Kamma and the Buddhist Hell

Abstract

As an extension of the Problem of Evil, the Problem of Hell poses further difficulties for the theodicy and eschatology of Western theist religions. This Problem of Hell, which presumes a transcendent divine entity, is, however, less applicable to an Eastern religious tradition like Buddhism. As a non-theist religion, Buddhism is not centered on an overpowering God but is predicated on the doctrine of kamma. Hell in Buddhism is conceived as one of the rebirth realms in the saṃsāra where beings are driven to by the force of their kamma. This kamma-based conception of hell has its own unique features, especially with regard to retribution and salvation. It also has a unique problem. As the doctrine of kamma is commonly understood as an endorsement of free will, it appears to conflict with another Buddhist doctrine, namely that of dependent origination, which is often interpreted as suggesting a deterministic worldview. This tension between doctrines of kamma and dependent origination is also known as the Buddhist free will problem, as it involves the controversy over the metaphysics of free will. Based on the Pāli scriptures, the essay tries to propose a compatibilist solution to the problem, defending kamma for the Buddhist hell.

1. Introduction

The reason why Buddhism is immune to the Problem of Hell is that it is a non-theist religion without the concept of an omnipotent, omniscient and compassionate God. The Buddha in Buddhism is not so much an equivalent to God in Christianity, as Buddha is an “awakened” person but not a transcendent divine entity who creates and controls everything. Although the highest realm of rebirth in the Buddhist saṃsāra is translated in English as the realm of gods, which comprises different heavens, gods here are the virtuous beings who enjoy a blissful existence in the heavens due to the good kamma accumulated from their past lives. They will still transmigrate to other rebirth realms, including hell, once their good kamma has been consumed in heavens. Since Buddhism is not centered around an overpowering God, there is no tension between God and an inescapable hell. The Problem of Hell therefore does not arise in the Buddhist context.

But like what has happened in other religions, hell in Buddhism is also a long-standing source of contention. Different Buddhist schools have varying conceptions of hell, reflecting differences in doctrinal, textual, and cultural traditions. The Theravāda tradition, for example, presented a more canonical and literal conception of hell based on the Pāli sources, while the Mahāyāna tradition more heavily emphasizes the psychological and symbolic aspects of suffering. The specific descriptions and interpretations of hell sometimes even parallel how things go in the temporal world. In Chinese Buddhist tradition, for example, hell is depicted as a bureaucratic system, which is similar to the one in the Tang Dynasty, with judges and kings who oversee the punishment of beings based on their actions (Jiang 2022). Despite the differing details, however, the core idea of hell as a realm of suffering due to kamma is shared across all Buddhist traditions. As it is perhaps impossible to provide a pan-Buddhist picture of hell, this essay will focus on the Pāli canon and provide a more tradition-specific discussion of relevant issues.

The Theravāda tradition of Buddhism also has a retributive conception of hell, although according to the Pāli sources, it is the law of kamma rather than a divine God that distributes the retributive suffering. Kamma is a Pāli word (Sanscrit karma) meaning “action” or “deed”, literally. The kamma doctrine, in its basic form, holds that all intentional actions (including attitudes, words, and acts) entail consequences that will rebound upon their agent at some point in the future. Good or virtuous actions generate positive kammic effects while bad or evil actions beget negative kammic effects. Together, these effects shape our present and future lives. The law of kamma is also closely connected with the idea of rebirth or reincarnation. Subject to the kammic force of their previous actions, sentient beings are reincarnated successively in the five realms of rebirth, i.e., the realms of gods (devā), humans (manussā), animals (tiracchānā), hungry ghosts (petā) and hell (Nirayā). These realms represent different states of existence or levels of suffering. From the Buddhist perspective, life in the cycle of rebirth, or saṃsāra, is deferentially painful no matter which realm one is reborn into. The ultimate goal of the Buddhist practice, therefore, is to achieve Nibbāna, or liberation from the cycle of rebirth. But before attaining Nibbāna, rebirths in the lower realms of hell, ghosts and animals will certainly be more painful than those in the higher realms of humans and gods. Hell in Buddhism is therefore not the only realm where beings experience painful suffering. But like the infernal region in other religions, hell in Buddhism is also the place where beings experience the most agonizing punishment for their most atrocious actions—it is where their worst kammic effects materialize.

Since the kamma doctrine offers a seemingly simple yet complete explanation for the operation of the Buddhist cosmos, of which hell is a constituting realm, it has aroused interest in the West and attracted both endorsements and skepticism. It has been lauded, for example, for its “consistency” and “metaphysical achievement” by Max Weber (Weber 1947, p. 359) and for its rationality by Peter Berger (Berger 1967, p. 65). Some scholars, like Arthur Herman, assert the superiority of kamma to Western theodicies (Herman 1976, p. 287) while others, such as Michael Stoeber, consider it “more plausible” than its counterpart Christian ideas (Stoeber 1992, p. 167). Despite its increasing popularity in the West, however, the kamma doctrine has also received some serious challenges. A recent criticism comes from Whitley R. P. Kaufman, who claims that the kamma doctrine “has still received comparatively little critical analysis in comparison with the scrutiny to which dominant Western ideas such as Original Sin or free will have been subjected” (Kaufman 2005, p. 16). By raising five moral objections, Kaufman casts doubts on the view that the kamma doctrine can be taken as a systematic and rational account of evil and suffering. Among these objections, the most fatal one concerns the free will problem, namely whether kamma, understood as a free will theory, is compatible with the deterministic worldview of Buddhism.

