MICHELMAN-BR WEBGALLEYS1.DOC 08/07/00 1:47 PM

1945

AND THE THINNING OF JUSTICE

For all its multiculturalist sympathies and flippable title,

postmodern loss of faith or withdrawal from strongly prescriptive

talk about justice in law.2 A formidable work of straight normative

political theory, the book acutely and elegantly restates a central

problem in the moral justification of political practice, canvasses

responses proposed by various eminent doctors, finds all of them

wanting, and ends by offering its own duly chastened response.

debate. The question is whether and how we may have liberal

legitimacy in democratic politics.3 As John Rawls puts the

question: “[I]n the light of what principles and ideals must we, as

free and equal citizens, be able to view ourselves as exercising [our

power as a collective body] if our exercise of [this power] is to be

justifiable to other citizens and to respect their being reasonable

and rational?”4 In other words: Given a general, prima facie moral

objection to coercion, what sort of guiding criteria for laws and

their applications might possibly justify the coercive practice of

democratic politics?5

1 MICHEL ROSENFELD, JUST INTERPRETATIONS: LAW BETWEEN ETHICS AND

POLITICS (1998).

167, 171, 177 (1990).

U. L. REV. 1009, 1009-10, 1014-18 (1999) (analyzing political-justificational aims of

contemporary liberal political philosophy).

4 JOHN RAWLS, POLITICAL LIBERALISM 136 (2d ed. 1996).

5 Rosenfeld rightly includes all episodes of lawmaking and law application under the

achieved . . . requires just interpretations of applicable laws and . . . laws that are

MICHELMAN-BR WEBGALLEYS1.DOC 08/07/00 1:47 PM

Various answers are proposed. Jürgen Habermas, for

example, offers a “democratic principle,” drawn from discourse

theory, to the effect that morally legitimate laws are those able to

gain the assent of everyone in a legally guaranteed, democratically

constituted, “discursive process” of “opinion and will-formation.”6

Rawls himself offers the idea of a “political conception of justice,”

a substantive doctrine of political rightness developed from ideal

notions, including a basic norm of reciprocity, that he finds already

resident in constitutional-democratic public culture.7 Michel

Rosenfeld proposes, instead, a substantive doctrine of the good he

calls “comprehensive pluralism.” Among my specific aims in this

review is to understand precisely how Rosenfeld means his

proposal to differ from that of Rawls.

Initially, I grant, there is little mystery about that. Rosenfeld

pointedly places the ethic of comprehensive pluralism, a

Habermasian procedural and Rawlsian substantive conceptions of

coercion.8 “[T]o the extent that comprehensive pluralism amounts

to yet another conception of the good,” he writes in what I shall be

treating as a key passage:

[I]t is clearly distinguishable from contemporary Kantian

pluralism is more than yet another conception of the good, in

that it unleashes an inevitable but ultimately resolvable dialectic

between first-order and second-order norms, it differs markedly

from theories based on building an “overlapping consensus,”

from all theories ultimately reducible to mere relativism.9

The meaning of this passage is not immediately and transparently

evident. To understand it fully, I shall suggest, is to peer deep into

Rosenfeld’s thoughts about the form of any possible answer to the

problem of liberal legitimacy in politics.

(1995).

6 JüRGEN HABERMAS, BETWEEN FACTS AND NORMS: CONTRIBUTIONS TO A

DISCOURSE THEORY OF LAW AND DEMOCRACY 110 (William Rehg trans., 1996).

COMPANION TO JOHN RAWLS (Samuel Freeman ed., forthcoming 2000) (describing

accompanying notes 59-60.

MICHELMAN-BR WEBGALLEYS1.DOC 08/07/00 1:47 PM

I. WHAT CRISIS?

here and now. Let us say, to keep things simple, that it’s a crisis

for contemporary Americans (although not for them alone).

Called a crisis of legal interpretation, is it also plainly to be

understood as a crisis of justice and political legitimacy. In

retracing the way in which Rosenfeld sets up the crisis, I shall use

the same terminology that he has used in his book. I warn the

repetition, of Rosenfeld’s text. Whether it is a just interpretation,

or rather is just an interpretation, will not be for me to say.

Rosenfeld divides all politically governed societies or

“polities” into the two classes of the “homogeneous” and the

“heterogeneous.” He subdivides heterogeneous polities into the

two subclasses of “modern” and “postmodern.” Rosenfeld

attributes the crisis to the most historically recent quadrant of this

society/polity space-time-that of the postmodern subtype of the

heterogeneous polities.

There is no crisis in homogeneous polities because these are

defined as polities in which ethical and ideological consensus over

ideas of the good and dominant common interests is so wide, deep,

and strong as to leave little or no dispute over the guiding criteria

for just legal interpretations.10 Even in such societies, Rosenfeld

expects, there must inevitably be some disagreement over

decisions of specific cases, but that disagreement will be swamped

in a widespread sense of confidence that all are applying the same

system of ethical and ideological commitments, the same

conceptions of what gives value to human life and what is in the

contrast, are those in which the depth of the differences among

numerous thriving and conflicting ethical value systems and

ideological formations precludes any sense of rooted agreement on

good, referring only occasionally to what I would call ideological divisions, or those over

dominant common interests-a salient example of which would be conflicts between the

outlooks of owners and workers in at least some forms of capitalism. Rosenfeld uses the

term “ideological” in a slightly different way, describing “labor” and “management” as

possibly sharing a “capitalist ideology,” even while holding differing “perspectives” within

difference are relevant to his diagnosis of a current crisis of legal interpretation.

Accordingly, I will often speak, in what follows, of “ethical and ideological” divisions and

41.

MICHELMAN-BR WEBGALLEYS1.DOC 08/07/00 1:47 PM

guiding substantive criteria for justice in legal interpretations,

much less on the justice of every interpretation actually reached

and applied by officials and citizens.11

Not every heterogeneous polity, however, is caught in the

crisis, because ethically and ideologically rooted agreement-

“substantive” agreement-is not the only kind of agreement we

might have on the justice of legal interpretations. Among

heterogeneous polities-those in which substantive agreement on

justice in law is impossible-some fall into the class of what

Rosenfeld calls modern polities.12 In such a polity, despite the

presence of an irreconcilable plurality of conflicting conceptions of

the good and of the common interest, there nevertheless can be

found a widely shared understanding about how to bridge the

resulting fissures so as to maintain social unity and coherence.

