28 ŚĀNTARAKS․ITA Climbing the Ladder to the Ultimate Truth Allison Aitken Introduction to Śāntaraksita’s Life and Works The scholar-monk and prolific author Śāntarakṣita (c. 725–788)1 left a lasting and significant impact on both Indian and Tibetan Madhyamaka Buddhist philosophy. He is known for his synthesis of Nāgārjuna’s (c. second century) Madhyamaka with elements of Dignāga (c. 480– 540 CE) and Dharmakīrti’s (c. seventh century) tradition of logic and epistemology as well as Yogācāra idealist ontology. Śāntarakṣita’s works are characterized by an emphasis on the indispensable role of rational analysis on the Buddhist path as well as serious and systematic engagement with competing Buddhist and non-Buddhist schools of thought. Śāntarakṣita is said to have been abbot of the great monastic university of Nālandā in Magadha (present-day Bihar, India) and counted Kamalaśīla (c. 740–795) and Haribhadra (late eighth century) among his most prominent students. Yet aside from this, few biographical details about Śāntarakṣita’s life in India remain. He did, however, play a central role in the early transmission (snga dar) of Buddhism to Tibet, and numerous semi-legendary reports of his activities there survive. The earliest accounts agree that, upon receiving an imperial invitation to Tibet from King Trisong Detsen (khri srong lde btsan) (742–797?), Śāntarakṣita oversaw the establishment of the first Tibetan Buddhist monastery at Samyé (bsam yas),2 serving as its abbot and ordaining the first Tibetan Buddhist monks into the Mūlasarvāstivāda monastic order (c. 779), whereupon he became known in Tibet as the “Khenpo (mkhan po)/ Ācārya Bodhisattva,” or “Abbot Bodhisattva.” According to Tibetan sources, Śāntarakṣita’s own ordination lineage proceeds as follows: Śāriputra → Rāhula → Nāgārjuna → Bhāviveka (c. sixth century) → Śrīgupta (c. seventh century) → Jñānagarbha (early eighth century) → Śāntarakṣita.3 This lineage also reflects philosophical affinities among these authors, with the later figures influenced by Bhāviveka’s interpretation of Nāgārjuna’s Madhyamaka (which would later come to be known as *Svātantrika-Madhyamaka).4 A sizeable corpus is attributed to Śāntarakṣita, spanning a range of genres and subject matters, including Madhyamaka metaphysics, logic and epistemology, Buddhist path literature, tantra, as well as several praises. His two most important independent treatises are the Compendium of True Principles (Tattvasaṃgraha, hereafter Compendium = TS) and the Ornament of the Middle Way (Madhyamakālaṃkāra, hereafter Ornament = MA) together with an autocommentary (Madhyamakālaṃkāravṛtti = MAV). Kamalaśīla authored lengthy commentaries ( pañjikā-s) 463 DOI: 10.4324/9781351030908-41 Allison Aitken on both the Compendium (TSP) and the Ornament (MAP). In practice, Śāntarakṣita’s basic texts (particularly the Compendium) are standardly read with the aid – and thus through the lens – of Kamalaśīla’s commentaries, such that Śāntarakṣita’s thought and that of his principal student are often inextricable. Several notable commentaries are also attributed to Śāntarakṣita, including one on Jñānagarbha’s The Distinction Between the Two Truths and its Autocommentary (Satyadvayavibhaṅga and Satyadvayavibhaṅgavṛtti = SDV and SDVV, Satyadvayavibhaṅgapañjikā = SDVP), as well as a commentary on Dharmakīrti’s The Logic of Debate (Vādanyāya) titled Commentary on the Logic of Debate: Elucidation of Its Meaning (Vādanyāyaṭīkā Vipañcitārthā).5 Of all the works attributed to Śāntarakṣita, only the Compendium, his commentary on Dharmakīrti’s The Logic of Debate, and a tantric-cum-epistemological work, Establishing the Truth (Tattvasiddhi),6 survive in Sanskrit. Śāntarakṣita’s references to his own works yield the following chronology of composition: Ascertainment of the Ultimate, Compendium, Commentary on the Logic of Debate, Commentary on the Ornament. This suggests that the Commentary on the Ornament represents his most mature thought and is his definitive work on Madhyamaka.7 As noted previously, Śāntarakṣita’s place in the history of Madhyamaka philosophy is perhaps most remarkable for his synthesis of Nāgārjuna’s Madhyamaka with elements from the Dignāga-Dharmakīrti tradition of logic and epistemology together with Yogācāra idealist ontology. Though not the first Mādhyamika to be influenced by Dignāga and Dharmakīrti, Śāntarakṣita looks to be the first Mādhyamika to author a commentary on one of Dharmakīrti’s works. The influence of the Dignāga-Dharmakīrti tradition of epistemology, according to which testimony as a source of knowledge is reduced to a form of inference, is reflected in Śāntarakṣita’s emphasis on the central role of rational analysis in the gradual progression toward a correct metaphysical view. As Śāntarakṣita repeatedly suggests, the ideal reader of his works and the ideal trainee on the Buddhist path is a discerning person ( prekṣāvat), that is, a rational epistemic agent.8 Prior to Śāntarakṣita, Śrīgupta is noteworthy for integrating the Dignāga-Dharmakīrti tradition of logic and epistemology into his presentation of Madhyamaka, but Śrīgupta rejects Yogācāra ontology without qualification.9 And while Jñānagarbha subsequently alludes to Yogācāra conceptual frameworks in his presentation of the Madhyamaka theory of two truths (satyadvaya), viz. the conventional truth (saṃvṛtisatya) and the ultimate truth ( paramārthasatya) (see, e.g., SDVV ad SDV 30), it is Śāntarakṣita who explicitly formalizes the incorporation of Yogācāra ontology into his presentation of Madhyamaka, though relegated to the domain of conventional truth. Śāntarakṣita accepts as conventionally true not only the Yogācāra doctrine that apparently external objects are merely mental in nature (cittamātra) (MA 91–93) but also the Yogācāra claim that cognition is reflexively aware (svasaṃvitti/ svasaṃvedana) (MAV ad MA 91). Following Śāntarakṣita, Kamalaśīla and Haribhadra likewise adopt the Dignāga-Dharmakīrti tradition of logic and epistemology while also taking a conciliatory approach to Yogācāra. Below, we will return to the question of how best to understand Śāntarakṣita’s Yogācāra-Madhyamaka synthesis. Both of Śāntarakṣita’s main treatises, the Compendium and the Ornament, exemplify his wide-ranging and systematic critical engagement with the philosophical views of competing Buddhist and non-Buddhist schools of thought. The Compendium – over 3,600 stanzas organized in twenty-six chapters – in some ways resembles a critical doxographical survey of the eighth-century Indian religio-philosophical landscape, yet such a description does not adequately reflect its dialogical structure or its in-depth engagement with these competing systems.10 The first twenty-three chapters of the treatise examine and ultimately reject a succession of cosmogonical theories, ontological categories, semantic theories, epistemological theories, and candidate sources of knowledge advanced by Nyāya, Vaiśeṣika, Mīmāṃsā, Sāṃkhya, Jaina, Vedānta, and 464 Śāntaraks∙ita Lokāyata/Cārvāka traditions, as well as Buddhist traditions such as the Vātsīputrīya.11 The final three final chapters that constitute nearly the second half of the treatise are largely aimed at Mīmāṃsakas, first rejecting their claim that the Vedas lack a human author (apauruṣeya), next critically examining their theory that veridical cognition is self-certified (svataḥprāmān ․ ya), and finally concluding with an argument in support of the possibility of omniscience. Given the breakdown of the text, the Compendium would seem primarily concerned with competing theories of the Nyāya and Mīmāṃsā, and Śāntarakṣita makes important contributions in formulating Buddhist responses to the Naiyāyika philosopher Uddyotakara (fl. c. 600), as well as the PūrvaMīmāṃsaka Kumārila Bhaṭṭa (fl. c. 660), both of whom were influential critics of Dignāga. The Ornament similarly rejects a range of positions from competing Buddhist and nonBuddhist systems, though it is more metaphysical in its focus than the Compendium, with roughly two-thirds of the ninety-seven stanzas devoted to the neither-one-nor-many argument (ekānekaviyogahetu), which sets out to demonstrate that nothing possesses an intrinsic nature (svabhāva). While the argument takes aim at entities advanced by many of the same nonBuddhist traditions addressed in the Compendium, the Ornament is concerned foremost with competing Buddhist views. In presenting and rejecting the views he addresses in both the Compendium and the Ornament, Śāntarakṣita utilizes a dialectical/pedagogical device of provisionally adopting what he deems to be successively more rational positions in order to reject less rational ones. Sara McClintock has influentially described this method whereby Śāntarakṣita argues from progressively shifting perspectives as a “sliding scale of analysis.”12 To illustrate, in the Ornament, Śāntarakṣita adopts the Sautrāntika representationalist theory of perception in order to reject the Vaibhāṣika direct realist theory of perception. He then assumes the Yogācāra idealist position on which mental representations have no external referents in order to reject Sautrāntika representationalism. Finally, he uses the Madhyamaka account of the nonexistence of fundamentally real cognition to reject Yogācāra theories of the mind and mental content. Although the Compendium includes several allusions to the superiority of the Madhyamaka perspective,13 it might be read as culminating in the Yogācāra perspective. By contrast, the Ornament, in which Śāntarakṣita presents his definitive account of Madhyamaka, devotes more critical attention to Yogācāra than to any other competing system. Yet it is in this same text that Śāntarakṣita presents his provisional endorsement of Yogācāra idealism on the level of conventional truth. The following sections will take up Śāntarakṣita’s contributions to the Madhyamaka theory of two truths in the Ornament. Ultimate Truth and the Neither-One-Nor-Many Argument The Madhyamaka central commitment, or ultimate truth, is the emptiness of intrinsic nature (svabhāvaśūnyatā), which might be glossed as the universal negation of ontological independence. In other words, according to Mādhyamikas, nothing lays claim to ontological selfsufficiency, which is commonly identified as a necessary condition for fundamentality and substancehood. The Madhyamaka view thus might be described as a kind of thoroughgoing anti-foundationalism as well as a form of substance nihilism. But if there are no ontologically independent or fundamental entities, then whatever there is depends for both its nature and its existence on something else. In his Ornament, Śāntarakṣita makes a major contribution to the Madhyamaka canon of arguments for emptiness with his presentation of the neither-one-nor-many argument. Although he expands on his predecessor Śrīgupta’s more condensed formulation of the argument in the Introduction to Reality, it is Śāntarakṣita’s influential Ornament that popularizes the argument 465 Allison Aitken in both Buddhist India and Tibet. In his Illumination of the Middle Way (Madhyamakāloka), Kamalaśīla presents the neither-one-nor-many argument among a set of five Madhyamaka arguments for emptiness, which subsequently became known in Tibet as the “five great arguments” for emptiness (gtan tshigs chen po lnga). Nāgārjuna articulates an early precursor to the neither-one-nor-many argument in his Precious Garland (Ratnāvalī), stating: Something is not a unity if it has multiple loci. There is nothing that lacks multiple loci. In the absence of any unity, neither is there a multiplicity. (Precious Garland 1.71) Here, Nāgārjuna argues that whatever is divisible into multiple discrete spatial or temporal loci does not count as a true unity. And everything, he claims, is so divisible. Just as each bit of matter – regardless of how minute – has a right side and a left side, a top and a bottom, and so on; likewise, each moment of time – no matter how brief – has a beginning, a middle, and an end. Otherwise, the existence of the spatially and temporally extended ordinary objects that populate our world – like computers, kangaroos, and cognitions – would be impossible. After all, the thought goes, how could fundamental building blocks that lack spatial/temporal extension ever yield anything that has spatial/temporal extension? And since a plurality presupposes unities as its basic constituents, if there is nothing that is truly one, then neither is there anything that is truly many. Śrīgupta's expanded formulation of the neither-one-nor-many argument makes explicit the implication that nothing has intrinsic nature, while also formalizing the argument and defending its soundness according to the Dignāga-Dharmakīrti tradition of logic and epistemology. Closely following Śrīgupta’s Introduction to Reality (TA 1), Śāntarakṣita articulates the central inference of the neither-one-nor-many argument in the opening stanza of the Ornament as follows: In reality, everything that is theorized by our own and other schools of thought is without intrinsic nature, due to lacking an intrinsic nature that is either one or many, like a reflection. (MA 1) The argument poses a destructive dilemma, which says: if anything has an intrinsic nature, then it is either one or many. Śāntarakṣita, in effect, argues that nothing can satisfy either disjunct of the consequent and therefore, by modus tollens, that nothing can satisfy the antecedent. Upon analysis, nothing possesses an intrinsic nature. Three features of the disjunctive property pair <one or many> are critical for the argument to go through. First, the terms translated as “one” and “many” here (eka and aneka in Sanskrit) are perhaps more precisely rendered as “unity” and “non-unity,” reflecting the fact that they are a mutually exclusive and contradictory pair, conforming to the conceptual, logical, and grammatical structure F and not-F. As Śāntarakṣita makes clear, if anything had an intrinsic nature, then on pain of violating the law of excluded middle, it would have to either be a unity or non-unity: Aside from unity and not-unity, an object’s having some other classification is impossible, since it is established that these two properties are mutually exclusive. (MA 62) 466 Śāntaraks∙ita Second, the operator, or qualifier, “in reality” (tattvataḥ) in the statement of the central inference clarifies that the target here is a true unity and a true multiplicity. This should be contrasted with a merely conventional status, like the unity of an aggregate such as a flock of sheep or a heap of sand. And, as indicated by Nāgārjuna, a true unity is defined as a mereological simple, that is, something that lacks proper parts, where something has proper parts just in case it is either physically or conceptually divisible. While physical divisibility is more or less straightforward, we can understand that x is conceptually divisible in the mereological sense just in case there are conceptually isolatable proper parts ys that compose x, such that x is the sum of the ys. A true multiplicity, then, is something that has proper parts, the most basic of which are themselves true unities. Finally, a third feature of this property pair is that, unlike most contradictories, unity and non-unity share not only a conceptual priority relation but also a metaphysical priority relation: the existence of a non-unity presupposes the existence of some unities. As Nāgārjuna pointed out in his Precious Garland, a plurality requires singular things as its building blocks. Śāntarakṣita explains, Thus, a “multiplicity” is defined as a composite of unities. If no unity exists, neither does a multiplicity, just like if no trees exist, neither does a forest. (MAV ad MA 61) But if a multiplicity depends for its existence on some unities just like a forest does on some trees, then a multiplicity is not a candidate for ontologically independent being after all. As it turns out, then, true unity is a necessary criterion for ontological independence. And this should not be so surprising: just like ontological independence, true unity is commonly cited as a necessary condition for fundamentality as well as for substantial reality. The neither-onenor-many