8

(Making this a separate post on suggestion from Sklivvz)

I have an issue with the "No Theoretical Answers" rule, because in its present wording, if I for instance refer to gravity and hinge my answer on that gravity exists, and that the theory of gravity is accurate, then the post will be deleted.

The problem: good answers get deleted

As stated with emphasis in this post, Skeptics SE do not subscribe to the school of thought "It's just a theory". In science, "theory" is the goal. "Theory" is the thing you get a Nobel Prize for. If in science you say "I have a theory", you rank among people such as Newton, Einstein, Hertz and Maxwell. "It's just a theory" is not a valid means by which to dismiss an answer.

Never the less the rule is an implementation of just that school of thought. The rule and the way it has been applied means that an answer of the sort "Looking at this particular scientific theory, the claim cannot be true" gets deleted under the "No theoretical answers" rule. I find this inherently problematic because to a reasonable person, such an answer is perfectly acceptable.

The users of Skeptics SE are missing out on good and valid answers because of the wording of this rule

Others have also raised the issue. Here is another example. And another one.

The root cause: conflating "theory" with "speculation"

When reading through the text of the rule, I find it obvious that the actual concept that the rule wishes to avoid is not "theory" but "speculation". The confusion comes from that — in daily parlance, by the average person that is not a scientifically minded person — the phrase "I have a theory" actually means "I have a hunch" / "I have a vague idea / "If I speculate a bit".

So the problem is that "theory" can mean two different things depending on context. One of the meanings is good and solid, the other is not.

I agree with the spirit of the rule: we do not want answers based on loose speculation. But I do not agree that scientific theory should also be tossed out, like the proverbial baby with the bath water, because then we lose good answers.

Proposed edits to resolve the issue

Here are a few edits I suggest to remove the issue and make it clear that theory is all right, but speculation is not. The strikeouts is the old wording, the boldface is the new wording.

Headline

FAQ: What are theoretical speculative answers?

Ingress

One of the premises of skepticism is the application of the scientific method: empirical proof evidence validates or disqualifies theoretical models. All questions we allow here We only allow posts that are empirical material in nature, thus answering via a purely theoretical model speculation is inappropriate: experiments are not "validated" by theory, but vice-versa.

Here is a list of common examples of types of unacceptable theoretical speculative answers.

Section "Back of the envelope calculations"

Answers based on simplified calculations instead of measurements are theoretical speculative. By their nature such calculations implicitly assume a mathematical model, but they generally fail to show that the model is adequate to the circumstances of the question. They also do not investigate their own inaccuracy. They are a form of Original Research.

[...]

Section "Research-level answers"

My suggestion is that this entire section is lifted out to its own rule, one that states that "Answers must be accessible to the audience of Skeptics SE". Stuff that requires an academic degree to take in are not accessible to the general public of Skeptics SE.

Section "Pure logic/pure maths answers"

Answers that rely only on logic and maths are theoretical speculative, because they do not connect the material nature of the question with the immaterial nature of the answer. All our questions are inherently referring to experimental evidence material reality. If your answer does not contain any material evidence, it is almost certainly not answering the right question.

Section "Common sense answers"

This section looks good as it is to me.

6
  • "Maths" is correct in many (if not most) dialects of English, so I don't think it's necessary to include that change. Also, perhaps because I can't see any of the deleted posts you reference, I don't see any evidence that there's any actual deletions happening specifically because of the difference in definition between scientific theory and "plain English" theory.
    – Kamil Drakari
    Jul 27, 2018 at 17:36
  • @KamilDrakari The word "Maths" (short for Mathematics) is present in the current wording. I have not changed that. Things I have taken out are marked with strike outs, additions with boldface. And I can assure you that posts have been deleted for pointing out that a claim would violate scientific theory if it was true. That fact that you have not dug hard enough to find them is not an argument.
    – MichaelK
    Jul 27, 2018 at 17:42
  • Under 'Section "Pure logic/pure maths answers"' in the blockquote where you made your proposed edits, the 's' in "maths" has a stirke out. I guess it was unintentional?
    – Kamil Drakari
    Jul 27, 2018 at 17:52
  • @KamilDrakari How silly of me, you are right. Fixed now.
    – MichaelK
    Jul 27, 2018 at 17:57
  • 2
    I think the concept of rejecting 'speculative' rather than 'theoretical' answers is a promising idea. In my opinion, the ban on answers that apply scientific theories is doing a fair bit of harm to this site. Using theoretical, 'back of the envelope' math with general scientific principles to determine if something is obviously impossible is one of the most useful tools a skeptic has.
    – Ask About Monica
    Aug 9, 2018 at 22:56
  • It's also unequally applied but that's part of a larger issue.
    – fredsbend
    Oct 29, 2018 at 14:44

