Translating Medicine Across Cultures: The Divergent Strategies of An Shigao and Dharmarakṣa in Introducing Indian Medical Concepts to China

Abstract

The Yogācārabhūmi, compiled by Saṅgharakṣa, was first introduced to China by An Shigao’s abridged translation (T607, Daodi jing 道地經), later, in 284 CE, Dharmarakṣa produced a more comprehensive version (T606, Xiuxing daodi jing 修行道地經). Lacking extant Sanskrit or Pali parallels, the text is difficult to interpret literally, and the differences between T607 and T606 add to the analytical challenges. However, a substantial section in both translations describing omens of impending death in the sick exhibits systematic parallels with Indian Āyurvedic texts, such as the Caraka-saṃhitā and Suśruta-saṃhitā. These parallels help clarify the ambiguous passages through comparative analysis. This study explores the translation strategies of An Shigao and Dharmarakṣa in introducing Indian medical concepts to China. An Shigao adopted a localization strategy, replacing foreign terms with analogous Chinese concepts. His terminology, corroborated by usage in Eastern Han or earlier Chinese texts—particularly excavated manuscripts—supports claims in the Chu sanzang ji ji regarding his expertise in medicine and divination. By contrast, Dharmarakṣa’s Xiuxing daodi jing sought greater fidelity to the Indian source material, offering a more detailed and systematic presentation of Āyurvedic knowledge. However, Dharmarakṣa did not entirely abandon An Shigao’s localization approach. He adopted a balanced strategy that combined faithful representation with cultural adaptation, reflecting the broader capacity of his more diverse and sophisticated audience to engage with complex and extensive foreign knowledge.

1. Introduction

Historians of medicine have long observed that the flow of healing knowledge between India and China tended to move in step with—and was often carried by—the networks that spread Buddhism. Joseph Needham’s magisterial Science and Civilisation in China already identifies Buddhism as the chief vehicle for ideas ranging from karma-based theories of disease to early variolation techniques.1 Scholarly attention therefore gravitates to the Tang dynasty, the high tide of Buddhist translation and patronage, when Chinese medical classics such as the Waitai miyao 外臺秘要 (Arcane Essentials from the Imperial Library) and Qianjin yifang 千金翼方 (Supplement to the Essential Prescriptions Worth a Thousand Gold Pieces) absorbed a wealth of Indian material (Ji [1991] 2008, pp. 82–84). By contrast, the pre-Tang era—particularly the earliest phase of Buddhist scripture translation—remains far more elusive: the surviving witnesses are fragmentary, philologically refractory, of uncertain provenance, and often laced with legend, leaving scholars divided over whether specific Indian medical concepts—such as the eight-branch system of Āyurvedic medicine—had yet entered China at all (Cai 1986).

Among the few early scriptures that can illuminate this contentious phase are the Daodi jing (道地經, T 607), translated by An Shigao (c. 148–180 CE), and the more expansive Xiuxing daodi jing (修行道地經, T 606), rendered by Dharmarakṣa in 284 CE. Demiéville (1954) first charted the transmission of these texts—which are both partial reflections of Saṅgharakṣa’s2 now-lost Yogācārabhūmi—and subsequent studies by Ui (1971) and Deleanu (1997) refined their doctrinal affiliations.3 However, the two Chinese versions—An Shigao’s in particular—are renowned for syntactic compression and obscure diction, while the Indic archetype has vanished, leaving many passages opaque.

One section of the Yogācārabhūmi catalogs omens of impending death, and its contents align very closely—though not word for word—with the corresponding chapters of Āyurvedic medical texts such as the Caraka-saṃhitā and Suśruta-saṃhitā. Only a handful of scholars focused on Buddhist medicine have so far remarked on the significance of this overlap. M. Chen (2005, pp. 52, 242–45) points out that Dharmarakṣa’s T606 preserves pediatric and obstetrical lore that has clear Sanskrit counterparts, while Salguero (2014, pp. 46–47) offers a overview of the medical content embedded in these early translations. An important parallel-text study by Yamabe (2013) further shows that these two handbooks contain the earliest recoverable form of a death-to-rebirth narrative: material first rearranged in the Garbhāvakrāntisūtra is carried into the handbooks, and the Manobhūmi chapter of the Yogācārabhūmi treatise 瑜伽師地論 then copies that very sequence, revealing a clear “sūtra → handbook → treatise” lineage. Yamabe’s comparison, however, is confined to doctrinal and narrative sources; to grasp T606 and T607 in full, one must read them against the wider backdrop of Āyurvedic medical science. Placed side by side with their Āyurvedic prototypes, these death-omen passages supply the philological keystone that T607 and T606 have long lacked. The Sanskrit text fixes the baseline wording, allowing us to unlock turns of phrase that have resisted interpretation for centuries, while the subtle mismatches—both between Sanskrit and Chinese and between An Shigao’s and Dharmarakṣa’s renderings—chart how unfamiliar medical ideas were incrementally understood as they crossed into a Han-dynasty milieu.

2. The Earliest Chinese Translation of Yogācārabhūmi

2.1. Background Overview

The Daodi jing (T607) translated by An Shigao is one of the earliest Buddhist texts introduced to China that has been preserved to this day. It is recorded in Dao’an’s 道安 seminal work, the Zongli zhongjing mulu 綜理眾經目錄, the earliest known catalog of the Buddhist canon, and its linguistic style closely aligns with An Shigao’s other translations, affirming its authenticity as one of his works.4 Among An Shigao’s translations, T607 is particularly notable for its linguistic complexity, reflecting his characteristic obscurity and difficulty. This complexity arises not only from the text’s use of rare or unconventional vocabulary and syntactic structures but also from the absence of a complete Sanskrit or Pali parallel text. Unlike most of An Shigao’s other translations, which primarily list basic Buddhist concepts5, T607 contains numerous descriptions related to secular life. These passages provide an invaluable resource for exploring how Chinese readers in the Eastern Han period first encountered and interpreted foreign concepts from India or Central Asia.

Due to the obscurity and complexity of T607, it is often necessary to consult another translation of the Yogācārabhūmi while reading it: T606, the Xiuxing daodi jing translated by Dharmarakṣa in 284 CE6. Dharmarakṣa’s version is notably more comprehensive, consisting of seven volumes, whereas T607 corresponds only to the first volume.7 Additionally, Dharmarakṣa’s Chinese is more fluent and easier to understand than An Shigao’s.8 Nonetheless, T606 also contains some perplexing terms, and the lack of parallel texts in other languages for comparison makes it difficult to explain many differences between T606 and T607.

Although a complete parallel text of the Yogācārabhūmi has not yet been discovered, some of the themes discussed within it have parallels in non-Buddhist Sanskrit literature. For instance, we find that part of Chapter 5 of T607, which discusses the signs of death in the sick (CBETA, T15, no. 607, pp. 232, a22–233, a15), is extensively discussed in several early Sanskrit medical texts, such as the Caraka-saṃhitā9 (henceforth Ca), the Suśruta-saṃhitā10 (Su), and the Aṣṭāṅga Hṛdaya11 (AH). Similarly, some content in the Purāṇas, such as the Agni Purāṇa12 (AP), the Vāyu Purāṇa13 (VP) (S. K. Sharma 2008), and the Skanda Purāṇa14 (SP), also overlaps with this theme.

In addition, there is another Chinese Buddhist text that discusses the theme of the signs of death in the sick in detail, but its parallels with the two Chinese translations of the Yogācārabhūmi have been less noticed and discussed. This discussion appears in a dialog between King Ajātaśatru and the renowned physician Jīvaka, found in the T374 Da Banniepan jing 大般涅槃經, Chapter 8 Fanxing pin 梵行品 (CBETA, T12, no. 374, pp. 481, b27–482, a29). This text was translated by Dharmakṣema 曇無讖 (385–433 CE). Unfortunately, a complete Sanskrit version of T374 has not survived.

The parallels between the content of T607, T606, and T374 and Sanskrit medical texts are particularly noteworthy. These connections not only provide critical insights for interpreting the enigmatic T607 but also illuminate how, over a span of more than two centuries, early Chinese audiences gradually understood and assimilated foreign concepts from vastly different cultural contexts, especially in areas such as medicine and divination.

2.2. Corresponding Content of Chinese Translations in Sanskrit Medical Texts

The discussions on “omens of imminent death in patients” in the Chinese translations T607, T606, and T374 follow an almost identical sequence, which can be broadly divided into three parts.

[A] The first part provides a detailed account of several signs indicating that a patient is beyond cure. These signs have systematic correspondences in the aforementioned Sanskrit texts and can be categorized into five types: (1) the patient has certain inauspicious dreams,15 such as dining or traveling with the dead, funeral workers, toilet cleaners, or monkeys;16 being naked and smeared with oil;17 or walking south with animals such as donkeys or dogs;18 (2) the messenger summoning the physician exhibits inauspicious traits and behaviors, such as being dirty,19 poorly dressed,20 or touching their hair;21 (3) the date, astrological signs, time of day, and the patient’s appearance on the day of the consultation are inauspicious (see Section 3.1); (4) the physician encounters inauspicious omens on the way to or at the patient’s home, such as predatory birds flying south22 or extinguished fires;23 (5) the patient displays certain physical symptoms, such as tangled eyelashes,24 a dry and curved nose,25 or insensitivity to hair being pulled.26

[B] The second part in T607 and T606 informs the reader that even the most skilled physician cannot cure such a patient. T606 is much more detailed in this section compared with T607, while T374 lacks this content altogether. The relationship between this section and the Āyurvedic medical system will be discussed in Section 3.2.

[C] The third part describes how physicians would choose to conceal the truth from the patient once they determine that the patient is incurable.27

Table 1. “Omens of Death” in the Yogācārabhūmi and their parallel passages in Chinese and Sanskrit texts.

