The asymmetry objection to political liberalism: evaluation of a defence

This paper evaluates Jonathan Quong’s attempt to defend a version of political liberalism from the asymmetry objection. I object that Quong’s defence relies on a premise that has not been adequately supported and does not look as if it can be given adequate support.


Introduction
In Political Liberalism, John Rawls proposes that the constitution in a liberal society should be justifiable to all reasonable citizens. More precisely, he proposes that the constitution should have a part called the essentials and it is the essentials which should be justifiable to all reasonable citizens (1993: 137). For convenience of expression, I shall write as if, for Rawls, it is the entire constitution which should be justifiable to all reasonable citizens. There are, of course, objections to Rawls's thinking. The purpose of this paper is to examine a defence of Rawls against one particular objection, known as the asymmetry objection. But before coming to the objection or the defence, it is necessary to introduce some clarifications.

Clarifications and objection
Let us start with the question, who are reasonable citizens? Rawls does not use "reasonable" in the ordinary sense of the word. On the interpretation of Rawls that I shall rely on, reasonable citizens in a liberal society are defined by having two beliefs (1993: 54;Quong 2011: 37). The first belief is that this society should be a fair system of social cooperation arranged for mutual benefit between free and equal people. But how are people equal because it does not seem that they have equal abilities? That they are even equal in moral abilities has been contested (Williams 1973: 234). Rawls's response is that adult members of society, except for some adults suffering from extreme psychological disorders, have a capacity for a sense of justice and a capacity for a conception of the goodfor forming a conception of what kind of life would be goodand that the presence of these two capacities makes people worthy of equal treatment (1993: 19). This statement of Rawls's thinking can be subject to further elaboration, but we can leave aside the task of elaborating it here.
The second belief which defines reasonable citizens is usually stated by relying on Rawls's specialist terminology. Using this terminology, the belief is that there is such a thing as the burdens of judgment, which gives rise to the fact of reasonable pluralism in a liberal society. The burdens of judgment are six obstacles to achieving agreement on moral, religious and metaphysical questions (1993: 56-57): (a) empirical and scientific evidence may be complex and conflicting; (b) people may disagree about the relative weight that different considerations should carry; (c) all concepts are to some extent inherently vague and subject to hard cases; (d) the way in which we assess moral and political values is inevitably shaped to some degree by our total life experience; (e) there are often different kinds of normative considerations on both sides of a question which fully rational people may not agree how to place; (f) social institutions are limited in the number of values they can incorporate, which will sometimes necessitate difficult choices.
Owing to these obstacles, Rawls thinks that in a liberal society there will inevitably be disagreement among reasonable citizens over moral, religious and metaphysical questions. Consensus is perhaps possible through oppression, but it would not occur in a liberal society, given the freedoms this kind of society grants citizens and the burdens of judgment (1993: 4).
To accept this point is to accept the fact of reasonable pluralism.
The constitution in a liberal society is supposed to be justifiable to all reasonable citizens. In other words, for any commitment of the constitution, one can say to a reasonable citizen, "Given that you have the two beliefs identified above, you should also accept this commitment, because it can be rationally inferred from these two beliefs." Rawls thinks that, starting from the two beliefs, we can justify commitments about social justice. These commitments specify what should and should not be done because otherwise the distribution of rights, duties and the advantages of social cooperation would be unfair. For example, slavery is prohibited because it is unfair for anyone to be a slave. The constitution, however, cannot explicitly include commitments about other things or be justified on the basis of such commitments.
Rawlsians regard commitments about these other things as commitments which some reasonable citizens are bound to reject. For example, even if some reasonable citizens believe the metaphysical doctrine that reality consists of minds and their ideas, others will not. In the literature on Rawls's political liberalism, the main focus is not on metaphysical disagreements. Rather it is on disagreements about the goodabout which kinds of life would be good and which kinds less good (Waldron 1999;Chan 2000: 21-22;Quong 2011). For example, if a parent judges that it would be good if their child becomes a doctor, a lawyer or an engineer, but not so good if their child becomes a writer of vampire novels, some reasonable citizens might agree, but others are sure to disagree.
However, cannot reasonable citizens also disagree about what social justice requires? What social justice requires can be subject to reasonable disagreement as much as the commitments which Rawls excludes from the constitution, so there are no grounds for treating commitments about social justice differently from other commitments in the way that Rawls does. To treat them in the same way is to treat them symmetrically, we can metaphorically say, hence this objection is referred to as the asymmetry objection. It says that Rawls lacks grounds for treating them asymmetrically. Jonathan Quong has offered a defence of Rawls against this objection and it is his defence that I aim to evaluate.

