The first challenge is answering the implicit "Why should I care about rationality?" question in a compelling way.
My suggestion is to decide what your answer is for that session, drop the "rationality" label, and talk about that instead. For example, if you're going to talk about the benefits of rationality for making decisions under uncertainty, then your topic is actually "Making decisions under uncertainty."
You may end up with multiple rationality talks... this isn't a bad thing.
Choose concrete examples wherever possible to make your point. Generalizing from a few concrete examples is fine, but starting from the general case can cause trouble. Make your examples either realistic or funny (or both!).
Tie endorsed behaviors to concrete results: "doing X makes Y more likely."
Don't try to cram in too much. Make one primary point, make it well. Have a couple of secondary points in your back pocket if your audience moves a lot faster than you expected, but don't expect to get to them.
Operationalize: give people things they can do now -- either right then, or right after the meeting -- that relate to the point your talk.
I have an idea-outline for a very public-friendly lecture on "why should I care about rationality," and sooner or later I'm going to make a power-point of this and hopefully present it.
Main point: probability calculations will help you make better life decisions.
In the movies, maybe the Chinese soldiers pull through, against all odds. But in real life, you'd rather be Genghis Khan. Most people do not succeed against all odds; people succeed because they've set up the odds to favor them.
The best way to predict what's going to happen to you is to look at what typically happens to people in similar circumstances. Thinking "In my case, it'll be different" is very common, but it usually results in failure. You want the statistics to be on your side. You want to be Genghis Khan.
Conditional probability. What you can predict, given special information that applies in your case. Bayes' Theorem.
Bayes' Theorem applied to a real-life scenario: what's the probability that you'll get a job, given that you've been offered an interview.
The power of back-of-the-envelope calculations.
Ask the room to estimate the number of fire trucks in New York.
Now break the problem down (number of streets, fire stations per square mile, fire engines per fire station.) Get a new estimate.
Breaking the problem down and multiplying makes you much more accurate than just guessing. Doing little calculations helps you get things less wrong. And you don't need to be a genius -- you just need to try it.
What if you looked at average earnings for a career before you embarked on it?
What if you gathered data on what you're eating and how much weight you lose, when you go on a diet?
You can make much better decisions if you take a quick look at the evidence.
Rationality is the art of treating all of real life as though it were science.
Make predictions about what will happen if you take a certain action.
Gather data that will make your predictions more accurate.
Break estimates down into smaller components and do a little arithmetic, so that your guesses will be more accurate.
Use Bayes' Theorem to update your beliefs when you find new facts.
Then act, to maximize your chances of getting what you want.
Have you ever tried something that you knew probably wasn't going to work?
Have you ever made a major decision and not even started to reason what the likely outcome would be? Did you do any arithmetic before you made your decision?
Have you ever avoided learning information because you were afraid it would be disappointing or make you feel bad?
If you answered yes, you have something to work on.
Yeah, it takes a little work to gather information and make probability estimates. Yeah, learning the truth can sometimes be disappointing or unflattering.
But in the end, you'll be much happier because you'll succeed at things more often.
And you'll worry less -- instead of hoping against hope that a plan will work, you'll have confidence that it's likely to work. Confidence based on real evidence, not wishful thinking.
People have common cognitive biases that make clear thinking more difficult and predictions more error-prone. Learning about these cognitive biases, and correcting for them, will help you be more rational.
Learning more math, especially probability theory, can help you make sounder decisions.
Practice making predictions -- if you have a belief, bet on it! See if you're right. Play games of strategy and chance to sharpen your skills.
The goal is to "make beliefs pay rent," to practice the mindset of trying to believe what's true, what matches up to reality. Try to catch yourself if you start to engage in wishful thinking. If you think something is true, test it against reality! Do an experiment, or look up the statistics, to see if you're right. It's always better to find out what's actually happening than to go by feel or intuition. Intuition sucks.
It's a long road to rationality. Once you start, you'll be working on getting better at it for the rest of your life. As you become more rational, you'll find out that a lot of what people around you do is incredibly irrational, and you'll see ways you can make the world a better place by adding a dose of clear thinking. Or you'll see ways you can make your life better and happier than most people's.
Auntie Mame: "Life's a banquet and most poor suckers are starving to death!" Rational people don't have to be suckers.
I hope you give the talk, and that it goes well!
