Does Anatman rationally entail altruism? On Bodhicaryavatara 8:101-103.
Next, I consider two reconstructions of Santideva's argument, by Mark Siderits and John Pettit, and argue that both these reconstructions fail to avoid the dilemma identified above. Two conclusions can be drawn from this failure to reconstruct a successful argument from anatman to altruism. First, it suggests that, for Madhyamaka Buddhists, an understanding of anatman does not entail an acceptance of the Bodhisattva path, but rather is instrumental to achieving it. Second, we should question whether the interpretation of Santideva as making an analytic argument for altruism is the correct one.
Santideva's Argument and Reconstruction by Paul Williams
What seems to be the heart of Santideva's argument occurs in chapter 8, verses 101-103 of his Bodhicaryavatara. Wallace and Wallace translate the verses as follows:
The continuum of consciousness, like a series, and the aggregation of constituents, like an army and such, are unreal. Since one who experiences suffering does not exist, to whom will that suffering belong? All sufferings are without an owner, because they are not different. They should be warded off simply because they are suffering. Why is any restriction made in this case? Why should suffering be prevented? Because everyone agrees. If it must be warded off, then all of it must be warded off; and if not, then this goes for oneself as it does for everyone else. (102)
Santideva's audience, presumably, is a Buddhist who accepts there is no self but rejects the characteristic Mahayana position that we ought to act altruistically. Throughout this essay, I will use "altruism" to refer to the position that one should have an impartial concern for one's own and others welfare, and should strive to remove all suffering, regardless to whom it belongs. (2) Verse 101 points out that since a person is a partite entity, accepting that partite entities do not exist entails accepting that persons are fictions. Suffering, if it exists, belongs to no one. Since we accept that selves do not exist, and since all suffering is alike in being negative, there is no rational reason to prioritize removing my own suffering. Verse 103 draws the conclusion that either we ought to commit to remove everyone's suffering or we ought to accept that it isn't worthwhile to remove any suffering. Because everyone, however, agrees that suffering ought to be removed, we ought to commit to altruism. (3)
The above is a brief sketch of what seems to be Santideva's general strategy. Commentators, however, have differed considerably on how to interpret the details. Paul Williams interprets Santideva as arguing that the moral imperative to place others's happiness alongside one's own in importance can be drawn from the ontological fact of the nonexistence of the self. Santideva then, pace Hume, is attempting to derive an ought from an is (Studies 104). Drawing upon the Tibetan commentarial tradition, Williams interprets Santideva's argument as an appeal to our rationality. Once we understand reality correctly, and accept the self is nonexistent, it will no longer be rational to remove one's pain before removing the pain of others (Studies 105). (4)
One puzzling feature of Santideva's argument is that it appears to depend upon the reductionist thesis that although partite objects like selves do not exist, momentary mental sensations like pains do. This is odd, since as a Madhyamika, Santideva should hold that nothing has intrinsic existence, but that all entities are convenient fictions depending on their causes and conditions and conceptual labeling for their existence. Although he is not explicit about this, Williams seems to concur with Mark Siderits in holding that Santideva here is provisionally accepting an Abhidharmika position ("Reality," 421). (5) Although Abhidharmikas reject the existence of partite objects, they do accept that there are really existing physical and mental events, fundamental constituents of the universe called "dharmas." Pain would be a species of sensation (vedana), one of these mental dharmic events; so pain, for the Abhidharmika, ultimately exists, even though the person does not. Santideva's argument, then, would be an example of skillful means (upaya) directed toward an Abhidharmika Buddhist, to show that even the ultimately incorrect Abhidharmika understanding of reality, in which pain is an ultimate constituent of reality, entails acceptance of the Mahayana tenet of altruism.
Both Abhidharmika and Madhyamaka Buddhists hold that although everyday entities like tables, chairs, and selves do not exist intrinsically, they still have conventional validity as transactional items in daily life. The conventional self, then, is not denied; rather it is the false conception of a self existing independently of its causes and conditions that is rejected. Williams, however, argues that to make sense of Santideva's argument, he must be seen as denying the reality even of this conventional transactional self (Studies 112). This is because accepting the conventional self would allow "all the normal everyday transactional distinctions to be made" (Studies 110). This would allow us, contra Santideva's intended conclusion, to prioritize our future conventional self over the selves of others. Williams puts the point as follows:
In order to give precedence to (a) this person over the interests and rights of (b) that person, all I have to do is be capable of making a distinction between (a) and (b). This distinction can be made in various ways, but in our common experience it rests on whatever normal everyday distinctions are indeed made between (a) and (b), between me and Archibald. Therefore if there is any difference at all between me and Archibald, if we are different persons, I can still be selfish. I can still put myself first. (Studies 111)
Williams, therefore, concludes that for Santideva's argument to work, it must be interpreted as an argument against not only an intrinsically existing self, but also a conventional transactional self. Williams, however, points out that this would be a disaster. It is within the everyday transactional realm that conventional selves perform salvific activities (Studies 164-165). In denying the conventional self, therefore, Santideva has destroyed the Bodhisattva path (Studies 174-176). (6)
It is true that Santideva claims the self does not exist, but he does not state explicitly whether only the intrinsically existing self, or both this self and conventional transactional self, is being denied. Since, as Williams points out, the results of denying not only the intrinsically existing, but also the conventional self would be disastrous, this view should not be attributed to Santideva unless we have strong reason to believe this is what he meant. In fact, in most of his text, Santideva makes liberal use of the conventional self, who is being instructed in the ways of the bodhisattva. (7) It seems, therefore, that this is unlikely to be what Santideva intended. (8)
Williams justifies attributing his interpretation to Santideva on the grounds that Santideva's argument will not work without it. As long as we can make a distinction between my future conventional self, and the conventional self of other persons, then it is still possible to prioritize my own welfare. Nevertheless, we need not interpret Santideva as making the implausible claim that realizing the truth of anatman somehow makes us no longer able to make distinctions among conventionally labeled entities. As Barbra Clayton insightfully points out, Santideva's reasoning here parallels traditional anti-discrimination arguments (91). Since all beings equally desire happiness, if it is ethical for me to prioritize the welfare of my future self, I ought to be able to point out some relevant distinction that justifies this. If my conventionally existing future self were ultimately identical to my current self, this would provide the relevant distinction. Accepting the truth of anatman, however, means accepting the identity of my current and future self is merely a convention. Therefore, assuming no other relevant distinction can be given, it follows that ethically I cannot justify prioritizing my own welfare above others; if I am committed to removing suffering, I ought to commit to removing all of it. (9)
An interpretation of Santideva's argument may be given, therefore, that does not attribute to him the implausible thesis that conventional selves do not exist even conventionally. Santideva can be seen as claiming that although conventional selves exist, when we realize they are only convenient fictions we will accept that we should not prioritize our own conventional self's welfare above that of other persons. (10) However, if we emphasize the lack of real identity between present and future conventional selves, it becomes open to Santideva's opponent to ask why we should care about the welfare of anyone at all, including of our future conventional self. (11) Since the identity of my present and future conventional self is merely a fiction, why should this understanding not motivate apathy instead of altruism?
