Moderate Idealization: Preserving Diversity
A central part of Kevin’s defense of convergence as a superior model of public justification depends upon how citizens are idealized. Radical idealization—where citizens are posited to be fully rational and fully informed—is problematic for Kevin’s aim of reconciling religious and liberal citizens. Radical idealization of citizens is problematic because the reasons that fully informed, rational citizens have may be vastly different than the reasons, we, here and now, have. Taking the reasons of radically idealized citizens as reasons for us, here and now, may challenge our integrity or may undercut some of our liberties. Moreover, were the convergence model to employ radical idealization the argument for restraint may be resurrected. Hence, it is crucially important for Kevin’s project that he provide a defensible account of moderate idealization of citizens that doesn’t carry the burdens of radical idealization.
The Chapter is dense in the sense that a lot of ground is covered. I will try to summarize concisely and raise a few concerns.
I. The Standard Conception of Idealization
Recall the Public Justification Principle (PJP): A coercive law L is justified only if each member I of the Public P has some sufficient reason(s) R to endorse L.
A conception of idealization gives not only an account of whom the members of the public are but also specifies their reasoning capacities and informational sets.
The standard conception employees a radical notion of idealization; citizens are taken to be fully informed and fully rational. They are also said to be reasonable, where being reasonable entails, at least:
i. A disposition to engage in public justification, and abide by justified principles.
ii. A disposition to recognize the “burdens of judgment.”
iii. A disposition to reject the repression of other reasonable points of view. (p.147)
Full rationality entails that citizens do not make common errors of reasoning. They have “flawless cognitive powers,” and their reasoning does not face the kinds of constraints that real-world reasoners face (e.g., time limits, high information collection costs).
The standard account doesn’t quite rely on idealized agents having full information; some restrict the kinds of information idealized agents have regarding their own particularity (e.g., Rawls’s veil of ignorance).
The advantages to the standard conception, noted by Kevin, are that it allows us to avoid “attributing reasons to citizens based on falsehoods” and generate determinacy of reasons.
II. An Argument from Incoherence
Here Kevin argues that radical idealization is incoherent. He argues there are two fatal flaws: standard idealization will not reduce errors in reasoning in the way it claims and radically idealized agents provide such “odd recommendations” that they fail to count as idealizations of real world agents.
Drawing on Pollock’s criticism of classic decision theory, Kevin argues that the gulf between idealized agents and real world agents is sufficiently large that the reasons for idealized agents do not count as reasons for real-world agents.
Ideal agents make optimal choices over a range of acts. Moreover, they have full information, so they always make warranted choices. For real-world agents to follow the recommendation of radically idealized agents would often be irrational, given the real world obstacles they face. An idealized agent would be justified in thanking Clark Kent for Superman’s actions, but a real-world agent would not be so justified, even if she were warranted. Given the current knowledge available for the real-world agent, it would be irrational for her to act on a recommendation to thank Clark Kent. Even more, real-world agents don’t choose optimal acts; they make choices about plans, which are defeasible and rarely fully worked out. The aim of a real-world agent is to choose good plans, given all the unknown contingencies they may face.
Kevin argues and ideal advisor account won’t save radical idealization from these criticisms. For space reasons, I leave this reply aside.
III. The Integrity-Based Critique of the Standard Conception
Given that a central aim of the book is to argue that the standard account of public justification, and so public reason, unduly burdens the integrity of religious believers and to rescue liberalism (and so public justification) from this criticism, if radical idealization unduly burdens the integrity of religious believers, that is a further reason to reject it. Kevin argues it does.
Idealization “imposes integrity costs to the extent that it revises an agent’s belief set.” And, since radical idealization involves a substantial revision to the belief set of agents, it imposes large integrity costs. We will do better to accept moderate idealization, and with it lessen the potential for integrity costs.
Since the whole point of idealization is to give an accurate account of citizens reasons, if idealization revises their belief set in a way that challenges or undermines (their presently held, actual) deep commitments, idealization may justify coercion that actual citizens reject as compatible with their deeply held commitments. And, this would unduly burden their integrity.
In developing this argument, Kevin draws on the work of Gilbert Harman and adopts the following principles to support the claim that “there is a presumption against revising one’s belief-value set in determining reasons for action”:
Clutter Avoidance: One should not clutter one’s mind with trivialities. 155
Interest Condition: One is to add a new proposition P to one’s beliefs only if one is interested in whether P is true (and otherwise reasonable for one to believe P). 155
Kevin revises the latter principle: revising our beliefs is justified insofar as doing so helps us achieve our goals.
Since citizens have an interest in “extending their effective power to act with integrity,” they have an interest in revising their beliefs to the extent it helps them achieve this goal. Radical idealization revises their beliefs too radically if the reasons of the radically idealized agent seem alien to actual agents or conflict with their deeply held commitments. Harman’s principles help demonstrate that citizens interest in idealization extends only to what is necessary to help them achieve their goals (one of which is living with integrity).
