There are a number of Meta posts about flags, both here on Meta.ELU and on Meta.SE.

  1. Do we have any guidelines for flagging comment as offense? (mELU)
  2. Why can’t I flag a question as offensive? (mELU)
  3. What is considered hate speech? (mSE)
  4. Official guidelines for flagging answer as offensive/hate speech (mSE)

These are quite old, and things may need restating or may even have shifted a bit. What’s the current guidance on how the moderators will deal with flags?

1 Answer 1


In this post, the ELU moderators attempt to consolidate all the advice in order that: it’s relevant for the English Language & Usage Stack Exchange site; it’s all in one place; it’s easily assimilated; and it provides a starting-point for future moderators as well.

Because this site actually discusses language, there are some quirks and edge cases which might not apply elsewhere. This particularly refers to flagging different categories of unwanted content.

While a lot of what’s here can be applied to Chat, that’s a fast-moving and less formal environment which will be dealt with separately. This post applies to the main ELU and Meta ELU sites.

The general rules

  • This is an academic site about language.

    Keep any content academic.

  • Civility is required at all times.

    Rudeness (for example in ad-hominem attacks) will not be tolerated.

  • Worldview is irrelevant to discussing language.

    We are international and cosmopolitan. You will encounter many people who think differently from you, perhaps even a majority. However, worldview is not relevant in a discussion about language, and need not be mentioned at all.

    Occasionally it may be necessary to accept the existence of a particular philosophical position in order to discuss the language of that philosophy objectively.

Practical applications of the general rules

Let’s create, for the sake of argument, two groups of people: those who like the colour blue, and those who do not.

It’s OK to disagree about liking blue. It’s OK to say you do or do not like the colour blue. It’s even OK to say that you don’t like people who don’t like the colour blue.

However, it doesn’t need to be mentioned at all if it’s not relevant to the question or its answers.

Comments can’t be altered, so flag a comment which is objectionable. Every flag will be looked into, but it helps the moderators if you use the right flag. Even if you use the wrong flag, if a comment needs some sort of action, the flag will be helpful; but the moderators may suggest that a different flag might have been more appropriate.

Posts (Questions and Answers) can be edited, so you may not need to flag at all.

Offensive language in comments and posts

  • What’s offensive is threatening behaviour, ad-hominem attacks, and using curses and swearing to intimidate.

    I think everyone who likes the colour blue should be shot.
    You are a typical bluey, and I wouldn’t even be surprised if you had blue blood.
    Just the thing a fucking blue-baby would write.

    Offensive flag

    “Offensive” flags should be limited to these three types: direct threats of, or support of, violence against the person; direct insults; swear words.

    You can edit such language out of Questions and Answers.

Comments which are otherwise inappropriate

  • What’s not constructive is suggesting that someone’s question or answer is good or bad based on some feature that is not relevant.

    Well, of course all blue-lovers think like this.

    The line between what’s “not constructive” and “rude or offensive” can be a fine distinction to make. Looked at objectively, this comment is a general comment about a group of people. It’s disparaging, but it’s not violent or hateful, a direct ad-hominem attack or swearing.

  • What’s too chatty is a comment which opens up discussion.

    omg, I love blue too! Have you seen that movie about the blue notes? So awesome.

  • Obsolete comments are no longer relevant, for example when a request for clarification or detail has been answered and the comment can be deleted.

Don’t forget that comments are intended for requesting clarification of a question, or (perhaps) pointing out a flaw in an otherwise good answer. Don’t engage in debate in comments: if you disagree with an answer, write your own!

Inappropriate language in Questions and Answers

All posts can be edited, and even members who cannot edit directly can suggest edits which are submitted for approval by higher-rep members.

If a worldview is expressed which is not relevant to the question, and it can be removed without destroying the post, edit it out.

Jones is behaving like a bluey objecting to perceived injustices.
Jones is behaving like someone objecting to perceived injustices.

The original poster might reject or rollback the edit. If that happens, don’t engage in comments, or re-edit the post starting an edit war. Flag the post as “needing ♦ moderator attention” and explain what has happened.

This isn’t a direct threat against the person, a direct insult, or swear words. Don’t use the “offensive, abusive or hate speech” flag for this sort of thing: the use of a bluey in the post is reflective of a particular worldview which is irrelevant and non-constructive and easily dealt with by a small edit. Just make the edit.

If you do use the offensive flag for something in a post which does not fit the definitions and is actually not constructive it could be declined.

Offensive posts which do contain swear-words, ad-hominem attacks or threats and which are dealt with using this flag carry a 100-rep penalty. For this reason, marking a post as offensive should be reserved for serious infractions where the post cannot be edited successfully. There’s more on this below.

Questions which are actually about offensive language

ELU needs to be able to discuss language, including offensive language. The and tags exist to assist with this.

This is one of the edge cases. It’s necessary to write offensive language in order to discuss it, which is where the use–mention distinction comes into play. Mentioning offensive words where it’s contextually necessary is not offensive per se; it’s part of the academic discussion of language.

When you are mentioning a word rather than using it, put it in italics. This helps people who read your post to put your use of the word into context.

The use of fucking in an earlier quote was an example of offensive language.

For titles, put mentions in double-quotes. Titles are already in italics, and double-quotes are prettified. It’s customary to play safe with titles and sanitise any that might be Not Safe For Work.

The use–mention distinction applies to all mentions, not just offensive words. Of course, using italics in an attempt to disguise offensive language as a mention actually emphasises it.

Questions which are predicated on a particular philosophical viewpoint

ELU needs to be able to discuss the language associated with various philosophies.

The way to deal with these questions is by following the general rules. Taking a particular example which was raised in one of the earlier Meta posts, there might be a question

The Hebrew word for God Elohim is plural, and the Christian God is a Trinity, so why doesn’t God ever take a plural verb?

This question assumes the validity of a philosophy that God exists. It may be a philosophy that you profoundly disagree with, but the general rules are

  • This is an academic site about language.
  • Your worldview is irrelevant to discussing language.

Questions about language should make it clear that they are about language. In the above example, the asker might be an atheist but still want to know about the language used. In this case, an answer or a comment along the lines of

“God” doesn’t exist, so discussing grammar is futile

is completely non-constructive. There is nothing requiring you to comment on or answer a question where you disagree with the philosophical premise.

A question, or equally an answer, might betray a particular philosophical worldview, perhaps even one which could be expressed in offensive terms.

What’s the word for someone who thinks Martians should be killed?
I hate Martians, and I think they should all be rounded up and gassed. My neighbours call me a Martiophobe. Is -phobe really the right suffix for someone who doesn’t actually fear what’s being referred to?

Rather than flagging, edit the post if it’s possible to make it obviously about language. In this case, quite a major edit is necessary to turn it into an objective question about language, but even an edit like this comes in at under the 400-character limit for anonymous users.

Does “-phobia” mean “dislike or hatred” or “fear” of something?
Is a martiophobic statement one which shows a fear of Martians, or one which expresses a dislike or hatred of them? I looked up the origin of -phobia and apparently it comes from the Greek for fear. But it seems that it’s often used about statements which don’t express much in the way of fear of Martians. Why is the word martiophobe?

If you don’t feel confident about making this sort of edit, flag for ♦ moderator attention. The moderators will look at the post, and edit it if possible. If it’s entirely irretrievable, they will mark it as offensive and the 100-rep penalty will apply.

If you do make the edit and it’s rolled back, that’s the time to flag as offensive, rather than start an edit war.

Other flags on posts, just for completeness

Flags on questions

Other flags on poor questions are likely to be self-explanatory:

  • Spam

    A link or advertisement unconnected with ELU and any question or answer in which it appears. Six spam flags on a post will ensure it’s deleted automatically.

  • Duplicate

    Another question, or an answer to another question, provides an answer

  • Another reason

    Use this flag if something is wrong which you can’t fix but which doesn’t require a moderator’s actions. Posts with this flag are offered to high-rep users to help with.

