The theory of tathāgatagarbha—most commonly, if not perfectly, translated into English as “buddha-nature”[1]—is generally thought by scholars to have first appeared around the third or fourth century CE and possibly as early as the second. Many Tibetan and Chinese scholiasts found justification for the ideas in various passages in the Pāli Canon, such as this from the Aṅguttaranikāya Sutta: “Luminous, monks, is this mind, but sometimes it is defiled by adventitious defilements. . . . sometimes it is free from adventitious defilements.”[2] The Vinaya contains a famous story in which the Buddha sends his gaze over all existence and perceives sentient beings as lotuses rooted in deep mud; the metaphor is taken as pertaining to the buddha-nature of all beings, destined as we are to attain perfectly pure enlightenment.
Traditional and modern scholars debate how much of a link can be found between early Pāli references to “luminosity” and buddha-nature.[3] Mainstream Pāli Buddhism considered consciousness to be one of the five skandhas, the building blocks of conditioned existence. Early exegesis of luminosity passages in the scripture seems to suggest that they were not, in fact, teaching that the mind is naturally pure or that it preexists the skandhas, but only that it has the potential to be made pure.[4] A related concept is bhavaṅga mind, meaning the substratum of consciousness that represents mind in its inactive state. This does not appear originally to have been intended as a permanent subconscious; at the moment the mind becomes active, bhavaṅga is cut off and the active mind (vīthicitta) takes over. Still, some scholars have pointed to the concept as a forerunner to the notion of luminosity.[5]
Although over the centuries Chinese and Tibetan scholiasts have categorized the concept of buddha-nature as either Yogācāra or Madhyamaka, there is sufficient reason to believe that the tathāgatagarbha theory developed independently: it is a cataphatic doctrine (that is, it uses positive language to describe the nature of reality), which distinguishes it from the apophatic approach of Madhyamaka; and it asserts that all sentient beings have an equal capacity to awaken, which contradicts the basic Yogācāra doctrine of different potentials for enlightenment. Instead, the rise of the doctrine was likely a result of Buddhist theorists grappling with long-standing core Buddhist conundrums such as the nature of mind; how to use language to describe what is by definition beyond the reach of language; how nirvāṇa, which is unconditioned and perfect, can arise out of saṁsāra; and how to make sense of various yogic experiences.
Early Appearances of the Term Tathāgatagarbha Scholars currently debate the earliest (surviving) appearance of the term tathāgatagarbha. The term itself appears in a handful of early scriptures but without elaboration, suggesting that the term had been coined but its meaning had not yet been fleshed out. These are documented by Karl Brunnhölzl in When the Clouds Part: Possibly the first appearance of the term tathāgatagarbha (though not in the sense in which it is used in the tathāgatagarbha sūtras) has been traced to the Mahāsaṃghika Ekottarikāgama (the Chinese recension of the Aṅguttara Nikāya): “If someone devotes himself to the Ekottarikāgama / Then he has the tathāgatagarbha. / Even if his body cannot exhaust defilements in this life / In his next life he will attain supreme wisdom.” The term is also used once in the Gaṇḍavyūhasūtra (which is dated prior to the Tathāgatagarbhasūtra) as an epithet of Sudhana, without further explanation. Furthermore, the Prajñāpāramitāsūtra in One Hundred Fifty Lines (Adhyardhaśatikāprajñāpāramitāsūtra) contains the sentence “all sentient beings possess the tathāgatagarbha because their entire being is that of the great bodhisattva Samantabhadra.” In When the Clouds Part Brunnhölzl also surveys the literature to which the earliest Indian treatise on tathāgatagarbha, the Ratnagotravibhāga, makes reference. These include the Dhāraṇīśvararājasūtra and the Ratnadārikāsūtra, among others. Although these scriptures do not use the term tathāgatagarbha, they provided the treatise’s author with much of the doctrinal basis for the explication of the theory. One such concept on which tathāgatagarbha theory relies is gotra, a Sanskrit term that refers to family unit by bloodline and is used metaphorically in Buddhism to refer to “class,” “lineage,” or “disposition.” Buddhist teachings since the early days of the religion discussed the various predilections of followers, a way of separating the children of the “noble” class—those who are sincere in their renunciation and diligent in their austerities—from the rest of humanity. In the Mahāyāna three basic classes of Buddhist practitioners were said to exist: śrāvakas, who will become arhats by following the Hīnayāna path; pratyekabuddhas, who will become arhats without being taught; and bodhisattvas, or those destined to become buddhas on the Mahāyāna path. An additional gotra was posited in some sūtras: that of the icchantika, who does not possess tathāgatagarbha and therefore has no possibility of becoming enlightened. Whether or not such a class of beings truly existed was one of the earliest controversies in buddha-nature theory.
Tathāgatagarbha Scripture A handful of texts that are sometimes collectively labeled “tathāgatagarbha sūtras” are generally agreed upon as the initial group of literature that developed the concept of buddha-nature as we know it today. These stand distinct from the Yogācāra scriptures such as the Saṃdhinirmocanasūtra, and from the Prajñāpāramitā literature that provided the foundation for the Madhyamaka; so much so that some historians have posited the existence of a third Indian Mahāyāna school alongside them: the Tathāgatagarbha school. Among the most important of these texts are the Mahāparinirvāṇasūtra, the Tathāgatagarbhasūtra, the Anūnatvāpūrṇatvanirdeśaparivarta, the Śrīmālādevīsiṃhanādanirdeśa, the Mahābherīsūtra, and the Aṅgulimālīyasūtra. While later Mahāyāna scriptures such as the Laṅkāvatāra and the Lotus Sūtra also teach tathāgatagarbha, the above-named scriptures predate the popular Ratnagotravibhāga, a fourth- or early fifth-century Indian treatise that systematized tathāgatagarbha theory, and so are considered the first wave of the doctrine. The dates of their creation are unknown, and there is as yet little consensus concerning the sequence of their appearances. The Mahāparinirvāṇasūtra and the Tathāgatagarbhasūtra are the likeliest candidates for the earliest surviving instance of the term tathāgatagarbha used in the sense that it has come down to us. Michael Radich dates the Mahāparinirvāṇasūtra to as early as the second century CE and claims that it is the earliest of the group, while Michael Zimmermann dates the Tathāgatagarbhasūtra to the third century; he once argued that this sūtra was the earliest of the group but has since backed away from that assertion in light of Radich’s findings. The Mahāparinirvāṇasūtra, like the Hīnayāna sūtra of the same name, is ostensibly a narrative about the final days of the Buddha, but this one extends into a discourse of Mahāyāna doctrine. The Buddha is depicted not as dying but as entering a nirvāṇa that is an enduring presence rather than an extinction. This seems to be the main thrust of the sūtra: to proclaim that the Buddha is ever-present and to equate parinirvāṇa with the eternal and all-pervading dharmakāya, which eventually came to be equated in the sūtra with tathāgatagarbha. The sūtra in fact inverts what are known as the four viparyāsas, or wrong views: that any phenomenon can be described as being free from suffering, permanent, pure, or endowed with a self. The Mahāparinirvāṇasūtra instead states that the Buddha, his enlightenment, and the buddhadhātu should all be properly described as blissful, permanent, pure, and endowed with a self. That permanent buddhahood, which is only masked by temporary stains, is tathāgatagarbha. (In typical parochial Mahāyāna fashion, the Mahāparinirvāṇasūtra reserves complete enlightenment for only those who have completed the Mahāyāna path; the nirvāṇa of the arhat is merely free of the stains, lacking the awareness of the buddhadhātu and bliss.) The Mahāparnirvāṇa’s liberal use of the term self (ātman) to describe tathāgatagarbha was controversial, flying in the face of one of the central doctrines of Buddhism, that of no-self, or anātman. As Christopher Jones points out, two additional tathāgatagarbha sūtras do the same, the Aṅgulimālīyasūtra and the *Mahābherīsūtra (both of which will be introduced below), leading him to speculate that opposition to these sūtras from within Buddhist communities was the reason later tathāgatagarbha sūtras dropped the use of the term ātman. In his study of the scripture Radich argues that the term tathāgatagarbha, which he glosses as “womb of a buddha,” was used to explain how a perfectly pure being such as a buddha could arise out of a polluted and degraded human being; how, in other words, the conditioned could give rise to the unconditioned. This line of argument remains one of the more popular defenses against the claim that buddha-nature theory is non-Buddhist; if sentient beings and buddhas do not share the same nature, defenders assert, the attainment of enlightenment cannot be explained. Either saṃsāra must be wiped away to reveal what is already present, or a spark of enlightenment that is part of a saṃsāric being’s essence is brought to fruition. Otherwise nirvāṇa is the result of some action and therefore determined by causes and conditions, a view that is abhorrent to any Buddhist; nirvāṇa is precisely the absence of any conditioning. As Brunnhölzl describes it, the Mahāparinirvāṇasūtra offers an ambiguous definition of tathāgatagarbha: it is an intrinsic pure nature that all sentient beings possess and of which they will become aware once obscurations are removed, and it is a seed or potential that will ripen into buddhahood once all conditions are present. Buddha-nature, it would seem, was from the very early days a doctrine that contained both an ontological and a soteriological assertion. In the first case it is a statement about the nature of reality: sentient beings are by nature perfect, but that perfection is obscured by stains that nevertheless do not impact its essence; that perfection is moreover equated with the nature of reality itself, and therefore buddha-nature becomes the basis for both saṁsāra and nirvāṇa. In the second case it is an ethical proposal relating to salvation: the potential for perfection is present in all sentient beings, but they must each strive to actualize it. This bifurcated definition would continue through all presentations, to the delight or consternation of many commentators. The Mahāparinirvāṇasūtra appears to have been compiled in at least three stages. As described by Liu Ming-Wood and also by Takasaki, the earliest section comprises the first five chapters, which read as a complete text and end with the final days of the Buddha. Chapters 6 through 9 clarify points made in the first, a commentary of sorts in the guise of a continuation, and the final section, chapters 10 through 13, add further explanation. As Christopher Jones explained, Japanese scholar Shimoda Masahiro suggested that the earliest core of the text was concerned with the Buddha’s permanent existence; rather than vanishing into nirvāṇa, here the buddha is permanent and omnipresent. The accretion of the tathāgatagarbha doctrine represents a transition of the Buddha’s body into that of sentient beings, the Buddha’s presence becoming the true self of ordinary beings. This suggests an interesting link between the early Buddhist concern with the relics—and lasting presence—of the Buddha with the doctrine of buddha-nature. In any case, the sūtra teaches, this innate buddha-body of sentient beings, which came to be called tathāgatagarbha, represents their true self. A primary divergence between the first and later sections of the Mahāparinirvāṇasūtra is in their positions on buddha-nature and the icchantikas, the class of beings who are beset with the gravest of flaws such that they can never become enlightened; by definition they are devoid of buddha-nature. The first section of the sūtra is adamant that the icchantikas do not have buddha-nature and can never become enlightened; they are a scorched seed that can never sprout. The second section is ambiguous on the subject, and the third states unequivocally that icchantikas do have buddha-nature and therefore do have the potential to become enlightened. By bestowing buddha-nature on the icchantika the additions brought the sūtra into full conformity with the Single Vehicle (Ekayāna) teachings of the Saddharmapuṇḍarīkasūtra—the Lotus Sutra—which influenced it and other early tathāgatagarbha sūtras, and concurrently into contradiction with fundamental Yogācāra doctrine of the three natures.
