# Reserving the use of the 'non-mainstream' close reason

In my opinion, the purpose of the 'non-mainstream' close reason is for non-mainstream theories. It's for people who have invented their own theory of everything, put it on Vixra, self-published a book about it, and want to have us look at it. This is a legitimate reason to close, as we get a couple of these people every day, and if we don't push them away quickly they'll just overwhelm the site.

But over time the scope of this close reason has increased to include naive questions. For example:

These are perfectly fine, if misguided questions from laypeople who want to know about physics. They can be answered within mainstream physics just fine, even though the answer will be along the lines of "this is not a valid thing to ask, because...". Can we be a bit more careful about closing questions like this?

• I concur. Besides, "mainstream" is a blunt instrument, and one that is sometimes abused. If you're critical of time travel, worm holes, multiverses etc, some people will say "that's not mainstream". But talking of overwhelming the site, the naïve questions are doing that, so maybe reviewers sometimes pick the wrong reason because they follow some other guy and they're in a hurry. – John Duffield Apr 7 '18 at 14:15
• As with many other “policies”, they are functionally guidelines as different people may have different interpretations, and actually depend on the quality of the question. The buffer against abuse is the threshold of $5$ votes to close so I’m fine with the idea that, if $5$ people find a post to be questionable without making a fine distinction between “misguided” and non-mainstream, the question can be closed. In case of real controversy, it can always be reopened. ... not saying I always agree with the VTC, but if $5$ agree it’s usually close enough for me to accept the decision. – ZeroTheHero Apr 7 '18 at 15:00
• @ZeroTheHero "they are functionally guidelines as different people may have different interpretations, and actually depend on the quality of the question" -- while this will always be true to some extent, we should strive to diminish it when possible, not use it as an excuse. That is precisely the purpose of these meta discussions. – Nathaniel Apr 10 '18 at 13:00
• @Nathaniel I absolutely agree. I'm intaking the position of others as myself I tend to be somewhat radical. – ZeroTheHero Apr 10 '18 at 13:07
• @Nathaniel It seems to me that while not all non-mainstream questions are crackpot all crackpot questions are non-mainstream. My experience is there are very few "good" non-mainstream questions so I'm not surprised that the distinction between "non-mainstream" and "crackpot" has been blurred. – ZeroTheHero Apr 28 '18 at 15:50
• I'm not sure how that relates to my comment. – Nathaniel Apr 29 '18 at 0:00

When it comes to non-mainstream topics, such as, e.g., faster-than-light travel, tachyons, time travel, warp-drive, wormholes, antigravity, multiverses, parallel universes, pre-big-bang, etc, the question should be of high quality, well-defined, well-researched, well-documented, and preferably referring to a specific theory, before Phys.SE should accept it.

The issue is that Phys.SE's reputation as a trustworthy & reliable physics Q&A site is at stake. The Phys.SE community must draw a line between mainstream & non-mainstream. This is best maintained by not being a soapbox. The Phys.SE community cannot rely on hoping that potential answers will always correct non-mainstream questions & views.

