Not just “philosophical mumbo-jumbo”: An experiment shows how facts may depend on the observer.
Davide Bonazzi/Salzman Art
Nearly 60 years ago, the Nobel prize–winning physicist Eugene Wigner captured one of the many oddities of quantum mechanics in a thought experiment. He imagined a friend of his, sealed in a lab, measuring a particle such as an atom while Wigner stood outside. Quantum mechanics famously allows particles to occupy many locations at once—a so-called superposition—but the friend’s observation “collapses” the particle to just one spot. Yet for Wigner, the superposition remains: The collapse occurs only when he makes a measurement sometime later. Worse, Wigner also sees the friend in a superposition. Their experiences directly conflict.
Now, researchers in Australia and Taiwan offer perhaps the sharpest demonstration that Wigner’s paradox is real. In a study published this week in Nature Physics, they transform the thought experiment into a mathematical theorem that confirms the irreconcilable contradiction at the heart of the scenario. The team also tests the theorem with an experiment, using photons as proxies for the humans. Whereas Wigner believed resolving the paradox requires quantum mechanics to break down for large systems such as human observers, some of the new study’s authors believe something just as fundamental is on thin ice: objectivity. It could mean there is no such thing as an absolute fact, one that is as true for me as it is for you.
“It’s a bit disconcerting,” says co-author Nora Tischler of Griffith University. “A measurement outcome is what science is based on. If somehow that’s not absolute, it’s hard to imagine.”
For physicists who have dismissed thought experiments like Wigner’s as interpretive navel gazing, the study shows the contradictions can emerge in actual experiments, says Dustin Lazarovici, a physicist and philosopher at the University of Lausanne who was not part of the team. “The paper goes to great lengths to speak the language of those who have tried to merely discuss foundational issues away and may thus compel at least some to face up to them,” he says.
Wigner’s thought experiment has seen renewed attention in recent years. In 2015, Časlav Brukner of the University of Vienna tested the most intuitive way around the paradox: that the friend inside the lab has, in fact, seen the particle in one place or another, and Wigner just doesn’t know what it is yet. In the jargon of quantum theory, the friend’s result is a hidden variable.
Brukner ruled out that conclusion in a thought experiment of his own, using a trick—based on quantum entanglement—to bring the hidden variable out into the open. He imagined setting up two friend-Wigner pairs and giving each a particle, entangled with its partner in such a way that their attributes, upon measurement, are correlated. Each friend measures the particle, each Wigner measures the friend measuring the particle, and the two Wigners compare notes. The process repeats. If the friends saw definite results—as you might suspect—the Wigners’ own findings would show only weak correlations. But instead they find a pattern of strong correlations. “You run into contradictions,” Brukner says. His experiment and a similar one in 2016 by Daniela Frauchiger and Renato Renner of ETH Zürich led to an outpouring of papers and heated discussion at conferences.
But in 2018, Richard Healey, a philosopher of physics at the University of Arizona, pointed out a loophole in Brukner’s thought experiment, which Tischer and her colleagues have now closed. In their new scenario they make four assumptions. One is that the results the friends obtain are real: They can be combined with other measurements to form a shared corpus of knowledge. They also assume quantum mechanics is universal, and as valid for observers as for particles; that the choices the observers make are free of peculiar biases induced by a godlike superdeterminism; and that physics is local, free of all but the most limited form of “spooky action” at a distance.
Yet their analysis shows the contradictions of Wigner’s paradox persist. The team’s tabletop experiment, in which they created entangled photons, also backs up the paradox. Optical elements steered each photon onto a path that depended on its polarization: the equivalent of the friends’ observations. The photon then entered a second set of elements and detectors that played the role of the Wigners. The team found, again, an irreconcilable mismatch between the friends and the Wigners. What is more, they varied exactly how entangled the particles were and showed that the mismatch occurs for different conditions than in Brukner’s scenario. “That shows that we really have something new here,” Tischler says.
It also indicates that one of the four assumptions has to give. Few physicists believe superdeterminism could be to blame. Some see locality as the weak point, but its failure would be stark: One observer’s actions would affect another’s results even across great distances—a stronger kind of nonlocality than the type quantum theorists often consider. So some are questioning the tenet that observers can pool their measurements empirically. “There are facts for one observer, and facts for another; they need not mesh,” suggests study co-author and Griffith physicist Howard Wiseman. It is a radical relativism, still jarring to many. “From a classical perspective, what everyone sees is considered objective, independent of what anyone else sees,” says Olimpia Lombardi, a philosopher of physics at the University of Buenos Aires.
And then there is Wigner’s conclusion that quantum mechanics itself breaks down. Of the assumptions, it is the most directly testable, by experiments that are probing quantum mechanics on ever larger scales. But the one position that doesn’t survive the analysis is to have no position, says another co-author at Griffith, Eric Cavalcanti. “Most physicists, they think: ‘That’s just philosophical mumbo-jumbo,’” he says. “They will have a hard time.”