A Rationalist's Guide to...

by devils_rights_advocate1 min read9th Aug 201834 comments


Value of Rationality
Personal Blog

The one asks "why aren't rationalists ruling the world?" Do you have an adequate answer? No excuses allowed. I found that when I first heard the question, I didn't at all expect rationalists to be systematically outperforming others (except perhaps in IQ tests and the things they correlate with). I find my own expectations disappointing. I don't truly even expect rationality to be life-changing, but I would like it to be.

Rationality currently may very well be a filter for some smart people with common interests. So, how do we improve on our instrumental rationality? How do we make sure rationality is actually helping people grow? I'm obviously not the first person to ask this. I'd say Eliezer's first few essays in The Craft and the Community lay out the problem well. I don't really see a way to change the grand problem quickly, but I have a sense that there's some low-hanging fruit.

On an episode of the Bayesian Conspiracy, I forget which one, their guest said something along the lines that rationalists should be coordinating better to address growth in specific areas, such as health or the ability to increase your own predictive abilities. I quite agree. I think we could put together guides to many different topics of interest. For instance, "A Rationalist's Guide to Physics", assembling the best-known sources for learning about physics, with people sharing what did and didn't work for them. I started to think of this when I noted that the Feynman Lectures are cited as extremely good works for learning the basics of physics in a meaningful way, but had heard of no similar books for chemistry or biology. Might someone else know? Hm...

Another example could be "A Rationalist's Guide to Exercise", with analysis as in-depth as Scott Alexander's "Much More Than You Wanted To Know" series. I bet there's a rationalist that's done the reading and has the experience with the subject. There are many many basic things that actually have scientific support that a rationalist ought to know. One of my favorites being the simple trick of standing up every regularly, which can outperform exercising regularly followed by rest periods.

I've also been reading through MIRI's guide to start learning about AI alignment. There's CFAR's reading guide. And there's random rationalist-recommended books strewn across the internet. Have you tried to track down all the books Eliezer has recommended? I have. It's annoying. I've been trying to cobble together study plans for topics that interest me. My first idea was to look at my university's degree catalogs for all the majors I'd like to learn through self-study, pick out the courses that interest me but still ensuring a breadth of topics, and try to track down top-recommended textbooks for each course cheaply. This is a large task, to say the least. I think it could be crowd-sourced. Probably something modeled on MIRI's guide above, with books that many rationalists have read and can recommend.

I just know there's a chemist here somewhere itching to share all the best ways to learn chemistry. I'm dying to know! So let's make it happen! If this project is already being done, please let me know. If we can do it well enough, perhaps the Overlords Who Pick Recommend Sequences will pin them at some point. What kinds of topics do you think would be good for this project? Would you like to make a guide of your own? If you make one, message me and I can link it here to try to keep a central catalog of them all. Any chance this is a good idea?


31 comments, sorted by Highlighting new comments since Today at 3:37 AM
New Comment

To me the piece of writing that most seems like a rationalist guide to achieving goals is Nick Winter's "The Motivation Hacker". While I think Nick Winter didn't post on LessWrong, I would count him among the top ten people of the Quantified Self movement.

A few years later Nick Winter managed to go into YC without going to formal YC interview. He went to an onstage office-hour session and Paul Graham liked what he saw so much that he declared he directly accepted them into YC (and that's not something that Paul Graham did previously).

I met him at the CFAR Alumni Reunion 2014, so he is definitely connected to the community.

(Excuse my english)

This is, so far, the list of books that have contributed to my enrichment as a rationalists:

[*] midly related to rationalism

[**] slightly related to rationalism

On Intelligence - Jeff Hawkins

What Do You Care What Other People Think? -Richard Feynman

**Allegro ma non tropo - Carlo Cipolla

Descartes Error - Antonio Damasio

** Sherlock Holmes Complete Canon - Arthut C. Doyle

** Freakonomics - Steven Levitt

You are not so smart - David McRaney

*Moonwalking with Einstein - J. Foer

*Mastermind - Maria Konnikova

*What Everybody is Saying - J.Navarro

*Louder than Words - J.Navarro.

**Ender's Game - Orson S. Card

**The Martian - Andy Weir

*The Future of Mind - Michio Kaku

Predictably Irrational - Dan Ariely

Spent - Geoffrey Miller

Why Everyone else is a Hypocrite - R.Kurzban

The Moral Animal - Robert Wright

*Our Inner Ape - Franz de Waal

All Sequences - Eliezer Yudkowsky

Sequence - Science of Winning at Life - Lukeprog

The Archaeology of Mind - J.Panksepp

Sapiens - Yuval Noah Harari

Sequence - Living Luminously - Alicorn

Sequences - Yvain's Excellent Articles

Why Zebras Don't Get Ulcers - R.Sapolsky

All LessWrong - Kaj Sotala's Posts

Origins of Human Communication - Tomasello

The Selfish Gene- Richard Dawkins

The Joy of Game Theory - P. Talwalkar

What Intelligence Tests Miss - Stanovich

The Third Chimpanzee - Jared Diamond

The Handicap Principle- Amotz Zahavi

The Mating Mind - Geoffrey Miller

The 10,000 year explosion - G.Cochran

Superintelligence - Nick Bostrom

The Body Keeps the Score - Van Der K.

Facing the Intelligence Explosion- Luke Muehlhauser

Intelligence: All That Matters. Stuart Ritchie

**Hackers and Painters. Paul Graham

Surfaces and Essences. Douglas Hofstadter

*Inside Jokes - Hurley Matthew

All Melting Asphalt Blog Posts K.Simler

The Psychology of Social Status- Cheng, Tracy

All Luke Muelhauser LessWrong Posts

The Replacing Guilt Series - N.Soares

Metaphors We Live By - George Lakoff

**Deep Thinking - Garry Kasparov

Blog -Mindlevelup - The Book

Conor Moreton 30 days of Rationality - LessWrong

Dark Lords’ Answer. E. Yudkowsky

Why Beautiful People Have More Daughters - A.Miller

Influence. Robert B. Cialdini

The Art of Thinking Clearly. R. Dobelli

Inadequate Equilibria. E. Yudkowsky

*Never Split the Difference. Chris Voss

The Elephant in the Brain.Simler.Hanson

Hive Mind-Garett Jones

**If Universe is Teeming with Aliens Where is Everybody - S.Webb

*The Edge Effect. Eric Braverman

Crystal Society. Max Harms

*Our Mathematical Universe. M.Tegmark

Visual Introduction Bayes. Dan Morris

Bayes: The Theory that would not Die. S. B. C

*Skin in the Game. Nassim Taleb

Smarter Than Us. Stuart Armstrong

Language in Thought and Action - Hayakawaw

I would like to read more books! I am ready for suggestions.

What do you want to know? And I can suggest.

Thank you, Elo. A list of books that helped you to become more rational. I don’t care about the subject.


2017 reading list. But I should warn you that my path is not your path and some of them will waste your time if you don't have direction. Don't copy me if you can skip some.

Could you share your list of Eliezer recommended books?

Well, I tend to throw them onto my general to-read list, so I'm not entirely sure. A few I remember are Godel, Escher, Bach, Judgment Under Uncertainty: Heuristics and Biases, Influence: Science and Practice, The End of Time, QED: The Strange Theory of Light and Matter, The Feynman Lectures, Surely You're Joking Mr. Feynman, Probability Theory: The Logic of Science, Probabilistic Inference in Intelligent Systems, and Player of Games. There's a longer list here, but it's marked as outdated.

I've been thinking about this too, and I'm not sure guide suffice. Getting in shape or learning about a topic are simple problems (not that can't be challenging in their own right) compared to the complexity of actually achieving something.

At this point, we don't even have good theories or hypotheses on why these things are hard. It's lot of small issues that aggregate and compound. Motivation is a big class of these issues. Not seeing clearly enough - failure to perceive danger, opportunities, alternative ways of doing things.

To achieve you have to get the strategy, the tactics and the operations right. There's a lot you can screw up at every level.

One key issue, I think, is that it's damn hard to hack yourself on some fundamental levels. For instance to "be more perceptive". You can't really install a TAP for that. I guess some mindfulness practice can help (although I'd be wary of prescribing meditation -- more like mindfulness on the move). Consuming self-help, insights, news, etc etc only seems to move the needle marginally.

So yeah, I don't know. Just throwing some ideas out there.

Something like this: https://www.lesswrong.com/posts/qwdupkFd6kmeZHYXy/build-small-skills-in-the-right-order might be a nice starting point. Maybe, just maybe, we're trying to lift heavy weights without having built the required muscles. Worth investigating and expanding.

I think this is largely correct and points at where some of the larger bottlenecks are.

It's not about finding a list of good resources. There are a lot of those already. It's about what happens next. Things like:

  • Getting yourself to actually read said resources.
  • Figuring out ways of making the material stick.
  • Looking for applications, tracking your progress.
  • Repeating all of the above, over and over.

I definitely agree that there's a bigger issue, but I think this could be a good small-scale test. Can we apply or own individual rationality to pick up skills relevant to us and distinguish between good and bad practices? Are we able to coordinate as a community to distinguish between good and bad science? Rationality should in theory be able to work on big problems, but we're never going to be able to craft the perfect art without being able to test it on smaller problems first and hone the skill.

So yeah. I think a guide putting together good resources and also including practical advice in the posts and comments could be useful. Something like this could be the start of answering Eliezer's questions "how do we test which schools of rationality work" and "how do we fight akrasia". That second question might be easier once we've seen the skills work in practice. Maybe I should make a guide first to get the ball rolling, but I'm not sure I know a topic in-depth enough to craft one just yet.

Give me a month to make a fitness one. I train a bunch of my friends including one rationalist friend that has been pushing me towards writing some analyses of studies; so I have a good amount of experience trying to find ways to get people into fitness who've had issues fighting against their baser urges just to sit down and conserve calories.

How about A Rationalist's Guide to Early Retirement?

Or, If you're so rational, why ain'cha rich?

OK, so some of you got lucky with the Bitcoins. Can we do any better than buy-and-hold the index? (Like Wealthfront etc.) Option spreads? Tax liens? Arbitraging junk on Ebay? Do you have to start a company, or can you do just as well as a consultant? Are there easy ways to start up passive income or do you need a rich uncle?

If I would have a really easy way to do a task that produces a good passive income, I would hire low wage people to do the task till the task is saturated instead of writing it up.

Most good ways to make money is by leveraging what Peter Thiel calls secrets that derive from local knowledge.

Also called the efficient market hypothesis.

OK, so go meta. How does one go about discovering these secrets? We don't all have to find the same one. How can a rationalist find these secrets better than average Joe?

I'd like A Rationalist's Guide to Signing Up for Cryonics.

Suppose you've made the decision and have the finances to do it. How do you go about it? Which institution would have better expected outcomes? Neuro or full-body? Which life insurance company? What do you tell your family? How can you best ensure that you actually get froze before your brain rots in case of your unforeseen accidental death, as opposed to a more convenient death due to age or disease in a controlled setting like a hospital? (Which we might expect in a younger-aged group.)

I would like to have A Rationalist's Guide to Reading Science.

Particularly, how to benefit from existing scientific publications and understand them well enough to write a Rationalist's Guide to X or Much More Than You Wanted to Know About X, where X is some field without common knowledge consensus, like medicine or diet or exercise or psychology.

Reading science news headlines seems suboptimal. How confident can we be in any particular study? We know there are some perverse incentives in science. Publish or perish, new discoveries more valued than replications, p-hacking, etc. What should we be wary of? How much training do we need in the field? Is this impossible without a degree in statistics?

This also interests me. Some of my hopes for the Ottawa Rationality Dojo are that we can assemble people who are interested in skill development, and that we can build curricula that are useful to similar groups. I'm not convinced that I'm the kind of person who could follow these curricula alone or with an online community, so I'm trying to build it locally.

One major concern that I have with trying to build a curriculum for instrumental rationality is that the art must have a purpose beyond itself. I believe that it is for this reason that CFAR has realigned itself from teaching instrumental rationality to teaching instrumental rationality for AI alignment.

At the upcoming local SSC meetups, I will be asking "For what purpose do you intend to develop the art?" If I get answers, I'll post them to LW.

Sounds awesome! A meatspace group would be great, I'm sure. One of my issues with self-study is having nobody to go to when I have questions or don't understand something. Having an empirical goal can also tell you if you've succeeded or failed in your attempt to learn the art.

Having a group of rationalists to talk to in person has been invaluable to me this year. It's helping me emerge from depression, overcome impostor syndrome, and find my project. The previous sentence reads like the battles have been won, but they are being fought constantly.

Check this list of upcoming meetups: https://slatestarcodex.com/2018/08/09/ssc-meetups-2018-times-and-places/

Right now is a really good time to start or join meatspace communities.

I would like A Rationalists Guide to Personal Catastrophic Risks.

We like to think a lot about Global Catastrophic Risks (especially the EA folks), but there are smaller problems that are just a devastating to the individual.

Should we wear helmets in cars? Should we wear covert body armor? Own a gun? Get a bug-out bag? An emergency cube? Learn wilderness survival?

And by how much should we be concerned about those "survivalist" topics vs less obvious longevity steps like flossing your teeth? Not everyone's risk profile is the same. How do we assess that?

How should we measure that? Dollars? QALYs? Micromorts? Should we use hyperbolic discounting? Do we expect to reach actuarial escape velocity (or be granted near-immortality after the Singularity) and how would that change the calculus?

Do anthropic effects matter to subjective survival? In the multiverse?

Consider also other catastrophes that don't kill you, like losing a limb, or going blind, or more social risks like identity theft or getting scammed or robbed or sued, etc.

First of all, I'd just like to take a moment to say that I quite appreciate your username.

Second, to take your initial question literally, I don't think there are that many rationalists who actually want to rule the world. The position sounds like it would involve paperwork and talking to uninteresting yet obstinate people, so speaking for myself I don't think I'd actually want the job. There are probably many rationalists who would take the position for instrumental reasons, but because it's an instrumentally useful job, the competition for it is fierce. I'm not saying you meant it literally, but I think the distinction points at something important; what is it we actually want?

I'd like to be more in-shape, to work on more interesting programming projects, and to go on more successful dates. I'd pretty cheerfully read a guide on those subjects, and would probably be amenable to contributing to such a guide. Somebody else might want to save more lives, or have a higher class lifestyle, or lead a more interesting and exciting life. Some skills are generically useful to a large range of goals (financial management, persuasion, etc) but something that might be crucial to my goals might be irrelevant to yours. In addition, the format of whatever we're learning from matters; when learning to work out a youtube video is probably more useful than text. I would love to see more essays in the vein of SSC's Much More Than You Wanted To Know, but audio lectures, videos, or illustrations are good too. (Have you ever tried learning martial arts from a textbook? It's not ideal.)

Lastly, something worth thinking about. We all have the internet, and can all ask google for information. What advantage does a rationalist repository of teachings have? I'm confident we have some (offhand, we have common jargon, possibly a willingness to do experiments, and the occasional dedicated specialist) but if we want to do more than toss a lot of blogs in a pot and stir, it might be good to keep the comparative advantages in mind.

The rationalists are winning. You are not looking carefully enough.

Can you be more specific? What evidence leads you to believe that rationalists are winning?

Elo's a nice guy, but I have no idea what he's talking about either.

Maybe rationality improves your quality of life or subjective well-being, there is certainly evidence for that.

But in terms of accomplishing more material and outwardly visible goals, you're right that the evidence is scant. CFAR and EA could be evidence, but there are a lot of non-rat institutions that perform well too.

I think overall the success of EA and Rationality is pretty visible. Open Phil has access to over 10 billion dollars, which makes them one of the 10 largest foundations in the worlds, we have successfully created safety teams at many of the world's top AI labs, have had many of the world's top entrepreneurs and researchers speak at our conferences, and generally seem to be doing much better at achieving our goals than I bet almost anyone would have naively expected had you asked them in 2010.

Obviously, not everyone who reads LessWrong suddenly develops superpowers, and generally as communities grow the average level of success or competence goes down, but in aggregate I think we are doing pretty well.

(Note: I don't think most of Open Phil would self-identify as rationalist, but in particular their focus on AI alignment seems heavily influenced by the rationality community, and in general I think that a lot of the staff at Open Phil are executing the kind of algorithms that we usually describe here as "the art of rationality" (and many of them have read LessWrong and found it quite valuable))

I agree with the thrust of this comment, but I'd like to push back against "have had many of the world's top entrepreneurs and researchers speak at our conferences" as a measure of success (although perhaps it's a predictor in the right context).

Agree that it's a weaker one, I guess it's one that comes up for me because I worked more directly on it :P

No. You will have to see for yourself. Of course you'd have to be looking for that to work.

...? "Winning" isn't just an abstraction, actually winning means getting something you value. Now, maybe many rationalists are in fact winning, but if so, there are specific values we're attaining. It shouldn't be hard to delineate them.

It should look like, "This person got a new job that makes them much happier, that person lost weight on an evidence-based diet after failing to do so on a string of other diets, this other person found a significant other once they started practicing Alicorn's self-awareness techniques and learned to accept their nervousness on a first date..." It might even look like, "This person developed a new technology and is currently working on a startup to build more prototypes."

In none of these cases should it be hard to explain how we're winning, nor should Tim's "not looking carefully enough" be an issue. Even if the wins are limited to subjective well-being, you should at least be able to explain that! Do you believe that we're winning, or do you merely believe you believe it?