In my regular discussions with people, as people run out of logical approaches to an argument, I frequently come across the argument:

...but if you were stuck on a desert island, would you...

The implication being that if you place the argument in a scenario that is extreme and unrealistic, and that argument fails in that scenario, the argument is invalid in the more frequent and realistic scenario.

My searches on Google seem to suggest this is called a Robinson Crusoe fallacy which seems apt but not formal.

Is Robinson Crusoe the correct term for this fallacy or is there a better one?

  • My usual response to this is "buy me an island and let's find out". :-p Jan 26, 2016 at 22:53
  • 3
    I'll try to write an answer in long form in a bit, but in short form, (a) there's no formal list of names for most informal fallacies. Thus, there's no correct term per se. Most broadly, this is a variation on "false premise" but that's not a very meaningful thing to say in philosophy.
    – virmaior
    Jan 26, 2016 at 23:01
  • 2
    Depending on use, it might be a valid approach. If one is claiming something is universal, and someone provides a situation where your claim did not hold, then it suggests there's a need to caveat the domain where the claim is valid. I do know the opposite of this "fallacy" shows up a lot, in the form of "the claim works in the most common scenarios, so clearly we should not consider the possibility that it might not be universal."
    – Cort Ammon
    Jan 26, 2016 at 23:40
  • Not on SO, people are arguing that this is not the Robinson Crusoe fallacy at all. Jan 27, 2016 at 2:50

4 Answers 4


I'm not sure if there's a named fallacy involved here. But it seems to me that the ring of Gyges might have some relevance or at least be interesting in this context.

The ring grants the wearer invisibility. This is the original "What would you do if you did not have to answer to society?" thought experiment. In the Republic, Glaucon uses the myth to show that no one with the ring would be immune to corruption, claiming that not even gods could resist it.

It might also be interesting to read Deleuze's short essay on desert (and deserted) islands. He gives a reading of Robinson Crusoe there, and is trying to show how the psychological movement towards the desert island is related to the geological movements that condition the island in the first place -- the continuous struggle between the land and the sea, the twin necessities of separation and creation (to really create, you had better be separate...)

  • As I understand it, the ring of Gyges is based around the concept of a human flaw. If we consider the need to survive as a human flaw, then this would be correct. However, in the story s/he is given a power, whereas in this argument the person usually has all power eroded to the point where they have only one option - to betray a commitment. I have to think on this one further as it's not quite sitting right as far as I can understand it. Jan 27, 2016 at 2:48

There is simply no fallacy here, taking assertions to their logical extremes is a good test of whether or not there are premises missing from the remainder of the argument.

If your argument does not hold on a desert island, or on a plane (or in a box or with a fox...) then there is some reason the ordinary case is the ordinary case, and it may very well be important that you find it.

Omitted premises do make for false arguments, and they are a very common cause for two reasonable people not to see situations in the same light. There may be ideas in play that you have not recognized, which will sometimes provide a basis for a much more direct argument.

  • I apologise if I've been misleading. The premise does actually hold on a desert island. The person arguing against it is implying that it wouldn't by taking it to extremes. Jan 27, 2016 at 4:25
  • 1
    @JeffY Half the things on that page are not fallacies. They are manipulative rhetorical devices. Fallacies result in invalid arguments and the mental state of the arguer is not relevant to validity. If your argument breaks under any of the 'Many Questions', it just was not valid to begin with.
    – user9166
    Jan 28, 2016 at 18:23
  • 1
    @tudor They take you off topic, rather than asserting something false. Fallacies assert falsehoods, e.g. ad-hominem tries to deduce the validity of the argument from the trustworthiness of its source -- it asserts 'you can't be trusted, so you must be lying in this case." But what he is calling informal fallacies and I would call sophistry or rhetorical manipulation, like asking for too many refinements to an argument, or seeking the boundaries of the premises, like this case, do not assert anything, true or false. They just distract from the main point and diffuse the focus.
    – user9166
    Mar 18, 2016 at 14:55
  • 1
    Fallacy (neither formal nor informal) is not about truth or falsehood, Formal fallacy is about incorrect reasoning, where informal fallacy is more about making deceptive arguments. But in both cases, a fallacious argument can have a true conclusion. (In such cases it is reaching a correct result via an incorrect route.)
    – Jeff Y
    Mar 18, 2016 at 15:34
  • 1
    Bear in mind that asserting something false can always result in a true conclusion. The argument does not determine the truth of the propositions, it only establishes and exposes it. This is one of the reasons a False in classical logic implies anything both true and false.
    – user9166
    Mar 18, 2016 at 17:48

Concept tester

There is nothing strictly fallacious (involving a logical error) in using the desert island scenario. It could be a useful thought experiment. For example, if one were alone on a desert island would the concept of justice have any application? Only, it would seem, if one can act justly or unjustly to oneself. Or could one act unjustly in this scenario if one took an unnecessarily large portion of the food supply, thus driving other animals to extinction? I have no idea what the right answer is to that question and the thought experiment has suggested a fresh ethical topic to me.

Heuristic value

The scenario may also be useful in introducing a non-ethical topic such as that of Wittgenstein's anti-private language argument (Philosophical Investigations, I §§243-315). The argument can be set out without the use of any such scenario but the scenario nonetheless would help 'concretise' the argument for those of us who tend to think visually.


> "The implication being that if you place the argument in a scenario that is extreme and unrealistic, and that argument fails in that scenario, the argument is invalid in the more frequent and realistic scenario."

The above technique is called reductio ad absurdum (reducing to absurdity), or appeal to extremes -- and it's no fallacy!

On the contrary, it is often the simplest and the most effective way to prove your point.

  • I suppose an example of this in reverse application would be to suggest that if it's ok to kill someone to survive on a food-starved desert island then it's ok to kill someone in in my highly populated and food-rich city to survive? :-p Nov 30, 2020 at 10:09
  • This sounds very similar to @GeoffreyThomas's "Concept Tester". Nov 30, 2020 at 10:10

You must log in to answer this question.

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