Quong's defence
Quong's defence against the asymmetry objection concedes that reasonable citizens may disagree about the requirements of social justice. But he thinks that these disagreements can be resolved by appealing to the beliefs shared by such people, beliefs without which they would not qualify as reasonable citizens by Rawls's definition (2011: 204). If two positions are involved in a disagreement about social justice and one position is incompatible with these beliefs, the incompatible one is mistaken. If both are incompatible, then both are mistaken.
(What if both are compatible? The process of resolution can be more complicated, involving voting and allowing for the state to act on one of the positions without showing the other to be mistaken. I am using the word "resolution" to cover this situation as well, because a decision is made regarding which position to act on and all reasonable citizens should abide by this decision. The nature of my objections means that more complex processes of resolution can be overlooked.) However, when reasonable citizens disagree about the good, there is a significant chance that there are no shared beliefs which provide a way of resolving the disagreement, because the beliefs which all reasonable citizens share will not enable a resolution. Both positions in the disagreement are compatible with these beliefs, in which case the beliefs do not reveal which position is mistaken. Furthermore, they do not enable any more complex process of resolution. If no other commitments enable a resolution, Quong characterizes the disagreement as foundational: Foundational disagreements are disagreements characterized by the fact that the participants do not share any premises which can serve as a mutually acceptable standard of justification… Reasonable disagreements about the good life are not necessarily justificatory and will almost certainly be foundational. (2011: 204) Quong thinks that the constitution of a liberal society can be based on judgments that reasonable citizens disagree on, but only if the two beliefs which reasonable citizens share provide a way of resolving these disagreements. Judgments about social justice supposedly meet this requirement, but judgments about the good supposedly do not.
Quong's argument for not appealing to judgments about the good can be reconstructed as follows: (1) If a reasonable citizen makes a judgment and it is possible for another reasonable citizen to disagree with that judgment and for the disagreement to be a foundational disagreement, then the constitution of a liberal society should not be based on that judgment.
(2) For any* judgment that a reasonable citizen makes about the goodabout which kinds of life, or pursuits in life, are good and which kinds less goodit is possible for another reasonable citizen to disagree with that judgment and for the disagreement to be a foundational disagreement. Therefore: (3) For any* judgment that a reasonable citizen makes about the good, the constitution of a liberal democracy should not be based on that judgment.
Next to "any" in (2) and (3) I have placed a "*" to indicate that Quong allows for a narrow set of exceptions: a narrow set of judgments about the good that all reasonable citizens should be committed to, in virtue of their reasonableness. This narrow set consists of judgments about the good that are obvious implications of Rawls's political liberalism, such as that a life of relating to fellow liberal citizens on fair terms is better than a life of not doing so. Henceforth the "*" symbol will be used to indicate that there is this narrow set of exceptions.