This shared understanding takes the form of a largely

proceduralized conception of justice, an agreement that justice is

largely a matter of the observance of institutional arrangements

and procedures that mediate appropriately among ethical value

systems and ideological claims. What people in a modern polity

manage to retain, if only barely, as Rosenfeld beautifully

explains-what gives them their social unity and cohesion-is a

somewhat precarious (“contingent”) sense of confidence that their

basic institutional setups really do effectuate a suitable division of

labor among the normative domains of law, ethics, and politics.13

A postmodern polity, in contrast, not only lacks any unity or

convergence on ethical values and common interests, but further

lacks any “broadly based shared understanding” of how to

maintain social unity while at the same time “preserving . . .

plurality and dealing fairly with the issues that arise as a

consequence of . . . diversity.”14 In other words, in a postmodern

polity, there is lacking not only a substantive, but also a

procedural, consensus. As Rosenfeld puts the case, not only have

“commonly held norms and values reced[ed] nearly completely,”

but also we have come to the point where every possible

“institutional means” for regulating conflicts among competing

conceptions of the good seems biased in favor of some substantive

competitors.15 So deep and sharp have our ethical and ideological

touching these divisions that are sufficiently determinate to hold

MICHELMAN-BR WEBGALLEYS1.DOC 08/07/00 1:47 PM

off unjust arbitrariness at the stage of applying them, without

introducing unjust partiality at the point of framing them. All

interpretations, then-all acts of political authority taken in the

name of the law-must be unjust (or at any rate cannot be known

or seen to be just), either because the laws and principles that

constrain them are unjust or because they are not credibly

constrained by any laws and principles.16 And that, then, is the

crisis from which Rosenfeld believes we now stand in need of

relief, in the form of a new-minted understanding of how law can

bridge the gap between self and other.

Rosenfeld’s ideal-typical classification of polities depends

crucially on a distinction he draws between two kinds of

disagreement about justice in legal interpretations. Obviously,

there can be disagreement over the prescriptive content of the idea

of justice, or what Rosenfeld calls the “normative criteria” to be

used in appraising the conformity to justice of any given

interpretation. But even when people are sure they agree about

those criteria, or what we would call the standards or principles of

justice, Rosenfeld says they can disagree about whether a given

instance of interpretation satisfies justice.17 In Rosenfeld’s view,

the problem of just legal interpretations is solved only when both

kinds of disagreement are adequately held in check. A society in

which too many legal interpretations are found unjust by too many

citizens or groups is, in his apparent view, ipso facto not a justly

governed society. In other words-and this is a point to bear in

mind-there is, in Rosenfeld’s apparent view, an internal

connection between the prevalence or non-prevalence of justice in

a given polity and the social distribution of belief in the justness of

the legal interpretations in that polity

17 I speak, as does Rosenfeld, of the justice (unmodified) of legal interpretations,

he does so only for analytical purposes and without ever implying that justice in legal

in the context of law,” he writes, “cannot be legitimately confined to justice according to

among other things), evincing due regard for positive law as it really is. It is, rather, to

subsume the doing of legal justice under the doing of justice. We assume, in other words,

that legal officials do not act justly at all if they fail to give due consideration to the

requirements of legal justice-meaning, by due consideration, whatever consideration

justice unmodified requires. The open question is: What and how can the idea of legal

justice-official decision and action according to law-contribute toward the achievement

of justice, or at least the reduction of injustice?

MICHELMAN-BR WEBGALLEYS1.DOC 08/07/00 1:47 PM

We must notice, now, that when Rosenfeld speaks of people

agreeing on the standards or criteria of justice but reaching

differing conclusions about what those criteria require in specific

cases, he is not just observing that people can disagree about the

specific facts of the cases before them, regarding the relevant

immediate acts and relations of the parties. He means they can

also disagree over how to apply an agreed measure of justice to the

decision of a case where the adjudicative facts-as legal theorists

would call them-are transparently clear and undisputed. For

example, the agreed standard might be a ban on laws prohibiting

the free exercise of religion and the case might be one of an Air

Force officer punished for (admittedly) having worn a yarmulke

while on duty, in (admittedly) direct and flagrant violation of an

(admittedly) duly promulgated military regulation.18

People, Rosenfeld says, can agree on the “free exercise”

standard-and be confident they are agreed on something very

important-even as they obdurately disagree over application of

the standard to a specific case. It thus seems that, in Rosenfeld’s

view, a normative standard can at least sometimes retain its

identity-can be held constant-under mutually contradictory

applications. People, in his apparent view, can at least sometimes

retain their certainty that they are all applying the same criterion

of judgment, even while they disagree about what result that

criterion requires in a case in which there is no dispute about how

to describe the relevant immediate facts. This decoupling of the

identity of a normative principle from its applications is an

absolutely crucial point for Rosenfeld. On it depends his

construction of the crisis.

For consider: If every disagreement about how to decide a

factually undisputed case in accordance with justice were taken to

disclose a hitherto latent disagreement about the applicable

principles or criteria of justice “themselves,” that would be the end

of the Rosenfeldian distinction between homogeneous and

heterogeneous polities. Without the idea that people can maintain

their sense of agreement on one and the same normative principle

even as they disagree diametrically over how to apply it, we would

never be able to say, as Rosenfeld says of homogeneous polities,

that there is constant agreement on the principles, even given the

inevitable disagreement about their applications. And yet there

appears to be a problem with putting the matter this way, for it

seems the very same thing could be said of our society.

In American society, for example, there is constant agreement

MICHELMAN-BR WEBGALLEYS1.DOC 08/07/00 1:47 PM

on the abstract principle of free exercise of religion, accompanied

by deep and chronic disagreement about the application of this

principle to cases such as that of the Air Force captain and his

yarmulke. There is constant agreement on the abstract principle

of equality of governmental concern and respect for every person,

accompanied by deep and chronic disagreement about the

application of this principle to decide questions about whether

government in the United States today is required to be

colorblind, or rather is sometimes permitted, or even required, to

act in a race-conscious manner.19 Generalizing from these

examples, can it not be said of us that we have a consensus on

substantive political values, provided only that we state these

values at a high enough level of abstraction from hot controversy,

whether ethically or ideologically fueled? Can it not be said of us

that our disagreements occur only on the level of applying our

agreed principles to particular controversies? If so, then how are

we Americans not, by Rosenfeld’s definition, a homogeneous

society?

Of course, Rosenfeld cannot tolerate any such inference, for

its plain result would be that the crisis he constructs has no

application to us. (And if not to us, then to whom?) And so we

have before us a question about how Rosenfeld can avoid this

inference that appears to be fatal for his project of prescribing for

a crisis.

Perhaps he could say that Americans imagine they agree on

basic normative principles of political justice when they really do

not. If we Americans lack even near-agreement on the application

of our “equality” principle to decision of the color-blindnessversus-

race-consciousness question, or on the application of our

religious-freedom principle to broadly applicable regulatory laws

that severely and exceptionally impede particular, minority

religious observances,20 or of our proprietary-liberty principle to

onerous workplace and environmental regulation,21 then what is it

in those general neighborhoods of normative concern that we have

to the purpose at hand, nothing that averts the crisis. But that

would be a leap from the pan to the fire. It would be, in effect, to

assert that disagreement over the application of a normative

PHILOSOPHICAL AND CONSTITUTIONAL INQUIRY (1991).

U.S. 872 (1990).

at 1061 (Stevens, J., dissenting).

MICHELMAN-BR WEBGALLEYS1.DOC 08/07/00 1:47 PM

principle impeaches all sense of agreement on the principle

“itself.” It would say, in effect, that disagreement over application

just is, or just does directly reflect, disagreement over what the

the other direction, the Rosenfeldian division of polities between

homogeneous and heterogeneous.

Before examining how that bears on Rosenfeld’s project of

prescribing for a crisis, we should have a glance at his use of a

substance-procedure distinction to subdivide heterogeneous

polities into modern and postmodern types. Marking the

difference between these two kinds of polities, Rosenfeld says, is

that only in the former do we find the makings of a procedural-

although not a substantive-consensus. Here, to remind you, is his

precise statement about what it is that modern polities have and

postmodern polities lack: “a broadly based shared understanding”

about how to “handle,” by the “institutional means” of a “division

of labor between law, ethics, and politics,” the very fact of ethical

pluralism that characterizes all heterogeneous polities-that is,

their lack of agreement on “legitimate normative criteria for

evaluating laws and adjudications.”22 If that formulation does not

point to a distinction between substantive and procedural

agreement, then I do not understand what difference Rosenfeld

means to posit between the modern and postmodern subclasses of

heterogeneous polities. If, on the other hand, it does point to a

distinction between substantive and procedural agreement, then

insofar as such a distinction cannot hold up, neither, it seems, can

Rosenfeld’s distinction between modern and postmodern polities.

And in fact it is far from clear that we can finally make sense of the

idea of agreement on a procedural level, in conditions of ethical

and ideological division so deep and wide as to place beyond reach

all hope of agreement on a substantive level.23

In sum, we must doubt the ultimate maintainability, in the

relevant context, of both the insulation of principles from their

applications and the idea of a procedural-not-substantive

consensus. If our doubts are well founded, they plainly threaten

the sturdiness of both of Rosenfeld’s ideal-typical divisions of

polities-between homogeneous and heterogeneous, and, within

the heterogeneous type, between modern and postmodern. That is

Doubts about flights from substance are a staple of recent American constitutional theory.

MICHELMAN-BR WEBGALLEYS1.DOC 08/07/00 1:47 PM

important, because it prompts us to consider that perhaps the

vexations of just legal interpretation are not a crisis at all, but a

nuisance ailment of the constitutional or law-governed state. They

would be, then, like mice in the attic, not a new disease of our latearriving

postmodern polities but a chronic affliction at least as old

as the house of constitutional government.

Suppose that constitutional-democratic societies, for quite

some time now, have been so ethically or ideologically diverse that

major disagreements over the application of nominal substantive

norms have been a part of their daily, common experience. If the

insulation of a norm’s identity from its major applications does not

hold up, if obdurate disagreements over major applications are

tantamount to lack of accord on the guiding substantive principles

of interpretation “themselves,” then those same societies are, and

long have been, heterogeneous ones in Rosenfeld’s taxonomy.

Now suppose that a great many societies are, and long have been,

so ethically or ideologically divided that choices among political

institutions and processes inevitably appear to be, or to implicate,

choices among competing ideological concerns or conceptions of

the good. In other words, the distinction between procedure and

substance does not hold up in these societies. Perhaps variations

in the strictness of observance, or intensity of application, of

ostensibly procedural principles, such as separation of powers and

federalism (“states rights”), are widely perceived as ethically or

ideologically biased. Any polity of which that has been true since

some moment in the past has, over that same span of time, been

“postmodern” in Rosenfeld’s classification, a social scene in which

even the procedural constitutional essentials are sucked into the

swirl of ethical and ideological controversy. But this means that

the so-called crisis is widespread and long-standing in the world.

In other words, the implication is that what we’re looking at here is

simply a part of a normal condition in the politics of constitutional

democracy, a condition that history and experience tell us is

manageable without tears. (Got mice? Get a cat and learn to like

it.) If we now think about the types of contemporary responses to

this “false-alarm” way of viewing the matter would point strongly

toward favoring the deflationary, pragmatist approaches to the

interpretation question, such as those advanced by Richards

Posner and Rorty.24

Richard Fallon’s account of the conventions of American constitutional-legal argument.

100 HARV. L. REV. 1189 (1987).

MICHELMAN-BR WEBGALLEYS1.DOC 08/07/00 1:47 PM

But let us not overstate our case. To say that the

principle/application and substance/procedure distinctions do not

describe significant relative differences respecting societal

of “just interpretation” has indeed, as Rosenfeld in effect says in

his book’s opening sentence, recently graduated from chronicnuisance

status to critical significance in the actual experience of

social life. “Interest in a subject increases dramatically,” he writes,

“when that subject becomes problematic.”25

what Rosenfeld means. He does not, after all, himself believe in

the analytical robustness of the distinctions out of which he builds

his table of polities. Regarding the principle/application

distinction, Rosenfeld remarks that “[i]t is only when . . . grasped

from the perspective of the conception of the good to which it

happens to be linked that the full import of [a] . . . norm can be

properly evaluated.”26 Thus, contemplating the social fact of

disagreement over the justice of killing physicians who perform

abortions, Rosenfeld finds it “unclear” whether what we witness is

consensus on a norm prohibiting “murder in general,”

accompanied by a subordinate disagreement over what constitutes

murder in certain particular contexts, or, rather, “an actual

relativism derived from an insoluble conflict among irreconcilable

conflicting conceptions of the good.”27 Rosenfeld knows that it is

not only endless variations in the adjudicative facts that stand

between a principle and its applications. Also in that space are

more general, experience-based clashes of perceptions and

understandings-“perspectives”-regarding the background

conditions, relations, needs, and interests of various groups in

society, and a nominal principle of justice driven by one of these

clashing perspectives to require race-conscious government action

(using another example) is not comfortably identical with “the

same” principle driven by another perspective to prohibit it.

