Damien Keown’s Buddhist Virtue Ethics and Its Difficulties

Keown’s work, The Nature of Buddhist Ethics (1992) is perhaps the most comprehensive discussion of Buddhist ethical theory that can be found in the market today. Keown rejects utilitarian readings of Buddhist ethics and advocates virtue ethics as a better interpretative framework. The closest western analogue to Buddhist ethics is not utilitarianism but Aristotelian virtue ethics. Aristotle’s ethical theory is “an illuminating guide to an understanding of the Buddhist moral system.” (1992: 21). As in Aristotelian ethics, in Buddhism there is a teleological summum bonum (eudaimonia in Aristotle, nirvana in Buddhism) to be achieved through the cultivation of virtues.

According to Keown, actions in Buddhism are not, as utilitarianism claims, good or evil because they lead to good or evil consequences. Rather, actions generate good or evil consequences because they are intrinsically good or evil. The intrinsic good or evil of actions derives from the mental states motivating and accompanying actions. Specifically, the intrinsic evil of actions derives from the three roots of the unwholesome (akusala). The three roots of the unwholesome are greed (lobha) or passion (rāga), hatred (dosa), and delusion (moha). Conversely, the intrinsic good of actions derives from the opposite mental states, the three roots of the wholesome (kusala), a term that Keown translates as “the good.” The three wholesome roots are the source of all the “wholesome mental factors” (kusaladhamma), which he translates as “good qualities.” Similarly, the three roots of the unwholesome are the source of all unwholesome mental factors (akusala-dhamma), which he translates as “bad qualities,” also known as defilements or afflictions (kilesa, Skt. kleśa). Good and bad qualities, wholesome and unwholesome mental factors “are perhaps best understood as corresponding to the Western notion of virtues and vices” (1992: 63). In fact, Keown calls the three wholesome roots the three Buddhist cardinal virtues, namely, nongreed (alobha) or nonpassion (arāga), nonhatred (adosa), and nondelusion (amoha), which he translates as “liberality,” “benevolence,” and “understanding.” The summum bonum of nirvana is achieved by purifying the mental stream (santāna) from vices and bad qualities, and by cultivating the three cardinal virtues and good qualities.

However, unlike Aristotle, Keown identifies the summum bonum with the good and the right: “Nirvana is the good, and rightness is predicated of acts and intentions to the extent which they participate in nirvanic goodness. The right and the good in Buddhism are inseparably intertwined. If an action does not display nirvanic qualities then it cannot be right in terms of Buddhist ethics whatever other characteristics (such as consequences) it might have” (1992: 177).

Keown’s identification of nirvanic virtues with the good and the right has far reaching and controversial implications. The first one is that the goodness of actions depends exclusively on the motivation or the mental states underlying actions. Moral actions are clear cut: When they display or participate of nirvanic good, they are right and virtuous, and when they do not display or participate of nirvanic good, they are wrong and nonvirtuous. This is so from the inception of action in the mind, and nothing can change it later. As Keown puts it, “An action is right or wrong from the moment of its inception -- its nature is fixed by reference to nirvanic values, and it cannot subsequently change its status.…In Buddhist ethics it is the motivation which precedes an act that determines its rightness. An act is right if it is virtuous, i.e. performed on the basis of Liberality (arāga), Benevolence (adosa) and Understanding (amoha)” (1992: 177-178).

The second implication is that the Buddhist criterion of goodness excludes from the moral domain actions not participating or not displaying nirvanic virtues. The problem is that many Buddhists, at least at the beginning of their spiritual practice, act morally not so much motivated by nirvanic virtues, but rather by nonnirvanic virtues such as craving for a proximate goal such as a good rebirth. Even practitioners who act ethically aiming at the ultimate goal of nirvana do so, at least on some occasions, motivated by subtle forms of spiritual greed. Aiming at nirvana out of spiritual greed and observing the five precepts out of craving for some worldly reward or fear of punishment after death cannot be said either to participate or display nirvanic virtues. However, it is problematic to suggest, like Keown’s criterion does, that such actions are morally wrong, evil, and outside the moral domain.

Keown’s identification and subsequent criterion of goodness lead him to marginalize the proximate goals of Buddhist ethics. Proximate goals such as a large fortune, good reputation, entering confident into any assembly, unconfused death, and rebirth in heaven are no longer part of the moral domain unless actions leading to them participate or display nirvanic virtues. In Keown’s words, the proximate goals are just “non-moral consequences of ethical action…secondary, contingent, consequences of moral actions,” and “apart from the final one [heavenly rebirth] there is nothing particularly Buddhist about them” (1992: 125). This marginalization of the proximate goals of Buddhist ethics, however, is also problematic. It is true that according to the Pāli Nikāyas, it is considered wrong to state that “whatever a person experiences, whether it be pleasant or painful or neither-painful-nor-pleasant, all that is caused by what was done in the past” (SN IV.230). Because not everything is due to past moral actions or karma, it is also true that the proximate goals one experiences are not necessarily the fruit of past moral actions. However, this does not justify Keown’s marginalization of the proximate goals for several reasons.

First, Buddhist texts and Buddhist practitioners do not see the proximate goals as nonmoral. On the contrary, in all cases it is understood that they are moral consequences of moral actions, and in some occasions it is explicitly said that they are so. For instance, in (SN I.92) it is said that as a result of a moral action (providing a Buddha with almsfood), a person “was reborn seven times in a good destination, in the heavenly world.” Similarly, it is also said that as a result of that moral action, he “obtained the position of financier seven times.” That is, after being reborn seven times in heaven, he was born another seven times as a human being who obtained in each life a good job and the material prosperity that goes with it.

