Usually the principles of physics are derived from successful scientific theories. For instance, Lorentz invariance which can be seen as the underlying principle on which special relativity is based, was originally derived from Maxwell’s equations. As we learn more about the universe and how it works, we discover more principles. These principles serve to constrain any new theories that we try to formulate to describe that which we don’t understand yet.

It turns out that the physics principles that we have uncovered so far, don’t seem to constrain theories enough. There are still vastly different ways to formulate new theories. So we need to do something that is very dangerous. We need to guess some additional physics principles that would guide us in the formulation of such new theories. Chances are that any random guess would send us down a random path in theory space with very little chance of being the right thing. An example is string theory, where the random guess was that the fundamental objects are strings. It has kept a vast number of researchers busy for decades without success.

Instead of making a random guess, we can try to see if our existing theories don’t perhaps already give us some additional hints at what such a guiding principle should be. So, I’ll share my thoughts on this for what it is worth. I’ll start with what our current theories tell us about substructure.

The notion of a substructure can already be identified in the work of Huygens, Fresnel, etc. on interference . It revealed that light is a wave. The physical quantity that is observed is the intensity, which is always positive. However, we need to break the intensity apart into amplitudes that can have negative values to allow destructive interference. In this very simple sense, the amplitude (which is often modeled as a complex valued function) serves as a substructure for that which is observed.

It is not a big leap from interference in classical light to come to the interference in quantum systems. Here the observation is interpreted as a probability, which is also a positive quantity. In quantum mechanics, the notion of a probability is given a substructure in the form of a probability amplitude which can be negative (or complex) to allow interference phenomena.

The concept of a substructure is today perhaps mostly associated with the notion of constituent particles. We know now that the proton is not a fundamental particle, but that it has a substructure consisting of fundamental particles called quarks, bound together via the strong force. Although it is not currently considered to be the case, these quarks may also have some substructure. However, the concept of this substructure may be different from the way it appears in protons.

A new idea that is emerging is the idea that spacetime itself may have a substructure. Ever since the advent of general relativity, we know that spacetime is affective by gravity. In our current formulation of particle physics, spacetime is the backdrop on which all the particle fields perform their dance. But when gravity is added, spacetime joins the dance. It makes the formulation of fundamental theories very complicated. The difference between the particles and spacetime becomes blurred. This leads to the idea that spacetime itself may have a substructure. In this way, it combines the two different ways to look at substructure. On the one hand it may be divided into two parts, perhaps to separate chirality, much in the way intensity separates into an amplitude and its complex conjugate. On the other hand the separation of spacetime may give some substructure to the particle fields, being described in terms of fluctuations in spacetime’s substructure.

Caution is necessary here. Even if these ideas turn out to be valid, they still leave much detail unspecified. It may not be enough to regard the idea of substructure as a physics principle. The importance it to keep to the standard practice in physics: mathematics is merely used to formulate and model the physics universe. It does not tell us something new about the universe unless this is somehow already logically encoded in what we start off with.

Perhaps an example would help to explain what I mean. Einstein formulated general relativity (GR) after he figured out the equivalence principle. So everything that we can learn from GR follows as inevitable logical consequences from this principle. It tells us that the mass-energy distribution curves spacetime, but it does not tell us how this happens. In other words, the mechanism by which mass curves spacetime is not known because it is not a logical consequence of the equivalence principle.

So, the idea is to come up with a general mathematical formalism that is powerful enough to model this kind of scenario without trying to dictate the physics. Remember, quantum field theory is a formalism in terms of which different models for the dynamics in particle physics can be modeled. It does not dictate the dynamics but allow anything to be modeled. Another example is differential geometry which allows the formulation of GR but does not dictate it. Part of the reason why string theory fails is because is a mathematical formulation that also dictates the dynamics. The formulation of a quantum theory for gravity requires a flexible formalism that does not dictate the dynamics.