The Zipper Problem

Try this. Without looking anything up, explain how a zipper works. Not what it does, but how it does it: what the slider actually does to the teeth, and why they stay locked once it passes.

Most people get about one sentence in before the confidence drains away. That moment has a name. In 2002, Yale psychologists Leonid Rozenblit and Frank Keil called it the illusion of explanatory depth (DOI).

They asked people to rate how well they understood everyday devices like zippers, flush toilets, and cylinder locks. Then they asked them to write out, step by step, how each one worked. Afterward, people rated their understanding again, and the ratings fell. Nothing had changed except that they had tried to explain.

What the study uncovered is a quiet substitution. I recognize the thing and can use it, so I assume I understand it. Familiarity poses as knowledge, and it does so convincingly, because from the inside the two feel the same.

The finding is new. The insight is not. Socrates ran this same experiment in the streets of Athens, questioning politicians, poets, and craftsmen who were sure they understood justice or courage, and asking them to explain. The explanations collapsed. His conclusion, as Plato tells it in the Apology, 21a–23b, was that his only advantage was knowing that he didn’t know.

The thing we never stop using

Now run the same test on something closer. Explain how you are aware of these words. Not which part of the brain handles vision, but how that processing becomes the experience of seeing them: the felt quality of it, the fact that there is something it is like to be you, reading, right now.

If the zipper stalled you at step two, this one stalls you before step one. Yet almost nobody walks around feeling confused about consciousness. We use it every waking second. It is the most familiar thing we have.

Augustine caught this sixteen centuries ago, writing about time in Book XI, chapter 14 of the Confessions. If no one asks me what time is, he said, I know. If I try to explain it to someone who asks, I don’t. Swap in consciousness and the sentence still holds.

That is exactly the problem. Rozenblit and Keil found the illusion was strongest for explanatory knowledge, the how-does-it-work kind, rather than for facts or procedures. Consciousness is the extreme case: total familiarity, no visible mechanism. In 1995, the philosopher David Chalmers named the gap the hard problem of consciousness in an essay that the Stanford Encyclopedia of Philosophy discusses as an open question: why physical processing is accompanied by experience at all.

The sages of the Upanishads saw why. The Kena Upanishad, 1.7 in Max Müller’s translation, describes the Self as that which the eye cannot see, but by which the eye sees. The instrument doing the looking can’t turn around and look at itself. Our familiarity with consciousness isn’t just close, it’s total, and that is what makes it so easy to mistake for understanding.

“It’s just the brain”

The first version of the illusion belongs to the confident materialist. Consciousness is what the brain does, the argument goes, so where is the mystery?

But “the brain does it” is a location, not a mechanism. It is like explaining the zipper by saying the slider does it. True, and it tells you nothing about how. Neuroscience has made real progress mapping which brain activity accompanies which experiences. Those are correlations. A map of where something happens is not an account of why it feels like anything.

None of this means materialism is wrong. It means the sentence “it’s just neurons” carries the feeling of an explanation without the content of one. Ask it for the next step and it has nothing to give.

“It’s the soul”

The same move runs in the other direction. For many religious people, consciousness is simply the soul, the immaterial self that animates the body. Mystery solved.

Except the word does the same job “the brain” did. It names where consciousness lives without saying how the soul produces experience, how it connects to a body, or why an injury to the brain can change the person. Ask for step two and you get silence, or another word.

I say this as someone who takes the sacred seriously and has spent close to forty years in contemplative practice. That is exactly why I distrust the shortcut. And the traditions themselves, at their deepest, distrust it too.

Christian mysticism built a whole path around refusing easy labels. Pseudo-Dionysius, writing around 500, taught that God is better approached by saying what God is not. The anonymous fourteenth-century author of The Cloud of Unknowing, chapter 6, wrote, “By love may He be gotten and holden; but by thought never.” The Brihadaranyaka Upanishad, 2.3.6 does the same with the Self: neti neti, not this, not this. Every description offered is set aside.

Those are not failures of explanation. They are traditions that saw the illusion of explanatory depth coming and built practices to resist it. A faith that says “the soul” and stops thinking has traded that inheritance for a label, and a label is easier to defend than wonder.

The machine question

The illusion is loudest right now in arguments about artificial intelligence. One camp says language models are obviously not conscious, since they only predict the next word. Another sees signs of an inner life in their fluent, reflective answers.

Both positions assume we understand consciousness well enough to check for it. We don’t. “It’s only prediction” is the zipper again: a description of the mechanism at one level, offered as if it settled what that mechanism does or doesn’t give rise to. “It sounds like it feels something” mistakes fluent output for the thing itself.

There is an irony here. The original illusion comes from mistaking a smooth surface for depth, and AI systems are extraordinarily good at smooth surfaces. Ask one to explain consciousness and you will get an answer that feels complete. That may leave us more confident than ever about a question no one has answered.

Asking for the next step

The answer is not to throw up our hands. Rozenblit and Keil’s finding carries its own remedy: the illusion broke when people tried to explain. Writing out the steps is what exposed the gaps.

So the practice is simple and uncomfortable. When someone, including you, says what consciousness is, ask for the next step. Then the one after that. Notice the exact point where the explanation turns into a word.

This is also one of the oldest methods in contemplative practice. The Buddha asked his students to look for the self they were so sure of. Examined closely, it breaks into parts: body, feeling, perception, mental formations, consciousness. None of them is the solid “I” they started with. The confident sense of self was familiarity, not understanding. The Anattalakkhaṇa Sutta (SN 22.59) lays out this examination of the five aggregates.

Zen tells of the monk Fayan, whose teacher Dizang asked him the purpose of his pilgrimage. Fayan said he didn’t know. Dizang replied that not knowing is most intimate. In the West, Nicholas of Cusa called the same posture learned ignorance: a not-knowing earned by pushing understanding to its limit, not by refusing to try. The dialogue is Case 20 of the Book of Equanimity.

Consciousness may be the one thing we can never step back from to examine, because the examining is done by it. It is the Upanishads’ eye that cannot see itself. That makes it the perfect home for this illusion. It also makes humility less a concession than a method, the only stance that keeps the question open long enough for anyone to actually answer it.