The idea of procedure transcending substance fares no better

with Rosenfeld. His critical treatment of Jürgen Habermas’s selfstyled

“proceduralist” approach is precisely aimed at dispatching

that idea. By way of illustration, Rosenfeld introduces a radical

feminist challenge against the Habermasian, proceduralist

HUM. RTS. L. REV. 249, 253 (1999).

MICHELMAN-BR WEBGALLEYS1.DOC 08/07/00 1:47 PM

paradigm of law-a challenge that shows, he says, that the

proceduralist paradigm is “buttressed by a particular conception of

law and rights.”28 “[A]ny alternative paradigm of law designed to

be consistent with the radical feminist challenge,” he says, could

not be merely a procedural one.29 “It would have to press

substantive feminist norms against masculinist objections,” and

thus would necessarily rely on “predialogical or extradialogical

sources of legitimacy.”30 “Legal interpretation,” Rosenfeld

concludes more generally, “cannot avoid reliance on substantive

normative commitments.”31

Given Rosenfeld’s own express denials of the analytical

robustness of the distinctions by which he differentiates crisis-free

homogeneous and heterogeneous-modern polities from crisisridden

heterogeneous-postmodern ones, it seems he must be

differentiating polities not in terms of categorical differences of

how law really does or does not work in them, but rather in terms

rule of law.32 Rosenfeld is best understood to say that, depending

principles transcending disagreement on applications, or a

substantive division, may or may not be capable of sustaining the

of which we speak, or would it be the sense or perception of

justice? Are they the same thing? Can the meaning and

possibility of justice really vary in this way from time to time and

from place to place, depending on how keenly and irrepressibly

aware people are of their ethical and ideological divisions and

corresponding normative disagreements? Indeed, it would appear

so, if justice and the perception of justice are internally connected

in the way that, now for the second time, we see implied by

very well transcend a specific set of substantive-normative divisions, while denying that

same point is indeed implicit in something I’ve already quoted from Rosenfeld’s text-that

is, his remark that, in postmodern polities, two developments occur in conjunction with

each other: “commonly held norms and values recede nearly completely” while, surely not

unrelatedly, “all institutional means for regulating intersubjective interaction” come to

“seem inevitably biased in favor of” one or another of the competing conceptions of the

postmodern polity as one of possibly gradual “erosion” of the “persuasive” power of what

had been, after all (in modernism), a strictly “contingent” overlap between the

interpretive implications of the impulses of “naturalism” and “positivism”).

MICHELMAN-BR WEBGALLEYS1.DOC 08/07/00 1:47 PM

Rosenfeld’s exposition of the crisis.

II. WHAT JUSTICE?

1. Reconciliation

In the terms favored by Rosenfeld, justice as it pertains to the

law of a pluralist society-one populated by a multiplicity of

individual and collective “selves” committed to various and

conflicting conceptions of the good and of common interest-is

part of the pursuit of a more “ultimate” good, consisting in “the

reconciliation of self and other in the least coercive and least

confining manner possible.”33 Law is the “external,” which is to

say the imposed, regulation of transactions and relations among

persons and groups. We may think of law at its best, democratic

terms for social coexistence and cooperation. Law thus involves

an “externalization” of the self toward the others. But “external”

is a two-way street. What is outgoing must also be incoming. In

an ethically and ideologically plural society, to think of democratic

law as outgoing from the self is also, unavoidably, to think of it as

incoming upon the self.34 (“We are all selves,” it might be said by

some president from the inaugural bandstand, “we are all others.”)

Binding terms that go out from any (or all) must also come to all,

and pluralist-democratic law, then, must also and always be an

irruption or intrusion by otherness upon the self.35 Thus, at risk in

the search for just legal interpretations is intrapersonal, as well as

social, unity-the threat to the unity of the self arising not only

from the fact that any self may at any time have potentially

conflicting ethical and ideological loyalties,36 but from the two-way

structure of the externalized character of law. “When the self

presses its claims in the name of justice,” Rosenfeld observes, it

does “violence” to other and different selves. Conversely, “when

the self restrains the pursuit of its own claims to do justice to the

make for a fair equilibrium between the respective needs, interests, and aspirations of the

entailed by a commitment to the all-out pursuit of self-other reconciliation).

the self (which from the vantage point of the other amounts to a mere extension of the

other) is but an illusion or an instrument of tyranny.”).

MICHELMAN-BR WEBGALLEYS1.DOC 08/07/00 1:47 PM

other, it does violence to itself.”37 In the domain of pluralistdemocratic

law-Rousseau would be the inspiration here38-

justice is the reach to make political authority and its exercise as

fully as possible internal to every individual and collective self in

the neighborhood. Or, to turn the point around, justice is the

reach to make political authority as little as possible an imposition

upon any self of what that self cannot recognize as according a full

measure of respect and sympathy to its own concerns, beliefs, and

commitments-indeed, so full a measure as almost-if-not-quite to

become or identify with those concerns. In sum, a just legal order

must be one that, in its content and by its practice, caters to

self/other empathy and reconciliation.

2. Decision

But how, then, is law to be made, or experienced as,

sufficiently “severable” from ethics to allow for reconciliation

through law without suppression of “ongoing” ethical conflict-

without, that is, suppression of at least some conceptions of the

good?39 Perhaps an answer lies in the very externality and

impersonality of legal norms of conduct, their appearance to daily

experience as force-backed impositions on members of society,

generally and indifferently. Perhaps that fact about the

experiential side of legal requirements is the very thing about them

that, in many cases (certainly not all) could allow for compliance

without self-loss or self-division, insofar as it allows for the

externalization of compliance from the innermost redoubts of the

self.40

That possibility would seem to be crucial to the hope that self

and other, freedom and political authority, might be reconciled in

and through the medium of law. For to imagine that is to imagine

that controversies so hot as to endanger the unity of both society

use of “violence” in contexts such as this. If this is violence, what word have we

[W]ithin the perspective of comprehensive pluralism, coercion is not only

constraining, but also liberating: it allows the self to . . . [become enlisted in the

pursuit of other’s objectives] without either of them being swept into the

perspective of the other . . . [and] without [the self’s] having to express allegiance

to the [other’s] good.

