Related: Should workarounds be suggested when the question implies an impossibilty?

That question is about questions that are (roughly) of the form, "In a world with feature X, how could phenomenon Y occur?" and how to respond when you think that phenomenon Y could not occur under those conditions.

I see a lot of questions of that general form, where the questioner presents a set of conditions for their world, and asks about the consequences. On these questions, it is not uncommon for someone to post an answer that rejects the premise of the question. That is, in response to the question, "Under X circumstances, how would Y happen?" someone says, "X can't happen."

For example, this question, which has the constraints

the original is most assuredly dead, and the copy is most assuredly not the original

has an answer that says, "you are still you," thus rejecting the constraints. The question Weapon developed in modern melee only world has a number of answers that say "people would use ranged weapons anyway". And consider how much debate came up around How would society react if the existence of a god was scientifically proven? regarding whether or not God actually does exist or can be proven.

How should we respond these answers, that reject the conditions of the question or say that those conditions can't occur?

My instinct is that in creating fictional worlds, we will necessarily have fictional elements. Therefore, answers that reject those elements are not helpful and should be rejected as "not an answer," especially if that is the entire content and no alternatives are suggested. After all, I can't imagine that someone responding to a question with "magic is not real" would be very welcome. Is this correct, or are these answers acceptable? (Or is there some other response to consider?)


It depends.

If a question is tagged and someone says that magic is not real, then it's reasonable to flag the response Not an Answer. However, if a question is marked and , then saying that something is not realistic is perfectly appropriate. There's a line somewhere in the middle.

A large part of worldbuilding is plausibility. In particular, the melee weapons only rule was trying to justify a claim. Knowing that there's a lot of pushback on certain parts of the explanation is useful in that case. It's not a plausible way to justify development of a specially compliant weapon unless there is some impetus to carry a weapon as well. If readers lose suspension of disbelief on a point like that, then the story stops carrying them.

How society would respond to proof of God's existence is more interesting. Arguing that God does not exist is useless there. I count three deleted answers, and I think rightfully so. It's possible that others should be deleted, e.g. this one, but even that answer illustrates an actual reaction to finding proof. Another reaction is of course to deny that the proof proves anything. So an answer like this one fits the question.

I think that TimB's answer to Implications of Respawning is helpful. It points out that some people would regard the clone as a continuation. I don't know that a story will get very far by assuming things that most people will regard as false. Authors can't force suspension of belief on readers. It must be earned. Now if you want to make the dominant position that the clone is not a continuation, that's possible. But it seems quite likely that most clones will consider themselves continuations. After all, what else could they be? They have all the memories of their first lives. When you awake, do you regard yourself as a different person from the one that went to sleep?

Most of the time, the best solution to a bad answer is a downvote. A deletion is a rather strong reaction and leaves the question vulnerable to others making the same point. A downvoted answer still shows to everyone and allows people to see the existing answer so that they don't have to repeat it.


My two cents:

I think these answers, while not totally in the spirit of the question, are quite often useful and should be left alone. I've seen a few instances where someone has answered in this fashion and the OP has commented something along the lines of

Thanks! I hadn't thought of this but it's an interesting direction to go in.

Worldbuilding requires creativity. Sometimes, I think if a premise in a question doesn't entirely make sense to an answerer but they can think of a similar alternative, the answer could actually help - whether by introducing a new direction or identifying an unviable one, it's still useful.

Obviously if the answers completely contradicts the question it should be deleted/flagged for deletion. In this case I don't think that's necessary as the answer is not that far off the question.

  • 2
    $\begingroup$ I agree. Challenging the frame of the questions is often quite useful. As long as the answers explain why they don't feel it would work, then they are providing value, though I don't expect them to score highly $\endgroup$ – Mourdos Mar 4 '15 at 8:48

I wanted to add one point and that is simply that, building an entire world from the universe/solar system, down to the social acceptance of a certain scenario is a massive undertaking.

None of us are experts on all of the topics that exist in that spectrum. To not help another user identify bad assumptions/logic/science is actually a disservice to the user asking the question.

For example in this question: What if the notion of 'Nationalism' was eradicated?

I responded that the premise the question was based on was flawed. I know this because I have studied it, my masters is in international relations.

Were I to not answer because to do so I have to tell the user their question is flawed I am certainly not helping them out. So as others have mentioned these kinds of answers certainly have their place...with the mentioned caveat that you can't (as mentioned above) say, "Your question about the implications of a deity is wrong because I am an atheist and god doesn't exist."

I think the difference between the two mentioned scenarios is pretty clear cut and I do not see that we currently have a problem in relation to this issue.


I might be guilty of this, in which case I apologize for those answers. At the very least I'm to blame for my ubiquitous "No, it can't" answers, although people seem to like those when I back it all up with facts, sources and calculations. Also, thanks for linking to the earlier meta question; that gave me some inspiration.

First off: The linked questions.

I'll be frank and say that while I'm a solid atheist, I know that any answer to How would society react if the existence of a god was scientifically proven? must take on the existence of a god as the premise. Period. There is no way around it. I really don't see how you can argue against it, especially because - as I'll get to later - the question does not have the or tags! So even though some of us disagree with the idea in real life, we have to cast that aside to discuss the issue. Were I to write an answer to that, I'd happily assume that a god exists.

I feel like the scenario in Weapon developed in modern melee only world is plausible (and I won't get into any arguments about it), so I reject the answers (but did not downvote) that forego the question based on the rationale of impossibility. But that's just my belief.

Implications of 'Respawning', too, didn't seem to have an outlandish presence. I was pleased to see that all but one answer (sorry, TimB!) took the premise as reasonable into consideration. Again, I see no reason why the scenario is implausible.

In general:

The first answer in which I acknowledged that I reject many premises is Could a civilization as advanced as humans exist on a planet/dwarf planet like Pluto?. I'm fine with this type of answer because

  1. The question asked if the scenario is possible, to which I duly responded.
  2. The question had the tag.
  3. My answer used facts and logic to back it up, and I think I did a decent job of proving my point.

If a question uses either the or tags, then I think it should accept answers that say that a given scenario is not possible (note that if refers to the universe in question, then some scenarios that are impossible in our universe may be possible in the given universe, and answers should reflect that).

On Physics, Astronomy and other science sites, I've seen instances where the user wants a scientific answer to a question but refuses to abandon an unscientific premise. Often the premise involves Creationism, to which I downvote, flag (occasionally, if the post is really bad and I think a mod won't see it) and wait for it to be closed or heavily modified. Fortunately, I've never seen a post along those lines on Worldbuilding, but the same logic applies: If you want a scientific answer, adopt a scientific premise.

I'll end on sources. I've always been big on adding in citations and links to give an answer some credibility and to make it clearer. Adding facts lends some heft to whatever point you're trying to make. If you say, "I don't think this is possible," you won't get as many people to believe you as if you say, "I don't think this is possible because this study says blah blah blah and this article says yada yada yada . . ." I'm not advocating mandatory citations and links in answers, but in cases like this, they can be very important.


Answers that directly contradict the premise of the question are not really helpful. However, answers that explain why the premise is impossible or unlikely (and especially answers that then offer new possibilities that stick as close to the intent of the original question as possible while fixing the flawed premise) seem to me to be more helpful than answers that accept an impossible premise without pointing it out.

So, basically, saying 'magic isn't real' deserves a downvote, but saying 'your concept of how magic works in your universe is problematic because x, however if you do y, you can solve your problem by doing z' is constructive and thus should be treated as a good answer.

Also, edits based on comments are usually encouraged. If the questioner feels that the answer is not constructive, they can say so.


You must log in to answer this question.

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