A Study of the Early-Stage Translations of Foxing佛性 in Chinese Buddhism: The Da Banniepan Jing大般涅槃經 Trans. Dharmakṣema and the Da Fangdeng Rulaizang Jing大方等如來藏經 Trans. Buddhabhadra

Abstract

The Da fangdeng rulaizang jing大方等如來藏經 (Skt. Tathāgatagarbha-sūtra), translated by Buddhabhadra佛陀跋陀羅 (358–429) is one of the early Chinese Buddhist canon texts where the term foxing佛性 (Jp. busshō; Buddha-nature) is clearly used to express Buddha-nature. However, the term foxing cannot be confirmed in other extant translations of the Tathāgatagarbha-sūtra. Another early text in the Chinese Buddhist canon, the Da banniepan jing大般涅槃經 (Skt. Mahāparinirvāṇa-mahāsūtra), translated by Dharmakṣema曇無讖 (385?–433), also used the term foxing, which cannot be correspondingly confirmed in the surviving Sanskrit fragments of this scripture. Some significant differences in foxing between the Sanskrit fragments and Dharmakṣema’s translation of this sutra belong to the first twelve fascicles of Dharmakṣema’s translation completed under his collaborators’ support when he had not mastered the Chinese language. It is very likely that Faxian法顯 (337–422) translated a version of the Mahāparinirvāṇa-mahāsūtra that featured buddhadhātu as foxing. Buddhabhadra, in the same period, translated a version of the Tathāgatagarbha-sūtra, in which he favoured the term foxing over a literal translation of the Sanskrit. As another contemporary monk with these two, Dharmakṣema translated the Mahāparinirvāṇa-mahāsūtra, going further than Faxian by using the term foxing regularly. These texts influenced the Dilun monastic tradition地論宗. Among these, the term foxing and its Sinicism explanations played the most significant role, influencing the whole of the Chinese and even East Asian Buddhist thought.

1. Introduction

In recent years, discussions in Buddhist scholarship have focused on the concept of “Buddha-nature” within all sentient beings, whether or not this concept is compatible with classical Buddhist teachings such as no-abiding-self or even those doctrines rooted in the Nikāyas/Āgamas. This controversy is not only relevant to East Asian Buddhism but also to the roots of this tradition in the Indian Mahāyāna sutras, which deploy the concept of tathāgatagarbha (Buddha-embryo or Buddha-womb).1 Moreover, in some cases, the term tathāgatagarbha is also used to describe sentient beings themselves (Skt. sarvasattvās tathāgatagarbhāḥ; all sentient beings are those who contain tathāgata). As is widely known, in the history of East Asian Buddhism, tathāgatagarbha (Chin. rulaizang如來藏) was sometimes considered a synonym of foxing佛性 (Jp. busshō; Buddha-nature).2 The relationship between these two terms was ambiguous in Chinese Buddhism because some monks and schools, such as the Nirvāna tradition (Chin. Niepan zong涅槃宗), declared that foxing is the same as rulaizang.3 Therefore, probing the early cases where the classical Chinese term foxing appeared in China is significant to clarifying the origin and development of these two concepts in East Asian Buddhism.

Two of the early translators who translated some terms as foxing were Buddhabhadra佛陀跋陀羅 (358–429) and Dharmakṣema曇無讖 (385?–433). Both of them worked on their texts in China in the first half of the fifth century.4 In this article, I investigate this issue based on the translations by Dharmakṣema and Buddhabhadra. In other words, cases of the term foxing that appeared during the Northern Liang北涼 dynasty (397–439) and the second half of the Eastern Jin東晉 (317–420) are the objects of this research.5 Among these two dynasties, the Northern Liang is much more important for my discussion because the full text of the Mahāparinirvāṇa-mahāsūtra [the Great Nirvāṇa Sūtra] was completely translated into classical Chinese and spread to the whole of China after this period.6

There is hardly research discussing both the translations of Dharmakṣema and Buddhabhadra to probe the origin of the term foxing as a Chinese term and its context in Chinese translation in the early fifth century, especially the lack of comparison with relevant Sanskrit fragments in the context of Chinese Buddhism remains, although it is evident that in Indian Buddhist texts, buddhadhātu indicates Buddha-nature, which has the meaning “nature of a buddha.”7 The scholars in Indian and Tibetan Buddhist Studies did not pay attention to this issue in the context of East Asian Buddhism. Conversely, many scholars in Chinese Buddhist Studies have hardly used the relevant Sanskrit and Tibetan texts to investigate the origin and development of the term foxing.8 In a sense, this is also one of the purposes of this article.

Through this study, we can presume that Faxian法顯 (337–422) translated a version of the Mahāparinirvāṇa-mahāsūtra that featured buddhadhātu as foxing. Buddhabhadra, in the same period, translated a version of the Tathāgatagarbha-sūtra, in which he favoured the term foxing over a literal translation of the Sanskrit, maybe a particular Sanskrit expression, or supplying foxing in place of diverse Sanskrit phrasings. As another contemporary monk with these two, Dharmakṣema translated the Mahāparinirvāṇa-mahāsūtra, going further than Faxian by using the term foxing regularly.

2. Dharmakṣema, Buddhabhadra and Chinese Buddhism at the Beginning of the Fifth Century

Unfortunately, most of the Buddhist literature produced during the Northern Liang dynasty has been lost. Only some quotes can thus be found from later treatises such as the Da banniepan jing shu大般涅槃經疏 [Commentary on the Mahāparinirvāṇa-mahāsūtra], written by Guanding灌頂 (561–632). In 421, it was recorded that Dharmakṣema held a Sanskrit text in hand and spoke Chinese to Daolang道朗 (?–?) when he translated the Mahāparinirvāṇa-mahāsūtra in 421. Daolang was one of the most accomplished monks in Hexi河西 area at that time and was guided by Dharmakṣema in Guzang姑臧 while receiving tuition from Dharmakṣema. According to the Chu sanzang ji ji出三藏記集 [Compilation of Notes on the Translation of the Tripitaka; or Collected Records concerning the Translation of the Tripiṭaka], 11 texts were regarded as Dharmakṣema’s translations, as follows.9

Da banniepan jing大般涅槃經 [the Great Nirvāṇa Sūtra], 36 juan, T374.

Fangdeng daji jing方等大集經 [the Sūtra of the Vaipulya Great Assembly], 29 juan or 30 juan, T397.

Fangdengwang xukongzang jing方等王虚空藏經 [the Sūtra of the King of Vaipulya and the Chamber of Space], 5 juan.

Fangdeng dayun jing方等大雲經 [Skt. Mahāmeghasūtra; the Sūtra of the Vaipulya Great Cloud], 4 juan (or Fangdeng wuxiang dayun jing方等無想大雲經, 6 juan), T387.

Beihua jing悲華經 [the Sūtra of Flower with Compassion], 10 juan, T157.

Jinguangming jing金光明經 [the Golden Light Sūtra], 4 juan, T663.

Hailongwang jing海龍王經 [the Sūtra of the King of Marine Dragons], 4 juan, T598.

Pusa dichi jing菩薩地持經 [the Sūtra of Stages of Bodhisattvas], 8 juan, T1581.

Pusajie ben菩薩戒本 [the Text on Precepts of Bodhisattvas], 1 juan (also regarded as a text translated in Dunhuang燉煌), T1500.

Youposai jie優婆塞戒 [Upasaka’s Precepts], 7 juan, T1488.

Pusajie jing菩薩戒經 [the Sūtra of Precepts of Bodhisattvas], 8 juan.

Pusajie youpo jietan wen菩薩戒優婆戒壇文 [the Treatise on Precepts of Bodhisattvas and Upasakas], 1 juan.

These eleven texts totally have 104 juan. In the dynasty of An emperor of Jin晋安帝, Indian monk Tanmochen曇摩讖10 (or Tanwuchen曇無讖) came to Western Liang prefecture西涼州 and translated these texts under the support of Juqumengxun沮渠蒙遜.

