Perception: A Buddhist Perspective on How You Know Anything At All

What Does It Mean to “Know” Something?

A Buddhist perspective on perception starts, oddly enough, with a Western theologian’s mistake. Rudolf Otto spent a career trying to name a particular kind of experience — the sense of standing in front of something vast, uncanny, not quite explicable, not quite bearable. He called it the numinous. And he called its object ganz Andere — wholly other. Something genuinely outside the self, making contact with the self from beyond its border.

Otto wasn’t wrong that something real happened to him. He was wrong about the address he mailed it to. “Outside the self” assumes there’s a wall there in the first place — a border the holy has to cross to reach you. Buddhism’s answer isn’t that Otto imagined the encounter. It’s that he mistook a reflection for a stranger.

Nine words, and it’s the whole correction. What Otto met wasn’t wholly other. It was his own nature, arriving at close enough range that it finally registered as something — and because nothing in his framework allowed for “this vast thing is also me,” he filed it under other by default. Not because it was other. Because that was the only category his epistemology had room for.

This is why the mistake was possible at all, and why it was still, in its way, an honest one. What Otto touched wasn’t new. It wasn’t visiting. It has never once arrived, because it was never once absent — present since beginningless time, prior to the self that supposedly stood apart from it, prior to the very category of “self” that made ganz Andere sound like a coherent thing to say.

So: not a mysterious contact between self and Other. A recognition, badly labeled.

Otto never renounced the wholly other. But in 1926, in Mysticism East and West, he sat Meister Eckhart and Śaṅkara side by side and did something his 1917 self couldn’t have allowed: he let them speak in each other’s words. Eckhart’s soul and God — “not similarity but identity.” Śaṅkara’s tat tvam asi — that art thou. Otto laid the two traditions out in parallel columns because, by his own account, the convergence went past resemblance into identity of language itself.

His alpha was separation, published and famous. His omega, nine years later, was two of history’s most exacting mystics telling him — in Latin and Sanskrit, independently, without ever having read each other — that there had never been a wall there to begin with. He documented the correction. He just never went back and rewrote the definition it corrected.

A similar collision happens closer to home in Western philosophy: see → → Meditating with Descartes — The Cogito, where the same search for unshakeable certainty runs into the same wall from a different direction.

That’s the gap this page starts inside of — not Otto’s error alone, but Otto’s own unfinished recognition of it. If knowing isn’t contact across a border, what is it? Mechanically, gate by gate, how does anything become known to mind at all? Buddhism has an answer that predates Otto by two thousand years, and it doesn’t start with mysticism. It starts with an inventory. That inventory — not mysticism, not metaphor — is what a Buddhist perspective on perception actually offers, starting with the twelve gates themselves.

The Twelve Āyatanas: Six Faculties, Six Objects

Before Cittāmātra ever showed up to argue that the external object doesn’t exist, Abhidharma had already done the paperwork on what mind has to work with regardless of whether that object is real. The system is called the twelve āyatanas (Tib. སྗྱེ་མཆོད་, kyeché, Wyl. skye mched) — literally “sources” or “gateways,” the doors through which anything becomes known at all. Six on the inside, six on the outside, paired:

• •Eye and form
• •Ear and sound
• •Nose and smell
• •Tongue and taste
• •Body and touch
• •Mind and mental phenomena

That’s the entire catalogue. Every perception, thought, memory, hallucination, and flash of insight in meditation you have ever had arrived through one of these twelve doors — no others available. Notice the sixth pair: mind is listed as a sense faculty on the same footing as the eye, and mental phenomena — thoughts, concepts, memories — are listed as its object, on the same footing as sound or smell. Abhidharma doesn’t grant thinking a privileged, non-sensory status. It’s a gate like the rest, with its own class of traffic passing through it.

This isn’t Cittāmātra’s system — it’s older, common ground across Abhidharma schools generally, Mahāyāna and non-Mahāyāna alike. Which makes it the right place to start counting: before asking whether the object behind the gate is real (Vaibhāṣika), inferred (Sautrāntika), or nonexistent (Cittāmātra), you have to know how many gates there are and what kind of traffic each one lets through.

The Eighteen Dhātus: Where Consciousness Enters the Count

Add one more layer and the twelve become eighteen. For each of the six sense pairs, there is also a resulting consciousness — eye-consciousness arising dependent on eye and form, ear-consciousness dependent on ear and sound, and so on through mind-consciousness, dependent on mind and mental phenomena. Six faculties, six objects, six consciousnesses: eighteen dhātus (Tib. ཁམས་, kham, Wyl. khams) — “elements” or “constituents,” the full accounting of what a moment of experience is actually built from.

This is where the two pages of this series physically join hands. The six consciousnesses named here are the first six of the eight named in → → Eight Consciousnesses, Five Skandhas: Buddhism’s Theory of Mind — eye-consciousness through mind-consciousness, counted exactly as in the āyatana/dhātu system, with manas (the afflicted mental consciousness) and ālayavijñāna (the storehouse) added as the seventh and eighth once Yogācāra extends the model. The eight consciousnesses were never a separate invention. They’re the eighteen dhātus, pushed one layer deeper.

Eighteen elements. No self mentioned anywhere in the list — not because the self was forgotten, but because Abhidharma is building an exhaustive inventory of what experience is actually made of, and nothing on the list requires one.

The Reducing Valve: Twelve Gates, Not Everything at Once

Twelve gates, not an infinite number. Worth asking why the architecture is limited at all — and it turns out a mid-century English novelist got there independently, working from a completely different tradition.

Huxley borrowed the underlying theory from Henri Bergson and C.D. Broad: the brain and nervous system don’t produce consciousness, they filter it, protecting the organism from being flooded with more reality than survival requires. Twelve gates, tightly rationed, built for utility rather than completeness. He names the mechanism from Bergson — but the vocabulary he reaches for once the valve opens, “sacramental vision,” overwhelming presence, is borrowed just as directly from Otto, whom Huxley cites by name in this same book.

This is the two-veils doctrine — kleśāvaraṇa (Tib. ྉོན་སྡྲིབ་, nyöndrib, Wyl. nyon sgrib), the obscuration of disturbing emotion, and jñeyāvaraṇa (Tib. ཤེས་སྡྲིབ་, shédrib, Wyl. shes sgrib), the obscuration of what can be known. Where Huxley names one mechanism and one metaphor, the Uttaratantra names two, distinguishes their function, and specifies what clears each — ethical discipline and meditative stability address the first, direct realization of emptiness addresses the second.

The reducing valve, in other words, already has a name on this site’s own doorstep — The Mechanism, sitting in the Kyerim/Dzogrim material without anyone drawing the line to Huxley until now, describing this exact architecture just without the Sanskrit.

A physical-level version of the same filtering claim shows up in → → Orch-OR: What Quantum Physics and Buddhism Agree On — a parallel, not an equivalence, same discipline as the rest of this site.Pramāṇa: Is Any of This Actually Valid?

Twelve gates, eighteen dhātus, a two-part filter narrowing what gets through — all of that explains how a cognition arises. None of it says whether that cognition is trustworthy. Buddhism built an entire discipline around it, centuries before anyone thought to ask a courtroom the same question: pramāṇa (Tib. ཚས་མ་, tsema, Wyl. tshad ma), “valid cognition,” founded by Dignāga (Tib. ཕྨོགས་ཀྱི་གལང་པོ་, Chokyi Langpo, Wyl. phyogs kyi glang po) and systematized by his successor Dharmakīrti (Tib. ཆོས་ཀྱི་གྤགས་པ་, Chökyi Drakpa, Wyl. chos kyi grags pa).

Their answer: only two means of knowledge count as valid. Direct perception — pratyakṣa (Tib. མངོན་སུམ་, ngönsḡum, Wyl. mngon sum) — non-conceptual, immediate contact through one of the twelve gates. And inference — anumāna (Tib. རྗེས་དཔག་, jedpak, Wyl. rjes dpag) — reasoning from evidence to a conclusion not directly seen, the classic example being smoke on a distant hill implying a fire you can’t see. Everything else — hearsay, assumption, a feeling that seems obviously true from the inside — doesn’t qualify, no matter how confident it feels.

That last clause is the trap, and it has a modern name: Michael Gazzaniga’s split-brain “interpreter.” In patients whose corpus callosum had been severed, Gazzaniga could feed information to one hemisphere while keeping the other — the one that does the talking — completely in the dark.

In his best-known version, the left hemisphere sees a chicken claw; the right hemisphere, cut off from language, sees a snow scene. Both hands correctly point to matching images — the right hand to a chicken, the left hand to a shovel. Ask the patient why the left hand chose a shovel, and the talking hemisphere — which never saw the snow, has zero information about it — doesn’t hesitate and doesn’t say “I don’t know.” It invents a reason on the spot: the shovel is to clean out the chicken shed. Confident. Coherent. Completely fabricated, and the patient has no sense whatsoever that anything was made up.

Gazzaniga’s point wasn’t really about split brains. It’s that this interpreter module runs constantly, in everyone, confabulating plausible reasons for feelings, choices, and actions it frequently has no genuine access to — and it never flags the difference between a reason it actually has and one it’s inventing to fill the gap. Which means most people’s account of their own motives wouldn’t survive cross-examination — not because they’re lying. Because the part of the brain answering was never in the room when the decision got made.

