Another in my aperiodic, quixotic rants intended to improve the quality of H:SE posts.

I've seen a few questions on meta that resemble "why did I get a downvote?". It struck me that I can't be the only person who has a set of pet peeves that will trigger a downvote. If we explicitly list those characteristics of a question that cause our mouse to lunge for the downvote arrow, perhaps we'll reach a stronger consensus on the site we want, and some guidelines to improve questions. I'd like to reach a state where instead of hitting downvote, I can enter a comment that effectively says, "This question is like bullet #x; if you could alter it to be in line with the recommendations for bullet #x, it is more likely to get upvotes and answers, and to cause warm and happy feelings in the cockles of the moderator's hearts.

Let me explicitly note that I do not intend to disrespect anyone in the examples I've chosen below. I'm trying to illustrate a point. While each of the examples has tempted me to hit the downvote button with vigor (and possibly repeatedly), some of them are actually fine questions that I've praised elsewhere. I am criticizing a feature of the question, not the value of the querent.

Please add your own pet peeves; I'll mark this community wiki. Yes, I know that this question violates my own #2. As I said, some questions tempt the downvote, but are actually fine.

Disclaimer; yes, I've said some of this before. It isn't my intent to be repetitive; I'm striving to get to a place where I'm tempted to upvote more often than I'm tempted to downvote. I hope that by changing the format of the material slightly, we can create something that will cultivate a stronger, more interesting H:SE

Please document your preliminary research

Cite every non-trivial assertion in your question.

Don't include anything you don't want to defend.

H:SE questions have an authoritative answer

Has there ever been anything like…

How do I write a good question?

I don't understand your question

I have to read the comment string to understand the question

I will downvote any question that misuses the term genocide.

I wonder if anyone can prove I'm wrong? / You've asked us to prove a negative

If you doubt the existing narrative, the burden is on you.

Is this true that…

Is your question in scope?

It is common knowledge that … knowledge isn't common

Out of scope questions

Questions about race

Quoting known Hitler apologists and/or Holocaust deniers as your primary/only source

Reconcile title with content in body


Someone once said…

Swords were in the past, so tell me about their physics

Terms are ill defined

The question is written in the passive voice

The question relies on a false aggregation

The scope of the question makes it difficult to answer

What evidence is there that the historical individual that I'm thinking of fits a historical theory that I read one time.?

What is the consensus of historians on… ?

What would google tell me if I asked?

When did X first happen?

Why didn't the world community bake me a birthday cake? (or stop Mugabe, or Bush, or Chavez, or whoever)?

Why were the majority of the population X at the time Y ok with Z when we know that Z is wrong? Is it true that Person S from time Y really did/believed in Z??! Horrendous!


You replied in comments

  • 10
    I've asked a handful of questions and honestly I find a lot of the reactions to them a bit arbitrary and unevenly distributed. Either the rules aren't clear, or when moderation makes a decision it's so subjective that it's hard to gauge one's own question. IMO, that's a sign that moderation is too restrictive. Maybe that's the site you want .. but if it is, I don't have the slightest clue how to word a question that meets all of the necessary criteria.
    – Cdn_Dev
    May 22, 2015 at 18:18
  • @MarkC.Wallace I somehow missed the ping for your comment: As the first question in my comment indicates, I know the purpose of CWs. I was and am still not sure of the purpose of this post. Is this intended to be only about questions or should DVs for answers also be covered here? If only Qs, someone like me might edit this Q, If As as well, someone like me might add an answer here… But 20 As seems like an unusual long thread already. Mar 22, 2018 at 12:30
  • Thx. Much clearer to me now. As said, and assuming it's not only your predisposition, and that there will be perhaps quite a few reasons pre-canned, I think a separate "Why was the answer DV'd" would reduce the clutter substantially.. Mar 22, 2018 at 13:40
  • 3
    People are just not educated on this site. Some people are nice, like the person who asked this question, but most of the people are just rude people who find joy and enlightment in downvoting other people. It's just a coward attitude, as if I was to downvote someone, I wouldn't do that without letting at least a comment, and a chance to improve the question. Stack Exchange is known to be one of the rudest community on the Internet.
    – Quidam
    Apr 25, 2019 at 19:16
  • 1
    There are a lot of As to this Q by now. That is good, & complicated if we want to refer to anyone specifically in case of incoming Qs covered by the As below. Therefore, 2 suggestions: 1. refer to this meta-post in comments more often, 2. to help in navigating this post, either link directly to the A here; or even better yet, also prefix all of the As here with a short-title, or abbreviation (not least to potentially save on comment chars). [Alternative solutions for advertising this post, structure & navigation, find-help, welcome] Mar 21, 2021 at 20:25
  • 2
    @LаngLаngС Nice edit.
    – justCal
    Oct 11, 2021 at 19:35
  • @Quidam It's not always worth it to engage with a clearly ignorant answer.
    – cmw
    Sep 22, 2023 at 2:35

35 Answers 35


Terms are ill-defined

This anti-pattern describes questions that rely on imprecisely defined terms; multiple contradictory answers are possible depending on how you define those terms. As a consequence it is not possible to provide an authoritative answer to the question.

For example, "When did the first civilization arise?" Reminds me of Gandhi's quip on Western civilization. "Civilization" is like "pretty" - discussion & comparison are only possible in the context of a shared definition. "Civilization" requires context for utility; it isn't defined in terms of some external objective standard. Economic historians might date "civilization" at the invention of money; Marxists will define it differently.


If a question is tagged with this anti-pattern, the comment should also identify the term or terms requiring definition. OP should edit the question to clarify how the term is defined within the context of the question - what definition will be used to select an authoritative answer?

The question will be strengthened if the definition is linked to a reference or external source - a definition that has been subjected to scholarly scrutiny to identify and resolve problems with the definition.

  • 1
    Strongly agree on this one. I've seen so many questions which I'd like to answer but can't, because I don't know what I'm supposed to be answering. Consequently, there are quite a number of questions with several answers using different interpretations. That said, there are also many questions which do have clear definitions & parameters, but with answers that do not respect what the OP has stated. Apr 17, 2021 at 2:01

I don't understand your question

We are a Q&A site. Built into that is the assumption that the Original Poster(OP) asks a question, and we do research to provide an answer.

If the title doesn't clearly convey a question, if it looks like word salad, or if the syntax is difficult to understand, then skilled and knowledgeable researchers may pass over the question to look for something more interesting.

This also sometimes a signal that OP is groping to find the right question; that OP recognizes that the topic is more complicated and they're searching for the right armature to do research.

In either case, the best approach is to revise the title to ask a question - begin with Who/what/Where/When/Which (be careful with "why"). If you can't ask a clear question, then best practice is to take a step back and ask a simpler question; once that is answered, it may provide the clarity to ask a more complicated question.

But ultimately, your question has to attract interest before it will get an answer.

This is closely related to

  • OP replies in comments (If the OP has to explain the question in comments, it is likely that the community doesn't understand; edit the question to address the comments

  • Question in the passive voice; primary schools and pretentious journals have taught us that academics speak in the passive voice. But outside of pretentious artificial environments, people speak in active voice; they ask questions that indicate what they want to know. If you want me to do research on your behalf, ask me like a person, not like you're talking down to me.


What is the consensus of historians on… ?

If there is in fact a secret cabal of historians that dictates opinions on history, nobody has informed me of it.

This anti-pattern may or may not generate a downvote, but I'm including it here to make it easier to reference.

History is a science, and like most sciences advances through conflict (through the dialectic).

  1. Sometimes there is a consensus (George Washington lived).
  2. Sometimes there is not a consensus. (From a historical perspective, Krishna may or may not have been an appropriated historical figure.)
  3. Sometimes we just don't know. (we had an excellent example of this recently, but I can't find it.
  4. Sometimes there is no authoritative answer to the question. I'd argue that the long running debate on the legality of Southern Secession is in this category, but there are other examples. The answer depends on interpretations and judgments. Depends on the specific definitions you give to terms. While arguments can be made for both sides, there is no test that would allow you to confirm or deny either hypothesis. These kind of questions are a slippery slope towards pilpuls and are dangerous for H:SE because they invite debate and rancor. These slide
  5. Frequently the answer depends on how you define the terms; if the question is based on imprecise terms, then multiple answers are possible.
  6. Sometimes the answer is context dependent (e.g. "What is the definition of “indigenous” for historical purposes?" et. al.) There is no universal definition; there is a definition that works in this context for this article or this study.

And sometimes these questions are actually a request for an argument; these merit a downvote.

How to fix?

If your question possibly fits this anti-pattern, then I would suggest that you

  1. Document your preliminary research
  2. Clearly define all terms
  3. Specify how you will select an authoritative answer. Will you accept an answer of "We don't know", if that answer is supported by strong evidence/reasoning?

We hate fun

Linking here for resource and to invite discussion (!) on what the guidelines for H:SE might be. In particular guideline #3 - few of us use the site for business. What would be the H:SE analogue to #3 (Yes, I'm opening discussion in an answer in the hopes of editing the answer to be useful. On the balance I think this is better than opening a new question and then incorporating the response here)


H:SE questions have an authoritative answer

H.SE is a Q&A site. We are based on the assumption that questions have authoritative answers. While "good question" is a spectrum, the best1 questions are those where any arbitrary participant can independently look at the answers and tell which of the answers best responds to the question.

