Can a physicist create a model of a universe that is not our own? originally appeared on Quora: the place to gain and share knowledge, empowering people to learn from others and better understand the world.
Answer by Paul Mainwood, Degrees in Physics and Philosophy, Doctorate in Philosophy of Physics, on Quora:
Can a physicist create a model of a universe that is not our own? We think so.
Physicists work with two types of states of affairs, both of which must be specified to model a physical system:
- Physical laws: These are the mathematical laws that tell us how arrangements of matter and energy evolve in space and time. (Examples: Newton’s law of gravitation, Schrodinger’s equation.)
- Initial/boundary conditions: The state of matter and energy at one particular time: most relevantly, at the start of the universe. (Examples: a pendulum bob being released to go at some given displacement from equilibrium, a hot, dense early universe.)
To construct models with different universes, we could change either the laws or the boundary conditions.
We can have a long set of arguments about whether physical laws are logically necessary or contingent. Most people think they are contingent — that is, it is logically possible for them to have been otherwise — and it certainly seems that we can just set them up differently (let’s just imagine that Newtonian physics is completely valid, or put an r3 in the law of gravitation).
But we might doubt this. As we continue to find deep connections between these physical laws and the basic symmetries of space and time, we might begin to worry. How deep do these links go? Maybe our laws of physics will eventually turn out to be necessary, and apparently consistent alternative laws will be shown to be subtly contradictory?
So, for the sake of argument, let us assume that we don’t know whether laws are necessary or contingent. Let’s play it safe — in order to be absolutely sure that any new imagined universe is logically consistent, we should keep the physical laws the same as in the actual one.
Still, in this situation, we can alter the boundary conditions. And physicists alter boundary conditions the whole time when they run simulations. They code up a set of physical laws (say, Einstein’s field equations of General Relativity) and set up their own boundary conditions (say, a large, slowly spinning cloud of dust) to see what happens next (spoiler: look around you).
So, to create a model of the universe that is not our own, but still guaranteed to be logically consistent, all we have to do is to keep the physical laws precisely the same as they are in the actual world, but change some boundary conditions. And physicists do this all the time. This seems to demonstrate that the answer to the question is a pretty clear yes.
But, let’s end with a word of caution. We have in the past been wrong about categorizing observed regularities into that broad set of physical laws and boundary conditions. For example, the second law of thermodynamics initially looked like a physical law, but as we have learned more about statistical mechanics, it now looks as though it is a very well-disguised expression of a boundary condition (the boundary condition being that the early universe was in a very low entropy state).
Could we be wrong in the other direction too? Could it turn out that what looks like boundary conditions right now, are in fact well-disguised consequences of physical laws? Certainly, this might appear in the feature of field theories, where the distinction isn’t quite so clear (there are some field configurations that are just plain banned by the dynamics). Then, let’s follow a few more suppositions. Suppose that all boundary conditions turned out to be really physical laws in disguise, and then suppose further that all these physical laws also turned out to be necessary, then (finally) the answer to this question would fall out as a no.
So that’s the word of caution. But it’s worth emphasizing that it is probably over-caution: it relies on a chain of very outlandish assumptions. And with our state of understanding at the moment, it’s very hard to see how the chain could all hold together. Simulations with non-actual-but-possible boundary conditions appear perfectly able to give consistent models of universes (and small bits of universes) that are not our own.