Santideva's response to this charge comes in verse 8:103: Why should suffering be prevented? Because everyone agrees. If it must be warded off, then all of it must be warded off; and if not, then this goes for oneself as it does for everyone else. (Wallace & Wallace 102)
We need not have a debate on whether suffering is worth removing, Santideva is claiming, because everyone is in agreement that it ought to be. The opponent, however, can respond that most people agree to this because they believe their self, and the selves of the ones they care about, endure. If the truth of anatman became widely realized, it might also be the case that there would no longer be widespread agreement that pain needed to be removed. There might be some altruistically minded persons who would be inclined to remove future pain, no matter who it belongs to; however, there might be others who would simply stop caring about their future conventional selves along with the selves of others. Here, we should note that in verse 103 Santideva gives us a choice: either we should commit to eliminating all suffering no matter to whom it belongs, or we should stop removing any of it. The opponent can ask, what is to stop us from choosing the option of removing no suffering at all?
An obvious response to this objection would be to point out that close ties of physical and psychological continuity between current and future conventional selves motivate caring for their welfare. If Santideva were to make this response, however, then he would be committed to the view that we have a reason that can justify prioritizing the welfare of our future conventional self, since our ties to our future conventional self are particularly strong. Santideva's argument, then, faces a dilemma. If he claims that the conventional endurance of selves provides rational motivation for removing future pain, then it would seem that my conventional identification with my future conventional self would be enough to rationally motivate my prioritizing its welfare over other selves. However, emphasizing the lack of any ultimate identity between present and future conventional selves opens the possibility that it is perfectly rational not to care about the future at all. Let us refer to this as Santideva's dilemma. (12) In short, emphasizing the importance of the conventional self opens the door to egoism, because my future self is conventionally identical to me. However, when we emphasize the fictitious nature of the conventional self, then the danger of nihilism arises, for the opponent can object that there is no longer reason to care for the suffering of anyone. (13) Rather than entailing altruism, Santideva's argument ends up providing a justification for total apathy.
Reconstruction by Mark Siderits
We have seen that there is a dilemma involved in reconstructing these passages as an argument that anatman entails altruism. On the one hand, emphasizing the ultimate nonexistence of the self leads us to doubt whether we are obligated to eliminate any suffering whatsoever. On the other hand, emphasizing the importance of conventionally existing selves leads us to prioritize the welfare of our future conventional selves, which are conventionally identical to our current selves.
However, this dilemma relies upon a hidden assumption: we are only obligated to remove pain if there are selves to experience this pain. Mark Siderits interprets Santideva as rejecting this assumption. He reconstructs a key portion of Santideva's argument as follows:
The person, being an aggregate, is ultimately unreal. Hence if pain is ultimately real, it must be ownerless or impersonal. It is universally agreed that pain is bad ("to be prevented"), although this agreement is typically restricted to one's own case. Either pain is ultimately and impersonally bad, or no pain (including what is conventionally one's own) is ultimately bad. But it is absurd that no pain should be ultimately bad. Therefore pain is ultimately and impersonally bad. (Personal 103)
While Buddhists hold that persons are ultimately unreal, at least some Buddhists hold that pain is ultimately real. (14) Since persons are unreal, this truly existing pain must be impersonal. There are now two possibilities. Either this impersonal pain is bad, and should be removed; or it is not bad, and need not be removed. Since there is common agreement that pain is bad, this point is not in dispute. Therefore, pain, which exists impersonally, ought to be removed.
In Siderits' interpretation, it is now a short step to the conclusion that we ought to make an equal commitment to removing the pain of everyone (altruism). It is our commitment to removing this pain, which ultimately exists impersonally, that leads to us adopting personhood conventions in which we identify our current and future collection of causal constituents as being the same person. Such a convention aids in removing pain. For instance, my current set of causally connected constituents are in a particularly advantageous position to prevent the pain of gingivitis from arising in my future causal continuum (Personal 103). (15)
In Siderits' view, the Buddhist holds that we conventionally identify ourselves as enduring persons solely as a stratagem for removing as much of this ultimately impersonal pain as possible. Very often the best stratagem to removing as much pain as possible is to pay particular attention to our future welfare, since the close causal connections between our current and future conventional selves allow us to have a particularly strong effect on its welfare. However, we should keep in mind that the personhood convention is merely a useful fiction adopted for maximal efficiency in removing pain. For this reason, there is no reason to give any precedence to our own happiness, unless doing so maximizes removal of pain overall. In cases where sacrificing my own happiness would result in a greater decrease overall in suffering, I should do that instead (Personal 103). (16) In other words, I should adopt a commitment to altruism, and commit to removing the suffering of everyone in as quick a manner as possible, regardless of its owner.
We could reconstruct the essence of Siderits's argument as follows:
1. Persons do not exist ultimately.
2. Pain does exist ultimately.
Conclusion 1: Therefore, pain must exist impersonally.
3. We should remove this impersonally existing pain, since it is bad.
4. Since persons exist only as a useful fiction to maximize removal of pain, we have no reason to prioritize removing our own pain.
