counting them off, or comparing them with a balance scale. Only Market Pricing (and the Rational-Legal mindset of which it is part) allows one to reason in terms of proportionality. The Rational-Legal model requires the nonintuitive tools of symbolic mathematics, such as fractions, percentages, and exponentiation. And as I have mentioned, it is far from universal, and depends on the cognition-enhancing skills of literacy and numeracy.
It’s no coincidence that the word proportionality has a moral as well as a mathematical sense. Only preachers and pop singers profess that violence will someday vanish off the face of the earth. A measured degree of violence, even if only held in reserve, will always be necessary in the form of police forces and armies to deter predation or to incapacitate those who cannot be deterred. Yet there is a vast difference between the minimal violence necessary to prevent greater violence and the bolts of fury that an uncalibrated mind is likely to deliver in acts of rough justice. A coarse sense of tit-for-tat payback, especially with the thumb of self-serving biases on the scale, produces many kinds of excess violence, including cruel and unusual punishments, savage beatings of naughty children, destructive retaliatory strikes in war, lethal reprisals for trivial insults, and brutal repression of rebellions by crappy governments in the developing world. By the same token, many moral advances have consisted not of eschewing force across the board but of applying it in carefully measured doses. Some examples include the reform of criminal punishment following Beccaria’s utilitarian arguments, the measured punishments of children by enlightened parents, civil disobedience and passive resistance that stop just short of violence, the calibrated responses to provocations by modern democracies (military exercises, warning shots, surgical strikes on military installations), and the partial amnesties in postconflict conciliation. These reductions in violence required a sense of proportionality, a habit of mind that does not come naturally and must be cultivated by reason.
Reason can also be a force against violence when it abstracts violence itself as a mental category and construes it as a problem to be solved rather than a contest to be won. The Greeks of Homer conceived of their devastating wars as the handiwork of sadistic puppeteers on high.220 That, to be sure, required a feat of abstraction: they lifted themselves out of a vantage point from which war is the fault of one’s eternally treacherous enemies. Yet blaming the gods for war does not open up many practical opportunities for mere mortals to reduce it. Moralistic denunciations of war also single it out as an entity, but they provide few guidelines on what to do when an invading army is at one’s doorstep. A real change came in the writings of Grotius, Hobbes, Kant, and other modern thinkers: war was intellectualized as a game-theoretic problem, to be solved by proactive institutional arrangements. Centuries later some of these arrangements, such as Kant’s triad of democratization, trade, and an international community, helped to drive down the rate of war in the Long Peace and the New Peace. And the Cuban Missile Crisis was defused when Kennedy and Khrushchev consciously reframed it as a trap for the two of them to escape without either side losing face.
None of these rationales for rationality speaks to Hume’s point that rationality is merely a means to an end, and that the end depends on the thinker’s passions. Reason can lay out a road map to peace and harmony, if the reasoner wants peace and harmony. But it can also lay out a road map to war and strife, if the reasoner delights in war and strife. Do we have any reason to expect that rationality should orient a reasoner to wanting less violence?
On the grounds of austere logic, the answer is no. But it doesn’t take much to switch it to yes. All you need are two conditions. The first is that the reasoners care about their own well-being: that they prefer to live rather than die, keep their body parts intact rather than have them maimed, and spend their days in comfort rather than in pain. Mere logic does not force them to have those prejudices. Yet any product of natural selection—indeed, any agent that manages to endure the ravages of entropy long enough to be reasoning in the first place—will in all likelihood have them.
The second condition is that the reasoners be part of a community of other agents who can impinge on their well-being and who can exchange messages and