Amartya Sen. Collective Choice and Social Welfare. Holden-Day, (San Francisco: 1970).
Amartya Sen in his monograph on Collective Choice and Social Welfare1 used the disagreement about the proper way to deal with the dilemma who should (and should not) read Lawrence's sensational book as a telling case of contradiction between personal liberty and social welfare.2 If, as Sen proves it, a free society is deeply paradoxical, then the belief in freedom of classical liberals (rather than 'liberals' in the modern American sense of Franklin Delano Roosevelt) is not logically consistent and therefore inherently weak. If we cannot hold individual freedom as the supreme value, then we in effect have lost the battle of ideas. Freedom becomes one of the many values of the good society, on a par with equality, good health, opportunity. It is telling that Sen ended his book on Collective Choice with the following phrase, which can be immediately applied to freedom: "while purity is an uncomplicated virtue for olive oil, sea air, and heroines of folk tales, it is not so for systems of collective choice." (Sen, page 200) If it turns out that Sen and the other practitioners of the collective choice philosophy are right, and liberal democracy is paradoxical and liberty to be valued sub conditione, then what are we classical liberals to do? Are we to become American liberals? You will agree the question is weighty.
Sen's Collective Choice was a valiant attempt to rescue welfare economics from the doldrums where the utilitarians had left it. Welfare economics is the branch of economics that aims at making the theory of economic policy scientific. The utilitarians could not find a way to extricate themselves from what G.E. Moore called "the naturalistic fallacy;" they were unable to find a way across the logical frontier traced by David Hume between 'is' and 'ought' propositions. Then, in 1932, came what seemed to be the final blow: Lionel Robbins powerfully argued that the happiness or welfare of one individual could not be compared with that of another. This made the summing of individual welfares demanded by Jeremy Bentham illogical, the notion of collective happiness inconceivable, and thus welfare economics impossible. So: the end of 'the greatest happiness of the greatest number'.3
The King is dead, long live the King! Welfare economists are perpetually engaged in a fight with Hume, whose interdict has reverberated through the ages. They pursue social welfare, but their starting point is individualistic, for which we should be grateful. They are well aware of the naturalistic fallacy but will not give up interpersonal comparison, about which we should be skeptical. The object of Sen's exercise, as that of a distinguished cohort of thinkers led by Abram Bergson, Paul Samuelson, and Kenneth Arrow, is to sail welfare economics out of the Humean doldrums.
This is not the place to discuss Sen's encompassing proposal to solve the problem of welfare economics by finding some factual bearings to make collective choice a practical craft for ethical voyages. I wish to concentrate on a partial point of Sen's imposing system of economic philosophy: his criticism of liberalism.4 With the example of whether individuals should be free to read D.H. Lawrence's Lady Chatterley's Lover, he wants to uncover a fundamental contradiction or paradox in liberal ideology and thus lay a depth charge under classical liberalism.
From a liberal point of view we must conclude that individuals should be free to read it or not depending on their personal preference, though the book was seen by many as obscene.5 Their preference in Sen's example however, not only regards themselves but the possible negative effects of the book on others (what an economist would call 'external effects'). A true liberal should not be nosey and still hold that the people concerned must be free to follow their preference even if it impinges on those of others. "Nosiness", as Sen calls it, is far from exceptional in a free society, but it makes the choice of individuals very much a social choice. Liberalism thus implicitly becomes the typical object for a collective choice rule, and here Sen proposes is where it becomes paradoxical.
Sen proposes that a minimum condition for a liberal society, which he denominates L, is that at least two6 of its members should be able freely to choose among three alternatives according to their preferences, and that what they choose should be accepted by the rest of society as its collective choice. In the Lady Chatterley example, two individuals, Mr. Prude and Mr. Lewd, must choose among the following three alternatives: (a), that Mr. Prude read the book; (b), that Mr. Lewd read the book; and (c), that no one reads it.
When transposed into a collective choice rule, the order of their preferences gives rise to a circle or contradiction. Thus, Mr. Prude prefers (c), that no one should read it but if it cannot be helped he would prefer (a), that he himself read it rather than let Mr. Lewd be further corrupted by the novel. On the other hand, Mr. Lewd is titillated by the thought of (a), Mr. Prude having to read it; if not that, then he prefers (b), that he himself should read it, since option (c), that no one should read it, would be a waste of a good tale.
In the following table P means 'preferred to':
In the table we see that Mr. Prude prefers that no one should read it, but in a pinch he is ready to read the book himself. As for Mr. Lewd, he would prefer that Mr. Prude should read the book and if not, he would be ready to see if he can glean some new ideas for his love life. The crux of the matter is that, on the basis of Principle L, the individuals' choice becomes the choice of the whole of society. In this case a cycle appears making the social preference is intransitive: society prefers c to a, a to b, and b to c and back again. This paradox ties the liberal society in knots.7
The first time Sen used Lady Chatterley for philosophical purposes was in a soberly titled essay "The impossibility of a Paretian liberal".8 What does the failure to be 'Paretian' imply in this context, and what does 'Paretian' mean?
The mysterious lore of welfare economics is a closed book for non-specialists when they try to debate actual policies 'rationally' or 'scientifically'. One of the more recondite concepts of that lore is the Pareto principle. This was proposed by the Italian economist and sociologist Vilfredo Pareto who thus ingeniously grappled with a fundamental difficulty of political economy: the impossibility, also alleged by Robbins, of fully measuring and comparing the utility, satisfaction, or happiness of different individuals. How can one say that a situation or policy is better than another when the utilities of the people it affects are incommensurable? Pareto proposed a criterion to aggregate individual utilities so that something could be said about society as a whole: a criterion that, although incomplete, can sometimes give us an indication of the way to follow.9
Simply put, a Pareto optimum is one where no one can be bettered without somebody else being worsened; and a Pareto improvement takes place when one individual is bettered without anybody else being worse off.10 A Pareto improvement should be uncontroversial, unless we find envy permissible. The Pareto principle, however, is incomplete, since it gives no guidance for situations where some gain and some lose.
The appeal of the Pareto principle is that it turns out to be unanimity under another garb. Logically speaking there are two ways of conceiving unanimity. One is the strict interpretation of everyone saying yes or no. The other, weaker, one is when a person (or in the Chatterley case, two) agrees, all the rest abstaining. This latter kind of unanimity is labelled nemine discrepante unanimitas—unanimity with nobody disagreeing. In our Chatterley case, Messrs. Prude and Lewd freely express their preferences, the combination of which should become a collective choice rule if the rest of society abstains. What could be more acceptable for a liberal democracy than a collective choice rule be induced from the preferences of two individuals, the rest of society abstaining?
Sen was putting across the idea that the principle of individual freedom could clash with the Pareto principle as a guide for defining a best decision for society; that is to say, he wanted to show that a liberal solution in the Lady Chatterley case would have to override democratic unanimity. The thesis of this essay of mine is that, although we must agree with Sen that cycles will happen, societies in fact avoid these contradictions with the help of institutional fire-doors.
Let me recall that I was a student of Amartya Sen when reading for my Master's degree at the London School of Economics in 1972-3. Quite some time has passed but I still remember why I chose his course despite its unappealing title of "Project Evaluation". I was impatient with the kind of theoretical political economy I had been taught and wanted to be trained in immediately applicable techniques. I found him one of the most inspiring teachers I ever been privileged to follow. His lectures lasted for one and a half hours without a break and not one of his students blinked. He covered the blackboard with mathematical logic derivations and explained them so clearly that we could all excitedly follow them. I especially profited from studying his rather advanced 1970 book. His kind of project evaluation was not a collection of recipes but a corollary to the collective choice theory he had helped to create. He explored countless new avenues in welfare economics before coming to discuss whether a bridge should be built across the river though it displaced the bargee. I had thought welfare economics had been killed by Lionel Robbins when he denied the possibility of interpersonal comparisons of utility. Sen tried to show otherwise. Rarely have I been treated more courteously in class than by Dr. Sen. I took a keen interest in his approach to collective choice and social welfare. In the end, however, I was not convinced and am still not convinced today.
Tough Sen does not say so, the kind of infernal circle he traced in the Chatterley case can lead to confrontation, with Prude demanding that the police impound the book and Lewd distributing free copies at school doors. In view of this. Sen suggested that
... the eventual guarantee for individual freedom cannot be found in mechanisms of collective choice, but in developing values and preferences that respect each other's privacy and personal choices. (page 85)
Indeed, habits and traditions can play a moderating role when the mechanism of collective choice fails. This was the guarantee suggested by Alexis de Tocqueville in the second volume of Democracy in America11 when he said that the mores of a free society could be a bulwark against the incursions of democracy into personal liberty.
I think I can go further than this. It is my belief that societies, especially democratic societies, evolve institutions that overcome the contradictions between personal and political freedom, on the one hand, and the democratic principle, on the other. Laying all one's hopes for personal liberty on the growth of civilized habits in democratic societies cannot be the whole story. We must see if human societies have self-healing capacities for 'Lady Chatterley' kinds of conflicts, in the form of the emergence and growth of institutions.
The sum of what Sen has to say about societies trying to be liberal and democratic at the same time is that they are prone to logical contradictions which will give occasion to conflict, unless an outside arbiter invents and applies the necessary rules.
My thesis is that societies evolve institutions that cut the cycle and stop conflict from spreading. The market is one of those institutions that emerge over time in societies to act as a 'fire-wall' to cut the spreading of cycles and thus avoid the occasions of conflict due to indecisive social situations of the kind we have just described. The market is the prime example of systems that harmonize individual preferences and at the same time fulfill the Pareto criterion of limited unanimity. Given sufficient competition, the market is the vehicle for separate individual agreements, where external effects, negative and positive, are channeled through prices. These prices are impersonal bits of information, which individuals take as data. Of course competition must be understood dynamically, in the sense that monopolies and oligopolies tend to be eroded over time, unless protected by government. From the point of view of social welfare, however, the market usually gets short shrift, and is seen as a very imperfect mechanism that needs to be constantly overseen by regulators.
For more on these topics, see "The Dangers of Majoritarian Democracy," by Pedro Schwartz, Library of Economics and Liberty, Dec. 5, 2013; and "Collective Choice at Work," by Anthony de Jasay, Library of Economics and Liberty, Dec. 3, 2012. See also Externalities by Bryan Caplan in the Concise Encyclopedia of Economics, and "Disputing Tastes", by John V.C. Nye, Oct. 6, 2003, Library of Economics and Liberty.
Another of these fire-wall institutions is private property. This is not the place to discuss the theories presented through the ages in defense of private property. My particular argument in this essay is that the institution of private property must be seen as another evolutionary response to self-destructive social conflict. I therefore find it little short of irresponsible that the institution of property does not get full attention in the collective choice literature. In fact, 'private property' is not even mentioned in Sen's book.
More generally, the kind of society that these collective welfare theorists outline is a kind of never-never land, devoid of institutions apart from voting rules, where disembodied individuals are organized by some all-knowing academic from Harvard or Princeton, totally oblivious of the operation of the forces of spontaneous order.
Let me show how the fangs of conflicts such as who is allowed to read Lady Chatterley's Lover can be drawn by the emergence of property rights. If Mr. Lewd possessed the only copy of Lady Chatterley in the market, all others having been burnt by a re-born Inquisition, then he could sell the copy he had just purchased on the black market to Mr. Prude for a (considerable) price, before he had had time to read it. Mr. Prude would then keep it uncut and under lock and key. That sale would be a Pareto improvement, as resulting from a voluntary contract, whereby Mr. Lewd had received a price high enough to pass on the property to Mr. Prude, who would reach his preferred state of no one reading the book. And vice versa if Prude was the original owner of the novel and needed funds for his charities. Simply by attributing the ownership of the book to one or the other of the two, the spell would be broken.12
I think most of my readers will agree that no kind of discussion of social can proceed without taking the preservation of human society into account. As David Hume put it, society cannot subsist without two kinds of moral duties being generally observed: "justice or a regard to the property of others, fidelity or the observance of promises".13 The need and convenience for the respect of property and contract have not been invented by some social welfare theorist but are the result of successful evolution. These observations of Hume's are wrongly interpreted as some kind of utilitarianism. I rather see them as an expression of the general doctrine of the Scottish moralists about the establishments nations stumble upon, "which are indeed the result of human action, but not the execution of any human design".14 I am here laying my own kind of depth charge under collective choice theory: Bergson, Samuelson, Arrow, Sen, all think they know and should re-read Friedrich Hayek's 1945 article on "The Use of Knowledge in Society".15
Amartya Sen. Collective Choice and Social Welfare. Holden-Day, (San Francisco: 1970).
Sen later expressed some regret at the sensation caused by his having used Lady Chatterley to illustrate the paradox of liberalism. See "The Impossibility of a Paretian Liberal," Journal of Political Economy, volume 78, 1970.
See "Jeremy Bentham's Mistaken Universalism" by Pedro Schwartz. June 1, 2015. Library of Economics and Liberty.
I will use the denomination of 'liberalism' in its classical sense of 'the system of individual freedom' and not as used in the United States after Franklin Roosevelt kidnapped it in the 1930s.
So much so that so that the publishers were taken to court in 1960, when a jury finally acquitted them.
If it is only one person to thus dictate the collective rule, then he would became a 'dictator', which is not very liberal.
Sen dissects the difficulties encountered in generating a collective choice rule out of the preferences of two individuals (in this case Messrs. Prude and Lewd) in "Conflicts and Dilemmas", chapter 6 of Collective Choice and Social Welfare (1970). The common sense conditions that cannot be fulfilled in the Lady Chatterley case are: that the social decision should be rational by not falling into a circle, and that the solution be applicable whatever the content of the dispute.
Journal of Political Economy, volume 78, 1970. He then expounded his arguments in his 1970 book.
Pareto also solved the difficulty of basing the negatively inclined demand curves in markets on the law of decreasing marginal utilities. Instead of demand curves he used 'indifference curves', whose points indicated bundles of goods regarding which individuals are indifferent. This avoided the need to measure or rank utility, as in ordinary demand curves.
In Sen's formulation it boils down to two rules: "(a) if everyone in the society is indifferent between two alternative social situations x and y, the society should be indifferent too; (b) if at least one individual strictly prefers x to y, and every individual regards x to be at least as good as y, then society should prefer x to y." See Chapter 2, "Unanimity" of Sen (1970).
Alexis de Tocqueville. Democracy in America, Historical-Critical Edition. Indianapolis: Liberty Fund, 2010. Available online at the Online Library of Liberty.
Notice that by defining property rights is how Ronald Coase showed that society spontaneously overcomes negative external effects. Remember "The problem of Social Cost", Journal of Law and Economics, Vol. 3 (October, 1960), pages 1-44.
David Hume, "Of the Original Contract," in Essays Moral, Political, and Literary. Indianapolis: Liberty Fund, 1987. Hume's italics. Online at the Library of Economics and Liberty.