135, 136 (1994).

MICHELMAN-BR WEBGALLEYS1.DOC 08/07/00 1:47 PM

and the self, if left for communal-ethical or majoritarian-political

resolution, can be cooled and defused by being sent instead to

law, a law that is clearly detached from ethics and ideology, so that

it can-as far as it goes-displace ethical and ideological

partisanship from legislative and judicial rule. Yet this discrete

law, this law that binds authority to nonpartisanship, in order to be

just, must be a law that acts with intense respect and regard for

each of the multiple, concrete selves it touches. Just law must be

reconciliatory even as it is decisive over ethical and ideological

controversy, decisive even as it is reconciliatory, and the two

demands do not sit easily together. Demanded is a law that, while

welcoming to all ethical and ideological views and hostile to none,

is nevertheless capable, more or less objectively, of pointing to

solutions to hot controversies fueled by ethical and ideological

differences. It seems inevitable-at least to Rosenfeld and to

if not most, of its norms in the form of what writers on

jurisprudence currently call “standards” as opposed to “rules.”42

accentuated social conditions of ethical and ideological pluralism,

of conceiving of a largely standard-based legal practice that can

satisfy both requirements together, avoiding divisive partiality

without falling into watery and useless indeterminacy, avoiding

indeterminacy without falling into partiality.

We seek guides to justice in the legal interpretations of a

postmodern polity. Required, it appears, are criteria neither too

thick nor too slack. Too thick a criterion will be divisive, a wedge

where we need a bond, while too slack a one will fail to provide

the special kind of external-arbitral antidote to social and personal

disintegration we can sometimes get (if Rosenfeld is right) from

routing hot controversies to legal instead of communal-ethical or

majoritarian-political resolution. Inherited criteria that appear too

CHI. L. REV. 527 (1999).

MICHELMAN-BR WEBGALLEYS1.DOC 08/07/00 1:47 PM

slack will have to be stiffened up; those that appear too thick will

have to be thinned out. Historically, a main direction of reform

has been toward thinning out, as the direction of social change has

been toward ethical and ideological diversification-as a “postmetaphysical”

age has succeeded its metaphysical predecessor43

and a “fact of reasonable pluralism” has emerged to proceduralize,

relatively speaking, our view of what political justice might

possibly be.44

Unmistakably, the search of the dominant, recent liberalminded

political philosophy spoken for by John Rawls and Jürgen

Habermas has been for the thinnest criteria that can still check the

disintegrative forces of diversity.45 Because it is the contemporary

alternative with which Rosenfeld’s own proposal bears closest

comparison, we may take as illustration John Rawls’s idea of a

culturally situated, political consensus on a small, substantive set of

“constitutional essentials.” The cultural situation envisioned by

Rawls is that of a wide, historical public culture of constitutional

democracy. By a “political” consensus on norms of justice, Rawls

means, among other things, one that is freestanding from all

“particular” conceptions of the good, where “particular” means

public culture.

The constitutional-democratic culture supplies a common

bond of recognition by each self of all others as equally the

possessors of moral capacities or “powers” and corresponding

interests in their exercise-capacities both for adopting, revising,

and pursuing one’s own conception of the good and for

recognizing and embracing fair terms of social cooperation among

persons thus constituted. “Reciprocity,” let us say, is recognition

by persons of each other as being “free and equal” through their

moral powers. Reciprocity, it is argued, allows for the emergence

of a very abstract and general, “thin” public conception of the

good as consisting in a person’s adequate stocks (so to speak) of

certain “primary goods”-rights, liberties, and opportunities;

income and wealth; the social bases of self-respect-that anyone

presumably would want, whatever his or her conception of the

good and related aims in life.46

Engagement in the spirit and practice of reciprocity, then, will

45 On the shared liberal-mindedness of Rawls and Habermas, see their well-known

MICHELMAN-BR WEBGALLEYS1.DOC 08/07/00 1:47 PM

further allow-according to Rawls’s argument-for arrival at a

broadly inclusive, “overlapping” consensus, among all who are

moved by reciprocity, regarding a set of basic terms of social

cooperation bearing on the distribution of these goods, or what

Rawls calls “principles of justice” for the basic structure of society.

the part of all participants, a moral judgment (about what is owed

to each other by persons reciprocally recognized as free and

equal), not just a strategic one (about what will secure the better

outcomes or more favorable conditions for the self’s pursuit of its

own conception of the good). Only such a consensus can allow for

powers that define them as free and equal claimants to justice-

their powers of recognition of, and compliance with, fair terms of

social cooperation as well as their powers of adoption and pursuit

of a vision of the good.47 What is “overlapping” about such a

moral consensus, among otherwise conflicting conceptions of the

good, is its dependency on-its presupposition of-the occupation

by reciprocity of a place in all the participating visions of the good.

Conceptions that have no place for reciprocity are simply excluded

as “unreasonable.”48 Justice, in Rawls’s view, is the set of

principles to govern decisions about the basic structure of society

that is emergent from an overlapping moral consensus among

conceptions of the good that include a place for reciprocity.

Rosenfeld believes the crisis calls for drawing the criteria for

just legal interpretations from a normative doctrine, called

“comprehensive pluralism,” which he presents in sharp contrast to

Rawlsian political liberalism. I do not here undertake to give this

proposal the full introduction it deserves, or even to begin to

indicate the refinement, care, and subtlety of Rosenfeld’s full

exposition of it.49 I treat it strictly and crudely with a view to a

particular reservation I want to raise about the proposal, based (no

doubt unfairly) on my particular curiosity about how Rosenfeld

precisely means to differentiate it-as he outspokenly does-from

the “political” liberalism of John Rawls.50

Comprehensive pluralism, to begin with, is a frankly

substantive, frankly ethical, frankly contentious doctrine, one that

(1994).

