« Idea Futures In Lumpaland | Main | No Safe Defense, Not Even Science »

May 17, 2008


"Let's see, billions spent via ordinary means, and millions offered in bounties, and it is the bounties they blame for Al-Qaeda's notoriety and failing to catch leaders? The billions are spent and gone, while the millions in bounties we only lose when they actually work. How then is this data suggesting we should prefer ordinary means to bounties?"

As usual, you've hit the nail right on the head. (And, as usual, the WaPo is written by fools.) So, do you want to post a reply to the article? Or should someone else do so?

Isn't it fairly reasonable that, if someone won't turn a terrorist suspect in for $50 million, he probably won't do it for any amount of money? Principle of diminishing utility, and all that...

John, if you can get them to publish your reply, more power to you.

Allan, large organizations can do many things individuals can not. With enough at stake, they may get involved. If not, what have we lost?

That's an interesting idea. If there was a sufficiently large bounty, a venture capital group could invest in a startup whose entire business was capturing bin Laden, or somesuch? It's a fascinating concept.

As was mentioned over at Marginal Revolution, even a failed bounty likely gives the USA some utility gain. If a bounty of $1 billion doesn't catch Osama, it almost certainly makes his life more difficult than it was before, at no cost to us.

Robin, although we're for the most part going to be the choir, I think the sources quoted in the WashPost piece may raise reasonable concerns.
1. "The State Department has declined to boost the reward for bin Laden, arguing that more money was unlikely to do any good and would only add to his notoriety."
2. "Perhaps this Post comment explains the real objection: This "price in his head", millions in rewards business has had a stench to it all along. It's evidence of our own raw materialism and reinforces the idea it's our enemies who occupy the moral high ground."

I think these are empirical questions. If bounties increase our existential risk more than they reduce it, then it wouldn't be rationally persistence maximizing for us to pursue them, even if heckler's (irrationalist's) vetos are a role in negating the value of bounties. I'm curious, Robin, why you feel State Department arguments aren't worthy of being treated as empirical questions?

$500 million is the estimated global food budget shortfall this year. So clearly, catching bin Laden is more important than feeding people. How exactly did our priorities get so screwed up?

Grant, good point.

Hopefully, yes of course data can illuminate these topics.

HA, how is the "moral high ground" an empirical concept?

There are a couple things to separate out with the "moral highground" argument quoted in Robin's OP. But right now I don't have time to be rigorous about it, so in a nutshell it could be something like, "the ability for the United States to engage in policies that reduce existential risk is contingent on global public opinion that the United States should be permitted to engage in such policies. When global public opinion reaches a certain threshhold level that the United States doesn't have sufficient "moral highground" to engage in a particular policy, there can be a backlash which harms United States ability to reduce existential risk even more than if the United States had never engaged in that policy to begin with. If that's the case, and if it was predictable in advance, then the existential risk minimizing action would have been for the United States not to engage in that particular existential risk minimizing policy in the first place, because of the heckler's veto of global public opinion. In this instance, one of the empirical concepts would be global public opinion on whether the United States has the "moral highground" in engaging in a particular policy, something that could be measured most simply probably by opinion polling.

Raising the bounty substantially also reminds people that there is one.



So all we need is $500 million and we can feed the world? Does that account for diminishing marginal returns? Factors of production? Is that before or after the U.S. and the world gets rid of its horrible agricultural policies?

Can bounties really have an impact on global public opinion about the United States and its morality? Maybe, but I don't think that impact is going to be significant. Global public opinion is made up of many individual opinions and surely there will be someone saying: oh, those nasty materialistic Americans, look what they are doing, offering rewards. And maybe those are the same people who will criticize America no matter what policies are adopted. For most people, other policies (like, you know, invading a country or not invading a country) will be more relevant to their moral evaluation of the United States. People worldwide mostly assume Americans, just like everyone else except a few religious fanatics, are pretty materialistic.

Then there are the anti-Americans, and the pro-Americans, who have a fixed assessment of whether the United States occupies the moral highground or the moral lowground, and won't change their opinion even if the government were to hand them the $500 million.

HA, I guess this isn't empirical enough... it's just one opinion. But I am speaking from outside the United States (I don't know if my English made this explanation unnecessary) and I can assure you that, at least in my country, bounties offered by the US government are not really a big issue in the media. (Still not empirical enough, perhaps, it's just one country)

I remember reading a suggestion years ago (possibly from Robin?) for an optimal way to structure bounty payments. The idea was to have a published schedule of bounty increases over time, with the bounty ultimately becoming large enough that it was essentially guaranteed that someone would give in eventually. But only the first person to provide the key information would get the whole reward. So this produces a race to the bottom among potential turncoats, each one afraid that someone else will beat him to the reward that is sure to go to someone eventually. The result is that the optimal strategy is to turn traitor at a very low reward level. I don't know if this method has ever been tried to see if it works out as well as it does on paper.

There's another problem with government putting a price on anyone's head: it creates a somewhat legal marketplace in murder-for-hire. Can our leaders really be so naïve as to assume that none of the resulting assassins will then go to work for other employers?

Bribing also has the stench of raw materialism. What do you think about speculation that bribing ethnic/regional/tribal leaders may have been more responsible for recent Iraq successes than the troop surge?

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