Doctrine & Reason II: Morality & Karma

4. Reasoning about Morality

4.1 Views

In this section of the essay I will extend my critique of karma by focussing on some general ideas related to reasoning and karma; and then in Part III, I will outline a more specific critique of Nāgārjuna's approach to karma.

One of the functions of Dharma study in the Triratna Buddhist Order is to help identify and quantify our existing views, or what Justin L Barrett (2004) calls non-reflective beliefs. This process is more effective if we experience cognitive dissonance. That is, when our beliefs are challenged by information that is true, but seems to contradict our worldview. Without the dissonance, we might never even know that we had a view, since many non-reflective views are transparent to introspection. Simply asking ourselves what our views are, is ineffective. Creating cognitive dissonance is a reliable method of identifying such views. Our negative reactions are a window into the unconscious, non-reflective belief system that shapes our worldview. This worldview can be quite different from our professions of belief and the doctrines that we recite as religieux. So our reactivity is an important clue to where our true beliefs lie. Which I find salutary. On the plus side, reactivity is part of the process as long as we reflect on it.

Ironically, converting to Buddhism often involves consciously taking on a whole raft of views. It can take many years to internalise these views, but one the goals of conversion is to achieve fluency in the jargon and behavioural quirks of a particular Buddhist group. Part of the goal is to develop what John Searle (1995) calls dispositions. Dispositions are generalised behavioural habits that allow us to behave in ways that are consistent with social norms, without us having to consciously follow rules all the time. As we internalise the rules, following them becomes a background capability. As Subhuti indicates (Part I), such beliefs and dispositions can be seen as constitutive of being Buddhist. The moment when one is acknowledged to have internalised the norms of the group and is accepted as a full member of the group is an important step in the religious life. In the Triratna Buddhist Order this corresponds with ordination.

When membership of the group is predicated on endorsing certain views, we can always find confirmation of such views from other members of the group. The resulting effect is called an echo-chamber (where we all reflect each other's view) and it can lead to groupthink, a cognitive bias in which we all go along with a perceived consensus for fear of being socially isolated. This is part of a broader psychological phenomenon called social proof, in which we judge the safest course of action to be doing what everyone else is doing (this may be related to the basic level of empathy, i.e., emotional contagion). On the one hand, these are behaviours we can expect to find in all social animals because they strengthen social bonds and promote security and, on the other, they are highly limiting for individuals and suppress reasoning. If our views are to be compatible with reasoning, then we cannot simply go along with what everyone else says or does, even if that means that our membership of the group is threatened.

Few people step outside and confront intelligent criticism of their views. In the case of Buddhism, very little intelligent criticism of Buddhist doctrine exists. So, even if one does step outside, one goes from an echo-chamber to a virtual vacuum. For example, in my collection of articles on karma I have just two that make any attempt to assess the idea on its merits. If one wanted to critically evaluate karma based on published sources, one would find two kinds of literature: religious apologetics and scholarly works that take Buddhism on its own terms. To my knowledge, there is no general survey of the dozen or so competing Buddhist accounts of karma, no critical or comparative studies of these views, no sense even that Buddhists might disagree on the subject of karma. There is no attempt to reconcile karma with modernity or to acknowledge the difficulty of such a project.

As Subhuti hints above, Buddhist karma is inextricably linked to rebirth. Karma refers to a variety of doctrines which boil down to: actions cause rebirth. The quality of our actions in this life determine the quality of our rebirth, unless we are liberated. The primary goal of traditional Buddhism is to end rebirth, either for oneself or for everyone. Since both virtue (dhamma) and vice (adhamma) lead to rebirth (See Thag 304), albeit better (sugati) and worse (duggati) rebirths, the Buddhist has to transcend all willed activity that might lead to rebirth. This way of looking at the Bauddhadharma has a flavour of the Jainadharma to it, as it was the Jains who saw all activity as karmic and resolved to do no action - the acme of which was to sit in meditation until one died, probably from dehydration or starvation.

Unfortunately, beyond this bare outline, almost every detail of karma doctrine is disputed, and some modern proponents of karma theory would dispute even this much. Karma is a rubric for a wide range of views on morality, many of which are mutually exclusive. So reasoning about karma is much more difficult that it seems at first glance, because we first have to establish which karma belief is being reasoned about.

4.2 Actions have Consequences

One common view on karma mistakenly equates it to cause and effect. Karma is not a theory of cause and effect. Karma only applies to our willed actions and the vast majority of events in the universe are not caused by willed actions. Humans are just one species, on just one planet, in an observable universe of two trillion galaxies, each with about 100 billion stars. So let's not overstate our role in matters.

Based on our past interactions with people, we can deduce that certain types of behaviours have desirable consequences and others have undesirable consequences. All social animals have a disposition to being prosocial, but each group has its own aesthetic norms that we must learn and internalise. A group is the sum of the personalities of the individuals that make it up, though a social group may be a structure that has emergent properties (culture).

By the time they reach adulthood, a young social mammal has to had have enough experience to generalise about what kind of interactions are favourable and which are not, in their group. They have to have internalised these as dispositions to enable them behave within social norms most of the time without having to laboriously reference rules consciously. Social animals typically acknowledge that gaining this knowledge and experience takes time. For example, social animals are typically very tolerant of infants, but have higher expectations as group members approach sexual maturity. Most social mammals have a very low tolerance of anti-social behaviour amongst adults.

We humans don't always get this right. For example, we may behave in an antisocial manner and be treated roughly, but come to the wrong conclusion. We may not have insight into our own role in the interaction, conclude that other people are mean, and continue to act in antisocial manner. Our modern, large impersonal cities regularly produce anti-social behaviour that becomes entrenched.

