Read Gauthier and Constructing Morality from Non-Moral Sources text version

Gauthier and Constructing Morality from Non-Moral Sources1 Robert Wavre University of York - Email: [email protected]

Abstract Constructivism is a major force in modern moral and political philosophy. It can be divided into two major schools of thought. One attempts to construct a moral system for society from already moralised grounds. The other, which I concentrate on in this paper, attempts the more ambitious project of constructing from non-moral sources. I examine this school's leading proponent, David Gauthier, and argue that his use of the contract metaphor invalidates his project as the bargaining situation is moralised. Indeed, I argue, there seems little reason for optimism that this problem could be overcome by anyone. It seems improbable that non-moral rationality will be of much use in specifying the "correct" terms of a hypothetical bargain. It appears the constructivist from non-moral sources would do well to leave the contract metaphor behind.


I would like to thank Matt Matravers, and Sue Mendus, for extremely useful criticism and advice on earlier versions of this paper. I would also like to thank participants at the Graduate Theory Workshop, University of York, Spring 2000, and the Graduate Political Theory Conference, University of Warwick, May 2000 for their feedback on versions of this paper.


This paper is about the role of the contract in moral constructivism in general and its role in Gauthier's work in particular. The paper is divided into three parts. The first is a brief examination of the role the contract plays in Rawls' work. The second is a more detailed look at the role it plays in Gauthier's work. I look at both to clarify the distinction I draw, following Matravers, between constructivism and faux constructivism2. Constructivism refers to the attempt to construct a moral system from the ground up, without appeal to any moral premisses; a project in which we can situate the majority of Gauthier's work. Faux constructivism refers to the use of the contract, most often considered part of the constructivist's arsenal, for a different project. One where the contractarian device is used to expand on, or clarify, moral premisses. This, I take it, is the way Rawls uses the contract. In the second section I intend to argue that the way Gauthier uses the contract smuggles in moral premisses. Thus, it performs much the same role as it does in Rawls' work. This is fatal for Gauthier's argument as, whilst a moralised contract may be suitable for a faux constructivist, it is clearly unworkable for anyone trying to construct morality from non-moral sources. In the third section, I look at the possibility of any contract being created which does not rely on moral premisses of some sort. I conclude that there seems little hope of any such contract being devised, and that the constructivist, as opposed to the faux constructivist, would do well to leave the contract metaphor behind.

Rawls and Faux Constructivism

I turn first, very briefly, to the faux constructivism of Rawls. Rawls uses his contract metaphor to expand and clarify a moral position. His contract depicts bargainers deciding on


The expression "faux constructivism" is taken from Matraver's (2000).


a moral system for a society from the original position, which is situated behind a `veil of ignorance'. (Rawls, 1971:12) As is well known, this veil of ignorance means that none of the bargainers is to know, "his place in society, his class position or social status..[nor] his intelligence, strength, and the like." (Rawls, 1971:12) The hypothetical bargainers then deliberate and choose Rawls' preferred theory of justice. However, Rawls neither assumes that this carries all the justificatory weight in the argument, nor that the stipulation of the bargaining position is free of moral influences. For Rawls, the original position is "understood as a purely hypothetical situation characterized so as to lead to a certain conception of justice." (Rawls, 1971:12) The hypothetical situation captures restrictions that it is reasonable to impose when reasoning about justice (Rawls, 1971:12-14). The basic idea is that we ought to move between our considered moral convictions, for instance that slavery is wrong, and the premisses and implications of our moral theory, until the two cohere with each other, or are in equilibrium. The idea being that the theory and the considered moral judgements can be mutually supportive of one another. It is as a role in this process that Rawls uses the contract metaphor. "By going back and forth, sometimes altering the conditions of the contractual circumstances, at others withdrawing our judgements and conforming them to principles, I assume that eventually we shall find a description of the initial situation that both expresses reasonable conditions and yields principles which match our considered judgements duly pruned and adjusted. This state of affairs I call reflective equilibrium." (Rawls, 1971:4)

Thus the contract has the role of systematising our considered judgements by expressing our moral convictions and testing particular intuitions against more general moral principles. It is an aid in a form of moral reflection that moves from moral starting points, rather than an 3

attempt to justify morality from non-moral premisses. The bargaining position is thus overtly built up from moral considerations. This fits it well for the task that faux constructivism gives it; the task of aiding in the explication of crucial moral premisses.