Based on the Pāli scriptures, this essay seeks to expound the Buddhist hell in terms of kamma and to offer a tentative solution to the Buddhist free will problem concerning kamma. It first argues that the kamma doctrine shapes a distinctive or even unique conception of hell, particularly in its framework of retribution and salvation (i.e., liberation from saṃsāra). With kamma as an elaborate explanatory system for the causes of suffering, the Buddhist hell offers a better version of moral exhortation which serves its soteriological goal. In addition, this essay argues that the kamma-based hell can be endorsed when a fatal challenge toward kamma is properly handled. With kamma interpreted as a free will theory, how to reconcile it with the causal determinism expressed by another Buddhist doctrine, i.e., the doctrine of “dependent origination”, has become a problem. By drawing on the Western thought on freedom, this essay reconstructs a conventional conception of freedom for Buddhism which is compatible with its ultimate conception of freedom. This essay argues that the conventional freedom is metaphysically compatible with determinism.

2. Kamma and Retribution

As indicated by the Pāli scriptures, the law of kamma is like a natural law, or “a law of physics”, that governs the changes in the cosmos, including the transmigration of beings between rebirth realms. Due to the workings of kamma, moral and immoral conducts are rewarded or punished, respectively. Evil actions cause suffering (dukkha) while virtuous deeds cause pleasantness (sukha), in due time of course. It is perhaps due to this close relationship between evil and suffering that some scholars have suggested that evil is interchangeable with suffering. Sebastian Gäb, for example, argues that “something is an evil if and only if it is actually or possibly the cause of suffering” (Gäb 2015, p. 347). The proposed interchangeability between evil and suffering may be innocuous for the purpose of his essay, but will not be embraced by this essay. In line with the kamma doctrine, evil is the moral cause and suffering is the outcome. This relationship is not to be reversed, for it is crucial for our understanding of the origin and nature of the kamma-based hell discussed in this essay.

One prominent feature of the kamma doctrine is that it sees evil, and the consequential suffering, as originating from the will of individual beings. Though kamma literally means “action”, in a strict sense, it refers to the will behind an action: “It is will (cetanā), O monks, that I call kamma; having willed, one acts through body, speech or mind” (Aṅguttara Nikāya III.415). Volition or intention is the inner motive that starts an action. Intentional actions, whether good or evil, will affect a person’s consciousness, shaping it in a certain way that leads to future consequences. Unintentional actions, however, will have less shaping impact on a person’s awareness and therefore result in discounted or even no kammic effects (Harvey 2000, p. 17). Suffering in hell is therefore the outcome of one’s own intentional evil actions. By illuminating the causal relationship between evil actions and suffering, the kamma doctrine enlightens people as to the origin and causes of hell.

The kamma doctrine also implies a “proportionality principle” that regulates all punishments and awards in the cycle of rebirth. When confined to punishments for evil actions, Buddhism is a religion that is concerned with a broad range of evil actions, from ethical lapses to severe sins. The proportionality principle, entailed by the kamma doctrine, assigns retribution in accordance with the causal sins. It requires that the gravity of retributive suffering be suitably proportioned to the severity of the wrongdoing. Due to such a principle, for example, actions of hatred and violence, which are the most severe actions, will incur rebirth in hell, while actions of delusion, confusion or greed, which are less severe ones, will incur rebirth in the realms of animals or ghosts (Harvey 2000, p. 15). Sufferings in these lower realms are similar to those experienced previously by others when a malicious action of the above-mentioned types was perpetrated towards them.

Besides the three lower realms, the proportionality principle also indicates that there is a further gradation of hell-incurring actions and their corresponding retribution in hell. Even among the deeds of hatred and violence, some are more heinous than others. Kathāvatthu, the earliest Buddhist scripture with a named author in the Pāli canon, mentions five sins that are more damnable than others (Kathāvatthu VII.7). These five cardinal sins are known as sins “having no intermediate” (ānantarika)—they cause the perpetrator to be reborn in hell immediately after death. They are matricide, patricide, killing of a perfected saint (arahant), causing a Buddha to bleed, and creating a division in the monastic Saṅgha. These heinous sins are punished in hell for the remainder of an eon (Papañcasūdanī IV.109-110), and no good actions can avert such a rebirth in the immediate next life (Aṭṭhasālinī 358). Killing or injuring sentient beings other than parents or saints is not listed among the five most deadly sins, but still ranks as one of the five basic sins in the Theravāda tradition (Braarvig 2009, pp. 268–269). Perpetrators of these five basic sins include someone who kills or injures (pāṇātipātī), who steals (adinnādāyī), who performs sexual misconduct (kāmesu micchācārī), who lies (musāvādī), and who indulges in intoxicants (surāmerayamajjapamādatthāyī). Although it is indeterminate when the kammic effects of these sins will ripen and materialize, the scripture indicates that the evildoers will be punished both by secular rulers and by rebirth in Niraya hell (Braarvig 2009, p. 269).

Besides the five deadliest sins and the five basic sins, there is also a wide range of wicked deeds that lead to rebirth in hell. For example, in Dhammapada of Sutta Pitaka, monastic hypocrisy and corruption are criticized and warned with hell (Dhammapada XXII, 306 ff). In Vinaya Pitaka, another collection of scriptures that consist of monastic principles for practitioners, the most serious misconducts are the pārājikas offenses, which include sexual misconduct (methuna dhamma), stealing (adinnādāna), killing of human beings (jīvitā voropana), and the claiming of magical abilities (uttarimanussadhamma) (Braarvig 2009, p. 266). These offenses no doubt involve such important codes of conduct for monks and nuns that offenders would undergo the punishment of being expelled from the monastic order. In addition to excommunication in this life, however, the pārājikas offenses would also incur falling to a lower state of existence such as hell (Braarvig 2009, p. 266). The Vinaya also informs us that the fifth ānantarika sin, i.e., causing division in the monastic Saṅgha, while incurring immediate rebirth in hell after death, is also punished in this life by suspension of membership for a period to be decided by the Saṅgha (Vinaya Piṭaka Su. S. X–XI). Attempts to usurp Buddha’s leadership by creating schism in the community of monks is certainly one of the most heinous sins. It was out of Buddha’s benevolence to use suspension of membership, but not expulsion from the Saṅgha, as a monastic punishment for the offense in the current life. Nevertheless, the Vinaya tells us that the perpetrator of such a deadly sin would still ultimately end up in the deepest hell, the Avici hell, and others who slandered Buddha would be reborn in the Lotus hell for the act of defaming Buddha (Vinaya Piṭaka SN I, 149).

Rebirth in hell is not only relevant to the severity and variety of evil actions, but also to the quantity of such actions. It is said that “By constantly committing evil deeds we are reborn in hell, by doing many we become spirits [i.e., ghosts], and when we do only a few we are reborn as an animal” (Guenther 1959, p. 79). This indicates that the more wrongdoing one perpetrates, the more likely one will be reborn in hell. Aggregation matters too, to some degree at least. Moreover, rebirth in hell is caused not only by committing evil deeds but also by encouraging others to do so, especially when this involves a large number of beings (Aṅguttara Nikāya v.306-308).

Corresponding to the severity, variety and quantity of evil deeds, retributive sufferings in hells also vary. The various hells actually indicate the assorted punishments that match their causal wrongdoings. Although the exact number and descriptions of hells vary to some extent with different scriptural texts and the later interpretive literature, early Pāli scriptures present a basic list of eight hells (Braarvig 2009, p. 264). Sañjīva is the hell of reviving, where beings are revived to undergo torments repeatedly until the negative kamma from their previous evil deeds is cleansed. Kālasutta is the hell where beings are cut or sawed apart after being measured with a black cord. Sanghātaka is where beings are crushed or ground into flesh and blood or devoured by fierce animals. Roruva is the “hell of roaring”, where beings cannot help roaring and begging for pity. Mahāroruvaniraya is the “hell of great roaring”, which is similar to Roruva hell but greater in scale and intensity. Tāpana and Patāpana are hells of “suffering” and “excessive suffering”, respectively. Finally, Avīci is the deepest and most horrific hell, where beings suffer with “no intermission”. What is common among these hells is that beings there have to undergo the most gruesome and repulsive torments or tortures. The intensity of suffering is worse in some hells than in others, but all are seen as appropriate to the malicious deeds which led to them.

Kamma determines not only the types of hellish retribution, but also the length of it. An important feature of the kamma-based hell is that it is not eternal. Inhabitants of hell would rise to higher states of existence once their bad kamma is consumed or canceled out in hell. But beings still suffer for prolonged periods of time in hell. In Buddhism, the word for describing time span is not “millennium”, but “eon” or kappa—a time period of enormous length. Hell is the worst place for rebirth and should be avoided as much as possible, not only because the retributive torments there are the most horrific, but also because it lasts for lengthened time spans—so lengthened that it is extremely difficult to getout of it and attain human rebirth again. The significance of attaining the human realm lies in the fact that it is where a person has a much greater chance to receive the teachings of the Buddha, advance their spiritual improvement, and be liberated from the horrors of saṃsāra (Harvey 2000, p. 30). The difficulty of attaining a human rebirth is often compared with the rarity with which a blind turtle coming to the surface of an ocean every hundred years puts its head through the hole of a piece of wood which floats everywhere on the sea (Braarvig 2009, p. 270). Although rebirth in hell is not eternal, its protracted suffering and the difficulty of extrication warn against anyone who dares to sin.

The above discussion shows that the kamma doctrine provides a simple yet consistent explanation for the origin of hell and the retribution there. Hell is incurred by the bad kamma generated by one’s past evil actions. Retributive sufferings in hell are therefore the kammic effects of one’s own actions, not punishments by a divine God. The kamma doctrine also entails a proportionality principle that renders the gravity of retributive suffering proportional to the severity, variety and quantity of evil actions. The kamma-based hell seems to have a rather sensible framework for retribution.

3. Kamma and Salvation

The kamma-based hell is also unique with regard to salvation. Although in Buddhism there is no such a day of reckoning when a transcendent God passes on the final judgment, the law of kamma determines whether beings will suffer from their negative kamma in the lower realms of rebirth, or enjoy their positive kamma in the higher realms. There are two types of salvation that can be interpreted from the Theravāda doctrines, i.e., quasi and genuine salvation. Rebirth as a human or god can amount to quasi salvation, for one no longer suffers from the horrible tortures in the lower realms and enjoys a mostly pleasant existence in the higher realms. But even rebirth as a god cannot count as genuine salvation, for no matter how long the life-span in those heavenly worlds lasts (some may last for tens of thousands of eons), one will still fall into a lower existence once their good kamma has been consumed there. The Four Noble Truths, which are the central teaching in the Theravāda tradition, teach us that life in the cycle of rebirth is suffering (dukkha), whichever form it is, and can only be transcended by attaining Nibbāna, which is an eternal state beyond space, time and dukkha (Harvey 1990, pp. 47–72). Only by achieving Nibbāna can one avoid rebirth in the saṃsāra, which is governed by the law of kamma. Such a two-fold conception of salvation allows hell to play some role in it. Instead of depriving the damned of chances of salvation, the Buddhist hell helps people cleanse their bad kamma and prepares them for a better rebirth, where they can have a greater opportunity to learn about the Dhamma and achieve Nibbāna. In this sense, the kamma-based hell bears some resemblance to the Catholic purgatory, where souls with venial sins undergo purification or spiritual cleansing before they can enter heaven.

In line with this kammic view of salvation, the Buddhist hell (and indeed any realm in the saṃsāra) cannot be permanent. As discussed earlier, the kamma doctrine explains not just a causal relationship between an action and its consequence, but also proportionality between the two. This proportionality principle indicates that finite kamma generated by intentional actions incurs only finite rewards or punishments. Since the wicked actions committed in the rebirth realms are finite in nature, the negative kamma generated will also be finite. It follows that the kamma-based hell must be non-eternal, like other realms in the saṃsāra, although it may last for an extended period of time measured in eons. This non-eternity feature is important, for it is another reason why the Buddhist hell is not held hostage to the Problem of Hell which accompanies the concept of an eternal hell in Christianity. In addition to the fact that there is no transcendent God in Buddhism, as mentioned at the beginning of the essay, the Buddhist hell which is commensurate with the finite kamma generated by previous malicious actions is also impermanent. There is therefore no tension in Buddhism between a powerful God and an eternal hell.

An important implication of the non-eternal hell, and indeed the non-eternal rebirth realms in saṃsāra, is that one will always have a chance of genuine salvation. Even beings in hell will be able to transmigrate to other rebirth realms through successive reincarnations as a result of their remaining kamma, where they may become awakened from their delusion or ignorance by learning the Dhamma and achieving Nibbāna. There is therefore the possibility, and hope, for those in hell to be released, enlightened and saved (liberated) someday. Being driven to hell by kamma is therefore not a deprivation of opportunity for ultimate salvation. It simply means that one is not able to achieve Nibbāna in this life. But there is always an opportunity for this in future lives. The door to Nibbāna is always open, even to those who go to hell by the force of their kamma. It is upon this understanding that Buddhism can be interpreted as entailing a “universalist” view of salvation (Weber 1947, p. 359). The kamma doctrine combined with the idea of multiple reincarnation offers more possibilities for salvation for everyone, since failure to achieve Nibbāna in this life does not prevent one from trying again in the future. This can really register as another merit of the kamma-based conception of hell, when compared with its counterparts in other religions. In orthodox Christian tradition, for example, there is an urgency to achieve redemption in this human life, since those who fail in this life will be doomed to eternal perdition, permanently loosing the chance to be saved. Compared with the one-time opportunity of redemption in traditional Christianity, Buddhism not only offers multiple opportunities for salvation, but also the possibility of universal salvation. Even the most heinous sinner in the inferno has the opportunity someday to receive the teachings of Buddha and be liberated from the sufferings of saṃsāra.

But the possibility of universal salvation is not to be interpreted as equal likelihood of being redeemed in all five rebirth realms. From the Buddhist point of view, it is mostly at the human level that beings are saved (Harvey 2000, p. 24). In the heavens, a higher realm than that of humans, the gods simply enjoy a blissful life that results from their previous virtuous actions. Therefore, they may not be motivated enough to seek further spiritual improvement for Nibbāna. In the lower realms of animals, ghosts and hell, beings may barely endure the torture, so they may have little extra intention to perform good or bad deeds. Even if they have such intentions, a good deed will have kammic effects that ripen in a future life, which will not help alleviate the pains they are now suffering. It is mostly at the human level, where there is adequate happiness as well as adequate endurance, that beings find both enough motivation to seek spiritual improvement and enough intention to perform virtuous or evil conducts. The human level is therefore a level most suitable for striving for self-improvement and attaining genuine salvation.

