Rediscovering the Textual Sources of the “De Dashizhi Pusa 得大勢志菩薩” in Cave 169 of Bingling Temple ^†

Abstract

Niche 6 in Cave 169 of Bingling Temple 炳靈寺 contains the earliest surviving Amitāyus sculpture triad in cave temples of China. This paper attempts to re-identify the textual sources of the ink inscription “De Dashizhi Pusa 得大勢志菩薩” (Skt. Mahāsthāmaprāpta Bodhisattva) from the Amitāyus sculpture triad in Cave 169. At first, it corrects the misidentification of the inscription “De Dashizhi Pusa” in recent decades. Then, it discusses the relationship between multiple scriptures and “De Dashizhi Pusa”. Considering the configuration of the niches, the theme of the sutra, the translation and transmission history of the sutra, and the content of the statue, this paper concludes that the inscription “De Dashizhi Pusa” and niche 6 should have been significantly influenced by the Lotus Sutra 法華經, although this paper does not deny the indirect and partial influence from the Pure Land texts on the Amitāyus sculpture triad during its restoration process. Such a discovery not only sheds light on the configuration of the Lotus Sutra images, but also provides art historical evidence for future exploration of the relationship between the Lotus Sutra and Pure Land Buddhism.

1. Introduction

Cave 169 of Bingling Temple 炳靈寺1 is one of the richest caves in the northwest China. Apart from the cave ceiling, most of the images were completed in the Western Qin dynasty (385–400, 409–431), with a few created in the Northern Wei dynasty (386–534) (Chang 1992, p. 416). The first contemporary survey of this cave was conducted in 1963 (Dong 1963, pp. 46–50), while studies over the past decades have been carried out from the perspectives of chronological analysis, identification of inscriptions, and artistic style of images.2 Of the 24 numbered niches in this cave, niche 6, which contains the earliest surviving Amitāyus sculpture triad 無量壽佛三聖3 (hereafter: AST) in Chinese cave temples, attracts extensive attention in the fields of Buddhist history and art history. In the past 60 years, it is generally accepted that the textual source of the AST in niche 6 is the Wuliangshou jing 無量壽經 (hereafter: SIL) (T12n0360)4. Such judgement is based on the fact that the “Wuliangshou fo 無量壽佛” (Skt. Amitāyus) in the inscription of the AST is consistent with the sutra name of the SIL. However, such an analysis is not supported by the evidence derived from the content of Buddhist texts and the history of sutra transmission. Moreover, considering the time of the image, scholars argued that the date recorded in the Jianhong Reign Inscription 建弘題記 (hereafter: JRI) corresponds to the time of the creation of the AST, i.e., 420 or 4245. This analysis seems reasonable; however, it ignores the phenomena of restoration and overlapping in this niche. These factors led to a fragmented identification of the textual sources and themes of the images. It is worth noting that previous studies also failed to correctly record the inscription “De Dashizhi Pusa 得大勢志菩薩” (Skt. Mahāsthāmaprāpta Bodhisattva) in the AST, which also affects the identification of its textual sources.

The early fifth century was a period of intense translation and dissemination of the Mahayana sutras in China. It needs to be noted that the sectarian model 宗派模式 was not formed during this period. The boundaries of the subjects covered by various sutras, and in particular the identity of images from the perspective of Buddhist art, are not very clear. However, this feature has not received much attention from art historians, and there is a potential for misinterpretation of the textual origins of the images of this period.

In terms of research methodology, there are two issues with previous studies when it comes to interpreting early cave temples of China. The first is over-interpretation. The second is that, in the absence of relevant textual evidence, the interpretation is based on a subjective connection between texts and images. This paper attempts to address the second issue by taking the inscriptions of Cave 169 as an example. It focuses on the textual origins of the “De Dashizhi Pusa 得大勢志菩薩” in niche 6 regarding the identification of inscriptions, the restoration of images, and the configuration of their content. This paper provides new ideas for future research on images of the Fahua jing 法華經 (hereafter: LS)6 and Pure Land Buddhism by exploring the textual sources of the “De Dashizhi Pusa 得大勢志菩薩” in Cave 169.

2. The Ink Inscription of “De Dashizhi Pusa 得大勢志菩薩” in the Amitāyus Sculpture Triad

Niche 6 is located on the north wall of Cave 169 and dates to the Western Qin dynasty (385–400, 409–431). It contains the images of the AST, Maitreya Bodhisattva, Śākyamuni Buddha, and a group of ten Buddhas (Figure 1). To the right of Śākyamuni Buddha there is the JRI (Figure 2), the earliest known inscription with a clear chronology in cave temples of China. Below the JRI, there are several murals of the donors7.

In niche 6, the one most discussed by scholars is the AST (Figure 3). The center of the AST is a seated Amitāyus Buddha, with a standing Avalokiteśvara Bodhisattva on the left and a standing Mahāsthāmaprāpta Bodhisattva on the right. The Buddha in the center is seated in a full lotus position and leaning against the backscreen.

The Buddha’s face and clothes are color painted, with hexagon-shaped patterns decorated on the sankaksika 僧祗支8. There are many celestial musicians in the aureole9. The identity of the three statues, especially Amitāyus, is revealed by the corresponding ink inscription (Figure 4).