	T607	T606	T374	Caraka-saṃhitā	Suśruta-saṃhitā	Aṣṭāṅga Hṛdaya	Purāṇa
[A.1] Dreams of a dying patient	232a22–b24	183c4–184a12	481b29–c9	5.5	1.29	2.6	AP.228 VP.19 SP.4.1
[A.2] Omens on the messenger sent to seek a physician	232b24–c2	184a13–26	481c9–13	5.12	1.29	2.6	AP.294
[A.3] Inauspicious dates, constellations, and times for seeking medical help	232c2–12	184a27–b13	481c13–24	5.12	1.29	2.6	AP.294
[A.4] Omens seen and heard by the physician in the home of a dying patient	232c12–17	184b14–18	481c24–482a9	5.12	1.29	2.5–6
[A.5] Physical symptoms of an incurable patient	232c17–233a10	184b18–c26	482a9–24	5.1–4, 7–12	1. 28, 31–32	2.5	VP.19 SP.4.1
[B] The most skilled physicians	233a10–11	184c26–185b8	/
[C] The benevolent lies of the physician	233a11–15	185b9–16	482a24–28	5.12		2.5

presents the parallel passages from the Chinese and Sanskrit texts, arranged following the sequence found in T607.

2.3. Did Dharmarakṣa Refer to An Shigao’s Translations?

Compared with An Shigao’s translation T607, Dharmarakṣa’s translation T606 is significantly more complete in terms of content. The content of the Daodi jing corresponds only to the first volume of the seven volumes of the Xiuxing daodi jing. Moreover, the Xiuxing daodi jing often includes additional details on shared content, many of which align with information found in Sanskrit medical texts (examples of which are discussed in Section 3.2). This evidence strongly suggests that Dharmarakṣa undertook a completely new translation based on an independent foreign source text, as explicitly recorded in the epilogue of T606 (see Note 6), rather than merely refining the linguistic shortcomings of earlier Chinese translations.28

This raises an important question: Did Dharmarakṣa consult An Shigao’s earlier translations while preparing his own work? While Dharmarakṣa is well known for revising or retranslating texts with reference to earlier versions, scholarly discussions have largely centered on his connections with the works of Zhi Qian 支謙.29 His potential engagement with An Shigao’s translations has received relatively little attention, and early Buddhist scripture catalogs do not provide direct evidence of such a link. Therefore, any connections must be inferred through a careful comparative analysis of the texts themselves.

Despite the greater completeness and fluency of Dharmarakṣa’s translations, many overlapping passages can still be identified between the Daodi jing and the Xiuxing daodi jing. These overlapping sections, highlighted in

Table 2. Overlapping sections between the Daodi jing (T607) and the Xiuxing daodi jing (T606).

	T607 Daodi jing	T606 Xiuxing daodi jing
1	從若干經得明堅，不老不死甘露 (CBETA, T15, no. 607, p. 231, b23)	從若干經採明要，立不老死甘露言 (CBETA, T15, no. 606, p. 182, c22–23)
2	如是在一貫珠，一時俱行，五陰更如是。 (CBETA, T15, no. 607, p. 232, a10–11)	如是五陰，如一貫珠，一時俱行 (CBETA, T15, no. 606, p. 183, b19–20)
3	自手摩抆鬚髮 (CBETA, T15, no. 607, p. 232, b27)	而數以手摩抆鬚髮 (CBETA, T15, no. 606, p. 184, a20–21)
4	復見小兒俱相坌土；復胆裸，相挽頭髮，破瓶盆瓦甌 (CBETA, T15, no. 607, p. 232, c15–16)	或見小兒，以土相坌，而復裸立，相挽頭髮，破甖瓶盆 (CBETA, T15, no. 606, p. 184, b17–18)
5	譬如人死時有死相，為口不知味，耳中不聞聲 (CBETA, T15, no. 607, p. 233, a6–7)	人臨死時，所現變怪：口不知味，耳不聞音 (CBETA, T15, no. 606, p. 184, c17–18)
6	中止者或住一日，或住七日止 (CBETA, T15, no. 607, p. 233, c7–8)	在中止者，或住一日，極久七日 (CBETA, T15, no. 606, p. 186, b9–10)
7	大腸、小腸、肝、肺、心、脾、腎，亦餘藏，令斷截 (CBETA, T15, no. 607, p. 233, b5–6)	大小之腸、肝肺心脾，并餘諸藏，皆令斷絕 (CBETA, T15, no. 606, p. 185, c12–13)
8	二十五七日，生七千脈，尚未具成 (CBETA, T15, no. 607, p. 234, b17)	二十五七日，生七千脈，尚未具成 (CBETA, T15, no. 606, p. 187, b14–15)
9	二十六七日，諸脈悉徹，具足成就，如蓮花根孔 (CBETA, T15, no. 607, p. 234, b17–18)	二十六七日，諸脈悉徹，具足成就，如蓮華根孔 (CBETA, T15, no. 606, p. 187, b15–16)
10	三十六七日，爪甲生 (CBETA, T15, no. 607, p. 234, b24–25)	三十六七日，爪甲成 (CBETA, T15, no. 606, p. 187, b22)
11	三十七七日，母腹中若干風起 (CBETA, T15, no. 607, p. 234, b25)	三十七七日，其母腹中，若干風起 (CBETA, T15, no. 606, p. 187, b22–23)
12	三十八七日，為九月不滿四日，骨節皆具足。兒生宿行有二分：一分從父，一分從母 (CBETA, T15, no. 607, p. 234, c4–6)	是為三十八七日。九月不滿四日，其兒身體、骨節，則成為人……其小兒體而有二分：一分從父，一分從母。 (CBETA, T15, no. 606, p. 187, c6–13)
13	如是并四百四病在身中。譬如木中出火還燒木 (CBETA, T15, no. 607, p. 235, a14–16)	凡合計之，四百四病，在人身中。如木生火，還自燒然 (CBETA, T15, no. 606, p. 188, c8–9)
14	是身為譬如餓鬼，常求食飲 (CBETA, T15, no. 607, p. 236, b12–13)	是身如餓鬼，常求飲食 (CBETA, T15, no. 606, p. 219, b14–15)

below, are particularly noteworthy because they do not consist of common terms or idiomatic expressions. In fact, many of these phrases are unique to these two texts and cannot be found elsewhere in the Chinese Buddhist canon.

It is essential to address and dismiss an alternative possibility: Could this “overlap” result from Dharmarakṣa revising An Shigao’s translation? The extant version of T607, while obscure, retains the distinctive hallmarks of An Shigao’s translation style, which suggests that it remained largely unaltered by later translators. Instead, a more plausible hypothesis is that Dharmarakṣa referred to T607 as a source during the process of translating T606. His approach involved reworking similar material into more coherent and rhythmically balanced four-character and seven-character sentences, which would be consistent with his own stylistic tendencies.

Although An Shigao’s translations were eventually superseded by newer versions, they continued to exert considerable influence, particularly up until the Eastern Jin dynasty. A testament to this enduring importance is the high regard in which the eminent Buddhist scholar Dao’an held An Shigao’s works. Dao’an composed annotations and prefaces for several of An Shigao’s texts, including the Daodi jing,30 praising it for its “pure and refined tone, simple yet unpretentious style: at times adapting the form to follow the text, at other times preserving the essence without embellishment. Truly, An Shigao has captured the essence of the teachings with precision (音近雅質，敦兮若樸，或變質從文，或因質不飾，皇矣世高，審得厥旨)”.31 Therefore, it is entirely reasonable to consider that Dharmarakṣa may have consulted and drawn upon the substantial earlier translation T607 during the process of translating T606.

In stark contrast to the evident intertextuality between T607 and T606, the relationship between Dharmakṣema’s T374 and the two Chinese translations of the Yogācārabhūmi appears to be limited to thematic parallels. There are no notable similarities in linguistic expression or phrasing (some of the comparisons can be found in Notes 16–27), suggesting that these texts are, in fact, independent translations.

3. Different Strategies for Translating Indian Medical Terminology

Given that Dharmarakṣa likely referred to An Shigao’s translation, the differences in their renderings of the same content merit particular attention. It is often dissatisfaction with earlier translations that motivates the production of new ones, and Dharmarakṣa’s approach reflects a clear departure from An Shigao’s. His version is not only more fluent and idiomatic in its Chinese but also demonstrates a distinct translation strategy regarding medical concepts. Whereas An Shigao tended to substitute Indian medical terms with corresponding Chinese concepts, Dharmarakṣa sought to preserve and introduce foreign concepts as comprehensively as possible, offering readers a more faithful representation of the original text.

3.1. Appropriate Timing for Seeking Medical Treatment

The three Chinese translations exhibit numerous detailed differences, many of which warrant further investigation. As a case study, this discussion focuses on a specific comparison: the “[A.3] Inauspicious Dates, Constellations, and Times for Seeking Medical Help” section in Table 1. By analyzing the three Chinese translations (see

Table 3. “The Timing of Sending Envoys to Seek Medical Treatment” in Chinese Buddhist translations and Sanskrit medical texts.