The possibility of universal salvation also should not be understood as implying a lack of urgency to seek salvation. Kaufman, for example, suggests that Buddhism does not provide sufficient imperative to seek redemption, “for all mistakes and misdeeds can be rectified in the fullness of future lives … [and] one’s choice here and now does not matter much in the greater scheme of things” (Kaufman 2005, p. 24). But there is no reason to think that the multiple-life theory removes the sense of urgency for salvation. It is common knowledge in Theravāda tradition that suffering is pervasive in all realms of rebirth, whether it takes the form of hellish agony, animal brutality, human pain/worry or the more subtle unsatisfactoriness of the gods in heaven. In addition, movement in the Buddhist cosmos is so slow that one could spend many eons undergoing excruciating suffering before being reborn into a better form of life. Knowledge about the nature and lengths of suffering would certainly create an imperative for a person to achieve liberation as quickly as possible. This sense of urgency is further reinforced by the idea, found in various Hindu and Buddhist texts, that human incarnation is rare and precious. Incarnations as humans are reached “only after lengthy and laborious spell in the subhuman realm and can just as quickly be followed by further spells in such a realm” (Chadha and Trakakis 2007, p. 546). This, combined with the view that liberation is most likely to occur in the human realm, would create a strong sense of urgency to seek liberation in this lifetime.

Some may suggest that the kamma doctrine should be interpreted as entailing an escapist, rather than a universalist, view of salvation. Since the kamma doctrine indicates that all beings, including those in hell, have the opportunity to achieve Nibbāna or liberation, this is similar to the escapist view of hell in Christianity according to which denizens of hell are also given the opportunity to leave hell and enter heaven (Buckareff and Plug 2015, p. 248). As an alternative approach to the Problem of Hell in Christianity, escapism has the merit of being consistent with God’s compassion and power. Denizens of hell are given the freedom to decide whether they want to accept divine grace to leave hell and be saved. To explain the possibility of escaping hell, Buckareff and Plug refer to the possibility of character transformation for the denizens of hell after they have experienced the torments there. While some denizens of hell have a “set character” which will not be modified by their post-mortem experience, others have a changeable character which may be transformed to the degree that they would become psychologically capable of accepting God’s grace and leave hell (Buckareff and Plug 2015, pp. 251–52). From Buckareff and Plug’s arguments, it seems that the escapist view of the Christian hell bears some similarities with the kamma-based conception of hell. The escapist hell is also non-eternal, at least for some, and there is also the possibility that everyone will be able to leave hell and be redeemed.

Despite the similarities between the escapist hell and the kammic hell, there are still reasons why the kammic hell entails a universalist but not an escapist view of salvation. First, the kamma doctrine is not committed to the view that some denizens of hell have a “set character” which renders them incapable of being transformed or redeemed. A common view shared by all Buddhist traditions is that beings in the cycle of rebirth are under the influence of delusion, confusion and cravings, which lead to wrongful conducts and bad kamma. The goal of the Buddhist practice is to help them be enlightened from such influences, so that they can come to understand the truth of dukkha and seek Nibbāna. Buddhism therefore presumes the possibility of character transformation, which is necessary for the Buddhist salvation. From the Buddhist point of view, there is no person with a so-called “set character”, but only people with a stubborn personality who may be difficult to educate or enlighten but can still be transformed if sufficient chances are granted or adequate time is given. Unlike the escapist view, the kamma doctrine does not preclude such possibilities. Secondly, and more importantly, while the escapist sees hell as eternal for some and non-eternal for others, proponents of kammic hell believe that hell is non-eternal for all. As mentioned above, the proportionality principle implies that one’s spell in hell is determined by their kamma. Since a person’s action and its subsequent kamma in saṃsāra are finite in nature, retributive sufferings in hell will be proportionally finite, though it could be long-lasting. Ultimately, even the most wicked sinner will be able to leave hell when their negative kamma has been cleansed. Hell is therefore everlasting for no one. Third, denizens of the kammic hell are not assumed to possess the freedom of choosing to leave hell or to stay. It is not up to them to decide, for this is determined by the law of kamma. For these reasons, what the karmic hell implies must be a universalist but not an escapist view of salvation, although escapism also indicates the possibility that all denizens of hell may choose to accept God’s grace and leave hell.