Whether or not it was the first tathāgatagarbha sūtra, most scholars agree that the Tathāgatagarbhasūtra was among the earliest of the group. It has been translated into English by Grosnick and again by Zimmermann, who also prepared critical editions of the Chinese and Tibetan which he published together with a lengthy study. Zimmermann explains in patient detail that there are two versions of the text, the first of which lacks much of the content of the second, later recension (see below in the section on translations into Chinese). Like the Mahāparinirvāṇasūtra, the Tathāgatagarbhasūtra offers both ontological and soteriological definition of tathāgatagarbha, although “definition” is probably not the right word: the short text simply lists nine similes to describe the concept. These include a golden statue covered in mud and a seed that is destined to grow into a tree, suggesting both an already perfected nature and the potential to become something that one is presently not. Zimmermann and others have noted that tathāgatagarbha theory may have initially been developed more for an ethical and soteriological purpose; the Tāthagatagarbhasūtra did not have to explain the idea with complicated philosophical arguments because it was intended to encourage and inspire, not convince. It is an appeal to emotion rather than the intellect. Madhyamaka and Yogācāra schools of doctrine were then in ascendance in Mahāyāna communities, and it is reasonable to hypothesize that practitioners were put off by the seeming nihilism of Madhyamaka; emptiness is too easily interpreted to mean that the ultimate is, terrifyingly, simply a void. Yogācāra, meanwhile, advocated a theory of “class” or “disposition” (gotra) in which only certain beings were said to be able to attain enlightenment. Such a doctrine might leave some of the faithful—not to mention potential converts—feeling left out. The early tathāgatagarbha literature countered both. It offered a positive description of the ultimate—buddha-nature, the true and real nature of both a person and reality—and it guaranteed complete and perfect enlightenment to all beings who were willing to strive for it (on the Mahāyāna path, of course). Yogācāra, it should be noted, also uses positive language to describe the ultimate—mind, at least in later Yogācāra scriptures, is said to be truly existent—and this has led some scholars to erroneously label tathāgatagarbha a Yogācāra doctrine. Another early tathāgatagarbha scripture, perhaps one of the most influential, was also an early instance of the concept of a single vehicle, merging all previous Buddhist doctrine into a single doxographical order. This was the Śrīmālādevīsiṃhanādanirdeśa, known in English as The Lion’s Roar of Queen Śrīmālā. Although it is no longer extant in Sanskrit, it was evidently highly influential in India, judging by the many references to it in other scripture. Diana Paul argues that it was composed at least by the year 350, to give it time to gain popularity in India and be brought to China, where it was translated in 435. In the sūtra Queen Śrīmālā of Ayodhyā is prompted by a letter from her parents to supplicate the Buddha, who appears before her and inspires her to teach. The main topics of her discourse are tathāgatagarbha, the single vehicle, and the Four Noble Truths. Although the sūtra affirms that all beings share the same buddha-nature, the Śrīmālā asserts that śrāvakas and pratyekabuddhas (that is, non-Mahāyāna Buddhists) cannot comprehend the steps needed to shed adventitious stains and reveal the intrinsic purity of mind; only bodhisattvas can. Thus while the sūtra proclaims a single vehicle and universal buddha-nature, it does so with some reservation, suggesting a Yogācāra influence. This inequality is probably an indication of the scripture’s early date: Mahāyāna communities were still in competition with the earlier Buddhist orders, and bodhisattvas and śrāvakas could not be depicted as equals. In When the Clouds Part Karl Brunnhölzl draws attention to a novel conception of emptiness in the Śrīmālā: tathāgatagarbha “is empty of adventitious stains but not empty of its limitless inseparable qualities.” With this the sūtra seems to be addressing the question of how tathāgatagarbha theory is to be addressed by Madhyamikas. If tathāgatagarbha is another name for emptiness, as some Madhyamaka theorists would argue, then buddha-nature ought to conform with Madhyamaka definitions of emptiness and lack its own qualities. Instead, buddha-nature is described as empty of all but its own characteristics, an early suggestion of a philosophical view that came to be known in Tibet as “other-emptiness.” The above three early tathāgatagarbha scriptures describe an ultimate nature that is naturally pure but is obscured by adventitious stains. The very short Anūnatvāpūrṇatvanirdeśaparivarta further expands the ontological aspect of buddha-nature. The Anūnatvāpūrṇatvanirdeśaparivarta, or “No-Increase, No-Decrease Chapter,” exists today only in Chinese translation, although it was eventually known to Tibetans through extensive quotations in the Ratnagotravibhāga (initial Tibetan translators of the Ratnagotravibhāga did not recognize the quotations and so failed to identify it as a sūtra). Jonathan Silk dates the text to at least before the early fifth century, after the Tathāgatagarbhasūtra and the Śrīmālādevīsiṃhanādanirdeśa, as does Diana Paul, while Takasaki argues for it having appeared after the Tathāgatagarbhasūtra and before the Śrīmālādevī. Like other tathāgatagarbha scriptures, this one is ambiguous as to whether tathāgatagarbha is a womb/seed or an intrinsic nature, a currently nonexistent potential or an already existent presence. Silk argues that the text understands the term in the latter sense. The title of the scripture comes from the discussion of whether the number of ordinary beings decreases when someone becomes a buddha. Such a question reveals dualistic thinking, the Buddha chides in the narrative, and is therefore flawed, because ordinary beings and buddhas are not fundamentally different in nature. As Brunnhölzl puts it, the text teaches that when the dharmakāya is obscured by adventitious stains, it is called “sentient being.” When this very same dharmakāya becomes weary of saṃsāra and practices the ten pāramitās and bodhisattva conduct, it is called “bodhisattva.” When it is free from all stains, it is called “buddha.” In other words, there is no essential difference between an ordinary being and a buddha, and to ask whether there is a change in population when a person attains enlightenment is nonsensical, not unlike asking whether there is a change in the number of water molecules when ice melts. It is simply that ordinary beings are afflicted by stains and a buddha is not, similar to a golden statue wrapped in rags compared to a statue on display in all its glory. (Note how in the above passage buddha-nature and dharmakāya are treated as synonyms.) One of the main contributions of the Anūnatvāpūrṇatvanirdeśaparivarta is the emphasis on faith as a necessary—though not sufficient—condition of enlightenment. Such a rhetorical move might suggest an admission that tathāgatagarbha is a notion that is not available for logical proof—it does, after all, raise the specter of a quasi-Hindu transcendent self, not to mention a mystical presence that is beyond the reach of language. The use of positive terms to describe tathāgatagarbha, much less its use of the word ātman, required a lot of exegesis to convince many that the doctrine was in accordance with current understandings of emptiness and did not violate the Buddha’s teaching of no-self. The Mahābherīsūtra, another tathāgatagarbha sūtra that like the Anūnatvāpūrṇatvanirdeśaparivarta is concerned with the issue of whether the sum total of sentient beings can increase or decrease, also makes extensive use of the term ātman. The sūtra was translated into Chinese in the fifth century by Guṇabhadra and was influenced by the Lotus Sūtra, which it mentions by name. Christopher Jones proposed that because the presentation of the issues in the Mahābherīsūtra is less sophisticated than in the Anūnatvāpūrṇatvanirdeśaparivarta, it should be considered to have been composed earlier. Jones also argues that with the Mahābherīsūtra and the Anūnatvāpūrṇatvanirdeśaparivarta came an important expansion of the meaning of tathāgatagarbha: earlier scriptures posited that all beings have buddha-nature in them and that they had the potential to become a buddha. Jones writes that these two sūtras were responsible for equating tathāgatagarbha with dharmakāya (as we saw above), the all-pervading true nature of all reality. Buddha-nature in this way is no longer just a potential or nature of the individual; it is the fundamental nature of reality shared by all beings. A final early tathāgatagarbha sūtra, if it can fairly be included in the category, is the Aṅgulimālīyasūtra, which like the Mahāparinirvāṇa examines earlier material through a Mahāyāna lens. Here is the story of the conversion of a bandit who has killed so many people and fashioned such an impressive necklace of their fingers that he has earned the epithet Aṅgulimāla, “Rosary of Fingers.” Like the Mahāparinirvāṇasūtra and the Mahābherīsūtra, the Aṅgulimālīyasūtra equates tathāgatagarbha with ātman, distinguishing it from non-Buddhist conceptions of the term. As Jones pointed out, referencing Kazuo Kano’s Japanese-language scholarship, the message of the scripture is not, as one might think, the universality of buddha-nature, even for those who commit heinous crimes. Aṅgulimāla is not actually converted in the Mahāyāna version of the Aṅgulimālīyasūtra; rather his killings are presented as illusory and the violence is justified as a defense of the dharma. Buddha-nature is rather incidental to this message.
