Julius Sensat / Philosophy 941: Rawls / Notes / Installment 4: Reasoning in the original position (1)
In Part III of JF, Rawls discusses the nature of the original position (OP) and the argument "from the original position" for his two principles of justice. As we shall see, that argument has two parts, the second of which is not given until we come to the topic of stability. But we get a fairly full statement of the first part of the argument in this part.

Features of the OP and its role in the defense of the principles

Rawls tells us that the original position is a "device of representation" that models two things:
  1. The conditions we (here and now) regard as fair for free and equal citizens to agree on terms of cooperation to be expressed as principles of justice.
  2. The restrictions that we (here and now) regard as appropriate on the reasons on the basis of which the parties in the OP can put forward or reject principles.
To illustrate the second part of this assertion: since we do not regard the fact that someone occupies a certain social position (is wealthy for example) as a good reason for that person to accept or expect others to accept principles of justice that favor those in that position, we do not allow the parties in the OP to know the social positions of the citizens they represent. It is important not to be misled by this restriction into thinking that the parties in the OP are actually arguing for or debating on which principles are correct principles of justice. They are not doing that. Rather, what the parties agree to is based on what each thinks most protects the fundamental interests of the citizen he represents. Thus the parties act as "trustees" on behalf of their respective client citizens. Note that this assumption itself should not be misunderstood as implying that the citizens are being represented as selfish or self-interested, as these terms are normally used. Whether people are selfish in this sense depends on the content of their final ends, and it is not being assumed that these ends are selfish ones (§ 24.2).

Another function of the OP device is that it enables us (here and now) to keep track of our assumptions in arguing for or deliberating about justice. As Rawls, says, we "can see what we've assumed by looking at the way the parties and their situation have been described" (p. 81). Ideally--though Rawls says this is not fully realizable--we should be able to deduce from that description which principles the parties would agree to from a given list of alternatives. Who specifies these alternatives? We do. We construct a list of what we take to be the known alternatives in contention. Anyone can propose an alternative, and it then must be added. If we determine that an alternative would not be chosen in the OP, then it is open to them to argue that the setup of the OP is flawed in some way, or even to argue against using that kind of device to determine the principles of justice.

Rawls's conjecture is that the most reasonable principles of justice for the basic structure of a modern democratic society are those that rational representatives of citizens, when subject to reasonable constraints, would themselves adopt. He offers the OP as a specification of which constraints are reasonable. It is up to us to decide whether on reflection we can agree with him.

Nature of the parties

In the OP, the parties rationally pursue certain aims. These aims, which could be grouped under a general description as the aim of protecting their repective client citizens' fundamental interests, are these:
  1. To secure conditions that will enable them to pursue their conception of the good (basic personal aims and commitments, view of the meaning of life, etc.).
  2. To secure conditions that will enable them to develop and exercise their first "moral power," namely their power to reflect on and revise their conception of the good for reasons that seem compelling to them.
  3. To secure conditions enabling them to develop and exercise their second moral power--the capacity to understand, apply and live by the conception of justice agreed to.
They try to secure all these conditions by deciding on principles that specify how the basic institutions will distribute prospects for primary goods and by deciding on a "companion agreement" (p. 89) on forms of reasoning and standards of evidence that citizens are to use when discussing and debating matters of political justice.

Constraints on the parties

I will list here what seem to me to be the most important constraints.
  1. Publicity: the parties are consciously choosing principles to serve as the public charter of a well-ordered society.
  2. Finality: the agreement is a commitment to live by the principles adopted (rather than an agreement that can be revoked by citizens dissatisfied with their lot).
  3. Veil of ignorance. The parties don't know their clients' place in society, their natural endowment or their conception of the good.
  4. On the other hand, the parties do know the general principles of noncontroversial science and common sense, including standards of evidence and inference.
  5. They also know that the circumstances of justice for a modern democratic society obtain.
Circumstances of justice:
  1. Moderate scarcity (cooperation required for all to have adequate life; constitutional democracy possible)
  2. Fact of reasonable pluralism and related facts (e.g., "fact of oppression").