4. Kamma and the Buddhist Free Will Problem

The above discussion shows that the kamma-based conception of hell has some unique features. Despite its merits, however, the kammic hell also faces some challenges, especially with regard to the metaphysical issues implicit in the doctrine of kamma. Kamma has been frequently praised as a freewill theory on the ground that it presumes the moral ability of an agent to choose to act ethically or unethically. Due to their moral choice, the agent deserves the good and bad of their life as these are the kammic consequences of his past deeds. When kamma is interpreted to imply moral agency and therefore free will, it seems to run into conflict with another important doctrine of Buddhism, namely “dependent origination/arising” (paṭiccasamuppāda). This doctrine holds that all phenomena (humans included) arise in dependence upon prior causes and conditions. It implies a deterministic relationship between causal conditions and results. The problem then is how people can be held morally responsible for their kamma in a world that is governed by the impersonal process of dependent arising? There is an obvious tension between kamma, interpreted as a free will theory, and dependent origination, interpreted as a deterministic theory. The tension is also frequently referred to as the Buddhist free will problem.

Like the teaching of kamma, that of dependent origination is also multi-faceted and can be expressed with varying degrees of complication. Each Buddhist school nuances its presentation of ontology, metaphysics and epistemology so as to create a coherent worldview. They also proposed different solutions to the tension between kamma and dependent origination. Luis Gōmez suggests that the Pāli sources imply a concept of “weak determinism”, a “middle way” between moral choice and determinism (Gōmez 1975). But Gōmez’s concept of “weak determinism” seems still rudimentary and falls short of solving the problem. Asaf Federman, who also relies on the Pāli sources, expounds a kind of free will that he contends is taught by the Buddha, and argues that it is compatible with the doctrine of dependent arising. But his reconstruction of the Buddhist free will uses the conceptual tools of psychology and his arguments for compatibilism sometimes lacks analytical clarity. The remaining part of this section will also rely on the Pāli sources and try to work out a Theravāda solution to the problem.

In the Pāli scriptures, the notion of “dependent origination” is expressed when the Buddha explains the doctrine of no self (anattā) to his disciples. In one sutta of the Saṃyutta Nikāya, Sāriputta explains the Buddha’s teaching on dependent origination (paṭiccasamuppāda) when he rejects the four extreme views that suffering is created by oneself, by another, by both, or by none (Saṃyutta Nikāya II.33). Bypassing the concept of the self, the Buddha teaches that suffering arises dependently upon prior conditions. The twelve steps that lead from the conditions to the generation of suffering constitute an impersonal process of causation, indicating the thesis of no self (anattā). In the Ānattalakkhana-sutta, which is one of the most important discourses of the Pāli canon, the Buddha also expounds the thesis of no soul (anattā) by referring to the five “aggregates” (khandha) that constitute the mind–body phenomenon (Saṃyutta Nikāya III.66). These five fundamental components are impermanent and subject to change, as explained by the Buddha, and for this reason none of them can be identified as the soul. But dependent origination is not just a principle of causation exclusively for explaining human experience and existence—it has been extended to the description of the world in general. It is perhaps most commonly expressed in the following short statement: “When this exists, that comes to be; with the arising of this, that arises” (Saṃyutta Nikāya II.33). Once the condition is present, the outcome will occur. In addition to the personal level, dependency relations also unfold on the cosmological level to the extent that “everything in this universe comes within the framework of causality” (Kalupahana 1976, p. 30).

If the doctrine of dependent arising implies no self (anattā), one might be tempted to think that this deterministic view may cancel out the free will problem for Buddhism altogether. Some other Pāli scriptures, however, seem to deny a wholesale deterministic interpretation of the Buddha’s teachings. In Dīgha Nikāya, for example, the Buddha rejects Makkhali Gosala’s fatalistic position, an act that has been interpreted as the Buddhist affirmation of free will (Stcherbatsky 1930; Federman 2010). Like the Buddha, Makkhali also believed that the distribution of pain and pleasure is governed by some cosmic principles. But he differs from the Buddha when he indicates that there is no role for the individual to play in the purification of their own pain or pleasure, as all the causes (hetu) or conditions (paccaya) lie not within them but outside of them (Dīgha Nikāya I. 53). He maintains the consistency of his fatalistic position when he claims that there is no agency (attakāra), will (vīriyaṃ), or human exertion (purisaparakkama) (Dīgha Nikāya I. 53). This denial of agency or will, however, runs counter to the Buddhist doctrine of kamma, which emphasizes the role that personal efforts play in shaping one’s destiny. While expressing a deterministic view in the doctrine of dependent origination, the Buddha has not suggested a downright denial of free will.