Tathāgatagarbha Sūtras in China Translation of the scriptures All the above tathāgatagarbha scriptures were translated into Chinese between the fifth and sixth centuries, during a period of intense translation activity. As evidence of the availability of Sanskrit manuscripts, forty-one Sanskrit fragments of the Mahāparinirvāṇa have been found in Central Asia, a primary region through which Buddhism was brought from India to China. Dunhuang in particular, it is believed, was a hub of buddha-nature transmission in China. Of the six scriptures described above, five were done in the fifth century. Three were translated by Guṇabhadra and two by Buddhabhadra. Dharmakṣema translated one of these, the Mahāparinirvāṇa, a second time, and may have also translated the Śrīmāla. Bodhiruci (of the Wei) translated the sixth scripture in the early sixth century. There are three recensions of the Mahāparinirvāṇasūtra, although only two are technically translations. The first, the Dabannihuan jing 大般泥洹經 (T376), was translated into Chinese in the southern capital of Jiankang 建康 around 416–418 by Buddhabhadra and Faxian 法顯. This consists of only the first five chapters (said to be the original core) of the sūtra. The second is the Dabanniepan jing 大般涅槃經 (T374), done around 421–432 by Dharmakṣema in the northern kingdom of Beiliang 北涼. This was revised in the 430s as Dabanniepan jing 大般涅槃經 (T375), also known as the “Southern Version,” produced in Liu Song 劉宋 by Huiyan 慧嚴, Huiguan 慧觀, Xie Lingyun 謝靈運 (385–433), and others. This is not technically a translation, as they did not consult a Sanskrit original. According to Diana Paul the prolific translator Guṇabhadra also later corrected Dharmakṣema’s translation. The Tathāgatagarbhasūtra may have been translated into Chinese as early as the late third century by a man named Faju 法炬 who was active at least between 290 and 306. This information is based on a catalog called the Chu sanzang ji ji 出三藏記集 by Sengyou 僧祐 (445–518), but it is far from certain, and the assertion has been discounted by some scholars; it would certainly push back the date of the sūtra’s creation. The early surviving recension of the Tathāgatagarbhasūtra is the Dafangdeng rulaizang jing 大方等如來藏經 (Taishō 666), which was translated into Chinese in the early fifth century by Buddhabhadra. A second recension, Dafanghuang rulaizang jing 大方廣如來藏經 (T667), was translated by Amoghavajra around the middle of the eighth century. The Śrīmālādevīsiṃhanādanirdeśa was translated twice, first in 435 by Guṇabhadra as the Shengmen shizi hou yicheng da fangbian fangguang jing 勝鬘師子吼一乘大方便方廣經 (T353). Guṇabhadra worked with a disciple of the great Central Asian translator Kumārajīva named Baoyun 寶雲, who was also a companion to Faxian. The second is Shengmen furen hui (T310), done by Bodhiruci (of the Tang) (572–727) between 706 and 713 as part of his translation of the Ratnakūṭa scriptural collection. Paul mentions that Chinese catalogs record two translations made prior to Guṇabhadra’s, one by and Dharmakṣema and the other done in 320 by a monk named Seng Fani 僧法尼. She suspects that the records may in fact point to the presence of Sanskrit manuscripts circulating in the region and several failed attempts at translations. Paul lists six commentaries to the Śrīmālādevī, the first being done before the year 500 and the last in 722, which attest to its popularity. The Mahābherīsūtra was also translated by Guṇabhadra, circa 435–436, as Dafa gu jing 大法鼓經 (T270), as was the Aṅgulimālīyasūtra, with the title Yang jue mo luo jing 央掘魔羅經 (T120), between 435 and 453. Finally, the Anūnatvāpūrṇatvanirdeśaparivarta, which, as mentioned above, exists only in Chinese, was translated by Bodhiruci (of the Wei Dynasty) in 520 in Luoyang with the title Foshuo bu zeng bu jian jing 佛 説不増不減經 (T668).
Influence of the Tathāgatagarbha Sūtras in China Of these six scriptures, the Mahāparinirvāṇa had the widest impact in terms of buddha-nature. Not long after its appearance in China, scholiasts merged its teaching of universal enlightenment with the indigenous Chinese Buddhist doctrine of Sudden Enlightenment (dunwu 頓悟). At the center of this synthesis was the Six Dynasties monk Daosheng (道生 360/375-434), a disciple of Kumārajīva. Daosheng had earlier developed the theory of Sudden Enlightenment based initially on Abhidharma teachings he received from the hermit Saṅgadeva in Lushan at the end of the fourth century and later expanded based on Ekayāna and Madhyamaka doctrines. Sudden Enlightenment holds that enlightenment is not causal; rather than being the result of a path requiring effort over innumerable eons, one need only recognize one’s own natural state. Universal buddha-nature, defined as that natural state, therefore had obvious appeal to Daosheng. His community, however, initially had access only to Faxian’s version of the Nirvāṇasūtra, in which icchantikas are not endowed with buddha-nature. In preaching universal buddha-nature Daosheng was accused by his colleague Huikuan 慧觀 of heresy, and in 428–429 he was banished from Qingyuan Monastery 青园寺 in Jiankang 建康, the capital of the Southern Dynasty. The following year, in 430, Dharmakṣema’s translation of the Mahāparinirvāṇasūtra, which extended buddha-nature to the icchantikas, became available in Jiankang and Daosheng was exonerated. Daosheng was not the first to link Sudden Enlightenment with the buddha-nature theory of the Mahāpanirvāṇasūtra. This honor goes to Hengrui 慧叡, also known as Sengrui 僧叡, who in his polemical Mahāyāna tract Treatise Clarifying Doubts (Yuyi lun 喻疑論), linked those who criticized Sudden Enlightenment to those who rejected universal buddha-nature and, somewhat incongruously, alluded to such people as icchantikas. Nevertheless, it was Daosheng who popularized the idea. There was, of course, resistance to Daosheng and his followers’ cataphatic embrace of buddha-nature. Jizang 吉藏 (549-623), the founder of Chinese Madhyamaka, wrote in his Treatise on the Mystery of the Mahāyāna (Dacheng xuanlun 大乘玄論) that “always it is necessary to oppose any definition of buddha-nature.” The Mahāparinirvāṇasūtra’s teaching of universal buddha-nature found congress with another, related, Chinese Buddhist doctrine: Original Enlightenment. As explained by Jacqueline Stone, Original Enlightenment theory was first articulated in three apocryphal Chinese scriptures: Jingang sanmeijing 金剛三昧經 and Renwangjing 任王經 (both of which were said to have been translated by Amoghavajra but were likely composed in China in the sixth century) and, most important, the wildly popular Awakening of Faith in the Mahāyāna, or Dacheng qixinlun 大乘起信論. In these works and their commentaries Chinese scholiasts adopted buddha-nature theory as a tool to explore the ontological basis on which ordinary sentient beings become buddhas. The author of Awakening of Faith merged buddha-nature theory with the Yogācāra doctrine of ālayavijñāna, which attempts to explain the nature of ignorance. This was done in order to, as Stone puts it, “clarify the relation between the mind, understood as originally pure, and ignorance.” Awakening of Faith treats originally pure mind and ignorance as merely two different perceptions of the same reality: buddha-nature, originally pure, is the ultimate reality of things—that is, emptiness; ālayavijñāna, the storehouse of saṃsāra, is reality as perceived through the lens of delusion and corresponds to conventional reality. And because even deluded perception is grounded in the dharmakāya, ignorance can be said to share the same fundamental nature as original purity Both tathāgatagarbha and ālayavijñāna theory describe the process of buddhification in terms of adventitious stains that must be cleansed over untold eons of ardent practice. In a cultural setting such as India, where a seemingly endless string of deaths and births was taken for granted, this probably was not a problem. Chinese audiences, however, appear to have been dismayed at the length of time needed to arrive at the goal of full enlightenment. The Awakening of Faith, by redefining ignorance and wisdom, shifted the nature of the path from the one to the other. There is, ultimately, no ignorance to purify. To attain buddhahood one needs only to recognize the original purity of one’s own mind, to realize the fundamental emptiness of all phenomena, including one’s own crass thoughts and experiences. The successful practitioner actualizes an enlightenment (shizhue 始覺) that is no different from the “original enlightenment” (benzhue 不覺). Original Enlightenment was hugely influential in Chinese Buddhist communities, and the Awakening of Faith became one of two main scriptures for the Huayan school, alongside the Avataṃsakasūtra. It was central in the writings of the great patriarchs Fazang (法藏 643–712) and Zongmi (宗密 780–841) and was key to the later Tiantai theory of the buddha-nature of inanimate objects: if all phenomena are in fact the dharmakāya, and the dharmakāya is the equivalent of buddha-nature, then even roof tiles have buddha-nature and can be said to preach the dharma. Sudden Enlightenment, meanwhile, came to be adopted in various degrees by most Chinese and later Japanese Buddhist schools and was contrasted with what the Chinese classified as the Gradual Path. The two were the subject of a famous eighth-century doctrinal debate in Tibet between the Indian Madhyamika Kamalaśīla (713–763) and the Chan master Heshang Moheyan 和尚摩訶衍 (discussed below). The difference between them was also given expression in the Platform Sutra of the Sixth Patriarch, where two monks leave dueling verses on a monastery wall: the one giving voice to the Gradual Path on which one strives to purify the obscurations, the second to the Sudden Path, with its expression of original enlightenment and admonition to not solidify the fog of saṃsāra: The body is the bodhi tree; The mind is like a bright mirror’s stand. Be always diligent in rubbing it— Do not let it attract any dust.