Dharmakṣema moved to Guzang姑臧 and translated these texts after 417. In the Chu sanzang ji ji, the Da banniepan jing大般涅槃經 [the Great Nirvāṇa Sūtra] and the Fangdeng daji jing方等大集經 [the Sūtra of the Vaipulya Great Assembly] are mentioned. Among these, the most influential text in China is the Da banniepan jing.11

According to Fuse Kōgaku布施浩岳, if the first ten fascicles of the Da banniepan jing are compared with the Nihuan jing泥洹經 [the Nirvāṇa Sūtra], which is a short Chinese translation, or originally a short version, of the Mahāparinirvāṇa-mahāsūtra translated by Faxian法顯 (337–422) several years before the Da banniepan jing and has six fascicles (Foshuo dabannihuan jing 佛說大般泥洹經, T 376), it is evident that the Da banniepan jing contains most of the content and information that is found in the Nihuan jing, although it is possible that they are based on different versions/recensions of the Mahāparinirvāṇa-mahāsūtra. Accordingly, following the completion of the Da banniepan jing’s translation, fewer and fewer Chinese readers read the Nihuan jing.12 This is a foundation of understanding the relationship between the Da banniepan jing and the Nihuan jing in Chinese Buddhism. Moreover, as it is well known, the southern text of the Da banniepan jing, containing thirty-six fascicles, was not a direct translation but an edited version based on the Nihuan jing translated by Faxian and the Da banniepan jing translated by Dharmakṣema. Faxian returned to China from India and launched his translation work after 410.13 It almost overlapped with the date of the translation of the Da banniepan jing. Faxian cooperated with Buddhabhadra and finished the translation of the Nihuan jing in 416.14 I will, therefore, ignore the southern text of the Da banniepan jing in this article. However, regarding the Nihuan jing, there will be a discussion of the Nihuan jing’s attitude to foxing, the purpose of my research, later in this article, since it is important for the study of what Dharmakṣema likely read.15

The original content of the Nihuan jing equates to the first ten fascicles of the Da banniepan jing translated by Dharmakṣema, which contains forty fascicles in total. After completing the first twelve fascicles of the translation of the Da banniepan jing, Dharmakṣema noted that the original Sanskrit text of the Da banniepan jing was not enough in China then; thus, he returned to India to seek an integrated version of this sutra. He arrived at Khotan于闐 by traveling through the southern path of Tianshan天山 and found the middle and later fascicles of the Mahāparinirvāṇa-mahāsūtra.16

Concerning the Sanskrit text, Takasaki Jikidō高崎直道 mentioned partial fragments of this text.17 Almost simultaneously, Matsuda Kazunobu松田和信 collected and translated all existing fragments18, and Habata Hiromi幅田裕美 used the previous research by Takasaki and Matsuda and provided a critical edition of the Tibetan text19, followed by a published monograph on all extant Sanskrit fragments of the Mahāparinirvāṇa-mahāsūtra.20

Ōno Hōdō大野法道 thinks the preface (Chin. jingxu 經序) by Hexi Daolang河西道朗 is more reliable. According to this account, Dharmakṣema first came to Dunhuang 燉煌 with a variety of scriptures, and then moved to the Northern Liang 北涼, where he translated the first ten juan of the Da banniepan jing, finishing in 421. He further states that it is certain that the source text for the remaining thirty juan incorporated into Dharmakṣema’s the Da banniepan jing came from Khotan.21 According to Ōchō Enichi横超慧日, conceiving that the same person or group translated at the same time all forty fascicles of the Da banniepan jing is difficult. Instead, they may have been edited by different people over several stages.22 In other words, some fascicles or sections of the forty-fascicle Da banniepan jing were not translated by Dharmakṣema. Feng Chengjun馮承鈞 states that Dharmakṣema used the Sanskrit text collected by Zhimeng智猛 (?–452), who did not engage in translating.23 In contrast, Chen Jinhua陳金華 objects to Feng’s conclusion. Chen asserts that Zhimeng not only brought the Sanskrit text of the Da banniepan jing to China but also participated in the translation.24 Hence, it is very likely that the translators of the Da banniepan jing were not only Dharmakṣema but included other people or groups, in addition to even some other materials or sources.25

Therefore, I intend to discuss the classical Chinese term foxing based on the Da banniepan jing attributed to Dharmakṣema in the framework of Chinese Buddhism. In addition, as a background to contemporary translation in China, the term foxing as found in the Da fangdeng rulaizang jing大方等如來藏經 [The Sutra of the Tathāgatagarbha] translated by Buddhabhadra佛陀跋陀羅 (358–429) is also the object of my discussion. In other words, as these two texts were translated into classical Chinese during almost the same and early period, I will talk of the use of the term foxing found in the Da fangdeng rulaizang jing at first and then move to the discussion on this topic in the Da banniepan jing.26 Finally, this study also slightly mentions the Da fangdeng wuxiang jing 大方等無想經 [Skt. Mahāmeghasūtra; T387] by Dharmakṣema and the renderings by Guṇabhadra 求那跋陀羅 (394–468) from roughly the same time.

According to the Fozu tongji佛祖統紀 [Entire Records of the Buddhas and the Ancestors], both Dharmakṣema and Buddhabhadra were engaged in the translation works in China during almost the same era.

In the fourth year of Yixi義熙, Huiyuan慧遠 was discontented with the uncompleted translations in Chinese Buddhism, the lack of meditation, and the incomplete canons of precept. He sent his disciples such as Zhifaling支法領 to India to collect more Sanskrit Buddhist texts. They met Buddhabhadra in India and asked him to return to China together. In the eighth year of Yixi, Dharmakṣema moved to Guzang姑臧. Juqu Mengxun沮渠蒙遜, the King of the Northern Liang, asked Dharmakṣema to stay there and translate the Mahāparinirvāṇa-mahāsūtra into the Da banniepan jing, which has forty fascicles. In the ninth year of Yixi, Buddhabhadra (Chin. Juexian覺賢), a monk from the Kapilavastu area, went to Lushan廬山 mountain and stayed there. Huiyuan asked Buddhabhadra to translate some texts of meditation.

義熙四年，遠法師以江東經卷未備，禪法無聞，律藏殘缺。乃令弟子支法領等往天竺，尋訪獲梵本。於于闐遇佛陀跋陀羅，乃要與東還。…八年，曇無讖至姑臧，涼王沮渠蒙遜留之譯《大般涅槃經》四十卷。…九年，迦維衛國沙門佛陀跋陀羅（此雲覺賢）至廬山入社，遠法師請譯禪數諸經。

27

That is to say, Huiyuan’s 慧遠 (334–416) disciples met Buddhabhadra in about 408. In 412, Dharmakṣema launched the translation of the Da banniepan jing in the Northern Liang. In 413, Buddhabhadra had met Kumārajīva鳩摩羅什 (344–413) in China and moved to Lushan Mountain to continue his translation works. Evidently, the periods of Dharmakṣema and Buddhabhadra acting in China overlapped. Therefore, in my opinion, it is very likely that they had similar circumstances from 410, such as mutual assistants in China and the same Chinese texts that had been translated.28 On the contrary, to say the least of it, even if these two translators did not actually share many assistants and they did not directly influence each other, we cannot deny that they were engaged in the translation works in China during the same era, a very early period for the appearance of the term foxing. For this reason, it is inevitable to discuss not only Dharmakṣema’s translations but also those by some other translators, such as Buddhabhadra, at the beginning of the fifth century.29

As has been mentioned above, there is hardly any research discussing both the translations of Dharmakṣema and Buddhabhadra to probe the origin of the term foxing as a Chinese term and its context in Chinese translation in the early fifth century, especially the lack of comparison with relevant Sanskrit fragments in the context of Chinese Buddhism remains. This is also one of the purposes of this article.

3. Foxing佛性 in the Da fangdeng rulaizang jing大方等如來藏經 (Skt. Tathāgatagarbha-sūtra) Translated by Buddhabhadra

One of the earliest Buddhist texts discussing tathāgatagarbha is the Tathāgatagarbha-sūtra. The Tathāgatagarbha-sūtra, a seminal text of tathagatagarbha doctrine, describes how tathāgatagarbha accounts for the possibility of transformation from a state of delusion to a state of enlightenment by uncovering the inherently pure nature within, referred to containing, store or “that which sentient beings possess.”30

Two recensions of the Tathāgatagarbha-sutra are extant in Chinese: the Da fangdeng rulaizang jing大方等如來藏經 (T vol. 16, no. 666), translated by Buddhabhadra佛陀跋陀羅 (358–429) in the Eastern Jin東晉 (317–420),31 and the Da fangguang rulaizang jing大方廣如來藏經 (T vol. 16, no. 667), translated by Amoghavajra (or Bukong) 不空 (705–774) under the Tang唐 (618–907).32 While the original Sanskrit sutra is not extant for comparison, the bKa’ ’gyur canon represents Tibetan recensions;33 one of them is titled Phags pa de bzhin gshegs pa’i snying po shes bya ba thegs pa chen po’i mdo, translated by Śākyaprabha and Ye-śes-sde (photographic print Tibetan Buddhist Canon 36, 240.1~245.5)34.

The Tathāgatagarbha-sūtra is a relatively short scripture that represents the point of a number of works in Indian Buddhism concentrating on the idea that all sentient beings are tathāgatagarbha. According to Michael Zimmermann, the hitherto accepted assumption that the Da fangdeng rulaizang jing reflects an Indian transmission, which has not undergone the textual alterations of later centuries, is only partly true because the source of the citations in the Ratnagotravibhāga [Chin. Jiujing yisheng baoxing lun究竟一乘寶性論; Treatise of the Jewel-nature of Ultimate Single Vehicle], a śāstra which was written at least fifty years before Buddhabhadra translated Tathāgatagarbha-sūtra into the Da fangdeng rulaizang jing, has turned out to be the recension represented in the Tibetan tradition.35 I agree with this view. That is to say, although it is possible that there are some differences between the underlying Sanskrit text of the Da fangdeng rulaizang jing and that of the Tibetan translation, we should not consider that the original Sanskrit text of the extant Tibetan translation had been substantially amended compared with that of the Da fangdeng rulaizang jing. Thus, in my opinion, it is still effective to investigate the unique way and purpose of Buddhabhadra through comparing his translation with the Tibetan text, especially about the Chinese term foxing, which appears only in Buddhabhadra’s translation.

