Chapter 1. The Maitreya treatises and the formation of Yogācāra¶

At the end of the preceding essay (Essay 5), Madhyamaka reached its logical limit. It declared everything empty and took conventional consensus as the sole basis of knowledge, leaving it unable to question that basis itself. This was less a failure of Madhyamaka than the inevitable destination of pursuing "refutation without establishing a position" to the end. It left a set of philosophical questions. If both external objects and consciousness lack intrinsic nature, and dependent arising and nominal designation are matters of convention, the subject of practice remains unexplained. So do the storage and transmission of afflictions and what liberation actually acts upon. Madhyamaka's account of emptiness suffices to dismantle mistaken attachments to views, but says almost nothing about the structure of cognition.

Yogācāra formed around these questions. It does not reject Madhyamaka wholesale; it directly addresses the problems Madhyamaka left behind. Once the work of dismantling attachment is underway, someone must explain how cognition and practice actually operate from the standpoint of emptiness. This chapter examines the school's origins: who founded it, which texts it drew on, and where its philosophy began.

1. Historical background: philosophical problems after Madhyamaka¶

The philosophy of the early schools, especially Abhidharma, offered a detailed map of cognition and practice. Dharmas each had their own characteristic; mental factors were enumerated one by one; and the classifications of the five aggregates, twelve sense bases, and eighteen elements provided usable coordinates for practice. Madhyamaka removed the metaphysical foundation of this framework, the real existence of intrinsic natures, in its entirety. Yet it supplied no new epistemological framework in its place. It said only that the old structure did not stand, that everything was nominally designated, and that conventional life went on as usual.

For a scholar teaching scripture at Nālandā, or a practitioner asking in meditation what kind of mind was arising at that moment, the answer was inadequate. They needed to know what an affliction depended on when it arose in a single moment. They also needed to know how seeds (bīja) were deposited through perfuming and how they produced karmic results. Madhyamaka's framework had to account for the phenomena observed in meditation, including the subtle flow of mind and the disappearance of external objects in deep concentration while perception continued.

Abhidharma remained available to them. Its phenomenology of mind, with its exhaustive classifications and rigorous deductions, had not disappeared. But Abhidharma alone was no longer sufficient, and it fell short in two directions at once.

First, Abhidharma was restrained at its core. Its ontology stopped at dharmas as the smallest units: mental and material phenomena arising and ceasing moment by moment. It did not ask why these dharmas operated as they did, but said only that "this is how they are." Philosophically, this restraint was quite honest. But the human mind, within Buddhism or outside it, is naturally dissatisfied with "this is how things are." The mind seeks a deeper mechanism to explain where seeds are stored and how they are preserved. It also seeks an explanation of their continuity in a stream of dharmas that perish moment by moment. Abhidharma offered a functional description, but no convincing underlying structure.

Second, after Madhyamaka denied Abhidharma's ontological foundation, the mind would inevitably grasp at a replacement. If the real existence of the intrinsic natures of all dharmas could not be established, "ultimately real dharmas" could no longer serve as the endpoint of explanation. Explanation would have to move in another direction, deeper still, to find an underlying structure outside the framework of intrinsically real existence that could nevertheless sustain the operations of practice.

The continual search for a deeper structure is the most distinctive feature of Consciousness-Only theory. It does not simply substitute another ontology; on the premise of emptiness, it constructs a more elaborate mechanism of production. At each successive level, something deeper than what is presently visible drives everything else. Storehouse consciousness offers the clearest example. When the six consciousnesses do not explain enough, a seventh is posited; when the first seven still cannot explain continuity, an eighth is posited.

This was the philosophical ground from which Yogācāra emerged. It sought to rebuild a theory of cognition and practice on the premise of emptiness. In its own terms, it sought to explain the "dependent nature" (paratantra-svabhāva), the cognitive process that arises in dependence on conditions.

2. Who was "Maitreya"? A historical problem¶

The lineage of Yogācāra is generally traced back to Maitreya. But a problem runs through the entire history of scholarship on the school: was this "Maitreya" a human being or a bodhisattva?