argument thus reduces to a rejection of true unities, which is to say a rejection of mereological simples. The Neither-One-Nor-Many Argument Against Material Simples Śāntarakṣita’s rejection of material simples closely follows Vasubandhu’s (c. fourth–fifth century) anti-atomist argument in his Yogācāra work, the Twenty Verses (Viṃśikā 11–15). This section of his neither-one-nor-many argument also features in the “Examination of External Objects” (Bahirarthaparīkṣā) chapter of his Compendium, wherein he assumes the Yogācāra perspective. Śāntarakṣita targets three kinds of views about how atoms aggregate to constitute composites, which recur in debates of this kind in pre-modern Indian philosophy: i. Each atom conjoins with surrounding atoms. ii. Atoms have interceding space between them. iii. Atoms are spatially continuous, neither conjoining with surrounding atoms nor having interceding space between them.14 To each of these views, Śāntarakṣita, in effect, poses the following dilemma: If matter is constituted by fundamental, simple particles, then those particles either face surrounding particles at one and the same locus or at spatially differentiable loci. If, on the one hand, fundamental particles did not have spatially differentiable loci at which to face neighboring particles, and were thus spatially unextended, then they could not compose an extended composite. If, on the other hand, fundamental particles did have spatially differentiable loci at which to face 467 Allison Aitken surrounding particles (e.g., a right side, a left side, etc.), then they would have spatially discrete parts, which means that they would be composites themselves and could not be fundamental unities after all. As Śāntarakṣita argues in both his Ornament and Compendium, Whether atoms are (i) conjoined, (ii) located at a distance from one another, or (iii) located continuously without interceding space, if the very same part of the central atom in a composite which is facing one atom were also imagined to be facing another atom, then the aggregation of atoms composing mountains, etc., would not be feasible. (TS 1989–90 = MA 11–12) If instead it were accepted that a different part of the central atom faced another atom, then how indeed could an atom like that, i.e., with distinct parts facing different atoms, be truly unitary? (TS 1991 = MA 13) He concludes that there is no account on which matter could be founded in simple particles. And given the metaphysical priority of true unities to a true multiplicity, in the absence of material simples, a material multitude is also precluded. The Neither-One-Nor-Many Argument Against Mental Simples While the Compendium restricts the subject of the neither-one-nor-many argument to external objects, the Ornament grants the argument a universal scope of application. Having rejected the true unity of any extramental entities within the first fifteen stanzas of the Ornament, Śāntarakṣita devotes stanzas 16–60 to rejecting the true unity of the mind, addressing a variety of Buddhist and non-Buddhist accounts of the mind and mental content, with the argument culminating in a sixteen-stanza section targeting Yogācāra theories. Śāntarakṣita introduces this section, remarking: Even though the Yogācāra view has merit, we shall consider whether such mental entities are to be accepted as real or as satisfactory only when not analyzed. (MA 45)15 The succeeding argument turns on an analysis of the relation between the mind and mental content qua cognition ( jñāna) and mental representations (ākāra). Śāntarakṣita targets two families of views from the Yogācāra tradition on the ontological status of mental representations: i. ii. Representational realism (*satyākāravāda): representations are real in the same way as cognition is taken to be.16 Representational antirealism (*alīkākāravāda): representations are unreal figments.17 It is important to keep in mind that for Yogācārins, who reject mind-independent material objects, a representation does not actually represent any extramental entity but is simply the intentional object of a cognition. Thus, the question of the ontological status of representations concerns not the represented content (like a desk or a dragon) but rather the representation itself as a feature of the mind.18 Representational realism is commonly associated with a second claim which says that cognition is necessarily and intrinsically endowed with representations 468 Śāntaraks∙ita (sākāra), while representational antirealism is commonly associated with the claim that cognition is not necessarily endowed with representations, and that invariably veridical enlightened cognition lacks representations (nirākāra).19 In addressing the first view, on which cognition is intrinsically and necessarily endowed with real representations, Śāntarakṣita takes it that, according to this theory, a cognition and its representation are non-distinct, constituting a single subject. Supposing that representations and cognition share a strict identity relation, his reasoning here turns on a version of the law of noncontradiction according to which contradictory properties cannot be predicated of the same subject. Śāntarakṣita observes that a moment of cognition seems to be indisputably simple, and yet the content of cognition looks obviously complex; in any given moment, ordinary experience presents us with a multiplicity of data – a white patch here, a blue patch there, and so on. And, indeed, perhaps the most intuitive representational realist view (the so-called citrādvaita, or “variegated nonduality” theory) says that unitary cognition is non-distinct from its multifaceted representation. But, Śāntarakṣita argues, given the law of noncontradiction, if cognition and representations are non-distinct, it follows that: i. ii. since a representation is manifold, so too is cognition, or else since cognition is truly unitary, so too is its representation. Śāntarakṣita lays out this argument from dilemma as follows: It is difficult to deny that: (i) due to being non-distinct from real representations, cognition must accord with the nature of representations and therefore be manifold. Or else, (ii) due to being non-distinct from unitary cognition, representations would have to be unitary in accord with the nature of cognition. On account of having contradictory properties, ultimately, representations and cognition would have to be distinct. (MAV ad MA 46) Śāntarakṣita takes up the second horn of the dilemma first, arguing that if a representation were truly unitary in accord with cognition, then absurd consequences would follow. For instance, since a simple representation could not be analytically divisible into phenomenal proper parts, we would be unable to conceptually isolate different aspects of our phenomenal field (like the right side and left side of this page). Furthermore, in the absence of phenomenal parts, it could never be the case that one element of our experience was in motion while another was at rest (MA 47–48). On the other hand, Śāntarakṣita reasons, the alternative that cognition is manifold in accord with its complex representation is susceptible to the same kind of argument that he leveled against material atomism: just as an extended material object could not be constituted by unextended material simples, a phenomenally extended representation (like the one you may have of this page) could not be composed of phenomenally unextended building blocks (MA 49).20 And given the metaphysical priority of unity to multiplicity, if there are no simple phenomenal parts, neither can there be a true multiplicity of them. The parts of cognition, then, could not exist in numerical parity with representational parts, since there can be no determinate number of them to which cognition might correspond. He thus concludes that cognition and a real representation could be neither truly one nor truly many. Śāntarakṣita next turns to the representational antirealist view on which cognition is not actually endowed with real representations, which only seem to appear to cognition due to an error (MA 52). This view, Śāntarakṣita argues, is incapable of accounting for ordinary experience, for how could we perceive anything if no percept exists (MA 53–54)? Indeed, he insists 469 Allison Aitken that there can be no cognition at all in the absence of an intentional object, since cognition is intentional by its very nature; to cognize is to have a cognition of something (MA 55). Moreover, Śāntarakṣita argues, an unreal representation could stand in no relation whatsoever with cognition, whether that be an identity relation or a causal relation (MA 57). If a representation stood in an identity relation with cognition, then given the law of noncontradiction, either: i. ii. since cognition is real, the representation too would be real, or else since the representation is unreal, cognition too would be unreal (MAV ad MA 57). Neither alternative is admissible for the representational antirealist. Furthermore, an unreal representation is no more capable of standing in a causal relation with cognition than an identity relation, since if a representation were caused, then it would be real, but if it had no cause, then there could be no explanation for its appearing with spatiotemporal determinacy or consistency (MA 58). Unable to get the semblance of a defeasible account of cognition up and running on this view, Śāntarakṣita does not even bother to apply the neither-one-nor-many analysis to the representational antirealist position. Having dismissed both the representational realist and antirealist views as untenable, Śāntarakṣita takes himself to have demonstrated that there is no account in which a mental entity could exist as a true unity or a true multitude, and thus concludes that nothing – whether material or mental – lays claim to ontologically independent being. Instead, the only kind of unity and being that exist are conventional and dependent. To flesh out precisely what Śāntarakṣita means by this, let’s turn to the second of the two truths. Conventional Truth and Yogācāra Ontology Upon concluding his neither-one-nor-many argument for the ultimate truth, Śāntarakṣita presents his account of conventional truth to clarify that the rejection of ontological independence does not entail an unmitigated nihilism. The term satya translated here as “truth” has a semantic range that is also inclusive of “reality,” and Śāntarakṣita’s account of the conventional is a theory of truth as well as an ontological theory. Yet Mādhyamikas, like Śāntarakṣita, affirm only an ultimate truth and not an ultimate reality qua ontological status. In fact, the ultimate truth as the emptiness of intrinsic nature might be interpreted as the claim that nothing is ultimately real.21 There is thus only one ontological status: conventional reality. According to Śāntarakṣita, whatever is conventionally real (i) has the capacity for causal efficacy (arthakriyāśakti/arthakriyāsamartha), (ii) is dependently originated ( pratītyasamutpanna), and (iii) satisfies our ordinary notions of unity and being only when not subjected to analysis into its final nature (avicāraraman ․ īya/avicāramanohara) (MA 64). Conventional truths, then, are pragmatically efficacious claims that concern conventionally real things and which may be verified by our epistemic instruments of perception and inference. With this account, Śāntarakṣita once again follows his predecessor Śrīgupta, who presents the earliest extant formulation of this threefold criterion,22 which was subsequently adopted by Jñānagarbha, Kamalaśīla, Haribhadra, the later Bhāviveka (c. eighth century), Atiśa (982–1054), and others.23 The first criterion for conventional reality – having the capacity for causal efficacy – is a repurposing of Dharmakīrti’s criterion for ultimately reality.24 Though an apparent subversion of Dharmakīrti’s intent, this criterion represents yet another Dharmakīrtian influence on this branch of the Madhyamaka tradition. We will return to the third criterion in treating the role of analysis in Śāntarakṣita’s account of conventional truth below, but it is with the second criterion, being dependently originated, that Śāntarakṣita incorporates Yogācāra ontology into his system. 470 Śāntaraks∙ita The claim that whatever is conventionally real comes into being in dependence on other things goes back to Nāgārjuna.25 The ontological dependence relation implicated here is inclusive of mereological dependence, mind-dependence, as well as causal dependence, such that every conventionally real thing: (i) comes into being in dependence upon its parts and those parts upon their own parts, ad indefinitum, (ii) is individuated as a conventional unity in dependence upon mental designation, and (iii) is a product of causes and conditions, each of which is in turn a product of its own causes and conditions, ad indefinitum. But, marking a significant departure from prior Madhyamaka accounts, Śāntarakṣita identifies all things involved in causal relations as mental in nature, thereby aligning his presentation of conventional reality with Yogācāra ontology and its central commitment that everything consists in cognition alone (vijñaptimātra):26 Whatever exists as cause and effect is, in fact, merely cognition ( jñāna). Whatever is established by cognition itself exists in cognition. (MA 91) In order to understand how Śāntarakṣita understands Yogācāra ontology to map onto conventional reality, it is necessary to first pin down what precisely he means by affirming the Yogācāra commitment that everything is merely mental (cittamātra). Here are some possible interpretations: i. ii. A phenomenological claim on which the only things relevant to our experience are mental A kind of skepticism which says that we cannot know whether or not there exist any extramental entities iii. An epistemological idealism which says that all objects of knowledge are determined by, or dependent on, the mind and the structure of thought iv. An immaterialism, or metaphysical idealism, on which there are no material things, and the only kinds of things that exist are mental The strongest claim, (iv) immaterialism, is an eliminative idealism insofar as it effectively eliminates, or precludes the existence of, extramental things. The former three options are varieties of non-eliminative idealism insofar as they grant some kind of primacy to the mental but leave open the possibility that extramental things exist. Some have argued for a version of the (i) phenomenological reading of Śāntarakṣita’s Yogācāra by pointing out that he seems to identify conventionally real things with appearances and thus exclusively with what lies within the domain of experience.27 Indeed, in a rather customary Madhyamaka move following an argument for emptiness, Śāntarakṣita insists that the conclusion of his neither-one-nor-many argument does not entail the denial of appearances (MA 78ab). To do so would be tantamount to an implausible thoroughgoing nihilism. But Śāntarakṣita’s nondenial of appearances should not be read as an anti-metaphysical, phenomenological turn; nor does it serve to restrict of the scope of knowledge to the domain of appearances, indicative of skepticism along the lines of view (ii). After all, he rejects as irrational and untenable both direct realist and representationalist theories of perception, which suppose mind-independent external objects to be the direct and indirect objects of perception, respectively. And he does not simply argue that such objects are unknowable or irrelevant to our experience. Rather, he insists that the existence of external objects founded in atoms is incoherent and thus metaphysically impossible. And just as Śāntarakṣita’s arguments rejecting a substantial self and a creator god, for example, are not intended to leave the back door open to 471 Allison Aitken the existence of such entities lying beyond the scope of our experience, ordinary cognition, or the reach of human knowledge, presumably the same is true of his rejection of material objects. That Śāntarakṣita’s Yogācāra is best read not as (iii) an epistemological idealism (in which all objects of knowledge are determined by, or dependent on, the mind) but as an eliminative (iv) metaphysical idealism is supported by his commentary on MA 91, where he states in no uncertain terms that external material objects do not exist and what we ordinarily take to be external objects simply are one’s own mind. This agrees with the Compendium chapter on the “Examination of External Objects,” wherein Śāntarakṣita assumes the Yogācāra perspective. Here, he lays out two main lines of reasoning against external objects, the first an epistemological argument against the possibility of having knowledge of the existence of external objects founded in atoms and the second an argument against the metaphysical possibility of such objects.28 He transitions from the epistemological to the metaphysical argument with reference to the discerning epistemic agent ( prekṣāvat), once again gesturing to the prominent place of rational analysis in his system: Be as it may that atoms are not established by any source of knowledge, there may nonetheless be doubt. But how should a discerning person come to have certainty about their nonexistence? (TS 1988) Śāntarakṣita aims to pull the rug out from under the external world by rejecting the existence of its purported foundations: material simples.29 That Śāntarakṣita takes the Yogācāra mindonly thesis to negate the existence (and not merely the epistemic accessibility) of real external objects is reiterated in his Yogācāra formulation of the neither-one-nor-many argument: Thus, it is appropriate for discerning individuals to ascertain that atoms are nonexistent, due to being empty of an intrinsic nature that is either one or many, like a lotus in the sky. (TS 1996) According to Śāntarakṣita’s Yogācāra, atoms – the purported building blocks of material objects – are no more real than lotuses growing in midair. Śāntaraksita’s Madhyamaka-Yogācāra Synthesis: An Instrumentalist Approach to the Ultimate With Śāntarakṣita’s characterization of Yogācāra ontology in place, let us turn now to the question of how precisely to understand his Yogācāra-Madhyamaka synthesis. That is, does Śāntarakṣita propose a genuine and coherent unification of these two systems, or does he have an instrumentalist story in mind? Given his identification of the conventional truth with Yogācāra ontology, this question is tied up with how best to understand the relation between the two truths, that is, between metaphysical idealism as the conventional truth and the universal negation of ontological independence as the ultimate truth. Are these two truths compatible or contradictory? One Madhyamaka story about the relation between the two truths sees them as contradictory inasmuch as (i) conventional truths are true according to the non-veridical beliefs and linguistic-conceptual norms of ordinary folks whose understanding is obscured by metaphysical ignorance, while (ii) the ultimate truth reflects the veridical cognition of an ideal epistemic agent. But this is not Śāntarakṣita’s story. The radically unintuitive claim that all things are 472 Śāntaraks∙ita merely mental in nature is plainly difficult to square with the commonly accepted view of the world.30 The conventional truth of metaphysical idealism can hardly be “read off ” our linguistic-conceptual norms. For Śāntarakṣita, the general consensus is not a guide to what is conventionally true or real. Instead, he insists that a discerning person should seek a correct understanding of conventional reality by employing analysis; rationality is king as a guide to both the conventional and the ultimate. Another Madhyamaka story about the two truths sees them as perfectly consistent and even synonymous. On this view, the ultimate truth as the universal negation of ontological independence is the obverse (viz. negative equivalent) of the conventional truth as the universal affirmation of dependent origination. Yet Śāntarakṣita does not see the two truths as wholly compatible either. This is evident from a comparative analysis of his Yogācāra and Madhyamaka versions of the neither-one-nor-many argument in the Compendium and Ornament. Not only are the subjects of the two arguments different (the Yogācāra iteration takes up atoms and the Madhyamaka argument concerns all things), but the predicate is also different, yielding distinct inference warranting entailment relations (vyāpti):31 Yogācāra entailment relation: Whatever is neither-one-nor-many does not exist. Madhyamaka entailment relation: Whatever is neither-one-nor-many lacks an intrinsic nature. How can Śāntarakṣita consistently maintain that this same neither-one-nor-many reason establishes the non-existence of atoms in his Yogācāra iteration of the argument and the absence of an intrinsic nature of all things in his Madhyamaka formulation? If, on the one hand, the Yogācāra entailment relation holds, then his Madhyamaka neither-one-nor-many argument – which applies this same reason to an unrestricted domain – commits him to a thoroughgoing nihilism. If, on the other hand, the Yogācāra entailment relation does not obtain, then Śāntarakṣita advances a fallacious argument in support of his account of conventional truth. Śāntarakṣita must surely reject the first alternative; nihilism is not an option. He thus looks committed to the second horn of the dilemma, that his argument in support of his view of conventional truth is fallacious. Yet this dilemma stands only if Śāntarakṣita intends to hold the Yogācāra and Madhyamaka perspectives simultaneously.32 But, as he points out, once one has realized the Madhyamaka ultimate truth that all things are equally devoid of ontological independence, one necessarily relinquishes the Yogācāra ontology that grants a privileged position to the mental: Those whose intellectual capacity is not slight and particularly those who are highly industrious will, upon analyzing whether the mind has a unitary or manifold nature, ultimately perceive no such entity. Thus, in reality the mind-only view is not accepted. (MAV ad MA 92) But why bother reasoning our way to metaphysical idealism if it is not ultimately true? According to Śāntarakṣita’s sliding scale of analysis, there are better and worse conventional truths, with more rational stories supplanting the less rational. But there is no definitive conventionally true story. In setting up his characterization of cause and effect as mental in nature in MA 91, Śāntarakṣita states: Whoever accepts the conventional reality of those things that stand in causal relations should analyze what those accepted conventional things are in order to respond 473 Allison Aitken to fallacious arguments: Are they merely in the nature of the mind and mental constituents or is their nature also extramental? He goes on to explain that among Mādhyamikas, there are two opinions on the matter. The first camp, exemplified by Bhāviveka, accepts conventionally real material and mental entities alike. He introduces the second view – that cause and effect are merely mental – as simply “the opinion of others.” Significantly, he does not reject the first opinion. He does, however, explain the pedagogical utility of the latter. Śāntarakṣita thus sets out an instrumentalist account of the relation between the two truths on which Yogācāra metaphysical idealism is provisionally accepted as conventionally true as an expedient means to arrive at an understanding of the Madhyamaka ultimate truth. Śāntarakṣita’s conventional truth is not determinate but contextual, as indicated by the third criterion for conventional truth that it does not withstand analysis. Importantly, Śāntarakṣita does not follow Candrakīrti (Introduction to the Middle Way, Madhyamakāvatāra 6.35) or Jñānagarbha (SDV 21) in articulating this criterion as the normative claim that one ought not to analyze conventional truths. This would be incompatible with Śāntarakṣita’s progressive path, which the discerning person traverses precisely by analyzing conventional truths. Instead, following Śrīgupta, Śāntarakṣita simply claims that conventional truths satisfy when not analyzed; a conventionally real thing satisfies our notions