Monday, March 31, 2008

How the World Works

"It is no easy task to be good. ... Anyone can get angry -- that is easy -- or give or spend money; but to do this to the right person, to the right extent, at the right time, with the right motive, and in the right way, that is not for everyone, nor is it easy; wherefore goodness is both rare and laudable and noble."
-- Aristotle, Nicomachean Ethics, Book II, Chapter 9.
Why are moral and political issues so difficult? Why isn't it easy to find answers to them that everyone can agree on?

Predicting the Future

One of the main sources of political disagreement -- and political incivility -- is disagreement about the way the world works. How will people behave if we pass a certain law? What will happen to the economy? Will incomes go up? Will unemployment go down? What will happen if we go to war? Will we win or lose? How many soldiers or civilians will die?

The empirical world -- the world we live in, observe and experience through sight and sound, etc. -- is hard to predict. Science has made great progress in studying the world and understanding the cause-effect relationships that govern it. But there is still an awful lot that we don't know about how the world -– and the people in it –- behave. (That is, there is a lot of vagueness in our knowledge of the empirical world.)

This creates a real problem when we face moral or political choices, because one of the things we take into account in order to make those choices is their consequences. Since we don't understand all the cause-effect relationships in the world, we don't necessarily know the consequences that different actions will have. Being unable to accurately predict the consequences of our actions makes it tough to figure out which action is the right one.

This is particularly apparent when we try to predict the behavior of people. It’s bad enough that we can’t always accurately predict earthquakes, chemical interactions, the weather, bee attacks, meteorites and milk in the refrigerator. We also have trouble predicting the behavior of parents, children, spouses, football coaches and little sisters. Even predicting the behavior of large groups of people –- elections, stock markets, sports teams –- is also difficult.

The way to fix this ignorance is with experiments, just like we do in physics, chemistry, and biology: Set up two identical sets of circumstances with only one difference between them so that you can isolate that factor from everything else. (In other words, you set up a control group.) Then you watch the two scenarios run and you find out what effect that one factor has.

But it's hard to run experiments (along with control groups) when it comes to people. It's not clear how you could ethically set up, say, military experiments to see which battle tactic is more effective. And economic experiments, where you test the effect of different tax rates or social spending, are very impractical. It's just not possible to set up two large, identical economies, and then give one of them a different tax rate to see what difference it makes.

Without experimental data, we only have historical data to look at. But such data is far from perfect. For instance, you can compare the histories of economies with higher tax rates to the histories of economies with lower tax rates, but these economies will differ in lots of ways other than tax rates, which makes it difficult to know how much differences in their performance is due to a difference in tax rates rather than a difference in something else. So it's hard to predict what effect a tax rate (or some other economic policy) will have if you implement it.

Figuring Out the Past

Along with the future, the past is also often a mystery. For instance, police investigators regularly face questions such as: Was a certain person murdered, or did they die of natural causes? If they were murdered, who was the murderer? And courts routinely face similar questions: Was this injured person harmed by someone else’s negligence or ill-will (which would make them eligible for compensation)? Or were they harmed by their own misbehavior (in which case they don’t deserve compensation)?

In order to make the right decision, we often need to know certain facts about the past. If those facts are unclear, then it's also unclear what the right thing to do is.

Humility and Certainty

Despite all this uncertainty, people in politics are incredibly confident in their knowledge of how the world works. The seem to know so clearly that their policies are right and will have good consequences that they feel justified in calling people who disagree with those policies names and deriding them. (After all, if it's so obvious that these policies are right, you'd have to be evil or stupid to disagree with them.)

And yet, these same people frequently avoid making specific, measurable predictions about the future. They'll say their policies will make things "better", but they're loathe to provide an actual number and tell us, say, how high the unemployment rate will be a year after their policies are enacted. Probably because they know there's a good chance they'd be wrong, right? (On the off chance a politician does give a numerical prediction, they often regret it: see Christina Romer's 2009 unemployment chart.)

After all, try predicting sporting events, elections, overseas political upheavals, wars, technological innovation, whether a relationship will last longer than a year, whether someone will like your outfit, or what the next fad among teenagers will be. If it were easy to predict these things, you could become a billionaire in no time.

But it's not easy. It's fantastically difficult. So, the next time you're compelled (or someone you know is compelled) to denounce someone for supporting "failed policies", think carefully about what evidence there is to back up what you're saying. Do you have experimental data? Is your evidence as iron-clad as, say, the evidence that citric acid cures scurvy? (This was discovered by James Lind in one of the first clinical trials -- i.e., medical experiments -- in the 18th century.)

If not, maybe it would pay to demonstrate some humility before resorting to insults (or at least try to run some experiments, first). It's bad enough that we live in a world that's hard to predict, there's no point making things worse by antagonizing people who disagree with us.


EXAMPLES

Below are examples of empirical uncertainties relating to some perennial moral issues:

Aiding the needy
  • Is providing aid directly –- for example, giving away food, blankets, shelter and so forth –- a more effective way to reduce need than giving some form of currency –- for example, cash or food stamps?
  • What sort of aid helps most in the short-term? What sort of aid helps most in the long-term?
  • What sort of aid will result in the recipient continuing to be dependent on others?
  • What sort of aid will result in the recipient being self-sufficient in the future, and no longer needing aid (i.e., “give a man a fish, you feed him for a day –- teach a man to fish, you feed him for a lifetime”)?
  • What sort of aid will result in the recipient being so prosperous that they will be in a position to help others who are needy?
  • What kind of aid is most likely to be abused or used fraudulently (for example, traded for drugs)? What kind of aid reaches the people who need it most, rather than being misused by people who aren't in need?

Character
  • Will this person or candidate fulfill their promises? Have they kept their promises in the past?
  • Have they lied in the past? Are they lying to us now? Will they lie to us in the future?

Climate Change and Global Warming
  • What is the current direction of climate change? Is Earth warming?
  • Did humans cause global warming, and do they have the power to reverse it?
  • What are the potential consequences of global warming? How much would sea levels rise? How much damage would be caused? What are the costs and benefits of adapting to global warming compared to the costs and benefits of reversing global warming?

Illegal immigration
  • What immigration policies will best prevent terrorists or criminals from entering the country?
  • What immigration policies will best fill the labor needs of the country?
  • What policies provide incentives or disincentives for people to break the law?

Punishment
  • What punishments will deter people from committing crimes?
  • What punishments will prevent people who have committed crimes from committing further crimes in the future?
  • What punishments are the most cost effective, giving the best results for the least money?
  • What punishments do the best job of educating wrongdoers -– for example, a misbehaving child, or a car thief, or a domestic abuser –- about what they have done wrong, and perhaps getting them to follow the rules in the future?

Taxes and Spending
  • How much money will be needed to pay for the spending programs the government has committed to?
  • What tax policies will bring in the revenue needed to pay for government spending?
  • What effects will the tax policy have on unemployment, individual incomes and savings, economic growth, consumer spending, income and wealth disparity, debt, investment?

Use of Coercion

  • Will employing economic and political sanctions against someone or some country –- for example, the current regimes in North Korea, China, Cuba, Iran; or past regimes in Nazi Germany, South Africa under apartheid, Iraq under Saddam Hussein –- decrease their bad behavior? Or will it create resentment and inspire violent defiance?
  • Will threatening the use of force against someone or some country decrease their bad behavior and instill order? Or will it create resentment and inspire violent defiance?
  • Will using force against someone or going to war with some country decrease their bad behavior and instill order? Or will it create resentment and inspire violent defiance?

No comments: