Consider a simple proposition: fire a single electron at a barrier with two narrow openings, and ask through which slit it passed. The question seems trivial. It is not. After more than a century of refinement, this experiment continues to resist any answer compatible with our ordinary notions of objects moving through space.

Richard Feynman declared the double-slit experiment contains the only mystery of quantum mechanics. This was not modesty about other puzzles—superposition, entanglement, decoherence—but a recognition that they all crystallize here, in this one apparently mundane setup. Understand the double slit, he suggested, and you have understood the strangeness in its purest form.

What unsettles is not that electrons behave like waves, nor that they behave like particles, but that they refuse to commit to either picture in any consistent way. The interference pattern built up one electron at a time on a distant screen forces us to confront a possibility our classical intuitions cannot accommodate: that the question which path did it take may not have an answer until we ask it, and that the asking changes what is asked about.

Send electrons toward the double slit one at a time, separated by minutes or hours, and each arrives at the detector as a single localized event—a discrete spot, unmistakably particle-like. There is no ambiguity at the moment of detection. Something arrives somewhere.

Yet as these individual arrivals accumulate, a pattern emerges that no classical particle ensemble could produce. Bands of high and low density form across the screen, the unmistakable signature of wave interference. Each electron, isolated from all others, somehow contributes to a structure that requires waves passing through both slits simultaneously.

This is the heart of the paradox. The pattern cannot be attributed to electrons interacting with one another, since they traverse the apparatus alone. Whatever produces the interference must be intrinsic to each individual electron's encounter with the slits. The wavefunction passes through both openings; the detection event registers at one location.

The mathematics handles this without contradiction. The probability amplitude for arrival at each point on the screen is computed by summing contributions from both possible paths, and the squared modulus yields the observed distribution. The formalism works flawlessly. What it does not provide is a story about what the electron is between emission and detection.

We are forced into an uncomfortable position. Either the electron is a thing that takes both paths in some literal sense, or it is something for which the very concept of path does not apply between measurements. Both options strain the inherited vocabulary of physics, which assumes objects have trajectories whether or not we observe them.

Takeaway

When a phenomenon resists every consistent picture we can draw, the failure may lie not in our imagination but in the assumption that a picture exists at all.

The natural response is to settle the question empirically. Place a detector at one of the slits and watch which path each electron actually takes. The experiment has been performed in countless variants, with increasingly delicate measurement schemes, and the result is invariant: the moment which-path information becomes available, the interference pattern vanishes. Two simple bands replace the fringes, exactly as classical particles would produce.

What makes this profound is that the disturbance argument—the idea that the detector physically jostles the electron—has been ruled out by quantum eraser and interaction-free measurement experiments. Even when the which-path information is encoded without any momentum kick to the particle, even when it is recorded only in principle and never actually read, the interference still disappears.

The relevant variable is not disturbance but distinguishability. If the universe contains, anywhere, a record sufficient to determine which path was taken, the interference is gone. Erase that record before the information propagates irreversibly, and the interference can return. The pattern responds not to physical interaction but to the structure of available information.

This forces a reconceptualization of measurement itself. Measurement is not a process by which we extract pre-existing facts about a system. It is a process by which certain facts become defined at the cost of others. Path-definiteness and interference-visibility are not two qualities the electron simultaneously possesses, with measurement merely revealing one of them.

Heisenberg saw this clearly: the trajectory comes into existence only when we observe it. The electron between slits and screen has no path in the classical sense, not because we are ignorant of it, but because the property is not instantiated. To demand a path is to demand a determinate answer to a question the world has not posed.

Takeaway

Knowledge is not a passive copy of reality but an active selection from incompatible possibilities, each excluding the others by its very actualization.

Bohr called the principle complementarity: certain pairs of descriptions are each indispensable for a complete account of phenomena, yet cannot be simultaneously applied. Wave and particle are the canonical pair. Each captures something real about quantum entities, and each excludes the other in any single experimental context.

The double slit makes complementarity tangible. The interference pattern requires the wave description—amplitudes superposing across both paths. The localized detection event requires the particle description—a quantum of energy delivered at a point. Neither alone suffices; together they form a coherent account only if we abandon the demand that both descriptions apply at once.

There is a temptation to seek deeper reality behind these complementary aspects, some hidden mechanism that would explain how the electron really behaves. Decades of effort, from de Broglie-Bohm pilot waves to many-worlds branching, have produced internally consistent frameworks. None has yielded a unique experimental signature distinguishing it from standard quantum mechanics. The mystery is not solved by interpretation; it is relocated.

Perhaps the lesson is that our intuitions, sculpted by macroscopic experience, were never going to fit the microscopic world. The categories of thing and process, of here and there, of before and after, are emergent features of a coarse-grained reality. At finer resolution, they fragment into something the language of classical physics cannot capture.

What survives is the formalism and the experimental predictions, both of staggering precision. What does not survive is the picture-thinking we inherited from a world of billiard balls and water waves. The double slit is the place where this loss becomes undeniable, where physics confronts the limits of what its concepts can mean.

Takeaway

Complementarity is not a deficiency of our knowledge but a structural feature of reality, suggesting that some truths can only be approached through pairs of descriptions that exclude each other.

The double slit endures as a touchstone because it asks a question physics cannot evade: what is happening when no one is looking? The honest answer is that our most successful theory does not say. It tells us what to expect when we measure, with extraordinary accuracy, but it offers no narrative of the unobserved.

This silence is not a temporary gap awaiting future theory. It may be the shape of the world itself—a reality in which definiteness is contextual, in which properties are relational rather than intrinsic, in which the boundary between observer and observed cannot be cleanly drawn.

To sit with the double slit is to acknowledge that the universe is stranger than the categories we brought to it. The mystery Feynman pointed to is not a puzzle awaiting solution but an invitation to think differently about what it means for something to exist at all.