10 June 2022

Prajñā And Sensory Deprivation

In the two preceding essays, I have explored the role that sensory deprivation (or monotony) might play in Buddhist practice, especially in relation to the threefold path: śīla, samādhi, and prajñā. Sensory deprivation occurs when we are deprived of sensory stimulation. Reading around the topic of sensory deprivation we find, for example, that Oliver Sacks (2012) describes some non-pathological hallucinations that sound identical to experiences described in Kamalashila's (1994) meditation manual. From this I infer that meditation does involve sensory deprivation and that we could interpret the hallucinations that meditators experience as resulting from the same mechanism. However, there is little or no mention of sensory deprivation in the scientific literature on meditation.

Having considered sensory deprivation and meditation, I then went back to reconsider śīla "conduct" in the light of this connection. This allowed me to make a connection with earlier research that showed the Spiral Path formulations to be an elaboration of the threefold path. Various kinds of correct conduct (kusalāni sīlāni) lead, progressively and cumulatively, to joy (pāmojja); joy opens the door to meditation (samādhi), which leads to knowledge and vision; this opens the door to paragnosis (prajñā) "knowledge from going beyond the sensorium". In this view, the whole process of liberation from rebirth is driven by good conduct and a clear conscience.

Lastly, I tried to make a case for the mechanism of non-pathological hallucinations being related to the brain's allostatic mode of functioning. Sensory experience is not simply the brain passively receiving sense stimulation and then reacting. Experience is as much prediction as perception. The brain is constantly active, constantly predicting what will happen next (at the level of patterns of neural activation) and comparing this with inputs in real time. The brain optimises its responses by minimising prediction error. Confronted with a discrepancy, the brain can either change the prediction or change the input (through actions). Past experience is our main guide to minimising prediction errors. Karl Friston has shown that minimising prediction error is mathematically equivalent to minimising the (informational) free energy, and is also (mathematically) related to Bayesian probability (aka The Bayesian Brain hypothesis).

Having dealt with the first two aspects of the threefold path, I now turn to paragnosis. I translate prajñā as paragnosis rather than "wisdom" for a couple of reasons. "Wisdom" is far too vague and seems unrelated to the texts at hand. Conze emphasised the word for the connections it afforded to Greek mystery religions and other esoterica of the type favoured by Madame Blavatsky and company. The English "wisdom" is cognate with vidyā and unrelated to jñā, which is cognate with Greek gnōsis, and Germanic know. My sense is that "wisdom" is simply wrong. Note also that "insight" has long been used to translate vipaśyanā, so that does not seem right, either. I feel it is important to make a conceptual break with that old fraud Conze and his magical thinking. So, having discovered it, I took the Greco-Christian term paragnosis "knowledge from beyond) and repurposed it to mean "knowledge that comes from going beyond the cessation of sensory experience". Repurposing words in a venerable Buddhist tradition that I have sometimes referred "Humpty Dumpty linguistics" (e.g. Attwood 2018a).

Life and Death in Ancient India

Buddhism was formulated during a period of radical socio-political change in India. Around the sixth century before the Common Era, new city-based kingdoms began to emerge from more distributed, village-based societies, resulting in what we call the Second Urbanisation (the first being the urban centres created by the Indus Valley civilisation). It was in and around these cities that religions like Buddhism, Jainism, and Ājivaka-ism were forged and tempered. It was also here that new forms of Brahmanical religion emerged, notably the idea of interiorized rituals that may well have kicked off the whole meditation craze. Buddhism shares many features with other Indian religions, including very often shared stories. For example, elements of the Dhammapāda and the Jātaka appear to have been drawn from a general pool of such stories that were common property and appear in the Mahābhārata.

An important shared feature of Indian religions was/is the cyclic afterlife (reincarnation/rebirth). This is not found in mainstream European cultures so was likely an idea the Indic-speakers picked up from the indigenous peoples they met and merged with  after they arrived in India. Accordingly, reincarnation plays no role in the oldest Ṛgveda stories, but it begins to creep in with the last book to be composed (Book 10). Rebirth features in the Bṛhadāranyaka Upaniṣad, composed around the very beginning of the second urbanisation. Michael Witzel and Signe Cohen (see Cohen 2018) have both argued that BU was composed in the city-state of Kosala, far away from the homeland of the Brahmins in the Punjab. They further suggest that BU reflects the concerns of a breakaway group who had migrated eastwards into the Kosala region and who challenged the hegemony of the Ṛgveda priests.

Another shared feature of different groups at this time was a form of philosophical explanation. In the literature these people left behind, explanations typically rely on analogy, often with nature. In particular, unseen processes are explained by analogy to seen processes. For example, karma is often analogised with growth of a seed into a mature plant. We refer to the consequence of an action as a "fruit" (phala) or as vipāka "ripe, mature". This reliance on analogy is important because I think there is an unspoken analogy at the heart of Buddhist soteriology:

the cessation of sensory experience is like death.

The ancient Indian literature is pre-scientific. The authors did have explanations for things. Some of their explanations are quite systematic, even. But this does not equate to science. For example, ancient Indians appear to have many misconceptions about human reproduction. In ancient Indian thought, a woman is (like) a field, which a man "ploughs" with his penis, planting his "seed" in the form of semen, which then (magically) grows into a human being inside the woman. This agricultural simile is found in both Buddhist and Brahmanical texts. Such similes appear to transcend religious differences, just as cultural ideas like a cyclic afterlife do. But this explanation is based on a raft of misconceptions that were not rightly conceived until the advent of empiricism in the eighteenth century. It took a couple of centuries to arrive at an objective account of human reproduction, and even now there is obviously some confusion amongst religious fundamentalists. 

