Copyright (c) 1994 The Board of Trustees of Leland Stanford Junior University
Stanford Law Review
46 Stan. L. Rev. 1807
LENGTH: 15353 words
REVIEW ESSAYS: The Subject of Liberalism
Political Liberalism. By John
Rawls. * New York: Columbia University Press. 1993. 371 pp. $ 29.95.
Professor of Philosophy, Harvard University.
Frank I. Michelman *
* Robert Walmsley University Professor, Harvard University. Thanks to
Joshua Cohen, Owen Fiss, and Michael Stocker for helpful discussion, and to Bruce Ackerman,
Joshua Cohen, and Michael Sandel for sharing their drafts on
Political Liberalism, the influential moral philosopher John
Rawls attempts in part to justify a particular, liberal conception of justice by
arguing that it is the conception which best corresponds to the way we perceive
our powers and motivations when we behave politically. ... When
Rawls builds a justificatory argument for justice as fairness, he starts from a
distinctly Kantian view of personality - the
"political conception of the person" - that seems to carry us well outside the irresistible core of thin liberal
subjectivity, into apparently much more controversial territory. ...
Political Liberalism, the influential moral philosopher John
Rawls attempts in part to justify a particular, liberal conception of justice by
arguing that it is the conception which best corresponds to the way we perceive
our powers and motivations when we behave politically. In this review essay,
Professor Michelman suggests that
Rawls' method may offer a way out of the debates over subjectivity and value
currently occupying many legal scholars. By starting from a depiction of the
liberal subject as the kind of person for whom constitutional democracy is the
appropriate form of social ordering, Professor Michelman observes,
Rawls constructs an
"internal critique" of American society which depends on the practical, cultural appeal of that
view of the person, not on its ultimate truth. However, Professor Michelman
concludes that whether
Rawls' approach is necessarily relativist - speaking only to those who already
believe in liberal constitutionalism - or instead produces universally
generalizable prescriptions remains unclear.
A. Meet the Political Subject (You?)
Try on for size this proposed description of yourself as a political person.
n1 To begin with, you possess two
n2 You have
"a capacity for a sense of justice,"
n3 that is, an ability to understand and act from some
"public conception that fixes fair terms of social cooperation." You also have
"a capacity for a conception of the good,"
n4 that is, an ability to form, revise, and rationally pursue your own conception
of what gives value to your life. Conscious and respectful of your moral
powers, you want to develop and use them; indeed, you rank their development
and exercise among your life's guiding or
n5 You have, in fact,
"conception-dependent desires," meaning desires to be the sort of person who lives the sort of life to which
[*1808] moral powers beckon: who acts according to principle and does so both
"rationally," in accord with your own conception of the good, and
"reasonably," in accord with a public conception of justice to which you are willingly
Thus are persons politically conceived, according to John
Political Liberalism. Does he mean you? Is this you?
What is Professor
Rawls trying to prove in his recent political philosophy? In particular, what is
Rawls doing with the set of descriptions that he styles
"the political conception of the person"? What is a
"political" conception of the person? These are key questions for those who ponder the
Rawls' work on their own professional and ethical concerns.
B. Lawyers Meet
Lawyers and legal theorists have fairly obvious, broad reasons for trying to
get at least roughly clear
Rawls' philosophic intentions and methods. Consider the suggestion that
constitutional jurisprudence from the Warren and Burger Courts is traceable in
Rawls' philosophy as set forth over twenty years ago in A Theory of Justice.
n7 Supposedly, law professors under
Rawls' spell helped spread his egalitarian-liberal ideas to the United States Reports
by propagating them in law reviews and in classrooms where judicial clerks are
spawned. On this point, the distinguished political theorist Alan Ryan wrote in
a recent review:
Rawls's ideas have crept into the law of the land. Liberalminded lawyers keep pushing
the envelope of the Constitution, trying to expand Americans' civil liberties,
but they don't at the same time encourage the courts to favor the rights of
property developers. One reason is that they have been taught that liberal
ideals of justice do embrace civil rights and economic equality but do not
embrace laissezfaire and the unfettered rights of property.
I do not know to what extent this influence thesis is true. The formation of
the characters and views of William Douglas, William Brennan, and other leading
liberal jurists predates A Theory of Justice. The Warren and Burger Courts
could have risen and flourished (insofar as they did) absent the influence of
Rawls or any comparable philosopher. The causes of those Courts' careers could lie
quite elsewhere, and the rise of the Warren Court itself may help to explain
the spread of egalitarian liberalism in the law schools.
It is nonetheless true that
Rawls' work has engaged law professors' attentions mainly as a (controverted) source
of support for egalitarian-liberal constructions of the Bill of Rights.
Rawls, however, has not been exclusively
[*1809] concerned with promoting a particular substantive conception of political
justice; he has also been digging deep into questions about how one can
possibly provide a philosophically adequate justification for any conception of
this kind in the apparent conditions of normative and visionary dissensus that
characterize modern liberal societies. His excavations, especially those made
in a series of 1980s writings beginning with his Dewey Lectures
n11 and now reworked and synthesized in
Political Liberalism, demand the attention of anyone exploring what it means to be engaged in
liberal political argument.
n12 Here I want to consider the possible bearing of
Rawls' work on a current debate in legal scholarship concerning the value and
defensibility of any attempt at systematic normative argumentation.
C. The Problem of the Subject
In law schools and law reviews today, debates about subjectivity are full upon
"Subjectivity" here connotes self-authentication
n14 and self-command. A
"subject" is a self in charge of its own thoughts and actions - a spontaneous author of
plans and doer of acts inspired by its own cognitions, calculations, and
desires. Subjectivity is banal: Of course lawyers argue, judges judge,
lawmakers legislate, and law professors theorize and prescribe as if it were
unquestionably true that humanity is composed of individual subjects. We
commonly speak and write without a care for the doubts cast by the
post-Enlightenment on the idea that all individuals really do stand severally
in control of their thoughts and behaviors.
n15 Some contemporary critics of subjectivity (and, relatedly,
n16 ) in legal scholarship perceive this
[*1810] heedlessness as a pathological condition that deserves attention. As they also
see, however, their intervention itself is problematic on its own terms. How
could anyone utter the call for attention, how could anyone give or have reason
to give attention to anything, if not apprehending himself subjectively, as a
consciousness sponsoring its own thoughts and exertions? The critique of
subjectivity performatively affirms the subjectivity it questions.
n17 In short,
"the problem of the subject," for legal scholarship generally, is the sense that we cannot abandon the idea
of our own subjectivity but also (perhaps) can't strongly defend it against
scientific and philosophical critique.
It is not my purpose here to attempt a resolution of the problem of the
n19 Rather, I propose that a study of
Rawls' work, particularly his argumentative deployment of a certain description of
individual personality, can shed some light on the problem.
Rawls visibly builds into his philosophy a particular, strong idea of the liberal
subject, but he does it so circumspectly, indeed, so ostensibly noncommitally,
as to suggest how the problem of the subject may more generally be managed in
contemporary liberal thought.
II. Justification, Personality, and Truth
A. Constitutional Justification and Conceptions of the Person
Political Liberalism in part to the justification of a particular
"political conception of justice" that he calls
"justice as fairness." By a political conception of justice
Rawls means, in the first place,
n21 a set of principles meant to regulate a society's
"basic structure" - its
"main political, social, and economic institutions, and how they fit together
into one unified system of social cooperation."
n22 Many will recall from A Theory of Justice (or from
[*1811] countless allusions to the book since its publication) the famous two
principles of justice as fairness:
a. Each person has an equal right to a fully adequate scheme of equal basic
liberties which is compatible with a similar scheme of liberties for all.
b. Social and economic inequalities are to satisfy two conditions. First, they
must be attached to offices and positions open to all under conditions of fair
equality of opportunity; and second, they must be to the greatest benefit of
the least advantaged members of society.
The principles in a political conception are designed to fix criteria for the
society's discursive resolutions of all questions of
Political Liberalism, to this extent, is an essay in normative political theory, offering
justification for a particular regulative conception of the political
constitution. No one, presumably, takes up such a work of justification who
does not think it matters what constitutional conception is in force. But how
do constitutional conceptions matter, if they do? Presumably, they matter
because of how we think political constitutions affect the values of people's
"under" them. It seems, then, that to justify one or another constitutional conception
must be to assert some sort of favorable relation between that conception and
the values of lives.
What sort of relation depends on how we conceive of human-life value. If value
signifies experiential satisfaction, then the better-justified constitution is
the one that is more conducive to the production of satisfaction. But if value
is of the sort that Ronald Dworkin calls
n28 - measured by people's success in living as they know they ought, as opposed
to living as they simply wish - then the better-justified constitution may be
the one that better serves our critical interest in living under conditions of
justice or in ways that express respect for human dignity.
Here we can leave this question open. Let the measure of life-value and the
correlative justificatory relation be what they may. We realize, in any case,
[*1812] in order to speak cogently of how one or another constitutional conception
connects with the values of lives, the speaker can hardly refrain from positing
some features of the persons whose lives they are.
