei (21.5 × 21.5 cm) to that of a monk named Xingsi (44.5 × 44.3 cm), which will be discussed in the next section.30 Though precise dating remains uncertain for most amulets, the earliest ones are believed to date from the mid-eighth century, coinciding with the second translation of the Mahāpratisarā Dhāraṇī Sūtra.31 According to Copp, the earliest of these amulets is the one that belongs to Madame Wei.32

Hand-written and partially hand-written amulets tend to be more personalised than printed ones. They often contain devotee names and Buddhist terms in Chinese characters alongside Indic scripts, presumably identifying the owners.33 For instance, the silk hand-written dhāraṇī amulet associated with Madame Wei includes her name in the sixth circle from the centre (Figure 3).34 Similarly, the silk amulet of Jiao Tie-Tou features his name alongside phrases such as All Buddha’s Heart Spell (yīqiè fó xīn zhòu), ablution (guàn tāng),35 and formation of enclosure (jí jiè)36 (Figure 4).

Conversely, printed amulets were mass-produced and lacked such customisation. Their standardised scripts reflect broader trends and ‘fashion’ in Buddhist material culture. The colophons of some printed amulets reveal their commercialised distribution across China by the ninth century.37 The most prominent example is an amulet discovered in a tomb in Xi’an, which took place in Chengdu, some seven hundred kilometres from Xi’an.38 Ma connects this amulet with another Chengdu-printed Mahāpratisarā Dhāraṇī amulet excavated at Sichuan University. The similarities in location and formatting suggest a shared origin. Since the dating of the Chengdu-discovered amulet is confirmed to be late Tang (late ninth century), Ma argues that the Xi’an amulet was likely produced in the second half of the ninth century.39

Figure 1. Timeline of dhāraṇī.

Figure 1. Timeline of dhāraṇī.

Figure 2. Basic timeline of the Mahāpratisarā dhāraṇī.

Figure 2. Basic timeline of the Mahāpratisarā dhāraṇī.

Table 1. List of Indic Script Mahāpratisarā dhāraṇī Amulets Discovered in China42.

Place of Discovery	Container	Name of Devotee	Name of Carver	Current Location	Estimated Date of Creation	Handwritten or Printed?
Turfan (Turfan 72TAM188:5)	No case (covered on the corpse)	NA	NA	Turfan?	Early mid-eighth century?	Handwritten
Turfan (Turfan 72TAM189:13)	No case (covered on the corpse)	NA	NA	Turfan?	Early mid-eighth century?	Handwritten
Xi’an (Fenghao Rd 西凤路高洼)	Armlet of gold-enameled bronze, 1 cm in width, with copper box riveted to it, 4.5*2.4 cm	Jiao Tie-Tou	NA	Shaanxi Provincial Museum (陕西历史博物馆)	Late eighth century?	Handwritten
NA	NA	Madame Wei	NA	Yale Art Gallery	Ninth or tenth century	Handwritten
Xi’an (Diesel machine factory)	Arc-shaped copper pendant, 4.5*4.2 cm	Wu De [ _ ]	NA	Xi’an?	Ninth or tenth century?	Partially
Xi’an (Fenghe 冯河)	Copper tube, 4*1 cm	Jing Sitai	NA	NA	Mid/late eighth century?	Partially
NA. Previously owned by Jiuxitang	Copper container?	NA	NA	Hangzhou Branch of the National Archives of Publications and Culture (杭州国家版本馆)	Mid/late eighth century?	Printed
Sichuan University / Jin River, Chengdu	Silver armlet	NA	NA	National Museum of China (中国国家博物馆)	Late ninth (post 841) or very early tenth century.	Printed
Xi’an (Sanqiao 三桥镇)	Copper armlet, 9 cm in diameter, 1 cm in width.	Monk Shaozhen	NA	Shaanxi Provincial Museum (陕西历史博物馆)	Late ninth century	Printed
NA. Previously owned by Bodhi-nature / Shanghai auction?	Metallic container?	NA	NA	NA	Ninth or tenth century?	Printed
Luoyang	Small tube (dimensions and material unknown) found near ear of corpse.	Xu Yin, Monk Zhi Yi	Shi Hong-Zhen	Luoyang Cultural Relics Work Team (洛阳文物工作队)	926	Printed
Mogao Cave 17, Dunhuang	No case	Li Zhi-Shun	Wang Wen-Zhao	British Museum and Musée national des Arts asiatiques - Guimet (Pelliot Collection)	980	Printed
Mogao Cave 17, Dunhuang	No case	Yang Fa	NA	Musée national des Arts asiatiques - Guimet (Pelliot Collection)	Late tenth century?	Printed
Ruiguang Si, Suzhou	Found in small pillar inside stūpa	Monk Xiu Zhang	NA	Suzhou Museum (苏州博物馆)	1005	Printed
Xiasha Outlet Gallery Museum	NA	NA	NA	Xiasha Outlet Gallery Museum	Ninth or tenth century?	Printed
Xiasha Outlet Gallery Museum	NA	NA	NA	Xiasha Outlet Gallery Museum	Ninth or tenth century?	Printed
Xiasha Outlet Gallery Museum	NA	NA	NA	Xiasha Outlet Gallery Museum	Ninth or tenth century?	Printed
Xiasha Outlet Gallery Museum	NA	NA	NA	Xiasha Outlet Gallery Museum	Ninth or tenth century?	Printed
Xiasha Outlet Gallery Museum	NA	NA	NA	Xiasha Outlet Gallery Museum	Ninth or tenth century?	Printed
Xiasha Outlet Gallery Museum	NA	NA	NA	Xiasha Outlet Gallery Museum	Ninth or tenth century?	Printed
Xiasha Outlet Gallery Museum	NA	NA	NA	Xiasha Outlet Gallery Museum	Ninth or tenth century?	Printed
Xiasha Outlet Gallery Museum	NA	NA	NA	Xiasha Outlet Gallery Museum	Ninth or tenth century?	Printed
Xiasha Outlet Gallery Museum	NA	NA	NA	Xiasha Outlet Gallery Museum	Ninth or tenth century?	Printed

Table 1. List of Indic Script Mahāpratisarā dhāraṇī Amulets Discovered in China42.

Place of Discovery	Container	Name of Devotee	Name of Carver	Current Location	Estimated Date of Creation	Handwritten or Printed?
Turfan (Turfan 72TAM188:5)	No case (covered on the corpse)	NA	NA	Turfan?	Early mid-eighth century?	Handwritten
Turfan (Turfan 72TAM189:13)	No case (covered on the corpse)	NA	NA	Turfan?	Early mid-eighth century?	Handwritten
Xi’an (Fenghao Rd 西凤路高洼)	Armlet of gold-enameled bronze, 1 cm in width, with copper box riveted to it, 4.5*2.4 cm	Jiao Tie-Tou	NA	Shaanxi Provincial Museum (陕西历史博物馆)	Late eighth century?	Handwritten
NA	NA	Madame Wei	NA	Yale Art Gallery	Ninth or tenth century	Handwritten
Xi’an (Diesel machine factory)	Arc-shaped copper pendant, 4.5*4.2 cm	Wu De [ _ ]	NA	Xi’an?	Ninth or tenth century?	Partially
Xi’an (Fenghe 冯河)	Copper tube, 4*1 cm	Jing Sitai	NA	NA	Mid/late eighth century?	Partially
NA. Previously owned by Jiuxitang	Copper container?	NA	NA	Hangzhou Branch of the National Archives of Publications and Culture (杭州国家版本馆)	Mid/late eighth century?	Printed
Sichuan University / Jin River, Chengdu	Silver armlet	NA	NA	National Museum of China (中国国家博物馆)	Late ninth (post 841) or very early tenth century.	Printed
Xi’an (Sanqiao 三桥镇)	Copper armlet, 9 cm in diameter, 1 cm in width.	Monk Shaozhen	NA	Shaanxi Provincial Museum (陕西历史博物馆)	Late ninth century	Printed
NA. Previously owned by Bodhi-nature / Shanghai auction?	Metallic container?	NA	NA	NA	Ninth or tenth century?	Printed
Luoyang	Small tube (dimensions and material unknown) found near ear of corpse.	Xu Yin, Monk Zhi Yi	Shi Hong-Zhen	Luoyang Cultural Relics Work Team (洛阳文物工作队)	926	Printed
Mogao Cave 17, Dunhuang	No case	Li Zhi-Shun	Wang Wen-Zhao	British Museum and Musée national des Arts asiatiques - Guimet (Pelliot Collection)	980	Printed
Mogao Cave 17, Dunhuang	No case	Yang Fa	NA	Musée national des Arts asiatiques - Guimet (Pelliot Collection)	Late tenth century?	Printed
Ruiguang Si, Suzhou	Found in small pillar inside stūpa	Monk Xiu Zhang	NA	Suzhou Museum (苏州博物馆)	1005	Printed
Xiasha Outlet Gallery Museum	NA	NA	NA	Xiasha Outlet Gallery Museum	Ninth or tenth century?	Printed
Xiasha Outlet Gallery Museum	NA	NA	NA	Xiasha Outlet Gallery Museum	Ninth or tenth century?	Printed
Xiasha Outlet Gallery Museum	NA	NA	NA	Xiasha Outlet Gallery Museum	Ninth or tenth century?	Printed
Xiasha Outlet Gallery Museum	NA	NA	NA	Xiasha Outlet Gallery Museum	Ninth or tenth century?	Printed
Xiasha Outlet Gallery Museum	NA	NA	NA	Xiasha Outlet Gallery Museum	Ninth or tenth century?	Printed
Xiasha Outlet Gallery Museum	NA	NA	NA	Xiasha Outlet Gallery Museum	Ninth or tenth century?	Printed
Xiasha Outlet Gallery Museum	NA	NA	NA	Xiasha Outlet Gallery Museum	Ninth or tenth century?	Printed
Xiasha Outlet Gallery Museum	NA	NA	NA	Xiasha Outlet Gallery Museum	Ninth or tenth century?	Printed
Xiasha Outlet Gallery Museum	NA	NA	NA	Xiasha Outlet Gallery Museum	Ninth or tenth century?	Printed

Figure 3. Madame Wei (魏大娘)’s handwritten and painted silk, centre image Vajradhara empowering Wei, Mahāpratisarā dhāraṇī, mid-eighth century (743-758 CE), now in Yale University Art Gallery, The Hobart and Edward Small Moore Memorial Collection. After Shi-Chang Ma, "大隨求陀羅尼曼荼羅圖像的初步考察," Journal of Tang Studies no. 十 (10) (2004), 566.

Figure 3. Madame Wei (魏大娘)’s handwritten and painted silk, centre image Vajradhara empowering Wei, Mahāpratisarā dhāraṇī, mid-eighth century (743-758 CE), now in Yale University Art Gallery, The Hobart and Edward Small Moore Memorial Collection. After Shi-Chang Ma, "大隨求陀羅尼曼荼羅圖像的初步考察," Journal of Tang Studies no. 十 (10) (2004), 566.

Figure 4. Jiao Tie-Tou (焦鐵頭)’s hand-written silk, centre picture showing devotee kneeling beside the six-armed bodhisattva Mahāpratisarā, Mahāpratisarā dhāraṇī, mid-to-late Tang, now in Shaanxi Provincial Museum (shǎnxī lìshǐ bówùguǎn, 陝西歷史博物館). After Shi-Chang Ma, "大隨求陀羅尼曼荼羅圖像的初步考察," Journal of Tang Studies no. 十 (10) (2004), 565.

Figure 4. Jiao Tie-Tou (焦鐵頭)’s hand-written silk, centre picture showing devotee kneeling beside the six-armed bodhisattva Mahāpratisarā, Mahāpratisarā dhāraṇī, mid-to-late Tang, now in Shaanxi Provincial Museum (shǎnxī lìshǐ bówùguǎn, 陝西歷史博物館). After Shi-Chang Ma, "大隨求陀羅尼曼荼羅圖像的初步考察," Journal of Tang Studies no. 十 (10) (2004), 565.

Figure 5. Block print, eight-armed bodhisattva Mahāpratisarā, Mahāpratisarā dhāraṇī, printed in Longchi Fang (龍池坊, Longchi Alley) Chengdu, discovered in former Sichuan University near Jin River, in silver armlet, very late ninth century. After Shi-Chang Ma, "大隨求陀羅尼曼荼羅圖像的初步考察," Journal of Tang Studies no. 十 (10) (2004), 571.

Figure 5. Block print, eight-armed bodhisattva Mahāpratisarā, Mahāpratisarā dhāraṇī, printed in Longchi Fang (龍池坊, Longchi Alley) Chengdu, discovered in former Sichuan University near Jin River, in silver armlet, very late ninth century. After Shi-Chang Ma, "大隨求陀羅尼曼荼羅圖像的初步考察," Journal of Tang Studies no. 十 (10) (2004), 571.

Figure 6. Monk Shaozhen (比丘僧少貞)’s block print on silkworm cocoon paper (絹, juān), with handwritten name in the centre, eight-armed bodhisattva Mahāpratisarā, Mahāpratisarā dhāraṇī, printed in Baoen Si (報恩寺, Baoen Temple) Chengdu, discovered in Xi’an, late ninth century, in copper armlet. After Shi-Chang Ma, "大隨求陀羅尼曼荼羅圖像的初步考察," Journal of Tang Studies no. 十 (10) (2004), 572.

Figure 6. Monk Shaozhen (比丘僧少貞)’s block print on silkworm cocoon paper (絹, juān), with handwritten name in the centre, eight-armed bodhisattva Mahāpratisarā, Mahāpratisarā dhāraṇī, printed in Baoen Si (報恩寺, Baoen Temple) Chengdu, discovered in Xi’an, late ninth century, in copper armlet. After Shi-Chang Ma, "大隨求陀羅尼曼荼羅圖像的初步考察," Journal of Tang Studies no. 十 (10) (2004), 572.

2.4. Indic Script in Chinese Amulets

Modern approaches to these amulets usually focus on Chinese inscriptions, imageries, formatting, storage, and usage. The Indic script found on them, meanwhile, is usually neglected. In my study, I will focus on the Indic scripts in the printed amulets. Therefore, the rest of this article, in subsequent the subsequent chapter, centres on printed amulets, examining in detail four xylograph amulets from the late Tang to early Song periods. My four xylographs are:

1) Xu Yin (徐殷)’s amulet

2) Li Zhi-Shun (李知順)’s amulet

3) Hangzhou National Archives’ amulet

4) Ruiguang Si (瑞光寺) amulet

The exploration will begin with a general background introduction to each sample before analysing them with, if applicable, some transliteration and translation of their Indic script.

3. Case Studies

3.1. Introduction

The dhāraṇī amulets examined in this chapter provide a distinctive perspective on the interaction between Chinese Buddhist artisans, donors, and the broader Sanskrit manuscript tradition. Rather than indicating that these texts were intended for a Sanskrit-literate audience, their presence within these objects suggests a connection to Sanskrit textual traditions through reproduction and adaptation. Rather than focusing on script forms or transliteration, this study examines how these amulets preserve Sanskrit textual material, sometimes in imperfect yet revealing ways, offering alternative sources for studying Sanskrit dhāraṇīs outside traditional manuscript traditions.

A key objective of this study is to determine whether original Sanskrit texts, potentially lost in their manuscript form, might be embedded within these dhāraṇī amulets. Currently, scholars often rely on Nepalese manuscript evidence for studying Sanskrit Mahāpratisarā Dhāraṇī texts, yet these amulets present an alternative and possibly earlier source for reconstructing textual transmission. By identifying parallels between the Sanskrit content in these artefacts and known manuscript traditions, this chapter seeks to clarify how accurately Sanskrit was copied, how faithfully it was preserved, and whether these texts reflect local adaptations or a direct link to transregional Buddhist networks.

This investigation, therefore, moves beyond the question of who could read these texts and instead asks: What do these inscriptions reveal about the transmission of Buddhist dhāraṇīs? How do these textual artefacts challenge existing manuscript-based reconstructions of Sanskrit Buddhist texts? And to what extent do these amulets serve as material witnesses to Indic scriptural traditions that may no longer survive in manuscript form? Through a close reading of the amulets' textual components, this chapter reexamines the role of script, copying practices, and transmission networks in shaping Buddhist material culture during this period.

3.2. The Dhāraṇī Amulets as Textual Artefacts

The dhāraṇī amulets examined in this study offer critical insights into the transmission and adaptation of Sanskrit Buddhist texts in China. These artefacts, often discovered in reliquaries, caves, and private collections, contain intricate combinations of dhāraṇīs, bījākṣaras, and donor inscriptions that provide valuable context for their production and use. Across different examples, we observe the interplay of textual fidelity, artistic execution, and ritual intention, raising important questions about how these amulets were copied, transmitted, and perceived by their users.

One of the earliest examples analysed in this study was unearthed in Shijiawan, Luoyang, dating to 926–927 CE (Figure 7). This amulet was commissioned by Monk Zhiyi (僧知益) of Baoguo Temple, inscribed by Shi Hong-Zhan (石弘展), and ultimately acquired by Xu Yin (徐殷), whose handwritten addition dates to early 927 CE. The main body consists of a large dhāraṇī square on the right, surrounded by sixteen bījākṣaras, with a Chinese inscription running down the left-hand side of the main square. The dhāraṇī itself is rendered in Siddhaṃ script, whereas the Chinese inscription provides an explanatory passage, stating that writing down and wearing this dhāraṇī would eliminate all bad karma and protect the wearer from disasters, in accordance with the scriptures. This particular amulet is significant not only for its meticulous carving but also for the additional layer of textual engagement provided by Xu Yin’s personal annotation, which allows us to trace both its initial printing (May 21, 926 CE) and subsequent acquisition (February 12, 927 CE). 43 The careful arrangement of eight circular seals, seven square sections of Siddhaṃ script, and four offering bodhisattvas further reflects its role as an object of devotion and ritual efficacy.

One of the earliest examples analysed in this study is an amulet discovered in Mogao Cave 17, Dunhuang, dated to 980 CE (Figure 8). This artefact, commissioned by Li Zhi-Shun (李知順) and carved by Wang Wen-Zhao (王文沼), presents an intriguing combination of Siddhaṃ bījākṣaras and Nāgarī script dating to the 6th–8th centuries. Unlike earlier dhāraṇī amulets, which predominantly relied on Siddhaṃ as the standard script for rendering Sanskrit texts in Medieval China, this amulet demonstrates the presence of Nāgarī script as an alternative means of writing Indic syllables. The use of Nāgarī in a Chinese Buddhist context suggests that Siddhaṃ was not the only script employed for transcribing Sanskrit dhāraṇīs, highlighting the more diverse landscape of script usage in medieval China than previously assumed. Whether this reflects direct manuscript influence, an experimental variation, or the scribal choices of the block carver remains open to interpretation. However, this example contributes to a broader understanding of the multiplicity of Indic scripts in Chinese Buddhist textual culture.

A further example, now housed in the Hangzhou Branch of the National Archives of Publications and Culture, was originally acquired through auction by a private collector, Jin Liang (Figure 9).44 While its exact date remains uncertain, the museum has tentatively attributed it to the 8th–10th centuries.45 This amulet is largely illegible due to preservation issues and poor printing quality, making its Siddhaṃ script unrecognisable. However, another piece, believed to have been printed from the same woodblock, is now preserved in the Xiasha Outlet Gallery Museum. This second specimen is far clearer, with legible Siddhaṃ characters and a handwritten inscription at the bottom of the lotus seat.46 The handwritten addition expresses a wish to be reborn in the Tuṣita Realm of Maitreya, a goal that the Mahāpratisarā dhāraṇī promises to fulfil.47

The existence of two nearly identical prints from the same block suggests a larger production network for such amulets. While the Hangzhou-branch example is too faded for detailed textual analysis, its presumed twin in the Xiasha Outlet Gallery Museum provides key insights into both the textual and ritual dimensions of these objects. Despite uncertainties about its precise provenance, this case further illustrates the continued use of bījākṣaras and dhāraṇīs in private devotional practice, reinforcing their role in esoteric Buddhist traditions. Further research is needed to clarify its place within the broader landscape of dhāraṇī circulation and transmission in medieval China.

A compelling example of dhāraṇī amulets in medieval China comes from the Ruiguang Stūpa (瑞光寺塔) in Suzhou, dating to 1005 CE (Figure 10). Two copies of the Mahāpratisarā dhāraṇī were found in a reliquary on the third floor of the stūpa: one written in swirling Chinese calligraphy (dated 1001) and the other in 9^th-10^th century Nāgarī script (dated 1005).48 This latter example, carved by Monk Xiu Zhang (沙門秀璋) and donated by Geng[…]Wai (耿[…]外), presents a distinct composition. Instead of a standard bodhisattva icon or an image of devotees, its central figure is Tejaprabhā, the Effulgent Buddha, surrounded by nine planetary luminaries and twelve zodiac signs. By the time this amulet was created, Tejaprabhā had become part of the Buddhist pantheon and was regarded as a counterpart to the Daoist deity, the Great Emperor of the Arch of the North Pole of the Purple Enclosure (Ziwei Beiji Dadi).49

Scholars such as Eugene Wang, Katherine R. Tsiang, and Paul Copp have examined this artefact in detail, noting its fusion of Daoist cosmology and Buddhist ritual imagery. The association between Tejaprabhā and the Daoist deity Ziwei Beiji Dadi suggests an ongoing process of syncretic adaptation, in which Buddhist protective dhāraṇīs absorbed liturgical elements from Daoist astrological traditions.

Beyond its iconographic significance, the use of Nāgarī script in this amulet is particularly noteworthy. While earlier Chinese Buddhist materials primarily employed Siddhaṃ script, this amulet, together with Li Zhishun’s amulet, reflects a gradual transition toward alternative Indic scripts in 10^th-11^th century China. The presence of Nāgarī alongside Siddhaṃ suggests an increasing openness to new calligraphic traditions, possibly influenced by international Buddhist interactions. This shift may indicate broader developments in textual transmission, script adoption, and evolving scribal practices in Chinese Buddhist communities.

Across these case studies, dhāraṇī amulets functioned not only as ritual objects but also as markers of textual and artistic exchange. Whether meticulously copied, imprecisely carved, or infused with new iconographic elements, these amulets reflect the multifaceted processes of adaptation and transmission that shaped the material culture of esoteric Buddhism in medieval China.

3.3. Sanskrit Texts

Amulet of the Hangzhou Branch of the National Archives of Publications and Culture

The Hangzhou National Archive Museum labels this print as a Mahāpratisarā dhāraṇī. 50 Although explicit documentation supporting this identification is absent from the Archives’ catalogues, the attribution remains plausible given the estimated date of the amulet and the widespread tradition of creating and using Mahāpratisarā dhāraṇī amulets during the 8^th to 10^th century. Despite the severe damage preventing conclusive textual identification, comparison with the clearer Xiasha amulet—which, although not definitively confirmed to have used exactly the same woodblock, clearly contains readable Mahāpratisarā dhāraṇī text in intelligible Siddhaṃ script—strengthens the attribution of the Hangzhou amulet as part of this dhāraṇī tradition.51

Xu Yin’s Amulet (926–927 CE)

Xu Yin’s amulet features sixteen bījākṣaras (seed syllables) arranged symmetrically around the central dhāraṇī. Each bījākṣara is seated on blooming lotuses, positioned in groups of four along each side, and separated by five vajras. These syllables are often understood as esoteric syllabic particles that constitute mantras or symbolise Buddhist deities.52 The reading order follows the standard mantra structure, beginning with "oṁ" at the bottom left corner and continuing upwards along the square. The transliteration is as follows:

(left) oṁ aṃ trāṃ haḥ

(top) hrīḥ dhaṃ māṃ raṃ

(right) kaṃ haṃ hūṁ vaṁ

(bottom) aḥ jaḥ hoḥ saḥ

The bījākṣaras on this amulet appear to have been printed using a different woodblock from the dhāraṇī itself, or possibly added onto the woodblock after the lotuses were carved. Some of their strokes overlap onto the lotuses, suggesting a possible secondary layer of carving. This variation raises the question of whether these bījākṣaras served as ritual additions rather than being intrinsic to the amulet’s original design.

Intriguingly, the outermost layer of the amulet's dhāraṇī maṇḍala contains four complete mantras:

oṃ amṛtā vare vara vara pravara vipujre hūṃ hūṃ phaṭ phaṭ svāhā ||

oṃ amṛtā vilokini garbha saṃraḥkṣaṇi akaṣiṇi hūṃ hūṃ phaṭ phaṭ svāhā ||

oṃ vajrāddhāna hūṃ jaḥ oṃ vimarajaya varī amṛte hūṃ hūṃ hūṃ hūṃ phaṭ phaṭ phaṭ phaṭ svāhā ||

oṃ bhara bhara saṃbhara saṃbhara idriya visodhani hūṃ hūṃ rūrūcari svāhā ||53

Given that some sounds from these mantras, such as “aṃ” from “amṛtā” and “jaḥ” from the penultimate mantra, correspond closely to certain bījākṣaras, it is plausible—though speculative—that the sixteen bījākṣaras might represent abbreviated or symbolic forms of these four mantras. However, each bījākṣara could carry multiple meanings, and thus, identifying a direct correspondence between specific bījākṣaras and individual mantras remains uncertain without additional evidence.

Li Zhi-Shun’s Amulet (980 CE)

Li Zhi-Shun’s amulet, unlike Xu Yin’s, contains twelve bījākṣaras, distributed between its inner and outer frame boundaries. The transliteration is as follows:

Outer frame: oṁ hūṃ rī traḥ gaḥ hrīḥ aḥ aḥ

Inner frame: trāṁ gīḥ kṛṁ hoḥ

These bījākṣaras on this piece of amulet were first studied by Matsumoto Eiichi, who suggested that the ones in the inner boundary were the representatives of the four inner offering bodhisattvas,54 the four at the corners of the outer boundary stood for the outer offering bodhisattvas,55 and the four remaining ones at the middle of each side resemble designate the four Pāramitā (perfection) bodhisattvas.56 He also provided a diagram to show whom they symbolise (Figure 11). A translation of this diagram is shown in Figure 14. Besides this suggestion, Ma notes that the bījākṣaras in the inner boundary are emblements of Sìshēn púsà (the Four-body bodhisattva).57 This structure indicates that the bījākṣaras on this amulet were not merely syllabic components of a mantra but also symbolic tokens representing deities, reinforcing their role in esoteric visualisation and ritual practice.

Another notable feature is the Chinese transliteration of Sanskrit mantras found on the sides of Li Zhi-Shun’s amulet (980 CE). In addition to the symbolic arrangement of bījākṣaras, the Li Zhi-Shun amulet also contains Chinese-transliterated Sanskrit mantras that invoke the eight tools held by Mahāpratisarā, the central figure in this amulet’s design. The transliteration and translation of these Chinese-transliterated Sanskrit mantras are as follows:

Right side frame:

唵縛日羅二合 唵縛日羅二合娑縛二合 唵播奢 唵竭誐

‘oṃ vajra’ twice, ‘oṃ vajra’ twice and ‘sarva’ twice, ‘oṃ pāśa,’ ‘oṃ khaṅga’

Left side frame:

唵真多麽抳 唵作羯羅 唵底哩戍哩 唵摩賀尾

‘oṃ kaṇṭhamaṇi,’ ‘oṃ cakra,’ ‘oṃ trisula,’ ‘oṃ maghava’

These mantras are mostly assigned with the name of the powerful object written in Chinese characters before them (in order):

chǔ (club, Skt: vajra), fǔ (axe), suǒ (lasso, Skt: pāśa), jiàn (sword, Skt. khaṅga), bǎo (jewel, Skt. kaṇṭhamaṇi), lún (wheel, Skt. cakra), jǐ (spear, Skt. triśūla), jiá (folder).

This direct association between the mantras and the eight ritual tools, also held in each of the central figure’s eight hands on this amulet,58 reinforces the idea that Mahāpratisarā was not only invoked through the dhāraṇī but also through her attributes, connecting textual and visual elements in a structured ritual composition. These findings suggest that the bījākṣaras and mantras functioned as ritual markers within a carefully structured tantric framework, aligning them with the esoteric Buddhist practice of visualisation and deity invocation.

Comparison and Analysis of the Bījākṣaras

The sixteen bījākṣaras on Xu Yin’s amulet are placed only along the sides, while the twelve on Li Zhi-Shun’s amulet are arranged both on sides and corners. Their patterns differ significantly, sharing only their initial and final syllables. Moreover, the four outer offering bodhisattvas are already decorated in the four corners of the dhāraṇī maṇḍala.59 Therefore, Matsumoto's hypothesis linking the bījākṣaras on Li Zhi-Shun’s amulet to offering and Pāramitā bodhisattvas is unlikely to apply directly to Xu Yin’s amulet. Any direct connection would require assuming a rapid change in symbolism within the fifty-three-year gap separating these two amulets.

Nevertheless, both amulets clearly reflect the concept of ritual space construction as suggested in Baosiwei’s and Amoghavajra’s translations of the Mahāpratisarā dhāraṇī sūtra. Ma has identified a shift in iconography among Mahāpratisarā dhāraṇī amulets, noting that earlier Tang-era examples frequently featured mudrās and Buddhist power instruments on their borders, whereas later amulets like Xu Yin’s emphasise bījākṣaras instead. 60 The Tang amulets often included elements such as mudrās on their frames—for example, the Xi’an suburb fragments (eight or ten mudrās),61 Wu De’s amulet (ten mudrās), 62 the Metallurgy Works amulet (ten mudrās), 63 Jing Si-Tai’s amulet (four mudrās),64 and the Longchi Fang and Baoen Si fully printed amulet (eight mudrās).65 The bījākṣaras on Xu Yin’s and Li Zhi-Shun’s amulets could be seen as evolving from these earlier visual mudrās, potentially taking over their ritual function as 'seals' (印, yìn) or keys to the dhāraṇī-portal, as explicitly prescribed in the Mahāpratisarā dhāraṇī sūtra translations.66 The scripture instructs that multiple seals should be drawn around the mantra for an amulet to be ritually effective.67 Each amulet embodies a ritual altar, visually representing powerful instruments—such as vajras, lotuses, and various offerings—around the central deity. This visual design directly follows the scriptural instructions, effectively making the amulet a portable ritual space for personal devotion. Thus, the most plausible theory for the configuration of bījākṣaras on Xu Yin’s amulet suggests that it may stem from an evolution—from mudrās, serving as “seals” for the “dhāraṇī-portal” to control its open and close, to the bījākṣaras, the need of “seals” on each side of the amulet which form the mantras and dhāraṇīs that bridge the profound knowledge, a power received from a new interpretation of Amoghavajra’s translation.

The Ruiguang Si Amulet (1005 CE)

Unlike Xu Yin’s and Li Zhi-Shun’s amulets, the Ruiguang Si amulet (1005 CE) does not feature bījākṣaras but introduces a Nāgarī script inscription that closely resembles 9th–10th century Western Indian inscriptions, as classified by Singh.68 This resemblance suggests a possible textual transmission route from regions such as Vidarbha, which falls within this classification.69

A particularly intriguing aspect of the Ruiguang Si amulet is its final invocation, which differs significantly from earlier dhāraṇī amulets. It contains an invocation of Vairocana, a feature absent from Xu Yin’s amulet (926 CE), followed by a mysterious concluding phrase:

oṃ amogha-vairocana-mahāmudrā-padma-pravarttani hūṃ phaṭ svāhā

(Translation: oṃ, O the Jewel-lotus that is the Great Seal of the Unfailing Vairocana, turn over, hūṃ phaṭ svāhā.)

Additionally, the final mantra on the Ruiguang Si amulet includes the unusual phrase:

vidani sulekha

This phrase is particularly interesting because it does not follow standard mantric structures, which typically begin with “oṃ” and conclude with “hūṃ,” “phaṭ” or “svāhā.”70 The term "sulekha" is the Sanskrit word for "well-written," reinforcing the idea that this phrase may have been a scribe’s mark or a ritual affirmation of the dhāraṇī's scriptural power.

However, the meaning of "vidani" remains uncertain. One hypothesis is that it represents a proper noun, possibly the carver’s name. Alternatively, it could relate to Vidarbha, aligning with the Nāgarī script’s resemblance to 9^th–10^th-century Western Indian inscriptions.71 This hypothesis is strengthened by the fact that Nāgarī 'ni' and 'bhi' look very similar. If this was a scribal error, and the word should instead be read as "vidabhi," it would closely resemble the name "Vidarbha," a historical, geographical region in Western India.72

These amulets illustrate how dhāraṇī practices in medieval China were not static but adapted dynamically, incorporating new deity associations, evolving script choices, and developing distinct regional mantra traditions that persisted beyond China into Japanese esoteric practices. Whether through the invocation of Mahāpratisarā’s attributes, the adaptation of deity-associated syllables, or the integration of newly emerging scriptural formulas, these amulets bear witness to a complex process of ritual and textual reinvention.

Figure 11. The seating positions on the Mahāpratisarā dhāraṇī maṇḍala (of Li Zhi-Shun’s amulet). After Eiichi Matsumoto, Tonkōga no kenkyū, 1985, 606.

Figure 11. The seating positions on the Mahāpratisarā dhāraṇī maṇḍala (of Li Zhi-Shun’s amulet). After Eiichi Matsumoto, Tonkōga no kenkyū, 1985, 606.

Figure 12. Transliteration of the bījākṣaras on Xu Yin’s amulet.

Figure 12. Transliteration of the bījākṣaras on Xu Yin’s amulet.

Figure 13. Transliteration of the bījākṣaras on Li Zhi-Shun’s amulet.

Figure 13. Transliteration of the bījākṣaras on Li Zhi-Shun’s amulet.

3.4. The Role of Copying Practices in Buddhist Transmission

Li Zhi-Shun’s Amulet

The Li Zhi-Shun amulet is significant in two primary ways: first, for its refined printing technology and embellished illustrations; second, for the script of the written dhāraṇī itself. While every character is clearly visible, the script employed diverges considerably from standard Siddhaṃ, rendering it difficult to decipher. The brushwork suggests an influence from Chinese calligraphy—characterised by predominantly straight strokes with abrupt turns—contrasting with the typical carving patterns of Indic scripts. Despite the familiar iconography, this dhāraṇī does not match any known versions of the Mahāpratisarā dhāraṇī.

For example, an attempt at transliteration of the innermost circle of the dhāraṇī reveals significant uncertainties, marked here as [?]:

baddhabadhi||[?][?]tathā[ga?]toyalaga[ja?][?]pra[rro?][?][?]hārvāhāyalidhāga[ya?]bharāyayadhā[?]badhi|li[?][?]|[?][?][?][?]rākṣa[?][ya?]dhara[?]traṇigabha[?]ṣaṇi[ga?][rdha?][?][ga?]

The phrase preceding the first double daṇḍa “||” may be “buddha-bodhi,” while the word immediately following the daṇḍas is “tathāgata.” However, most Mahāpratisarā dhāraṇīs (see Hidas’ transliterations and the one recorded in T1153_.20.0632b01) begin with a veneration phrase including the term “namaḥ” (to bow, honor, or salute). If this salutation is missing, the subsequent phrase still does not appear in any known versions of the Mahāpratisarā dhāraṇī. Due to these uncertainties, further research is required to establish the textual basis of this dhāraṇī.

The Hangzhou Amulet

A key distinguishing feature of the Hangzhou amulet is the absence of a Chinese colophon, which is typically crucial for identifying printed dhāraṇīs. Despite its blurred state, the Siddhaṃ script provides two important insights regarding the structure and orientation of the dhāraṇī maṇḍala:

• Unconventional Script Orientation: The central circular maṇḍala follows an ‘inside-out’ writing pattern, whereas the surrounding square-shaped maṇḍalas are oriented ‘outside-in.’ This contrast in orientation is not observed in comparable dhāraṇī amulets, such as those by Xu Yin, Li Zhi-Shun, or those in the Pelliot Collection (MG 17689). However, the Xiasha amulet shares this exact script orientation, suggesting that the Hangzhou amulet may belong to the same production tradition or scribal lineage. This unique layout may indicate regional variation, a specific ritual function, or the stylistic preferences of the block carver.
• Esoteric Symbolism and Script Blur: The amulet maker may have viewed the entire dhāraṇī maṇḍala not as a readable script but as a talismanic image designed for divine interaction rather than human interpretation. The indistinct yet Siddhaṃ-resembling characters suggest an intentional shift towards esotericism in Buddhist dhāraṇī talismans. This conclusion is further supported by the initial blurriness of the print, which indicates that the printer was likely more concerned with the symbolic and ritual function of the amulet rather than ensuring the legibility of the text itself.

Xu Yin’s Amulet

Xu Yin’s amulet presents an intriguing case within Buddhist copying practices. Unlike the other amulets discussed, this piece appears to have been purchased or received from Huguo Temple (护国寺), yet the exact location of this temple remains uncertain. While the amulet was discovered in Luoyang, it is possible that it originated from a Huguo Temple in Yuncheng, more than 200 km away. However, due to the lack of definitive records, we cannot rule out the possibility that there was a Huguo Temple in or near Luoyang at the time. This raises broader questions about the movement of dhāraṇī amulets across monastic networks and commercial exchange within religious institutions.