MICHELMAN-BR WEBGALLEYS1.DOC 08/07/00 1:47 PM

posits a distinct, substantive theory of value or of the good, of what

is worth pursuing in human life. It conceives of the good as

success in the promotion of “peaceful coexistence among[] as

many competing theories of the good as possible.”51 We are not to

understand this as expressing a quasi-aesthetic sensibility that finds

beauty, stimulation, or richness of life in the vision of a thousand

flowers blooming. Rosenfeld is no liberal “perfectionist,”52 and his

comprehensive-pluralist conception of the good is not in that way

a freestanding one. It is rather parasitic on the “pluralism in fact”

that characterizes postmodern polities. The value-the good-

conceived by comprehensive pluralism lies exactly in solicitude for

the multitude of individual and collective selves that in fact there

are, and are bound to be, in all such polities. It lies in the very

avoidance, insofar as possible, of divisive and oppressive intrusion

upon those selves. The point about “pluralism as norm,” as

Rosenfeld succinctly put it, is that it is “best for pluralism in fact.”53

Comprehensive pluralism is all the same a “non-neutral”

conception of the good. It competes directly with all of the other

ethical conceptions that jostle for space and privilege in

postmodern polities. Seemingly paradoxically, Rosenfeld posits a

particular, partisan, fighting faith, the observance of which may

nevertheless bring a valuable kind of unity to a society inhabited

by many other partisan fighting faiths. Obviously, such a result is

not possible unless there is something to differentiate the ethic of

comprehensive pluralism categorically from all the other ethics

and ideologies whose reconciliation comprehensive pluralism

supposedly is going to effectuate. There must be some way in

which comprehensive pluralism occupies a different conceptual

plane from that of the other competitors. At the same time, the

differentiating notion must be one that does not disguise the

character of comprehensive pluralism as itself a nonnecessitated,

partisan and particular matter of ethical substance.54 It is to meet

this need that Rosenfeld has introduced the idea of a class of

“second-order” ethical conceptions, the members of which (but, of

course, there is only one of these, comprehensive pluralism) stand

in a different relation to all of the first-order ethical conceptions

than does any one of them to the rest (but, of course, the special

relationship can’t be that of neutrality between itself and any of

52 Compare JOSEPH RAZ, THE MORALITY OF FREEDOM (1996).

the best possible normative criterion for the reconciliation of self and other within a

pluralistic in fact society, in a way that maximizes the potential for justice while

minimizing that for violence.”).

MICHELMAN-BR WEBGALLEYS1.DOC 08/07/00 1:47 PM

them). But then what can this second-orderness possibly be?

How can comprehensive pluralism’s demand for placing all

conceptions of the good on an equal footing prevail over rival

ethical claims, given that its call is, after all, to place all

conceptions of the good on an equal footing?55

Is that, after all, such a hard question? There you are,

committed to a normative doctrine of maximum equal solicitude

for all conceptions of the good. You want to set that doctrine to

ruling over all conceptions of the good, in the sense of mediating

and arbitrating conflicts among them in accord with its precept of

maximum equal solicitude. It seems you can easily do so without

any hint of paradox or contradiction. You simply refuse to classify

your super-doctrine as a doctrine of the good. You classify it,

instead, as a doctrine of the right. That is how John Rawls does it,

to take a notable example.

Rawls, but not Rosenfeld. Rosenfeld decidedly refuses this

easy answer to what must then remain, indeed, a very hard and

puzzling question. Why does he? And what are the

consequences?

1. A Similarity of Opposites?

On the surface, at least, Rosenfeldian comprehensive

pluralism bears many striking resemblances to Rawlsian political

liberalism. At the core of both doctrines is the teaching that

justice in law and politics depends on our finding a higher-order

normative conception that, while frankly substantive, is also

sufficiently thin to accommodate a broad range of extant ethical

views. Rosenfeldian second-orderness is, in this respect, a gesture

to thinness, structurally comparable to Rawlsian overlapping

consensus.

Both theories also include a resigned acceptance, even so, of a

filter for “unreasonable” ethical views (as Rawls calls them), views

insufficiently committed to what both authors call “reciprocity,”

meaning acceptance of the full equality of the claim of every

individual (and, for Rosenfeld, collective) self to respect and

accommodation from others.56 Both theories use a notion of

(“[A] crusading religion, for which conversion of the infidel, by force if necessary, is a

213-14 (calling reciprocity a central value in comprehensive pluralism).

MICHELMAN-BR WEBGALLEYS1.DOC 08/07/00 1:47 PM

primary goods.57

But of course it is those things that are structurally most

similar that also stand to be most sharply opposed in meaning-

and opposition, not identity, of meaning is how Rosenfeld

understands the relation between the comprehensive pluralism he

sponsors and Rawlsian political liberalism. The whole story lies

compacted in the key passage I quoted near the outset:

[T]o the extent that comprehensive pluralism amounts to yet

another conception of the good, it is clearly distinguishable

from contemporary Kantian visions such as those Habermas or

extent that comprehensive pluralism is more than yet another

conception of the good, in that it unleashes an inevitable but

ultimately resolvable dialectic between first-order and secondorder

norms, it differs markedly from theories based on

building an “overlapping consensus,” such as that elaborated by

reducible to mere relativism.58

2. The Right and the Good

We need some conceptual tools for unpacking this dense

passage. One that will come in handy is the Rawlsian notion of

“the priority of the right over the good.”59

The question of “the right” asks what ought to be done, all

morally relevant factors considered, while the question of “the

good” asks what is of value to a person, group, or society, or what

contributes value to a person’s life, a group’s existence, or a

society’s history. In the field of social policy, at least, these two

questions have appeared to many not to be identical. A chief

reason is that, even if we had universal agreement on, say, glory as

the ultimate good for humankind, we would still have to face a

distributional question: Are we to aim for the glory of society as a

whole, or for that of the individuals and groups composing it? If of

individuals, is our aim to be the greatest achievable aggregate sum

of individual glory scores? The greatest average glory of

individuals? Equality of glory among individuals? The greatest

glory of the least glorious person? Or what? Resolution of these

questions of aggregation and distribution seems necessarily to

depend on considerations that cannot already be contained in the

idea of the ultimate goodness of human glory, and those

considerations, then, are considerations of the right as

MICHELMAN-BR WEBGALLEYS1.DOC 08/07/00 1:47 PM

distinguished from the good.

Speaking more intuitively, the notion of the priority of the

right over the good is that there are certain constraints on action,

including political action, that persons owe to each other out of

reciprocal regard for one another as free and equal individuals,60

regardless of what idea of the good anyone (or everyone) may

happen to hold. Compliance with the demands of the right is thus

conceived as preconditional to anyone’s and everyone’s normative

permission to pursue the good, in something like the way in which

compliance with the legal norms in a constitutional bill of rights is

preconditional to legal permission to pursue public policy

objectives through lawmaking. The term “deontological” is often

used to designate the class of normative theories that posit such a

relationship of right-before-good preconditionality.

