In one of the first questions on this site, "If I was" vs "If I were", there are a slew of answers.

The question had a somewhat more recent (Feb 2018) spate of activity with a very authoritative, nuanced, and reliable answer.

Up to that time (and even up to yesterday) the highest voted answer, done on the day of the question 9 years ago, upvoted at a whopping 93, was basically Just use 'were'.

But this is all in the past. Sometime in the past day or so, the answer was flagged (I am not a mod so I don't know the flag reason), and a mod deleted the answer.

For the life of me, as much as I might disagree in fact with the answer, I see nothing that is flaggable or deletable in the answer. It's not very deep, not really the best quality, not exactly supported with data. But it is a highly upvoted answer, and I think that should stand for something.

Now, I hardly know the difference between a spate and a slew, but I think I know when the subjunctive is used in English, and that is... well it's complicated. Some people still use it, mostly young people have never heard it and bridges still seem not to collapse any more often. Sometimes my lumbago really flares up when I hear kids trampling my lawn saying "If I was a rich man...", but when I put on my linguistics hat, the subjunctive is mostly dead in English (just like 'whom'!), so I think of the answer as 'mostly wrong'.


But it seems like a misuse of flagging. What is there to flag? There's nothing rude or spammy, it's totally an answer.

Don't go against the operation of the system, use the system. Downvote that sucker if you don't agree with it. Those guys back then only had one vote each like you, they all liked the "If I were..." always answer. Just like you might on another question. You want your votes deleted? More importantly, you want 100 upvotes for your answer deleted?

So, please undelete that answer. Comment on the answer and down vote it and say it is wrong (but please give reasons and don't just say 'you are wrong nyah nyah nyah') and up vote the answer you think is right (It's tchrist's. Really).

There is a case to be made that a judge should be able to override a jury. But if thats the case, the precedent you want to make, there's a boatload of old questions, starting from Aug 2010, that we have to go through and delete because those old people are effing wrong.

I'd like people not to vote here about the subject, subjunctive, but about whether a mod should hammer delete a highly upvoted question that some people in comments disagree with. And I hope you vote here to have the question reinstated so we can downvote it and upvote ones you think are right.

Update Mar 25, 19: Thank you mods for undeleting. Everyone: act on that question as you would.