That said, I think probably every one I know would consider the proposition that actions have consequences to be a self-evident statement of fact. Which suggests that it is a belief at the non-reflective level, albeit one that is accessible to introspection. For most people it is simply axiomatic in human relations. Even people who are habitually antisocial seem to understand that their actions have consequences. Often, the problem is that they do not identify with the (often large, amorphous) group that is trying to obtain their compliance, but have allegiance to a local, more personalised group. Indeed, the acceptance of actions have consequences seems to be universal amongst humans and to have analogues (at least) amongst other social animals.

Some of us take the axiom, actions have consequences, to be a summary of the doctrine of karma. But this is not what the traditional view was. Traditionally, karma is always linked to rebirth. Karma usually says that we are reborn according to the weight and kind of our deeds in life; or if liberated, we are not reborn at all. That is: karma causes rebirth. In some versions of the karma doctrine, karma may also ripen as an experience (vedanā) within a rebirth, but rebirth is still the primary manifestation of karma. Karma without rebirth is not karma. Those who leave rebirth out, are often adjusting karma to fit with a secular humanist outlook. I have some sympathy with this approach, but calling it "karma" seems to miss the point. All karma is moralistic; but not all morality is karmic.

"Actions have consequences" is about the vaguest true statement about morality that one could possibly make. It is true, but it doesn't say much. Everybody knows it and the vast majority internalise it as infants. There is nothing particularly Buddhist about it. If we are talking about karma then there is something missing (apart from rebirth). By this I mean that there is no implication that the consequences will be appropriate to the action. And this leads us naturally into considering fairness and justice.


4.2 Fairness and Justice

"Actions have consequences" is the minimum one needs to know to understand how to behave in social groups. Probably all social mammals have some grasp of this concept. But, on its own, it is hardly good enough to explain morality. Morality, according to Frans de Waal, is based on two qualities found in social animals: empathy and reciprocity. I've essayed these two qualities (See The Evolution of Morality), but here will focus on reciprocity and its implications for karma.

Social animals understand reciprocity. It is a feature of the social lifestyle that animals share what they have. At a minimum, they club together to share food resources, for defence, and especially in social primates, for access to mates. And part of reciprocity involves keeping track of sharing: who shares with us, in particular. We preferentially share with those who share with us and we don't share with those who don't share with us. As I noted in my exposition of Frans de Waal's ideas on the evolution of morality, this entails a basically generous disposition: in principle, each individual must be prepared to start off sharing, else no one would share with anyone. This disposition to generosity and sharing means than when another group member does not share we are attuned to this as a kind of threat to the group. When things are unfair we feel it.

The classic example of this is the much watched video of the capuchin monkey who, perceiving that his comrade is receiving better rewards for the same task, angrily throws his food back at the researchers and screeches at her. All of the social animals which have been tested seem to grasp basic fairness related to rewards for tasks. They all notice another individual getting too much reward. Chimps also notice an individual getting too little and are able to reason out that the one getting too little may retaliate later on.

Even for these animals, who are trained to perform actions for rewards and thus fully comprehend that actions have consequences, the consequences cannot be arbitrary. They will perform the task for a lesser reward as long as the others are getting the same reward. They expect the rewards to be fair.

For consequences to be fair they must be appropriate to the action and appear in a timely manner. These two restrictions (niyāma) are explicit in Buddhist karma doctrine, though they are not always formalised, and even then not until relatively late. In Buddhaghosa's conception of niyāma, appropriateness is symbolised by the rice seed giving rise to a rice plant and is called the germinal-restriction or bīja-niyāma. Results appearing in a timely manner are symbolised by the timely arrival of the monsoon or the season for flowering or fruiting; and this is called the seasonal-restriction or utu-niyāma. Buddhaghosa added a third restriction which was that consequences of willed actions were inescapable, which he called the action-restriction or kamma-niyāma. As I explored in my article Escaping the Inescapable (Attwood 2014), this restriction was deprecated by Mahāyānists who proposed that consequences could be avoided through religious exercises. The idea of avoiding karma through religious exercises probably came from Jainism, since it is a characteristic of their religion, but absent from early Buddhist accounts of Buddhism.

Justice involves the idea that unfair situations can be made fair by taking actions. And as morality is based on reciprocity, justice is often seen in terms of balance or debt. An antisocial action upsets the balance or creates a debt. It must be balanced out by pro-social actions, or the debt must be paid in kind. A criminal has to "pay their debt to society". Arguably, the capuchin, by displaying and flinging food, was taking action to restore fairness and was thus pursuing justice.

It is not enough for actions to have consequences; we have to add some restrictions, some niyāmas. The consequences of actions have to have be appropriate and timely for the situation to seem fair. This principle is starting to look sufficiently sophisticated to account for morality. It is certainly broad enough to encompass many definitions of fairness and justice. Some accounts of karma stress that one gets the rebirth one deserves; one's life is a logical consequence of actions in a previous life.; and so on. Where we get squeamish is when someone who is not obviously evil is struggling with some burden like congenital illness. We wonder what they can have done in a past life to deserve such a fate. Blaming the victim for their misfortune is an unfortunate aspect of the just-world fallacy.

This brings up the major problem that we have. The whole point of the idea that actions have consequences is that I suffer the consequences of my actions; and you suffer the consequences of your actions. Additionally, I may suffer from your actions and vice versa, and this must figure in any rational moral theory, but karma doctrines emphasise the way my actions lead to rebirth for me. The connection between action and consequence is specific and completely non-random. There is no question that if you suffer the consequences of my actions, that you will perceive this as unfair and unjust. The problem is that karma is restricted by dependent arising. And dependent arising, as usually interpreted, disallows any persistence of effect beyond the cessation of its condition (imassa nirodhā idaṃ nirujjhati). The best we can offer is that the person who suffers the consequences of the action arises in dependence on the agent of the action. This is a rather distant relationship compared to most moral formulas.