This isn’t a claim about whether your actions are free. It’s narrower than that, and sharper for pramāṇa’s purposes: the felt sense of “I know why I did that” fails Dignāga’s own test. It isn’t direct perception — you don’t perceive your own motives the way you perceive a shape. It usually isn’t valid inference either — there’s no visible chain of evidence, just a story that arrives fully formed and certain. By the oldest standard Buddhist epistemology has, most of what passes for self-knowledge doesn’t clear the bar.

The same question, aimed at a very different kind of mind: → → Could AI Ever Wake Up? Buddhism vs Artificial Intelligence — if an AI’s outputs can’t clear pramāṇa’s bar either, what does that say about the bar, or about the AI?

Where the Gates Lead: The Four Kāyas

Twelve gates. Two veils. A narrator making things up as it goes. That’s the whole mess this page has built — and here’s where it stops being a mess.

Tibetan doctrine maps this in four stages, the four kāyas (“bodies”), each one naming a different degree of obscuration rather than a different place or being:

Nirmāṇakāya (Tib. སྤྥུལ་སྐུ་, tulku, Wyl. sprul sku) — the emanation body. Ordinary perception, unpurified, still running on the twelve gates exactly as described in H2 2.

Sambhogakāya (Tib. ལོངས་སྐུ་, longku, Wyl. longs sku) — the enjoyment body. Visible only once the grosser obscuration — kleśāvaraṇa, from H2 4 — has substantially cleared.

Dharmakāya (Tib. ཆོས་སྐུ་, chöku, Wyl. chos sku) — no gate left to describe. The collapse of the entire perceiver/perceived structure this page has been mapping — nothing left standing outside anything else to perceive it as other.

Svābhāvikakāya (Tib. ངོ་བོ་ྉིད་ཀྱི་སྐུ་, ngowo nyi kyi ku, Wyl. ngo bo nyid kyi sku) — not a fourth thing added to the other three. The inseparability of all three, stated as its own category.

Read this way, the four kāyas are the direct continuation of everything this page has built toward — not a cosmology bolted onto the epistemology, but its natural endpoint. The twelve gates never actually close. What changes, stage by stage, is how much of the two veils still stands between the gate and what’s on the other side of it — until, at dharmakāya, “the other side” stops meaning anything, because there was never a wall there to begin with.

That’s not a new claim on this site. It’s the same one made about rigpa on “Consciousness: A Buddhist Perspective” — dharmakāya and rigpa naming the identical non-dual ground, one vocabulary built from epistemology and perception, the other from Mahāmudrā and the nature of mind. Two pages, two different doors. Same room on the other side of both of them.

These four stages are also where practice picks up where philosophy leaves off: → → Six Yogas of Naropa — Master Page maps the actual completion-stage practices aimed at sambhogakāya and dharmakāya directly.

Frequently Asked Questions

What is a Buddhist perspective on perception, exactly?

Not a claim that perception is illusory, but an inventory of its actual mechanism — twelve gates, eighteen constituent elements, a two-part filter, and a discipline for testing whether any given cognition is even valid. A Buddhist perspective on perception starts with plumbing, not mysticism.

What are the twelve āyatanas?

The six sense faculties (eye, ear, nose, tongue, body, mind) paired with their six objects (form, sound, smell, taste, touch, mental phenomena) — the full inventory of what knowing has ever had to work with, per H2 2.

What is pramāṇa in Buddhist epistemology?

The discipline of valid cognition founded by Dignāga and Dharmakīrti, recognizing only two valid means of knowledge: direct perception and inference. Most of what passes for self-knowledge, per H2 5, doesn’t actually qualify.

What’s the difference between the twelve āyatanas and the eighteen dhātus?

The eighteen dhātus are the twelve āyatanas plus the six resulting consciousnesses — same inventory, one layer deeper, and the exact structural link to the eight-consciousness system on the companion page.

How does the reducing valve relate to Buddhist doctrine?

Huxley’s reducing valve names one mechanism; the Uttaratantra’s two-veils doctrine names two, and specifies what clears each. Same claim, Huxley four decades ago, Buddhist doctrine roughly fifteen centuries earlier.

Further Reading:

For more from this Buddhist perspective on perception, and the pages it stands on:

→ Buddhist Perspective on Consciousness — dharmakāya and rigpa naming the identical non-dual ground. Parent Page

→ The Eight Consciousnesses, Five Skandhas: Buddhism’s Theory of Mind

What the Hell is Awareness? →

→ Meditating with Descartes — The Cogito

→ Could AI Ever Wake Up? Buddhism vs Artificial Intelligence

→ Orch-OR: What Quantum Physics and Buddhism Agree On

→ Six Yogas of Naropa — Master Page

Outbound:

→ Dharmakīrti and the Roots of Pramāṇa Valid Cognition

→ The Idea of the Holy — Otto’s Original Case for the Numinous