The second quantum revolution
20 June 2007
NewScientist.com news service
Michael Brook

Adjusting to this new universe was a tall order. Some physicists constructed elaborate philosophies to deal with the implications. Albert Einstein, on the other hand, famously rejected entanglement as "spooky". He was convinced that quantum entanglement couldn't be real because the implications would run too deep: any effort to produce a unified theory, one that tied quantum mechanics together with relativity and other physical theories, would need to reconcile the weirdness of entanglement with relativity's rather more practical grasp of time and space. That just seemed altogether too hard.

Not that he ever gave up on a theory of everything. Einstein spent the latter part of his life trying to construct a unified universe, without success. He also continued to wrestle with quantum spookiness on his own. To most physicists, quantum theory was useful if you wanted to design a laser or transistor, but it didn't do to think about it too deeply.

That attitude prevailed even among those who wanted to understand the intimate workings of the universe. So the "foundations" of quantum theory - the assumptions behind how it describes particles, fields and reality itself - has taken a back seat in the quest for a unified theory. "Einstein's strong belief that the foundational issues in quantum mechanics are a necessary part of solving the problem of unification got suppressed and lost," says Lee Smolin of the Perimeter Institute in Waterloo, Canada.

But what was lost has now been found again. At the core of this renaissance is a growing tally of results showing that entanglement has profound implications for our view of reality. Recent experiments led by a group at the University of Vienna, Austria, provide the most compelling evidence yet that there is no objective reality beyond what we observe. This idea, that our measurements create reality, is controversial and scarcely new, but the mounting evidence for it could have major implications in the search for a theory of everything. Indeed, we are at the "conceptual beginnings of a second quantum revolution", according to Alain Aspect of the Institute of Optics at Palaiseau in France.

The original quantum revolution came in the 1920s. Einstein's big problem with quantum mechanics was that it contradicted the intuition supported by all other theories of physics. In our experience, objects have definite locations in

space and a limited range of influence. According to quantum theory, however, a pair of particles would be able to share information about each other's quantum states - and sometimes influence them - even where the distance and timing involved meant that no signal could have passed between them.

This suggested to Einstein that when it came to describing physical reality, quantum theory was lacking something. It is not that the information about the particles' states is shared in a spooky link between them, he thought - it is simply that we don't know where to find all the factors working on a particle to determine its momentum, say. Uncover these "hidden variables", Einstein said, and all the mystical spookiness would melt away. In its place would be a quantum theory that operates by the rules of common sense.

True to form, Einstein didn't leave it at that: he formulated a mathematical argument to bolster his case. Working with Boris Podolsky and Nathan Rosen, he threw down the gauntlet to the quantum camp.

In 1935, the three theorists published the Einstein-Podolsky-Rosen thought experiment, known as EPR. It said that if quantum theory was correct and complete, you should be able to perform an experiment in which a measurement on one entangled particle instantaneously affects the quantum state of its distant twin. At the time, this seemed to violate the known laws of physics and cast doubt on whether quantum theory could be considered to be a complete description of reality.

Much gnashing of teeth and triumphalism followed. Erwin Schrödinger, who had also wrestled with the implications of quantum mechanics, gleefully told Einstein the EPR paper had caught quantum theory "by the throat". No one knew how to turn the thought experiment into a real one, however, so the two camps - Einstein's opposition was spearheaded by the fearsome Danish physicist Niels Bohr - spent the next two decades shaking their fists at each other.

In 1964, nine years after Einstein's death, physicist John Bell worked out a scheme to test EPR. He believed, as Einstein did, in the intuitive idea of "local realism": that a particle cannot be instantly influenced by a distant event, and that its properties exist independently of any measurement.

Bell derived a mathematical formula that quantified what you would get if you made measurements on an entangled pair of particles. If local realism was correct, the correlation between measurements made on one of the pair and those made on its partner could not exceed a certain amount, because of each particle's limited influence. The stage was set for a definitive experiment.

Many years passed before Aspect built the necessary equipment in his basement laboratory at the University of Paris, but by 1982 he had a result: Bell's formula did not agree with quantum experiments (New Scientist, 24 November 1990, p 43). The world, Aspect announced, could not be both local and real - Einstein was wrong. But which idea had to go, realism or locality? Do particles only acquire real properties when they are measured? Or are distant, instant influences possible between particles?

The answer would come from another source. In 1976, well before Aspect had carried out his experiment, physicist Anthony Leggett had what he calls "the kernel" of an idea to rework Bell's formula with a twist: he quantified what you would get if you made measurements on entangled particles, assuming that distant, instant influences were in fact possible. Leggett eventually published this formula in 2003, the year he won the Nobel prize in physics for his work on the quantum properties of helium-3.

Enter a team of Austrian and Polish physicists, who have now done experiments on pairs of entangled photons to test Leggett's formula (see "The end of reality"). The team, led by Markus Aspelmeyer of the Austrian Academy of Sciences and Anton Zeilinger of the University of Vienna, managed to reduce the noise in their set-up by a necessary factor of 10, compared with Aspect's work. They published their results in April (Nature, vol 446, p 871).

What they found is that Leggett's formula is violated as well: even if you allow for instantaneous influences, quantum measurements do not fit with the idea of an objective reality. This is surprising because you might expect that, once any spooky "non-local" action is allowed, you could account for almost any relationship between two particles, and there would be no reason to ditch our concepts of reality. "This is not the case," says Aspelmeyer.

Although some loopholes remain - not all non-local models have been ruled out - we now have to face the possibility that there is nothing inherently real about the properties of an object that we measure. In other words, measuring those properties is what brings them into existence. "Rather than passively observing it, we in fact create reality," says quantum researcher Vlatko Vedral of the University of Leeds, UK.

This idea may not be new, but the evidence for it is, and it could have serious implications for a theory of everything: it tells us that what we think of as real is not necessarily so. "We know from our experience that there is a 'real' world with 'real' physical events, starting from clicks in a detector in the laboratory to experiencing a headache after too many beers," Aspelmeyer says. But that doesn't mean our physical theories ought to slavishly follow that experience, he points out - perhaps they need to dig deeper.

While quantum researchers may find this satisfying, it raises deep concerns for anyone attempting to unify the universe. General relativity, Einstein's theory of gravity, is fully realistic - it relies on things existing independent of measurements. So the search for a theory of everything, which involves uniting quantum physics with general relativity, may be even more difficult than we thought. "It is not at all clear how to construct a theory of gravity that is not real, which is what we need to do if we want to quantise gravity," Vedral says.

Spooky space-time

As Einstein suggested decades ago, entanglement could be the key. "Understanding entanglement means understanding a great deal about the principles upon which physical theories are based," Aspelmeyer says. His Vienna colleague ÇCaslav Brukner goes further. For more than two decades, Brukner points out, people have been saying that physics is almost wrapped up, yet if anything we seem to have reached a stalemate. "We need to rethink and radically revise our basic physical concepts before we can make the next big breakthrough in physics," he says.

Some physicists working on unified theories are well aware of this. In terms of "new ideas about quantum gravity", says Smolin, "non-locality is certainly at the core". His particular area, loop quantum gravity, does not presume that space and time behave as Einstein's relativity dictates (New Scientist, 12 August 2006, p 28). As well as allowing for spooky instantaneous signalling across space-time, those working on similar models are revisiting the fundamental aspects of quantum mechanics. A recent paper Smolin wrote with Perimeter colleague Fotini Markopoulou points out, for instance, that loop quantum gravity may conflict with common notions of entanglement (www.arxiv.org/abs/gr-qc/0702044).

Translating these studies into a deeper theory won't be easy. Physicist David Deutsch at the University of Oxford warns that even re-examining entanglement might not help us find the path to a theory of everything. According to Deutsch, we are blocked by something even more fundamental than that.

Entanglement is real, he says, but it tells us more about how information can be extracted from quantum systems than the nature of the physical universe. All the philosophical hand-wringing over entanglement is based on the "delusion" that we have a basic grasp on quantum theory, he adds. Just because we have made one leap away from the classical world doesn't mean we've reached the heart of quantum truth. "This local realism stuff is all to do with whether it is possible to have a classical world view," Deutsch says. "It's a completely pointless controversy that should have ended in the 1950s."

While his world view is clearly quantum, all Deutsch will say about a theory of everything is that it is likely to come from uniting quantum theory and relativity at a more fundamental level than current entanglement experiments allow.

This is, of course, where we are still scrabbling for clues. "The whole underlying problem, ultimately, is that we lack experimental observations in the region where quantum and gravitational effects both matter," Vedral says. "Gravity works well in its own domain and so does quantum physics." What we have to decide, he says, is whether gravity or quantum theory is more fundamental.

So does the universe exist independently of measurements? That is a question we will have to face. Maybe it is time to revisit Einstein's lost quest, if we are serious about uncovering the basic laws of the universe; the money spent on particle smashers such as the Large Hadron Collider certainly suggests we are. Perhaps we need to move quantum entanglement and the nature of reality to the centre of the quest to find a theory of everything. What was once a quirky sideshow may yet prove to be the main event.

The end of reality

Want to prove that everything you thought to be true actually isn't? Start with some common sense assumptions: that things have real, measurable properties before anyone measures them. Then try a quantum experiment.

Take the polarisation of photons. Assume that if a photon is polarised, its electric and magnetic fields oscillate in well-defined directions. Now assume that each photon's polarisation will create a predictable effect: when you put a polariser in front of the beam, you can predict what light intensity you'll get out.

These assumptions have been tested by Markus Aspelmeyer and Anton Zeilinger's team at the University of Vienna, Austria, in a landmark experiment (Nature, vol 446, p 871). The researchers examined pairs of entangled photons emitted by a crystal (see Diagram). Whereas previous experiments have looked at polarisations in one plane, the Aspelmeyer experiment happens in 3D, which allows it to rule out more options regarding the objective reality of the photons.

Underlying the set-up is a mathematical relation between the sum, difference and product of polarisation measurements, formulated by Anthony Leggett of the University of Illinois, Urbana-Champaign. This reduces to a set of results called a correlation function, which tells you how linked polarisation measurements on the entangled photons will be.

Two experimenters, called Alice and Bob (A and B), make the measurements. Alice can choose one of two angles for her polariser, and Bob can choose one of three. Both choose between planes of polarisation that are at right angles to each other. This enables you to test realism - the idea that a particle's state exists in objective reality - and non-locality, the idea that distant, instant influences are possible between particles.

The trick is to see if quantum theory plus the assumptions can predict the correlations. Leggett's formula says that a combination of the measurements should depend on the angles that Alice and Bob choose. If the predictions agree with what you observe, your assumptions are fine. If not, you have to give up your view of reality. "If you take into account all the theories in question, there will be a certain range of allowed values," says ÇCaslav Brukner, one of the Vienna researchers. "If the measured value is outside of that range, you know that nature is not described by those theories."

It turns out that assuming the photons have distinct polarisations prior to measurement stops the numbers adding up. However, the predictions match up perfectly if you assume that quantum theory is right, and that you can only describe properties statistically. The researchers take this to mean we have to abandon the idea of an objective reality. "Maybe Bohr and Heisenberg were right after all," Aspelmeyer says. "Physics doesn't tell us how nature is, it only tells us what we can say about nature."