From: Dan Fabulich (daniel.fabulich@yale.edu)
Date: Mon Jan 11 1999 - 15:36:45 MST
KPJ wrote:
>|>I suggest that the meme rules of conduct be commutative.
>|>
>|>Commutativity: If the rule is "X shall not perform act P on Y"
>|> then also the rule "Y shall not perform act P on X" shall hold.
>|
>|You have more or less enunciated Kantianism. Kantianism has various
>|problems, however. Take killing, for example. We normally regard killing
>|in self-defense as OK. Must I simply relent if someone tries to kill me?
>|Similarly, attacking another person is usually considered wrong. Must I
>|turn over my wallet to an armed thug so as to avoid a fight?
>
>If X does not wish to be killed, then X would try to avoid killing Y.
>
>If Y tries to kill X anyway, then X would lack the reason to avoid killing Y.
Eh? X does not lack the reason to avoid killing Y: the reason not to kill
Y is because the commutative rule is correct (by assumption). Y shouldn't
kill X, so X shouldn't kill Y, even if Y does what Y shouldn't do. The
reason X is to avoid killing Y simply IS the commutative rule, (ie the
categorical imperative) which doesn't go away when Y breaks the rules. If
it did, then when Y stole from X, it would be OK for X to steal from Y, etc.
>Would Kantianism define this the correct behaviour?
Kantian deontology is not in perfect agreement here. <guess>Kant himself
probably would have disagreed; that even if your own life or the lives of
others are in jeopardy, that this does not make it right to break the
rules.</guess>
>Perhaps the above rules would better be described as part of game theory.
Again, if you're going for game theory, the commutative rule will often
result in inefficient outcomes for both parties.
For example, the categorical imperative, it is often said, tells us that if
you are at home and your mother is upstairs, and a madman comes to your
door armed with a shotgun and asks "Is your mother at home? If she is, I
will murder her," you should answer "Yes," because she is, and lying is
always wrong.
Now suppose you've got two players, Alice and Bob, in rooms A and B
respectively. A madman comes up to Alice and asks "Is Bob in room B? If
so, I'll kill him." And another madman comes up to Bob and asks "Is Alice
in room A? If so, I'll kill her."
Both players must answer "Yes" under the commutative rule, or else your
commutative rule does not rule out lying to someone unrelated to the game,
Cindy. But if the rule demands that the players answer "No," ie if the
commutative rule eliminates lying, then it results in an obviously
inefficient outcome: in this game, both Alice and Bob die.
So either it's OK to lie, or we must accept an inefficient outcome. If you
think inefficient outcomes of this kind are sometimes bad, then you must
reject the commutative rule.
Anyway, IMO utilitarianism is the thing you're looking for if you're trying
to couch your argument in game theory. Kant argued that following the
categorical imperative was right even when it would result in decreased
utility; utilitarianism would never reach that conclusion, and would
naturally and easily conclude that both players should answer "no," because
that would maximize the utility of both players.
-Dan
-GIVE ME IMMORTALITY OR GIVE ME DEATH-
This archive was generated by hypermail 2.1.5 : Fri Nov 01 2002 - 15:02:47 MST