As mentioned above, the Da fangdeng rulaizang jing translated by Buddhabhadra is a text early in the known literary history of the term foxing. At the beginning of this chapter, the following paragraph must be discussed prior to others.

In the same way, sons of good family, I see with my buddha-vision that all sentient beings, within the afflictions of desire, hostility and delusion, possess the knowledge, vision and body of a tathāgata sitting cross-legged, dignified and motionless. Sons of good family, all sentient beings, although situated in many kinds of rebirth, in the midst of their afflictions possess the tathāgatagarbha, always permanent and undefiled, replete with excellent characteristics no different to my own. Sons of good family, for example, it is like this, a person with divine-vision/eye (Skt. divyacakṣus; Chin. tianyan 天眼) inspects calyxes and find that the tathāgata-body within the flowers can be revealed if the drooped petals had been moved away. In the same way, sons of good family, the Buddha, seeing all sentient beings to already be tathāgatagarbha, desiring to cause this to be revealed, explains the dharma, destroying their defilements and manifesting their buddha-nature. Sons of good family, the true nature (Skt. dharmatā; Chin. faer法爾) of all buddhas is this: whether or not buddhas appear in the world, in all living beings the store [or womb] of a tathāgata is at all times present without change.

如是善男子！我以佛眼觀一切眾生，貪欲恚癡諸煩惱中，有如來智、如來眼、如來身，結加趺坐儼然不動。善男子！一切眾生雖在諸趣，煩惱身中有如來藏，常無染污、德相備足，如我無異。又善男子！譬如天眼之人觀未敷花，見諸花內有如來身結加趺坐，除去萎花便得顯現。如是善男子！佛見眾生如來藏已，欲令開敷為說經法，除滅煩惱顯現佛性。善男子！諸佛法爾，若佛出世若不出世，一切眾生如來之藏常住不變。

36

[Tibetan translation] In the same way, sons of good family, also the Tathagata, the Honorable One and Perfectly Awakened One, [perceives] with his insight (prajñā), knowledge (jñāna) and tathāgata-vision that all the various sentient beings are encased in myriads of defilements, [such as] desire (rāga), anger (dveṣa), misguidedness (moha), longing (trṣṇā) and ignorance (avidyā). And, sons of good family, [he] perceives that inside sentient beings encased in defilements sit many tathāgatas, cross-legged and motionless, endowed like myself with a [tathāgata’s] knowledge and vision. And [the Tathāgata], having perceived inside those [sentient beings] defiled by all defilements the true nature of a tathāgata (tathāgatadharmatā) motionless and unaffected by any of the states of existence, then says: “Those tathāgatas are just like me!” Sons of good family, in this way a tathāgata’s vision is admirable, [because] with it [he] perceives that all sentient beings contain a tathāgata (tathāgatagarbha). “Sons of good family, it is like the example of a person endowed with divine vision [who] would [use this] divine vision to look at such unsightly and putrid lotuses, not blooming and not open, and would [owing to his vision] recognise that there are tathagatas sitting cross-legged in their center, in the calyx of [each] lotus, and [knowing that, he] would then desire to look at the forms of the tathāgatas; [he would] then peel away and remove the unsightly, putrid and disgusting lotus petals in order to thoroughly clean the forms of the tathāgatas. In the same way, sons of good family, with the vision of a buddha, the Tathāgata also perceives that all sentient beings contain a tathāgata (tathāgatagarbha), and [therefore] teaches the Dharma [to them] in order to peel away the sheaths of those sentient beings [encased in such] defilements [as] desire, anger, misguidedness, longing and ignorance. And after [those sentient beings] have realized the [Dharma, their] tathāgatas [inside] are established in the perfection [of the tathāgatas] (ma rig pa’i nyon mongs pa’i sbubs dbye ba’i phyir chos ston te/de sgrub pa’i de bzhin gshegs pa rnams ni yang dag pa nyid du gnas so).” Sons of good family, the essential law (dharmatā) of the dharmas is this37: whether or not tathāgatas appear in the world, all these sentient beings at all times contain a tathāgata (tathāgatagarbha).38

It is notable that in the Tibetan translation, there is a sentence stating “de bzhin gshegs pa rnas ni yan dag pa nyid du gnas so,” rather than the Da fangdeng rulaizang jing stating “xianxian foxing顯現佛性 (manifesting Buddha-nature).” In other words, in the Tibetan translation, it is difficult to identify the term corresponding to foxing in this paragraph, while it is very possible that this Tibetan translation is not translating the same Indic text as Buddhabhadra.

A similar example can also be found in the following passage.

They [women] do not know it. Therefore, the Tathāgata teaches widely the Dharma for living beings saying: “Sons of good family, do not denigrate yourself! In your own body, you all have the buddha-nature. If you practice diligently and diminish all evil, then you will attain the designations ‘bodhisattva’ and ‘exalted one.’ You will guide and save innumerable living beings!”

如彼女人而不覺知，是故如來普為說法，言：善男子！莫自輕鄙，汝等自身皆有佛性，若勤精進滅眾過惡，則受菩薩及世尊號，化導濟度無量眾生。

39

[Tibetan translation] Then, though the element of a tathāgata has entered into sentient beings and is present within, those sentient beings do not realize [it]. Sons of good family, in order that sentient beings do not despise themselves, the Tathāgata in this [connection] teaches the Dharma with the [following] words: “Sons of good family, apply energy without giving in to despondency! It will happen that one day the tathāgata [who has] entered [and] is present within you will become manifest. (rig kyi bud ag khyed bdag nyid sro shi bar ma byed par khyed brtson ’grus brtan par gyis shig dang/khyed la de bzhin gshegs pa zhugs pa yod pa dus shig na ’byung bar ’gyur te) Then you will be designated “bodhisattva,” rather than “[ordinary] sentient being (sattva).” [And] again in the [next stage you] will be designated “buddha,” rather than “bodhisattva”.”40

Here, Buddhabhadra translated “rudeng zishen jie you foxing汝等自身皆有佛性 (you all have Buddha-nature),” compared to the statement, “it will happen that one day the Tathāgata who has entered and is present within you will become manifest. Then you will be designated a bodhisattva, rather than ordinary sentient being (sattva)” in the Tibetan translation. We can only find tathāgata and sattva in the Tibetan translation, rather than a proper term matching the Chinese term foxing.

Similarly, the following passage is also typical of the difference between these two translations.

In the same way, with the vision of a Sugata (buddha) I can see that although living beings are covered over by defilements, their tathāgata-nature is indestructible. I teach the Dharma with appropriate means in order to let living beings attain buddhahood. Because their buddha-nature has been covered by defilements, I intend to remove the defilements to make their buddha-nature purified rapidly.

善逝眼如是，觀諸眾生類，

煩惱淤泥中，如來性不壞。

隨應而說法，令辦一切事，

佛性煩惱覆，速除令清淨。

41

[Tibetan translation] In the same way I can see that also all sentient beings have for a long time been constantly overpowered by defilements, but knowing that their defilements [are only] accidental (āgantuka), [I] teach the Dharma with [appropriate] means in order to purify [their] intrinsic nature (prakṛti). (de dag gi nib lo bur nyon mongs shes/rang bzhin sbyang phyir thabs kyis chos ston to)42

The term foxing佛性 appears in Buddhabhadra’s translation again. If we check the Tibetan translation, the corresponding term for the Tibetan is likely to be “prakṛti (intrinsic nature).” There does not seem to be a term corresponding to at least the Chinese character fo佛.43

In particular, among Chinese renderings of the Tathāgatagarbha-sūtra, the classical Chinese term foxing can only be found in the Da fangdeng rulaizang jing, which was translated by Buddhabhadra. It is difficult to accurately confirm the relevant term for foxing or corresponding Tibetan terms, such as sangs rgyas kyi khams/dbyings, in both Amoghavajra’s classical Chinese and the Tibetan translation. Similarly, as mentioned above, there are various terms related to the Chinese term rulaizang in the Tibetan translation, rather than a fixed term.

As is argued by Zimmermann (2002) and some other scholars, since it seems that Buddhabhadra has translated a different recension of the Tathāgatagarbha-sūtra, the fact that its content is different should not surprise us. I also concur with this opinion. The reason for discussing the Da fangdeng rulaizang jing by Buddhabhadra here is to reconsider the importance of this rendering for the history of the term foxing and Buddha-nature thought in Chinese Buddhism—that is, either the different Sanskrit recension of the Tathāgatagarbha-sūtra, which was read and used by Buddhabhadra, or Buddhabhadra’s own creation influenced Chinese Buddhist thought at an early stage. This point was, to my knowledge, seldom emphasized by scholars in East Asian Buddhist studies.

To summarize, although a lack of clarity about Buddhabhadra’s reasoning and motivation remains, as an early classical Chinese Buddhist canon text, the Da fangdeng rulaizang jing used the term foxing, which cannot be confirmed in other extant translations of the Tathāgatagarbha-sūtra.