The fact that Xu Yin’s amulet was initiated by a monk also aligns with patterns seen in the Ruiguang Si amulet, where monastic figures played a role in producing and circulating these sacred objects. This suggests that amulet production may not have been purely devotional but may have also involved an element of commercialised religious practice within Buddhist monasteries.

The Ruiguang Si Amulet

The Ruiguang Si amulet, found stored inside a stūpa, presents an unusual departure from the more common wearable format of Mahāpratisarā dhāraṇī amulets. Typically, these amulets were designed for personal use, in accordance with the dhāraṇī sūtra’s prescription that they be worn on the body for protection. The stūpa placement of the Ruiguang Si amulet suggests a possible fusion of ritual practices between different dhāraṇīs. Notably, the Sarva Durgati Pariśodhana Uṣṇīṣa-vijaya Dhāraṇī, another protective Buddhist spell, was frequently enshrined in stūpas and pillars, yet it also appeared in wearable amulet forms similar to Mahāpratisarā dhāraṇīs. This cross-influence between dhāraṇī traditions may have contributed to the decision to enshrine this particular Mahāpratisarā amulet within a stūpa.

The involvement of monastic figures in commissioning both Xu Yin’s and Ruiguang Si’s amulets further reinforces the idea that Buddhist amulet production was embedded within monastic economies. Monasteries likely played an active role in facilitating both devotional and commercial circulation of these sacred objects, allowing dhāraṇīs to reach a wider audience beyond immediate temple communities.

Conclusion

The copying practices observed in these amulets highlight the fluid nature of Buddhist textual transmission. The significant variations in script styles, orientations, and textual fidelity suggest that these amulets were not merely textual reproductions but also ritual objects shaped by evolving devotional and esoteric practices. Whether through the unconventional calligraphic influences seen in Li Zhi-Shun’s amulet or the blurred script of the Hangzhou amulet, these objects reflect the broader landscape of Buddhist textual adaptation and transmission in medieval China.

4. Conclusion

This study has demonstrated that dhāraṇī maṇḍala amulets, long dismissed as ritual ephemera, hold untapped value for the study of Buddhist textual transmission, scribal practice, and cultural exchange. Focusing on four representative examples from the late Tang and early Song periods—those associated with Xu Yin, Li Zhi-Shun, the Hangzhou National Archives, and Ruiguang Si—I have argued that these artefacts not only preserve otherwise unattested versions of the Mahāpratisarā Dhāraṇī, but also reveal the evolving forms and functions of Buddhist visual-linguistic expression in medieval China.

Though not produced for active recitation or reading, these amulets preserve a wealth of textual data. Their use of Indic scripts, especially Siddhaṃ and Nāgarī, provides material evidence of script transitions and scribal networks stretching between South and East Asia. Moreover, their layouts, bījākṣara arrangements, and use of mantra formulations allow us to reconstruct lost or otherwise unrecorded ritual practices. In particular, the discovery of Nāgarī script in tenth- and eleventh-century Chinese amulets challenges previous assumptions about script standardisation and reveals a continuing dynamism in Sino-Indian textual relations well into the second millennium.

Each of the four amulets contributes unique findings:

Xu Yin’s amulet (926–927 CE) provides one of the most complete and clearly dated examples of Mahāpratisarā recensions in Siddhaṃ, featuring carefully arranged bījākṣaras that likely functioned as seals or symbolic representations of the mantras.

Li Zhi-Shun’s amulet (980 CE) presents a sophisticated use of both Siddhaṃ and early Nāgarī, and includes rare Chinese-transliterated mantras. It also contains one of the earliest known structured visual mappings of bījākṣaras to Buddhist deities, suggesting highly developed ritual knowledge.

The Hangzhou amulet, although partially damaged, is part of a larger corpus of amulets preserved in the Xiasha Outlet Gallery Museum, revealing regional print networks and unexpected textual parallels that demand further editorial analysis. It also hints at script visualisation becoming an object of devotion itself.

The Ruiguang Si amulet (1005 CE) stands out for its integration of Nāgarī script, planetary deities, and the figure of Tejaprabhā, showing how Chinese Buddhist imagery absorbed and reinterpreted Indic and Daoist elements in new ritual contexts. Its mysterious ending phrase “vidani sulekha” opens the door to future research on scribal culture and textual closure.

Looking forward, three urgent areas of research emerge. First, a comprehensive comparative study of Mahāpratisarā Dhāraṇī recensions is now possible, especially thanks to newly discovered amulets in the Xiasha Outlet Gallery Museum. Editorial work across multiple recensions can clarify textual evolution, transmission paths, and ritual adjustments. Second, we need a dedicated lexicon or multilingual dictionary of bījākṣaras. These potent syllables recur across Buddhist and Hindu traditions but remain poorly understood outside of performative contexts. Scholarly tools are necessary to decode their meanings, functions, and patterns. Third, the Mahāpratisarā amulets should be studied alongside other major dhāraṇī amulets from the same period, such as the Uṣṇīṣa-vijaya dhāraṇī. Such cross-textual analysis could yield insights into genre formation, ritual overlap, and the dynamics of dhāraṇī culture in East Asia.

Ultimately, this study is not only about four amulets. It is about challenging our assumptions regarding what counts as a text, who gets to produce it, and where meaningful textual histories might be found. The dhāraṇī maṇḍala amulets may be small in size, but they carry the weight of expansive historical processes—of translation, ritualisation, visualisation, and transmission—that continue to shape Buddhist cultures to this day.