However, the contract has long been associated with the more radical constructivist project of building morality from non-moral premisses. In the next section I turn to Gauthier's work.

Gauthier's Contract

Gauthier argues that morality "can be generated as a rational constraint from the non-moral premisses of rational choice." (Gauthier, 1986:4) In this reasoning "our concern ... is to do this without incorporating into the premisses of our argument any of the moral conceptions that emerge in our conclusions." (Gauthier, 1986:6)

There are two main elements to Gauthier's project. The first is to show that it could be rational, in certain circumstances, for someone who is not already motivated by morality to choose to acquire that motivation. That is to say the idea of rational, amoral, actors agreeing to moral rules that constrain their behaviour is a real possibility. The second, is the working out of exactly to which rules these rational bargainers would agree. The first element is seen in Gauthier's argument for the transition from straightforward maximisation to constrained maximisation.

Gauthier identifies straightforward maximisation with Hobbes' Foole. The Foole would be willing to break his word whenever he judges it in his best interest. So, if you paid a 4

straightforward maximiser for a job in advance, he would not do the work (assuming he did not want to win a future contract from you). This work would be a cost for which he would receive no extra benefit. Gauthier starts from this position because, as he says in his essay "Reason and Maximisation", "in our society the received conception [of rationality]...does identify rationality with maximizing activity. And this identification is usually expressed by the condition of straightforward maximisation." (Gauthier, 1990:231) However, he argues, rationality would be better expressed through the idea of constrained maximisation. A constrained maximiser is, in Gauthier's words:

i) Someone who is conditionally disposed to base her actions on a joint strategy or practice should the utility she expects were everyone so to base his action, be no less than what she would expect were everyone to employ individual strategies, and approach what she would expect from the cooperative outcome determined by minimax relative concession

ii) Someone who actually acts on the conditional disposition should her expected utility be greater than what she would expect were everyone to employ individual strategies. (Gauthier, 1986:167)

We need not worry about minimax relative concession here. All that need be noted is that it is the conclusion of the bargain to which the constrained maximisers are supposed to agree. For the purposes of this paper I also intend to accept that a move to a form of constrained maximisation, whereby the actor may constrain his later interactions on the basis of an agreement made with other constrained maximisers, is possible. I am solely interested in the 5

constraints imposed on the bargainers, in Gauthier's bargaining position, that lead them to accept his bargaining solution. Before moving on to this, however, there is one other aspect of constrained maximisation that we need to be aware of. Namely, Gauthier's argument that constrained maximisers would be narrow rather than broad compliers.

Gauthier defines narrow and broad compliance as follows:

Let us call a person who is disposed to cooperate in ways that, followed by all, yield nearly optimal and fair outcomes, narrowly compliant. And let us call a person who is disposed to cooperate in ways that, followed by all, merely yield her some benefit in relation to universal non-cooperation, broadly compliant. (Gauthier, 1986:178 - original emphasis)

Broad compliers, Gauthier claims, will be taken advantage of, given translucence, as others will have a reason to give them terms that are only slightly better than if there were no bargain. Whilst:

the narrowly compliant person is always prepared to accept cooperative arrangements based on the principle of minimax relative concession, she is prepared to be cooperative whenever cooperation can be mutually beneficial on terms equally rational and fair to all. In refusing other terms she does not diminish her prospects for cooperation with other rational persons, and she ensures that those not disposed to fair cooperation do not enjoy the benefits of any cooperation, thus making their unfairness costly to themselves, and so 6

irrational. (Gauthier, 1986:178-9)

So, it is rational to be a narrow complier. Note that narrow compliance is defined in relation to minimax and an idea of fairness. This will be problematic if the argument for narrow compliance is used as a premiss of the bargain that leads to the minimax bargaining solution. For it is the bargaining solution that is supposed to provide an account of fairness.

Now let us turn to Gauthier's contract. The question for Gauthier is to which moral system of cooperation would constrained maximisers' agree, if they were bargaining with other constrained maximisers? Clearly how the interaction is structured is the key point. With different starting points and different abilities attributed to the actors, different moral structures would be accepted. The actors are assumed to have equal rationality, but otherwise to have whatever abilities nature gave them. However, the situation they bargain from is subject to some constraints. Due to limited space I intend here to examine only one of these, the Lockean proviso.

This proviso "prohibits bettering one's situation through interaction that worsens the situation of another...The base point for determining how I affect you, in terms of bettering or worsening your situation, is determined by the outcome you would expect in my absence." (Gauthier, 1986:204-5) What is of interest is that Gauthier uses the proviso to generate a noncoerced baseline. However, I shall argue that this move is very problematic.

Given rational maximisers bargaining with others to increase their utility one would think that they would use whatever methods were available to increase their share of the resource 7

distribution. One of these methods would seem to be coercion; certainly it does not take a detailed study of history to see this conjecture gain empirical support. So why does Gauthier believe that fully rational actors would eschew such unpleasantness, and bargain from a baseline that excludes coercion? One way into this question is to examine Gauthier's example of the masters and slaves (Gauthier, 1986:190-191). Here I should stress that I am looking at this for what it tells us about the structure of the bargainers' interactions, not as an argument about motivation and the rationality of constrained maximisation. In the story a set of masters became disillusioned with the quality of their slaves' work, due to the slaves' lack of motivation, and the overall costs of repression. One day a young master came forward with a theory taken from one of his wise university professors, who had propounded the theory of Morals by Agreement. He argued that they could make a rational deal with the slaves whereby the slaves would become paid servants (not well-paid of course) doing much the same job as they were, and that it would be rational for these new servants to comply with the bargain. The masters would thus remove the costs of coercion, whilst leaving themselves with willing workers, and the slaves would be freer and less repressed. The slaves accept the deal, but, once the means of repression have been dismantled, renege. (Gauthier, 1986:190191) As one of the slaves said:

What those [masters] never understood was that this bargain was coercively based. It was only because of the power they held over us that it seemed a rational deal. Once that power was taken away, it became obvious that the fruits of cooperation weren't being divided up in accordance with that fancy principle of minimax relative concession. And so there wasn't any reason to become willing servants. (Gauthier, 1986:191) 8

This, Gauthier believes, gives us a reason to agree that broad compliance is irrational, as we want to benefit fully from cooperation, not partially. So a deal that is not based on minimax3 will be unstable (granted rational actors).

Does the example show this? What the example actually shows is a society where the masters were doing really well for themselves until they were irrational enough to sacrifice their power, without guaranteeing compliance with the bargain. Bargaining strength is not determined by what you would bring to the table in counterfactual situations where what you have accumulated is that the acquirement of which did not violate the Lockean proviso. Your bargaining strength is determined by what power you have here and now. If so, then the link between rationality and Gauthier's bargaining solution is severed. This conclusion is supported by Barry who claims that, "Gauthier would like to say that slavery could be opposed on his own premisses. But unless slave owners were making a mistake about where their interests lay, there is no room for criticism." (Barry, 1998:220-221) Moreover, Gauthier seems to admit the possibility of such cases when he discusses technological differences between groups. He says that,"a superior technology enables its possessors rationally to maintain, and requires others rationally to acquiesce in, arrangements that rest on differential rights in clear violation of the proviso." (Gauthier, 1986:231) So whether rationality and his moral system tie in together will be a contingent matter depending on the power distributions of the groups and individuals involved. This is not necessarily problematic for Gauthier. He would accept that in certain circumstances it may not be rational to be moral. However, it would be problematic for Gauthier if it was found to be rational to constrain our actions on