IV. The Diversity-Based Critique of the Standard Conception
The second of the twin desiderata of integrity and diversity for an account of public justification (and public reason) provides further reason to reject the standard model of idealization.
The standard conception generates determinacy (determinate recommendations for common laws) by normalizing agents—giving them the same full reasoning capacities and information sets. However, if radical idealization normalizes too much, it undercuts the range of diversity. Kevin argues that we don’t need to radically idealize to achieve determinacy. And since radical idealization narrows the range of reasonable pluralism, a moderate conception is preferable if it can achieve determinacy without sacrificing diversity.
V. A Model of Moderate Idealization
Kevin offers a possibility proof for a moderate conception of idealization, aiming to show it is superior to a radical model and does the work needed while reconciling religion and liberalism.
The Rationality Dimension: Moderately idealized agents “rarely optimize, they engage in nonterminating reasoning and they make defeasible inferences.” They make justified choices, rather than warranted ones. They avoid clutter and “only affirm values and beliefs required to satisfy their interests.” 161
Moderately idealized agents aim for coherence or consistency in their reasoning.
They are rational if they have epistemically justified beliefs and arrive at those beliefs via sound rule of inference.
The Informational Dimension: Agent’s practical interests (in their projects and goals) specify what kinds of information they need. Information ascribed to moderately idealized agents is attributed to them on the basis of relevance, collection costs, and processing costs.
Reasonable Dimension: An agent is reasonable if she has:
i. A disposition to engage in public justification, and abide by justified principles.
ii. A disposition to recognize the “burdens of judgment.”
iii. A disposition to reject the repression of other reasonable points of view.
Moderately idealized agents reasons more closely resemble our own reasons for action than does the reasons of a radically idealized agent.
Moderate idealization doesn’t generate determinacy by normalizing agents in the way that a radical conception does. Moderately idealized agents have their own particular reasons given their interests/goals/beliefs, etc. We cannot generate determinate content of reasons by moderate idealization alone. Thus, we need an account of how determinacy is generated.
Kevin identifies four mechanisms for generating determinacy: deliberation, bargaining, adjudication, and evolution or convention making.
Deliberation is self-explanatory: citizens deliberate about proposals aiming to agree on an outcome.
Bargaining involves agents aiming to trade and cooperate with others to achieve the outcome she regards best.
Adjudication involves a third party, i.e., judge, resolving conflict to generate convergence.
A convention can generate convergence spontaneously without deliberation or bargaining in cases where a common standard is needed for social cooperation. For example, the convergence of consumers on the choice of a product (VHS v. Betamax) can lead to a spontaneous agreement without any explicit bargaining or deliberation.
Convention based convergence faces three challenges. First, such norms need a feedback mechanism in order for them to reliably generate publicly justified laws. Punishment is that mechanism. Kevin argues it is likely for moderately idealized agents to be induced to punish convention-breakers insofar as humans have a deep seated and near universal inclination to punish violators. Second, we need an account of how laws can evolve. Judge made common law is an example of legal evolution. Third, some may worry that convention based convergence may converge on unjustified laws. He replies: assuming we can determine which laws are publicly justified (see early arguments), then moderately idealized agents will gravitate towards those laws because they are motivated to comply with justified laws.
VII. Five Concerns about Moderate Idealization (and Kevin’s replies)
1. Moderate Idealization will collapse into radical idealization.
Reply: The account of moderate idealized agents only idealizes agent’s beliefs when it is in the practical interest of the agent to do so, and so it doesn’t succumb to the same pressures as radical idealization to normalize agents.
2. Moderate idealized agents may be still too idealized to generate reasons for us, here and now.
Reply: The aim of moderate idealization is to facilitate the ability of real world agents to achieve their projects and goals. In this way, moderate idealization does not abstract way from citizens’ actual plans, goals, and projects.
3. Moderate Idealization allows for the coercion of marginal members of the public—persons who had they reasoned a bit more would have rejected the law in question.
Reply: Yes, his view does allow coercing the marginal member, but it’s not clear there are any better alternatives. Moreover, if there are good reasons to reject a law after it is considered initially publicly justified, we should consider those reasons and perhaps take action.
4. Moderate idealization may not allow us to distinguish between a cult member and a religious person.
Reply: The requirement that agents be reasonable rules out cult members. Moreover, cultish beliefs are won’t survive rational examination.
5. Given there are various possible accounts of moderate idealization may accused of assigning reasons to citizens on an ad hoc basis.
a. Reply: While there can be various accounts of moderate idealization, they will likely assign similar reasons. They problem of various accounts doesn’t appear dehabilitating.
VIII. Discerning What is Justified
Radical idealization makes discerning what is publicly justified rather easy insofar as it normalizes agents they all have the same reason. Moderate idealization allows that citizens are diverse and disagree, so determining which of their reasons will be publicly justified is more difficult.
Kevin distinguishes two ways of thinking about the problem of public justification. The first way involves constructing principles of justice a priori. He claims this is a plausible way of interpreting Rawls’s project in A Theory. The second way sees public justification as aimed at resolving conflicts as they arise. This second way is more amenable to moderate idealization because we take issues as they emerge and evaluate laws “in a piecemeal fashion.”
