Despite being popularly positioned as leading advocates of opposing political philosophies, the signature works of public choice founder James Buchanan (with co-founder Gordon Tullock) and philosopher John Rawls share the same foundational approach. While Buchanan became more critical of Rawls’s work when A Theory of Justice finally appeared in 1971, his criticisms are more tempered than many readers would expect. Buchanan also admitted that his criticism of Rawls likewise indicted the approach that he and Tullock developed a decade earlier in The Calculus of Consent.

Both Buchanan and Rawls cast large shadows across the academy in the last generation. Rather than blithely assuming that political decisions by politicians and voters aim neutrally to advance the common good, Buchanan recovered and advanced the realist Madisonian insight that “modern legislation … involves the spirit of party and faction in the necessary and ordinary operations of government.” In recognition of the significant influence Buchanan had on both academic and popular thought, Buchanan won the Nobel Memorial Prize in Economics in 1986. 

Rawls cast a similarly outsized shadow, although in a different direction. While A Theory of Justice can be viewed as justifying the post-WWII mixture of equality and liberty in mainstream liberalism, its signature contribution has been in providing a philosophical rationale for the modern redistributive welfare state. A Theory of Justice has dominated academic and popular political theory since its publication over fifty years ago.

Because Rawls’ argument is widely taken to justify government activism precisely where Buchanan’s approach justifies caution, they are commonly conceived as occupying opposite and rival positions on the political spectrum. 

Yet throughout the 1960s, after publication of The Calculus of Consent, Buchanan and Rawls communicated frequently and warmly with each other, drawn together intellectually by commonalities in their work. Not only did both apply strikingly similar social contract methodologies, but the implications they drew from their theories overlapped more than one might anticipate. 

A Shared Social Contract Methodology: Choosing Behind a Veil of Ignorance

While the questions Buchanan and Rawls sought to address differed—Rawls focused on deriving general principles of justice while Buchanan and Tullock focused on the constitutional “emergence of democracy”—each addressed their questions using a social contract approach that shared strikingly similar attributes. Both theories developed an argument in which rational individuals at a hypothetical initial stage would unanimously consent to a framework that obliged them to accept subsequent policy choices even when those choices ran contrary to their individual policy preferences at the later time. 

Both theories employed a similar methodological move by constructing a representative decision maker who made initial decisions regarding justice or constitutional structures without full knowledge of their socioeconomic positions and conditions.

Of the two works, Rawls’ approach is best known, given his quotable label of this methodological move as decision-making behind a “veil of ignorance.” In the original position in which principles of justice are chosen, people do not know their socioeconomic status, intelligence, strength, sex, the conditions of the society in which they live, or even their “conception of the good” or preferences over risk. This uncertainty plays a crucial role in the development of Rawls’ social contract theory: “The veil of ignorance makes possible a unanimous choice of a particular conception of justice.”

Buchanan criticized Rawls for pushing the content of the principles of justice chosen behind the veil of ignorance in the original position further than Buchanan thought analytically appropriate.

Widespread reference to and invocation of Rawls’ “veil of ignorance” bears striking testimony to the power of a well-turned phrase, even in academic writing.

Buchanan and Tullock made an identical methodological move nine years earlier in The Calculus of Consent. Identical, that is, in all but name. Where Rawls coined the evocative-sounding phrase “veil of ignorance,” Buchanan and Tullock simply noted that their theory assumed “that the individual is uncertain as to what his own precise role will be in any one of a whole chain of later collective choices that will actually have to be made.”

Needless to say, Buchanan and Tullock’s line did not resonate as nimbly as Rawls’ “veil of ignorance.” 

To be sure, Buchanan and Tullock did not originate the concept either. Rawls appealed to a much thinner version of the idea (and without calling it a veil of ignorance) in a 1958 article in The Philosophical Review. And Rawls cited a 1953 article by John Harsanyi in which Harsanyi argued that political and policy impartiality could be guaranteed if decisions were made “in complete ignorance” of a person’s ability to gain or lose from a decision.

Surprisingly, in his 1958 article, Rawls does not assume uncertainty or ignorance of personal attributes on the part of decision-makers in the original position. Rather, he argues that impartiality in the selection of the principles of justice would be sufficiently induced in the original position because those principles would be “binding on future occasions.” It is only the uncertainty of “future circumstances the peculiarities of which cannot be known” which, according to Rawls, would induce persons deliberating on principles of justice to be impartial in selecting those principles. 

Unlike the person behind the veil of ignorance in his 1971 book, those participating in the hypothetical original position in Rawls’ 1958 article would not be deracinated; they would know their current social, economic, and historical positions as well as their personal characteristics and beliefs. Rawls defends this by tapping into intuitions about the creation of recreational games. Consider a group of children seeking to create their own game on a playground. The rules for these games would be accepted as fair and would receive unanimous consent only if they did not asymmetrically reward a predominant attribute already possessed by one of the children creating the game. 

In The Calculus of Consent, Buchanan and Tullock wrap their initial-position decision makers in a cloak of uncertainty that is closer in spirit to Rawls’ veil of ignorance than it is to Rawls’ 1958 framework. They assume the individual at the initial constitution-making stage does not have “a particular and distinguishable interest separate and apart from his fellows”; the individual is “uncertain as to what his own precise role will be in any … of later collective choices that will actually have to be made.”

Buchanan and Tullock deploy the idea of uncertainty regarding individual attributes at the constitution-making stage for the same purposes as Rawls: it allows them to argue that unanimous consent would be provided to an appropriately calibrated constitution. Unanimity at the initial stage also accounts “for the emergence of democratic institutions.” Buchanan and Tullock develop an argument in which a group of rational individuals would unanimously agree to be bound by non-unanimous policy choices in (some) post-constitutional policy areas.

While representative individuals at the constitution-making stage (under uncertainty) would recognize the likelihood of being on the losing end of some policy choices made under the constitution, nonetheless, the overall gains from allowing non-unanimous collective action in certain areas would be greater than the overall losses of forbidding all non-unanimous collective decisions.

In this way, Buchanan and Tullock provide an “economic … explanation for the emergence of democratic political institutions” and answer the constitutional question of “what activities should be left in the realm of private choice and what activities should be collectivized [through assignment to non-unanimous government decision making].” Their rationale for the unanimous acceptance of a constitution at an initial stage allows for non-unanimous policy decisions in subsequent operational stages.

The constitution-making individuals of The Calculus of Consent fall somewhere between Rawls’ 1958 argument and his later argument in A Theory of Justice when it comes to the amount of information available to them. They must have sufficient information about personal attributes and socio-economic conditions to be able to calculate and compare external and decision costs. It is these comparative interdependence costs that allow individuals at the constitution-making stage to recognize that some externalities should be left to lie undisturbed where they fall initially, some externalities should be dealt with through private contracting, and some externalities would optimally be dealt with through collective action. 

In contrast, in selecting broad principles of justice from behind the veil of ignorance, Rawls argues in A Theory of Justice that the deracinated decision-makers would be unable even to assign probabilities to the different circumstances in which one might potentially find oneself on the other side of the veil of ignorance. This deep uncertainty plays a pivotal role in Rawls’ subsequent analysis; he argues that individuals would respond to this level of uncertainty by adopting maximin preferences. That is, individuals behind the veil of ignorance would (unanimously) choose those principles of justice that would select the optimal among the worst possible outcomes for everyone.

Rawls then argues that decision-makers applying maximin preferences behind the veil of ignorance would unanimously adopt two principles of justice:

First: each person is to have an equal right to the most extensive basic liberty compatible with a similar liberty for others.

Second: social and economic inequalities are to be arranged so that they are both (a) reasonably expected to be to everyone’s advantage, and (b) attached to positions and offices open to all.

These two principles are lexically ordered, meaning that tradeoffs cannot be made across the principles. That is, at the operational stage, policies cannot impose restrictions on equal liberty (the first principle) in order to promote overall economic gain (the second principle).

Rawls famously argues that the second principle of justice implies what he terms the “difference principle,” that is, “that the social order is not to establish and secure the more attractive prospects of those better off unless doing so is to the advantage of those less fortunate.” 