However, the identification of the ink inscription of Mahāsthāmaprāpta Bodhisattva in previous studies is inaccurate. The inscription was read as “DaShizhi Pusa 大勢至菩薩” in the early stages of the discovery of Cave 169 by scholars (Du 1985, p. 86; Zhao 1994, p. 147; Sun and Gao 2009, p. 57). Subsequent studies generally identify it as “De Dashizhi Pusa 得大勢至菩薩”10 (Zhang 1992, p. 12; Rhie 2010, p. 174; Zheng 2021, p. 126). When I magnified the Chinese character of the inscription of Mahāsthāmaprāpta Bodhisattva in niche 6, I could recognize the word “De 得” (Figure 5). However, the Chinese character “Dashizhi 大勢志” (Figure 5) is not the “Dashizhi 大勢至”11. Both the omission of the “De 得” and the misidentification of the “zhi 志” in previous studies have influenced the judgment on the textual origin of the AST.

3. Text Consideration of the Amitāyus Sculpture Triad in Niche 6

3.1. The Consideration from the Da Amituo Jing 大阿彌陀經

Due to the presence of the ink inscription “Amitāyus 無量壽佛” in niche 6, much of the discussion of this cave has centered on the SIL. Almost all scholars considered the canonical source of the AST to be the Pure Land sutras, and more specifically, they pointed to the SIL (Rhie 2010, p. 302; Wang 2020, p. 49) and the Guanwuliangshou jing 觀無量壽經 (hereafter: ADS) (T12n0365)12 in the absence of scriptural evidence (Lai 1990, p. 169).

The SIL is often written as the Da Amituo Jing 大阿彌陀經 (hereafter: LSukh)13 to distinguish it from the Xiao wuliangshou jing 小無量壽經 (hereafter: SSukh)14 in ancient Chinese script. The records of different versions of the LSukh from the Chusanzangji ji 出三藏記集 (hereafter: CSZJ) (T55n2145)15, the Kaiyuan shijao lu 開元釋教錄 (hereafter: Kaiyuan Lu) (T55n2154)16, and the Gaoseng zhuan 高僧傳 (hereafter: BEM) (T50n2059)17 are not consistent. For instance, the CSZJ incorrectly mix the sutra names of the LSukh and the Amituo Jing 阿彌陀經 and classify both of them under the sutra name of the LSukh. The Kaiyuan Lu (T55n2154, pp. 586, 626, 701) records eleven translations of the LSukh, four of which are mentioned as being in the collection, while seven of them are lost.

The language and content of various versions of the LSukh have been discussed in detail by scholars18. In terms of the timing of translation, only three versions may be linked to Cave 169. They are the Amituo sanyesanfo saloufotan guodu rendao jing 阿彌陀三耶三佛薩樓佛檀過度人道經 (hereafter: Rendao Sutra) (T12n0362), the Wuliangqingjing pingdengjue jing 無量清淨平等覺經 (hereafter: PEM) (T12n0361), and the SIL (T12n0360). The focus of this section is to discuss the potential connections between these surviving three versions of the LSukh and the AST in niche 6.

The Rendao Sutra (T12n0362) consists of two volumes. The first volume describes the eight directions with the upper and lower worlds but does not mention the names of the Buddhas in these directions. It focuses on the twenty-four vows of Amitābha and the merits and majesty of Amitābha Buddha Land. Volume two describes three levels of rebirth, five evils, and the importance of eliminating doubts of Amitābha Buddha Land. In terms of content, although Maitreya (Bodhisattva) appears in the second volume, the sutra contains no reference to Avalokiteśvara and Mahāsthāmaprāpta, nor does it mention the AST. Moreover, the sutra uses Amitābha instead of Amitāyus.

In contrast to the Rendao Sutra, the PEM (T12n0361) has four volumes. However, the length and content for both are basically the same. Neither Avalokiteśvara nor Mahāsthāmaprāpta appear in the PEM. Moreover, there is a mixed use of Wuliangqingjing fo 無量清淨佛19 and Amitābha in the PEM. The Amitāyus is also not available in the PEM. As discussed by Harrison et al. (2002, p. 180), “they (the PEM and Rendao Sutra) are not entirely separate renditions, but are, in most respects, different versions of the same text.” Considering the contents of the sutras, the Rendao Sutra and the PEM do not fit in with the images and inscriptions of the AST.

Most scholars argued that the AST originated from the SIL. The content of the SIL differs significantly from previous two surviving versions. As explained by Harrison et al. (2002, p. 181), “a third recension (T12n0360) of the LSukh which agrees with no known version of the text”, which means the source of the original text of the SIL is not identical with the PEM and the Rendao Sutra.

Regarding the connection between the AST and the SIL, it can be judged from the Chinese characters in inscriptions. First, compared with the Rendao Sutra and the PEM, only the SIL explicitly mentions Avalokiteśvara and Mahāsthāmaprāpta. However, not only does the “Dashizhi 大勢至” in the SIL lack the word “De 得”, but the Chinese character “zhi 至” in the sutra does not match the “zhi 志”in the inscription. Second, the eight directions plus the upper and lower worlds in the Rendao Sutra and the PEM become the six directions in the SIL. They are east, south, west, and north (four dimensions) plus upper and lower worlds. None of them are connected to the murals of ten Buddhas on the top area of this niche.

Thus, considering the degree to which the SIL matches the AST and the rest of niche 6, I consider that the influence of the SIL on the AST cannot be ruled out. However, since there is a significant distinction between the “Dashizhi 大勢至” in the SIL and the “De Dashizhi 得大勢志” in the inscription, the influence of the SIL on the AST should be indirect. There must be other textual sources that influenced the composition or restoration of the AST.

