« Caplan-Hanson Debate Tuesday | Main | Rational Paranoia »

April 11, 2009


While I think the underlying insight of this post is quite important ("X-lovers should be primarily buying X, not selling X", roughly), I'd be concerned that the division isn't quite as clear as one might like it to be... one could certainly imagine truth-buying becoming a form of conspicuous consumption wherein you sell yourself by buying. Of course, this effect may not be particularly strong and even if it was this behavior would still be good for generating truths, if not separating genuine truth-lovers from others.

The main question this brings to mind for me would be how exactly we'd extract the signal from a given "order for truth" at a certain price. Can a lover of truth properly value the discovery of some truths more than others; and if so, what rules should govern the appropriate levels at which truths should be priced?

Fine except that truth is (as you observe implicitly) not a commodity, like a cow, which a large group can put an approximate value on. Truth is both idiosyncratic and ephemeral, and it's value is at least bilateral -- there is no useful truth that cannot be both offered and accepted. Auctions are not the right venue for discerning truth.

Many truths are public goods and so markets under-produce them. You can't conclude from this, however, that few people love truth especially since many people do spend lots of money (by purchasing education or non-fiction books) to learn truths others have discovered.

Many who work on Wikipedia without any expectation of compensation or even notice love spreading truth.

Michael -

The commodity being bought in this case would not be some truth itself, but the discovery process that leads to the desired truth. As for it being "idiosyncratic and ephemeral", well... there are hashed-over epistemological arguments to be made on this point that I'll leave to others. Suffice to say, even if the discovery process in question does not lead to some Knowledge of the Absolute, we could at least imagine the utility of markets that give us this illusion. I imagine that your real worry is that people will disagree over whether a given proposed discovery process is actually valid, but there are presumably unbiased ways of evaluating this... that's pretty much the entire point of this blog.

And if nothing else, I doubt that that criticism extends to discoveries of mathematical propositions, which we already have organizations offering money for. (See: The Millennium Prize Problems, etc.)

Overall, I think your assessment is too harsh. While there are tricky issues here which I can't even label very well (epistemic? economic? moral?), I think the general principles of resolving a signal extraction problem apply. For example, ceteris paribus, a marginal increase in someone's willingness to pay for a truth indicates an increased desire to know that truth, ignoring the conspicuous consumption factors I mentioned above.

There might be some underlying problems which prevent some truth-buying equivalent of eBay being recognized, but it'd probably be fruitful to pinpoint the exact nature of the problem here rather than simply pointing out some potential issues and then throwing up one's hands in despair. I imagine that analogous criticisms were made of prediction markets in their infancy.

James -

That's a good point, but we could still imagine contingent contracts being made to overcome the public goods problem. Do you think that a lack of these is really the main thing that holds people back from bidding for discoveries?

Contrary to your conceptualization, it's often a nonbinary question. So if you're looking at buyers, you'll see many who will settle for information that's close enough and reasonably priced (when the extra cost for getting closer to the absolutely true truth doesn't seem worthwhile). Ask market researchers.

And what James D. Miller said.

What if you wanted to convince others that you were actually devoted primarily to truth about some topic, and to an unusual degree?

Why would anyone who wished solely to use truth be concerned with signaling that this was the case? Shouldn't the next step of the reasoning rather say, "What activity would convince me that someone was actually devoted primarily to truth about some topic"?

People who are genuinely infatuated with truth might not have any kind of capital (besides truth itself) to offer in exchange.

it is me, I love truth the most

I think you're selling the hobbyists a bit short here. Hobbyists do, in fact, typically spend a great deal of money buying the objects of their hobby, and their seeking of respect is typically limited to the respect of other people who also spend a great deal of time and money on the hobby.

The main criticism one can level at hobbyists is that they overvalue their hobby relative to potential competition or alternative means to similar ends; e.g., car hobbyists and enthusiasts are likely to be especially dismissive of the value of bicycles or public transit as a means of transportation.

What "truth hobbyists" might overvalue the truth relative to is left as an exercise for the reader.