1 Answer 1

Reset to default
1

Thanks for your proposal. There are good bits that are certainly worth investigating further, and some other parts that are a bit puzzling.

There are many things that we need to address together, so let me try to clarify a bit why things are how they are so we can see how to proceed.

Here are a few examples of answers that are notoriously problematic and we need to be able to nuke.

  1. Totally speculative

Q: Does it take 10s for a coin to drop from the Empire State Building?

A: Yes, I think it's about correct.

  1. High school physics

Q: Does it take 10s for a coin to drop from the Empire State Building?

A: No, sqrt(381 m / 9.8 ms-2) = 2s

  1. University physics

Q: Does it take 10s for a coin to drop from the Empire State Building?

A: {a huge series of formulas which include air resistance coefficents etc.}

  1. Back-of-the-envelope

Q: Does it take 10s for a coin to drop from the Empire State Building?

A: No, it falls at terminal velocity (~10m/s) which means more than 30s.

  1. Home experiments and andecdotes

Q: Does it take 10s for a coin to drop from the Empire State Building?

A: No, I went the other day and timed it at 22s

  1. Broken reductio ad absurdum

Q: Does it take 10s for a coin to drop from the Empire State Building?

A: No, t = sqrt(s/a) so if that were true then gravity would be a mere 3m/s-2!

Some notes:

  • all these answers might contain links and they would be allowed if these links are the source of the answers. If the links are merely to the height of the Empire State Building or the high school physics formula book page, they are not allowed.

  • often the difference between 2, 3 and 4 is hard to tell, which is why they are bundled together. We want to get out of discussions whether an answer is at a high school level or not. All levels are bad.

I don't know which answers we are "losing" as you state since we don't really want any theoretical answer. Which of these are covered by your proposal?

22
  • 1
    "We want to get out of discussions whether an answer is at a high school level or not. All levels are bad." That is just horrible. So you are saying that rule exists just to avoid discussion?! That we are excluding some definitive and valid answers just to avoid the gray area cases? I am appalled, truly I am. Whatever happened to Stack Exchange being "all about getting answers"? This is amputating the foot to fix an ingrown nail.
    – MichaelK
    Jul 28, 2018 at 2:24
  • Counter-example: if someone claims "A coin falls from the top of the Empire State Building in 1s", then we can use physics to give a definitive "No" to that question. But you say we may not answer that because there might be a controvery in the 10s case, and we do not want to have to deal with that? That is terrible! I challenge you to show me at least a handful of examples of these "10 second coin drop" type of questions there have been, and controvery avoided "thanks to" this rule, and compare that to valid answers that have gotten tossed out. I say the loss is worse than the gain.
    – MichaelK
    Jul 28, 2018 at 2:37
  • 1
    @MichaelK: We have a kind of natural experiment. Here is Quora which doesn't have our rules. I think only one answer would pass here.
    – Oddthinking Mod
    Jul 28, 2018 at 8:13
  • 2
    @Oddthinking What is your point? "Any rules are good riles because having no rules is bad"? I say again: show me posts where Skeptics SE have benefitted from being this restrictive. The rules are supposed to make SE better, not be a cheap get-out-clause for its moderators. Also I think that this upnosed attitude of... "* sniff * We are so much better than Quora or Reddit" is distasteful. The aim of SE is not to be the biggest snob on the hill.
    – MichaelK
    Jul 28, 2018 at 8:24
  • @MichaelK: Please remember the Be Nice rules.
    – Oddthinking Mod
    Jul 28, 2018 at 8:48
  • @Oddthinking I am nice. I am also bringing forth critique. I honestly think that Skeptics SE have a problem. The rules are heavy-handed and strike down good answer for no or little benefit. I also think there is an air of snobbery on Skeptics SE. I mention no names because I think we all carry a shared responsibility in it. The goal of SE is not "to be better at this than Quora and Reddit", meaning that every reference to that is just an empty argument. The goal of SE is to be — as clearly stated — "all about the answers". We miss that goal if we strike answers for no good reason.
    – MichaelK
    Jul 28, 2018 at 8:55
  • 2
    If you claim that the present wording is necessary to prevent a morass of weak answers and accompanying controversy that puts a high burden on moderators and lowers the quality of Skeptics SE, then I respond to that with: present evidence for the statement. Show us examples — from Skeptics SE, not Quora — where we would have had such problems if it was not for the rule that let us prevent it. To be a wee bit of a wise-cracker: until you show evidence for that claim... it is only theoretical/speculative.
    – MichaelK
    Jul 28, 2018 at 9:09
  • @MichaelK first of all: I don't mind critique. Second Skeptics doesn't have a problem: you have a problem with it. Which is fine, interesting and useful -- if framed correctly. The large majority of community and answers benefits from the rules that annoy you.
    – Sklivvz
    Jul 28, 2018 at 9:29
  • We cannot show you the kind of content you want to see, because it's deleted content. Get 10k rep to see it, like all other users. That said if you have in mind another form of evidence that would convince you, we are here to help.
    – Sklivvz
    Jul 28, 2018 at 9:30
  • Third: yes discussions due to rules which are not readily understood by the users are bad. I thought this was also the point of your suggestion -- which according you your own statement, is due to a few meta examples of discussion. I can assure you that there are hundreds of discussions in comments we avoided.
    – Sklivvz
    Jul 28, 2018 at 9:34
  • Fourth: here's some evidence, a question with 8 theoretical answers based on "theory" which were wrong and their OP refused to fix, to which I've had to personally answer with a proper, science based answer. I was simply late to intervene there: skeptics.meta.stackexchange.com/questions/4124/…
    – Sklivvz
    Jul 28, 2018 at 9:37
  • 2
    @Sklivvz "We cannot show you the kind of content you want to see, because it's deleted content." Yes you can because you can Copy & Paste into an answer. " I can assure you that there are hundreds of discussions in comments we avoided". I am not interested in assurances. Show me the evidence. And show me that these discussions could not have been avoided with my suggested wording. Because all I have right now is the bare assertions from two moderators — that wrote the rule — that say "You are wrong. Everything is fine with the rule. We do not need to do anything about it".
    – MichaelK
    Jul 28, 2018 at 10:16
  • 1
    -1: Because of the "Get 10k rep to see our evidence" - that's not how skepticism is supposed to work. This response comes across overall as "just asking questions" rather than an honest attempt to answer OP's request with a yes or no. Edit it to actually say yes or no and then we can talk about whether it is right. But for now, it's not even wrong.
    – Kevin
    Oct 28, 2018 at 19:55
  • 1
    @matt_black what we are getting are actually wrong answers because the assumptions are wrong (there is air resistance after all) -- but I agree IF the assumptions were stated correctly AND verified then it would be acceptable in some cases to have a theoretical model. Then the problem becomes who has the skills to validate that the model is correct, the assumptions are valid and so on?
    – Sklivvz
    Dec 7, 2018 at 21:00
  • 1
    I am a trained physicist. I know how to make these models, and I can tell you for sure that the vast majority of such models here are of terrible quality. Heck, even on Physics there are a huge amount of mistakes in the answers and the models provided! The reason is that finding these mistakes is hard, and the Stack Exchange model tends to favor immediate judgements instead. So software answers are fine: just try out the code and see if it works. Complex models, instead, can only be ultimately validated by experiment and observation.
    – Sklivvz
    Dec 7, 2018 at 21:03

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .