Epistemology, Scheduling, Bias and Iteration

Epistemology is about how to learn. I know you just wanted to know about cooking or getting a raise or something specific, but you need some way of learning about those topics. Well, you already have a way – you do learn some – but it’s mediocre, so it’s worth improving. Your way of learning is not consciously thought out; you don’t consciously understand it; you didn’t study how to make it good and design it; and it sometimes doesn’t work. E.g. maybe you regard yourself as “not a math person” because your current epistemology didn’t do a good enough job of enabling you to learn math. You don’t learn some things and give up on some discussions.

The key to epistemology is error correction. Most people are in favor of error correction already. Yeah, we should try to think critically. Yeah, we should try to find and fix mistakes. So what difference does learning more philosophy make?

Those attitudes to error correction are inadequate. A more rigorous approach is needed. Not just lip service to error correction but big, clear steps that you do.

This article will talk about a few philosophical points that can help.

Scheduling

For example, leave your schedule one third empty. That is a concrete, practical step which is significantly different than what most people do right now. That is part of what taking error correction seriously means. You have to leave time in your schedule for things to go wrong and to deal with those errors. That’s called buffer, safety or margin for error. Having a large buffer, like a third, is what it takes to have like an 80% or 90% chance to finish a task on time. The amount of extra time needed, to some extent, randomly varies. But if the expected time (50% chance to finish in this time) is 60 minutes, then maybe you have an 85% chance to finish in 90 minutes. Leaving a third of 90 minutes unscheduled means 60 minutes for the task and 30 minutes for buffer. So if you leave a third of your time unscheduled, then you can have time to finish most of your tasks instead of half of your tasks.

What about the good days when tasks finish on time or early? Take a break. Or do something that doesn’t require scheduling in advance (e.g. read/watch/listen to something, talk with a friend, meditate or write in your journal). Or work on your backlog of tasks you’re trying to get done which aren’t finished yet.

What if you run out of backlog? Great! It’d be good for you to stop being rushed all the time. Stop putting out fires and get ahead of the game. It’s way more efficient. You’ll get more done if you aren’t behind on a bunch of stuff all the time. If you want, you can schedule a little more aggressively once you’re fully on top of things. Or you could use the time for philosophy.

Another huge benefit of scheduling fewer activities is you control what you do better. Put the most important stuff on your schedule. What people often do is put too much stuff on their schedule and some of it never gets done. But they don’t consciously choose which things never get done. It’s better if you make conscious choices about which stuff is worth going on your schedule and which isn’t, rather than just seeing in retrospect which stuff didn’t happen (which is often due to bias and due to preferences you don’t admit having, and your messy schedule helps you fool yourself).

Combatting Bias

Another philosophical idea that makes a big practical difference is thinking about reasons you don’t do something. People commonly won’t do stuff, like think about or discuss a certain topic, and don’t say why. They dodge certain critical questions and explain nothing. Maybe they generically say they are too busy to answer everything. This allows bias.

People are biased about what they do and don’t do. Thinking about (and even sharing and discussing) reasoning about their policies would be better and allow for bias to be visible and criticized. That’d help with error correction.

This applies to public intellectuals but also e.g. to talks with your spouse. It applies to a boss who gives orders to subordinates or to a teacher who ignores some doubts of students. And it applies when thinking about stuff by yourself. Being open and honest with yourself is more important than what you do with others. When people lie to or mislead others, most of the time they were lying to or misleading themselves first.

You can give specific reasons you don’t answer a particular question or think about a particular topic. Or, alternatively, you can give a general purpose policy for how you spend some of your time and then that can be applied to many different potential activities. A general purpose policy should have visible anti-bias stuff in it. For example, you should answer some questions from your critics/enemies and some questions at random, not just choose everything to address yourself. That way some hard questions can get past your biases (even if you don’t know what your biases are).

Letting some of your allies or friends pick some things, instead of you, is good too. You’ll generally be happy with their choices (they are similar to you) but they won’t have the identical biases that you do. They’ll choose some issues for attention (and potential error correction) that you would have dodged. Also, you’ll choose some issues to address that they wouldn’t have; choosing some yourself is good too. Don’t have all the questions you face be screened by a handler who is trying to avoid the hard questions (lots of politicians are guilty of this) or screen them all yourself (you have biases you don’t know about).

If you say why something is boring or not a priority to spend time on, that gives potential for error correction for you and others. Someone could learn from you why to drop the matter and spend their time on better issues. And, in the alternative, someone could say why you’re wrong and what you’re missing. This dual benefit is typical when you rationally deal with error correction – you should be saying a lot of stuff where someone else could learn something and/or correct you. Exposing ideas to criticism also exposes them for others to learn from; it’s two good things at once.

Iterating

To deal with bias well, you need to do multiple iterations (back-and-forths) on stuff. You can’t just answer someone’s preliminary questions and then say “OK, you had your turn.” This applies whether “you” is another person or a voice in your head with doubts.

For most people, a critic can’t get anywhere with you with just the first one to three rounds of questions or arguments. Why not? First, because you didn’t detail your own ideas enough publicly for the critic to respond to. He can suspect there’s a disagreement, but he has to ask you to share more ideas, see what you say, and then (if there really is an important disagreement) finalize and share his argument.

Also, ideas you share have ambiguities, so he has to use questions clarifying some points of your position before he can challenge it. (Alternatively, he can make risky guesses to try to speed up the process.) And he may say something ambiguous and have to follow up to clarify it for you.

In general, people have lots of misunderstandings, so some iteration is needed to clarify things.

In general, with two smart and knowledgeable people, it takes a bunch of quick back-and-forth to get to something new. You say something I’ve heard before. I say my standard response. I don’t know what you will say back to it (there are perhaps three standard options) so I can’t skip steps even though it will turn out that I’ve heard your view before. You’ve heard my first idea before too, but you couldn’t predict it because there are a few other things you’re also familiar with that I might have said. So we should quickly go through stuff we’re both already familiar with until we get to the first thing where someone has to stop and think and say “That’s new to me; I hadn’t already thought about that before.” But most discussions end before that. Public intellectuals do podcasts where they take listener questions but they only take one question per person and it never gets to deep stuff. They’re always repeating their main points without being meaningfully challenged.

There are some intellectuals who’ve said enough publicly that a critic can say something substantive in reply to their published work, without needing a bunch of back and forth first to clarify their position. What typically happens next? Those questions get ignored in Q&A periods as “too detailed” or something that would take too long to answer. Is that intellectually honest? Consider how many public intellectuals write a monthly blog post responding to the hardest critical question they got all month (in their own opinion) and giving it a serious answer. I don’t know of anyone who does that… Nor do people write even a yearly blog post responding to a randomly selected criticism or a criticism selected by their prominent critics.

If you do well at anti-bias techniques, you could have an advantage over almost everyone.