Bodhi is fundamentally without any tree; The bright mirror is also not a stand. Fundamentally there is not a single thing— Where could any dust be attracted?
The popularity of the Awakening of Faith led to the composition in Korea of the the *Vajrasamādhisūtra, or Kǔmgang sammae kyǒng in Korean (Jingang sanmei jing / Kongō sanmaikyō / 金剛三昧經) at the end of the seventh century. The great seventh-century Korean scholiast Wǒnhyo (617-686), believing it to be the scriptural basis of the Awakening of Faith, wrote the earliest and most famous commentary, simply titled Kǔmgang sammae kyǒng non, Commentary on the *Vajrasamādhisūtra. (This was completed after his commentary on the Awakening of Faith.) The *Vajrasamādhisūtra presents a synthesis of Chan subitism and scholastic Buddhist traditions of medieval Korea. The sūtra, as described by Robert Buswell, represents a fusion of early Chan explorations of original enlightenment and buddha-nature with Korean Hwaǒm (Huayan) teachings, then the mainstream tradition there. The scripture is firmly in the Chinese Yogācāra tradition of Paramārtha, and contains an extensive discussion of "immaculate consciousness" (amalavijñāna), the ninth consciousness which unites saṃsāra and nirvāṇa in a "single taste." According to Wǒnhyo's biographer Zanning (919-1001), the sūtra came into the world when a Korean king sent an envoy to China for medicine to cure his ailing wife. The envoy was waylaid and brought to the palace of the dragon king, deep under the ocean. The dragon king explained to the envoy that the illness and mission was all a ruse to deliver the *Vajrasamādhisūtra into the envoy's hands. Handing him a disheveled pile of papers, he instructed him that a monk named Taean should redact the scripture and that Wǒnhyo should write a commentary. The text found wide appeal in East Asian communities, and was accepted until the twentieth century as a translation of an Indic original which had been done in the fourth century and then lost. Modern scholars noted similarities between passages in the *Vajrasādhisūtra and an early Chan text, Er ru si xing lun 二入四行綸, reputedly authored by Bodhidharma, and this discovery directed attention to the sūtra as an early source for Chan, the reasonable assumption being that the śāstra author drew from the sūtra. In the mid twentieth century Japanese scholar Mozuno Kōgen reversed this relationship, arguing that in fact the Bodhidharma text came earlier and the *Vajrasādhisūtra in fact had borrowed from the Bodhidharma. Mozuno thereby argued that the *Vajrasādhisūtra is an apocrypha, a theory that has been largely accepted. Although subsequent scholars reasonably proposed that Wǒnhyo himself was the author of the text, Robert Buswell has proposed that the sūtra was composed in Korea between 680 and 685 by a monk named Pǒmnang, a Korean student of Daoxin, the reputed founder of the East Mountain School of Chan. Other scholars have proposed Taean and Wǒnhyo himself, as well as Wǒnhyo's teacher Hyegong. Tathāgatagarbha Sutras in Tibet Tibetan translations of the Tathāgatagarbha Sūtras Very little research has been done on the early transmission of tathāgatagarbha theory in Tibet. We know that most of the important buddha-nature sūtras were translated as part of the first transmission of Buddhism in Tibet, as they are listed in the early ninth-century Magpie Catalog (Ldan dkar ma) of all known translated scriptures. Dorji Wangchuk points to a number of native Tibetan compositions that make reference to the doctrine, such as Yeshe De’s Differentiations of Views (Lta ba’i khyad par) and Pelyang’s Lamp of the Mind (Thabs shes sgron ma), an early proto-Dzogchen treatise. Nubchen Sanggye Yeshe quoted the Tathāgatagarbhasūtra, albeit without mentioning the term tathāgatagarbha, in his Lamp for the Eyes in Contemplation (Bsam gtan mig sgron). The terms tathāgatagarbha and *sugatagarbha also appear in tantras belonging to the Nyingma classifications of Mahāyoga, Anuyoga, and Atiyoga. Despite the early appearance Wangchuk finds little evidence that tathāgatagarbha became widespread—the term appears, but there seem to have been no attempts to explain or integrate the idea until the translation of the Ratnagotravibhāga in the late eleventh century. Regarding the translations, the Tathāgatagarbhasūtra was translated around the year 800 as ’Phags pa de bzhin gshegs pa’i snying po zhes bya ba theg pa chen po’i mdo (D258). The Tibetan Yeshe De, who was later apotheosized into one of the twenty-five disciples of Padmasambhava (his mystical ability was said to be flight) is credited with the translation, alongside Śākyaprabha. One recension of the translation alternately credits the work to Jinamitra and Dānaśīla, prolific translators who are well represented in the Tibetan canon. There is reason, however, to doubt the association—Michael Zimmermann reasonably surmised that this was an attempt on the part of the scribe to connect the translation with more established figures. The Tibetan corresponds to the Chinese T667. The sūtra inspired at least one commentary, Golden Key (Gser gyi lde mig) by the great Butön Rinchen Drub. The Mahāparinirvāṇasūtra was translated into Tibetan three times, one of which was from the Chinese. This is the ’Phags pa yongs su mya ngan las ’das pa chen po’i mdo (D119), translated by a Chinese man whose name was Tibetanized as Wangpabzhun (wang phab zhun) together with Gewai Lodro and Gyatso De, none of whose dates are known. They possibly worked from T374, as Radich states that the two correspond. A second translation, ’Phags pa yongs su mya ngan las ’das pa chen po theg pa chen po’i mdo (D120) was done in the early ninth century by Jinamitra, Jñānagarbha, and Devacandra. This translation corresponds to T376. A third, ’Phags pa yongs su mya ngan las ’das pa chen po'i mdo (D121) was done later, at the start of the second diffusion period, by Kamalagupta and Rinchen Zangpo. The Tibetan canon has only one translation of the Śrīmālādevīsiṃhanādanirdeśa, the ’Phags pa lha mo dpal phreng seng ge’i sgra shes bya ba theg pa chen po’i mdo (D92), which is part of the Ratnakūṭa collection translated by Jinamitra, Surendrabodhi, and Yeshe De. The Aṅgulimālīyasūtra was translated as ’Phags pa sor mo’i phreng ba la phan pa zhes bya ba theg pa chen po’i mdo (D213), in the late eighth or early ninth century by Dharmatāśīla, Śākyaprabha, and a monk named Tong Ācārya who was either Indian or Chinese. The Mahābherīsūtra was translated as ’Phags pa rnga bo che chen po’i le’u zhes bya ba theg pa chen po’i mdo (D222) in the ninth century by Vidyākaraprabha and Pelgyi Lhunpo. Christopher Jones writes that the Tibetan is longer than the Chinese version (T270) and is altered in ways that suggest to Jones that the translators sought to make sense of difficult passages. As mentioned above, no Tibetan translation of the Anūnatvāpūrṇatvanirdeśaparivarta exists, if it was ever made. Ngok Lotsāwa Loden Sherab did not recognize the title as a sūtra when he and Sajjana translated the Ratnagotravibhāga, and so they integrated the words of the title into the passage. Later Tibetans, however, did know that it was a sūtra title, proving at least that it was known to Tibetans, whether through Chinese translation or Sanskrit, or even a possible lost Tibetan. Influence of the Tathāgatagarbha Sūtras in Tibet To the degree that buddha-nature theory provided the foundation for the Sudden/Gradual debate, the so-called Council of Lhasa or Samye Debate played an important role in popularizing the ideas in Tibet, albeit framed in a negative perspective. According to Tibetan and Chinese sources, some time at the end of the eighth century an Indian monk and a Chinese Chan master met to debate the nature of enlightenment and the requirements of the path. The Indian was the highly educated Kamalaśīla, a monk from the great Indian monastic university Vikramalaśīla and a student of Śāntarakṣita, the monk who ordained the first Tibetan monks. The Chan master is known only as Monk Mahāyāna—Heshang Mohoyan in Tibetan, or, in Chinese, Heshang Moheyan 和尚摩訶衍. Although their encounter is commonly depicted as a meeting between the two monks in the presence of the Tibetan king Tri Songdetsen, the historical record suggests it was more of “a haphazard series of indirect confrontations,” in the words of Luis Gómez. Documentation rests almost entirely on Kamalaśīla’s Bhāvanākrama, which was the basis for Tibetan historian Buton Rinchen Drub’s account of the contest, and a Chinese work, Dunwu Dacheng Zhenglijue 頓悟大乘正理決 that is ostensibly by Moheyan’s disciple Wangxi 王錫, which consists of a series of questions and answers on points of doctrine. Various popular narratives of the encounter give different outcomes; most have Moheyan losing and returning to China in humiliation, but some accounts have him victorious. Regardless of the context for the encounter, it is clear that Indian and Chinese Buddhist ideas were circulating in Tibet, particularly after the Tibetan empire conquered Dunhuang, then a center of Chan activity. Chan subitism may have been parodied in Tibetan literature—Moheyan is said to have taught that all activity perpetuates saṃsāra, and so liberation can only be gained through utterly ceasing to act or think—but there is ample evidence that many Tibetans took Chinese notions of Sudden Enlightenment seriously. As documented by Sam van Schaik in his book Tibetan Zen, the Chan tradition did have a presence in Tibet, although earlier claims of its influence on Dzogchen have, he argues, been overstated.
