Ethan Siegel's misconceptions on quantum gravity

Ethan Siegel is such a reasonable man when it comes to "ordinary" topics in physics and science. But he didn't hesitate to write about something that self-evidently goes well beyond his expertise, namely quantum gravity and string theory.

The result is a repetition of the most widespread, cheapest, and most incorrect laymen's misconceptions that you may hear from generic dirty homeless beggars on the street, except for the smartest and cleanest ones. ;-)

His

Ask Ethan #91: Does Quantum Gravity Need String Theory? (medium.com)

starts with an important quote by Edward Witten:

I just think too many nice things have happened in string theory for it to be all wrong. Humans do not understand it very well, but I just don’t believe there is a big cosmic conspiracy that created this incredible thing that has nothing to do with the real world.

Witten wants to say that even without definite empirical proofs, the mathematical properties of string theory make us certain that it is an incredibly tight, rich, and unique mathematical structure that seems to contain the ideas compatible with physics as well as many new structures and relationships that came as surprises and taught us to think about physical and mathematical concepts in new ways.

If string theory happened to be a wrong theory of physics, a curious researcher familiar with it would still face the surprising "anomaly" waiting to be explained – namely why such an unused or useless unique structure that resembles everything we need physics, but is not the right theory of physics, exists. If string theory were known to be wrong sometime in the future, the question from the previous sentence wouldn't quite be physics. But it would still be one that an intelligent, curious researcher interested in the logical relationships between physical and mathematically physical ideas simply couldn't ignore.




Ethan responds to Witten's quote by words that make you think that he agrees with Witten but he clearly doesn't. If the following quote is meant to confirm something about Witten's quote (and the words "there's no doubt that" do sound like he wants to agree with something), it is as distorted and demagogic as you can get. Ethan tells us:

There’s no doubt that, from a mathematical viewpoint, there’s no shortage of incredible, beautiful, elegant frameworks out there.

But Witten's point was almost exactly the opposite one. There is a huge shortage of such theories. As far as we know, their number is exactly one. String theory is an absolutely unique accumulation of tightly connected mathematical ideas that are seemingly relevant for physics. There is nothing such as another "counterpart" of string theory that would deserve similar words as Witten's words for similar reasons.




Ethan continues:

But not all of them are relevant for our physical Universe.

Not all mathematical ideas are relevant for the mathematical Universe. But string theory is completely unique so it is totally inadequate to clump the question about its physical validity with the questions about the validity or relevance of completely different mathematical ideas and structures. The question is whether string theory is relevant for our physical Universe, not what is the percentage of the relevant ideas in some random group of mathematical ideas most of which are not string theory.

It seems like for every brilliant idea that accurately describes what we can observe and measure, there’s at least one equally brilliant idea that attempts to describe the same things that turns out to be completely wrong.

Well, then the world isn't what it "seems" to you. Whether string theory is right or wrong as the most fundamental theory of physics, there isn't any analogous other, equally "elegant" theory or idea. You may be and other laymen may be ignorant about this conclusion of the scientific research but this ignorance doesn't make the uniqueness of string theory any less true. Ethan copies a quote by a boy called Kent:

I hope you do have time to devote an article to quantum gravity sometime soon. In particular, I would like to know if any progress has been made in this field in the last five to ten years. From my non-expert perspective, it seems like the field has been stuck for a while since since string theory started falling out of favor for testability reasons and having \(10^{500}\) possible solutions. Is this true, or is there progress being made behind the scenes that just hasn’t gotten as much main-stream press?

Dear Kent, first, string theory didn't start to "fall out of favor", except among imbeciles who have no clue about the discipline. If you were told something else, then you were fed feces and you should better throw up as soon as possible. Also, almost no scientific papers talk about the number of \(10^{500}\) solutions. It was an estimate of some semi-well-defined set that appeared in one paper and its several followups – an estimate that may be accurate or less accurate (and it is absolutely irrational to view one answer or another emotionally as "good news" or "bad news": they are just possible answers to a question!) – but thousands of other string theory papers that were written before that paper and after that paper study completely different issues than this number.

Whether the research in theoretical physics is stuck or not is relative and subjective. But what is not relative or subjective is the fact that in the recent decade just like in the previous two, virtually all viable findings and calculations that have something to say about the unified quantum theories that include gravity either directly followed from string theory or are at least string-theory-inspired (like PR, ER=EPR, Ryu-Takayanagi, glue-entanglement etc.) in the sense that string theory is either viewed as the only specific example of some more general and more vague ideas; or the string-inspired ideas were directly extracted from some features of string theory phenomena.