4. Foxing佛性 in the Da banniepan jing大般涅槃經 (Skt. Mahāparinirvāṇa-mahāsūtra) Translated by Dharmakṣema

The universality of emptiness (śūnyatā) and the doctrines of no-abiding-self (anātman) and impermanence (anitya) are some basic Buddhist teachings. Conversely, one still finds texts such as the Śrīmālādevī-sūtra and the Mahāparinirvāṇa-mahāsūtra that use terms such as ātman.44 This is, in a sense, one of the most basic issues in Buddhist Studies.45 This section focusses on which term the translators used to express the meaning such as ātman in the Chinese translation of the Mahāparinirvāṇa-mahāsūtra.

Before discussing Dharmakṣema’s translation, Faxian’s rendering should be mentioned first. Faxian’s version seems to be a short Chinese translation of the Mahāparinirvāṇa-mahāsūtra if we merely read the Chinese translations, but it is better understood to be a Chinese translation of a shorter version of the Mahāparinirvāṇa-mahāsūtra. Following Habata, the term sangs rgyas kyi khams, which is a translation of buddhadhātu in the Tibetan rendering, is found 23 times in the Tibetan version of the Mahāparinirvāṇa-mahāsūtra. This number is relatively small because we find many instances of foxing in the Chinese translations. Among these 23 references, Dharmakṣema translates 17 instances of foxing, compared to only 8 of foxing in Faxian’s rendering. Moreover, there is no example where Faxian translates sangs rgyas kyi khams as foxing but Dharmakṣema does not.46 For this reason, I will focus on Dharmakṣema’s rendering in this section.

Takasaki Jikidō notes that the underlying term of the foxing in the Da banniepan jing, translated by Dharmakṣema, refers to the nature of tathāgata (Chin. rulai如來).47 Both Shimoda Masahiro下田正弘 and Michael Radich state that in the Mahāparinirvāṇa-mahāsūtra, a strong connection exists between buddhadhātu (Buddha-nature) and tathāgatagarbha (the embryo of Buddha), related to stūpa (relic-chamber).48 Kanō Kazuo加納和雄 also asserts that both buddhadhātu and tathāgatagarbha refer to the content of a stūpa. Furthermore, two kinds of meaning in dhātu, containing both body and relics, are present. Beings possess buddhadhātu, understood as a Buddha’s relic, which evokes the interior of a stūpa at which a relic generally sat.49 Saliently, the term buddhadhātu (Chin. foxing), regarded as the most significant term in the Da banniepan jing (Skt. Mahāparinirvāṇa-mahāsūtra), cannot be found in the extant Sanskrit fragments of this scripture.50 Buddhadhātu, as noted by Takasaki and Radich, in the Mahāparinirvāṇa-mahāsūtra, is considered a synonym of tathāgatagarbha. Alternatively, strictly speaking, tathāgatagarbha may be a way of referring to the presence of buddhadhātu. Meanwhile, we should not ignore the cases where the Chinese term foxing is not explained by referring to tathāgatagarbha.51

An interesting fact appears: the classical Chinese term foxing, emphasized in various classical Chinese translations of the Mahāparinirvāṇa-mahāsūtra, cannot be found in the existing Sanskrit fragments of the Mahāparinirvāṇa-mahāsūtra.52 Therefore, the statement that in the Mahāparinirvāṇa-mahāsūtra, a strong connection existing between Buddha-nature and tathāgatagarbha, which was pointed out by Shimoda and Radich, is mainly based on the Tibetan and Chinese translations of the Mahāparinirvāṇa-mahāsūtra. In this case, it is meaningful to reconsider the original terms and the reasons they were translated into the Chinese term foxing by these translators, including Dharmakṣema.

Kamalaśīla (ca. 740–797) quotes the following in the Bhāvanākrama.

Thus, the Mahāparinirvāṇa-mahāsūtra states the following. Śrāvakas fail to see the lineage (rigs; gotra) of tathāgata in themselves because their meditation (samādhi) is strong, compared to their weak wisdom. Bodhisattvas can merely see an undefined lineage of tathāgata because their wisdom is strong, compared to their weak samādhi. Tathāgata can see all of these because he possesses both meditation and wisdom.

de’i phyir ’phags pa yongs su mya ngan las ’das pa chen po’i mdo las kyang nyan thos rnams kyis ni de bzhin gshegs pa’i rigs mi mthong ste/ting nge ’dzin gyi shas che ba’i phyir dang/shes rab chung ba’i phyir ro//byang chub sems dpa’ rnams kyis ni mthong mod kyi mi gsal te/shes rab kyi shas che ba’i phyir dang/ting nge ’dzin chung ba’i phyir ro//de bzhin gshegs pas ni thams cad gzigs te/zhi gnas dang lhag mthong mtshungs par ldan pa’i phyir ro zhes bska’ stsal te/

53

The bodhisattvas of the ten abodes possess strong wisdom but little samādhi (meditation), so that they cannot clearly see foxing (Buddha-nature). Śrāvakas and pratyekabuddhas possess strong samādhi but little wisdom, so that they cannot clearly see Buddha-nature. Buddhas can clearly see Buddha-nature because they have both meditation and wisdom and have achieved buddhahood without any obstacle.

十住菩薩智慧力多，三昧力少。是故不得明見佛性。声聞縁覚三昧力多，智慧力少。以是因縁不見佛性。諸佛世尊定慧等故，明見佛性，了了無礙。

54

Commenting on this, Yoshimura Shūki芳村修基 contends that Kamalaśīla’s quote corresponds to the thirty-first fascicle of the Da banniepan jing translated by Dharmakṣema.55 Except for small differences in their depictions of the śrāvaka and bodhisattva, these two translations correspond with each other very well. Matsuda denied the possibility that Kamalaśīla knew about the existence of the Da banniepan jing translated by Dharmakṣema.56 If so, Kamalaśīla merely employed the Sanskrit text and Tibetan translations to quote the sentence that śrāvakas fail to see the lineage (rigs; gotra) of tathāgata. Conversely, the Da banniepan jing, translated by Dharmakṣema, clearly states that śrāvakas and bodhisattvas cannot see foxing (Buddha-nature). Hence, the lineage (gotra) found in this Tibetan quote was translated as foxing in the Da banniepan jing.57

Some similar cases appear in the Jiujing yisheng baoxing lun究竟一乘寶性論 [Skt. Ratnagotravibhāga]; namely, gotra was translated as zhenru foxing真如佛性 (Buddha-nature in thusness).

In summary, all beings, according to the Buddha, are always tathāgatagarbha according to three meanings: the tathāgata’s dharmakāya (Dharma-body) is omnipresent in all beings; there is no difference in the tathāgata’s tathatā (thusness); and the gotra of tathāgata (the cause for Buddhahood) exists.

samāsatas trividhenārthena sadā sarvasattvās tathāgatagarbhā ity uktaṃ bhagavatā/yad uta sarvasattveṣu tathāgatadharmakāyaparispharaṇārthena tathāgatatathatāvyatibhedārthena tathāgatagotrasaṃbhavārthena ca/(RG, 26, 7–9)

58

This passage indicates three meanings. Tathāgata, therefore, the Tathāgata taught that all beings always have and share the embryo of Buddha (Skt. tathāgatagarbha; Chin. rulaizang 如來藏). What are these three kinds? First, tathāgata’s dharmakāya (Dharma-body) is omnipresent in all beings. It is said fo fashen bianman佛法身遍滿 (Dharma-body of Buddha is omnipresent). Second, there is no difference in tathāgata’s zhenru (thusness). It is said zhenru wu chabie真如無差別 (there is no difference in thusness). Third, all beings have zhenru foxing真如佛性. It is said jie shiyou foxing皆實有佛性 (all beings possess Buddha-nature).

此偈明何義。有三種義，是故如來説一切時一切眾生有如來藏。何等為三？一者，如來法身遍在一切諸眾生身，偈言佛法身遍滿故。二者，如來真如無差別，偈言真如無差別故。三者，一切眾生皆悉實有真如佛性，偈言皆實有佛性故。

59

In the Sanskrit text, the third part of the definition of tathāgatagarbha is tathāgatagotra. In the classical Chinese translation, this term is translated as zhenru foxing真如佛性. Crucially, the three parts of tathāgatagarbha’s definition, namely, dharmakāya, tathatā and gotra, have been modified in the classical Chinese translation into: fashen法身・zhenru真如・zhenru foxing真如佛性, respectively.60 This kind of translation of gotra in the Jiujing yisheng baoxing lun61, translated by Ratnamati勒那摩提 (6th century CE; ?-508-?), is the same as in the Da banniepan jing translated by Dharmakṣema.62 The Jiujing yisheng baoxing lun was translated into classical Chinese nearly a century after the Da banniepan jing. Furthermore, both were translated during the northern Chinese dynasties. For these reasons, the monks of Dilun地論 tradition were very likely to have been influenced by the terms and concepts of the Da banniepan jing.