The Ratnagotravibhāga and the Later Spread of Buddha-Nature Theory in India The Ratnagotravibhāga While the tathāgatagarbha sūtras were being translated and circulated in China, someone in India—we do not know who—composed a treatise that systematized the doctrine for the first time. This is the Ratnagotravibhāga Mahāyānottaratantraśāstra, which roughly translates as “The Final Teaching (uttaratantra) of the Mahāyāna, A Treatise (śāstra) Analyzing (vibhāga) the Source (gotra) of the [Three] Jewels (ratna).” In root and explanatory verses, with a prose (auto)commentary, the treatise provides a philosophical basis for supporting buddha-nature theory, quoting extensively from most of the aforementioned sūtras. The date of the Ratnagotravibhāga is not known; some scholars have speculated as early as the mid third century, but it was most likely composed in the mid to late fifth century. We know that the Ratnagotravibhāga existed in India by at least the year 498, the earliest date that Chinese records give for the arrival in China of its translator, Ratnamati. Although it appears to have been largely forgotten in India between the sixth and the tenth centuries, it was revived there in the eleventh century and promulgated in Central Asia, moving from there into Tibet in the early eleventh century. The Ratnagotravibhāga is divided into five chapters, the first four of which cover what it calls the seven vajrapadas, or “vajra points”: the Buddha, dharma, saṃgha, tathāgatagarbha, enlightenment, buddha qualities, and buddha activity. Chapter 1 deals with the first three and tathāgatagarbha, chapter 2 with enlightenment, chapter 3 with buddha qualities, and chapter 4 with buddha activity. The final chapter concerns the virtues and benefits of teaching tathāgatagarbha and the customary Mahāyāna exhortations to promulgate the text. The bulk of the first chapter, following a brief discussion of the Three Jewels, is dedicated to tathāgatagarbha. This topic is introduced with two famous verses that set forth the three reasons that justify the claim that all sentient beings possess buddha-nature. As translated by Brunnhölzl, verses 1.27 and 1.28 read as follows: Since buddha wisdom enters into the multitudes of beings, Since its stainlessness is nondual by nature, And since the buddha disposition is metaphorically referred to by [the name of] its fruition, All beings are said to possess [buddha-nature].
Since the perfect buddhakāya radiates, Since suchness is undifferentiable, And because of the disposition, All beings always possess [buddha-nature].
In other words, all sentient beings have buddha-nature because the dharmakāya is the essential nature of all beings, because sentient beings’ nature is not different from the Buddha’s, and because all sentient beings have the same disposition (gotra) as the Buddha and therefore have the potential to attain an identical enlightenment. These three reasons, abbreviated as the essence, qualities, and disposition, became a primary framework for the exposition of buddha-nature theory in Tibet. The Ratnagotravibhāga does not elaborate on these verses; its author instead states that the similes taken from the Tathāgatagarbhasūtra (a kernel of grain in its husk, a kingly being in the womb of a destitute woman, and so forth) will serve to explicate their meaning. Yet before moving to the nine analogies the text addresses ten topics (tathāgatagarbha’s nature, cause, fruition, function, endowment, manifestation, phases, all-pervasiveness, changelessness, and inseparability of qualities), the main thrust of which appears to be the classification of all beings into three categories, only one of which has the actual potential to attain enlightenment. These categories are those who desire saṃāric existence, those who long to be free of saṃsāric existence, and those who do not desire either extreme. Ordinary sentient beings are in the first category; śrāvakas, pratyekabuddhas, and non-Buddhist renunciates are in the second; and bodhisattvas are in the third. Only these bodhisattvas, the text asserts in typical Mahāyāna fashion, can attain enlightenment, although the text goes to some length to clarify the exact type of bodhisattva who will be victorious. Thus although the treatise affirms that buddha-nature is universal, it does not assert that all sentient beings have the actual potential to realize their nature and attain enlightenment. There is no promise, as in the Lotus Sūtra, that even those currently on the Hīnayāna path will eventually convert and join the bodhisattvas. For this reason interpreters who have taken the Ratnagotravibhāga to have belonged to the Yogācāra school can be forgiven. While the treatise overall does not reflect standard Yogācāra positions, certain statements on emptiness made it difficult for adherents of the Madhyamaka to incorporate the Ratnagotravibhāga into their literature, primarily because the text asserts that the teaching on buddha-nature is superior to that of emptiness. The Ratnagotravibhāga accepts the teaching of the emptiness of conventional reality, but not of buddha-nature, stating in verse 1.155 that buddha-nature is “not void of unsurpassable qualities.” Although buddha-nature is unconditioned, it possesses the perfect characteristics of a buddha and therefore cannot be said to be utterly void, as all other things are. It is likely that the author, whoever he was (he was almost certainly a he), was an adherent of neither the Yogācāra nor the Madhyamaka doctrinal schools, although clearly well acquainted with the doctrines of both. We also gather, from textual evidence, that transmission of the Ratnagotravibhāga in India between the late sixth and the tenth centuries was minimal. It is not quoted in any surviving composition, indicating that it was not widely read. Scholar Kazuo Kano offers three indications that transmission of the text survived, however sporadically: a Dunhuang manuscript quotes from it, and two texts—the Sarvatathāgatattvasaṃgraha (late seventh to eighth century) and a text called Bṛhaṭṭīkā ascribed to Daṃṣṭrasena and translated in Tibet by Surendrabodhi and Yeshe De in the ninth century—both used language similar to that found in the Ratnagotravibhāga.
Late Indian Sources on Buddha-Nature Regardless of the initial failure of the Ratnagotravibhāga to find an audience in India (or at least other authors who would quote it explicitly), due to the popularity of the tathāgatagarbha literature the theory itself spread widely in India, eventually seeping into both Yogācāra and Madhyamaka schools. Sūtras outside the main tathāgatagarbha scriptures also taught buddha-nature, increasing the doctrine’s popularity among the more established Buddhist schools. Chief among these would be the Laṅkāvatārasūtra, a central text in the promotion of “mind-only” theory and one of the main scriptures of Chan/Zen. The sūtra either was produced after the Ratnagotravibhāga or was unknown to its compilers, as there are no quotations from it in that treatise. As described by Brunnhölzl, the Laṅkāvatāra provided language that allowed the integration of buddha-nature theory into both Yogācāra and Madhyamaka communities, by equating tathāgatagarbha with both ālayavijñāna and emptiness. In the sūtra the Buddha travels to the island of Laṅka, where he teaches the bodhisattva Mahāmati that all phenomena are manifestations of mind, outlining the eight consciousnesses as taught in the Yogācāra school, and equating tathāgatagarbha with ālayavijñāna, which although by nature pure is obscured by adventitious stains. The Buddha explains that the nature of the ālayavijñāna is emptiness, and so tathāgatagarbha, ālayavijñāna, and emptiness are all different words for the same thing. The Laṇkāvatārasūtra is ambivalent around these definitions, however, most likely a result of having been patched together over many years, similar to so many other Buddhist scriptures. The sūtra also equates tathāgatagarbha with the mind's quality of innate awareness and states that buddha-nature is the true understanding of emptiness, as distinct from nihilistic interpretations: “by virtue of the purity of natural luminosity, [buddha-nature] is primordially pure, endowed with the thirty-two major marks, and hidden within the bodies of all sentient beings.” Yet the sūtra also presents buddha-nature as a provisional teaching, given by the Buddha to ease the fear of no-self among the Buddhists and to encourage non-Buddhists, who are attached to the wrong view of an existent self, to embark on the Buddhist path. Such a statement, of course, puts the sūtra in contradiction with the Ratnagotravibhāga. The debate over the definition of buddha-nature—whether it is luminosity or emptiness, and whether it is a provisional or definitive teaching—would continue through to the final days of Buddhism in India and be taken up with vigor in Tibet. Kano gives as an example of buddha-nature in Yogācāra literature the Central Indian translator and Yogācārin adherent Paramārtha (499–569), who was so steeped in its ideas that they informed his translations into Chinese of several Yogācāra classics. Madhyamaka patriarchs Bhāviveka (circa 490–570), Candrakīrti (circa 600–650), and Kamalaśīla (circa 740–795) all wrote on buddha-nature, attempting to define it in a way that makes it conform to Madhyamaka doctrine. This was done in a myriad of ways; Candrakīrti downgraded tathāgatagarbha to provisional status (along with ālayavijñāna), while Kamalaśīla used it to bolster his argument for single-vehicle doctrine, and so classified it as definitive. Buddha-nature also was taken up by Tantric communities. Kano gives several examples: the Adhyardhaśatikā Prajñāpāramitā, a Yogatantra text, states: “All sentient beings have Buddha-nature, owing to [their] identity with Samantabhadra, the Great Bodhisattva,” a passage that is quoted by Dharmamitra in his Abhisamayālaṃkāra commentary. Another, the Amoghapāsakalparāja, has: “He (who worships Avalokiteśvara, is empowered by Avalokiteśvara, and is equated with Avalokiteśvara) should be known as having Buddha-nature (or possessing a buddha inside).” The Tantric embrace of buddha-nature would be exponentially expanded in Tibet. The Ratnagotravibhāga in China The Ratnagotravibhāga was translated into Chinese by Ratnamati, who possibly arrived in China from Madhyadeśa (zhongtianzhu 中天竺) between 498 and 508 and worked on the translation in Luoyang between 511 and around 520. He may or may not have brought the manuscript with him and may have been assisted by Bodhiruci.