T607	亦如是諱日來呼：[i] 若四、若六、若九、若十二、若十四來至到，復觸忌諱日，人所不喜醫，[ii] 復何血忌、上相 1、四激、反支來喚？是亦不必日時、漏刻、星宿、須臾，……(CBETA, T15, no. 607, p. 232, c2–5) … or seeking the physician on inauspicious days, such as the fourth, sixth, ninth, twelfth, or fourteenth days (of each month), or on other unfavorable days when it’s not advisable to consult a physician, such as on dates of Xueji, Shangxiang,* Siji, and Fanzhi. However, there are situations where one doesn’t necessarily have to adhere to specific dates, times, constellations, or fortune-telling, …
T606	[i] 又其日惡，若四日、六日、十二日、十四日，以此日來者皆為不祥，醫即不喜，[ii] 以觝星宿，失於良時，神仙先聖所禁之日。醫心念言：“雖值此怪星宿吉凶，或可治療。” (CBETA, T15, no. 606, p. 184, a29–b3) If the day is inauspicious, such as the fourth, sixth, twelfth, or fourteenth days (of a month), seeking medical attention on these days is considered unlucky. The physician won’t be pleased because it conflicts with the celestial alignments and is not a favorable time. These are days that the sages (ṛṣi) have prohibited. However, the physician may think, “Even though it’s an inauspicious time astrologically, the patient might still be treatable.”
T374	[i] 復作是念：“使雖不吉，當復占日，為可治不？” 若四日、六日、八日、十二日、十四日，如是日者，病亦難治。[ii] 復作是念：“日雖不吉，當復占星，為可治不？” 若是火星、金星、昴星、閻羅王星、濕星、滿星，如是星時，病亦難治。[iii] 復作是念：“星雖不吉，復當觀時。” 若是秋時、冬時，及日入時、夜半時、月入時，當知是病亦難可治。(CBETA, T12, no. 374, p. 481, c13–20) He further reflects, saying: “Even if the messenger is inauspicious, should I observe the day to determine if treatment is possible?” If it is the fourth, sixth, eighth, twelfth, or fourteenth day (of a month), then on such days, the illness is also difficult to treat. He further reflects, saying: “Even if the day is inauspicious, should I observe the stars to determine if treatment is possible?” If it is under the influence of Mars, Venus, the Pleiades, Yama’s star, the “Moist Star”, or the “Full Star”, then during such times, the illness is also difficult to treat. He further reflects, saying: “Even if the stars are inauspicious, should I observe the timing?” If it is during autumn, winter, sunset, midnight, or moonset, it should be understood that the illness is also difficult to treat.
Ca.	5.12.68 [iii] …-asandhyāsv agraheṣu ca |      [ii] adāruṇeṣu nakṣatreṣv anugreṣu dhruveṣu ca || 5.12.69 [i] vinā caturthīṃ navamīṃ vinā riktāṃ caturdaśīm |     [iii] madhyāhnamardharātraṃ ca … (The physician should regard the messenger of the following description as auspicious, i.e., … arriving at times other than) those marked by the two twilights, inauspicious conjunction of planets, constellations which are unstable and of a fierce or baleful aspect, the “void” days of the fortnight comprising the fourth, ninth and fourteenth days, the mid-day and the midnight… 2
Su.	1.29.17       [iii] madhyāhne cārdharātre vā sandhyayoḥ kṛttikāsu ca |      [ii] ārdrāśleṣāmaghāmūlapūrvāsu bharaṇīṣu ca ||      [i] caturthyāṃ vā navamyāṃ vā ṣaṣṭhyāṃ sandhidineṣu ca | (A messenger, seeking the interview of a physician …) at noon or at midnight, at morning or at evening, or during the happening of any abnormal physical phenomenon, or at an hour under the influence of any of the following asterisms (lunar mansions), viz. the Ārdra, the Aśleṣā, the Maghā, the Mulā, the two Pūrvas, and the Bharaṇī, or on the day of the fourth, ninth, or the sixth phase of the moon (whether on the wane or on the increase), as well as on the last days of months and fortnights, should be considered as a messenger of evil augury. 3
AH	2.6.11–12   [iii] … tathārdharātre madhyāhne saṃdhyayoḥ parvavāsare ||      [i] ṣaṣṭhīcaturthīnavamī- [ii]rāhuketūdayādiṣu |      [ii] bharaṇīkṛttikāśleṣāpūrvārdrāpaitryanairṛte || So also he, who approaches (the physician) at midnight, midday, sunrise and sunset, on a crucial (bad) day; on the sixth, fourth, and ninth days (of the two fortnights), on days of rise of rāhu and ketu, on days of stars like Bharaṇi, Kṛttikā, Aśleṣā, Pūrvā, Ārdrā, Paitra (Maghā) and Nairṛta (Mūla).

¹ The character “xiang” (相) might be erroneous; see Section 3.1.4 for details. ² The English translation here is based on P. V. Sharma (1994, pp. 527–28). ³ The English translation here is based on P. V. Sharma (1999, p. 291).

), it becomes evident that T607 and T606 describe two circumstances under which it is considered unsuitable to send a messenger to seek medical help:

[i] “Dates”—specific inauspicious days within a lunar month.

[ii] “Constellations”—periods when the moon occupies certain dangerous nakṣatras (lunar mansions).

In contrast, T374 adds a third condition:

[iii] “Seasons and Times”—particular seasonal periods, such as autumn and winter, as well as critical times of day, including noon, midnight, dawn, and dusk, when transitions or peaks occur.

These taboos, as outlined in the three translations, show significant parallels with Sanskrit medical texts, which also highlight the inauspicious influence of specific times, dates, and constellations for seeking medical assistance.

When comparing the conditions of “dates” and “constellations” across the three versions, the translations of “dates” are relatively clear in all versions, though discrepancies exist in the specific dates mentioned.32 However, the translation of “constellations” varies significantly across the texts.

(a) The latest version, Dharmakṣema’s T374, provides the most specific and detailed account. Dharmakṣema named each inauspicious constellation explicitly, some of which might correspond to the Sanskrit originals. However, the correspondences appear inconsistent: some are literal translations, such as Shi xing 濕星 (“Wet Star”), which may correspond to Ārdrā (the fourth or sixth Nakṣatra), as ārdra can mean “wet”; others are conceptual equivalents, such as Mao xing 昴星 (Pleiades) which might correspond to Kṛttikā; still others reflect the presiding deity of the constellation, such as Yanluowang xing 閻羅王星 (“Yama Star”), which may correspond to Bharaṇi (whose deity is Yama). Man xing 滿星 (“Full Star”) is perplexing in both the Chinese astrological context and the Sanskrit source. I speculate that it may result from a mistranslation of Pūrvā as Pūrṇa.

(b) Dharmarakṣa’s T606 takes a more general approach, merely noting the condition of “forbidden days under the influence of constellations for sending a messenger to seek medical help.” This omission of specific constellations may stem from the source text itself lacking such details (similar to Caraka-saṃhitā 5.12.68), or it may have been an intentional effort to simplify the translation and maintain brevity. The former seems more plausible, as Dharmarakṣa’s other works, such as his translation of the Śārdūlakarṇāvadāna, demonstrate his familiarity with Indian astrological theories.33 Furthermore, in T606, Dharmarakṣa generally provided more detailed translations of Indian medical terms than other translators (as discussed in Section 3.2).

(c) The earliest and most ambiguous version, An Shigao’s T607, replaces the Indian constellation names entirely with localized Chinese terms, making it impossible to establish connections with Indian texts without the contextual information provided by T606 and T374.

As one of the earliest translators of Buddhist scriptures, An Shigao faced the formidable task of communicating medical, astronomical, and astrological concepts across two vastly different cultural contexts. To understand his approach, it is necessary to reconstruct the meanings of several terms he used in T607. His translation employs terms such as Xueji 血忌, Shangxiang 上相, Siji 四激, and Fanzhi 反支, all of which refer to inauspicious days influenced by deities and spirits. These deities and spirits were associated with certain constellations or Heavenly Stems and Earthly Branches, which were believed to govern fortune and misfortune in daily life. This system of beliefs was highly prevalent during An Shigao’s era and remained influential into the modern period.34 Although these terms, rooted in folk beliefs, are rarely mentioned in official or classical texts, recent discoveries of Qin and Han dynasty bamboo (or wooden) slips and silk manuscripts have preserved extensive material that can help reconstruct their historical context. For instance, among the Juyan Xinjian 居延新簡 (newly unearthed Juyan bamboo slips) is one labeled EPT65:425, dated to the “Second Year of Yongyuan” 永元 (90 CE, approximately half a century before An Shigao’s arrival in China). This slip mentions three of the four terms used in T607—Xueji 血忌, Siji 四激, and Fanzhi 反支 (see Appendix A, Figure A1). Such a striking overlap is unlikely to be coincidental, indicating that the terminology in T607 draws directly from this indigenous Chinese belief system.

To understand why An Shigao replaced Indian medical and astronomical terminology with these four deities and spirits, it is essential to examine their specific meanings within the historical context and how they were believed to influence human activities.

3.1.1. Xueji

Xueji (literally “Blood Taboo,” referring to days when bloodletting is prohibited) is a commonly mentioned deity in Qin and Han dynasty bamboo slips. The Jianshui Jinguan 肩水金關 Han slips associate the Xueji with the following twelve Earthly Branches:

血忌：丑、未、寅、申、卯、酉、辰、戌、巳、亥、午、子35

Xueji (appearing on the following Earthly Branches): Chou, Wei, Yin, Shen, Mao, You, Chen, Xu, Si, Hai, Wu, Zi.

The Huangdi hama jing 黃帝蝦蟆經 (The Yellow Emperor’s Scripture of the Toad),36 composed approximately between the third and fourth centuries, builds on this association by correlating the twelve Earthly Branches with specific days in each month, retaining the same order as in the Han slips:

正月丑、二月未、三月寅、四月申、五月卯、六月酉、七月辰、八月戌、九月巳、十月亥、十一月午、十二月子。右十二日，忌血日也。一名“致忌”。一名“禁忌”。其日不可灸刺，见血忌凶。

First month: Chou, second month: Wei, third month: Yin, fourth month: Shen, fifth month: Mao, sixth month: You, seventh month: Chen, eighth month: Xu, ninth month: Si, tenth month: Hai, eleventh month: Wu, twelfth month: Zi. These twelve days are known as the “blood taboo” days. They are also called Zhiji or Jinji. On these days, acupuncture and any activity involving blood are considered inauspicious.