Conclusion 2: We ought to act altruistically, that is, we ought to remove all pain as quickly as possible, regardless of its owner. (17)
There are two potential problems with this argument. First, as Williams points out, it is not clear a Buddhist would accept the antecedent of premise 4 ("Response to Siderits" 437). Buddhists hold that we identify ourselves as enduring persons due to the influence of beginningless ignorance, and rather than helping remove pain, this very identification causes a great deal of suffering. A Buddhist, then, might claim that rather than being a useful stratagem for reducing pain, identifying ourselves as persons conditions the existence of any suffering whatsoever.
To this objection, the Buddhist can reply that suffering arises only as a result of misidentifying the series of mental and physical moments (skhandas) as an enduring person (atman). So long as one recognizes "person" is only a convenient designation (prajnapti) for the series, however, no clinging toward a person arises, and suffering does not result. (18) Siderits clearly intends only this conventional designation of person to be adopted for pragmatic purposes, so it is no longer clear his account departs from Buddhist orthodoxy.
Second, Santideva's opponent can respond that insufficient defense has been given to premise three. At first glance, this premise seems unobjectionable. People generally accept that pain and suffering are negative and, all things being equal, ought to be removed no matter to whom they belong. Here, there are two arguments the opponent may make in response to Santideva. Both responses attempt to use Santideva's invocation of anatman against him.
In the first, the opponent might claim that pain, impersonally conceived, is not really negative; it is only because we falsely identify ourselves as enduring selves that pain becomes suffering. Plausibly, a distinction can be made between pain, which can be conceived of as a neutral experience, and suffering, which is inherently negative. Consider the experience of running a marathon. For most of us, the feeling of exhaustion experienced while running would be one of great suffering; yet for some athletes, the same set of sensations might be relished as integral to the satisfying experience of pushing one's body to its furthest limit. Similarly, persons have different levels of pain tolerance; sensations one person might find distressing another may find inconsequential. From these examples, the opponent might draw the conclusion that it is only sensations of pain that are interpreted as negative that are experienced as suffering. He could then conclude that, if it is only these painful sensations that are ultimately real, and pain is not of itself negative, pain need not be removed. Suffering, further, would be unreal, so no effort needs to be made to eliminate it.
Such an argument might not be convincing for several reasons. First, it might be claimed that certain experiences of the sensation of pain are inherently negative--for instance, the excruciating pain of a dentist performing a root canal without any kind of anesthetic. Likewise, even if all physical pain only becomes suffering when interpreted as negative, it seems that mental pain is intrinsically negative. It is hard to see how Santideva's opponent could claim that a moment of deep depression, of itself, is a neutral experience that becomes suffering only when interpreted in a certain way. Yet Abhidharmika Buddhists recognize mental factors such as a moment of depression as dharmas that have ultimate existence. Therefore, plausibly, Santideva might respond that at least some painful experiences are intrinsically negative, and therefore provide some reason to be removed.
Let us assume with Santideva, then, that there are some mental states that are intrinsically negative, regardless of whether or not they belong to persons, and turn to the second hypothetical objection. Here, the opponent can question whether the existence of these negative mental states entails the ethical conclusion that we ought to prevent or remove them. (19) In verse 8:102, Santideva claims we should remove suffering simply because it has the nature of being suffering. The phenomenal feel of suffering is negative and as such it is to be removed. Nevertheless, the opponent may object that simply pointing out the fact that a mental state is negative does not obviously entail the normative conclusion that we ought to prevent or remove it. Of course, generally we are motivated to remove these negative mental states in ourselves, those we love, and even strangers, but arguably this may be the case only because we falsely believe ourselves and others to be enduring persons. Once anatman is accepted, it is no longer clear why we should automatically transition from the fact that suffering exists to the normative conclusion that we ought to remove it. The opponent, here, need not make the stronger claim that it is irrational to remove suffering; rather, all he or she needs to do is to claim that accepting anatman makes it no longer clear why we are ethically obligated to remove anyone's suffering.
Santideva anticipates this response, and in the following verse he appeals to commonly held intuitions to shift the burden of proof to his opponent. "Why should suffering be prevented? Because everyone agrees" (Wallace & Wallace 102). We need not have a discussion about whether suffering needs to be removed, he is claiming, because there is very little disagreement on the matter. If the opponent wants to dispute this claim, since he is going against common opinion, it is he who owes an argument as to why we are not ethically required to remove suffering.
Santideva, then, appeals to common ethical intuitions to support his position that suffering ought to be removed. Here, however, we need to remember that common intuitions support not only the claim that we ought to remove suffering, but also the claim that it is ethically allowable to prioritize one's own happiness over the happiness of other persons. There are, then, two principles supported by commonplace ethical intuitions that are relevant to Santideva's argument. Let us call them, respectively, the "Principle of Moderate Benevolence" and the "Principle of Moderate Egoism."
Principle of Moderate Benevolence (PMB): Provided it would not be overly demanding, we ought to remove suffering, no matter to whom it belongs. Principe of Moderate Egoism (PME): It is not immoral to prioritize our own welfare above that of other persons.
There is no obvious conflict in accepting both these principles, and indeed commonplace ethical intuitions seem to support the acceptance of both. Most people hold both that we ought to care about the suffering of others, and that there is nothing unethical about giving somewhat greater priority to our own welfare. In his argument, Santideva invokes the principle of anatman to undercut the ethical intuitions supporting PME. Santideva argues that since there are no enduring selves, it would be irrational, and therefore unjustifiable, to prioritize the welfare of my own enduring self above the welfare of other persons. The opponent, however, may claim that the invocation of anatman also undercuts the intuitions supporting PMB. It is at least plausible that the common belief that we should remove suffering is also bound up with the belief that suffering belongs to enduring beings.