B. Liberal Personality: Minimal and Beyond
As normally conceived in liberal throught, persons, I suggest, have the
following attributes: They are ethically several, interest-bearing,
self-activating, communicative, and self-conscious (or self-reflective). To
Ethically several: Each subject leads a life of its own that is conceptually
distinct from the lives of other subjects, and each of these several lives is a
conceptually distinct field of value.
"Conceptually distinct" means that in contemplating the values of lives we take them one by one. This
does not rule out causal linkages or require that the values assigned to any
one subject's actions or experiences be held independent of values assigned to
others' actions or experiences.
Interestbearing: Various events may differentially affect each subject for
better or worse, so that we can judge events as enhancing or detracting from
the thriving or wellbeing of subjects taken one by one. Again, this does not
rule out empirical linkages between the interests of one subject and the
interests of others.
Selfactivating: Each subject can contemplate its interests, discover and
present to itself corresponding reasons for action, and act or alter its state
accordingly. This does not rule out joint, coordinated, reciprocal, or
otherwise co-responsive reasons and actions.
Communicative: Each subject can intentionally affect other subjects'
perceptions of reasons for action and may have its own perceptions
intentionally affected by others.
Selfconscious (reflective): Each subject is aware of itself as an ethically
several, interestbearing, selfactivating, communicative subject. This does not
rule out that individual selfconsciousness may depend on reciprocal awareness
by subjects of each other.
Since all of these attributes would be indisputable common ground for every
liberal, spanning the range from the proto-liberal Hobbes
n29 to the super-liberal Unger,
n30 we may call this conception of the subject a minimal or
n31 liberal conception. A more robust conception of the person would be one that
drew additional attributes into subjectivity, ones not so conspicuously
irresistible to inhabitants of liberal culture as the ones we have so far
compiled. In contemporary egalitarianliberal political philosophy, two sorts of
more robust subjectivities appear, which we can somewhat loosely call
"Kantian." On the romantic side, we find a subject so conditioned that thriving for it
essentially consists of selfcreative accomplishments such as authoring
[*1813] one's life or setting apt challenges to oneself and meeting them well.
n32 On the Kantian side appears a subject so conditioned that it cannot thrive
without having before it principles of regard for the interests of particular
fellow subjects or of all subjects - principles of right or morality or justice
- to which it can precommit itself and from which it can then draw preemptive
reasons for action.
Suppose, for the moment, that we take these romantic and Kantian motivational
attributions as claims about how people just are,
"by nature," so to speak. Then, while the attributions may not seem outlandish, neither do
they seem irresistibly self-evident. Many will find closer to irresistible
(although still adding to the minimal liberal conception) Hobbes' picture of a
humanity moved only by aversion to pain, attraction to ease, and imaginatively
inflected calculations of the means and prospects of avoiding the one and
securing the other.
n34 By comparison, the romantic and Kantian depictions may seem to stand out as
specialty flavors of humankind.
C. Conception and Truth
To speak of the liberal subject is to call to mind the sort of being theorists
have in view when they undertake to justify a liberal constitutional conception
by showing some benign relation between the favored conception and the values
of human lives.
n35 Liberal constitutional justifiers do irresistibly think (and confidently
expect their audiences to think) of persons as subjects. That is, we
(justifiers and audience) irresistibly view ourselves and other individuals
each as a point of origin of consciousness, choice, and action, and this
self-understanding closely shapes our constitutional prescriptions.
n36 One of
Political Liberalism's striking features is its apparent attribution to humankind, as a basis for its
claims on behalf of liberal constitutionalism, of features going well beyond
"default" subjectivity. When
Rawls builds a justificatory argument for justice as fairness, he starts from a
distinctly Kantian view of personality - the
"political conception of the person"
n37 - that seems to carry us well outside
[*1814] the irresistible core of thin liberal subjectivity, into apparently much more
Yet notice this interesting possibility: Perhaps it needn't be that we hold
the more fulsome Rawlsian conception of the person (or even, for that matter,
minimal liberal subjectivity) to be a simple fact of the human condition.
Perhaps all that's required to justify a constitutional conception is that we
treat a certain depiction of personality as true of us and our co-citizens just
insofar as we're either engaged in conceiving and defending a normative
doctrine of the state, or we're involved in focused participation as citizens
or officials in the range of political activities that such a doctrine
n38 This, then, would be one way to qualify as
"political" the Rawlsian conception of the person.
Political Liberalism takes this possibility seriously. The book is (or includes) an essay in
fitting a specific constitutional conception to a specific conception of the
person, while avoiding any claim that the personal conception is true in the
sense in which we hold something to be true in our cognitive dispositions as
inquirers into the ultimate - as scientists, religionists, or first-things
n39 Sufficient for the purposes of liberal constitutional justification,
Rawls aims to show, is that the personal attributes of subjectivity, moral powers,
and corresponding conception-dependent desires are true of us
"politically speaking." By this he means that these attributes register how we do and must regard
ourselves, on due reflection, when what's specifically on our minds is how we,
as citizens in a liberal (constitutional-democratic) regime, stand vis-a-vis
political society and its institutional manifestations of state and law.
Needless to say, the principles of justice as fairness
n41 are liberal principles. They feature strong commitments to basic individual
rights and liberties, to limited government, to acceptance of difference and
even conflict among people's conceptions of the good in human life, to securing
broad latitude for people's endeavors to live their lives according to their
varying conceptions of the good, to a rule of law and equality before the law,
and to entrenchment of these commitments in a political constitution.
Equally obviously, justice as fairness is a doctrine of the political,
concerned with the political fundamentals - the basic structure, constitutional
essentials, and questions of basic justice. For
Rawls, however, the
"political" character of a normative conception resides not only in the politicality of
its subject matter. Rather, the conception must also be
"freestanding" vis-a-vis what
"comprehensive" moral, philosophical, or religious doc
n42 A conception is political in this second sense insofar as it is expounded
without reference to any broader doctrine of ethics in general, of human
flourishing, or of the good life for humankind.
n43 Of course, it is controversial how far political concerns can ultimately be
expounded and defended without appealing to ethical and metaphysical ideas.
Rawls finds compelling reasons to hold political conceptions as distinct as possible
from comprehensive views of reality and value, so that justifications for the
former depend as little as possible on partisan commitment regarding the
In addition to subject matter and expository self-sufficiency,
Rawls cites a further respect in which a normative conception can qualify as
"political." A conception is political in this third sense insofar as
"its content is expressed in terms of certain fundamental ideas seen as implicit
in the public political culture of a democratic society."
n46 A political culture is composed of
"the political institutions of a constitutional regime and the public traditions
of their interpretation."
n47 It is from constitutional-democratic political culture that
Rawls expressly draws the core of the political conception of the person. We start,
Rawls says, by looking to this culture for a shared fund of basic normative
n48 This shared fund includes the ideas that persons are free and equal, and that
as such they intend their political association - in and through a state - as a
fair scheme of cooperation.
n49 Reflecting on these two ideas about ourselves as we stand politically,
Rawls says that we are led to ascribe to persons the two moral powers.
Let us now try to gather some thoughts. As a political conception of justice,
justice as fairness aims to regulate society's determinations of constitutional
essentials and basic justice. It is, thus, a normative conception of the
political constitution. Accordingly, the philosopher commending justice as
fairness as the most fitting conception
n51 would seem to be setting himself a task of attributing features to persons,
"us" for whom the conception is said to be most fitting and the features of us by
reason of which this fittingness is said to obtain.
Rawls has seemingly attended to this task, and the ostensible result of his
attentions is the political conception of the person as endowed with the two
moral powers and correlative interests and motiva
n52 A reader can hardly help taking this depiction as
Rawls' presentation of the liberal subject, the archetype to whose attributes and
Rawls' favored political conception (justice as fairness) is meant to be fitting. Yet
Rawls ecumenically refrains from asserting the truth of this depiction. Instead, he
claims for it a different sort of credential - recognizable correspondence with
how subjects of constitutional-democratic culture represent themselves
Political Liberalism has on offer an intriguing project in constitutional justification: one that
does argue for the fittingness of a particular constitutional conception to
persons according to a certain conception of them, yet does not assert that
this conception of persons is (comprehensively) true. The rest of this essay
attempts to gauge
Rawls' success in this endeavor. Complete success, it seems, could potentially cut
much of the ground from under contemporary critiques of subjectivity in legal
III. The Status of the Subject
A. Politicality and Possibility:
"The Problem of
At the outset of his new book, Professor
"the problem of
political liberalism," cast in terms of a question about possibility:
How is it possible that there may exist over time a stable and just society of
free and equal citizens profoundly divided by reasonable though incompatible
religious, philosophical, and moral doctrines? Put another way: How is it
possible that deeply opposed though reasonable comprehensive doctrines may live
together and all affirm the political conception of a constitutional regime?
If one reads this formulation to make political stability (on liberal terms)
the ultimate quest for liberal constitutionalism, it may be misleading.