On the other hand, due to the edition and research on the Sanskrit texts of the Ratnagotravibhāga and the Laṅkāvatāra-sūtra, it has been clarified that yichanti一闡提 (Skt. icchantika; beings who cannot achieve the buddhahood) is the translation of the Sanskrit term icchantika. Saliently, as Mizutani noted, the term icchantika, which was used in the Mahāparinirvāṇa-mahāsūtra at a very early stage, cannot be found in any surviving Buddhist scripture established prior to the Mahāparinirvāṇa-mahāsūtra.63 In East Asia, numerous monks and scholars have attempted to demonstrate the possibility that icchantika achieve buddhahood.64 The most important issue, however, is the controversy about gotra and Buddha-nature (Chin. foxing).

As introduced above, Matsuda edited the existent Sanskrit fragments of the Mahāparinirvāṇa-mahāsūtra, which can be used to further research the classical Chinese translation. In addition, Habata Hiromi edited the extant Sanskrit fragments and provided a new translation in 2019, which is more in-depth. In these Sanskrit fragments, there is one section, as stated below.

Icchantikas do not see (na paśyanti) virtuous deeds (kalyāṇakṛta). They see blame and evil (papa). Virtuous deeds (sukṛta) mean Bodhi (or enlightenment).65 Not coming means not approaching. The esoteric (or intended) meaning means what is virtuous (kalyāṇa). Who is far away from esoteric deeds (sandhākarma)? Auspicious deeds (bhadrakarma) do not approach the icchantika. Who is far away from a good mind? A good mind does not approach icchantika because they are not wholesome beings due to their arrogant attitudes. What is the basic branch (mūlāṃga)? It means abandoning (or rejecting) this sutra (sūtrapratikṣepa). It is terrible because abandoning (or rejecting) the sutra is frightful. ……Who does not see (or understand) virtuous deed (kṛta)? Evil icchantika does not understand virtuous deed. Icchantikas do not see (or understand) virtuous deed until the end of their transmigration. I will summarize these meanings. Therefore, we should take these terrible things seriously because it is of the utmost frightfulness. At the time when all beings, after having become of one mind, will recognize the ultimate enlightenment (anuttarāṃ saṃmyaksaṃbodhi); it will be possible for icchantika to recognize Bodhi (enlightenment) at that time. However, icchantikas do not see (or understand) virtuous deed. The people who does not see enlightenment and virtuous deed should understand the fact. Namely, the deed of the Tathāgata will not end (or destroyed) unless all beings involved in transmigration recognize the ultimate enlightenment. At that moment, the Buddha will come to complete final nirvāṇa. Along with the final nirvāṇa (atyantaparinirvāṇa), the Buddha will become changeable and absent, like fire and a lamp.66

icchaṃti[kāḥ ka]lyāṇakṛtaṃ na paśyata(?): paśya(ṃ)ti tu pāpaṃ nidiśtuṃ(!) garhituṃ(!) [ca] …

sukṛ(ta)[ṃ b]o[dh]i[r] it[y] arthaḥ (/) na vyaiti n[ā]gacchatīty arthaḥ (/) [san]dheti kalyāṇam ity arthaḥ (/) saṃndhākarmaviśiṣṭakalyāṇaṃ kasya nāgacchati (/) bhadrakarma icchaṃtikasya nāga[ccha]ti (/) +++++++ laṃ satva icchaṃtikā iti i. +++++++++++ [ki]ṃ mūlā(ṃ)gaṃ sūtrapratikṣepaḥ (/) [ta]smād bhetavyaṃ sūtrapratikṣepako hi dāruṇaṃ …

[kaḥ] kṛtaṃ na paśyati (/) saṃsārakoṭyāṃ sa na paśyati (/) arthaṃ bhāṣiṣye: [saṃkṣepa]samuccayaṃ tasmād bhetavy[āḥ] (pa)[ra]madāru○ṇā[ḥ] (/) yadā sa[rvbasa]tvā [e]kamanaso bhūtvā anuttarāṃ saṃmyaksaṃbodhim abhisaṃbotsyate.

tadā [i]cchaṃ[tika](ḥ pā)po ’pi saṃ(bo)[tsyate] … dā [pa○rāṃ] bodhi(ṃ /) sarvba[kṛ](taṃ sa) na paśyati evaṃ jānīṣva viśārada (/) kasya kṛtaṃ na paśyati tathāgatas[y]a (/) [yadā] sa[rvbasa]tvā anuttarāṃ saṃmyaksaṃbodhim abhi○saṃbotsyate saṃsāra … [tā] tadā tathāgatasya kṛtaṃ na vinakṣya(ti/ta)dā [par]inirvbāyātyaṃtaparinirvbāṇe[na anit]yo bu[ddho bhaviṣya]ti [d]īpa ivendhana … ā[dagdhir iva… /

67

Good man, [regarding icchantikas,] “not seeing” refers to not seeing the buddha-nature. “What is good” is anuttarā samyaksaṃbodhi itself. To say “they will not do it” refers to [an icchantika] being unable to approach a good friend who can guide him spiritually. “Only seeing” refers to seeing without good reason. The word “bad” here refers to their repudiation of the well-balanced Mahāyāna sutras. And “this they may do” refers to the fact that icchantikas do say there are no well-balanced [sutras]. The meaning of the verse is simply that icchantikas do not think in a way that advances them toward the pure and good dharma. What is the pure and good dharma? It is nirvāṇa itself! To advance toward nirvāṇa refers to the capacity to cultivate practices that are wise and good, yet icchantikas have no practices that are wise and good. This is why they are incapable of progressing toward nirvāṇa. “On that basis, one should be afraid” refers to repudiating the true-dharma. Who should be frightened? … In addition, one may also speak of “not seeing what has been done” in reference to the fact that icchantikas do not admit to themselves the host of bad things they have done. Because the icchantikas are arrogant, even though they often do things that are harmful, while doing them they initially have no sense of fear. This is why icchantikas are unable to attain nirvāṇa; they are like monkeys grabbing at the [reflection of the] moon in the water. Good man, if all living beings, however innumerable, were to all at once attain anuttarā samyaksaṃbodhi, the tathāgatas would still not see the icchantikas attaining bodhi. This is also the meaning of what I have called “not seeing what has been carried out.” Furthermore, not seeing whose deeds were carried out means not seeing that carried out by the Tathagata. The Buddha has expounded the existence of buddha-nature for the benefit of living beings, but icchantikas transmigrate through saṃsāra unable to discern what this is. It is in this sense that I used the phrase “not noticing what has been done by the tathāgatas.” Icchantikas will also look at the complete nirvāṇa of the Tathāgata and say to themselves, “This truly shows impermanence, nothing more than a lamp going out when its oil is extinguished.”68

不見者謂不見佛性。善者即是阿耨多羅三藐三菩提。不作者所謂不能親近善友。唯見者見無因果。惡者謂謗方等大乘經典。可作者謂一闡提説無方等。以是義故，一闡提輩無心趣向清淨善法。何等善法。謂涅槃也。趣涅槃者謂能修習賢善之行。而一闡提無賢善行，是故不能趣向涅槃。是處可畏謂謗正法，誰應怖畏？……復次不見所作者謂一闡提所作眾惡而不自見。是一闡提憍慢心故，雖多作惡，於是事中初無怖畏。以是義故，不得涅槃。喩如獼猴捉水中月。善男子，假使一切無量眾生一時成於阿耨多羅三藐三菩提已，此諸如來亦復不見彼一闡提成於菩提。以是義故，名不見所作。又復不見誰之所作，所謂不見如來所作。佛為眾説有佛性，一闡提輩流轉生死，不能知見。以是義故，名為不見如來所作。又一闡提見於如來畢竟涅槃，謂真無常。猶如燈滅，膏油俱尽。

69

The Sanskrit fragment states that icchantikas do not see (or understand) virtuous deeds (kalyāṇakṛta). Since this Sanskrit fragment is very likely later than what Dharmakṣema would have translated, we cannot know that this Sanskrit was what Dharmakṣema translated. The only thing I can say here is that Dharmakṣema translates this as “bu jian zhe wei bu jian foxing不見者謂不見佛性”, which differs from that in our surviving Sanskrit fragment, in his classical Chinese rendering, no matter what underlying term or phrase he read.70 The Sanskrit fragment states “the deed of the Tathāgata will not end (or destroyed) unless all beings involved in transmigration recognize the ultimate enlightenment. At that moment, the Buddha will come to complete final nirvāṇa. Along with the final nirvāṇa (atyantaparinirvāṇa), the Buddha will become changeable and absent, like fire and a lamp.” Alternatively, the classical Chinese translation by Dharmakṣema states that “bu jian shui zhi suozuo不見誰之所作 (not seeing whose deeds were carried out)” means ignoring tathāgata’s deeds. Although Buddha explained foxing for beings, icchantikas cannot recognize foxing due to their transmigration. Thus, it is called “bu jian rulai suozuo不見如來所作 (not seeing that carried out by the Tathagata).” Seeing that the Tathāgata has gone into the ultimate nirvāṇa, the icchantikas mistakenly thinks that the Tathāgata is impermanent, just like the light that goes out when the oil is exhausted. Specifically, in the Sanskrit fragment, “dhātu” does not appear in this passage. On the contrary, Dharmakṣema and his collaborators translated something as “foxing,” which is the Chinese translation term of “buddhadhātu” or “dhātu” in many cases.

