Chapter 4. The later influence of tathāgatagarbha in China and Tibet¶

The previous three chapters discussed tathāgatagarbha within Indian texts and philosophical frameworks. The present chapter turns to China and Tibet to examine its fortunes after leaving India. In both places, tathāgatagarbha exerted an influence far greater than it had in India, but in sharply different ways and with different emphases. China absorbed it into the schools' doctrines, making it a shared basis. Tibet made it a central point of philosophical contention, determining its place through precise debates over the classification of teachings.

1. Two routes of transmission¶

Tathāgatagarbha thought entered China early and had a wide influence. The Nirvāṇa Sūtra (translated by Dharmakṣema, approximately 412 to 421), the Śrīmālā Sūtra (translated by Guṇabhadra in 435), and the Laṅkāvatāra Sūtra (translated in the twentieth year of the Song dynasty's Yuanjia era, 443) appeared in succession. The Awakening of Mahāyāna Faith (composed around the sixth century; Chapter 3 of this essay) followed soon afterward. The works had already deeply influenced Chinese Buddhist thought before the Tang dynasty. Direct conflict between the Faxiang school (consciousness-only) and tathāgatagarbha briefly intensified in the early Tang. Thereafter, Faxiang declined, and the tathāgatagarbha framework became the shared basis of Chinese Buddhism with almost no competition.

Tathāgatagarbha entered Tibet somewhat later, chiefly through the arrival of Śāntarakṣita and Kamalaśīla in the eighth century. The Ratnagotravibhāga was transmitted as a core textbook. How to interpret it, especially whether it was "definitive" (nītārtha, expressing final truth) or "provisional" (neyārtha, an expedient accommodation), became the focus of centuries of philosophical debate among Tibetan Buddhist schools.

2. China: the unfolding of Buddha-nature theory¶

Daosheng and the Nirvāṇa Sūtra¶

The earliest figure in China to engage directly with the problem of Buddha-nature was Daosheng (Zhu Daosheng, 360 to 434). Before the complete Nirvāṇa Sūtra reached China, Daosheng independently proposed, on the basis of the partial translation then available, that "even icchantikas can become Buddhas." Even beings wholly without faith in the Buddha's teaching possess Buddha-nature and can ultimately attain Buddhahood. This claim was highly controversial at the time, and Daosheng was even ostracized for it. When the complete Nirvāṇa Sūtra subsequently arrived, it confirmed his claim, greatly enhancing his reputation.¹

The episode has a significant historical implication. Daosheng's judgment arose chiefly from working out the internal logic of tathāgatagarbha thought, rather than from an already available textual warrant. It shows how rapidly the universal claim that all sentient beings possess Buddha-nature was taken to its full consequences once it reached China. Even reservations in the original texts about icchantikas could hardly restrain such deductions.

The Three Treatise school and Jizang¶

Jizang (549 to 623), the great synthesizer of the Three Treatise school, critically examined the Buddha-nature theories current in his day on the basis of the three Madhyamaka treatises: the Middle Treatise, the Hundred Treatise, and the Twelve Gate Treatise. He explicitly criticized treating Buddha-nature as something that truly exists. Buddha-nature, he held, is not an identifiable existent. It describes sentient beings' inherent possibility of becoming Buddhas; its true meaning is the "middle way," not some entity.²

Jizang's treatment came closest to a Madhyamaka critique of tathāgatagarbha in China. He did not deny that sentient beings possess Buddha-nature, but refused to reify it. His approach moved in the same direction as the Prāsaṅgika strategy of incorporating tathāgatagarbha into emptiness, although he lacked Candrakīrti's rigorous tools of reductio argument.

The influence of the Three Treatise school declined after the Tang dynasty. Jizang's critical position that "Buddha-nature is not a real entity" did not form a continuing tradition in later Chinese Buddhism. More affirmative accounts of tathāgatagarbha came to dominate.

Tiantai: the doctrine of inherent inclusion¶

Tiantai, represented by Zhiyi (538 to 597), developed the doctrine of inherent inclusion, the most philosophically original form of tathāgatagarbha thought in China.³

The basic Tiantai claim is that the nature of sentient beings (tathāgatagarbha or Buddha-nature) is not merely "pure." It inherently includes every possibility of good and evil. A Buddha's tathāgatagarbha contains evil as well as good, and beings' Buddha-nature contains afflictions as well as purity. The claim revises the mainstream tathāgatagarbha position of original purity accidentally defiled. Inherent inclusion holds that defilement, too, is an aspect already included in one's nature and cannot be counted as "accidental, adventitious dust."

