706
SERGIU HART
by the moral rules. But it’s also part of the equilibrium, because the whole thing
would fall apart if he weren’t.
There is, incidentally, a phenomenon like this also among the “Mitnagdim,”
like me. There is a person in Israel called Rabbi Firer, who is absolutely the
top source of medical information in the whole country, possibly in the whole
world. And he is not a physician. Anybody who has an unusual or serious medical
problem can go to him, or phone him. You make a phone appointment for, say,
1:17
A
.
M
., you describe your problem, and he tells you where to go for treatment.
Often the whole thing takes no more than a minute. Sometimes, in complicated
cases, it takes more; he will not only direct you to a treatment center in Arizona,
he’ll arrange transportation when necessary, make the introductions, etc., etc. The
whole point is that he is not a physician, so he has no special interests, no axe
to grind. How it works is that he, like the Khassidic rabbi, gets information from
everybody, patients and doctors alike, and he is also unusually brilliant. And he is
deeply religious, which, again, is what keeps him honest. I have made use of him
more often than I would have liked.
Up to now we have been discussing the normative side of game theory—advising
individuals how to act—but there are also other sides. One is “public normative.”
The religion will not tell you how to conduct elections, or when to cut the discount
rate, or how to form a government. It will not tell you how to build a distributed
computer, or how to run a spectrum auction, or how to assign interns to hospitals.
Still another side is the “descriptive.” Religion will not explain how evolution
formed various species, or why competition works.
But I must immediately correct myself: the Talmud does in fact discuss both
evolution and competition. Evolution is discussed in the tractate Shabbat on page
31a. The sage Hillel was asked why the eyes of certain African tribesmen are
smaller than usual, and why the feet of other African tribesmen are broader than
usual. Hillel’s answers were adaptive: the eyes are smaller because these tribesmen
live in a windy, sandy region, and the smallness of the eyes enables them better
to keep the sand out; and the feet are broader because that tribe lives in a swampy
region, and the broad feet enable easier navigation of the swamps.
Competition is also discussed in the Talmud. In the tractate Baba Bathra 89a,
the Talmud says that the authorities must appoint inspectors to check the accuracy
of the weights and measures used by marketplace vendors, but not to oversee
prices. The twelfth-century commentator Samuel ben Meier (Rashbam) explains
the reason: if a vendor overcharges, another vendor who needs the money will
undercut him, all the customers will go to him, and the original vendor will have
to match the lower price. The invisible hand—six hundred years before Adam
Smith!
Other game-theoretic and economic principles are also discussed in the Talmud.
The nucleolus makes an implicit appearance in the tractate Kethuboth 93a [46];
risk aversion shows up in Makkoth 3a [80]; moral hazard, in Kethuboth 15a, and
the list can be made much longer.
But of course, all these discussions are only the barest of hints. We still need the
game theory to understand these matters. The Talmud speaks about adaptation,
INTERVIEW WITH ROBERT AUMANN
707
but one can hardly say that it anticipated the theory of evolution. The Talmud
discusses competition, but we can hardly say that it anticipated the formulation of
the equivalence theorem, to say nothing of its proof.
Besides, one needs game theory to explain the ethical and moral rules them-
selves. Why not steal software? Why have accurate weights and measures? Why
love one’s neighbor as oneself? How did it come about, what function does it
serve, what keeps it together? All these are game-theoretic questions.
Finally, let’s not forget that the world is very far from being—to use your
phrase—populated by religious people only.
In short, the Bible and the Talmud are fascinating documents, and they cover
a lot of ground, but there still is a lot of room for game theory—and for all of
science.
H: So, to summarize this point: game theory definitely has a place in a religious
world. In the “micro,” the rules of conduct are principles that cover only certain
issues, and there is “freedom of decision.” In the “macro,” the structures that arise,
and the rules of conduct themselves, are subject to game-theoretic analysis: how
and why did they come about?
Is your view a common view of religious people?
A: Maybe not. One doesn’t discuss this very much in religious circles. When I
was young, there were many attempts by religious people to “reconcile” science
and religion. For example, each of the six days of creation can be viewed as
representing a different geological era. There was—and perhaps still is—a view
that science contradicts religion, that one has to reconcile them. It is apologetic,
and I don’t buy it.
H: Take, for example, the six days of creation; whether or not this is how it
happened is practically irrelevant to one’s decisions and way of conduct. It’s on a
different level.
A: It is a different view of the world, a different way of looking at the world.
That’s why I prefaced my answer to your question with the story about the round-
ness of the world being one way of viewing the world. An evolutionary geological
perspective is one way of viewing the world. A different way is with the six days
of creation. Truth is in our minds. If we are sufficiently broad-minded, then we
can simultaneously entertain different ideas of truth, different models, different
views of the world.
H: I think a scientist will have no problem with that. Would a religious person
have problems with what you just said?
A: Different religious people have different viewpoints. Some of them might
have problems with it. By the way, I’m not so sure that no scientist would have a
problem with it. Some scientists are very doctrinaire.
H: I was just reminded of Newcomb’s paradox, with its “omniscient being.” We
both share the view that it doesn’t make much sense. On the other hand, perhaps
it does make sense in a religious world.
A: No, no. It’s a little similar to this question of the omnipotence of G-d. If G-d
is omnipotent, can he create an immovable object? Atheists will come up with a
question like that, saying, here, I’ve disproved the whole idea of religion.