How, then, can we reconcile the doctrines of kamma and dependent origination, both of which were explained by the Buddha, as recorded in the Pāli scriptures, and yet seem to contradict each other? If the teachings of dependent origination are true and ultimately there is no self (anattā), then what is the point of talking about the free will of a self that does not exist? If the teachings of kamma are true and the self is capable of making a choice and taking responsibility for his own fate, then the self must possess a free will that transcends the impersonal causal processes leading up to it. Sometimes, however, seemingly contradictory doctrines might stem from the profundity of the Buddha’s teachings. The simultaneous affirmation of contradictory doctrines could be a skillful and practical method for pedagogical as well as soteriological purposes. It seems that despite the ultimate truth of no self (anattā), the Buddha is also concerned with the non-ultimate self—a self that may be an illusion in the Buddha’s view, but a conventional entity that can be recognized by people. Admission of such a less ultimate identity is necessary, for it is the Buddha’s aim to teach the conventional self how to achieve enlightenment and liberation. This distinction between the ultimate truth (paramattha-sacca) and conventional truth (sammuti-sacca) seems capable of providing a valid way to reconcile the above-mentioned contradictory doctrines (Abhidhamma Piṭaka, particularly the Dhammasaṅgaṇī and the Paṭṭhāna). While the Buddha teaches the ultimate truth of no self (anattā) through the doctrine of dependent arising, it is at the level of conventional truth that he acknowledges the existence of the self, without which the idea of kammic consequences rebounding upon the person who performed the initial action would be nonsensical. The tension between kamma (free will) and dependent origination (determinism), therefore, occurs at this conventional level.

It must be admitted that even with the idea of conventional self, The Buddhist understanding of free will is not entirely isomorphic with the Western concept. As reflected in the Pāli scriptures, free will in Buddhism is more concerned with the removal of internal limitations. From the Buddhist perspective, free will is the will that is free from mental constraints. In Majjhima Nikaya, for example, the three primary roots of suffering (i.e., craving, hatred and delusion) and the five hindrances to spiritual progress (i.e., sensual desire, ill will, sloth, restlessness and doubt) are described as mental or psychological factors that confine the mind and hinder liberation (Majjhima Nikaya 9&10). When a person’s mind is limited by these unwholesome factors, the consequential thoughts, words or deeds are also considered unfree in that sense. As we all know, the Buddhist path to salvation is the path to freedom (vimuttimagga)—free from those mental limitations. The term nibbāna conveys the same sense as it is defined as the “absence of [mental] constraints” (Kalupahana 1992, p. 91). In Theravāda Buddhism, it therefore seems that the ideal of free will is achieved only when samsāric beings have extirpated all spiritual impediments. Those who have not purified themselves spiritually do not possess genuine free will, and their actions, which are also unfree, will generate kamma, binding them to the cycle of rebirth and preventing them from being liberated therefrom.

If pure free will is defined in the Pāli sources as the will purged of mental defilement, few persons on the conventional level will be able to possess it. The fact that they are bound to the saṃsāra proves that they are still controlled by the force of kamma which was generated by their spiritual imperfections. The will that the conventional self possesses, if there is one, must then be impure, tainted with unwholesome mental properties. If this is the case, it would then be difficult to think of the conventional self as possessing a free will because a will constrained by mental imperfections is not considered free from the Buddhist perspective. But it is also undeniable that the Buddha affirms free human agency through the doctrine of kamma. Willed actions count as they affect one’s future life. Although human existence is dependently arisen and is therefore the outcome of an impersonal process of causation, it does not follow that the conventional self has no ability at all to make a choice. The conventional conception of free will or freedom must be able to convey our everyday thinking about what it means to be free, without necessarily presuming the kind of pure free will in the Buddhist sense. The question then is how conventional free will or freedom can be conceived such that it can be compatible with the deterministic view expressed by dependent origination.

This essay argues that the concepts of freedom in Western philosophy may offer some help with regard to this question. The conventional freedom that people in saṃsāra enjoy can be conceived with reference to some Western thoughts on freedom. If the Buddhist conception of freedom is more inward-looking and self-purifying, the Western conception is more outward-looking and self-asserting. In Western philosophy, the most well-known descriptive conception of freedom is presented by Isaiah Berlin who draws a distinction between two senses of freedom—negative and positive (Berlin 1969). The negative sense of liberty requires the absence of external, but not internal or mental, constraints. According to Berlin, a person is negatively free when no one else “interferes with my activity” (Berlin 1969, p. 122). Freedom in this sense is simply “an area within which a man can act unobstructed by others” (p. 122). If a person is prevented by others from doing what they would otherwise do, they are, to that degree, unfree. This concept of negative freedom reflects our most immediate thought about what it means to be free. Freedom in the negative sense does not require one to remove mental constraints or to purify their will. No matter how much they are influenced by spiritual imperfections, they are still free, according to Berlin’s negative sense of freedom, if no other person hinders them from doing what they want to.

Distinguished from negative freedom, which features an absence of external constraints, positive freedom is the ideal of self-mastery or autonomy (Berlin 1969, p. 131). It reflects the wish on the part of the individual to be their own master: “I wish my life and decisions to depend on my self, not on external forces of whatever kind. I wish to be the instrument of my own, not of other men’s, acts of will. I wish to be a subject, not an object; to be moved by reasons, by conscious purposes, which are my own, not by causes which affect me, as it were, from outside” (Berlin 1969, p. 131). A person is positively free when they are capable of conceiving goals of their own, acting upon their own reasons or implementing their own ideas. Their decision or choice is not influenced or controlled by forces outside of them. Besides asserting one’s subjectivity, the concept of positive freedom also emphasizes that one should not be a slave to their own nature, i.e., their “unbridled” passions (Berlin 1969, p. 132). To be a true master of oneself, one must allow their higher self, identified with reason, to restrain their lower self, identified with irrational impulse and uncontrolled desires. Berlin thinks that only when the rational, real, higher self dominates over the irrational, uncontrolled, lower self can one have real freedom. This positive sense of freedom again seems to conform with our conventional understanding of freedom. A person is positively free when they are their own master, with their actions motivated by their own will. They believe it is their own will because it identifies with their own reason, whether their ability of reasoning is fettered by various unwholesome factors or not. In other words, positive freedom does not require the ability for reasoning to be perfect or the will behind the action to be free from all mental constraints. Although it is concerned with the internal causes for actions, it tries to assert one’s subjectivity, rather than the degree to which one is unconstrained by unwholesome mental or psychological factors.