The Chinese title of the Ratnagotravibhāga is Jiu jing yi cheng bao xing lun 究竟一乘寶性論 (Taishō no. 1611, vol. 31), which Takasaki has reconstructed as Uttara-ekayāna-ratnagotra-śāstra and which translates to something like “Treatise on the Superior Jewel Family of the Single Vehicle.” Kano, however, reasonably suspects that the yicheng 一乘 is a mistake for dacheng 大乘, or Mahāyāna. If this is the case then the title would back-translate to a more familiar form (note that the Chinese does not contain the word tantra). The first section (lines 813a8–820c20) consists of eighteen opening verses that are not found in Sanskrit or Tibetan translation; their origin is not explained. The treatise itself is lines 820c21–848a27 complete with root verses, commentarial verses, and prose commentary.
Kano documents the influence of the Chinese translation of the Ratnagotravibhāga through three texts that exist only in Chinese. The first of these is the *Mahāyānadharmadhātunirviśeṣa said to be written by Sāramati, the same man to whom the Chinese tradition assigns authorship of the Ratnagotravibhāga, and translated by the Khotanese monk Devendraprajña in 691. Kano suggests it might have been composed around the fifth century, if in fact the attribution to Sāramati is correct, soon after the Ratnagotravibhāga. The text develops the ten buddha-nature topics as outlined in the Ratnagotravibhāga and expands the discussion of bodhicitta, here equated with dharmadhātu which is said to abide in all beings naturally. Kano describes two additional works, both of which were translated and possibly composed by Paramārtha, a prolific translator of Yogācāra works. The first is titled *Anuttarāśrayasūtra, a reworking of the Ratnagotravibhāga in the form of a Mahāyāna sūtra. The second is titled *Buddhadhātuśāstra and is ascribed to Vasubandhu. Both texts also rework the ten buddha-nature topics from the Ratnagotravibhāga, while the second asserts that buddha-nature theory is consistent with Yogācāra three-nature theory. As seen above, buddha-nature theory had already spread widely in China by the time the Ratnagotravibhāga was translated in the early sixth century. Western scholars do not appear to have explored the role of the treatise in the further development of the doctrine in China.
The Ratnagotravibhāga in Late Indian Buddhism As Kazuo Kano has described in his magisterial book on the Ratnagotravibhāga, in the early eleventh century there was a flurry of interest in India in the treatise, largely centered around the monastic university Vikramalaśīla. Although not all the commentators he named used the Ratnagotravibhāga specifically to advance buddha-nature theory, the increased interest in the treatise certainly had the effect of spreading buddha-nature across the diverse philosophical traditions of Indian Buddhism. The compositions of these authors often vary in their presentations of buddha-nature only along the subtlest points of definition, reminding us that classical Indian writers, like the Tibetan commentators who succeeded them in Mahāyāna Buddhist exegesis, can easily be compared to contemporary academic writers who make their names in part by differentiating themselves from their predecessors and peers. According to legend the Ratnagotravibhāga was rediscovered in a stūpa by a monk from Vikramalaśīla named Maitrīpa (born 1007 or 1010). This person is known only through hagiographies and legends around the recovery of the treatise; different versions give the name of the man who found the text as Maitreyanātha or Aṇarakṣita. He is credited as the author of multiple works, however. Maitrīpa left no writing on buddha-nature, but he did quote the Ratnagotravibhāga in one of his compositions, the Pañcatathāgathamudrāvivaraṇa, and was possibly the first commentator to do so. His position, Kano explains, was basically Madhyamaka, but with decided Nirākāravāda-Yogācāra leanings. Kano identified seventeen subsequent Indian and Kashmiri authors who composed commentaries or otherwise left remarks on the Ratnagotravibhāga between the eleventh and thirteenth centuries. Those based around Vikramalaśīla included Jñānaśrīmutra (circa 980–1030), a teacher of Maitripa who quoted the Ratnagotravibhāga in his Sākārakiddhiśāstra, a work on Sākāravada-Yogācāra doctrine, as well as in its summary, Sākārasaṃgraha. His use of the Ratnagotravibhāga, as detailed by Kano, was largely to defend his position that the dharmakāya and the saṃbhogakāya are identical in nature, the first being a conventional quality of the second, which is a definitive body of a buddha. He further argued that buddha-nature likewise is a quality of the ultimate and a synonym of both the dharmakāya and emptiness. Kano notes that Jñānaśrīmutra is generally described as a teacher of Maitripa, and so unless Jñānaśrīmutra received the transmission of the Ratnagotravibāga from an unnamed source, he would have to have received it from his own student. Both Maitripa and Jñānaśrīmutra equated buddha-nature with emptiness and understood it to be universal. A contemporary of theirs, Ratnakaraśānti (late tenth century to early eleventh century), who was the main Indian teacher to Drokmi Lotsāwa, took roughly a more mainline Vijñānavāda-Yogācāra view, restricting buddha-nature to bodhisattvas alone. Prajñākaramati (early eleventh century), another teacher of Drokmi, quoted the Ratnagotravibhāga in a commentary on Śāntideva's Bodhicaryāvatāra. Atiśa Dīpaṃkaraśrījñāna (982/986–1054/1065) received the Ratnagotravibhāga at Vikramalaśila and taught it in Tibet, at Tangpoche, most likely the first time the treatise was taught there. Kano points out that although he produced the first translation of the Ratnagotravibhāga into Tibetan, Atiśa’s position on buddha-nature is difficult to gauge, as the translation is lost and the sole surviving quotation is simply to establish the single-vehicle doctrine. Atiśa’s contemporary Yamāri (first half of the eleventh century) used the Ratnagotravibhāga to defend his Sākāravāda position and refute the Alīkākāravāda stance. A man named Vairocanakṣita (there were two, both living from the eleventh to twelfth centuries) composed a brief gloss of the Ratnagotravibhāga titled Mahāyānottaratantraṭippaṇī; the eleventh-century Vikramalaśīla-based Vairocanakṣita wrote a commentary on the Bodhicaryāvatāra in which he quoted the Ratnagotravibhāga, a work that Kano surmises was based on Prajñākaramati's work mentioned above. A disciple of Maitrīpa, Rāmapāla, who lived in the eleventh century, quoted the Ratnagotravibhāga in a commentary to his teacher’s Sekanirdeśa, in a discussion of the bodies of the Buddha. Another of Maitrīpa’s disciples, Sahajavajra, also quoted the Ratnagotravibhāga in a commentary of one of his teacher’s compositions, the Tattvadaśaka, in a discussion of erroneous perception. In the early twelfth century Abhayākaragupta, one of the last known paṇḍitas of Vikramalaśīla, composed over a dozen works on Tantra, Prajñāpāramitā, and Madhyamaka. He quoted the Ratnagotravibhāga repeatedly in several of these, taking a clear Madhyamaka position that buddha-nature is to be equated with emptiness and that there is but a single vehicle to liberation. Other late Indian Buddhist writers who quoted the Ratnagotravibhāga include *Śāntyākara, Daśabalaśrīmitra, Ratnarakṣita, and Vibhūticandra. Contemporary with the above authors based at Vikramalaśīla was a line of teachers in Kashmir: Sajjana, Mahājana, Amṛtākara, and Jayānanda. Sajjana, who helped Ngok Lotsāwa translate the Ratnagotravibhāga into Tibetan in the late eleventh century, received the transmission of the treatise from *Ānandakīrti, who is said to have received it from Maitrīpa. His disciples Ngok and Tsen Khawoche are credited with the two main interpretive strategies of buddha-nature and the Ratnagotravibhāga in Tibet: the “analytic” and the “meditative” (discussed below). Sajjana’s Mahāyānottaratantraśātropadeśa consists of thirty-seven verses summarizing the main points of the Ratnagotravibhāga and laying out his own position regarding the various points of doctrine. This includes the assertion that there is no contradiction between emptiness as presented in the Prajñāpāramitāsūtras and buddha-nature: everything conditioned is empty, but because buddha-nature is not conditioned, it is not empty of its own qualities. Moreover, Sajjana appears to equate buddha-nature with luminous mind, which generates itself from moment to moment. An interlinear note to a surviving Sanskrit manuscript of the text reads: In this statement, the conditioned (saṃskṛta) is comprehended (abhivyāpta) as being empty, whereas the luminous mind (prabhāsvaraṃ cittaṃ) is not conditioned. This is because [in the luminous mind] there is nothing to be done through cause and conditions coming together, based on the fact that the origination of the [luminous] mind in the succeeding moment depends on [the mind] that was generated by its (the mind's) own kind (sajāti) in the previous moment.