Some people do try to buy truth with money. Scott Adams is a good example. Google with the 10^100 contest, the founders of givewell, Peter Thiel, and the sponsors of some essay contests seem like good examples. Those who pay for investment newsletters or for intelligence agencies are other examples. Maybe even the public in paying for public defenders. The caveat of "he who would pay the piper must know the tune" limits their ability to do this however and reduces the relevance of comparative advantage. This means that most truth-seekers should be expected to be trading time for knowledge of truth rather than money. Unfortunately, it is relatively easy to estimate how concerned a person is with money and how much of their money they are selling for truth. By contrast, people who spend a great deal of time on something are (rightly?) suspected of inadequately valuing their time rather than valuing, say, celebrity gossip a great deal. An indicator of truth-loving might thus be time spent on reading/study/web-browsing of purported non-fiction by those who a) change their opinions fairly regularly and b) obviously value their time highly. Many caveats apply, but this guideline does seem useful.

"The people who just want to know things because they need to make important decisions, in contrast, usually say little about their love of truth"

This could be in part because they don't, actually, love truth at all. (We do not usually describe someone who uses X purely as a means as someone who 'loves' X.) If their attitude is purely instrumental in the way you describe, then they would happily disregard the truth as soon as something better comes along (happiness-boosting self-deception, say). The instrumentalist who presently relies on truth-seeking might, at least for the moment, be a very reliable truth-seeker. But he may not be as robust a truth-seeker as the hobbyist who would pursue truth even if it were useless.

@Peter Twieg

I'm not sure its possible for truth discovery to take place without the active participation of who we are imagining might be a bidder.

I think Brian has the best answer.

Nicely put. I love how libertarians publish Reason, Ayn Rand called her philosophy Objectivism, and Eliezer is some sort of priest of Rationality.

Recently I came across this quote from a TED lecture, attributable to Sent Ts'an, a Zen practitioner from 8th Century BCE. Here it is, for what it's worth:

"If you want the truth to stand clear before you, never be for or against.
The struggle between 'for' and 'against' is the mind’s worst disease."

That strikes me as similar to a line of the main post as follows: "The people who just want to know things because they need to make important decisions, in contrast, usually say little about their love of truth; they are too busy trying to figure stuff out."

folks most vocal about loving "truth" are usually selling something

Oh this is silly—aren't you one of the more vocal people around?

I am impressed that you condemn preachers while standing on a soapbox yourself.

Carson Pitcher, Hanson explicitly mentioned academics.

Chip Smith responds to this post by listing his position on a number of conspiracy theories or odd beliefs. I like that he did that, though I find it odd that he is so much more skeptical of the holocaust than 9/11 (I'm with the conventional wisdom on both).

Truth is not always valuable.

On short term, value of truth depends on heavily what other people believe to be true. And because there are so few contrarians, transaction costs for contrarianness are high.

For example, I started investing heavily against the market 4 years ago. I ended up neither winning or losing much, because investing against the crowds and being "wrong" for 3 years was expensive.

Public belief in illusions seems to grow fastest just before the illusion ends. Thus, value of anything that is based on illusion goes up the most before it goes down. It is very difficult to time the end of illusions and profit from truth.

Peter, yes folks can signal raw wealth via buying anything, but usually they try to combine such signals with signals of other things.

James, yes folks should focus less on public good truths and more on private good truths. But if they pay little to buy any truths, I'll conclude as I described.

Eliezer, great questions.

Michael V., I think you are giving your associates and heros too much benefit of the doubt.

Who loves overcoming bias the most?

What if you wanted to convince others that you were actually devoted primarily to truth about some topic, and to an unusual degree?

Which others are you trying to convince? If they aren't devoted to truth themselves, they'll mistake credentials or proclamations of truth-seeking or whatever crude markers for truth-seeking.

If you're trying to convince fellow truth-seekers, then "gladly learn and gladly teach" is probably the handiest method. I'd say that one of the best markers is someone changing their mind about something important for a good reason. This has limits-- it's more likely to apply to people who've been around long enough to run into good reasons to change their ideas.

Anyone know enough history of science to know whether having a taste for insulting people is inconsistent with truth-seeking? I'd like to believe that insults do so much more to drive away thin-skinned people than to raise standards that anyone who's seriously interested in truth wouldn't use them, but I suspect that life isn't that simple.

Your standard of spending money on getting truth is interesting, but the vast majority of people don't have enough money to make a noticeable splash that way. Also, even someone who spends money on truth might be kidding themselves about how much truth they'll listen to.

Experimentation is probably a good marker, and can be done cheaply on a small scale.

It'll be hard to find solitary truth-seekers who aren't especially interested in conveying what they've learned.