Evidently, Dharmakṣema and his collaborators translated something, compared with kalyāṇakṛta and saṃmyaksaṃbodhi found in the extant Sanskrit fragment, as foxing in the Da banniepan jing. The case that this kind of translation was made by Dharmakṣema’s hand is doubtful. Furthermore, the material corresponding to the sentence “yichanti jian yu rulai bijing niepan一闡提見於如來畢竟涅槃” cannot be found in the Sanskrit fragment. Accordingly, compared to Dharmakṣema’s translation, it is difficult to identify the direct evidence that icchantika can also achieve buddhahood in the existing Sanskrit fragments.

Notably, as the above section, the difference between the Sanskrit fragment and Chinese translation is located in the ninth fascicle of the Da banniepan jing, which is attributed to a rendering by Dharmakṣema himself. According to previous research, after the finishing of the translation of the first twelve fascicles of the Da banniepan jing, Dharmakṣema stayed in Guzang姑臧 and learned the Chinese language for three years.71 In other words, the section discussed above, where the difference in foxing between Sanskrit and Chinese appears, was translated by Dharmakṣema when he was not proficient in the Chinese language. For this reason, it would be understandable if his collaborators and disciples inserted some personal views, or removed agency from Dharmakṣema, into their translations.72

According to the Gaoseng zhuan高僧傳 [Biographies of Eminent Monks], Dharmakṣema was engaged in the translation work of the Da banniepan jing from 414 to 421.73

Dharmakṣema intended to go abroad because there was a shortage in the original text of the Mahāparinirvāṇa-mahāsūtra. However, due to his mother’s death, he had no choice but to stay for several years. After that, he actually went to Khotan and found the middle portion of the Mahāparinirvāṇa-mahāsūtra. Dharmakṣema then returned to Guzang and translated it. Finally, he sent people to Khotan and found the latter portion of the Mahāparinirvāṇa-mahāsūtra. This was translated into the thirty-three fascicles of the Da boeniapan jing. This translation work was launched during the third year of Xuanshi玄始 and finished on the twenty-third of October of the tenth year of Xuanshi, namely the second year of Yongchu永初.

讖以《涅槃經》本品数未足，還外国究尋。値其母亡，遂留歳餘。後於于闐更得經本中分，復還姑臧譯之。後又遣使于闐尋得後分，於是續譯為三十三巻。以偽玄始三年初就翻譯，至玄始十年十月二十三日三袠方竟，即宋武永初二年也。

74

Furthermore, according to the Guanding灌頂 (561–632)’s record, from 414 to 416, Dharmakṣema was engaged in translating the Da banniepan jing by collaborating with Zhimeng智猛 (?–452).75

When Dharmakṣema arrived at the western Liang state西涼州, Juqu Mengxun沮渠蒙遜 dominated the Longhou隴後 area and the Xuanshi玄始 reign began. During the third year of Xuanshi, Juqu Mengxun asked Dharmakṣema to translate the Da banniepan jing. Dharmakṣema translated five fascicles of its original text into twenty fascicles of the classical Chinese translation. After that, due to the shortage in the original text, Juqu mengxun sent people abroad and found eight fascicles. They were the chapters of bingxing病行品, shengxing聖行品, fanxing梵行品, yingerxing嬰兒行品, dewang徳王品, shizihou師子吼品, jiashe迦葉品 and chenru陳如品. Dharmakṣema translated them into twenty fascicles and spread them across northern China. During the fifth year of Xuanshi, the translation work of the Da banniepan jing was complete.

到西涼州，値沮渠蒙遜割拠隴後，自号玄始。其号三年，請曇無羅讖共猛訳五品，得二十巻。遜恨文義不圓，再遣使外国，更得八品。謂病行、聖行、梵行、嬰兒行、德王、師子吼、迦葉、陳如等品。又翻二十巻，合成四十軸，伝於北方。玄始五年乃得究訖。

76

If these documents are examined together, they record that Dharmakṣema clearly translated the original text of the Mahāparinirvāṇa-mahāsūtra with Zhimeng and the support of other collaborators at least twice.77 If so, the significant difference in foxing between the extant Sanskrit fragments and Dharmakṣema’s Da banniepan jing discussed in this section belongs to the first twelve fascicles of his translation, which was translated by Dharmakṣema and his collaborators when he had not yet mastered the Chinese language. Accordingly, the role of Dharmakṣema’s collaborators and disciples in his translation of the Da banniepan jing is important. Needless to say, it is also likely that Dharmakṣema had seen Faxian’s translation, and its use of foxing, and himself reasoned that this was a good way of communicating what the Mahāparinirvāṇa-mahāsūtra was teaching.

Regarding my hypothesis mentioned above, one of my reviewers once noted: “Even if Dharmakṣema was not familiar with the Chinese language at that time, in my opinion, a translator’s limited proficiency in the target language does not make it more likely that the translator would insert into his translation something that is not there in the source language. Probably the reverse argument can also be made, namely, a translator who has excellent proficiency in the target language would then be more likely to insert into his translation something that was not there in the source language. Alternatively, I don’t think it makes much sense for the author to suggest that, since his Chinese language was not good enough, then his collaborators and disciples inserted some personal views into their translations. Were Dharmakṣema’s colleagues to add anything into the translation of the Da banniepan jing, this inserted notion must had already become quite popular before the text was translated.” I have to add something of my response towards this query here. In my view, it is likely that either Dharmakṣema or his collaborators and disciples translated some other terms as the Chinese term foxing. There are further two possibilities. One is that they created the term foxing here. Another possibility is that they used the term foxing, which had already become popular before the translation of the Da banniepan jing. For the second possibility, when I say Dharmakṣema’s collaborators and disciples inserted some personal views into their translations, needless to say, it is also possible that their personal views had been influenced by some terms which had already become popular then. In other words, I do not think that there is a fundamental contradiction between my reviewer’s hypothesis and that of mine, although both of our views are merely assumptions.78

Additionally, some sections of the Da banniepan jing translated by Dharmakṣema state that icchantikas also possess foxing.

For those who are icchantikas, although they possess buddha-nature, they are held down by the stain of their innumerable transgressions, unable to get free, like silkworms inside of cocoons. Because of their karmic conditions, they cannot produce a marvellous cause that would lead to bodhi and instead transmigrate through saṃsāra with no end in sight.79

彼一闡提雖有佛性，而為無量罪垢所纏，不能得出。如蠶所繭。以是業緣，不能生於菩提妙因。流轉生死，無有窮已。

80

As mentioned above, Chinese translators used the term foxing to correspond with various original terms. In this section, they translate “bi yichanti sui you foxing彼一闡提雖有佛性.” Furthermore, although the Chinese translation states that icchantikas can merely float in the stream of birth and death without becoming free from transmigration, according to its interpretation, icchantikas definitely possess foxing. This statement has strongly influenced even wider East Asian Buddhist thought.

Conversely, the Nihuan jing translated by Faxian states the following.

Icchantikas are separated from the rulai xing如來性 (lineage of the tathāgata) forever due to committing the crime of criticizing the Buddhist Dharma. It is like the cocoon created by bugs which controls the bugs themselves, so do icchantikas. They cannot stimulate their origins of Bodhi in the lineage of the tathāgata, so that they cannot become free from transmigration during all lives.

彼一闡提於如來性所以永絶，斯由誹謗作大惡業。如彼蠶虫綿網，自纏而無出處。一闡提輩亦復如是。於如來性不能開發起菩提因，乃至一切極生死際。

81

Evidently, the rulai xing如來性 found in this section of the Nihuan jing corresponds to the foxing stated in the Da banniepan jing.82 However, the Nihuan jing clearly states that icchantikas are forever separated from the rulai xing due to committing the crime of criticizing the Buddhist Dharma. Although Faxian also used the term foxing in the Nihuan jing, there were fewer uses of the term than those in the Da banniepan jing. In this section, Faxian used the term rulai xing instead of foxing.

Regardless, it remains necessary to confirm whether the original Sanskrit text of the Nihuan jing and that of the Da banniepan jing are identical or not. From the perspective discussed above, on the relationship between icchantika and rulai xing (or foxing), it seems likely that the assertions of Dharmakṣema’s Chinese collaborators also influenced the translation more or less.83 Identical to the classical Chinese translation by Dharmakṣema, the Tibetan translation of the Mahāparinirvāṇa-mahāsūtra states that Buddha-nature is also within the icchantikas’ bodies; among various translations of the Mahāparinirvāṇa-mahāsūtra, the only version that states the icchantika without Buddha-nature is the Nihuan jing. It is very likely that the translators of the Tibetan translation consulted Dharmakṣema’s classical Chinese translation.84

According to Ōchō, the Nihuan jing denies the possibility that icchantikas can achieve buddhahood, whereas the Da banniepan jing states that icchantikas can achieve this if they successfully see their shanxin善心 (good mind).85 The discussion above also clarifies Ōchō’s assertion. The sentence “icchaṃtikāḥ kalyāṇakṛtaṃ na” in the extant Sanskrit fragment corresponds to “bu jian zhe wei bu jian foxing不見者謂不見佛性.” “Icchantikas fail to see virtuous deed (kalyāṇakṛta)” in this Sanskrit fragment corresponds to “icchantikas can achieve buddhahood” in the Da banniepan jing. While we cannot know what Dharmakṣema was seeing in his Sanskrit text, I contend that this translation strongly supported the theory that icchantikas can achieve buddhahood in East Asian Buddhism.86

Concerning this issue, Takasaki notes that the Da banniepan jing translated by Dharmakṣema, alongside the Nihuan jing translated by Faxian, also states that icchantikas do not possess foxing before the eleventh fascicle. In contrast, after this fascicle, the Da banniepan jing admits the possibility that icchantikas could finally achieve buddhahood. Saliently, the above portion can only be found in the classical Chinese translation rather than the Sanskrit or other texts.87 Furthermore, Matsumoto Shirō松本史朗 asserts that in the Da banniepan jing, the you foxing有佛性 (possessing Buddha-nature) does not mean jie chengfo皆成佛 (accomplishing buddhahood for all beings).88 Both Takasaki and Matsumoto were aware of the difference between the first twelve and subsequent fascicles of the Da banniepan jing. Clearly, their assertions reinforce my opinion.

