Economics is supposed to be a special science with ceteris paribus laws. Those laws posit probabilities or tendencies that aren’t observed necessities only because the system in question isn’t isolated from background systems which are outside the scope of the special science.

That seems self-contradictory to me. What is the role of “ceteris paribus” in that description if you are going to say that things outside the system affect it? For me, the cetera are in the system, not outside it.

Physics is no more “universal” than economics; it just describes less chaotic aspects of the universe. Is the uncertainty principle a universal law, or is it a denial of universality? If science can’t tell us whether a hypothetical cat is alive or dead, how “universal” is it?

There’s a big difference, however, between allowing for some exceptions to a norm, and _prescribing_ an outcome based on an unrealistic idealization or utopian vision.

People purporting to do economics can be wrong, just like people purporting to do science. Laissez-faire economists are benighted economists. Describing them as frauds or cultists, depending on whether they are evil or stupid, is fair game. But they are no different logically or philosophically from those who coined the term Jüdische Physik. Was “German Physics” not a fraud or a cult?

The same holds for game theory, by the way. When you say game theoretic models are counterfactual, in that they’re about what we would do under certain conditions, you mean to say game theory’s assumptions are only ceteris paribus, not idealizations or prescriptions.

Not exactly. (Please let me determine what I mean to say.) I don’t know what “idealizations” means, but game theory is often used prescriptively. Like when the Mudville manager sent Casey up to bat. That was the right play, because Casey was the team’s best available hitter. That’s game theory. It’s not a guarantor; it’s an optimizing technology.

I wouldn’t describe game theory as “scientific.” But if Dawkins finds that birds’ traits are distributed in ways that game theory would have predicted the distribution to evolve, then game theory is not a cult, either. It may well turn out that game theory is more useful in predicting natural history than human history, but that does not demote it to “fraud” or “cult.” And it certainly does not affect whether the Prisoners’ Dilemma “game” distills to its essence the dynamic tension between selfishness and cooperation in the universe and provide insights into where the pressure points are for affecting outcomes beneficially.

But in actuality, game theorists take onboard at least some of the neoclassical assumptions about hyperrationality.

What some people do with some tools is not the tools’ problem. I don’t doubt for a minute that game theory improves the approximations of careful economists. But I don’t see how game theory supports laissez-faire economics, because I include politics as a rational economic response to the excesses of capitalism. Is it not rational for those with limited bargain power to agree among themselves that they will appoint a committee to determine a collective bargaining strategy to which all will adhere because that is the rational thing to do to maximize the effect of the strategy? Self-regulation is irrational.

Now as humans with big brains we clearly have some stupendous rational capacities, but our rational decisions are exceptions, not norms. Normally, according to cognitive science, we operate nonrationally. We don’t calculate probabilities, but rely on intuitions, emotions, and on authorities who exploit our cognitive weaknesses.

We do not have to ratiocinate to behave rationally. Most “rational” economic behavior evolves from the harsh lessons of experience. The most rational thing most of us do is let the pilot fly the plane. But we don’t “think” about doing otherwise.

Grouch Marx was asked by a friend to write a blurb for a book he had not read. He wrote: “Outside of a dog, a book is man’s best friend. Inside of a dog, it’s too dark to read.” Based on the facts presented by Groucho, a rational person would not attempt to read inside of a dog, and lo and behold, no one does. Rationality prevents the attempt. But nowhere is anyone deciding not to try to read inside a dog. The question appears never to come up. And yet, if we do a retrospective study, we find that people behave “hyperrationally” with respect to the choice.

As we move along the behavioral spectrum from trying to read inside of a dog to choosing between paper and plastic at the check-out counter, only the complexity changes. At some point, “Look before you leap” meets “He who hesitates is lost,” and prediction becomes very difficult, especially, as Yogi says, when you’re talking about the future. But there is a current of rationality underpinning the whole enterprise, and I see nothing unscientific about exploring it if the object is to discern tendencies and probabilities. (Did you notice how Paul Volcker raised interest rates and a recession ensued? Is it cultish to claim cause and effect based on rational behavior in the face of higher rates?)

That’s the reality, and neoclassical economists and game theorists mean to prescribe an alternative to reality, based not on science, but on cultish convictions or on cons that are useful to big business.

This is exactly backwards. Figures never lie, but liars often figure. The misuse of economics is not economics’ fault. You have a beef with dishonest and stupid people. So do I.

Hobbes came out of a civil war that created misery in chaos, so he prescribed a powerful coordinating entity, misjudging how difficult a thing that is to create with human beings, as history shows. But Hobbes was right about how bad life is without a coordinating authority, which is the lesson game theory offers. Game theory does not prescribe a political solution; game theory identifies the benefits of a political solution, which provides the theoretical impetus for trying to find one. The US Constitution was a major step forward in that effort. The Federalists understood, without the benefit of Nash or Von Neumann, the fractal nature of Hobbes’s jungle. The internal state of the coordinating entity cannot itself be a war of all against all. The system needs “checks and balances,” coordination mechanisms for the coordination mechanism, custodes for ipsos custodes. That seems to me very rational.

Incidentally, to say that any behavioural pattern in retrospect can seem rational is to concede that the definition of rationality is pseudoscientific (unfalsifiable, abstract, and faith-based).

That sounds overbroad. I am aware of Texas Sharpshooter statistics — you claim to have been aiming at whatever you hit — but if Dawkins can look at birds and find game theoretically predicted distributions of traits, I don’t see why we can’t scientifically do the same thing with human behavior. Either that predictable recession happened, or it didn’t. There is nothing cultish, superstitious, or fraudulent about believing that a particular correlation, which would arise if people behaved rationally, cannot be explained by a collection of drivers best approximated by the fiction of intentional “rationality,” because the actual drivers were shaped by forces that conform to what rationality would prescribe.