  • Sometime in the past day or so, the answer was flagged How do you know it was flagged? How can you tell?
    – Mari-Lou A
    Commented Mar 20, 2019 at 18:25
  • 1
    @Mari-LouA I am not a mod so I don't see flags so I don't know for certain. But in the comment thread of the deleted question, someone said "Moderators keep declining my flags, but this is such a nonsense.". That was yesterday and then today it was deleted. So I inferred tentatively a connection.
    – Mitch
    Commented Mar 20, 2019 at 18:46
  • 4
    I flagged it. The reason I left was as follows: Going on ten years old, this answer falls short of all modern requirements: research, sources, and explanation. There are many other answers in this thread that do meet these requirements, making this answer completely dated and unnecessary.
    – user231780
    Commented Mar 21, 2019 at 3:02
  • 5
    @BlackandWhite then you should also flag the question for exactly the same reason, it too lacks any effort and attempt at research. It too does not meet the minimum "requirements" laid out by EL&U. It too ought to be deleted or migrated to ELL and a researched question should be posted in its place.
    – Mari-Lou A
    Commented Mar 21, 2019 at 6:54
  • 2
    This conversation prompted this meta post. Commented Mar 21, 2019 at 9:41
  • Oh, this one is a tricky situation.
    – NVZ Mod
    Commented Mar 21, 2019 at 13:27
  • @Mari-LouA Since when do questions require "research"? They're questions, after all.
    – user231780
    Commented Mar 21, 2019 at 17:00
  • 11
    @BlackandWhite Basic research by the OP is a highly recommended suggestion that has been found over the years here to make much better questions. Research, or the record of at least the attempt at answering ones own question, is the recommended basis for a good question. It tells the answerers at what level they should answer, how much detail they should provide, etc. It also eliminates questions that could be easily answered by accessing general references (like a dictionary or thesaurus).
    – Mitch
    Commented Mar 21, 2019 at 18:20
  • @BlackandWhite very "old" and under researched questions along with their highly upvoted answers have been deleted, some were later undeleted when they were detected and mentioned on meta, but others have vanished.
    – Mari-Lou A
    Commented Mar 22, 2019 at 7:46
  • 1
    As an aside, one of my higher scored answers is one that started out wrong! After receiving negative feedback on it, I edited it to explain how my original thinking was wrong and it turned out to be one of those “long tail” answers that every so often attracts an upvote even though it’s years old. Improving our answers is an important part of participation here. If an author starts getting a bunch of downvotes maybe they will edit their answer.
    – ColleenV
    Commented Mar 22, 2019 at 17:51
  • 2
    I'll give it until Monday in case people with opinions don't see this until the weekend, but I expect to be undeleting the post after the weekend. Commented Mar 22, 2019 at 18:04
  • @Mari-LouA ELL is not our dumping ground for poorly researched questions! They have the same quality requirements as we or any other SE site. If something isn't good enough for us, then it isn't good enough for them either. We should only migrate questions that ask things obvious to any native speaker but not to non-natives. And we should always follow the golden rule of migrations: don't migrate crap. So "not good enough for us" is never a good reason for migration.
    – terdon
    Commented Mar 25, 2019 at 11:38
  • 2
    @terdon you got the wrong end of the stick, maybe you should read the comment again. I also said deleted and it was obviously a provocation aimed at Black and White, who must have flagged the answer several times over two years or so. Well, if someone is going to do that, flag the question as well because that is also "crap".
    – Mari-Lou A
    Commented Mar 25, 2019 at 12:53
  • 1
    @Mari-LouA I was referring to this comment where you suggested it "ought to be deleted or migrated to ELL". Since there is a longstanding issue with us sending our crap to ELL, which they're understandably not very pleased about, I wanted to point out that the migration bit was bad advice.
    – terdon
    Commented Mar 25, 2019 at 13:20
  • 2
    @terdon I actually think ELL would be delighted to have that question (which you define as "crap") along with its answers. Have you seen the questions they have been receiving lately? The vast majority lack context, research and effort, exactly like the "was/were you" question which was never, presumably, flagged. I'll now get a comment from either Colleen V or J.R. telling me that my affirmation is an exaggeration but it's true. BTW I have been an EL&U user for almost 6 years, I think I know my way around the place by now and I wasn't giving any advice I was pointing out the double standard.
    – Mari-Lou A
    Commented Mar 26, 2019 at 8:34

5 Answers 5


So I’ve looked over the guidance I can find about incorrect answers and what to do about them (especially if they’ve been accepted or have attracted a lot of votes) on the main meta. The consensus appears to be :

  1. Write a correct answer
  2. Downvote incorrect answers
  3. Leave comments on incorrect answers explaining why they are incorrect

The answer to Should I nominate incorrect answers for deletion? seems to be “downvote don’t delete”. Even for suggestions like Requesting review of technically inaccurate answers the answer is downvote, comment, and maybe ask a moderator to put a notice on the answer.

Deleting an answer that has no other problem than it is old, highly scored and possibly incorrect does not seem to be a supported method of dealing with this well-known issue. Not one of the highest scored answers to these questions suggests taking steps to delete an incorrect answer.

I would also suggest that even though I personally don’t believe anything untoward was going on, there is an appearance of a conflict of interest when a moderator has a competing answer. At a minimum, the possibility of deleting the answer should have been discussed on meta before it happened. There was no urgency in removing it (as far as I can tell) so I don’t understand why it wouldn’t be brought up for discussion.

