Thanks to Darrin McMahon, whose truly stimulating essay sent my mind off on more tangents than I can hope to share in a short reply. It will be more than enough to riff off just this one passage in Darrin’s essay:
From Jeremy Bentham’s efforts to apply “arithmetical calculation to the elements of happiness” to Marx’s quest to secure “real happiness,” the dream of scientifically managing social welfare has enticed theorists and technocrats alike. The fact that all have failed, of course, does not ipso facto mean that present and future attempts will similarly come to naught. But that knowledge should at the very least give us pause.
I am sure Darrin is right, and I’d like to pause with him before rushing to empower political elites with the authority to nudge us tenderly, coercively toward happiness.
There are at least three enormous problems with the quest for a scientific politics of happiness. The first is that happiness is but one among many values. There is also beauty, truth, love, and freedom, to name a few good ones. Jeremy Bentham believed — as Richard Layard professes still to believe — that all these have value only as instruments to happiness, but I don’t buy it, and I suspect Darrin doesn’t either. We do and think things for lots of reasons, and some of our reasons are good. But not all our good reasons are reasons of happiness. Beauty, truth, love, and freedom can be good reasons, too, and we can have good reason to choose them over happiness, if we have to choose. I don’t think this is self-evident, but I do think it is evident if you pay close attention to the texture of human moral life, as a good empiricist should. At the very least, there is nothing especially scientific about choosing happiness as the master value. Why not a scientific politics of science, aimed at truth?
The second enormous problem: What is happiness anyway? Have economists and psychologists finally nailed down its elusive essence? Should we alert the philosophers? As I note in my new Cato Policy Analysis, “In Pursuit of Happiness Research: Is It Reliable? What Does It Imply For Policy,” even the biggest names in “happiness research” aren’t agreed about what happiness is, which makes it hard to see how the government is going to helpfully measure it. This isn’t surprising if you have had the chance to read Darrin’s wonderful book, Happiness: A History, which helps you to see that happiness has a history — that it is a cultural and historical moving target, and at any given time embodies shifting ideals about the good life. But let me set aside the thorny problem of definitions and move on to the problem I think is most interesting.
The third enormous problem with the quest for a scientific politics of happiness is that it so smoothly shades into the “dream of scientifically managing social welfare” — a dream of pseudoscience. One of the main conclusions of 20th century social science is that social welfare cannot be effectively scientifically managed, neither in command and control dictatorships nor in liberal democracies. Kenneth Arrow proved that there is no satisfactory way of aggregating individual preferences into a single summed social good. Friedrich Hayek showed us why government managers will rarely have the information they need effectively to manage. James Buchanan and Gordon Tullock showed that the interests of policymakers and the people often diverge. Anthony Downs showed us that voters will remain largely ignorant of politics, and so cannot be counted on to choose the most effective administrators. Mancur Olson revealed the depressing logic behind the policymaking dominance of special interests and lobbying groups. I’m glad Darrin mentioned Bruno Frey and Alois Stutzer’s outstanding paper “Should We Maximize National Happiness?” since Frey and Stutzer are among the distressingly few active happiness scholars who intimately understand the great advances of 20th century political economy. A truly scientific politics is one that grasps the nature and limits of politics.
But a scientific politics of happiness need not be conceived as the scientific management of social welfare. The American Founders, it is worth pointing out, were self-consciously involved in the Enlightenment project of developing a scientific politics of happiness, but one based more on artfully limiting power than in providing broad scope for benevolent public administration. Gouverneur Morris, who penned the final draft of the Constitution, called politics “the sublime science which embraces for its object the happiness of mankind.” John Adams in his 1776 “Thoughts on Government” styled politics “the divine science of social happiness” and wrote, in grand Enlightenment fashion, that “the happiness of society is the end of government, as all divines and moral philosophers will agree that the happiness of the individual is the end of man.” Yet Adams pointedly maintained that the structure or “form” of the institutions of government is the paramount concern for a proper science of social happiness:
Pope flattered tyrants too much when he said,
“For forms of government let fools contest,
That which is best administered is best.”
Nothing can be more fallacious than this. But poets read history to collect flowers, not fruits; they attend to fanciful images, not the effects of social institutions. Nothing is more certain, from the history of nations and nature of man, than that some forms of government are better fitted for being well administered than others.
It is plausible to see Founders such as Adams, Madison, Hamilton and Morris as “institutionalists” who conceive the “science of social happiness” as a science of the incentives and equilibrium properties of various institutional arrangements. The point for the institutionalist is not effective, ongoing rational management using the latest scientific knowledge. Instead, the point is a stable, general structure of rules under which individuals can form reasonable, long-term expectations and successfully coordinate with others to achieve their aims in accordance with whatever conception of happiness happens to animate them.
A scientific politics of happiness focused on social coordination is not objectionable in the same way such a politics focused on the aggregation and maximization of happiness may be. The great 20th century liberal philosophers John Rawls and Robert Nozick both rightly argued that Bentham-style utilitarianism, which takes the maximization of happiness as its aim, fails to duly recognize the “separateness of persons” by failing to recognize in-principle constraints on what people may do to one another. This is a compelling moral objection to maximizing doctrines, but it is also an excellent practical objection to policies that actively try to maximize. While bits of happiness (hedons? utils?) are the units of maximization, persons and their plans are the units of coordination, and they will not effectively coordinate without a common framework of expectations that takes their inviolability for granted. Without confidence in the security of our bodies and property, and in the stability of the ongoing rules of social engagement, we will fail to commit to the forms of cooperation most likely bring the greatest mutual rewards — the greatest prosperity, the greatest happiness.
The wonderful thing about a coordination-focused scientific politics of happiness is that it is able to completely sidestep my first two “enormous problems.” If you are trying to create a general framework for effective social coordination, you have no choice but to deal with what John Rawls called “the fact of reasonable pluralism,” which is the unavoidable fact that reasonable people differ about the nature of the good life. You don’t have to be an actual pluralist about value, like me, to acknowledge the fact of reasonable pluralism. A stable system of social coordination in a cosmopolitan, pluralist society is going to have to coordinate the behavior of people who are moved by different values and aiming at different ends. The need to coordinate persons and plans under conditions of moral diversity tends to push the terms of association in the direction of greater generality and neutrality, in which case no one has much reason to mind if the overall system happens to generate a great deal of happiness on the way to generating whatever it was that variously motivated individuals were trying to generate. And, in which case, the difficulty of pinning down the definition of happiness is moot, since there was no need to define it or measure it in order to produce it.
I don’t worry so much about people “pursuing happiness too hard,” to use Darrin’s words, although I do agree that the over-self-conscious pursuit of happiness can sometimes be self-undermining. I believe, as Ruut Veenhoven shows, that we are in fact getting happier — at least according to one notion of happiness. Indeed, I suspect that the dominant survey techniques for measuring happiness understate how much better we now feel thanks to the growth of wealth and the forward march of science and technology. Furthermore, according to Veenhoven’s data, it appears that most of the happiest places on Earth are stable, wealthy liberal democracies with high levels of political and economic liberty — places where the overall rules of the game most closely approximate the schemes of optimal social coordination laid down by the great classical liberal political theorists.
There should be no doubt that we can do better. We can become happier, and we shouldn’t shrink from trying. However, the quest for a scientific politics specifically aimed at happiness is either dangerous or superfluous. It either ends as harmful pseudoscience or merges into a more general science of coordination that is more or less indifferent to happiness.
Thanks again to Darrin for his thought-provoking essay. I had meant to say something about why I think people worry so much about happiness these days, but that will have to wait until the next round of conversation.
Will Wilkinson is a policy analyst at the Cato Institute and managing editor of Cato Unbound. His paper, “In Pursuit of Happiness Research: Is It Reliable? What Does It Imply for Policy?” was published last week.