• The key distinctions seem to be soapbox (not acceptable even when pretty mainstream?), the not-even-wrong questions, the speculative but real physics based (wormholes), and the actual non-mainstream physics. I think it would be useful to have a bit more fine-grained way to describe the close. – Anders Sandberg Apr 7 '18 at 21:39
• This seems not to be addressing the question asked, which is about questions that are naive, explicitly not the soapbox ones. – Nathaniel Apr 10 '18 at 13:00
• @Nathaniel Qmechanic’s answer is very much “barbarians at the gates”. After seeing some of the comments on this question, some recent questions and answers on the main site, and thinking about it some more I changed my mind and agree with his general philosophy, hence the accept. – knzhou Apr 11 '18 at 10:35
• @knzhou that's your prerogative of course, but for me the last paragraph makes no sense in terms of the questions you're asking about. Yes, we should not be a soapbox, and yes, we should not hope that potential answers will correct non-mainstream views. But naïve questions are neither soapboxes nor the result of non-mainstream views, they are just requests for information. – Nathaniel Apr 11 '18 at 11:01
• You list a bunch of topics that you call "non-mainstream," but actually some of them are perfectly reasonable topics in relativity. Tachyons have been extensively studied in a variety of theoretical contexts, and there have been experimental searches for them. "Time travel" is not a phrase you see in the literature, but only because people use the term CTC. Wormholes have been extensively studied by relativists. "Warp drive" is Star Trek, but the Alcubierre metric is in some ways conceptually similar. The fact that your list is so inaccurate suggests to me the danger of false positives here. – Ben Crowell Apr 11 '18 at 15:22
• @BenCrowell even if there are false positives, there's still the reopen queue that would remedy the issue. Likely non-optimal, but better than permanent closure. – Kyle Kanos Apr 12 '18 at 11:07
• @BenCrowell I think Qmechanic's point is that it doesn't matter if the Alcubierre drive is technically mainstream, it comes with so much popsci baggage that it would decrease the overall quality of the site to count it as mainstream. Actually, this goes for a lot of things in GR. – knzhou Apr 18 '18 at 16:31
• More than "general comments" they sound like commandments. I'm part of this community, and I'm open to any kind of questions, I'm happy with not drawing a line between mainstream and non-mainstream, and I rely on the usefulness of the answers to give insights to a questioner. It seems that StackExchange would just become a "Technical FAQ" otherwise. – pglpm Apr 21 '18 at 4:36

I think you may be right. I'm not sure the scope of the close reason is meant to be quite as narrow as (it sounds like) you're proposing here, but it is good to be careful not to overuse it.

We have an FAQ post describing the use of the non-mainstream close reason. And for further context, I went back and dug up the original proposal of this close reason as well as the chat session where we hashed out the wording. The main focus of all these resources is on this site not being a substitute for peer review. A question that asks us to broadly evaluate the general correctness of some idea, in the manner that a reviewer would, is supposed to be off topic. There's a notion of broadness there, in terms of what is being asked, that may have been neglected over the years, and we should probably pay more attention to that going forward.

Part of the issue is that questions about fictional physics (i.e. questions that start from a premise that violates a theory and then asks us to use that theory to predict a result) are also meant to be off topic, as described in the help center. We don't currently have a standard close reason that applies to these questions, so they wind up getting grouped under non-mainstream physics. Perhaps it's worth having a separate discussion about how to handle those questions.

• I like this question and answer. Recently, I was closed up a question about multitemporal theories (that are not originally mine, despite the fact I think I am one of the few who wrote about them on the blogsphere!). Even, I felt that multitemporal theories should not be considered non-mainstream. There are even studied by people in string theroy (F-Theory, S-theory,...) or were studied by solid researchers like Arefeva et al., who "explained" the observed small cosmological constant with a two-time theory. I see the non-mainstream closing reason a censorship tool from time to time... – riemannium Apr 9 '18 at 20:26

I'd like to address your second example (here for 10k+; screenshot) and questions along that vein. In a nutshell, that question follows the broad strokes of a rather common class of questions that get closed as non-mainstream and which basically read

I know that the laws of physics say that X can never happen.

Suppose that X happened. Then what do the laws of physics predict as a consequence?

Frankly, I have no sympathy at all for those questions, and their closure is simply a good riddance. We're here to discuss physics, not hypotheticals based on some ill-defined concept of how the laws of physics might be modified to indulge Random Internet User's idea of what reality should be like. But more importantly, this class of question is, by construction, a contradiction, and we have much better places to funnel the time and attention of this site's userbase than to threads that are fundamentally unanswerable.

And, similarly, while the non-mainstream close reason's original reason for being was along a different vein than these ones, the overall spirit (that this site is not here to be all things to everyone and that in order to have high-quality content we explicitly restrict our scope from considering all possible variations of the laws of physics) fits these counterfactuals quite closely. I think it's perfectly well-applied here.

This sort of censorship over non-mainstream theories always makes me uneasy.

Fallacies in science are discussed and resolved using logic and experiment. Not using censorship. Censuring is just putting a lid on the fallacies, and they'll keep on growing under the lid.

Censuring is like saying "Shut up". Well, that's not a scientific argument. A scientific argument is "What you're saying is wrong, as shown experimentally/logically by X in section Y of reference Z". Arguments like "Shut up", or "You're wrong, because X said otherwise", or "You're wrong, what you said appeared in a non-respectable journal" show a lack of scientific arguments.

We usually condemn the Church for making Galilei retract his theories, and condemn the Church's censorship and its Index librorum prohibitorum. And then go on doing this kind of censorship ourselves. We think we have the right to, because we see mistakes in non-mainstream theories. We are right, they are wrong. But the Church censored Galilei because they saw mistakes in his theories too – they gave reasons why his works had to be retracted. They were the "peer-review" of the time. They represented what was accepted and acceptable. In their eyes, they were right, he was wrong. Do we lack a historical sense?

I prefer the point of view attributed to Voltaire: "I wholly disapprove of what you say – and will defend to the death your right to say it".

An important message in Galilei's work, and in the work of others from that period, like Francis Bacon and Giordano Bruno, was that every human being has the right to openly put forward a thesis, and other human beings have the right to approve it or criticize it and argue why it is false, and in turn put forward their own. And every human being has the right to examine this kind of exchange, think about it, and come upon their own conclusions. There's only one true peer reviewer, and that is your own brain. Accepting or rejecting a scientific thesis just because of someone's authority, whoever that might be, is not science, it's religion – we're back at the time of "because Aristotle said so".

Sure, we can say "people who support non-mainstream theories can very well put them elsewhere on the internet, nobody forbids them to do so", and give some reasons why they shouldn't appear on StackExchange. Yet... I don't know, to me it sounds like "democracy is very difficult to handle, so let's instore a dictatorship instead".

When we think of "peer-review" I suppose we ideally think of something like this: someone proposes a theory, and the whole scientific community examine it and "votes" in its favour or against it, from logical and experimental consideration. But actual peer-review doesn't work that way. It's about 1–4 persons making a decision. Statistically speaking that's not a representative sample of the whole community (on top of that they sometimes disagree with one another).

I apologize for the fuzziness in these thoughts. It's an uneasy feeling that I can't put properly into words, related to this kind of discussions about censorship...

By the way, I see a lot of answers in Physics Stack Exchange that don't give any references at all.