Similes and metaphors play a central role in these analogies. The basic form of explanation in this: X is like Y, where Y is something that is familiar. In this way any unknown can be, and was, understood with reference to the known. This form of explanation has broad applicability. The explanations that emerge from it are often colourful and entertaining. Considerable creativity has gone into these explanations. We can certainly appreciate the intent and the achievement.

Buddhists themselves were clearly not entirely convinced by early Buddhism, however, since they universally felt they had to change it. And they changed it to the point of being unrecognisable. This includes, by the way, the Theravādins. Having abandoned āyatana practice for jhāna practice, Theravādins eventually also abandoned jhāna for analytical style meditations, and then abandoned meditation entirely for many centuries, before reinventing meditation in the eighteenth and nineteenth centuries. Those who see early Buddhist texts as authoritative are left hanging by the fact that all Buddhist sects moved on, developed and innovated doctrines and practices, and they kept on doing so right down to the present. Buddhisms like Zen or Pureland or Tantric cannot be traced back to early Buddhism. Rather these are inventions of a medieval culture that rejected early Buddhism as a "defective vehicle" (hīnayāna).

In the early Buddhist milieu, human life was widely believed to be cyclic. The basic idea, which one can easily discern in Pāli, is that one is born, lives, and dies in this world (ayaṃ loko), and after death one is born, lives, and dies in the other world (paraṃ loko). Again, this invites agricultural similes as explanations. This cyclic afterlife appears to have been a regional feature of cultures in and around the Ganga Valley. The evolution of this afterlife in Buddhism is described in great detail in Gananath Obeyesekere's book Imagining Karma (2002). The other world split into good destinations (sugati) and bad destinations (duggati), which Obeyesekere attributes to the idea of a right and wrong way to live, i.e. puṇyakarma and pāpakarma. For Buddhists, the ultimate problem was rebirth. Even the best possible rebirth is still problematic. Buddhists, and other religieux of that time and place, all conceived of ending rebirth as the summum bonum of their religion, though they disagreed on how to end it. For Brahmins liberation meant merging back into the eternal and unchanging Brahman, with the loss of all personal identity. For Buddhists it was left undefined: we cannot say anything about a person who escapes from rebirth. They are beyond all conception; neither existent, nor nonexistent.

Given this, we might ask how the cessation and absence of sensory experience would have functioned as the source domain of a cognitive metaphor for Buddhists, in other words, as the basis of an analogy. I say "might" because I don't know for sure, and I'm pretty sure no one else does either. I think what I'm about to say is plausible, given what we do know.

Buddhism and Rebirth

Modern English-speaking Buddhists often talk about rebirth as though it is inconsequential; as though we can just dispense with it. Although I don't believe traditional Buddhist accounts of rebirth any more than I believe traditional Buddhist accounts of human reproduction, I have always tried to acknowledge that from a traditional point of view, Buddhism is completely tied up with rebirth and the ending of rebirth. Fundamentalists see me as apostate, but fundamentalist Buddhism is more of a contradiction in terms than not believing in Iron Age afterlife theories. Even worse, many people seem to think of rebirth as a kind of backdoor to immortality. Europeans tend to see rebirth as a good thing. Buddhism always takes the opposite view, rebirth is the central problem that Buddhism addresses and it does this by eliminating rebirth. The Buddha was not reborn after his death and this caused problems for Buddhism  (but that's another story). 

By the late 1900s, many of the scholars involved in collating and publishing Pāli texts were rationalists in search of a replacement for Christianity, which they saw—in the spirit of the Enlightenment—as hopelessly mired in superstition. In presenting Buddhism to the world in English, they routinely bowdlerised it to fit their own preconceptions. They presented Buddhism as far more rational than it really is. David McMahan drew attention to this in his influential book The Making of Buddhist Modernism. While the concern for rationalism is not universal, it is a powerful current in modernity generally and can be felt in, for example, attempts to claim that Buddhism and science are consistent bodies of knowledge (they are emphatically inconsistent).

I don't know how much of the Pāli suttas would remain if we cut out all the supernatural stuff, all the superstition, myth, and magic. But I want to say about 10%. Something like that. A minority, a small minority, of early Buddhist texts are concerned with objectivity. Indeed, as I have quoted many times, Bodhi says:

“The world with which the Buddha’s teaching is principally concerned is ‘the world of experience,’ and even the objective world is of interest only to the extent that it serves as that necessary external condition for experience.” (Bodhi 2000: 394, n.182)

Buddhism emerged in a culture in which a cyclic afterlife was a given. While there were pockets of different approaches to this issue, the mainstream never doubted rebirth, and Buddhists portray non-believers as wrong-headed and foolish.

There is a complex dynamic that reinforces belief in an afterlife, which I have outlined in this way (see Attwood 2018b)

  • The certainty of the death of individuals creates cognitive dissonance in the self-aware living being.
  • Most of us find the duality of mind and body intuitive because over-active agent detection makes disembodied minds plausible.
  • According to testimony, certain experiences appear to demonstrate that identity and personality are not tied to the body, but can exist independently.
  • The idea that something might survive the death of the body and continue to live seems plausible.
  • Emotional weighting of facts (salience) creates a cognitive bias in favour of the afterlife.
  • Since the finality of death causes intense cognitive dissonance, post-mortem survival seems preferable.
  • We make the leap from probable/preferable to actually true and it feels satisfying because we have resolved the dissonance created by the fact of death and is consistent with our other beliefs.
  • Having made the leap, we fall victim to confirmation bias. We filter incoming information, accepting anything that reinforces our view and rejecting anything that challenges it. The view comes to seem more plausible over time.
  • Our community reinforces our belief, or even makes a profession of the belief a condition of group membership. Believing an afterlife is normative. Apostasy is punished with rejection from the group.
  • Over many generations, the afterlife view is reinforced until it seems to be the only possible view.

So, in thinking about early Buddhism, we have to keep in mind what rebirth is, how it is said to work, and how this belief shaped Buddhist soteriology. But we also have to keep an open mind. We are not compelled to adopt this Iron Age worldview or to treat it as any more grounded than other religious world-views from that time. We are allowed to apply critical thinking to everything, including (especially) religious beliefs.

How, in light of belief in this cyclic eschatology, did Buddhists view death?

Cessation and Death

The cessation of sensory experience is not something I can talk about from personal experience. I haven't undergone cessation. Still, there are many literary and anecdotal accounts of cessation and we can easily grasp the concept. We're all intelligent folk who understand that there is a distinction between our concept of a thing and the thing itself. So let's not get bogged down in anticipating mistakes that none of us make just because that's how Buddhists traditionally conduct arguments (to my enduring frustration).

It's no great stretch of the imagination to say that losing one's sense of self in meditation is a dramatic occurrence (at least the first couple of times). A common comment from people who have undergone this change is that it "feels like death". Interestingly, a recent study of meditation-induced near-death experiences that might shed light on this. William Van Gordon and his team studied meditation-induced near-death experiences (MI-NDE) in 12 intensive meditators. They used a standard definition of NDE and noted reports of them in non-life-threatening situations.

A near-death experience (NDE) is a reported memory of a pattern of experiences that can occur when a person is close to dying (e.g. life-threatening situations, asphyxia, near-drowning, stroke, etc.), when they believe they are close to dying (e.g. shock due to loss of blood) and in the period between clinical death and resuscitation (e.g. due to cardiac arrest) (Van Gordon et al. 2018).

NB: "clinical death" is medical a term indicating the cessation of breathing and blood circulation. But actual death only occurs with the absence of brain activity. There is a short period between the cessation of our heart beat and actual death. No one who is revived was actually dead. They were, in the inimitable words of Miracle Max, "only mostly dead". Although "clinical death" is a term used by medical people, clinical death is evidently not a form of death. Experiences associated with this state are called "near-death experiences", not "actual death experiences". If you are having experiences, even hallucinations, you are not dead. You cannot be dead because dead people don't have experiences. But you might decide, and most people do, that the experiences you had in that near-death state, gave you insights into death.

That Buddhists use meditation to prepare for death is common knowledge. Meditation on death is said to loosen one's attachment to the body, though meditating on death is not always a simple proposition. Compare the mass suicide reported in the Vesālī Sutta (SN 54.9) after the Buddha taught a group of monks to meditate on decomposing corpses. Attachment to being embodied is seen as a major driver of rebirth, and letting go of that attachment is a factor in being liberated from rebirth. But the idea goes further, as Van Gordon et al (2018) suggest:

Thus, in Tibetan Buddhism and to a lesser extent in other Buddhist traditions such as Theravada Buddhism, there exists the view that some advanced meditators can use meditation in order to gain insight into the state of consciousness that manifests after death.

It seems that some meditators are able to voluntarily and intentionally enter a state that substantially overlaps in phenomenology with near-death experiences. And that those who do these practices arrive in mental states that they interpret as death-like, while retaining awareness. Many meditators of my acquaintance solemnly assure me that they know their mind can operate independently of their body. I suspect that this is an artefact of such meditative states. It is worth once again citing Thomas Metzinger on the effects of hyperreal out-of-body experiences:

For anyone who actually had [an out-of-body experience] it is almost impossible not to become an ontological dualist afterwards. In all their realism, cognitive clarity and general coherence, these phenomenal experiences almost inevitably lead the experiencing subject to conclude that conscious experience can, as a matter of fact, take place independently of the brain and body. (p.78)

Meditation can induce hallucinations due to sensory deprivation. And they are likely due to the allostatic mode of the brain.

Van Gordon and company did not consider the impact of sensory deprivation in their study, though it is undoubtedly present and accounts for some of the phenomenology they observed. Indeed, Van Gordon et al have an almost emic approach, perhaps not surprising when the lead investigators are "experienced meditation teachers" as well as neuroscientists. They use the term "spiritual insight" quite a lot for example, but it's not clear what this means. They also use language like this:

A feasible explanation for these observations is that in order to reach advanced stages of meditative development, Buddhist meditators have to embrace the principle of “boundlessness” and seek to transcend relative concepts such as time and space (Van Gordon et al. 2016b).

How is this an explanation, let alone a feasible explanation? What does it mean to label space and time "relative concepts"? It sounds a lot like Madhyamaka metaphysics. Van Gordon et al appear to simply accept the insider accounts of mediation, perhaps because as "experienced meditation teachers" this is the personal orientation as well as the language used by their informants. No attempt is made to decode this Buddhist jargon. Which is fine in a religious text, but out of place in a notionally scientific publication. 

While I don't think the emic orientation of the article vitiates the point it seeks to make, we do have to be cautious of these insider-researchers because they are often setting out to prove something (in a religious epistemic mode) rather than observing what can be observed and adopting whatever the best explanation is.

No one who is primarily concerned with object accounts of the world can possibly take "spiritual insight" seriously as a category (much less when the conclusions take the form of incoherent metaphysical assertions). The problem here is that explanations that draw on meditative traditions are not objective. Such explanations tend to claim that no one can be objective, that there is no objective world, or that the objective world is not real. And none of this is science. Too much credence is given to emic categories and interpretations. Van Gordon et al acknowledge this only in passing under the heading Limitations: "...interpretations of the MI-NDE arising due to religious predispositions were not controlled for." No kidding. Seems like the religious predispositions of the researchers were not controlled for, either.

Despite some limitations, the study conducted by Van Gordon highlights the connection that Buddhists have intuited between meditation and death. The view seems to be that, in meditation, one can rehearse one's death because in death one is cut off from sensory experience. Cessation is like death in that sense. That cessation is often preceded, or accompanied, by the loss of a sense of self only reinforces this interpretation.

The trick is that, for Buddhists and other Indian religieux, death is not connected with the cessation of conscious states, rather is it axiomatic in this worldview that conscious states are not tied to the body and continue on in a disembodied form after death. This intuition about conscious states after death is one that occurs in children. They easily intuit that being dead a person does not need to eat, but they imagine that the dead person will get hungry (c.f. Emmons and Kelemen 2014). That is to say they intuit that their mind is distinct from their body and that this is true for others as well.

I don't think of this as a Buddhist view as such, since this is not how most Buddhist texts talk about it. A notable and influential exception is the Bardo Thödöl or Tibetan Book of the Dead, which is effectively describing reincarnation rather than rebirth. The Bardo Thödöl represents people as having personal continuity in the after-death state.

Note, I've started actively avoiding the term "consciousness" in this context because it is an abstraction. Abstractions are ideas about things, especially things that humans lump together in categories. Categories are also abstractions. This is not problematic, per se, except when we reify the abstraction as most people do with consciousness. It is apparent that almost all non-philosophers and not a few philosophers use the abstract term "consciousness" to refer to an entity; a thing. In this view, consciousness is metaphorically an object. This is why we can think of transferring consciousness from one body to another, whether by rebirth or through "uploading". This why we often talk about a person "having consciousness". In English consciousness can also metaphorically be a container for experience. I've discussed how this cognitive metaphor doesn't seem to appear in Pāli texts. It can be surprising how many people both assert that consciousness is not an entity and that they have experienced disembodied consciousness. In any case, another name for the kind of entity that people associate with "consciousness" is "soul". Europeans are often trying to sneak the soul in the back door of Buddhism.


Since the cycle of rebirths is the central issue of Buddhism, it should come as no surprise that death has long fascinated Buddhists. The contemplation of death, dying, and the decay of a human body are all part of the standard Buddhist collection of techniques. Often they are associated with loosening attachment to one's body, since this attachment to being embodied is seen as a crucial factor in being reborn.

But Buddhists seem to perceive a deeper connection between cessation and dwelling in the absence of sensory experience. The operative cognitive metaphor is: CESSATION IS DEATH. By undergoing cessation, and particularly with the cessation of one's sense of self, one undergoes a kind of death. That is to say, one undergoes an experience, and in seeking to categorise it, Buddhists classed cessation with death. Unlike death, however, cessation and absence do involve awareness. The death of cessation was a kind of death from which one returned with many ideas about what had just happened. The different stories about what cessation reflects, gave rise to a series of religions during the second urbanisation.

The little death of cessation is distinguished from dreamless sleep because one is aware. It is distinguished from dreaming by seeming to be hyperreal and to be hyper-significant. So-called "spiritual experiences" seem to be interpretations of various kinds of hallucinations experienced in the course of withdrawing attention from sensory experience, i.e. in sensory deprivation. Hallucinations arising from sensory deprivation can be seen in the light of the allostatic brain model. The brain expects input and puts a lot of effort into predicting what the next input will be, and then trying to minimise the prediction error through changing expectations or changing the input (through actions).

The combination of neuroscience insights from Metzinger, Sacks, and the others gives us not just an objective explanation of is actually happening during cessation and dwelling in the absence of sensory experience, they also help to explain why meditators start to assert either a dualist metaphysics or a nihilistic metaphysics (Madhyamaka) as a result. These interpretations are what strike people as plausible, not because of the experience per se (since it often has minimal content), but because of the cultural and personal context in which the meditator exists.

If we are lucky enough to have these kinds of rare and unusual experiences, then we inevitably give them the spin that we have been taught to give them. The classic example for me is Gary Weber (especially this interview). Weber describes his awakening in quite sincere and plausible terms. I have no doubt that his experience of the world is now quite different to mine. But Weber sees his experiences as confirming Advaita Vedanta doctrines from medieval India. His interpretation of awakening is very different to Buddhist accounts; though he invokes "emptiness" he doesn't do so in a plausible way (I don't think he understands mainstream Buddhist discourses around emptiness, let alone the Prajñāpāramitā take on it). The same experience can be interpreted in many different ways.

What I have tried to show in these three essays is that sensory deprivation is something that needs to be taken into account when we explain the process and outcomes of meditation. I've tried to show that, though it is not explicitly talked about, there is an awareness of the phenomenology of sensory deprivation in Buddhist texts. This is why Buddhist śīla takes the form it does, rather than the form of ethical treatises, i.e. śīla is primarily preparation for sensory deprivation in meditation and helps smooth the way beyond the destabilising effects of the brain trying to fill in the gaps in predicted experience. Success in meditation is predicated on the brain predicting that nothing is going to happen next. While it predicts something, there is always an unacceptable gap between prediction and present experience.

I have not attempted to be comprehensive. I've consulted a fairly small sample of the literature. Those who do research on meditation would no doubt see this in a different light. We can only hope that the experts do begin incorporate sensory deprivation into their accounts of mediation (though these essays are unlikely to have much impact on that sphere). Whether religieux will do the same remains to be seen. Buddhists are generally averse to deflationary accounts (i.e. accounts which deflate Buddhist exceptionalism).



