Tom J.F. Tillemans
Āryadeva (Tibetan: ‘phags pa lha; Chinese: di po 提婆, sheng tian 聖天) is a second-third century C.E. Indian Buddhist author who, along with his teacher Nāgārjuna, is considered the co-founder of the school of thought known as the “Philosophy of the Middle” (madhyamaka). This Great Vehicle (mahāyāna) philosophy is characterized by its use of dialectics to reject any attribution of intrinsic natures (svabhāva) to things, an intrinsic nature being a type of aseity, what a thing is in itself completely independently of other factors. And while possessing an intrinsic nature is thought by most Indian thinkers to be a necessary condition for anything to be a fully real entity (bhāva), Āryadeva, Nāgārjuna and their followers, the Mādhyamikas, held that all things are what they are only dependently. The conclusion: things are “empty” (śūnya) of intrinsic natures; it is impossible for there to be fully real entities and impossible to ascribe any fully real properties to them. Like Nāgārjuna, Āryadeva is thus not a metaphysical realist (Tib. dngos smra ba = *bhāvavādin); his thought is in sharp contrast with that of his Indian opponents and most of his coreligionists. I will take up his philosophical perspectives on metaphysics and ethics in more detail below.
Before going further, we need a modicum of clarity on historical and textual matters. I thus naturally begin with what we know of his life. Following Candrakīrti (sixth century C.E.), one of Āryadeva’s Indian commentators, Āryadeva was born in Siṃhaladvīpa as a prince. He subsequently traveled from Siṃhaladvīpa—likely Śrī Laṅka— to South India and became a disciple of the great Mādhyamika thinker, Nāgārjuna, the author of “the sixfold corpus of [[[Madhyamaka]]] reasoning” (Tib. rigs tshogs drug), including the Verses on the Philosophy of the Middle (madhyamakakārikā). After supposedly founding monasteries in South India, he moved to the Northern Indian monastery of Nālandā. His being a native of Śrī Laṅka may possibly be confirmed by references in the Ceylonese chronicles Dīpavaṃsa and Mahāvaṃsa to a “Deva” who lived in the second half of the third century at the time when the Indian Vetullavāda sect of Great Vehicle Buddhism was temporarily implanted in Śrī Laṅka (Seyfort Ruegg 1981, 50).
All accounts concord on his skill in defending Buddhism from non-Buddhist thinkers. And indeed, his major work, The Four Hundred Verses (catuḥśataka), does show a very strong familiarity with non-Buddhist doctrines of the time, especially those of Brahmanical schools such as the Sāṃkhya and Vaiśeṣika, which it subjects to trenchant criticism. In a famous debate, supposedly with a Mātṛceta or Durdharṣakāla—whom the Tibetan historian Tāranātha implausibly claims to become the Buddhist poet Aśvaghoṣa (Chimpa and Chatopadhyaya 1970, 131-132) —he supposedly outdid his opponent with dialectics, mantras, and supernatural feats, converting him finally to Buddhism (see Cabezón 2008, 75-78). The Chinese biography goes a step further: Āryadeva ended his days murdered by the disciples of a non-Buddhist philosopher whom he had defeated in debate (Taishō 2048, Robinson 1967, 27-8, Lang 1986, 7). As is frequent in hagiographical versions of Indian debates, we learn about the buildup and the denouement, the magical tricks and the perfidy, but relatively little of philosophical content (Cabezón 2008, 89-90).
The philosophies one attributes to a second or third century thinker will significantly differ depending upon the texts one regards as authentic. And such questions of textual authenticity are generally not simple matters. Various works, besides the Four Hundred Verses, have been attributed to Āryadeva, especially in the Tibetan and Chinese canons (see Tillemans 1990, 6-7). In fact, there can be little doubt that the name “Āryadeva” was widely applied to authors of works that could not have been written by the second or third century author of the Four Hundred Verses. A case in point is the Destruction of Errors about Madhyamaka (*madhyamakabhramaghāta), assigned to an Āryadeva by Tibetans: this text copiously borrows from the Verses on the Heart of Madhyamaka (madhyamakahṛdayakārikā) and Torch of Dialectics (tarkajvālā) of Bhāviveka, a celebrated Mādhyamika who lived in the sixth century (i.e., 500-570 C.E.).
Another such case is the Compendium on the Essence of Knowledge (jñānasārasamuccaya), a text which the Tibetan canon ascribes to Āryadeva, but which gives the fourfold presentation of Buddhist doctrine typical of the doxographical (siddhānta) literature, a genre which considerably post-dates the third century (transl. Mimaki 1976). Less clear are the Hand Treatise (hastavālaprakaraṇa) and its Commentary (hastavālaprakaraṇavṛtti), both of which the Tibetan canon attributes to Āryadeva, although the Chinese canon most likely has it right in attributing them to the fifth century author Dignāga (transl. Thomas and Ui 1918). Finally, the Chinese canon has five works attributed to Āryadeva, the longest two of which are known as “Hundred Verse Treatises”: Taishō 1570, the Extensive Hundred Verse Treatise (guang bai lun 廣百論), is the last half of the Four Hundred Verses (catuḥśataka); Taishō 1569, the Hundred Verse Treatise (bai lun 百論 = *śata(ka)śāstra; transl. Tucci 1929), bears many thematic affinities with the Four Hundred Verses, but is nonetheless a different work and not
simply an extract of the Four Hundred. The Hundred Verse Treatise became a key text of the Chinese Madhyamaka school known as the “Three Treatise” (san lun 三論) tradition (May 1979; May and Mimaki 1979). If the identification of the author Qing mu 青目 with Āryadeva should be correct, then the author of the Four Hundred Verses may have also written a commentary on Nāgārjuna’s Verses on the Philosophy of the Middle (May 1979, 481; Tillemans 1990, 7, fn. 17). We shall, in what follows, examine aspects of the thought of the author of the Four Hundred Verses. I consider this treatise—and, to a much lesser degree, the Hundred Verse Treatise—as the best representative of Āryadeva’s Madhyamaka, although it may well be that at least some other texts that critique non-Buddhist systems, e.g., the Hundred Letter Treatise (akṣaraśataka; transl. Gokhale 1930), are by the same historical individual too. Tibetan tradition identifies the major Mādhyamika authors— i.e., Nāgārjuna, Āryadeva and Candrakīrti— with later tantrikas having the same namesake. In particular, the Mādhyamika Nāgārjuna is
identified with tantric Nāgārjunapāda, also known as Ārya Nāgārjuna (c. seventh century C.E.); Āryadeva is said to be the disciple of Ārya Nāgārjuna and the author of the Lamp that Integrates the Practices (caryāmelāpakapradīpa; transl. Wedemeyer 2007), which furthers the Ārya (noble) tradition of exegesis on the Tantra of the Esoteric Community (guhyasamājatantra); he is also said to be the author of the Treatise on the Purity of Mind (cittaviśuddhiprakaraṇa) connected with Ārya Nāgārjuna’s Five Stages (pañcakrama). To be clear, however, one cannot follow the Tibetan tradition in attributing tantric texts to the same Mādhyamika Āryadeva; hence, one badly needs a differentiation between at least two individuals, Āryadeva I being the Mādhyamika and Āryadeva II a much later tantric author (Mimaki 1987; Tillemans 1990, 6-7).
