30

I have encountered this reasoning quite frequently:

Somebody posits the hypothesis that an event X can happpen. A recent example I encountered was "vinegar and salt in the boiling water make eggs easier to peel afterwards", but I have seen many more. Another person imagines a way in which it could happen, judges that way to be impossible, and claims that X can never happen. In that example, the reply could be "salt and vinegar cannot cross the shell, so they will not make the egg easier to peel".

I have seen two ways in which this argument can be false. First, the replying person might just have wrong knowledge (maybe vinegar or salt can pass the shell?) - but I don't think this is a problem of the reasoning process, it is just correct reasoning based on a false assumption. What I find more interesting is that frequently, there can be other mechanisms for X to happen - maybe vinegar doesn't have to pass the shell to make the egg peelable? - but the person making the argument overlooks that possibility and goes from "the mechanism for X that I have in mind doesn't work" to the conclusion "X is completely impossible".

Another example would be people who insist that placebo does not help against pain, because per definition, it contains no chemically active substance that acts on pain receptors.

Has this flaw in reasoning received a recognizable name? I am asking specifically about fallacies as they apply to a formal reasoning process, not a general reason why arguing against vinegar in egg water would be a wrong position to take. (In fact, I don't even know if vinegar in egg water has any effect).

14
  • 3
    I don't think this is always a fallacy. Sure, the "I can't imagine it happening, therefore it can't happen" variant is, but if the person is a subject matter expert (on... vinegar and salt and eggshells, I guess, haha) who actually knows all of the mechanisms through which something could happen, it seems different. That's perhaps less applicable in biology than something more fixed, though.
    – anon
    Dec 11, 2018 at 20:10
  • 2
    The title and body seem to ask two different questions: the title asks about not being able to imagine any mechanism, while the body is about imagining a single mechanism and then disproving that. Which is it? Dec 11, 2018 at 20:19
  • 3
    This is a variation on "absence of evidence is not evidence of absence", but I agree with @NicHartley, in proper context it is not a fallacy at all: "In some circumstances it can be safely assumed that if a certain event had occurred, evidence of it could be discovered by qualified investigators. In such circumstances it is perfectly reasonable to take the absence of proof of its occurrence as positive proof of its non-occurrence." (Copi, Introduction to Logic) Replace "evidence" with "mechanism".
    – Conifold
    Dec 11, 2018 at 23:15
  • 2
    @Conifold you are making an interesting point. Your citation is subtly different, I think, because it introduces the matter of evidence of X occuring. My question assumed that the person making the argument does not know if there is evidence of X or not, and just claims that it cannot happen anyway, so is not interested in seeking such evidence. In reality, I have seen cases where evidence for X having happened was presented to the person making the claim, and that person denied that the evidence can be true, based on their reasoning of "it should never happen". So the claimer cannot...
    – rumtscho
    Dec 12, 2018 at 10:39
  • 2
    ... be said to be reasoning on the lines of "if it were possible, it would have been observed by now" in these cases - since their reasoning is used to deny the truth of the observation having happened.
    – rumtscho
    Dec 12, 2018 at 10:40

5 Answers 5

30

This fallacy is generally called argument from incredulity, or argument from failure of imagination. https://rationalwiki.org/wiki/Argument_from_incredulity It also is a subset of the argument from ignorance. Informal fallacies often overlap, and bleed into each other.

This fallacy is a common one, even among respected philosophers, as it is a characteristic of people that we are poor at identifying our own blindspots. An example of this mistake in a major philosopher is Jaegwon Kim's Pairing Problem argument, where point 3b and 4 is an argument from failure of imagination:

  1. For substance dualism to be intelligible, the mental and the physical must be able to causally interact.
  2. If there is causal interaction between any two events (say the mental act of the immaterial entity and the physical act of the material en tity), then there must be a pairing relation between them.
  3. a) In all physical cases, the specific spatial relations between relata allow us to pick out the appropriate pairing relation, either by having direct spatial relations or by being connected by a spatially contiguous chain of events. b) There seems to be nothing else that can play this role.
  4. Thus, it is highly likely pairing relations are only possible when both entities involved occupy specific spatial coordinates.
  5. The events in mental-to-physical causation in substance dualism do not both occupy specific spatial coordinates. Thus, there seems to be no way to construct the appropriate causal pairing relation in these cases.
  6. Thus, substance dualism is not intelligible.

What makes this a fallacy by Kim is because he does not test and support his assumption that location, and location alone, ("There seems to be nothing else that can play this role" from 3) allows for pairing. For example, matter pairs, even though at the heisenberg scale -- matter is not fully localized. And even though matter is localized at the macro scale, it does not always pair -- as waves and fields can and do superimpose. Plus, with stream of consciousness, reasoning, and math, we have examples of apparently non-localized exclusions and causation (hence pairing). What all these examples show is that while location is often useful to achieve pairing, it is not always sufficient, nor the only way to do pairing.

7
  • 2
    I edited the formatting, but I don't know where "3b" came from. Perhaps it is a typo. Please roll back if I did this incorrectly. Dec 11, 2018 at 19:31
  • 3
    Point 3 has two sentences. The first is a generalized statement of facts. The second one is an argument from ignorance/failure-of-imagination, which then is repeated in 4. I hoped that by labeling it 3b, I would make it clear that the first sentence in 3 is not this fallacy. But ... good intentions are often not successful ...
    – Dcleve
    Dec 11, 2018 at 19:37
  • so if I apply modus ponens in an argument, or any other common inferential rule, because I can't imagine it being false and take it for granted, am I guilty of this fallacy?
    – Hasse1987
    Dec 12, 2018 at 1:38
  • @Hasse1987, I don't think I understand what you are asking. Modus Ponens is not a fallacy. Inference is not a fallacy. Making an inference, and asserting its conclusions are true, while ignoring other logical possibilities -- that is a fallacy.
    – Dcleve
    Dec 12, 2018 at 1:52
  • 2
    Can you elaborate as to how stream of consciousness, reasoning, and math are non-localized exclusions, and particularly causation? In particular, I think math isn't understood to be causal in itself: it's a formal system that can be physically manifested in a variety of (presumably causal) settings. Unless you mean the mental processes underlying mathematical reasoning?
    – Obie 2.0
    Dec 12, 2018 at 17:23
13

Argument from incredulity (aka: appeal to common sense) :

I cannot imagine how P could be true; therefore P must be false.

10

The argument you discuss in the title is a form of argument from ignorance, but the example you give of

Another person imagines a way in which it could happen, judges that way to be impossible, and claims that X can never happen.

is a form of denying the antecedent: "If Y were true, then X could happen, but Y is not true, therefore X can't happen." Argument from ignorance and denying the antecedent are somewhat similar fallacies, and argument from ignorance could be considered to be a form of denying the antecedent: "If I knew of evidence that X is true, then I should accept X, but I don't know of such evidence, so I shouldn't accept X."

1
  • It could also be considered a straw man argument - they're taking your argument that X occurs, declaring that it must happen by some process Y, and then refuting Y as a way to refute X. Dec 13, 2018 at 14:45
9

Because a claim is made from ignorance (or even partial ignorance) this argument could be classified as an argument from ignorance.

However, not all arguments from ignorance are fallacious. Some may simply be weak arguments. Douglas Walton notes the following: (page 272)

In an inquiry, the argumentum ad ignorantiam for a particular proposition becomes stronger and stronger as the knowledge cumulated by the inquiry becomes more and more firmly established. However, even at the beginning stages of an inquiry, where very little is known about a subject or hypothesis, the argumentum ad ignorantiam can still be a correct, even if weak, kind of argument. It can be a speculative argument that only yields a small degree of plausibility for its conclusion. Even so, in such a case, it can be a correct, as opposed to an erroneous argument.

Where do such arguments go wrong? It is not just the presence of the argument but "its misemployment as a tactic to make an argument seem (unjustifiably) stronger than it really is." (page 274)

Locke's insight brought out this tactical element very well when he wrote (Hamblin, 1970,60) that men use the argumentum ad ignorantiam to "drive" others and "force" them to "submit" in debate.

Here is the question: Has this flaw in reasoning received a recognizable name? I am asking specifically about fallacies as they apply to a formal reasoning process...

An appropriate name would likely be "argument from ignorance". Such an argument might be weak. For it to be considered erroneous or a fallacy, however, requires a context where the one using the argument from ignorance is driving others to submit in debate. Just using the argument by itself is not enough for there to be fallacious reasoning.


Walton, D. (1996). Arguments from ignorance. Penn State Press.

1

Straw Man

The wording "Another person imagines a way in which it could happen, judges that way to be impossible" is what makes this a form of straw-man argument. The person imagining Y is the mechanism by which X is to be accomplished has thus changed the proposition from "X" to "X by means of Y", and thus by refuting Y, he believes he's refuted X as well.

If, historically, people have tried to accomplish X by method Y (and failed due to defects in Y), this is not an unreasonable reaction to X, so X proponents need to consider clarifying the original statement of X to exclude Y as the method, and if possible offer method Z to distinguish this X advocacy from those historical XY failures. If unable to do the latter, it's still true that even if Y is proven to be unable to produce X, that doesn't provide any information about other X methods.

4
  • I honestly think your post does not add a slight bit of information to be even tangentially referred to as an answer. I said this without a slight intention to offend you. I am not downvoting it solely because you tried to explain something at least. Dec 14, 2018 at 6:43
  • A straw man is an attack on a different argument than the one presented. The question is regarding an attack on the mechanism through which the claimed action occurs, which is directly related to the claim. It is therefore not a straw man.
    – Baldrickk
    Dec 14, 2018 at 8:39
  • @Baldrickk By adding a specific way, it's making the argument different from the one presented. Imagine "refuting" the statement "it is possible to drive from Calais to Dover" by saying "you can't drive on the English Channel, so you're wrong!" Clearly, the response is to a different argument ("it is possible to drive from Calais to Dover on the English Channel") from the one originally presented. Dec 27, 2018 at 20:39
  • @BertrandWittgenstein'sGhost Which part is hard to understand: 1) What's a straw-man argument? 2) By adding a method Y not stated in the original argument, the response alters the argument, and thus is a straw man. Dec 27, 2018 at 20:43

You must log in to answer this question.

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