Sajjana’s son Mahājana was also an author of commentaries that cite the Ratnagotravibhāga, as was his (possible) contemporary, Amṛtākara, whose exact position in the transmission lineage is not known; Kano places him in Kashmir based on the fact that the only surviving copy of his Catuḥstavasamāsārtha is preserved in a manuscript written in proto-Śāradā, a script used only in the region. The Kashmiri logician Jayānanda, who lived between the late eleventh and early twelfth centuries, used the Ratnagotravibhāga to defend the single-vehicle doctrine. An ardent Prāsaṅgika-Madhyāmika, he used it as a weapon to refute Yogācāra; Jayānanda declared buddha-nature to be provisional, writing, “It was taught that just as the teaching of Buddha-nature is provisional, the teaching of mind-only is also provisional,” a position he repeats in several passages. Yet buddha-nature for Jayānanda was also another word for emptiness: “Therefore, just as the purpose of teaching Buddha-nature is nothing other than [to teach] emptiness, so too the teaching of mind-only is also [merely] a means of entering emptiness.”
The Ratnagotravibhāga in Tibet The Ratnagotravibhāga was brought to Tibet first by Atiśa in the middle of the eleventh century and then by Ngok several decades later. Atiśa translated the Ratnagotravibhāga with the help of Naktso Tsultrim Gyelwa, who had himself studied under a disciple of Maitripa’s named Vajrapāṇi. As with all Tibetan translations save that of Ngok, Atiśa and Naktso’s has been lost, although it survived at least until the fifteenth century, when Go Lotsāwa Zhonnu Pel used it in preparing for his commentary. Atiśa's translation was done at Yerpa Barang in Taktse Rinchen Gang, near Samye, at the request of Ngok Jangchub Jungne, the monastery's founder, between Atiśa’s arrival at Samye in 1047 and his death in 1054. As mentioned above, Atiśa taught the Ratnagotravibhāga at Tangpoche at the request of a man named Khuton, the head monk of the monastery. Chronologically, the next translation of the Ratnagotravibhāga is the only one that remains extant, that of Ngok Lotsāwa and Sajjana. It was produced in Kashmir, a city known in Tibetan as Peme (dpe med) and which Kano reconstructs as *Anupamapura. Kano dates the translation to between 1076 when Ngok left Tibet, and around 1092, when he returned home. Ngok and Sajjana’s translation was widely adopted, remaining the standard translation today; Kano points out that later translations appear for the most part to be simply revisions of it. In the Tibetan canon, the Kangyur (bka’ ’gyur), Ngok and Sajjana’s translation is divided into two texts, one comprised solely of verses and the other complete with verses and prose. The name of the extracted verses is Theg pa chen po rgyud bla ma’i bstan bcos, which back translates into Sanskrit as Mahāyāna-uttaratantra-śāstra (D4024). The complete text, however, is titled Theg pa chen po rgyud bla ma’i bstan bcos rnam par bshad pa (D4025), which reconstructs as *Mahāyāna-uttaratantra-śāstra-vyākhyā and translates to “A Commentary on the Treatise on the Final Teaching of the Mahāyāna.” It is important to note that the title Ratnagotravibhāgavyākhyā—or any version with vyākhyā—is not attested in any surviving Sanskrit redaction; Kano surmised that the root verses were extracted by a disciple of Ngok and given the title of the work, at which point the prose section was deemed to be a commentary and therefore given the status of “vyākhyā.” With two texts, Tibetan tradition decided that there would be two authors: Maitreya for the verses and Asaṅga for the commentary. Takasaki’s comment on how this happened was simply that “the point is undecided.” Western scholars on the Ratnagotravibhāga, coming entirely out of the Tibetan Studies field, have largely followed Tibetan tradition and divided the text in two, speaking of root verses and “a vyākhyā.” While the translation was a vital moment in the transmission of buddha-nature theory into Tibet, Ngok’s commentary, A Guide to the Mahāyānottaratantra, perhaps the first Tibetan exegesis on the topic, was of equal importance. Ngok’s interpretation, generally known as the analytic tradition, in which buddha-nature is equated with emptiness, became the dominant Tibetan position. Ngok was himself a follower of the Madhyamaka, and he regarded the Ratnagotravibhāga as belonging to that tradition; as such he considered it to be definitive (on the other hand, the four other of the so-called five books of Maitreya, clear Yogācāra treatises, he deemed to be provisional). For Ngok buddha-nature had three aspects: causal, inherent, and resultant. The causal was the aspect possessed by saṃsāric beings, while the resultant was unique to the buddhas. All sentient beings have buddha-nature in the sense that we have the potential to become perfect, not in the sense that we contain within us an already-existent perfection. Buddha-nature for Ngok is the seed that grows into a mango tree, not the golden statue covered in mud. For this reason buddha-nature does not possess the buddha qualities, and it is therefore empty. Ngok extends the metaphor: After allaying the “heat” of defilements with the “cool rainwater” of repeated study, carried by the “clouds” of good teachers; and after moistening the “seeds” of Buddha-nature (bde gshegs snying po), you should cultivate “crops” of perfect Buddha-qualities.
Because buddha-nature as taught in the Ratnagotravibhāga was the equivalent of emptiness, for Ngok the treatise could be in agreement with Madhyamaka scriptures and therefore of definitive meaning. After Ngok, Sakya Paṇḍita Kunga Gyeltsen was probably the most prominent adherent of the analytic tradition. A follower of the Prāsaṅgika Madhyamaka position, he addressed buddha-nature in his famous Clear Differentiation of the Three Vows (Sdom gsum rab dbye) and appears to have been the first Tibetan scholiast to deem the Ratnagotravibhāga and the teachings on buddha-nature to be of provisional status. A hundred years later the Sakya lama Buton Rinchen Drub followed Sakya Paṇḍita in classifying buddha-nature as provisional and in rejecting the position that it was a synonym of emptiness. His disciple Dratsepa Rinchen Namgyel took the same position. Not all Kadam and Sakya authors adopted Ngok and Sakya Paṇḍita’s positions on buddha-nature. Monlam Tsultrim, for example, an abbot of Nartang, composed several instruction manuals in which he expressly defined buddha-nature as the natural luminosity of mind. Chomden Rikpai Reldri, a student of Sakya Paṇḍita’s at Sakya Monastery, composed an Ratnagotravibhāga commentary that is firmly in the meditative tradition. Chomden had also studied with Monlam Tsultrim and, as a result, in contrast to Sakya Paṇḍita he considered the Ratnagotravibhāga and the teachings of buddha-nature to be definitive and in agreement with both Madhyamaka and Yogācāra teachings. He defined buddha-nature as “the natural luminous mind that is inseparable from dharmatā” and equivalent to the “ultimate truth, which is unconditioned and primordially existent.”
As seen above, Ngok’s teacher Sajjana appears to have advocated a different position on buddha-nature than Ngok; for Sajjana, buddha-nature was not an emptiness that was a nonaffirming negation but instead was equivalent to luminous mind, which is unconditioned and not empty of its own qualities, a view certainly supported by the Ratnagotravibhāga. Ngok’s fellow disciple Tsen Khawoche followed this line of exegesis and is credited by Tibetan historians for giving birth to the so-called meditative tradition (sgom lugs), or Tsen tradition (btsan lugs). According to the Sakya philosopher Śākya Chokden, the analytic and meditative traditions are not mutually exclusive; after all, they stemmed from the same teacher. They are instead merely different points of emphasis. After Tsen Khawoche the earliest author of a Ratnagotravibhāga commentary who can be said to be associated with the meditative tradition is Marpa Dopa Chokyi Wangchuk, who studied with Parahitabhadra in either Kashmir or Toling, and possibly with Sajjana himself. He is said to have encountered Marpa Chokyi Lodro in Nepal and to have been blessed by Nāropa in India. A disseminator of the Cakrasaṃvāratantra in Tibet and a student of Vajrapāṇi (born 1017) and other disciples of Nāropa, Marpa Dopa, as Brunnhölzl puts it, “represents a clear connection between the Mahāmudrā lineage and the lineage of the Maitreya texts.” A commentary on the Ratnagotravibhāga titled Commentary on the Meaning of the Words of the Uttaratantra (Rgyud bla ma’i tshig don rnam par ’grel pa) claims to preserve his and Parahitabhadra’s teachings and is translated in When the Clouds Part. The text presents a strong Yogācāra position, asserting that enlightenment is the realization of the natural luminous nature of mind freed from the adventitious stains. Not long after Marpa Dopa’s teachings were set down, Gampopa Sonam Rinchen, who merged Marpa Chokyi Lodro’s tantric teachings with the Kadam monastic practices and thereby initiated the Kagyu tradition, is said to have embraced the Ratnagotravibhāga, asserting that it was in accord with Mahāmudrā and definitive in its presentation of emptiness. Many Kagyu commentaries on the Ratnagotravibhāga followed, chief among them those of the Third Karmapa, Rangjung Dorje, whose lost commentary had considerable influence on later Kagyu writing. His surviving writings provide ample entry into the early development of the Kagyu synthesis of Mahāmudrā and buddha-nature teachings. Subsequent Karmapa incarnations, including the Seventh, Chodrak Gyatso (who wrote about other-emptiness) and Eighth, Mikyo Dorje, also famously wrote on buddha-nature; both accepted the perspective of the meditative tradition, albeit with a slightly different emphasis. Buddha-nature theory has remained an important aspect of Kagyu teachings. The great fifteenth-century historian and translator Go Lotsāwa Zhonnu Pel wrote an influential commentary on the Ratnagotravibhāga, which Mathes states is nearly identical to that of the fourteenth-century Drukpa Kagyu master Barawa Gyeltsen Pelzang. Karma Trinle Chokle Namgyel wrote a commentary on the Third Karmapa’s composition, but it unfortunately has been lost. Karma Trinle, a student of the Seventh Karmapa and a teacher to the Eighth, did write elsewhere about buddha-nature, however, including in a commentary on the Abhisamayālaṃkāra and the Third Karmapa’s Profound Inner Principles. Another disciple of the Seventh Karmapa was Tashi Wozer. Born in Kham and trained in central Tibet, he was a disciple also of the Third Situ, Tashi Peljor, and a primary teacher to the Eighth Karmapa. In his commentary on the Ratnagotravibhāga titled Heart of the Luminous Sun, he equates buddha-nature with the ālayavijñāna. Later prominent Kagyu authors who have touched on buddha-nature include the Fourth Drukchen, Pema Karpo; the Eighth Situ, Chökyi Jungne; and Jamgon Kongtrul.