To summarize, as a classical Chinese Buddhist canon text translated at the beginning of the fifth century, the Da banniepan jing used the term foxing, which cannot be correspondingly confirmed in the surviving Sanskrit fragments of the Mahāparinirvāṇa-mahāsūtra. Foxing was most naturally translated buddhadhātu, but the Sanskrit fragments do not mention buddhadhātu. Those sections where the difference between the Sanskrit and the Chinese translation of foxing appears belong to the translation made by Dharmakṣema before he was proficient in the Chinese language.89 For this reason, it is not impossible that his collaborators and disciples may have inserted some personal views into their translations. It is possible that his inserted notion had already become popular before the translation of the Da banniepan jing. Meanwhile, it is also likely that Dharmakṣema had seen Faxian’s translation, and its use of foxing, and himself reasoned that this was a good, shorthand way of communicating what the Mahāparinirvāṇa-mahāsūtra was teaching.

Moreover, over half of the Da banniepan jing, after the first 10 juan, is unique to this version. We have no Sanskrit fragments corresponding to its content and no Tibetan, apart from a Tibetan translation made from Dharmakṣema’s Chinese translation.90

5. Foxing in the Da fangdeng wuxiang jing 大方等無想經 and Guṇabhadra’s Renderings

In this short section, as a supplement for this study, I intend to slightly mention the Da fangdeng wuxiang jing 大方等無想經 [Skt. Mahāmeghasūtra; T387] by Dharmakṣema and the renderings by Guṇabhadra 求那跋陀羅 (394–468) from roughly the same time.

As is mentioned in the first section of this article, according to the Chu sanzang ji ji出三藏記集, 11 texts were regarded as Dharmakṣema’s translations. It is impossible to analyze the term foxing in all of them in this space-limited article. Since we have a Tibetan translation of the Da fangdeng wuxiang jing, the Sprin chen po’i mdo (Derge no. 232; Peking no. 898), I merely have a look at one case of foxing in this text.

The Mahāmeghasūtra is a tathāgatagarbha doctrinal sutra, overlapping with the Mahāparinirvāṇa-mahāsūtra. The Da fangdeng wuxiang jing, the Chinese rendering of the Mahāmeghasūtra, was also translated by Dharmakṣema. In the Da fangdeng wuxiang jing and the corresponding Tibetan translation, we can find the following example:

Da fangdeng wuxiang jing大方等無想經 (Taishō no. 387, 1102b2–3)	Sprin chen po’i mdo (Derge no. 232, 194b)
猛風起者，喻如來常。風入毛孔者，喻諸眾生悉有佛性。	de bzhin du ’dir yang ting nge ’dzin gyis de bzhin gshegs pa’i yon tan rtag pa nyid kyi yon tan gyis bsgos pa’i rlung nyon mongs pa’i nam mkha’ la ldang bar byed cing/

In the Tibetan rendering, confirming a reasonable corresponding term to the Chinese term foxing here is a little difficult. We cannot find de bzhin gshegs pa’i snying po (buddha-nature), but only de bzhin gshegs pa’i yon tan or rtag pa nyid kyi yon tan. We also confirm some similar cases in the Da fangdeng wuxiang jing like this.91 In other words, Dharmakṣema uses the term foxing in the Da fangdeng wuxiang jing, while the corresponding term or phrase in the Tibetan translation is unclear.

Although the Da fangdeng wuxiang jing may appear to be a partial translation of the Mahāmegha-sūtra, the real situation was more likely that the original text was still incomplete when Dharmakṣema brought it to China.92

Guṇabhadra was born in central India to a brāhmaṇa family and departed from Sri Lanka for China, arriving in Guangzhou by sea in around 435.93 He translated some famous Mahāyānist sutras, including the Yangjuemoluo jing央掘魔羅經 [Pāli. Aṅgulimāla-sutta; Skt. Aṅgulimālīyasūtra; T120] and the Da fagu jing 大法鼓經 [Skt. Mahābherīhārakasūtra; T270], in which the term foxing can be found. The Da banniepan jing translated by Dharmakṣema, however, was brought to Jiankang建康, present-day Nanjing南京, becoming the foundation of the southern version of this scripture’s Chinese rendering. Huiyan 慧嚴 (363–433), Huiguan 慧觀 (4th to 5th centuries CE) and Xie Lingyun 謝靈運 (385–433) edited this scripture into the southern version in 436.94 Moreover, with their help, Guṇabhadra translated some texts.95 For this reason, the term foxing found in Guṇabhadra’s renderings, which were translated later than the renderings translated by Faxian, Buddhabhadra and Dharmakṣema, was probably more or less influenced by the Da banniepan jing.96

6. The Interpretations of Foxing in Later Chinese Buddhism

Dharmakṣema was proficient at incantation and respected in many countries.97 Finally, he was assassinated by Juqu Mengxun沮渠蒙遜 (368–433), the King of Northern Liang北涼. Northern Wei destroyed Northern Liang very soon afterwards. Dharmakṣema’s disciples and collaborators moved to Pingcheng平城, the capital of Northern Wei北魏. Furthermore, the Da banniepan jing translated by Dharmakṣema was brought to Jiankang建康, becoming the foundation of the southern version of this scripture’s classical Chinese translation. Hence, in my opinion, the translations and concepts in Dharmakṣema’s system strongly influenced Buddhism during the Northern Wei period, especially Bodhiruci菩提流支 (6th century CE; active in China after 508) and Ratnamati勒那摩提 (6th century CE; active in China after 508). The Da banniepan jing, which was sufficiently researched in Northern Wei, became the foundation of the doctrines of the Dilun tradition地論宗, including Huiyuan of the Jingying temple淨影寺慧遠 (523–592).98 The influence of the Da banniepan jing on Chinese Buddhist thought is apparent. As abundant amount of research already exists on this issue99; I will, therefore, merely discuss the cases of Huiyuan of the Jingying temple and Guanding of the Tiantai tradition天台宗 in this section.

In the Da banniepan jing yi ji大般涅槃經義記 [Meaning of the Great Nirvana Sutra], Huiyuan’s commentary on the Da banniepan jing, he states the following.

There is a type of Buddha-nature that the icchantikas have but those who possess wholesome roots do not. [Namely, the icchantikas] have the unwholesome nature, and hence they lack wholesome nature. Due to dependent origination [based upon] the Buddha-nature, unwholesome aggregates arise. Hence unwholesome aggregates are named Buddha-nature, which the icchantikas have. There is another type of Buddha-nature that those who possess wholesome roots have but the icchantikas don’t. Those who have advanced above the first [bodhisattva-]stage are called people with wholesome roots. Or more broadly, the bodhisattvas above the stage of buddha-gotra (Chin. zhongxing di 種性地) are named wholesome human beings (i.e., people with wholesome roots). They have wholesome nature and lack unwholesome nature. There is a type of Buddha-nature that both [of the above two groups of people] have, namely, they both have the nature as the principle (lixing 理性is a short form for li foxing 理佛性). There is another type of Buddha-nature that both [of the above two groups of people] do not have, namely, neither of them has the nature as the result (meaning that they have not attained Buddhahood).

或有佛性，一闡提有，善根無者，有不善性，無其善性。佛性緣起為不善陰，故不善陰名為佛性。闡提有此。或有佛性，善根人有，闡提無者，初地已上名善根人，通則種性已上菩薩斯名善人。彼有善性，無不善性。或有佛性，二人俱有，俱有理性。或性，二俱無，俱無果性。

100

Huiyuan’s interpretation contains vital information. The statement that icchantikas also possess Buddha-nature (foxing) is clearly influenced by the Da banniepan jing. According to Huiyuan’s explanation, icchantikas have the arising Buddha-nature and principal Buddha-nature. Among these two, the arising Buddha-nature is only possessed by icchantikas. Thus, it is clear that Huiyuan was deeply influenced by the Da banniepan jing and regarded it as the foundation of his theory of Buddha-nature.101 The most significant connection in this passage is the term foxing, which cannot be correspondingly confirmed in the current Sanskrit fragments of the Mahāparinirvāṇa-mahāsūtra.

