4

According to WP's article on abductive reasoning:

[A]bduction is formally equivalent to the logical fallacy of affirming the consequent [citation needed] because of multiple possible explanations for b.

Is abductive reasoning necessarily a logical fallacy, and if not, why is considered a form of important logic?

As far I know A implies B does not mean B implies A. For example, abductive reasoning:

P1. Whenever it rains, the streets get wet.
P2. The streets are wet now.
C. Therefore it must have rained.

5
  • You have inferred an explanation, but have you inferred the best explanation (abduction)? Perhaps not. It is a "logical fallacy" to say you must affirm the antecedent in order to infer the best explanation, even if you can deduce the explanandum from the best explanation. Abduction is not deductive logic.
    – user62090
    Commented Aug 7, 2022 at 18:51
  • Two good references in SEP: plato.stanford.edu/entries/abduction and plato.stanford.edu/entries/abduction/peirce.html
    – J D
    Commented Aug 7, 2022 at 19:17
  • There is a distinction between formal and informal fallacies. Abduction is (obviously) formally fallacious, its form of inference does not guarantee that true premises never produce false conclusions. But, when properly applied, it is informally valid, i.e. provides rational grounds for provisionally accepting the conclusion absent contrary evidence and subject to revision should it emerge, see SEP, Defeasible Reasoning.
    – Conifold
    Commented Aug 8, 2022 at 12:14
  • @JD The whole medical science is based on abductive reasoning A person having a heart attack tends to have a feeling of choking, pain in the chest radiating to the left shoulder and arm, abnormal perspiration, breathlessness, and nausea. Fanny Jenkins has just experienced a feeling of choking, pain in the chest radiating to the left shoulder and arm, abnormal perspiration, breathlessness, and nausea. Therefore it is reasonable to conclude that Fanny Jenkins had a heart attack
    – quanity
    Commented Aug 8, 2022 at 17:45
  • You betcha! en.wikipedia.org/wiki/Differential_diagnosis?wprov=sfla1
    – J D
    Commented Aug 8, 2022 at 18:01

5 Answers 5

5

Is abductive reasoning a logical fallacy?

Abduction is certainly not a logical fallacy.

Abduction is the reasoning that if ϕ ⇒ ψ is true, then ϕ is ipso facto a possible explanation for the fact that ψ is true.

There is absolutely nothing fallacious about that. This is just human logic at its best.

If you don't like it, don't do it, but the truth is, humans cannot stop their own brain doing it.

As far I know A implies B does not mean B implies A.

Indeed, and this has no relation to abduction whatsoever.

Abductive reasoning- Whenever it rains, the streets get wet. The streets are wet now. Therefore it must have rained.

This is a misrepresentation of abductive reasoning. Don't believe everything you find on the Internet.

5

I'll weigh in validating Cort Ammon's answer (it depends on the context of the abduction) and provide some examples.

P1. Whenever it rains, the streets get wet.
P2. The streets are wet now.
C. Therefore it must have rained.

Logical fallacy, since a construction truck may have recently watered the street to keep dust from being stirred up by passing vehicles. Here the logic presumes deductive certainty of the conclusion where none can be had. This example is a fallacy.

P1. Whenever it rains, the streets get wet.
P2. The streets are wet now.
C. Therefore it probably has rained.

Not a logical fallacy, since from a frequentist's perspective, the majority of times streets are wet all things being equal is from rain and not another source. Hence, the logic presumes inductive certainty. This example is not a fallacy.

Note that some consider there to be a distinction between abduction and inference to best explanation. Another example of abduction could be:

P1. Whenever it rains, the streets get wet.
P2. The streets are wet now.
C. Therefore many members of the the local community may have simultaneously had leaky water service lines resulting in a temporary flood.

Also inductively certain but manifesting a less likely frequentist's probability. Thus, one might extend arguments 2 and 3 to address why one conclusion is more probably than another to determine which inference to explanation is the best. This example is not a fallacy, but is an uncogent inference and therefore not a candidate for "best" explanation.

4

It depends on how you're selling it.

A formal logical fallacy requires that you make an assertion that does not follow from the rules of inference you are using. If you make an abductive inference and claim it to be deductive, then you have committed a fallacy. If you recognize what you are doing as abductive, it is not a formal fallacy.

A similar situation can occur with inductive logic. Many mathematical proofs come in the form of "prove the statement is true for n=0, and prove that "if it is true for n=k, it is true for n=k+1. If both are true, then the statement is true for all natural numbers", a so called inductive proof. This is recognized as logical in most mathematical settings, but would be a fallacy if you are specifically working without the law of induction.

It could be an informal fallacy if the listener of the argument is not one who is swayed by abductive logic. The different kinds of inferences do lead us to different outcomes.

All that being said, know thy listener. In my opinion, most people who throw around the term "logical fallacy" are referring to fallacies in deductive reasoning. As such, any inductive or abductive reasoning will be considered to be a "fallacy." But check with the listener. Make sure you both agree on what the term means before proceeding!

3
  • Glad to see you reposted, @quanity. I'd give extra weight to this answer. Abduction is far more complex a notion than deduction, and the devil lies in the details.
    – J D
    Commented Aug 7, 2022 at 17:38
  • 3
    Mathematical induction is not the same as inductive logic or inductive reasoning. It's still taking a serious of deductive steps. Though the wider point still stands that abduction and induction would be formal fallacies when being presented as deduction, but don't need to be if they're presented for what they are.
    – haxor789
    Commented Aug 7, 2022 at 22:13
  • 1
    Maybe an interesting rephrasing of the core of this answer: A fallacy is a stated falsehood. You claim that something is definitively so, yet it isn't universally correct even though you claim that it is. The claim is the fallacy. An abductive reason is an educated guess. As long as the abductive reason is clearly identified as such, it is not claimed to be a definitive truth - it's an uncertainty with some subjective estimate as to which outcome is the most likely. It only appears to be a fallacy to anyone who interprets is as (or claims it to be) a definitive truth.
    – Flater
    Commented Aug 8, 2022 at 9:04
2

Abudctive reasoning is not a logical fallacy. Abductive reasoning is drawing probable inference from context and experience. Note the word probable. Abductive reasoning does not produce logical necessity, as does deductive reasoning.

The second question is a highschool level intro to logic type thing. The statement "A implies B" does not necessarily mean that "B implies A".

0

It's stated right there in the wikipedia article: https://en.wikipedia.org/wiki/Abductive_reasoning#Abduction

As such, abduction is formally equivalent to the logical fallacy of affirming the consequent because of multiple possible explanations for b.

And yes from "all yellow cabs are yellow" you cannot just follow "All yellow things are yellow cabs".

14
  • 3
    It seems to me the wikipedia article is wrong here (note the claim is uncited, it's just some random editor's opinion), abduction does not automatically assume that if A->B, an observation of B should then justify an inference of A. After all there may be multiple contradictory conditions that could imply B, say A->B and C->B, abduction does not mechanically require you to endorse all these conditions at once. It's also my understanding that abduction is supposed to be specifically about causal explanations, not just about correlated facts like "X is a cab"->"X is yellow".
    – Hypnosifl
    Commented Aug 7, 2022 at 16:00
  • 1
    Yes NOW there's a [citation needed] because you added that in the Wikipedia article, the Edit pages literally shows your nickname :D That being said, fallacies only really apply to deductive reasoning because it's the one arguing from a general rule to a specific example, the abduction and induction are inferring rules from examples which is by default more of an educated guess so there's no point in calling that a fallacy unless they pretend it's more than that. Also the German wiki seems more instructive: de.wikipedia.org/wiki/Abduktion#Vergleich_der_Schlussweisen
    – haxor789
    Commented Aug 7, 2022 at 17:38
  • 1
    @haxor789 Deduction is not necessarily arguing from general to specific. Mathematical induction is deductively certain and moves the other way. Deduction is about the certainty of outcomes. And the idea that fallacies can't be informal has never been uttered, to the best of my knowledge, by any contemporary logician ever.
    – J D
    Commented Aug 7, 2022 at 17:44
  • Context matters in the application of fallacy, and one is only affirming the consequent if one claims one conclusion is certain. In the context of non-monotonic logic, like this that model defeasible inference, such a claim isn't even possible. CSP's work was developed in such a context.
    – J D
    Commented Aug 7, 2022 at 17:48
  • 3
    Just to clarify I wasn't being sneaky, when I said "the claim is uncited" I just meant the article provided no source for the claim, I added the "citation needed" to the article as an afterthought after posting the comment.
    – Hypnosifl
    Commented Aug 7, 2022 at 17:50

You must log in to answer this question.

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