Before you read, go watch: http://www.youtube.com/watch?v=zORv8wwiadQ.
Alright, for those of you who didn’t actually watch, here’s the synopsis of the man’s argument: because the worst-case cost of not preparing for global warming is worse than the worst-case cost of preparing for global warming, we should prepare for global warming.
Now, some of you might recognize the type of argument as close to Pascal’s Wager, and those of you who have had any amount of game theory also see the resemblance there, especially to the commonly known “Prisoner’s Dilemma“.
For those who aren’t familiar with it, the prisoner’s dilemma goes the same way – the Nash Equilibrium (the guy from A Beautiful Mind, if you remember that), or the stable state of the game, is such that both prisoner’s tattle, thus minimizing their worst-case scenarios.
On the surface, the video seems entirely logical – in any given scenario, there are two “extreme” strategies: either you maximize your potential wins, or you minimize your potential losses, without respect to the other one. Generally, our day-to-day decisions are variations on these two, often denoted by phrases such as “betting the farm” or, in the opposite, “the safe bet”.
A Nash Equilibrium, at the highest level, is simply a combination of choices such that each player is making the best choice regardless of what the other player(s) choose. In this world (much as in real life), you must choose your strategy beforehand, and you are not allowed to change your strategy.
Pascal’s Wager was a similar argument to the global warming argument above, attempting to show that it is only logical to believe in God (of Abraham, of course :)). The short of that story is that the worst case of believing in God (looking foolish?) is much more pleasant than the worst case of not believing in God (eternal damnation, hellfire, general unpleasantness in the afterlife). The choice here, then, seems obvious: would you rather be uncomfortable, or be eternally damned?
My earlier parenthetical about this being a question of believing in the Christian god shows a bit of the weakness of this argument – what if the true god is the Amitabha Buddha? Shouldn’t we include him in the game? What about the gigabazillion other gods and goddesses we have ever believed in? This, perhaps, we can consider analogous to the global warming argument by pointing out the fallacy of the “yes”/”no” dichotomy (to “fixing” global warming).
Barring that (sound) argument, we should go back to Pascal’s Wager and ask, “Why doesn’t every logical person believe in God, then?”
In the prisoner’s dilemma and other similar two-player games, in order to find points of equilibria, you always need to consider possibility. Often times, you consider the possibility to be even for all choices of your opponent, and this simplifies the game greatly – if it is equally possible, for each of your choices, to get the worst possible case, then you want to try and minimize that worse possible case.
But what about this: let’s say that there is a meteor falling towards your house right now. Now, you can buy my fancy house protector for only $100 and put it on your roof (which will, of course, stop the meteor), or you can let your house be destroyed, possibly with you and your family in it.
Now, let’s look at the choices. If the meteor is falling, then not buying the protector means you are gonna die (ick.). If you do buy the protector, you are out $100 and you get to live (yay!). Now, if the meteor is not falling, then not buying the protector means you are gonna live (yay!), but buying the protector means you are gonna live but are out $100 (still pretty yay!).
So, the worst case of buying the protector is that you are out $100, and the worst case of not buying the protector is, well, dieing. The choice is obvious – you don’t want to die, do you?
When you put it like that, you begin to see the obvious fallacy here. The likelihood of a meteor falling on your house and the protector being able to stop it is slim to none. Sure, the worst cases are clear, but that isn’t the end of the story.
The same holds for Pascal’s wager – sure, the worst cases are obvious, but I would venture that the people who don’t believe in God do so because they think his existence unlikely or impossible, thus the little bit of annoyance isn’t worth the possibility of protecting against eternal damnation.
And, finally, to global warming. The guy’s argument has the following assumptions, which he either ignores or explicitly tells us to ignore, but change the entire outcome of the decision:
- Global warming is equally possible or impossible (he mentions this, but says that it doesn’t matter) – the probability of each is vital
- That trying to fix global warming will work (i.e. the probability of an outcome happening based upon our choice)
- More accurate outcomes – perhaps the slightly less bad but more probably “worst-case scenarios” are different – maybe the worst case of global warming is mass extinction, or maybe the worst case of no global warming is global societal collapse. Of course, the worst cases of each might be that we end up but better off (nothing says the “worst” case can’t be good)
- Along with the previous, that global warming or no global warming are bad or good things.
So, even though the guy tells you otherwise, the probabilities of both global warming and the efficacy of our work are incredibly important, and the characterizations of the outcomes right along with them.
Of course, if you don’t believe me, send me a check for $100, and I’ll ship you that house protector. Seriously. It’s gonna get you. Only $9.95 shipping and handling!