Huw Price, Director of Time

The British astrophysicist Sir Arthur Eddington might not be a household name currently, but in the early half of the twentieth century, he was one of the most well-known science popularisers – imagine Carl Sagan in a three piece suit.

In 1919, Eddington offered some of the first proof of Einstein's general theory of relativity by confirming that light bends around the Sun, during a total solar eclipse. But relativity brought with it a revolutionary new conception of time that Eddington struggled to buy into. Time, it says, is just another dimension to be added to our three-dimensional view of the world, building up a four-dimensional block universe that maps out all locations at all times. In this scenario, the past, present, and future all exist on an equal footing, while our everyday perceptions of a distinguished now, time's direction and its flow, all seem to vanish from view.

These concerns caused the astrophysicist to raise what philosopher and FQXi member Huw Price calls "Eddington's challenge" – asking believers in the block universe to come up with a convincing description of the universe in which events appear to unfold in a way that defies time's arrow. Price embraces Eddington's challenge. He has developed a model of retrocausality, in which causes from the future can travel back in time and affect the past, that could resolve some paradoxes at the heart of quantum mechanics. It could even help bring the spooky theory into line with relativity – an achievement that has eluded physicists for decades.

Price certainly has the credentials to talk about time. He is currently the Director of the Centre for Time at the University of Sydney, in Australia, and in September 2012, he will become the Bertrand Russell Professor of Philosophy at Cambridge University, UK. For him, Eddington's concerns are just as important today as they were when the astrophysicist first set them down. In 1927, in his popular book, The Nature of the Physical World, Eddington coined the term, "the arrow of time," noting that over the course of history, the universe has on average been moving toward a state of greater entropy, or disorder, which defines the direction of this arrow.

Price describes Eddington’s Challenge in this FQXi video:

Eddington thought this required a fundamental, one-way process in time itself – "becoming," as he called it – that is completely missing in the block universe picture. Without this, there would be no objective sense in which the universe was running from past to future, rather than future to past. Hence his challenge to supporters of the block universe view: "He says that if you don’t believe in becoming, you should be happy to run your physics in reverse – to describe planets 'unforming' into dust, for example, which seems ridiculous," says Price. Although Price agrees with Eddington about the formation of planets, and other systems in which differences of entropy are important, he thinks that the astrophysicist was wrong about microphysics. At that level, says Price, we should take more seriously the idea that physics doesn’t have a direction.

You Are Here (and Now?)

Although we think of time as split between the past and future by a special moment marked "now," Price argues, the universe does not. Using Eddington's metaphor of a four-dimensional map, he argues that we could indeed put a small, red X on our particular moment in time just as cartographers often like to put small, red X's on maps to show where you are standing. As Price points out, the map is still complete without a small, red X spatially, so why do we think that we need one temporally? Plus, as Price notes, though on a spatial map the "You Are Here" X helps us locate what is near and far, up and down, those are all a matter of perspective, which changes wildly depending on our location and the scale we are interested in. For instance, we might say that our environment is oxygen rich, but that only holds locally, when you consider the scale of the cosmos. And notions of up and down on Earth do not hold true at different scales; what does "up" or "down" even mean for a galaxy or for a proton? asks Price.

Extending this argument further, Price adds, "Should we assume that particles in physics know about the past but not the future? That turns out to be a crucial assumption in some fundamental results important to understanding what quantum mechanics tells us about the world. But is it an assumption we are entitled to make?"

In particular, Price is referring to the issue of entanglement – the process by which two or more quantum particles can become intertwined in the lab, such that measurements carried out on one seemingly instantaneously affect the properties of their partners. In the 1960s, Irish physicist John Bell devised a strategy for working out whether these correlations could be explained by something like plain old classical physics – for instance, perhaps the blueprints for how the particles should respond to future measurements were set at the point they were entangled, encoded into so-called "hidden variables" that we simply do not know about – or whether something more mysterious was connecting these objects. Since then, experiments have tested "Bell's theorem" and confirmed that the level of correlations between entangled particles defy simple hidden-variable explanations.

But perhaps things are more complicated, says Price: "Bell's theorem explicitly assumes what amounts to a No Retrocausality condition. It assumes that if there are any hidden variables, then their values must be independent of what is going to happen to the particles in the future, including things we experimenters can control like measurement settings." Price points out that this is an asymmetric, or one-way assumption: everyone takes for granted that hidden variables can depend on measurement settings in the past. But, inspired by the block universe and Eddington’s challenge, Price believes there may be room for retrocauality – the idea that the future can impact the past just as the past impacts the future – in quantum mechanics.