The opponent, then, claims that accepting anatman may likely undercut existing intuitions that suffering should be removed. The objection will not work if our intuitions support the conclusion that impersonal suffering should be removed. However, it is not clear that we have any intuitions about impersonal suffering. Consider a world where suffering exists in isolated mental flashes completely disconnected from any other mental events. Do our intuitions support the conclusion that we ought to remove the suffering in this world? As far as I can tell, I have no intuitions about this world. Although I grant its logical possibility, it is not clear to me that I have any ethical obligations toward it. (20) Presumably, commonplace ethical intuitions hold that the suffering of persons, all things being equal, ought to be prevented. It is not clear, however, that these same intuitions can be transferred to support the conclusion that momentary instances of impersonal suffering also ought to be prevented.
The opponent's charge, in brief, is that if anatman is invoked to undercut ethical intuitions, then it is unfair to simply cherry pick which ethical intuitions will be undercut, and which will remain intact. Of course, this is not an argument that PMB is incorrect. Rather, what is really at issue is the question of burden of proof. Santideva has shifted the burden of proof to his opponent by pointing out that there is common acceptance that suffering ought to be removed. The opponent, however, may respond that once anatman has been accepted, it is irrational to accept any intuitions that arise in dependence on the belief in enduring persons. Further, it seems likely that existing intuitions that suffering ought to be removed are dependent on the belief in enduring persons. Therefore, the burden of proof is returned to Santideva, and we are owed a new argument for why impersonal suffering ought to be prevented. (21)
Of course, such an argument for removing suffering could be fashioned by appealing to the importance of facts that are conventionally labeled "persons," such as close causal connections and continuity of consciousness. Emphasizing the normative value of such impersonal facts, however, brings to light the other horn of Santideva's dilemma. Since my current moment of mind is located in one such closely related stream of mental and physical events, it would seem that I have a good reason to prioritize the welfare of those moments of mind most closely related to my current moment of mind. Emphasizing the normative significance of the conventional person provides a justification for moderate egoism, even if we understand such persons are ultimately reducible to impersonal facts.
Our discussion thus far allows us to now state more clearly what lies behind this dilemma. Ordinarily, ethical arguments incorporate premises based upon commonly accepted intuitions. Santideva himself does this by arguing that we ought to remove suffering, giving as a reason everyone's agreement that it should be removed. In other words, it is simply an ethical intuition that suffering is bad and ought to be gotten rid of when possible. However, Santideva also uses a premise, anatman, which is counterintuitive, and if true, might undercut many of our basic ethical intuitions. Indeed, he uses anatman to argue that the commonly accepted intuition that we are justified in prioritizing our own welfare is false. The difficulty is that, as we have just seen, accepting the truth of anatman also throws into question ethical intuitions Santideva wants to appeal to in arguing that we ought to remove suffering. (22)
The underlying cause of the difficulty faced by Santideva's argument, then, seems to be his incorporation of premises taken from the levels of both conventional and ultimate truth. Ethical intuitions that support the limited egoistic position that we are justified in prioritizing our own welfare, as well as the position of limited benevolence that says we ought to have at least some concern for the welfare of others, most plausibly arise in dependence on conventional truth and its pragmatic employment of personhood. It is less clear that if we eliminated persons from our conception of the world, that any obligation to remove any suffering would be intuitively recognized. On the other hand, Santideva appeals to the Abhidharmika understanding of the way things ultimately exist by incorporating the truth of anatman into his argument. By switching between these levels, Santideva is able to both argue that we ought to honor our intuitions that suffering should be removed (level of conventions) and also argue that since selves do not exist, it would be irrational to prioritize our own welfare (level of ultimate truth).
What I have called "Santideva's dilemma," in turn, may be seen as a way of reminding Santideva of the baggage that comes with arguing from one level or the other. If one argues from the level of ultimate truth, some new justification will be needed for our obligation to remove suffering; appealing to intuitions based on conventions will not suffice, since these may be undercut by an acceptance of anatman. On the other hand, if one argues from the level of conventions, some new justification is needed for the obligation to act altruistically, since our ordinary intuitions do not support this conclusion.
Our discussion here raises the intriguing question of to what extent a Madhyamaka Buddhist's argumentation should be based in conventional truth, and to what extent he may legitimately draw upon considerations derived from the viewpoint of ultimate reality. For our purposes, we must restrict ourselves to noting that employing premises drawn from both levels in a single argument will likely result in friction. In this particular case, Santideva's appeal to the ultimate truth of anatman results in undermining both the conventionally based intuitions he wants to reject, the egoist's appeal to our common belief that we are justified in prioritizing our own welfare, as well as the conventionally based intuition that suffering should be removed that he requires to complete his argument.
There are, however, ways of reconstructing Santideva's argument that do not obviously depend upon this seemingly illicit switching between the levels of ultimate and conventional truth. John Pettit, in his response to Williams, provides such an interpretation, and it is to this that we now turn.
Reconstruction by John Pettit
As I have above, in his reply to Williams, Pettit also accuses Williams of misidentifying Santideva's object of negation in these verses, although his analysis differs from my own. Pettit suggests that Williams takes Santideva's object of negation to be the Atman of Upanisadic philosophy (132). This is the ultimately existing, independent and self sufficient self that all Buddhists alike deny. Pettit agrees that Santideva would deny the existence of any such self, but claims his true object of negation is the innate conception of self (132). Drawing upon the later Tibetan philosopher Tsong kha pa, he describes this self as the innate sense of "I" that arises, for instance, when we are verbally abused. It is this innate self grasping which is the source of our conflicting emotions, and would therefore be the primary object of Santideva's concern (133). By contrast, the conventional self would not be denied by Santideva. This conventional self is a convenient fiction, the mere name that is given to the continuum of physical and mental events in close association. Such a conventional self would not give rise to conflicting emotions, and therefore need not be negated.
Pettit is certainly correct in pointing out that Madhyamaka thinkers make a distinction between the self believed in as a result of faulty philosophical thinking, here named the Upanisadic self, and the innate self-clinging common to all unliberated sentient beings. It is not clear to me that it is this innate self grasping, rather than the Upanisadic self, which is the primary object of negation in the verses we have looked at. However, it seems plausible that Santideva meant to deny the existence of both these selves in these verses, and so I will not argue the point.
In any event, Pettit believes that once we identify the innate sense of ego grasping as the correct object of negation, then Santideva's argument becomes plausible. This is because once innate ego clinging has been eliminated, there will no longer be any impediments to putting the happiness of others alongside one's own in importance.