3.2. The Amituo jing 阿彌陀經 (T12n0366)

As recorded from the CSZJ (T55n2145, p. 14), there is “a single volume of the Wuliangshou jing 無量壽經 from Kumārajīva……a single volume of the Wuliangshou jing 無量壽經 from Guṇabhadra 求那跋陀羅”. The single-volume version of the Wuliangshou jing 無量壽經 mentioned in the CSZJ, translated from Kumārajīva 鳩摩羅什 (344–413) and Guṇabhadra 求那跋陀羅 (394–468), respectively, is not the LSukh, but the Amituo jing 阿彌陀經, also known as the SSukh 小無量壽經 in some Buddhist records for the purposes of distinction.20

The version translated by Guṇabhadra 求那跋陀羅 is lost. He arrived in Guangzhou in 435 and began translating the sutras. It suggests that his translation of the SSukh would have been after 435. For example, he translated the Za’ahan jing 雜阿含經 (T02n99)21 and the Lengjia jing 楞伽阿跋多羅寶經 (T16n0670)22, both in 443. Given the date of translation, there should be no direct connections between his (lost) version of the SSukh and niche 6.

The version translated by Kumārajīva 鳩摩羅什 is also named the SSukh in the BEM (T50 n2059, p. 332). The Kaiyuan Lu records Kumārajīva translated a single volume of the Amituo jing 阿彌陀經 (hereafter: AMS) (T12n0366)23, “also known as the Wuliangshou jing 亦名無量壽經” (T55n2154, p. 595). The Wuliangshou jing 無量壽經 in this occasion is not the SIL (T12n0360), but the AMS (T12n0366).

The AMS describes the Land of Utmost Bliss 極樂淨土 (Skt. Sukhāvatī) and the importance of making vows to be reborn in the Western Pure Land. It does not describe Amitābha Buddha too much, nor is there any reference to two Bodhisattvas: Avalokiteśvara and Mahāsthāmaprāpta. Moreover, the AMS uses the Amitābha instead of Amitāyus, and it focuses on the description of majesty and merit of Amitābha Buddha Land. Since Amitāyus appears in niche 6 as the AST, the content of the AMS does not quite match the depiction of this niche. Moreover, the AMS has very clear descriptions of the names of the Buddhas in the six directions—east, south, west, north, lower, and upper—which is not consistent with the names of ten Buddhas in the upper part of niche 6. Therefore, the AMS (T12n0366) cannot be accepted as the source of the AST from the perspective of textual content.

3.3. The Guanwuliangshou Jing 觀無量壽經 (T12n0365)

Scholars claimed that the AST were created according to the ADS and other visualization sutras. Thus, they hold the view that these images were used for visualization. For instance, Lai (1990, p. 169) argued that the source of the halo of statues such as Amitāyus in Cave 169 comes from sutras such as the ADS. The author tried to establish a relationship between various meditation methods and images. However, this argument could only be taken as a speculation with strong personal preference and lacks scriptural evidence to support it.

For Chinese translation, the BEM (T50n2059, p. 343) records that Kalayashas 畺良耶舍 (383–442) translated the ADS in the early years of Yuanjia 元嘉之初 but does not indicate the exact time. The Fozu tong ji 佛祖統紀 (T49n2035, p. 344)24 records the date of its translation as the first year of Yuanjia 元嘉元年 (424). However, such a date is disputed by scholars. For instance, Fukuhara and Shi (2017, p. 14) claimed that the translation took place during the Yuanjia period (424–453) but did not give a specific date. Hiraoka (2018, p. 71) claimed that the ADS was translated during 430–442; however, the author did not provide sufficient evidence to support his argument.

Chronologically, even if the translation is dated to 424, the earliest date shown in various records, it is unlikely that the AST in Cave 169 is associated with the ADS. First, since niche 6 was completed in 424 at the latest, it is unlikely that the sutra was translated after August 42425, spread from Jiankang 建康 to Bingling Temple, and the statue creation completed in the same year. More critically, neither the content nor the style of the iconography presents sufficient connotation with the ADS.

In terms of the content, the visualization method/object in the sutra also does not match the details of the images. First, there are many objects to be observed in the ADS; for example, the thirteen contemplations start with a simple contemplation of the setting sun 日觀 and become complex, gradually moving to the contemplation of Amitāyus and two Bodhisattvas 雜想觀. Each of these contemplations has a specific/major object. For instance, in the 10th contemplation (Eng. Inagaki and Stewart 2003, p. 76; Chi.T12n0365, p. 343): “Bodhisattva Avalokiteśvara… body is the color of purple-gold, and on the top of his head is a mound surrounded by an aureole with a radius... he wears a heavenly crown made of śakra-abhilagna-maṇi-gems, on which stands a transformed buddha”. In the 11th contemplation (Eng. Inagaki and Stewart 2003, p. 78; Chi.T12n0365, p. 344): “The mound on his head, shaped like a lotus bud, has a jeweled vase in front... The rest of the characteristics of his body are exactly the same as Avalokiteśvara’s”. In the 13th contemplation (Eng. Inagaki and Stewart 2003, p. 79; Chi.T12n0365, p. 344): “The figures that he manifests are all of the color of pure gold. The transformed buddhas and jeweled lotus flowers in the aureole of each manifested form are like those described above”.

The above texts regarding the 10th, 11th, and 13th contemplations are directly related to the visualization of Amitāyus, Avalokiteśvara, and Mahāsthāmaprāpta in the ADS. However, the key depictions from those contemplations such as the transformed Buddha on the heavenly crown of Avalokiteśvara, the lotus-like vase on the head of Mahāsthāmaprāpta, the true golden color of the image of Amitāyus, the round light transformed Buddha, and the lotus flower are not represented in the AST in niche 6. In addition, the Bodhisattva “Dashizhi 大勢至” in the sutra does not correspond to the ink inscription “De Dashizhi 得大勢志”.

Thus, no direct link can be constructed between the ADS and the AST in Cave 169, either in terms of the timing of translation or the depiction of images.

3.4. The Consideration from the Foming Jing 佛名經 (T14n0441), the Wuliangshoujing Yiji 無量壽經義記 (T85n2759), and the Fahua Wenju Ji 法華文句記 (T34n1719)

The ink inscription “De Dashizhi 得大勢志” in niche 6 does not appear in any of the existing Chinese Tipitaka in the period of Sixteen Kingdoms (304–439) and Northern dynasty (439–581). In addition to “Dashizhi 大勢至”, which is adopted by the SIL, there are two variants of “De Dashizhi 得大勢志” available for reference before Pure Land Buddhism became popular. They are “Dashizhi 大勢志” and “De Dashi 得大勢”. The “Dashizhi 大勢志” is found in the 30-volume version of the Foming Jing 佛名經 (hereafter: BNS) (T14n0441)26, the Wuliangshoujing yiji 無量壽經義記 (T85n2759), and the Fahua wenju ji 法華文句記 (T34n1719).

Few scholars have discussed the relationship between the BNS and the AST in niche 6. Moreover, the sequence of several versions of the BNS and the authenticity for some of them are also in doubt.

The BNS is available in various versions: 12-volume version, 16-volume version, 18-volume version, 20-volume version, and 30-volume version. Among them, the “Dashizhi大勢志” is only available in the 30-volume version (unknown date/translator, T14n0441). Although the translation time of the 30-volume version cannot be fully determined, there is some indirect evidence that the 30-volume version is the latest version of the BNS. For instance, the 12-volume version is recorded in the Lidaisanbao Ji 歷代三寶記 (T49n2034)27. It is recorded as a translation by Bodhiruci 菩提流支 in the Zhengguang period 正光年間 (520–525) (T49n2034, p. 45). Inokuchi (1964) pointed out that the 18-volume text and the 16-volume text were produced from the 12-volume text, and then the 30-volume text from the 18-volume text. Fang (1990, p. 474) considered the relationship between various versions as: 12 volumes → 20 volumes → 16 volumes → 30 volumes. Zhang (2008, p. 47) pointed out that the 20-volume text originated from the Sui dynasty (581–518), and he argued that the 20-volume text was written based on the 12-volume text. He also pointed out that the beginning part of the 20-volume text was also copied from the 12-volume text. Thus, based on the corresponding time between different versions, the 12-volume one should be the earliest, with a noted translation period and translator; the 30-volume version should be the latest one, and it should have appeared in or after the Sui dynasty (581–518). Moreover, it is generally accepted that the 30-volume version of the BNS is an apocryphal sutra 偽經 because it is not found in the Chinese collection of Tripitaka, nor in any Chinese sutra records. It is only found in the Tripitaka Koreana 高麗大藏經 and available in the Taishō Tripitaka (T14n0441).

The authenticity of other versions of the BNS also has been questioned by scholars. For instance, the Dunhuang jieyu lu 敦煌劫餘錄 (Chen 1931) categorized the BNS in the Bei-Tu Tripitaka collection 北圖藏 into the 12-volume version text translated by Bodhiruci 菩提流支, and another three lost texts, including the 16-volume version text, the 20-volume version text, and the 30-volume version text. Inokuchi (1959, pp. 607–9; 1964, pp. 397–437) challenged Chen’s view and argued that the 20-volume version text did not exist, and that what did exist was the 18-volume version text. Zhang (2008, pp. 1–2) argued that the 20-volume, the 16-volume, and the 30-volume texts are all apocryphal sutras.

Thus, according to the research and judgement of scholars, the 30-volume version of the BNS would be the latest edition, which is judged to be roughly dated in or after the Sui dynasty (581–618). Moreover, it is generally accepted that the 30-volume version of the BNS is an apocryphal sutra. Therefore, there is no direct connection between “Dashizhi 大勢志” in the 30-volume version of the BNS and “De Dashizhi 得大勢志” in niche 6 in terms of the period of sutra production and its authenticity.

Except the BNS, the “Dashizhi 大勢志” is also available in the Wuliangshoujing yiji 無量壽經義記 (T85n2759) and the Fahua wenju ji 法華文句記 (T34n1719). Since the Wuliangshoujing yiji 無量壽經義記 is a fragmentary volume, the author and the period in which it was written are no longer available. As claimed by Mochizuki (1974, pp. 129–30): “Like the Guanjing yiji 觀經義記, it bears the title ‘Yiji 義記’28 and has a similar style of interpretation, so we believe that it also should be considered a work of Lingyu 靈裕 (517–605).” However, Liu (2000) claimed that they were not written by the same person, because there are cases where the two Yiji 義記 have completely contradictory views on the same subject. Liu (2000) assumed that the Wuliangshoujing yiji 無量壽經義記 was probably written between the Southern and Northern dynasties or even the Sui and early Tang dynasties.

In terms of content, this article tends to support Liu’s argument. However, the production of Yiji should be dated to the sixth century or later. While Yiji contains several references to the SIL, it also places great emphasis on the content of the Ten Grounds 十地. The Shidijing lun 十地經論 (T26n1522)29 is an interpretation of the Shidi ping 十地品30 of the Dafangguang fo huayan jing 大方廣佛華嚴經 (hereafter: AS) (T09n0278)31 by the great Indian Buddhist scholar Vasubandhu 世親. It was translated by Ratnamati 勒那摩提 and Bodhiruji 菩提流支 in the fourth year of Yongping (511). Only then did Huayan 華嚴 scholars in China begin to devote themselves to studying the Ten Grounds of the AS.

Thus, there is no connection between the AST in niche 6 and the Wuliangshoujing yiji 無量壽經義記, either in terms of the timing when the text was created or in terms of its content. Since the Fahua wenju ji 法華文句記 (T34n1719) is written by Zhanran 湛然 (711–782) of the Tang dynasty (618–907), it is too remote to our discussion.

Therefore, since “De Dashizhi 得大勢志” does not exist in any of the surviving Chinese Buddhist text, this section attempts to identify a relevant source of its variation: “Dashizhi 大勢志”. Although it appears in the 30-volume version of the BNS, the Wuliangshoujing yiji 無量壽經義記 and the Fahua wenju ji 法華文句記, they are not considered to be the textual sources of “De Dashizhi 得大勢志” in niche 6, either in terms of timing of translation or the content of the text.

4. The New Discovery Based on the Configuration of the Buddha Niches Surrounding the Amitāyus Sculpture Triad

In niche 6, the inscription “De Dashizhi 得大勢志” is not available in any of the extant Buddhist textual sources originating by the early fifth century. For this reason, we try to identify a source from the Pure Land sutras that is most frequently discussed in the previous studies. However, a number of Pure Land sutras that have been discussed by scholars in recent decades, such as the LSukh, the AMS, and the ADS, cannot be considered as the direct source of the AST. After checking multiple versions of the BNS, I found “Dashizhi 大勢志”, which is a variant of “De Dashizhi 得大勢志” in Chinese, in the 30-volume version of the BNS. However, not only did this version come out at least 150 years later than the creation of niche 6, but it has been considered by scholars as an apocryphal scripture. In this case, therefore, we need to further explore the textual origins of the AST in the context of other clues in this cave, especially the configuration of niche 6 and adjacent Buddhist niches with relevant scriptures.

Regarding the textual origins of the images in niche 6, previous studies have focused on constructing the relationship between the Pure Land sutras and the image of Amitāyus (Rhie 2010; Wang 2020); the LS and the image of two Buddhas seated side by side 二佛並坐 (Wang 2005; Zhang 2017); the Weimojie jing 維摩詰經 (Skt. Vimalakīrtinirdeśa) and the image of Mañjuśrī’s Call on Vimalakīrti 文殊問疾 (Zhang 1998; Jin 1989; An 2022). Such discussions are basically limited to a single image or a single niche and are highly fragmented. Moreover, previous studies have relied heavily on the content of the inscriptions to identify a textual basis for them. This ignores not only the overall configuration of niche 6, but also the configuration of its adjacent Buddhist niches.

In niche 6, in addition to the AST, there are other images of Buddhas and Bodhisattvas with ink inscriptions. To the right of the AST, there is a standing mural entitled “Maitreya Bodhisattva 彌勒菩薩”, and to the right of Maitreya Bodhisattva there is another standing mural entitled “Śākyamuni Buddha 釋迦牟尼佛”. Above the standing images of Bodhisattvas Mahāsthāmaprāpta and Maitreya, there is also a mural of ten Buddhas. The name of ten Buddhas is still recognizable from ink inscriptions and is considered to be derived from the Rulai minghao ping 如來名號品32 from the AS (T09n0278) (Zhang 1992, pp. 12–13) (Figure 6).

Niche 7 is adjacent to niche 6, which mainly contains the image of three Buddhas, with only the middle Buddha and the arm of the left Buddha preserved. Below the eastern side of three Buddhas, there was originally a mural of one Buddha and two Bodhisattvas. But only one Bodhisattva remains, inscribed “Dashi□pusa 大勢□菩薩” (Figure 7). The rest of the content in this niche is no longer identifiable. In terms of size, this group of one Buddha and two Bodhisattvas is small in scale, compared with other images. It is not the main focus of niche 7, but only the subordinate of three Buddhas. The nearby niche 9 is also created in the same period. The main subject of niche 9 is also three Buddhas (Figure 8). The halos of niches 7 and 9 differ from each other, but the shape remains similar. The three Buddhas in niches 7 and 9 are considered as the Buddhas of Three Times 三世佛 from scholars (Zhao 1994; Lai 2002).

Niche 8 is under niche 6. The main item is a group of two seated figures (Figure 9). They were created in the Northern Wei dynasty (386–535). Only the top area (head) of this two-seated statue is well preserved. The left side of the statue is a seated figure with a Dhyanā-mudrā33; the right side of the statue is a crossed-leg seated figure with a crossed hands gesture.

The style of this statue is markedly distinct from other Western Qin statues in niches 6 and 7, and it is also overly scribbled compared to other seated Buddha statues from the Northern Wei in niche 5. Considering the overall layout of the east wall, although niche 8 is not a Western Qin niche, the statue of two figures is most likely a restoration of the statues originally created in the Western Qin. Considering the center area of the backscreen and the residual content of the lower area, it is conceivable that the statue may have been restored several times because of the collapse of the lower part of this niche.

From my view, there are two ways of interpreting the statue of niche 8, which bears the traces of restoration. On the one hand, it may be regarded as the statue of one Buddha and one Bodhisattva, with a seated Buddha on the left and a cross-legged Bodhisattva statue on the right. The Buddha on the left is slightly larger than the Bodhisattva on the right. The touji 頭髻 (Skt. jaṭā) of the statue on the right is closer to the style of Bodhisattva. Since the figure on the right side is a cross-legged statue, it is likely to be a statue of Maitreya. So, the content of niche 8 could be deemed as the Śākyamuni Buddha on the left and the Maitreya Bodhisattva on the right. This interpretation is based on the surviving statue.

On the other hand, based on the details of the figures, particularly the upper and lower edges of the backscreen and the cyan area between two figures, it suggests that they were probably not restored in one operation. They may have been completed by two craftsmen at different times due to the collapse of the lower side of the niche. The restoration in the Northern Wei may have originated from the Western Qin statue. Due to the severe damage and the presence of more than one collapse of the lower part of this niche, the statue as it stands is the result of multiple restorations. Given the configuration of the content in the nearby niches 6, 7, and 9, it is likely that the two figures in niche 8 are derived from the depiction of two Buddhas seated side by side from the LS. The differences between the left and right statues show the restoration styles of different craftsmen at different times.

Either of these interpretations is not inconsistent with the content of the LS. On the contrary, both the statue of Maitreya Bodhisattva and two Buddhas seated side by side are closely related to the contents of the LS in niche 6 and the theme of three Buddhas (Buddhas of three times) in the adjacent niches 7 and 9.

Based on these considerations, the combination of subjects in niches 6–9 (Figure 10) is thus: “AST + Maitreya Bodhisattva + three Buddhas (Buddhas of three times) + ten Buddhas + two seated figures (Śākyamuni Buddha with Maitreya Bodhisattva or two Buddhas)”. This combination goes far beyond the scope of the Pure Land sutras translated by the early fifth century.

5. The Consideration from the Lotus Sutra

According to the Kaiyuan Lu (T55n2154), there are six Chinese translations of the LS, three of which are surviving: the 10-volume version of the Zheng fahua jing 正法華經 (Hereafter: ZLS) (T09n0263) translated by Dharmarakṣa 竺法護 (229–306) in the Western Jin dynasty (266–316), the 8-volume version of the Miaofa lianhua jing 妙法蓮華經 (hereafter: MLS)(T09n0262) translated by Kumārajīva 鳩摩羅什 (344–413) in the Later Qin dynasty (384–417), and the 7-volume version of the Tianpin miaofa lianhua Jing 添品妙法蓮華經 (T09n0264) translated by Jñānagupta 闍那崛多 (523–600) and Dharma-Gupta 達摩笈多 (unknown–619) in the Sui dynasty (581–618). The first two versions fall within the scope of our discussion. Among the surviving texts, Kumārajīva’s version is the most widespread and influential. It was also the basis for the doctrinal interpretation of the Tian’tai School 天臺宗, which emerged from the Sui dynasty (581–618). Due to this, the ZLS has not received much attention from art historians.

Based on the evidence from surviving Dunhuang manuscripts, the ZLS was distributed in Turpan 吐魯番 as early as 399 (or even earlier) (Ikeda 1990, p. 78; Fang 1997, p. 215). The earliest surviving Dunhuang manuscript of the MLS was found in Kucha 庫車 and is dated to 411 (Ikeda 1990, p. 81; Fang 1997, p. 217). Since the translation of the MLS in 406 (T55n2154, p. 512; T55n2145, p. 10), the written manuscript of the ZLS almost disappeared. Thus, the influence of the two versions of the LS in the Hexi Corridor ebbed and flowed during the Yongkang (412–419) and Jianhong period (420–427). An examination of the murals, scriptures, and inscriptions shows that entire Western Qin niches in Cave 169 was influenced by a combination of both versions of the LS.

Given the configuration of this cave, the Buddha and Bodhisattva images of niche 6 and adjacent niches is consistent with the LS. The following builds on this foundation by further interpreting the textual sources of the AST.

5.1. The Phenomena of Restoration in Niche 6

First, and the most easily overlooked, the phenomena of restoration are observed in several niches of Cave 169, including niche 6. According to Zhang’s (1998, p. 195) identification, the statue of Avalokiteśvara on the left side of the AST has rat holes and traces of burns. In addition, the word “Chonghui 重暉” in the JRI also suggests that the JRI is a text of a vow from the lay Buddhist related to restoration. According to this, the restoration area of niche 6 extended beyond the AST to the area of the Maitreya Bodhisattva, Śākyamuni Buddha, and the JRI.

With reference to the depiction of other Western Qin niches, the touji 頭髻 (Skt. jaṭā), the timing of the translation and transmission of the SIL, Wei (2018, pp. 246–47) claimed that the JRI, ten Buddhas, Maitreya Bodhisattva, and Śākyamuni Buddha in niche 6 were created in the Western Qin, while the AST were probably created (restored) close to the period of the first and second phase of the Yungang Grottoes. Wei (2018) suggested that the repainting of the AST should have been inspired by the original iconography as well as the SIL. Regarding the translation of the SIL, Harrison et al. (2002, p. 180) claimed that the SIL was translated by Buddhabhadra and Baoyun 寶雲 in 421. Karashima (2016, p. 344) agreed in principle with Harrison’s argument. After examining the lexical and grammatical aspects of several sutras, Karashima also concluded that the SIL should have been translated around 421. This view is widely acknowledged34.

Combined with the images of ten Buddhas and the AST in niche 6, I intend to claim that both the AS and the SIL were probably translated by Buddhabhadra in 421 and then transmitted by the same route from Jiankang to the northwest China concurrently. Therefore, from the point of view of the history of sutra translation and transmission, I am more inclined to conclude that the year of Xuanxiao 玄枵 (424) is the date when the JRI was created, which was also the year when niche 6 was restored. The assumption that the two sutras were translated in Jiankang in 421 and spread from Jiankang to Bingling Temple in 424 (or a little earlier) is consistent with the traffic conditions regarding the dissemination of the sutras at that time and location.

On this basis, I claim that the restoration of niche 6 should be considered into two parts. The first part is the area around the JRI, including the JRI, Maitreya Bodhisattva, and Śākyamuni Buddha. This part should be less burnt, and the restoration should be as faithful as possible to the original creation of the images and inscriptions. The second part is the images of the AST and ten Buddhas. The ink inscriptions of the AST were likely to be unrecognizable at the time of restoration. In around 424, the inscriptions of the AST and ten Buddhas at the time of restoration probably refer to the names of the Buddhas and Bodhisattvas in the newly introduced SIL and the AS, respectively. This area would therefore have been influenced by a combination of the original creation and the newly introduced sutras at the time of restoration. The superimposition of some of the murals and statues in niche 6 is also the evidence of restoration.

5.2. The “De Dashi 得大勢” Bodhisattva in the LS

The above sections have ruled out the possibility that the Pure Land sutras, especially the SIL, could be the exclusive textual basis of the AST; it has also ruled out the possibility of the BNS as its textual source. In connection with iconographic configuration, the paper analyzes the influence of the LS on niche 6 as well as adjacent niches. This section further discusses the relationship between the LS and the AST.

As far as the Chinese character of Mahāsthāmaprāpta is concerned, the wording in the ZLS is “De dashi 德大勢” and the wording in the MLS is “De dashi 得大勢”. Neither version has a “zhi 志”, but both have a “De 德” or “De 得”, although the “De dashi 得大勢” in the MLS is closer to the “De Dashizhi 得大勢志” in the inscription.

This paper suggests that the inscription “De Dashizhi 得大勢志” in niche 6 is similar to the case of “Made on the twenty-fourth day of the third month of the first year of Jianhong, the year of Xuanxiao 建弘元年歲在玄枵三月廿四日造” in the JRI, that is, there is a writing error in Chinese characters in the inscription. The possibilities for the textual sources of the inscription “De Dashizhi 得大勢志” have been considered from various perspectives. Although this paper suggests that the inscription “De Dashizhi 得大勢志” is originated from a writing error, it is still feasible to explore its textual sources based on the surviving evidence.

First, considering the details of the image, the absence of a vase in the crown of the head of “De Dashizhi 得大勢志” Bodhisattva is clearly distinguishable from images of the Mahāsthāmaprāpta Bodhisattva in the later period. Moreover, the image of the AST did not exist from the same or adjacent periods in other Chinese cave temples. Therefore, it is reasonable to assume that ink inscriptions of the AST were probably influenced by the Pure Land sutras, such as the SIL, in an indirect manner during the restoration process. However, the original statues and original inscriptions were not derived from the surviving Pure Land sutras, but from the LS.

Second, in view of the configuration of the niches, niche 6 and adjacent niches should have been influenced by the LS in its entirety. Accordingly, it is inferred that the initial image of the AST before the restoration may have been associated with the LS. This is also consistent with the date of translation and transmission of the LS. The restoration of niche 6 would have been completed around 424, according to the nearby JRI. The original images in niche 6 were made earlier than 424, which is consistent with the translation time of the MLS (406).

In addition, multiple Buddhist records show that after the translation of the MLS by Kumārajīva, its circulation territory had radiated from Chang’an to the northwest China, while Western Pure Land Buddhism had not yet become an independent Buddhist faith during this period. This is also reflected in the mural of donors in niche 6. For example, there are two rows of donors painted at the bottom of the JRI. Of them, one mural is inscribed with “Image of Bhikshu Daorong 比丘道融之像” (Figure 11).