Or another `fair' distribution of some sort (as Gauthier no longer accepts minimax).


entirely different grounds than those towards which Gauthier points. For instance, if the bargain was based on coercion and still led to constrained maximisation. Different rational pacts will be formed from many different starting positions. Could more than one pact share the same formal properties of being a system of constraint on interactions, with which all involved have reason to acquire a disposition to comply? Specifically, could coerced bargains give rational actors a reason to constrain their maximisation with the bargains to which they agree? Could it be rational for these rational actors to keep to an agreement made under duress, even if that coercive power is removed at a later date? If so, what would it be that makes bargains formed from non-coerced baselines suitable for Gauthier's constructivist project, and the rest not?

It could be that, in the case of the masters and the slaves, the original coerced position did not produce a bargain, in Gauthier's sense. Morals can only arise from rational bargains, and it would not have been rational for the slaves honestly to make, and to comply with, the bargain that they did. Even if they would have been better off, the idea of narrow compliance is meant to show that their compliance would have been irrational. Therefore, coercion meant that no rational deal could have been struck. However, this is not true. People are meant to choose to be narrowly compliant, and therefore not accept unfair bargains, because otherwise others will see that they can abuse their broadly compliant disposition by offering unfair terms in the future (Gauthier, 1986:226) So, if we accept a deal from a coerced baseline we could be opening ourselves up to such coercion in the future. However, the deal does bring other factors to the table that affect what the `fair' distribution is, such as our intelligence and strength (of the non potentially coercive variety). Accepting these, will, of course, also affect future bargains. 10

Let us compare two possible bargains, one which is based on coercion, and the other on intelligence, or strength. Let us assume that the payoff in utility terms for the actors in the two scenarios is the same. With the stronger bargainers in both situations doing equally well, and the weaker bargainers doing equally badly. Gauthier wants to delineate a distinction between legitimate and illegitimate aspects of bargaining strength. Where factors such as intelligence and strength are legitimate, whilst the use of coercive power is illegitimate. Whilst Gauthier believes that illegitimate bargaining strength can be excluded on the grounds of rationality alone, I wish to show that his account of rationality cannot do the necessary work.

Gauthier's attempt to give a rational basis to the exclusion of `illegitimate' forms of bargaining strength relies on the argument that the second, non-coercive, bargain will be stable while the first, coercive, bargain will not. However, I argue, the prudential reasons to reach the bargain, and keep to it, are identical in the two cases. The first instance, if we apply it to the masters and slaves example, shows that the slaves would gain from making the bargain. However, according to Gauthier, they would not keep the bargain because it was based in coercion. Yet, on Gauthier's own account of the reasons to move from being a straightforward maximiser to a constrained maximiser, there is a problem. Constrained maximisation is rational only given translucence. If we apply this idea to the potential bargaining situation in the case of the masters and the slaves, the young master who had been taught Morals by Agreement, should have stressed to the elders that the deal with the slaves should only be made if they could be sure that the slaves are disposed to comply with the bargain. The slaves, when debating whether they should be disposed to comply with the 11

coerced bargain, would take into account that they would run the risk of coercion in the future. This would mean future agreements may see them making proportionally smaller gains than if the other party had no coercive potential. However, a bargain based on intelligence would have the same problem. If the stupid accept that what is brought to the bargaining table affects distribution, they will do worse in future bargains than if they rejected the idea that what they bring should have an effect. Both face a similar situation, and both have exactly the same reasons to develop a disposition to comply. Therefore, the slaves would be rational to make the bargain and, given translucence and equal rationality for both the slaves and the masters, would be rational to acquire a disposition to comply with that agreement.4

My claim is that there is no significant difference between a bargain based on coercion, and one based on intelligence. So it must be some other idea that marks the difference between them. It is a reasonable conjecture that what is doing the work is a moral commitment to the impermissibility of coercion. What else could it be that tells us that coercion should not affect the distribution?

This conjecture is given further support by Barry's argument. Rather than an internal criticism of Gauthier's theory, he points to the historical stability of systems of coercion. He believes that at times Gauthier argues that if everyone believed in his bargaining conclusions then "the only kind of society that could be stable would be one that conformed to it."(Barry, 1998:220) As Barry points out, we can say the same for any theory of morality or justice.


The master-slave example may appear extreme, but given what Gauthier says about technological differences (Gauthier, 1986:231), we can see that such coercive considerations could affect many, if not most, bargains struck in the real world.


Barry quotes Gauthier and replies to him as follows:

Gauthier says "those who believe, correctly, in justice as mutual advantage, will recognise that the practices and institutions of their society do not satisfy the criterion of its purposes and will not take themselves to have any reason to conform to the rules of their that afforded by their coercive imposition. (Gauthier, 1986:124) If we reject Gauthier's own transparently circular line of argument, I can see no rationale for saying that utility maximisers must eschew coercion....Over the course of history, slavery and the systematic oppression of one racial or ethnic group by another have proved stable over long periods of time. (Barry, 1998:220)

So, we see that Barry accuses Gauthier at the practical level of what I have accused him of at the theoretical level. The idea of fairness or morality that he defends must be accepted by most of those affected, before the defence of that same fairness or morality can be mounted.

So, insofar as Gauthier defends his moral theory thus, we see that undefended moral assumptions about impartiality and fairness; along with assumptions of equal rationality, and broadly equal starting points, underlie the project. The constraints are not derived from rationality at all. At a guess, it is the moral idea of proportionality to contribution that underlies the proviso.



Gauthier's project therefore fails. We have seen that he uses the conclusions of the bargain, which give us fairness and the idea of minimax relative concession, to justify the constraints on the bargainers. Thus the conclusions are used as premisses. Narrow compliance is defined in terms of compliance with fair agreements, and is a constraint on the bargain that produces an idea of fairness.

Gauthier is one of the best, and certainly one of the most famous, contemporary constructivists. Yet he has been shown to be unable to combine his constructivism with his contractarian methodology. Rather, the hypothetical contract has been shown to fit far more comfortably with the faux constructivist project. While it is not part of my project here to prove the impossibility of combining genuine constructivism with contractarianism, it seems worthwhile to point to a reason for pessimism. The hypothetical bargain is problematic, for the genuine constructivist, as what hypothetical people will agree to is determined by their philosophical creators. Obviously, the specification of the bargaining situation affects the outcome. We must ask how rationality alone could select the `correct' specification of a fictional bargain? Other, presumably moral, grounds must come into play if the selection is not arbitrary. For instance, in both Rawls' and Harsanyi's work the bargaining situation is very similar. However, even subtle differences have a huge impact. Are the bargainer's risk averse? If so we shall reach the two principles of justice. Are the bargainers willing to take risks? If so we shall accept the average utilitarianism of Harsanyi. As the bargainers are hypothetical people how can one choose between the two characterisations, except on the grounds of which resultant moral theory one prefers?

Of course, for the Rawlsian faux constructivist this is precisely the point. The idea of the 14

contract is to model our moral convictions to aid reflection and the search for reflective equilibrium. However, for the genuine constructivist, like Gauthier, this is not an option as he aims to ground morality in non-moral premisses and thus cannot use morality as a constraint on the contract, if the contract is to do any work.

Thus, I conclude, the contract may be suitable for faux constructivism, but it would appear incompatible with genuine constructivism. If we think the constructivist project is an interesting one then we would be well advised to look for a device other than the (hypothetical) contract to further the project5.


By this I do not mean to imply that I believe a genuine, historical, contract would give us more grounds for optimism.



Barry, B., Theories of Justice, Hemel Hempstead: Harvester Wheatsheaf, 1989 Barry, B., Justice as Impartiality, Oxford: Clarendon Press, 1995 Barry, B., `A Reply', in Kelly, P., (ed.) Impartiality, Neutrality and Justice, Trowbridge: Edinburgh University Press, 1998 Copp, D., `Contractarianism and Moral Skepticism', in Vallentyne, P., (ed.) Contractarianism and Rational Choice, New York: Cambridge University Press, 1991

Daniels, N., Justice and Justification: Reflective Equilibrium in Theory and Practice, New York: Cambridge University Press, 1996 Gauthier, D., `Justice as Social Choice', in Morality, Reason and Truth, Copp, D., & Zimmerman, D., (eds.) New Jersey: Otowa, 1984 Gauthier, D., `Morality, Rational Choice, and Semantic Representation', in Paul, Miller, Paul, & Ahrens (eds.), The New Social Contract, Oxford: Blackwell, 1988 Gauthier, D., Morals by Agreement, New York: Oxford University Press, 1986 Gauthier, D., `Mutual Advantage and Impartiality', in Kelly, P., ed., Impartiality, Neutrality and Justice, Trowbridge: Edinburgh University Press, 1998 Gauthier, D., `Political Contractarianism', Journal of Political Philosophy, vol. 5,2, 1997 p.132-48 Gauthier, D., `Uniting Separate Persons', in Sugden & Gauthier, (eds.), Rationality, Justice and the Social Contract, Harvester Wheatsheaf, 1993 Gauthier, D., `Assure and Threaten', Ethics, vol. 104(4), 1994, p. 690-721 Gauthier, D., `Why Contractarianism?', in Vallentyne, P., (ed.), ibid., 1991 Gauthier, D., `The Best of Times (Universality, Individuality, and Democracy), Political Theory Workshop, York, 2000 Goodin, R.E., `Equal Rationality and Equal Endowments', in Sugden & Gauthier, ibid., 1993 Hampton, J., `Equalizing Concessions in the Pursuit of Justice: A Discussion of Gauthier's Bargaining Solution', in Vallentyne, ibid, 1991 Hobbes, T., Leviathan, Avon: Cambridge University Press, 1991 Hollis, M., `The Agriculture of the Mind', in Sugden & Gauthier, ibid, 1993 Kymlicka, W., `The Social Contract Tradition', in Singer, P. (ed.), A Companion to Ethics,


Oxford: Blackwell, 1993 Matravers, M., Justice and Punishment: The Rationale of Coercion, Oxford: Oxford University Press, 2000 Moore, M., Foundations of Liberalism, Oxford: Oxford University Press, 1993 Sayre-Mcord, `Deception and Reasons to be Moral', in Vallentyne, 1991, ibid., 76-97 Scanlon, T.M., `Contractualism and Utilitarianism', in Sen, A., & Williams, B., (eds.) Utilitarianism and Beyond, New York: Cambridge University Press, 1982 Scanlon, T.M., What we Owe to Each Other, Cambridge, Massachusetts: Bellknapp Press of Harvard University Press, 1998 Smith, H., `Deriving Morality From Rationality', in Vallentyne, ibid, 1991 Vallentyne, P., `Gauthier's Three Projects', in Vallentyne, 1991, ibid., 1-11 Vallentyne, P., `Contractarianism and the Assumption of Mutual Unconcern', in Vallentyne, 1991, ibid., 71-75 Velleman, J.D., `Deciding How to Decide', in Ethics and Practical Reason, New York: Oxford University Press, 1997



Gauthier and Constructing Morality from Non-Moral Sources

17 pages

Find more like this

Report File (DMCA)

Our content is added by our users. We aim to remove reported files within 1 working day. Please use this link to notify us:

Report this file as copyright or inappropriate


You might also be interested in

Gauthier and Constructing Morality from Non-Moral Sources