So, I was going through and flagging new answers to old questions (because that's fun) and it dawned on me, are opinions really answers? The obvious answer is "yes" and that wrong opinions should be downvoted. But let me present an example and then I'll open the floor for discussion and laying of guidelines.

Here's an example question that I just thought up (no doubt it's been asked before but that isn't the point of this):

Q: How can two massless photons interact to form a massive particle-antiparticle pair? And where does all that extra energy for the new masses come from?

Now, there's three distinct types of answers you can get for this. First, there's the answer based on physics, expertise, and referencing:

A1: Pair production occurs through multiple processes. The zero-loop Feynman diagrams (shown below) for this process show the most common of them. The energy for the final pair comes from......... (You get the idea)

The second type of answer combines knowledge and opinion. It is usually from someone that doesn't know the exact physics, but knows enough to make it up as they go along (although it may not be right):

A2: I'm not a particle physicist but I think it probably has to do with quantum mechanics and the energy of photons. Quantum mechanics says anything that can happen does happen and this is something that can happen with a non-zero probability. The energy for the final pair probably comes from the energy of the photons transferred to mass by E=mc^2. And then the momentum of the new particles must be whatever's left over.

The third type of answer is purely opinion and rarely has anything to do with physics:

A3: I thought about this myself and it's my personal belief, which seems fairly obvious once you think about it, is that God had to make all the particles at some point, so he allowed this to be possible so he could get the energy from somewhere. That's the simplest answer, the physicists just have made it all confused to hide that they can't quite grasp how it really works.

Okay, here's my analysis. A1 isn't in question; no opinion, all answer. A2 uses opinion but it also uses physics and intuition to form, what I'd call, a valid if not completely correct answer. A3 is pure opinion. Obviously the thing to do is downvote it and maybe leave a comment to the effect of "this isn't a place for personal opinions". But is it truly an answer?

I'd love to hear what everyone thinks on whether things like A3 are really answers (N.B. By "like A3" I'm not referring specifically to religious stuff. Anything that's so obviously pure opinion and no physics). To start off the discussion, I'll provide two points of view, one for and one against. Perhaps we can discuss and, if necessary, amend guidelines.

Yes, they're answers

Perhaps the easiest to argue, A3 attempts to answer the question. It asks how the process can happen and it gives an answer. The answer is admittedly unscientific, wrong, and generally poor, but it presents an idea that, if true, would sate the asker's curiosity.

To call it not an answer would mean we could flag it and remove it as such. This opens us up to a wide range of potential abuse. Where do we draw the line between purely opinion and an acceptable mix of opinion and physics? This should be downvoted, but not deleted.

No, they're not answers

An answer to the question must do just that; answer the question. The question is asking "How does this happen?", not "What is your opinion of how this happens?". If I asked "Why is the sky blue?" and you responded with "In my opinion, it's blue because it reflects the ocean." I'd say "I didn't ask you what your opinion was, nor do I care what it is. I want to know the accepted explanation, your opinion is irrelevant." (And then I might literally bite your head off too).

This is the essence of what I'm saying. If it asks for the physics explanation, does it really count as an answer if someone gives their opinion? Shouldn't A3 be flagged as not an answer? After all, the appropriate comment to be made is "The OP asked for the explanation of this process not your opinion of what it might actually be." If the question were "What is your opinion of X?", it would be closed as primarily opinion-based. Doesn't that suggest the question "How does X work?" is sufficiently different and that a valid answer to the former would be sufficiently different from a valid answer to the latter?

Okay, there's two points of view. I think both have their strengths. I'm leaning on the side of "Yes", but I wouldn't be making this post if I didn't have strong enough doubts to seriously consider "No". So let's have a fun discussion (no heated debates, please). What do you think? And if you also think A2 shouldn't be considered an answer, I especially want to hear from you.