Stability is indeed one of
Rawls' main concerns, but he conceives that stability may flow from either of two
types of agreement. One is a true, heartfelt, and conscious moral consensus on
the regnant political conception. The other is a strategic compromise or
balance of forces ("modus vivendi") among participants conscious of moral divisions that preclude affirmation by
all of the morality of one and the same constitutional conception. What
Rawls seeks is agreement of the first kind, which he calls
"overlapping consensus." The aim is not only that adherents of opposed comprehensive doctrines should
all affirm the same political conception, but that they should all affirm that
the political conception is truly just for their plural society, hence as not
only rational but also reasonable in the sight of them all.
n56 Only then,
Rawls evidently means, have we a fully satisfying solution to the problem of
Behind the problem lies what
Rawls calls the general fact of reasonable pluralism: that under conditions secured
by free institutions, there inevitably arise and persist
"a diversity of conflicting and irreconcilable - and what's more, reasonable -
n57 A related general fact, one that plainly helps drive the Rawlsian attempt at a
form of constitutional justification that doesn't depend on the comprehensive
truth of any correspondent conception of persons, is the fact of oppression:
that only savage application of state power could possibly bring all the
inhabitants of a modern plural society to heel behind a single, spiritually
unifying, comprehensive religious, philosophical, or moral view.
Now, one might take
Rawls' question about possibility in at least three distinct senses that we may call
practical, transcendental, and ideological. Most straightforwardly, one can
take the question as a practical one, simply asking for a solution to the
problem it sets out. One form of a solution would be a substantive conception
of political justice on the order of justice as fairness. A statement of
principles for governing constitutional essentials and matters of basic justice
would practically answer the possibility question insofar as one could show
that the statement, publicly acknowledged and respected in practice, could
bring and hold together a society of conflicting comprehensive views on terms
that all could accept as reasonable.
Alternatively, one can take
Rawls' possibility question in what philosophers might call a transcendental sense.
n60 The transcendental question asks not what is to be done, but rather what would
hypothetically (if unascertainably) have to be true of humankind in order that
a constitutional-democratic ordering for a modern plural society could persist
peacefully and stably through time. The question asks not about the content of
a solution to the problem of
political liberalism but about conceptual possibility conditions for a solution. Just as
"the political conception of justice as fairness" evidently stands as
[*1818] swer to his possibility question taken in the practical sense, so
"the political conception of the person" may be his answer (or a part of it) to the same question taken
Rawls' possibility question seems an inquiry headed for counterfactual speculation,
destined to land in the always somewhat mysterious zone of the necessary but
unascertainable. But transcendentally is not the only possible way in which to
understand a question about conceptual preconditions for a solution to the
political liberalism. This might also be an empiricist's question, directed to social facts of
belief. Construed in that way, the political conception of the person stands as
Rawls' account of certain ideas and a certain image that citizens of liberally
constituted political societies do actually hold of themselves,
n62 It is the holding of these ideas and the particular image by the citizens,
then, that makes possible a solution to the problem.
To call this an account of ideology is to think of
Rawls himself seeing through the citizens' self-image he describes. It is to see
Rawls thinking of citizens prompted by a sense of moral need to conceive of
themselves and their fellows in a certain, Kantian way: This image, appearing
to the citizens as commonly shared, enables them, despite their awareness of
deep moral, philosophical, and religious divisions, severally to affirm as
reasonable and rational the same set of constitutional principles.
To gain an adequate grasp of
Rawls' political philosophy, I believe we must understand him as intending his
answers to the possibility question in all three senses - practical,
transcendental, and ideological.
B. Possible Senses of the Political Conception of the Person
It would follow, then, that
Rawls intends his
"political" account of
"the person" in two senses simultaneously. The first (transcendental, counterfactual) sense
is that of a statement of attributes that would hypothetically have to be true
of persons in a population if both of the following were also to be true of
that population: (a) the facts of reasonable pluralism and oppression apply to
it (as do the burdens of judgment), but (b) there exists for it, even so, the
possibility of solving the problem of
political liberalism. The second (empirical, ideological) sense is that of a statement of something
that inhabitants of constitutional-democratic political culture in fact believe
about themselves (politically speaking) that enables them to find a solution to
These two possible intended senses - transcendental and ideological - of
Rawls' Kantian attribution to persons of the moral powers and correlative motivations
do not exhaust the possibilities. A complete survey would obviously have to
include at least one further possible sense that we can call
Rawls would intend the political conception of the person ontologically insofar as
he meant it to assert a truth of human nature. To be precise, there are two
possible variations here. By
"human nature" we could mean simply the ineluctable mode of being of persons as we always and
presently find them, or we could mean a human telos - a potential but not
(necessarily) actual state of fully developed human being. On either variation,
an ontological construction of the political conception of the person would be
more straightforward than either of the transcendental or ideological readings.
Moreover, an ontological reading is the kind that responds directly to our
intuitive idea about political justification: that it consists in showing a
favorable relation between a certain constitutional conception and the values
of the lives of persons according to a certain conception of how persons, by
nature, are or have reason to be.
Despite these attractions, an ontological reading of the Rawlsian conception
of the person is not readily available.
Rawls seeks a constitutional conception that practically responds to the problem of
political liberalism. That means navigating around the points of divisive conflict among the
reasonable comprehensive views - moral, religious, and philosophical doctrines
- expected to inhabit a modern, free society. Would
Rawls call unreasonable every doctrine that conceives of persons, comprehensively
speaking, as actually or ideally constituted rather differently from how
Rawls conceives them? Unless he would, he cannot urge his conception of persons
ontologically without making the resulting conception of justice an
unacceptably partisan and not a
In other words, the conception of the person is part of the exposition of
justice as fairness, and
Rawls avowedly intends justice as fairness as
"political" in the sense of its expositional independence from comprehensive ethical and
n65 The conception of the person, therefore, must itself meet the test of
freestandingness, and it's hard to see how it can do so if taken ontologically.
An ideological reading easily satisfies the test: Ideologically construed, the
conception of the person describes how (certain) people think of themselves
when they have politics, broadly speaking, on their minds. Perhaps the
transcendental reading, too, may meet the test of freestandingness.
Transcendentally construed, the conception of the person makes no claim to
factuality or validity (or even to significance of any kind) outside the
expository project of
political liberalism. The philosopher says:
"This is what we must somehow believe about persons, in order to make credible
and coherent our (relatively) fixed beliefs about the basic rights and wrongs
of politics." The same philosopher is then making claims on your beliefs, to be sure, but
claims that still are relative to your disposition to respond in certain ways
[*1820] tively political concerns. This philosopher's engagement could be limited to
helping you unpack and organize the beliefs you have about politics,
propounding certain facts about your beliefs relative to politics (although not
necessarily about the objects of those beliefs). There seems, by contrast, no
comparable way of restricting to a limited domain of the political a direct
ontological declaration that persons are possessed of the two moral powers and
correlative motivations. One can always tack
"politically speaking" onto the end of the sentence, but what would this appendix now mean? I, at any
rate, cannot find any clear sense for it. It seems that only by running the
sentence through the filter of belief, either actual (the ideological
construction) or necessary (the transcendental construction), can we give
"politically speaking" any force.
Are we, then, ready to conclude that
Rawls must intend either or both of the transcendental and ideological constructions
of the political conception of the person? Not quite, because there is a fourth
possibility still to be considered:
Rawls might intend his conception of the person in a sense we can call
"strictly constructive." Pending more extensive exploration, here is the rough idea: To regard a
conception of the person as strictly constructive is to see it as a
philosopher's tool for helping to elicit some cognition other than a cognition
of the conception's own truth, its own correspondence to any other object
(whether that other object be the world or the human condition itself or
anyone's beliefs about the world or the human condition). For example, the
philosopher solicits our cognition of the rightness (or the fittingness for us)
of the political conception of justice as fairness. In the course of working us
around to this cognition, the philosopher puts before us a certain conception
of the person. This conception enters into his argument in something like the
way in which (the analogy is very loose) a scaffold enters into the
construction of a building; like a scaffold, the conception's claim on our
attentions ceases when the projected cognition (the building) finally stands.
To speak of the conception's validity, then, would mean only that it serves its
cognition-building purpose, nothing more. Further explanation requires a brief
foray into rarefied methodological regions of Rawlsian political philosophy. We
need at least a superficial glance at the justificatory method that
C. Political Constructivism
1. The original position as a regulative part of justice as fairness.
As a starting point for this excursion, consider that - as constitutional
lawyers will rush to tell you - the regulative content of a constitutional
[*1821] is not and cannot be fully or adequately expressed by abstract black-letter
principles like the two principles of justice as fairness. The full regulative
conception must include some kind of shared background understanding of the
motivation and derivation of the principles, which can channel disputes over
concrete applications of the principles forcefully enough to sustain confidence
among disputants that all are appealing to one and the same public conception
of justice. In justice as fairness, some of this requisite background
understanding is provided by the famous hypothetical-contract construction that
Rawls calls the
n67 The original position is a conclave of representative
"parties" hidden by a
"veil of ignorance" from any knowledge of their constituents' actual social stations, economic
lots, moral, religious, and metaphysical convictions, and particular life
projects. The parties' task is to represent their constituents in choosing a
conception - a set of principles - to regulate the distribution of
"primary goods" in their society.
n68 Portions of A Theory of Justice and
Political Liberalism explain why, as
Rawls contends, the parties rationally would choose the principles of justice as
fairness. Shared understanding among citizens of how their constitutional
principles are rationally derivable from the original-position construct
enters, along with the black-letter principles, into the regulative content of
justice as fairness to which citizens appeal in resolving disagreements over
constitutional essentials and questions of basic justice.