However two objections may be lodged against this method: it’s overly conservative and it may be relativistic. Kevin answers the conservatism charge by claiming that revolutionary change my sometimes be necessary, we can evaluate the complaints of challenges to the entire system. However, revolutionary change often produces a cure worse than the illness, and so we should be cautious. He answers the relativism charge by claiming that his view is not problematically relativistic, for the PJP does show some laws are unjustified as they unduly restrict freedom or equality; there is an objective basis to discern justified from unjustified laws.
Other methods of discernment include: deliberation, studying the evolution of norms in response to social changes, and forms of media. These public fora provide opportunities for the discovery of social norms and points of convergence.
Comments and Questions:
1. I disagree with Kevin’s interpretation of Rawls’s project. I will try to be brief here in order to get to Kevin’s own account. Rawls has made clear, in his later writings, that Justice as Fairness—the account of the OP, VI, and the way it generates principles of justice—is but one way of formulating an account of justice. There are other reasonable ways of doing so, and other reasonable political conceptions of justice. Hence, it is one way of generating content for potentially publicly justifiable principles; it is not the only way. Neither it is an account of public reason. Public reasons are generated from reasonable political conceptions of justice, but there may be several and reasonable citizens won’t necessarily agree as to which one is preferable. Any reasonable account of political justice must include three features, limited by the criterion of reciprocity: a list of basic rights and liberties, an argument for the special priority of those rights, and measures to ensure citizens effective use of their freedoms. Various reasonable accounts of political justice will generate different content to public reason (“Idea of Public Reason Revisited, in Collected Papers, p. 582). The point of this is that political liberals need not be committed to radical idealization (of the kind reflected in the O.P. and V.I.), and their account of public justification does not turn on a commitment to Justice as Fairness. Nonetheless, political liberals are committed to idealization; perhaps more radical than Kevin thinks is compatible with integrity.
2. I wonder if Kevin gives far too much weight to the value of one’s present ends and projects. In other words, the way he talks about persons’ projects and goals make them sound constitutive, in a strong sense, of persons. In specifying citizens interests in pursing their ultimate aims and projects with integrity as the basis for idealization, Kevin argues that idealization is necessary only to the extent that is serves that goal (protecting their given ends and goals). However, if you think that citizens also have an interest in finding fair terms of social cooperation with their co-citizens; they may have conflicting interests. Moreover, if you think that citizens interest in finding fair terms of social cooperation is either a. one of their ultimate aims and goals or b. a higher priority aim (this could be rendered plausible if you think that their ability to achieve other aims an interests in dependent on fair, stable cooperation with others), then citizens may have reasons to accept stronger idealization and may have reasons to revise or alter their other (presumably, nonpolitical) goals or aims.
3. A third point I wish to raise has more to do with the convergence model of public reasons than with the account of idealization, but its something that has been troubling me throughout the book. On the convergence account, public justification occurs when citizens converge on a particular law and each has sufficient reason to endorse the law in question. From which it follows, they need not have shared or even accessible reasons. Kevin identifies deliberation and adjudication as two mechanisms by which we can determine the content of our reasons. However, I wonder whether a convergence account is sufficient to allow these mechanisms to work. Consider in that adjudication (both in the context of common law and statutory law), judges are trying to determine whether a law applies to a new case, whether the case is sufficiently similar to past cases, and so on. In so doing, they often need to appeal to the underlying principles or rationale for the law (or constitutional principle). The underlying rationale for the law or principle is taken to reflect our deeper shared commitments, and it is those shared commitments that provide guidance to the new case or allow us to determine whether is relevantly similar. An example: suppose citizens converge on something like the First Amendment. For their various, particular reasons they all think that “Congress shall make no law respecting an establishment of religion, or prohibiting the free exercise thereof; or abridging the freedom of speech, or of the press;…” In deliberating or adjudicating cases, it is often necessary to invoke the underlying rationale, the role that such freedom plays in a democratic society for example (and so a particular and presumably shared vision of what that amounts to), to settle the interpretative issues or to determine the principles’ extension. How is convergence sufficient to settle such questions?
4. Finally, I wonder what motivates Kevin’s commitment and definition of what it is to be reasonable. For Rawls, the account of reasonableness is generated in part by the assumption that citizens want to find fair terms of social cooperation and regard such terms as having value in it self. Citizens value and desire to find fair terms of social cooperation, not just to facilitate their own ends, but because this is part of their vision of respectful cooperative living. On Kevin’s account, it looks like citizens motivation to converge on terms of cooperation is grounded in their desire to pursue their own life plans with integrity, and social cooperation is necessary for that. But this sees social cooperation as instrumental, and doesn’t generate the dispositions to reasonableness in the way that a commitment to seeking fair terms of social cooperation for their own sake does. It may be that this is what generates the supposed conflict with integrity of citizens where their ultimate ends are always trumps. In contrast, in Rawls’s view political justice must trump comprehensive values (or we must show how its possible that political justice trumps comprehensive values).