No metastable type IIB de Sitter vacua

[Urs Schreiber]

Take note of today's Obied-Ooguri-Spodyneiko-Vafa 18.

Curious situation: After it had influenced much the development of the string theory community’s attention and almost all of its public media perception since 2003, after what must be thousands of followup publications, the argument of KKLT 03 for a large finite number of metastable de Sitter spacetime vacua of type IIB string theory is in the process of being abandoned for being plainly mathematically wrong.

Apparently the decisive blow was the focused and comprehensive analysis in Danielsson-VanRiet 18 of all the available arguments, which concluded that all this work is based on a mathematical assumption that was never checked and for which there is mounting evidence that it is just false.

If at the next social gathering smalltalk with your colleagues you need a good argument that physics does need mathematically airtight proof after all, you’ll have a monumental example here.

What a waste of energy and time and of media attention! On the other hand, good to see that scientific scepticism is coming back, even if with 15 years of delay.

 

[bapowell]

Thanks Urs. Any chance you can summarize what the assumption was?

 

[Urs Schreiber]

Any chance you can summarize what the assumption was?

The problematic assumptions is the stability and consistency of the anti-D3-brane probes, which KKLT-like constructions bring into give a positive cosmological constant.

Here is maybe the main points in Danielsson-VanRiet 18:

top of p. 21: "We now highlight 3 problems with the IIB constructions that have been pointed out in
the recent literature: 1) Issues with anti-brane backreaction on the internal geometry [119–
130], 2) issues with anti-brane backreaction on the 4D moduli, which was found recently by
Moritz, Retolaza, and Westphal [131] and 3) issues with very basic assumptions of the moduli-
stabilisation scenario as discussed by Sethi [132]."

p. 22: "The source of the debate/worry about KKLT stability came from the observation in [119] that the
solution, within the approximations made, contains an unphysical singularity near the anti-D3 at the
tip of the throat geometry."

p. 23: "the 4D interpretation of a 10D computation carried out recently in [131]. This 10D computation revealed that a single gaugino condensate together with anti-D3-branes cannot lead to dS vacua, thereby contradicting the original KKLT paper"

p. 24: "for general worries about the validity of an effective field theory description of anti-brane uplifting we refer to [156]."

p. 26: "It is not unthinkable that dS space is simply a space that cannot exist quantum mechanically. [...] This has been claimed before in several papers that study QFT in curved space [161–168] and if those papers are correct, then doing a proper string computation should reveal that dS vacua cannot exist."

p. 30: "We therefore think that the most natural assumption, at this point in time, is that string theory conspires against the existence of dS space."

 

[mitchell porter]

I'm very skeptical about this. Why wouldn't these arguments rule out inflation as well?

 

[bapowell]

If correct, they once again make string theory inhospitable to inflation. Mathematical proofs can’t rule out inflation, which is a model based on effective theories to describe observable phenomena.

 

[Urs Schreiber]

I'm very skeptical about this. Why wouldn't these arguments rule out inflation as well?

That sounds as if you are saying: "I am skeptical of the argument, because it contradicts what I would like to conclude". I suppose it's clear why this is a fallacy, or would be one if that's what you really mean?

Both Danielsson-VanRiet 18 as well as Obied-Ooguri-Spodyneiko-Vafa 18 point out that there are other potential mechanisms to obtain the desired conclusion. The problem really is that, due to lack of decent analysis, at the moment nobody has a clue what the theory really yields. 15 years wasted.

From p. 26 of Danielsson-VanRiet 18:

"From this analysis we conclude that string theory has not made much progress on the problem of the cosmological constant during the last 15 years [...] string theorists have not been up to the challenge."

 

[mitchell porter]

There's a huge literature on inflation in string theory, and those papers say nothing or almost nothing about it. The word inflation doesn't even occur in Obied et al, while the only substantial comment in Danielsson & Van Riet is in a footnote on page 29, where it is remarked that "in the context of inflation, the dS isometries are always a little bit broken", but that dS/CFT can apparently be used as an approximation, insofar as conformal symmetry still exists.

