For England, like France, had its century of liberalism: and to the century of the French Revolution corresponded, on the other side of the Channel, the century of the Industrial Revolution: to the juristic and spiritualistic philosophy of the Rights of Man corresponded the Utilitarian philosophy of the identity of interests. The interests of all individuals are identical. Every individual is the best judge of his own interests. Therefore it is necessary to break down all artificial barriers which traditional institutions set up between individuals, and all the social restraints based on the supposed necessity of protecting individuals against each other and against themselves. It is a philosophy of emancipation, very different in its inspiration and principles, but akin in many of its applications, to the sentimental philosophy of J.-J. Rousseau. The philosophy of the Rights of Man eventually led, on the Continent, to the Revolution of 1848; in England at the same time the philosophy of the identity of interests resulted in the triumph of the free-trade doctrine of the Manchester School.--Élie Halévy (1901) The Growth of Philosophic Radicalism, Translated by Marry Morris, London Faber & Faber (1928), p. xvi.
Despite the different terminology, Élie Halévy's distinction anticipates the Rawlsian distinction between a social contract and a utilitarian tradition. According to Halévy both traditions are a "philosophy of emancipation." In Halévy (1870 – 1937) the opposition of utilitarianism to the social contract is introduced via Hume's criticism of Locke which Bentham takes over from Hume and then also applies against Blackstone (and the inconsistencies in Beccaria) in chapter 2. It's worth noting, again, that Foucault, treats the first as the "axiomatic, juridico-deductive approach" (he drops its connection to sentimentalism--Halévy is clearly seeing the influence of Mandeville on Rousseau) focused on contracts and rights, which Foucault explicitly associates with Rousseau, and the second Foucault associates with "English radicalism" or "the problem of utility" or interest (The Birth of Biopolitics, 17 January 1979, pp. 39-40), My interest here, however, is not in labeling.
Rawls knew Halévy's book. After introducing "the distinction between the constitutive rules of an institution, which establish its various rights and duties, and so on, and strategies and maxims for how best to take advantage of the institution for particular purposes," and explaining this distinction (Rawls does not cite Buchanan & Tullock here, but rather Searle and Anscombe),* Rawls goes on to cite Halévy's Growth (from the French original) as a source on Bentham and Smith in A Theory of Justice:
In designing and reforming social arrangements one must, of course, examine the schemes and tactics it allows and the forms of behavior which it tends to encourage. Ideally the rules should be set up so that men are led by their predominant interests to act in ways which further socially desirable ends. The conduct of individuals guided by their rational plans should be coordinated as far as possible to achieve results which although not intended or perhaps even foreseen by them are nevertheless the best ones from the standpoint of social justice. Bentham thinks of this coordination as the artificial identification of interests, Adam Smith as the work of the invisible hand.3 It is the aim of the ideal legislator in enacting laws and of the moralist in urging their reform.--Rawls A Theory of Justice (Revised edition), p. 49 [it's also in the original 1971 edition.]
I take it that Rawls is describing his own program when he writes that in designing and reforming those social arrangements which are "the constitutive rules," which just are (a part of) the basic institutions of society, "the conduct of individuals guided by their rational plans should be coordinated as far as possible to achieve results which although not intended or perhaps even foreseen by them are nevertheless the best ones from the standpoint of social justice." As I recount in my book, the distinction between intended and foreseeable consequences is rather important to a proper understanding of Smith's account of the invisible hand, but I leave that aside here.
As an aside, it's worth noting that the terminology of the 'ideal' here (recall) echoes the manner by which Knight and Arrow refer to what they call the 'idealist tradition,' which in the very notes that Rawls calls attention to (see p. 233 and 314 n. 16 of TJ) in Arrow, Arrow mentions Rousseau, Kant, T.H. Green and Milton in the context of discussing the 'Idealist' tradition. (On p. 233, Rawls tacitly divides the Idealist tradition, which he ordinarily only associates with Bradley (in the sense familiar to us), into a social contract variant.))
Be that as it may, in the accompanying footnote to the passage from p. 49, Rawls claims that "The phrase “the artificial identification of interests” is from Elie Halévy’s account of Bentham." Rawls cites pp. 20-24 from volume 1 of the French edition. In the English translation this is pp. 15-18 (which basically runs from Mandeville to Helvetius). The key passage is worth quoting:
But there is yet another argument which can be used: while still admitting that individuals are chiefly or even exclusively egoistic, it is yet possible to deny that their egoisms will ever harmonise either immediately or even ultimately. It is therefore argued that in the interest of individuals the interest of the individual must be identified with the general interest, and that it is the business of the legislator to bring about this identification. This may be called the principle of the artificial identification of interests. Hume approved the maxim of political writers according to which every man should, on principle, be held a knave, and, once this principle had been laid down, concluded that the art of politics consists in governing individuals through their own interests, in creating artifices of such a kind that in spite of their avarice and their ambition they shall co-operate for the public good. If politics are not carried on in this way, it is vain to boast of possess-ing the advantages of a good constitution ; for it will in the end be found that a man’s sole guarantee of his liberty and property consists in the good-will of his rulers, which amounts to saying that he has no guarantee at all. — Now this is the form in which Bentham first adopted the principle of utility. It is true that he occasionally applied, by accident, the principle of the fusion of interests. It is true that in political economy he adopted, with the ideas of Adam Smith, the principle of the natural identity of interests. But the primitive and original form in which in his doctrine the principle of utility is invested is the principle of the artificial identification interests. Bentham appealed to the legislator to solve…the great problem of morals, to identify the interests of the individual with the interests of the community. -Élie Halévy (1901) The Growth of Philosophic Radicalism, pp. 17-18.
(For the roles of knaves in public choice see this lovely paper by Kliemt; and my response.) Notice, first, that Halévy does not claim here explicitly, as Rawls does, that the invisible hand solves the meta-coordination problem that Bentham ascribes as a task legislator. Rawls' position or gloss is explicable, of course, that if (in Bentham) a legislator is needed to identify and coordinate artificial identification (of true/authentic, well-considered) interests, then no such legislator would be needed if such identification is natural (and, hence, can be left to the market mechanism or the invisible hand, or as Halevy, phrases it sometimes in his treatment of Smith, "the spontaneous harmony of egoisms." (p. 89)) Halévy explicitly suggests that Smith invisible hand links egoism to general interest of society or civilization on p. 90, where no "wisdom of a legislator" is said to be required. So, Rawls' echoes Halévy's nicely which suggests he read more of the book than just the pages cited in TJ.
Shortly after the passage Rawls explicitly cites (and before p. 89), Halévy points out that for Bentham, "Either the man whose actions I intend to direct is myself, in which case morals is the art of governing myself or private morals: or else the men whose actions I intend to direct are men other than myself. In this case, if they are not adults, the art of governing them is called education, which is either private or public: if they are adults, the art of directing their actions so as to produce the greatest happiness of the greatest number appertains to legislation, if the acts of the government are of a permanent kind, or to administration, if they are of a temporary kind dictated by circumstances." (p. 27)
So, for Bentham (go check out Bowring, vol. i, pp. 142-143, partially cited by Halévy), the (outer directed) "art of government" directs the actions of "other human beings" and is divided in legislation and administration. The terminology of 'legislation' as the permanent kind is a bit misleading to modern ears, which often associate it not with the constitutional rules or basic structure, but with particular laws. (Bentham's terminology here follows Smith's something Halévy emphasizes throughout.)
In Rawlsian terms the Benthamite art of government just is the setting up of constitutive rules (that is legislation) such that "men are led by their predominant interests to act in ways which further socially desirable ends." If these rules work properly, then "the conduct of individuals guided by their rational plan" produces what in context Rawls calls "social justice." (We later learn quite a bit is packed into the idea of a rational plan for Rawls.) Somewhat surprisingly, then, the design of the basic institutions behind the veil of ignorance in the original position in which we imagine we pursue a rational plan (without knowing our station) just is what Bentham would call the 'art of government.' (A phrase Rawls never uses I think. But I return to Rawlsian art of government some other time.)
I should stop here. But I want to begin to argue that Halévy is non trivial to Foucault's overall argument in Birth of Biopolitics. In Foucault's account the art of government in the history of liberalism (in The Birth of Biopolitics), Hume is the crucial originary figure (against the confusions of Blackstone) that helps set up a separation between the juridical rights based approach and the one focused one interests (see 28 March 1979, pp. 273-274; recall this post). Chronologically, Foucault's argument is a bit dubious (Hume's crucial material seems to be published ahead of Blackstone.) But conceptually, Foucault is clear and he is clearly echoing Halévy's narrative (see, especially, p. 133 in the English translation; but it's anticipated between pp. 12-25!) And, in fact, Foucault's subsequent claim that the subject of interest and the subject of right cannot be superimposed on each other (p. 276) is a nice summary and crystallization of Halévy in memorable language!+
And, then, after developing the point through Mandeville and Condorcet, Foucault turns to Smith's invisible hand and claims "that kind of bizarre mechanism which makes homo oeconomicus function as an individual subject of interest within a totality which eludes him and which nevertheless founds the rationality of his egoistic choices." (p. 278) Now, as a straight reading of the invisible hand passage of Wealth of Nations this is very implausible. But if we return to the very same passage from Halévy that Rawls cites and uses to frame the nature of constitutive rules above quoted above, Foucault's claim is, in fact, a lovely interpretation of Halévy's account of Smith!
TBC+
*The language of 'Constitutive rules' does not play an important role in Rawls' subsequent argument. Its presence harks back to early Rawls' interest in ordoliberal and Hayekian thought (conceived in terms of rules of the game) as documented by Katrina Forrester, see p. 33 and the material leading up to it.
+In his introduction to the Foucault Effect, Colin Gordon calls attention to Halévy's treatment of the "radical discord between the economic and juridical register" (see p. 21). In the paragraph before Gordon goes over the material I had discussed in my earlier blog post on Hume and Foucault. So, I don't want to claim originality in recognizing the significance of Halévy here.
Comments