The Book of the New Sun, the Self, and Finding Rationalism in Science Fantasy

    I recently finished my first reading of Gene Wolfe's The Book of the New Sun. It was vivid, profound, difficult, much more literary than I expected - perhaps the best fantasy I've ever read. I won't talk about the plot here. Read it! I can say that it is strewn with insights that Wolfe casually tosses about like old clothes. One for example, outlines tensions between the individual self, companionship and the collective self:

"I have never had much need for companionship, unless it was the companionship of someone I would call a friend. Certainly I have seldom wished the conversation of strangers or the sight of strange faces. I believe rather when I was alone I felt I had in some fashion lost my individuality; to the thrush and the rabbit I had not been Severian, but Man. The many people who like to be utterly alone, and particularly to be utterly alone in a wilderness, do so, I believe, because they enjoy playing that part. But I wanted to be a particular person again, and so I sought the mirror of other persons, which would show me that I was not as they were."

I found thematic echoes of these ideas, the desire for individuality, the need and the constraint of companionship in a couple of movies I saw recently during the Omicron wave: Wit and Nomadland. Good stuff.

Another quote I found recalls a Scott Alexander article entitled "What Are We Arguing About When We Argue About Rationality?"

    To outline Alexander (I promise this will eventually get to Wolfe), he thinks that Rationalist and Anti-Rationalist arguments have been oversimplified by their opponents: Rationalists are characterized as advocates for deep logic and quantitative reasoning, whereas Anti-Rationalists are seen as proponents of heuristics and traditions instead. Alexander acknowledges that neither of these positions are as absolute as they might be characterized by their opponents. That's the thing about an argument, sometimes you need to steamroll nuance in order to find something to argue against. 

    Eliezer Yudowsky conceptualizes rationalism as a kind of "systematized winning". This way of thinking about the subject rolls fidelity to truth into consequentialism. By this conception, rationalism explicitly allows the use of heuristics and rewards strategies that work. Not an exhaustive definition, but also not an exhausting one either. 

    Alexander notes that this definition trivializes the idea of rationality into simply seeking the best outcome, so he reformulates, emphasizing the systematized aspect: "Rationalism is the study of truth-seeking, i.e. the study of study." He points out that it is easy to become confused between the object level of study and the meta level (study of study).

    He then argues that the reason Rationalists tend to be drawn to science and quantitative methods is that they need a level of transparency and replicability to effectively study a process. Heuristics and traditions may lead to desired outcomes, but it is hard to generalize these ideas or learn from them without understanding why they work.

    Or you could have just read The Book of the New Sun:

"They could tell everybody what to do if they know real magic."

I only shook my head to that, but I have thought much about it since. It seems to me that there are two objections to the boy's idea, though expressed in a more mature form it must appear more convincing.The first is that so little knowledge is passed from one generation to the next by the magicians. My own training was in what may be called the most fundamental of the applied sciences; and I know from it that the progress of science depends much less upon either theoretical considerations or systematic investigation than is commonly believed, but rather on the transmittal of reliable information, gained by chance or insight, from one set of men to their successors. The nature of those who hunt after dark knowledge is to hoard it, even in death, or to transmit it so wrapped in disguise and beclouded with self-serving lies that it is of little use or value. At times one hears of those who teach their lovers well, or their children; but it is the nature of such people seldom to have either, and it may be their art is weakened when they do. 

The second [objection] is that the very existence of such powers argues a counter-force. 

So what is to save the world from dark sorcerers? Bad record-keeping. If you squint, you can see an argument about how the pursuit of power at the expense of truth and transparency costs all progress. I must say that I'm not really cherrypicking here. These aren't the best insights from Wolfe's books. There are dozens if not hundreds more. That prolificness feels like luxury.