:: 2/28/2003 ::

Btw, libertarian = concerned with freedom, liberal = concerned with choice. that guy = wanker, latter, no?

:: phil 00:37 [link] ::

~~~~~~~~~~~~~~~

Btw, Gibbard was one of the brightest of Rawls pupils and Barry was Rawls' major rival as "king of justice"!

:: alan 12:45 [link] ::

~~~~~~~~~~~~~~~

My script from a seminar i gave yesterday on justice. Post the session i had a lot more thoughts but here he were i got to before hand. The second session was led by libertarian guy on risk - proposing exclusively "private" solutions. He got absolutely pilloried, and i was proud to be a part of it! He went for me over my justification for the presumption of equality (founded in terms of a lack of coherent account of property rights meaning that inequalities must be justified - just having things is not justification in itself). The original article isn;t available for free but i can email anyone who fancies it the pdf.


Constructing Justice by Allan Gibbard, Philosophy and Public Affairs, Vol. 20, Issue 3 (Summer, 1991), 264-279. All quotes from this unless otherwise attributed. Quotes described as “Barry” refer to Brian Barry Treatise on Social Justice.


Brian Barry’s Treatise on Social Justice considers “two great rival theories step[ping] forward to vie for the hand of Justice”


Firstly, Justice as Mutual Advantage, described as;

“the constraints on themselves that rational self-interested people would agree to as the minimum price that has to be paid in order to obtain the cooperation of others.” Justice “is simply rational prudence pursued in contexts where the cooperation (or at least forbearance) of other people is a condition of out being able to get what we want.” (Barry pp. 6-7)

Barry notes that this is the model Plato was attempting to reject and agrees with him that this approach is somewhat flawed.

Plato thinks that justice is something that is both inherently good and good in respect of its consequences. He rejects descriptions of it as the advantage of the stronger or the good of another and even in terms of reciprocity – that is giving every man his due. He thinks it resides in harmony between the three parts of the soul, where reason rules over appetitive and spiritual concerns.


Secondly, Justice as Impartiality, is described as;

“not reducible to even a sophisticated and indirect pursuit of self-interest” instead it is “the desire to act in ways that can be defended to oneself and others without appealing to personal advantage.” (Barry pp. 7-8, 361-64)

Like Plato, Barry thinks justice is something more than a matter of social convenience but an “original principle of human nature” (p.265)


The first conception is meaningful but it lacks a full account of what Justice is Barry thinks, “True justice is Justice as Impartiality.”

Game theory provides sustenance to both camps but “in the end it fails to gain for itself a secure lackeyhood in the castle of justice”. Barry notes attempts to capture the insights game theory provides within “precise moral theories” and pays particular attention to what he calls constructivist endeavours where “The theorist specifies an ideal, hypothetical situation in which people choose the principles that govern them.” The conclusions or principles arrived at through this process are then said to be valid principles in virtue of this process.

Of course the specific account of what factors make a given situation ideal are subject to some dispute…

Barry draws a distinction between what he calls hard constructivism and soft constructivism. Examples of those using the former approach are Harsanyi and Rawls, whose concluding parties’ egoism, or self-interestedness, is veiled by ignorance, and Gauthier, whose parties are egoistic.

Barry rejects hard constructivism preferring instead his soft version in which valid principles of justice are selected from hypothetical possibilities by parties who are both egoistic, or self-interested, but not purely so – they are “reasonable”. The notion of reasonableness is a moral consideration and thus game theory has nothing to contribute here.



Gibbard feels that Barry does not give sufficient consideration to the conception of Justice as Fair Reciprocity.

I. The Motive of Fair Reciprocity;

On Barry’s account Rawls wavers between the notion of impartiality and mutual advantage but Gibbard thinks that he also considers Justice as Reciprocity as a distinct third alternative because, “

“…in a nutshell… on the one had it is distinct from Justice as Mutual Advantage because it draws on non-egoistic motives [favours are returned not always in order that one might receive further favours but also because we want to reward favours that have been granted to us – Gauthier disputes this, attributing such behaviour to straightforward prudence]. On the other hand, it is distinct from Justice as Impartiality because it says that a person cannot reasonably be asked to support a social order unless he gains from it.”


“Rawls proposes that justice is fairness in exchange, but on a grand scale: it is fairness in the terms governing a society-wide system of reciprocity.” In which citizens receive benefits and likewise contribute his motivation structured by his belief in the fairness of the system.