Accordingly, after Northern Wei extinguished Northern Liang, the Da banniepan jing translated by Dharmakṣema, and his assistants’ interpretations, were likely conveyed to Pingcheng, the Northern Wei capital, and influenced the Dilun monastic tradition, including monks such as Huiyuan.

Moreover, Guanding states in the Da banniepan jing shu大般涅槃經疏 [Commentary on the Mahāparinirvāṇa-Sūtra] as below.

First, hearing (Chin. wen聞) is divine ear (Skt. divyaśrotra; Chin. tianer 天耳). Seeing (Chin. jian見) is divine eye (Skt. divyacakṣus; Chin. tianyan 天眼). They relate to the jishen tong即身通 (penetrating understanding with the body). Second, the ninth stage is hearing, in which one can see Buddha-nature. The tenth stage is sight, in which one can complete and clarify himself through seeing Buddha-nature. Achieving the ninth stage by liberation of wisdom is the particular hearing which is manifested without normal hearing. Achieving the tenth stage from the ninth stage is the particular seeing which is manifested without normal seeing. Achieving the buddhahood stage from the tenth stage is the particular achieving which is manifested without normal achieving.

一云，聞即天耳，見即天眼，至即身通。二云，九地為聞，見佛性，十地為眼。見佛性，具足明了。今因慧解脱至第九地，是不聞而聞。因九地至十地，即不見而見。因十地至佛地，為不至而至。

102

Guanding mentions the term “jian foxing見佛性” as found in the Da banniepan jing translated by Dharmakṣema in his commentary and states that one would see foxing if he has achieved the ninth stage of bodhisattvas’ stages. As has been discussed above, Dharmakṣema and his assistants translated something, which is reported as the gotra of tathāgata (rigs) and good deeds (kalyāṇakṛta) in the surviving Sanskrit fragments of the Mahāparinirvāṇa-mahāsūtra, as jian foxing (seeing Buddha-nature).103 Evidently, the gotra of tathāgata (rigs) means those people who will or have achieved the boundary of the tathāgata. However, Guanding only used the term jian foxing, while zhongxing (Skt. gotra; lineage/caste) cannot be found. Furthermore, he attempted to integrate jian foxing with the theory of the stages of the bodhisattvas, especially the ninth stage.104

Evidently, the Da banniepan jing was translated by Dharmakṣema with his assistants’ interpretations, influenced not only the Dilun monastic tradition, but also the Tiantai monastic tradition and monks such as Guanding. The term “jian foxing” found in the Da banniepan jing was emphasized by Guanding, the direct disciple of Zhiyi智顗 (538–597).105 As is well known, this term greatly influenced the later Tiantai tradition, the Huayan華嚴 (Jp. Kegon) tradition and Chan禪 (Jp. Zen) Buddhism through some of the early Tiantai monks such as Guanding. According to Whalen Lai, the Tiantai tradition, based on the Lotus Sūtra (Chin. Fahua jing法華經), superseded the Nirvāna tradition涅槃宗 by incorporating many of its ideas.106 We can therefore imagine the wide influence of the Da banniepan jing (Nirvāna Sūtra) and the Nirvāna tradition. Needless to say, the most important idea of the Nirvāna tradition is the theory of foxing.107 However, also as discussed above, the extant Sanskrit fragments of the Mahāparinirvāṇa-mahāsūtra states that icchantikas do not see good deeds (kalyāṇakṛta). Instead, the term or phrase in this corresponding place was rendered as “bu jian foxing不見佛性” in the Da banniepan jing. The most important term foxing cannot be found as a fixed term in our current Sanskrit fragment.

While probably correct from the perspective of those Indic original texts, I suppose that a crucial point exists. That is, from the perspective of a Chinese reader, in all these cases there is only one single term—the Chinese word foxing佛性.108

7. Conclusions

In East Asian Buddhism, rulaizang (Skt. tathāgatagarbha) is sometimes considered a synonym of foxing (Buddha-nature) because the relationship between these two terms was ambiguous in Chinese Buddhism since some monks and schools declared that foxing is the same as rulaizang. The early translators who emphasized some translated terms as foxing were Buddhabhadra and Dharmakṣema, two Indian Buddhist monks living in China in the first half of the fifth century. That is to say, the cases of the Chinese term foxing appeared during the Northern Liang dynasty (397–439) and the second half of the Eastern Jin (317–420) are probably the key to probing some early cases where the term foxing appeared.

The Da fangdeng rulaizang jing (Skt. Tathāgatagarbha-sūtra) translated by Buddhabhadra is a very early classical Chinese Buddhist canon text where the term foxing is clearly used to express Buddha-nature. However, the Chinese term foxing is difficult to confirm in Amoghavajra’s classical Chinese translation. Although a lack of clarity remains about Buddhabhadra’s motivation, as an early classical Chinese Buddhist canon text, the Da fangdeng rulaizang jing used the term foxing, which cannot be confirmed in other extant translations of the Tathāgatagarbha-sūtra.109

Compared to the Da fangdeng rulaizang jing, the Da banniepan jing (Skt. Mahāparinirvāṇa-mahāsūtra) translated by Dharmakṣema has exerted a much greater influence on Chinese Buddhist thought. As another early classical Chinese Buddhist canonical text, the Da banniepan jing also used the term foxing, which cannot be correspondingly confirmed in the surviving Sanskrit fragments of the Mahāparinirvāṇa-mahāsūtra.110 The sections where the differences between the Sanskrit fragment and the Chinese term foxing appear belong to Dharmakṣema’s early translation before he was proficient in the Chinese language.

Furthermore, it is very unrealistic to believe that the same person or group simultaneously translated all forty fascicles of the Da banniepan jing. Different people would have edited these fascicles in several stages. Notably, buddhadhātu, the original Sanskrit term of the Chinese term foxing, which is regarded as the most significant term in the Da banniepan jing, cannot be found in the extant Sanskrit fragments of this scripture. Dharmakṣema translated the original text of the Mahāparinirvāṇa-mahāsūtra with Zhimeng and the support of other collaborators at least twice. The significant difference between the Sanskrit fragments and the classical Chinese translation of the Mahāparinirvāṇa-mahāsūtra in this article belongs to the first twelve fascicles of Dharmakṣema’s translation aided by his disciples and collaborators when he had not yet mastered the Chinese language. Therefore, we should not ignore the role of his assistants. Of course, it is also likely that Dharmakṣema had seen Faxian’s translation and its use of foxing and, himself, reasoned that this was a good, shorthand way of communicating what the Mahāparinirvāṇa-mahāsūtra was teaching.

Meanwhile, we frequently find sangs rgyas kyi khams/dbyings, which is a translation of buddhadhātu in the Tibetan rendering of the Mahāparinirvāṇa-mahāsūtra. This leads us to presume that Faxian and Dharmakṣema both read versions of the Mahāparinirvāṇa-mahāsūtra that used this term and translated it and other terms, including those I mentioned in this article, with foxing.

It is likely that Faxian translated a version of the Mahāparinirvāṇa-mahāsūtra that featured buddhadhātu as foxing. Buddhabhadra, in the same period, translated a version of the Tathāgatagarbha-sūtra. In some passages, he had favoured the term foxing over a literal translation of the Sanskrit. As a contemporary monk with Buddhabhadra and Faxian, Dharmakṣema translated the Mahāparinirvāṇa-mahāsūtra, going further than Faxian by using the term foxing regularly. Our Sanskrit fragments of the Mahāparinirvāṇa-mahāsūtra are surely of a later date.111 We can suspect that both Dharmakṣema and Buddhabhadra employ foxing as a non-literal translation, after Faxian.

Moreover, after Northern Wei extinguished Northern Liang, the Da banniepan jing translated by Dharmakṣema, and the interpretations of his collaborators and disciples were likely conveyed to Pingcheng, the Northern Wei capital. These two texts translated by Buddhabhadra and Dharmakṣema respectively, especially the Da banniepan jing, deeply influenced the Dilun monastic tradition. Among these, the term foxing and its Sinicism explanations played a highly significant role, influencing the whole of East Asian Buddhist thought. Needless to say, the controversies focusing on the concept of “Buddha-nature” within all sentient beings in East Asian Buddhism, including the theory of tathāgatagarbha, are closely related to the term foxing and its Sinicism explanations discussed in this article. However, it is difficult to clarify the accurate origin of the Chinese term foxing at least at the beginning of the fifth century in the relevant Sanskrit and Tibetan fragments and texts at present.112

The aim of this article was not to be exhaustive or comprehensive but to provide some additional reflections on the term foxing represented in the Da fangdeng rulaizang jing and the Da banniepan jing, two contemporary classical Chinese renderings, suggesting possible further research. Although it is a little difficult to say that the Da fangdeng rulaizang jing and the Da banniepan jing are the earliest classical Chinese Buddhist canon texts where the term foxing is clearly used to express Buddha-nature, these two Chinese renderings are very early-stage translations in this sense. It is hoped that this study can make a small contribution to reconsider the origin and background of the Chinese term foxing within the historical context of Chinese Buddhist translation.