1.3 Translations and editions of the Four Hundred Verses
Here are the translations and editions of the text I am taking to best represent the thought of Āryadeva I. The Four Hundred Verses has sixteen chapters, each with twenty-five verses. The work has been translated into English in Lang 1986, who uses the Sanskrit fragments and the Tibetan; Sonam 2008 translates the Four Hundred Verses and rGyal tshab rje’s commentary from the Tibetan; chapters nine to sixteen are translated into Italian in Tucci 1925 based on the Sanskrit fragments, Tibetan, and Chinese (Taishō 1570). Lang 2003 introduces and translates the first four chapters along with Candrakīrti’s commentary; May 1980-1982 provide a French translation of chapter nine of the Four Hundred Verses and Candrakīrti’s commentary; Tillemans 1990 contains a translation of chapters twelve and thirteen along with Candrakīrti’s and Dharmapāla’s commentaries; Johnson 2012 translates chapter fifteen and Candrakīrti’s commentary. An edition of the Sanskrit fragments and their corresponding Tibetan translation is found in Suzuki 1994, following upon the works of Haraprasād Shāstrī 1914 and Bhattacharya 1931.
1.4 Overview of the Four Hundred Verses
Before going further, one also needs a whistle-stop tour of the content of the Four Hundred Verses. Chapters one to eight discuss ethical themes, many of which are common to all Buddhist schools, whether of the Great Vehicle or of the so-called “Lesser Vehicle” (hīnayāna). The first four chapters, for example, criticize the four basic illusions (viparyāsa) governing our psyches, viz., our mistakenly taking transitory life as permanent, what is painful as pleasurable, what is dirty as clean, and what is selfless as having a self. The next chapters deal respectively with the bodhisattva’s practices leading to enlightenment, the elimination of the passions (kleśa) that impede such practices, and the elimination of attachment to the objects of the senses; the eighth chapter deals with the proper practices of Buddhist disciples.
It is in the latter half of the Four Hundred Verses that we find the detailed refutations of various ontologies, Buddhist and non-Buddhist, as well as instruction on methodological principles for Madhyamaka thought. Chapters nine, ten, and eleven refute respectively the reality of permanent entities (nitya), personal selves (ātman), and time (kāla); chapter twelve refutes “heretical views” (dṛṣṭi), i.e., inter alia, non-Buddhist views on liberation, scripture, asceticism, and the importance of high caste birth; chapter thirteen refutes Buddhist and non-Buddhist positions on the reality of the sense faculties and their objects (indriyārtha); chapter fourteen refutes man’s deep-seated thinking in terms of dichotomies, or “extremes” (antagrāha); chapter fifteen refutes the reality of conditioned things (saṃskṛtārtha); chapter sixteen, entitled “Cultivating understanding in the master and disciple” (slob dpon dang slob ma rnam par gtan la dbab pa bsgom pa), is on Madhyamaka method and logical issues that arise in executing that method. In what follows, I will first seek to better understand Āryadeva’s philosophical orientation and method in the latter half of the Four Hundred Verses and then take up the first half on ethics and religion; I close with the question of the systemic coherence of the work.
2. Quietism about ontology
Āryadeva’s orientation in the last chapters of the Four Hundred Verses and the Hundred Verse Treatise can, in my opinion, best be described as quietistic. He neither advances nor defends any theses that are ontologically committed, i.e., presuppose acceptance of fully real entities (bhāva). This includes positive statements as well as negations and all conjunctions or disjunctions of the two. In having no theses and no debates about them he closely follows Nāgārjuna, who had famously said in verse 50 of his Sixty Verses on Reasoning (yuktiṣaṣṭikā):
Now, it might be objected that quietism is, in one way or another, a relatively widespread theme in Buddhism, Madhyamaka or not. There are, for example, a variety of meditative techniques to arrive at an irenic state beyond conceptual thought, at nirvāṇa, a state of peace, or the “silence of the noble ones” (āryatūṣṇīṃbhāva). One finds some form of quietism in the Pāli canon, Tantra, and Zen. True, but the particularity of the Mādhyamikas, including Āryadeva, is that they emphasize philosophical analysis as a method leading to that quietening of thought, diagnosing rationally and with dialectics where it goes wrong in making ontologically committed truth claims and debating about them. That is why it should be of special interest to philosophy.
In fact, however, it is inadequately recognized in much modern scholarship that both Nāgārjuna and Āryadeva discuss and promote two markedly different ways in which a Mādhyamika philosopher does not have theses, that is to say, one way that does not clearly concern the propriety or legitimacy of philosophical debate and another that certainly does. They are follows: (1) the Mādhyamika does not have a thesis, because for him there is not anything fully real, including thesis-statements. (2) the Mādhyamika does not have a thesis, because he does not make or personally endorse truth claims presupposing anything fully real, all such claims being refutable. It is the second sense of “no thesis” which most clearly leads to quietism, as it embodies the normative stance that one should not be committed to the content of theses-statements nor debate about that content. The second also lends itself to a semantic interpretation of the no-thesis stance, viz., that a Mādhyamika does not accept or debate about any theses whose terms range over fully real entities (bhāva).