So it is desirable to understand the relationship between stringy inflation and this critique of the landscape. Van Riet's theme has been that KKLT etc employs a supergravity approximation, that might be invalid when all string and brane degrees of freedom are included. Is there a similar threat to the validity of stringy models of inflation? Conversely, can slightly breaking the dS isometries, save the KKLT landscape? Assuming that it needs saving, which still seems to be far from established.

 

[Urs Schreiber]

So it is desirable to understand

Yup.

 

[mitchell porter]

Now we have a new "Obied et al" which does actually address the issue of inflation. One of the new authors is Paul Steinhardt (who helped originate inflation, and now rejects it, for rather bad reasons in my opinion). I wonder how that collaboration came about...

But what really provokes me now, is this claim that you can't have positive vacuum energy in string theory. Isn't that what this amounts to - a claim that all string vacua have negative or zero vacuum energy. For that, arguments about anti-D3-branes aren't enough. What about Bousso-Polchinski, for example? That was the original proposal for how to realize Weinberg's scenario in string theory, and yet it barely rates a mention in Danielsson and Van Riet.

edit: Vafa's talk at Strings 2018 (PDF, video) this week.

 

[haushofer]

What does this mean for dS-holography, as used in e.g. Verlinde's theory of gravity?

 

[Urs Schreiber]

What does this mean for dS-holography, as used in e.g. Verlinde's theory of gravity?

It means that it looks even more far fetched.

 

[Urs Schreiber]

Now we have a new "Obied et al" which does actually address the issue of inflation. [...] edit: Vafa's talk at Strings 2018 (PDF, video) this week.

Thanks for the pointers!

While it is interesting to see the sociology of the traditional lore being abandoned, it seems dubious to me that the solution to the problem is to abandon one improperly proven claim just to embrace the next best improperly substantiated conjecture. At some point one will have to admit that the only way to make progress here is to make a solid analysis of the theory, and to actually build the (non-perturbative) theory where it is currently missing. Otherwise we'll just enter the next 15 year cycle without gain.

 

[atyy]

Otherwise we'll just enter the next 15 year cycle without gain.

I don't understand why you think it's 15 years without gain. First, it does seem that these points were known in the string community earlier. For example, Van Riet wrote a guest blog (with interesting comments section) on Motl's site in 2014: An evaporating landscape? Possible issues with the KKLT scenario. Second if it takes that long to understand that the problems with KKLT may be intractable, then that is still progress.

 

[ohwilleke]

Both Danielsson-VanRiet 18 as well as Obied-Ooguri-Spodyneiko-Vafa 18 point out that there are other potential mechanisms to obtain the desired conclusion. The problem really is that, due to lack of decent analysis, at the moment nobody has a clue what the theory really yields. 15 years wasted.

Both D-VR 18 and OOSV 18 seems to go well beyond Type IIB String theory. For example, OOSV 18 also takes aim at 11 dimensional supergravity, and conjectures that an inequality inconsistent with deSitter space applies to all string theories, and D-VR 18 puts the various string theories on a branching tree and plausibly suggests how the result for Type IIB might extend more generally.

OOSV 18 also seems to suggest that while its hypothesis that deSitter space is inconsistent with all string theories doesn't rely on supersymmetry, that it seems particularly easy to apply to any supersymmetric theory.

Is this impression correct?

 

[Urs Schreiber]

it does seem that these points were known

Doubts had been raised early on, that didn't take 15 years, but the main stream ignored them. The remarkable aspect of Ooguri-Vafa's article is not any mathematical insight they add, in fact they add conjecture on conjecture, but that it's Ooguri-Vafa agreeing that the KKLT construction is most dubious. This is the sign that now the message has arrived in the inner circle of "the community".

 

[Urs Schreiber]

Is this impression correct?

Sounds good to me. I felt D-VR-18 were trying to be careful, appreciably, to state what they have, and not to overstate.

 

[Islam Hassan]

I am obviously not at all qualified to comment on this development, having a basic “String Theory for Dummies” understanding of the subject. What is striking however to any reader of the present thread —be they qualified or not— is the repeated occurence of the number 15...

If Danielsson-VanRiet 18 does indeed turn out to be a sound analysis, it would be quite astounding, to think that such a ‘group think’ phenomenon could be so long-lived as mainstream science...IH

 

[PAllen]

I am obviously not at all qualified to comment on this development, having a basic “String Theory for Dummies” understanding of the subject. What is striking however to any reader of the present thread —be they qualified or not— is the repeated occurence of the number 15...

If Danielsson-VanRiet 18 does indeed turn out to be a sound analysis, it would be quite astounding, to think that such a ‘group think’ phenomenon could be so long-lived as mainstream science...IH

The history of science is littered with centuries long group thinks. Nothing really unusual at all. Just consider Ptolemaic solar system model. It seem very healthy to me that insiders in the field are the ones finding the errors. THAT is unusual in a positive way in the history of science.

 

[Islam Hassan]

I guess you're right. We tend to consider the era we live in as the greatest, the biggest, the fastest, the most this or that, etc. We are biased towards our era as well as ourselves.

History helps...IH

 

[ohwilleke]

Perhaps beyond the scope of this topic, but, suppose that none of the vacua in M-Theory are de Sitter, and that this problem (which maybe standing alone could be cured with a quintessence theory of dark energy, the papers seem to suggest) and other problems with String Theory, taken together, mean that M-Theory is officially NOT in any of its manifestations capable of being a Theory of Everything that describes our universe, or even a Grand Unified Theory that explains the Standard Model.

What components of M-Theory, if any, are most amenable to being repurposed for theoretical physics in non-M-Theory directions? Or, is M-Theory really such a deeply interrelated set of concepts that it is really an all or nothing affair?

Also, even if de Sitter vacua are ruled out, are there any possible "flat" vacua in M-Theory (or at least Type IIB String Theory) that are neither de Sitter or anti-de Sitter? It didn't seem like there were but I'm not confident that I have correctly understood that point.

 

[Ken G]

Reading through all this with little background expertise, but accepting the arguments at face value, I am led to wonder if the most natural conclusion isn't, as D-VR 18 say, that "string theory conspires against the existence of dS space," but rather, that the existence of dS space conspires against the usefulness of string theory. And similarly, rather than concluding "string theory has not made much progress on the problem of the cosmological constant during the last 15 years," it is perhaps more tempting to conclude that string theory is not the right course at all, when deSitter space seems to have much more going for it. I realize D-VR 18 are not trying to conclude anything beyond what they can show, but it still seems to me that the language used is placing string theory on a kind of pedastol, such that we should blame KKLT 03 for being careless and wasting 15 years, rather than blaming string theory itself, and using the KKLT 03 affair as a very productive means of helping us decide to try something other than string theory (or at least type IIb string theory, I don't really know the alternatives there). I'm saying that if it took the analysis of the last 15 years to understand that string theory is a red herring for understanding inflation and/or dark energy, then that may be the most significant discovery in theoretical physics in that timeframe. When everyone is doing something, the most constructive possible thing is to find out as quickly as possible if they should be doing something else!

 

[Urs Schreiber]

if it took the analysis of the last 15 years to understand

It didn't take 15 years to see that KKLT had problems, that took at most a few months, e.g. here. But if a field does not proceed by rigorous argument, then some handwaving may survive a long long time before being abandoned.

It's hard to draw any conclusions from this regarding the actual theory, since the point is that the poor theory has been neglected all along, in favor of fantasies.

 

[Ken G]

But that conclusion, that only the KKLT 03 flawed version of string theory faces difficulties, is what I don't see here. I see that you are saying KKLT 03 was nonrigorous and is likely deeply flawed, and I am accepting this as fact. But the flaw was not that KKLT 03 didn't work well with dS space, whereas string theory might, the flaw was that KKLT 03 did work well with dS space, and string theory might not. The fervent hope for string theory, and the reason KKLT 03 had such impact, is that it would all play well with the dS space, as dS space seems to be our best means of understanding a plethora of befuddling astronomical observations. So critiquing KKLT 03 on the grounds that a more rigorous argument should not play well with dS is a much deeper issue than simply rejecting KKLT 03 and all the work done following it. It seems to me it must shake string theory to its core if it does not work with dS space, so if KKLT 03 stimulated arguments that demonstrate string theory (or some popular form of it) does not do inflation, then it was an error that was of tremendous value. You could certainly argue that some might have already known that string theory doesn't like to do inflation even without KKLT 03, but would it have the same widespread impact as getting everyone to believe something, and then show it isn't true? The rejection of KKLT 03, as framed here, sounds awful close to a proof by contradiction that string theory won't explain the need for inflation to understand astronomical observations.

