Rationality: From AI to Zombies by Eliezer Yudkowsky

Rationality was an incredibly beneficial book for me. It was a bit of a slog to get through, but absolutely worth it. My cognition has improved, and I came away with a stronger mindset for learning and reasoning. The book itself is engaging and polarized; Eliezer Yudkowsky writes with an unapologetic style that can be irritating, educational, dramatic, and intuitive all at once.

What it’s about

Rationality is about a few different topics united by the theme of reducing cognitive bias and learning to think probabilistically (though not in the way that you might think—Yudkowsky emphasizes something called Bayes’ Theorem rather than the statistics you might have learned in Stats 101). He makes examples of how to apply this to subjects ranging from the lottery, to evolutionary biology, to religion.

Why I picked it

It’s basically a compilation of the Sequences on Less Wrong, which are composed of the blog posts that Yudkowsky wrote for the Overcoming Bias blog late in the first decade of the 2000s. I heard about Less Wrong on the CollegeInfoGeek blog by Thomas Frank, and the thought of improving my cognition and meta-cognitive patterns appealed immensely to me.

What I got out of it

I learned a lot. The most widely applicable concept that’s stuck with me is that humans, being irrational entities that believe we’re rational agents, have lots of ways of we perceive things and make decisions, and not all of them are logically sensible or even things we do consciously. It’s taught me a lot about how to think about how others think, and how to make less errors in my own reasoning. I’ve caught myself catching beliefs of mine that are inconsistent with each other.

Learning to think this way was a bit painful at first, but has been a net gain; learning to reason better means that I work through my emotions more calmly and more accurately, and am better at examining and learning from my mistakes.

In fact, I had a dream where I used a technique I developed after reading this. I dreamt that a significant other had been disloyal, and where I actually expected to feel livid and furious, I found dream-me saying “Yes. It hurts.” That “Yes, it hurts.” is an acknowledgement I’ve been using to face things that don’t go my way. It hurts, but accepting it doesn’t make it hurt more; it just frees up the ability to actually perceive things and deal with them.

Who I’d recommend it to

I’d recommend it to anybody who desires to improve their thinking, and is comfortable or at least willing to lay aside knee-jerk reactions to “esoteric” topics like evolutionary biology, cryonics, philosophy, and artificial intelligence (e.g. move past the initial reaction of “Artificial intelligence? Won’t that make something like Skynet?”). (That’s what’s known as the “Fallacy of Generalization from Fictional Evidence.”)

I would not recommend it for theists who take strong offense to strongly entrenched anti-theist viewpoints; Yudkowsky has a tendency to harp on religion (perhaps as a side-effect of being a former theist). A point in his defense is that he actively thinks about alternatives that provide the same benefits he sees in religion. I also would not recommend it for people who have a history of going easy on themselves. Yudkowsky expects a lot of himself, a viewpoint that you can learn a lot from, and will resonate with people who also do, but will turn off people who don’t.

I’d especially recommend it to college students with a basic background in math and physics; some novice familiarity with calculus and probability, and with electromagnetism, will help you get far more out of this book. Engineers, applied mathematicians, and similar students will get a lot out of it. (Engineers especially; something he stresses is that it doesn’t matter if you “tried”, if you had a responsibility for lives and you failed to live up to it, you screwed up.)

A few of my favorite quotes:

I realized that once I could guess what my answer would be—once I could assign a higher probability to deciding one way than other—then I had, in all probability, already decided.

What is true is already so. Owning up to it doesn’t make it worse. Not being open about it doesn’t make it go away. And because it’s true, it is what is there to be interacted with. Anything untrue isn’t there to be lived. People can stand what is true, for they are already enduring it. —Eugene Gendlin

Buehler et al.: asked students for estimates of project completion at 50%, 75% and 99% probabilities of being complete. 13% finished by the time they gave 50% probability they’d be done at, 19% at the time given a 75% probability level, and only 45% (less than half!) at the 99% probability level time.

So remember the Litany Against Being Transported Into An Alternate Universe: If I’m going to be happy anywhere, Or achieve greatness anywhere, Or learn true secrets anywhere, Or save the world anywhere, Or feel strongly anywhere, Or help people anywhere, I may as well do it in reality.

If you once tell a lie, the truth is ever after your enemy.

Leave a Reply

Your email address will not be published. Required fields are marked *