of reality, independence, and unity when its ultimate nature is not subjected to analysis. We may arrive at progressively more rational conventional truths through analysis, but there is no final conventionally true theory; metaphysical inquiry into the nature of things has no termination point. Still, if there is no determinate conventional truth, what makes one conventionally true theory more rational than another? Śāntarakṣita implies that since nothing withstands analysis, a theory that posits fewer ontologically independent entities is more rational than one that posits more insofar as it is closer to the ultimate truth. In both his Compendium and Ornament, Śāntarakṣita uses analysis to gradually eliminate ontological categories, with the Yogācāra sparse ontology of “mindonly” being, as it were, the last man standing. But at the end of the day, this category too does not withstand analysis. The primary utility of Yogācāra isn’t in what it affirms but in what it denies. From an idealist, rather than a dualist view of conventional reality, it simply takes fewer steps to arrive at an understanding of the ultimate truth, which undermines the fundamentality of anything – whether material or mental. Śāntarakṣita introduces his Yogācāra-Madhyamaka synthesis, stating: Based on the perspective of the mind-only system (cittamātra), one should understand that there are no real external objects. Based on this Madhyamaka system, one should understand that the mind too is utterly selfless. (MA 92) Those who hold the reins of rationality while riding the chariot of these two systems will thereby achieve the state of a genuine proponent of the Mahāyāna. (MA 93) Śāntarakṣita recommends approaching the ultimate truth via the Yogācāra view, but it is just that – an approach to the ultimate, not a definitive or unrevisable claim about the final nature of things. Thus, on this picture, one cannot definitively claim that the two truths are either compatible or contradictory. The conventional truth is a moving target on shifting sands. That Yogācāra idealism is just one among other instructively efficacious conventional stories by 474 Śāntaraks∙ita way of which one might arrive at the ultimate is supported by the following remark from Kamalaśīla’s Illumination of the Middle Way: Thus, one who is unable to instantaneously realize the fact that all things without exception lack an intrinsic nature should, by temporarily relying on the mind-only system, proceed in stages, beginning with understanding that external objects lack an intrinsic nature. (D 3887, 157a) In this same instrumentalist spirit, Śāntarakṣita (MAV ad MA 70) cites Bhāviveka’s famed metaphor of conventional truth as a ladder to the ultimate truth: Without the ladder of conventional truth, it would not be possible for the learned to ascend to the pinnacle of the palace of reality. (Verses on the Heart of the Middle Way, Madhyamakahṛdayakārikā 3.12) In Śāntarakṣita’s presentation of the two truths, the Yogācāra idealist ontology – though not the exclusive, determinate, or preeminent conventional truth – is nonetheless a highly efficacious penultimate steppingstone to understanding the Madhyamaka ultimate truth. Conclusion Śāntarakṣita remained a Madhyamaka authority to be reckoned with well into the final period of Buddhism in India. His direct students, Kamalaśīla and Haribhadra, adopted and developed his Madhyamaka-Yogācāra synthesis, and he was recognized as a principal source on the Madhyamaka neither-one-nor-many argument throughout the succeeding centuries by authors such as Dharmamitra (fl. ca. 800), Jitāri (late tenth century), Bodhibhadra (fl. c. 1000), Prajñākaramati (ca. 950–1030), Atiśa, Prajñāmokṣa (ca. eleventh century), Abhayākaragupta (late eleventh–early twelfth century), and so on. Later prominent Yogācāra philosophers including Ratnākaraśānti (ca. eleventh century) and Jñānaśrīmitra (ca. eleventh century) also cited and responded to Śāntarakṣita’s neither-one-nor-many argument, indicating that they considered his attack on Yogācāra theories of the mind and mental content a serious enough threat to necessitate critical engagement several hundred years later. In Tibet, Śāntarakṣita’s Ornament together with Jñānagarbha’s Differentiation of the Two Truths and Kamalaśīla’s Illumination of the Middle Way came to be known as the major works of the so-called “three Mādhyamikas of the East” (dbu ma shar gsum), with commentaries composed on the Ornament by such philosophically and temporally diverse luminaries as Chapa Chökyi Senggé (Phya pa Chos kyi seng ge, 1109–1169), Tsongkhapa Lobsang Drakpa (1357–1419), and Jamgön Ju Mipham Gyatso (1846–1912). Śāntarakṣita’s emphasis on the role of rationality and a progressive path, as well as his synthesis of Madhyamaka with the Dignāga-Dharmakīrti tradition of logic and epistemology, left a lasting and definite impact on Indian and Tibetan Buddhist philosophy.33 Notes 1 See Frauwallner 1961, 141–43. 2 See, for example, the Records of the Ba Clan (sba/dba’ bzhed) for one of the earliest sources (Wangdu and Diemberger 2000). For a much later account from a compilation of sources, see Butön Rinchen Drup’s (bu ston rin chen grub, 1290–1364) History of Buddhism (chos ’byung gsung rab rin po che’i 475 Allison Aitken 3 4 5 6 7 8 9 10 11 12 13 14 gter mdzod), translated in Obermiller (1932, 187–92), and Gö Lotsawa Zhönnu Pel’s (’gos lo tsā ba gzhon nu dpal, 1392–1481) Blue Annals (deb gter sngon po), translated in Roerich (1949, 41–44). For a recent compilation of sources on the life and work of Śāntarakṣita, see Eltschinger 2019. Portraits of the members of this lineage were painted on the walls of Samyé; see, for instance, Obermiller 1932, 190; Roerich 1949, 34. Tibetan doxographies commonly present Śāntarakṣita as an exemplar of the so-called *YogācāraSvātantrika-Madhyamaka school of thought and Bhāviveka as the representative of *SautrāntikaSvātantrika-Madhyamaka, while Buddhapālita and Candrakīrti are standardly cited as paradigmatic proponents of *Prāsaṅgika-Madhyamaka. “Yogācāra” in Śāntarakṣita’s doxographical designation indicates his synthesis of Yogācāra ontology into his account of conventional truth, while the *Svātantrika label signifies (in part) his style of argumentation which, following Bhāviveka, deploys independent inferential arguments (svatantrānumāna) to establish his theses rather than exclusively utilizing reductio ad absurdum arguments ( prasaṅga) to undermine the theses of his opponents. On the Tibetan doxographical assignments of Śāntarakṣita, see Seyfort Ruegg 1981, 87–100; Blumenthal 2004, 41–470, and for a critical analysis of this doxographical assignment, see McClintock 2003. For discussions on the historical development of the *Svātantrika and *Prāsaṅgika categories, see Dreyfus and McClintock 2003 and Seyfort Ruegg 2006. Śāntarakṣita’s authorship of this work has been called into question by Tsongkhapa Lobsang Drakpa (Tsong kha pa Blo bzang grags pa, 1357–1419) and Tāranātha (1575–1634) owing to the fact that Kamalaśīla appears to argue against the author of the SDVP in the TSP; see Eckel 1987, 27–31; Blumenthal 2004, 29; Seyfort Ruegg 1981, 68 n. 224. The attribution of Tattvasiddhi to Śāntarakṣita has been called into question by Steinkellner (1999, 356–57). In addition to the texts already mentioned, the other texts attributed to Śāntarakṣita in the Tengyur are: two short praises, the Praise of the Eight Tathāgatas (*Aṣṭatathāgatastotra) and Praise of the Bhagavan: Song of Śrī Vajradhara (*Śrīvajradharasaṃgītibhagavatstotra) together with autocommentary (ṭīkā); two tantric works, Five Great Instructions on Kurukulla Arisen from Hevajra (Hevajrodbhavakurukullāyāḥ Pañcamahopadeśa) and Ritual for the Extensive Recitation of the Previous Aspirations of the Seven Tathāgatas Collected from Sūtras (Saptatathāgatapūrvapraṇidhānaviśeṣavistārakalpavacanavidhi-sūtrāntasaṃkṣepa); and a commentary (vṛtti) on Candragomin’s (seventh century) Twenty Verses on the Bodhisattva Vows (Bodhisattvasaṃvaravṃśaka). In his Commentary on the Ornament, Śāntarakṣita appears to allude to another text he authored titled Ascertainment of the Ultimate (*Paramārthaviniścaya) (Ichigō 1985, 330). As Eltschinger (2019, 384–85) points out, Śāntarakṣita also seems to allude to this work in TS 2083. See McClintock 2010, 58–62, 2013; Tillemans 2011, 153–54, 2016, 143–44 on Kamalaśīla’s elaboration on this concept, particularly in the context of discerning the correct understanding of conventional truth; on the term, prekṣāvat, see also Eltschinger 2007, 137–50, 2014, 195 n. 17, 219–34. On Śrīgupta’s Introduction to Reality (Tattvāvatāra = TA) and accompanying autocommentary (-vṛtti = TAV), see Aitken (Forthcoming). For a short synopsis of the Compendium, see Seyfort Ruegg 1981, 89–90, and for a compilation of secondary scholarship on this work, see Steinkellner and Much 1995, 56–63; Eltschinger 2019, 385–86. Śāntarakṣita presents a summary of the topics of the text in the first six stanzas of the Compendium, with the main points of the first twenty-three chapters glossed as modifying dependent origination and those of the final three chapters modifying the Buddha who taught dependent origination. On the two-part structure of the Compendium, see McClintock 2010, 97–98. On the application of this framework for describing Śāntarakṣita’s method, see McClintock 2003, 2010, 85–91; on this same concept applied to the work of Dharmakīrti, see Dreyfus 1997 and Dunne 2004, 53–79. Blumenthal (2004, 43, 44, 46) articulates this same approach of Śāntarakṣita’s in terms of a “graded ascent of philosophical views,” “multiple levels or stages of provisionality,” and an “ascent through provisional views.” McClintock (2003, 68–76) points to TS 1916–17 as instances where Śāntarakṣita acknowledges the superiority of Madhyamaka. Śāntarakṣita additionally nods to the Madhyamaka tradition in the framing of the text, mirroring Nāgārjuna’s opening to his Fundamental Verses on the Middle Way (Mūlamadhyamakakārikā) in laying out the purpose and structure of the Compendium in its opening stanzas. (i) Vaiśeṣikas defend the conjoined view, (ii) Vaibhāṣikas such as Saṅghabhadra (c. fourth–fifth century) defend the interceding space view, and (iii) Vasubandhu defends the spatially continuous view in the Commentary on the Treasury of Abhidharma (Abhidharmakośabhāṣya) ad 1.43d2, where 476 Śāntaraks∙ita 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 he attributes the position to the Kaśmīri Vaibhāṣikas such as Vasumitra (c. second century CE); Śubhagupta (c. 720–780) also defends this view in his Proof of External Objects (Bāhyārthasiddhi) 50, 52, 53, 56, although Śāntarakṣita cites Śubhagupta in connection to the second view. In his autocommentary, Śāntarakṣita goes on to praise Yogācāra for its theory being confirmed and known by a source of knowledge ( pramāṇa), for serving as a corrective for mistaken views, for its rejection of material simples, and for its agreement with scriptural sources (MAV ad MA 45; Ichigō 1985, 124). MA 46–51; see Dharmakīrti’s Explanation of the Sources of Knowledge (Pramāṇavārttika) 3.209–22 and Śrīgupta’s TA 4. MA 52–60; see TA 5–6, TAV transitional stanzas (antaraśloka) 1–4. In Cartesian terms, the question here concerns the formal reality of thoughts themselves rather than the objective reality of whatever might be represented in thought. *Satyākāravāda (rnam pa bden par smra ba) and *Alīkākāravāda (rnam pa brdzun par smra ba) (literally “theory/proponent of real representations” and “theory/proponent of unreal representations”) are not attested in extant Indic doxographies, where we instead find the Sākāravāda-Nirākāravāda distinction. While these labels were imposed onto diverse sets of thinkers in contriving subschools of Yogācāra, they are nevertheless useful for clarifying the structure of Śāntarakṣita’s argument and the dialectical lay of the land as he understood it. One should be careful to distinguish the use of the terms sākāra vs. nirākāra in the Yogācāra context from the use of this same pair of terms to designate, respectively, representationalist vs. direct realist accounts of ordinary perception among realists about external objects. To the contrary, Berkeley (Principles in Works vol. 2, 98) and Hume (Treatise 1.2.4), for instance, both argue for theories of a minima sensibilia, a kind of phenomenal atomism on which a perception is reducible to indivisible, unextended simples. This account can be traced back to Nāgārjuna, who argues that nothing lays claim to the Abhidharma ontological category of ultimate reality ( paramārthasat), or substantially reality (dravyasat). TA 11; see Aitken (Forthcoming) and (2021a) for an interpretation of Śrīgupta’s version of this threefold criterion. For a comparison of Śrīgupta and Śāntarakṣita on the two truths, see Aitken (2021b). See, for example, Kamalaśīla’s MAP ad MA 64, Haribhadra’s Illuminating the Ornament of Realization (Abhisamayālaṃkārālokā) (Wogihara 1932–35, 594.18–25), Bhāviveka’s Compendium on the Meaning of the Middle Way (Madhyamakārthasaṃgraha) 9–11 and Jeweled Lamp for the Middle Way (Madhyamakaratnapradīpa) 1.4, and Atiśa’s Introduction to the Two Truths (Satyadvayāvatāra) 4. Jñānagarbha also sets out versions of these three criteria (SDV 8, 12, and 21). Explanation of the Sources of Knowledge 3.3ab and Essence of Reasoning (Nyāyabindu) 1.15. Śāntarakṣita cites Nāgārjuna’s Fundamental Verses on the Middle Way 24.18 in support of this criterion. Importantly, Śāntarakṣita does not regard this as an innovation, citing Nāgārjuna’s Sixty Stanzas on Reasoning (Yuktiṣaṣṭika) 21 and 34 (among other sources) in support of this presentation. For example, Garfield (2016) argues that Jamgön Ju Mipham Gyatso (’Jam mgon ’Ju Mi pham rgya mtsho, 1846–1912) reads Śāntarakṣita’s Yogācāra phenomenologically. See Saccone’s 2018 edition, translation, and analysis of this chapter. It is this second, metaphysical argument that follows Vasubandhu’s previously discussed anti-atomist argument from his Twenty Verses 11–15. It is worth pointing out that Śāntarakṣita’s rejection of atoms does not in principle preclude the possibility of conventionally real external objects that are not founded in atoms, though he does not explicitly consider such a scenario, presumably owing to the fact that all external world realists in his intellectual milieu were atomists. To the contrary, Berkeley famously makes a case for subjective idealism as part of his project to restore commonsense. This difference is also reflected in the different examples cited in the two inferences (in which the entailment relation is instantiated), with the Madhyamaka example being a reflection, which lacks an intrinsic nature, and the Yogācāra example being a lotus growing in the sky, which is a nonexistent thing. Since atoms are stipulated by their proponents as partless, and therefore true unities by definition (TS 1992ab), the rejection of the unity of atoms would necessarily preclude their existence. The problem here lies in the sweeping nihilistic implications of the generalized Yogācāra entailment relation when applied to the Madhyamaka all-inclusive subject. 477 Allison Aitken 33 Although Śāntarakṣita’s conciliatory approach to Yogācāra was rejected by many later Tibetan Mādhyamikas in favor of Candrakīrti’s account of conventional truth, his synthesis of Madhyamaka with Dignāga and Dharmakīrti’s logic and epistemology remains influential in Tibetan Madhyamaka cutting across traditions. Bibliography Aitken, Allison. Forthcoming. Introduction to Reality: Śrīgupta’s Tattvāvatāravṛtti. Harvard Oriental Series. Cambridge, MA: Harvard University Press. ———. 2021a. “No Unity, No Problem: Madhyamaka Metaphysical Indefinitism.” Philosophers’ Imprint 21 (31): 1–24. ———. 2021b. “The Truth About Śrīgupta’s Two Truths: Longchen Rabjampa’s ‘Lower Svātantrikas’ and the Making of a New Philosophical School.” Journal of South Asian Intellectual History 3 (2): 185–225. Blumenthal, James. 2004. The Ornament of the Middle Way: A Study of the Madhyamaka Thought of Śāntarakṣita. Ithaca: Snow Lion Publications. ———. 2009a. “Dynamic and Syncretic Dimensions to Śāntarakṣita’s Presentation of the Two Truths.” Asian Philosophy 19 (1): 51–62. ———. 