The revision seeks to avoid the dualistic tension between original purity and actual defilement. If one's nature is pure, its relationship with actual defilement always needs explaining, as Chapter 2 of this essay discussed. Inherent inclusion states directly that one's nature has always included both good and evil. Since defilement does not come from outside, its source no longer needs to be explained.

Zhiyi's "three thousand realms in a single thought" develops the philosophy of inherent inclusion. The calculation in fascicle 5 of the Mohe zhiguan (Great Calming and Contemplation) has four steps. "One mind includes ten dharma realms; each dharma realm in turn includes ten dharma realms, making one hundred dharma realms. Each realm includes thirty kinds of worlds; one hundred dharma realms therefore include three thousand kinds of worlds."⁴ One mind's inclusion of ten dharma realms is the first step. Each realm in turn includes all ten, so their mutual inclusion yields one hundred realms in the second step. The third gives each realm thirty kinds of worlds, combining the ten suchnesses with the three kinds of worlds. The fourth gives each of the one hundred realms these thirty kinds of worlds, yielding three thousand.

The step of mutual inclusion determines whether the doctrine of inherent inclusion can stand. The hell realm includes the other nine realms, and the Buddha realm does likewise. The realms of sentient beings and of Buddhas include one another, so good and evil coexist within one mind. Their mutual inclusion supports the doctrine's next claim that a Buddha inherently includes evil. Without mutual inclusion, multiplying the ten dharma realms, the ten suchnesses, and the three kinds of worlds yields only three hundred. The figure of three thousand cannot be reached.

"Three thousand realms in a single thought" does not mean that one can think of three thousand things in one thought. It means that the very existence of this thought is internally interwoven with three thousand states of existence. Every possibility in the universe is inherently included in each present thought.³

The doctrine of inherent inclusion produced a famous controversy within Tiantai over whether a Buddha possesses "inherent evil." Zhiyi answered yes. Buddha-nature includes both good and evil, but a Buddha does not "cultivate evil" or actualize its possibility. Later Tiantai thinkers of the Shanjia faction, such as Zhili, developed the claim, while the Shanwai faction questioned it. It became the sharpest point of internal debate over inherent inclusion.

Tiantai's affirmative answer to the question "If one is originally a Buddha, why must one still practice?" lies in its classification of stages, called the six identities. Fascicle 1 of the Mohe zhiguan explains that they were established to block misreadings in two directions at once: "When wisdom and faith are complete, one hears that a single thought is itself [Buddha]. Because of faith one does not disparage it; because of wisdom one does not fear it. Beginning and end are both so. Without faith, one places the sacred realm on high, beyond the reach of one's own wisdom; without wisdom, one becomes arrogantly presumptuous and claims to equal a Buddha. Beginning and end are both mistaken. For this reason one must know the six identities."⁵ The six stages are, in order, identity in principle, identity in name, identity in contemplation and practice, identity in resemblance, identity in partial realization of truth, and ultimate identity. The first, identity in principle, precedes hearing the teaching: the text says, "Identity in principle means that a single thought of mind is identical with the principle of tathāgatagarbha." The second, identity in name, begins with hearing: "Though one is identical in principle, one does not know it in daily life."⁵

The structure of the six identities deserves careful attention. "Identity" preserves horizontal sameness. Sentient beings and Buddhas are not two in principle, which affirms inherent possession. "Six" preserves vertical sequence, with six levels extending from not yet having heard the teaching to the ultimate stage. The sequence gives the necessity of practice its place. A teaching of a single step provides its own six levels of attainment. Inherent possession and realization through practice can thus coexist in the same account, without abolishing one to preserve the other. The text's two stated safeguards likewise form a pair. Without faith, one places the sacred realm so high that Buddhahood has nothing to do with oneself, effectively giving up on oneself. Without wisdom, one claims equality with a Buddha and mistakes identity in principle for ultimate identity, making a presumptuous claim to attainment. Zhiyi's concluding analogy is a poor person's treasure. A treasure already lies at home (identity in principle), and someone points it out (identity in name). The person clears weeds and digs (identities in contemplation and practice, and in resemblance), then gradually approaches it (identity in partial realization of truth). Finally the treasure is opened and used in full (ultimate identity).⁵ Inherent possession and digging it out are the two ends of the same matter in this analogy.

The scheme belongs specifically to the Perfect Teaching. It classifies stages within Tiantai's own system of doctrinal classification and should not be transferred directly into the vocabulary of Chan or Huayan. Chan is treated separately below in this section. The scheme's structure nevertheless deserves the attention of the tathāgatagarbha tradition as a whole. In China, the compatibility of inherent possession with practice and realization received a systematic, direct answer. The method distinguishes "identity" and "sequence" as two accounts, each stated independently without obstructing the other.

Huayan: dependent origination of the dharmadhātu¶

Huayan, represented by Fazang (643 to 712), developed from tathāgatagarbha thought a cosmological framework of "dependent origination of the dharmadhātu."⁶

In Huayan's dependent origination from tathāgatagarbha, all dharmas share tathāgatagarbha (dharma-nature or suchness) as the foundation of their existence. Each dharma appears as the dharmadhātu (tathāgatagarbha) manifests according to conditions. It depends on and mutually includes every other dharma. Huayan expresses the relationship in its famous image of Indra's net. Each jewel, or dharma, reflects all the other jewels, each of which in turn reflects all the jewels. Their mutual reflections are endless and their overlap infinite. No dharma exists independently or in isolation.

Fazang analyzes the relationship through the "perfect interfusion of the six characteristics." Fascicle 4 of the Huayan yisheng jiaoyi fenqi zhang (Treatise on the Divisions of the Doctrine of the One Vehicle of Huayan) defines each characteristic using a house and its contributing conditions, such as rafters and tiles. Totality is "one house with many properties," and particularity is "the many properties are not one." Both concern the relationship of whole to parts between the house and its rafters, tiles, doors, and windows. Identity means "the many aspects do not conflict and together form one totality." The conditions cooperate to make a house, jointly forming it without opposing one another. The point is cooperation, not their being identical to one another. Difference means "the many aspects are each different in relation to one another." Rafters and tiles retain their respective forms and kinds, differing from one another. Fazang explicitly answers the question of how difference differs from particularity. Particularity distinguishes the conditions from the whole, as a rafter differs from the house. Difference distinguishes the conditions from one another, as a rafter differs from a tile. Their points of comparison differ.

Formation means "arising and being formed through these conditions," while nonintegration (坏相) means "the various aspects each abide in their own dharma without moving." The pair does not mean establishment and differentiation. Fazang explains, "If a rafter became a house, it would lose its original dharma as a rafter, and the house could not be formed." If the rafter turned into the house, its character as a rafter would be lost. With no rafter available, the house could not be formed at all. The condition of composition is precisely that each part retains its form and its own dharma. Formation and nonintegration therefore hold simultaneously and condition each other.⁷

The perfect interfusion of the six characteristics exemplifies the path by which tathāgatagarbha thought changes from a theory of practice into cosmology. Tathāgatagarbha no longer serves merely as the basis of practice; it also becomes the fundamental principle explaining why all dharmas in the universe include one another in this way.

Chapter 7.1 identified one source of the philosophical problem in Huayan's dependent origination from tathāgatagarbha. The Śrīmālā Sūtra equates tathāgatagarbha with the unconditioned dharma-realm. A further point follows from that equation. Tathāgatagarbha expands from the seed of Buddha-nature within sentient beings into the ontological foundation of all dharmas in the universe. The philosophical expansion is substantial, far beyond the original intent of Indian tathāgatagarbha texts.

The debate over Buddha-nature in insentient things took shape in the eighth century, between Tiantai and Huayan. Tiantai's Zhanran (711 to 782) wrote the Jin'gang pi (Diamond Scalpel) to argue that insentient things also possess Buddha-nature. His basis was Huayan's own principle of following conditions while remaining unchanged. He explicitly stated that "the teaching of following conditions while remaining unchanged comes from the great teaching." All dharmas are suchness because of its unchanging character, and suchness is all dharmas because it follows conditions. Zhanran judged it self-contradictory to accept this principle yet confine Buddha-nature to sentient beings.⁸

In the section on general and restricted scope in his Tanxuan ji (Record of Exploring the Mysteries), Fazang had already written that it "pervades even the insentient," extending the scope of nature-origination to insentient things.⁹ It was Chengguan (738 to 839) who narrowed this scope. His Huayan jing suishu yanyi chao (Subcommentary Expanding the Meaning of the Commentary on the Avataṃsaka Sūtra) devotes a passage specifically to excluding the claim that insentient things possess Buddha-nature. Its affirmed position is that "nature and characteristics are neither identical nor different; the sentient and the insentient are likewise neither identical nor different." It also says that although Huayan speaks of insentient things becoming Buddhas, "this does not mean that insentient things also possess an awakened nature and become Buddhas in the same way as sentient beings."¹⁰ Zhanran opposed Chengguan's narrowing of the scope; Fazang's scope accords with Zhanran's position.

Chan: seeing one's nature and becoming a Buddha¶

Chan, especially the Southern school after the Sixth Patriarch Huineng, became the most influential home of tathāgatagarbha thought in China. Chan itself, however, rarely uses tathāgatagarbha terminology, preferring more everyday expressions such as "original face," "one's own mind," and "Buddha-nature."

"See one's nature and become a Buddha" is the Southern school's central formulation. "Nature" refers to Buddha-nature or tathāgatagarbha, which is already there, already Buddha, merely obscured. "Seeing" is a sudden intuition, not gradual analysis. The significance of practice lies neither in accumulating merit nor in cultivating a lineage. It lies in directly recognizing that one has always been a Buddha.¹¹ In this book's judgment, the step takes the Awakening of Faith's "original awakening" to its furthest point. In that treatise, original awakening is the starting point of practice. Here it is also the endpoint, because original awakening is Buddha, and recognizing it through seeing one's nature is becoming a Buddha. The extension is this book's own comparative reading of the two texts. For original awakening in the Awakening of Faith, see Chapter 7.3.

An internal tension remains. If one is already a Buddha, why is practice needed? Chan answers that practice merely "removes delusion," eliminating false discrimination, rather than "acquiring awakening" or obtaining something. The answer places the problem within an account of practice and avoids addressing metaphysics directly. How a being that is already a Buddha remains in saṃsāra receives no direct answer in Chan's own vocabulary. Tiantai gives a direct answer to the same question through the six identities discussed above in this section. "Identity" acknowledges inherent possession, while "six" orders the stages. Inherent possession and realization through practice occupy their respective levels within the classification of attainment. The comparison identifies Chan's choice. It gives up the distinction between inherent possession and attained realization within a classification of stages, in exchange for directness in its account of practice. Chan's strength lies in practical efficacy, not metaphysical consistency.

3. The overall picture in China¶

The Chinese source diagram connects the Nirvāṇa Sūtra to Daosheng and connects the Laṅkāvatāra Sūtra to Xuanzang’s school. The discussion distinguishes that school’s conflict with universal Buddha-nature and subsequent decline from the doctrinal influence of tathāgatagarbha on Huayan, Tiantai, and Chan. The lines in this overview do not establish transmission lineages.

Tathāgatagarbha thought spread through Chinese Buddhism as a presupposition shared by the schools, without philosophical criticism of Candrakīrti's kind to define its boundaries. Original purity, the all-pervasiveness of Buddha-nature, and practice as the disclosure of what is inherently present became the accepted background of Chinese Buddhism. Yet the schools differed greatly in their philosophical treatments. Tiantai emphasized inherent inclusion, which includes both good and evil. Huayan emphasized perfect interfusion, the mutual inclusion of all dharmas, while Chan emphasized sudden awakening, being Buddha here and now. The different emphases reflect different understandings of how tathāgatagarbha operates in the present.

4. Tibet: increasing precision in debates over the classification of teachings¶

The three turnings of the Dharma wheel and the classification of definitive teachings¶

Tibetan Buddhism inherited the Indian framework of the "three turnings of the Dharma wheel" (trīṇi dharmacakrāṇi) for classifying teachings:

The central question concerns which turning is "definitive," a final truth admitting no further interpretation, and which is "provisional," an expedient teaching requiring further interpretation.

