Has theoretical physics moved too far away from experiments? Is the field entering a crisis and, if so, what should we do about it?

Not in most areas (e.g. LHC and field theory, Quantum info/ Quantum foundations, experiments and cosmological theory is working well), string theory is the outlier. As we are getting to deeper and more subtle and intricate theories the going is slower, so we aren’t having breakthroughs all the time.

Historically we have had breakthroughs every 25 years or so, so we shouldn’t be expecting things too often.

Increasing specialization means that people are less aware of other subfields. This divide also appears to be increasing between theorists and experimentalists, so that one group is not as aware of the cutting edge and breakthroughs of the other side. This is a bigger problem in some subfields than others. Is this inevitable as theories develop and become more specialized and advanced? How can we bridge this divide? For example we go to highly specialized conferences.

Is this divide built into our education? Eg how we learn newtonian mechanics, we skip dynamics and the three body problem after doing 2-body problem we go straight to Hamiltonian / Lagrangian mechanics. Could deeper understanding of these types of problems help in forming new theories of QG.

A larger problem in the academic institution. Postdocs -> tenure tracks -> faculty. You are forced to become highly specialized to have a competitive CV if you want to get a job after a post-doc. Half your time is spent applying for funding or the next academic position, and very little time to learn new areas and increasing your breath of knowledge, let alone do your own research.

The way that physics evolved throughout the years has changed in the recent decades. In the past centuries theoretical physics has been driven by experimental evidence.  Typically, experiments offered observations that were challenging and they gave birth to theoretical descriptions whose validity was measured according to the degree of agreement of their predictions and the experimental observations. In some fields of contemporary physics, it seems that experiments are now driven by theory. This may be a fundamental problem as many of the experiments are designed to test or disprove a particular theory. This often requires to take a look to very particular things that will make experimentalists overlook some features not contemplated in the present theoretical descriptions. Fields like mathematical physics, string theory and quantum gravity, seem to be going away from experiments. However,  fields like quantum information, condensed matter and cosmology are moving closer towards them. Hence, the field is not entering into a definite crisis. What it looks like is that certain fields may enter a crisis in the sense that their validity may not be determined fully because there are no experiments engineered to test these descriptions. Nevertheless, not everything is lost. Recently there have been interesting theoretical developments that relate different fields that a priori have nothing to do with each other. For example, AdS/CFT correspondence offer a promising   path towards an experimental test that will prove or disprove string theory descriptions of reality. Similarly, there have been recent developments that have build a path between condensed matter and string theory which may bring string theory closer to an experimental test.  Although it is clear that theoretical developments are necessary for the development of physics, one has to take into account that a theory is valid as a physical description when it successfully goes through an experimental test. There is no guarantee that the field will undergo a crisis in which what theoretical physicist will do is pure math whose validity is no given y experiments. In order to avoid getting into this abyss, it is important to encourage that any physical theory most be obliged to have a connection to experiments.

The first comment came from condensed matter: even in this field there are lots of activities concentrating on models which are very exotic and may not be realized in nature.

But who decides on the direction theoretical physics should take and should one really restrict the freedom of research?

This raised the question what criteria one should apply for “good theoretical physics”. It certainly is not only to produce useful gadgets. It was pointed out that physics should however strive to explain nature and that parts of physics might be currently to far removed from that goal. This problem should be addressed in the curriculum for graduate students.

The question was raised whether string theory is neither physics nor mathematics and whether this is a danger to the field. Indeed it was said that a current theme in string theory is about recognizing patterns and that it could take a long time to make rigorous mathematics out of it.

We are moving to an era that data (experiments) is expensive. We need theories that are consistent internally and close to data.

Building theories and exploring new ideas and possibilities are important, testing theories are important too.

Nowadays, physics starts from abstract mathematical principle and connecting to tangible physical parameters becoming harder and harder.

Using the word “crisis” is not good, physics is advanced and questions are hard.

New theories, historically, were developed from existing problems or in a decade could be confirmed. This is not the case now.

In the lack of data, efficiency of theories can be used. For example sun-center model of solar system is far simpler than earth-center model.

Some fields are..

Bias towards elegance – hubris of theorists

Lead to inflation, string theory, both ran into

Are we missing some basic ideas?

Do we just have to make a lot of mistakes?

How do we know when to stop, eg susy: should we give up on it?

Funding as a bottleneck

Creating a culture which allows new directions/new ideas

Are there examples of problems where the “natural” approach didn’t more or less work? (what about QFT and renormalization)

Victim of past success

Theoretical physicists, especially in quantum gravity, don’t have almost any data on which to base their theories. However, we still need a theory of quantum gravity if we want to answer many of the fundamental questions about our universe. One possible guiding principle, in the absence of experimental data, is the elegance of the theory, but this is a subjective human notion. A perhaps more practical principle is mathematical consistency. If a theory can be proven to be the unique theory that reduces to GR at the appropriate limit, then it makes sense to assume that it must, in fact, be the correct one. This is the current paradigm among many string theorists, although there is no proof that string theory is the only possible approach. In addition, string theory incorporates the string landscape which makes it very flexible; it’s hard to pinpoint any specific discovery (or non-discovery), for example in the LHC or Planck, which will undisputedly either verify or debunk string theory, since it seems that the theory can accommodate anything.

There is indeed a crisis in physics, since in the absence of data, theorists are forced to work blindly. Even if we happened to stumble upon a true theory along the way, we won’t be able to recognize it by comparing it to experiment. Theory today is progressing much faster than experiment, unlike in the past. There could be a so-called desert all the way up to the Planck scale, and we won’t find anything new unless we find a way to probe this scale. If the LHC doesn’t see anything, it might mean that instead of building more and more powerful accelerators, what we actually need is a shift in thinking, or maybe some new way to do experiments that we haven’t thought of yet.

SUBMIT ADDITIONAL COMMENTS

Comments will be accepted until June 29.