The Virtue of Stubborn Conviction, lecture notes
The Virtue of Stubborn Conviction:
Ethics Through the Lens of Emergence
I
am interested in dissecting conscience using the language of emergence.
How do the beliefs of individuals affect the dynamics of their
communities at the macro level? If you could design a society and
stipulate the moral values held by each of its members, which values
would you choose? How should we translate the answers to these
questions into our actual beliefs and behavior?
I'm proposing:
The content of people's consciences should be whatever produces
functional societies (whatever makes it safe for us to trust each
other).
The Problem:
The Tragedy of the CommonsFour
farmers, each owning two cows, share a plot of grass that can sustain
nine grazing cattle. The week of the Village Cow Sale, Farmer Becky
gets another cow, thinking, "our plot can support one more cow, and I
could sure use the milk." Unbeknownst to Becky, her three colleagues
each bought another cow, thinking the very same thing. Subsequently,
the field is ravaged and all thirteen cows die of starvation.
Buying
a cow appeared to each farmer to be the rational thing to do. The same
reasoning endorses great deal of immoral behaviors:
- Lying
- Stealing
- Committing fraud
- Drinking milk out of the carton when no one else is home
There are a lot of things which common sense says we should do, but which these ways-of-reasoning repudiate:
- Voting
- Jury duty
- Good-Samaritan acts
- Putting in a new roll of toilet paper when the old one runs out
This method of reasoning (naive preference maximization) has the (emergent?) effect of generating dysfunctional communities.A naive economist might justify good behavior by reference to
conscience.
We should do the Right Thing (tm) because we know that the pangs of our
consciences will make the immoral choice unbearable, and so its
marginal rewards are outweighed.
Can we construe conscience as normative, rather than descriptive (and accidental)?
Let's look at the prisoner's dilemma (a two-player Tragedy of the Commons):
| Cooperate | Defect |
Cooperate | 5, 5 | 1, 7 |
Defect | 7, 1 | 3, 3 |
Traditionally,
economists have said that in a one-shot prisoner's dilemma, defection
is the rational strategy. This puts players in the fourth cell. It
should be clear that the first cell is more utilitous, both for the
individual and the group. But, the economist will protest, the first
cell is unstable: each player is tempted to defect.
- He knows that he can win more if he defects while his opponent cooperates.
- He
also knows that his opponent has a similar incentive to defect, and
wants to protect himself from landing on the short end of a win-lose
game.
My goal is to find a way to stabilize the first cell.Speaking
more generally, I want to find the set of beliefs (S) which, when
committed to by all (or most) individuals in a society, produces a
functioning social system (where people serve on juries and don't
commit fraud).
My Proposal
Suppose we recast the situation as follows:
To act in some way, X, is to affirm the statement, "For any person in a situation identical to mine, it is rational to do X."
More,
even--- to do X is to affirm the statement: "For any person in a
situation sufficiently similar to mine (one that differs only in
irrelevant details), it is rational to do X."
If players in the
prisoner's dilemma thought this way (and could trust their opponents to
do the same), they would ignore the second and third cells:
| Cooperate | Defect |
Cooperate | 5, 5 | 1, 7 |
Defect | 7, 1 | 3, 3 |
With
the question reframed this way, players could settle in the first cell,
and (knowing that their opponents in the PD are not tempted to defect)
they would need feel no temptation to defect either.
I'm curious
about finding ways-of-thinking that produce this kind of good behavior.
Let us use "S" to refer to any belief set that produces socially
healthy behavior.
Some candidates for S:- I should act only according to rules which I would be willing to generalize, and
- I should always treat people (both myself and others) never simply as behavers, but always as decision-makers, and
- I should act as though I were, through the rules that motivate my actions, legislating what it is to be rational.
- Do unto others as you would have them do unto you.
- It is wrong to pretend that I am the exception to the rule.
- No matter what, don't be hypocritical. (Mark's suggestion)
How do you get people to believe in things like that?
> Convert them to some mainstream religion ...?
> Make them read Kant's Groundwork of the Metaphysics of Morals ...?
> Embed them in a society that brainwashes them with anti-hypocrisy rhetoric ...?
> Explain this whole story to them ...?
Advantages of my approach:
- It's intellectually therapeutic.
Our
consciences are no longer irrational or arational, but (in Douglas
Hofstadter's terminology) superrational. According to this story, is
not only acceptable, but actually mandatory to behave cooperatively.
- It gives us a reason to care about other people's beliefs.
Since
we're both locally and globally better-off living in a society with
lots of people who believe in S, we should try to convince other people
to believe S.
- It gives us a fairly concrete framework through which to make decisions:
Should I buy antibacterial soap?
Does
the marginal benefit of (probably) reducing my disease-spreading
potential outweigh the marginal harm of (possibly) producing resistant
strains of bacteria? Is my situation importantly different from most
people's in this regard (for instance, am I a butcher? A doctor? Do I
have an immunological problem?)
- It carves out a research program for Ethics.
- It motivates interesting questions.
For
instance, it opens up a lot of interesting research programs for
psychology and meme theory. What kinds of belief systems can we fit S
into to make it catchier? To give it more holding power?
- The story gives a mechanism for its own modification.
- The contents of "S" are not set in stone. They can be modified with their emergent properties in mind.
- Neither, for that matter, are contents of the contents of "S".
Consider the Categorical Imperative
My
story gives us a way of finding out whether something like "never lie"
should count as a maxim. (Basically, we would ask whether a given rule
or level of analysis fruitfully discriminates between socially helpful
and socially hurtful behavior, and look for the one that does so best.)
Bugs & Features:
Some objections that could be lodged against my solution:
Variations on "it won't work":- Individuals
will only adopt these beliefs if they know that everyone around them
believes the same way. But no such guarantee can ever exist in the real
world.
- A system like the one I describe relies too much on
individuals' unenforceable commitments to their beliefs. If anyone
defects, she can take advantage of the system.
I'm willing to let those objections stand. Although I don't think they're fatal, they do seem to weaken my analysis somewhat.
Objections that I find more interesting:- I insist that people formulate their beliefs irrationally.
This
is true, in a way. I don't permit agents to doubt the contents of S.
One could say (as my roommate does) that this makes me monstrously
elitist: I seem to think that people won't choose for themselves to
believe in S. I'm happy to bite this bullet, largely because I think my
character is restored in the next point.
- I don't care about the motivation for people's beliefs.
I
consider this a feature, not a bug. You don't have to be an expert in
philosophy to be able to behave morally, according to my story. You can
believe in S because your pastor (or your mother, or your therapist)
told you to. You're saved, so to speak, as long as you believe in S. So
my story is egalitarian!
- My story is antirealistic about ethics.
I
don't want people to believe in S because I think it's the truth about
ethics; rather, I want them to believe in S because, whether or not
they know it, society is better off if they believe in S. This whole
story feeds into (and is fed by) a pragmatist view of reality
(at least w.r.t. ethics): it suggests that we should choose our beliefs
based on their usefulness, not their truth.BUT, if you hate pragmatism and you still want to get on board...
My
program can be fit into a realist framework, if you think that think
that the truth of the matter about ethics is that S is what people
should believe, where S is that set of beliefs that helps society the
most.
Links & References:
NY Times Article: The Evolution of Cooperative BehaviorThe Stanford Encyclopedia of Philosophy's article on the Prisoner's DilemmaThe SEoP's article on Evolutionary Game TheoryHardin, Garrett. "
The Tragedy of the Commons,"
Science, Vol. 162, December 1968, pp. 1243-1248.
Hofstadter, Douglas. "Dilemmas for Superrational Thinkers, Leading Up to a Luring Lottery," in
Metamagical Themas, New York: Basic Books, 1985.