Buddhist philosophy has been evolving for some 2,500 years, in India, China, Tibet, Japan, amongst other Asian countries, and continues to do so as it moves into the West. In this time it has acquired complexities and subtleties equal to any philosophical tradition in the West – along with numerous different schools and the in-house (and out-of-house) debates these bring. In this volume, Jan Westerhoff brings us an account of what he terms – with some justice – the Golden Age of Indian Buddhism, starting with the rise of the Abhidharma schools about the third century bce, and ending with developments in metaphysics and epistemology about the sixth century ce.
Westerhoff's knowledge of the subject, his scholarship, and his philosophical acumen are patent. I'd be hesitant to recommend the book to someone new to the subject, for fear that they would be overwhelmed by the wealth of detail. A novice might be better off reading Siderits (Buddhism as Philosophy, 2007) or Carpenter (Indian Buddhist Philosophy, 2014). However, I do strongly recommend this book for anyone who has a basic knowledge of Buddhist philosophy, and who wishes to engage in the subtleties of its history and development. They will surely learn much from it.
So, unlike many histories of Buddhist thought, Westerhoff's book skips over the early sūtra period of Buddhism (the Four Noble Truths, etc.). There are also only scattered remarks concerning Indian Buddhist philosophy after the sixth century (and thinkers such as Śāntarakṣita and Śāntideva). It also focuses heavily on Buddhist metaphysics and epistemology: its comments and analysis of Buddhist ethics are very occasional. (Those who want more of ethical matters could consult Carpenter, Indian Buddhist Philosophy.) Within this framework, it takes us through Buddhist thinkers, schools, and their philosophical ideas, providing a sympathetic but critical analysis of these. (It also recounts details of the tradition's mythology and legends – though Westerhoff declines to call them such out of deference to the tradition.)
The book starts with an analytic table of contents, and a very helpful timeline of the thinkers and schools that feature in the book. The Introduction then explains the parameters within which Westerhoff has chosen to frame his account. Chapter 1 takes us into the details of the Abhidharma schools which arose in pre-Mahāyāna Buddhism, such as Pudgalavāda, Sarvāstivāda, and Sautrāntika. This requires a good bit of historical reconstruction, since many of the original texts are lost, and things have to be inferred from what others said about them. Topics mereological and temporal come in for detailed discussion.
Chapter 2 deals with the Mahāyāna-generating Prajñapārāmetā Sūtras, which first appeared around the turn of the Common Era. It analyses the appearance of the Madhyama school in the work of Nāgārjuna and his Mūlamadhyamakakārikā, and the subsequent commentarial tradition to which this gave rise. Much attention is paid to the central metaphysical concept of emptiness and its ramifications.
Chapter 3 moves to the other school of Indian Mahāyāna, Yogācāra. Sūtras such as the Lankāvatāra Sūtra are discussed. The chapter then turns to the philosophical positions of Asaṅga and Vasubandhu. Philosophical notions such as root consciousness (ālayavjñāna) and the three natures (trisvabhāva) are expounded and analysed. The connections between these ideas and both meditative practices and some non-Buddhist schools, such as Advaita Vendānta, come in for scrutiny.
The final substantial chapter, chapter 4, looks at the metaphysics and epistemology of the sixth-century philosophers, Diṅnāga and, especially, Dharmakīrti. Much attention is paid to the ambiguous relation of the thought of Dharmakīrti to previous schools of Indian Buddhism. There is then a brief discussion which contrasts his views on language and epistemology with the Hindu Mīmāmṣā School. A brief concluding chapter draws some threads of the book together.
In the rest of this review I will discuss just two of the philosophical issues that appear in Westerhoff's narrative. The first concerns inconsistency in Nāgārjuna's Mūlamadhyamakakārikā (MMK). Westerhoff notes, as have many previous commentators, that the Prajñapārāmetā Sūtras abound in contradictions. It is therefore unsurprising to find them appear in the MMK.
In his discussion of the way in which contradictions figure in the MMK, Westerhoff notes that one case is when Nāgārjuna deploys the catuṣkoṭi. Applying this early Buddhist trope, Nāgārjuna considers the possibility that certain things are true, false, both, or neither. Since this is in the context of a fourfold reductio, and Nāgārjuna rejects all of these possibilities, however, this provides no reason to suppose that he accepts contradictions.
More thorny are places where Nāgārjuna does appear to endorse explicit contradictions. Westerhoff notes that such contradictions may be defused by appealing to the standard Buddhist distinction between conventional and ultimate truth. So a statement, A, may be taken as true conventionally, but false ultimately. He worries that this relegates much Buddhist doctrine to irrelevance. There is, however, a much greater worry for a consistent reading of the text. Though Nāgārjuna endorses the thought that there are two truths, he also says (MMK, XXV: 19–20), notoriously, that there is not the slightest difference between saṃasāra and nirvāṅa. Since saṃasāra is how things are conventionally, and nirvāṅa is how things are ultimately, this implies that there is no difference between conventional and ultimate reality. So if A is conventionally true, it is also ultimately true; so, then, A is ultimately true and false. Nor will it help to say that when Nāgārjuna says that there is no difference between the two truths, he just means that there is no ultimate difference between the two truths. (There may yet be a conventional difference.) For then, since it is ultimately true that there is no difference between conventional truth and ultimate truth, it is ultimately true that A has a contradictory status.
