I was having a discussion with a friend the other day, and his way of arguing struck me as rather weird. He would constantly say things like:

Well that's just like this other thing (which I already have an opinion about)

When making this statement, he seemed to assume that I would share his view in this other example. The opposite thing he said constantly was:

Well, if you think that, do you also think this?

Again he tries to make an analogy of something he thinks is clear cut.

My thought was, that instead of "that's just like this other thing", one ought to provide the arguments that support the opinion of both things.

Is this an actual problem, or a legit way of arguing? If so, what is this fallacy called?

This is the actual discussion, if memory serves me right.

Example 1

Me: Just because more people think we shouldn't pay for smokers' healthcare, doesn't make it true, or right.

Him: But that is democracy.

This puts me in a position where I need to argue against democracy, which I am willing to do, but I believe it will be a lot harder to convince him of that, than the subject at hand.

Example 2

Me: We ought to want to pay for people who get sick, even from smoking.

Him: Do you also want to pay for people who break their legs skydiving?

Let me reiterate, this is not about whether the analogy is okay or not, simply if it is okay to use it like this

Scott Adams (in this video) seems to recognize this issue, but at the same time make it sound like it is something he made up.

This is different from Faulty Analogy in that the analogies might be perfectly corrolated.

  • I'd say this is an inductive argument: if man X is rich and man Y has $1 less, man Y must also be rich. He's starting from something you agree with and extending it until he finds your "line in the sand". Once he does, he would point out your line is arbitrary.
    – user935
    Sep 22, 2017 at 14:13

2 Answers 2


It's both. Analogy is never officially logically valid, but it is rhetorically cogent. And reasonable logic is never so precise that some degree of analogy is totally absent.

It is a natural part of every argument that is not about a simple, individual case. No useful generalization is made strictly upon the defined attributes of the things being generalized over outside of mathematics, and even then we generally do the generalization first, and then find the attributes that help us extend it further.

An analogy is an allusion to a parallel argument that could presumably be made by translating existing arguments for the analogous case. If the analogy is very strong, for instance, simply substituting a different individual for a given individual in the case at hand based upon a number of similarities between their circumstances, we would never expect the entire parallel argument to be made over again.

But what makes an analogy strong?

This is an aspect of the problem that Quine captures in discussing Natural Kinds. We do not know when two things are so much alike that they are substitutable and which things are not. Trying to reduce this kind of thing to rules fails in general. But we know how to navigate substitutions and other transformations needed to stretch existing cases to more general uses. We do it so naturally without analysis that when we attempt to subject it to analysis, we find very little ground to stand on.

We have to have ways of determining how broadly an argument based upon parallel properties actually applies. But when we have passed over from useful paraphrase to mere metaphor is something that is not a solvable problem.

  • I might not have made this completely clear. The analogies are arbitrarily strong. I agree completely with my friend, that "this is indeed just like that other thing". But all he did was find another thing to disagree about, sidetracking the discussion when he tried to find common ground. I'll add an example to my question. Aug 28, 2017 at 13:30
  • 1
    I just wanted to point out that in Common Law analogical reasoning is extremely important also since a Precedent can be set in one case which then forms the basis for a decision in similar cases later. Defining rules for which cases are similar enough to be treated the same is a difficult problem however and just because these arguments are valid in Law does not necessarily mean they are also valid for logical reasoning Aug 28, 2017 at 15:11

Me: We ought to want to pay for people who get sick, even from smoking.

Him: Do you also want to pay for people who break their legs skydiving?

This is a perfectly valid question. I don't even think it qualifies as reasoning from analogy.

The question here is, what fundamental principles do you hold?

You say that you are willing to argue against democracy, but believe it would be harder to sway him on that point. An interesting stance.

My own take would be that it would only be "more difficult" to sway him on it if either (a) your own views are not completely formed and clear in your mind, or (b) you don't inquire sufficiently about his views.

If you do believe that democracy is fundamentally broken, and he believes that democracy is fundamentally sound, then you are unlikely to get anywhere trying to convince him of an opinion that can only be founded in the brokenness of democracy.

To explain another way: you have made a statement which is, according to his basic principles, outrageous. There are only two ways he is likely to respond to that. (There are many possible responses but they are exceedingly unlikely.)

One is to directly and bluntly refute your statement. ("I think that possibly this is not really a chair I'm sitting on." "Yes it is.")

The other (which he is doing) is to make some sort of a guess at what your basic assumptions might be, and ask about them. That's what he's done.

Or rather, he's taken the basic belief that he holds (that people should not be forced into responsibility for the actions or conditions of others) which your statement violates, and has given you another (more extreme) statement which also violates that belief, to see if you will agree with that, too.

You could possibly shortcut the process by explaining your own basic principle which you believe supports the one statement (paying for sickness of smokers) but not the other.

My own belief is that arguing is not a good way to get people to change their minds, and that FRIENDLY conversation on another topic entirely is actually more likely to produce a change in beliefs than tackling someone head on about some bothersome illogical nonsense they are stubbornly asserting.

I would suggest you master how to communicate with another person before you worry about "logical fallacies" and whether the person's response is "legitimate." There are no illegitimate responses in a conversation.

  • I totally agree that, that is indeed what he is doing (The, take it to the extreme part). Yet it feels like sidetracking, with gaining little, ie. of cause my answer is yes to the skydiving, when it is to smoking. I make a claim, which I am willing to back up, but instead of asking why we ought to pay for smokers, he takes it to an extreme in what felt like an attempt at making me sound ludicrous. Do you mind elaborating on "there are no illegitimate responses in conversation"? Sounds unreal to me. Aug 29, 2017 at 6:51
  • I guess some of my issue comes from "But that is democracy". Is that really an argument? Aug 29, 2017 at 7:06
  • @ChrisWohlert, I missed these comments originally. "Legitimate" means "conforming to the law or to rules," according to the New Oxford American Dictionary. There are no "rules" governing a conversation in the usual sense (although there are certain natural laws)—but I believe you're applying the definition "able to be defended with logic or justification," whereas a conversation just as such is not a debate. (cont'd)
    – Wildcard
    Feb 2, 2018 at 3:13
  • @ChrisWohlert, also, if your answer is "yes" to the skydiving and yet you fear sounding ludicrous if you admit it...then I admit I am puzzled at your stance. Unless you view the two beliefs as entirely disrelated. (I wrote in my answer how they are likely connected in your friend's view.)
    – Wildcard
    Feb 2, 2018 at 3:16

You must log in to answer this question.

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