Although it seems to have been important in Mādhyamikas’ debates with their Brahmanical contemporaries, what I am calling the first sense does not turn on the general stance that all propositions are somehow refutable. Nor is it about the impossibility of making claims via statements with a realist semantics. It turns on the existence or nonexistence of just one statement, viz., the Mādhyamika’s own fundamental principle that nothing has any intrinsic nature (svabhāva). Let’s look at the textual data a bit closer. In Nāgārjuna’s treatise, the Dismissal of Disputes (Vigrahavyāvartanī), a non-Buddhist metaphysical realist—a Naiyāyika or Vaiśeṣika— argues in the first verse that Nāgārjuna’s own statement that nothing has any intrinsic nature is self-refuting, because if true, it would imply that this very thesis-statement itself does not really exist and therefore could not do anything; it could not mean anything, prove anything, nor refute anything. Nāgārjuna replies in verse 29 of the Dismissal:
Here Nāgārjuna recognizes that if he had a fully real thesis-statement to the effect that nothing is fully real, then the existence of that very statement would falsify what it asserts; but, so his reply goes, he does not have any such real thesis-statements, precisely because nothing is real, and hence does not fall into self-refutation. Indeed, the Dismissal repeatedly argues that reasoning, statements, etc., can function, all the while being unreal, just as two illusory or magical projections can do things (or, more accurately, seem to do things) to each other. Āryadeva in chapter sixteen of the Four Hundred Verses, and in a broadly similar passage in the Hundred Verse Treatise chapter ten (Tucci 1929, 82), takes up that charge of self-refutation. He says in the Four Hundred Verses 16.2:
“[Objection:] Since the [[[Mādhyamika]]] speaker, what he states [and his own words] would also be [unreal], then his saying that [all] is empty would be incoherent. [We reply that] That [i.e., the speaker, assertion and words] which arises dependence is not existent in any of the three.” Instead of appealing to reasoning as simply an efficacious play of illusion, he invokes the basic Madhyamaka idea that things— including speakers, what is stated, and words— are all mutually dependent upon each other. However, for our purposes, the objection and reply are the same as in Nāgārjuna’s Dismissal. In both cases, the no thesis stance invoked is not the generalized quietistic refusal to make truth claims about the real; it is about the mere existence, or illusoriness, of one statement and its speaker. In short, this is a somewhat specialized debate that should not be mistakenly conflated with Madhyamaka’s generalized rejection of truth claims or his rejection of any semantics that presupposes fully real entities (see Oetke 2003 against Seyfort Ruegg’s interpretation of Dismissal verse 29 in such a generalized fashion). The undeniable philosophical interest and importance of Madhyamaka’s rejection of realist semantics does not mean that the specific passages in the Dismissal, like verse 29, are to be read as themselves expressing that rejection. The same critique would apply to semantic readings of Āryadeva’s 16.2.
Let’s turn to the second sense of “no thesis.” The stance that one should not make or endorse any truth claims that imply commitment to fully real things, or that have a realist semantics, is at its clearest in Āryadeva’s systematic refutation of each of the “four alternatives,” or the tetralemma (catuṣkoṭi), i.e, is…; is not…; both is…and is not…; neither is…nor is not…. These four supposedly exhaust the possible formulations of any metaphysical thesis (on the tetralemma, see Seyfort Ruegg, 2010, chapter 3). Here is how Āryadeva described Madhyamaka method in verse 22 of chapter fourteen in the Four Hundred Verses: “Being, nonbeing, [both] being and nonbeing, neither being nor nonbeing: such is the successive method that the wise should always use with regard to identity and the like [i.e., all other theses].”
Typically, each of these four is refuted by reductio ad absurdum arguments, so-called prasaṅga, i.e., consequences that would be unacceptable to the proponent of the alternative in question. Āryadeva’s extensive use of absurd consequences is thus taken by many as evidence that he is a “Consequentialist Mādhyamika” (Tib. dbu ma thal ‘gyur pa; Skt. *prāsaṅgikamādhyamika), as is supposedly Nāgārjuna–roughly, a Mādhyamika who only refutes others, but does not positively prove any ontological positions of his own.
Consequentialist Madhyamaka is the way to read Āryadeva’s work via the commentary of the sixth century Mādhyamika Candrakīrti, the Catuḥśatakaṭīkā, and it does lead to a refusal of theses, for the would-be intrinsic natures of entities turn out to be riddled with inconsistencies on analysis, so that no-one can rationally make fullfledged truth claims about them. While Tibetan thinkers, and indeed most modern scholarship, have embraced some variant of this Candrakīrtian interpretation of Āryadeva, it is certainly not the only way to understand his thought and refusal of theses. The other important way is in terms of Yogācāra idealism, by relying upon the Yogācāra’s trademark stance, viz., the three nature (trisvabhāva) theory.
While Consequentialist Madhyamaka rejects any and all intrinsic natures (svabhāva) and all truth claims that presuppose their existence, Nāgārjuna and Āryadeva, if we take a Yogācāra interpretation of their thought, reject only one type of nature among three: their quietism is thus correspondingly circumscribed. In particular, they refute the existence of subject-object dualities and the linguistically conditioned, in favor of “mind alone” (cittamātra). More generally, they reject all intrinsic natures conditioned by conceptual thought and language as just “imagined natures” (parikalpitasvabhāva) that are unreal and nonexistent. On the other hand, they leave the ineffable dependent nature (paratantrasvabhāva) and the perfect nature (pariniṣpannasvabhāva) intact as existent.
It is quite probable that four out of the so-called eight great commentators on Nāgārjuna—i.e., the four Yogācāra commentators, Guṇamati, Gunaśrī, Devaśarman, Sthiramati—would have held something like this three-nature interpretation of Nāgārjuna, although the surviving textual fragments of the first three are very meager. In the case of Sthiramati (an hui 安慧; c. 475-555), his commentary (i.e., Taishō 1567 da cheng zhong guan shi lun 大乘中觀釋論) remains only in a difficult Chinese translation (Tillemans 1990, 57-58) by Wei jing 惟淨. That said, the Yogācāra interpretation of early Madhyamaka was clearly a very strong contender in India. Yogācāra-Madhyamaka synthesis continues all the way to the eleventh century thinker Ratnākaraśānti and beyond to Tibet, its followers being designated under a huge variety of names—“those who profess emptiness of what is other (gzhan stong pa)” and other appellations. Pertinently, Ratnākaraśānti’s self-definition was as a *Trisvabhāva¬mādhyamika (rang bzhin gsum gyi dbu ma pa) (see Luo 2018).