There is yet another worry about a consistent reading of the MMK, which Westerhoff does not consider. Nāgārjuna appears to tell us (e.g. MMK, XVIII: 9, and elsewhere) that ultimate reality is free from mental fabrications. That is, it cannot be characterized by concepts. It is, hence, ineffable. As is clear from even a cursory inspection of the MMK, however, Nāgārjuna says much about it. The text, then, is implicitly committed to contradiction. Moreover, since the contradiction is so obvious, one can hardly claim that Nāgārjuna failed to notice it! (For more on all this, see Priest, The Fifth Corner of Four (2018), part 2.)
Westerhoff, as do some interpreters of Madhyamaka, holds that it is not committed to an ineffable ultimate reality; indeed, that this is the ‘crucial divide’ (212) between Madhyamaka and Yogācāra, which is so committed (210). When a Mādhyamika says that one cannot say anything about ultimate reality, they are simply rejecting the claim that there is an ultimate reality to be talked about (211). As hardly needs to be said, however, this is hardly compatible with Nāgārjuna's explicit statement that there are two realities, a conventional and an ultimate (MMK, XXIV: 8–10).
I now turn to the second matter. This concerns Westerhoff's account of Dharmakīrti's graded analysis of Buddhist doctrines. Dharmakīrti appears to adopt a progressive critique of metaphysical systems, ending with the one he takes to be correct. Four systems are at issue. I: A naive realism, according to which reality contains common-sense objects, composed of parts. II: A Sarvāstivāda view, according to which the ultimate parts (dharmas) are real enough, but their mereological wholes are mere conceptual constructions. III: A Sautrāntika view, according to which the ultimate parts (svālakṣaṇas) are ineffable, and all ways of categorizing them (even spatial and temporal) are mere conceptual impositions. IV: A Yogācāra view, which dispenses with material objects altogether; such things are all conceptual illusions.
Westerhoff suggests that it is the same sort of argument which moves one through this progression. This is the neither-same-nor-different (NSND) argument. (He refers to this, as is standard, as the neither-one-nor-many argument. However, this seems inappropriate, since only in the first case is a plurality of objects at issue.) Criticizing I, it is argued that a mereological whole can be neither the same as nor different from its parts. This motivates taking the whole to be non-existent, moving us to II. Criticizing II, it is argued that the F-ness of an object can be neither the same as nor different from the object. This motivates eliminating any categorized object as non-existent, taking us to III. Criticizing III, it is argued that the subject who cognizes an object can be neither the same as nor different from the object cognized. This motivates eliminating the objects altogether, leaving us with Yogācāra idealism.
This is a fascinating conceptual analysis of the progression, and may well be textually correct. However, the cogency of the NSND argument is highly suspect. The first step argues that an object cannot be identical with its parts, since the object is a single thing, whilst the plurality is a multitude. This obviously appeals to a version of Leibniz's Law: objects with different properties are not identical. There has recently been a substantial literature on how one should deal with plural objects/predication/quantification. Whilst it is true that there is still, perhaps, no consensus on the issue, it is usually agreed that if the relation between an object and its parts is one of identity, one cannot endorse Leibniz's Law. The Law applies only when the identity is between two objects or between two pluralities. (See e.g. Priest, One (2014), 6.10, 6.11.)
In the other two cases, the non-identity wing of the argument is less problematic. Most would be happy to grant that Socrates’ snub-nosedness is not the same as Socrates himself; and that when I see a cat, I am not the same as the cat that I see. However, in all cases, there is a problem with the other wing of the argument. In each case, this is of the form that the things in question cannot be distinct because one can never find the one without the other. This form of argument appears simply fallacious. One can never find a north pole of a magnet without a south pole, and vice versa. Or again, one can never find a husband without a wife (or however one wants to express this in gender-neutral terms), and vice versa. North pole and south pole, husband and wife are, nonetheless, distinct. (Siderits, Buddhism as Philosophy, 6.1, has a more extended discussion of the NSND argument, which exposes other problems, though here is not the place to go into them.)
Westerhoff suggests, further (258–259), that the NSND argument may locate a crucial difference between Madhyamaka and Yogācāra. A Mādhyamika can apply the argument in an indefinite process of analysis. For a Yogācārin, either stage IV provides the final analysis of matters, giving an ultimate reality of consciousness; or, if this is not the ultimate stage then ‘this must be replaced by something that analyses the mental away in the same manner as the mental analysed the particular away’ (though Westerhoff does not discuss how this might be executed), and ‘any stage [GP: my italics] after the idealist stage will be non-dual and ineffable’ (though, again, why so is not discussed). Neither possibility, says Westerhoff, is acceptable to Madhyamaka, however. Now, it is true that a Mādhyamika can apply the NSND argument to any object and the things on which it depends. Indeed, Śāntarakṣita does so. (See Blumenthal, ‘Śantarakṣita’, in Stanford Encyclopedia of Philosophy (2018), 1.2.) However, this does not show that there is no ultimate level, merely that it is as empty as anything else (the ‘emptiness of emptiness’). Moreover, as I have already noted, Madhyamaka seems stuck with ineffability at the ultimate level. If this is so, then there may not be the contrast between Madhyamaka and Yogācāra that Westerhoff suggests.
In the second part of this review I have addressed two of the philosophical issues that Westerhoff's acute exegesis and analysis brings to the surface. These are but two of many. Westerhoff's book is a rich and significant contribution both to Buddhist scholarship in the West and to its philosophical richness.
Many thanks go to Westerhoff for correspondence, which much improved this review.