For whatever it's worth, I think your strongest example is the NY firetrucks (because it makes the audience feel a little bit silly...silly enough to want to remember to do it the right way, but not so silly as to be humiliated), and your weakest example is the Chinese garrison. Unless you've recently seen Mulan or whatever, it's tricky to empathize with a Chinese garrison, and it's certainly not funny. I would see if you can find another, more light-hearted example to get across the idea of wanting statistics to be on your side. I like that the example tries to force you to choose whether you would rather be a garrisoner or a raider, and that both options feel emotionally available, i.e, most Americans are not pre-committed to be loyal to one side or the other. I'm concerned, though, that there are just so many inferential steps there that the audience won't quite get the message.
Either way, go give this talk, and let us know how it goes!
Don't give a talk about "rationality" -- it's a big topic, and just invites blank states.
Here are good topics which would be "friendly" to people with large inferential distances:
"What is Bayes' Theorem and Why Should I Care?" (Following, roughly, along the lines of EY's "Intuitive Explanation" [and giving proper credit!])
"Human Cognitive Biases"
"What Is Occam's Razor, and Why The Lady Down the Street Is Probably Not a Witch"
All of these are sneaking in rationality, but without scaring people, and giving them a reason to care.
One thing I noticed: You did not say how long the talks are. I'm assuming 1hr-2hrs, I guess.
You did not say how long the talks are. I'm assuming 1hr-2hrs, I guess.
I don't know myself. I'll have to go along to one or two to see what the format is. They're advertised as "conversations", so it won't be just a matter of talking for most of the time and then taking questions.
How do you approach the matter
I gave a fairly successful talk at the Center for Inquiry's Student Leadership Conference this last summer on Bayes' Theorem and cognitive biases that was really a trojan horse for introducing fellow young atheists to the idea of rationality as an actual area of study.
one-to-a-small-roomful, with the general public
There were about sixty student group leaders of college and highschool atheist organizations in attendance.
how far can you expect to get?
Quantifying "fairly successful", I was approached afterward by at least six people who were interested in discussing my presentation and learning more. At least four more told me that they appreciated my presentation later that day or weekend.
There have been several posts and comments on basic rationality stuff for the general public. Is there a suitable page on the wiki, or one that can be started that people might actually edit?
I'll second that motion.
Eliezer is currently working on a book, but I'd love to see an open-source attempt at making an elevator pitch for rationality.
Nice idea! If you're targetting the general public, I think rationality per se might be a bit hard to swallow (at once); maybe explain things in a bit more indirect way, for example:
cognitive biases - there's plenty of material to choose from, and depending how entertaining you'd like to make you could use stuff like the 'invisible gorilla', some Dan Ariely examples, and so on.
logical fallacies - take some well-known examples; again, there's quite some material to choose from.
from there on, you could go through ways to think more effectively - e.g., how rationalists deal with pseudo-science. Maybe relate this to bayes' rule.
now, when we have established what rational thinking is, you could discuss some more advanced topics, such as FAI.
In general, the general public knows little about these matters, you'd probably need a lot of light-hearted/amusing examples to convey our message (and avoid rolling eyes)
Meetup groups; some members of the "general public" like going to public lectures, and it's not too expensive to rent a space (~$50 USD for an hour in New York.) Add a friend who owns a projector, and you're in business.
I recently discovered that in my home town (Norwich, England) there is currently running a series of "Café Conversations" in which some faculty members of the university that I work at are giving talks/hosting discussions on various topics. The meetings take place in a café that I know, which has room for about 20 people, and are open to the general public. Titles of some of the meetings already arranged are "What is infinity?", "Increasing happiness, decreasing consumption", "Bioplastics: wasteproduct or gold mine?", one on the nature of boredom (really about how old, retired people find things to do), and several on environmental topics. I have not been to any of them -- in fact, only the first of them has happened so far.
The obvious thing for me to do is to volunteer something on rationality, but quite apart from whether I would be able to do that at all (not being the friendly and outgoing, charismatic sort suitable for leading such a meeting), a problem that I foresee is this: these meetings are intended for the general public. This is nothing like a LessWrong meetup, or a Singularity conference, or delivering a lecture which, while nominally open to the general public is not actually intended for them.
Has anyone here had experience of communicating about rationality, one-to-one or one-to-a-small-roomful, with the general public? How do you approach the matter, and how far can you expect to get?
Norfolk, by the way, has given the world the expression "Normal for Norfolk". Go on, Google it.