"The people who just want to know things because they need to make important decisions, in contrast, usually say little about their love of truth; they are too busy trying to figure stuff out. These are the "truth lovers" I most respect in the sense of trusting their efforts to be directly targeted to actually uncovering truth. Sellers, hobbyists, and do-gooders are instead more likely to pretend to seek truth while actually seeking cash or respect."
I agree completely! (BTW, I am glad I found your site. I added it to my PDA news feeder).
Two days ago, I went to my wife's church for a "special" service. I do not belong to any church - nor do I plan to join any in the near future. My wife respects my stance on this particular issue. To put it simply, I consider myself a "free thinker" - especially when it comes to topics concerning religion and spirituality. (And that is not to imply - in any way - that those who are members of religious groups are not "free thinkers.").

At any rate, I agreed to join her for this once a year church occasion. After the service ended, she introduced me to several "elders" of her congregation. My wife''s religion is extremely conservative. (I guess "opposites" do attract.lol). She actually proselytizes "door to door" with others from her church. I respect her commitment to her faith.
One of the elder's (who was extremely polite) and I began talking about how the technology of digital cameras is improving at a fast rate. But then he said something like: "But you know, the Bible has always stayed great in its present form." I was thinking "ugh." This is his way of trying to began a dialogue on the Bible. After all, the setting of our conversation was inside a church. His intentions (however "pure") seemed to be focused on his truth - not THE truth.
Any feedback would be greatly appreciated.



Confounding factor: "Truth" comes in the form of facts (i.e. specific true bits of knowledge), but facts are valued more on their usefulness than their truthfulness.

Worth considering: people spend a lot on truth-seeking in opportunity cost. E.g. whatever your feelings about the noisy "public intellectual" types of academics, there are plenty of people honestly looking for true things in academia, who generally take reduced wages in return for being there (especially during the protracted study for undergraduate and graduate degrees---usually around 9 years).

"The people who just want to know things because they need to make important decisions, in contrast, usually say little about their love of truth; they are too busy trying to figure stuff out. These are the "truth lovers" I most respect in the sense of trusting their efforts to be directly targeted to actually uncovering truth."

I totally agree; these people need to find out the truth, because they have a financial incentive to do so. If they don't, they may loose their job and a lot of money.
People who have financial incentives to find out the truth about something are most likely to know the truth. For example, long-term successful investors are most likely to know where the economy is going. On the other hand, economists in academia and government have little incentives to know the truth, the probability that they'll loose their job or a lot of money if they get it wrong is very small. They can allow themselves to be blinded by their ideological and other biases.

Good financial investors are able to change their mind quickly, they don't suffer from path dependency. An example is George Soros:

"The French playboy trader Jean-Manuel Rozan discusses the following episode in his autobiography (disguised as a novel in order to avoid legal bills). (...) One weekend, Saulos exhibited in his discussion a large amount of bearishness, (...) A few days later, the market rallied violently, making record highs. The protagonist worried about Saulos, and asked him at their subsequent tennis encounter if he was hurt. "We made a killing," Saulos said. "I changed my mind. We covered and went very long."

Nassim Taleb: Fooled by Randomness, second edition. p. 239



I've often wondered why, if people on the transhumanist blogs and mailing lists are so concerned with truth, they seem to want to spend so much time on the internet trying to advertise this fact. Surely if they really were such 'Bayesian Masters' there would be no need - they'd be out making insight after insight after insight in scientific circles, and winning over and over and over in betting markets etc etc?

Put it this way: I don't need to go running to some board pleading for a pay increase and having to suffer being directed to an Internet 'food voucher' service. I simply win as much cash as I need off Betfair horse racing markets by calibrating the probabilities better.

As to AGI, if ever I get it it's probably the last any of you would hear of me.



I might be. But what makes you say that?

Nancy, the vast majority of people have enough money to buy many other things they want. Yes seeing their beliefs follow random walks would be a nice indicator, but few folks publish their belief changes often enough to show this.

MPL, reduced wages in academia could be for many other compensations; it isn't obviously for truth.

Andrej, yes investors have incentives to know certain truths, just as engineers have incentives to know others.

Michael M., in my previous comment I meant Michael V.; I edited the comment to be clearer.

What about people who publicly announce a change to their political philosophy?

Anybody who switches their philosophy is clearly a truth seeker (no matter what they switch to). And doing it publicly announces "many things I have said before are wrong". Those who put truth above pride, ego, etc... would love truth more than others. Revealed preferences, as they say.