The above discussion suggests that freedom in Buddhism can also be discussed on both the ultimate and conventional levels. While ultimate free will is the absence of mental constraints, conventional free will falls short of that for it is tainted with delusions, hatred or cravings. Although it is not pure free will in the strict Buddhist sense, conventional free will can be conceived in a way that embodies the tangible freedoms people can recognize in every life. Under the conventional conception of free will, the self is still able to make a choice, despite the fact that the self is dependently arisen. The choice would reflect the individual’s current state of mind, i.e., the extent to which they are mentally checked or unchecked by selfish desires or ignorance. Due to such mental limitations, their cognitive and reasoning abilities would be impaired and their subsequent actions imperfect. But they may not be able to realize this and would therefore identify with the choice they have made, believing it to reflect their own free will. This conception of conventional freedom has both negative and positive senses, as illustrated above. A person is free in both senses when they are able to act in line with their own reasons, unimpeded by external interference. Conventional freedom does not require one’s reasoning ability to be perfect, in the sense the it is free from all kinds of mental constraints. The will generated by imperfect rationality can still count as a will of their own if they identify with it.

If conventional free will is the will with which the person identifies given their level of cognition and reasoning, is it then compatible with the determinist view expressed by the doctrine of dependent origination? As presented above, the will the person identifies with is their higher self or reason, but not the unbridled passions of the lower self. While a person’s reasoning ability, including their cognitive ability, is relevant to the amount of education one receives, it is also true that the ability is a natural endowment which differs from one to another. This is evidenced by the fact that people would still end up with different levels of reasoning and cognition even if equal access to education is guaranteed. That the faculty of reason, like many other aspects of the human existence, is dependently arisen is then undeniable. This might suggest that the agential will, which identifies with reason, is also a product of impersonal processes of causality and thus cannot be considered free will. But it is one thing for the faculty of reason to be causally determined; it is another for the outcome of reasoning to be so. For even if the faculty of reason is dependently arisen, the output of reasoning may still be unpredictable, within some appropriate scope of course. The activity of reasoning involves not just the faculty of reason, but also the reasons that happen to bear on a certain issue and how the faculty would respond to them. The resulting decision, which is unforeseen to certain degree, will be identified with by the agent and, for that reason, count as the embodiment of his conventional free will.

What follows is a further note about the compatibility between conventional free will and determinism. The faculty of reason, which is dependently arisen and to that extent predetermined, is not perfect. It, as previously discussed, may be subject to all kinds of mental limitations. But one’s reasoning ability is also not fixed or unchangeable, but open to the possibility of improvement. With the kamma doctrine, the Buddha teaches that a person’s intended actions will have shaping effect on their future life, including their future faculty of reason. This is why practicing in line with the Buddha’s teachings is so important, for in this way one can gradually remove their mental constraints and improve their reasoning ability, as well as their overall destiny, through successive reincarnation until someday they achieve awakening and liberation. In other words, practicing in accordance with Dhamma will guide people to ultimate freedom, i.e. the absence of spiritual impairments, and ultimate liberation from the cycle of rebirth. From the Buddhist perspective, those who have not learned and practiced the Dhamma may not have the opportunity of improving their cognitive and reasoning faculties and will therefore be stuck with whatever faculties that have dependently arisen. Although they still possess conventional freedom, they will not be able to achieve ultimate freedom and will therefore continue to be bound to the cycle of rebirth.

5. Conclusions

In contrast to that described in theist religions such as Christianity, the Buddhist hell is unique in that it relies on the doctrine of kamma to provide a rational explanatory basis. Kamma explains the origin of hell as lying in the evil actions of the agent, and requires the retribution in hell be proportional to the severity, variety and quantity of the wrongdoing. It also entails non-eternity of the Buddhist hell, as well as a universalist view of salvation. Although the karmma-based hell is not held hostage to the Problem of Hell that exists in Christianity, it also has a problem of its own. The tension between the doctrines of kamma and dependent origination is known as the Buddhist free will problem, as it involves the metaphysical discussions about free will and determinism. By drawing on Berlin’s two senses of freedom in Western philosophy, this essay builds a conventional conception of freedom for Theravāda tradition of Buddhism and proposes a compatibilist solution to the problem. Hopefully, this defense of the kamma doctrine can be an endorsement of the kamma-based hell.