Second, not all Buddhist traditions follow what the Pāli Canon says about karma, and accept that Canon as the authority to define what is or what is not moral in Buddhist ethics. In fact, there are Buddhist traditions, for instance Tibetan traditions, that believe that everything, including the proximate goals, are always the consequence of past actions or karma. So at least from the point of view of these Buddhist traditions, the experience of proximate goals is necessarily a moral consequence of past moral actions. In these Buddhist traditions, where everything is seen as being the consequence of past karma, Keown’s marginalization of the proximate goals is unjustifiable.

Third, even if one were to ignore other Buddhist understandings of karma and concede to Keown and the Theravāda orthodoxy that the proximate goals are not always and necessarily consequences of moral actions, it is true, as we have already said, that according to the Pāli Nikāyas they are so on some occasions. Consequently, even if it is admitted that they are contingent consequences of moral actions, there are no grounds to characterize them as nonmoral. When they are consequences of moral actions, they are necessarily moral, as moral as the actions that produced them. Only when the proximate goals are not caused by moral actions can one properly say that they are not moral consequences. However, according to the Pāli Nikāyas, the workings of karma, and therefore whether or not the proximate goals are moral consequences of ethical actions, is something that only Buddhas can know (MN I.74), not scholars of Buddhist ethics.

Furthermore, Keown’s marginalization of the proximate goals is questionable from a descriptive point of view. Buddhist ethics in practice seems to unanimously consider the proximate goals as part of the Buddhist moral domain. It has been extensively documented that the proximate goals play an important ethical role in actual Buddhist practices. The proximate goals are the most common pattern of validation in traditional Buddhist communities. In Theravāda communities, the proximate goals of prosperity, fame, accumulation of merit, wholesome karma, and a happy rebirth usually validate moral and religious actions independently of the validation based on the ultimate goal of nirvana (Swearer, 1995; Gombrich, 1971; Bunnag, 1973; Keyes, 1983). Similarly, in Tibetan Buddhist ethics, even though the bodhisattva ideal is prevalent among monks, nuns, and lay people, the most common pattern of validation of ethical conduct is not the ultimate goal of Buddhahood, or nonabiding nirvana (apratiṣṭhita nirvāṇa), but the more proximate goals of merit, wholesome karma, and a good rebirth (Tucci, 1980; Ekvall, 1964; Samuel, 1993; Tatz and Kent, 1977; Lichter and Epstein 1983).

Keown’s marginalization of the proximate goals is also questionable on textual grounds. Already in the Pāli Nikāyas, it is possible to detect texts where the justification of ethical conduct based on proximate goals functions independently of the justification of ethics based on the ultimate goal of nirvana. (DN II.85; MN III.165; MN III.203f; MN I.400f; AN IV.247-248; AN V.306-308). Similarly, great Buddhist thinkers use the doctrines of karma and rebirth to justify moral behavior without mentioning nirvana or the display of nirvanic virtues as a necessary requirement for moral actions. See for instance Nāgārjuna’s Precious Garland (I. 12-23; IV.310) and Śāntideva’s Bodhicaryāvatāra (I. 17; IV 5, 9, 12, 14; V. 20; VI 72-74). Even the Dalai Lama discusses the doctrines of karma and rebirth and uses them as a justification of ethical conduct, without speaking about nirvana or the need to display nirvanic virtues (1994: 68-69).

Because the marginalization of the proximate goals and the exclusion of nonnirvanic actions from the domain of the good are not warranted by early Buddhist sources, classical Buddhist thinkers, or Buddhist ethics in practice, I do not see enough empirical grounds for Keown’s identification and subsequent criterion of goodness.

One might defend Keown by saying that he is engaged in normative ethics. That is, he is not trying to describe Buddhist ethics as it is or as it has been commonly practiced, but rather as it ought to be practiced. Keown’s criterion of goodness would be intended to apply only to certain ideal types of Buddhist ethics but not to all types of Buddhist ethical practice. Therefore, it would be unfair to reject his ideal view of Buddhist ethics from an empirical point of view.

It is true that Keown wants to go “beyond simple descriptive ethics,” but it is also true that following a methodology inspired by Little and Twiss, he claims to be doing both “description and analysis” (1992: 5-6). To state that the proximate goals are nonmoral consequences that are not particularly Buddhist, as Keown does, is an improvement from saying that they are “a concession to spiritually retarded individuals” (Little and Twiss 1978: 244). Nevertheless, I believe there is still room to improve our descriptions and analysis of Buddhist ethics. For instance, there is nothing wrong in saying from a descriptive point of view that there is a gap between the ideal practice of virtue ethics and the actions of Buddhists aiming at proximate goals without displaying nirvanic virtues. However, I do not think it is accurate to identify ideal types of practice with the nature of Buddhist ethics, especially when that nature entails the marginalization of important aspects of Buddhist ethics and the exclusion from the moral domain of less ideal types of ethical practice.

Expanding Keown’s Conception of Virtue Ethics

Keown’s rejection of utilitarian readings of Buddhist ethics is intended to overcome what he calls the “transcendency thesis,” that is, the view according to which morality is a preliminary stage of the Buddhist path to be transcended once the goal of nirvana is attained. Keown’s major contribution to the field of Buddhist ethics is perhaps his conclusive refutation of the transcendency thesis. A particular example of the transcendency thesis is what Keown calls the King-Spiro hypothesis, that is, the understanding of Buddhist ethics as composed of two originally independent and ultimately incompatible ethical systems, karmatic and nirvanic Buddhism (King 1964; Spiro 1970). According to the King-Spiro hypothesis, karmatic Buddhism is primarily practiced by lay people and nirvanic Buddhism by monks and nuns. Whereas the aims of karmatic Buddhism are the accumulation of good karma and happy rebirths, the aims of nirvanic Buddhism are the cessation of karma and rebirth. Karmatic Buddhism is related to ethical practice, and because karmatic Buddhism is a means to nirvanic Buddhism, it follows that ethics is also a means to nirvana and not a constitutive part of the ultimate goal. In other words, ethics is merely instrumental and transcended in nirvanic Buddhism.

Building upon the work of Harvey Aronson (1979; 1980) and Nathan Katz (1982) Keown challenges the anthropological studies of King and Spiro and shows how karmic Buddhism and nirvanic Buddhism are integrated; they are neither separable nor incompatible. Comparing Buddhist ethics to Aristotelian ethics, Keown argues that moral virtues are constitutive parts of the ultimate goal of nirvana. According to Aristotle, human flourishing includes not only the cultivation of intellectual virtues but also of moral virtues. Similarly, the Buddhist ideal involves both cognitive and affective virtues, wisdom as well as love and compassion. I agree with Keown on this, and his comparison of Buddhist ethics and Aristotle is very illuminating. However, the price to pay for Keown’s excellent refutation of the transcendency thesis and the King-Spiro hypothesis is a criterion of goodness that generates several difficulties from a descriptive point of view, specifically, the exclusion of nonnirvanic actions and the marginalization of the proximate goals.

I believe there is a way to avoid the trancendency thesis and the King-Spiro hypothesis without generating these descriptive difficulties. I also believe that that way is more consistent with the Aristotelian model of virtue ethics than Keown’s criterion of goodness.

The difficulties that Keown’s criterion of goodness generates can be solved by introducing a heuristic distinction between instrumental and teleological actions. By instrumental actions I mean actions leading to favorable conditions for cultivating nirvanic virtues and by teleological I mean actions actually displaying nirvanic virtues or virtues characteristic of the Buddhist ideal of sainthood.

Actions not displaying nirvanic virtues, such as observing the precepts out of craving for a proximate goal, participate in the good because they are instrumental for attaining the highest good of nirvana. The nonnirvanic moral actions are also good because they accord with the Dharma. When the proximate goals are the consequence of good actions leading to favorable conditions to cultivate nirvanic virtues or good actions actually displaying nirvanic virtues, they are part of the moral domain of the good. Even when the proximate goals are not the consequence of moral actions, they are part of the good because they are favorable conditions for the cultivation and actual display of nirvanic virtues.

Keown’s identification of the good with nirvanic virtues is inconsistent with the Aristotelian model of virtue ethics on which he explicitly bases his view of Buddhist ethics (1992: 21). Aristotle identifies eudaimonia with the highest good of human flourishing, but not with the moral domain of the good. Aristotle speaks about a variety of intrinsic goods, some of them constitutive of the highest good of human flourishing, such as moral and intellectual virtues, and others external or instrumental, such as honors, fortune, sensual pleasures, and friendship (Kraut 1999: 82-83; Sherman 1989: 125-127). Similarly, in Buddhism the good is never identified with the summum bonum of nirvana. As Keown himself acknowledges, the terms kusala and puñña, which denote the good in a wide sense, are predicated not only of good moral actions and dispositions, but also of the consequences of moral activity (1992: 123), which include the proximate goals.

Keown’s criterion implies that instrumental goods (nonnirvanic moral actions and the proximate goals) are not part of the moral domain of the good and not characteristic of virtue ethics. For Keown, only teleological goods (actions participating or displaying nirvanic virtues) seem to qualify as good and characteristic of virtue ethics. However, it is important to notice that with Keown’s criterion of goodness not even the Aristotelian system would qualify as virtue ethics because Aristotle speaks of both instrumental and teleological goods. If Aristotle is the paradigmatic representative of virtue ethics and he himself admits the existence of both instrumental and teleological goods, I do not see why Buddhist virtue ethics cannot have instrumental goods. How does the acceptance of instrumental goods compromise virtue ethics? If the acceptance of instrumental goods does not so compromise the virtue ethics of Aristotle, why would such acceptance compromise Buddhist virtue ethics? The acceptance of instrumental goods does not make Aristotle a utilitarian or an advocator of the transcendency thesis. Similarly, the acceptance of instrumental goods within Buddhism does not entail a commitment to utilitarianism or the transcendency thesis.

In different ways, the two kinds of actions that I am distinguishing here are related to the highest good of nirvana. While actions leading to favorable conditions to cultivate nirvanic virtues are instrumental, actions currently displaying nirvanic virtues are teleological. In other words, whereas actions leading to favorable conditions to cultivate nirvanic virtues are good because of their consequences, actions actually displaying nirvanic virtues are good because of their participation in mental states characteristic of the Buddhist ideal of sainthood.

The distinction between actions leading to favorable conditions to cultivate virtues and actions actually displaying virtues is inspired not only by Aristotelian ethics, but also and primarily by the early Buddhists concepts of kusala and puñña.

