Does quantum physics need an observer?

The experiment everyone has heard of

Send electrons one at a time toward a wall with two narrow slits, and put a screen behind it. Each electron arrives as one dot. Send enough of them, and the dots build bands: light, dark, light, dark. These are interference fringes, the pattern a wave makes when it passes through both slits and meets itself on the far side.

Now put a detector at the slits to learn which one each electron went through. The fringes vanish. The dots pile up in two plain bands, one behind each slit.

This is where the popular story begins: looking changes the result. Watch the electron and it behaves one way; leave it unwatched and it behaves another. From there it is a short step to the idea that the observer, perhaps even the observer’s mind, has a hand in what happens.

The story has a weak point. A detector is a physical thing, and it touches the electron’s wave. So the fair question is whether the looking matters, or only the touching. And if only the touching, which kind?

Take the observer away

Here is the test. Remove the detector. Fly the electrons past a plain slab of material, close to one of the two paths, with nothing connected to it: no wires, no readout, no one looking. If the observer is what matters, the fringes should stay.

The experiment has been done. In 2007, Sonnentag and Hasselbach at the University of Tübingen sent electrons through an interferometer a few micrometres above a plate of doped silicon. The closer the electrons flew to the plate, and the farther apart the two paths, the weaker the fringes became. Near the surface they were gone. Nothing read the plate.

In 2020, Kerker and colleagues, also at Tübingen, repeated the experiment with two slabs: doped silicon again, and gold, whose resistivity is about a million times lower. Beside the silicon, the fringes died. Beside the gold, they barely changed. One theory, with no adjustable factor, fit both, and three competing models were ruled out.

What the slab did

The electron carries charge, so its field reaches into the slab as it passes. In silicon, that field stirs the slab’s own electrons, and the stirring is lost as a tiny amount of heat. Where the heat is left depends on which path the electron took. The slab has kept a trace of the path. That trace is a record, even though no one will ever read it.

Gold behaves differently. Its electrons follow the passing charge almost without loss, so almost nothing is left behind as heat. Almost no record is kept, and the fringes survive. An ideal superconductor, with no loss at all, should leave them alone.

So the slab did what the detector did. The detector kept a record of the path, and so did the silicon. The fringes vanish whenever a record of the path exists, whether or not anything reads it.

Touching is not enough

One might think any interaction would do. It does not. In 2005, Perreault and Cronin sent atoms through a grating of nanoscale bars. The atoms felt the walls: the van der Waals force shifted the phase of their waves. Yet the fringes stayed. The interaction left no record of the path, so it destroyed nothing.

The gold slab makes the same point. The electron meets the gold through the same electric force it meets the silicon with. What differs is whether a record is kept.

Interaction is not the criterion. A record is.

Rocks are everywhere

Once you see it, the same thing turns up all through the laboratory record. Any body that keeps a trace of the path plays the part of the slab. We call it a rock.

• A gas is a rock. Hornberger and colleagues (2003) sent C₇₀ molecules, cages of seventy carbon atoms, through an interferometer and let a little gas into the chamber. The fringes faded with the pressure, as collision theory predicts. Each gas molecule that bounced off carried a trace of the path away.
• A molecule’s own glow is a rock. Hackermüller and colleagues (2004) heated the C₇₀ molecules before they flew. Hot molecules give off thermal light, and that light carried the record away. The hotter the molecules, the weaker the fringes.
• One photon is a rock. Chapman and colleagues (1995) scattered single photons off atoms inside an interferometer. One scattered photon destroyed the fringes. When they kept only the photons whose direction left the path undetermined, the fringes came back. The record, not the kick, was what mattered.

None of these had an observer. Each had a record. The Evidence page lists nineteen experiments under the Rock, each with what it shows and what it does not.

So what is a detector?

A detector is any body that keeps an irreversible record of the path. Geiger counters and cameras are detectors. So is a slab of silicon, a puff of gas, or a single stray photon. Whether a person later reads the record makes no difference to the fringes.

Physicists call this effect decoherence. The surroundings — the electrons and vibrating atoms of the slab, the gas, the light — are called the bath. The bath takes up a trace of what passed through it, and the interference fades. Decoherence theory predicts the fading in each of these experiments, and the measurements bear it out.

The answer to the question is therefore no. Quantum physics needs no observer. It needs a record.

What decoherence leaves open

Decoherence explains why the fringes fade. It leaves one thing open: each electron still makes one dot, at one place and one time. Something happens on the screen. Decoherence describes how the parts of the wave stop interfering; it does not say why one dot appears, rather than none or many.

Anchored Causality Theory (ACT) starts from that question. Its answer fits in one line: the wave is what exists; the particle is what happens.

The electron is a wave from start to finish, never a little ball on a path. The bath around a slab or a detector keeps a record that tells the alternatives apart. When the record has passed beyond recall, so that nothing the system will actually undergo can restore the interference, the record is released, and one alternative becomes an event: a mark in the one actual history. The dot on the screen is such a mark. No observer is needed, and the Schrödinger equation is never modified.

Seen this way, the Rock does more than remove the observer. It finds the source of the decoherence, the slab’s own bath. And it gives that bath its first measured property: for a slab, the record follows the dissipation — strong in silicon, weak in gold, expected to vanish in a superconductor. No dissipation, no record. That is the first step of ACT’s argument, and it is Section 2 of the flagship paper.

Where to go next

1. P. Sonnentag & F. Hasselbach, Phys. Rev. Lett. 98, 200402 (2007).
2. N. Kerker, R. Röpke, L. M. Steinert, A. Pooch & A. Stibor, New J. Phys. 22, 063039 (2020).
3. J. D. Perreault & A. D. Cronin, Phys. Rev. Lett. 95, 133201 (2005).
4. K. Hornberger et al., Phys. Rev. Lett. 90, 160401 (2003).
5. L. Hackermüller et al., Nature 427, 711 (2004).
6. M. S. Chapman et al., Phys. Rev. Lett. 75, 3783 (1995).
7. K. Sonderegger, Anchored Causality Theory: A Record-History Ontology for Open Quantum Systems, preprint, version 3.6 (2026), doi:10.5281/zenodo.23811463.