Attwood, Jayarava (2018a). "Defining Vedanā: Through the Looking Glass." Contemporary Buddhism, 18(3): 31-46. https://doi.org/10.1080/14639947.2018.1450959

——. 2018b. Karma and Rebirth Reconsidered: An Inquiry into the Buddhist Myths of a Just World and an Afterlife. Visible Mantra Press.

Cohen, S. (2019) The Upaniṣads: A Complete Guide. Routledge.

Emmons, N. A. and Kelemen, D. (2014) "The Development of Children’s Prelife Reasoning: Evidence From Two Cultures." Child Development. 2014 Jul-Aug;85(4):1617-33. doi: 10.1111/cdev.12220. Epub 2014 Jan 16.

Gananath Obeyesekere (2002) Imagining Karma: Ethical Transformation in Amerindian, Buddhist, and Greek Rebirth. University of California Press.

Kamalashila. (1994). Meditation: The Buddhist Way of Tranquillity and Insight. Windhorse Publications.

McMahan, David. (2008) The Making of Buddhist Modernism. Oxford University Press.

Sacks, Oliver. (2012). Hallucinations. Picador.

Van Gordon, W., Shonin, E., Dunn, T.J. et al. (2018) "Meditation-Induced Near-Death Experiences: a 3-Year Longitudinal Study." Mindfulness 9: 1794–1806. https://doi.org/10.1007/s12671-018-0922-3

03 June 2022

Buddhist "Ethics" and Sensory Deprivation

In my previous essay, I argued for a link between meditation and sensory deprivation: more specifically that meditation, viewed as withdrawal of attention from the sensorium, causes (non-pathological) hallucinations in the same way that sensory deprivation does. In this essay, I want to broaden the scope to look at śīla "conduct" in the light of this link and its place in the threefold path: śīla, samādhi, and prajñā. Along the way I introduce the Spiral Path (which ought to be familiar to my readers) and Karl Friston's free energy principle. 

It is common to translate śīla as "ethics", even though we have known for some time that Buddhism has no ethics. That is to say, traditional Buddhism produced no treatise on ethics. Here, I am making the philosophical distinction between morality as rules of conduct (pañcaśīla, praṭimokṣa) and ethics as the principles or meta-rules that guide the formation of rules. Alternatively we could think of ethics as the reasons for following moral rules. The lack of ethics in Buddhism has been a problem for those of us who wish to adapt Buddhism to modern life. On one hand, we have no guidance for forming new rules and, on the other, we have yet to write our own ethical treatises. Europeans, much like their Buddhist counterparts in Asia, have filled in the lacuna by drawing upon their own culture. European Buddhists tend to be liberals (of various types including, more recently, neoliberals) and as David Chapman discovered in relation to US "Consensus Buddhism" we tend to see liberal values as Buddhist values, because after all these are our values and we are Buddhists. It's a shame David McMahan did not include liberalism amongst his important factors of modernity, in Buddhist Modernism

Here I want to revisit some research I did before getting sucked into the Heart Sutra vortex.

Ethics and the Spiral Path

In Triratna we emphasise what Sangharakshita called the Spiral Path, for which we tend to use his book The Three Jewels (1967) as the principal source. I published a scholarly essay on this formulation of the Buddhist path (Attwood 2013). I also created an infographic (right) which describes the main features and variants of these texts.

Sangharakshita based his exposition on the Upanisā Sutta (SN 12:23) which is something of an oddity. Ayya Khema (1991) and Bodhi (1980) also composed commentaries on the Upanisā Sutta, the latter directly in response to Sangharakshita (who used Sanskrit technical terms rather than Pāli). However, in my view, the locus classicus for this doctrinal formulation should be considered the first five suttas of the Dasakanipātapāḷi and the Ekādasakanipātapāḷi of the Aṅguttara Nikāya (i.e. the Tenfold Chapter and Elevenfold Chapter)*.

* The reason we cite both is that, while otherwise the texts in the different chapters are identical, in the Tenfold Chapter, nibbidā and virāga are treated as one step, while in the Elevenfold Chapter. they are treated as two.

The Spiral Path describes a form of conditionality that has a different emphasis from the one that most Buddhists are focussed on. The presence of one condition naturally (effortlessly) gives rise to the effect, which then becomes the condition for the next effect, in a cumulative manner. In some cases it is accompanied by the image of water overflowing from one container to the next like a champagne fountain. 

This doctrine is less systematic than its counterpart the twelve nidānas and, at least in Pāli, shows little standardisation, especially in the early stages. The Chinese counterparts of these texts, collected in the Chinese Madhyamāgama translation (MĀ 42-55), have been homogenised to a greater extent, though some variations remain.

When I looked at all of the forty or so Pāli suttas that have some version of the Spiral Path, I could see a familiar pattern. I was able to show that the Spiral Path is in fact an elaboration of the threefold path: śīla, samādhi, and prajñā, which we routinely see translated as "ethics, meditation, and wisdom" One of the salient features of the threefold path in these suttas is the two liminal states that link the three phases. So in śīla we see a range of techniques that culminate in pāmojja "joy" (also pāmujja; Skt prāmodya) which is a result of śīla rather and an example of śīla. In her exposition of the Spiral Path, Ayya Khema (1991) refers to pāmojja as a necessary prerequisite for meditation. Pāmojja is the culmination of practising sīla and the doorway to samādhi. Similarly, the stages of meditation in the spiral path texts—pīti, passadhi, sukha, and samādhi; which seem to approximate the stages of jhāna—lead to another liminal state: yathabhūtañāṇadassaṇa "knowledge and vision of things as they are". Knowledge and vision opens the door to the prajñā stage of the threefold path, i.e. one becomes fed up with sensory experience (nibbidā) and rejects it (virāga), resulting in either liberation (vimokkha) and the knowledge of liberation (vimokkhañāṇa), or the cutting off of karma-making (āsavakkhaya) and the knowledge of this (āsavakkhayañāṇa).

