Yes, a blog.

by Academian3 min read19th Nov 2010108 comments


Intellectual Progress via LessWrongSite MetaCommunity

When I recommend LessWrong to people, their gut reaction is usually "What? You think the best existing philosophical treatise on rationality is a blog?"

Well, yes, at the moment I do.

"But why is it not an ancient philosophical manuscript written by a single Very Special Person with no access to the massive knowledge the human race has accumulated over the last 100 years?"

Besides the obvious? Three reasons: idea selection, critical mass, and helpful standards for collaboration and debate.

Idea selection.

Ancient people came up with some amazing ideas, like how to make fire, tools, and languages. Those ideas have stuck around, and become integrated in our daily lives to the point where they barely seem like knowledge anymore. The great thing is that we don't have to read ancient cave writings to be reminded that fire can keep us warm; we simply haven't forgotten. That's why more people agree that fire can heat your home than on how the universe began.

Classical philosophers like Hume came up with some great ideas, too, especially considering that they had no access to modern scientific knowledge. But you don't have to spend thousands of hours reading through their flawed or now-uninteresting writings to find their few truly inspiring ideas, because their best ideas have become modern scientific knowledge. You don't need to read Hume to know about empiricism, because we simply haven't forgotten it... that's what science is now. You don't have to read Kant to think abstractly about Time; thinking about "timelines" is practically built into our language nowadays.

See, society works like a great sieve that remembers good ideas, and forgets some of the bad ones. Plenty of bad ideas stick around because they're viral (self-propagating for reasons other than helpfulness/verifiability), so you can't always trust an idea just because it's old. But that's how any sieve works: it narrows your search. It keeps the stuff you want, and throws away some of the bad stuff so you don't have to look at it.

LessWrong itself is an update patch for philosophy to fix compatibility issues with science and render it more useful. That it would exist now rather than much earlier is no coincidence: right now, it's the gold at the bottom of the pan, because it's taking the idea filtering process to a whole new level. Here's a rough timeline of how LessWrong happened:

Critical mass.

To get off the ground, a critical mass of very good ideas was needed: the LessWrong Sequences. Eliezer Yudkowsky spent several years posting a lot of extremely sane writing on, and then founded, attracting the attention of other people who were annoyed at the lower density of good ideas in older literature.

Part of what made them successful is that the sequences are written in a widely learned, widely applicable language: the language of basic science and mathematics. A lot of the serious effort in classical philosophy was spent trying to develop precise and appropriate terminology in which to communicate, and so joining the conversation always required a serious exclusive study of the accumulated lingo and concepts. But nowadays we can study rationality by transfer of learning from tried-and-true technical disciplines like probability theory, computer science, biology, and even physics. So the Sequences were written.

Then, using an explicit upvote system, LessWrong and its readers began accelerating the historically slow process of idea selection: if you wanted to be sure to see something inspiring, you just had to click "TOP" to see a list of top voted posts.1

Collaboration and debate.

Finally, with a firm foundation taking hold, there is now a context, a language, and a community that will understand your good ideas. Reading LessWrong makes it vastly easier to collaborate effectively on resolving abstract practical issues2. And if you disagree with LessWrong, reading LessWrong will help you communicate your disagreement better. There was a time when you couldn't have a productive abstract conversation with someone unless you spent a few days establishing a context with that person; now you have LessWrong sequences to do that for you.

The sequences also refer to plenty of historical mistakes made by old-school philosophers, so you don't necessarily have to spend thousands of hours reading very old books to learn what not to do. This leaves you with more time to develop basic or advanced skills in math and science3, which, aside from the obvious career benefits, gets you closer to understanding subjects like cognitive and neuropsychology, probability and statistics, information and coding theory, formal logic, complexity theory, decision theory, quantum physics, relativity... Any philosophical discussion predating these subjects is simply out of the loop. A lot of their mistakes aren't even about the things we need to be analysing now.

So yes, if you want good ideas about rationality, and particularly its applications to understanding the nature of reality and life, you can restrict a lot of your attention to what people are talking about right now, and you'll be at a comparatively low risk of missing out on something important. Of course, you have to use your judgement to finish to search. Luckily, LessWrong tries to teach that, too. It's really a very good deal. Plus, if you upvote your favorite posts, you start contributing right away by helping the idea selection process.

Don't forget: Wikipedia happened. It didn't sell out. It didn't fall to vandals. Encyclopedic knowledge is now free, accessible, collaborative, and even addictive. Now, LessWrong is happening to rationality.


1 In my experience, the Top Posts section works like an anti-sieve: pretty much everything on there is clever, but in any one reader's opinion there is probably a lot of great material that didn't make it to the top.

2 I sometimes describe the LessWrong dialogue as about "abstract practicality", because to most people the word "philosophy" communicates a sense of explicit uselessness, which LessWrong defies. The discussions here are all aimed at resolving real-life decisions of some kind or another, be it whether to start meditating or whether to freeze yourself when you die.

3 I compiled this abridged list of sequence posts for people who already have a strong background in math and science, to accomodate a faster exposure to the LessWrong "introductory" material.

4 This post is about how LessWrong happened as a blog. For recent general discussion of LessWrong's good and bad effects, consider When you need Less Wrong and Self-Improvement or Shiny Distraction?