2009b. “The ‘Neither-One-Nor-Many’ Argument of Śāntarakṣita: A Classical Buddhist Argument on the Ontological Status of Phenomena.” In Buddhist Philosophy: Essential Writings, edited by William Edelglass and Jay L. Garfield, 46–60. New York: Oxford University Press. ———. 2015. “Two Topics Concerning Consciousness in Śāntarakṣita’s Yogācāra-Madhyamaka Syncretism.” In Madhyamaka and Yogācāra: Allies or Rivals?, edited by Jay L. Garfield and Jan Westerhoff, 242–51. New York: Oxford University Press. Dreyfus, Georges. 1997. Recognizing Reality: Dharmakīrti’s Philosophy and Its Tibetan Interpretations. Albany: State University of New York Press. Dreyfus, Georges, and Sara McClintock. 2003. The Svātantrika-Prāsaṅgika Distinction: What Difference Does a Difference Make? Boston, MA: Wisdom Publications. Dunne, John. 2004. Foundations of Dharmakīrti’s Philosophy. Boston: Wisdom Publications. Eckel, Malcolm David. 1987. Jñānagarbha on the Two Truths: An Eighth Century Handbook of Madhyamaka Philosophy. Albany, NY: State University of New York Press. Eltschinger, Vincent. 2007. “On 7th and 8th Century Buddhist Accounts of Human Action, Practical Rationality and Soteriology.” In Pramāṇakīrtiḥ: Papers Dedicated to Ernst Steinkellner on the Occasion of His 70th Birthday, Part 1, edited by Birgit Kellner, Horst Lasic, M. T. Much, and Helmut Tauscher, 135–62. Wien: Arbeitskreis für Tibetische und Buddhistische Studien Universität Wien. ———. 2019. “Śāntarakṣita.” In Brill’s Encyclopedia of Buddhism, Volume II: Lives, edited by Jonathan A. Silk, Richard Bowring, Vincent Eltschinger, and Michael Radich, 383–90. Leiden and Boston: Brill. Frauwallner, Erich. 1961. “Landmarks in the History of Indian Logic.” Wiener Zeitschrift für die Kunde Süd- und Ostasiens 5: 125–48. Garfield, Jay L. 2016. “Cittamātra as Conventional Truth from Śāntarakṣita to Mipham.” Journal of Buddhist Philosophy 2: 263–80. Ichigō, Masamichi, ed. 1985. Madhyamakālaṃkāra of Śāntarakṣita with His Own Commentary or Vṛtti and with the Subcommentary or Pañjikā of Kamalaśīla. Kyoto: Buneido. ———. 1989. “Śāntarakṣita’s Madhyamakālaṃkāra.” In Studies in the Literature of the Great Vehicle: Three Mahāyāna Buddhist Texts, edited by Luis O. Goméz and Jonathan A. Silk, 141–240. Ann Arbor: Center for South and Southeast Asian Studies, University of Michigan. ———. 2016. “Śāntarakṣita on Satyadvaya.” Journal of Indian and Tibetan Studies 20: 166–94. Jha, Ganganatha. 1937–1939. The Tattvasaṅgraha of Śāntarakṣita with the Commentary of Kamalaśīla. 2 vols. Baroda: Oriental Institute. Kapstein, Matthew. 2001. Reason’s Traces: Identity and Interpretation in Indian and Tibetan Buddhist Thought. Boston: Wisdom Publications. ———. 2009. “Śāntarakṣita’s Tattvasaṃgraha. A Buddhist Critique of the Nyāya View of the Self.” In Buddhist Philosophy. Essential Readings, edited by William Edelglass and Jay L. Garfield, 320–33. Oxford: Oxford University Press. Keira, Ryusei. 2004. Mādhyamika and Epistemology: A Study of Kamalaśīla’s Method for Proving the Voidness of All Dharmas: Introduction, Annotated Translations and Tibetan Texts of Selected Sections of the Second Chapter of the Madhyamakāloka. Wiener Studien zur Tibetologie und Buddhismuskunde, Heft 59. Wien: Arbeitskreis für Tibetische und Buddhistische Studien Universität Wien. 478 Śāntaraks∙ita McClintock, Sara L. 2003. “The Role of the ‘Given’ in the Classification of Śāntarakṣita and Kamalaśīla as Svātantrika-Mādhyamikas.” In The Svātantrika-Prāsaṅgika Distinction: What Difference Does a Difference Make? edited by Georges B. J. Dreyfus and Sara L. McClintock, 125–71. Boston: Wisdom Publications. ———. 2010. Omniscience and the Rhetoric of Reason: Śāntarakṣita and Kamalaśīla on Rationality, Argumentation, and Religious Authority. Boston: Wisdom Publications. ———. 2013. “Kamalaśīla and Śāntarakṣita on Scripture and Reason – The Limits and Extent of ‘Practical Rationality’ in the Tattvasaṃgraha and Pañjikā.” In Scriptural Authority, Reason, and Action: Proceedings of a Panel at the 14th World Sanskrit Conference, Kyoto, September 1st-5th 2009, edited by Vincent Eltschinger and Helmut Krasser, 209–38. Wien: Verlag der Österreichischen Akademie der Wissenschaften. Obermiller, Eugene. 1932. The History of Buddhism in India and Tibet by Bu Ston: Part II. Heidelberg: Harrassowitz. Roerich, George. 1949. The Blue Annals, Part I. Calcutta: Royal Asiatic Society of Bengal. Saccone, Margherita Serena. 2018. On the Nature of Things: A Buddhist Debate on Cognitions and Their Object. Vienna: Arbeitskreis für tibetische und buddhistische Studien Universität Wien. Sāṅkṛtyāyana, Rāhula. 1935/1936. “Dharmakīrti’s Vādanyāya. With the Commentary of Śāntarakṣita.” Journal of the Bihar and Orissa Research Society: 21–22. Seyfort Ruegg, David. 1981. The Literature of the Madhyamaka School of Philosophy in India. Wiesbaden: Otto Harrassowitz. ———. 2006. “The Svātantrika-Prāsaṅgika Distinction in the History of Madhyamaka Thought.” IndoIranian Journal 49 (3/4): 319–46. Sferra, Francesco. 2004. “Śāntarakṣita: Vātsīputrīyaparikalpitātmaparīkṣa, Disamina del se immaginato dai seguaci di Vātsīputra (Tattvasaṃgraha, 336–49 con il commento [Pañjikā] di Kamalaśīla).” In La Rivelazione del Buddha, vol. II, edited by Raniero Gnoli, 1155–74. Milan: Arnoldo Mondadori. Shastri, Dwarka Das. 1968. Tattvasaṅgraha of Ācārya Shāntarakṣita with the Commentary Pañjikā of Shri Kamalashīla. 2 vols. Varanasi: Bauddha Bharati. Steinkellner, Ernst. 1999. “Yogic Cognition, Tantric Goal, and Other Methodological Applications of Dharmakīrti’s kāryānumāna Theorem.” In Dharmakīrti’s Thought and Its Impact on Indian and Tibetan Philosophy, Proceedings of the Third International Dharmakīrti Conference, Hiroshima, November 4–6, 1997, edited by Shoryu Katsura, 349–62. Vienna: Austrian Academy of Sciences Press. 2001. “Is the Ultimate Cognition of the Yogin Conceptual or Non-Conceptual? Part I: A Critical Edition of the Tantric Tattvasiddhi, Final Section.” In Le Parole e i marmi: Studi in onore di Raniero Gnoli nel suo 70° compleanno, edited by Raffaele Torella, 835–52. Rome: Istituto Italiano per l’Africa e l’Oriente. ———. 2014. The Edition of Śāntarakṣita’s Vādanyāyaṭīkā Collated with the Kundeling Manuscript. Vienna: Arbeitskreis für tibetische und buddhistische Studien Universität Wien. Steinkellner, Ernst, and M. T. Much. 1995. Texte der erkennt-nistheoretischen Schule des Buddhismus: systematische Übersicht über die buddhistische Sanskrit-Literatur II. Göttingen: Vandenhoeck and Ruprecht. Tillemans, Tom T. F. 1982. “The ‘Neither One Nor Many’ Argument for śūnyatā and Its Tibetan Interpretations: Background Information and Source Materials.” Études de Lettres (University of Lausanne) 3: 103–28. ———. 1983. “The ‘Neither One Nor Many’ Argument for śūnyatā and Its Tibetan Interpretations.” In Contributions on Tibetan and Buddhist Religion and Philosophy, edited by Ernst Steinkellner and Helmut Tauscher, 305–20. Vienna: Arbeitskreis für Tibetische and Buddhistische Studien. ———. 1984. “Two Tibetan Texts on the ‘Neither One Nor Many’ Argument for Śūnyatā.” Journal of Indian Philosophy 12: 357–88. ———. 2011. “How Far Can a Mādhyamika Reform Conventional Truth? Dismal Relativism, Fictionalism, Easy-Easy Truth, and the Alternatives.” In Moonshadows: Conventional Truth in Buddhist Philosophy, edited by the Cowherds, 151–66. New York: Oxford University Press. ———. 2016. How do Mādhyamikas Think? And Other Essays on the Buddhist Philosophy of the Middle. Somerville, MA: Wisdom Publications. Wangdu, Pasang, and Hildegard Diemberger. 2000. dBa’ bzhed: The Royal Narrative concerning the bringing of the Buddha’s Doctrine to Tibet. Vienna: Austrian Academy of Sciences Press. Westerhoff, Jan. 2015. “Nāgārjuna’s Yogācāra.” In Madhyamaka and Yogācāra: Allies or Rivals? edited by Jay L. Garfield and Jan Westerhoff, 165–83. New York: Oxford University Press. Wogihara, Unrai, ed. 1932–1935. Abhisamayālaṃkārāloka Prajñāpāramitāvyākhyā (Commentary on Aṣṭasāhasrikā-Prajñāpāramitā) by Haribhadra Together with the Text Commented On. Tokyo: The Toyo Bunko. 479