An inadequately supported premise
My objections to Quong's defence concern premise (2) of the argument from him that I have reconstructed: the premise that for any* judgment that a reasonable citizen makes about the goodabout which kinds of life, or pursuits in life, are good and which kinds less goodit is possible for another reasonable citizen to disagree with that judgment and for the disagreement to be a foundational disagreement. Why accept this premise?
The premise is not self-evident, and it is not clear from Quong's article on the asymmetry objection or his book defending Rawls what exactly the argument for it is. As far as I can see, the relevant section of his article is just unsupported assertions (2005: 313-314), as is the relevant section of his book (2011: 206-207). At present, premise (2) is inadequately supported.
Although Quong does not explicitly state his reason for endorsing premise (2), my impression is that his reason consists of two propositions: (2a) If one is able to do the following three things, then there can be a foundational disagreement between reasonable citizens about a certain judgment: (i) one is able to conceive of a reasonable citizen making that judgment; (ii) one is able to conceive of another reasonable citizen who disagrees with that judgment; and (iii) one is able to conceive of the details of the disagreement in such a way that one cannot see how to resolve the disagreement by appealing to commitments shared by the parties to it. (2b) One is able to do these three things for any* judgment about the good that a reasonable citizen might make. Although Quong's defence against the asymmetry objection looks promising, there are objections to both of these premises.
There are at least two objections to (2a). The first objection is that, even if no one can at present see how to do so, it does not follow that there is no way of resolving the disagreement by appealing to the shared beliefs of reasonable citizens. Many arguments that proceed by making inferences from shared beliefs are difficult to anticipate beforehand. For example, a person who accepts the axioms of Euclid's geometry may well not be able to deduce the theorems. Similarly, perhaps there are difficult-to-anticipate arguments from the beliefs shared by reasonable citizens and perhaps some of these arguments will serve to resolve disagreements about the good.
(Note that Rawls himself makes an argument from these beliefs that is difficult to anticipate. His argument is that given the two beliefs of reasonable citizens, it follows that the original position is a suitable method for determining the principles of a liberal society (1993: 24), and the application of this method reveals two principles which the major institutions of a liberal society should be organized to realize. So why not be open to difficult-to-anticipate arguments about the good, which start from these two beliefs?) The second objection to (2a) is that one cannot appeal to it when defending Rawls's political liberalism, in light of Rawls's prohibition on appealing to metaphysical commitments. Metaphysical commitments are, for the most part, doctrines about the general nature of reality. Some doctrines of this type concern themselves with possibility, for example the doctrine that what is conceivable is possible. This doctrine cannot be appealed to when defending political liberalism. Proposition (2a) is not as general as this doctrine. It lacks the generality that is normally associated with a metaphysical commitment. But it does not make sense to endorse this specific proposition without endorsing a more general theory of possibility, presumably some qualified version of the doctrine that what is conceivable is possible. Quong's way of explaining to reasonable citizens why a liberal constitution cannot be based on any of their judgments about the good therefore implies this more general theory. Since appeal to this general theory is prohibited, appeal to (2a) should be prohibited as well.
What about (2b)? It seems that Quong endorses this proposition on the basis of an induction. He takes a sample of conceivable disagreements between reasonable citizens about the good, tries to resolve them but cannot see how, and so concludes that all such disagreements are unresolvable. An objection to his use of this method is that his sample is too small. He offers only one example of a foundational disagreement about the good. Other objections arise when we look carefully at the example.

