Einstein vs. Bohr: The Debate Over What Quantum Physics Says Is Real

Albert Einstein and Niels Bohr agreed that quantum mechanics predicts experiments superbly. They argued about what it tells us is real. Einstein believed physics should describe a world that exists whether or not anyone looks, with no instant influences across space. Because quantum mechanics seemed to describe only what we will observe, and only in probabilities, he concluded it must be incomplete. Bohr believed the theory was complete. In his view, we can only describe atoms through whole experimental setups, using pictures that complete each other but cannot be merged into one.

Later experiments went against the fuller theory Einstein hoped for, one built on hidden, local ingredients. But the deeper question the two men argued over, what quantum mechanics says about reality, is still open.

What were they really arguing about?

Einstein’s famous line about dice makes it sound as if his only problem was chance. Scholars see two worries. First, the theory seemed to give up on describing nature as it is when no one is watching. Second, its probabilities looked basic, not just a sign of missing details (Stanford Encyclopedia of Philosophy). Some scholars argue the first worry ran deeper. In a letter to Max Born dated 31 March 1954, the physicist Wolfgang Pauli wrote that Einstein’s starting point was “realistic” rather than “deterministic” (Del Santo).

Bohr started from a different place. At a congress in Como, Italy, in September 1927, he introduced an idea he called complementarity (Bohr, 1949). Some descriptions, such as an exact position and an exact motion, need setups that exclude each other. Both are needed for a full account, but never in the same experiment. Think of a can of soup. From above it looks like a circle; from the side, a rectangle. Each view is true, and neither is complete alone. The analogy is rough, and Bohr himself never gave complementarity a precise definition (Stanford Encyclopedia of Philosophy).

What happened at the Solvay meetings of 1927 and 1930?

In October 1927, leading physicists met in Brussels for the Fifth Solvay Conference. Einstein described a thought experiment. Electrons pass through a small hole and spread toward a curved photographic screen. If the spreading wave is the complete story, why does it suddenly make a single dot in one spot? It was as if a signal flashed instantly across the screen, telling every other spot not to light up (Stanford Encyclopedia of Philosophy). Bohr later recalled that Einstein teasingly asked whether they truly believed God plays dice. Bohr replied by urging great care in giving Providence qualities in everyday language (Bohr, 1949).

You may have seen a snappier version: “Einstein, stop telling God what to do.” We could not find that line in Bohr’s or Einstein’s own writings. It appears to be a later retelling; Bohr’s own account is more careful. A full English edition of the proceedings also shows the meeting ended without agreement, not with a clear win for Bohr (Bacciagaluppi and Valentini).

At the 1930 Solvay meeting, Einstein brought his sharpest challenge, the photon box. Picture a box full of light hanging from a spring scale, with a clock inside that opens a shutter for an instant and lets one photon out. The clock gives the exact time. Weighing the box before and after gives the photon’s energy, because energy and mass are linked by E = mc². If both could be known exactly, a basic quantum limit on pinning down energy and time together would fail.

Bohr answered with Einstein’s own theory of gravity. To weigh the box, you must watch it shift slightly on the spring. But general relativity says a clock’s rate depends on its height in a gravitational field. The more precisely you fix the weight, the less precisely you know the time, so the quantum limit survives. Bohr recalled that Einstein himself helped work this out. Yet their friend Paul Ehrenfest later told Bohr that Einstein was still far from satisfied (Bohr, 1949).

How did EPR and Bohr’s reply change the debate?

By 1935 Einstein had changed his approach. Instead of trying to catch the theory in a contradiction, he argued that it was incomplete. On 15 May 1935, with Boris Podolsky and Nathan Rosen, he published “Can Quantum-Mechanical Description of Physical Reality Be Considered Complete?” The paper, known as EPR, imagines two particles that interact and then separate. Measuring one lets you predict the matching property of the other, far away, with certainty. EPR assumed that each particle has its own reality, and that nothing done in one place can instantly change what is real in another. From there, they concluded that quantum mechanics leaves something out (Stanford Encyclopedia of Philosophy).

In a June 1935 letter to Erwin Schrödinger, Einstein complained that Podolsky’s write-up had smothered the main point in formalism. In those same exchanges, Schrödinger named the strange link between the particles: entanglement (Stanford Encyclopedia of Philosophy).

Bohr answered under the same title on 15 October 1935. He agreed that measuring one particle does not physically disturb the other. But he argued it still affects the very conditions that define what can be predicted about it. Many readers found this hard to follow, and in 1949 Bohr admitted his wording had been inefficient. Einstein, for his part, came to see quantum mechanics as a description of the average behavior of many similar systems, not a complete account of any single one (Bohr, 1949).

So who was right?

In 1964 John Bell showed that a “complete” theory of the kind EPR pointed to, with local hidden variables, must disagree with quantum mechanics in certain tests. Experiments from 1972 onward sided with quantum mechanics, loophole-free tests followed in 2015, and the 2022 Nobel Prize in Physics honored this line of work (NobelPrize.org). Our articles on quantum entanglement and how scientists tested Einstein’s hunch tell that story.

So Einstein’s specific hope failed. Does that make Bohr right about everything? Thoughtful people differ. These are perspectives, not settled findings:

• A Bohr-leaning view: quantum mechanics has passed every test for a century. Asking what a particle “really” is between measurements may be the wrong question. The Copenhagen outlook, drawn from Bohr and Werner Heisenberg, is still the most popular single choice among physicists surveyed.
• An Einstein-leaning view: a good theory should say clearly what exists, not only what we will observe. Pilot-wave theory, which adds hidden but definite details and allows distant influences, and many-worlds views, where every outcome happens, both try to meet that demand while matching the data (Stanford Encyclopedia of Philosophy).

What do we still not know, and what would change the answer?

No one knows which interpretation, if any, is correct. In 2025, Nature surveyed researchers and received more than 1,100 responses. The Copenhagen interpretation led with 36 percent, far from a majority, and only 24 percent felt confident their favorite was correct. Respondents also split on whether the wave function is something real (36 percent) or a useful tool (47 percent) (Nature).

What could move the debate? An experiment showing quantum mechanics failing, for instance in larger and larger objects, would point to new physics. An interpretation that made a distinct, testable prediction, and got it right, would gain ground. Until then, choosing among interpretations rests partly on philosophy: what we think a good explanation must include.

Bohr remembered their decades of debate as good-humored, and their disagreement made physics sharper. Which side of the table would you sit on, and what would it take to move you? Keep asking, and follow the story from one letter to a Nobel Prize in The Dice Letter at 100.