Wednesday, June 27, 2018

Another Big Problem For String Theory

A diagram of string theory dualities. Yellow arrows indicate S-duality. Blue arrows indicate T-duality. These dualities may be combined to obtain equivalences of any of the five theories with M-theory (Image and caption via Wikipedia).

A schematic illustration of the relationship between M-theory, the five superstring theories, and eleven-dimensional supergravity. The shaded region represents a family of different physical scenarios that are possible in M-theory. In certain limiting cases corresponding to the cusps, it is natural to describe the physics using one of the six theories labeled there (from the same source).

Another new paper has largely established that one popular to theorize about class of string theories (Type IIB) has no variation ("vacua") that are consistent with our observations that the universe has a positive cosmological constant (technically speaking "de Sitter space"), rendering fifteen years of research attempting to apply this class of string theories to the real world a waste.

There are several other classes of string theories, so all hope for string theory is not lost yet. But, even more concerning for sting theorists is the possibility that since all of the different classes of string theories have deep connections to each other though S-dualities and T-dualities, that bind them to a common M-theory, that this result might be later proved to hold for all versions of sting theory, effectively falsifying the entire string theory enterprise.

Another recent paper, which I blogged previously, has made similar claims, and explores the impact of this result on non-Type IIB string theories for which there has not been as much research regarding the question that seems to have made it possible to falsify Type IIB string theory.

This is on top of some other long standing problems with string theory, such as the non-detection of any experimental evidence for supersymmetry (which is an element of almost all classes of string theory), the non-detection of more the four dimensions of space-time (which is an element of almost all versions of string theory), and the lack of any identified string theory vacua that correspond to the Standard Model and general relativity compatible universe in which we live.

String theorists are hoping that it might be possible to solve the lack of deSitter vacua with quintessence, i.e. by putting something into the vacuum rather than changing the nature of the vacuum itself.

This also leaves me wondering about questions like: how different phenomenologically from each other are (1) classical general relativity from (2) a spin-0 massless graviton in flat Minkowski space (with all of the properties apart from spin-2 of a standard spin-2 massless graviton) from (3) a spin-2 massless graviton in flat Minkowski space from (4) a spin-2 massless graviton in de Sitter space. My intuition is that options (2) and (3) are much less different phenomenologically from (1) and (4) than is conventionally assumed.

There is a short anthology of the literature on this new problem at this site, which I reproduce in the pertinent part below the fold (without reformatting).

On the other hand, discussion casting doubt on the existence of a large number of de Sitter spacetimeperturbative string theory vacua includes the following:
Implications of the possible non-existence of de Sitter vacua in string theory are explored in
But then:
  • Jakob Moritz, Ander Retolaza, Alexander Westphal, Towards de Sitter from 10D, Phys. Rev. D 97, 046010 (2018) (arXiv:1707.08678)
See also


neo said...

SO when it comes to QG what are we left with? on PF, urs shreiber and mitchell porter call LQG a mistake, an error. urs said spinfoam has no relation to GR

sabine describes some of the shortcomings of AS.

knock against entropic gravity is gravity is reversible entropic forces are not.

andrew said...

I'm thinking that a massless graviton in flat 4-D Minkowski space (possibly spin-9 rather than spin-2, but maybe not) is looking pretty interesting at the moment.

andrew said...

spin-0 not spin-9, that was just a slip of the finger. Nobody thinks that there are spin-9 gravitons.

neo said...

what's the uv completion of that theory?

by spin-0 are you alluding to Brans Dicke theory?

IMO create a theory of gravity that reproduces GR, perhaps some variation of a scalar tensor, but can be canonically quantized with better UV behavior

or GR is classical and is emergent and arises from QM ala Antoine Tilloy

andrew said...

"what's the uv completion of that theory?

by spin-0 are you alluding to Brans Dicke theory?"

I'm not thinking very specifically about any particular theory, really more a process of elimination. My reason for thinking about spin-0 is that it is much simpler and I'm not sure how much it adds phenomenologically to add the spin-2 to a spin-0 model. I wonder if some of the tensor features in GR couldn't be captured by spin-0 particle interactions .

neo said...

the only spin-0 that is experimentally verified is the higgs. which doesn't couple to massless particles like photon.

btw what's your take on LQG?

it seems string theory is "wrong" given all the aforementioned problems, and perhaps the realization there is no stable desitter space

maybe if string theorists switch to LQG, the outstanding problems of LQG can be solved more quickly. the kodama state gives rise to desitter and the problems witten found perhaps could be overcome.

andrew said...

I am agnostic about LQG.

There have been efforts to unify LQG and String Theory.

Some of the criticism of LQG is plainly off base.

LQG was certainly making progress as late as 2012.

LQG ideas about the imperfection of locality are potentially important.

The argument that QG via LQG could give you MOND-like rather than pure GR results is encouraging. This link also explains quite well why I am so ethusiastic about Alexander Deur's approach to QG.

LQG doesn't have to imply a breakdown of Lorentz-invariance.

It is a very natural way to include a cosmological constant.

In general, criticisms of LQG are overstated and usually demonstrated unfamiliarity with the state of the field.

neo said...


well i've also asked Sabine Bee on her blog similar question about LQG,

since if we're on a game show and behind door labeled strings is probably wrong for reasons mentioned in this post, door labelled loops might have a chance for success.

i pointed out to bee that even if some genius string theorist can come up with a way to make strings compatible with deSitter, as KKLT even urs scheiber agrees is debunked, it seems desitter is more natural in LQG.

after all smolin wrote a paper QG with a + cc, based on the kodama state.

if these smart string theorists instead work on ironing out the problems with loops, such as the semiclassical limit and hamiltonian, perhaps loops can progress more quickly, as LHC found no SUSY, to concrete predictions