2. Procedures of construction (devices of representation) and reflective
Rawls' use of the original-position construct exemplifies the form of justificatory
argument that he labels political constructivist. A constructivist argument
pivots on a
"procedure of construction" or a
"device of representation." This means a situation of choice of a constitutional conception, laid out so
that it aptly reflects certain attributes of society and persons, namely, the
ones that (the procedure's proponent stands ready to argue) ought to enter into
a conception of political justice.
Rawls, this means a conception suited to a deeply pluralized society committed to
free democratic institutions and in that way responsive to the problem of
n70 The original-position construct is meant to model both the problem and the
form of a possible response to it. Perhaps we can join reasonable pluralism to
political peace and freedom by showing how the principles of justice as
fairness could have issued from (i) unanimous, self-protectively motivated
rational choice in (ii) an initial situation aptly constructed to represent
politically significant attributes of per
[*1822] sons as (iii) those attributes are drawn from a broadly shared political
culture of constitutional democracy.
Now, the rightness of
Rawls' procedure of construction depends on more than the aptness of his ascriptions
to society and persons; it also depends on his translations of these
descriptions into features of the constructive procedure.
Rawls has no way (or at any rate makes no attempt) to prove the rightness of his
particular way of designing the procedure of construction, he cannot (nor does
he claim to) prove the rightness of justice as fairness simply by showing that
its principles pass the political constructivist test according to his
particular procedural design.
n72 So it cannot be for the sake of proving rightness that
Rawls undertakes his demonstration that the principles issue from the procedure.
Rather, it is for the sake of giving shape and sinew to the conception for
which the black-letter principles stand, filling in the conception's visionary
inspiration and normative content, while at the same time providing support for
the argument that the conception is a reasonable - indeed the most reasonable -
one for us.
Acceptance of a constructivistically generated political conception as right
or most reasonable finally depends, then, on a reflective process in which
"we" the audience apply the candidate conception's abstract black-letter principles
of justice to various concrete issues and match the results against our direct
intuitions of the answers required by justice. When we find an apparent
mismatch, we must then determine whether we should revise the principles or our
n74 The better the principles seem to be doing over all, the more we may come to
find that discrepant intuitions have to give way. When we find them refusing to
give way - when they prove to be
n75 at odds with the principles - we then have to choose between revising the
principles and revising our initial interpretations of them. What would set the
line, then, between principle and interpretation? The answer is knowledge of
how the principles are derived from a certain construction into which certain
conceptions of society and persons, drawn from a certain political culture,
have been modeled. This background knowledge is itself a part of the complete
political conception of justice under scrutiny,
n76 capable (as lawyers know) of constraining choice among verbally admissible
readings of the black-letter principles. Now, it might happen that no reading
of the black-letter can be found that squares with both the interpretative
background and our stubbornly resistant concrete convictions. In that event, it
will be the black-letter and,
[*1823] behind it, the procedure of construction, and the correspondent conception of
society and persons that will have to be revised. Ideally, we move back and
forth among justice intuitions, interpretations, black-letter principles, the
procedure of construction, and conceptions of society and person, adjusting
here and there, until we finally come to rest in the idealized state of
"reflective equilibrium." At that point, we will have identified the political conception of justice
that is right or most reasonable for us.
3. The original position and the two moral powers.
A procedure of construction, we said, is a situation of choice laid out to
represent attributes of society and person that ought to help shape a
conception of political justice. In
Rawls' view this means a conception for a deeply pluralized society committed to free
democratic institutions, which in turn means that the relevant attributes are
those given by the correlative ideas of society as a fair scheme of cooperation
and of free and equal persons motivated by two moral powers. In the
original-position construct, the idea of society as a fair scheme of
cooperation is represented by the conditions of choice. Not only must the
parties choose a unified and abstractly formulated set of principles to apply
for all time to all basic structural questions, they must do so from behind the
veil of ignorance. These conditions represent the idea of the fairness of the
n78 The veil makes each party into a representative for any and all members of a
modern plural society.
How, then, are the two moral powers of persons and their correlative
motivations modeled in the original-position constructive device? According to
our intuitive notion of constitutional justification,
n80 we might have expected
Rawls to have the parties simply decide which principles of justice are best
designed to contribute value to the lives of persons conceived as possessing
both of the moral powers and as concerned with developing and exercising both.
Strikingly, that is not what he does. Rather, it is only the second moral power
- that of rationally holding to, revising, and pursuing one's own conception of
the good - that
Rawls thus straightforwardly builds into his construct. The construct directs the
parties who choose a conception of justice to do so in a rationally calculative
way, as if the sole concern of their constituents is to optimize the prospects
for successful exercise of the second moral power.
Rawls expressly forbids the parties to assume that development and exercise of
[*1824] a capacity for a sense of justice is a part of the conception of the good
actually held by all (or indeed any) of the persons on whose behalf they are
supposed to be making a rationally preferred choice.
How, then, is the original position, as a device of representation, supposed
to represent the first moral power, the capacity of persons politically
conceived to act upon a public conception fixing fair terms of cooperation?
Rawls says that the situation of choice is so constructed as to give the parties
reason to take account of their constituents' capacities for justice or, in
other words, their propensities for civil cooperation on terms that all can
accept as reasonable. This may seem surprising, given that the parties must
choose strictly with a view to optimizing their constituents' prospects for
realizing their own several conceptions of the good, and furthermore may not
treat the manifestation or exercise of justice as a stock component of
individual conceptions of the good for themselves. But
Rawls says that the parties in the original position, rationally pursuing
self-protective ends on behalf of their constituencies, will be instrumentally
well-advised to take into account the (ascribed) fact that citizens have
propensities for justice that can motivate them, regardless of whether anyone
is supposed to count this fact about herself as directly contributory to her
own well-being or the value of her life. If people in fact have such a
propensity, regardless of whether they value intrinsically their having it or
their exercise of it, then tapping into the propensity is rational for parties
concerned about the pursuit of whatever it is they do value. Appealing to this
propensity can help secure
"the great advantage to everyone's conception of good of ... a just and stable
scheme of cooperation."
n83 In this way, engagement of peoples' capacities for justice can serve as an
"to advance the determinate conceptions of the good of the persons [the choosing
n84 And so this is how, even though as
"rationally autonomous representatives" the parties are not moved by reasons of justice itself, it nevertheless does
matter to them that
"persons as citizens" are conceived as being moved in this way.
4. Characters and motivations.
Rawls' political constructivist argument posits a number of characters or
standpoints, each with its distinctive motivational ascriptions. The characters
and standpoints include (1) the representative
"parties" who choose in the original position, motivated to do the best they can for
those whom they represent as deputies; (2) those who are thus represented, the
"citizens," each motivated by individual prospects for successful exercise of the second
moral power; (3) politically conceived
"persons" motivated by both moral powers - justice as well as pursuit of the good - and
correlative conception-dependent desires; and (4) the audience for the
political philosopher's justificatory effort,
"you and I" assessing the reasonableness for us of the total conception. This total
[*1825] is composed of (i) the principles the parties would choose, (ii) the layout of
the parties' situation of choice, (iii) the manner of modeling into that layout
the motivational attributes of politically conceived persons, and (iv) the
interpretation of democratic, constitutional culture from which the political
conception of the person is drawn.
Notice, now, that while motivational features are specifically ascribed by
Rawls to the characters in classes (1), (2), and (3), none are separately specified
for you and me, the audience constituting class (4). Motivational attributions
for characters in the first three classes are certainly parametric for the
procedure of construction that organizes
Rawls' politicalconstructivist argument in support of justice as fairness. Without
specification of attributes for these characters, the argument is obviously
incomplete. But neither does it seem complete without some motivational
imputation to class (4). What, then, if anything, does
Rawls' conception imply about the motivational attributes of you and me, the
audience? It is implicit, no doubt, that our motivations are such that the
philosopher's politicalconstructivist argument is appropriately addressed to
us. But what are these motivations? The question seems important. Considering,
after all, that the philosopher is inviting
"us" to submit our fates (politically speaking) to justice as fairness, we might
want to know whether his argument is addressed only to those of us who perceive
"our" motivations (politically speaking) as matching those of
"persons" according to the Rawlsian political conception of them, the motivations that
Rawls has modeled (if somewhat circuitously) into his recommended procedure of
5. Taking stock.
Since to most readers it will be the first moral power, the capacity for a
sense of justice, and correlative motivations that seem most dubious
ontologically, let us focus on it. What attributions is
Rawls making to us the audience with respect to the capacity for justice and the
interest in exercising it? Is he saying that we have them in fact? That we in
fact hold some belief about our having them? That if we don't hold some such
belief, then we can't coherently subscribe to constitutional democracy? Or are
we now, perhaps, in a position to conclude that
Rawls is not, after all, attributing to us anything at all regarding the capacity or
motivation for justice - not a fact, not a belief, not even any sort of need
for a belief? If
Rawls has made a strictly constructive use of the first moral power, then he has
sought to move us toward approval of the principles of justice as fairness by
showing how they would issue from a procedure of construction whose design or
workings were shown to be inspired in some way by the idea of the first moral
power, the capacity and motivation for justice. This is certainly not the same
as asking us directly to validate that idea in any way (either as true in fact
or true to our beliefs, whether speaking politically or more broadly), for any
purpose beyond its contribution to the project in constructivist political
justification for which it was specifically drawn.
But we move too fast. Even such a roundabout use of the political conception
of the person suggests strongly that the conception is one that
Rawls expects you and me to find attractive. And we have stronger grounds than that
for concluding that
Rawls must be attributing to us the audience some sort of belief (at least) in the
first moral power.
Rawls asks us to envision that in the original position the parties rely on
citizens' capacities for justice to assure themselves of the effectiveness and
stability of the principles of justice as fairness.
n87 This assurance is a distinct component of the persuasion intended by
Rawls' use of the constructivist method. Somewhere out there (although not in
Political Liberalism), a persuasive case may be found for justice as fairness that presents itself
as independent of belief in the capacity and motivation for justice. However,
if we want to hang on to this persuasion or profess to be persuaded by it, then
we can't now toss overboard all belief in this capacity and motivation. To be
Rawls is (in part) to be moved by either the transcendental-deductive proposition
that constitutional-democratic politics presupposes the idea of the capacity
for justice (in which case commitment to the one demands belief in the other),
or the empirical-ideological proposition that we enable ourselves to commit to
constitutional democracy by endorsing the idea of this capacity. In either
case, this idea's status in
Rawls' argument is stronger than that of a scaffold-like constructive tool whose
contribution ends with the process of construction. Yet the conception's status
may still be that of an idea of reason internal to a certain project of
political justification. It may still be that the only validity claimed for the
idea of the capacity and motivation for justice is rational necessity, relative
to the demand for a solution to the problem of political possibility that our
author has propounded.
IV. Toward Moral Consensus
A. Stability and Reproduction
Rawls makes stability a secondstage test for a conception of justice that appears in
other respects attractive to reflective equilibrium. A conception is stable if
it can hold the freely given loyalties of adherents to all moral, religious,
and metaphysical doctrines that can be expected to arise and survive critical
reflection in the public culture that the conception itself is expected to
sustain. A conception that fails this test is not reasonable at all, much less
can it be right or the most reasonable for us.
Does the demand for stability inject (unwanted) ontological force into the
political conception of the person? Is not
Rawls, after all, contending something like the following? First, the principles of
justice as fairness are understood to issue from a procedure of construction
designed with reference to persons endowed and motivated in specific ways.
Second, on independent psychological grounds we can predict that a society
effectively and publicly governed by the principles will tend toward
reproduction of inhabitants with like
[*1827] motivations. Third, since the persons thus engendered by the regime will
understand how the regime's conception of justice was constructed with them in
mind, they will be at home in the regime, and it will be stable.
I don't think this is exactly
Rawls' argument, but suppose for a moment that it were. Then one of the things
Rawls would be saying in favor of justice as fairness is that putting the conception
into effect builds characters like the ones described by the political
conception of the person. True, our author would not be advancing this
character-building consequence as the motivating aim or point of his favored
constitutional conception, but he would still be resting a part of his defense
of the conception on a prediction of this consequence. So the claim that
adoption of a certain institutional conception will lead to people being a
certain way is a load-bearing part of the philosopher's justification of the
conception. And hadn't that then better be a way that the philosopher stands
ready to affirm as good for people to be, according to their natures?
The answer (we should know by now) is that we don't have to understand the
process of stabilizing reproduction as reaching to people's ways of being in
any ultimate sense, or for all of life's purposes. We can understand it as a
process of reproduction of people's self-concepts relative to politics. The
argument for stability, then, is that a regime governed by principles issuing
from a constructive procedure designed to reflect a conception of persons,
politically speaking, as self-authenticating sources of valid claims
n89 will be conducive to people's regarding themselves as self-authenticating
sources of valid claims, politically speaking. The argument sounds in ideology,
B. The Good of Justice: (Overlapping) Consensus and (Full) Autonomy
What is the good of justice, according to
Rawls? Justice signifies a solution to the problem of
political liberalism, peace and freedom without oppression. Is that all? Wouldn't it be enough?
Rawls avowedly intends a regulative mission for justice as fairness. Within the
broad practice of constitutional democracy, many important issues remain
obstinately unsettled. One task of political philosophy,
Rawls says, is to examine whether some underlying basis of agreement can be
uncovered for resolving some of these questions or at least narrowing
n90 To that end, some degree of abstraction is a virtue in a regulative conception
of political justice. When we are striving to resolve disagreements with
Rawls says, we must look for premises that all can
"accept for the purpose of establishing a working agreement."
n91 To that end, it can sometimes be helpful to postpone concrete issues pending
establishment of common ground at higher levels of abstraction.
Rawls is at least as sensitive to a converse point: Parties who disagree hopelessly
"the highest things" may happen to find
[*1828] agreement at lower levels.
"Different premises," as
"may [support] the same conclusions."
n93 In constitutional democratic societies,
Rawls believes, we have the best chance of finding this sort of midlevel overlap
when we focus on the main political, social, and economic institutions of the
regime, that is, the basic structure.
n94 This suggests that a publicly affirmable conception of justice should aim in
the first instance at criteria for the basic structure, framed at a moderately
high level of abstraction (witness the two principles of justice).
A political conception of justice, such as justice as fairness, is of course
meant to supply higher-order criteria for lawmaking, both constitutional and
n95 Given the fact of reasonable pluralism,
Rawls thinks the best way to find consensus on a public regulative conception is to
confine the conception's scope to the basic structure. He hopes it is possible
to work out a moderately abstract, regulative conception of basic structure
without having to take positions on ethical and metaphysical issues over which
comprehensive doctrines must divide.
n97 A further hope is that if we can find a moderately abstract, clear, and simple
expression of a conception of political justice that is acceptable to every
reasonable comprehensive view, then political results issuing from this
conception should be broadly acceptable.
Consensus, then, is a pivotal concern for
Rawls, but not all consensuses are alike in value.
Rawls wants a public regulative conception of justice to be based on a true moral
consensus and not on a diplomatic modus vivendi. To base the conception on a
Rawls says, is to make it
"political in the wrong way."
n99 But what, precisely, is the difference, and why is it so important?
What makes an agreement a
"mere" modus vivendi is that each party regards the agreement as strictly
instrumental to some good of his or hers apart from any good found in the
agreement itself. In a modus vivendi, the agreement has no intrinsic,
noninstrumental value; parties participate in it to further their own ulterior
interests. Contrast the case in which the parties regard participation in the
agreement as good for them for its own sake, as, say, a moral good. Each person
would then regard agreement as both a common aim or social good and as a
constituent of each of their goods taken severally. This is
[*1829] an especially intensive way - this responsiveness to moral interest or moral
motivation - in which overlapping consensus can support justice in a society
where it is found.
n100 The difference from a strictly instrumental agreement seems plain. What is not
so apparent, however, is why people would find intrinsically valuable their
participation in a social consensus over principles to regulate the basic
structure of their society. On what understanding of these people's natures
would it make sense for them to find intrinsic - not just instrumental - value
in this state of affairs?
An answer may be found in the conception of the person as possessed of
higher-order interests in the development and exercise of both the moral
powers, the capacity for justice as well as that for self-determined pursuit of
the good. On that understanding, overlapping consensus on a public regulative
conception of justice is an intrinsic good for every participant because it
puts within each one's reach a certain enabling condition of a good life
(insofar as the conduct of politics in one's society bears on the goodness of a
life). Overlapping consensus enables satisfaction of the conceptiondependent
desire to realize in one's person an ideal conception of liberal citizenship.
n101 Overlapping consensus is the condition upon which the first moral power, that
of grasping and acting upon a public conception of justice, can be realized
without grievously infringing the second moral power, that of holding to one's
own autonomously determined conception of the good. That is what
Rawls calls a person's full autonomy.
n102 Now, if full autonomy is a good, for whom is it a good? Apparently, for
persons conceived as the political conception of the person conceives them.
To recapitulate, the strongest argument that justice as fairness has
regulative force for a society appears to rest in some part on the claim that
justice as fairness supports a special kind of consensus that members of the
society have good reason to value intrinsically.
n103 But the claim that any sort of consensus can possibly have intrinsic value for
anyone depends, in turn, on what appears to be a quite specific and distinct
idea of the flourishing or perfection of a person. In sum, these two features
of the total conception suggest that the full
[*1830] regulative force of justice as fairness is available only to those for whom
this idea of personal flourishing is already a correct or compelling one.
Accordingly, questions loom: How might it be that this specific account of the
person is commonly compelling upon interlocutors coming from a deeply divided
plurality of comprehensive ethical and metaphysical views? And what is the
political method for soliciting endorsement from such a diverse constituency of
principles geared to such a seemingly strong conception of the person?
n105 The answer seemingly can't be that the conception is true. That is not, at any
rate, the answer
Rawls gives. He relies, rather, on the expectation that the conception is
intuitively salient to diverse parties all ensconced in the culture of
Note how this points to a strong inferential connection between the
hypothetical ability of those living under justice as fairness to reach
consensus on the legitimacy of that conception and
"our" ability now to discern the conception's fittingness to ourselves as we are now
or wish to become. The political conception of justice as fairness is supposed
to represent a compelling organization of fundamental political values as we
now know them. The conception of justice and its correspondent conception of
the person are tendered as clarifications of ideas and beliefs already implicit
in extant constitutional, democratic culture. But that gives us reason to
expect that something resembling the extant range of comprehensive views would
flourish under justice as fairness. It follows that if we did not recognize the
fittingness of justice as fairness to our condition (actual or desired) as we
consciously perceive it, that would call into serious question the claim that
justice as fairness is a stable conception.
One sees, then, why
Rawls once wrote that what ultimately justifies a political conception of justice is
"congruence with our deeper understanding of ourselves and our aspirations, and
our realization that, given our history and the traditions embedded in our
public life, it is the most reasonable doctrine for us."
n107 One sees, too, why
Rawls called on his readers to imagine not that citizens
"choose" these aspirations (for that would be to represent citizens ontologically - and
deeply controversially - as transcendental subjects, pure wills existent prior
to their ends
n108 ) but rather that they discover themselves already
"holding" these aspirations as
"ideals that they have taken in part from the culture of their society."
V. Conclusion: Relativity and Subjectivity
We have seen the ways in which
Rawls puts a certain motivating conception of the person to use in the constructive
political liberalism. So far as we have been able to find out, these ways have required no claim of
an ontological status for the conception, independent of the constructive
endeavor in which it appears. For all we have seen, the conception states no
more than a possibility condition for the endeavor. The argument, then, is that
the philosophical project of
political liberalism is possible for us (if it is) because, as it happens, a certain conception of
the person appeals to us as an account of a truth of our being.
This reading takes us to the brink of the debate over the alleged relativism
n110 Does the book really mean no more than to commend
political liberalism and justice as fairness to those for whom, by reason of cultural situation,
the Rawlsian conception of the person is regulative already? Taking the book to
mean no more than that, the conception of the person is comfortably internal to
the project. But there are also those who would take
political liberalism, the project, to be a more globally commendable guide to the conduct of
n111 As for them, it seems that by adopting
Rawls' justificatory arguments they would also buy into some strong, distinctive
ideas about what gives value to lives, humanly speaking. It seems that in order
Political Liberalism to be a globally relevant guide to what is right institutionally, the
conception of the person must be correspondingly viewed as a case of what
Martha Nussbaum calls
"internal essentialism" - that is, a product of evaluative inquiry (with or without reference to
"metaphysical foundations") into
"what we really think about ourselves [qua human] and what holds our histor[ies
as human] together."
It is a fair question whether the book can respectfully be read to lay claim
to anything less. For surely (the question would posit) the political
conception of justice as fairness must be meant to have prescriptive and
critical bite somewhere on Earth. Indeed, it must. But couldn't this
"somewhere" be right here, in constitutional-democratic cultures whence the Rawlsian
conception of the person is drawn? A leading aim of
Political Liberalism is to persuade an audience, already knowing themselves as broadly committed to
liberal constitutionalism ("constitutional democracy") that justice as fairness correctly interprets their broad commitment. But
within the family of liberalisms, justice as fairness is after all a somewhat
contentiously egalitarian member.
n113 Constitutional democracy as practiced today in the United States, for example,
[*1832] certainly does not satisfy the difference principle (and quite arguably does
not satisfy the liberty or opportunity principles, either).
Now consider the form of
Rawls' argument: (1) We hold constitutional democracy (broadly conceived) to be the
right form of political ordering for us; (2) correspondingly, we think of
social life as a fair scheme of cooperation and of ourselves as beings free and
equal in our possession of the two moral powers and the correlative ethical and
moral motivations; (3) but from these ways of thinking of society and persons
follows the egalitarian content of justice as fairness, along with its implicit
critique of contemporary constitutional practice in the United States. No doubt
we can strenuously dispute the soundness of the inferences from (1) to (2), and
from (1) and (2) to (3). Insofar as the inferences are compelling, however, the
political conception of the person figures here as an intermediate term in an
internal critique of constitutional democracy. By
"internal critique" I mean a demonstration of how concrete practice (in the United States, for
example) deviates from norms that said to be are rationally reconstructible
from the broader concept of the practice.
n114 Figuring in this internal-critical capacity, the conception of the person
would have critical bite. Interestingly, the conception in this capacity would
also apparently be secure against the postmodernist critique of subjectivity:
Here we would have
Rawls claiming to derive a certain depiction of personsassubjects from the practical
concept of constitutional democracy or from certain empirical cultural content.
There we would have the critics of subjectivity claiming that the depiction is
(ontologically!) wrong, mistaken, illusory. The two claims, it seems, would
pass each other as ships in the night.
Political Liberalism fend off the critique of subjectivity. Or rather, thus does it if we read the
book, on its prescriptive side, as intending only an internal critique of
constitutional democracy - prescribing to those who practice that political
form but prescribing not to those who practice it not. Does this mean, then,
that relativism is (in this case, at any rate) the price of evading the
critique of subjectivity? I don't know. Who can do better than to live the best
she can according to the sort of being she finds herself believing herself to
be - believing, that is, reflectively, unshakeably, recurrently (which is not
to say incessantly, at every waking moment)? And what is (avoidably)
relativist, then, about a prescription to live that way? If nothing, then how
Rawls setting up for relativist prescription by offering the political conception of
the person as an account of how persons, qua participants in constitutional
democracy, reflectively and recurrently find themselves believing themselves to
Suppose you were determined to read
Political Liberalism for ethical wisdom on a personal level. The wisdom then would be: For you as a
participant in constitutional democracy, living the best you can according to
the way that you reflectively and recurrently believe yourself to be means
living under the
[*1833] constitutional conception of justice as fairness. But if you do read the book
that way, then for you its implicit response to the critique of subjectivity
will not be evasion, it will be confrontation. Confrontation by challenge. The
challenge is to divest yourself, if you can, of belief that constitutional
democracy is, in fact, the right form of political ordering for you and your
n115 Nothing in
Rawls' argument would require you to answer the question that obviously now looms:
whether (for this purpose) your kind is anything less than the human kind.
Nevertheless, you might feel some pressure to respond.
n1. To find out what is meant by
"political," read on.
n2. P. 19.
n3. P. 19.
n4. P. 19.
n5. See pp. 73-74, 105-106.
"Conception-dependent desires" are desires to act from the principles we see as
"belonging to, and ... helping to articulate, a certain rational or reasonable
conception, or a political ideal." P. 84.
Rawls, A Theory of Justice (1971).
n8. Alan Ryan, How liberalism, politics, come to terms, Wash. Times, May 16, 1993,
n9. It's true that law clerks come from law schools as graduates. It's also true
that numbers of them return to law schools as professors.
n10. See John Hart Ely, Democracy and Distrust 58 (1980) ("We like
Rawls, you like Nozick. We win, 6-3. Statute invalidated.").
n11. These lectures appeared in revised form in a 1980 article. John
Rawls, Kantian Constructivism in Moral Theory: The Dewey Lectures 1980, 77 J. Phil.
515 (l980). In
Political Liberalism, Rawls reports that his reflections on the dissensus prompted something of a break
with his earlier ideas in A Theory of Justice. Pp. xv-xvii.
n12. They also affect important topical issues in political theory, such as
communitarian critiques of liberalism and anxieties over rights talk. I do not
pursue these matters here. For prior discussions, see Frank I. Michelman,
Justification (and Justifiability) of Law in a Contradictory World, in NOMOS
XXVIII: Justification 71, 82-94 (J. Roland Pennock
& John W. Chapman eds., 1986); Frank I. Michelman, Super Liberal: Romance,
Community and Tradition in William J. Brennan Jr.'s Constitutional Thought,
77 Va. L. Rev. 1261, 1306-1312 (1991); Margaret Jane Radin
& Frank I. Michelman, Pragmatist and Poststructuralist Critical Legal Practice,
139 U. Pa. L. Rev. 1019, 1035-39 (1991).
n13. See Postmodernism and Law: A Symposium,
62 U. Col. L. Rev. 489 (1991); Symposium, Beyond Critique: Law, Culture, and the Politics of Form,
69 Tex. L. Rev. 595 (1991); Symposium, The Critique of Normativity,
139 U. Pa. L. Rev. 801 (1991).
n14. According to
"regard themselves [politically] as self-authenticating sources of valid claims" and
"as being entitled to make claims on their institutions so as to advance their
conceptions of the good." P. 32.
n15. See, e.g., Pierre Schlag, Normativity and the Politics of Form,
139 U. Pa. L. Rev. 801, 805 (1991) ("Legal thinkers ... represent themselves as ... selfdirecting individual liberal
humanist subjects at once rational, morally competent, and in control of their
own situations, the captain of their own ships, the Hercules of their own
empires, the author of their own texts. It isn't so.").
"Normativity" broadly refers to the unabashedly prescriptive character of typical legal
scholarship and its focus on questions of how some personal or institutional
agent of the law ought to (choose to) act or what the agent ought to (choose
to) do. See, e.g., Pierre Schlag, Normative and Nowhere to Go,
43 Stan. L. Rev. 167, 171, 177 (1990). More narrowly defined, the term refers to
"comprehensive normative rationality," or the aim of producing
"a norm that is complete, self-sufficient, separable, trans-situational,
non-contradictory, and non-paradoxical within its intellectual or legal
" Mark V. Tushnet, The Left Critique of Normativity: A Comment,
90 Mich. L. Rev. 2325, 2326 (1992) (quoting Schlag, supra note 15, at 839). A prohibition on comprehensive
normative rationality does not foreclose the practice of
"thin" normative judgment. Id. Yet the critique of subjectivity implies a relentless
critique of normativity
"all the way down,"
id. at 2347, because if subjectivity
"isn't so," then (it would seem) normative scholarship is aimless and idle, an imposition
on us all, and scholars had better concentrate on discovering and describing
our situation. See, e.g., Richard Delgado, Norms and Normal Science: Toward A
Critique Of Normativity In Legal Thought,
139 U. Pa. L. Rev. 933, 959-60 (1991) (urging that legal scholars make
"law" their subject matter,
"actually observing and describing it"); Schlag, supra note 15, at 889 (criticizing scholars for their failure to
"the context of academic legal thought").
n17. See Radin
& Michelman, supra note 12, at 1021-22, 1053
& n.125 (noting that Schlag himself emphasizes the issue of
"performative contradiction" in the critique of normativity).
n18. Cf. Pierre Schlag, The Problem of the Subject,
69 Tex. L. Rev. 1627, 1730-31 (1991) ("In his most educated moments, [the] liberal subject understands that he is
socially and rhetorically constructed, but nonetheless retains his autonomy to
decide just how constructed or autonomous he really is.").
n19. For some views on this problem, see Radin
& Michelman, supra note 12.
n20. Mark Tushnet interprets
Rawls' recent work as
"accepting the critique of comprehensive normative rationality but arguing that
the critique does not undermine ... the claims of traditional liberalism." Tushnet, supra note 16, at 2336. My proposal parallels Tushnet's reading, but
it shifts the focus from normativity to the problem of the subject.
n21. This notion is more fully explained in the text accompanying notes 41-53 infra.
n22. P. 11. Of course,
Rawls has in mind that it is the regulative conception of justice that joins the
various institutions together into a conceptually unified system.
n23. P. 291. I have used
Political Liberalism's formulation of the principles. The respects in which this formulation differs
from that in A Theory of Justice are not material to this essay.
"equal basic liberties" are specified by a list as follows:
"freedom of thought and liberty of conscience; the political liberties and
freedom of association, as well as the freedoms specified by the liberty and
integrity of the person; and ... the rights and liberties covered by the rule
of law." P. 291.
n25. The first principle of justice (principle
"a"), also called the
"liberty" principle, lexically outranks the second principle (principle
"b"). Within the second principle of justice, the first subprinciple, the
"opportunity" principle, lexically outranks the second subprinciple, the
"difference" principle. See p. 6; see also
Rawls, supra note 7, at 30203.
"Lexically outranks" means that a higherranking principle must be fully satisfied before attempting
to satisfy the next principle. The analogy is to a dictionary, in which no
words beginning with
"b" (not even
"Baal") can appear until after
"azure" has been given its due. See id. at 4243
Rawls distinguishes two kinds of
"constitutional essentials." The first are principles that dictate the organization and powers of
government, as well as the political process. The second are principles that
"equal basic rights and liberties of citizenship that legislative majorities are
to respect." P. 227.
n27. Principles of basic justice regulate the
"basic matters of distributive justice, such as ... equality of opportunity
[and] social and economic inequalities." P. 228. These issues represent the bulk of the territory controlled by
Rawls' second principle of justice.
n28. See, e.g., Ronald Dworkin, Life's Dominion 201-04 (1993).
n29. See Thomas Hobbes, Leviathan or the Matter, Forme and Power of a Commonwealth
1-52 (Collier Books 1962) (1651).
n30. See Roberto Unger, Passion: An Essay on Personality (1984).
n31. See Schlag, supra note 18, at 1731 (speaking of the
n32. See, e.g., Dworkin, supra note 28, at 20006; Joseph Raz, The Morality of
Freedom 20405 (1986).
n33. See, e.g.,
Rawls, supra note 7, at 574-75 ("The desire to express our nature as a free and equal rational being can be
fulfilled only by acting on the principles of right and justice as having first
priority.... How far we succeed in expressing our nature depends upon how
consistently we act from our sense of justice as finally regulative."); Michael Sandel,
107 Harv. L. Rev. 1765, 1773 (1994) (book review) ("The political conception of the person ... closely parallels the Kantian
conception of the person, with the important difference that its scope is
limited to our ... identity as citizens.").
n34. See note 29 supra.
n35. I should say something here about my use of the third person in referring to
liberal theorists. But for the awkwardness, I would not hesitate to say
"we." I gladly claim membership in the family.
n36. Cf. Schlag, supra note 18, at 1730 ("The [liberal] supposition is that the individual subject is essentially ...
autonomous, coherent, selfdirecting, integrated, rational, and originary."). By contrast, one might take the view that consciousness and agency are
attributes of God alone, of whom our several
"selves" are but contingent manifestations, or that all behavior is determined and all
consciousness fabricated by socialsystemic processes beyond the control of
n37. See text accompanying notes 16 supra.
n38. See p. 33 (explaining that the political conception of the person describes
"the way in which citizens regard themselves in a democratic society when
questions of political justice arise").
n39. See, e.g., p. 29.
n40. See pp. 29-35.
N41. See notes 23-25 supra and accompanying text.
"A ... conception is ... comprehensive when it includes conceptions of what is
of value in human life, and ideals of personal character, as well as ideals of
friendship and of familial and associational relationships, and much else that
is to inform our conduct, and in the limit to our life as a whole." P. 13.
n43. See pp. 12-13 ("[A] distinguishing feature of a political conception is that it is presented as
freestanding and expounded apart from, or without reference to, any such wider
n44. See, e.g., Raz, supra note 32, at 46; Ronald Dworkin, Foundations of Liberal
Equality, in XI The Tanner Lectures on Human Values 1, 36 (1990).
n45. See text accompanying notes 55-58 infra.
n46. P. 13.
n47. Pp. 13-14.
n48. P. 8 ("We collect such settled convictions as the belief in religious toleration and
the rejection of slavery.").
n49. See pp. 13-15.
n50. See pp. 18-19; text accompanying notes 24 supra.
n51. See text accompanying note 107 infra.
n52. See text accompanying notes 16 supra.
n53. See text accompanying note 40 supra.
n54. See text following note 114 infra.
n55. P. xviii.
n56. In an overlapping consensus, a political conception appears morally right to
each one of a plurality of conflicting comprehensive views. Each of the
participating comprehensive views converges on the moral rightness of one and
the same political conception without having to
"bracket" the political from the rest of life, ethics, or the truth of the world. See,
e.g., pp. 150-54. This is not, however, equivalent to saying that no bracketing
occurs at any stage of thought. The criterion of
"freestandingess" for a political conception of justice, see notes 42-43 supra, suggests that
every comprehensive view that finds such a conception morally right must also
find it morally right to accept some bracketing of the political. Perhaps it's
precisely the sharing of this feature by the participant views that qualifies
"reasonable" and eligible for participation in the overlapping consensus, thus making the
moral consensus possible.
n57. P. 36; see pp. xvi-xvii. The potentially divisive nature of reasonable
pluralism is aggravated by
"burdens of judgment," as
Rawls calls certain ineluctable causes of disagreement even among persons who
ostensibly affirm the same governing principles. The following are causes of
this disagreement: (a) difficulties in assessing conflicting, complex,
empirical evidence; (b) disagreement over the relative weight accorded to
competing considerations; (c) ambiguities in defining moral and political
concepts; (d) conflicts arising from different life experiences and
perspectives; (e) incommensurability of relevant competing considerations; and
(f) the impossibility of accommodating all recognized goods in any one
political system. See pp. 54-58.
n58. See p. 37.
n59. For further discussion of the conditions
Rawls posits as necessary to achieve this consensus, see text accompanying notes
67-68, 76 infra.
n60. The term derives from Immanuel Kant, whose philosophy undertook to
"deduce" from certain immediately gripping and undeniable aspects of selfconscious
human experience (such as perception and discrimination of external objects or
impulses of obligation and duty) the existence of certain other conditions of
being that could not themselves be directly known or intuited. See H.J. Paton,
The Categorical Imperative: A Study in Kant's Moral Philosophy 2023, 20203
Rawls at one point classifies the political conception of the person as a
"conception of practical reason,'
" explaining that this conception responds to the question:
"What must persons be like to engage in practical reason?" Pp. 107-08.
n62. See text accompanying note 40 supra.
n63. Cf. William Powers, Jr., Constructing Liberal Political Theory,
72 Tex. L. Rev. 443, 464-65 (1993) (reviewing
Political Liberalism) (concluding that
Rawls formulates motivating values and traits for his argument by combining a
"desired results" with description of
"traits commonly accepted in our culture as aspects of moral ... persons"). Note that the transcendental and ideological senses are logically both
separable and combinable. So far as logic is concerned,
Rawls could intend the transcendental sense but not the ideological, the ideological
sense but not the transcendental, or both senses simultaneously.
n64. See text accompanying notes 26-28 supra.
n65. See text accompanying notes 41-45 supra.
n66. Compare Powers, supra note 63, at 450:
Political Liberalism, any autonomy achieved through adherence to the principles of justice is
political, not moral, autonomy. By adopting the principles of justice, we
"reasonable' political nature, that is, our desire to engage in cooperative
social life." The first sentence's sense is unclear, simply because it's unclear what work
"political" does in the second sentence. No doubt we can sometimes advance analysis by
subdividing the class of desires into numerous subclasses, and
"political" may well be a fit name for a subclass containing the desire for engagement in
social cooperation. Nevertheless, to say that the (political) desire for
cooperative social engagement is a salient part of our nature would be to speak
comprehensively about us - hardly different, in that respect, from someone
saying we are creatures fit by nature for life in cities.
n67. See pp. 2228.
n68. Primary goods are advantages and resources of which anyone supposedly would
wish as many or as much as possible, regardless of his or her particular aims
in life. These
"all-purpose means" include, but are not limited to, rights and liberties, occupational and
office-holding opportunities, income and wealth, and social bases of individual
self-respect. See pp. 181, 227.
n69. See pp. xx-xi, 89-98.
n70. See text accompanying note 55 supra.
n71. We shall soon look more closely at some details of the translation. See text
accompanying notes 78-85 infra.
n72. The details of the Rawlsian constructivist design are controversial even among
Rawls' egalitarian-liberal kindred spirits. See Bruce A. Ackerman,
Political Liberalism, 91 J. Phil. - (forthcoming 1994) (book review).
n73. See Powers, supra note 63, at 464; id. at 457 ("Constructivism ... holds that humans construct moral values, but if we are
careful about the conditions under which the values are constructed, they can
transcend individual biases and preferences." (citing T.K. Seung, Intuition and Construction 17 (1993)).
n74. See p. 45.
n75. P. 45.
Rawls calls this background knowledge the publicity condition. P. 67.
n77. P. 28.
n78. Pp. 304-05.
n79. See pp. 24-25. In
Political Liberalism, Rawls takes pains to explain the veil of ignorance as a device for representing a
situation of full normative reciprocity. This veil allows citizens to choose
sets of regulative principles for their society on the understanding that the
only sets available for choice are those that every member can reasonably
accept, regardless of differences of social station, economic lot,
"comprehensive view," or particular life projects. Pp. 24-25, 27.
n80. See text accompanying note 28 supra.
n81. See p. 305. The parties understand, though, that the second moral power
includes capacities to choose and revise a conception of the good, as well as
to pursue rationally a currently held conception. How a person exercises the
power to revise a conception of the good may itself be an aspect of the good as
the person conceives it. See pp. 310-15. We might call this the romantic strain
in the Rawlsian political conception of the person. See text accompanying note
n82. Pp. 315-16.
n83. P. 316.
n84. P. 318.
n85. P. 315.
n86. See pp. 27-28.
n87. See text accompanying note 83 supra.
n88. See pp. 140-42.
n89. See note 14 supra.
n90. See pp. 8-9, 48.
Rawls, Justice As Fairness: Political Not Metaphysical, 14 Phil.
& Pub. Aff. 223, 229 (1985).
n92. See pp. 115, 192; John
Rawls, The Priority of Right and Ideas of the Good, 17 Phil.
& Pub. Aff. 251, 261-62 (1988).
Rawls, The Idea of an Overlapping Consensus, 7 Oxford J. Legal Stud. 1, 9 (1987); see
also p. 4 ("The most intractable struggles ... are ... for the sake of the highest things:
for religion, for philosophical views of the world, and for different moral
conceptions of the good."). For a recent expression of a like view in the context, see Cass R. Sunstein,
On Analogical Reasoning,
106 Harv. L. Rev. 741, 771-73 (1993).
n94. See pp. 11-13; Kurt Baier, Justice and the Aims of Political Philosophy, 99
Ethics 771, 772-73 (1989) (summarizing
Rawls' belief that in order for a nonutopian conception of justice to gain the
support of an overlapping consensus, it must satisfy four conditions: it must
be capable of bypassing philosophy's longstanding controversies; it must be
political; it must be liberal; and it must steer a course between two extreme
strands of liberalism). On justice as fairness as a body of range-restricted,
midlevel principles, see Thomas E. Hill, Jr., Kantian Constructivism in Ethics,
99 Ethics 752 (1989).
Rawls, supra note 93, at 5-6.
n96. See text accompanying note 48 supra.
n97. See pp. 10-11, 152.
n98. See pp. 35, 156, 230.
n99. P. 40; see also pp. 141-42.
n100. See pp. 147-48, 168, 208, 316-17; see also
Joshua Cohen, Democratic Equality, 99 Ethics 727, 748-49 (1989) ("To have a sense of justice is to have the aim of coordinating action in ways
that provide favorable conditions for the expression of a wide range of human
powers in pursuit of a plurality of aims."); Gerald Doppelt, Is
Rawls's Kantian Liberalism Coherent and Defensible?, 99 Ethics 815, 831 (1989) ("Rawls's Kantian citizens do not regard a just way of social life as a mere means to
the realization of their individual ends. Rather, the realization of their
commonly valued Kantian capacities is inconceivable apart from a just structure
of social interaction and reciprocal recognition without which they could not
be the persons they take themselves to be, in the Kantian view."); William A. Galston, Pluralism and Social Unity, 99 Ethics 711, 715 (1989)
(explaining that because the exercise of the two moral powers is experienced as
good, not only is justice a highest-order moral power and interest, but also
there is an intrinsic impulse to develop and employ it in society).
n101. See note 6 supra and accompanying text.
n102. See pp. 77-78.
Rawls writes that full autonomy is
"a political and not an ethical value," p. 77, and he writes that citizens, by honoring principles of justice,
"show themselves autonomous, politically speaking." P. 78. But see note 66 supra and accompanying text.
n103. See p. 207.
n104. See Galston, supra note 100, at 714, 717.
n105. See Jean Hampton, Should Political Philosophy Be Done without Metaphysics?, 99
Ethics 791, 804 (1989).
Joshua Cohen, Moral Pluralism and Political Consensus, in The Idea of Democracy 270, 281
(David Copp, Jean Hampton
& John E. Roemer eds., 1993).
Rawls, supra note 11, at 519; cf. pp. 90-95 (contrasting the features of rational
intuitionism with those of political constructivism).
n108. Commentators have criticized A Theory of Justice for presenting the
"self," ontologically, as having an unencumbered essence existing prior to all
relationships and commitments. See Michael J. Sandel, Liberalism and the Limits
of Justice 5-9, 19-23 (1982). In
Political Liberalism, Rawls expressly denies that his account either of persons or the original position
has this ontological significance. Pp. 26-27.
Rawls, supra note 11, at 568-69; see also pp. 13-14.
n110. See, e.g., John Gray, Can We Agree to Disagree?, N.Y. Times, May 16, 1993, 7,
at 35 (reviewing
Political Liberalism) (describing
Rawls' product as
"not a political conception of general human interest, but an apology for
American institutions as they are perceived from the politically marginal
standpoint of American academic liberalism").
n111. See, e.g., Brian Barry, Good for us, but not for them, Guardian, Aug. 14,
1993, at 23 (reviewing
n112. Martha Nussbaum, Human Functioning and Social Justice, 20 Pol. Theory 202, 208
(1992); see also id. at 229 (contending that without a humanly essential
account of the good,
"we have no adequate basis for ... justifying the claim that any ... [practice]
we encounter is unjust").
n113. On the egalitarian content of justice as fairness, see Cohen, supra note 106,
n114. For a discussion of the possibility of such critique with specific reference
Political Liberalism, see Frank I. Michelman, On Regulating Practice With Theories Drawn From Them:
A Case of Justice As Fairness, in NOMOS XXXVII: Theory and Practice (Judith
& Ian Shapiro eds., forthcoming 1995).
n115. Down this path lies the idea of nonfoundationalist, humanly universal,
"recursive" grounding for
Rawls' model conception of the person. This would consist in showing that the
"account of our self-conception" is not only latent in a certain public culture, but is also
"the best account that can be given of the basic capacities presupposed by the
practice" of political justification - or deliberation or discourse - itself. Kenneth
Bayres, Constructivism and Practical Reason in
Rawls, 14 Analyse
& Kritik 18, 30 (1992); cf. note 61 supra.
Prepared: January 24, 2003 - 5:02:29 PM
Edited and Updated, January 25, 2003
Philosophy 111A Page