Buchanan’s Reservations on A Theory of Justice, and on The Calculus of Consent

Buchanan reviewed A Theory of Justice after it came out in 1971. He criticized the argument on two fronts. First, Buchanan criticized Rawls for pushing the content of the principles of justice chosen behind the veil of ignorance in the original position further than Buchanan thought analytically appropriate. Second, Buchanan noted that he was “less of a contractarian” than he was when he and Rawls first began interacting in the early sixties. Poignantly, Buchanan noted that his reservations about the idealism in Rawls’ social contract theory also applied to the approach he and Tullock applied in The Calculus of Consent.

Buchanan’s first criticism of A Theory of Justice is not what we might expect: that the difference principle is inconsistent with justice or fairness. Buchanan agrees that the difference principle is consistent with fairness. Rather, Buchanan’s criticism of the difference principle is that it is only one of many possible principles that would be consistent with fairness and that Rawls is unjustified in drawing attention uniquely to the difference principle.

To his first principle of equal liberty, Rawls appends lexicographically his second “difference principle.” This states, specifically, that allowable distributional inequalities among persons are acceptable only to the extent that their existence benefits the least-advantaged members of the community.

I should accept the hypothesis that a socio-political-economic structure embodying the difference principle meets widely accepted criteria of “fairness.” But I should not be prepared to elevate this principle into the ideal position accorded it by Rawls. There may be many other distributional rules that qualify within the acceptable set. 

Buchanan’s concession that Rawls’ difference principle is among the set of acceptable criteria of fairness is unexpected given the common perception that Buchanan and Rawls occupy opposite poles of the political-economic spectrum. Rather, Buchanan’s criticism is that Rawls’ theory arbitrarily latched onto just one of a broader set of fair principles. 

Buchanan sketches a second reservation to A Theory of Justice, although, again, perhaps surprisingly, he notes that his reservation “is as applicable to [Buchanan’s] own earlier conception of the contractarian position” articulated in The Calculus of Consent as it is to Rawls’ argument.

Buchanan expresses dissatisfaction with the idealism of the assumption of “uncertainty” in The Calculus of Consent and the “veil of ignorance” in Rawls. He asks, “What point is there in talking as if persons will ‘think themselves’ into some idealized version of an ‘original position’ … when we know that, descriptively, the men who must make social choices are not likely to make such an effort?”

In essence, Buchanan makes a public choice criticism of the approach he and Tullock took in The Calculus of Consent. This reflects the rapid evolution of public choice in the decade after publication of The Calculus of Consent.

The central insight of the public choice school, as it developed, was that people making political decisions descriptively face the same incentives people face in other economic and social relationships. People making governmental and political decisions do not act with immunity to the tug of self-interest. Thus, an account of foundational constitutional decisions applicable to the real world must account for these incentives and take people as they are, rather than starting with an idealized version of decision makers.

Ironically, Buchanan then advocates an approach similar to the one Rawls took in his 1958 article in The Philosophical Review. Buchanan observes that when people select more-or-less permanent rules for a more-or-less open time period, they may be “uncertain as to just where their own self-interest lies” and so will “be motivated to opt for rules and rules changes that embody ‘fairness.’” Nonetheless, this motivation for impartiality cannot be idealized because framing social institutions “will reflect the struggle among interests, and will rarely, if ever, qualify as ‘just’ in accordance with any idealized criteria.” 

During the decade after publication of The Calculus of Consent, Buchanan seems to have embraced a more thorough-going Madisonianism. Rather than resting content with the idealized approaches of what Madison dismissed as “theoretic politicians,” he essentially conceded Madison’s insight that “the regulation of these various and interfering interests forms the principal task of modern legislation and involves the spirit of party and faction in the necessary and ordinary operation of government.”

This does not mean that justice and fairness are vain pursuits, nor does it mean that impartiality is not an appropriate criterion for judging political decisions and outcomes. But it does mean that social scientists and institutional designers, in studying and crafting institutions, must take humans as we are rather than starting, and ending, with idealized versions of ourselves behind a veil of ignorance.

Source link