PROGRESS in physics often comes about by discarding the bias that humans are at the centre of everything, the most obvious example being the repositioning of our planet from the centre of the universe. But might there still be such anthropocentrism lurking in our best models of reality? Experience and instinct make it natural to have such biases; the difficulty is recognising them and finding a more objective vantage point from which to evaluate them. And there is one particular bias that has resisted this evaluation for far too long.
We have grown accustomed to the idea that there is no centre of the universe. The space to our left is no different from the space to our right. But our instincts balk when this comparison shifts from space to time. Our immediate future seems somehow different to our immediate past. We can fight these instincts with careful logic, realising there is nothing special about “now”, because every time we have ever experienced seemed like “now” at the time.
The importance of putting the past and future on an equal footing is particularly clear in Einstein’s general theory of relativity. But these arguments still seem instinctively wrong. After all, we don’t know the future, and we can’t act to change the past. Our human condition has given rise to an anthropocentric bias when it comes to time.
Advertisement
Many physicists will tell you that we have purged this bias along with the others. All of the microscopic laws of physics are time-symmetric, and general relativity represents space and time together as a four-dimensional “space-time block”. Nevertheless, there is a related bias still hiding in modern physics, and it has been there since Newton – an instinct so natural that it’s hard to even notice. This is the assumption that the universe solves problems in the same way that we do – that the universe works like a computer.
“We can’t help thinking the universe solves problems in the same way we do”
Humans are always trying to compute the future. Given that all of our experience is of the past, there is really only one way we can do this: take information about the past, manipulate it using some rules, and then use the result to forecast the future. Mechanical computers process data in the same fashion.
So it is not surprising that when Newton first laid out how to do physics, he framed it with the same computational “schema”: 1) Map present reality onto some mathematical state; 2) Input that state into some dynamical equation; 3) Map the equation’s output back onto a future reality. As we know, this process works quite well – many predictions made in this manner actually come to pass.
But even though we have moved well beyond Newtonian physics, we haven’t yet moved beyond the Newtonian schema. The universe, we almost can’t help but imagine, is some cosmic computer that generates the future from the past via some master “software” (the laws of physics) and some special initial input (the big bang). Note that this is very different from the claim that the universe is a computer simulation.
After 400 years of solving physics problems in this way, it’s only natural that we have incorporated this schema into our world view. This is the case even when it backs us into an impossible corner, as when we try to use it to explain quantum phenomena.
The key point is that Newton’s schema naturally arose from our human experience of time, and it is arguably out of sync with what we have discovered since. The notion of the cosmic computer is itself an anthropocentric bias. This doesn’t mean that it’s wrong, but it does mean that it should be evaluated from an unbiased vantage point. This evaluation has not yet occurred.
That could soon change. It may be surprising to hear that there is already a wildly different alternative to Newton’s 3-step schema. The “Lagrangian” approach, largely laid out in 1788 by the mathematician , turns out to be of crucial importance to both relativity and quantum theory. A simple example of a Lagrangian-style approach is , which describes how light rays travel.
Fermat supposed that when a ray of light travelled from X to Y it would always take the quickest path of all paths available. In a uniform material this is a straight line, but it is a different matter for rays that pass from air to water, where light travels more slowly. Just as a smart lifeguard will run a crooked path to rescue a drowning swimmer – diagonally for the fast portion on the beach, and then less-diagonally for the slow portion through the water – light rays do the same.
Fermat’s principle makes it much easier to explain certain phenomena. Take mirages: light travels faster just above a hot surface, so the light from the sky bends in such a way that it appears to be coming from the ground. But stories like this don’t follow the Newtonian schema – they don’t require dynamical equations, just two endpoints, X and Y, and a process to determine the fastest of all the possible paths between them. Crucially, this style of physics is not as blind to the future as we are ourselves.
Remarkably, these principles can be extended to all of classical physics, and are especially valuable to quantum field theory. But despite using the Lagrangian approach, physicists tend to view it as a mathematical trick rather than as an alternative framework for how our universe might really work. This attitude may be an indication of our biases, where our cosmic computer assumption is so deeply ingrained that we don’t even realise we are making it.
This is not irrelevant metaphysics. Our assumptions frame our best models in physics, and for quantum physics in particular, the models have deep problems. For example, quantum predictions are fundamentally uncertain, and Newton’s schema doesn’t work so nicely with uncertain inputs or uncertain equations. So modern quantum theory effectively removes this initial uncertainty, postponing it until the final output step when it can no longer be ignored. Adhering to the Newtonian schema then leads to a ridiculously impossible “collapse”, when all the built-up uncertainty suddenly emerges into reality.
Contrast this with the mirage example, where the uncertainty of the actual path between X and Y was smoothly spread out and elegantly solved by Lagrange’s methods. Nevertheless, if such stories can’t be translated into the Newtonian schema, no one seems to take them seriously as a template for how our universe might operate.
Well, almost no one. I am proposing a modification of the most Lagrangian-friendly formulation of quantum theory, such that the mathematics could be taken literally (). If it works, it could provide an underlying realistic explanation of quantum phenomena, but without any corresponding Newtonian schema version. Confronting our biases is a prerequisite for even considering this style of explanation. While it’s far too soon to see whether this modification will be successful, it is about time we tried setting aside our anthropocentric notion of the cosmic computer and at least see what the alternatives might look like.
Who knows? Quantum theory may have more in common with mirages than anyone could have guessed, and our universe may not be a computer after all.
Profile
is a quantum physicist at San José State University in California. This article is based on his essay ““, which won third prize in the 2012 essay contest
This article appeared in print under the headline “Against the cosmic computer”