Whereas Justice as Impartiality is motivated by;

“the desire to act in accordance with principles that could not reasonably be rejected by people seeking an agreement with others under conditions free from morally irrelevant bargaining advantages and disadvantages.”(p. 8)

Gibbard suggests that there is little difference between this view and fair reciprocity, “Perhaps fair terms of reciprocity are whatever terms every reasonable person would find acceptable.”

Barry rejects the aspects of Rawls that characterise Justice as Impartiality as a type of Justice as Fair Reciprocity.
Barry and Rawls both reject “the justice of inequalities based on morally arbitrary advantages” but Barry takes this to mean that Rawls’ view of society as a “cooperative venture for mutual advantage” is mistaken since any bargaining outcome is skewed by the advantages and disadvantages of its participants – the outcome is consequently unjust.

Rawls‘ difference principle allows some inequalities to exist, subject to there being in the best interests of the worst off, and Barry and Rawls concur that not all inequalities, though they are ultimately derived from morally arbitrary causes, are necessarily unjust.

[Consider Locke’s account of property rights, and the entitlement “labour mixing” gives an individual to the fruit of the toil where there is “enough and as good” for others]

“The central issue in any theory of justice is the defensibility of unequal relations between people” (Barry p. 3)… “If it is just that unequal endowments lead to unequal outcomes, that must be because the social structure is just, we cannot just assume at the outset that unequal endowments give rise to unequal desert or entitlement.”

Gibbard thinks that this claim stands in need of justification perhaps Rawls is asking not so much “Why expect inequality?” but instead “Why limit myself in pursuit of my own advantage?” to which he, in effect answers;

“You have what you have only because others constrain themselves, in ways that make for a fair cooperative venture for mutual advantage. Constrain yourselves by those rules in return, and you give them fair return for what they give you.”

However, a participants response to this will depend on his feelings about fair reciprocity. Gibbard’s example of participants living solo on islands which are either fertile or not – it may be a matter of luck but it does not follow that individuals should share since in this circumstance no-one has either cooperated or restrained themselves.

[whilst this makes a point that conceptions of justice as fair reciprocity/mutual advantage may be incomplete, I think its fair to say that the example is not analogous since the participants are outside the circumstances of justice i.e. they are not cooperating in circumstances of scarcity)


II. The Content of Fair Reciprocity

Is the difference principle acceptable to those who are motivated by the morality of Fair Reciprocity, within a society conceived as a venture for mutual advantage?

Barry thinks not, because he feels fair reciprocity would require mutually advantageous outcomes to be advantageous to an equal degree – in other words that outcomes should or would reflect (morally arbitrary) distinctions in participants initial bargaining positions (where they are at the “disagreement point”).

In other words “Introducing the requirement of mutual advantage threatens to unravel Rawls’s theory.” (Barry p. 249)

Gauthier’s disagreement point is one which apparently ignores lawless egoism, prompting Gibbard to ponder;

“Why should anyone starve peaceable just because he would do badly in a Lockean world with magic police? [Gauthier seems to imagine market constraints will be self regulating – no conflict will arise from rule enforcement]

“Gauthier awards the person who gains least the most possible gain” where gain is the entire surplus derived from cooperation. Gibbard points out that were Gauthier’s disagreement point to be “one of general egoism” then given that everyone is roughly equally in terms of primary goods at the disagreement point then the maximum gain available is also equal and consequently, given broadly similar utility curves, Gauthier’s approach produces Rawls’ difference principle – Gauthier’s maximin approach to utility gains becomes a maximin distribution of prospects iro primary goods.

Also problematic is, as Allen Buchanan (Justice as Reciprocity) points out, the risk of exclusion.

“It is not only cooperation that is at stake in questions of justice, but mutual restraint: nonaggression and respect for a system of property rights… To claim Justice as Fair Reciprocity, one must offer fair reciprocity.”

Two other concerns Barry identifies are;

1. groups incapable of voluntary restraint – babies can be tortured to death without violating demands of fair reciprocity.

2. Slavery – people are excluded not because they are useless but because they can be controlled.

Rawls denies these exclusions but is it possible to do so within Justice as Fair reciprocity?



III. Justice as Mutual Advantage

Barry adapts this position to one Gibbard calls Justice as Ideal Bargaining Outcome, where egoistic concerns of participants are replaced by “what ideally rational bargainers would finish up with in any given situation.” (Barry p.11)

However, Thomas Schelling (The Strategy of Conflict, 1960) objects that actual bargaining depends on “salience”, that is the “subjective prominence of that outcome” for participants. “Salience lends itself poorly to fruitful idealisation, and if we take salience away, nothing is left to determine the outcome of bargaining.“

[This strikes me as similar to Rawls participants in the original position – idealising according to the “thin theory of the good”]

There is also an objection iro the fact that bargains take place multilaterally – there are multiple participants and thus coalitions may form;

“We cannot fruitfully ask what would happen with an ideal hypothetical bargain that abstracts away from the contingencies of history.” (p.274)

and these coalitions will depend on “salient aspects of group identity.” which are ignored by game theory – skin colour, dialect, kinship, ritual etc.

“Once we have despaired of anything more inspiring, a strategy with great appeal is to find what is systematic in the waste-preventing aspects of egoistic interactions, and label it justice.”

“Justice as Ideal Bargaining Outcome is responsive to bargaining power” and what exactly this is remains unclear expect that it is related to the “possible combinations of utilities the bargainers can obtain” – in something like the Nash model, but in a way that is “not easily idealised.” perhaps because of the uncertainties involved.

[the example of the rich man and the poor man bargaining]

In summary, “Justice as Ideal Bargaining Outcome cannot be made coherent and convincing. Too much in actual bargaining depends on features of the situation that are not subject to standard game-theoretic idealisations.”



IV. Justice as Impartiality

“Rawls thinks that behind [the] veil of ignorance, rational self-interested choosers would act with maximum risk aversion.” (p. 276) Barry is in favour of the difference principle but rejects the original position argument for it. Like John Harsanyi he thinks the “test of self-interest from behind the veil of ignorance yields utilitarianism.”

Barry is more interested in an argument Rawls gives as auxiliary to the rather dramatic original position argument in Chapter 2 of A Theory of Justice where he “pushes the rationale for equality of opportunity to the limit, and then allows a move to Pareto efficiency. “ posing the question “Should we demand equality of opportunity and push this demand to the limit?”

Given that this could be rather costly to implement Gibbard suggests that;

“A requirement of strict equality might be waived when it is to everyone’s advantage to do so; this seems entirely reasonable. I argued above, however, that when a difference between people is morally irrelevant, what follows is not that this difference cannot be allowed to influence outcomes. What follows is that the social structure cannot be justified on the grounds that it makes outcomes depend on this property. If, though, the social structure can be justified in some other way, the fact that, under it, outcomes depend on morally irrelevant contingencies is no objection.”

Barry then returns to his constructivist distinction. A theory is constructivist if;

“What comes out of a certain kind of situation is to count as just.” and there is no standard of justice independent of this test. (Barry p. 266) In addition the test need not actually occur, it may be hypothetical.

Barry identifies his own conception as soft constructivism in that participants have a “capacity and preparedness to be moved by moral considerations” (Barry p.350)

As Gibbard puts it “They have human decency, with an emphasis on human.”

Whereas Rawls had aimed to construct an all encompassing moral model Barry thinks this is not possible – that we cannot “hope for anything like a deductive proof.” (Barry p.345)

The problem he is left with is then what soft constructivism has to offer considering the acid test of what sort of agreements might be reasonably rejected – a term which Barry thinks cannot be defined but must be intuitively grasped. But he maintains that his approach amounts to more than simply substituting the vagueness of what can be considered reasonable for the vagueness of what can be considered justice since;

“It encourages – indeed forces – one to make a distinction between what one would personally support and what one believes could not be reasonably rejected.” (Barry p. 352)

But Gibbard is left wondering;

“Argument stops somewhere, to be sure, but I hope it does not have to stop with brute intuitions of what is reasonable and what is not. If someone tells me a demand is reasonable, I want to ask why.”







Contractualism, from Kant to Rawls (cat imp to veil of ignorance)
Contractualism from Hobbes to Gauthier (rational decision theory, in egoistic agents interests to become “constrained maximiser” – someone who respects the interests of others, argument not from notions of morality but from purely self interested motives)

T.J.Scanlon, acceptable rules those that could not be “reasonably rejected” by individuals seeking unforced cooperation. What is reasonable associated with notions of fundamental human rights.

:: alan 12:44 [link] ::

~~~~~~~~~~~~~~~