4.2.1 Fairness in Buddhism

In fact, this distance between action and consequence must have troubled Indian Buddhists as well. We know this because their literature is full of morality tales, in which actions in a past life have consequences for the same person in their present life. These are the jātaka stories (which occur throughout the literature as well as in a specific collections such as Jātakapāḷi and Apadāna). The typical framework is that something happens, then the Buddha tells a story of a deed done in the past, and concludes with identifying how the event in the present is affecting people now because of their behaviour in previous lives, explicitly identifying them as the same people. In jātaka stories there is a direct line from action in the previous life to consequence in this one. It seems to have escaped the sustained attention of scholars that this is a direct contradiction of dependent arising. Despite the contradiction, jātaka stories became enormously popular in Indian Buddhism. They are the main theme of stūpa decorations during the Asoka period, for example. And they are the main vehicle for teaching morality in Theravāda countries down to the present.

So Buddhists maintain two distinct narratives in relation to morality, one which emphasises pragmatic morality and the other which emphasises a negative metaphysics of self.

1. My actions have appropriate and timely consequences for me, especially rebirth.
2. There is no "I" who can will actions.

The generally unacknowledged corollary of 2. is that without an "I" there can be no morality. Nāgārjuna does more or less acknowledge it at the end of Chp 17 of Mūlamadhyamaka-kārikā, but it earns him the sobriquet of nihilist from the rest of the Buddhist world, to the extent they were aware of him (he appears to have had zero influence on Theravāda Buddhism, for example). A closer look at Nāgārjuna's answer to the problem of karma makes up the bulk of Part III.

Modern day Buddhists tend to segue between these two narratives so seamlessly that it took me 20 years to notice it happening. In talking about morality we insist actions have appropriate and timely consequences for the individual. In talking about metaphysics we assert that the self is at best an illusion, and therefore the one experiencing the consequences is not the same as the agent who performed the action, but not different. When this threatens to undermine morality, we slip back into talking about actions and consequences.


4.3 Just World Fallacy

In our social world we expect the consequences of our actions to have appropriate and timely consequences. In other words, we expect other members of our group to act predictably. Social living is predicated on us all having predictable responses to social situations. It doesn't matter so much what the norms of the group are, with some limitations; as long as most members of the group follow the rules most of the time, the group will thrive. How a group of humans live is as much a matter of aesthetics as morality and practicalities.

This expectation is rational within a social setting, but we go a step further. We infer that if social actions have appropriate and timely consequences, then the whole world ought function this way. This is the just-world fallacy. It is a fallacy because, although the world does follow rules (or least rule-like paths), it does not follow our social rules. The universe does not care about us any more than a landslide, earthquake, or volcano cares about us. Where humans are constantly modifying their behaviour in response to each other, the universe never modifies its behaviour in response to humans. The universe has no trolley problem; it would never swerve to avoid killing anyone.

A contributing factor to this fallacy may be animism, which is the most ubiquitous supernatural belief. In this view the world is full of supernatural beings, often called "spirits". In Burma they are nats; in Japan kami; in India devas, and so on. Most people, for most of human history, seem to have believed that spirits inhabited the world around them. And these spirits are commonly seen as part of the community. Special people called, shaman, had the role of mediating between physical beings and spirits. It was only natural that any expectations of our social group would extend to these non-material members as well, though they are often more capricious than humans. The extension of this expectation to nature as a whole is no stretch of the imagination.

However it came about, it is common for people to assume that the rules of social interaction apply across the board; i.e., that the world itself is, or ought to be, fair. What this means is that if we are, on the whole, good, we expect good things to happen; not only amongst our group, but generally in the world. We expect to be lucky, for example, to have good fortune, to avoid misfortune. The corollary of this is that if we experience misfortune, it may be that we have inadvertently transgressed or that someone has used magic against us.

It does not take a genius to see that the world is not fair. Consequences of actions are not always (I would say seldom) appropriate and timely. Another aspect of a pre-scientific worldview that is almost universal is belief in life-after-death. An afterlife seems plausible for any number of reasons, including out-of-body experiences, near-death experience, dreams, hallucinations, etc., that give credence to the necessary Cartesian-style mind-body dualism. Also, we have a very strong desire to continuing living, which operates on many levels. So, in this sense, anything which seems to confirm the existence of an afterlife is willingly accepted and any counterfactual information is quietly buried.

The presumed existence of life after death provides a neat solution to the problem created by consequences which are inappropriate and/or untimely. Everything is balanced out in the afterlife. This may involve literal weighing up of actions, or a ledger of good and bad deed or, in the case of karma, some mechanism which is unclear, but produces the right results (what I call a black-box function)

Having dealt with a number of generalities regarding morality, in the next part of this essay I will turn to the specific subject of Nāgārjuna's approach to karma as found in Chapter 17 of the Mūlamadhyamakakārikā.

~~oOo~~

Preceded by Part I | Continued by Part III

Bibliography

Attwood, Jayarava. (2014). Escaping the Inescapable: Changes in Buddhist Karma. Journal of Buddhist Ethics, 21, 503-535. http://blogs.dickinson.edu/buddhistethics/2014/06/04/changes-in-buddhist-karma

Barrett, Justin L. (2004). Why Would Anyone Believe in God? Altamira Press.

Kalupahana, David J. (1986) Nāgārjuna, The Philosophy of the Middle Way: Mūlamadhyamakakārikā. SUNY.