In the BEM, there are two references to Daorong 道融, one of which is (Eng. Yang 2022, p. 266; Chi. T50n2059, p. 363): “Kumārajīva asked Daorong to expound the New LS (MLS). Listening to his sermon, Kumārajīva admired aloud: ‘The prosperity of the Dharma will rely on Daorong.’” The second record is (Eng. Yang 2022, p. 268; Chi. T50n2059, p. 363): “Daorong later arrived in Pengcheng and often held consecutive sermons to preach the Dharma; over a thousand listeners came to follow him……He passed away in Pengcheng at 74 (years old), leaving his commentaries on the LS (MLS), the Mahāprajñāpāramitā-sūtra, the Golden Light Sutra, the Daśabhūmika-sūtra, and the Vimalakīrti-nirdeśa-sūtra for posterity.” Both of these accounts are closely related to the MLS and not to the contents of the Pure Land sutras. Therefore, the mural of the donor also testifies to the influence of the LS in this period. Moreover, the texts of the BEM (T50n2059, p. 364; 368; 371) and the CSZJ (T50n2059, p. 83; 364) also show that monks such as Tandi 曇諦, Tanying 曇影, Sengrui 僧叡, and Huiguan 慧觀 preached the MLS in the early fifth century. In addition, the CSZJ (T55n2145, p. 83) records that Huiyuan 慧遠 and Sengrui 僧叡 wrote a preface and a postscript to the MLS, respectively. In addition to these historical records, the Fahuajing yiji 法華經義記 (T33n1715) by Fayun 法雲 (467–529) is the second earliest surviving commentary on the LS35 which also includes the description of ”De Dashi 得大勢”, Amitāyus 無量壽佛, and Avalokiteśvara 觀世音. Even though the Fahuajing yiji 法華經義記 was created much later than the niche 6, it is an indirect reflection of the fact that the LS was more popular than Pure Land texts by the end of the fifth century.

In contrast, there are few records of the dissemination of the Pure Land sutras or the AST belief in Buddhist record generated by the early fifth century. For example, the most widespread Pure Land event of this period was the founding of the White Lotus Society 白蓮社 by Huiyuan 慧遠 (334–416) at Donglin Temple 東林寺. However, it was not until the middle and late Tang dynasty that this event began to spread. It is found in the literary works of Dai Shulun 戴叔倫 (732–789), Lingche 靈澈 (746–816), and Bai Juyi 白居易 (772–846). The earliest historical record of the event is from the Lushan ji 廬山記36 by Chen Shunyu 陳舜俞 (unknown–1074) of the Song dynasty. So, the possible influence of Huiyuan 慧遠 on Bingling Temple in the early fifth century is also excluded.

Thus, based on Buddhist records, the spread of Pure Land Buddhism in the early fifth century was limited, especially in the northwest China. The content related to the Pure Land Buddhism in Cave 169, in my view, is coincidental. The configuration of niche 6 with adjacent niches suggests that the depiction of the AST is not a separate representation of the Pure Land scriptures but was influenced by a combination of Pure Land Buddhism and the LS, and that the influence of the LS is spread throughout this niche as well as in adjacent niches.

Therefore, considering the configuration of the niches, it is suggested that the surviving inscription “De Dashizhi 得大勢志” contains a writing error in the Chinese characters, which is a similar case to the handwriting error of the date recorded in the JRI. The word “De 得” is probably derived from the MLS, while the word “zhi 志” might be a writing error reflecting the indirect influence of the Pure Land texts, such as the SIL or another lost version of the LSukh. However, such an indirect influence is limited and appears only in the ink inscriptions of the AST. Viewed in the context of niche 6 and adjacent Buddhist niches as a whole, the AST is not inconsistent with the rest of the images of the surrounding niches and belong to the depiction of the LS.

6. Limitations

In traditional discussions of Chinese cave temples, few scholars have discussed the influence of Sanskrit or other ancient Indian texts on iconography. In the case of Cave 169 at Bingling Temple, there is also no reason to discuss the Sanskrit text. First, the Qifu 乞伏 regime of Western Qin (385–400, 409–431) pursued a policy of Sinicization. Therefore, during this period (and the Northern Wei dynasty), there are no traces of Buddhist texts in ancient Indian languages circulating around the Bingling Temple. Second, the Sanskrit texts or fragments related to the textual sources of Cave 169, such as the LS and the LSukh, all appear in the 8th or even after the 10th century, which is too remote to our discussion. Most notably, all inscriptions in the Western Qin and Northern Wei dynasties in this cave are in Chinese characters. Therefore, none of the images and inscriptions within the scope of discussion have any detectable connections to the Sanskrit material. It also should be noted that while it cannot be ruled out that the inscriptions may have been influenced by other lost scriptures, the discussion in this paper is limited to the surviving texts that correspond to the cave functions37.

As Cave 169 is extremely complex and rich in content38, it is not realistic to cover all inscriptions39/images in one article. The artistic style of the images in this cave is only touched upon in this paper because scholars have been studying this area in depth and systematically over half a century,40 which does not fall into the category of understudied research.

7. Conclusions

This paper attempts to correct the misidentification of the inscription “De Dashizhi 得大勢志” in Cave 169 of Bingling Temple and re-identify its textual sources. Based on the textual studies, Appendix A summarizes the textual sources of the “De Dashizhi 得大勢志” and its variants “Dashizhi 大勢志” and “De Dashi 得大勢” discussed in this paper. Although most scholars agree that the AST in niche 6 comes from the SIL, this view is not consistent with the Chinese characters of the inscription, the translation and theme of the Pure Land texts, or the configuration of the niches. Considering these aspects, this paper concludes that the inscription “De Dashizhi 得大勢志” contains a writing error and has been significantly influenced by the LS, although this paper does not deny the indirect and partial influence from Pure Land Buddhism on the AST during its restoration process. The discovery in this paper not only sheds new light on the configuration of images of the LS in the early fifth century, but also provides art historical evidence for future exploration of the relationship between the LS and Pure Land Buddhism in the Sixteen Kingdoms (304–439) and Northern dynasty (439–581).