Game Theory And Moral Philosophy

The previous post had me confused about the bright line separating psychopathy from a good logician. The neuropsychology of most games shows an evolutionary intuition towards the Nash equilibrium, against unfairness, and towards long term maximization through qualified cooperation. We treat all our interactions, at some level, as games to be optimized, seeking dominant strategies as well as equitable outcomes that increases the probability of our opponents also choosing the outcome that would be best for us. None of this is really manipulation. So when does it cross over?

1. Level 1: Manipulate the conditions of the game such that the opponent chooses an outcome that is best for us, that they might not have chosen otherwise. We buy flowers, play soothing music in the background, or take our date to the amusement park after having read a Cosmo article saying something about cotton candy sticks reminding us of our fathers.
2. Level 2: Manipulate the payoffs and costs such that the opponent considers them higher than they would have otherwise. The entire sales and marketing industry
3. Level 3: Play your own game. You’d never call a chess player manipulative because he’s trying to predict everything his opponent will do and nudge him towards a favorable direction in a zero-sum game. The difference is that the opponent knows the rules and what he’s signed up for. There is total transparency. This is not at all clear in real life though. If you’re playing a game with someone who doesn’t know they’re in a game or have no idea what the rules are, I don’t love your chances of being normal.
4. Level 4: Play to destroy. There is a sort of unsaid ethic in competitive games that are unequal. If you’re Real Madrid playing Barcelona, all bets are off, if you’re up 5–0 at halftime, you take a deep breath and gear up for another 5. But if you’re playing Osasuna in 20th place struggling to stave off relegation, you ease off at 3–0 and don’t crush their spirit. And if it’s a 6th division side, from the blind football league, playing with just 1 goalkeeper after the other 10 players died in a plane crash, in the plane you had sent to pick them up for the game, then you’d hope to let that goalkeeper save all your shots then dribble that ball through your team and score for a miraculous victory, all this is of course assuming Ronaldo is benched and not salivating at the prospect of wrapping up the Pichichi trophy in a single game.

This isn’t a backward induction principle that is different, the payoff matrices themselves have been rewritten to reflect the impact of your moral philosophy on the utility from different actions. In a Barcelona game, the payoff matrix could be something like (5,-7) for a W, (2,-1) for a D, and (-3, 5) for an L, given the specific conditions in the league, whose stadium it is etc. The utility of the margin of victory might not be linear, but certainly increases by margin, perhaps 5 for a 1–0, 7 for a 3–0, and 10 for a 5–0.

For the Osasuna game, you’d expect not only the utility of a W to be lower (with a larger downside for a D/L), you’d also expect the utility from margin of victory to be much shallower, if not constant. More, you’d hope that the utility starts dipping after a certain margin of victory where guilt and shame start to take away from the payoff as people start calling you a cruel bully. This would be most pronounced against Braille FC, where anything less than deriving pleasure from the pleasure of your opposition is cruel.

This leads me to think that the sign of psychopathy isn’t the capability and willingness to use induction and logic to secure the best outcomes, but instead a totally bizarre payoff matrix logic. A psychopath derives pleasure from your pain, so there are no independent variables and there are no possible cooperative strategies. The larger your loss, the larger his gain. This looks like a zero-sum game, and people do it all the time. Why are they not psychopaths? Maybe the nuance is that psychopaths have no underlying resource to gain from the zero-sum game apart from the opponent’s loss. They therefore convert optimally cooperative games into zero-sum games that are totally counter-intuitive. Evolution works both by competition as well as cooperation and symbiosis, a psychopath would choose to engage in a zero-sum game when cooperation is the clearly dominant strategy.

But that sounds bad for survival, why do we still have psychopaths? Bacteria that don’t find a certain equilibrium with their host and instead become overly virulent, will kill their host before they have a chance to be spread to another host. They eventually die out from the population, leaving behind the more cooperative strains. Under certain circumstances though, like when cities grow to unmanageable levels of human density, poor sanitation, and ample livestock, virulent psychopathic bacteria are able to spread faster hence running riot in their current host confident of being able to find a new one before they die. The bacteria haven’t become more virulent, it’s just that the more virulent ones can survive. We’ve conquered nature and broken free the gates that have held our gene pool within a tight band of optimal combinations, it’s allowed us all the diversity in traits that used to be trade-offs, both good and bad. I suppose this phenomenon is itself a trade-off we’d accept at a certain equilibrium. If evolutionary disadvantages like psychopathy are a result of our sudden decadence or superiority as a species, so are autistic savants who gave us the most illuminating theories in mathematics or the most stunning works of music.

The scary thing of course is to consider the contrarian view, that psychopathy is actually a superior strategy, and they would make for more successful organisms in the species.