First off, there’s a big difference between the idea of quantum gravity, the String Theory solution (or proposed solution), and other alternatives.

Right. The difference is that all the other alternatives are known to be wrong while string theory has passed all the tests so far. But if one only talked about "currently known and not yet excluded" candidates for a theory of quantum gravity (in \(d\geq 4\)), then all the research findings are compatible with the thesis that "string theory" and "quantum gravity" describe the very same structure, albeit with different emphases.

After some unoriginal introduction to general relativity and the Standard Model, Ethan writes things like this:

[T]here are multiple physical instances where we need a quantum theory of gravity, all having to do with gravitational physics on the smallest of scales: at tiny distances.

Physical phenomena at extremely short (Planckian) length scales obviously do require the "full theory of quantum gravity". But even though Ethan explicitly wrote it, it is not true that other questions – involving physics at longer distances – never involve questions about quantum gravity. All the recent activity concerning PR, ER=EPR and "firewalls" deals with – in the community, controversial – questions about events and information that take place at macroscopic length scales, length scales much longer than the Planck length.

The reason is that the apparent "mysteries" about the localization of quantum information in the presence of the black holes don't start just near the singularity; they already start near the event horizon. That's why the proposed "paradoxes" deal with the low-curvature regions just outside or just inside the event horizon. The idea that the research of quantum gravity only has to deal with the ultrashort distance scales is a misconception of the laymen such as Sabine Hossenfelder and it is sad to see that Ethan is confused about these basic matters, too.

Does the gravitational field pass through both slits? Through one-or-the-other? In General Relativity, there’s no way to account for this.

This is ludicrous. First of all, the gravitational field is a different name for the metric tensor – a classical field in the classical theory; a quantum field in an effective quantum field theory. This field exists everywhere so it surely exists in both slits. For the "gravitational field" to be investigated on par with the electrons, you need something like particles. For example gravitons.

Our established knowledge of general relativity and quantum mechanics surely allows us to answer questions such as "Whether gravitons may create an interference pattern in the double slit experiment". The answer is a resounding "Yes". Due to the weakness of gravity, it's extremely hard to detect gravitons. But physics knows that the superposition principle applies to all systems in Nature and gravity just cannot be an exception. Many experiments combining quantum mechanics and gravity, e.g. the neutron interference experiments in the gravitational field, have actually been made and all the theorists' predictions have always been confirmed.

The laymen are often misled by popular texts when they are effectively told that "we do not have clue what happens when quantum mechanics is combined with gravity". This assertion is hugely overstated. We know almost everything that is relevant for all such mundane questions – such as those in the double slit experiment. Quantized general relativity works well to solve all similar homework exercises. When we want to solve "every" possible problem, we find out that the naively quantized general relativity breaks down and is inconsistent – for high-energy or short-distance processes, or even for complicated enough yet fine measurements of quantum information at long length scales.

But that doesn't mean that we don't know anything. Even without string theory, we can surely solve all well-defined problems of Ethan's type that he sells as complete mysteries.

It’s thought that there must be a quantum theory of gravity to account for these and other problems inherent to a “smooth” theory of gravity like General Relativity. In order to explain what happens at short distances in the presence of gravitational sources — or masses — we need a quantum, discrete, and hence particle-based theory of gravity.

All the adjectives except for "quantum" are just totally unjustified and almost certainly wrong. The right theory of quantum gravity almost certainly cannot be discrete – even Sabine Hossenfelder began to understand why – and the right theory almost certainly isn't a point-like particle-based theory in the same spacetime. We don't have the full rigorous proofs but we have lots of partial ones and the scientific research seems to imply exactly the opposite of Ethan's adjectives. The right theory must admit some continuous description which makes the continuity of the spacetime – which may emerge – manifest. And the right theory of quantum gravity almost certainly isn't just a local theory of point-like particles.

After some elementary words about the list of particles in the Standard Model plus the graviton and references to the existence of the Newtonian limit, he writes:

The big question, of course is how? How do you quantize gravity in a way that’s correct (at describing reality), consistent (with both GR and QFT), and hopefully leads to calculable predictions for new phenomena that might be observed, measured or somehow tested.

The verb "quantize" (gravity) is a widespread piece of sloppy jargon. The word "quantize" refers to a particular procedure of reconstructing a quantum theory from its classical limit – effectively the usual procedure of adding hats, promoting degrees of freedom to operators, making their commutators nonzero, and so on. But we pretty much know that the correct quantum theory of gravity cannot be obtained (from classical GR) by such a simple quantization procedure.