The practice of associating these days with the twelve Earthly Branches remained largely unchanged in Ming and Qing dynasty almanacs. However, the Kongjiapo 孔家坡 Han slips reveal a rare method of linking Xueji days to constellations:37

春心，夏輿鬼，秋婁，冬虛，不可出血若傷，必死。血忌，帝啓百虫口日也。甲寅、乙卯、乙酉不可出血，出血，不出三歲必死。

(For the transcription and annotation of this slip text, see Hubei Provincial Institute of Cultural Relics and Archaeology and Suizhou Archaeological Team (2006, p. 179)).

In spring: Heart constellation, in summer: Ghost constellation, in autumn: Bond constellation, in winter: Void constellation. (When influenced by these constellations,) bloodletting or injuries are forbidden, as they are believed to be fatal. Xueji days are also the days when the “Di” opens the mouths of a hundred insects. On the days Jiayin, Yimao, and Yiyou, bloodletting is forbidden; if blood is let, death will occur within three years.

The prohibition of bloodletting on Xueji days is further corroborated by Wang Chong 王充 (27–ca. 97 CE) in his Lunheng 論衡 (Balanced Discourses). Known for his rationalist and materialist views, Wang Chong offered the following critique of such folk superstitions:

如以殺牲見血，避血忌、月殺，則生人食六畜，亦宜辟之。海内屠肆，六畜死者，日數千頭，不擇吉凶，早死者，未必屠工也。天下死罪，冬月斷囚，亦數千人，其刑於市，不擇吉日，受禍者，未必獄吏也。（王充《論衡·譏日》）

If it is believed that the slaughter of livestock and the sight of blood should be avoided on Xueji and Yuesha days, then people consuming livestock should also avoid these days. Across the nation, thousands of livestock are slaughtered daily in butcher shops without regard for auspicious or inauspicious days, yet those who die prematurely are not necessarily the butchers. Similarly, thousands of criminals sentenced to death are executed in the winter months without choosing auspicious days, yet those who suffer misfortune are not necessarily the prison officials. (Wang Chong, Lunheng [Balanced Discourses], “On Criticizing the Selection of Auspicious Days.”)

Because treatments such as acupuncture often involve bleeding, medical procedures were considered inappropriate on Xueji days. This notion is consistently mentioned in the Han slips, Lunheng, and later works, including medical texts, Daoist scriptures, and almanacs for selecting auspicious dates.38

3.1.2. Siji

In the unearthed Qin and Han bamboo slips, Siji四激 is also written as “四儌” or “四敫.” These variant characters (激, 儌, and 敫) are likely phonetic loans for jiao (交), meaning “intersection,” referring to the transitional periods between the four seasons.39 The Shuihudi Qin slips provide detailed records of these days:

入月七日及冬未、春戌、夏丑、秋辰，是胃（謂）四敫，不可初穿門、爲户牖、伐木、壞垣、起垣、徹屋及殺，大凶；利爲嗇夫。（《日書》甲種《門》）

On the seventh day of each month, as well as the Wei day in winter, the Xu day in spring, the Chou day in summer, and the Chen day in autumn, are the so-called “Four Jiao”. On these days, it is highly inauspicious to dig doorways, construct doors and windows, fell trees, demolish walls, build walls, dismantle houses, or kill (livestock?). (However,) they are deemed appropriate for agricultural officials (to carry out their duties) (Rishu jiazhong [Daybook A], “Men [Door].”)40

夏三月丑敫。春三月戊（戌）敫。秋三月辰敫。冬三月未敫……凡敫日，利以漁邋（獵）、請謁、責人、摯（執）盗賊，不可祠祀、殺生（牲）。（《日書》甲種《臽日敫日》）

… In the third month of summer, the Chou day is a Jiao day; in the third month of spring, the Xu day is a Jiao day; in the third month of autumn, the Chen day is a Jiao day; in the third month of winter, the Wei day is a Jiao day… On these Jiao days, fishing, hunting, paying respects, pursuing fugitives, and capturing thieves are appropriate activities, but worship and killing livestock are prohibited. (Rishu jiazhong [Daybook A], “Xianrijiaori [Xian day and Jiao day].”)41

The Qin slips’ “Four Jiao” (四敫) days correspond to the “Four Ji” (四激 or 四擊) mentioned in later texts (L. Liu 1994, pp. 178–79). The difference lies only in the written form, not the meaning. From the medieval period onward, “Four Ji” days were frequently cited in medical texts as unsuitable for medical treatment. The Huangdi hama jing notes the following: 春戌、夏丑、秋辰、冬卯。右四時忌日，不可灸刺。(“Spring: Xu, Summer: Chou, Autumn: Chen, Winter: Mao. These four seasonal taboo days are unsuitable for acupuncture”). The earliest surviving Japanese medical text Ishinpō 醫心方 (Heart of Medicine Formulae), quotes two relevant texts, both of which emphasize that Four Ji days are unsuitable for preparing or taking medicine. The first, from the Hama jing 蝦蟆經, states the following: 四激日：春戌、夏丑、秋辰、冬卯。有四時忌日，古今傳諱，不合藥服藥也。 (“These four seasonal taboo days have been historically avoided and are unsuitable for preparing or taking medicine”). The second is from the Yangsheng yaoji 養生要集 (Essential Collection on Nourishing Life): 四激、破、除、未日時，不中合藥、服藥。(“On the days of Four Ji, Broken Days, Transition Days, and Wei Days, it is inappropriate to prepare or take medicine”).

Even Ming and Qing almanacs, such as the Xieji Bianfang Shu 協紀辨方書 (The Book of Harmonizing the Seasons and Distinguishing Directions), maintained the calculations for Four Ji days from Qin and Han manuscripts, though they emphasized their inauspiciousness for military campaigns and border defenses (其日忌出軍、邊防等事).42

3.1.3. Fanzhi

The taboo surrounding Fanzhi 反支 days gained prominence during the Han dynasty. Although official records provide little direct information about the meanings and practices associated with this folk custom, frequent critiques suggest that it was widely observed among the general populace and even influenced the elite. Wang Fu 王符 (ca. 85–ca. 163 CE), for instance, recounted the following in his Qianfu lun 潛夫論 (Essays of a Recluse):

孝明皇帝嘗問今旦何得無上書者？左右對曰：“反支故。”帝曰：“民既廢農遠來詣闕，而復使避反支，是則又奪其日而寃之也。”乃勅公車受章，無避反支。（王符《潛夫論·愛日》）

Emperor Xiaoming once asked why no memorials had been submitted that morning. His attendants replied, “It is because today is a Fanzhi day.” The Emperor said, “The people have already abandoned their farming and traveled far to the capital, and now we make them avoid Fanzhi days, thus depriving them of their time and causing them unjust suffering. “He then ordered the officials to receive petitions regardless of Fanzhi days. (Wang Fu, Qianfu Lun [Essays of a Recluse], “On Cherishing the Day.”)

Annotations by Li Qi 李奇 on Ban Gu’s 班固 Han Shu 漢書 (Book of Han) highlight the dangers of adhering to such taboos:

“及王莽敗，二人俱客於池陽，竦爲賊兵所殺。”（班固《漢書·陳遵傳》）顏師古注引李奇注：“竦知有賊當去，會反支日，不去，因爲賊所殺。桓譚以爲通人之蔽也。”

“When Wang Mang was defeated, the two men were guests in Chiyang, and Song was killed by bandits.” (Ban Gu, Han shu [Book of Han], “Biography of Chen Zun.”） Yan Shigu’s commentary quoted Li Qi’s annotations: “Song knew there were bandits and planned to leave, but it happened to be a Fanzhi day, so he stayed and was killed by the bandits. Huan Tan believed that even wise people could be misled (by such beliefs).”

Criticism of the belief in Fanzhi days continued into later periods. Yan Zhitui 顏之推 (531–591 CE) in Yanshi jiaxun 顏氏家訓 (Yan’s Family Instructions) lamented the folly of such superstitions:

世傳術書，皆出流俗，言辭鄙淺，驗少妄多。至如反支不行，竟以遇害；歸忌寄宿，不免凶終：拘而多忌，亦無益也。（顏之推《顏氏家訓·雜藝》）

The divination and fortune-telling books circulating in the world are all produced by common, unsophisticated people, and their content is crude and shallow, with few predictions coming true and many being false. For example, not leaving home on Fanzhi days ultimately leads to harm; encountering taboo days for travel or return and staying overnight en route cannot prevent dying an unnatural death. Adhering too rigidly to these superstitions is ultimately of no benefit. (Yan Zhitui, Yanshi jiaxun (Yan’s Family Instructions), “Miscellaneous Skills.”)

The Qin and Han bamboo slips provide detailed calculations and taboos for Fanzhi days.43 For instance, the Fangmatan Qin slips describe Fanzhi days as follows:

子朔巳亥，丑朔子午，寅朔子午，卯朔丑未，辰朔丑未，巳朔寅甲，午朔寅申，未朔【酉卯，申朔酉卯，酉朔戌辰，戌朔戌辰】亥朔巳亥，是胃（謂）反支。以徙官，十徙；以得憂者，十喜；以亡者，得十；𣪠（繫）囚，亟出。不可冠帶、見人、取（娶）婦、嫁女、入臣妾。不可生歌樂鼓橆（舞），殺畜生見血，人死之，利以出，不利以入，得一失十，以受賀喜，十憂。以[去]入官者，必去。以歐（敺）治（笞）人者，必蓐（辱）。（《日書》乙種《反支》）44

If the first day of a lunar month is Zi, then Si and Hai days are Fanzhi days. If the first day is Chou, then Zi and Wu days are Fanzhi days. If the first day is Yin, then Zi and Wu days are Fanzhi days. If the first day is Mao, then Chou and Wei days are Fanzhi days. If the first day is Chen, then Chou and Wei days are Fanzhi days. If the first day is Si, then Yin and Jia days are Fanzhi days. If the first day is Wu, then Yin and Shen days are Fanzhi days. If the first day is Wei, then You and Mao days are Fanzhi days. If the first day is Shen, then You and Mao days are Fanzhi days. If the first day is You, then Xu and Chen days are Fanzhi days. If the first day is Xu, then Xu and Chen days are Fanzhi days. If the first day is Hai, then Si and Hai days are Fanzhi days. These are termed “Fanzhi” days. On these days, those who are transferred to new positions will be transferred tenfold (in the opposite direction). Encountering sorrowful events will be met with tenfold joy. Lost items will be recovered tenfold. Prisoners who are bound will be released immediately. One should not wear a hat or belt, meet people, marry, arrange marriages for daughters, or purchase male and female servants. Singing, playing music, killing livestock, and seeing blood are also prohibited, as these actions can lead to death. It is favorable to give things away on this day but not to receive them; gaining one thing will result in losing ten. Receiving congratulations will bring tenfold sorrow. Officials holding office on this day are certain to lose their positions, and those who beat or whip others will inevitably be disgraced. (Rishu yizhong [Daybook B], “Fanzhi.”)