According to Premasiri, the terms kusala and puñña in early Buddhism refer to different types of actions with different soteriological consequences: “the term that is invariably used in specifying the good actions which lead to the spiritual bliss of nibbāna is kusala, whereas the term more frequently used for specifying the good actions which lead to sensuous enjoyment and happiness in saṃsāra is puñña” (Premasiri, 1976: 69, quoted by Keown 1992: 122). Keown rejects Premasiri’s distinction between kusala and puñña actions because both terms overlap in early Buddhism and because there is not scriptural evidence for the distinction. In his words, “If they were opposed in some way and had such different soteriological implications there is little doubt that the Buddha would have taken care to point it out.” In opposition to Premasiri’s view, Keown contends that “puñña and kusala do not describe two kinds of actions but emphasize different aspects of one and the same action” (1992: 123). However, even though the Buddha did not literally distinguish between kusala and puñña actions, Lance Cousins’ study of kusala in the Pāli tradition shows the existence of textual grounds for Premasiri’s distinction. Cousins states explicitly that he essentially agrees with Premasiri in this point and “although there is some overlapping, puñña is most often used in regard to actions intended to bring about results of a pleasant kind in the future. It is almost exclusively kusala which is used in relation to the Buddha’s path” (1996: 154). Using Keown’s own argument, if it were true that puñña and kusala referred to one and the same action “there is little doubt that the Buddha would have taken care to point it out.”

Because the early texts do not say explicitly that puñña and kusala refer either to two actions or one and the same action, one has to analyze the different usages of the terms to reach a conclusion. The most comprehensive study of these usages seems to indicate that the terms refer to two different kinds of actions. At least this is what Cousins concludes, agreeing with Premasiri. I prefer to follow their view and disagree with Keown on this point.

Theravāda Buddhist ethics, in practice, seems to maintain a clear distinction between actions leading to the accumulation of puñña and the experience of pleasant consequences within saṃsāra, and kusala actions leading to nirvana. This fact could be interpreted as an indication of the canonical origins of these two kinds of actions. One might dispute whether or not the contemporary Theravāda emphasis on puñña over kusala is consistent with the Pāli Canon, but one cannot deny that there are early textual grounds for speaking about two kinds of actions, at least when the meanings of kusala and puñña do not overlap. What can be denied, in agreement with Keown, is that the soteriological outcome of the two actions is unrelated. Keown’s rejection of the distinction between puñña and kusala actions presupposes that the two kinds of actions are opposed to each other and that they lead to two qualitative different soteriological goals. This, however, is not necessarily the case because kusala and puñña can also be interpreted as leading in different ways (teleological and instrumental) to one and the same soteriological goal, namely, nirvanic virtues. That is, one can consider puñña and the proximate goals as stepping stones towards kusala and the ultimate goal of nirvana.

However, in order to avoid the exegetical issues around the terminology of kusala and puñña, I prefer the heuristic distinction between instrumental actions (leading to favorable conditions to cultivate nirvanic virtues) and teleological actions (actually displaying nirvanic virtues). In fact my distinction is not exactly equivalent to the distinction between puñña and kusala. Actions actually displaying nirvanic virtues are similar to kusala if kusala signifies actions participating in wholesome mental states characteristic of the Buddhist ideal of sainthood. However, they are dissimilar if kusala actions in early Buddhism denotes primarily, as Cousins has shown, actions associated with special states produced by wisdom and related to a meditational context (Cousins 1996: 145). Actions leading to favorable conditions to cultivate nirvanic virtues are similar to puñña actions in that they both lead to pleasant or happy results in the future. However they are dissimilar in that puñña actions are not explicitly subordinated to the cultivation of nirvanic virtues in the future.

My distinction highlights the unity and continuity between these two kinds of actions, and between instrumental and teleological goods in Buddhist ethics. Instrumental and teleological actions have the same soteriological goal, both lead to nirvanic virtues and the cessation of suffering. Strictly speaking, it could be maintained that the soteriological goals are different because favorable conditions for cultivating nirvanic virtues is not the same as the actual display of these virtues. However, I prefer to emphasize that both kinds of actions ultimately converge in one and the same goal.

Like Keown, I do not support the King-Spiro hypothesis. The distinction between instrumental and teleological actions should not be confused with the distinction between karmatic and nirvanic Buddhism. The distinction I am proposing between these two kinds of actions is precisely intended to render the categories of karmatic and nirvanic Buddhism unnecessary. According to my distinction, both kinds of actions are karmatic and nirvanic at the same time. Both kinds of actions generate karma either instrumentally or teleologically related to nirvanic virtues. Liberated beings could be interpreted as performing actions that are only nirvanic in nature, but this is misleading because liberated beings are not beyond the moral domain of the good or beyond the karma of other living beings. The actions of liberated beings display nirvanic virtues, and in this sense they are part of the moral domain of the good. Similarly, actions of liberated beings can lead others to favorable conditions by teaching or inspiring them to cultivate or display nirvanic virtues.

Like Keown, I do not support the transcendency thesis, and my distinction does not imply that ethics in Buddhism is a preliminary practice to be transcended in advanced stages of the spiritual path. Keown is perfectly right on this point, and I share his view of moral virtues as a constitutive part of nirvana, together with cognitive virtues. In order to avoid the transcendency thesis, Keown proposes a version of virtue ethics where only teleological actions are part of the moral domain of the good. However, the acceptance of instrumental actions does not necessarily entail the transcendence of ethics once nirvana is attained. Ethics is never transcended in Buddhism. Although actions leading to favorable conditions to cultivate nirvanic virtues are instrumental, they are never transcended. In fact, liberated beings and perhaps all practitioners are supposed, at least ideally, to act to generate favorable conditions for the cultivation or display of nirvanic virtues in others. What is transcended in Buddhist ethics is the need to accumulate more karma to attain awakening, but not the performance of actions displaying nirvanic virtues, nor actions leading others to favorable conditions to cultivate nirvanic virtues.

The Early Buddhist Criteria of Goodness

At first sight, Keown’s analysis of morality in the Dīgha Nikāya seems to be grounding his account of Buddhist ethics in the Pāli Nikāyas. However, the third chapter (Ethics and Psychology) and the refutation of utilitarian readings of Buddhist ethics in chapter nine, make clear that Keown’s criterion of goodness is inspired by Abhidharma thought. In his words, “For utilitarians there are not intrinsically good motives, while for Buddhism action inspired by the three Cardinal Virtues is intrinsically good. In terms of Buddhist psychology, as we saw when discussing the Abhidharma in Chapter 3, the locus of good and evil is to be found in the human psyche -- not in the consequences of actions in the world at large.” (1992: 179).

However, is Abhidharma psychological ethics the yardstick to measure the nature of Buddhist ethics? Can Abhidharma ethics be generalized and extrapolated to all manifestations of Buddhist ethics? Here I limit myself to discuss whether or not Keown’s Abhidharmic criterion of goodness corresponds to the criteria of goodness found in early Buddhism, by which I mean the Buddhism of the Pāli Nikāyas.

Perhaps the more straightforward early Buddhist criterion of goodness can be found in (M.I.415-419) within the Ambalaṭṭhikārāhulovāda Sutta. There the Buddha advises his recently ordained son Rāhula to reflect before, during, and after performing a bodily, verbal, or mental action: whether or not an action may “lead to my own affliction, or to the affliction of others, or to the affliction of both; it is an unwholesome bodily action…verbal action…mental action with painful consequences, with painful results…it is a wholesome bodily action…verbal action…mental action with pleasant consequences, with pleasant results” (Bhikkhu Ñāṇamoli and Bhikkhu Bodhi 1995: 524-526).

The criterion has two parts. The first one is clearly utilitarian because it explicitly states that one should consider whether or not the consequences of an action might lead (saṃvatteyya) to one’s one affliction (attavyābādhāya), to the affliction of others (paravyābādhāya), or to the affliction of both (ubhayavyābādhāya). In other words, the first part of the criterion tries to minimize suffering for the greatest number, which here is referred to as the affliction of oneself, others, and both oneself and others. The second part of the criterion, which Peter Harvey does not mention in his own discussion of this text (1995), can be interpreted as characteristic of either moral realism or virtue ethics, depending on how the terms unwholesome (akusala) or wholesome (kusala) are understood.

The standard definition of the unwholesome and the wholesome appears in (MN I.47), within the Sammādiṭṭhi Sutta. There, the unwholesome is defined as the ten unwholesome actions: killing, taking what is not given, misconduct in sensual pleasures, false, divisive, harsh, and frivolous speech, covetousness, ill will, and wrong view. The roots of the unwholesome are said to be greed, hate, and delusion. The wholesome is defined as the opposite of the ten unwholesome actions, and the roots of the wholesome as the opposite of the three roots of the unwholesome.

Because the unwholesome and the wholesome refer to both certain external bodily and verbal actions and certain internal mental actions or mental states, it is not clear which is the primary meaning. If unwholesome refers primarily to the intrinsic proprieties of certain external actions such as killing, stealing, lying, and so on, then the second part of the criterion is characteristic of moral realism. If unwholesome refers primarily to internal mental actions or mental states such as covetousness, ill will, and wrong view, then the second part of the criterion is characteristic of virtue ethics.

The fact that the three roots of the unwholesome and the three roots of the wholesome are mental states does not preclude the existence of certain external actions that are intrinsically unwholesome or wholesome. Certainly mental states add unwholesomeness or wholesomeness to actions, but it is important to highlight that (MN I.47) does not state that the first six external bodily and verbal actions are unwholesome or wholesome depending on just the mental root from which they originate. This idea appears in Abhidharma literature but not in the Pāli Nikāyas.

Because the referent of unwholesome is external bodily and verbal actions as well as internal mental actions, I interpret the second part of the criterion as a combination of moral realism and virtue ethics; moral realism because certain external actions are unwholesome or wholesome, and virtue ethics because certain internal actions are unwholesome or wholesome.

The second part of the criterion also states that unwholesome and wholesome actions have respectively painful or pleasant consequences, painful or pleasant results. However, this mention of consequences in relation to the wholesomeness or unwholesomeness of actions is not a sign of utilitarianism. The point here is not to consider the consequences in order to minimize suffering, but rather realize that certain actions produce certain kind of consequences.

The second part of the criterion does not refute my characterization of the first part as utilitarian because it is nowhere stated that the wholesomeness or unwholesomeness of actions is what leads to pleasant or painful consequences. The second part of the criterion is not to be universalized but supplemented by a consideration of the consequences of actions. If what the criterion conveys as a whole were that the good or evil of actions depends exclusively on their wholesomeness or unwholesomeness, then it would be unnecessary to consider the consequences of actions; at best it would be redundant. Because the criterion as a whole does require a consideration of the consequences of actions and not just of their wholesomeness or unwholesomeness, I think it is better to interpret it as implying that the goodness of actions depends not only on their wholesomeness but also on their consequences.

It is true that the wholesomeness or unwholesomeness of actions is a very important factor to determine the goodness of actions. However, this does not mean that all the good or evil of actions derives from their wholesomeness or unwholesomeness. In the Pāli Nikāyas, the consideration of the wholesomeness or unwholesomeness of external bodily and verbal actions (moral realism) and internal mental actions (virtue ethics) is to be supplemented by the consideration of the consequences of actions for the happiness of oneself and others (utilitarianism).

If something can be inferred from the criterion found in the Ambalaṭṭhi-kārāhulovāda Sutta, it is not that the good or evil of actions depends exclusively on their wholesomeness or unwholesomeness. What the Ambalaṭṭhi-kārāhulovāda Sutta indicates is that before, during, and after performing an action one has to take into account two basic things: the wholesomeness or unwholesomeness of actions, and the consequences of these actions for the happiness or suffering of oneself, others, or both self and others. The moral realism and the virtue ethics of the second part of the criterion is combined with the utilitarianism of the first part. The second part of the criterion emphasizes the pleasant or painful consequences of wholesome or unwholesome actions, and in that way restates from a different angle (moral realism and virtue ethics) the first utilitarian part of the criterion. While the utilitarian part tries to minimize suffering for the greatest number, the moral realist and virtue ethics part remind the practitioner the intrinsic good or evil of certain actions.

I believe that the complexity and richness of this early Buddhist criterion of goodness is seriously compromised when one tries to reduce it to a form of virtue ethics where only the intentions or the mental states underlying actions count. Similarly, I believe that reducing this early Buddhist criterion of goodness to utilitarianism is equally reductionistic because consequences are not the only thing that matters. Following Roy Perret, I believe that the opposition between intentionalism and consequentialism in Western ethical theory does not figure in Buddhist ethics (1987). In other words, the goodness of actions does not depend exclusively on either the goodness of intentions or the goodness of consequences. Virtue ethics is certainly present in early Buddhism, and evidently the internal mental state or motivation underlying actions is very important to determine the overall goodness of actions, perhaps the most important from a Buddhist point of view. However, the intrinsic wholesomeness of certain external bodily and verbal actions, as well as the consequences of these actions for the happiness or suffering of oneself and others, are also extremely important and necessary to assess the goodness of actions.

Another text indispensable to understand the early Buddhist criteria of goodness appears in (AN I.186-187) within the Kālāma Sutta. There, the Buddha not only proposes a criterion to accept the truth of a doctrine as it is commonly interpreted, but also a criterion to ascertain the good or evil of a practice. The Buddha advices the Kālāma people not to accept something because it is tradition, hearsay, a sacred scripture, a logical reasoning, something said by a competent speaker, or even because it is said by one of their teachers. Then the Buddha says, “But when you know for yourselves, ‘These things are unwholesome, these things are blamable; these things are censured by the wise; these things, if undertaken and practiced, lead to harm and suffering.’ Then you should abandon them.” Then the Buddha asks whether greed, hatred, and delusion lead to welfare or harm, and he says that a greedy, hating, and deluded person will kill, steal, sexually misbehave, and lie. The opposite is said about what is wholesome and the three roots of the wholesome.

Like the criterion of the Ambalaṭṭhikārāhulovāda Sutta, the criterion of the Kālāma Sutta does not advise one to consider only the unwholesomeness of actions, literally of things (dhammā), which can refer to doctrines and/or practices. Besides the unwholesomeness (akusala) of things, the Buddha advises to consider three more things: whether or not they are blamable (sāvajja), censured by the wise (viññūgarahita), and if “practiced and undertaken lead to harm and suffering” (samattā samādinā ahitāya dukkhāya saṃvattanti). The opposite is also to be considered, that is, the wholesomeness of actions, whether or not they are blameless, praised by the wise, and lead to harmless and happy results.

The first two items to be considered, namely, whether or not actions are wholesome or unwholesome, blamable or blameless, are similar in that they all refer to the intrinsic goodness of certain bodily, verbal, and mental actions. The Kālāma Sutta, like the criterion of the Ambalaṭṭhikārāhulovāda Sutta, mentions internal and external actions, that is, mental actions such as greed, hatred, and delusion; and external bodily and verbal actions such as killing, stealing, sexual misconduct, and lying. Again, it is nowhere stated in the Kālāma Sutta that the four external actions are unwholesome due to the three mental roots of the unwholesome. Strictly speaking, the text only says that a greedy, hating, and deluded person will perform four kinds of external actions, and that a nongreedy, nonhating, nondeluded person will abstain from these actions. The unwholesome or the wholesome can refer to either the three roots or the four external actions. That is, it can be argued that the criterion combines features of moral realism (intrinsic good or evil external actions) with features of virtue ethics (intrinsic good or evil mental states).

The third item to be considered, namely whether actions are censured or praised by the wise, can be interpreted either as virtue ethics (virtuous persons determine what is good or evil) or as a social convention (society or a particular group of society decides what is good or evil). The fourth item to be considered clearly refers to the consequences of actions. Again, the harmful and painful consequences are neither said to be derived from the unwholesomeness of actions, nor from greed, hatred, and delusion. So, it can be argued that the fourth item of the criterion is utilitarian because it takes into account the resulting happiness or suffering of actions.

Another Pāli text relevant for our discussion of the early Buddhist criterion of goodness can be found in (MN II.114), within the Bāhitika Sutta. There the king Pasenadi of Kosala asks the monk Ānanda whether or not the Buddha performs actions censured by wise recluses and brahmins. The king asks what action is censured by the wise, Ānanda replies that unwholesome action, which is further explained in subsequent questions as action that is blamable, action that brings affliction, action with painful results, and action that “leads to one’s own affliction, or to the affliction of others, or to the affliction of both, and on account of which unwholesome states increase and wholesome states diminishes. Such bodily…verbal…mental behavior is censured by wise recluses and brahmins” (Bhikkhu Ñāṇamoli and Bhikkhu Bodhi 1995: 724-726).

It seems possible to interpret the text as suggesting a criterion of goodness characteristic of virtue ethics. The goodness of actions derives from the wholesome virtue or root mental motivation associated with them; it is due to this intrinsic unwholesomeness of actions that they are blamable, bring affliction and painful results to oneself and others, and that is why they are censured by the wise. I disagree with this interpretation because the text does not say that the consequences derive from the unwholesomeness of actions. The three roots of the unwholesome are not mentioned, and, as I have already said, the term unwholesome refers not only to internal mental actions (the last three of the ten unwholesome actions), but also to external bodily and verbal actions (the first six of the ten unwholesome actions, and the four precepts).

It seems also possible to interpret the text as suggesting a criterion of goodness characteristic of consequentialism. In fact, this is precisely the interpretation of Premasiri (1987). I also disagree with his consequentialist interpretation because the text does not explicitly state that the unwholesomeness of actions derives from their consequences.

In my view, neither a consequentialist nor a virtue ethics reading of the text do justice to this early Buddhist criterion of goodness. If it were true that only the consequences count, I do not see why the text mentions the unwholesomeness and blamefulness of actions. Conversely, if only the unwholesomeness and blamefulness counts, why are the consequences so much emphasized? I do not think that the point of the text is to justify any exclusive interpretation of Buddhist ethics, be it virtue ethics or consequentialism. I prefer to interpret this criterion as the other criteria I have discussed, that is, as advising the consideration of actions from different angles. These angles generate diverse but complementary perspectives of the goodness of actions. From one point of view, one calculates the happiness or suffering resulting from actions. From another point of view, one assesses the motivation or mental states underlying actions. From still another point of view, one ponders the kind or content of actions. The overall goodness of actions depends on several factors, not just on one of them, be it the consequences, the mental states, or the nature of actions.

In conclusion, the criteria of goodness in the Pāli Nikāyas does not correspond to a form of virtue ethics as Keown defines it, but rather to a particular system of virtue ethics with features of utilitarianism and moral realism. This does not mean that early Buddhist ethics is a form of moral particularism with a plurality of moral theories, namely, virtue ethics, utilitarianism, and moral realism (Hallisey, 1996). Early Buddhist ethics is sui generis, that is, one of a kind, different from other traditions of virtue ethics known in the West. As Peter Harvey rightly says “the rich field of Buddhist ethics would be narrowed by wholly collapsing it into any single one of the Kantian, Aristotelian or Utilitarian models.” (2000: 51).

Early Buddhist ethics considers mental action as more important than bodily and verbal action (MN I.373): “I describe mental action as more reprehensible for the performance of evil action, for the perpetration of evil action, and not so much bodily action and verbal action.” Similarly, early Buddhism equates intention (cetanā) and karma (AN III.415), which does not mean that only intention or motivation constitutes moral actions, but rather that without intention actions do not generate karma. That is, intention is the basic requirement for speaking about moral actions within Buddhism. This primacy of the mind and intention in early and classical Buddhism seems to indicate that the mental states behind actions are the most important factor to determine the goodness of actions. I do not deny this classical tenet of Buddhist ethics. However, this primacy of mental action and intention does not mean that within early Buddhist ethics the consequences or the content of actions are irrelevant for determining the goodness of actions. Generally speaking, it might be said that Abhidharma ethics tend to emphasize the mental states behind actions, Mahāyāna ethics the consequences of actions for the suffering of all living beings, and Vinaya ethics the rules and the content of actions. Early Buddhist ethics, however, tend to integrate in its criteria of goodness the three factors: motivation and content of actions (wholesomeness, blamelessness) and their consequences (harmless and happy results for oneself and others).

Early Buddhist ethics can be considered a system of virtue ethics if virtue ethics is understood more in keeping with the Aristotelian model and Christian traditions of virtue ethics. That is, as a system where the summum bonum is not identified with the moral domain of the good and the right, where there are both instrumental and teleological goods, where observing certain sets of precepts or laws is morally good even when the motivation is not virtuous, and where the consequences of actions are also taken into account to assess the overall goodness of actions. However, if virtue ethics is understood as Keown does, that is, as an ethical theory where the good and the right is identified with nirvana, where only actions displaying or participating in nirvanic virtues are good, and where motivation or intention is the only factor to determine the goodness of actions, then I think that calling early Buddhist ethics a system of virtue ethics is problematic.

I believe that the interpretation of the early Buddhist criteria of goodness I have developed in this article has important consequences for our understanding of Buddhist ethics as a whole. First, if what I have said about the early Buddhist criteria of goodness is plausible, then Keown’s view, according to which actions in Buddhism are good or evil depending on just the agent’s motivation, has to be modified. Besides motivation, Buddhists, at least those who consider authoritative the Pāli Nikāyas, have to take into account the consequences of actions for the happiness of oneself and others, as well as the intrinsic good or evil of certain actions. Second, the widespread belief according to which Mahāyāna ethics constitute a radical departure from early Buddhist ethics has to be qualified. It is true that Mahāyāna ethics is more consequentialist than earlier Buddhism, but it is also true that Mahāyāna ethics builds upon the utilitarian features of early Buddhism. Like the utilitarian features of early Buddhism, the consequentialist features of Mahāyāna ethics do not imply that the goodness of actions depends exclusively on the consequences. The content of actions matters, and the bad karma associated to certain actions is acknowledged even in Mahāyāna texts where someone is killed in order to preserve the happiness of the greatest number (Harvey 2000: 135-136).

Similarly, I believe that the heuristic distinction I have introduced in this article has also important repercussions for our descriptions of Buddhist ethics. First, the distinction privileges ideal types of ethical practice without excluding or marginalizing less ideal but equally important types of Buddhist practice. Second, the distinction expands Keown’s conception of virtue ethics, making it even more consistent with Aristotelian ethics and other Western traditions of virtue ethics.

Abbreviations

All references to the Pāli texts are to the edition of the Pāli Text Society, Oxford. References to the Aṅguttara, Dīgha, Majjhima and Saṃyutta Nikāyas are to the volume and page number.

AN Aṅguttara Nikāya

DN Dīgha Nikāya

MN Majjhima Nikāya

SN Saṃyutta Nikāya

Bibliography

Aronson, H.B. “The Relationship of the Karmic to the Nirvanic in Theravāda Buddhism.” in: Journal of Religious Ethics. Vol. 7.1. 1979, p. 28-36.

Aronson, H.B. Love and Sympathy in Theravāda Buddhism. Delhi: Motilal Barnasidass, 1980.

Ñāṇamoli, Bhikkhu and Bhikkhu Bodhi. The Middle Length Discourses of the Buddha. A New Translation of the Majjhima Nikāya. Kandy: Buddhist Publication Society, 1995.

Bunnag, J. Buddhist Monk, Buddhist Layman. Cambridge: Cambridge University Press, 1973.

Cousins, L. S. “Good or Skilful? Kusala in Canon and Commentary.” in: Journal of Buddhist Ethics. Vol. 3, 1996, p.133-164.

Dalai Lama. The Way to Freedom. San Francisco: Harper, 1994.

Dharmasiri, G. Fundamentals of Buddhist Ethics. Singapore: Buddhist Research Society, 1986.

Ekvall, R. B. Religious Observances in Tibet. Chicago and London: University of Chicago Press, 1964.

Gombrich, R. Precept and Practice: Traditional Buddhism in the Rural Highlands of Ceylan. Oxford: Clarendon Press. 1971. See also “Merit transference in Sinhalese Buddhism.” in: History of Religions Vol. 11, 1971, p. 203-219.

Hallisey, C. “Ethical Particularism in Theravāda Buddhism.” in: Journal of Buddhist Ethics. Vol. 3, 1996, p. 32-34.

Harvey, P. “Criteria for Judging the Unwholesomeness of Actions in the Texts of Theravāda Buddhism.” in: Journal of Buddhist Ethics. Vol. 2, 1995, p.140-151.

Harvey, P. An Introduction to Buddhist Ethics. Cambridge: Cambridge University Press, 2000.

Kalupahana, D. Buddhist Philosophy: A Historical Analysis. Honolulu: University Press of Hawai’i, 1976.

Katz, N. Buddhist Images of Human Perfection. Delhi: Motilal Barnasidass, 1982.

Keown, D. The Nature of Buddhist Ethics. New York: Palgrave, 1992.

Keyes, C. F. “Merit-Transference in the Kammatic Theory of Popular Theravāda Buddhism.” in: Karma: An Anthropological Inquiry. Ed. C. F. Keyes and E. V. Daniel. Berkeley: University of California Press, 1983, p. 261-286.

Keyes, C. F. “Buddhist Economics and Buddhist Fundamentalism in Burma and Thailand.” in: Fundamentalisms and the State. Ed. Martin E. Marty and R. Scott Appleby. Chicago and London: The University of Chicago Press, 1993, p. 367-409.

King, W. L. In the Hope of Nibbāna: An Essay on Theravada Buddhist Ethics. Chicago: Open Court Publishing, 1964.

Kraut, R. “Aristotle on the Human Good: An Overview.” in: Aristotle’s Ethics. Ed. Nancy Sherman. Lanham: Rowman & Littlefield, 1999, p. 79-104.

Lichter, D. and Epstein, L. “Irony in Tibetan Notions of the Good Life.” in: Karma: An Anthropological Inquiry. Ed. C. F. Keyes and E. V. Daniel. Berkeley: University of California Press, 1983, p. 223-259.

Little, D. and Twiss, S. B. Comparative Religious Ethics. New York: Harper and Row, 1978.

Little, D. “Ethical Analysis and Wealth in Theravāda Buddhism: A Response to Frank Reynolds.” in: Ethics, Wealth, and Salvation: A Study in Buddhist Social Ethics. Ed. R.F. Sizemore and D. Swearer. Columbia: University of South Carolina Press, 1990, p. 77-86.

Perret, R. “Egoism, Altruism, and Intentionalism in Buddhist Ethics.” in: Journal of Indian Philosophy. Vol. 15, 1987, p. 71-85.

Premasiri, P. D. “Early Buddhist Concept of Ethical Knowledge: A Philosophical Analysis.” in: Buddhist Philosophy and Culture: Essays in Honor of N.A. Jayawickrema. Ed. D. Kalupahana and W.G. Weeraratne. Colombo: N.A. Jayawickrema Felicitation Volume Committee, 1987, p. 37-70.

Samuel, G. Civilized Shamans: Buddhism in Tibetan Societies. Washington, London: Smithsonian Institution Press, 1993.

Sherman, N. The Fabric of Character: Aristotle Theory of Virtue. Oxford: Clarendon, 1989.

Spiro, M. E. Buddhism and Society: a Great Tradition and its Burmese Vicissitudes. London: George Allen and Unwin, 1971.

Swearer, D. K. The Buddhist World of Southeast Asia. New York: State University of New York Press, 1995.

Tatz, M. and Kent, J. Rebirth: The Tibetan Game of Liberation. New York: Anchor Books, 1977.

Tucci, G. The Religions of Tibet. London: Routledge and Kegan Paul, 1980.

Copyright 2004

Return to beginning of article

---

Index | Back Issues | FAQ | Global Resources | Current Issue | Scholarly Resources | Pali Canon | About the Journal | PDF | Submissions | Subscribe | Online Conferences | Book Reviews | Search