This is unfortunately not just a terminological imperfection. People who use this flawed language almost always make incorrect assumptions about "what the quantum theory should look like".

1.) String Theory. String Theory is an interesting framework — it can include all of the standard model fields and particles, both the fermions and the bosons. It includes also a 10-dimensional Tensor-Scalar theory of gravity: with 9 space and 1 time dimensions, and a scalar field parameter. If we erase six of those spatial dimensions (through an incompletely defined process that people just call compactification) and let the parameter (ω) that defines the scalar interaction go to infinity, we can recover General Relativity.

It is a very outdated description. First of all, if he discusses the 10D Minkowski vacua, they contain other massless fields – the gravitinos (one or two), the \(B\)-field (except for type I), the Ramond-Ramond fields (for type I/II i.e. non-heterotic theories), and the \(E_8\times E_8\) or \(SO(32)\) gauge multiplets (in the heterotic case).

Already in these examples, the massless spectrum is diverse, depending on which of the five 10-dimensional string vacua we take. But the field content predicted by string theory is even much more dynamical and "flexible" than these differences would indicate. For example, we know that string theory allows a solution/vacuum/description – one known M-theory – that looks like the 11D Minkowski one. The massless spectrum is that of the 11D supergravity. There is no scalar. Instead, there is one gravitino, the metric tensor, and a 3-form potential.

Once we considered compactified vacua, the number of different transformations – and appearance and disappearance of massless or light fields of many types – becomes even richer. Ethan thinks of string theory in terms of a "fixed" field content that is just compactified but this is indeed a point-like particle-based view and it is not adequate for a proper modern understanding of string theory.

But there are a whole host of phenomenological problems with String Theory.

There are no phenomenological problems with string theory that could be accurately described by this bold, universal sentence. Individual vacua or classes of vacua may have problems but there are known vacua that seem to suffer from no problems, too.

One is that it predicts a large number of new particles, including all the supersymmetric ones, none of which have been found.

It's not a sin for a theory to predict something. Predictions are a part of the package, an important part. Superpartners haven't been found but they haven't been excluded, either, which is why it's logically incorrect to use the unknown answer to this question for or against a theory. Moreover, superpartners may have been created at the LHC this week and we may hear about the discovery soon. They're almost certainly the #1 justification of this run of the Large Hadron Collider.

To pretend that "superpartners are silly or can't exist" means to be deluded not only about modern theoretical particle physics but about the modern experimental particle physics, too. Supersymmetry is an extremely important and extremely physical theory in particle physics – indeed, one with important implications for quantum gravity, too – and whoever doesn't appreciate it is an uneducated brainwashed savage.

It claims to not need to need “free parameters” like the standard model has (for the masses of the particles), but it replaces that problem with an even worse one. When Kent refers to “\(10^{500}\) possible solutions,” these solutions refer to the vacuum expectation values of the string fields, and there’s no mechanism to recover them; if you want String Theory to work, you need to give up on dynamics, and simply say, “well, it must’ve been anthropically selected.”

The replacement of continuously adjustable parameters by a countable – and, as Ethan says, even finite – set of possibilities surely doesn't make the degeneracy worse because the number of real numbers in \(\RR^{19}\) was infinite. The increase of predictivity in string theory is a huge improvement. If the number of possibilities is countable or even finite, it is in principle possible to isolate the right solution exactly – indeed, to exactly calculate all the vevs – and calculate all answers to physical questions absolutely precisely. A finite number of bits extracted from the experiments – that may isolate the correct vacuum – is the sufficient input.

Whatever is the number of solutions (one of which is ours) predicted by a fundamental theory, it is the right answer. To say that "one doesn't like if the number is this or that or too high" only shows the critic's huge bias and irrationality. In Nature, we have hundreds of isotopes of nuclei – and almost googolplexes of possible DNA codes, possible species – and people could say that it's too many, too undetermined etc. But that's what the science implies and if you don't like how the Universe works, you should better apply for asylum in a different Universe.

Whether the right vacuum is only selected "anthropically" is a pure speculation. There is no credible research that would establish that no other criteria or mechanisms are actually picking the right vacuum. In the absence of clear evidence, we may only talk about sociology. I would bet that more than 50% of string theorists think that the right explanation of the vacuum selection isn't just "the pure anthropic principle". But we don't have solid evidence in one way or another which is why it's wise for people to remain open-minded instead of inventing ideological rationalizations of their randomly chosen views.

