Why be moral? #
Two questions
- Descriptive: would you be moral if you could always get away with immorality?
- Normative: should you be moral if you could always get away with immorality?
- Normativity clarification: this is a rational should, not a moral should.
- Is morality rational?
Rational choice theory
- People have a stable, ordered set of preferences.
- Actions have utility insofar as they satisfy these preferences.
- Rational agents always choose the action that has the highest expected utility in light of their preferences.
Expected utility
- Discount the value of something by the probability it will occur.
- E.g., you can buy a lottery ticket for $10. Has a 1% chance of winning $100, and a 99% chance of winning $0. Whats the expected utility?
- ((0.01 \cdot 90) + (0.99 \cdot -10))
Is morality (altruism) rational?
- Why does this matter? Why try to show that it can be in our interest to be altruistic?
- Remember, preferences are just given.
- And it turns out that most people are naturally altruistic.
- So, according to rational choice expected utility theory, altruism is rational.
Not a satisfying answer
- Still want to know if its rational to have such preferences in the first place
- If you could take a pill that would eliminate altruistic ends, would you have reason to do so/not to do so?
- If you were a sociopath and you could take a pill that would imbue you with altruistic prefernces, would you have a reason to do so/not do so?
Prisoner’s Dilemma
Imagine you have 2 prisoners, A and B
B is silent | B testifies | |
---|---|---|
A is silent | Both get 6 months | A gets 10 years, B goes free |
A testifies | A goes free, B gets 10 years | Both get 5 years |
Another version
Player 2 cooperates | Player 2 defects | |
---|---|---|
Player 1 cooperates | 1: WIN, 2: WIN | 1: LOSE BIG, 2: WIN BIG |
Player 1 defects | 1: WIN, 2: LOSE BIG | 1: LOSE, 2: LOSE |
Cooperation is a positive-sum, but defection is a dominant strategy
- Dominant strategy: a strategy that produces the best results for a player regardless of what other players do
- Other player will either defect or cooperate:
- If she defects, I do better by defecting
- If she cooperates, I do better by defecting.
- So I should always defect.
Connection to morality
- Generally, morality is a positive-sum game.
- Life without morality is a negative sum game.
- But theres always a strong temptation to cheat.
- Should I?
Iterated games
- What if you play PD multiple times with different players who play a wide variety of ways?
- What strategy works best then?
Robert Axelrod’s computer simulations
- Best strategy is tit for tat.
- Cooperate until the other player defects, then do whatever the other player did last.
Variations: reputation
- Play one PD round with the person next to you
- Record total score
- Hold up whatever card you played (so everyone can see)
- Find a new player to play with – must agree to play next round.
- Play another round and record the total score
- Repeat 3-5
- One person is randomly selected and paid a tally of their score
An argument that it’s rational to be mroal
- Life is a series of PD interactions, where we have informatino about each other
- So, “maximize my utility” doesn’t maximize your utility.
- It’s better to adopt a credible, public strategy to restrain yourself, provided others do so as well.
- Therefore, it’s rational to internalize a moral code that requires restraint and disposes one to cooperate.