The traditional account holds that, after long practice, Asaṅga (c. fourth century) was able to visit Tuṣita Heaven in spirit. There he received teachings directly from the bodhisattva Maitreya, the Buddha's successor, who now dwells in Tuṣita's inner court awaiting his descent into this world. Asaṅga brought back five treatises, known as the "five treatises of Maitreya" (also called the "five treatises of the Benevolent One"). This is the consistent account in the Indian Buddhist tradition, and Tibetan Buddhism continues to accept it.

Modern scholars disagree. Some hold that "Maitreya" was a historical person: a master called Maitreya, or Maitreyanātha, whose disciple was Asaṅga. On this view, Maitreyanātha wrote the five treatises, and Asaṅga edited and developed them. The evidence cited includes substantial differences of literary style within the five treatises, making it difficult to regard them as the work of a single person in a single period; passages in historical materials preserved in Tibetan translation that do distinguish Maitreya from Asaṅga; and biographical accounts left by Xuanzang and other travelers to India.¹

Other scholars consider the textual evidence for establishing "Maitreya" as a historical person rather weak. Accounts of Maitreyanātha are mostly indirect citations whose original sources are difficult to trace. There is also evident continuity of thought between the five treatises and Asaṅga's other works. It is therefore reasonable to attribute their composition and compilation as a whole to work led by Asaṅga alone.²

For this book, the question remains unresolved. We cannot rule out the existence of "Maitreya" as a historical person, Maitreyanātha, but the available evidence is insufficient to give him a secure, separate place as an individual philosopher. This book uses the conventional designation "the five treatises of Maitreya," treating them as the foundational corpus of Yogācāra without deciding the precise identity of their human author. In discussing particular teachings, it attributes explicitly to Asaṅga those parts that can confidently be assigned to him.

3. The five treatises of Maitreya¶

Whatever their authorship, these five treatises are foundational texts of Yogācāra thought. Each deserves a brief introduction:

The five treatises differ considerably in style and emphasis. The Yogācārabhūmi is an encyclopedic manual of practice, longer than all the other treatises combined. The Madhyāntavibhāga and Dharmadharmatāvibhāga are the most philosophical. The Abhisamayālaṃkāra has deeper connections with Madhyamaka and is used independently in Tibetan lineages of Prajñāpāramitā study. If Maitreyanātha was a historical person, the Madhyāntavibhāga and Dharmadharmatāvibhāga are the works most likely to be his. Much of the Yogācārabhūmi, by contrast, more clearly reflects the breadth of Asaṅga's own learning and his ability to organize it.

4. Asaṅga: from a celestial visit to a system¶

Asaṅga (c. 310–390; the dates are disputed) was the actual founder of Yogācāra. However the dispute over the authorship of Maitreya's five treatises is resolved, Asaṅga's own writings and organizational work were what enabled the school to become a system.

Tradition holds that he first studied within the early schools, specifically the Vibhajyavāda, though accounts differ. After a long period without a breakthrough in meditative practice, he ascended to Tuṣita Heaven to meet Maitreya. From this encounter he gained insight and began teaching Consciousness-Only. Whether or not this legendary narrative is literally true, it shows that Yogācāra's understanding of itself was deeply bound to meditative practice from the outset, rather than to philosophical debate alone. It was not speculation in a study; it understood itself as arising from observation of the mind's structure in meditation.

Asaṅga's most important works are:

• The Compendium of the Mahāyāna (Mahāyānasaṃgraha): short, yet the clearest outline of the entire Consciousness-Only system. Its chapters discuss storehouse consciousness (ālaya-vijñāna), the three natures (trisvabhāva), the bodhisattva practice of the six perfections, and transformation of the basis. It is one of the principal sources for the next chapter (6.2).⁶
• The Compendium of Abhidharma (Abhidharmasamuccaya): reorganizes Abhidharma's classification of mental factors from the standpoint of Consciousness-Only. It can be regarded as a Yogācāra version of Abhidharma.

After Asaṅga, his younger brother Vasubandhu (c. fourth–fifth centuries) turned from Abhidharma scholarship to Consciousness-Only. He wrote the Twenty Verses and Thirty Verses, becoming the most rigorous exponent of its philosophical arguments.

A scholarly dispute needs to be explained here. In 1951, Frauwallner proposed a highly influential hypothesis: there were actually two historical figures called "Vasubandhu." The elder (c. 320–380) was Asaṅga's younger brother and wrote Yogācāra works on Consciousness-Only. The younger (c. 400–480) was a scholar in the Abhidharma tradition who wrote the Abhidharmakośa and other works. Both bore the name Vasubandhu, rendered in Chinese as Posoupandou, and later generations confused them.³ This "two Vasubandhus" theory was once widely accepted and reshaped the overall chronology of Indian Buddhism.

Later research, however, has increasingly tended to question this hypothesis. The texts themselves show a fairly clear development of thought from the Abhidharmakośa to the Consciousness-Only works. They could readily have been written by one person at different stages of his life: first mastering the Abhidharma of the Sarvāstivāda and Sautrāntika, then turning under Asaṅga's influence to Mahāyāna Consciousness-Only. The traditional sources present this as a coherent narrative that does not require dividing its subject into two people.⁴ This book's position is that Vasubandhu was one person, that his philosophical transformation was real, and that the Abhidharmakośa and Thirty Verses were the work of the same scholar at different periods. A detailed discussion of Vasubandhu is reserved for Chapter 3 (6.3).

5. Two names: Yogācāra and Consciousness-Only¶

The school has two names in parallel use, each emphasizing a different aspect:

Yogācāra: literally "practice in accordance with yoga" or "practitioner of yoga." Here yoga primarily means meditation (dhyāna) and spiritual practice; ācāra means a mode of practice. The name emphasizes the school's origins in a tradition of practice. Its theories developed from observation of the mind in deep concentration, and its account of liberation rests on progressively purifying the structure of consciousness in meditation.

Consciousness-Only (vijñaptimātra or vijñānamātra): literally "nothing but consciousness" or "only consciousness." This is the school's central philosophical claim. The world that we ordinarily suppose to exist externally is actually a manifestation through the transformation (pariṇāma) of consciousness (vijñāna). What we call an "external object" is an object constructed by consciousness in a particular way, not a material entity existing independently of it.

These names do not designate two wholly separate things. It is precisely the discovery in deep concentration that external objects disappear while perception continues that leads to the philosophical conclusion "external objects are actually constructions of consciousness." And precisely because the problem lies in the structure of consciousness, practice must transform consciousness at its basis ("transformation of the basis," āśraya-parāvṛtti). This book will use the two names interchangeably, as the context permits.

A Chinese controversy: Consciousness-Only or Mind-Only?

The distinction between these two names occasioned an interesting debate in China, which merits a brief aside.

In Sanskrit, the central terms are vijñaptimātra, which Xuanzang translated as weishi, "Consciousness-Only," and vijñānamātra, translated as either "Consciousness-Only" or "Mind-Only." Vijñapti in the first term emphasizes "cognitive discrimination" or "representation," the expressive or presentational aspect of consciousness. Vijñāna in the second is the broader "consciousness" or "mind." Indian texts sometimes use the terms together and sometimes give them different emphases. In China, choices of translation led to a philosophical divergence.

Before Xuanzang (602–664), Paramārtha (499–569) had already translated a large body of Consciousness-Only texts, including the Mahāyānasaṃgraha and the Treatise on Consciousness-Only, that is, Vasubandhu's Twenty Verses. Paramārtha used a broader vocabulary. "Consciousness-Only" and "Mind-Only" could sometimes be interchanged in his translations, and his interpretation bore a clear tathāgatagarbha, or buddha-womb, coloring. He tended to identify the "pure aspect" within storehouse consciousness (immaculate consciousness) with the tathāgatagarbha, thereby establishing a connection to the possibility of liberation.

Xuanzang brought back from India the stricter Consciousness-Only position of the lineage of Dharmapāla (c. 530–561). He deliberately distinguished vijñapti (cognitive discrimination, what consciousness manifests) from vijñāna (consciousness). He insisted on translating the doctrine as "Consciousness-Only," not "Mind-Only," because the latter was too broad. It could easily lead toward traditional Chinese theories of mind and nature, conflating "consciousness" with some ontologically foundational "mind" or buddha-nature. Xuanzang held firmly that the doctrine concerned "the manifestations of consciousness," not the presence of a universal "mind."

After returning to China, he organized a large program of scriptural translation. His disciple Kuiji (632–682) systematized the teachings into what became known as the "Faxiang school" or "Consciousness-Only school." Faxiang means "the characteristics and properties of dharmas." Precise analysis of the nature and categories of all dharmas was the school's central methodological commitment.

The fortunes of the Faxiang school in China, however, took a number of turns. Its position was strict. It upheld the doctrine of five lineages (pañcagotrā): sentient beings differed in their capacities, and icchantikas could not become buddhas. This directly conflicted with the belief that "all sentient beings have buddha-nature," which by then had taken deep root in China and was upheld by the Buddha-Nature, Tiantai, and Huayan schools alike. With few successors of comparable ability after Xuanzang and Kuiji, the Faxiang school had largely declined following the Huichang persecution of 845 in the late Tang. It failed to maintain a continuous lineage of teaching in China.⁵

Its fate is revealing. Yogācāra's most rigorously articulated philosophical form proved ill-suited to Chinese conditions. What continued to exert influence in China was a form of Consciousness-Only diluted by tathāgatagarbha thought and blended with the intuition that "the nature of mind is originally pure." It stood further from the original texts, but better met China's religious needs. The tension between these two directions will receive fuller treatment in the chapter on Chinese Buddhism in Essay 10.

6. The philosophical starting point: where consciousness is, there the problem lies¶

To understand why Yogācāra argues as it does, we must first examine the problem from which it begins.

Madhyamaka's analysis of emptiness begins with external objects: they have no intrinsic nature and are nominally designated. Yogācāra asks in a different direction, approaching from the side of the knowing subject. Practitioners in deep concentration observe that, when meditation reaches a certain depth, so-called "external objects" disappear, yet cognition, or perception itself, continues. Yogācāra asks why external objects, if they really exist independently of consciousness, disappear in deep concentration while perception continues. It also asks what that continuing perception is perceiving.

From the observation, Yogācāra argues that the so-called "external object" is merely content that consciousness presents to itself in a particular way. This content, the object aspect (nimitta-bhāga), is not an independently existing thing. This is the starting point of the Consciousness-Only argument, not its conclusion. Vasubandhu refined it into a more formally rigorous version (6.3).

Yogācāra must explain how consciousness manifests the illusion of an "external object" and describe the internal structure of consciousness. Within that structure, it must account for the accumulation of afflictions, the perfuming of seeds, and transformation of the basis. Ālaya-vijñāna, storehouse consciousness, is introduced to perform the function of "storing and transmitting seeds." The next chapter turns to this central concept.

Storehouse consciousness: 6.2. The three natures and transformation of the basis: 6.4.

7. The relationship with Madhyamaka¶

Yogācāra is not a rejection of Madhyamaka. This needs to be clear from the outset.

A more accurate description of the relationship between these two Mahāyāna schools is that each addresses a different side of the same set of problems. Yogācāra does not cancel Madhyamaka's analysis of emptiness: everything lacks intrinsic nature and is nominally designated. Asaṅga and Vasubandhu themselves both explicitly uphold emptiness. Madhyamaka largely stops after saying "empty." Yogācāra maintains that the operational structure of consciousness must still be explained on the premise of emptiness. Otherwise, practitioners would have nothing to work with.

It was precisely this relationship that later made possible an Indian tradition of synthesis called "Yogācāra-Madhyamaka" (Yogācāra-Mādhyamika). Its representative figures were Śāntarakṣita (c. eighth century) and Kamalaśīla. They sought to integrate the epistemological basis of Consciousness-Only with Madhyamaka's analysis of emptiness into a single system. This was an important philosophical development in late Indian Buddhism (see 8.3).

Candrakīrti's refutation of Consciousness-Only (5.5, Section 4), and the latter's dissatisfaction with Madhyamaka's "refutation without establishing a position," show that real philosophical disagreements existed between the schools. This debate will receive a separate discussion in Chapter 5 of this essay (6.5). Before that, we need to see Yogācāra's own philosophical system clearly, beginning with Asaṅga's storehouse consciousness.

References¶