Defining Śīla.

One approach to defining śīla would be to list all the practices that lead up to pamojjā in the Spiral Path literature and look for patterns. From my 2013 article here is such a list:

  • saṃvara/saṃvuta "restraint"
  • indriyesu guttadvāra "guarding the sense doors"
  • yoniso-manasikāra "wise attention"
  • appamattassa vihārato "dwelling vigilantly"
  • sati sampajañña "mindfulness and attentiveness"
  • hiri-otappa "shame & scruple" (AN 8.81)
  • sīla "behaviour"
  • kusalāni sīlāni "virtuous behaviour"
  • saddhā "faith".

Of these, the only practices with an explicit moral character are shame (hiri) and scruple (ottapa). Like mindfulness (sati) and full attention (sampajañña), these two terms often occur together and are often poorly distinguished in the literature. They are said to reflect one's active recognition of an unskilful act, and the fear of being judged unskilful by the discriminating (viññū) members of the Buddhist community. And keep in mind that these are not passive feelings about actions, these are practices or, in other words, these are actions that one consciously and voluntarily engages in for the purposes of achieving the goals of Buddhism, including the ultimate goal of ending rebirth. 

The odd one out is "faith" (saddhā). Despite a great deal being written about faith in Buddhism, I have yet to see a modern English account that accurately reflects saddhā as I encounter it in Pāli. In the suttas, saddhā is that feeling when one has listened to the Buddha talk about his teaching, it makes you want to try what he's talking about. What saddhā means, then, is less about faith (unreasoning and/or unreasoned belief) than it is about enthusiasm combined with intention. Enthusiasm for Buddhist practices based on understanding an explanation by an expert, and an intention to try it out. This is not an unreasoning belief as faith in Christianity is supposed to be. Nor is this confidence in the teachings based on experience. The quality of having confidence in the practices as described is called aveccapasāda "perfect clarity". Saddhā is "faith" in the reasoned sense that having listened to a presentation on Dharma, one finds it plausible enough to want to go ahead and try it. Unreasoned faith comes into Buddhism only with the advent of so-called "Pure Land Buddhism", with the focus of the early centuries being initially on Akṣobhya and his realm of Abhirati, before moving to Amitābha and Sukhāvati. Neither is part of our universe but both are able to appear in our universe to succor those who have sufficient faith, with a gradually lowering bar on "sufficient". 

The bulk of these ideas, by contrast, are not obviously moral in character and, indeed, they involve minimising, to the point of elimination, interacting with other people rather than improving the quality of those interactions. More precisely the practices listed as comprising śīla appear to aim to limit our relationship with sensory experience. Here, for example, we are counselled to reduce input, to stop ourselves from seeking out sensory experience, to guard the sense doors (against the intrusion of sensory experience), to attend to the sensorium wisely, i.e. to not be intoxicated (appamāda) by experience but to remain indifferent (virāga) to it. These are not instructions on how to treat other people well (morality), nor are they principles for treating people well (ethics). They mostly involve avoiding and ignoring other people as much as possible. The idea for Buddhists being the solitary retreat. 

From the point of view of modern Buddhists, śīla is about our moral conduct. We link śīla not to these practices of restraint, but to the precepts (sikkhapāda) or monastic rules (vinaya, paṭimokkha) or some other formulation. The Spiral Path texts, by contrast, don't mention precepts or rules. So there is a disconnect between the received tradition of Buddhist rules of conduct and this formulation of the path of Buddhism. Treating people well is not really considered in this context. Rather one is largely concerned with maintaining social isolation and having a sober relationship with sensory experience; concerned with reducing sensory input. It seems to me that śīla as a moral teaching came along after the fact as a sop for people who were not spending their days in samādhi.

If śīla is not principally about moral conduct, if Buddhism moral rules are an afterthought, then what is śīla concerned with? Here I foreground the idea of restraining the senses and try to relate it to the idea that sensory deprivation is an important component of Buddhist meditation. 

Śīla as Preparation for Meditation.

The threefold way formulation of Buddhism includes the idea that śīla is preparatory. Specifically, we learn that śīla prepares us for the samādhi "meditation" phase. Moreover, the Spiral Path texts suggest that the significant result of our preparation is pāmojja "joy". In this view, śīla leaves us with an untroubled conscience which then facilitates our entry into jhāna and other altered states". And this is not wrong. It is explicitly included in the Spiral Path texts in AN 11.1, where the benefit of wholesome conduct (kusalāni sīlāni) is non-regret (avippaṭisāro) which in turn is the condition for pāmojja, which, as we have seen, is the doorway to meditation. It's not hard to interpret this as relating to morality. "Non-regret" appears to tie in with seemingly moral practices such as shame (hiri) and scruple (ottapa). In this context, "wholesome conduct" seems to suggest the practices of restraint that we find in the other Spiral Path texts.

In most cases, the idea that good behaviour contributes to liberation is largely concerned with obtaining the right kind of fortunate rebirth. It needs emphasising that, for Buddhists, good behaviour alone provides no escape from saṃsāra. Good actions (kusalāni sīlāni) result in a good rebirth destination (suggati), but the acme of Buddhism is the cessation of rebirth. This can only be accomplished once we stop making new karma. And we only stop making karma when we see things as they are.