"Two parts" and "two comparisons"

Rawls says that the argument for his principles from the original positions contains two parts. In the first, the parties assumethat citizens will not be subject to certain special psychologies: envy, spite, peculiarly high aversions to risk and uncertainty, a strong will to dominate others, etc. The parties aim at selecting the principles of justice that best secure those citizens' fundamental interests, given this assumption. In the second part of the argument, the parties go on to consider whether the chosen principles, when they are actually implemented, would generate their own support, or on the contrary would generate the aforementioned special psychologies to a degree that the conception of justice breaks down.

The first part of the argument itself includes two "fundamental comparisons." In the first, the parties choose between Rawls's two principles and the principle of average utility, which says that the basic structure should be designed to maximize the average welfare of the population, from now into the foreseeable future. If Rawls's two principles win in this comparison, they are then compared in the OP with an alternative set that agrees with them except that they substitute the principle of average utility, combined with a stipulated social minimum, for the difference principle. Rawls uses the second comparison to zero in on special issues connected with the desirability of the difference principle.

Though Rawls does not say it explicitly, his setup implies that it might be useful and even important to make other comparisons as well, for example his two principles against principles that define a libertarian alternative.

A thought experiment

To see how Rawls envisages the reasoning in these comparisons, especially in the first, suppose you are offered a choice between the following gambles on the outcome of a drawing of a ball from an opaque urn: You are not told how many balls are in the urn. You are told that they are either red or black, but you are not told the percentages (they could be all red or all black or anything in between). Which would you choose?

Suppose that the following conditions hold:

  1. G1 has a possible outcome that would be disastrous.
  2. You care relatively little about gains above $1000
  3. You have no evidence for assigning probabilities.
Some or all of these conditions could work to make it rational to "play it safe" and choose the second gamble.

Some jargon:

The foregoing three conditions can be restated as general conditions that can make it rational to follow the maximin rule in making decisions under uncertainty:
  1. You have no basis for assigning probabilities.
  2. The maximin option guarantees a certain minimum (call it the guaranteeable level) that is highly satisfactory, so that you care relatively little about possible gains above that level.
  3. The alternatives to the maximin option could have a disastrous outcome (an outcome considerably worse than the guaranteeable level).

First fundamental comparison

In the argument for the first fundamental comparison, Rawls's strategy is to convince us that these three conditions obtain to a sufficiently great degree that it is rational of the parties in the OP to be guided by the maximin rule. He then claims that justice as fairness will be chosen over average utilitarianism, since it is the maximin option.

More precisely, he focuses on the second and third conditions as doing the most work. That is, he tries to show that using the maximin rule is rational when the second and third conditions are realized to a high degree, even if the first condition is not fully satisfied. This relative avoidance of the first condition is something that stands out in JF compared to TJ. He may have come to believe that to argue that the first condition is fully satisfied he would have to appeal to claims about probability that are controversial. Next week, I will try to give you some sense of what is at stake in the debate over the first condition.

The important sections for the first fundamental comparison are the following:

§29 The Argument Stressing the Third Condition
§30 The Priority of the Basic Liberties
§31 An Objection about Aversion to Uncertainty
§32 The Equal Basic Liberties Revisited
§33 The Argument Stressing the Second Condition
Notice that Rawls seems to be dividing the argumentative strategy into two parts: one arguing that the third condition is satisfied, and another arguing the same for the second condition. That may be true analytically, but he does not keep the discussion of the two parts entirely separate, as the table of section headings might suggest. In particular, in §29, he argues that both conditions are satisfied, though the emphasis is on the third condition. Then, in §33, the emphasis is on the second condition, though even here he says some things in favor of the third condition.

Why are the sections on the basic liberties and their priority interspersed among these sections? The answer is that the explicit requirement of the basic liberties, along with their required priority over the second principle, is important to the highly satisfactory character of the guaranteeable level as well as to the fact that the principle of average utilitarianism, which does not explicitly endorse the liberties or their priority, has possible outcomes significantly below the guaranteeable level.