  • $\begingroup$ Possibly relevant discussion (from "related" to the right): meta.physics.stackexchange.com/questions/627/… $\endgroup$
    – Kyle Kanos
    Commented Jan 27, 2015 at 18:25
  • $\begingroup$ @KyleKanos Oooh, I like that post too. It's good supplementary reading. Too bad these answers wouldn't fall under the pseudo-answer category. Otherwise this whole issue would be solved already $\endgroup$
    – Jim
    Commented Jan 27, 2015 at 18:28
  • $\begingroup$ @Jim, just to see where you fall on it, how would you classify my answer here? I would say it's an A2 answer because I have never once studied/performed/etc MD simulations. But I know enough about computational science to effectively "speculate" why it would be a problem -- as it turned out, the answer is right. But I can't provide more domain-specific information as suggested in the comments. $\endgroup$
    – tpg2114
    Commented Jan 27, 2015 at 20:28
  • $\begingroup$ @tpg2114 I'd say that's an A1.5, maybe an A1.8. But I've never done MD simulations either, so my rating isn't very reliable. $\endgroup$
    – Jim
    Commented Jan 27, 2015 at 20:43
  • $\begingroup$ I wonder what the downvotes are trying to express. Were they disagreeing with one of the stances taken? Or did they really feel this is a worthless topic to bring up? $\endgroup$
    – user10851
    Commented Jan 28, 2015 at 22:01
  • $\begingroup$ @ChrisWhite I was wondering the same thing. I doubt it's the former. It could be that they feel this issue (or one sufficiently like it) has been discussed before and thus it isn't a good post anymore. But who knows $\endgroup$
    – Jim
    Commented Jan 28, 2015 at 22:03

5 Answers 5


I think A2 is an acceptable answer. The votes (up or down) depend on the accuracy (which probably requires knowledge on the actual answer to the question), but I think leaving comments along the lines of This answer could be better if ..." would be helpful to the OP and to future readers (again, this probably requires knowledge on the actual answer).

I think A3 is an unacceptable answer; it should be downvoted to oblivion. I am hesitant to say that it should be deleted because our LQ deletion reasons are:

  • Comments:
    A3 doesn't quite fit here because it is an "attempt" to answer the question, it's just fundamentally wrong.
  • Thanks:
    Also does not apply as the user is not thanking the OP
  • New Question:
    Does not apply
  • URL only:
    Usually not the case, but can be sometimes applied
  • No comment necessary:
    This, to me, is supposed to be used when someone else has already left a delete-comment as above*. So this one too doesn't seem to apply either.

So nothing, as far as I can read into it, in the LQ delete reasons can justify deleting the answer. The Official policy is supposed to be "Looks Okay" when reviewing these (see also What to do with Low Quality posts that you just can't delete, asked here on Physics by Jim, and the link before the LQ delete reasons.)

*Though I have done it in the past, I think using this as a means to "get around" the other comments and delete A3 answers is a little underhanded--unless the community feels this is an appropriate course of action.

  • 1
    $\begingroup$ Kudos for linking me to my own post. Don't forget though that there's also the option of writing your own comment and then selecting "No comment necessary" $\endgroup$
    – Jim
    Commented Jan 27, 2015 at 21:45
  • $\begingroup$ @Jim: There is that, but I am of the opinion that the "No comment necessary" option is supposed to be used if someone else has already made the appropriate comment. Though I've done it in the past, I'm think deleting without a comment is a little dirty/underhanded. $\endgroup$
    – Kyle Kanos
    Commented Jan 27, 2015 at 21:47
  • 1
    $\begingroup$ Could whoever downvote express what aspect they're disagreeing with? (e.g., is it my saying A3's shouldn't be deleted, or that A2 is acceptable, or what?) $\endgroup$
    – Kyle Kanos
    Commented Jan 30, 2015 at 21:52
  • $\begingroup$ Maybe it's the same mysterious downvoter as the 3 I have on the question $\endgroup$
    – Jim
    Commented Feb 2, 2015 at 14:39
  • $\begingroup$ Please could you explain what you mean about the "no comment necessary" reason for not deleting. I don't understand what you mean by "someone else has already left a delete-comment as above." Otherwise, I am in complete agreement with you. $\endgroup$ Commented Jul 19, 2016 at 21:30
  • $\begingroup$ @Sammy: if you aren't the first reviewer and the previous reviewer already left a message about deleting the answer. $\endgroup$
    – Kyle Kanos
    Commented Jul 19, 2016 at 22:18
  • $\begingroup$ Sorry, I'm still not clear about this. Are you saying that if two reviewers recommend deletion, then deletion becomes an option for the Mods? $\endgroup$ Commented Jul 19, 2016 at 22:35
  • $\begingroup$ Not at all. What I mean by that statement is that you can skip leaving a comment if the other reasons apply and someone else has left a comment $\endgroup$
    – Kyle Kanos
    Commented Jul 19, 2016 at 22:50
  • $\begingroup$ @KyleKanos You visited <3 $\endgroup$
    – Jim
    Commented Jul 20, 2016 at 10:42
  • $\begingroup$ @Jim: that I did. I did say that I'd probably respond to comments, Sammy asked for clarification as I was headed out the door from work. $\endgroup$
    – Kyle Kanos
    Commented Jul 20, 2016 at 11:19

I really hate opinion type answers, and indeed many of the more banal answers I post are because I considered an existing answer inadequate and I didn't want the OP to go away with the wrong impression. Since I'm philosophically opposed to downvoting, posting a better answer is usually the only way to show my displeasure with the current answers.

So just in case it's not clear, I consider both A2 and A3 wholly unacceptable. People come to us for an authoritative answer from an expert in their field. That it's possible for a young physics nerd, as I was forty (!!!) years ago, to get answers from a working string thorist is frankly astonishing. Indeed, in my days as a teen physics nerd it was utterly inconceivable. Given what a fantastic service we offer I feel we should strongly resist diluting it.

But, I would not downvote or otherwise sanction any answer given in good faith, unless it was grossly incorrect in which case I normally vote to delete it. If an answer is unsatisfactory then the very best solution is to post a better answer and show how it should be done. If you do this you are not just educating the OP, but you are also educating the person posting the unsatisfactory answer.

  • $\begingroup$ So would you say they count as unacceptable answers, but answers nonetheless? That is, not open to flag or vote for deletion as "not-an-answer", but simply reprehensible. Or perhaps are you in favour of voting to delete all answers of type A2 and/or A3? $\endgroup$
    – Jim
    Commented Jan 27, 2015 at 19:55
  • 1
    $\begingroup$ @Jim: I wouldn't vote to delete (or downvote) an A2 answer, but it would spur me to write a better one. Once a better answer has been written I can see that an A2 answer may have merit as an introduction to the rigorous answer i.e. reading the A2 answer may help beginners appreciate what the rigorous answer is saying. $\endgroup$ Commented Jan 27, 2015 at 19:58
  • 3
    $\begingroup$ I generally try to reserve my A2 answers as comments instead because if I'm not an expert, I don't really want to present an answer as "an answer". On the other hand, sometimes those A2 answers are actually perfectly correct (because I got lucky, or because it does work like something else I do know) and then it's bad because it's an answer in a comment which shouldn't happen. So... sometimes it's unclear on which action to take. $\endgroup$
    – tpg2114
    Commented Jan 27, 2015 at 20:22
  • $\begingroup$ Your Answer illustrates some of the dichotomies of the site : 1. Answers are expected to address the OP's needs (at his/her level), but also to come from an expert and be authoritative according to the needs of a hypothetical future user. 2. Voting is a reward/punishment affecting the rep of the user who posts a Qn or Answer, but also an indication of the community's opinion on the correctness or value of the post. I am not suggesting that these dichotomies need to be resolved, just recognised. $\endgroup$ Commented Jul 19, 2016 at 21:48

My comment under the original question highlights why A2 answers may be acceptable. I asked OP to classify my answer to the following question: Time length molecular dynamics

I would call it an A2 answer. OP ranked it as an A1.5 or A1.8 type answer (in other words, opinion/speculation but fairly strongly supported by evidence from another area). I'm an engineer, and I'm going to say 1.5 and 1.8 have too many significant digits and just call it an A2.

I think this exercise serves to illustrate that things aren't always clear cut with that middle classification of answers. Because it's possible something is 100% correct and if you removed the "I am not certain but..." from the start of the answer, one may never know that the author didn't know.

I'd argue A2 are still answers, and possibly good ones, but they require close scrutiny from the community. This is where John Rennie's opposition to downvoting I think is flawed. It's precisely answers like this that require up, and down, votes to establish accuracy. In my example answer above, the information is correct and accurate (particularly for a "here's an overview to a topic you don't know must about" level of sophistication). But it could just as easily been inaccurate. And people with domain-specific knowledge need to vote accordingly.

The downvotes are particularly important. OP may have upvoted an A2 answer that was not completely accurate because they don't know any better and it sounds close enough. Maybe other people with marginal knowledge of the domain also upvoted for similar reasons. So people with actual, in depth knowledge, must downvote if it is inaccurate.

But I will agree with John Rennie for the rest of his response -- A2 answers should be supplanted by A1 answers from domain experts to fill in the real, gritty details that tend to be glossed over in A2 answers.

Regarding A3 answers, I would flag as low quality (or not a real answer depending on the specific wording), downvote, and go about my day.

  • 2
    $\begingroup$ Right, with the sig digs. I always forget the sig digs $\endgroup$
    – Jim
    Commented Jan 27, 2015 at 21:05
  • $\begingroup$ @Jim Should have called them A1.0, A2.0 and A3.0! $\endgroup$
    – tpg2114
    Commented Jan 27, 2015 at 21:06

Speaking with my moderator hat on ...