Given these considerations, Williams' thesis that 'the ought' of unselfishness simply does not follow from the 'is' of anatman can be brought seriously into question. If one's rigid sense of separation (via emotional clinging and rejection) from others can be dissolved through demolishing the misconception of a vulnerable emotional self, then naturally one's selfishness, stemming as it does from a false belief in that self, will likewise be eliminated. (133)
Pettit is certainly correct that one of Santideva's primary concerns in the Bodhicaryavatara is eliminating disturbing afflictions, and the innate sense of self grasping which is their root. Further, he is right to point out that Santideva holds that understanding the truth of anatman has instrumental value in eliminating the obstacles to compassionate action. Nevertheless, his configuration of Santideva's argument does not bridge the is/ought gap; further, eliminating innate ego clinging, in itself, does not normatively entail altruistic action, any more than eliminating the mistaken belief in the Upanisadic self would.
To see why, we first need to disambiguate what Pettit means when he refers to "the ought of unselfishness" that supposedly follows from the "is" of anatman. One possibility is that "unselfishness" refers to the psychological state of being without ego grasping that results when our innate conception of self has been eliminated. This, however, is a description of a psychological state, and is not itself a normative conclusion. Pointing out that a state of emotional equanimity would result from eliminating our innate conception of self does not, of itself, tell us what actions we are obligated to perform. In this interpretation, Santideva would have derived an is (emotional equanimity) from an is (lack of self), and a further argument is needed to explain why an emotionally equanimous person ought to act altruistically.
On the other hand, it is also possible to interpret Pettit as meaning by "unselfishness" the unselfish activity of removing the suffering of others. This would be a normative conclusion: once we realize the absence of any innate self (what is the case), then we must accept that we should eliminate the suffering of others (what ought to be the case). Un fortunately, the conclusion does not follow from the premise. It is true that if we eliminate the innate sense of self grasping, we will have eliminated the most significant obstacle to the altruistic attitude of putting the welfare of others on par with our own. However, this of itself does not entail that we ought to remove the suffering of others--only that we are able to do so. Removing ego grasping, then, is a necessary, but not a sufficient condition for wholly committing to the altruistic path. What is still needed is an obligation, or at least a motivation, to care about the suffering of any beings whatsoever. (23)
Here, we can remember the second horn of Santideva's dilemma framed above. Let us say we have negated both the Upanisadic conception of self, and the innate self grasping which is the root of conflicting emotions. The emotions of craving and greed no longer act as an impediment to our concern for others, and we have achieved perfect equanimity in regard to self and others. Here, we can remember, Santideva has presented us with a choice: either all suffering must be eliminated, or none of it. Again, we can ask, why should we be required to choose the first option? Complete altruism and total apathy are alike in being apparently rational states for an equanimous being to adopt.
We should conclude, therefore, that pace Pettit, even identifying Santideva's object of negation as innate self grasping will not successfully complete an argument that rationally, we ought to act altruistically if we accept the truth of anatman.
Conclusion: Did Santideva Really Provide an Argument for Altruism?
Above, I have considered two reconstructions of Santideva's purported argument that accepting the truth of anatman entails one should act altruistically; we have seen that the success of both arguments is doubtful. Of course, further reconstructions of Santideva's position might be attempted, and would have to be evaluated on their own merits. Nevertheless, these reconstructions would have to find some way of resolving Santideva's dilemma identified above. On the one hand, if we emphasize the merely imputed and fictitious nature of the conventional self, then it will be open to the opponent to inquire why we ought to care about anyone's welfare, even that of the future being conventionally identified as identical to my present self. In this horn of the dilemma, it becomes unclear why understanding the truth of anatman should entail altruism rather than apathy. On the other hand, we can reestablish a motivation to help all sentient beings by emphasizing the importance of the conventional self. However, since my future conventional self is conventionally identical to my current conventional self, it would seem that it would not be irrational to prioritize its welfare over other conventional selves. As I emphasized in my response to Mark Siderits, at the heart of this dilemma is a vulnerability to which Santideva's method of argumentation opens him. Emphasizing the nonexistence of an enduring self may throw into doubt the ethical intuitions the egoist relies on in establishing his position that it is rational to prioritize one's own welfare above others; however, it also threatens to undercut ethical intuitions Santideva relies upon in claiming we ought to care about the welfare of anyone.
From these considerations, we need not conclude with certainty that no argument for altruism using anatman as a premise can be successful; perhaps arguments incorporating other Buddhist principles, such as the (conventional) joy of benefiting others, or obligation to those who have taken care of us in the past, as additional premises might succeed. From my observations above, I draw the more conservative conclusion that it seems unlikely any argument for altruism appealing solely to the principle of anatman, as well as commonly accepted ethical intuitions, is likely to succeed. The verses considered here within Santideva's meditation chapter cannot be characterized in isolation as providing a successful argument from the premise of anatman to the conclusion that we ought to act altruistically.
There are two remaining questions to explore before bringing my own consideration to a close. First, assuming Williams, Siderits, and Pettit were correct in interpreting Santideva as providing in these verses an analytic argument for the entailment of altruism from anatman, should the failure of this argument be troubling to Madhyamaka Buddhists? It seems to me that the answer is no. Instead, the failure of Santideva's argument encourages us to reconsider the relationship between altruism and anatman. Rather than conceiving of anatman as rationally entailing altruistic action, we should view realization of the truth of anatman as instrumental to the foundational Mahayana commitment of altruistic action. As Pettit forcefully argues, it is only by deeply realizing the nonexistence of any enduring self, and eliminating the deep seated grasping that arises from this error, that we are able to give equal priority to the welfare of other beings. Realizing anatman, then, is a necessary condition for completion of the Bodhisattva path, although it does not of itself constitute a reason for entering it.