The three-nature reading of Madhyamaka is indeed the interpretation promoted by Dharmapāla (hu fa 護法; c. 530-561 C.E.), who commented upon the last half of the Four Hundred Verses and whose work does not survive in Sanskrit, was never translated into Tibetan, but only comes to us in Xuanzang’s Chinese translation (Taishō 1571 guang bai lun shi lun 廣百論釋論). It was Dharmapāla’s commentary on the last chapter of Āryadeva’s Four Hundred Verses that provided his responses in a crucial debate with the Mādhyamika Bhāviveka, one that set Yogācāra and Madhyamaka on significantly different trajectories. (For a translation of key parts of this chapter of Dharmapāla’s commentary, see Hoornaert 2004. ) Āryadeva was thus the focal point for a major parting of the ways in India. Interestingly enough, he was also probably a focus for the Tibetan synthetic movements —his commentator Dharmapāla sometimes figures in their lineages too. What textual evidence might have prompted Dharmapāla to see Yogācāra in Āryadeva? As Hoornaert 2004, n. 11 pointed out, it seems that Dharmapāla was working with a Sanskrit manuscript of the Four Hundred Verses that was significantly different from that on which Candrakīrti commented; Dharmapāla’s Sanskrit source must have contained some key Yogācāra terminology lacking in that used by Candrakīrti. In particular, Dharmapāla’s three-nature stance on Āryadeva comes out most clearly in his commentary to Four Hundred Verses 16.23, a verse that can be translated as follows from the Chinese:
“[Objection:] If things are empty of intrinsic natures (ben xing 本性 = svabhāva), what benefit [do we gain] by seeing that emptiness (kong 空 = śūnyatā)? [Response:] It is the realization of emptiness that eliminates the fetters of false conceptualizations (xu wang fen bie 虛妄分別 = abhūtaparikalpa). The term abhūta¬parikalpa is a notoriously important Yogācāra term, one that easily evokes the three-nature theory. False conceptualizations are themselves dependent natures (paratantra-svabhāva), as they arise causally and produce effects, but their content consists in imagined natures (parikalpitasvabhāva). Indeed, Dharmapāla does go into an exposition of the Yogācāra three- nature theory immediately after this verse. If, however, we translate the same verse as it figures in Tibetan in Candrakīrti’s commentary—the actual Sanskrit of the passage is unfortunately lost—it is apparent that abhūtaparikalpa would not have occurred. Indeed, we simply have the basic and recurrent Buddhist idea that conceptualization obscures or blocks direct seeing of the truth. “[Objection:] If entities exist by their intrinsic natures, what benefit do we gain in seeing emptiness? [Response:] Seeing is fettered by conceptualizations (rtog pa = kalpanā); they are what is to be refuted here [in the Four Hundred Verses].”
It is also clear that the rest of verse 23 is very different in the manuscript used by Candrakīrti, and thus not at all surprising that it led to a very different commentary from that of Dharmapāla. In Candrakīrti’s explanation there is no clear reference to Yogācāra. The crucial point of Dharmapāla’s interpretation of Āryadeva is to avoid what he sees as a nihilistic notion of ultimate truth (paramārthasatya). While many Mādhyamikas construe the ultimate truth, emptiness (śūnyatā), as a simple negation (prasajyapratiṣedha), or an absence of any real intrinsic nature whatsoever, Dharmapāla, in his commentary to the Four Hundred Verses 16.23, stresses that it is only the linguistically conditioned imagined natures that are customary truth, while the ineffable dependent and perfect natures are ultimate truths. Moreover, he makes it clear (f. 247c) that those linguistically conditioned natures are the same as universals, or common characteristics (sāmānyalakṣaṇa), while the dependent natures are particulars (svalakṣaṇa). There is thus a deliberate rapprochement with the nominalist philosophy of the Epistemological school of Dignāga.
“Words designate only common characteristics (共相, sāmānyalakṣaṇa), but the individual characteristics (自相, svalakṣaṇa) of things are beyond the scope of words. Individual characteristics are not nonexistent (非無), and common characteristics are not existent (非有).” (247c. Translation Hoornaert 2004) The upshot is that ultimate truth for him is certainly not the simple absence of all intrinsic nature, as it is for Mādhyamikas from Bhāviveka and Candrakīrti on; it is the combination of the ineffable dependent nature—i.e., particulars perceived without any linguistic conditioned natures—and the perfect nature. The two philosophical interpretations of Āryadeva thus differ markedly: the Yogācāra sees him as accepting a wordless, concept-less reality composed of particulars, whereas for a Mādhyamika Āryadeva’s ultimate is the denial of any reality whatsoever.
What are we to make of Āryadeva’s ethics in the Four Hundred Verses? Importantly, if Āryadeva, like Nāgārjuna, rejected Buddhist and non-Buddhist claims that involve ontological commitment, he did not think that he also had to reject or remain quietistic about ethical positions and religious notions of enlightenment, bodhisattva paths, and the like. Indeed, as we saw in our outline of the chapters of the Four Hundred Verses, half of the work argues for basic Buddhist ethical views on impermanence, suffering, purity, selflessness, altruistic intentions, freedom from attachments, and the conduct of bodhisattvas. Āryadeva thus clearly thought that such argumentation did not fall victim to his metaphysical critiques or quietism about having theses. The result is a complex position that reoccurs, in one way or another, in major Madhyamaka writing, from Nāgārjuna’s Jewel Garland (ratnāvalī) to Candrakīrti’s commentaries (vṛtti, ṭīkā) on Nāgārjuna and Āryadeva, Śāntideva’s Engaging in the Practices of a Bodhisattva (bodhicaryāvatāra) and on to the Path and its Stages (lam rim) of Tsong kha pa (1357-1419). In these writers’ works, we find a more or less nuanced rejection of ontologies, coupled with a robust advocacy of the canonical Buddhist ethics and religious practices promoted in the Discourses (sūtra) traditionally attributed to the Buddha, the Scholastic Teachings (abhidharma), Code of Discipline (vinaya), or the Great Vehicle literature laying out the bodhisattva path.
3.1 Descriptive versus revisionary
The Oxford philosopher P.F. Strawson made a now well-known distinction between descriptive and revisionary metaphysics—“Descriptive metaphysics is content to describe the actual structure of our thought about the world, revisionary metaphysics is concerned to produce a better structure” (Strawson 1959, i). We could extend Strawson’s descriptive-revisionary contrast to ethics. An ethics that prones supererogation, moral sainthood, universal responsibility, and other very substantial betterments to our present views and behaviour will be said to be revisionary while one based more on how people generally do think and act will go in the direction of description. What would the Madhyamaka of Āryadeva, Nāgārjuna, Śāntideva et al. look like in this light?
