Quantum Mechanics and the Measurement Problem

The Success and the Strangeness

Quantum mechanics makes predictions that have been tested to extraordinary precision. The anomalous magnetic moment of the electron — a small deviation from the value predicted by classical electrodynamics — is predicted by quantum electrodynamics to agree with experiment to eleven decimal places. No other scientific theory has achieved anything close to this. Quantum mechanics is, by this measure, the most successful physical theory ever constructed.

It is also, by widespread acknowledgment among physicists who have thought carefully about it, profoundly unclear about what it says. Not unclear in the way a new theory might be unclear while awaiting further development — unclear in a way that persists after seventy years of debate and appears to be intrinsic to the structure of the theory.

The strangeness is not the colloquial version — “particles are waves,” “things can be in two places at once,” the routine wonders of popular science writing. The real strangeness is more specific and harder to dissolve into comfortable imagery.

The Formalism

In quantum mechanics, the state of a system is described by a wave function — a mathematical object that encodes the probabilities of finding the system in various configurations when measured. The wave function evolves deterministically according to the Schrödinger equation. If you know the wave function at time t, you can compute it at any later time t’. This evolution is smooth, continuous, and deterministic — more like classical mechanics than you might expect.

The strangeness enters when you interact with the system — when you measure it. Before measurement, a particle can exist in a superposition of multiple states: spin up and spin down simultaneously, in some sense. The wave function assigns probability amplitudes to each possibility. When you make a measurement, you get a definite outcome — spin up, or spin down. The wave function, which had been smeared across possibilities, collapses to the definite state you observed.

The collapse is not described by the Schrödinger equation. The Schrödinger equation evolves wave functions smoothly — it doesn’t produce sudden collapses to definite states. Measurement is treated as a special process, outside the normal quantum dynamics, that produces definite outcomes from indefinite superpositions. The rule for this process — the Born rule — says the probability of getting a particular outcome is proportional to the squared amplitude of the wave function for that outcome. This rule is confirmed experimentally with extreme precision.

The Measurement Problem

The measurement problem is: why should measurement be a special process? Measurement is a physical interaction between a system and an apparatus. Both the system and the apparatus are physical objects. If quantum mechanics applies to all physical objects, it applies to the apparatus. When the apparatus interacts with the system, the combined system + apparatus should evolve under the Schrödinger equation — and the Schrödinger equation will produce a superposition of (system in state A, apparatus showing A) + (system in state B, apparatus showing B). No collapse; just a larger superposition.

You can extend this to the observer: the observer is also a physical object, and the Schrödinger equation will produce a superposition of (system in A, apparatus showing A, observer seeing A) + (system in B, apparatus showing B, observer seeing B). The observer never seems to experience a superposition — they always get a definite outcome. Where does the superposition end and the definite outcome begin?

This is the measurement problem. It is not resolved by pointing to the technical apparatus of decoherence — the process by which quantum superpositions become effectively invisible through entanglement with the environment. Decoherence explains why macroscopic superpositions are practically unobservable. It does not explain why there is a definite outcome at all rather than a superposition that merely looks classical.

The Interpretations

There are multiple, mutually inconsistent interpretations of quantum mechanics, each resolving the measurement problem differently. They all predict identical experimental outcomes — this is what makes the problem so hard to resolve empirically.

The Copenhagen Interpretation, associated with Niels Bohr and Werner Heisenberg, says: don’t ask what happens between measurements. Quantum mechanics is a tool for predicting measurement outcomes; it is not a description of what is happening in between. The wave function is a mathematical device, not a physical reality. This is not a coherent physical account — it is a principled agnosticism about asking certain questions. Most working physicists operate in a vague Copenhagen-adjacent mode: use the formalism, don’t worry about the foundations.

Many-Worlds, proposed by Hugh Everett in 1957, takes the Schrödinger equation seriously and applies it universally without collapse. When a measurement occurs, the world branches: in one branch, the observer sees outcome A; in another, they see outcome B. Both branches are real. Both observers are real. The observer’s experience of a definite outcome is correct — but there are other versions of them who got the other outcome. The wave function never collapses; it just gets very large. This interpretation preserves the determinism of the Schrödinger equation but multiplies observers without any mechanism for doing so, and faces serious technical problems explaining why the Born rule probabilities have the values they do.

Pilot-Wave Theory (de Broglie-Bohm) says that particles are real, with definite positions at all times, but are guided by a real wave field — the pilot wave — whose evolution is the Schrödinger equation. Measurement reveals the pre-existing particle position; no collapse needed. This is explicitly non-local: the pilot wave for an entangled pair of particles extends over all of space, and changing the configuration at one location instantaneously updates the pilot wave everywhere. It reproduces all quantum mechanical predictions exactly but at the cost of non-locality and the need for a preferred reference frame — in tension with special relativity.

Relational Quantum Mechanics, developed by Carlo Rovelli, says that the wave function is not a description of a system’s intrinsic state but of its state relative to a particular observer. Different observers have different, equally valid wave functions for the same system. Quantum mechanics is about relations between systems, not absolute states.

Heisenberg’s Uncertainty Principle

The uncertainty principle is the most popularly misunderstood result in physics. The common version — you disturb a particle when you measure it, so you can’t know both position and momentum — is wrong, or at least deeply incomplete. The uncertainty is not about measurement disturbance. It is about what states are available.

The formal statement: for any quantum state, the product of the standard deviation of position (σₓ) and the standard deviation of momentum (σₚ) satisfies σₓσₚ ≥ ℏ/2. A quantum state that has a very definite position must have very spread-out momentum, and vice versa. This is a feature of the wave function, not of measurement techniques. A photon in a very precisely defined momentum state has a wavelength with no defined position — it’s spread over all of space. You can’t prepare the photon in a state with both definite position and definite momentum, because no such state exists.

The uncertainty is ontological, not epistemological — it is not ignorance of values that are there; it is the absence of definite values in states where they aren’t defined.

What the Weirdness Is Pointing At

The empirical core of quantum mechanics is not in dispute. Every experiment confirms the Born rule probabilities. Entanglement is real — correlations between distant particles violate Bell inequalities by amounts that can’t be explained by any local hidden-variable theory, a result proven by John Bell in 1964 and confirmed experimentally by Alain Aspect in 1982. Whatever quantum mechanics is describing, it is something genuinely non-classical about the structure of the world.

The interpretive question — what it says about what the world is like — is the one that remains open. This is unusual in the history of physics. Usually, once a theory is confirmed to this level of precision, its ontology is reasonably clear — even if approximate and superseded by later theories. Quantum mechanics is different: the better confirmed it becomes, the harder it is to say what it means.

Richard Feynman’s famous remark — “nobody understands quantum mechanics” — was not false modesty. It was an accurate description of where the field stands. The tool works perfectly. The account of what the tool is describing is genuinely contested.