Second, since on close examination all configurations of these passages as an argument fail to be convincing, we should wonder whether it is correct to interpret these verses as an analytic argument trying to derive the conclusion of altruism from the premise of anatman. The passages are placed not in the "Wisdom" chapter of the Bodhicaryavatara, where Santideva argues against opponents' views, but in a chapter detailing various meditations designed to reduce attachment and generate compassion. (24) My consideration of these passages as an argument has been haunted by a dilemma: either we emphasize the ultimate nonexistence of conventional phenomena like persons, in which case it is unclear why we are obligated to remove suffering at all; or we emphasize the importance of their conventional existence, in which case it seems rational to prioritize our own future happiness, since that self is conventionally identical to us. However, taken as meditational practice, one might switch between viewpoints, depending on one's current mental state. If we feel attachment to self arise, we remember that selves are nonexistent, and the attachment is lessened. If we feel apathetic toward helping sentient beings, then we can focus on their conventional suffering, and empathy will arise. Such a strategy would make sense if, as I have suggested, the bodhisattva has already committed, for independent reasons, to removing all suffering equally.
In fact, in his "Wisdom" chapter, Santideva explicitly endorses this strategy of switching between meditation on the conventional welfare of beings and their ultimate emptiness as a method of respectively nurturing compassion or reducing attachment.
9:76: [Qualm] If there is no sentient being, whose is the task?
[Madhyamika:] True. The effort too, is due to delusion. Nevertheless, in order to alleviate suffering, delusion with regard to one's task is not averted. (Wallace & Wallace 124).
9:77: However, grasping onto the "I," which is a cause of suffering, increases because of the delusion with regard to the Self. If this is the unavoidable results of that, meditation on identitylessness is the best. (Wallace & Wallace 125)
The first verse suggests that there may be times when taking on the provisional belief in an ultimately nonexistent sentient being will be advantageous because this can allow the arising of compassion toward this (ultimately empty) being. Nevertheless, if attachment begins to arise as a result of this provisional acceptance of the person, then one should switch to meditation on anatman instead. Here we see Santideva endorsing the general strategy I have suggested; this accounts, I believe, for the set of verses appearing in the "Meditation" chapter.
Additional support for this reading is provided by the identification of another set of verses in the Bodhicaryavatara in which Santideva switches between focusing on the conventional existence and the ultimate nonexistence of persons as a method of respectively increasing our commitment to developing virtuous states of mind and decreasing our negative mental states. These verses are located in Santideva's chapter on developing patience:
6:31: Thus, everything is dependent on something else, and even that on which something is dependent is not autonomous. Hence, why would one get angry at things that are inactive, like apparitions? (Wallace & Wallace 65)
In this verse, Santideva points out that there is no intrinsically existing entity who is the cause of our anger. Rather, selves are conceptual imputations upon causes and conditions, and therefore exist like apparitions. (25) Here, Santideva appeals to the lack of intrinsic existence of persons (anatman) as a means of quelling our anger toward them. The strategy parallels his appeal to anatman to eliminate our attachment to prioritizing our own welfare in chapter eight, verses 101-103. Santideva next considers an objection quite similar to the objection made by the opponent in the eighth chapter who asks, given the nonexistence of persons, why suffering should be eliminated at all.
6:32: [Qualm] Averting anger is inappropriate, for who averts what?
[Response] That is appropriate, because it is a state of dependent origination and is considered to be the cessation of suffering. (Wallace & Wallace 65)
Here, the opponent claims that since anger and the angry person lack intrinsic existence, it does not make any sense to talk of eliminating anger. Santideva's reference to dependent origination in his reply indicates he is responding from the viewpoint of conventional existence; things do exist, conventionally, in dependence upon their causes and conditions. Therefore, this conventionally existent anger ought to be removed in order to end the conventionally experienced suffering of the conventionally existent person.
One might here object that Santideva cannot have it both ways. If the conventional existence of persons, suffering, and anger is enough to motivate our removal of this conventionally bad state of affairs, then the conventional existence of the person harming us should likewise justify our (conventional) anger against that person. Taken as a meditational technique, however, this tension dissolves. Santideva is not here ontologizing about the existence of mental states and persons, but rather is providing meditational instructions to be used for one already committed to the Bodhisattva path. When anger arises, meditating on the lack of intrinsic existence of one's adversary can be a reliable way to make that anger dissipate. If one's meditation on emptiness begins to corrode one's commitment to creating virtuous states of mind, however, then focusing on the conventional existence of persons will reinstate one's virtuous motivation.
It may be the case, therefore, that no argument in the strict sense of a logical entailment of a conclusion from premises was intended in the set of verses considered above in the eighth chapter of the Bodhicaryavatara. Rather, it may be that Santideva's primary intention was to provide exactly what the title of his chapter on meditation suggests: effective meditational methods of eliminating egoistic selfishness, and nurturing compassion toward others. (26)
Clayton, Barbra (2001). "Compassion as a Matter of Fact: The Argument from No-self to Selflessness in Santideva's Siksasamuccaya." Contemporary Buddhism. 2:1, 83-97.
Parfit, Derek (1984). Reasons and Persons. Oxford: Clarendon Press.
Pelden, Kunzang (2007). The Nectar of Manjushri's Speech: A Detailed Commentary on Shantideva's Way of the Bodhisattva. Padmakara Translation Group (trans). Boston & London: Shambhala.
Perrett, Roy (1987). "Egoism, Altruism and Intentionalism in Buddhist Ethics." Journal of Indian Philosophy. 15, 71-85.
--(2002). "Personal Identity, Minimalism, and Madhyamaka." Philosophy East and West. 52:3, 373-385.
Pettit, John (1999). "Paul Williams: Altruism and Reality: Studies in the Philosophy of the Bodhicaryavatara." Journal of Buddhist Ethics. 6, 120-137.
Santideva (1997). A Guide to the Bodhisattva Way of Life. Wallace, Vesna and Wallace, Alan (translators). Ithaca, New York: Snow Lion Publications.
Siderits, Mark (2000). "The Reality of Altruism: Reconstructing Santideva." Philosophy East & West. 50(3), 412-424.
--(2003). Personal Identity and Buddhist Philosophy. Burlington, VT: Ashgate.
--(2007). Buddhism as Philosophy: An Introduction. Indianapolis, IN: Hackett Publishing.