These philosophers will end up with a significant tension in their overall stance: they adopt a largely descriptive approach to all matters ontological, but a strongly revisionary approach to ethics. Ethics and bodhisattva practices concern customary (saṃvṛti) truths, or customarily real matters about which one can and should argue, while ontology is about ultimately (paramārtha) real things about which one should not take a position, nor argue; instead of bettering the world’s customary views on what there is one supposedly limits oneself, faute de mieux, to description about what the world acknowledges (lokaprasiddha) and what things are in keeping with its epistemic procedures. Such is the position of Candrakīrti in his commentary to Four Hundred Verses, especially in his arguments in Chapter thirteen against the logicians’ misplaced attempts to revise worldly epistemic standards (see Tillemans 1990, vol. 1, 176-179 ). Indeed, in his Clear Words (prasannapadā) he quoted a famous sūtra passage in support of reliance on worldly acknowledgement:
“The world (loka) argues with me. I don't argue with the world. What is agreed upon (saṃmata) in the world to exist, I too agree that it exists. What is agreed upon in the world to be nonexistent, I too agree that it does not exist.”
If Candrakīrti is right, this is also how Āryadeva proceeded on both ontological and epistemological issues: he left the world’s customary truths about what is and is not and how we know largely intact.
The ethical discussions in the Four Hundred Verses, however, do not follow that model: they do not leave the worlds’ ethical views largely intact, but rather seek to show that the world should very significantly better its views and behaviour. In effect, Āryadeva legitimizes the revisionism inherited from canonical non-Madhyamaka Buddhist sources. A problem of systemic coherence then arises: if Madhyamaka tells us to do little more than describe what the world acknowledges on customary matters concerning what is and is not – essentially because there is no realist ontology that would provide leverage for anything better—then why the radical revisionism in ethics? That philosophical tension between description and revision is not just limited to Āryadeva: it is what we find in Nāgārjuna, Śāntideva and many others. Let’s look at how it plays out in more detail in the Four Hundred Verses. The picture, in my opinion, may well need some significant rethinking if Madhyamaka Buddhism is to be persuasive to a modern reader.
Indeed it must be said from the outset that for many readers nowadays the ethical chapters in Āryadeva’s Four Hundred Verses will meet with stiff resistance in part because arguments for passionless detachment are often focused on strong deprecation of women and sexuality. Misogyny was, no doubt, widespread in the ascetical monastic milieu in which much of the Great Vehicle (mahāyāna) historically originated. (Contrary to how Mahāyāna has been typically depicted, it was not a predominantly egalitarian lay movement (see Boucher 2008: 50-52, Nattier 2003, 96-100)). It is attested in Pāli Discourses (sutta) of early Buddhism too, like the Numerical Discourses (aṅguttaranikāya) (see Bhikkhu Bodhi 2012, 683-684), although, revealingly enough, the most virulent forms seem more frequent in the Great Vehicle (Sponberg 1992, 21). In any case, Āryadeva manifestly inherited monastic asceticism and deprecation of women, bolstering it with questionable dialectics in chapter three of the Four Hundred Verses; Nāgārjuna had done likewise extensively in his Jewel Garland (ratnāvalī).
The origins of Āryadeva’s deprecation of women are thus partially explicable historically. Nonetheless, the disturbance it poses to those who think about his ideas philosophically will not simply go away with basic facts about the early Mahāyāna context. There is, in effect, a question as to what is core and what is peripheral in this ethical thought. Minimizers of the disturbance might well try to point out that no less an ethicist than Aristotle defended condemnable institutions such as slavery as just and natural in Politics I.iii-vii, and that his promotion of it does not vitiate the many important ideas in the Nicomachean Ethics, nor virtue ethics in general, as it is somehow unessential and peripheral to them. The Four Hundred Verses’ deprecation of women may, however, present a more intractable problem than Aristotle’s acceptance of slavery, as it is a vivid, emotionally charged theme that seems to fit very closely into a larger Buddhist advocacy of renunciation and contemptus mundi: Āryadeva is against desire, the wrong idea that the female body is clean, romantic love, love for family members and grief at their death (1.12-13); even work and all other exertions are devoid of anything pleasurable (2.18). I leave it up to the reader of Āryadeva’s and Candrakīrti’s first four chapters (see Lang 2003) to judge if deprecation of women is only peripheral to their ethics and whether the parallel with Aristotle’s unfortunate rationalization of slavery would hold.
3.3 Intuitions, consensus, moral disagreement
Āryadeva’s method in the first eight chapters of the Four Hundred Verses seems to be largely an attempt to plomb people’s ethical intuitions, showing that if we were not in the grip of self-grasping, knew the facts, and were consistent with fundamental intuitions, we would accept canonical Buddhism’s ideas, and would change much of what we now think. Is that approach, based on an actual or ideal consensus, or perhaps just convergence of vews, likely to justify Āryadeva’s revisionary conclusions? Not easily, for justifying canonical Buddhism’s ethics by means of a meta-ethics of intuitions and convergence is a very tall order (see Finnigan 2015). While consistency with intuitions can lead to changes of beliefs without appeals to metaphysics, scripture, omniscient buddhas, divine will, or some other source of argumentational leverage, the changes would usually leave the larger more fundamental framework of beliefs untouched. Strongly unacceptable radical conclusions will instead lead people to say that premises are false or that the inferences are non-sequiturs. Why then did Āryadeva himself think that rational people would eventually come around to Buddhist customary truths on ethics?
Some might well say that, in any case, there are genuine, insurmountable moral disagreements amongst rational people—they might then argue that the existence of such disagreements are evidence for historical and cultural relativism. Or they would be evidence for some version of moral skepticism, viz., that there are no truth values for moral statements at all. Or, finally, it might be said that the genuine moral disagreements show that moral statements are all erroneous; it might even be added that an error theory would be the position most in keeping with Āryadeva’s Buddhism, for (so it would be argued) Mādhyamikas do not hold any statements of customary truths (saṃvṛtisatya) to be true, only that people think them true for some sort of religious or practically advantageous result. Such arguments would raise a number of complicated issues, including whether genuine disagreement does imply moral skepticism (see e.g., Wedgwood 2014), and whether Āryadeva, and Madhyamaka in general, intended to promote truths or only convincing semblances of them (see Cowherds 2015).