What if the point of śīla in this context was not morality? What if these behavioural norms were actually intended to help us cope with the intense sensory deprivation of samādhi meditation?

Experience shows that difficulty applying meditation techniques is often correlated to how busy one's life is. The Buddha almost always recommends that a meditator should detach themselves from society, to find a deserted spot in the wilderness, far away from the intense sensory stimulation of town, village, or monastery. Buddhist monastics are supposed to definitively severe their social connections and live adrift from society. As the Buddha says in a memorable Vinaya passage,

"Monks," said the Bhagavan, "you have no mother and no father to care for you. If you don't care for each other, then who will care for you? If you would care for me, then tend to the sick." (Vin I 301).

The ideal Buddhist meditator is solitary, isolated, and disenchanted with sensory experience.

As noted in the previous essay, it is very common for meditators to meet hallucinations when they practice, almost ubiquitous. Even complete beginners can meet with hallucinations the first time they try it (I've known people for whom this was true). For the most part, these gross hallucinations are a barrier to deeper concentration. We experience them as a kind of turbulence or disturbance that distracts us from the focus of the meditation. At least in traditional Buddhist ideas about mind, hallucinations are a form for sensory experience in which the object is mental rather than physical. Buddhism treats the mind as a "sense" in this worldview. So internally generated experiences, with no objective counterpart, are not seen as special, they are just experiences, just the kind of thing that stops with cessation. And the goal of Buddhist meditation is to make experience cease (I'm going to get into this in a subsequent essay).

Perhaps it is worth saying that Pāli clearly does allow for an objective world that follows different rules to sensory experience. There is no talking of breaking down the subject/object distinction. There is no talk of non-dualism. And to my mind, had non-dualism been part of early Buddhism, it would have been mentioned. Not being dualists, I doubt such a doctrine would have appealed to the authors of the Pāli suttas.

The Allostatic Brain

If we take śīla in the sense that I have been suggesting, then we can begin to see how it helps. It comes back to the brain and allostasis. If our brain were simply reacting to sensory input, it would never be able to keep up, even with the relatively coarse-grained representation that reaches awareness in our first person perspective. It is true that life requires that various parameters of our body are kept within limits conducive to life. This is called homeostasis, which effectively means "keeping things the same". Homeostasis is achieved using feedback loops. The basic process is represented by the simplest version of feedback, the kind of mechanical switches we see in thermostats. Heat causes a bar composed of two metals to bend as they expand at different rates. The bending breaks the contact, switching the heating off. The room cools and the metal bar straightens out, and eventually makes contact and switches on the heating.

Our body has various, far more complex and interrelated, feedback loops that help to keep things like the composition of our blood in the optimal range. But the feedback process is reactive and thus can only change things after the fact. And reactive processes can't account for what the brain does. Rather, neuroscientists have shown that the brain anticipates changes and sometimes takes preemptive actions. We could go so far as to say that what the brain does is predict future inputs and try to minimise any discrepancies between what it predicted and what is happening. Moreover, this can be seen as a specific application of a principle, first enunciated by Karl Friston and called the Free Energy Principle. Friston has given this idea its own mathematical formalism, but also showed that the resulting formulas recreate results from statistical mechanics and information theory. Thus the FEP appears to unify a number of seemingly disparate fields. I suspect that eventually this will tie into work by David Deutsch and Chiara Marletto on constructor theory, but that is another essay entirely. 

There are two ways the brain can minimise prediction errors (or in Friston's terms to minimise the free energy of the system): it can alter its prediction, or it can alter the input (which is usually accomplished by intervening in the world through some action).

It's worth emphasising that this is not a process open to introspection. And therefore, it is not intuitive. Moreover, when we talk about the brain "predicting inputs" we are not talking about the high-level, coarse-grained experiences that we are aware of. The only "inputs" to the brain are 1) electrochemical signals from peripheral nerves; and 2) electrochemical signals from within the central nervous system. With respect to 2), Lisa Feldman-Barrett notes that 90% of the incoming connections to the visual cortex are from other parts of the brain, rather than from the eyes. There are millions of such inputs. So what the brain is predicting is patterns of electrochemical signals, i.e. variations in signal strength and frequency across millions of individual inputs. Similarly the only output the brain has is firing neurons that connect outwards to the body, sending identical electrochemical signals that cause the body to move in various ways.

Let's use an example from Feldman-Barrett's account: standing up. Moving from sitting to standing involves a lot of coordinated muscle activity, all of which is stimulated at a fine-grained level by brain activity. Our coarse-grained first person perspective on this is very different in scope and detail. One of the problems we face is that suddenly standing up causes a drop in blood pressure in the head, which means insufficient blood reaches the brain and the brain does not work at optimum. In other words, standing up threatens to disrupt homeostasis.

Before we consciously decide the stand up, the brain is balancing the odds. If it predicts that we are going to stand up, then it initiates actions that at a coarse-grained level amount to raising our blood pressure (and neither the prediction nor the preemptive action are available to introspection). At best we become aware of this process when it fails, i.e. we stand up too quickly and get dizzy due to transient low blood pressure in the head. Of course, how we get from the fine-grained view of nerve cells storing energy and releasing it in pulses, to coarse-grained first person perspective, is still unclear. But there is no other viable explanation.