Wetlesen, Jon (2002). "Did Santideva Destroy the Bodhisattva Path?" Journal of Buddhist Ethics. 9, 34-88.
Vasubandhu (1988). Abhidharmakosabhasyam. Poussin, L. De La Vallee (translator). English translation by Leo M. Pruden. Berkeley, CA: Asian Humanities Press.
Williams, Paul (1989). Mahayana Buddhism: The doctrinal foundations. London and New York: Routledge.
--(1998). Studies in the Philosophy of the Bodhicaryavatara: Altruism and Reality. Delhi, India: Motilal Banarsidass.
--(1999). "A Response to John Pettit." Journal of Buddhist Ethics, 6, 138-153.
--(2000). "Response to Mark Siderits' Review." Philosophy East & West., 50(3), 424-453.
(1) Department of Philosophy, University of New Mexico. Email: Sharris4@unm.edu.
(2) Here, we may adopt with minor qualification, one of Jon Wetlesen's individualistic definitions of altruism. An altruistic person "has an impartial concern for the welfare of all parties concerned, without discriminating between the welfare of himself or herself and others" (41). The necessary qualification is that an altruistic person might still focus on removing his own suffering in some cases. This is because his concern is to remove as much suffering as possible, regardless to whom it belongs. Therefore, since he is often in a particularly advantageous position to care for his own welfare, he would sometimes focus on this first.
(3) In the conclusion of this essay, I will question whether Santideva is really making an argument in these passages. Part of my evidence for considering this possibility is that no reconstruction of these passages as an argument seems to be successful. Therefore, in this and the next section, I provisionally assume he is making an argument and develop what I take to be the most reasonable version of this argument, in order to show why it fails.
(4) For another version of an argument that realizing the truth of anatman entails altruistic action, see Perrett, "Egoism."
(5) For a different interpretation that claims Santideva does not argue from an Abhidharmika position, see Wetlesen. I agree with Wetlesen's view that for most of his text, Santideva does not argue from an Abhidharmika position; however, it seems to me that Siderits is correct in claiming this particular argument depends on Abhidharmika presuppositions. In the verses quoted above, Santideva explicitly refers to sufferings that exist without an owner, a possibility that an Abhidharmika, but not a Madhyamika, would accept. Therefore, if we are to take these passages as an argument from the premise of anatman to the conclusion of altruism, they are best interpreted as a kind of skillful means in which Santideva argues from premises his opponent accepts. In general, I am sympathetic to Wetlesen's argument that Santideva stresses the interdependence of sentient beings in this text. Nevertheless, at least the verses under consideration here do not seem to support Wetlesen's underlying thesis that Santideva was arguing for a holistic conception of self, in which one considers the selves of others as part of one's own self.
(6) Cf. Clayton 84-86.
(7) Evidence for this point can be found by talking almost any verse from the Bodhicaryavatara at random. Consider, for instance, the first verse of the eighth chapter: "Upon developing zeal in that way, one should stabilize the mind in meditative concentration, since a person whose mind is distracted lives between the fangs of mental afflictions" (Wallace & Wallace 89). Here, as in numerous verses throughout the text, Santideva is clearly directing his comments to the conventionally existing self.
(8) Here, I am in agreement with Barbara Clayton (90-91).
(9) In the fifth chapter of his Studies in the Philosophy of the Bodhicaryavatara, Paul Williams argues in great detail against the Buddhist claim that partite entities, including persons, exist only as a result of conceptual imputation. The chapter illustrates a masterly blend of scholarship rooted in the Indian and Tibetan commentarial tradition, and Williams' application of work by contemporary analytic metaphysicians. For my purposes, however, I am simply assuming the Buddhist position of anatman is coherent, since my own interest is in questioning whether a commitment to altruism would follow from its acceptance. See Siderits's "Reality" for a thoughtful response to Williams' critique.
(10) Williams elsewhere expresses his point as follows: "I argue on logical grounds that the negation of the Atman will not eliminate selfishness because denying the Atman is compatible with being selfish." ("Response to Pettit" 147). Here, I think Clayton's suggestion that we think of Santideva as making an antidiscrimination argument is helpful. Williams is correct that one could both deny the intrinsic existence of the self and be selfish, by prioritizing the welfare of the conventional self. However, the force of Santideva's argument would be to claim that such prioritizing would be irrational, in that there is no good justification for prioritizing one's own future welfare, once anatman is accepted.
(11) Williams makes a similar point in reference to verses ninety-seven and ninety-eight in chapter eight of the Bodhicaryavatara, verses that deal with the topic of altruism and rebirth. Here, Santideva suggests that if I have concern for my own future rebirths, which are not identical to my current self, I should likewise be concerned for the welfare of contemporary others. Williams points out that a potential difficulty for Santideva's argument is that his opponent might use these observations to suggest we need not care about the welfare of our future rebirths, since they are not really identical to me. Williams here stresses the lack of physical and psychological continuity between my current self and future rebirths as reasons the opponent might use to suggest we need not be concerned with their welfare. Williams insightfully points out that in these verses Santideva and his commentators face a dilemma: "The more they stress otherness between this life and future lives, the more they open themselves to the reply that there is no need to concern ourselves with future lives. The more it is argued that there is a need to concern ourselves with future lives because it will be us, the less grounds there can be for arguing a concern with contemporary others" (Studies 50). Williams goes on to suggest that this issue of altruism and rebirth differs from the question of whether we should care for our future selves in the same lifetime, because there will be considerable physical and psychological continuity between our current and future selves in the present lifetime that will be lacking between my current self and future reincarnations. See Studies, chapter 2, especially 49-51. Nevertheless, it seems to me that a slightly modified version of this dilemma identified by Williams applies also to verses 101-103 of chapter eight that are the subject of this essay. I develop this idea below.
(12) My presentation of Santideva's dilemma is an adaptation of the one Williams himself presents. Williams sees Santideva as caught in the dilemma of either accepting the existence of the conventional self, in which case egoism is still possible, or denying the existence of the conventional self, in which case the idea of removing pain becomes incoherent. See "Response to Pettit" 147. I find it both too uncharitable, and too out of keeping with the remainder of the text, which frequently refers to conventional selves, to interpret Santideva as denying the existence of the conventional self. Instead, I suggest that he accepts the existence of the conventional self, but argues that given its lack of intrinsic identity in any ultimate sense with my future conventional self, the existence of the conventional self does not provide rational grounds for prioritizing my own welfare. This leads to the dilemma I call "Santideva's dilemma": Emphasizing the lack of identity between my present and future conventional selves raises the question of why I should care for the welfare of any future conventional self, including my own. Emphasizing the importance of the relationship between my present and future conventional self, on the other hand, offers a reason to prioritize my own welfare.
(13) There are obvious parallels between Santideva's argument, and Derek Parfit's work on the moral implications of accepting the truth of reductionism about personal identity, the view that "a person's identity over time just consists in the holding of certain more particular facts" that can be "described in an impersonal way" (210). Parfit himself considers the question of whether accepting the truth of reductionism might entail we would have either no reason to be concerned about our own future, or, given that there would remain a great deal of psychological connectedness between different stages of our lives, "special" concern for our future welfare might still be justified. Interestingly, Parfit claims that both of these views are defensible, and is agnostic as to whether either ought to be accepted. See Parfit 307-312. See Perrett ("Personal," especially 381) for an interesting discussion of the Madhyamaka view of personal identity that also considers the relationship between this view and altruistic action.
(14) As explained above, Santideva appears to be arguing from the perspective of an Abhidharmika, who holds that pains are among the ultimate constituents of existence.
(15) See also Siderits "Reality," 415.
(16) See also Siderits "Reality," 415-416.
(17) For a slightly different presentation, see Buddhism 82.
(18) See here Abhidharmakosabhasyam 9:1, 1313.
(19) We have already seen that Paul Williams reconstructs Santideva's argument as an attempt to bridge the is ought gap from the fact of anatman to the normative conclusion of altruism. Appealing to the intrinsic negativity of mental states of suffering simply shifts the attempt to bridge this gap from the fact of anatman to the fact of suffering (duhkha). Here, the fact of intrinsically negative states is appealed to in drawing the normative conclusion that we ought to prevent or remove these states of suffering. As will become apparent, it remains unclear that the gap can be bridged, at least simply by appealing to the intrinsic negativity of suffering.
(20) Of course, this is not a Buddhist world. In a Buddhist world, flashes of mental suffering are closely connected to other physical and mental events in such a way that the suffering of one mental flash can influence and be recognized by closely associated mental moments of mind. One might claim, then, that such connected mental states of suffering, though belonging ultimately to no one, ought to be prevented, and that our intuitions support this conclusion. Here, however, Santideva's dilemma reappears. If I emphasize the connections between these mental states as having normative significance, then I must also acknowledge that my present moment of mind is situated in one of these distinct causal streams of mental moments. Therefore, since my current moment of consciousness is situated within a distinct casual stream, it seems rational for me to prioritize the prevention of suffering most closely connected to my present moment of consciousness. Only the Principle of Moderate Benevolence, and not altruism, follows.
(21) A third argument Santideva's opponent might make is the one raised by Williams, that the notion of pains existing without a subject is incoherent. See Williams Studies 147-164, and the reply by Siderits ("Reality" 418-421) for considerations of this objection. For a different response to the issue of free-floating pains, see Clayton 86. For the sake of argument, I assume that pains may be free-floating.
(22) This, I think, also serves as a response to Barbra Clayton's reconstruction of Santideva's argument. As I have above, Clayton takes Santideva to be arguing that, given the lack of an intrinsically existing self, I have no rational reason to justify prioritizing removing my own pain above others. Clayton then suggests Santideva holds it is simply the nature of pains that they are abhorrent, and they should be removed for that reason. Here, the opponent could make similar objections to the ones I offered above. First, he might claim that, once we deny the intrinsic existence of selves, it becomes doubtful as to whether free-floating pains are really abhorrent. Second, he might claim that even granted free-floating pains are abhorrent, it is no longer clear why we are ethically obligated to eliminate them, given that the commonly accepted ethical intuition that pain ought to be removed has been thrown into doubt by the acceptance of anatman. See Clayton 91.
(23) Certain schools of Buddhism believe that our innate nature is compassionate, and so hold that merely removing ego clinging will leave us in a state psychologically motivated to remove everyones's suffering. For these Buddhists, simply accepting that prioritizing one's own welfare is irrational might be sufficient, since they hold humans by nature are intrinsically motivated towards altruism. See Williams' (Mahayana chapter five) overview of the Tathagatagarbha tradition for one example of such a development. It is less clear, however, that Madhyamaka Buddhists could successfully employ this strategy, since they are antiessentialist, and should therefore reject that there is any intrinsic human nature. It seems to me that Madhyamakas in general urge us towards compassionate action, without relying on any essentialist tendencies towards altruism.
(24) Siderits also notes the location of these passages is the end of the meditation chapter. He continues to interpret them as an argument, suggesting their location explains why Santideva employed premises he did not ultimately accept. See "Reality" 421.
(25) Kunzang Peldon (207) explains that in this verse, Santideva is pointing out that there are no intrinsically existing agents that are the bearer or recipients of anger. This is because hatred arises owing to various causal conditions, which themselves arise due to other causal conditions and so on. Therefore, no independent agent who acts as the creator of anger may be located. For this reason, Santideva refers to the bearers and recipients of such causally conditioned anger as illusionary.
(26) I am grateful to Richard Hayes, John Taber, Roy Perrett, Ethan Mills, Jay Garfield, Daniel Cozort, Venerable Khenmo Trinlay and an anonymous referee for helpful comments on various versions of this essay.
University of New Mexico
|Printer friendly Cite/link Email Feedback|
|Publication:||Journal of Buddhist Ethics|
|Article Type:||Critical essay|
|Date:||Jan 1, 2011|
|Previous Article:||Paternalist deception in the Lotus Sutra: a normative assessment.|
|Next Article:||Buddhism Beyond the Monastery: Tantric Practices and their Performers in Tibet and the Himalayas.|