To summarize my concerns:

  1. The vast majority of guidance, particularly discussions where Stack Exchange employees have been involved, discourages deleting answers because they are wrong. I don't see a clear and compelling reason to contravene that guidance, especially when the recommended methods of dealing with the issue were already applied.
  2. The question has been protected to prevent answers from new users, but it's not clear to me why that protection is permanent.
  3. The accepted answer has a good score and also has a bounty awarded to it which gives it plenty of credibility and weakens the argument that the other answer is dangerously wrong.
  4. A moderator decided to remove an 8 year old highly scored answer without clear consensus from the community, based on a flag from another user whose answer to the same question was criticized, and where a moderator has a competing answer, which makes the whole thing look a bit shady (even though I am confident that there were only good intentions involved here).
  • 1
    I disagree that we are talking about answers that are possibly wrong. The answer is definitely wrong, so I went with meta.stackexchange.com/q/120298/147791 Commented Mar 21, 2019 at 8:34
  • 2
    @MattE.Эллен you allowed the possibility of being mistaken, (so my judgement could easily be off and It could easily be seen by someone else like me, who doesn't know what the correct answer is) but in reality you knew precisely what you were doing and you were wholly convinced that you did the right thing, so be clear and say that.
    – Mari-Lou A
    Commented Mar 21, 2019 at 8:43
  • 7
    @MattE.Эллен There’s a big difference between “horribly wrong” and “after consulting with experts I believe it’s wrong”. The vast majority of guidance I was able to find doesn’t favor deletion even for answers known to be incorrect.
    – ColleenV
    Commented Mar 21, 2019 at 11:02
  • 1
    @ColleenV "The question will recur and will get the same kinds of idiotic answers because there are more idiotic answers available than sane ones" is a comment from John Lawler. If "idiotic answer" isn't equivalent "horribly wrong", what would you call it? Commented Mar 21, 2019 at 11:05
  • I realise I've got a bit more information than you can see, so I apologise for that. Commented Mar 21, 2019 at 11:06
  • 3
    @MattE.Эллен I do have to ask though why the comment by Lawler and a complete and correct answer isn’t enough. Wouldn’t it be better to let the wrong answer stand with rebuttal than to hide it?
    – ColleenV
    Commented Mar 21, 2019 at 11:24
  • @ColleenV Yes, I suppose it could be that deleting the rest of the comments and just leaving Lawler's might suffice. I still think deleting it is the right thing, as it removes the ambiguity. 93 vs 41 creates unnecessary doubt, If the answer was just a joke, we've lost nothing. If the answer was serious then we've gained clarity. Commented Mar 21, 2019 at 11:41
  • 7
    @MattE.Эллен If so many people upvoted the wrong answer, it probably needs to be shown to be incorrect. I can't judge, because it was deleted before I got a chance to see the discussion about it, but it sat there for years, so I don't understand the urgency. Removing the answer is removing information that should lead people to a better understanding of the topic because it's incorrect. We often learn better from mistakes. If you don't leave it there, the votes have no chance of normalizing. We should let folks judge for themselves, even though some won't reach the right conclusion.
    – ColleenV
    Commented Mar 21, 2019 at 13:10
  • 5
    @@MattE.Эллен: I don't like to argue with mods any more than I like to argue with John Lawler, but there are times when it's necessary (JL, for example, has made what I consider untenable assertions regarding usages like I'll open you the door and If I had have lost you). And I find your answer is definitely wrong assertion equally untenable - but even if I tone that down to questionable, that doesn't alter the fact that you actually deleting the answer was a misguided unilateral act. Commented Mar 22, 2019 at 14:02
  • Good luck with leaving comments about why it's incorrect. I have done that and the OP hounded me and I ended up being sarcastic and having my knuckles rapped by the mods. This site has still not come to grips with what to do about blatantly wrong answers (not the case here, I know) with tons of upvotes.
    – Lambie
    Commented Mar 26, 2019 at 6:09
  • 1
    @Lambie if someone is hounding you, flag it instead of responding. It takes two to tango, so reject their invitation to dance and save yourself the annoyance. On some discussion forums I participated in I would set my signature to “Don’t try to teach pigs to sing, it wastes your time and annoys the pigs.” to help remind myself that I don’t really need to rebut every argument that crosses my path. Try setting a timer and waiting 30 minutes before deciding whether to respond. I’ve found that makes it easier for me to not get sucked in to debates that usually end up just making me cranky.
    – ColleenV
    Commented Mar 27, 2019 at 14:08
  • @ColleenV That was just one case. Hounding is not always immediately apparent.
    – Lambie
    Commented Mar 27, 2019 at 14:50
  • great call @ColleenV ! thanks.
    – Fattie
    Commented Mar 28, 2019 at 9:57
  • 3
    Just BTW, it is utterly, astonishingly, totally, completely bizarre to assert that that answer is "particularly bad" on this site. ELU is a fantastic site, but it famously has staggeringly incorrect, bad, hopeless, utterly-wrong answers: as a commonplace. It's just completely bizarro to, in particular, point to that one totally innocuous, unremarkable answer. There is nothing, whatsoever, special about it in terms of "badness". To assert that that (particular) answer is "definitely! wrong!" - on this site - is just totally whacky.
    – Fattie
    Commented Mar 28, 2019 at 10:04

I understand that it is disturbing when a dubious answer far outscores a careful and authoritative answer. However, a number of factors suggest to me that the present case is not an especially hard one to judge.

Essentially, the key elements in the controversy are a no-research-shown question posted on August 13, 2010; a brief (four lines) but confident-sounding no-research-shown answer also posted on August 13, 2010; and a lengthy, serious, carefully argued answer posted on February 24, 2018. There are also nine other live answers that have received at least one upvote, and five other answers that were deleted in the more distant past after earning net scores of 1, 0, –1, –3, and –4.

It is noteworthy that of the nine also-ran answers, two (one posted on August 13, 2010, and the other on September 8, 2010) are only three lines long and two others (coincidentally also posted on August 13, 2010, and on September 8, 2010) are only four lines long each. The extreme brevity of these answers and their paucity of reasoning or corroborating evidence reflect the standards of the time, obviously. But it bears noting that the only indisputable qualitative difference between the now-deleted brief post and the four still-live brief ones is that the deleted one had received a net upvote score of 93 (103 up and 10 down), while the four extant shorties have net scores of 30 (30 up and 0 down), 13 (26 up and 13 down), 13 (16 up and 3 down), and 4 (5 up and 1 down).

On this record, it appears to me that the 93-vote short answer was deleted not because it was worse than the other four short answers, but because it was more popular than they and the other posted answers were—so popular, in fact, that it was getting in the way of the obviously superior (and longer and much more recent) answer that arguably deserves to have the most votes.

But by remarkable good luck the original poster came back at least seven and half years after posting the question and awarded the green check mark to the long, high-quality answer. As a result, the long, better answer now sits at the top of the series of answers, in immediate view of anyone who wants a detailed analysis of subjunctive use.

Under those circumstances, I don't see how we can argue that the short, superficial answer is keeping the long, deep answer down. It simply isn't in a position to do that.

To the extent that anyone is disturbed that the old, short, opinion-based answer has substantially outscored the better longer answer, all I can say is, that's how it goes at this site: a lot of people (including a number who cast upvotes and downvotes) prefer short superficial answers to long closely argued ones. (In fact, one voter disliked the long answer enough to downvote it, leaving it at the moment with a net score of 45 upvotes [46 up and 1 down].) If you don't like the results, I recommend that you (1) upvote the answer you think is good, (2) downvote (if you must) the ones you don't like, and (3) call attention to the answer you think is especially meritorious—perhaps by presenting it with a bounty for being exemplary and deserving of more recognition (it already has had one bounty awarded to it).

I do not think the deleted answer should have been deleted. I think it should be reopened and subjected to the same vicissitudes of the voting public's opinions that all other answers are, assuming that they satisfy the criteria for answers prevailing when they were posted. In my opinion, if the deleted answer were reinstated, the longer and much better informed answer would gradually gain ground on it anyway, thanks to its inherent merits and to the boost it will continue to receive from being the answer accepted by the original poster.