 

[haushofer]

It means that it looks even more far fetched.

That's what I also think, but somehow not a lot of people seem to worry about that. Maybe I'm missing the relevant papers on this.

 

[mitchell porter]

are there any possible "flat" vacua in M-Theory (or at least Type IIB String Theory) that are neither de Sitter or anti-de Sitter?

Vacua with unbroken supersymmetry have zero vacuum energy and are therefore flat. Until 1998 (discovery of dark energy), string phenomenologists hoped that some unknown mechanism could still cancel or absorb vacuum energy contributions in the realistic vacua with broken supersymmetry. But after 1998, people took the problem more seriously, and after KKLT in 2003, opinion moved to the idea that there is no unknown mechanism that damps the gravitation of the vacuum energy, there's just a coincidental-looking, but anthropically necessitated, near-cancellation between negative and positive contributions to the vacuum energy.

Now apparently we are going to think about whether e.g. an undiscovered generic instability of quantum de Sitter space would decay all those would-be dS vacua after all. In that case it's back to the pre-1998 situation, but with the post-LHC knowledge that there's something wrong with the old ideas of supersymmetric phenomenology too.

If dark energy isn't vacuum energy, the next candidate is quintessence, and Vafa has his idea for how it fits into string theory (a modulus, e.g. size of a Calabi-Yau "handle", that only interacts with dark matter or hidden sector fields). But the problem with supersymmetry-breaking creating a vacuum energy still exists. Vafa's supervisor Witten wrote a few papers speculating as to how supersymmetry might somehow be unbroken after all (and we have a long-running thread here in "Beyond the Standard Model" where that is also a theme). Then there are string models which are non-supersymmetric from the beginning (but which have a kind of echo of supersymmetric properties).

String Pheno 2018 opens in Poland today...

 

[Urs Schreiber]

String Pheno 2018 opens in Poland today...

Thanks for the alert. From the webpage there seems to be one contributed talk on the subject:

Timm Wrase:
"de Sitter extrema and the swampland"
Abstract: Following the recent claim in 1806.08362 that dS extrema are in the swampland, I will review and scrutinize existing constructions of type II flux compactifications that use only classical ingredients. I will review some no-go results against dS extrema but also show that explicit constructions of (unstable) dS extrema exist that are inconsistent with the conjecture in 1806.08362. I will scrutinize all potential assumptions that go into these existing constructions.

(Interesting how the attribution goes, now the claim is attributed to Ooguri-Vafa et. al...)

Probably Wrase will be comparing to

Diego Gallego, M.C. David Marsh, Bert Vercnocke, Timm Wrase.,
"A New Class of de Sitter Vacua in Type IIB Large Volume Compactifications"
(arXiv:1707.01095)

 

[Urs Schreiber]

Danielsson now exploring possible alternative constructions:

Souvik Banerjee, Ulf Danielsson, Giuseppe Dibitetto, Suvendu Giri, Marjorie Schillo,
"Emergent de Sitter cosmology from decaying AdS"
(arXiv:1807.01570)

"Since the discovery of dark energy two decades ago, string
theory has been faced with the challenge of reproducing
a small positive vacuum energy. The dominant approach
has been the reliance on a landscape of different vacua [1]
equipped with a transition mechanism such that the an-
thropic principle selects our vacuum [2]. This approach
became calculable in string theory with the construc-
tion of KKLT [3], in which one can achieve a landscape
of scale-separated vacua, with both positive and nega-
tive cosmological constant (CC), by tuning flux numbers.
However, issues have recently been raised which indicate
that the na ̈ıve application of supersymmetry-breaking
and non-perturbative effects necessary in the construc-
tion of the landscape is insufficient [4–8]. These concerns
suggest that not even a single rigorous string vacuum has
actually been constructed and further hints that string
theory abhors de Sitter space and any solution with a
positive vacuum energy will suffer from instabilities.

