Session 6 · Capstone — the doorway to the Pramāṇavārttika

Kinds of Knowing

識 — how do you actually know something is true?

Before class: Sessions 1–5. This session brings all four moves to their hardest target yet: knowledge itself.

Same toggle as before — every debate line below has both versions.

Part 1

The Puzzle

You're standing at the base of a mountain. Over its ridge, you can see a thin column of smoke rising. You cannot see any fire — the ridge blocks it completely.

Do you know there is fire on that mountain?

Most people say yes, instantly. But notice what you're actually doing: you are not seeing the fire. You are seeing smoke, and reasoning your way to a fire you've never laid eyes on. That's a completely different mental act from, say, looking directly at a candle flame in front of you. Both count as "knowing there's fire" — but they arrive at that knowledge by two entirely different roads. This session is the one that finally puts a name on the difference between those two roads — and it's the exact question Dharmakīrti's Pramāṇavārttika spends an entire treatise unpacking.

Part 2

The New IdeaTwo roads to knowledge

Quick recap — last time with these four:

"So…""It follows that…" "That's not true.""The reason is not established." "That doesn't prove it.""There is no pervasion." "Yes, I agree.""I accept it."

Seeing it directly現量 (mngon sum) — direct perception

Knowledge through direct contact — sight, sound, touch, taste, smell — right now, no reasoning required.Looking straight at a candle flame

Reasoning to it比量 (rjes dpag) — inference

Knowledge through a sign, plus a known connection between the sign and what it proves.Smoke (seen directly) + "smoke means fire" (known already) → fire (never seen)

Every session in this pilot has secretly been building toward this. Session 1's "true reason, no pervasion" trap, Session 4's cause-and-effect test — both were rehearsals for exactly this: an inference is only valid if the sign genuinely, reliably connects to what it's supposed to prove. Smoke passes; a rooster's crow (Session 4) does not.

Part 3

Worked ExampleDon't confuse the sign with the thing itself

Challenger

"So, you know there's fire on that mountain, because you can see smoke rising, and smoke always means fire." "It follows that fire on the mountain is known, because smoke — a reliable sign of fire — is directly perceived."

A directly perceived sign, plus a solid, known connection to the conclusion.

Defender

"Yes, I agree." "I accept it."

A textbook valid inference — this is exactly what makes an inference count as knowledge, not just a guess.

Challenger

"So, you directly see the fire itself, because you can see the smoke." "It follows that the fire is directly perceived, because the smoke is directly perceived."

The trap: smuggling inference in and calling it perception.

Defender

"That doesn't prove it!" "There is no pervasion."

Seeing smoke is real — but it's direct perception of smoke, not of fire. The fire is only ever known by inference here.

The habit six sessions have been building toward

Notice which road you're on. Are you seeing the thing itself, or seeing a sign and reasoning your way to a conclusion? Conflating the two is one of the most common — and most invisible — reasoning errors people make.
An inference is only as good as its pervasion. Every session's central skill — telling a true-but-toothless reason from one that actually forces its conclusion — was always secretly about this.

Part 4

PracticeSame three-move check, one last time

1. "So, you have direct perception of your friend's headache, because she told you about it."

"That doesn't prove it!""There is no pervasion." — being told is real, but you're inferring her pain from her words, not perceiving it directly. Her headache is never something you can see, hear, or touch yourself.

2. "So, you directly perceive the fire on that distant mountain, because you can feel its heat from here."

"That's not true!""The reason is not established." — the reason is simply false: a mountain that far away, you cannot actually feel its heat. No perception happened at all here.

3. "So, you know it rained last night, because the streets are wet this morning."

"Yes, I agree.""I accept it." — a directly perceived sign (wet streets) plus a reliable connection to its cause. Same shape as the smoke-and-fire inference.

Pilot complete

You've done, in six sessions, what usually takes a monk several years to begin.

You now carry the same four moves a Geluk monk carries into every debate courtyard: challenge, false reason, no pervasion, accept. You've pointed them at colors, identity, definitions, causation, and negation — and finally at knowledge itself.

That last move — telling direct perception apart from inference, and knowing exactly what makes an inference valid — is not a side topic. It's the front door of Dharmakīrti's Pramāṇavārttika (《釋量論》), the great Indian treatise on valid reasoning that every one of these six sessions was quietly preparing you to read. If you want to go further, that's where "further" actually leads.

← Course overview Start Reading the Pramāṇavārttika →