• We're not the Church prohibiting Prospective User X from publishing their thoughts anywhere online or offline (which is what happened to Galileo), we're a single specific publisher refusing to carry certain specific subjects. All modern scientific publishers (or at least, the ones that have a reasonable chance of being taken seriously by any meaningful audience) do this. – Emilio Pisanty Apr 21 '18 at 19:33
• Moreover, I have yet to see any proponent of the “speculative” theories produce experimental evidence to support their claim. This is very much unlike Galileo, where the people refused to accept his experimental evidence - v.g. Galilean satellites of Jupiter showing not everything moved around the Earth. – ZeroTheHero Apr 21 '18 at 20:10
• @ZeroTheHero Exactly, and that's how "speculative" theories are countered: by openly showing their logical fallacies or experimental shortcomings. Not by censorship. "Shut up", or "I'm right, you're wrong", or "Nobody believes you", or more long-winded verbose statements that amount to the same as these, are not scientific arguments. – pglpm Apr 21 '18 at 22:11
• @pglpm I respectfully disagree. It’s the other way around: the onus is on the supporters of the speculative theories to show these solve some unexplained data. If not then people will spend their time forever disproving groundless speculations. – ZeroTheHero Apr 21 '18 at 22:13
• @ZeroTheHero So I suppose you include some high-energy theories or theories about "many-worlds" among those. – pglpm Apr 21 '18 at 22:18
• @EmilioPisanty Voltaire supposedly said "I wholly disapprove of what you say—and will defend to the death your right to say it". But obviously we're more illuminated today than he was. We aren't the Church, we are the Pontius Pilate of the situation: "Such speculations can be published elsewhere; I don't want to deal with them". – pglpm Apr 21 '18 at 22:22
• @pglpm predictions are by their nature speculative and one may or may not have a positive view of some high energy theories, but these theories are based on methods proven to be fantastically successful. Even those willing to call such work speculative agree that the field theory premise is proven true. – ZeroTheHero Apr 22 '18 at 0:08
• @ZeroTheHero So you're adding an additional criterion for a theory to be worthy of consideration, besides "solving unexplained data". You see, that's the problem: we tend to add this or that additional clause as need arises, just to keep the theories we like in, and those we don't like out. What are the methods you refer to? and who are the people you refer to? In the end the criteria we give looks very circular; are they different from "Because I like it" or "Because I say so"? – pglpm Apr 22 '18 at 6:08
• @ZeroTheHero Theory judgement is tricky. For example, the Ptolemaic theory can explain all the data that the Keplerian theory can (the former is like a series expansion of the latter). The aether theory + Lorentz-FitzGerald contraction can explain all the data that special relativity does (cf. Dirac, Is there an æther? Nature 168 (1951), 906) . The reason we initially chose Kepler's and Einstein's theories is not that they explained unexplained data, and their methods were controversial. Many of today's mainstream theories started as "speculative", some were even mainstream-censured. – pglpm Apr 22 '18 at 6:34
• The format of the site is "Question & Answer". How do you propose we run the questions and answers if the answers are unknown because the questions haven't been studied, or verify answers if they are based on concepts that physicists aren't familiar with? – JMac Apr 24 '18 at 12:37
• @pglpm Your idealism is touching but it speaks much more loudly like the viewpoint of someone who's not actually had to do what you're proposing or seen its draining effects first-hand. This site has finite resources (in terms of people, expertise, time, energy, and reputation among a broader public) and what you're proposing is that we squander those resources on trying to educate an audience that has shown, time and time again, that it's very rarely interested in honest scientific debate. Call me misguided but I would rather the site focus on quality physics instead. – Emilio Pisanty Apr 25 '18 at 13:34
• @pglpm - by definition, based on the SE rules, yes, the people of this site do get to choose. I would add that your focus on references just moves the discussion to what a quality reference would be, which for most physics communities would be answered with a similar 'we know them when we see them' (and 'we know crap when we see it'). – Jon Custer Apr 25 '18 at 19:53
• @pglpm I'm sure you have noble intentions, but they're just too naive. Who's to decide any kind of standards in any society or group whatsoever? Are you against the existence of any laws whatsoever? – knzhou Apr 28 '18 at 16:19
• @pglpm Once you get past very small groups, 'unanimous agreement' becomes completely impossible as a mechanism for deciding laws. On a site like this, not even 'majority agreement' (i.e. democracy) is feasible, because users who know physics (~1000 of us) are massively outnumbered by users that don't (100k, and more if you count HNQ visitors). If we listened to them, we would discuss absolutely nothing but time travel, warp drives, and the multiverse. If this were allowed, all actual physicists would quickly leave, and we'd degenerate into a cesspit like Worldbuilding.SE. – knzhou Apr 28 '18 at 16:20
• @pglpm Instead, "who gets to decide" is the set of users that get 3k reputation. That's not a very high bar, given that you can get halfway there with a single popular question or answer. It's low enough that even posters who know no physics at all can get there in a few months. In this question, I am talking to this set of users. – knzhou Apr 28 '18 at 16:23