Friday, November 01, 2019

Artificial Intelligence, Natural Stupidity, and Theoretical Particle Physics

For decades ago I encountered the concept pair Artificial Intelligence (AI) - Natural Stupidity (NS). Much later I realized that the notion of NS allows to get overall view about the developments of theoretical physics during the last four decades. One can imagine also other letter combinations - Natural Intelligence (NI) and Artificial Stupidity (AS). I will say few words also about these.

Four decades ago standard unifying electroweak and strong interactions in terms of standard model had emerged. Standard model was so called non-Abelian gauge theory generalizing electrodynamics - first to a model of electroweak interactions by adding weak gauge fields and then to a model of also strong interactions by adding gluons. The model was rather successful but had also problems.

  1. Higgs mechanism as a tool to make particles massive was a central aspect of the model. It allowed reproduction of fermion masses but did not predict them. With the advent of NS approach this was soon forgotten: who would dare to be skeptic about God particle?

  2. One could describe strong interactions at high collisions energies using the notion of asymptotic freedom but could not say much about hadrons themselves since the perturbation series failed to converge for large values of the gauge coupling. Also color confinement binding quarks and gluons to hadrons remained poorly understood as also the spin of proton.

    QCD predicts so called strong CP violation due to the instantons - a non-perturbative effect predicted by non-Abelian gauge theories. NS solution to any problem is to invent a new particle and strong CP problem led to the introduction of axion which has not been however observed. These problems suggest that there might be something wrong even with the standard model.

    This feeling becomes stronger when one realizes that there is no idea why standard model symmetries are what they are and why the couplings are so strange looking and violate chiral symmetry in the manner they do. One could not of course predict the couplings. With the advent of NS these problems were forgotten since the attention was directed to much more ambitious goals.

The period of NS

Standard model can be said to still represent the period of NI in theoretical particle physics and theoretical physics in general. The emergence of GUTs meant the beginning of NS era.

  1. New powerful tools of NS led first to the proposal of GUTs as a purely formal recipe generalizing standard model. The highly demanding, painfully slow, and lonely intellectual activity that was needed to create Relativities and Quantum Theory was replaced by a mechanical application of the recipes of Lie group theory. Identify the Lie group as gauge group, select the group representations to which particles of matter are assigned, and calculate the coupling constant evolution using existing recipes.

    The only non-mechanizable part of the TOEing was to invent how to prevent proton from decaying too fast. Even a Hand-Book of TOEr was published providing the needed Lie theoretic recipes. Suddenly everyone could build his or her own TOE! The age of Einsteins and Newtons was over and replaced with complete democracy! This generated an impressive amount of literature and using this as an objective measure for progress one can say that Golden Age of theoretical physics had begun.

    GUTs initiated a period of enormous progress also in cosmology. Inflation theory postulating inflatons as Higgs like particles was invented to solve the horizon problem of standard cosmology. It however turned out that there is not a single inflation model free of problems and there is inflation of inflation models. But what was perhaps lost in quality was gained back in quantity! Eventually it turned out that inflation theory led to what was christened as multiverse meaning a total loss of predictivity. Superstring theory led to an analogous breakthrough and what was called "multiverse mania" started. In the new vision about good theoretical physics the loss of predictivity became a victory.

    GUTS provide also new insights to dark matter problem and endless variety of dark matter candidates as particles having very weak interactions with ordinary particles were invented. All experiments trying to discover these exotics failed but the research continues.

  2. Next great step in made possible by the methods of NS based progress was the discovery of super symmetry and the building SUSYs started. The simplest SUSY with minimal supersymmetry forced to give up fermion number conservation since for the needed Majorana fermions fermion number was conserved only modulo two. One could however guarantee conservation of the sum of B and L at low energies at least. Again a vast literature emerged. Large Hadron Collider demonstrated that there was no sign of SUSY at energies at which SUSY was believed to solve some key problems of unifications. In NS vision this in not however experienced as a problem and there is no doubt that also the colliders of future will search standard SUSY.

  3. After failing to solve the problem of unification of gauge interactions, it was natural to adopt even more ambitious goal: to fail to unify gravitation with standard model or its failing extension. The attempts to quantize general relativity had failed. Also string model of hadrons had failed but it was thought to be a pity that such amount of literature would be in vain.

    The idea of superstring models was to give up 4-D space-time altogether and to replace it with 2-D string world sheets. Superstring seemed to work only in dimension 10 (later it became clear that there are also other critical dimensions). Superstring models indeed led to what was believed to be a calculable theory free of infinities that had destroyed the hopes about quantizing general relativity. 5 different scenarios emerged but they had nothing to do with the real world since space-time is 4-D rather than 2-D. Despite this small shortcoming it was however thought that within 10 years proton mass could be calculated with the desired number of decimals.

    1. So called spontaneous compactification was proposed as the solution of the problem posed by the 4-D nature of space-time, and historians of science will probably see spontaneous compactification as the most impressive achievement of NS period. It was decided that the 6 big dimensions of 10-D Minkowski space experience so called spontaneous compactification and become very small so that the resulting 10-D space looks like a 4-D Minkowski space. Spontaneous compactification would take place by "non-perturbative effects". This predicted huge number of different 10-D space-times and it began to look like the theory could not make any predictions. First this was experienced as a problem.

    2. There was also a second unpleasant discovery. The observation that the expansion of cosmology is accelerating forces in GRT framework non-vanishing but very small cosmological constant. String models assuming spontaneous compactification predicted cosmological constant but it was too large by an immense factor and had also wrong sign.

      Then people started to see indications for the existence to spontaneous compactification with a small cosmological constant having a correct sign. Soon this became an established fact. Unfortunately, this led to the prediction that there are of order 10500 different string model vacua identified as so called Calabi-Yau manifolds so that the theory cannot predict anything. This space of solutions was christened as landscape. This looked first like a real problem.

    3. Now came one of the largest breakthroughs in the NS based developments: a complete philosophical turn-around. That theory cannot predict anything was earlier regarded as a catastrophe but suddenly it became the best that theory can achieve. One could say that M-theory was really unique among competing theories developed by people still applying the old-fashioned NI strategy. No-one of course new n about these theories because of the censorship in arXiv.org and prestiged journals.

    4. The minor difficulties did not end with this. A couple of years ago it began to look that this huge landscape of solutions with desired cosmological constant does not exist at all! There would be only the so called Swampland consisting of the existing solutions, whatever they might be. Now the research is continuing to dig the correct solution representing our Universe from the enormous Swampland and literature is generated with an amazing pace.

  4. In light of the impressive successes of NS it was natural that practically all funding went to NS. The Breakthrough Prices - about 3 million dollars - for the gurus of NS for making possible the memorable period in theoretical physics expressed in beautiful manner the gratitude of ordinary citizens.

The fate of NI after the victory of NS

When some-one wins, some-one also looses. I had the fate of being the loser. Being somehow slow-minded I could not adapt to the development and continued to base my attempts to do theoretical physics on NI.

The basic strategy of NI is simplistic: start from a problem and try to solve it. Try to find some really important theoretical problem or experimental anomaly and look what its solution requires. At best the solution could force to rewrite text-books.

  1. My starting point discovery was what I called the energy problem of general relativity (GRT). Conservation laws are lost because space-time becomes curved in the presence of matter, and this leads to a loss of translational, rotational and Lorentz symmetries (Poincare symmetries): Noether theorem cannot be used to construct conserved energy, momentum, and angular momentum.

  2. Conserved energy defines Hamiltonian in quantum field theory and if one does not have it, one cannot construct time evolution operator giving S-matrix coding the predictions of quantum theory. Hence the failure quantize GRT.

This kind of argument was of course hopelessly out-of-fashion and no-one took it seriously it and even less the theory that I developed in my thesis 1982 and continued to develop further and am doing so still but now by applying TGD to anomalies popping up almost on daily basis and developing the mathematical details of TGD.
  1. The idea was that space-time at fundamental level is representable as a 4-D surface of higher-D space-time, which is 4-D Minkowski space M4 of special relativity with points replaced with certain small internal space S. M4 allows the above mentioned Poincare symmetries becoming therefore symmetries of imbedding space M4× S rather than space-time surface. General relativity and standard model emerges at QFT limit when many-sheeted space-time is replaced with a regions of M4 made slightly curved. The problem is solved.

  2. Even more, of S is chosen to be CP2 one obtains standard model symmetries and unification of gravitation and standard model generalizing the superstring model by replacing strings with 3-D surfaces. The choice of H is unique from standard model symmetries. Many years later it became clear that twistor lift of TGD allows only this choice of H. Also number theoretic vision forces H as the only possible choice.

  3. The development of TGD led to an identification of numerous other problems and anomalies of the existing theory and led to a generalization of quantum theory involved modification of quantum measurement theory to solve its basic paradox and extending to a theory of consciousness. Also quantum biology emerged as application of new quantum theory and new view about space-time and classical fields.

Of course, I humbly admit that my little "achievements" are nothing as compared to the achievements of superstring models: the ability to predict nothing is much more impressive than ability to predict everything since it makes theory completely unique and I can only admit my defeat. Against this it is quite natural that during these four decades I have not received a single coin of funding. The money must go where the progress is.

To sum up, the transition from NI to NS meant a profound revolution in theoretical physics. By analogy one might expect the transition from AI to AS to take place. There are indeed encouraging signs about the AS revolution. It has been already proposed that theoretical physicists could become altogether obsolete. Computer programs would construct theories of physics and deduce their predictions. Intelligence - both artificial and natural - would leave theoretical physics altogether as it seems to be leaving also other areas of life.

For a summary of earlier postings see Latest progress in TGD.

Articles and other material related to TGD.

1 comment:

Anonymous said...

Excellent article.