But despite what you may have heard, String Theory isn’t the only game in town.

Despite what you may have heard from the aromatic homeless folks on the street and read on crackpots' blogs and in books that should better be recycled quickly, string theory is the only game in town. This is not just a slogan. There are numerous papers that have found nontrivial evidence in favor of various technical interpretations of this playful slogan in numerous contexts.

2.) Loop Quantum Gravity. LQG is an interesting take on the problem: ...

No, it's not.

...rather than trying to quantize particles, LQG has as one of its central features that space itself is discrete.

No, this is a completely flawed description. LQG is actually the most canonical, and in some sense, unique example of a (failed) attempt to obtain a theory of quantum gravity by directly quantizing particles. That's what Ashtekar did. LQG was always meant to be "the canonical quantization" of gravity.

The "discreteness of space" is usually presented as a derived feature in LQG but it's only derived if one changes the boundary conditions on the phase space to different, "periodic" ones. In this way, the discreteness is "put in" as an assumption. Properly quantized fields in GR imply that no such discreteness exists.

But aside from these modified boundary conditions – the metric tensor is "rewritten" in terms of bulk gauge fields and the Wilson lines/monodromies (taking values in compact group manifolds) describe the most relevant observables (this is how Ashtekar's formalism may bring you to loops, and that's what Smolin proposed) – LQG is a straightforward quantization of point-like particle-based effective theories, like GR, and that's a way to see that it cannot work, too.

Independently, we know numerous reasons why the discreteness of space instantly kills the theory. Most importantly, finite-density discrete structures in the spacetime always break the Poincaré symmetry, see e.g. Backreaction if credible sources are not enough for you. Such a discrete structure in the spacetime would also fail to be unique and therefore carry a nonzero entropy density. This is also a problem, not just for thermodynamics but for relativity, too: entropy density is the time component of a 4-vector so if this 4-vector is nonzero, the Lorentz symmetry is broken – another proof of that.

[A playful description of some LEGO trampoline.] Space might be the same way!

No, it might not. You may only think that this possibility exists before you learn and/or understand about any relevant evidence.

[Comments on the wrong spin networks.] ...one that saw a tremendous leap forward made in just 2007/8, so this is still actively advancing.

There have been no results that would change the essential point that discrete structures don't have a limit that looks like a smooth Poincaré-invariant spacetime.

3.) Asymptotically Safe Gravity. This is my personal favorite of the attempts at a quantum theory of gravity.

Good for you, Ethan. Unfortunately, they don't seem good for physics.

[Basic comments about asymptotic freedom of QCD and a hypothesized analogy for gravity.] ...All coupling constants change with energy, so what asymptotic safety does is pick a high-energy fixed point for the constant (technically, for the renormalization group, from which the coupling constant is derived), and then everything else can be calculated at lower energies.

Except that while a complete proof doesn't exist, it seems very unlikely that such a fixed point exists. Why?

A "fixed point" is a scale-invariant theory. The beta-functions (numbers indicating how much a coupling constant depends on the scale) for all couplings have to vanish. However, the perturbative non-renormalizability of the Einstein-Hilbert action makes it rather clear that these theories should be viewed as theories from an infinite-dimensional class of theories labeled by infinitely many couplings.

It seems very unlikely that all these (infinitely many) beta-functions vanish for a finite value of all of them. Well, just to be sure, it is not "insanely" impossible because the vanishing of the beta-functions is a set of infinitely many equations for the exact same infinite numbers of variables. But the known list of scale-invariant theories make it rather clear that interacting (not free, not Gaussian) fixed points are very rare and there always seem to be some symmetries or special structures that justify their existence.

The scientific literature offers no nontrivial evidence that the fixed point of gravity exists. Jacques Distler's explanation why all the proposed "evidence" in favor of the existence of the fixed point is actually vacuous is as valid today as it was in 2008.

At least, that’s the idea! We’ve figured out how to do this in 1+1 dimensions (one space and one time), but not yet in 3+1 dimensions.

Gravity has no local excitations – gravitons – in 1+1 (and even 2+1) dimensions so many serious problems of "actual gravity with gravitons" do easily hide under the wall of triviality in these lower dimensions.

Still, progress has been made, most notably by Christof Wetterich, who had two groundbreaking papers in the 1990s. More recently, Wetterich used asymptotic safety — just six years ago — to calculate a prediction for the mass of the Higgs boson before the LHC found it. The result? [126 GeV]

Nice. The LHC measures the mass more accurately and according to the newest data, 125 GeV is almost certainly closer to the correct mass than 126 GeV.

At any rate, the root of the argument in favor of 126 GeV above is almost equivalent to the observation that 126 GeV is close to the minimum mass for which the Standard Model may be stable up to the Planck scale. If this coincidence is real, it's really an argument for the derivation of the electroweak theory from a theory that is marginally consistent at the Planck scale. However, the behavior of the actual gravitational degrees of freedom has virtually nothing to do with this coincidence, anyway.

Amazingly, what it indicated was perfectly in line with what the LHC wound up finding.

Not really. The measured value is 125 GeV and the error margin is said to be 0.3 GeV or so. So even by pure numerology, it is not looking too well.

[One extra paragraph of completely irrational hype.]

It’s not only very promising, it has many of the same appealing properties of string theory: quantizes gravity successfully, reduces to GR in the low energy limit, and is UV-finite.

The difference is that string theory actually has a (patch-wise) definition as well as these properties; the existence of the fixed point and its desirable properties is nothing else than a wishful thinking. When it comes to the virtues mentioned above, the difference between string theory and asymptotic safety is the same as the difference between the real world and a fairy-tale. Perhaps, the bottomless mug with the infinite amount of milk in it exists somewhere but the partial evidence we have makes it much more likely that it doesn't exist.

Moreover, the asymptotic safety shares many lethal problems with all attempts to explain gravity in terms of a local quantum field theory (in the gravitating bulk spacetime). In particular, it violates holography as the density of high-energy states fails to be proportional to the surface of the black hole. Thanks to the assumed "exact locality" at arbitrarily short length scales, it also directly contradicts the conservation of the information after a black hole radiates away.

In addition, it beats string theory on at least one account: it doesn’t need a whole crapload of new stuff that we have no evidence for! And that’s why it’s my favorite candidate so far.

What is referred to as a "crapload" are new remarkable predictions and organizing principles and concepts that follow from the theory – they are not being assumed. There is nothing wrong about a theory that predicts lots of things, and indeed, string theory is an example of a highly predictive theory. Quite on the contrary, these new insights are among the main motivations that drives people's research.

If a theory produces no new insights or concepts or predicts no new objects – no new "crapload" – it only shows that it is a worthless, content-free set of musings that don't mean much. It is closer to a quasi-scientific formulation that is just meant to defend people's prejudices. A theory that works and has some internal logic and predictive power unavoidably makes predictions or implies some "crapload" that wasn't obvious from the beginning. If it doesn't, it's too bad. So Ethan's open admission is that he always prefers worthless and meaningless theories that don't teach us anything!

It's very ironic because people like Ethan – and Ethan himself – sometimes criticize string theory for making no predictions. But they also criticize it for predictions because to predict something means to produce "crapload". So which way it goes, Ethan? Are you really incapable of seeing that your slogans are not only wrong but internally inconsistent, too?

Darwin's evolution predicts the existence of various extinct (and otherwise seemingly unnecessary) animal and plant species – or fossils proving that those species existed. You may call these predicted entities "crapload" but that only shows that you are utterly irrational. If/because the theory is correct, the things simply exist – whether you have seen them or not – and you can't do anything about this fact.

Ethan doesn't mention what he exactly counts to his "crapload". He doesn't have to because his text is only addressed to the crappy readers who don't need any evidence or any sentences that make sense. Offensively stupid slogans and catchwords are enough for them. But if he were specific, he would have to count something like D-branes or fluxes or dualities or M-theory or matrix-model description of composite systems in terms of blocks or new formulae for viscosities or non-Abelian gauge groups arising on singularities or giant gravitons or ER bridge visualization of entanglement or lots of other things that we know from hundreds of important stringy papers.

I assure you that if he suggested that these things were a "crapload" in front of a genuine theorist, not necessarily a string theorist but maybe a formally theoretically oriented condensed matter physicist, Ethan would be considered a "crappy loser" instead. These are extremely valuable discoveries and concepts – in string theory and outside string theory – and only a crappy readership could believe that they may be called a "crapload".

4.) Causal Dynamical Triangulations. This idea, CDT, is one of the new kids in town, first developed only in 2000 by Renate Loll and expanded on by others since. It’s similar to LQG in that space itself is discrete, but is primarily concerned with how that space itself evolves. One interesting property of this idea is that time must be discrete as well!