I have not been acting on examples of A2 or cases that are borderline between A2 and A3: that's up to the community to establish by consensus.

Outright example of A3 I delete after they have accumulated a net score of around -4 and no significant upvote contribution. That is, those that the community has shown it's disapproval of.

Speaking with my user hat on ...

I down-vote A3s.

Often mercilessly, though I've been known to hold back if I think the post suggests the users (1) didn't know and better and (2) can be taught. Whether I vote or not I am likely to leave a comment summarizing where the answer goes wrong or to upvote and existing comment of that kind.

We want to strongly discourage these kinds of posts.

I'd like to downvote many A2s, but I don't always.

I tend to be lenient if it is the first answer to a question that has gone unanswered for several hours, or if the poster provides a reasonably well reasoned argument for why an analogy with some other physics makes sense.

I'd like to discourage these posts if better answers are present, but if no expert answer is available an example of how another kind of physicists would approach the problem is better than letting it languish unattended for days or weeks. Or so it seems to me.

  • $\begingroup$ With your moderator hat on, would flagging A3s as not an answer and letting the LQ queue sort it out be an acceptable alternative, or do you consider it best that us mortal users just downvote and let you delete it later? $\endgroup$
    – Jim
    Commented Jan 27, 2015 at 20:48
  • 1
    $\begingroup$ I'll revise the last sentence, and flagging so that it goes to the low quality queue is perfect, I just don't take diamond powered action on those posts until they've been buried by collective action. $\endgroup$ Commented Jan 27, 2015 at 20:50

I agree with Kyle Kanos that A2 and A3 are answers, however poor in quality, since they attempt to address the Question.

I agree with tpg2114 that A2 Answers require development. This can be done more conveniently if comments are addressed to this Answer, rather than to the Question.

I think we would all agree that A3 Answers have no value on this site. However if, as Kyle points out, A3 Answers do not fall within the official criteria for deletion - eg because they are comments - then I think they must be - officially - acceptable. This situation is not ideal.

My Suggestion

To get round this, I suggest that acceptable Answers on Physics SE should not only address the Question but also use mainstream physics to do so. A3 would then be unacceptable and meet the criteria for deletion.

This criterion corresponds with the requirement that Questions must be on topic and can be closed and deleted if off topic. Answers must likewise be on topic in my view. If Questions which ask about Chemistry or Mathematics or Biology or Technology etc can be deemed off topic then Answers which do the same - with no physics content - ought also to be off topic.

The difficulty with my suggestion is that fairness probably requires Voting to Close - ie the judgement of one's peers - as with Questions. But there is currently no facility to do so.

  • $\begingroup$ As you pointed out, your suggestion is prone to difficulties in at least fairness. I'd also say there is a difficulty in determining what constitutes appropriate physics. It would be a nice way to determine off-topicness, and I'm fairly on the fence about all of this discussion. For the record, however, how do you define "mainstream physics"? Physics that has been published in papers? Physics currently accepted by a majority of physicists? Does it only include modern physics? Or does historically accepted physics count as mainstream? Where exactly would you draw the line? $\endgroup$
    – Jim
    Commented Jul 20, 2016 at 10:42
  • 1
    $\begingroup$ @Jim : For 'mainstream physics' I would use the same definition we apply in the reasons for Voting to Close questions, where this criterion is already in use. It is best to be consistent between Questions and Answers, I think. $\endgroup$ Commented Jul 20, 2016 at 10:55
  • $\begingroup$ Oh, I definitely agree that consistency is key. And when closing questions, the definition I choose for "non-mainstream" is "unpublished personal theories". Is that the same definition you'd like to apply? If so, I'd say that voting to delete all answers with non-mainstream content is a good idea. However, I seem to recall a time when I raised the question of why we didn't delete them and I got an answer from a mod or something that put me in my place. Perhaps it was on meta, I'll try to find it for you $\endgroup$
    – Jim
    Commented Jul 20, 2016 at 10:58
  • $\begingroup$ @Jim : Perhaps meta.physics.stackexchange.com/questions/5995 as cited by Kyle Kanos at the end of his post above? $\endgroup$ Commented Jul 20, 2016 at 11:05
  • $\begingroup$ No, I don't think that was it. That question was about a post that I knew was undeleteable $\endgroup$
    – Jim
    Commented Jul 20, 2016 at 11:09
  • 1
    $\begingroup$ I think this was it: meta.physics.stackexchange.com/q/6032/23473 $\endgroup$
    – Jim
    Commented Jul 20, 2016 at 11:10

You must log in to answer this question.

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