You can also identify truth seekers by searching for those who constantly look for disconfirming evidence of their views.

Clearly betting and markets signals would work, but not for people without money.

On your view then, does this follow?

The Nobel Prizes should a) not pay money to the best X of year Y in field Z, but rather b) select that person to give out the money to the person who, for some time frame, discovers important truths in field Z, with X as the judge?

I'm surprised that nobody has mentioned the story of Simon Magus in the biblical account of Acts. The word "simony" comes from this story of this man's ignoble attempt to pay money for the truth. The story is intended to teach a profound truth, which is also repeated in other "truth-seeking" traditions like Buddhism. Many truth-seeking traditions believe that wisdom can be earned, but none believe that wisdom can be bought.

In fact, "truth" in the sense you are using it, is exactly the opposite of "information", since "information" can be purchased, and "truth" cannot. If you are talking about truth which can be acquired for a fee, you are talking about "information", and should use the proper term.

Additionally, truth proper is held by most traditions to be a non-rival, non-excludable economic good; so the suggestion of valuing in a market exchange seems somewhat absurd. It would be like saying that we need to make people start paying for air, to see who likes air the most.

Finally, if we were to take this proposal seriously, we would have to conclude that the elderly people who give all of their money to greedy televangelists, love truth the most.

Where do you draw the line? Does a man love a $300 hooker more than a $200 hooker, and more than his wife?

Besides, lovers and truth-seekers have already leapfrogged this proposal thousands of years ago. Romeo, Juliet, and young Werther would propose that willingness to die is a measure of one's love; as would the random suicide bomber argue that it was a sign of their commitment to truth. Such twisted but plausible examples should demonstrate the folly of placing an economic cost on love, truth, or love of truth.

Finally, anyone who seeks to stack-rank other human beings to figure out which of them "love the truth" more, is arranging angels on the head of a pin. The very premise is a bit odd.

I am surprised that the group of "folks visibly concerned that the poor don't have enough cars" is of any significant size and I would be more surprised if these people publicly declared that they love cars (rather than that they love people). I have never seen a self-declared car-lover care about availability of cars to general public; for me the car lovers and philanthropists are almost perfectly disjoint sets. Can anybody provide a link prooving that such people really exist?

By the way, I agree that in general car sellers don't actually love cars - they love the money earned by selling them and could replace cars by anything comparably profitable with ease. But the car hobbyists seem to love cars genuinely. Why should it matter whether cars are loved for their main purpose? Especially in analogy with truth, which I wouldn't say has a main purpose.

Accidentally I happen to be a bit of railway fan, so in terminology of this post I love trains, although I would never say it so. I like to travel by train, read texts about different locomotives and history of railways. To get from A to B I frequently choose train over bus even if the bus is slightly cheaper and faster, and even sometimes make a trip only to see an interesting narrow-gauge line or railway museum. Therefore I think I'm not interested in trains only as a means to get from A to B.
And I don't participate in any railfan group and don't speak about it often, so signalling is not an important issue. I think that, more often than not, car lovers are of this kind and the car sellers try to signal this sort of love for cars rather than "we like to efficiently get from A to B at lowest cost and maximum safety". With the latter, they wouldn't sell a single Ferrari.

"The people who are loudest are bad, bad people" is a mantra I've heard and read for several years now -- filtering through people who believe in earnest that they have cleverly picked up on some "truth" (ironically). Of course, you find an analogy that works for you -- but you want the reader to believe that it applies to all people/circumstances? In a blog called Overcoming bias no less?

What about investigative reporters? You think they're more quiet the more they believe in their truth?

What about people who are honestly convinced of an existential threat (Manhattan Project, e.g.)? Do you suppose they sit on their hands or have their say?

I think you have a point, but that you're way, way overselling it -- as if you have some truth to peddle.

Heh, I love Richard's counterpoint (echoed by others). People who talk a lot about a subject may be selling it - or they may be genuine, passionate hobbyists. Sure, the former are biased, but the latter are often better at a subject than those just using it.

The comments to this entry are closed.

Less Wrong (sister site)

May 2009

Sun Mon Tue Wed Thu Fri Sat
          1 2
3 4 5 6 7 8 9
10 11 12 13 14 15 16
17 18 19 20 21 22 23
24 25 26 27 28 29 30