  • Needs moderator attention

    Anything else.

Poor answers

  • Low quality

    Only use this flag if an answer attempts to answer the question but is completely unsalvageable even with editing. If an answer makes sense or can be edited to make sense, but you don't agree with it, or it’s not particularly useful, vote instead.

  • Not an answer

    The “Not an answer” flag should be reserved for answers which make no attempt to answer the question, for example because they are “Me too!” or “Thanks!” or a supplementary question, or an “answer” which is unrelated to the question, including gibberish.

    The “Not an answer” flag is not appropriate for answers which do attempt to answer the question but are wrong or poor in some other way. That’s what the “Not useful” downvote is for, and users with the required privilege can also vote to delete.

The bottom line

  1. Follow the rules and avoid being flagged.
  2. Fix things you can fix.
  3. Vote where you can vote.
  4. Flag things only if there’s no other option.
  • 1
    Nice! On the section about use-mention, could you add something to say that it is not (usually) flaggable (the parenthetical being that it shouldn't be an excuse to use offensive language)?
    – Mitch
    Jul 10, 2014 at 2:14
  • @Mitch You're right. Although that edge case was implied, something explicit is better.
    – Andrew Leach Mod
    Jul 10, 2014 at 9:10
  • 1
    I'm not too keen on "Your worldview is irrelevant to discussing language". The fact of the matter is many aspects of usage are predicated on worldviews (often, reflecting perspectives which aren't shared by all, and/or are hopelessly out-of-fate). Sometimes this needs to be pointed out, because it's not always obvious. Jul 10, 2014 at 20:31
  • We've needed something like this for a while... thanks.
    – phenry
    Jul 10, 2014 at 22:40
  • @FumbleFingers Yes, and 'bluey' is an unfortunate choice of word in the examples. I'm surprised it hasn't been flagged as 'offensive' :) bluey : SA and Aus. slang for an alcoholic - from the colour of methylated spirits No need for a citation as that's not from a book or website.
    – Frank
    Jul 11, 2014 at 14:18
  • 1
    @FumbleFingers I believe that's what Andrew is getting at later in the post. Worldview is irrelevant when it's irrelevant. Mentioning that your worldview has possibly had an impact on your answer is relevant.
    – Kit Z. Fox Mod
    Jul 13, 2014 at 12:49
  • 1
    @KitFox: At least four people give credence to the idea that Pharisee is potentially offensive to Jewish people. I find that intriguing, but I can't easily take it any further without knowing at least something of the worldviews of the commenter and three upvoters there. Jul 13, 2014 at 16:46
  • This post needs to be updated a little: 1) There is no offensive flag anymore. It’s rude/abusive now. 2) IIRC there was some discussion on extending NAA/VLQ to one-line answers on this site. This should be addressed (whatever the outcome). 3) General consensus is that total gibberish is to be flagged abusive (it’s abusing the site).
    – Wrzlprmft
    Jul 27, 2016 at 9:39
  • Also: Only use this flag if an answer attempts to answer the question but is completely unsalvageable even with editing. – How would an answer look like that is completely unsalvagable but clearly attempts to answer the question? Also, what about edits by the author?
    – Wrzlprmft
    Jul 27, 2016 at 9:40
  • You can update this answer to reflect the updates on MSE: meta.stackexchange.com/a/17365/309993
    – NVZ Mod
    Jul 22, 2017 at 11:13
  • @NVZ Or, indeed, you can!
    – Andrew Leach Mod
    Jul 22, 2017 at 11:33
  • @AndrewLeach I would, but this is a comprehensive and detailed answer and I'll have to go through it all in order to make sure any edit I make don't spoil the overall style. Also, you are the moderator here, and I think you should be the one to update it (when you have the time). :)
    – NVZ Mod
    Jul 22, 2017 at 11:35
  • Comment flags have been changed so this needs rewriting.
    – Laurel Mod
    Sep 23, 2018 at 21:51

You must log in to answer this question.

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