In this article, Allen Buchanan offers a provocative critique of Rawls’ difference principle (DP), according to which social and economic inequalities ought to be arranged so as to benefit the worst off. He begins by suggesting a similarity between his critique and Nozick’s well-known critique of ‘end-state theories’ (end state theories are one’s which recommend a specific distributive pattern). Nozick holds that end state theories require “continuous interference into people’s lives” that violates liberty. Buchanan suggests that Nozick’s basic point is right, even if the point about liberty is wrong.
Buchanan first argues that Rawls’ difference principle “makes the justness or unjustness of societies, actions and expectations alarmingly transient” (420). To illustrate, suppose at T1 the DP is satisfied by basic structure b, and institutions are arranged accordingly. But, suppose at T2, some new facts come to light – for example, a new technology that makes new arrangements feasible. In light of these new facts, b might not maximize the benefit to the worst off. A new structure b’ might do so. Since b is now unjust, Rawls’ DP requires that we bring about b’.
As a result, Buchanan suggests, applying the DP may require ‘gross and frequent disruptions of individuals’ framing and execution of long-term plans’ (420) even though those plans were not previously unjust. This is so because citizens’ expectations are legitimate if and only if they issue from a just basic structure; if the basic structure is no longer just, their expectations no longer have a claim to be honored.
Buchanan gives two ideal theory examples:
- A t T1 benefiting the worst off requires a new health service, which requires a training program together with certain incentives. At T2, the program is scrapped because new equipment makes it a hindrances to maximizing the expectations of the worst off. What claim do the individuals who participated in the program have based on their legitimate expectations which arose as a result of encouragement by past policy?
- At T1 the DP is satisfied by a regime of ‘property-owning democracy,’ while at T2 the DP is satisfied by a regime of ‘market socialism’. What of the expectations of property owners?
Buchanan then contrasts these with a non-ideal case:
- In order to rectify past injustice, a society institutes a quota system on minority applications to law school. The non-minority applicant complains that the introduction of a quota system ‘late in the game’ undercuts his effectiveness as a rational planner.
In cases (3), we can reply that the agent’s past expectations were founded on unjust arrangements [note: Buchanan takes the legitimacy of the agent’s expectations in (3) much more seriously than I do. One way to define privilege might as the assumptions that one’s expectations continue to be met, where such expectations are the product of unjust institutions and doing so requires continual injustice]. But in case (1) and (2), we cannot say the past arrangement was unjust.
Buchanan considers various responses:
- The appeal to rule of law as ‘providing a stable, relatively permanent framework of expectations.’ Buchanan suggests this fails because what is missing is ‘a theory of institutional change,’ not merely a set of principles for cases wherein institutions remain constant.
- The appeal to the ideal/non-ideal distinction. This doesn’t help because the objection applies equally in ideal theory. The problem is how to maintain the justness of a scheme over time, where maintaining justness requires institutional change.
Buchanan then suggests a possible Rawlsian solution: Rawls could appeal to the difference principle itself. The DP primarily distributes long-term life prospects and expectations, and if we include institutions for institutional change in the basic structure, then we should apply the DP in making decisions about how to maintain justice. Many disruptive changes may not be to the benefit of the worst off, since much depends on stable arrangements. Furthermore, determining whether a given disruption would be to the benefit of the worst off might itself require vast expenditure of resources, which might not be to the benefit of the worst off (similar remarks might be made contra Nozick’s objection).
Buchanan’s conclusion is tentative: “Until a corresponding ideal theory of institutional change is adequately developed, the acceptability of a principle of distribution simply cannot be determined” (425). That might be. But the basic tenant of Buchanan’s solution – namely, applying the principles of justice to institutional change itself –is promising. This is especially so when one considers that, in a Rawlsian well-ordered society, participants (1) would presumably know that maintaining justice would require institutional change, and (2) would accept the principles of justice as a public conception of justice for making such changes. If they know (1) and (2), then, presumably, it would be reasonable to regard citizens as capable adjusting their expectations accordingly.
Buchanan, Allen. “Distributive justice and legitimate expectations.”Philosophical Studies 28.6 (1975): 419-425.
Political Not Metaphysical 