3. Deontology and Oppression

With this notion in hand, we return to our key passage. Let us

focus for a bit on the first sentence. It says, in effect, that

Rosenfeld is not a deontological theorist-that Rosenfeldian

comprehensive pluralism is not to be regarded as a doctrine of the

right, as such to be differentiated from, or prioritized over, any and

all doctrines of the good. True (anticipating the second sentence

in our key passage), the pursuit of comprehensive pluralism has

“second-order” status as compared with the first-order status of all

other ethical commitments, but the first sentence, by its

repudiation of “Kantian” views, tells us that the second-order/firstorder

deontological theories posit between compliance with the right

and the pursuit of the good. It cannot be, if comprehensive

pluralism is what Rosenfeld tells us it is-a conception of the good,

competing, as such, with other conceptions of the good.

Of course, this first sentence does not stand alone, and its full

meaning cannot be extracted without consideration of its

qualifying successor. It is nevertheless worth pausing briefly,

before moving on, to ask what might motivate this rather preciouslooking

insistence on Rosenfeld’s part on distancing his theory

from the deontological (“Kantian”) tradition. A likely answer lies

in a concern about oppression. When we do come to the next

sentence, we shall find embedded in it a cautiousness about the

way in which ostensible doctrines of impartiality among

perspectives may end up giving some perspectives the upper hand

over others. Worries of that sort light easily upon normative

MICHELMAN-BR WEBGALLEYS1.DOC 08/07/00 1:47 PM

doctrines that appear to privilege themselves, as doctrines of right,

over others that are plainly and frankly mere doctrines of the

good. It is therefore no surprise to find Rosenfeld rejecting any

such claim of privilege for comprehensive pluralism.

And yet, as a criterion for just legal interpretations, defined as

interpretations that “make for a fair equilibrium between the

respective needs, interests, and aspirations of the self and those of

the other,”61 comprehensive pluralism obviously cannot be just

another conception of the good amid all the rest. It must also be

something special, second-order, reminiscent of a Rawlsian

overlapping consensus. And so it is. But the burden of the second

sentence of our key quotation is that the structural similarity to

Rawls is decidedly of the oppositional kind.

In that sentence, Rosenfeld points to the “dialectic” that

comprehensive pluralism sets going between itself and all of the

first-order ethical conceptions as the factor that differentiates

comprehensive pluralism from all competing theories of justice for

postmodern polities. On that basis, he sets comprehensive

pluralism in opposition to “theories reducible to mere relativism.”

conflict or contention; a “dialectical” relation between norms is

always and interminably a contentious one. True, the term may

also signify a perception that the conflict occurs within a higherlevel

harmony, but we don’t need that wrinkle to see how an

ethical norm that can enter into contentious relations with others

stands opposed to relativism. “Relativism,” after all, is defined by

semantic contact, contentious or non-contentious, between any

pair of ethical norms. It is the view that “all value preferences are

ultimately purely subjective and so contextually bound to the

conception of the good from which they emerge that it would be

meaningless to seek to gauge them from the standpoint of any

other perspective.”62 Comprehensive pluralism is obviously antirelativist,

then, in the sense that it is itself a committed ethical

stance from which to “gauge” and deal with conflicts among other

ethical views and between any of them and itself.

pondering the debate over the moral permissibility of abortions, or of their prevention by

the killing of those who perform them, where the question is cast as that of whether either

class of acts is correctly subsumed under a commonly accepted norm prohibiting murder.

The case, he suggested, may be better described as one of “an actual relativism derived

from an insoluble conflict among irreconcilably conflicting conceptions of the good” than

MICHELMAN-BR WEBGALLEYS1.DOC 08/07/00 1:47 PM

Rosenfeld speaks not just of differentiating comprehensive

pluralism from relativist theories, but, more complicatedly, of

differentiating it from “theories based on building an ‘overlapping

liberalism’s membership in the family of theories that are

“ultimately reducible” to relativism. But we ought not thus to

read it. Critiques of political-not-comprehensive liberalism occur

elsewhere in Rosenfeld’s text, and they show clearly that

Rosenfeld cannot intend an accusation of relativism against the

Rawlsian doctrine.

It is precisely in relation to the idea of a political-notcomprehensive

normative (“overlapping”) consensus that

Rosenfeld expressly develops his concerns about the oppressive

potential of normative doctrines that present themselves as

different in kind, or quality, from frankly partisan conceptions of

the good. Rosenfeld rejects, as incapable of sympathetic response

to the full range of ethical and ideological commitments that

postmodern polities have to accommodate, a conception of justice

as “mere” reciprocity. Such a conception works by “reducing” self

and other to the sameness of being each a “possessor of a

perspective,” while glossing over the concrete realities of those

perspectives. It excessively “stresses identities at the expense of

differences.”63 It will fail to achieve justice because it “lacks the

means to constrain a dominant self to acknowledge the needs and

aspirations rooted in the perspective of the other as opposed to

those traceable to the other possessing a perspective.”64 In other

words, the strategy of prescinding from concrete conflict to a

search for more abstract principles that all can accept in the spirit

of reciprocity is objectionable for its indulgence of the self in

refusal to confront and come to terms with otherness in all its

living specificity and concreteness.65 Such a strategy, Rosenfeld

believes, is bound to end, when push comes to shove, in concrete

cases, by giving some perspectives the upper hand over others

under an inevitably false guise of neutrality among perspectives.66

(The question of holding principles separate from their

applications is now upon us with a vengeance.) In sum, Rosenfeld

finds potentially oppressive Rawls’s claim that a quasi-universally

SELF 148, 158-59 (1992).

MICHELMAN-BR WEBGALLEYS1.DOC 08/07/00 1:47 PM

cognizable constitutional-democratic culture-the common

culture, as Rawls conceives it, of law-governed, free societies-can

support a substantive conception of justice that is nevertheless

strictly “political,” not itself reflecting any conception of the good

so distinctive or “particular” that members of society will

experience it as a partisan, fighting faith among other partisan,

fighting faiths. Rosenfeld cannot, therefore, be thinking of the

political-liberal stance as one that is tantamount to relativism, the

meta-ethical doctrine most apparently designed to be guaranteed

free of all such oppressive potential.

To the contrary, anti-relativism is a shared feature of the

justice theories of Rosenfeld and Rawls, one to be added to the list

the relativist route to avoidance of potential for oppressiveness.

And the question still hangs, then, about what the crucial

difference, in this respect, is supposed to be between them.