3.4 Scripture and the humanly unknowable
While the existence of moral disagreement and the relativist or skeptical consequences that might ensue are important philosophical issues, it is also important to see that they are more or less red herrings when it comes to understanding Āryadeva’s own stance : such disagreement simply was not genuine for him; the apparent disagreements would and should disappear with full knowledge of facts. Āryadeva, in effect, granted that justification of Buddhist revisionary ethical positions could not only be based on the religiously unschooled intuitions of the common person and the unproblematically observable or inferrable facts recognized by the world. Instead, ethics required belief in special kinds of facts described in Buddhist scriptural statements, i.e., “radically inaccessible” (atyantaparokṣa) facts about which actions lead to which results as moral retribution—for him, as for subsequent Mahāyānists, much of the workings of that subtle, karmic, retributive causality will be unfathomable by any ordinary human beings in that it is humanly unobservable, not inferable from anything observable, and thus only understandable through trustworthy scriptures authored by individuals with extraordinary understanding of the suprasensible (Tillemans 1999 chapter 1, 2).
Would Buddhist insistence upon such special facts make rational individuals radically revise their beliefs? The obvious question is why would rational individuals believe that Buddhist scriptures are trustworthy on unfathomable matters, especially if Mādhyamikas take customary truths to be those that are acknowledged, and hence fathomable, by the world? Four Hundred Verses 12.5 attempts a response: Āryadeva tells us that when, in an ethical deliberation, there is doubt about the veracity of the Buddha’s descriptions of imperceptible karmic consequences, we should nonetheless believe what he said about them because of his track record in other significant areas, notably his teaching on emptiness. This response is crucially important in the history of Buddhism and is still current; it is invoked regularly by Tibetans, including the Fourteenth Dalai Lama. Dharmapāla and Candrakīrti develop an account of scriptural trustworthiness on the basis of 12.5, and, if we accept Tibetan accounts, that Āryadevan account was the same as what we find later in Dharmakīrti’s discussions of scripture, i.e., the “triple analysis” (Tib. dpyad pa gsum) of the reliability of scriptures (see Tillemans 1999, chapters 1 and 2; 2000, 78-79). It may well be that Āryadeva played a major role in strongly grounding ethics on the suprasensible.
Here is what Āryadeva said:
“When someone entertains doubt concerning the imperceptible things (parokṣa) taught by the Buddha, he should develop a [[[Wikipedia:rationally|rationally]] founded] belief in these very things on account of emptiness (śūnyatā).”
His point is that humans supposedly can, with their own critical acumen, determine that the teachings on emptiness are an example (dṛṣṭānta) where the Buddha got something very important perfectly right. Therefore, so it is argued, because of his reliability on something essential like emptiness, it is also rational to believe his statements, even when we cannot ourselves determine their truth. There is, in effect, a transfer of credibility from one subject to another. Candrakīrti, in his commentary to 12.5, will go further and claim that transferring credibility is in keeping with worldly practices and intuitions – it is, as he puts it, following “your very own principles alone (svanayenaiva).” In other words it follows intuitions and epistemic practices that are acknowledged by the world (lokaprasiddha). We regularly believe people who are proven experts.
As I have argued in more detail elsewhere (Tillemans 2016, chapter 8) that argument, with a little reflection, can be seen to be badly flawed. Having attested competence and credibility in one subject does not transfer to another unless the two are demonstrably closely related. Probably we are all familiar with the trap of wrongly according extended credibility to people simply because they happen to be highly qualified in one important field. Those would-be universal experts become unreliable, not because they hadn’t at some point understood some significant things, but because they overstep their qualifications; their expertise is made to extend to subjects not related to the area in which they have been rightly recognized. The flaw in the appeal to the Buddha’s track record is similar: it is not at all clear that trustworthiness concerning important general principles like emptiness does reasonably transfer to explanations concerning the details of karma in all their specificity and complexity, because the relationship is not clear. While emptiness of intrinsic nature, as a general principle, may be closely linked with the general feature that phenomena arise dependently due to causes and conditions—as Buddhists from Nāgārjuna on have stressed —knowing that much would hardly suggest that one somehow knows the specific details of what is retribution for what.
In the end, I think that verse 12.5, Candrakīrti’s construal of it, or the closely related Dharmakīrtian triple analysis, will not provide a publicly debatable proof of scriptural trustworthiness to convince people outside an already committed Buddhist community. Indeed, Dharmakīrti and his commentators seem to say as much. Dharmakīrti was quite skeptical about the probative status of “scripturally based inferences” establishing such things as subtle karmic causality, and he said clearly in his Own Commentary (svavṛtti) to verse 1.217 that reasoning “in this way is not a bona fide inference at all (na khalv evam anumānam anapāyam) as there is no necessary connection between words and states of affairs (anāntarīyakatvād artheṣu śabdānām).” In short, there are just too many counterexamples where people are correct on one important set of things, but fall down hopelessly on others. Śākyabuddhi and Karṇakagomin give us another angle: “it is not an inference because of factual matters” (vastutas tv ananumānam); it is only one “on account of the thought of people who want to engage [on the spiritual path] (pravṛttikāmasya puṃso ‘bhiprāyavaśāt).” (see Tillemans 1999, 43 and n. 11). In other terms, it is a type of faith-based reasoning that will be used among religious aspirants, but not one that could be invoked in a neutral, public context of unbelievers. Dharmakīrti, to his credit, recognized that limitation. It seems unlikely that Āryadeva and Candrakīrti did.
3.5 Epistemic humility?
A larger problem then arises. Is basing ethics, in part or in whole, on a retributive moral causality that is nonetheless outside the range of human beings’ own knowledge a coherent, rational standpoint at all? Arguably, for Buddhist metaphysical realists of the non-Madhyamaka schools, like Dharmakīrti, it is not an incoherent position to say that actions have humanly inaccessible dispositions to lead to certain results, dispositions that are as they are irrespective of all beliefs and intuitions people might have. Such realist philosophers could, in effect, accept a type of epistemic humility about karmic facts—unknowability does not count against real existence, but only calls for acceptance of our personal limitations.