The Gelug classification, represented by Tsongkhapa, holds that the second turning, Prajñāpāramitā emptiness, is definitive. The third, consciousness-only and tathāgatagarbha, offers expedient teachings for certain capacities. Positive formulations of tathāgatagarbha, such as its inherent possession of the qualities of Buddha-nature, must be reduced to "tathāgatagarbha is emptiness" to yield their true meaning. The affirmative language of the third turning is an expedient devised for beings unable to accept emptiness directly.¹²

The Jonang classification, represented by Dolpopa, holds that the third turning is definitive. The second is a teaching devised to dismantle attachment to real existence, not the final position. The affirmative content of tathāgatagarbha, the real and permanent existence of the qualities of Buddha-nature, is the final truth. The emptiness of the second turning is "self-emptiness" (rang stong), the emptiness of dharmas in themselves, taught to break attachment. Ultimate tathāgatagarbha is "other-empty" (gzhan stong). It is empty of extraneous, adventitious defilements but not empty with respect to tathāgatagarbha itself.¹³

The dispute over the classification of teachings goes beyond textual interpretation to a fundamental disagreement over the final direction of Buddhist thought. Practice might lead toward a thoroughgoing emptiness in which "there is nothing at all," or toward the disclosure of "a pure essence that was already there."

The Gelug Prāsaṅgika Madhyamaka reading¶

In works such as the Great Treatise on the Stages of the Path to Awakening and the Essence of Eloquence Distinguishing the Definitive and Provisional, Tsongkhapa (1357 to 1419) set out the Gelug school's systematic position on tathāgatagarbha.

His central judgment is that claims concerning tathāgatagarbha's "inherent possession," "pure real existence," or "permanence" fall into eternalism if understood as asserting some positive reality beyond conventional designation. Correct understanding requires reducing tathāgatagarbha to the claim that the nature of mind is emptiness. Sentient beings' inherent possession of tathāgatagarbha means that their minds have always been empty; practice consists in thoroughly seeing this emptiness. Tathāgatagarbha language is merely a poetic expression of emptiness and makes no independent ontological claim.¹²

Tsongkhapa's position parallels the Prāsaṅgika treatment of consciousness-only. Candrakīrti criticizes the "dependently arisen consciousness" of consciousness-only as a residue of realism. Similarly, Tsongkhapa criticizes Jonang other-emptiness as a variant of the doctrine of a truly permanent reality. In the Gelug classification, Prāsaṅgika Madhyamaka is the highest view. Tathāgatagarbha discourse becomes definitive only when reduced to Prāsaṅgika emptiness.

Jonang and other-empty Madhyamaka¶

Dolpopa Sherab Gyaltsen (1292 to 1361) was the most systematic exponent of Tibetan "other-emptiness." His position was introduced in Chapter 2 of this essay; here we add its historical and social background.

The Jonang school's view of other-emptiness was an influential position in Tibetan Buddhism during the fourteenth to fifteenth centuries, with the Ratnagotravibhāga as its central authority. Dolpopa insisted that tathāgatagarbha's qualities of Buddha-nature truly exist. Their existence is ultimate, real in the highest sense. "Other-emptiness" here means emptiness of other dharmas, namely afflictions and conditioned dharmas. It does not mean self-emptiness, or emptiness of itself.

In the seventeenth century, the Fifth Dalai Lama supported Gelug power, and Jonang suffered varying degrees of suppression in Tibet proper. Monasteries were converted to other affiliations, and the principal transmission moved to what is now the Aba region of Sichuan. Twentieth-century research brought renewed attention to the Jonang tradition, and translations and studies of Dolpopa's works began to appear.¹⁴

Philosophically, other-emptiness comes closest to the wording of tathāgatagarbha thought, but for that very reason faces the sharpest criticism. How can "truly existent, permanent qualities of Buddha-nature" be distinguished from the non-Buddhist ātman? Jonang responds that the causal relationships differ, that this is not a self within saṃsāra, and so forth. The replies resemble those of the Ratnagotravibhāga described in Chapter 2 of this essay, but critics hold that they do not fundamentally resolve the problem.

Nyingma: Dzogchen and tathāgatagarbha¶

The Nyingma school's Dzogchen, or Great Perfection (Tibetan rdzogs pa chen po), has a deep internal connection with tathāgatagarbha thought, but expresses it more radically.

