Wavepacket Blog

Wavepacket Blog
displaying only one specific post

Wed May 6 22:41:32 2009
Is Torture Worth It?
The Economics of Collateral Damage

Good investments?
Image courtesy of Crux (wiki)

Lately, the torture memos have been making news (and see more information about the memos at the ACLU's website).

The left claims that torture is immoral, the right claims that Bush kept us safe. I have no intention or desire to write a political blog, so I'll stay out of that.

But a state's decision to use torture has economic (game theoretical) implications, and it's worth considering those. And it's more than just torture: how far do you go to respect human rights? For instance, we have killed a lot of civilians in the Iraq war alone, and just today we apologized to Afghanistan for killing so many of their citizens in airstrikes.

Here is one thought experiment: suppose a large terrorist faction hides people and material in a large United States city. The government discovers that the faction exists in the city, but doesn't know exactly where. Is it okay if the government bombs city blocks it suspects may contain terrorists, but isn't sure, even if there are a lot of innocent US civilians also there?

I don't think the government would do such a thing. I'm guessing risking US civilian lives would not be an option at all.

However, a different standard clearly applies for non-US citizens. We bombed large sections of Baghdad, killing hundreds of civilians, based only on rumors of Saddam Hussein's whereabouts. (In the end, all those civilian deaths were in vain, since Saddam was captured without a shot in Tikrit). And we continue to shell Afghan villages with less safety checks than we would use for US towns.

Game-Theoretical reasons the US should not respect human rights:

It may save lives. There are arguments against this (see below), but it is possible that risking the deaths of non-US civilians could save the lives of US civilians and military personnel, just because discounting civilian lives opens up more options against the enemy and allows for faster decision-making.
It's cheaper. Respecting human rights is very expensive. Can you imagine if we treated every Iraq airstrike request as if it was occurring within the US? There would be endless confirmations, approvals up and down the chain of command, verifications and re-verifications of targets, etc.
US laws don't apply. You can't be sued for wrongful death, since the attacks aren't happening on US soil. Those civilians aren't protected by the US Constitution anyway. And that saves a lot of court cases.
Deters would-be terrorists. There are arguments against this (see below), but if the US says it will torture or generally maltreat all enemy combatants, that may scare away people who would otherwise be tempted to take up arms against us.

Most of the above reasons are only valid in the short run. If an "abandon all human rights" policy prolongs or intensifies the conflict, then any short-term savings in lives or money are likely to be swamped by the costs (in money and lives) of a longer or more intense conflict.

And #4 (would-be terrorists will think twice) assumes that terrorists are rational, and they almost always aren't.

I don't think a strategy will get very far if it assumes terrorists are rational. A better (and probably safer) strategy is to assume that terrorists aren't rational, except in a twisted way that they want to inflict as much damage as possible. For those sorts of worst-case terrorists, #4 (promising them punishment) won't help.

Instead, the approaches should focus on the people who are rational. These are typically the civilian populations that house, feed, and supply terrorists, and the politicians in those areas. The civilians involved are either coerced or sincerely believe they are doing the right thing by helping the terrorists. Politicians as a whole are shrewd calculators and strategies that open up political advantages to cracking down on or disavowing terrorism will likely yield results.

Game-Theoretical reasons the US should respect human rights:

It protects our own military and civilian personnel. By upholding the Geneva Conventions for all of our enemies, terrorist or not, we have a better chance of having our own personnel and civilians so treated. Given the mindset of terrorists, it's not clear how much this would sway their treatment of our civilians, but it would likely sway the greater foreign populations which house and (indirectly) support those terrorists. And that leads to the next point:
It lays claim to high moral ground. Again, terrorists may be beyond reasoning, but most of the world's population is not terrorists, and will never be. If the majority of a country's population is on your side, whoever you are, it is far less likely to support or even harbor terrorists against you.

For these sorts of asymmetric conflicts, claiming the moral ground may actually be a key to winning. Our short-term costs, in terms of money and time, may be higher (and we may have to spend more money to protect lives!) but over the long run, this is the approach that will win allies and even the terrorists' host populations. Any strategy that antagonizes the host populations runs the risk of prolonging the conflict, and any strategy that antagonizes the host populations and our allies is almost certainly doomed to failure in the long run.

So based on what I hope is a relatively dispassionate consideration, respecting human rights is the best long-term way to win the conflict. And by "respecting human rights," I'd recommend treating every civilian on the planet as if they were a US citizen, at least when considering military options. It makes military operations much more complex and expensive to plan, but has the best chance of keeping the conflict short and winning many allies.

This argument does assume one thing: that there is a moral high ground. If all morals are relative, then none of this applies and we're back to a most-aggressive-combatant-wins strategy. And in that case, "terrorist" is also a relative term. So I'll stick with my assumption that there is a moral high ground.

Comments

Related:
  economics
  geopolitics

Unrelated:
  books
  energy
  environment
  lists
  mathematics
  predictions
  science

Links: