I'm a bit confused by the criticism here.

That is, this argument has no problems with logical sequence and if the premises are correct, then the conclusion will also be correct? Why do many experts consider it unsuitable?

The principle of uniformity of nature and preservation of symmetry is also an argument. by analogy or are these inductive arguments?

Thank you.

Thank you.

  • 1
    Argument from one's own analogy is naive (some would even call mind projection fallacy) and not unlike the famous private language argument wherefrom no others can check the said analogy in an objective way. Also in the case of existence of other mind such analogy is extending one’s knowledge from a single analogical case which may not be reliable. Lastly do you really know you yourself have a mind founded upon Descartes' dualistic cognito in the first place, not a p-zombie with phenomenal illusions under physicalist monism?... Sep 8, 2023 at 5:21
  • You are right: the argument is very very plausible. But it is an argument and not a proof. Sep 8, 2023 at 6:15
  • 1
    "If the premises are correct then the conclusion is correct" is true of formal arguments only. Nothing substantive can be proved by such arguments, they just recombine the premises. So people use informal arguments, by analogy is one of them, and those do not have such infallibility. Any analogy, by definition, is imperfect in some way, and it may turn out to be just the way that matters. Informal arguments provide, at best, only plausibility, and may well derive false conclusions from true premises. SEP cites arguments why this analogy is particularly imperfect, even for plausibility.
    – Conifold
    Sep 8, 2023 at 9:51
  • If, instead of analogy, we use induction based on the principle of uniformity of nature and support it by the fact that we have no reason to violate symmetry or the principle of uniformity. In this case, it will be more reliable? If so, please show how this argument will look.
    – Arti
    Sep 8, 2023 at 10:24
  • Symmetry could be violated in nature as manifested by the wellknown parity breaking in quantum physics, and there’s no unfalsifiable physical theory of everything yet to prove that almost all phenomena in nature has to be reducible to a few simple recursive fundamental laws. Ergo based on our limited contingent experiences so far in this Saha world we cannot infer or induce the absolute induction of uniformity of nature to arbitrary things especially those with obviously high complexity such as a so-called mind, though such inference may be the best explanation so far in a posteriori sense… Sep 9, 2023 at 4:32

2 Answers 2


Induction is argument by analogy.

  • We see X holds in case A
  • Case B is similar enough to case A in relevant characteristics
  • Thus, we think X will hold in case B too.

That is induction, and also analogy.

Another word for analogy is "pattern-matching." We see that X holds in case A, and case A matches pattern P; case B also matches pattern P; thus, we think X will hold in case B too.

A neuron does nothing but match patterns. It fires under some circumstances, and not under others. The circumstances under which it fires are the pattern recognized by the neuron. We might equivalently say that the neuron is constantly inspecting the surrounding circumstances to see if they are similar enough to the circumstances under which it should fire, and if so, it fires.

Because neurons just match patterns, arguably practically everything the brain is doing is analogy.

More specifically, in regard to your question, the argument from analogy for other minds could also be called induction, or pattern-matching, whichever you want to call it. And it does provide some partial (probabilistic) evidence for other minds.

The SEP article lists the objection to the argument from analogy, that other minds cannot be directly checked. Well, neither can the presence of a table!

  • We look at the table (it sure looks like a table; an analogy.)
  • We touch the table (yes, it feels like a table; an analogy.)
  • We disassemble the table (yes, those look and feel like parts of a table! an analogy)

But we can never directly check that there is a table there. Only that it looks and feels like a table, perhaps also smells and tastes like a table, weighs the same as a table, appears to hold plates like a table, and on and on. All by analogy. There is always the possibility that the apparent table might be some clever illusion, that we could be trapped in a dream, or something like that. This is precisely the same situation we are in when it comes to other minds.

All confirmations are probabilistic, never perfectly definite. Everything we know about the empirical world is by analogy/induction/pattern-matching. The argument from analogy for other minds has no more difficulty in this regard than any other claim about the world.

  • But all arguments in favor of the existence of other minds rely on their own case, which makes them arguments by analogy, but in other arguments, analogy is used as an explanation that is justified by further premises. Can an argument by analogy be used to justify the existence of other minds as an initial reason that can then be supported by other arguments?
    – Arti
    Sep 8, 2023 at 7:37
  • @causative "A neuron does nothing but match patterns" Could you add some academic references? Sep 8, 2023 at 15:09
  • 1
    @causative "We look at the table (it sure looks like a table; an analogy.)" Not an analogy. Looking at a table, we are certain it is a table. No analogy. The brain says it is a table. Analogy starts when looking at something we think is not a table (it doesn't look like it), but is somehow analogous to a table, say, it has four legs, perhaps a dog. We are certain it is a dog, but we elect to think in terms of the analogy with the table. You are mixing up irrelevant considerations about neurons with logical reasoning. This is a category error. Sep 8, 2023 at 15:22
  • @Speakpigeon If you want to split hairs, it's not the table that looks like a table, it's the sensory data we are receiving that looks like a table. You are not certain it is a table; it's just sensory data. It could be an illusion, you could be in a dream, you could be fooled by a demon, you could be having a stroke. But the pattern of the sensory data is similar to the sensory data resulting from tables you've seen in the past, so that's what you reasonably assume it is.
    – causative
    Sep 11, 2023 at 5:08
  • @causative "the pattern of the sensory data is similar to the sensory data resulting from tables you've seen in the past," Exactly, which is why we say not that what we see is analogous to a table but that it is a table (and possibly the same table). We don't say that this is an analogy because this is not what we mean by analogy. I suspect the word "analogy" is not even used by anyone specialised in neuronal processes. Sep 11, 2023 at 16:13

The problem with an argument-by-analogy in this particular case, as I see it, is that the assumption of solipsism necessarily renders any analogy moot. Analogy works by creating a similarity-class: A and B are similar in this respect, and so A and B can be treated as analogous in this respect. But by definition solipsism questions the existence (i.e., similarity) of other minds. No similarity-class can be created without violating the first principle, so no analogy can hold.

It's a bit like the Liar's Paradox: we assert that all other minds are false and ask for a means to prove they are true. That de-evolves into a pointless snark hunt.

  • "No similarity-class can be created without violating the first principle" I don't understand your point. Assuming a solipsist could exist to begin with, he could deny the existence of other minds and still be able to see analogies, for example between a dog and a table because they both have four legs. Sep 8, 2023 at 15:26
  • @Speakpigeon: But the specific analogy asked for here is that other people are like me in various ways, therefore other people have minds like me. A table/dog comparison involves two objects, but this self/other comparison involves (what the solipsist sees as) an object and a subject. It's like saying "dogs and tables both have four legs, therefore tables can think like dogs". Sep 8, 2023 at 15:37
  • 1
    "It's like saying "dogs and tables both have four legs, therefore tables can think like dogs"" But this is the point of a reasoning by analogy. If tables don't think, the argument is fallacious, but the analogy and the reasoning by analogy don't disappear. That both dogs and tables have four legs is too poor an analogy to produce good inferences, but from the better analogy between humans and dogs we infer very reasonably that dogs have a mind. Sep 8, 2023 at 16:02
  • 1
    Whether we can check that dogs really have a mind is irrelevant. If we had to be able to check that the conclusion is true to be certain of the validity of the argument, logic would be totally useless. We would just leave logic alone and check directly our facts. Sep 8, 2023 at 16:04

You must log in to answer this question.

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