In my last essay, I floated the idea that sensory deprivation is an issue for meditators. In this view, one of the consequences of doing mediation techniques can be, often is, hallucinations. In this view, hallucinations are internally generated distractions and while they may bother meditators at first, they are generally transient. My too-small sample of meditation instruction books suggested that as we become more profoundly cut off from sensory experience, as we may experience more subtle hallucinations that are more tempting to assign meaning or specialness.

Even with a tenuous grasp of Friston's free energy principle (which mine admittedly is)  we can see hallucinations due to sensory deprivation in the context of what the brain expects to see. If one lives a hedonistic lifestyle, one's brain comes to expect high levels of sensory stimulation. One outcome of this is insensitivity to more subtle stimulus. 

It seems to me that we can now state a hypothesis: dropping from high levels of stimulation to very low levels of stimulation is likely to produce a much stronger response that moving from low levels of stimulation to very low levels. Trying to meditate after a busy day at work, is likely to throw up a lot of internal stimulation (I'm going to avoid calling this self-stimulation). 

At least some Buddhists have been concerned to attain what neuroscientists are various calling "contentless experience", "contentless awareness" or "minimal phenomenal consciousness". If our aim is to bring sensory experience to a halt, then reducing the gap between the starting level of stimulation and the aimed-at level would make sense. That is to say, if we aim to achieve contentless awareness, then starting from a much reduced level of stimulation would be advantageous. 

My meditation teachers always emphasised preparation. It's no use going from indulging in the senses to trying to cut oneself off from them. It just creates misery and doubt. Rather one must actively reduce input, reduce stimulation. 

I often find myself explaining that for Buddhists, good behaviour or good karma has little soteriological value. No one is saved by good works. Rather the key to Buddhist soteriology is ending karma-driven rebirth; this requires ending karma. Yes, this does require being reborn as a human and having access to instruction in Buddhist techniques, which in Buddhist soteriology we achieve through good karma, That this is as far as good karma can get us, the mere opportunity to escape. The acme of Buddhism is the end of rebirth. Something similar applies to sensory experience. There is no mileage, from a Buddhist perspective, in indulging the senses. Indeed, since karma is intention (cetanā) in early Buddhist thought, indulging in sensory perception is counterproductive since it gives rise to greed and hatred. 

If śīla is seen as preparation for Buddhist meditation, then it would make sense if it were aimed at reducing sensory stimulation. This is, to some extent at least, intuitive. If Karl Friston is right, and he does seem to be, then the free energy principle gives us a deep explanation for this. It's not simply a useful heuristic, though it is that. Rather it reflects something built into the human mind. We can cope with low levels of stimulation, we can avoid distracting hallucinations, if and when our brain comes to expect low levels of stimulation. And the brain can be habituated by, for example, long intervals of not talking for example, or periods of doing nothing (which is much harder than it sounds).

Thinking about traditional Buddhism in these ultra-modern ways, incorporating cutting edge science, is not straightforward. One has to be aware of projections. I'm not suggesting that Buddhists pre-empted or even prefigured the free energy principle and its detailed mathematical formalism. Buddhism is not scientific and certainly not proto-scientific. There is a kind of systematic approach to Buddhism, but it is all presented in a thoroughly religious context. The central problems of Buddhism are much the same as in all religion, i.e. the afterlife and the problem of evil (i.e. it is based on interpreting a metaphysical speculation as though it accurately reflected reality). 

Rather, what I am trying to show here is that modern perspectives, especially scientific perspectives, often give us better explanations than Iron Age or Medieval religious texts. Which ought to come as no surprise. My sense is that Buddhists, for example, are highly critical of Christian fundamentalism, but their arguments against Christianity often amount to Buddhist fundamentalism, i.e. your religious book is not right, because our religious books say so. I don't think looking to the Iron Age for solutions to living in the twenty-first century makes much sense. Since I have taken the time to learn Pāli and read those Iron Age texts in that language (and compared them with Sanskrit and Chinese counterparts) I find the anachronisms in Buddhism even harder to take seriously. Our texts are no more an accurate reflection of reality than any other religious text. 

In the final analysis, if samādhi is your goal, then you will need to have some understanding of, and strategies for dealing with, sensory deprivation. Traditionally this amounts to reducing sensory input across the board. And this process of reducing sensory input preparatory to meditation is called śīla.

Anyone interested to follow up the connection between meditation and the free energy principle could try reading the recent paper by Ruben Laukkonen and Heleen Slagter (2021) which uses Friston's model to propose a new way of understanding what meditation does. It's not easy and I don't fully understand it myself, but it seems very promising as an addition to how we think about what we wish to achieve as Buddhists. 



Attwood, J. (2013). "The Spiral Path or Lokuttara Paṭiccasamuppāda." Western Buddhist Review 6, 1–34. http://www.jayarava.org/texts/the-spiral-path.pdf

Bodhi. (1980). Transcendental Dependent Arising: a Translation and Exposition of the Upanisa Sutta. (The Wheel Publication no.277/278.) Buddhist Publication Society. Online: http://www.accesstoinsight.org/lib/authors/bodhi/wheel277.html

Khema (1991). When the Iron Eagle Flies. Penguin.

Laukkonen, Ruben E. and Slagter, Heleen A. (2021) “From many to (n)one: Meditation and the plasticity of the predictive mind.” Neuroscience & Biobehavioral Reviews, 128, 199-217.

Sangharakshita. (1967). The Three Jewels: An Introduction to Buddhism. Rider.

Woods, T.J., Windt, J.M. & Carter, O. (2022). "The path to contentless experience in meditation: An evidence synthesis based on expert texts." Phenomenology and the Cognitive Sciences.  (Open access version unpaginated) https://doi.org/10.1007/s11097-022-09812-y

Related Posts with Thumbnails