Induction and Critical Rationalism

What problem is the theory of induction intended to solve? What’s it for? What’s the goal or purpose?


This is a companion discussion topic for the original entry at https://criticalfallibilism.com/induction-and-critical-rationalism/

People often make mistakes about introspection even when they’re trying to be honest; understanding everything that goes on in your mind, in detail, is extremely hard.

If it wasn’t hard large parts of epistemology wouldn’t be so hard.


Overall I think the article wasn’t that useful for me. I don’t think it was bad, I just think that I’m not the target audience, I was already familiar with the content. I think it would be good for people who know very little about CR.

Based on the title I thought it would be more about logic and arguments of induction and CR, so I was a bit disappointed when it was more about the history and meta of the ideas. I’m not at all entitled to a different article, I had just built up some different expectations.

@Elliot, was that feedback useful?

There were still interesting parts that told me new things:

However, for many centuries, most people saw induction as the only way to avoid skepticism and rationalism, so they came to equate defending induction with avoiding skepticism or rationalism.

I don’t think I had thought of it in those terms. I knew that inductivists have opposed skepticism and that belief in induction has lead to/been accompanied by a pro-reason attitude. I think this is some useful historical context to know.

I haven’t read much Popper so I think this one is also useful:

Not all of this is super clear in Popper’s writing and he wrote a lot of other stuff too.

1 Like

Yeah the title isn’t great; it’s pretty vague. Titles can be hard. I don’t know what would have been better.

Yes the article is at a pretty introductory or overview level. I also thought a short, organized, clean overview could be useful to people who already know a lot. I also think it’s useful to have this article exist so I can link it to people like @actually_thinking; I don’t think it’s a duplicate of a previous article.

I don’t think it’s a duplicate either. I guess it’s just the story I’ve picked up on reading a bunch of your stuff.

The part about Popper critics would also have been new and very useful if I hadn’t read your recent blog posts on that topic.

I think the overview was short, organized and clean, and that seems useful for knowledgeable people too. Maybe I’m underestimating the value it had specifically to me because I had different initial expectations.

I heard Deutsch talking about this many years ago and have been trying to understand it. I understand that genes encode for the creation of biological organisms and dictate how the proteins should fold and I recognize this as a sort of information. But this sort of information (that for the creation of biological systems) seems fundamentally different from the information we get from sense-data that we then use to abstract from to form concepts.

Would you agree with this formulation so far?

Agreeing with that formulation would require agreeing that we do in fact abstract concepts from sense data (i.e via induction), which Elliot doesn’t believe.

Why does our human knowledge seem different to you than knowledge that genes have?

They don’t seem different in terms of what they allow their bearers to do. Genes use their knowledge of optics to create working eyes, knowledge of chemistry to build biochemical processes in their organisms, knowledge of their local environment to tell their organisms how to search for food, shelter, find mates, and many other things. This is also what our knowledge allows us to do. (We build cameras, chemical plants, and maps etc.) The difference is that we can create new knowledge in our minds and genes have to go through the long wandering protracted process of natural selection. And their selection criteria is fixed, but ours isn’t (we can decide to evolve ideas for any selection criteria i.e building cameras).

The problem that Darwin solved was: what process could create new organisms? He came up with the theory of evolution by natural selection.

Popper realised that Darwin’s problem was an epistemological problem, and that his own epistemology explained how Darwin’s solution worked generally. Popper showed that human knowledge can be created by the same variation and selection that biological knowledge is.

And these differences aren’t fundamental. In both cases knowledge is still created by evolution (variation and selection).

The connection between epistemology to biological evolution I think is a really important for the case for Popperian epistemology. Here you have a philosophical theory that underlies the logic one of the most well regarded scientific theories, a theory which is today is basically uncontroversial and unchallenged.

The connection means that if Popper were wrong about how knowledge was created then Darwin would have to be wrong about evolution, and vice versa. It’s the same underlying epistemological theory that connects them.

Genes knowing how to instruct proteins to fold and knowledge of how to hit a flop shot are completely different in that one is encoded and the other is learned through experience. Both are similar in the fact that they are formed through trial and error but differ in the fact that one is volitional and the other is not. One is conceptual and one is material.

Just to recap your position for myself: I’m reading this as saying that you think knowledge in animals genes and human knowledge are two distinct types of knowledge. That they are a “fundamentally different” “sort of information” to each other.

I don’t agree that that distinction is fundamental. Sure, you’re born with a gene to do x, and you later learn how to do a flop shot after you’re born. That’s a difference. But I think the process that creates each piece of knowledge is the same. The issue is whether evolution is or isn’t the process that underlies their creation.

Wait, don’t you disagree that conceptual knowledge is formed via trial and error i.e by evolution? Don’t you think that we abstract concepts from sense-data using induction? Or do you think we sometimes use evolution and sometimes use induction?

I also don’t think this is a fundamental distinction between the two pieces of knowledge (animal genes and human ideas). I brought this point up and said so just previously:

There are lots of differences you could point out, but the issue is is there an epistemological difference in terms of the how they’re created.

Popper’s claim is that all knowledge, genes and human idea, etc, is created by the same process. But you say but those kinds of knowledge are fundamentally different, what’s your point? That they’re created via different processes? That evolution creates some things that aren’t knowledge? That one should be considered knowledge and not the other?

If you only can respond to one thing my questions about trial and error is what I’m most interested in.

I think that’s the common sense view. I think people in the past used to think animals have instincts, and that instincts are fundamentally different from human knowledge. Modern people are more inclined to believe animals have intelligence like humans do but are just dumber. Even if they thought intelligence was genetically determined most wouldn’t say genes have knowledge. The prevailing theory is that knowledge is justified, true, belief.

I’m guessing he wouldn’t think of animal genes as knowledge.

If I didn’t already know about epistemological evolution I don’t think I would be convinced animal genes were knowledge from this.


@actually_thinking I suggest that you read Fallible Ideas – Evolution And Knowledge. Do you see the similarity between Paley’s problem and the question “where does knowledge come from?”?

No. Abstracting sense data to form concepts sounds like induction. And you don’t say what’s the same about the information/knowledge in genes and minds.

I did here.

But I will expand on the differences since this was also asked by LMD

Applying the same model–call it variation and selection, or trial and error– to both genes and theories is a category error. Biological evolution (BE) and epistemological evolution (EE) work in completely different ways. BE happens with genes and organisms and are blindly selected by environmental fitness. EE happens with ideas and is volitionally selected using criticism and reasoning. This is distinct line which Popper blurs.

Trial and error cannot explain how concepts arise, either. Biological variation is random and valueless with conceptual variation being purposeful and driven by values. When Popper says that everything is evolution he avoids confronting concept formation, and conflates the valueless with the purposeful.

I know that Popper opposed pragmatism and instrumentalism while defining truth to mean “the intuitive idea of truth as correspondence with the facts,” (Conjectures and Refutations, 1962, p.223) but, again, he seems to be blurring the line between utility and veracity. Popper seems to suggest that ideas survive because they are useful and not because they are true. His nuanced (and in my eyes vague) definition of truth invokes intuition which takes place at the perceptual level and not at the conceptual level. This is, again, reducing the volitional to the automatic. If ideas survive because they are useful then that is pragmatism. If ideas survive because they are true then that must be determined using reason and not intuition if we are to recognize our conceptual faculty.

If I were to sum up my view in a single sentence it would be this: Popper equivocates the term evolution to refer to fundamentally different things as if they were the same, namely the volitional and the non-volitional.

That comment doesn’t show understanding of what CR/CF thinks are the same about them. And I thought you were trying to focus on understanding the CR/CF position, which I thought that was a good idea. But now after getting a negative response about whether you were understanding it correctly so far, you’re arguing with it again.

I thought that I did understand it and then showed why it doesn’t make sense. My understanding is that the CR/CF position gives credit to evolution –through replication with variation and selection– for the creation of knowledge. And genetic knowledge and conceptual knowledge are the same. Evolution is what replaces induction with creating knowledge.

I think this is the argument. Did I miss anything fundamental?

This statement gives some conclusions, with minimal detail, without the reasoning or arguments that would persuade anyone.

I think your response here could be better.

What’s being analyzed here is CR’s position. And there is a claim of a difference which brings up two issues to compare two things. The claim seems to be that the answer is different between the two things for both issues. So you should consider what CR says about each issue for each thing (four total evaluations). Like this:

According to CR, is genetic knowledge encoded?
According to CR, is genetic knowledge learned through experience?
According to CR, is human knowledge encoded?
According to CR, is human knowledge learned through experience?

I view this as skipping steps: it’s better to have more clarity and agreement about what the similarities and differences are before deciding whether they’re “fundamental” (which I’d want to define and clarify the relevance and importance of – maybe that’s not the right issue to consider).

You conclude it’s a category error by giving a non-CR/CF summary of the issues. But do you know what CR/CF says about this? Is it a category error given CR/CF premises?

Is it ok if i try to predict the rest of some sentences? I wanna see where we differ.

1st quote:

they make arguments in favor of it.

In favor cuz they want it to work. They don’t see the downsides as something major that makes it not work. They don’t see that the primary issue is that it even works at all.

Actual quote:

theyre trying fix it? Seems like it’s broken. “defend it” is similar to what i said(“make arguments in favor of it”). Biased against alternatives? didnt think about that. That must get in the way of finding something that works

2nd quote:

method that doesn’t reach the two extremes.

By reach i think i meant like doesnt do skepticism and doesn’t do rationalism

the actual quote:

I see, using observation is important in the knowledge creation method. Since in rationalism one can use pure reason to get knowledge. Whats lacking in rationalism is that it doesnt use observation in its method

3rd quote:

it’s a working method that people can use

I say it’s a working method cuz the article was talking about I think how inadequate induction is. idk why i said “that people can use.”

actual quote:

Oh ok i like how persuasive the quote is. It sounds like a proper conclusion. I think it answers the “so what?” question of making an essay. So what? so people should be eager to give popper’s epistomology a try or just even consider it.

this part looks important that it was included cuz the essay was talking about how not good enough induction is.

1 Like