Mercier, Hugo & Sperber, Dan. (2011). Why Do Humans Reason. Arguments for an Argumentative Theory. Behavioral and Brain Sciences. 34: 57 – 111. doi:10.1017/S0140525X10000968. Available from Dan Sperber's website.

Searle, John R. (1995). The Construction of Social reality. Penguin.

Subhuti (2007) There are Limits or Buddhism with Beliefs.

Yang, J. H., Barnidgeb, M. and Rojasa, H. (2017) The politics of “Unfriending”: User filtration in response to political disagreement on social media. Computers in Human Behavior 70, May 2017: 22–29

Doctrine & Reason I: Reasoning.

In my forthcoming book on karma and rebirth I cite one of the leaders of the Triratna Buddhist Order on the importance of beliefs in Buddhism. In his 2007 essay, There Are Limits, Subhuti says:

“These essential principles of the Dharma set out how existence works and are therefore the basis for a Buddhist life. Just as a Christian life is based on belief in God’s creation, Christ’s sacrifice, and salvation through faith in him and works in accordance with God’s commandments, so a Buddhist life is based on belief in conditionality, karma (including ‘rebirth’), and the Path – albeit Buddhist belief being provisional, compatible with reason, and capable of direct verification. Without conviction that these are the essential mechanics of life, one will not practice the Dharma.” [Emphasis added] Subhuti (2007)

In this essay I will focus on the phrase, compatible with reason. What is reason, or more specifically, what is the activity of reasoning? What would it mean for a religious belief to be compatible with reason? Having addressed these general questions, I will use the example of the Buddhist belief in karma. I choose karma because Subhuti mentions it and because I know the various karma doctrines of karma fairly well.

The first snag that we hit comes almost immediately because, based on discussions over the years, I can identify around a dozen different views on karma currently held by members of our Order, some of which are mutually incompatible. So belief in karma is not a simple matter. The threads of karma doctrine form a tangled mess that take a book length project to unravel (not that my book will do this). In this essay I will, therefore, take one view of karma that is fairly common in our Order and test its compatibility with reason. This view is the one deriving from Nāgārjuna's treatment of the subject in his Mūlamadhyamakakārikā, though this source is seldom acknowledged and the view is generally absorbed by reading modern day Tibetan Madhyamaka philosophy of one kind or another.

1. Reason

The second snag is that what Subhuti means by "reason" is not entirely clear. My Dictionary of Philosophy opens its entry on reason by explaining that reason is,

"A word used in many, various, often vague senses, with complex and sometimes obscure connections with one another."

The dictionary goes on to note that one important distinction is between reason and other mental qualities such as "imagination, experience, passion or faith." I think Subhuti probably has something like this distinction in mind. The implied comparison with Christian articles of faith reinforces the impression. In other words, Subhuti seems to be referencing the common distinction between faith and reason as basis for belief. This distinction has been a feature of Christian theology and the focus of a lot of debate about religion in modern times.

Actually, in theology, faith and reason are both authorities for belief. Faith is usually considered to be the basis of belief, but some theologians have attempted to use reason to prove articles of faith. Faith is clearly important in Buddhist life. In the classical Pāḷi texts, faith (saddhā) arises when one hears a dhamma teaching, i.e., at DN 2 (dhammaṃ suṇāti), and at AN 10.61 and AN 10.62 (saddhammassavana). Faith, here, is faith in the Buddha (tathāgate saddhā). All too often, Buddhists (and particularly Triratna Buddhists) insist that Buddhism does not involve blind faith and that saddha (Skt śraddhā) is not blind faith. The Pāḷi texts make it clear that saddhā is precisely faith in the words of a religious teacher, lacking demonstrability, at least for the for the moment.

Having practised the methods of Buddhism with success (by which is generally meant, becoming a stream entrant) one may also develop another quality, aveccapasāda ‘confirmed confidence’ (or perhaps ‘perfect clarity’). So, until stream entry, until we join the āriyasaṃgha, our motivation to practice is based on faith. Beyond this we see people taking Buddhism on faith all the time. Most Buddhists take the possibility of enlightenment on faith, and have to, because there are no enlightened people around. For Buddhists, as for other religieux, belief is based on faith.

As the Internet Encyclopedia of Philosophy (IEP) says of reason, "Some kind of algorithmic demonstrability is ordinarily presupposed." Here we see why Subhuti might have included "capable of direct verification" as a criterion. To go beyond faith we have to have a procedure to test our belief and see where it leads us. Note that though Subhuti's actual expression appears to be that of a logical positivist, this was probably unintentional. What he means is that the promised results of legend are said to be attainable by anyone. Of course, this statement is also an article of faith.

Subhuti has said only that belief must be compatible with reason; not that it must be based on reason. This implies that belief may still be based on faith, as long as reason does not subsequently dis-confirm it. This gives us a little more room to manoeuvre. Most rationalists see faith and reason as antagonistic, at best and as polar opposites, at worst. However, in so-called Natural Theology, for example, "Articles of faith can be demonstrated by reason, either deductively (from widely shared theological premises) or inductively (from common experiences)." (IEP).

There is a subtle move here from reason as a faculty of the mind, to reasoning as a method for producing knowledge through the application of logical inferences (deduction and inductions). This is not necessarily problematic, because reason is often associated with the ability to employ the methods of reasoning. However, it is worth noting the tacit shift from reason as a faculty that exists (ontology) to reasoning as a method of obtaining knowledge (epistemology). The confusion between the ontology and epistemology is a major problem in philosophy.