A Mādhyamika who disavows metaphysical realism and professes exclusive adherence to what is customarily acknowledged by the world (lokaprasiddha) does not, however, have the ready option of such humility. Candrakīrti’s argument appealing to “your own principles alone” does not work and it is hard to see that any other like it will. For a Mādhyamika the conclusion then is not epistemic humility, but is one that is potentially more disturbing to traditional Buddhists: if subtle karmic cause and effect is not acknowledged as existent by most of us in the world and not justifiable by the world’s epistemic procedures, much talk of karma-based distinctions in religious ethics risks to be deemed false. In the end, what the unresolved tension between description and revision in the two sections of the Four Hundred Verses shows us is that a Madhyamaka philosophy that emphasizes quietism about ontology and a strong reliance on worldly intuitions will have to move in a different direction from Buddhist traditions that have heavily relied on the suprasensible and the faith-based, discussable only within protected confines. The price to pay for the philosophy of the second eight chapters of the Four Hundred Verses would seem to be much of the ethics of the first eight. Instead of Āryadeva’s positions, Madhyamaka’s emphasis on the world’s ideas would seem to require a much more secularly oriented, publicly debatable ethics that values what people do think with their own autonomous reason.
CS = Catuḥśataka of Āryadeva. CST = Catuḥśatakaṭīkā of Candrakīrti. GBLSL = Guang bai lun shi lun 廣百論釋論. Taishō 1571. Dharmapāla’s commentary on the last eight chapters of CS. Taishō = Taishō Shinshū Daizōkyō. The Tripiṭaka in Chinese. Ed. by J. Takakusu and K. Watanabe. 100 volumes. Tokyo, 1925-1935.
Bhattacharya, Vidhushekhara 1931. The Catuḥśataka of Āryadeva: Sanskrit and Tibetan texts with copious extracts from the commentary of Candrakīrti. [Chapters 8-16.] Part 2. Calcutta: Visvabharati Bookshop. Bhikkhu Bodhi 2012. The Numerical Discourses of the Buddha. A Translation of the Aṅguttara Nikāya. Somerville, MA: Wisdom Publications. Boucher, Daniel 2008. Bodhisattvas of the Forest and the Formation of Mahāyāna. A Study and Translation of the Rāṣṭrapālaparipṛcchāsūtra. Honolulu: University of Hawaii Press.
Cabezón, José I. 2008. “Buddhist Narratives of the Great Debates.” Argumentation 22: 71-92. Chimpa, Lama and Alaka Chattopadhyaya 1970. Tāranātha’s History of Buddhism in India. Simla, H.P.: Indian Institute of Advanced Study. Cowherds, The (= Georges Dreyfus, Bronwyn Finnigan, Jay Garfield, Guy Newland, Graham Priest, Mark Siderits, Koji Tanaka, Sonam Thakchoe, Tom Tillemans, Jan Westerhoff) 2011. Moonshadows: Conventional Truth in Buddhist Philosophy. New York: Oxford University Press. Feer, Léon 1890. The Saṃyutta-nikāya of the Sutta-piṭaka. London: Published for the Pali Text Society by Henry Frowde, Oxford University Press. Finnigan, Bronwyn 2015. “Madhyamaka Buddhist Meta-ethics : The justificatory grounds of moral judgments.” Philosophy East and West 65.3: 765-785. Finnigan, Bronwyn 2018. “Madhyamaka Ethics.” In Daniel Cozort and James Mark Shields (eds.) The Oxford Handbook of Buddhist Ethics. Oxford: Oxford University Press: 162-183.
Gnoli, Raniero 1960. The Pramāṇavārttikam of Dharmakīrti. The First Chapter with the Autocommentary. Text and Critical Notes. Serie Orientale Roma 23. Rome: Istituto Italiano per il Medio ed Estremo Oriente. Gokhale, Vasudev Vishwanath and Budhiruci 1930. Akṣara-Çatakam: The Hundred Letters, a Madhyamaka Text. After Chinese and Tibetan materials translated by Vasudev Gokhale Materialen zur Kunde Buddhismus 14. Heidelberg: O. Harrasowitz. Haraprasād Shāstrī 1914. “Catuḥśatika of Ārya Deva.” Memoirs of the Asiatic Society of Bengal, Vol. III, 8 (Calcutta): 449-514. [[[Sanskrit]] fragments of CS and CST]
Hoornaert, Paul. 2004. “The Dharmapāla-Bhāvaviveka Debate as Presented in Dharmapāla’s Commentary to Catuḥśataka XVI.23.” Kanazawadaigaku Bungakubu Ronshū, Kōdōkagaku Tetsugaku hen 24—Studies and Essays, Behavioural Sciences and Philosophy, Faculty of Letters. Kanazawa University, 24: 119-149. Johnson, Dennis A. 2012. Refuting the Conditioned—The Saṃskṛtārthapratiṣedha of Candrakīrti’s Catuḥśatakaṭīkā: Introduction, Translation and Summary together with a Critical Edition of the Tibetan Translation. MA Thesis, University of Vienna. Available at : https://othes.univie.ac.at/20891/ accessed 2019.01.18. [Translation of chapter 15 of CS and CST]
Keenan, John P. 1997. Dharmapāla’s Yogācāra Critique of Bhāvaviveka’s Mādhyamika Explanation of Emptiness. The Tenth Chapter of Ta-ch’eng Kuang Pai-lun Shih Commenting on Āryadeva’s Catuḥśataka Chapter Sixteen. Lewiston: Edwin Mellen. La Vallée Poussin, Louis de 1970. Mūlamadhyamakakārikās (Mādhyamikasūtras) de Nāgārjuna, avec le commentaire de Candrakīrti. First published in St. Petersburg, Bibliotheca Buddhica IV, 1903-1913. Reprinted Osnabrück: Biblio Verlag, 1970. La Vallée Poussin, Louis de 1971. L’Abhidharmakośa de Vasubandhu: traduction et annotations, 6 vols. Paris: Geuthner. Louvain: J.-B. Istas, 1923-31. New edition prepared by É. Lamotte, Mélanges chinois et bouddhiques 16. Brussels: Institut Belge des Hautes Études Chinoises, 1971. Lang, Karen 1986. Āryadeva’s Catuḥśataka. On the Bodhisattva’s Cultivation of Merit and Knowledge. Indiske Studier VII. Copenhagen: Akademisk Forlag. [Translation of CS.]
Lang, Karen 2003. Four Illusions. Candrakīrti’s Advice to Travelers on the Bodhisattva Path. Translation and Introduction by Karen C. Lang. New York: Oxford University Press. [Translation of chapters 1-4 of CS and CST] Langton, Rae 2004. “Elusive Knowledge of Things in Themselves.” Australasian Journal of Philosophy 82.1, 2004: 129-136. Lindtner, Christian 1982. Nāgārjuniana. Studies in the Writings and Philosophy of Nāgārjuna. Indiske Studier 4. Copenhagen: Akademisk Forlag, 1982. Reprinted Delhi: Motilal Banarsidass, 1990.