The interpretation of buddha-nature as either emptiness or luminosity accorded to a large degree with the social setting of the teachers and their patrons. The royal patrons in western Tibet who initiated the Second Propagation were primarily interested in sūtra-based Buddhism. Buddha-nature in these circles was almost always equated with emptiness and Madhyamaka. The Kadam tradition that arose out of their work and the Geluk tradition that supplanted it continues to maintain the “analytic” interpretation. But the clan-based translators who brought in Tantric systems that gave rise to the Kagyu and, through their innovation of treasure revelation, the Nyingma traditions, were interested primarily in the meditative interpretation, in which buddha-nature was aligned with practice, and positive terms such as the natural luminosity of mind were needed to describe meditative experiences. Sakya Paṇḍita Kunga Gyeltsen, who was wary of most Tantric systems then circulating in Tibet, was dismissive of buddha-nature, and so the Sakya largely held to Ngok’s interpretation; as Tsongkhapa was a disciple of Sakya Paṇḍita, this served to strengthen the Geluk position that buddha-nature was a provisional teaching. The shifting political winds of Tibet have helped to determine the ebb and flow of buddha-nature popularity. After the end of Sakya rule of Tibet in the fourteenth century there was a flourishing of Kagyu and Jonang commentaries on the Ratnagotravibāga, as Tsang in particular became a region of great doctrinal exchange and innovation. This ended with the start of Geluk hegemony in the sixteenth century, which brought about the repression of the Jonang and a clamping-down on Kagyu and Nyingma activity, at least in central Tibet. Buddha-nature as luminous mind has continued to thrive in areas not tightly controlled by the Geluk, such as Kham.
A third important early Ratnagotravibhāga transmission vector was through Parahitabhadra, who was a student of Mahāpaṇḍita Somaśrī and Ratnavajra, and a collaborator with Sajjana on the translation of the Mahāyānasūtrālaṃkāra. Despite his involvement in Yogācāra literature, he appears to have primarily been a Madhyāmika, as were most paṇḍits in Kashmir of the period. He also traveled to Tolung, and it was either there or in Kashmir that he gave teachings to Ngok, Patsab Nyima Drak, and Marpa Dopa Chökyi Wangchuk, who was perhaps one of the earliest to combine Mahāmudrā teachings with those of buddha-nature.
By the fourteenth century, commentary on the Ratnagotravibhāga had become sophisticated enough to give rise to doctrinal innovations particular to Tibet. The great Dolpopa Sherab Gyeltsen, a fourteenth-century monk from the Tibetan-Nepali border region who became abbot of the monastery Jonang, is credited in Tibet with the theory of “other-emptiness,” or zhentong (gzhan stong), a fairly radical interpretation of emptiness based largely on buddha-nature theory. Although primarily trained in the Sakya tradition, Dolpopa took the Kālacakra as the highest Tantric system; he is said to have awoken to the other-emptiness view of ultimate reality while practicing Kālacakra in retreat, and he framed his buddha-nature teachings around that tantric system. Dolpopa also received teachings on the Ratnagotravibhāga from the Third Karmapa. He was openly critical of Sakya Paṇḍita and of his own contemporary Buton, whose students took particular care to refute Dolpopa in return. Dolpopa’s student Sabzang Mati Paṇchen wrote a commentary on the Ratnagotravibhāga that largely echoed that of his teacher. The primary inheritor of Dolpopa’s mantle was Taranātha, the twenty-sixth abbot of Jonang Monastery. Late-period Kadam and early Geluk scholiasts largely followed Sakya Paṇḍita in classifying buddha-nature as a provisional teaching and vehemently rejected Dolpopa’s view of buddha-nature and ultimate truth. Rendawa Zhonnu Lodro returned to Ngok’s position of equating buddha-nature with emptiness, although deeming the teaching to be provisional. One of the major commentaries on the Ratnagotravibhāga in terms of impact on buddha-nature theory in Tibet was that of Gyeltsabje Darma Rinchen, one of several close disciples of Tsongkhapa Lobzang Drakpa, who together initiated the Geluk tradition. Because Tsongkhapa himself left no commentary on the Ratnagotravibhāga, many scholars have looked to Gyeltsabje to understand his master’s positions on the topic. Gyeltsabje agreed with Ngok that buddha-nature was emptiness but only, like Rongtön (and Tsongkhapa?), in the causal sense, as did Paṇchen Sönam Drakpa; the view has long since been Geluk orthodoxy.
A contemporary of Gyeltsabje’s, Rongton Sheja Kunrik, accepted Ngok’s view that buddha-nature and emptiness were the same, but only insofar as emptiness was the potential for buddhahood. A student of the Sakya master Yakton Sanggye Pel and also the Fourth Karmapa, Rolpai Dorje, Rongton was the founder of Nalendra Monastery. He is famous in part for composing seven commentaries on the Abhisamayālaṃkāra, all with a clear “self-emptiness” interpretation of the ultimate. In line with his Sakya heritage he rejected Dolpopa’s other-emptiness and the natural luminosity of mind. Nor did he accept that buddha-nature was the dharmakāya in the sense of pervading all things; rather, only a fully awakened buddha can be said to be imbued with the dharmakāya. Rongton, as a fifteenth-century Sakya teacher, can be said to have occupied a sort of middle ground between the extremes of Geluk negation and Jonang (and, to some degree, Kagyu) affirmation. That space was expanded by two of his most famous students, Gorampa Söonam Sengge and Śākya Chokden. While their presentations may have differed, they each attempted to categorize the schools of thought and define their terms in a way that allowed for the full span of Sajjana’s teachings, everything that is otherwise divided between the analytic and the meditative.
Through it all, Tibetan scholiasts struggled to stake out a position that conformed to either Madhyamaka, Yogācāra, or a synthesis of the two. Ngok’s tradition, however, was adopted by the disciples of Tsongkhapa and has become dominant, propagated by the Geluk and Sakya traditions. The other-emptiness-inflected meditative tradition, which conforms with Tantric systems of Mahāmudrā and Dzogchen, is stronger in the Jonang, Kagyu, and Nyingma traditions. Although there are significant differences among the philosophical positions taken by Tibetan scholiasts who have written on buddha-nature, the categories of meditative and analytic can be used, if cautiously, to describe Tibetan commentators up through the present day.
- Buddha-nature is actually an English translation of a Chinese term, foxing 佛性. This term appears to have been invented in China to translate buddhadhātu, possibly also buddatā, tathatā, prakṛtivyadadāna, and other terms. See King, Buddha Nature, 173–74, n5. The most common Sanskrit term, tathāgatagarbha, means something like “womb/essence/seed (garbha) of the one who has gone/come (gata / āgata) to thusness (tathā; i.e., enlightenment).” The Chinese translation of tathāgatagarbha is rulaixing 如來性. The Tibetan equivalents of buddha-nature include rang bzhin gnas rigs and sangs rgyas kyi snying po. Tathāgatagarbha is translated as de bzhin gshegs pa’i snying po: “the essence of those who have gone/come to thusness.”
- Morris, The Aṅguttara-Nikāya, i.10, 11–16, as quoted in Silk, Buddhist Cosmic Unity, 39. For more early scriptural passages on the mind’s natural luminosity, see Skorupski, “Consciousness and Luminosity.”
- Jonathan Silk, for example, (Buddhist Cosmic Unity, 39) points out that the compilers of the Anūnatvāpūrṇatvanirdeśaparivarta appear to have been aware of the Aṅguttaranikāya passage, as they integrated it into that Mahāyāna sūtra nearly verbatim.
- For scholarship on luminosity in early Buddhism, see Shih, “The Concept of ‘Innate Purity of Mind’ in the Agamas and Nikayas”; and Williams, The Reflexive Nature of Awareness.
- See, for example, Collins, “Momentariness and the Bhavaṅga Mind,” in Selfless Persons; Harris, “The Problem of Idealism” in The Continuity between Madhyamaka and Yogācāra in Indian Mahāyāna Buddhism; and Harvey, “The Brightly Shining Bhavaṅga Mind,” in The Selfless Mind: Personality, Consciousness and Nirvāṇa in Early Buddhism.