"Given these difficulties, it is reasonable to consider the
possibility that neither metastability nor scale separa-
tion can be achieved in string theory in the way envi-
sioned. It seems, therefore, that something completely
different is needed. In order to construct an alternative,
we will take motivation from work that received consid-
erable attention around the turn of the millennium just
before the idea of the string landscape started to flour-
ish: braneworlds. In this context, the cosmology we see
as 4D observers is not due to vacuum energy, but rather
arises as an effective description on a dynamical object
embedded in a higher dimensional space.

"In the scenario developed by Randall and Sundrum
(RS)[9, 10], two identical AdS5 vacua are glued together
across a three-brane. The 5D graviton has a zero mode
confined on the brane that gives rise to an effective 4D
gravity despite the existence of large extra dimensions;
this solves the issue of finding scale-separated vacua. We
will consider a variation of this scenario that starts with
a metastable false AdS5 vacuum that non-perturbatively
decays to a supersymmetric true AdS5 vacuum through
bubble nucleation. Here, a spherical brane separates
the two phases with an inside and an outside, and 4D
observers confined to the brane experience an effective
dS4. This scenario is further motivated by recent argu-
ments that all non-supersymmetric AdS vacua must pos-
sess such an instability in a consistent theory of quantum
gravity [11, 12]

"In section 2, we discuss the physics experienced by an
observer riding on an expanding bubble that nucleated in
a first-order phase transition. In section 3 we discuss the
restriction of gravity to the brane, performing a consis-
tency check that the 4D Newton’s constant matches the
expected value. In section 4 we sketch a concrete con-
struction of a de Sitter braneworld using explicit type
IIB string vacua; a metastable vacuum breaking all su-
persymmetries first obtained by Romans [13] is shown to
have a decay channel to an orbifold of the maximally su-
persymmetric vacuum. We conclude by commenting on
some future developments of this scenario."

 

[Ken G]

Interesting, so they are trying to save string theory in cosmology by eliminating elements that string theory is intolerant to, and replacing them with higher-order structures that string theory is more tolerant to. I would call that a rather striking break from the concept of Occam's Razor, where if the simplest description of some phenomenon is a positive vacuum energy, then we should seek theories that are compatible with that simplest description. But I realize that "what is simplest" is a somewhat subjective determination, and for those already married to string theory, maintaining what advances they already perceive is always going to seem like the simplest path. Ultimately, any scientific theory must be judged not on what we see that it can explain, but rather what we have not yet seen that we would only expect to see if the theory is good. We must await that type of confirmation of all these notions. I don't know if any of us will see that in our lifetimes, however, but perhaps.

 

[Islam Hassan]

Interesting, so they are trying to save string theory in cosmology by eliminating elements that string theory is intolerant to, and replacing them with higher-order structures that string theory is more tolerant to. I would call that a rather striking break from the concept of Occam's Razor, where if the simplest description of some phenomenon is a positive vacuum energy, then we should seek theories that are compatible with that simplest description. But I realize that "what is simplest" is a somewhat subjective determination, and for those already married to string theory, maintaining what advances they already perceive is always going to seem like the simplest path. Ultimately, any scientific theory must be judged not on what we see that it can explain, but rather what we have not yet seen that we would only expect to see if the theory is good. We must await that type of confirmation of all these notions. I don't know if any of us will see that in our lifetimes, however, but perhaps.

That’s an interesting comment...I wonder if Occam’s is a universal, foolproof principle...general relativity is very complex...but correct...one needs just enough complication to get it right...and no more...