For example, before materialists ask "What is real?" they divide the world into mental and physical phenomena based, as all such divisions are based, on the epistemological differences engendered by our perceptual apparatus. They conclude that only physical phenomena are real. But this result has confused ontology with epistemology. "Mental" and "physical" are epistemological distinctions. The question is like asking: "Which is more real, hearing, vision, smell, or taste?" Which is to say the question is nonsensical. Also, the question of which kind of experience corresponds to reality is predicated on using mental phenomena to judge the truth. If mental phenomena are not real, then how can they produce accurate judgements on what is real? And so on.

If there is an equivalence between compatibility with reason and conforming to the procedures of reasoning the we have an obvious way to test Subhuti's assertion. Can we, for example, derive the details of the Buddhist belief in karma from first principles? That is to say, can we arrive at a doctrine of karma by applying the various modes of reasoning? In order to answer this question we need to look more closely at how reasoning works.

2. Reasoning

In the passages that follow I'll look at the three most common methods associated with reasoning: deduction, induction, and abduction. All these logic words derive from Latin ducere 'to lead' hence: de-duce, 'to lead down' or derive; in-duce 'to lead onwards' or predict; and ab-duce 'to lead away' or explain. We can see why abduct is a synonym for kidnap! From the verb ducere we also get Mussolini's title Il Duce i.e The Leader; other titles such as duke and dux; and a whole raft of other English words: adduce, conduce, conduit, douche, duct, ductile, educate, induct, introduce, produce, product, redoubt, reduce, seduce, subdue, and traduce.

Deduction, induction, and abduction are all methods of inferring new knowledge from something already known. I'll begin, as most philosophers do, considering these activities as solo events, but I will also reconsider them as collective activities, which Mercier and Sperber (2011) have argued is the natural context for reason. Although I will not recapitulate Mercier and Sperber's arguments here, I will have them constantly in mind. Their most important observation, which is by no means original or new, is that in solo reasoning tasks most people score so badly that they cannot be said to be reasoning at all. Instead, they rely on cognitive bias and logical fallacies. Mercier and Sperber point out that, by contrast, when critiquing someone else's argument in a small group setting, most people do very much better. In other words, when producing arguments we don't use reason, but when evaluating someone else's argument we do. Thus, they argue that reasoning is argumentative. A corollary of this is that confirmation bias is a feature (and perhaps even a necessary feature) of argument production, though not of argument evaluation.

2.1 Deduction

Using deductions, we try to infer conclusions based on our set of axioms about how the world works. These axioms are what Justin L. Barrett (2004) has called our non-reflective beliefs.* Non-reflective beliefs include our views on such metaphysical concepts as time, space, and causation. These are the beliefs that we absorb while we are growing up, both from our experience of interacting with objects and from interacting with people. We may not know we have these beliefs and they may not even be immediately accessible to introspection. Nevertheless, these axioms are central to how we understand the world.

* I discussed Barrett's ideas in a two part essay called Why Are Karma and Rebirth (Still) Plausible (for Many People)?

This kind of reasoning involves asking ourselves, in the light of our axioms about the world, what event, or sequence of events could have occurred to bring about the current state of affairs. For example, most of us non-reflectively believe that there are agents behind most events. So, based on the available information, we may try to deduce what kind of agent was responsible and what their motivations might have been, based on our internal models of what agents exist and what kinds of events they can cause. So we might hear an eerie cry in the night and experience horripilation, but deduce that this is the kind of noise a fox makes and conclude we are safe. Since most of us include supernatural elements in our non-reflective beliefs, it often seems intuitive, or at least minimally counter-intuitive, to conclude that an experience has a supernatural cause.

The IEP citation above referred to "widely shared theological premises". This highlights a problem with reasoning with respect to religious beliefs. A deduction from widely shared theological premises is likely to reinforce those same widely shared theological premises. If our widely held theological premise is that the Christian God exists, then deductions we make about, for example, how the world came into being or what is moral, are predetermined by our axioms. We may well perform a perfectly logical deduction from our premise, but this signifies little because the starting premise or axiom was not arrived at by reasoning.

Similarly for our theological premise that karma is, in Subhuti's words, how existence works. What we have done is decide a priori that karma is how existence works and then set out to look for confirmation of this axiom. This is a cognitive bias called confirmation bias. As noted, Mercier and Sperber (2011) have argued that confirmation bias is ubiquitous in argument production, but seldom in found in argument assessment, unless one already agrees with the argument. So, getting a believer to critique and argue for karma is pointless. To get rational, objective feedback, one must get feedback from a non-believer, but not one who is so hostile the belief that they cannot think rationally about it.

Looking for confirmation of our beliefs is not rational, because of the Black Swan Effect. This means no matter how many times we confirm our view, some evidence may still come along that falsifies it. Thus, a tenet of rationality is that one ought to seek falsification rather than confirmation, which for most people is counterintuitive. Most of us, set a problem in which we have a choice between seeking confirmation and seeking falsification of a belief, opt to seek confirmation. We look for evidence to support our argument. We only try to disconfirm arguments produced by others.This is an important observation: what is intuitive is not necessarily rational and vice versa.

If our in-group is Buddhist, then our argument is typically with out-group non-Buddhists. Within the group we tend to confirm and reinforce each other's views (which is not compatible with reason), while without we argue against the other's views (which is compatible with reason). This suggests that most of the time in-group beliefs won't be compatible with reason; and that reasoning about our views can only be found with those who disagree with us. This failure of groups to produce an internal critique can lead to groupthink, another cognitive bias in which the desire for harmony or conformity overwhelms reasoning in a group. In this sense, the wide range of incompatible views on karma in the Triratna Buddhist Order is a good thing, or it would be, if people were willing to argue about their views (there is some resistance to arguing with me about views, I find).