Luo, Hong 2018. “Is Ratnākaraśānti a gZhan stong pa?” Journal of Indian Philosophy 46: 577-619. https://doi.org/10.1007/s10781-018-9353-7. May, Jacques 1979. “Chūgan.” In Sylvain Lévi, J. Takakusu, Paul Demiéville, Hubert Durt et al. (eds.) Hōbōgirin, Dictionnaire encyclopédique du bouddhisme d’après les sources chinoises et japonaises V, p. 470-493. Paris : Librairie d’Amérique et d’Orient. Tokyo : Maison franco-japonaise, 1929- May, Jacques 1980. "Āryadeva et Candrakīrti sur la permanence (I)." In In¬dianisme et Bouddhisme, Mélanges offerts à Mgr Etienne Lamotte. Louvain: Publica¬tions de l'In¬stitut orientaliste de Louvain. [[[Wikipedia:French|French]] translation of chapter 9 of CS and CST] May, Jacques 1981a. "Āryadeva et Candrakīrti sur la permanence (II)." Bulletin de l'École française d'Extrême-Orient. Tome 69 à la mémoire de Paul Demiéville: 75-96.
May, Jacques 1981b. "Āryadeva et Candrakīrti sur la permanence (III)", Études Asiatiques XXXV, 2, Hommage à Constantin Regamey: 47-76. May, Jacques 1982. "Āryadeva et Candrakīrti sur la permanence (IV)." Études de Lettres, University of Lausanne, 3: 45-76. May, Jacques 1984. "Āryadeva et Candrakīrti sur la permanence (V)." Acta Indologica, Studies of Mysticism in Honor of the 1150th Anniversary of Kōbō Daishi's Nirvāṇam. Narita: Naritasan Shinshoji. pp. 115-144.
May, Jacques and Katsumi Mimaki 1979. "Chūdō." Hōbōgirin V, Paris-Tokyo: 456-470. Mimaki, Katsumi 1976. La Réfutation bouddhique de la permanence des choses (sthirasiddhidūṣaṇa) et la preuve de la momentanéité des choses (kṣaṇabhaṅgasiddhi). Publications de l’Institut de Civilisation Indienne, fascicule 41. Paris : Institut de Civilisation Indienne. Mimaki, Katsumi 1987. “Āryadeva.” In Vol. 1, p. 431a-432a of Mircea Eliade et al. (eds.). The Encyclopedia of Religion. New York: Macmillan Publishing. Nattier, Jan 2003. A Few Good Men. Honolulu: University of Hawaii Press.
Sonam, Ruth 2008. Āryadeva’s Four Hundred Stanzas on the Middle Way, with commentary by Gyel-tsap. Translated by Ruth Sonam with additional commentary by Geshe Sonam Rinchen. Ithaca NY: Snow Lion Publications. [Translation of CS, CST and rGyal tshab rje’s commentary] Sponberg, Alan 1992. “Attitudes toward Women and the Feminine in Early Buddhism.” In José I. Cabezón (ed.). Buddhism, Sexuality, and Gender. Albany: State University of New York Press: 3-36. Strawson, Peter F. 1959. Individuals: An Essay in Descriptive Metaphysics. London: Methuen. Suzuki, Kōshin 1994. Sanskrit Fragments and Tibetan Translation of Candrakīrti’s Bodhisattvayogācāracatuḥśatakaṭīkā. Tokyo: Sankibo. [[[Sanskrit]] fragments of CS and CST with their corresponding Tibetan.] Thomas, Frederick W. and Hakuju Ui 1918. “The Hand Treatise, A Work of Āryadeva.” Journal of the Royal Asiatic Society: 267-310. Tillemans, Tom J.F., 1990. Materials for the Study of Āryadeva, Dhar¬mapāla and Candrakīrti. The Catuḥśataka of Āryadeva, chapters XII and XIII, with the commen¬taries of Dhar¬mapāla and Candrakīrti: Introduction, translation, Sanskrit, Tibetan and Chinese texts, notes. 2 Volumes. Wiener Studien zur Tibetologie und Buddhismuskunde 24.1 & 24.2. Vienna: Ar¬beitskreis für Tibetische und Buddhistische Studien Universität Wien. Reprinted in one volume by Motilal Banarsidass Publishing Co., Delhi 2008. [Translation of chapters 12 and 13 of CS, CST and GBLSL.] Tillemans, Tom J.F. 1999. Scripture, Logic, Language: Essays on Dharmakīrti and His Tibetan Successors. Studies in Indian and Tibetan Buddhism. Somerville MA: Wisdom Publications. Tillemans, Tom J.F. 2000. Dharmakīrti’s Pramāṇavārttika: An Annotated Translation of the Fourth Chapter (parārthānumāna), vol. 1 (k. 1–148). Vienna: Verlag der Österreichischen Akademie der Wissenschaften. Tillemans, Tom J.F. 2016. How do Mādhyamikas Think? And other essays on the Buddhist Philosophy of the Middle. Studies in Indian and Tibetan Buddhism. Somerville MA: Wisdom Publications.
Tucci, Giuseppe 1925. “Studi Mahāyānici. La versione cinese del Catuḥçataka di Āryadeva confrontata col testo sanscrito e la traduzione tibetana.” Rivista degli Studi Orientali (Rome: Casa Editrice Italiana) 10: 521-590. [[[Wikipedia:Italian|Italian]] translation of chapters 9-16 of CS and some extracts of GBLSL.] Tucci, Giuseppe 1929. Pre-Diṅnāga Buddhist Texts on Logic from Chinese Sources. Gaekwad’s Oriental Series, no. XLIX. Baroda: The Oriental Institute. [Translation of Āryadeva’s Śatakaśāstra (bai lun), Taishō 1569.] Wedemeyer, Christian K. 2008. Āryadeva’s Lamp that Integrates the Practices (Caryāmelāpakapradīpa): The Gradual Path of Vajrayāna Buddhism According to the Esoteric Community Noble Tradition. American Institute of Buddhist Studies. New York: Columbia University Press. Wedgwood, Ralph 2014. “Moral Disagreement among Philosophers.” In M. Bergmann and P. Kain (eds.). Challenges to Moral and Religious Belief: Evolution and Disagreement. Oxford: Oxford University Press, 2014: 23-39.