After finishing David Deutsch’s book I was impressed that he was so impressed with the book "Conjectures And Refutations" by Sir Karl Popper. I live next door to a library to which can be summoned over the internet any crusty tomes from the depths of the deepest nerd archives in the region. I thought I’d take a look. What a brilliant book! Now I too am extremely impressed with Sir Karl.

Of all the famous German speaking intellectuals named Karl who moved to England to write their famous works, Marx is by far the better known. But the world would be a much nicer place, I think, if Popper was better known. Popper puts a fine point on it by ripping Marxism to small bits in the book. He’s no capitalist lackey however. He just wanted to repudiate the bad behavior of various Reds by pointing out that Marxism is not scientific. It is pseudo-scientific despite what Marx believed.

While the book lurched and veered from topic to disparate topic (it is a collection of papers and lectures), the main theme of the work as a whole is science. Sir Karl is a philosopher so the question asked and answered is, what exactly is science? If you call yourself a scientist I highly recommend that you read this book and learn about the philosophical foundations of your art.

In teenager detention centers (high school) one is taught a bunch of almost laughable nonsense called "The Scientific Method". Even modern "scientific" papers have an abstract, methods, results, etc., in a formulaic ritual that is supposed to magically impart the endeavor with Official Science essences. I think that unintelligibly boring prose is also supposed to heighten the magical effects. Sir Karl handily destroys this cult.

How does Popper show that many of the trappings of modern science are simply weird traditions? He uses a combination of two things. First, the guy is extremely knowledgeable about systems of thought and belief throughout history. And second, he uses the conception of science which he himself proposes to properly explain science itself.

Popper believes (and now I do too) that science is a progression of conjectures and refutations, just like it says on the tin. He takes, for example, the conjecture that the structured rituals of modern science and scientific methods are the valid necessary and sufficient conditions for science; he then refutes it with crystal clear example after example.

Another interesting boat he rocks that will surprise many people who have not studied philosophy is that he takes apart scientific induction. A lot of science people believe that scientific knowledge is gained by doing lots of stuff (experiments) and figuring out what the overarching principles are. Nonsense says Sir Karl!

Popper believes (and now I do too) that science, good and interesting science, is really just making up some crazy shit. Or as Sir Karl says, "Knowledge is an adventure of ideas." It’s that simple. I have to say that many of the science people I know could definitely improve their ability to make up crazy shit, an ability squeezed out of them by their rigid education. I feel like I’m actually pretty good at coming up with interesting conjectures but sadly, one does not win any NSF grants that way.

Wait a sec, you’re saying… Isn’t making up crazy shit what wing nuts and cranks do? Sure. The difference between science and pseudo-science isn’t that one is created with a fancy structured method and one is pulled out of your ass. The difference is that science is easy to test. The easier it is to make an unambiguous test, the more scientific a claim is. If the test turns out to actually refute the claim it is neither pseudo-science nor science - it is just, in fact, wrong. Move along.

On the topic of practical science tips, I thought this advice to scientists was really insightful and should probably be more seriously applied.

All this means that a young scientist who hopes to make discoveries is badly advised if his teacher tells him, "Go round and observe," and that he is well advised if his teacher tells him: "Try to learn what people are discussing nowadays in science. Find out where difficulties arise, and take an interest in disagreements. These are the questions which you should take up." In other words, you should study the problem situation of the day.

Although Popper’s major points are pretty clear, the implications are far ranging. The book covered all kinds of interesting areas. Of interest to people contemplating the future of AI (like Deutsch and me), Popper has many thought provoking ideas. Here’s an example which is as interesting today as when proposed in 1953.

To sum up this logical criticism of Hume’s psychology of induction we may consider the idea of building an induction machine. Placed in a simplified "world" such a machine may through repetition "learn", or even "formulate", laws of succession which hold in its "world". If such a machine can be constructed (and I have no doubt that it can) then, it might be argued, my theory must be wrong; for if a machine is capable of performing inductions on the basis of repetition, there can be no logical reasons preventing us from doing the same.

The argument sounds convincing, but it is mistaken. In constructing an induction machine we, the architects of the machine, must decide a priori what constitutes its "world"; what things are to be taken as similar or equal; and what kind of "laws" we wish the machine to be able to "discover" in its "world". In other words we must build into the machine a framework determining what is relevant or interesting in its world: the machine will have its "inborn" selection principles. The problems of similarity will have been solved for it by its makers who thus have interpreted the "world" for the machine.

In Popper’s time there was the Red Scare. Today people worry about super intelligent AI causing trouble as it pursues its own alien sense of Utopia. This bit discussing why Utopianism (of a Marxist flavor, for example) can’t work seems to me to also apply to AI design. He’s dispelling why the "ends" probably do not justify the means.

Any rational and non-selfish political action… must be preceded by a determination of our ultimate ends… That it is self-defeating is connected with the fact that it is impossible to determine the ends scientifically.

One could imagine the "political action" being an AI policy that involved the extermination of all humans. The idea is that if this AI is so smart and operating scientifically it won’t be acting based on strange emotional whims of some envisioned utopia. I’m only mentioning the idea which is much more carefully supported in the book.

Sir Karl is definitely one of the intellectually bravest thinkers who’ve ever lived. His basic premise is that good stuff only happens if you weed out bad stuff. It’s a big fat invitation, a reminder, to refute his idea of conjectures and refutations. But that’s hard to do because it’s a pretty good idea. I thought that this was a good micro summary of Popper’s whole school of thought.

…reasonableness is readiness to listen to criticism.

And Popper is ready. While that quotable line is all you need to know, you can rest assured that he is very careful about every detail (what is "reasonableness", "reason", "criticism", etc).

One surprising message that I was struck by was Sir Karl’s profound respect for tradition. I’m reminded of when my foreign friends ask me about the premise for Halloween; I tell them that the reason we celebrate Halloween is because that’s what we did last year. Basically Popper believes that tradition is not just tolerable, but essential to science and progress. He believes that everyone should understand and respect traditions. His critical departure from, uh, traditional thoughts on tradition is that he is anti-dogmatic. Basically he believes that traditions are always good until they are shown to be bad. And at that point, Karl would advise you to get a new tradition. That indeed sounds like a pretty reasonable way to live.

I don’t know if Sir Karl wrote this in English or had one hell of a translator but his writing was extremely clear and readable to me. I was reminded of the style of Bertrand Russell which both Karl and I admire immensely. It was a long book but I’m really glad I read it. I learned more about what science is from reading this book than I have from working with teams of research scientists for the last 12 years. That’s why I highly recommend that anyone involved in science take a look at this underrated and important work.