Exactly how is it that committed, anti-relativist, comprehensive

pluralism avoids the potential for oppressiveness that committed,

anti-relativist, political liberalism involves? Taking Rosenfeld at

his word, the answer lies in the “inevitable but ultimately

resolvable dialectic between [itself] and [other] ethical

conceptions” that comprehensive pluralism sets up but that a

consensus-based, deontological doctrine like political liberalism

does not.

Of course, that response, while asserting a crucial difference

between the two normative doctrines before us, also again reminds

us of the structural similarity between them. Both are two-level

constructions. Both locate the possibility of justice and legitimacy

in politics in the relationship they contemplate between the norms

that respectively occupy their higher and lower levels. For

Rosenfeld, that means the openly contentious relationship

between the uniquely second-order ethical conception of

comprehensive pluralism and its first order competitors. For

Rawls, it means the ideally inclusionary relationship between the

“political” consensus on principles of justice and the

comprehensive ethical views participating in the consensus.

We have seen that it is this very holding-out of the possibility

of a public conception of the right in politics, to which all

“reasonable” comprehensive views can willingly and

wholeheartedly subscribe, that Rosenfeld finds dangerous in

Rawlsian theory. And yet the puzzle remains, for Rosenfeld, too,

needs a second-order normative conception that will, at the

MICHELMAN-BR WEBGALLEYS1.DOC 08/07/00 1:47 PM

moment of application to contested cases, achieve “reconciliation”

among the contesting first-order ethics and ideologies. He needs,

as his words indicate, a second-order conception whose

contentious relation to any of the first-order ones is ultimately as

“resolvable” as it is “inevitable.” If Rosenfeld is not speaking in

riddles, only one possibility remains: In Rosenfeld’s parlance,

“reconciliation” (or “resolution”) between comprehensive

pluralism and the first-order ethical conceptions over whose

conflicts it mediates does not consist in any sort of agreement on

the latter’s part to the former’s claim to be correct as a theory of

the good.

But how, then, are we to make sense of this notion of ethical

reconciliation without ethical agreement, as a matter of the actual

experiences of selves in society? On an experiential level, how can

any self-respecting self, confronting others whose projects conflict

with its own, allow itself to be “reconciled” to legislatively and

adjudicatively imposed arbitrations, when those arbitrations

avowedly have been cast to meet the requirements of a conception

of the good with which that self decidedly does not agree? What

do we imagine going on with that self at the moment of

reconciliation?

from impossible to understand. What is fair among persons

committed to sundry conceptions of the good is one sort of

question. What the good is for me, for you, or for humankind is

quite obviously an entirely different sort of question. One need

not in the least accept the claims of comprehensive pluralism to

state a correct theory of the good, in order believe that

comprehensive pluralism delivers fair results when elaborated into

a set of guides for the arbitration of disputes among conflicting

selves, defined in terms of the conflicting conceptions of the good

to which they hold.

I put the matter as I have in hopes of showing how the

dialectical kinship of Rosenfeld and Rawls persists, even to this

penultimate point in our analysis. Rawls, after all, would be the

last to deny the possibility of every (reasonable) self’s finding, in a

normative doctrine that many of these selves do not accept as a

in the arbitration of conflicts among their diverse ethical

MICHELMAN-BR WEBGALLEYS1.DOC 08/07/00 1:47 PM

commitments. Nothing, Rawls might well say, could more

good) and the “priority” of the former over the latter. What

Rosenfeld has done, Rawls might well say, for all intents and

purposes, is to make comprehensive pluralism figure for the self in

question as a doctrine of the right, privileged in the way that any

such doctrine always is with respect to doctrines of the good.

Can a reconciliation of Rosenfeld and Rawls really be had on

those terms? Does not the first sentence of our key quotation

exclude it, by rejecting, on anti-oppression grounds we have

reviewed, all reliance on the Kantian idea of the priority of the

right over the good? Maybe so, maybe not. If the only exegetical

test here is one of analytical coherence and consistency, then

perhaps the first sentence appears satisfiable by splitting the

standpoints of dispensers and receivers of justice. We could

stipulate that comprehensive pluralism, and the second-order

doctrines of the good and only as doctrines of the good. They

and not of the good.

Interestingly, Rosenfeld’s text does ask us to hold these two

standpoints apart. “[C]omprehensive pluralism,” Rosenfeld

writes, “maintains a split between demands for justice predicated

on first-order norms and legitimate endeavors to dispense justice,

which must look, above all, to second-order norms.”68 It “requires

the legal interpreter to be guided by second-order norms while

urging those who submit claims . . . to remain as true to the firstorder

norms to which they are committed as is possible consistent

with genuine acceptance of the dictates of legitimate judicial

adjudication.”69 As far as they go, those words seem compatible

Rawlsian overlapping consensus, among multiple conceptions of

potential that Rosenfeld finds in the idea of the priority of the

right would seem to be at least in some measure averted.

It may be a strained reading, but consider the alternative.

Rejecting it would force us to rewrite in starker form, as follows,

our experiential rendition of the Rosenfeldian notion of

MICHELMAN-BR WEBGALLEYS1.DOC 08/07/00 1:47 PM

escape by having the justice-receiving self regard comprehensive

pluralism as an above-the-fray, noncompetitive doctrine of the

right and not of the good.) And then what we would have is an

exquisite expression not of the priority of the right over the good,

but of something else, something opposite-namely, a Rawlsian

by all of a principle or principles for the guidance of legal

interpretation, on strictly strategic grounds and without any

supporting normative agreement, whether on a conception of the

good perceived to underlie the principles or on the principles

themselves as an apt expression of the right.

In postmodern conditions, it seems, we do not look for

convergence on ideas of the good. Normative convergence is

possible, if at all, only on an idea of the right standing free from

ideas of the good. The difference between Rosenfeld and Rawls, if

there is one-a question I leave open-is that, where Rawls sees a

way to reconciliation through agreement on principles understood

to be principles of right and not of good,70 Rosenfeld does not.

The idea of the right prior to the good is, for him, too fraught with

risk of subverting the achievement of a “fair equilibrium between

the respective needs, interests, and aspirations of the self and those

of the other,” or what he calls justice.71

If that is finally the way we read him, then Rosenfeld’s

message is that, in the pursuit of social unity and cohesion in

more is unduly to risk subverting the pursuit. The aim is to

redeem what we can from postmodern ethical and ideological

question to ponder.