But then we are far from knowing if ST in any of its guises is correct or not...IH

 

[Ken G]

I agree, the relation between Occam and string theory is still unclear, and Occam is more useful in hindsight than foresight. All I'm saying is, there is a danger when a community builds an expertise that has as yet not proved to be a good theory, there can be the danger of square pegs in round holes. But they are the best ones to figure out which it is at this point-- if they can be objective.

 

[Islam Hassan]

I agree, the relation between Occam and string theory is still unclear, and Occam is more useful in hindsight than foresight. All I'm saying is, there is a danger when a community builds an expertise that has as yet not proved to be a good theory, there can be the danger of square pegs in round holes. But they are the best ones to figure out which it is at this point-- if they can be objective.

Agreed, except perhaps that objectivity can sometimes be a wobbly issue...from what I gather mathematical objectivity is perhaps the major focus of ST today...with the hope that this will one day yield a testable physical one...in fine, objectivity is also defined in hindsight...IH

 

[Urs Schreiber]

Ken G, some information:

The KKLT 03-constructions that are being abandoned (for being wrong) were so complicated that Susskind 03 famously referred to them as "not at all simple. They are jury-rigged Rube Goldberg contraptions" (p. 5). Hence abandoning KKLT likely means to reduce the perceived complexity, not to increase it.

On the other hand, beware that there are general arguments in quantum gravity, independent of string theory, that global de Sitter spacetime is inconsistent, see e.g. Rajaraman 16 and references given there. If true, this means that no quantum consistent model for observed cosmology will be totally straightforward, all of them will have to realize a de Sitter cosmology as an effective phenomenon on the backdrop of non-de Sitter cosmology.

Hence if you are intent on doing fundamental physics by way of slogans, remember the Occam's slogan has an addendum: "as simple as possible, but no simpler".

 

[Ken G]

Ken G, some information:

The KKLT 03-constructions that are being abandoned (for being wrong) were so complicated that Susskind 03 famously referred to them as "not at all simple. They are jury-rigged Rube Goldberg contraptions" (p. 5). Hence abandoning KKLT likely means to reduce the perceived complexity, not to increase it.

That's well and good, and I don't doubt you are correct, but that's not the reduction in complexity I'm talking about-- I'm talking about the cosmological constant. It sounds like you are saying that the reason KKLT existed in the first place was because it looked like it supported a positive vacuum energy. You are saying that was never a good enough reason to adopt what is essentially wrong physics, but in arguing it is wrong, you are pointing to studies that conclude type IIB string theory is not terribly compatible with positive vacuum energy. So that's the irony here-- the killing blow to KKLT is that it concludes something that string theory prefers to violate, yet it is the thing that KKLT is trying to conclude that is the simplest explanation for astronomical observations. This in turn suggests that string theory may not be terribly compatible with Occam's Razor when applied astronomically.

On the other hand, beware that there are general arguments in quantum gravity, independent of string theory, that global de Sitter spacetime is inconsistent, see e.g. Rajaraman 16 and references given there. If true, this means that no quantum consistent model for observed cosmology will be totally straightforward, all of them will have to realize a de Sitter cosmology as an effective phenomenon on the backdrop of non-de Sitter cosmology.

That does sound significant, though of course it will also present the choice: accept awkward modifications to quantum gravity to get de Sitter, or just go with a very different approach to gravity that is built to play well with de Sitter from the get go. How much of current quantum gravity is regarded as successful, in the sense of producing expectations of things we actually see? Any of it?

Hence if you are intent on doing fundamental physics by way of slogans, remember the Occam's slogan has an addendum: "as simple as possible, but no simpler".

Calling Occam a "slogan" hardly rescues you from the problem here: physics must be done so as to give the simplest possible explanations to what we actually see. That isn't a slogan, it's the definition of scientific progress.

 

[Urs Schreiber]

What I meant to indicate regarding "slogans" is -- given that the field has suffered from people being prejudiced about what the theory should yield, without being careful about doing the actual math -- that it is wrong to react to this by just changing the prejudice about what ought to be the case. It's time to do careful analysis and let the prejudices rest for a while.

 

[Urs Schreiber]

Danielsson himself writes about the issue discussed here:

U. Danielsson: "Is string theory in crisis?" (July 8, 2018)