It's been stated again and again that Stack Exchange is not designed to handle questions with many answers. It doesn't need to be stated at all; it should be obvious from the design. The sorting, the pagination, the Accepted Answer concept; real questions have answers, not items or ideas or opinions. meta.stackexchange.com

We assume that questions have authoritative answers. "Authoritative" means both that the answer relies on authority, and that multiple people comparing the answers given will probably select the same answer as "best".

  1. Subjective answers aren't authoritative. "What do you think of X" or even "Discuss X in the context of Y..." cannot have authoritative answers.
  2. Argumentative answers are usually not authoritative. If you ask "What caused the Great Depression?", you'll get one answer from a monetarist, and a second answer from a Keynesian, and a third answer from a Marxist. There will probably be a third answer from someone still waving the banner of the Austrian school. And if you're lucky you'll get another answer from someone examining the same question through the distinction of the Washington consensus vs the Beijing consensus. Each of them will present research and evidence and clear theories. But they won't agree. There is no way to select an authoritative answer.
  3. Homework questions have authoritative answers, but only the teacher who assigned knows what the answer is. Homework questions aren't designed to test your knowledge of the subject, but your ability to regurgitate what the teacher has curated in their lesson plans. These will probably be closed pre-emptively.
  4. List questions are not authoritative (in the general case). If you ask "Who signed the Declaration of Independence, there is a canonical list and the answer has to match that list. But if you ask "Give me some examples of left handed red headed historical figures", there is, by definition, no way to achieve an authoritative answer. If the first answer gives you examples "A,B" and the second answer gives "C, D", which of those answers is better? What if the third answer is "B,D"? which is the authoritative answer? (and there is never an end to the answers, which is another problem.)

How to fix

Difficult to provide general guidance, but there are a couple of common patterns

Can X be considered Y?

(e.g. "Can Commodus be considered King of the Britons?") This is intrinsically subjective; if the first answer says "yes" with evidence and the second answer says "no" with evidence, both are valid answers to the question.
Revise the question into a hypothesis. A better question would be, "What evidence supports granting Commodus the title "King of the Britons"?" The question text can request that the answer address counterarguments (a good answer will address counterarguments); the authoritative answer is the best support for (or opposition to) that hypothesis.

Question requests a secondary reference

See Are requests for references appropriate on History Stack Exchange?. But one of the core problems with reference requests is that (with the exception of canonical references), what I think is a good reference, may not be a good reference for you.

This is another example of an anti-pattern that may not merit a downvote, but poses problems for history.stackexchange.com

How to fix.

  1. Don't ask list questions; revise the question to ask what you want to know. Why are you making a list?.
  • If you really want evidence to support a historical theory then revise the question to ask about that theory
  • If you must ask about a list, then explain how you'll evaluate the answers and how you will identify the single best answer from all those provided.
  1. Don't ask for references; ask a question about the information you want to find. All answers should include references, which will give you the information you want to find.

  2. Re-read your question and double check that it is possible for someone (other than you) to identify the best, most authoritative answer.

  1. Aside: That doesn't mean that questions without an authoritative answer are bad; many of them are fascinating to discuss. There are questions I discuss in the pub with my mates, and there are questions that I examine in a scholarly dissertation. Neither question is "better" or "worse". Questions with authoritative answers are appropriate for H:SE; questions without authoritative answers should be elsewhere
  • 1
    The latest addition is starting to diverge from official SE policy? Not only are accepted As now unpinned on many sites, 'the design' also explicitly calls for 'competing answers', from which I conclude that 'more than one answer' is desirable? Now, I know that ideally a Q, even on H:SE should_strive to have/get an authoritative answer, 'strive':always, but that is sometimes unrealistic, sometimes _un-desirable? Oct 10, 2022 at 7:30
  • Not quite sure I follow.
    – MCW Mod
    Oct 10, 2022 at 11:47
  • The application of this answer to questions on H:SE (emphasised with the newly inserted quote (checked answers aren't even any longer obligatory pinned) are not always compatible with (H:SE) reality. It reads as if there should be (and only could be, reasonably) '1 & only 1 A'. All of SE has to be able to deal with more than 1, with alternative answers. Of course, one should be come out as 'best', but seldom are therefore the other As 'to be deleted(?)' (& agreement often not reached). (For example for the concept 'good subjective' this answer says: 'impossible by system's design') Oct 10, 2022 at 12:14
  • Fair point and worth of consideration. I think the quoted author is a bit .. polarized., The distinction between "discussion" and "Q&A" is not binary. What I take from the quote is the notion that the quest for authoritative answers is a design feature of SE; it isn't something that H:SE added, it is the way this is supposed to work.
    – MCW Mod
    Oct 10, 2022 at 13:36
  • 1
    "best example of X" would be subjective in many cases. Just to illustrate with your suggestions how to fix a certain question: "best example of 'literary inquisition'?" May 20, 2023 at 12:26

You must log in to answer this question.

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