One of the buzz-words of the day is echo-chamber. This compound word was coined to refer to the mistaken view that our social media environments tend to restrict our exposure to dissenting political views, so that we end up only seeing and hearing views which seem to confirm our own view. Yang et al (2017) showed that, in fact, social media exposes people to more dissent rather than less. Making deductions about the world based on widely held religious premises is only ever going to result in our conclusions echoing our existing beliefs.

Deduction is a useful tool for reasoning, but it has rather severe limitations. When it comes to reasoning about beliefs, that limitation becomes catastrophic if our articles of faith are taken as axiomatic. Since articles of faith are treated as axioms and not themselves arrived at by reasoning, the danger is that our conclusions simply reflect our existing beliefs. Logic and reason are not always the same thing. Deductions logically derived from irrational axioms can and will be irrational.

So in this sense I disagree with the Natural Theology crowd that deduction enables us to reason about belief. Deduction is completely dependent on what we believe.

2.2 Induction

We use inductive reasoning to arrive at generalisations about experience and to form rules of thumb for dealing with similar experiences. Generalisations are possible because experience has patterns. Experience has patterns because the world evolves in regular ways, our minds operate in regular ways and experience is a function of both. A lot of induction relies on the general principle that the future will most likely be like the past. Probabilities are an important form of generalisation about the future.

For a generalisation to be valid does not require that all experiences confirm it. It's not like a law of nature. If 80% of experiences fall into known categories, then it can be more efficient to proceed as if they all will and be alert for exceptions, than to have to assess each experience individually. It's like a compression algorithm that only notes the parts of a video that are changing. There's no need to compute the whole picture every time if large chunks of the background are not changing. Of course, if we don't notice the exceptions, then we are led into error by generalisations.

One thing I want to flag here is the problem of generalising from a single or rare experience. A made up example might be that I try mint and licorice ice-cream and I conclude that I do not like ice-cream. This is an over-generalisation, because mint and licorice is an unusual flavour and there are more conventional choices that I probably would like.

Another problem is when we combine this with confirmation bias. For example, astrology may seem to make sense if we generalise from the predictions that confirm our belief and ignore those that do not. A random prediction is likely to be right some of the time. By filtering out all the times the prediction is wrong, we come to the conclusion that astrology is generally pretty accurate. This aggressive filtering of experience is not only possible, but very likely to happen. And it explains the persistent popularity of irrational claims like those made by astrologers.

These kinds of generalisations from one experience, or just a few experiences, are extremely prone to cognitive bias. And many of our experiences in meditation are unique or unusual. But even if they are not, they tend only to coincide with being in an altered state of consciousness. Thus, we ought to be wary of generalising on the basis of them. However, Buddhists often rush to the conclusion that is supported by the norms of the group. A vision in meditation is not an hallucination, but a confirmation of the transcendent reality that our latter-day Buddhists metaphysics describes.

Another problem we Buddhists face is the premise that what applies in meditation is applicable everywhere; i.e., that features of our awareness that we identify in the altered states achieved in meditation are general features of awareness or, indeed, general features of reality. If we stop to consider this, it is quite a bizarre inference to make. The effort required to get into the altered state is considerable and the state itself is so qualitatively different from any other kind of experience. The very fact that I can describe these as altered states, reflects that they are unusual rather than common. Why would we choose to infer knowledge about reality on the basis of unusual experiences instead of usual experiences? Since it is common to be completely absorbed in these states and completely cut off from sensory perceptions of the world, why would we infer that they reflect the world more accurately?

Inductive reasoning is even more susceptible to bias than deductive, or at least susceptible to more kinds of bias that skew the conclusions we come to. One of the common biases is to see ourselves as less biased than other people (bias blind spot). Wikipedia has a list of almost 200 cognitive biases, most of which apply to the process of inductive reasoning. In his Cognitive Bias Cheat Sheet, Buster Benson has provided a more structured overview of cognitive bias that I find quite useful. Biases creep in when we have too much information, or too little, or when we are trying to arrive at an answer via a shortcut (which we do most of the time). Too much information creates the secondary problem of what to remember.

So again, induction is not a very reliable way of reasoning about belief. Beliefs themselves create cognitive biases (like confirmation bias) that distort the reasoning process in favour of what we already believe. In fact, most of the time we arrive at a belief or a decision and then, and only then, we look for reasons to retrospectively justify our belief or decision. So, when you ask a Buddhist why they believe in karma and the answer is, "Because it seems intuitive", the first suspicion must be that it seems intuitive because it's what that person believes. Belief itself makes the belief seem intuitive and thus we will tend to infer that our belief is rational.

We have one more approach to reasoning. Is it any better?

2.3 Abduction

Abduction is the process by which we infer explanations from observations, and use these explanations to make predictions. Where deduction proceeds to a certain conclusion, and induction to rules of thumb, abduction seeks to produce the best explanation given some facts that do not allow for a certain conclusion. Whenever we "jump to a conclusion" we are using abduction. And in this lies the downside of abductive reasoning. Many of our shortcuts are motivated by cognitive bias or logical fallacy. So if we hoped for a fool-proof approach, we aren't going to find it in abduction.

One of the most famous applications of abduction is the quote by Sherlock Holmes that:

“When you have eliminated the impossible, whatever remains, however improbable, must be the truth.”

Eliminating impossible explanations is an important process in abductive reasoning. One of the reasons philosophers frequently refer to Occam's Razor (aka the principle of parsimony) is that it places a useful limit on how we should go about the process of producing explanations. Attributed to William of Ockham (c. 1287–1347), though it definitely existed before him, Occam's razor takes many forms, but the basic form is that the explanation which makes the least assumptions is best. This is sometimes over-simplified and presented in terms of "the simplest explanation is best". However, this version is not very useful. Sometimes a complex explanation is best because it makes fewer assumptions.

A great example this is the reasoning behind Jan Nattier's argument for the Heart Sutra being composed in China. The larger Prajñāpārmitā text (LPT) is taken from India to China via Central Asia, where it is translated, extracted, and framed to create the Heart Sutra, whereupon it is exported back to India, back-translated into Sanskrit, lengthened, and then re-transmitted to China. This is by no means a simple scenario. But it makes very few assumptions compared with other possible explanations of the available evidence. Red Pine, for example, has to assume that in addition to the Sanskrit and Chinese versions of the texts that we still possess, a separate large Prajñāpārmitā text with different wording was composed, transmitted to China, and then lost in both Sanskrit and Chinese, leaving only the Heart Sutra as a record of it. But this is hardly credible.

Also, some assumptions are more likely than others. If our explanation of events requires a miracle, then Hume's comments on miracles become pertinent:

"...no testimony is sufficient to establish a miracle, unless the testimony be of a kind, that its falsehood would be more miraculous, than the fact, which it endeavours to establish..."

Science at its best is the epitome of abduction at work. And science at its best also involves a process which we have only mentioned in passing, i.e., comparing notes. Since this is often left out of accounts of reasoning, I want to highlight it here. But first a word about salience.

2.4 Salience

Clearly, reasoning has some limitations. On our own, we may not reason at all, but take some shortcut or invoke a rule of thumb ,instead. Humans are poor at solo reasoning tasks because we fall victim to many cognitive biases and logical fallacies. However, even if we were competently reasoning, there are many cases in which given the same information two people would come up with entirely different conclusions, generalisations, and explanations. Nowhere is this more obvious than in the case of politics.

The Political Compass website assesses political affiliation on two axes: progressive-conservative (or economic left-right) and authoritarian-libertarian. But no matter which quadrant you end up in, anyone who takes the test has access to pretty much the same information. The differences come about because of salience. George Lakoff's (1995) explanations of the different underlying political metaphors of our divisions are very salient to my understanding of politics. They relate to the kind of family that feels right for us.

For example, conservatives conceptualise the nation as a self-contained family with a strong father-figure in charge, who is strict, strong, and calm. Part of being self-contained is always paying off debts promptly. Happiness is found by everyone playing their role and following the rules. Children are taught obedience and self-reliance. Liberals, by contrast, see the nation as a family of two equal parents in which everyone is cared for and loved. Happiness is found by playing positive roles in the community and work. Children are taught to love their parents and to care for themselves and others.

These metaphors underpin reasoning in the political domain. The conclusions, generalisations, and explanations produced by political reasoning are powerfully shaped to fit. So, given, to take a topical example, a shortfall of funding in the National Health Service, conservatives will tend to want to cut costs and balance the books, whereas liberals will want to insure that everyone is looked after. We used to pay a lot more in tax and the government had a number of income generating assets and businesses. But the Neoliberal (really a conservative libertarian) forced the government to lower taxes and to sell off assets, because of an ideological commitment to small government and minimal government involvement in individuals' lives. As Ronald Raygun framed it, it was about relieving the tax burden, tax here being a burden imposed on the individual, rather than a way of the community taking care of its own. But with low taxation and eliminating other sources of income comes a crisis in funding the health service. But the health service conflicts with conservative values in fundamental ways: it does not encourage self-reliance, people get something for nothing (creating a debt that cannot be paid), and people who injure themselves through carelessness or poor lifestyle choices get the same treatment as those who are careful and who make good lifestyle choices.

Different people reasoning about the same situation, with the same information, but coming to very different conclusions, making very different generalisations, and explaining the situation using very different principles.

I have often written about this: when we assess information, we may go through a cognitive procedure to assess its veracity, but we register how important the information is to us emotionally. This importance is what I call salience. The emotionality of salience is what enables us to have a "gut reaction" to news or to make "intuitive" choices.

Salience is also an aggressive filter on what we consider when making decisions. Given a wealth of information, we all filter it. Dealing with too much information is a major source of cognitive bias. If we always had to evaluate every option, we would be unable to make decisions at all. The solution to the problems of bias, error, and salience is comparing notes, to which I will now turn.

2.5 Comparing Notes

The evidence is that humans are extremely poor at solo reasoning tasks (Mercier and Sperber 2011). It's not very credible to assume that humans can reason things out on their own under normal circumstances.

In fact, we seem to have evolved reasoning in the context of decision making in small groups. And this means that all this reasoning needs to be reframed as a group activity. And, after all, we are social animals, we evolved to live in communal groups. This social aspect of human beings is all too often simply left out of accounts of how our minds work. When we look at other social mammals, many social relations are fully functional without language or abstract reasoning. This has led me to suggest that, although we typically see the hierarchy of science as going from individual psychology to collective sociology, in fact, sociology is more fundamental and so profoundly shapes our psyches that it ought to be the other way around; i.e., out of biology emerges sociology, which shapes the minds of individuals. Indeed, we are so attuned to our social environment that "individual psychology" may be an oxymoron.

Some years ago Sean Carroll's Twitter bio read,

"I'm sure if the blind guys had compared notes they'd have figured out it was an elephant."

For me this captured something important; not only about our search for knowledge, but the stories we tell about our search for knowledge. Comparing notes (in the form of literally comparing notes, but also of presenting results in seminars and conferences where they can be discussed, and in formal peer review prior to publication) is one of the things scientists do that makes science an effective knowledge seeking activity. Knowledge seeking is typically a collective activity. "Science is sold as facts and it's not, it's process. And that process is mainly arguing." (Edwards 2017)

It is precisely when we do not compare notes that we are most at risk of falling into some logical fallacy or cognitive bias. By comparing notes and, well, arguing about what they mean, we are more likely to be rational. Of course, groups are also prone to cognitive bias, so even then we must proceed with caution. For example, simply comparing notes in a naive way can be unproductive. We can uncritically accept the other's conclusions, generalisations, and explanations because they support our own.