In spin foams, time is basically discrete, too. All these theories are excluded by the Poincaré invariance of the Universe around us, as mentioned above. CDT is really a prototype of the most dysfunctional generic framework that may be used to demonstrate that everything breaks down as soon as one deviates from the rules of string/M-theory that are strictly required for consistency of quantum gravity.

It might be able to explain gravity, but it isn’t 100% certain that the standard model of elementary particles can fit suitably into this framework.

There is absolutely no known scientific reason to think that CDT or any other non-stringy proposal for a theory of quantum gravity could be at least reconciled with, let alone imply, the existence of the non-gravitational interactions.

5.) Emergent gravity. Probably the most speculative, recent of the quantum gravity possibilities, it only gained prominence in 2009, when Erik Verlinde proposed entropic gravity, ...

Gravity is known not to be an entropic force because if it were one, the motion caused by gravity would have to be irreversible – but the orbiting along ellipses is reversible – and the interference in various experiments e.g. with neutrons in the gravitational field (those were done already 40 years ago!) would break down. It's been discussed in numerous TRF blog posts and papers on the archive.

In fact, the seeds of emergent gravity go back to the discoverer of the conditions for generating a matter-antimatter asymmetry, Andrei Sakharov, who proposed the concept back in 1967. This research is still in its infancy, but as far as developments in the last 5–10 years go, it’s hard to ask for more than this.

It's complete nonsense that the research of entropic forces (or gravity) is in its infancy. It's been done for a very, very long time and it's been known to imply irreversibility from the beginning, making the entropic explanation of forces like gravity impossible. It's insulting to (bizarrely) link Andrei Sakharov to these manifestly wrong claims about the entropy. Physicists trained at Moscow State University generally received a very good background in concepts like entropy and almost none of them would ever do as rudimentary mistakes as Erik Verlinde. What Sakharov proposed in 1967 was something completely different, "induced gravity", where the spacetime curvature tensor is a collective field arising from a Bose-Einstein condensate. Such condensates cannot have this huge, black-hole-like entropy envisioned by Verlinde.

So that’s where we stand on Quantum Gravity today, Kent (and everyone).

What Ethan described shows where research on quantum gravity does not stand. However, it is a very good summary of almost all the misconceptions and myths that were spread among the laymen by lousy popular books and newspaper articles in the recent decade.

String Theory is the best studied of the five, ...

Even this seemingly positive sentence is deeply misleading because it makes it look like if the dominance of string theory were some sociological effect caused by having many people. Nothing can be further from the truth. There is about 1 string theorist among 5-10 million people in the world. But the reason why string theory is the only one that is discussed seriously is that despite the tiny number of string theories on Planet Earth, its mathematical and physical properties make it so likely, special, and intriguing. So what the string theorists have uncovered and are uncovering actually has a huge value and can be built upon.

Lots of other people would love to prefer "competitors". The number of people "pushing" for the competitors may easily be 10 or 100 times higher than the number of string theorists. The problem is that the competing theories just don't work so the overall scientific contribution of the "non-string theorists" is orders of magnitude smaller than the string theorists' contribution. They're just on the wrong track where nothing interesting can be found – and they're usually on the wrong track simply because they're not comparably smart to string theorists. If one is capable of studying a physical theory deeply enough and if he actually does so, he may verify these statements of mine. The "competitors" are only being sold by people who have not analyzed the questions carefully enough, usually because they lack the required skill (or integrity). The status of the "competitors" is arguably worse than that of the creationist "competitor" of Darwin's theory. Many creationists have actually learned to work within Darwin's theory. No "non-string theorist" has mastered string theory yet.

The idea that these competitors may describe certain things may be excluded, either definitively, or through a large amount of circumstantial evidence. And at any rate, nothing interesting has ever come out of these proposals.

Thanks to everyone for the submissions for Ask Ethan this week (send yours in here), and thanks also to all who encouraged me to tackle this difficult topic.

Ethan is thanking others for helping him to embarrass himself. I think it is a bad idea for you to write about rather technical things that are brutally outside your expertise – then your texts are bound to be analogously flawed as the texts of the people who believe in warp drives and EM drives everyone around us because these people understand basic physics as badly as you understand quantum gravity or string theory.

You know, all these failures are completely analogous. Just like you are convinced that all the qualitative insights about gravity that we have accumulated – e.g. in string theory – may be avoided and completely different theories could be right, well, the champions of EM drives believe similar things. Anything goes. All knowledge may be ignored. But it is only being ignored by inadequate kibitzers, not by serious scientists.