Why can't time run backwards

(Nobel Laureate Public Lecture by Tony Leggett)

This is really one of the great unanswered questions in physics. It is our common experience that time seems to "flow" only from the past to the future. We say this because we remember only our past and we can affect only our future. This is not just a semantic issue because we can all agree that there is such a thing as the past and such a thing as the future. Why such a divide exists is not at all easy to explain.

Our current understanding of physics doesn't seem to help us much. Why? Because all the laws of physics we know so far are totally symmetric with respect to the flow of time. If you let time run backwards in our equations, equally valid systems result. This is true for Newton's physics as it is for relativity or quantum mechanics.

However, it is quite easy for us to tell whether a certain roll of film is being run backwards, because some events can only believably happen in one direction. You never see broken eggs spontaneously form or fallen dominoes that suddenly stand up.  So how is it that we are able to recognize a particular direction of time when physical laws don't compel us to? (It is not physically impossible for our broken egg to re-form, for example)

Prof. Leggett mentions five different ways in which we can describe the flow of time: thermodynamic, psychological, biological, electromagnetic, and cosmological. The "broken- egg reformation" process is eliminated by the thermodynamic flow, which says that physical systems tend to go to greater disorder. In physics textbooks this is usually quoted as the second law of thermodynamics. (More precisely, the measure of disorder  of a system, which we call entropy, always increases.)  The psychological flow of time characterizes the fact that we remember only the past and influence only the future. This is somewhat connected to our consciousness, and because this is not within the realm of the physical world (yet) we do not have any proper physical theories for it, only philosophical ones. The biological flow describes how living things are born, grow old, and then die. As far as we know, no organism becomes younger. The electromagnetic flow of time says that electromagnetic radiation always moves outwards, away from a source, which explains why we never see light going into a bulb. (Yes, materials can absorb radiation but they do so passively. They don't naturally suck light in) And finally, the cosmological flow tells us that the universe is expanding, beginning with the singularity of the Big Bang up to the large expanse it is today.

In a certain sense, one can think of these different flows of time as forming a hierarchy, the previous level somewhat a consequence of the next level. At the lowest level we have the thermodynamic phenomenon of ever-increasing disorder. The second law of thermodynamics gives us a reference for what is in the past or in the future. We recognize such a reference because we are psychological wired to do so. Why are do we perceive the past or future? Maybe it's a consequence of our biological development, which dictates the sequence in which we grow and grow old. The existence of life is ultimately due to the sun, which would not be able to provide the energy living things require if the sun didn't naturally emit light away from itself. The sun and all the planets and all the galaxies are here because of how the universe evolved from the Big Bang. 

Of course even Leggett admits these connections are a bit tenuous but the generally idea is that maybe the flow of time can ultimately be explained by cosmological arguments. Now if we turn to cosmologists today we find that we still can't give us anything definite. The simple reason is there is no compelling evidence why at the Big Bang, the universe should have been in the condition of perfect order. We have speculations here and there but even the simple issue of why  universal fundamental constants such as the electron charge or mass of the proton have the value they have. It seems the best explanation so far we have is the so-called anthropic principle. Crudely speaking, this says that the universe is the way it is because it wasn't then we wouldn't even be here to ask why. Well, so much for a scientific reason but this arguably nonscientific idea has been making quite a revival recently.

So what other explanations do we have? It seems the only other plausible alternative is that all times--past, present, future--exist simultaneously, in what we call the block-universe. The analogy here is a loaf of bread where each slice is a particular moment in time. The flow of time is nothing but an illusion of our limited conscious experience. The present is like a special slice of the universe-bread that our minds prefers to look at. (We can then remember the past because our mind can store memories of the slice we have previously chosen). This requires that the way we perceive our environment involves a mental process than unavoidably puts events in a particular time sequence. This is really more philosophy than science but at least it would be a consistent extrapolation of Einstein's relativity. Many hate this model of the universe because it rips away free will, since things that ever were or ever will be have always been there all the time.  

It's rather unsatisfying to bring up an interesting question and leave it unresolved but there is where we are still at. Leggett concluded his talk by mentioning an experiment that illustrates the bizarre phenomenon of quantum entanglement. In simplest terms, it says that there are some systems of two or more distinctly (spatially) separate objects can only be described together.

A popular analogy for the entanglement experiment uses a pair of socks. Put each sock in a different box and take the two boxes to a very long distance from each other, say bring them to different galaxies, When you look at one box and see a left sock, you know the other box has a right sock. That seems simple enough but the situation with quantum objects such as electrons or photons is way more bizarre. Take, for example, the electron spin. Imagine an experiment where you have a source that emits pairs of electrons that will have oppositely-oriented spins. (An analogous experiment with photons has been performed numerous times but for illustration purposes, the concept of electron spin is slightly less complicated.) So far you can just take the sock-analogy to understand that when you measure spin-up in one electron, the other must be spin-down. Until you wonder which direction is up? The answer is, simple, you can measure along any axis. If one electron is spin-up in your chosen axis, the other is spin-down in that axis. Take a different axis and the same thing happens. If you use the sock-analogy, it's as if the sock can have different ways of being a left or a right sock. Clearly our classical analogy can not capture this purely quantum feature of things like electrons.

So where does this fit in? To explain the result of entanglement experiments, at least one of three previously accepted truths about the nature of physical systems have to be abandoned. These three are objectivity, locality, and causality. Objectivity is that a physical system has carries the values of measured quantities before you even measure them. Locality means things happening in one place can not instantly affect what's happening in some farther away place. (This embodies the speed of light limit.) Causality refers to cause-and-effect time ordering of events. The former two have been extensively explored and there are enough arguments around to believe physical laws are either not objective or nonlocal (or in some cases both). But anti-causality arguments have been few. Perhaps our perception of the flow of time has made us unwittingly biased in this regard. 

What then is a good non-causal theory? Leggett describes one option with an analogy to iron. Iron is ferromagnetic because of so-called magnetic domains. These domains define regions in  iron where electron spins point in a particular direction, giving iron magnetic properties in the right environment. However, it wouldn't be surprising that even in the best conditions a handful of the spins will point in the wrong direction. Leggett mentions a possible similar effect with time. Maybe in some tiny regions of space, time flows backward. Such a situation can help explain entanglement by allowing certain effects which would look to us as if the effect came before the cause. This is a radical departure from conventional lines of thinking and so it has attracted none but a few cavalier researchers.

Revolutions in science have come at the expense of ideas previously thought to be common sense: that the earth is flat, that planets move around the earth, that time flows at the same rate for everyone. Maybe we need another revolution to explain why time only flows forward. Perhaps it goes backward or doesn't move at all? It sure would help if we know what consciousness is all about. Then we can at least figure out once and for all if the direction of time is real or a mere by-product of our minds.