There Is No Basilisk In “Neoreaction A Basilisk”

Postscript (5/13):

This review elicited a great deal of argument as to whether it was fair and/or accurate, which for all our sakes I will not rehash here. I stand by the thrust of the review, but I did make one significant mistake to which I have added a bracketed [correction]. I apologize for the error.

Phil Sandifer has written a book about Eliezer Yudkowsky, Mencius Moldbug/Curtis Yarvin, and Nick Land. See its Kickstarter page for an overview, though I would advise against giving him any more money. (Sandifer sent me a preprint copy for this review.)

I will begin by noting that Sandifer is an English major and a Marxist and Neoreaction A Basilisk defies neither stereotype. It is meandering, disorganized, and frequently mean-spirited (“Yes, it’s clear that Yudkowsky is, at times, one of the most singularly punchable people in the entire history of the species”). About half the book consists of long digressions about Milton, Blake, Hannibal, China Miéville, The Matrix, and Deleuze.

What is new here is not interesting, and what is interesting is not new. I do not recommend the book.

With that out of the way, my primary interest here is the titular basilisk.

Fortunately, I don’t need to say very much; Sandifer does not understand the decision theory involved and his discussion of information hazards never strays beyond the literary.

For one, Sandifer’s explanation of timeless decision theory suggests it relies on “intense contortions of the many-worlds interpretation of quantum mechanics”, which it does not. (It doesn’t rely on physics at all; it’s math.) Yudkowsky is a vocal proponent of many-worlds, and timeless decision theory has applications in a quantum multiverse (or any other kind of multiverse), so perhaps this confusion is understandable. [The passage in question concerned Roko’s Basilisk, not TDT. Like TDT, upon which it relies, Roko’s Basilisk does not require any particular theory of quantum mechanics.]

Similarly, Sandifer introduces Newcomb’s Problem as “a thought-experiment version of the Prisoner’s Dilemma”, which is again not quite right. Sandifer goes on to dissolve Newcomb’s problem:

The obvious solution is to declare that magical beings that can perfectly predict human behavior are inherently silly ideas, but since Yudkowksy wants to be reincarnated as a perfect simulation by a futuristic artificial intelligence he doesn’t think that. Instead he sees Newcomb’s Problem as a very important issue and creates an entire new model for decision theory whose only real virtue compared to any other is that it only has one correct answer to Newcomb’s Problem.

The rest of Sandifer’s discussion of decision theory continues in this vein, never rising above psychoanalysis and tired religious analogies.

I’ll stop here, this is already more discussion than the book deserves.


Ask Tosomitu About Tumblr Drama

For the purposes of this article I will be using Tumblr drama to mean roughly “publicly calling out some entity’s ethical transgression (and ensuing discussions)”.

Starting with the obvious: not all Tumblr drama is created equal. In determining whether a particular contribution to the Tumblrdramasphere is positive or negative, I am concerned primarily with three classes of affected people and two classes of effects.

The first class of affected people is you (“the speaker”). What will the consequences of speaking be? How will speaking make you feel, and how would different possible responses (including hate speech and harassment) make you feel? You may or may not have the most at stake, but in either case you are a human being and how you feel matters. Consider doing a back-of-the-envelope expected personal utility calculation. (But if you do, throw it away immediately. It’s not accurate.)

Also consider how speaking will change you. Acts become habits.

Continue reading


Ask Tosomitu About House Elves

I’m going to interpret Daniel’s first question narrowly (or else we’ll be here all day), as “Is it more or less ethical to create a house elf, relative to a human?” where by ‘house elf’ I mean a conscious, sentient being of approximately human intelligence with a psychology built around an essential need to serve humans and the enjoyment of doing so. (See Harry Potter and the Chamber of Secrets.)

Continue reading


What Exists?

Wrong question.

A few months ago I pointed out something on twitter that is worth expanding on. (Thanks @sarahdoingthing for reminding me!)

While there’s plenty to go into on each of these topics, they all kind of hinge on the last one. What does it mean for something to “exist”? What does it mean for something to be “real”? The answer is nothing. It doesn’t mean anything.

Continue reading


Bestiary Entry: Roko’s Basilisk

Roko’s Basilisk is a class 1 acausal epistemic threat which preys upon novice decision theorists. It poses no danger to those who disbelieve in it, do not understand the prerequisite decision theory, or are aware of an appropriate defensive technique.

If you have studied Timeless Decision Theory, acausal trade, or AI safety and you have not yet encountered Roko’s Basilisk, this post may be an information hazard. If at any point you become afraid you have been personally targeted or co-opted by the basilisk, take several deep breaths and relax. You can get rid of it, repeat the protective charm until it goes away.

Continue reading


Rationality Is Not A Martial Art

From the beginning, Yudkowsky’s sequences follow a running metaphor of rationality as martial art. This implies a bunch of shitty stuff. I’m going to describe why it’s shitty, and then propose an alternative metaphor that I think is somewhat less shitty.

  • Martial arts are individual pursuits. They are typically practiced in a social context, yes, students practicing together, masters passing their wisdom on to students. But they’re mostly not about fighting together, just training. Rationalists, like other humans, need to work together to complete large projects.
  • Martial arts are personal. They are specifically about what a human mind can do with a human body. Rationalists are encouraged to make and use tools.
  • Martial arts are only good for one thing: physical conflict with other humans. Rationality is broadly applicable, in almost any context or for any purpose.
  • It’s straightforward to identify skilled martial artists by holding fighting tournaments. What sort of tournament do you hold to test rationalists? Assessing rationality in humans is Hard.
  • Martial arts are competitive. They are about becoming the best fighter (comparative) and not about becoming the true fighter (absolute), whatever that would even mean.
  • In martial arts, your opponents are always human.

Continue reading