The underlined section describes the method for calculating “Fanzhi” days. If the first day of a lunar month is Zi (子), then the 6th, 12th, 18th, 24th, and 30th days (alternating between Si 巳 and Hai 亥) are designated as “Fanzhi” days, following a 12-day cycle. Since the twelve Earthly Branches are believed to correspond to twelve directions (see Figure 1), “Fanzhi” days always involve two days that are diametrically opposed in position (e.g., Si 巳 and Hai 亥, Zi 子 and Wu 午). Consequently, actions undertaken on Fanzhi (“Reversed Branch”) days are believed to yield opposite outcomes.45

Medical practices were also affected by these beliefs. The Huangdi hama jing explicitly warns readers with the following: 凡反支日……不灸刺、服藥。大凶。(“On Fanzhi days, … avoid acupuncture and taking medicine. Great misfortune [will result]”)

3.1.4. Shangxiang: An Early-Stage Scribal Error

Among the four terms under discussion, the precise meaning of “Shangxiang” (上相) is the most enigmatic. Unlike Xueji, Siji, and Fanzhi, Shangxiang does not appear in contemporary date-selection texts from the Han dynasty. Given its association with the other terms, it is plausible that it also refers to a type of deity or spirit. I propose that Shangxiang (上相) is likely a scribal error for Shangshuo (上朔). The characters Shuo (朔) and Xiang (相) exhibit similar forms in the handwritten scripts of the Han dynasty, as illustrated in

Table 4. The handwritten forms of “Shuo” (朔) and “Xiang” (相) in the Eastern Han dynasty.

Shuo 朔	EPT52 1: 194	EPT43 2: 62	EPT43: 99
Xiang 相	EPT56: 77B	EPT59: 524A	EPT57: 64

¹ This bamboo slip bears the date 建始元年 “First Year of Jianshi” (32 BCE). ² This bamboo slip bears the date 更始二年 “Second Year of Gengshi” (24 CE).

.

Unlike Xueji and Siji, which have been consistently transmitted from the Han dynasty through the Qing dynasty with minimal variation, the calculation methods for Shangshuo days exhibit considerable differences between extant calendrical texts and unearthed manuscripts.

For example, the Qing dynasty’s Xieji bianfang shu describes the calculation of Shangshuo as follows: 陽年以年干加寅，順數至亥，陰年以年干加丑，順數至巳也。47 (“In Yang years, add the Heavenly Stem of the year to Yin 寅, counting forward to Hai 亥; in Yin years, add the Heavenly Stem of the year to Chou 丑, counting forward to Si 巳.”) On Shangshuo days, activities such as banquets, weddings, long journeys, and official appointments are prohibited (上朔日忌宴會、嫁娶、遠行、上官).48 However, Han dynasty texts such as Wang Chong’s Lunheng only mention 上朔不會衆 (“On Shangshuo days, gatherings are not permitted”).49

The excavated Mawangdui silk manuscripts, particularly Xingde A (Xingde jiapian刑德甲篇), Yin yang wuxing A (陰陽五行甲篇), and Yin yang wuxing B (陰陽五行乙篇), offer more comprehensive information on Shangshuo, diverging from the calculation methods in later texts such as the Xieji bianfang shu. Recent studies suggest that Shangshuo was understood in two distinct ways during the Han period: the annual “Yearly Shangshuo” (年上朔) and the bi-monthly “Qi Shangshuo” (氣上朔, also referred to as 月上朔, “Monthly Shangshuo”). These two systems correspond, respectively, to the “Xingde dayou” (刑德大遊) and “Xingde xiaoyou” (刑德小遊),50 both of which relate to the movements of the deities Xing (刑, Punishment) and De (德, Virtue).

The Mawangdui silk manuscripts associate Shangshuo days with the goddess Nüwa, portraying them as unsuitable for most activities, particularly those involving women. For instance, the Yin Yang Wuxing A states:

〼【爲】上朔，名爲女淉（媧），不可祭祀、訶（歌）藥（樂）、行、作事、𢱭（拜）受爵，必有大咎，百事莫可，女子之事蜀（獨）甚。（《陰陽五行甲篇》雜占之二《上朔》）51

… is the Shangshuo day, (and the deity influencing this day) is called Nüwa. It is forbidden to perform sacrifices, sing, play music, travel, undertake tasks, or receive titles. Such actions bring great misfortune, and all activities should be avoided, especially those related to women. (Yin yang wuxing A, Miscellaneous Divinations II, “Shangshuo.”)

Although these manuscripts do not explicitly mention medical prohibitions on Shangshuo days, medieval Chinese medical and Daoist texts often warned against preparing or consuming medicine on these days, citing ineffective or harmful results:

又云：月建、月殺、反支、天季、上朔、自刑日，此不可用。自刑日者，如寅生人不得用寅和藥、服藥，他准此。（《醫心方》卷二《針灸服藥吉凶日第七》引《大清經》）52

It is also said: Yuejian, Yuesha, Fanzhi, Tianji, Shangshuo, and Zixing days are unsuitable (for medicine). On Zixing (self-injury) days, such as when a person born on a Yin day cannot prepare or take medicine on a Yin day, this rule applies accordingly. (Taiqing jing cited in Ishinpō, Volume 2, “Seventh Section on Auspicious and Inauspicious Days for Acupuncture and Medicine.”)

Another medieval Daoist text similarly lists numerous inauspicious days, including Shangshuo, where medicinal activities are explicitly discouraged:

春戊辰，夏丁已、戊申、己巳、丑未辰，秋戊戌、已亥、辛亥、庚子，冬戊寅、己卯、癸酉、未戌，及壬丙、戊丁、亥土、戌癸、辛巳、日建、日殺、反支、天季、孟仲季月收閉、晦朔、上朔、八魁、往亡、留後日，皆凶，作藥不成矣。……（《三十六水法·作丹忌日》）53

In spring: Wuchen; in summer: Dingyi, Wushen, Jisi, Chouwei; in autumn: Wuxu, Yihai, Xinhei, Gengzi; in winter: Wuyin, Jimao, Guiyou, Weixu; as well as Renbing, Wuding, Haitu, Xujue, Xinsi, Riji, Risha, Fanzhi, Tianji, the end of the first, second, and third months of each season, dark moon, new moon, Shangshuo, Bakui, Wangwang, and Liuhou days are all inauspicious and will result in failed medicine preparation… (Sanshiliu shui fa [A Guide on Preparing Thirty-six Types of Divine Water], “prohibited days for making elixirs.”)

3.1.5. Early Chinese Buddhism and An Shigao’s Expertise

The terms Xueji, Shangshuo, Siji, and Fanzhi in An Shigao’s translations do not directly correspond to the Indian astrological concept of “inauspicious stars for seeking medical assistance.” As one of the earliest translators of Buddhist texts into Chinese, An Shigao likely adapted his translations to accommodate an audience largely unfamiliar with Indian astronomical and astrological traditions. To make these texts more accessible, he replaced Indian terms with analogous Chinese concepts—deities or taboos associated with specific Stems, Branches, or constellations—commonly believed to influence the auspiciousness of human activities. Notably, these four terms were all recognized in medieval China as indicators of inauspicious days for medical treatment.

Beyond these substitutions for the specific names of inauspicious stars, a more precise example of An Shigao’s adaptive approach is evident in his use of the term Xuyu (須臾) as a divinatory method within the phrase “是亦不必日時、漏刻、星宿、須臾” from his T607 (see Table 2). The Houhanshu 後漢書 (Book of the Later Han) identifies Xuyu as a specific form of divination, and Li Xian 李賢 (655–684 CE) in his commentary mentioned a now-lost work titled Wuwang xuyu 武王須臾, which belonged to this category.54 Although transmitted texts offer little insight into the precise nature of such practices, unearthed Qin slips from Shuihudi include texts titled Yu xuyu 禹須臾,55 likely related to Wuwang xuyu. The use of names such as “Yu” (Yu the Great) and “Wuwang” (King Wu) suggests that these forms of divination were attributed to ancient sage kings, lending them legitimacy and authority. The term xuyu, which literally means “moment” or “short time,” likely describes a method for expediently determining auspicious or inauspicious outcomes based on specific dates and times, often presented in tabular formats. For example

戊己、丙丁、庚辛旦行，有二喜。甲乙、壬癸、丙丁日中行，有五喜。庚辛、戊已、壬癸餔时行，有七喜。壬癸、庚辛、甲乙夕行，有九喜。己酉從遠行入，有三喜。（《日書》甲種《禹須臾》）

If traveling at dawn on Wuji, Bingding, Gengxin days, two auspicious events will occur. If traveling at noon on Jiayou, Renji, Bingding days, five auspicious events will occur. If traveling at dusk on Gengxin, Wuxi, Renji days, seven auspicious events will occur. If traveling at night on Renji, Gengxin, Jiayou days, nine auspicious events will occur. If entering home from a long journey on a Jiyou day, three auspicious events will occur. (Rishu jiazhong, “Yu xuyu.”)

The inclusion of divinatory terms such as Xuyu in An Shigao’s translations is unlikely to be coincidental. The divinatory terminology found in T607 should not be regarded as a result of personal idiosyncrasy—or mistranslation—on the part of An Shigao. Rather, such language must be understood within the broader intellectual and cultural context of early Buddhism’s transmission to China. At the time, Buddhist doctrines were frequently mentioned alongside Huang-Lao thought, and foreign śramaṇas were often perceived as akin to fangshi 方士—masters of esoteric arts and techniques (fangshu 方術). This association is clearly reflected in imperial rhetoric; for instance, Emperor Ming of the Han (r. 57–75 CE) referred to Buddhist worship as a form of moral cult comparable to the teachings of Huang-Lao, stating that “the King of Chu recites the subtle words of Huang-Lao and reveres the benevolent cult of the Buddha (楚王誦黃老之微言，尚浮圖之仁祠).” This overlap is also evident in the lexical strategies adopted by early translators such as An Shigao, Lokakṣema, and Dharmarakṣa, who frequently borrowed terms from Daoist texts to render Buddhist concepts—for example, using daoren 道人 to translate brāhmaṇa, dushi 度世 and wuwei 無為 for nirvāṇa, and ziran 自然 for svabhāvatā. What distinguishes T607, however, is that many of its expressions—such as xueji, shangshuo, and xuyu—do not derive from pre-Qin philosophical texts such as the Laozi 老子 or Zhuangzi 莊子, but rather originate in popular religious beliefs and calendrical customs.

An Shigao’s connection with Han dynasty fangshu (esoteric techniques and arts) is far from incidental. Rather than leaving behind only fragmentary traces of such practices in his translations, An Shigao emerges in the historical record as a figure deeply imbued with the aura of divinatory and medical expertise. The earliest account of his proficiency in these domains appears in the preface to the Anban shouyi jing 安般守意經 (Ānāpānasmṛti-sūtra), attributed to Kang Senghui (ca. ?–280 CE), a translator of Kangju origin who was active only a few decades after An Shigao’s career in Luoyang. Kang Senghui’s emphasis on An Shigao’s knowledge of medical and prognostic techniques is not incidental: the meditation method taught in the Anban shouyi jing—focusing on the observation of breath—bears a striking resemblance to the contemporary Chinese practices of breath regulation (daoyin 導引) that are employed by both Daoist and medical traditions for cultivating health and longevity.56 This resonance likely contributed to the text’s early popularity in China, making it one of the first Buddhist scriptures to gain widespread acceptance among Chinese readers:

其為人也，博學多識，貫綜神谋、七正、盈縮、風氣、吉凶、山崩、地動、鍼䘑諸術，覩色知病，鳥獸鳴啼，無音不照。57

He was a man of prodigious learning. He had mastered the full range of divinatory and therapeutic arts: celestial “numinous calculations,” the Seven Regulators that chart the movements of the heavens, techniques for gauging (or prolonging) the span of human life, wind-based augury that reads the four-quarters’ currents for omens of good or ill, the scrutiny of auspicious and inauspicious signs, portents of landslides and earthquakes, and the skills of acupuncture and moxibustion. By a mere glance at a person’s complexion he could diagnose hidden illnesses, and from the cries of birds and beasts he could interpret every omen—there was no sound whose meaning eluded him.

This preface was later adopted by Sengyou 僧祐 (445–518 CE) as a source in An Shigao’s biography, with only minor editorial changes:

七曜五行之象、風角雲物之占、推步盈縮，悉窮其變；兼洞曉醫術，妙善鍼䘑，覩色知病，投藥必濟，乃至鳥獸嗚呼，聞聲知心。58

He excelled in divination through observing the seven stars, five elements, wind directions, and cloud colors. He was proficient in astronomical calendars and predicting lifespans. Additionally, he was skilled in medicine and acupuncture, able to diagnose illnesses by observing a person’s complexion, and his prescribed medicines were always effective. He could even understand the sounds of birds and beasts.

This description was significantly abridged in the Gaoseng zhuan高僧傳 (Biographies of Eminent Monks) compiled by Huijiao 慧皎 (497–554 CE).59 As Salguero (2014, p. 134) notes, figures such as An Shigao—portrayed as monks skilled in healing—appear frequently in medieval Buddhist hagiographies, where they are often credited with mastery of classical medical techniques and concern for lay welfare. However, these accounts were shaped by later compilers with specific ideological aims and frequently rely on now-lost sources.

In this context, An Shigao’s case stands out as particularly valuable: the extant materials not only date relatively close to his lifetime but are also supported by internal evidence from his own translations. The translation techniques that are evident in the Daodi jing lend credence to the Chu sanzang ji ji’s account of An Shigao’s medical proficiency, suggesting that it is not merely hagiographic embellishments but instead retains a substantial degree of historical accuracy. Among An Shigao’s extant works, the Daodi jing may be one of the very few reliable sources for examining this aspect of his legacy. By contrast, many other scriptures attributed to him that concern medical topics—such as T553 Nainü qiyu yinyuan jing 㮈女祇域因緣經 (Āmrapālī and Jīvaka Avadāna Sutra) and T701 Wenshi xiyu zhong seng jing 溫室洗浴眾僧經 (The Sutra on Bathing the Sangha in the Bath House)—are now considered of doubtful authenticity by modern scholars.60

As Salguero (2014, pp. 92–94) observes, Buddhist translators often employed localization strategies to make their texts more accessible to broader audiences, incorporating familiar terminology drawn from indigenous Chinese texts and cosmological frameworks. An Shigao’s deliberate localization of medical terminology in T607, coupled with his documented medical expertise, suggests that he may have been addressing a specific audience: readers who, despite having limited familiarity with Indian astrological traditions, were highly reliant on divination to guide their daily decisions regarding auspiciousness and misfortune. This approach, which contrasts with the critical stance of educated Chinese intellectuals toward such practices, aligns more closely with the common lifestyle of the general populace during the late Eastern Han period. During this time—marked by widespread warfare and disease—practical knowledge, such as medicine and divination, likely served as a key medium for transmitting Buddhist teachings.

In contrast, An Shigao’s other works, such as T1508 Ahan koujie shier yinyuan jing 阿含口解十二因緣經, targeted a more specialized and focused readership. As Zacchetti (2004, p. 221) notes, T1508’s technical language, esoteric style, and lack of rhetorical appeal indicate that it was intended for a well established, possibly smaller Buddhist community, rather than for attracting new followers. These texts exhibit little accommodation to Chinese thought. However, even in T1508 (and other reliably attributed works of An Shigao), Chinese medical terminology occasionally surfaces. For instance, as Loukota (2020) highlights, the term wochu 臥出, which also appears in the Huangdi neijing黃帝內經 (Esoteric Scripture of the Yellow Emperor), is present. More broadly, terms shared between Daoist texts, medical treatises, and Buddhist scriptures—such as Tongyang 痛癢 (“pain and itch”) or Xihua 細滑 (“fine and smooth”)—are often used metaphorically to express Buddhist concepts.

3.2. Medical Specializations and Renowned Physicians

In Table 1, the section “[B] The most skilled physicians” appears exclusively in the translations by An Shigao and Dharmarakṣa, highlighting a significant divergence in their approaches. Following the enumeration of various omens signaling impending death, An Shigao’s T607 concludes with a concise remark referencing Bian Que (扁鵲), a renowned figure from ancient Chinese tradition:

如是病痛相，不可治，設鶣鵲亦一切良醫并祠祀盡會，亦不能愈是。61

After the patient has shown the aforementioned symptoms, it means that he is beyond saving. Even if renowned physicians and wizards like Bian Que were to come, they could not cure him.

The inclusion of “Bian Que” in T607 reflects An Shigao’s deliberate localization strategy, aimed at aiding comprehension for Chinese readers who were likely unfamiliar with the Indian medical system. This approach aligns with his practice of substituting Indian astrological terms with their Chinese counterparts. The choice of Bian Que is particularly significant. In Han dynasty China, Bian Que was not merely a historical figure from the Warring States period (known by his real name, Qin Yueren 秦越人) but was also venerated as a divine physician. He was depicted in Han tomb reliefs as a bird-headed figure with one hand diagnosing the pulse and the other holding a needle (Figure 2), symbolizing his role as a guardian deity ensuring well-being in the afterlife. Additionally, Bian Que was associated with the Western Queen Mother (Xi Wangmu 西王母) as one of the immortals residing in her paradise.62 The mythological attributes associated with Bian Que bear a resemblance to the ṛṣis, such as Ātreya, revered as early transmitters of Indian medical knowledge.

In addition to the mystical attributes associated with Bian Que, An Shigao’s reference to him may also relate to another notable feature—Bian Que was regarded during the Han dynasty as an expert capable of practicing different branches of medicine, such as gynecology, ophthalmology, and pediatrics, depending on regional needs.63 While this aspect is not elaborated in T607 and can only be understood as contextual background rather than definitive intent, Dharmarakṣa’s T606 explicitly provides a comprehensive account of the Indian medical system, corresponding to the first five branches of the Āyurvedic “eightfold medicine” (cikitsāyām aṣṭāṅgāyāṃ):64

• “Body treatment” (療身病, equivalent to kāyacikitsitā, general internal medicine);
• “Treatment of ears and eyes” (主治耳目, śālākya, otorhinolaryngology);
• “Treatment of wounds” (瘡醫, śalya, surgery);
• “Pediatrics” (小兒醫, kaumārabhṛtya);
• “Demonology” (鬼神醫, bhūtavidyā, exorcism and psychiatry).

For each branch, he enumerated representative figures, concluding the extensive list with the following statement: “Even if all these physicians, along with practitioners of magic and sorcery, are united, they cannot prevent death” (正使合會此上諸醫，及幻蠱道并巫呪說，不能使差，令不終亡).65

Dharmarakṣa’s translation uses paraphrasing for the names of these physicians, making it challenging to precisely match them with the ṛṣis referenced in Sanskrit medical texts. Furthermore, the source text for Dharmarakṣa’s translation likely contained numerous elements of Middle Indic languages rather than standard Classical Sanskrit,66 further complicating attempts to identify these figures accurately.

To illustrate these complexities, the discussion below will focus on the first branch, “body treatment”(see

Table 5. Renowned Physicians in the Treatment of Bodily Diseases (Kāyacikitsā).

T606 (CBETA, T15, no. 606, p. 184, c26–185, a8)	English Translation
古昔良醫，造結經文，名曰：	In ancient times, there were many skilled physicians who wrote scriptures, and their names were:
於彼、除恐、	“In There (=Ātreya),” “Remove Fear (=? Abhayadatta),”
長耳、灰掌、	“Growth-ears (=Jatūkarṇa),” “Ashen Palms (=Kṣārapaṇi),”
養言、長育、	“Cultivate Language,” “Nurture (=? Janaka),”
急教、多髯、	“Eagerly Educate,” “Many Beards,”
天又 1、長蓋 2、	“Divine Blessing,” “Long Umbrella,”
大首、退轉、	“Big Skull,” “Retreat,”
燋悴 3、大白 4、	“Emaciated,” “Venus (=? Uśanas),”
最尊、路面、	“Most Revered (=? Vasiṣṭha),” “Road Surface,”
調牛、岐伯、	“Taming Bull (=Gautama 5),” “Qi Bo,”
醫徊、扁鵲，	“Physician Wandering (=Caraka),” and “Bian Que.”
如是等輩， 悉療身病。 於是頌曰：	All of these individuals specialized in treating physical ailments (kāyacikitsitā). A verse praises them in this way:
於彼之等類， 尊法梵志仙， 正救所有果， 及餘王良醫。 此為主成敗 6， 博知能度厄， 愍以經救命， 猶如梵造法。	Brahmin sages revered in the law, like “In There,” and other royal physicians can rescue all from the fruits of karma. They are well-learned, decisive in success or failure, able to assist in overcoming calamities, compassionate in their hearts, saving lives according to the scriptures, just as crafted by Brahma himself.

¹ There are numerous variant readings of “天又.” In the Yuan and Ming editions, it appears as “天友”; in the Imperial Household Agency edition, it is “天人”; in the Shogozō edition, it is rendered as “天反.” ² “長蓋” appears as “長益” in the Song, Yuan, and Ming editions. ³ “燋悴” appears as “顦顇” in the Song, Yuan, Ming, and Imperial Household Agency editions. However, this is merely a difference in character usage and does not affect the meaning of the term. ⁴ “大白” appears as “太白” in the Song, Yuan, and Ming editions and as “太帛” in the Imperial Household Agency and Shōgozō editions. These are common examples of interchangeable homophones, with “太白” being a more plausible reading (referring to Venus). As the name of an ancient physician, it may correspond to Uśanas, who is mentioned in the Nāvanītaka as an ancient physician. See M. Chen (2005, p. 47). ⁵ Here the consonant -t- in Gautama is probably confused with -d- and thus Gautama is rendered as Gau-dama “subdued bull” (or “to subdue the bull”). ⁶ “主成敗” appears as “至誠財” in the Song, Yuan, Ming, and Imperial Household Agency editions. This is likely a scribal error caused by similarity in character forms, although both readings seem to have a certain degree of plausibility.

), as a case study.

3.2.1. Leading Figures in Each Branch

Despite the somewhat perplexing nature of the passage in T606, particularly regarding the names for which only tentative translations can be proposed, there is nonetheless evidence of a connection between T606 and Āyurvedic texts. In the section on “body treatment” (療身病), Dharmarakṣa identified a distinguished physician named “Yubi” (於彼) as the foremost practitioner. While “Yubi” may not immediately seem like a personal name without the Sanskrit reference, a comparison with the Śārdūlakarṇāvadāna and its two Chinese translations, particularly Dharmarakṣa’s T1301 Shetoujian taizi ershiba xiu jing 舍頭諫太子二十八宿經, reveals that “Yubi” corresponds to the illustrious physician Ātreya (see

Table 6. Translations of the name of the physician Ātreya.

Śārdūlakarṇāvadāna 1	kiṃ pūrvaḥ? āha: Ātreyaḥ.
T1301 Shetoujian taizi ershiba xiu jing, translated by Dharmarakṣa 2	時弗袈裟聞說如是，則逆問曰：“仁何種姓？”答曰：“於是。” 3
T1300 Modengjia jing, anonymous 4	蓮華實言：“汝姓何等？”曰：“姓三無”。 5
English Translation	Puṣkarasārin, upon hearing him speak this way, counter-asked, “To which family do you belong?” The reply came, “I am an Ātreya (offspring of Atri).”

¹ Edited by Mukhopadhyaya (1954, p. 44). ² See Note 33. ³ CBETA, T21, no. 1301, p. 415, a17–18. ⁴ The Taishō Tripiṭaka currently attributes this translation to Zhi Qian, but the Chu sanzang ji ji does not record this text. Hayashiya (1945, pp. 524–43) suggested that the translation of this text dates to a period later than Kumārajīva, possibly after the Song and Qi dynasties (541 CE). ⁵ CBETA, T21, no. 1300, p. 404, b12.

).

In the later anonymous translation T1300 Modengjia jing 摩登伽經, Ātreya, etymologically meaning “descendant of Atri,”67 was rendered as “Sanwu” (三無) based on a interpretation of Atri as the negative prefix a- plus tri (“three”). In contrast, Dharmarakṣa’s earlier translation T1301 traced the etymology of Ātreya to atra (“there”), resulting in the translation “Yushi” (於是), which is synonymous with “Yubi” (於彼) in T606.

Ātreya is regarded as the founder of kāyacikitsitā,68 explaining his prominence in T606 under the category of “body treatment.” Following this pattern, we can see that the first name mentioned under each branch in T606 is typically the founder or a key figure in that field. For instance, “Yanshun” (眼眴, literally “blinking eye”) under “treatment of ears and eyes” (主治耳目) likely refers to Nimi,69 a representative figure of śālākya; “Zun Jiaye” (尊迦葉, “Revered Kaśyapa”) under pediatrics (小兒醫) evokes Kaśyapa, who is associated with the kaumārabhṛtya branch and the pediatric text Kaśyapa-samhitā.

The remaining two branches are more complex. “Facai” (法財, “Dharma Wealth”) under “treatment of wounds” (瘡醫) might correspond to Dhanvantari, the founder of śalya, given that the Sanskrit dhanvan (“bow”) and dharma are easily confused in Middle Indic languages (e.g., both could be pronounced as dhamma in Pali). The rationale for cai (財, wealth) remains unclear. Additionally, “Daihua” (戴華, “Adorned with Flowers”) and “Bu Shihuo” (不事火, “Non-Fire Worshipper”) under “demonology” (鬼神醫) are difficult to identify definitively.

3.2.2. Other Notable Figures

Building on these identifications, further connections may be drawn between other figures listed in T606 and individuals mentioned in early Āyurvedic texts. Some of these names might refer to disciples of the aforementioned founders. In the same list of the “body treatment” branch, the name “Zhang Er” (長耳) follows “Yubi” (於彼). While this name might readily be interpreted as “one with long ears,” pronounced Chang’er in Chinese, further research may reveal that this is not the case. Given its likely synonymy with “Sheng Er” (生耳) as found in T1301 (see

Table 7. Translations of the name Jātūkarṇya.

Śārdūlakarṇāvadāna	pūrvabhādrapadānakṣatraṃ dvitāraṃ padakasaṃsthānaṃ triṃśanmuhūrtayogaṃ māṃsarudhirâhāramahirbudhnyadaivataṃ jātūkarṇyaṃ gotreṇa ｜
T1301, by Dharmarakṣa	前賢迹宿者，有二要星，相遠對立，行三十須臾，而侍從矣，餅肉為食，主人是天，姓生耳。(CBETA, T21, no. 1301, p. 415, c26–27)
T1300, anonymous	室有二星，形如人步，一日一夜，與月共行，血肉祠祀，其宿屬在富單那神，姓闍罽那。(CBETA, T21, no. 1300, p. 405, a20–22)
	Pūrvabhadrapāda, two stars, shaped like footprints, correspond to thirty muhūrtas. Meat and blood are its food, Ahirbudhnya is the ruling deity, and it belongs to the Jātūkarṇya gotra

), it should be more appropriately read as Zhang Er, conveying the meaning “growing ears.”70

In T1301, “Sheng Er” (生耳) is an interpretive translation of the gotra Jātūkarṇya, where “sheng” (生) and “zhang” (長) are synonymous. According to the Suśrutasaṃhitā, Jātūkarṇa (or Jātūkarṇya) was one of Ātreya’s six disciples71 and was credited with authoring a now-lost compendium (saṃhitā) on kāyacikitsā (Meulenbeld 1999, p. 161). As a result, “Zhang Er” (長耳) follows “Yubi” (於彼) in T606, also classified under “body treatment” (療身病).

Following this pattern, we can identify “Huizhang” (灰掌, “Ash Palm”), who appears after “Zhang Er,” as Kṣārapaṇi, another disciple of Ātreya. Similarly, “Yi Huai” (醫徊) could correspond to Caraka, a successor of Ātreya’s disciple Agniveśa, given that both “Huai” and “Caraka” literally convey the meaning “to wander.” Table 5 provides annotations for some of these hypothesized identifications; however, it is crucial to note that many of these names remain challenging to verify with certainty.

3.2.3. Dharmarakṣa’s Translation Strategies and Audience Adaptation

Despite the absence of the Yogācārabhūmi’s original source text, the names listed in T606—though perplexing at first glance—become more intelligible when situated within the framework of Indian traditional medicine. Viewed through this lens, the text reveals a clear structure of medical specializations and representative figures for each discipline. Dharmarakṣa appeared to have made a concerted effort to introduce this system to his audience, a stark contrast to An Shigao’s approach, which replaced such references with the familiar figure of the Chinese physician Bian Que and treated the subject in a cursory manner.

This divergence can, in part, be attributed to the differing historical contexts and target audiences of the two translators. The social networks surrounding An Shigao after his arrival in Luoyang remain largely undocumented, apart from the claim of a single Chinese disciple, Yan Fotiao 嚴佛調 (also written Yan Futiao 嚴浮調), who identified himself as a follower. By contrast, the biographical section on Dharmarakṣa in the Chu sanzang ji ji, along with various prefaces to his translations, provides considerable detail about his extensive network of disciples and assistants. It describes him as being surrounded by a community that “numbered in the thousands, comprising monks and followers from near and far, all gathered to revere him (於是德化四布，聲蓋遠近，僧徒千數，咸來宗奉).”72 This diverse group included Chinese scholars and commoners as well as individuals from regions such as Kucha, Parthia, Sogdia, Khotan, Gandhāra/Kashmir, India, and the Yuezhi.73

It is plausible that, within this multicultural milieu, especially among disciples from India and Central Asia, there were individuals with some knowledge of Āyurvedic medicine, even if only in its most rudimentary form. For these disciples, references to ancient Indian physicians in the Yogācārabhūmi might have been as familiar and resonant as Bian Que was to Eastern Han Chinese readers. More importantly, by Dharmarakṣa’s time, the primary audience for his Chinese translations had developed a far greater capacity for engaging with lengthy and complex Buddhist texts compared with audiences during the late Eastern Han period. This evolution in readership is evidenced by Dharmarakṣa’s ability to translate an impressive 154 Buddhist texts over approximately four decades, including lengthy works such as the T222 Guangzan jing 光讚經 (Larger Prajñāpāramitā) and T263 Zhengfa hua jing 正法華經 (Saddharmapundarīkasūtra). This accomplishment reflects not only the extraordinary efforts of Dharmarakṣa and his team but also the growing demand among Chinese readers for comprehensive Buddhist scriptures.

It is also noteworthy that T606 features renowned ancient Chinese physicians such as Bian Que and Qi Bo 岐伯 (see Table 4), figures frequently cited in that classical Chinese medical literature, analogous to the way that Ātreya is referenced in Sanskrit medical texts. The inclusion of “Bian Que” in T606 is likely the result of Dharmarakṣa drawing upon An Shigao’s earlier translation (see Section 2.3), but the addition of “Qi Bo” was clearly Dharmarakṣa’s own decision. This demonstrates that Dharmarakṣa did not uncritically accept elements from An Shigao’s translations; rather, he adopted and adapted An Shigao’s localization strategies to suit his own objectives.74 Dharmarakṣa’s translation strategy appears to have been a balanced one: on the one hand, he sought to convey the original text’s medical system comprehensively and accurately, ensuring that Chinese readers could gain a thorough understanding of its content; on the other hand, he selectively incorporated localized elements to enhance accessibility.

Early Buddhist translators were not merely engaged in linguistic conversions but actively sought to bridge cultural and historical gaps between Indian and Chinese traditions by employing analogous concepts. Traditionally, such practices have been critiqued as the result of misunderstandings or limitations imposed by linguistic and cultural differences. Some scholars have gone further, questioning the authenticity of these texts due to the presence of “Chinese elements.”75 However, such simplistic judgments may overlook the complexity of the issue and underestimate the agency of translators such as Dharmarakṣa. The translation in T606—undoubtedly a translated text rather than an apocryphal work—demonstrates that Dharmarakṣa’s inclusion of ancient Chinese physicians was a deliberate and strategic choice: while he strove to faithfully transmit a complex array of foreign medical concepts—despite these being embedded within the Yogācārabhūmi’s detailed descriptions of the suffering associated with birth, aging, illness, and death, where a more concise treatment, as in An Shigao’s translations, might have sufficed—he did not shy away from integrating localized elements. Given the nature of his audience, we can envision a readership that was already equipped to engage with lengthy and complex Buddhist texts—one that was diverse in ethnicity, language, and intellectual backgrounds. For such an audience, Dharmarakṣa’s strategy was likely the most effective way to ensure both accessibility and comprehension.

4. Conclusions

The parallels between the Yogācārabhūmi and the Sanskrit medical texts provide significant new insights into these sources. At a fundamental level, the complexities and ambiguities present in the Chinese translations, especially An Shigao’s T607 and Dharmarakṣa’s T606, can now be substantiated with compelling corroborative evidence. Notably, the archaic and intricate linguistic style of An Shigao’s translations has rendered them difficult for later scribes, editors, and publishers to modify based on contextual understanding, preserving numerous linguistic features from the Eastern Han period. These include rare terms that fell out of use in later periods, as well as instances of confusion between “xiang” (相) and “shuo” (朔) in the handwritten scripts of the time.

The early translation dates and reliable documentation of T607, T606, and T374 offer substantial evidence for the early development of Āyurvedic medical concepts. Despite the fact that the extant Caraka-saṃhitā and other Sanskrit medical texts were compiled several centuries after An Shigao’s time, the sections concerning “omens of death” exhibit a significant degree of alignment with T607, suggesting that these ideas had already reached a formative stage by the mid-2nd century CE.

The discovery of parallel Sanskrit passages also enriches our understanding of early Buddhist translators. While most of An Shigao’s translations are brief and primarily focus on listing Buddhist terminology, T607 stands out for its engagement with various aspects of secular life, providing valuable insights into his translation approach. As one of the earliest translators, An Shigao largely replaced foreign medical and divinatory terms with indigenous Chinese concepts to make his translations more accessible to readers unfamiliar with Āyurvedic knowledge.76 His apparent familiarity with contemporary Chinese practices of auspicious date selection and divination is corroborated by his biography in the Chu sanzang ji ji, which describes him as skilled in divination, acupuncture, and pharmacology.

Although Dao’an did acknowledge that An Shigao’s translations could at times be difficult to comprehend,77 he consistently expressed high regard for An Shigao’s translation. Dao’an praised An Shigao’s faithful rendering of Indian texts, valuing clarity and simplicity over rhetorical embellishment.78 His appreciation was not merely stylistic but deeply devotional: Dao’an repeatedly extolled An Shigao’s translations for their elegance, precision, and spiritual depth.79 He even claimed that receiving An’s teachings in person would have been tantamount to meeting the Buddha himself.80 This high esteem from Dao’an stands in stark contrast to the assessments of some modern scholars. Zürcher (1991, p. 11), for instance, describes An Shigao’s translation style as “erratic, crude, full of vulgarisms, often chaotic to the point of unintelligibility,” suggesting that such praise from later figures such as Dao’an may have been motivated more by traditional reverence for ancient works than by a critical assessment of their actual merit (Zürcher [1972] 2007, p. 34). Such conflicting evaluations may partly stem from the vastly different linguistic and cultural contexts in which An Shigao operated—contexts that may remain obscure to modern researchers. However, it is evident that An Shigao was deeply familiar with and well adapted to the cultural milieu of Han dynasty China. For his contemporaneous audience, his translations represented a groundbreaking achievement, one so influential that their historical significance endured even after Dharmarakṣa produced more polished and faithful translations.

In comparison, Dharmarakṣa’s translations exhibit a much higher level of fluency and idiomatic accuracy in Chinese while also adhering more closely to the Indian source texts in terms of content detail. In terms of translation quality, Dharmarakṣa clearly represents a significant step forward. However, it is only fair to consider the differing historical contexts and audiences that they faced, which likely account for the distinct approaches in their translation strategies.

While Dharmarakṣa made extensive updates to An Shigao’s translations, he also preserved and even deliberately adopted some practices that, by modern standards, might seem questionable—for example, rendering Indian physicians as the Chinese figures Qi Bo and Bian Que. His simultaneous use of both domestication and foreignization—two seemingly opposing translation strategies—offers a glimpse into how this multilingual “Dunhuang Bodhisattva” achieved what might otherwise seem impossible: attracting a large and diverse group of disciples and assistants from entirely different ethnic, cultural, and linguistic backgrounds and leading them to produce an extraordinary number of significant Buddhist translations at an astonishing pace.

The relationship among the three Chinese parallel texts is both complex and revealing. The Chu sanzang ji ji includes a preface to the Yogācārabhūmi written by Dao’an, which states that the text was compiled by Saṅgharakṣa through the synthesis and elaboration of multiple scriptures, extracting their essential teachings.81 A similar description appears in the 384 CE preface to T194 Sengqieluocha suoji jing 僧伽羅剎所集經 (The Collected Life Stories of the Buddha by Saṅgharakṣa).82 These accounts suggest that the Yogācārabhūmi functions as a compilation of various scriptures. Its “omens of death” section likely reflects Saṅgharakṣa’s integration of medical knowledge with Buddhist reflections on the suffering of death. What is particularly noteworthy, however, is that although the Yogācārabhūmi translations by An Shigao and Dharmarakṣa and the T374 Mahāparinirvāṇa Sūtra translated by Dharmakṣema, do not, in their entirety, belong to the same textual tradition—and Dharmakṣema appeared not to have consulted or relied heavily on the earlier translations83—these three Chinese versions exhibit a consistent structural correspondence in the “omens of death” section. This consistency suggests that they may have been influenced by a shared medical tradition. It is plausible that this specific content originally circulated as a standalone “physician’s manual” before being incorporated into larger texts such as the Yogācārabhūmi and the Mahāparinirvāṇa Sūtra.