When we compare notes uncritically we get a consensus reality. For example, part of consensus reality is the supernatural. If I have an experience and describe it as supernatural to someone who has come to a similar conclusion about some experience they have had, we may both reinforce the delusion of the other. Critical comparing of notes leads to what I call a collective empirical realism. In this approach there can be no unquestioned axioms. All axioms are up for discussion and criticism. Other people who participate in the comparing of notes critique methods as well as conclusions. By being sceptical about axioms, methods, and results, we can begin to eliminate the illogical and irrational elements that inevitably creep into our narratives, along with the other purely subjective elements.

What science does, that other forms of knowledge seeking do not, is to look at why different observers come to different conclusions or explanations. Scientists try to get at the underlying principles of our beliefs to see which are most consistent with reality. Hence, for the first few centuries of science, the emphasis was on reductionism. Given the human propensity for bias and error, we had to really get clear on the underlying substance and principles under discussion. And note that in the general population bias and error are still dominant forces. Supernatural beliefs are de rigueur, for example. Even within science, bias and error cannot be eliminated except by retrospectively subjecting results to collective criticism and weeding. Wrong results and claims are published all the time. But the approach of science means that before a result can be widely accepted it must be replicated and shown to fit in with the system of knowledge that has developed.

3.0 Compatible With Reason

The concept of reason is by no means straightforward. When we say that our beliefs are compatible with reason we are making some big assumptions. We assume we are capable of reasoning and capable of understanding when some belief of ours is compatible with reason. Looked at in the cold light of day these are doubtful assumptions. Our beliefs are much more likely to be unreflective assumptions based on bias and fallacy. Which may explain why our expectations and intentions are so very often thwarted.

Clearly, if humans are poor at reasoning, then a lot of what is said about reason is bunk. If you look up popular quotes on the subject, it is variously supposed to be what separates us from other animals (we called ourselves Homo sapiens); our highest faculty, a kind of pure and abstract virtue, the quality that helps us triumph over nature, etc., etc. But this is all bunk. Most of us don't reason, but instead rely on irrational rules of thumb and shortcuts. It's not that we are incapable of reason. We are certainly capable, but we prefer not to and have other means of arriving at decisions that we prefer to use instead. Yes, we can, if called upon, give reasons for our beliefs and decisions, but the overwhelming likelihood is that we did not use reason when arriving at them. For most of us, the best we manage on a day to day basis is post hoc rationalisation for beliefs we already hold or decisions we've already made without the benefit of reason.

Reason and reasoning have been widely misunderstood in history. For the most part they are still widely misunderstood. What is called "reason" is often something else entirely. All too often, it is simply ideology or some kind of Freudian wish-fulfilment fantasy. Those people who come across as more severely rational are almost always simply good at hiding the emotional basis of their decision making and good at persuading people. Most top politicians fall into this category: irrationally committed to an ideology, emotionally self-contained (and thus impervious to criticism), and highly persuasive. All qualities we might also associate with psychopaths.

If our religious identity resides in adopting certain beliefs, and that identity is important to us, then our ability to think clearly about belief is severely compromised. If we have made great sacrifices in our religious life—the extreme example is refraining from sexual activity—then our reasoning is always motivated towards confirming the value of our sacrifice. Which is why monks are such vocal apologists for Buddhism. Others will be inspired by such sacrifices and also want to confirm the value of them, since they get status by association. Outsiders can never appreciate the true meaning or significance of religious identity and their opinions hardly matter. Thus, religious belief becomes a self-sustaining process within a religious group.

We can see that compatible with reason is a very high bar to reach. Having explored the general issues surround reason and reasoning, in Part II of this essay, I'll begin to look at the reasoning behind morality, such as it is, because karma is the Buddhist explanation for morality. Part III will focus on assessing whether a particular version of karma doctrine is compatible with reason.

~~oOo~~

Continues | Part II | Part III |

Bibliography

Attwood, Jayarava. (2014). Escaping the Inescapable: Changes in Buddhist Karma. Journal of Buddhist Ethics, 21, 503-535. http://blogs.dickinson.edu/buddhistethics/2014/06/04/changes-in-buddhist-karma

Barrett, Justin L. (2004). Why Would Anyone Believe in God? Altamira Press.

Edwards, Tamsin. (2017) Inside Science [Interview on explaining science]. BBC Radio4. 12 Jan 2017.

Kalupahana, David J. (1986) Nāgārjuna, The Philosophy of the Middle Way: Mūlamadhyamakakārikā. SUNY.

Lakoff, George (1995) Metaphor, Morality, and Politics, Or, Why Conservatives Have Left Liberals In the Dust. http://www.wwcd.org/issues/Lakoff.html

Mercier, Hugo & Sperber, Dan. (2011). Why Do Humans Reason. Arguments for an Argumentative Theory. Behavioral and Brain Sciences. 34: 57 – 111. doi:10.1017/S0140525X10000968. Available from Dan Sperber's website.

Subhuti (2007) There are Limits or Buddhism with Beliefs. Privately Circulated. [This essay is not included on Subhuti's Website, nor is it included in the collection of his essays entitled, Seven Papers.]

Yang, J. H., Barnidgeb, M. and Rojasa, H. (2017) The politics of “Unfriending”: User filtration in response to political disagreement on social media. Computers in Human Behavior 70, May 2017: 22–29