Does the Eye Travel to What It Sees?

The whole chapter is generated by a question the Outsider did not realize he had answered twice. His formula has two elements — a seeing performed by the eye, and a knowing that arises from the conjunction — and it never says which comes first.

Put the seeing first and the knowing is unemployed: the color has already been taken in, the work is done, and the knowledge that shows up afterward has nothing to do. Put the knowing first and it has arrived before its own conditions, which contradicts the formula itself, since the knowing was defined as what arises from the conjunction.

Notice that this is the same question — 復何用, what further use? — that ran through the refutation of the self in Part 6. There it exposed a self with no work left after the mind had done the knowing. Here it exposes a knowing with no work left after the eye has done the seeing. The treatise is applying a single test to whatever gets posited: once you have described the process fully, is there anything for the extra item to do?

The three-way test is the treatise's standard instrument and it is worth having in one place. Anything said to arise is either already there, not there at all, or half and half. If it is already there, arising adds nothing. If it is not there, there is nothing to do the arising. And "half arisen" is not a third option but a contradiction wearing a disguise — the same thing said to be and not to be in the same respect.

The last clause is the sharpest and is easy to read past. The Outsider chose simultaneity to preserve the tight fit between seeing and knowing. But strict simultaneity destroys dependence: if the two arise together, neither is waiting on the other, and the formula "knowledge arises from the conjunction" is abandoned. He has paid for the fit with the very relation the fit was supposed to explain — which is the same fault the Twelve Gate Treatise names with the horns of an ox, which grow at once and so cannot cause each other.

The moon and the pot make an argument that is genuinely empirical, and it is worth noticing that the treatise is willing to reason this way. Step outside, look at a pot at your feet and the moon overhead: they arrive together. If seeing required a journey, the moon should lag, and by an enormous margin. It does not. So nothing travels.

But the reply immediately runs the opposite objection with equal force, and this is what makes the section more than a physics argument. If nothing travels — if distance is simply irrelevant to seeing — then why is there any limit at all? Why does a wall stop sight, and why can you not see what is on the far side of the world? Distance manifestly matters, and manifestly does not behave like a journey. Both models are refuted by ordinary observation, and neither party has a third.

Then the two dilemmas that finish it, and they are the elegant part. First, whether the travelling eye has seen before it sets out: if it has, the journey is pointless; if it has not, it is blind, and a blind traveller cannot find what the mind wanted — 趣東則西, it sets out for the east and goes west. Second, and best, the question of location. Wherever the eye is, seeing happens there and not elsewhere. Send it to the moon and your face is empty. Keep it in your face and nothing is present at the moon. The taking-hold has to occur where both parties are, and on either model they are never in the same place.

This is the same structure the treatise ran against the lamp in Part 2 — the lamp that never reaches the darkness — and it is the structure the Mūlamadhyamakakārikā gives at length in its chapters on motion and on seeing. Contact between two things, each of which has to be somewhere, turns out to have no place to happen.

The fire returns for the fourth time in this treatise, and by now its role is fixed: it is the model of a mark that is genuinely inseparable from what bears it. Fire is not hot on the outside and cool within; the heat is everywhere the fire is, which is exactly why heat can serve as fire's mark. Make seeing the eye's mark in that strong sense and the eye should be no less illuminated to itself than the room is.

The fingertip is the Outsider's best reply and it fails on a distinction the treatise has already established. In Part 5 the hand was offered as a self that is a self even when not grasping, and the answer was that grasping is the hand's act and not its mark. The same answer is available here without modification: a fingertip is recognized by its shape, not by its touching, so its inability to touch itself takes nothing away from what it is. The Outsider, however, has explicitly said that seeing is the eye's mark — its identifying nature, not something it does — and marks are not intermittent and do not have exceptions.

This consistency across chapters is one of the treatise's virtues and it repays attention. The mark-versus-act distinction was established against the self in Chapter Two and is simply picked up and used here against the eye. The book accumulates tools.

The corpse is the decisive image, and it is decisive because both parties have already agreed to it. Back in Part 5 the Outsider himself argued for a self from the difference between a living body and a dead one, and in Part 6 he explained that a dead man does not see because he has no mind. So mind-in-the-body is what makes a body alive, on his own account. Send the mind to the moon and you have described a corpse looking at the moon.

The observation about memory is a small piece of phenomenology and very well made. When you remember last summer, nothing of you goes to last summer; the remembering happens now, here, complete. Reaching a distant object in thought is not a journey, and the Outsider's whole model treats it as one.

The final proposal is his most ingenious and is worth stating fairly: mind and eye stay home, and what travels is the eye's light, dispatched under the mind's power to make contact with the color out there. Every party is where it should be, and the conjunction is preserved through an intermediary. It is a genuinely good repair. What it cannot survive is the question the next section asks — not how the parties are connected, but which of them, once connected, is doing the seeing.

This is the chapter's real conclusion and it is a demand for an inventory. Three things are in contact. Go through them one at a time.

The mind is not the eye — the Outsider has insisted on this himself, since he needs them to be two things that conjoin — so seeing is not the mind's mark, and the mind does not see. The eye is a physical organ, built from the four great elements, so knowing is not its mark, and the eye does not know. And the color is just a color: it neither sees nor knows anything. Three items, none of which can do the job, however tightly you press them together. Conjunction does not manufacture a capacity that none of the parties has.

The force of this depends entirely on the Outsider's own dualism, and that is what makes it more than a trick. He needed the mind and the eye to be really distinct — otherwise there was nothing to conjoin, and his whole account of perception collapsed at the start. But really distinct things keep their own natures, which is the principle the treatise established back with the salt in Part 2 and the ox in Part 4: 鹽相鹽中住, the mark of salt abides in salt. Marks do not migrate across a conjunction. So the very distinctness that made the mechanism possible guarantees that the mechanism cannot work.

Two small textual points, since readers of the Chinese will notice them. The 內曰 that opens this section is not followed by the colon that appears everywhere else in the chapter — a small irregularity in the received text, which we have translated as though the colon were present. And the received punctuation reads 非見相故，不能見眼。四大造故，非知相。, attaching 眼 to the preceding clause. Since it is the eye and not the mind that is made of the four great elements, the character plainly begins the next sentence, and we have translated it so; the punctuation, not the text, is what we depart from.

Finally, the closing sentence is worth registering for how much it does in seven characters: ear, nose, tongue and body are to be refuted in this same way. One faculty has been examined and the other four are dispatched by analogy. This is a habit worth watching for — the same thing happens to the forest and the army in Part 7. The treatise argues one case thoroughly and then asserts the generalization, and the reader is expected to have understood the argument well enough to run it themselves.