An occasional TRF guest blogger Gordon Kane of Michigan has posted an edited, written version of his talk at the Munich workshop on the "testability of string theory etc." that was organized by Richard Dawid:
The paper argues that the qualitative character of testability of string theory is exactly the same as it has been for all respectable theories in the history of physics. And it uses the predictions for particle physics from M-theory on \(G_2\) holonomy manifolds as a detailed example of all these claims. It's impressive how this framework has been developed to say something about all aspects of physics – including the nature of dark matter, patterns of superpartner masses, solution to the CP-problem, and many others.
Gordon Kane admits that like others, this scenario has no solution for the cosmological constant problem – why the CC is so small – so he assumes that this problem is solved by something that is "orthogonal" to the mechanisms deciding about the rest of particle physics.
He has written down lots of details about the history of M-theory and its \(G_2\) holonomy compactifications and the detailed predictions that follow from it – winos are above \(600\GeV\), binos are LSPs, gauginos are light, gravitons and light moduli are close to \(50\TeV\), squarks and sleptons just a little bit lighter, gluinos at \(1.5\TeV\) plus minus 10-15 percent, with cross sections somewhat lower than what is assumed in most of literature, and so on. Before the Higgs discovery, the Higgs mass was calculated as \(126.4\GeV\); unfortunately, Gordon couldn't quantify the error margin.
The setup allows you to meaningfully incorporate some cosmological epochs, inflation, non-thermal evolution in cosmology, dark matter, and other things.
It's being argued that string theory indeed seems to be the only game in town that shows the promise to unify gravity with other required mechanisms and interactions and particles of matter and there exist vacua that do that. They seem to do everything that is needed. No non-stringy attempt has ever come anywhere close to these achievements.
Even if the number of vacua is large, it doesn't mean in any way that string theory is untestable. What is to be compared with the experiments is a particular compactification to 4 dimensions, Kane stresses, but this disclaimer is in no way different than it was in the past. People always needed specific models to make the theory testable. His oldest example is Newton's \(F=ma\) that is very deep but it's also testable only once you enrich it by additional formulae for the force (e.g. the inverse square gravitational force).
People who talk about "testing of string theory" in general – or even testing string/M-theory as a 10- or 11-dimensional theory – generally don't know what they are talking about. And the large number of solutions is not a problem. After all, I would add, the goal of science was always to locate the viable or right theory as a tiny subset of the a priori possible but actually wrong (and to-be-excluded) competing hypotheses. String theory is no different – it just organizes lots of the "wrong theories" as "physically irrelevant, other vacua/solutions of the same correct equations". But you still have to pick the right vacuum before you test things.
It may be useful to add that Kane's favorite M-theory on \(G_2\) holonomy manifolds is something like "1/5" of the string/M-theoretical phenomenology – even though he and some of his collaborators would argue that this picture is so good that it covers more than "50% of the potential" to connect string theory with experiments. I could personally place (at least sometimes) heterotic \(E_8\times E_8\) models, their strongly coupled Hořava-Witten limits, Vafa's F-theory models with localized standard model in the extra dimensions, and some other braneworlds "above" the \(G_2\) holonomy manifolds. But sometimes, I do feel that that the \(G_2\) scenario is the best one. It starts with the maximum number of non-infinitesimal spacetime dimensions and compactifies the excessive ones in the most "democratic" way, and this could be a virtue making these scenarios natural.
Off-topic: Dutch physicist Andreas Wahl shot himself with a gun. The only problem was that he was in the water. The principle he demonstrated, namely the friction in water, is a bit less deep than M-theory on 7-dimensional manifold of exceptional holonomy, however, despite the label "Viten" in the corner.
Sometimes, string theorists may find the right vacuum as well as the reasons why this vacuum had to be the right one – some cosmological vacuum selection mechanism. But even if the latter step (which would really bring us a big step closer to the "face of God") never materializes, string theory is still exactly as testable and scientific as older theories. Unrealized vacua may be considered on par with "wrong candidate theories" and the goal is simply to find the right one – the right vacuum of string theory. Old, recent, as well as future experimental discoveries will surely influence this search – just like they always have. The number of "wrong competing theories" or "wrong vacua" may be very large but it has always been very large. String theory doesn't make these matters any worse or "less scientific". I am happy that not everyone has lost his mind. Gordon Kane clearly agrees with me that this whole discussion about "string theory as an activity abandoning science" is wrong from the very beginning.
Aside from the visibility of many totally incompetent people who decided to speak on the issue that hugely transcends their intellectual capabilities, one less outrageous but still unfortunate reason why these fallacious statements about string theory were able to spread this much is, as Gordon also mentions, that most of string theorists are formal theorists, not particle physics phenomenologists. Within the broad string theory community, the string phenomenologists represent a very small fraction. And they also give a small fraction of the talks at recent pan-string-theory conferences. I do think that it's unfortunate. And it's even more counterproductive if and when the non-phenomenological string theorists sometimes talk as if they were assuming that string phenomenology doesn't exist.
It exists and it has achieved and still is achieving amazing things. Even the first sketch of the "real world of particle physics within string theory" that was outlined in the mid 1980s was stunning. But there's been lots of progress – even if the progress hasn't been "definitive" so far.
Let me tell you that when I was publishing papers in journals, they were papers about formal string theory. It was natural for me to focus on this subdiscipline – especially after I made some important enough contributions to Matrix theory etc. I am and I was greatly intrigued by the inner structure of string theory, by the conceptual questions of quantum gravity in string theory, and so on.
But it may be fair to say that even at the moment when Tom Banks invited me to the U.S., I may have been even more impressed by string phenomenology than by the formal theory. I think it's fascinating that the models work so well, I believe that there is no reason why one of the string vacua shouldn't be "perfect" and why we shouldn't be able to find the correct one increasingly accurately (or "perfectly"), and I would love to be one who demonstrates that this perfectly correct vacuum is included in string theory. Around 1995, I was particularly intrigued by the heterotic string models, especially those in the free-fermionic construction, and so on, but I don't want to say too many things that could be too idiosyncratic.
The kibitzers surrounding theoretical physics – and it's indeed a vastly more accurate phrase to describe the likes of Lee Smolin than "physicists" – frequently offer their conspiracy theories about the suppression of some ingenious researchers and their research directions. I hate these victimist clichés (especially because they're just plain lies in almost all cases that we actually hear about) and those who use them to serve their own, personal, egotist interests. And you probably don't remember me making similar statements, at least not often.
But let me tell you something. The whole string phenomenology subcommunity is one of the actual examples of these "real-world Cinderellas". One of the reasons why I have never spent too much time with writing papers about string phenomenology was indeed my feeling that people who do this amazing work and the work itself are heavily underappreciated.
Off-topic: Zika virus was said to have exploded. Flu-like symptoms, rashes, muscle weakness – and especially the shrunken heads (and retardation) of newborns – are the symptoms. No fatalities are occurring. But due to the dependence on mosquitoes, the virus is predicted to spread to Florida and the "hot wet third world".
Dozens of extremely active and brave people keep on working on string phenomenology, anyway. Some of them are partly phenomenologists, partly formal theorists etc. They are developing models that are qualitatively analogous to the models within effective quantum field theory that are being developed by more ordinary "model builders". The string phenomenologists could probably work on any of these "normal model building" activities. But they work on string phenomenology because their skills and knowledge are deeper than those of the "ordinary model builders". This "elite within elite" status is the only true reason why string phenomenologists represent a small percentage both among string theorists and among the model builders. Feel free to check that when I talk about string phenomenologists in this way, it's an opinion coming from an impartial person.
Within quantum field theory, one may discuss things like supersymmetry, supersymmetry breaking, dark matter, patterns in fermionic masses, inflation etc. But these individual aspects of beyond-the-Standard-Model physics are "mostly" independent of each other. String/M-theory basically unifies all of those. It forces you to discuss all these things at the same time. It requires you to know the effective field theory methods because much of the experimentally accessible physics following from string theory may be described by the approximation known as effective field theory. But it also requires you to know something more – string theory and the new relationships between physical phenomena (and new "geometrizations" of the old phenomena) that string theory has brought us and is still bringing us.
Sometimes it's being said that string phenomenologists don't use all the insights from "formal string theory" too much. I have the same feeling. But you must understand that it's not really the primary goal of string or particle physics phenomenology to maximally utilize all cute and seemingly fundamental ideas that have been found. The goal is to describe phenomena in particle physics.
If one looks at the things fairly but from any perspective, it's damn obvious that string phenomenology is at least as great and as scientific as other approaches to similar questions (or simpler questions). Even though I have never made living by full-fledged string phenomenology, I find it obvious that all the individuals who openly contaminate the public discourse with their "criticisms" of string phenomenology as a whole are piles of worthless pseudointellectual junk.