Rawls's principles as the maximin option

On which grounds could it be claimed that Rawls's principles provide the maximin option in a comparison with average utilitarianism? If the parties choose the Rawlsian principles, then the outcome for their clients is sure to be no worse than that of the least-advantaged members of a Rawlsian society. In this situation, they would have equal basic liberties, their political liberties would have a fair value, they would enjoy fair equality of opportunity, and, because of the difference principle, their prospects for social and economic benefits would be as high as it is possible to make the prospects of the least advantaged. In such a situation, they are sure to be at least as well off as the position of the least advantaged might be in a utilitarian society. For one can imagine realistic social circumstances in which average utility is more effectively promoted through a denial or restriction of basic liberties to some citizens, or through departures from fair value of political liberties and fair equality of opportunity (since those principles would require resources that might be more effectivel if put to other uses), or through a system of incentives that violates the difference principle by making some worse off than they would be under equal distribution of social and economic prospects. In saying that the worse possible outcome under Rawlsian principles is better than the worse possible outcome under utilitarianism, in what sense do we mean it is better? Here we have to refer to the aims of the parties as trustees looking out for citizens' fundamental interests in developing and using their moral powers and in pursuing their determinate conception of the good. It is these aims that determine what is better and what is worse. The idea is that the Rawlsian principles more effectively protect these fundamental interests.

The third condition

Given that Rawls's principles constitute the maximin option in the first comparison, why does Rawls claim that the third condition obtains to a high degree? That is, why does Rawls claim that utilitarianism's worst outcomes are significantly below the guaranteeable level? The guaranteeable level provides rights and liberties, opportunities, powers and prerogatives, income, wealth and social bases of self respect that are fully adequate for realizing the fundamental interests of free and equal citizens. The parties know this because they know that circumstances obtain that make possible a constitutional regime in which everyone can satisfy those interests. Given that this alternative is available, the parties' role as trustees requires them to regard some of the possible outcomes associated with utilitarianism as unacceptable and intolerable (p. 100). This is especially true of limitations of political liberties and freedom of conscience. These are too fundamental to risk their loss for the sake of a chance at a larger lot of material means for satisfying their ends (p. 102).
Were the parties to gamble in that way, they would show that they did not take seriously the religious, philosophical, and moral convictions of the persons they represent. Indeed, they would show that they did not understand the nature of religious belief, or philosophical or moral conviction. This remark is not an argument: it simply calls attention to the special place of such beliefs and convictions, and to the fact that for those who affirm them, they are regarded as nonnegotiable [p. 105].

The second condition

On what grounds does Rawls claim that the second condition obtains to a high degree in the first comparison? We've just seen one reason: the guaranteeable level provides resources that effectively protect citizens' fundamental interests. In §§32-33 Rawls adds some considerations associated with the kind of social world that corresponds to the two principles. This is such an attractive social world, he thinks, that even those in the least advantaged positions can readily give their allegiance to it. This part of the argument enters into stability considerations, which are most fully treated in part IV. Nonetheless, Rawls stresses three ways in which his principles promote social union and therefore stability:
  1. By fixing once and for all the basic rights and liberties, and assigning them a special priority, the principles realistically deal with the dangers associated with the fact of pluralism. They do this by removing the basic liberties and their priority from the political agenda of political parties. It insulates them from the "calculus of social interests" thereby lowering the stakes and potential suspicion and hostility of public life, and it promotes mutual recognition of citizens on a footing of equality.
  2. The Rawlsian principles provide a clear, workable basis of public reason, especially compared to the principle of utility
  3. The two principles encourage the cooperative virtues of political life, especially by expressing an idea of reciprocity that is lacking in the principle of utility and by stressing the need to use arguments that appeal to reasons that all can freely accept.
SENSAT HOME     |     PHILOSOPHY DEPT      |     UWM         Updated February 12, 2004