Retro Retrocausality

Retrocausality is not a new idea among physicists. The French physicist Olivier Costa de Beauregard suggested it as early as the 1950s as a way to solve some of the puzzles thrown down by quantum mechanics. The notion first caught Price's imagination in 1977 when he was a graduate student at Oxford University, UK. He was listening to a lecturer who stated that "of course" the two sides at play in quantum entanglement experiments could not be affected at the present by any measurements in their future. To Price, already a fan of the block universe, that seemed like an enormous assumption to make with no apparent justification.

Price adds that most physicists concur that fundamental physics is time-symmetric, in quantum mechanics as in classical physics. "But then they just go on assuming that quantum particles know about past measurements but not about future measurements!" he says.

To see how retrocausality can help resolve some quantum paradoxes, take for instance the aspect that bothered Einstein: "spooky action at a distance." Even if two entangled particles are shot millions of light years apart into space, a measurement of one still seems to have a subtle influence on the results of the measurements on the other, in a way that is very hard to reconcile with Einstein's theory of special relativity. Retrocausality allows a way out of this conundrum, however.

"Retrocausality gives a means to decompose that spooky non-local action into two local actions, the first one backwards in time," said Price. In other words, according to Price, the properties of the entangled particles are correlated backwards in time, from the point in spacetime where the measurement is made, to the point in spacetime where they become entangled, allowing the correct hidden variables to be encoded into the pair of particles. "In a case with retrocausality, the measurement that one observer makes affects her particle 'backwards' to the point where they are together," says Price. "There's a zigzag path connecting the two particles through time and space and no need for instantaneous action at a distance."

The double slit experiment: The top picture shows the interference pattern created by waves passing though the slits, the middle picture shows what you'd expect to see when particles are fired through the slits, and the bottom picture shows what actually happens when you fire particles such as electrons through the slits: you get the interference pattern you expect from waves, but the electrons are registered as arriving as particles.

Ken Wharton, a professor in the department of physics and astronomy at San Jose State University and a collaborator with Price, has also done significant work on this topic. In particular, he has looked at the implications of retrocausality on interpreting another paradoxical result seen in the double-slit experiment. In this set-up, physicists shoot light from a laser toward to a screen. Between the laser and the screen is a thin wall with two tiny slits in it, which diffracts the light into two separate waves that interfere with each other, creating a distinctive interference pattern on the screen. From this, you would happily conclude that light is a wave.

So far, so good. But an odd thing happens if you change the set-up slightly: Take away the screen and replace it with two photon detectors, one lined up with each slit. Now, when you fire the laser, the detectors register the photons passing through the slits like bullets, with no apparent interference. From the second experiment, you would think that light is made up of particles, not waves.

The puzzle is that the laser light seems to know ahead of time whether it is to meet a screen or a pair of photon detectors on the other side of the wall with the slits, and it appears to modify its behavior on hitting the wall—acting as either a wave or as a particle—accordingly, based on this future knowledge. But if retrocausality is true, then it is perfectly possible for the experimenter's choice of how to measure the light later in the experiment to cause a change in the light's behaviour earlier on, says Wharton. "The double slit experiment baffles people, but in a retrocausal model it makes sense, and is more natural," he says.

Despite potentially removing some quantum puzzles, retrocausality is so counter-intuitive, physicists may still be reluctant to subscribe to it. But there are other reasons to take that leap, says Wharton. Retrocausality could actually help solve one of the most troublesome problems in current physics: the unification of quantum mechanics with relativity.

Wharton has watched physicists try to quantize gravity – coming up with a description in which the force of gravity is treated in a similar manner to the other fundamental forces – but this is difficult because quantum theory has thrown out traditional spacetime locality (the notion that an object can only be affected by its immediate surroundings, with influences constrained by the speed of light). Instead, he argues that we should leave relativity alone and try to fit quantum mechanics into spacetime. And, retrocausality would be the most elegant way to make a spacetime representation of what is happening at a quantum level.

Human intuition may be getting in the way of solving the classic Einstein-Podolsky-Rosen paradox, through which Einstein explicitly expressed the conundrum behind entanglement and concluded that either quantum mechanics was missing some hidden variables or that it was incompatible with special relativity because the phenomenon seems to allow communication faster than the speed of light. Many physicists, Wharton says, are prepared to give up the principle of locality to solve this dilemma, but Wharton believes giving up our one-way notion of causality would actually be better. "If experiments say we had to give up one most think that it would be locality not causality, but that’s just instinct," says Wharton.

Matt Leifer, an independent quantum physicist and a 2006 FQXi awardee at University College London, believes that retrocausality is a solution that has too often been ignored. "It is certainly worth looking at," says Leifer. "I'm not certain it's the solution, but it is an option and too many people dismiss it out of hand."

The most obvious problem people have with retrocausality – chiming with Eddington's concerns about the block universe – is that it contradicts how we experience the world. But Price hopes that scientists will look beyond just their physical senses to find the true nature of the universe. "The naïve view that people have is that the universe is a one way track – things 'happen' in one direction, from the past to the future," says Price. "There is an asymmetry for creatures like us. But, how fundamental do you think that is? What if this is just a reflection of our viewpoint and not an objective feature of reality?"

Free Will

But if retrocausality is true, then maybe scientists do not have any choice about what they believe; the idea that free will cannot exist in a universe in which the future is prewritten and affects the past, is another pressing criticism. In fact, when a young Price wrote Bell a letter in 1988 proposing his notion of retrocausality, Bell replied saying that when he thought of retrocausality he lapsed "quickly into fatalism" (see below).

John Bell's letter to Huw Price about the retrocausal blow to free will.

Price, trained in philosophy, is less troubled. "Would it be the past controlling us or us controlling the past? That's a subtle question, but one best left to philosophers, who have been arguing about free will for 2000 years," says Price. "It's going to feel just the same, either way!”

As Price adds, "This might be one of those rare occasions where progress in physics depends on a bit of help from philosophy – in this case, help in seeing that the worries about free will that discouraged Bell and others from exploring this path could be completely groundless."

"In a world where past systems depend on the future strange things result," admits Wharton. "One way to keep from making paradoxes and having things get really strange is if you are incapable of knowing certain features of the present that depend on the future. So, if there is an element of reality that we cannot know, and that element is caused by the future, then all the paradoxes go away while still being retrocausal." For Wharton, this paradox-preventing element already exists as the Heisenberg Uncertainty principle, which states that you cannot simultaneously know certain pairs of particle properties, for instance its position and its momentum.

Leifer adds that many confuse retrocausality with super determinism, the notion that everything could have been figured out at the Big Bang if only someone had been around to write it down. In super determinism, a measurement determined through some random event, say a coin flip, would be known in advance and predetermined. That's not how retrocausality works though, argues Leifer. "In retrocausality, the coin is independent just correlated with the measurement device, that information from the measurement device is what propagates back within time," says Leifer. "It's not the case that there is a conspiracy at the beginning to correlate; it gets correlated through a common cause, but that cause is in the future and we are used to thinking of them in the past."

Still, it will take more solid results to persuade the physics mainstream that the future can affect the past. Wharton is on the case: His current research program focuses on finding a local, retrocausal toy model that could match key features of quantum mechanics. He is finding it by looking at a class of physics theories using a mathematical (Lagrangian) strategy commonly used in analytical mechanics to constrain or define certain quantities at the end of the problem. It's a trick used to fix future conditions, but one which draws analogies to retrocausality and which Wharton hopes to use to find answers to the mysteries of quantum theory.

Meanwhile, Price's move to Cambridge, England, later this year will bring about a chance to continue his work in a new setting—the place where Eddington spent his academic life. In The Nature of the Physical World Eddington wrote:

"Unless we have been altogether misreading the significance of the world outside us…we must regard the feeling of 'becoming' as…a true mental insight into the physical condition which determines it."

In accepting Eddington's challenge, Price is turning that logic on its head and asking whether we have been misreading the significance of the world: "It might be that physicists are still making mistakes about what is actually in the world and what comes from the human perspective," says Price. "We need to shake things up a bit, and taking Eddington's challenge seriously is a good way to do that."

About this article

This article first appeared on the FQXi community website. FQXi are our partners in our Stuff happens: the physics of events project, within which we explored the nature of time. Click here to find out about the different views of time in cosmology.