A problem with the example
Quong tries to provide us with a fictional example of a foundational disagreement over the good between two reasonable citizens. One of the persons involved is called Mike: Mike believes recreational drug use is immoral because it involves seeking pleasure for pleasure's sake-it follows from a hedonistic view of what makes a good human life. Mike thinks this view of human life is completely mistaken. He thinks human life is created by God and thus the proper function of every human life is not the pursuit of pleasure, but devoted service to God's commandments. (2011: 204-205) As Quong interprets his example, Mike's standard for evaluating judgments about the good is whether or not such judgments follow from or are at least consistent with God's commandments. The other person involved in the disagreement is Sara. I will not refer much to her below. Briefly, Sara's standard for evaluating judgments about the good involves the following commitment: if an action that a person does has no effect on anyone else, then it cannot be subject to moral evaluation. She thinks that using recreational drugs does not have an effect on others, so she denies that it can be subject to moral evaluation (2011: 205).
Quong says that this is a foundational disagreement because "there is no deeper standard of justification that both Mike and Sara accept that could serve as a basis for adjudicating their dispute. " (2011: 205) There are various questions that can be raised about the details of this example. The question I shall raise is this: is Mike a reasonable citizen? If not, then we have not been provided with an example of a foundational disagreement between two reasonable citizens. Quong asserts that Mike, as he is described above, could be a reasonable citizen (2011: 206). To challenge this assertion I shall adapt a famous Old Testament story.
Imagine that Mike has a son and one day perceives that God commands him to sacrifice his son. Mike prepares to do so. Is Mike a reasonable citizen? One thing that might be said is that he is simply not a reasonable citizen, by Rawls's definition, even before preparing to sacrifice his son, because he accepts the following commitment: if God commands me to sacrifice my son, I should do this. For it is doubtful that such a commitment is consistent with the beliefs that define Rawlsian reasonable citizens. Presumably, an advocate of Rawls's political philosophy would agree that there is an inconsistency. Perhaps though it will be said that Mike is reasonable while the commands he interprets God as giving are reasonable, but as soon as he interprets God as giving unreasonable commands, then he is unreasonable.
From imagining this situation, we can see that either Mike is not a reasonable citizen, owing to what he is prepared to do if God commands it, or else he is a very tricky case of a reasonable citizenfurther argument is needed to establish that he is reasonable. Consequently, Quong does not put us in a position to say that there can be foundational disagreements between reasonable citizens, because it is just not clear that the example he gives is an example of a disagreement between two reasonable citizens.

Another problem with the example
Even if both parties in Quong's example somehow qualify as reasonable, it is open to doubt that what he presents is a foundational disagreement. Rawls can only justify the principles of his proposed constitution to all reasonable citizens by implying statements of the form "If you accept X, you should accept Y, because Y follows from X." It seems then that Rawls understands reasonable citizens as sharing more than just the two beliefs identified earlier. They also share a commitment to valid reasoning as a criterion by which to evaluate arguments.
In light of this point, if both parties in Quong's example are reasonable, then there is a shared commitment which allows Sara to object to Mike's standard for evaluating what is good. Recall how Quong describes Mike: He thinks human life is created by God and thus the proper function of every human life is not the pursuit of pleasure, but devoted service to God's commandments. (2011: 204-205) The word "thus" here is Mike making an inference: from the premise that human life is created by God, he infers the conclusion that the proper function of every human life is devoted service to God's commandments. But from this premise alone, the conclusion does not follow (Hume 1739: 3.1.1.27;Kelsen 1959: 108). For it is no violation of logic to accept the premise yet reject the conclusion. Sara can therefore say to Mike that his standard for evaluating what is good should not be accepted by any reasonable citizen, at least on the basis of Mike's argument, because it is the result of invalid reasoning. His standard is the result of an inference which does not follow. Consequently, it is unclear whether this really is a foundational disagreement. If both parties are reasonable, then they both accept valid reasoning as a criterion by which to evaluate arguments, and this commitment is something that Sara can appeal to in order to object to Mike's standard.
Quong's defence against the asymmetry objection is based on the idea that if a judgment is made by a reasonable citizen and another reasonable citizen might disagree with that judgment and the disagreement could potentially be a foundational disagreementa disagreement where there are no shared commitments which enable a resolutionthen the constitution should not be based on that judgment. Judgments about the good are supposed to fit this description. But, in addition to the objections identified two sections ago, it is unclear whether even the example Quong gives of a reasonable yet foundational disagreement about the good really is a foundational disagreement. Regarding judgments about what is valuable in human life, Quong writes, "metaphysical beliefs often provide the explanation for why those judgements are valid or true." (2011: 13) In which case, the problem identified in this section will arise in a number of other cases.