Dzogchen holds that the mind of sentient beings, also called the "ground" (gzhi), has always been in a state of completely perfect awareness. Practice does not acquire an awakened nature where none existed. It recognizes this already perfect awareness. The Tibetan term rigpa, cognition or awareness, is its central concept. Dzogchen takes the logic of original awakening to its furthest point. Beyond saying that sentient beings inherently possess Buddha-nature, it says that the essential state of their minds has always been Buddha. The ground is the fruit.¹⁵

The main difference between Dzogchen and tathāgatagarbha is that Dzogchen bypasses argument for tathāgatagarbha's positive claims and appeals directly to experience in practice. "Recognizing rigpa" is a direct event in practice, not primarily a philosophical position. Dzogchen can thus accept tathāgatagarbha's affirmative claims while avoiding a direct confrontation with philosophical criticism. To the criticism that it asserts the real existence of Buddha-nature's qualities, it replies that it is describing experience in practice rather than making a metaphysical proposition.

Nyingma thinkers have specifically discussed the relationship between Dzogchen and Prāsaṅgika Madhyamaka. Some, such as Longchenpa (1308 to 1364), seek to show that Dzogchen does not conflict with Prāsaṅgika; others hold that Dzogchen transcends its level. This discussion continues in modern times.

Kagyu: Mahāmudrā¶

The Kagyu tradition of Mahāmudrā (Chag chen) likewise centers on the mind's luminous and pure nature, its intrinsic purity or clear-light mind. Its structure closely overlaps with tathāgatagarbha's "original purity." Mahāmudrā practice centers on directly recognizing the nature of mind in meditation. The nature is described as "empty and luminous," combining emptiness with cognitive luminosity.

Kagyu thinkers approach Mahāmudrā's relationship to philosophical traditions, especially Madhyamaka and consciousness-only, in different ways. The tantric methods transmitted in the lineage of Marpa and Milarepa proceed through stages involving channels, winds, and drops. Gampopa's sūtra Mahāmudrā instead emphasizes the teacher's direct pointing out. It relies on direct perception rather than inference and recognizes the coemergent mind as the dharma body. Gampopa himself classifies gradual teachings as provisional and teachings of sudden entry as definitive.¹⁶

5. The overall Tibetan classification of teachings¶

Top: the Ratnagotravibhāga enters Tibet. Gelug (Tsongkhapa): other-emptiness is eternalism; tathāgatagarbha is an expression of emptiness; the second turning is definitive. Jonang (Dolpopa): other-emptiness—the qualities of buddha-nature truly exist; the third turning is definitive. Nyingma: Dzogchen; recognition of rigpa transcends philosophical disputes. Kagyu: Mahāmudrā; the empty, luminous nature of mind. The upper dashed connector is labelled “criticizes other-emptiness as eternalism”; the lower is labelled “criticizes self-emptiness: tathāgatagarbha is redundant.” The chapter attributes the criticism of other-emptiness to Tsongkhapa. The question whether self-emptiness makes tathāgatagarbha redundant is posed in Chapter 7.2 as the author’s philosophical dilemma; its placement by a directional arrow does not establish a historical attribution.

Tibetan debates over tathāgatagarbha are more precise than their Chinese counterparts and more explicitly situated within a framework of doctrinal classification. Whose teaching is definitive and whose is provisional directly determines the ultimate goal of practice. This precision is both a strength of Tibet's scholarly tradition, which systematically inherited Indian philosophy, and a source of deeper philosophical divisions. The schools formed clear barriers in this debate and have reached no consensus to this day.

6. Comparing China and Tibet¶

The fortunes of tathāgatagarbha in China and Tibet offer an interesting contrast:

China. Tathāgatagarbha became a shared basis. Each school accepted it in its own vocabulary, but direct philosophical debate was less common. Consensus outweighed dispute, at the cost of blurred philosophical boundaries. The claim that all sentient beings possess Buddha-nature became an unquestionable premise. The schools differed chiefly over how to realize it through inherent inclusion, dependent origination of the dharmadhātu, or sudden awakening and seeing one's nature. What Buddha-nature ultimately is was less at issue.

Tibet. Tathāgatagarbha became a central point of contention in the classification of teachings, with each school's interpretation directly determining its philosophical position. Debate was precise, but at the cost of sharply divided positions that were difficult to integrate.

The contrast is partly a product of historical conditions. China lacked a critical tradition of Prāsaṅgika Madhyamaka like Candrakīrti's to put pressure on tathāgatagarbha. Tibet imported Candrakīrti's tradition alongside tathāgatagarbha, especially through Tsongkhapa's systematization. It therefore had to confront the tension between tathāgatagarbha and Prāsaṅgika directly.

7. The historical influence and philosophical legacy of tathāgatagarbha¶

Let us return to the starting point of this essay. Tathāgatagarbha thought offers a substantive response to a real philosophical problem concerning what guarantee at the level of existence makes liberation possible. Madhyamaka's emptiness is negative, while consciousness-only's transformation of the basis is transformative. Both answer the question of why Buddhahood is possible only indirectly. Tathāgatagarbha answers that the Buddha was already there, so practice reveals rather than creates. The answer has a powerful intuitive appeal and is also full of philosophical tension.

As Chapters 7.1 to 7.3 have made clear, the answer brings difficulties that still lack complete solutions, including how purity and defilement coexist and how tathāgatagarbha differs from ātman. Chapter 7.4 adds an account of how this answer was received historically. China accepted it as a postulate, and Tibet made it a subject of debate. Chan made it the basis of sudden awakening, while Huayan made it a cosmological principle. In Dzogchen, it became the object of recognition in practice. In Gelug, it became an expedient teaching that is definitive only when reduced to emptiness. Each mode of reception reflects a different answer to what practice ultimately leads toward.

We can now look back over a longer span.

In Essay 1's account of early Buddhism, the core of the theory of liberation was the three characteristics of impermanence (anicca), suffering (dukkha), and not-self (anattā). Practice meant seeing them clearly and thereby ceasing to grasp at physical and mental processes. Craving then ceases, suffering ends, and nirvāṇa is attained. Nirvāṇa itself was mostly described negatively, as unconditioned, neither arising nor ceasing, and involving no further birth. The language concerned "what is absent," not "something wonderful that is there." Nothing was presupposed to have been present all along. The practitioner's direction was to see things as they actually are and let go, not to recognize one's original state.

With tathāgatagarbha, the language of liberation changes to permanence (nitya), bliss (sukha), self (ātman), and purity (śuddha). The Ratnagotravibhāga and the Nirvāṇa Sūtra establish them as the four virtues of nirvāṇa. Each of these four terms reverses the early Buddhist account of the three characteristics. Impermanence becomes permanence, suffering becomes bliss, and not-self becomes self. Impurity, or neutrality, becomes purity. Practice is no longer directed toward seeing clearly and letting go. It turns instead to removing obscurations and revealing what has always been present. "Original awakening," "seeing one's nature and becoming a Buddha," and "recognizing rigpa" share a presupposition. At the deepest level there is something pure, perfect, and originally so, which one need only recognize.

The change is substantial. The whole development of Mahāyāna separates the two positions. One says, "Nothing has been there all along. There is only the process of conditions gathering and dispersing, to be observed as it is and relinquished." The other says, "There is an essence that has always been Buddha, and practice consists in recognizing it." The development passes through Prajñāpāramitā emptiness, Madhyamaka's dismantling of attachment, and consciousness-only's transformation of the basis before reaching this point. The core of early Buddhist liberation, not-self, impermanence, cessation, and letting go, has become only a faint remnant in tathāgatagarbha language. The tradition still speaks of "liberation" and "practice," and still points toward some cessation of suffering. Yet its supporting metaphysics has changed beyond recognition into something else entirely.

The intent here is to state clearly that the shift occurred in history, without judging the tathāgatagarbha tradition as good or bad. Each step had its reasons. The needs of liberation theory called for metaphysical support, and the practical demands of the language of practice called for positive descriptions of the endpoint. Seeing the distance clearly is a requirement of an honest reading of history.

Tathāgatagarbha thought continues to evolve. Modern Tibetan Buddhist and Western scholars continue to discuss other-emptiness and self-emptiness, the distinction between tathāgatagarbha and ātman, and the philosophical status of Dzogchen. Scholars are reexamining Chinese Chan and Tiantai thought from new perspectives. The work involves both historical study and unfinished philosophical inquiry.

References¶