Update (3/28/19): In the eight days since Mitch opened this Meta discussion, the detailed, comprehensive answer has received 5 additional upvotes and no additional downvotes; the brief, highly upvoted, temporarily deleted answer has received 2 additional upvotes and 6 additional downvotes. The gap in net upvotes between the two answers has thus dropped from 48 to 39. It will be interesting to see how voting on these answers proceeds in the longer term, now that the immediate controversy has subsided.

  • 6
    Bravo! And I'd also say that the shorter deleted answer elevates tchrist's one, users can see that new answers can always be posted on old questions especially if they debunk older answers and prejudices. In other words, someone somewhere will post a better answer if the answers are objectively weak and unsubstantiated.
    – Mari-Lou A
    Commented Mar 22, 2019 at 5:31
  • 5
    And in any case, "history" shouldn't be deleted because today you're embarrassed or dismayed by it.
    – Mari-Lou A
    Commented Mar 22, 2019 at 7:48
  • It's not voting, that's just the marketing term for it. It's a slightly liking system, which usually works, but doesn´t in cases like these. There is nothing deomcratic about it. One particular point here: thousands of passers by who live on other sites got to upvote only on this question. Their rep doesn´t allow them to downvote. So there is nothing meritocratic about an answer which the vast majority of voters have only beeen able to upvote and not downvote. Who knows how many downvotes this would have had by now if up and downvotes were available to the same type of voters. Commented Mar 25, 2019 at 18:10
  • 3
    @Araucaria That’s an argument to reset the scores, not to delete the content IMO.
    – ColleenV
    Commented Mar 25, 2019 at 18:51

I am disturbed by this deletion. The answer may not have been a good one (although the number of upvotes clearly disputes that)—but, in all other cases, whether an answer (or question) is good is reflected by the community. Only if an answer (or question) clearly violates the code of conduct should it ever be deleted by a moderator.

And, even then, there is nothing in the rules of this site that says that X number of downvotes should result in deletion. Further, in this case, it was the opposite. This answer received a great many upvotes. The community, as a whole, appreciated it and thought it was a good answer.

If a moderator objects to something at a personal level, they should do what everybody else does: comment on it, provide an answer of their own, downvote it, or vote to close it. That is no different than what anybody can (and should) do. For a moderator, personal opinion should weigh no more nor less than personal opinion for anybody else. Just because a moderator has the ability to wield moderator tools that should not give them the right to do so.

Only in the most extreme of situation should an answer actually be deleted. This answer doesn't meet any of the objective criteria for actual deletion. It doesn't matter if it wasn't liked by some people, or if some people thought it shouldn't have been provided. Deletion of answers (and questions) should only be done for objectively obvious violations of posted and acknowledged rules. Being considered poor is simply not something that makes a post worthy of deletion. (Even if a majority of people consider it to be poor. And, in this case, that clearly wasn't the case.)

I'm glad to see the community weighing in to have this undeleted. At least on the face of it, I am appalled that it was deleted in the first place.

Having said all of that, I want to reaffirm my claim that things may have gone on behind the scenes that I'm not aware of, and there may have been justification for its deletion that can't be made public. I certainly don't want my answer to be taken as any kind of personal attack. I just want to express what my take on this is, based on the facts I'm aware of.

  • Only if an answer (or question) clearly violates the code of conduct should it ever be deleted by a moderator. - Full stop, end of story.
    – Mazura
    Commented Mar 27, 2019 at 2:47
  • That's if the code of conduct explicitly says Don't spam. (As far as I can see, it doesn't.) There are more reasons for deleting a post, including the fact a user posts the same question after it was closed as off-topic, or a user posts an answer that is copy-pasted for different questions, even the ones for which the answer is irrelevant.
    – apaderno
    Commented Aug 31, 2019 at 0:17

For users who do not have 10K rep and cannot see the deleted answer, here it is. And, yes, I had already downvoted it.

screenshot of the deleted answer transcribed below

When in doubt, always use the subjunctive mood:

If I were you...

It will make you sound smarter and it is technically correct since "the subjunctive mood is used to express a wish or possible situation that is currently not true."

Moreover, the top upvoted answer took its time before it hit the 90 mark

  November 24, 2011 it had a score of 22.
  September 24, 2013 it had a score of 41.
  September 23, 2014 it had a score of 48.
  April 04, 2015 it had a score of 51.
  June 29, 2018 the score rose to 80.

  • 3
    That is very consistent growth of ~9 net votes a year. A very clear example of consistent approval (and average ~1 downvote per year). I expected that since it was one of the first questions ever on the site that it would have gotten most hits within a week. Nice research.
    – Mitch
    Commented Mar 23, 2019 at 13:58
  • Honestly, you guys... Doesn't anybody realize how bizarre and hilarious this whole event and its reified resurrection here is? ROFL Commented Jan 30, 2022 at 19:11

I admit I'm no linguistic scholar, not by a long shot, so my judgement could easily be off.

My reasoning for deleting the answer was that many our members that I believe to be professionally involved in linguistics thought the answer was misleading and the answer had a lot of upvotes.

The comment that really caught my attention (available to 10kers)

The question will recur and will get the same kinds of idiotic answers because there are more idiotic answers available than sane ones.

It could easily be seen by someone else like me, who doesn't know what the correct answer is, as the correct answer. Which meant to me that the answer was noise, and should be silenced, because it would be a long time before anyone got around to down voting it.

Edit for clarity: If the community is against what I've done, I'll undo it.

  • 7
    In other words, those 103 users who upvoted the answer were all wrong?
    – Mari-Lou A
    Commented Mar 20, 2019 at 17:52
  • 4
    @mariloua that is what I'm lead to believe from reading the discussion around the answer, as well as tchrist's answer. Commented Mar 20, 2019 at 17:54
  • 3
    And deleting answers that have earned 103 upvotes and only ten downvotes is in the moderators manifesto, is it? Is it? Because if it is then it's absolutely rubbish what SE explicitly says that the best answers rise to the top. Well? Are people not allowed to to vote because of what you or tchrist may think of the answer?
    – Mari-Lou A
    Commented Mar 20, 2019 at 17:57
  • 9
    Speaking as an ELU lurker and not a moderator, I think there should be a LOT more transparency around deleting a highly upvoted answer because it’s highly upvoted. It’s not being deleted because it’s wrong, because it wouldn’t be deleted if it was wrong and had one upvote. Yes, I know meta discussions about judgement calls are messy and annoying, but there should not be a small group deciding this behind “closed doors”.
    – ColleenV
    Commented Mar 20, 2019 at 17:58
  • 1
    I think you're underestimating people's intelligence, They can see which answer has greater depth and which one contains a summary and a quip.
    – Mari-Lou A
    Commented Mar 20, 2019 at 18:00
  • 2
    @Mari-LouA I was the whistleblower. Any rubbish answer like that published today would be met with intense criticism from moderation. Why? Because it has no sources (which you have unironically criticized me for in the same hypocritical breath). And no research. And no attempt at any explanation.
    – user231780
    Commented Mar 21, 2019 at 3:07
  • 4
    Thank you for finally deleting this answer. No post should receive special treatment from ELU requirements because it has 103 upvotes.
    – user231780
    Commented Mar 21, 2019 at 3:08
  • @colleenv I agree. I prompted Mitch to ask the question in chat. This is the first time I can think of where this has come up, so we're deliberately not doing this behind closed doors. Commented Mar 21, 2019 at 8:28
  • @MattE.Эллен please provide relevant link to the chat, so we can see the dynamics.
    – Mari-Lou A
    Commented Mar 21, 2019 at 8:53
  • 1
    @Mari-LouA I've added it to the comments on the question Commented Mar 21, 2019 at 9:58
  • 6
    @MattE.Эллен I appreciate that I’m calling for a meta discussion about a topic in a meta discussion about that topic, however the discussion should take place before the deletion. Some of the folks that have an interest can’t see deleted posts. I’m guessing you probably did the right thing, my complaint is about the process.
    – ColleenV
    Commented Mar 21, 2019 at 11:06

You must log in to answer this question.

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