COHEN - Reflections On Habermas On Democracy PDF

Ratio Juris. Vol. 12 No.
4 December 1999 (385416)
Reflections on Habermas on
Democracy
JOSHUA COHEN*
Abstract. Jrgen Habermas is a radical democrat. The source of that self-designation
is that his conception of democracywhat he calls discursive democracyis
founded on the ideal of a self-organizing community of free and equal citizens, coordinating their collective affairs through their common reason. The author discusses
three large challenges to this radical-democratic ideal of collective self-regulation:
1) What is the role of private autonomy in a radical-democratic view? 2) What role
does reason play in collective self-regulation? 3) What relevance might a radicaldemocratic outlook have for contemporary democracies? The author addresses these
questions by considering Habermas answers, and then presenting alternative
responses to them. The alternatives are also radical-democratic in inspiration, but
they draw on a richer set of normative-political ideas than Habermas wants to rely
on, and are more ambitious in their hopes for democratic practice.
I. Radical Democracy
Jrgen Habermas is a radical democrat (Habermas 1996a, xliiiii). The source
of that self-designation is that his conception of democracywhat he calls
discursive democracyis founded on the abstract ideal of a self-organizing community of free and equal citizens, coordinating their collective
affairs through their common reason (Habermas 1996a, 7). In this paper, I
discuss three large challenges to this radical-democratic ideal of collective
self-regulation:
1) What is the role of private autonomy in a radical-democratic view?
2) What role does reason play in collective self-regulation?
* I am grateful to Oliver Gerstenberg and Kenneth Baynes for discussion of the material in
section II, to Sebastiano Maffettone for making available a draft of his essay on Liberalism and
Its Critique, and to Leonardo Avritzer and Joshua Flaherty for extensive discussion of many
themes in this essay. I presented earlier versions at a Political Theory Workshop at Nuffield
College, Oxford, and to the McGill University Philosophy Department. I also wish to thank
Jrgen Habermas for comments on an earlier draft. My broader intellectual debt to Habermas
should be clear from virtually everything I have written. I am pleased to have this occasion for
expressing that debt.
Blackwell Publishers Ltd 1999, 108 Cowley Road, Oxford OX4 1JF, UK and 350 Main Street, Malden, MA 02148, USA.
386
Joshua Cohen
3) What relevance might a radical-democratic outlook have for contemporary democracies?

I will address these questions by considering Habermas answers, and then
presenting alternative responses to them. The alternatives are also radicaldemocratic in inspiration, but they draw on a richer set of normativepolitical ideas than Habermas wants to rely on, and are more ambitious in
their hopes for democratic practice.
1.
Habermas offers two lines of argument in support of his radical-democratic
ideal of discursive democracy: In brief, he claims that it is rooted in reason
and practically relevant to contemporary political societies. First, then,
Habermas locates the bases of democracy in a general, post-metaphysical
theory of human reason, which he presents in the theory of communicative
action, and of argumentation as the reflective form of such action. The
intuitive idea is that democracy, through its basic constitution, institutionalizes practices of free, open-ended, reflective reasoning about common
affairs, and tames and guides the exercise of coercive power by reference to
those practices. To be sure, democracy does not guarantee the subordination
of sovereign will and the coercive power it guides to the force of the better
argumentwhat could guarantee that practical reason guides political
power?but it establishes conditions favorable to such subordination. Moreover, the promise to subordinate political will to practical reason is a justifying ideal underlying democratic practice. By requiring a more complete
subordination of political will to practices of reasoning, then, we hold democracy to its own internal standards.
Second, Habermas aims to show how the old promise of a community
of free and equal members, guiding their collective conduct through their
common reason, can be redeemed if it is reconceived under the conditions
of complex societies (Habermas 1996a, 7). He offers such redemption by
elaborating the content of the democratic idealhe describes the rights that
citizens must assign to one anotherand showing how it can serve as a
practical guide once it is reinterpreted in light of modern conditions of social
and political complexity, including a market economy and an administrative
bureaucracy.
2.
I will say very little about the philosophical bases of democracy in the
communicative account of reason, and concentrate instead on the content of
Habermas conception of democracy and its implications. I steer clear of the
wider philosophical frameworkHabermas post-metaphysical theory of
Blackwell Publishers Ltd 1999.
Habermas on Democracy
387
human reason, communicative action, and argumentationbecause I think

that political argument should not be made to depend, or presented as
dependent on, a philosophical theory about the nature of reason. Philosophical theories about the nature and competence of reason do not provide
the common ground for equal citizens that is desirable in public argument in
a democracy. An appeal to reason cannot help us get behind the plurality
of competing moral, political, religious, metaphysical outlooks, because the
nature and competence of reason is one matter on which such outlooks
disagree. Thus, a post-metaphysical conception of reason, which ties the
account of reason to the presuppositions of argumentation, will not find
favor with a natural law theorist who believes that reason delivers substantial metaphysical truths and insights about the best human life.
Instead, I accept (with Rawls) the relative autonomy of political reason. Political reason is autonomous in that it can and should proceed in articulating a
conception of democracy without relying on an encompassing philosophy of
life or claiming to resolve the controversies among them, including controversies about the nature and competence of reason. It is only relatively
autonomous, because autonomous political argument needs to make sense
in light of the diverse and conflicting encompassing philosophies that (at
least some) citizens endorse: Citizens must judge, from within those separate
philosophies, that autonomous political argument is appropriate, and accept,
as a public matter, that the diversity of such philosophies recommends an
autonomous political reason. Political reason, we might say, lacks public
foundations, because there is no single, publicly authoritative basis for its
principles and modes of argument. But it may well have a plurality of nonpublic foundations, different for different citizens.
3.
So I will put to the side claims about the bases of democracy in a theory
about the nature and competence of reason, and come back to the substance
of Habermas radical account of democracy. That account takes its fundamental orientation, I said, from the idea of a self-organizing community of
free and equal citizens.1 Radically understood, democracy is not simply a
matter of selecting among competing elites (through regular elections), nor
simply a matter of ensuring, through such selection, a protected framework
of private liberties, founded on antecedent liberal commitments. Instead
democracy is a form of self-rule, and requires that the legitimate exercise of
political power trace to the free communication of citizens, expressed through
law. For the radical democrat, the fundamental fact of political sociology is
1
For this reason, Habermas understands his view as having important affinities with anarchist and
socialist ideas, once the normative core of those ideas is properly understood. Thats because
he supposes (correctly, I think) that the normative core is provided by the ideal of a free association among equals, guiding the exercise of their collective power through their common reason.
388
Joshua Cohen
not the distinction between a decision-making elite and others subject to the
decisions of that elite, and the consequent need to organize the exercise of
power by that elite, but the horizontal, communicative relation among equal
citizens; democracy establishes a framework for that relation and makes the
exercise of collective power sensitive to it.
Such a conception of democracy has two components. First, one must
describe the content of the abstract conception: What, more precisely, is it for
a political society to be a self-organizing community of free and equal citizens, and for the exercise of collective power to trace to the free communication of citizens? Assume as background that the conception is addressed
to a pluralistic society, whose members embrace competing philosophies of
life; a reflective culture, that self-consciously embraces a distinction between
the fact that a practice is socially accepted and the legitimacy of the practice
(between facticity and validity); a society whose complexity, size, and pluralism preclude social coordination through communication alone, as distinct
from market exchange and administrative power; and a society whose members engage in strategic action (Habermas 1996a, 25). What could popular
self-organization and self-government possibly amount to under these conditions? How could free communication among citizens play a regulative
role in the political life of such a society? Perhaps under these conditions the
ideal of a self-organizing community of free and equal citizens loses its
capacity to guide social and political arrangements. The first task, then, is to
address this concern: to show how a radically democratic republic might
even be conceived today (Habermas 1996b, 471).
Second, one needs to consider whether such a society is possible. Here we
take the content of the normative idealsay, of Habermas discourse model
of democracywhich is developed on the social-political assumptions just
noted, which include no unfavorable assumptions about power and human
motivation. And we ask: Can this ideal be realized, given the realities of contemporary power and human motivation? Or do sociological and psychological realism imply that we must reduce our normative expectations, and
adopt a more minimalist understanding of democracy, according to which
democracy is a system of competitive elections in which citizens chose who
will rule, rather than in any more substantial sense a system of self-rule?
Of the three questions that I propose to discuss in this essay, two fall under
the problem of content, the third under the problem of possibility.
1) I begin with the role of rights of private autonomy in a democratic constitution. The place of such rights in a radical democratic view is uncertain.
One might think that a radical democrat, concerned with the self-rule of
citizens, will make the protection of personal liberties dependent on how the
people choose to exercise their collective power. But a radical-democratic
view that cannot provide personal liberties with a secure basis will seem, to
that extent, unreasonable. In response to this concern, Habermas argues that
rights of private and public autonomy (rights of participation) are equally
389
fundamental (co-original): Indeed, each is required to explain one another.

More particularly, both kinds of rights of autonomy are founded on the
conjunction of the rule of law and the discourse principlea requirement of
impartiality that provides the basis for judgments of the legitimacy of law.
Though I agree with Habermas conclusion, I find the argument for it
unpersuasive, and I outline an alternative view that shares radical-democratic
inspiration but founds rights of private autonomy on ideas of deliberative
justification and reasonable pluralism, both devised for democratic conditions. My alternative strategy of argument makes richer normative assumptions than Habermas does: reasonable pluralism instead of mere legality,
deliberation among persons understood as free and equal rather than the
impartiality required by the discourse principle. But I think such richer
assumptions are necessary, and also defensible if our aim is to articulate a
conception of democracy, and not to found that conception on a general
philosophical theory of reason and action.
2) Next, I consider a pair of related questions about the conception of democracy itself. First, why should a radical democrat insist on reason: Assuming mass participation, why is it important for democracy to be deliberative?
And second, once we decide to insist on a requirement that law be reasonable
(rooted in practices of argumentation), why is it important for deliberation to be
democratic: Assuming reasonable outcomes, why insist on mass participation?
To explore Habermas answers to this pair of questions, I sketch his twotrack discourse model of democratic process. Democracy, thus conceived,
comprises both an informal track of free public communication, founded on
the dispersed associations of civil society, and a formal track of deliberative
decision-making by conventional political institutions that are responsive to
the informal discussion of the first track. Working together, the two tracks
suggest a way to combine mass participation, through the informal public
arena, with competent and reasonable political decisions, through deliberation in formal politics. By displacing the principal locus of participation
from formal politics to the informal public sphere, Habermas suggests a way
that the public can come into politics, without requiring small-scale states or
large, long meetings.
The answers to the questions about democracy and deliberation that
Habermas proposes on the basis of this model are suggestive, but once more
I think that conceptions of deliberative justification and reasonable pluralism,
both suited to democratic conditions, provide more compelling responses.
3) Finally, I discuss the possibility problem: Is the radical democratic ideal
of any practical relevance to the exercise of power in a modern political society?
If Habermas two-track scheme is to describe a way to join mass participation
through the informal public sphere with competent and reasonable formal
decision-making, then it must be possible for associations in the opinionforming public sphere to exercise autonomous influence on politics.
Otherwise radical democracy dissolves into a scheme in which open-ended
390
Joshua Cohen
debate among citizens proceeds in splendid isolation from the exercise of

political power.
Habermas makes a case for the possibility of such autonomous influence, and
his case has some force. Understood as an account of democracy in its most
compelling form, however, I think it is unnecessarily restrictive. To sharpen
the point, I contrast Habermas model of discursive democracy with a more
institutionalized version of radical democracy, based on an idea of directlydeliberative polyarchy that Charles Sabel and I (1997) have presented
elsewhere. This conception ties practices of deliberation more closely to the
exercise of collective power than does Habermas model of separate tracks.
This is a very full plate, and I cant hope here to discuss any of these issues
in detail. Instead, my aim is to provoke further debate about certain fundamental elements of Habermas statement of the radical democratic outlook.
I share the fundamentals of that outlook, but think that some of its elements
can be presented in more compelling ways. In general terms, Habermas
account is insufficiently explicit about the normative substance of radical
democracy, in part because he seeks to found it on a general theory of human
reason rather than the political values associated with democracy, and, in
turn, insufficiently ambitious in specifying possible institutional ideals that
are suggested by radical democracy.
II. Co-Originality and Private Autonomy

1.
According to Habermas, political philosophy has always misconceived the
relationship between civic autonomy, and the equal political liberties associated with it, and private autonomy, and the equal personal liberties associated with it: Thus far no one has succeeded in satisfactorily reconciling
private and public autonomy at a fundamental conceptual level, as is
evident if we consider the tensions between ideas of human rights and
popular sovereignty in social-contract theory (Habermas 1996a, 84).
Liberalism, in Habermas stylization, defends public autonomy in terms
of its capacity to protect private autonomy, thus turning democracy into
an instrument for the protection of private liberties: Democracy is the
response to tyranny, understood as the systematic deprivation of basic
personal liberties. Republicanism makes the protection of private autonomy
contingent on democratic collective decisions, thus rendering liberty dependent on popular judgments about the best means for achieving collective aims
or on the collective commitments contingently embraced by a particular
community. Stuck between these two options, political philosophy has
never really been able to strike a balance between popular sovereignty and
human rights, or between the freedom of the ancients and the freedom of the
moderns (Habermas 1998a, 258).
391
Habermas rejects the idea that either public or private autonomy is more
basic: The requirement of ensuring private autonomy cannot legitimately
be imposed on a people, but a legitimate legal order cannot fail to protect
private autonomy. Instead, he argues that civic and private autonomy are
co-originalequally fundamental: The universal right to equal liberties
may neither be imposed as a moral right that merely sets an external constraint on the sovereign legislator, nor be instrumentalized as a functional
prerequisite for the legislators aims (Habermas 1996a, 104). In Habermas
own explanation of co-originality, each form of autonomy is required to
explain the other; they are, as it were, co-originating, as well as co-original.
But the claim about co-origination is best understood as a theory about why
the two forms of autonomy are co-original, and not as identical to the thesis
of co-originality itself.2
More particularly, the notion of co-originality implies the following: A
democratic process of legitimate lawmaking must ensure a variety of equal
liberties to citizens, including both communicative-participatory liberties
and personal liberties. Providing both is constitutive of a process of legitimate law-making. So, for example, just as a process of legitimate law-making
cannot ensure rights of political participation, association, and expression
only for some, it cannot establish a system of legal rights in which the rights
of conscience, privacy, or bodily integrity required for personal independence in pursuing a private conception of the good are available only to
some citizens. Though the specific rights of private autonomy that receive
protection are not given by the principle itself, but need to be specified
through a democratic process, liberties of both kinds are constitutive of a
process of legitimate lawmaking.
The argument for this conclusion proceeds (schematically) as follows.3
2
Thus, Rawls agrees that both forms of autonomy are equally fundamental but argues for this
conclusion by connecting each to a fundamental aspect of the moral powers of citizens, rather
than by showing that each is in some way required by the other (see Rawls 1995; 1987, esp. secs.
5, 6). Dworkin (1996, 1926, esp. 256), too, endorses the idea that both forms of autonomy are
equally fundamental, arguing that (roughly) democracy fosters freedom only if the subjects of
the laws can also regard themselves as its authors. But to regard themselves as its authors, they
must identify with the political communityunderstand themselves as its moral members.
And they can understand themselves as moral members only if they preserve independent
judgment about the values that will govern their own individual lives and about the quality of
the communitys decisions. Personal liberties are, in turn, required for this requisite
independence. Dworkins account seems close to at least part of what Habermas identifies as
the intuitive idea behind his account of the mutually presupposing character of public and
private autonomy: That [. . .] citizens can make adequate use of their public autonomy only if,
on the basis of their equally protected private autonomy, they are sufficiently independent
(Habermas 1998a, 261, emph. added). Still, Rawls and Dworkin present explicitly normative
treatments of the importance of such individual independence, whereas Habermas theory aims
to derive the requirement of independence, and associated rights of private autonomy, from the
need to institutionalize popular sovereignty and democratic process through law. My own
discussion is also explicitly normative, and draws on the idea of respect for those who hold
views that are reasonable, politically speaking. See below, sec. III.
3
I am not confident that I have the argument right. I draw particularly on discussions in Habermas
1996a, 1998, and on discussions with Joshua Flaherty, Kenneth Baynes, and Oliver Gerstenberg.
392
Joshua Cohen
Step 1. Begin with the fact of law: that coordination and regulation under
modern conditions proceed through law. This is a basic fact of modern social
life, following from the functional requirements of a complex society, with
a considerable degree of decentralized decision-making.
Step 2. The rule of law leads to a scheme of minimal personal liberty. Two
aspects of the rule of law lead to this result. First, law is Janus-faced: Law
is a distinctive form of social coordination in that it permits individuals to
choose whether to comply for strategic or normative reasonsfrom fear of
sanctions or from respect for the laws legitimacythus assigning latitude
to act according to personal preferences. By leaving reasons for compliance
open to choice, and by rejecting the idea that individuals can be held
accountable for their reasons for compliance, legal regulation establishes a
minimal order of liberty, the liberty not to give an account of reasons for
conduct: Private autonomy extends as far as the legal subject does not have
to give others an account or give publicly acceptable reasons for her action
plans (Habermas 1996a, 120).
Moreover, it is a feature of a legal order that individuals are at liberty to
act as they wish unless the law prohibits it: Modern law as a whole implements the principle that whatever is not explicitly prohibited is permitted
(Habermas 1998a, 256). That is, individuals are to be free specifically from
coercive collective power unless it is used to enforce valid law.
I dont propose to focus on these claims about what is ingredient in the
rule of law as such, though I do want to draw attention to two limitations of
the scheme of liberty that follows from the rule of law. First, the claim is not
that legality as suchthe very existence of a legal codegives us a requirement of equality or a principle of equal subjective liberties, according to
which each person is entitled to the same liberties as others. Instead, the
existence of a legal code implies only that some individuals have some rights
of private autonomy. Furthermore, the rights of private autonomy that
emerge from the principle that whatever is not prohibited is permitted are
very weak in that there are no limits on what might be prohibited, or for
what reasons.
Step 3. Next, moving from legality as such to legitimate law, we introduce
a principle of legitimacy: The discourse principle, an interpretation of the
idea of impartiality, according to which practical norms, whether legal or
moral, are legitimate if and only if all possibly affected persons could agree to
them as participants in rational discourses (Habermas 1996a, 107). This
Principle explicates the claim to justifiability or rightness characteristic of the
Janus-faced law as such, one face of which looks to legitimacy.
I say that we introduce this principle, but the claim is that the discourse
principle explicates the claims to normative validity characteristic of the
(Janus-faced) law as such. If that is right, then the discourse principle is laws
own implicit standard of validity, and any implications that follow from
applying the discourse principle to the legal medium are implicit elements
393
of legality as such (though the connection between legality and those implications is not analytic).
Step 4. A legal code, which must establish some system of rights, can be
approved by all affected parties (approved by them through rational discourses) only if that code assigns equal liberties to each person, which
strengthens the assurance of personal autonomy; for only if the code
incorporates this equal liberty principle can the addressees of the legal code
also regard themselves as its authors:
Norms appearing in the form of law entitle actors to exercise their rights or liberties.
However, one cannot determine which of these laws are legitimate simply by looking
at the form of individual rights. Only by bringing in the discourse principle can one
show that each person is owed a right to the greatest possible measure of equal liberties
that are mutually compatible. (Habermas 1996a, 123)
The precise argument for this claim about how the discourse principle (the
requirement of impartiality) leads to equal liberties is not entirely clear,
though the basic idea is familiar, at least since Hobbes derivation of the
second law of nature: Assuming that individuals are legitimately concerned
to protect their own fundamental interests, we cannot expect universal agreement on the code unless it provides equal protection of personal liberty. Thus,
if law as such implies a minimal order of liberty, legitimate law requires a
stronger scheme of rights to liberty.
Step 5. To apply the discourse principle to lawusing it to judge the
acceptability of legal regulationsrequires that law be available as a medium
for collective regulation. But citizens can only apply the discourse principle
to law if that same legal order already ensures their rights of public autonomy: That is, they can only judge whether those affected could consent after
reasoned consideration if they have rights to reflect, to communicate, to
associate, and to bring their judgments to bear on proposed regulation. So,
we get a requirement of democracy, as the way in which the discourse principle is brought to bear on evaluating proposed laws.
So public autonomy requires private autonomy because public autonomy
requires a legal order, which order is legitimate only if it ensures equal
liberties; and private requires public, because the legal regulation of private
autonomy is legitimate only if it emerges from a discursive process that
ensure political rights. Thus we have co-originality.
The equal liberty principle that comes with legitimate law only gives us
the requirement that there be some system of equal liberties for all; it does not
give determinate content to that system. In particular, specifically liberal rights
to conscience, bodily integrity, privacy, property, etc.do not emerge simply
from the requirement that the legal code be specified through a process that
satisfies the discourse principle, but emerge instead (if they do) from the actual
exercise of civic autonomy under particular historical conditions: basic rights
must be interpreted and given concrete shape by a political legislature in response
394
Joshua Cohen
to changing circumstances (Habermas 1996a, 125). Such exercisedemocratic

processsaturates (ibid.) the otherwise abstract principle of equal subjective
liberties, and gives us a system of rights that can be understood as embodying the equal liberty principle. But, as comparative and historical reflection
on constitutional democracies suggests, there may be many such systems.
So there is an important difference in status between specifically liberal
rights (to conscience, and personal privacy) and the abstract principle of
equal subjective liberties. Satisfying the equal liberty principle by establishing some determinate system of equal liberties is required for actual decisionmaking to be discursively democratic: Antecedent to any actual exercise of
public autonomy, we can say that the system of rights adopted through
democratic discourse, whatever its precise content, must ensure equal liberties. If it does not, then collective decision-making would not count as an
application of the discourse principle to the lawjust as the system of
collective decision-making would not count as an application of the discourse principle to the law if it failed to ensure the political liberties necessary
for public autonomy. In contrast, the specification of the concrete liberties
say, the liberal libertiesessentially involves actual discourse: The conjunction of legality and discourse simply does not yield a determinate system of
private liberties, only the requirement that some system of equal private
liberties for each must be adopted.
The argument for the constitutive status of the equal liberty principle is
based, so to speak, on the theorists or reflective persons own application of
the discourse principle: put otherwise, it is based on hypothetical discourse
rather than actual discourse. This, I believe, is the force of the idea that
private autonomy is at first abstractly posited (Habermas 1996a, 121).
We, as theorists or reflective citizens thinking about constitutional issues, ask
what system of rights is normatively justified: What rights must citizens
accord one another if they want to legitimately regulate their common life
by means of positive law (Habermas 1996a, 82). To answer this question, we
ask what kind of system can be impartially justified; and we approach this
issue by asking what system the addressees of the law could agree to under
idealized conditions. We then argue, by appeal to the discourse principle,
that they (or we) could only agree, with reason, to a system of equal liberties
for all. Of course we may bring this argument to actual discourse: But the
argument we would make is an argument about what idealized discourse
would deliver, together with an argument to the effect that idealized
discourse reconstructs our understandings of normative validity.
In contrast, the specification of the concrete libertiessay, the liberal libertiesessentially involves actual discourse through a democratic process.
The conjunction of legality and discourse simply does not yield a determinate
system of private liberties, but only the requirement that some system of equal
private liberties for each must be adopted: Specificity results inasmuch as
the external perspective taken initially by the theorist is, in the course of
395
elaboration, internalized in the system of rights (Habermas 1996a, 122). So

we know from the argument at Step 4 that if actual political decision-making
does not yield a constitution that meets the equal liberty principle, then that
decision-making is not suitably discursive: The argument at Step 4 shows
that satisfying the equal liberty principle is constitutive of actual
discursiveness. To underscore, I am not claiming that actual decision making
is discursive only if citizens already operate within a legal order that satisfies
that principle of equal liberty, but that actual decision-making is discursive
an application of the discourse principle (as implicitly understood or
explicitly articulated) only if participants endorse the equal liberty principle.
2.
I am in general sympathy with this line of thoughtwith the idea that both
forms of liberty are equally fundamental, the associated claim that personal
liberties are constitutive of a process of legitimate lawmaking, and the idea
that this constitutive role flows from the requirement (expressed in the
requirement of discursive justification) that the addressees of the law must
be able to see themselves as its authors.
Still, I have three concerns about (perhaps objections to) this line of argument: I am not sure why the legal form itself plays an essential role in the
argument; I dont find the equal liberty principle itself compelling, as distinct from a principle that assigns special importance to basic or fundamental liberties; andthe point I propose to concentrate on hereI dont see
how the discourse principle gives us equal liberties. The problem is that the
discourse principle, which states, again, that practical norms are legitimate if
and only if all possibly affected persons could agree to them as participants
in rational discourses, appears to rely on a highly generic account of
reasonsnot an account restricted to political argument in a democracy of
equal members. But with no restriction on what can count as a reason, and
with the full panoply of pragmatic, ethical, and moral reasons in play in the
relevant forms of discourse, it would seem that anything could come from
discourse. If all we need is that all possibly affected persons could agree to
them as participants in rational discourses, and there are no constraints on
acceptable reasons, then what constrains the discursive equilibrium in the
way that Habermas proposes?
3.
Let me suggest, then, an alternative argument for the constitutive role of
non-political liberties, based on two central ideas: the idea of reasonable
pluralism and a deliberative conception of political justification, framed by
the fundamental democratic idea of citizens as free and equal.4 Though these
4
The discussion that follows draws on Cohen 1998.

396
Joshua Cohen
assumptions appear to be normatively more substantive than the notions of

impartiality and Janus-faced legality that Habermas officially relies on, I
dont think they (or similarly richer normative ideas) can be avoided in a
successful case for liberties (and co-originality).
1) I begin with the fact of reasonable pluralism: The fact that there are
distinct and incompatible philosophies of life to which people, who are
reasonable politically speaking, are drawn under favorable conditions for
the exercise of practical reason. By a philosophy of lifewhat Rawls calls a
comprehensive doctrineI mean an all-embracing view, religious or
secular in foundation, liberal or traditionalist in substance, that includes an
account of all ethical values and, crucially, provides a general guide to conduct, individual as well as collective. People are reasonable, politically speaking,
only if they are concerned to live with others on terms that those others,
understood as free and equal, can also reasonably accept: only if they accept
what Rawls calls the criterion of reciprocity (see Rawls 1999, 578).
I say reasonable, politically speaking, because the relevant notion of
reasonableness is suited to political questions: Generically speaking, a reasonable person is someone who gives due attention to the considerations
that bear on an issue, and acts in light of that attention. So the notion of being
reasonable, politically speaking, is a matter of giving due attention to the facts
about the political relation of citizens in a democracy: the fact that political power
is the collective power of citizens, understood as equals. The fact of reasonable pluralism, then, is that conscientious, good-faith efforts in the exercise of practical reason, by politically reasonable people (thus understood),
do not converge on a particular philosophy of lifethat such philosophies
are matters on which (politically) reasonable people disagree.
2) A deliberative conception of democracy puts public reasoning at the
center of political justification. According to the deliberative interpretation
of democracy, then, democracy is a system of social and political arrangements that institutionally ties the exercise of collective power to free
reasoning among equals. This conception of justification through public
reasoningthe core of the deliberative democratic idealcan be represented in an idealized procedure of political deliberation, constructed to
capture the notions of free, equal, and reason that figure in the deliberative
democratic ideal. The point of the idealized procedure is to provide a model
characterization of free reasoning among equals, which can in turn serve as
a model for arrangements of collective decision-making that are to establish
a framework of free reasoning among equals. Using the model, we can work
out the content of the deliberative democratic ideal by considering features
of public reasoning in the idealized case, and then aiming to build those
features into institutions.
3) Thus, in an ideal deliberative procedure, participants are and regard
one another as free: Recognizing the fact of reasonable pluralism, they
acknowledge that no comprehensive moral or religious view provides a
397
defining condition of participation or a test of the acceptability of arguments

in support of the exercise of political power. To represent participants as free
is not to say that their philosophy of life is, morally or metaphysically
speaking, a matter of choice. To someone who has a religious view and takes
Gods laws as the touchstone of morality, for example, believing the view is
a matter of believing what is true and acting on it a matter of fulfilling
obligations that are not self-legislated, and are perhaps more fundamental
than political obligations. But politically speaking, citizens are free in that it
is open to them to accept or reject such views without loss of status.
Moreover, participants regard one another as formally and substantively
equal. They are formally equal in that the rules regulating the ideal procedure
do not single out individuals for special advantage or disadvantage. Instead,
everyone with deliberative capacitieswhich is to say, more or less all
human beingshas and is recognized as having equal standing at each of
the stages of the deliberative process. Each, that is, can propose issues for the
agenda, propose solutions to the issues on the agenda, offer reasons in
support of or in criticism of proposed solutions. And each has an equal voice
in the decision. The participants are substantively equal in that the existing
distribution of power and resources does not shape their chances to contribute to deliberation.
In addition, they are reasonable in that they aim to defend and criticize
institutions and programs in terms of considerations that others, as free and
equal, have reason to accept, given the fact of reasonable pluralism and on the
assumption that those others are themselves reasonable.
4) Which considerations count as reasons? Generically speaking, a reason
is a consideration that counts in favor of something: in particular, a belief, or
an action. That is not meant to be illuminating analysis of the concept of a
reason: I doubt that illuminating analysis is available, or that it would be
helpful in answering our question. What is needed is not an account of what
a reason is, but of which considerations count as reasons. And the answer to
this question depends on context: Whether considerations count in favor in
the relevant way depends on the setting in which they are advanced. Applying this point to the issue at hand: A suitable account of which considerations count as reasons for the purposes of an account of democratic
deliberation will not take the form of a generic account of what a reason is,
but a statement of which considerations count in favor of proposals within a
deliberative setting suited to the case of free association among equals,
understood to include an acknowledgment of reasonable pluralism. This
background is reflected in the kinds of reasons that will be acceptable:
meaning, as always, acceptable to individuals as free and equal citizens.
I have specified the relevant deliberative setting as one in which people
are understood as free, equal, and politically reasonable, and as having
conflicting, reasonable philosophies of life. Under these conditionswithin
the idealized deliberative setting that captures themit will not do simply
398
Joshua Cohen
to advance considerations that one takes to be true or compelling. For such

considerations may well be rejected by others who are themselves reasonablein being prepared to live with others on terms that are acceptable
to those others, given their different comprehensive viewsand endorse
conflicting comprehensive views. One needs instead to find reasons that
are compelling to others, where those others are regarded as (and regard
themselves as) equals, with conflicting reasonable commitments. Considerations that do not meet these tests will be rejected in the idealized setting and
so do not count as acceptable or sufficient political reasons. Lets say then
that a consideration is an acceptable political reason just in case it has the
support of the different comprehensive views endorsed by reasonable
citizens.
5) These observations about reasonable pluralism, and the role of background understandings of citizens as free, equal, and reasonable in constraining the set of political reasonsthus giving content to democracys public
reasonplays an important role in understanding the essential role of nonpolitical liberties within the account of democracy.
First, people hold some of their commitmentsfor example, religious
commitmentson faith, and those commitments impose what they take to
be overriding obligations. Such commitments are not, as such, unreasonable.
To be sure, faith transcends reason, even as reason is understood within
the tradition to which the commitments belong. Still, beliefs held on faith
perhaps beliefs in what are understood to be revealed truthsare not as
such unreasonable. But such beliefs can reasonably be rejected by others,
who rely on the darkness of an unconverted heart. So they cannot serve to
justify legislation. And the fact that they cannot will impose pressure for
personal libertiessay, religious, expressive, and moral liberty.
Second, acceptable considerations will have different weights in political
justification. And the weight will depend on the nature of the regulated
conduct, in particular the weight of the reasons that support the conduct.
Take considerations of public order, for example. They provide acceptable
reasons for regulating conduct. Different views have different ways of
explaining the value of public order: utilitarians will found it on considerations of aggregate happiness, Kantians on the social preconditions of
autonomous conduct, others on the intrinsic value of human life and human
sociability. Moreover, people are bound to disagree about what public order
requires. But it will not be acceptable to suppose that, as a general matter, the
value of public order transcends all other political values. Except perhaps in
the most extreme circumstances, for example, a state may not impose a
blanket prohibition on alcohol consumptionincluding consumption in
religious servicesin the name of public order. The reasons that support
such consumption include considerations of religious obligationmore
generally, considerations of fundamental obligation, which are normally
overridingwhich will provide a suitable basis for rejecting a justification
399
cast in terms of the value of public order, except in the most extreme
conditions. To be sure, not all citizens acknowledge the obligations in question. But even those who do not can see the weightiness of those reasons,
within the outlooks of other politically reasonable citizens.
As these two observations indicate, pressure for liberty comes from at
least two sources: The pluralism of philosophies of life among political reasonable citizens leads to the rejection of some bases of restriction as politically weightless; other bases of restriction will not be weightless, but insufficient
to outweigh the reasons that can be acknowledged, consistent with reasonable pluralism, as commending or commanding conduct. Taking these
two considerations together, we have the basis for a strong case for religious,
moral, and nonpolitical expressive liberties: Conduct in these areas is supported by strong (perhaps compelling) reasons, as when religious exercise is
a matter of obligation according to a persons reasonable religious outlook.
Moreover, standard reasons for restrictionreligious and sectarian moral
reasonswill often be weightless.
Given this deliberative rationale for personal liberties, we can see why
their protection would be constitutive of democracy, and how, therefore, we
get co-originality. For imposing regulations in the name of reasons that are
either weightless or of insufficient force to override reasonable demands is a
violation of the fundamental democratic idea that the authorization to
exercise state power must arise from the collective decisions of the equal
members of a society who are governed by that powerthat it must be
supported by reasons that can be shared by the set of politically reasonable
citizens over whom power is exercised. Decisions to regulate are not suitably
collective, for the addressees of the regulations cannot all be included in
their collective authorization.
III. Discursive Democracy

1.
Habermas conception of discursive democracy provides an idealized, normative account of democratic process. Set within a constitutional order that
protects personal and political liberties, discursive democracy ties together
two elements or tracks of a process of collective decision-making: The
informal discussion of issues in an unorganized, wild, decentered (not
centrally coordinated) public sphere that does not make authoritative collective decisions, and a more formal political process, including elections
and legislative decision-making, as well as the conduct of agencies and
courts. In the formal process, candidates and elected legislators deliberate
about issues, make authoritative decisions by translating the opinions formed
in the informal sphere into legal regulations, and monitor the execution of
those decisions by administrative bodies. Whereas discourse in the public
400
Joshua Cohen
sphere is open-ended, the decision-making procedures are subject to conditions of deliberative-democratic legitimacy: for example, that decisions are
to be founded on reasons; that the processes are to be open and fair; that they
are to be free of coercion; and that results are to be determined by the better
argument (Habermas 1996a, 3056).
This discursive model of democratic process appears to be founded on an
hypothesis about the connection between idealized discourse and actual
democratic decision-making (understood as proceeding along both tracks).
The central idea is that democratic procedures should produce rational
outcomeswhere rational outcomes are those that would emerge from
idealized discourse. Suppose, then, that we think collective decision-making
as a form of problem solving: The production of legitimate law through
deliberative politics represents a problem-solving procedure that needs and
assimilates knowledge in order to program the regulation of conflicts and
the pursuit of collective goals (Habermas 1996a, 318). Then, a discursively
democratic process of decision-making provides grounds for expecting reasonable solutions to problems:
The democratic procedure is institutionalized in discourses and bargaining processes
(assumed to be fair) by employing forms of communication that promise that all
outcomes reached in conformity with the procedure are reasonable []. Deliberative
politics acquires its legitimating force from the discursive structure of an opinionand will-formation that can fulfill its socially integrative function only because citizens expects its results to have a reasonable quality. (Habermas 1996a, 304, 296)
More generally,
democratic procedure makes it possible for issues and contributions, information
and reasons to float freely; it secures a discursive character for political will-formation; and it thereby secures that fallibilist assumption that results issuing from proper
procedure are more or less reasonable. (Habermas 1996a, 448)
In achieving such reasonable results, the two tracks of deliberative politics

play distinct roles, which correspond to different stages in an idealized process of problem-solving. Informal communication in the public sphere
provides a close-to-the-ground and unregulated arena for detecting new
problems, bringing them to public view in a non-specialized language, and
suggesting ways to address those problems: Because information is not
controlled and communication is unrestricted, new problems situations can
be perceived more sensitively (Habermas 1996a, 308). Thus the communicative structures of the public sphere constitute a far-flung network of
sensors that react to the pressure of society-wide problems and stimulate
influential opinions (Habermas 1996a, 300). It is founded on a network of
associations that specialize [] in discovering issues relevant for all society,
contributing possible solutions to problems, interpreting values, producing
good reasons, and invalidating others (Habermas 1996a, 485).
401
Formal political processeselections, legislatures, agencies, and courts

provide the second stage in an idealized problem-solving system. They provide institutionally regulated ways to assess ideas: to deliberate about
proposals under fair conditions, evaluate alternative solutions, and make
authoritative decisions after due consideration. So on the second, institutional track we have a disciplined testing through reason of proposals that
emerge from open-ended public discussion:
The operative meaning of these regulations consists less in discovering and identifying problems than in dealing with them; it has less to do with becoming sensitive
to new ways of looking at problems than with justifying the selection of a problem
and the choice among competing proposals for solving it. The publics of parliamentary bodies are structured predominantly as a context of justification. These bodies
rely not only on the administrations preparatory work and further processing but
also on the context of discovery provided by a procedurally unregulated public sphere
that is borne by the general public of citizens. (Habermas 1996a, 307)
Thus the case for the two-track process is founded on the claim that it will
generate rational outcomes. And that claim is based on the interplay in discursive democracy between an open-ended exploration of problems and
possible solutions, which influences the premises of judgment and decisionmaking in the political system (Habermas 1996b, 4867), and a disciplined,
rational assessment of proposed solutions. This interplay between discovery
and justification supports the presumption that the results will conform to
idealized, discursive problem solving. Because the two phases of reasoning
in the actual process conform to idealized reasoning, the actual process will
generate results like those that idealized discourse would generate:
Thus the normative expectation of rational outcomes is grounded ultimately in the
interplay between institutionally structured political will-formation and spontaneous,
unsubverted circuits of communication in a public sphere that is not programmed to
reach decisions and thus is not organized. (Habermas 1996b, 485)
Thus, Habermas interprets popular sovereignty procedurally, as the possible

influence on authoritative political decisions of public discourses in an
autonomous communicative network, rather than as the direct control of
legislation by a determinate and coherent popular will. In this way, the twotrack idea identifies a way that authorship of the terms and conditions of
political association by free, equal, reasonable citizens can be made compatible with the modern organization of social and political power.5 Think of
the achievement this way: Habermas has shown that the pluralist critique of
sovereignty and of a state-centered conception of politics, and associated
insights about the importance of social organization in modern democracy
5
See Habermas 1998b, 251, on popular sovereignty as consisting in interactions between
legally institutionalized will-formation and culturally mobilized publics.
402
Joshua Cohen
(about the social bases of democratic governance), can be freed from the
pluralists own theory of politics as bargaining between and among groups
that represent well-defined interests. Those insights can be wedded instead
to a conception of politics in which reasoning about the basic terms of association plays a central role. But the marriage requires the idea of the informal,
discursive public sphere, in which all can freely participate, and which has
the capacity to influence opinion through argument, and thereby shape the
agenda of formal politics. This strikes me as a fundamental contribution to
democratic thought: a remarkable reconception, with redemptive promise.
2.
So much for Habermas view. Now I come to the pair of questions I identified earlier: Why should democracy be deliberative? And why should
deliberation be democratic?
First, then, why is it important for democracy to be deliberative? Why
should collective decision-making involve the giving of reasons of suitable
kinds, rather than simply a fair aggregation of citizen interests? One rationale,
already suggested, arises from concerns about the impartial justifiability of
(or, as Habermas puts it, the rationality of) outcomes: the concern captured
in Habermas discourse principle. Suppose we have a hypothetical test of
validity: Outcomes are justified only if they could be accepted by people
who give suitable weight to the reasonable objections of others, assuming
those others to be free and equal. Deliberative democracy, then, may seem a
natural way to achieve such impartially justified outcomes. For it is a form
of democracy that aims to mirror hypothetical conditions of good information, attentiveness to reasons, and regard for others as equals by requiring,
in particular, that the exercise of power be justified by appeal to considerations that others acknowledge as reasons, and assuming a shared commitment to such justification. Bargaining under fair background conditions may
also produce rational outcomes, but deliberation generates a stronger presumption because it requires attentiveness to reasons.
Rawls suggests this thought about the relationship between a hypotheticalcontractual notion of justice and actual political decision-making when he
remarks that his principle of participationrequiring fair political equality
transfers the requirement of equal standing that defines the original position
into the design of the constitution: we have an effective political procedure
that mirrors the fair representation of persons in the original position
(Rawls 1993, 330). Deliberative democracy might be seen as giving this idea
of connecting contractual and actual a Scanlonian twist. Scanlons contractualism (1998, chap. 5) presents an idealized model of moral reasoning
rather suggesting that rational choice under conditions of ignorance can
provide a substitute for such reasoning. Correspondingly, then, instead of
merely transferring a requirement of equal standing or fair representation,
403
deliberative democracy institutionalizes the concern for justifiability to

others from their standpoint that defines Scanlons ideal contractualism,
moving that concern from the contractual to the actual, and applying it to the
special case of binding collective choice.
Brian Barry (1995, sec. 16, 100) has a very illuminating discussion of this
idea. He considers what he calls the circumstances of impartiality: the
social-political conditions that approximate those of a Scanlonian original
position. Borrowing this term, then, we might think of deliberative democracy as an essential part of the circumstances of impartiality. The idea is that
if we wish to realize impartial justicesay, to satisfy Habermas discourse
principlethen we must embrace in our actual collective decision-making a
commitment to mutual reason-giving (and institutional conditions that express
and sustain that commitment) of a kind that approximates the idealized
practice of mutual reason-giving that determines the requirements of justice.
Put simply, impartial justice must, arguably, be aimed at to be achieved; and
here, aiming at it means approximating its procedures.
If the requirements of justice are fixed by a kind of impartial reasoning
under hypothetical conditions, then, even if we do not know what would be
agreed to, we will, arguably, only achieve the requirements of justicethe
outcomes that could or would be agreed toif we make collective decisions
using our best actual approximations to impartial reasoning. We cannot
simply trust the achievement of justice to the pursuit of interests even under
ideally fair conditions, for those fair conditions themselves are likely to
erode without a commitment to democratically-deliberative decision-making.
I think this argument has much to be said for it, and it seems to be
Habermas idea about the relationship between the standard of justification
stated in the discourse principle and deliberative-democratic practice:
The democratic procedure is institutionalized in discourses and bargaining processes
(assumed to be fair) by employing forms of communication that promise that all
outcomes reached in conformity with the procedure are reasonable []. Deliberative politics acquires its legitimating force from the discursive structure of an
opinion- and will-formation that can fulfill its socially integrative function only
because citizens expect its results to have a reasonable quality. (Habermas 1996a, 304,
296, 448)
They have such expectation because actual decision-making, under the conditions of the discursive model, approximates (in ways noted earlier) idealized
deliberation.
Still, the case for the importance of deliberation need not proceed solely in
terms of the requirements on a system of collective decision-making that is
to match the results that would be achieved were decision-making to be
ideally deliberative. The virtues of the deliberative view are also more intrinsic, and allied closely with its conception of binding collective choice, in
particular with the role in that conception of the idea of reasons acceptable
404
Joshua Cohen
to others whose conduct is governed by those choices, and who themselves

have reasonable views. By emphasizing the importance of reasons acceptable to all citizens, the deliberative view expresses an especially compelling
picture of the possible relations among people within a democratic order;
moreover, it states a forceful ideal of political legitimacy for a democracy. I
take up these two points in turn.
First, the deliberative conception offers a forceful rendering of the fundamental democratic ideathe idea that decisions about the exercise of state
power are collective. It requires that we offer considerations that others
whose conduct will be governed by the decisions, and who are understood
to be free, equal, and reasonable, can accept, not simply that we count their
interests, while keeping our fingers crossed that those interests are outweighed. The idea of popular authorization is reflected not only in the
processes of decision-making, but in the formand as we have seen, the
contentof political reason itself.
This point about the attractions of the deliberative interpretation of
collective decisions can be stated in terms of ideas of political autonomy and
political community. If a political community is a group of people sharing a
comprehensive conception moral or religious view, or a substantive national
identity defined in terms of such a view, then reasonable pluralism ruins the
possibility of political community. But an alternative conception of political
community connects the deliberative view to the value of community. To see
how, notice first that by requiring justification on terms acceptable to others,
deliberative democracy provides for a form of political autonomy. Without
denying the coercive aspects of common political life, it requires that all who
are governed by collective decisionswho are expected to govern their own
conduct by those decisionsmust find the political values that provide the
bases of those decisions acceptable, even when they disagree with the details
of the decision.
Through this assurance of political autonomy, deliberative democracy
achieves one important element of the ideal of community. Not because
collective decisions crystallize a shared ethical outlook that informs all social
life generally, nor because the collective good takes precedence over liberties
of members. Rather deliberative democracy is connected to political community because the requirement of providing reasons for the exercise of
political power that are compelling to those who are governed by it itself
expresses the full and equal membership of all in the sovereign body
responsible for authorizing the exercise of that power, and establishes the
common reason and will of that body.
The deliberative conception of democracy also presents an account of
when decisions made in a democracy are politically legitimate and how to
shape institutions and forms of argument so as to make legitimate decisions.
Generally speaking, we have a strong case for political legitimacy when
the exercise of political power has sufficient justification. But, as a conceptual
405
matter, a person can believe that the exercise of power is well-justified

therefore legitimatewhile also acknowledging that others over whom it is
exercised reject the justification. As a conceptual matter, legitimacy does not
require that the relevant justification be acknowledged as such by those who
are subject to the legitimate power: there need be no justification to them. But
the background of democracythe idea of citizens as free and equaland
the fact of reasonable pluralism are important in characterizing a more
limited conception of justification: Because of these conditions, the relevant
justification must be addressed to citizens, by which I mean that its terms
must be acknowledged as suitable by those subject to political power. Given
that citizens have equal standing and are understood as free, and given the
fact of reasonable pluralism, we have an especially strong showing of legitimacy when the exercise of state power is supported by considerations acknowledged as reasons by the different views endorsed by reasonable citizens,
who are understood as equals: No other account of reasons is suited for this
case. The deliberative conception articulates an account of political legitimacy suited to democratic conditions, and through the ideal deliberative
procedure it aims to specify the content of those conditions.
3.
I turn now to the second question: Why should deliberation be democratic?
Assume, arguendo, that the discourse principle can only be satisfied by
deliberative decision-making. Still, we need to ask why deliberative political
decision-making needs to be democraticto satisfy the principle of political
equality, with its guarantees of universal political rights. The mere fact that
the outcomes are to match those that could be accepted by all under idealized conditions does not seem to lead to this conclusion: not, anyway,
without further argument. It might be argued that an ideal deliberative
procedure is best institutionalized by ensuring well-conducted political
debate among elites, which enables citizens to make informed choices among
them and the alternatives they represent. Why does a deliberative view such
as Habermas require equal political liberties? How does it connect to concerns about participation and political equality? Why, in short, does the
discourse principle become the democratic principle, once it assumes legal
shape?
I am not sure that I understand Habermas answer to this question. In at
least one place, he notes that his view has a dogmatic core in its commitment to an idea of autonomy according to which human beings act as
free subjects only insofar as they obey just those laws they give themselves
in accordance with insights they have acquired intersubjectively (Habermas
1996a, 446). A different line of thought, that does not depend on this normative understanding of autonomy, runs parallel to the argument about why
democracy needs to be deliberative. Here the idea would be that the best
406
Joshua Cohen
way to determine what would be agreed to by all in idealized discourse is to

see what is actually agreed to in actual democratic discourse, in which all
have a right to participate. All we need to get this result is to add a nonnormative assumption about personal autonomy, e.g., the thesis that individuals are the best judges and most vigilant defenders of the interests and
concerns that they would have in idealized deliberation: Nothing better
prevents others from perspectivally distorting ones own interests than actual
participation. It is in this pragmatic sense that the individual is the last court
of appeal for judging what is in his best interest (Habermas 1990, 67). A
third argument is that equal political liberties are required because thats
what applying the discourse principle implies: no democracy, no rational
approval in idealized discourse (Habermas 1996a, 127).
Here, again, I think the first two points have some force, but that the
bridge between an idealized account of political justification and actual
democracy could be strengthenedand freed from a philosophy of life that
assumes the supreme value of autonomy and from the empirical assumption
of autonomyby developing the third. And that means presenting a more
explicit account of the nature of idealized justification and the kinds of reasons suited to it, given the background ideas of reasonable pluralism and
members as free and equal. In particular, three considerations are important
in an account of why deliberation should be democratic.
First, if we assume the equal liberty principle (or some analog to it, requiring personal liberties), the deliberative view can appeal to traditional instrumental reasons in support of institutions that ensure equal political rights. In
particular, such rights provide the means for protecting other basic rights
for example, those that are protected under the equal liberty principle.
Though such instrumental reasons are not the sole basis for equal political
rights, part of the case for them turns on their protective role.
A second consideration turns on the issue of acceptable reasons. Consider
conventional, historical justifications for exclusions from or inequalities of
political rights. Those justificationswhether of formal exclusion or unequally weighted voteshave typically been based on considerations about
racial, gender, ethnic, or religious differences. But such considerations will
not provide acceptable reasons in public deliberation, given the background
conception of members as free and equal, and so arrangements of collective
decision-making cannot be justified by reference to them.
The third consideration is analogous to a central point that figured in the
case for private liberties. A characteristic feature of different philosophies of
life is that they each give us strong reasons for seeking to shape our politicalsocial environment: for exercising responsible judgment about the proper
conduct of collective life. The theories underlying those reasons cover a wide
range: Aristotelian views about the central role of civic engagement in a
flourishing human life; Rousseauean claims about the connection between
realizing the personal autonomy that is essential to human nature and
407
political participation in a democratic polity; and views, founded on religious

convictions, about the commanding personal responsibility to ensure social
justice and the corresponding personal sin of failing in that responsibility.
Common ground among these competing, reasonable philosophies is that
citizens sometimes have substantial, sometimes compelling reasons for addressing public affairs, and therefore a fundamental interest in favorable
conditions for forming judgments about the proper directions of policy, and
acting on their judgments.
The failure to acknowledge the weight of those reasons for the agent and
to acknowledge the claims to political opportunities that emerge from them
reflects a failure to respect the background idea of citizens as equals. We
acknowledge the weight of these reasons is reflected in part by equal rights
of participation.
4.
In my remarks about both personal liberty and democracy, I have been
emphasizing in effect that we need to build into the actual process of political decision-making the conclusions of idealized, hypothetical deliberation,
where the idealizations arguably articulate and organize ordinary understandings of acceptable political argument, under democratic conditions. To
which Habermas might object that I am not giving suitable weight to actual
deliberation. He says: The justification of norms and commands requires
that a real discourse be carried out and this cannot occur [] in the form of
a hypothetical process of argumentation occurring in the individual mind
(Habermas 1990, 68).
Here, I want in part to agree. It is not sufficient for political justification
that outcomes be rationalizablethat the deliberative process issue in
decisions for which appropriate reasons could be cited, and that it be left to
another institution, say, a court, to determine whether that condition is met.
Outcomes in a deliberative democracy are to be arrived at through discussion in which reasons of the appropriate kind are given by participants.
Four considerations support the importance of actual deliberation:
1) Though deliberative justifiability itself is important, it mustas the
BarryHabermas argument about the circumstances of impartiality suggests
be aimed at to be achieved; that is, it will not in general be true that results
achieved through a process of exchange or bargaining (under fair conditions), or outcomes that reflect a balance of power, will be defensible by
reasons of an appropriate kind. So requiring actual deliberation helps to
establish a presumption that results can be defended through reasons, and
thus a presumptive legitimacy for outcomes of collective decision-making.
2) Offering reasons to others expresses respect for them as equal members
of a deliberative body. So actual deliberation plausibly helps to foster mutual
respect, which in turn encourages citizens to confine the exercise of power as
408
Joshua Cohen
the deliberative idea requires. No similar result can be expected if we assign

the job of assessing the justifiability of outcomes to a separate institution.
3) Actual deliberation is a way to acquire and master fundamental political principles and their rationale by drawing on those principles and having
to defend them in open argument. The fact that the principles can be defended
in hypothetical discourse of course does not suffice for their understanding
or motivational impact.
4) In actual reason-giving, citizens are required to defend proposals by
reference to considerations that others acknowledge as reasons, and not
simply by reference to their own interests. To the extent that such public
reasoning shapes preferences, conflicts over policy will be reduced, as will
inclinations to strategically misrepresent circumstances. A crucial point here
is that the extent of preference-diversity is not fixed, not given prior to
political deliberation. Not that the aim of such deliberation is to change
citizen preferences by reducing their diversity: The aim is to make collective
decisions. Still, one thought behind a deliberative conception is that public
reasoning itself can help to reduce the diversity of politically relevant preferences because such preferences are shaped and even formed in the process
of public reasoning itself. And if it does help to reduce that diversity, then it
mitigates tendencies to distortion even in strategic communication.
So actual deliberation is important. But an account of democracy as the
source of legitimate law must give some account of what the relevant democratic background is, such that deliberation under democratic conditions,
thus specified, results in legitimate law. And we cant simply say that the
correct specification of those legitimacy-establishing conditions is itself to be
the product of actual democratic deliberation, because we need an account
of the conditions that make deliberation democratic and that make democracy deliberative. To be sure, the account of those conditions may receive
support from actual deliberation, as citizens master its principles and the
reasons for them; indeed, if they do not achieve such mastery and understanding, if the ideal is not actualized in the reflective political thought of
citizens, there may well be problems about democratic stability. So actual
deliberation can (perhaps must) renew the constitutive conditions of a
democratic process of legitimate lawmaking. But it cannot bear the full
weight of specifying those conditions.
IV. Problem of Possibility6

1.
Finally, I come to Habermas answer to the question of how radical democracy is possible. Given the realities of social and political power, how is
6
This section draws substantially from Cohen and Sabel 1997.
409
the abstract ideal of a self-organizing community of free and equal

citizens, coordinating their collective affairs through their common reason,
of practical relevance? Habermas answer draws on the two-track discursive
model.
The two-track model indicates how (communicative) power might flow
from citizens, reasoning in a dispersed network, through a deliberative legislature, to administration. But this flow from dispersed publics to administrative implementation is threatened by the control, perhaps manipulation,
of formal and informal public discussion by organized social power and
political agencies (including parties and interest groups) with interests and
modes of argument fixed independently from the concerns and opinions of
freely communicating citizens. The possibility of the proper flow, in turn, is
founded on the capacity of associations in the informal, unspecialized public
sphere autonomously to identify issues and concerns, including encompassing social problems (Habermas 1996a, 365), that lie outside the agenda
of formal politics, bring those issues and concerns to wider public attention,
propose solutions to them, and, by moving public opinion, influence the
operations of the formal political system.
The key is autonomously (Habermas 1996a, 375; 1996b, 484). The discovery, articulation, and exploration of concerns, as well as the formulation
on new understandings of reasonable practice, must not itself be subject to
the initiation or subsequent control of organized political or social powers,
with their specialized interests, routines, and vocabularies. Only when
initiative and subsequent organized influence on legislative and administrative power come from outside institutionalized, routinized poweronly
if it breaks free from the unofficial circulation of this unlegitimated power
(Habermas 1996a, 328)can we say that the flow of power moves from
equal citizens, through law, to administration (Habermas 1996a, 380). And if
it can, then democracy is possible, despite the realities of organized social
and administrative power.
The requirement of outside initiative strikes me as ill-conceived: Lots of
political movements are initially provoked by developments internal to
conventional institutions and actorsfor example, by competition between
and among elites who mobilize popular support with the expectation that
that mobilization can be controlledeven though the subsequent evolution
of those movements proceeds independently; when it comes to popular
movements, genesis is not identity. But this is largely a matter of detail
though it does underscore the difficulties of giving empirical content to the
relevant notion of autonomy.
My larger concern with Habermas answer to the possibility problem
begins from the observation that this answer is, as Frank Michelman (1996,
38) has put it, a dispiriting meltdown of popular sovereignty. On
Habermas account, radical democracy is possible largely because of the
sporadic bursts of energy by social movements that, in their role as dispersed
410
Joshua Cohen
sensors, detect popular concerns that are off the public agenda, suggest
novel solutions to them, and perhaps influence legislation (and ultimately
administration). To demonstrate that possibility, it suffices to show that
under certain circumstances civil society can acquire influence in the public
sphere, have an effect on the parliamentary complex, [] and compel the
political system to switch over to the official circulation of power
(Habermas 1996a, 373). And to make this case, it suffices to show that in
a perceived crisis situation, the actors in civil society [] can assume a
surprisingly active and momentous role (Habermas 1996a, 380).
In saying that this conclusion isto use Michelmans worddispiriting, I
do not disparage at all the momentous role of the social movementsfor
example, feminist and environmentalthat Habermas here has in mind
(Habermas 1996a, 381). But the argument does make democracy, as reconceived, foreign to the settled institutional routines of a modern polity. Except
for the exceptional conditions in which associations break free from the
institutionalized circuit of power, so to speak, the system rules: a reconception with limited redemptive force.7
Before going further, a qualification is in order. The conclusions we should
draw from Habermas account of democratic possibilityhow dispiriting
we should find itdepend on which of two purposes we assign to the argument. On one construction, the aim is simply to show that the old-fashioned,
radical-democratic ideal of a self-governing association of free and equal
citizensauthors of the laws, not merely their addresseesstill can connect
to modern politics, thus turning back realist arguments for less demanding
accounts of democracy. Interpreted this way, the argument succeeds, even if
Habermas is only able to point to occasional disruptions of the normal
routines of institutionalized power. The disruptions suffice as proof of
possibility.
Suppose instead that the purpose of the two-track model, with its sharp
distinction between free-floating discourse in a network of autonomous
associations and institutional decision-making and exercise of power, is to
identify democracys most attractive possibilities. Then the view strikes me
as less compelling. Perhaps because he is principally concerned with the
issue of possibility, Habermas thinks it suffices to make the case for autonomous influence flowing from the periphery, under conditions of crisis. But
once that case is on hand, we can ask whether there are other forms of citizen
participation that would more fully achieve the radical democratic promise.
Those forms would need to meet three conditions: They must permit and
encourage inputs that reflect experiences and concerns that may not
occupy the current agenda (sensors, rooted in local experience and information); they must provide disciplined assessment of proposals through
7
In this respect, Habermas view bears some resemblance to Bruce Ackermans (1991) account
of dualist democracy, with its distinction between normal and constitutional politics.
411
deliberation that encompasses fundamental political values; and (here we go

beyond Habermas emphasis on social movements in periods of crisis), they
must also provide more institutionalized, regularized occasions for citizen
participation in collective decision-making (and perhaps, by so doing,
improve the quality of discourse in the informal public sphere). In brief,
they must be autonomous, deliberative, and institutional.
2.
Sabel and I (1997) have recently suggested some ideas along these lines,
captured in the idea of a directly-deliberative polyarchy. The fundamental idea
is to institutionalize direct problem-solving by citizens, and not simply to
foster informal citizen discussion with promises of possible influence on the
formal political arena. In directly-deliberative polyarchy, collective decisions
are made through public deliberation in arenas open to citizens who use
public services, or who are otherwise regulated by public decisions. But in
deciding, those citizens must examine their own choices in the light of the
relevant deliberations and experiences of others facing similar problems in
comparable jurisdictions or subdivisions of government. Ideally, then, directlydeliberative polyarchy combines the advantages of local learning and selfgovernment with the advantages (and discipline) of wider social learning
and heightened political accountability that result when the outcomes of
many concurrent experiments are pooled to permit public scrutiny of the
effectiveness of strategies and leaders.
This conception is suggested by a range of political experiments, and
reflection on how their separate energies might be combined.8 Consider, for
example, community policing: A strategy for enhancing public security that
features a return of police officers to particular beats, regular discussions
between them and organized bodies in the communities they are policing,
and regular coordination between those bodies and agencies providing
other services that bear on controlling crime. Or consider forms of school
decentralization thatwhile shrinking school size and permitting parents to
choose schoolsalso replace close controls by central bureaucracies with
governance mechanisms in which teachers and parents play a central role.
Or arrangements for local and regional economic development, that include
strong components of training and service provision, and whose governance includes local community interests, service providers, representatives
of more encompassing organizations, as well as local representatives of
regional or national government.
These new arrangements are not conventionally public because, in solving
problems, they operate autonomously from the dictates of legislatures or
8
For discussions of such experiments, see Meares and Kahan 1999; Luria and Rogers 1999;
Meier 2000; Sabel, Fung, and Karkkainen 2000.
412
Joshua Cohen
public agencies; they are not conventionally private in that they do exercise
problem-solving powers, and their governance works through discussion
among citizens rather than the assignment of ownership rights. Moreover,
they are attractive because they appear to foster two fundamental democratic valuesdeliberation and direct citizen participationwhile potentially
offering advantages as problem-solvers that programs conceived within the
limits of conventional representative democracies do not.
Stated without much detail or nuance, the fundamental idea comprises
the following three elements:
1) Local problem solving through directly-deliberative participation, which
is well-suited to bringing the relevant local knowledge and values to bear in
making decisions. Direct participation helps because participants can be
assumed to have relevant information about the local contours of the problem, and can relatively easily detect both deception by others and unintended consequences of past decisions. Deliberative participation helps
because it encourages the expression of differences in outlook, and the
provision of information more generally: The respect expressed through the
mutual reason-giving that defines deliberation reinforces a commitment to
such conversational norms as sincerity and to solving problems, rather than
to strategic angling for advantage (perhaps by providing misleading information); furthermore, if preferences over outcomes themselves are shaped
and even formed by discussion, and mutual reason-giving reduces disagreements among such preferences, then being truthful will also be good
strategy.
2) With an eye to addressing the narrowness commonly associated with
localism, an institutionalization of links among local unitsin particular, the
institutionalization of links that require separate deliberative units to consider
their own proposals against benchmarks provided by other units. Because
practical reasoning requires a search for best solutions, decision-makers
need to explore alternatives to current practice. A natural place to look for
promising alternativesincluding alternatives previously unimagined in
the local settingis in the experience of units facing analogous problems.
Thus alongside directly-deliberative decision-making we need deliberative
coordination: deliberation among units of decision-making directed both to
learning jointly from their several experiences, and improving the institutional possibilities for such learning. Extending deliberation across units
allows each group to see its viewpoints and proposals in light of alternatives
articulated by the others: in effect, it ensures that the exercise of practical
reason is both disciplined and imaginative.
3) Responsibility for ensuring that deliberation within and among units
meets these conditions, vested ultimately in authorizing and monitoring
agencieslegislatures, agencies, and courts. In contrast to the conventional division of deliberative labor, this responsibility, under conditions of
directly-deliberative polyarchy, is to be discharged by ensuring that the
413
relevant decision-making bodies act deliberatively, notso far as possible

by substituting for their decisions.
As this observation indicates (and as the term polyarchy is meant to
signal), directly-deliberative polyarchy assumes the continued presence of
the legislatures, courts, executives, and administrative agencies, controlled
by officials chosen through free and fair elections, in which virtually all
adults have rights to suffrage, office-holding, association, expression, and
face alternative, legally protected sources of information. Though the
operation of these institutions and arrangements changes, they remain and
continue to serve some of the political values with which they are conventionally associated: peaceful transitions of power, restraints on unbridled
power, fair chances for effective influence over authoritative collective
decisions, opportunities to develop informed preferences.
The shift in the locus of problem-solving, however, changes the operations
and expectations of basic political institutions. Consider the role of legislatures. Directly-deliberative polyarchy is animated by a recognition of the
limits on the capacity of legislatures to solve problemseither on their own
or by delegating tasks to administrative agenciesdespite the importance of
solutions. The role of the legislature in directly-deliberative polyarchy is to
empower and facilitate problem-solving through directly-deliberative arenas
operating in closer proximity than the legislature to the problem. More
particularly, the idea is for legislatures to declare areas of policy (education,
community safety, environmental health) as open to directly-deliberative
polyarchic action; state general goals for policy in the area; assist potential
deliberative arenas in organizing to achieve those goals; make resources
available to deliberative problem-solving bodies that meet basic requirements on membership and benchmarking; and review at regular intervals
the assignments of resources and responsibility.
This changed role for legislatures does not exclude national solutions
through legislative enactment when uniform solutions are preferable (because
of limited diversity among sites) or when externalities overwhelm local
problem-solving. Instead, the availability of alternative methods of problemsolving imposes on legislatures a greater burden in justifying their own
direct efforts: They must explicitly make the case that the benefits of those
efforts suffice to overcome the advantages of direct-deliberative solutions.9
Administrative agencies, in turn, provide the infrastructure for information exchange between and among unitsthe exchange required for benchmarking and continuous improvement. Instead of seeking to solve problems,
the agencies see their task as reducing the costs of information faced by
different problem-solvers: helping them to determine which deliberative
bodies are similarly situated, what projects those bodies are pursuing, and
9
For related discussions of federalism, see Gardbaum 1996, and the account of the
commandeering problem in Dorf and Sabel 1998.
414
Joshua Cohen
what modifications of those projects might be needed under local

conditions.
3.
This is the barest sketch of the idea of directly-deliberative polyarchy, but I
hope it is clear even from the sketch that it offers a different redemptive
project than we find in Habermas response to the problem of possibility.
Here, I want to emphasize two points of difference, both focused on the
conception of the public sphere. First, in directly-deliberative democracy
(and, by extension, directly-deliberative polyarchy) the public arena is organizationally dispersed in that public opinion crystallizes not only in reference to
the national legislature, but also in the work of the local school governance
committee, the community policing beat organization, and their analogs in
areas such as the provision of services to firms or to distressed families.
Nevertheless, the pieces of this dispersed public sphere are connected by the
requirements of reason-giving, in particular the demand to respect basic
constitutional values; the need for explicit comparison with other units which
are themselves conducting similar comparisons; and a wider public debate
informed by such comparisons and focused on national projects. In short,
we do have a public sphere in directly-deliberative polyarchy: both because
citizens participate in solving problems, and because of the deliberative,
reason-giving terms of that participation.
Second, and more fundamentally, the public arena is the place where practicality in the form of problem solving meets political principle in the form
of deliberation through reason giving among free and equal citizens. In
directly-deliberative polyarchy, with direct problem-solving by groups of
affected citizens, public deliberation cuts across the distinction between
reflection on political purposes and efforts to address problems in light of
those purposes. This marriage of principle and problem-solving might have
the effect both of sharpening discussion in the informal public sphere; more
immediately, it promises an effectiveness to public engagement that is absent
from Habermas account.
For Habermas, discussion within the communicatively fluid public sphere
comprises all manner of topic and question, and is guided by experiential
concerns to which citizens themselves are attentive. So the dispersed network of communication that constitutes the public comes as close as can
reasonably be hoped to a free community of equals, autonomously debating
the terms of their collective lifeas close as can be hoped, if we take as an
assumption that the principal political, problem-solving institutions remain
fixed in design and conception, and that citizens are to discuss encompassing political directions, and not solve problems. Inevitably, then, the
capacity of the publics contributions to subsequently steer the state remain
an open question. The freer the communication within the public, the greater
415
clarification it can attain. But even the most radical extension and deepening
of the public sphere will be of limited consequence precisely because the
technical demands, to which administration, parliament, and party must
respond, limit the direction that might issue from a more encompassing,
unrestricted discussion among citizens: Communicative power cannot supply a substitute for the systematic inner logic of public bureaucracies. Rather,
it achieves an impact on this logic in a siege-like manner (Habermas 1992;
1996b, 486). In the end, radical democracy on this conception serves more as
a series of remindersthat human communication need not be narrowly
technical, that unsolved problems remain outside the purview of conventional institutionsand a source of new ideas in periods of crisis, than a
program to redirect the ensemble of institutions to ensure a controlling role
for the communicative power of free and equal citizens. I see no compelling
reason for that self-limitation: We should not confuse a proof of possibility
with a redemption of promise.
MIT
Department of Political Science
E53-473
Cambridge, MA 02139
USA
References
Ackerman, Bruce. 1991. We, The People: Foundations. Cambridge, MA: Harvard
University Press.
Barry, Brian. 1995. Justice as Impartiality. Oxford: Oxford University Press.
Cohen, Joshua. 1998. Democracy and Liberty. In Deliberative Democracy. Ed. J. Elster.
Cambridge: Cambridge University Press.
Cohen, Joshua, and Charles Sabel. 1997. Directly-Deliberative Polyarchy. European
Law Journal 3, 4: 31342.
Dorf, Michael, and Charles Sabel. 1998. Constitution of Democratic Experimentalism.
Columbia Law Review 98, 2: 267473.
Dworkin, Ronald. 1996. Freedoms Law. The Moral Reading of the American Constitution.
Cambridge, MA: Harvard University Press.
Gardbaum, Stephen. 1996. Rethinking Constitutional Federalism. Texas Law Review
74: 795838.
Habermas, Jrgen. 1990. Discourse Ethics: Notes on a Program of Philosophical
Justification. In J. Habermas, Moral Consciousness and Communicative Action. Trans.
C. Lenhardt, and S. Weber Nicholsen. Cambridge, MA: MIT Press.
. 1992. Further Reflections on the Public Sphere. In Habermas and the Public
Sphere. Ed. C. Calhoun. Cambridge, MA: MIT Press.
. 1996a. Between Facts and Norms. Contributions to a Discourse Theory of Law and
Democracy. Trans. W. Rehg. Cambridge, MA: MIT Press.
. 1996b. Appendix 1. Popular Sovereignty as Procedure. In J. Habermas,
Between Facts and Norms. Contributions to a Discourse Theory of Law and Democracy,
46390. Trans. W. Rehg. Cambridge, MA: MIT Press.
416
Joshua Cohen
. 1998a. On the Internal Relation Between the Rule of Law and Democracy. In
J. Habermas, The Inclusion of the Other. Trans. C. Cronin, and P. De Greiff.
Cambridge, MA: MIT Press.
. 1998b. Three Normative Models of Democracy. In J. Habermas, The Inclusion
of the Other. Trans. C. Cronin, and P. De Greiff. Cambridge, MA: MIT Press.
Luria, Daniel, and Joel Rogers. 1999. Metro Futures. Ed. J. Cohen, and J. Rogers.
Boston, MA: Beacon Press.
Meares, Tracey, and Dan Kahan. 1999. Urgent Times. Ed. J. Cohen, and J. Rogers.
Boston, MA: Beacon Press.
Meier, Deborah. 2000. Schooling a Democracy. Ed. J. Cohen, and J. Rogers. Boston, MA:
Beacon Press (forthcoming).
Michelman, Frank. 1996. Democracy and Positive Liberty. Boston Review 21: 38.
Rawls, John. 1987. The Basic Liberties and Their Priority. The Tanner Lectures on Human
Values. Ed. S. McMurrin. Salt Lake City, UT: University of Utah Press.
. 1993. Political Liberalism. New York, NY: Columbia University Press.
. 1995. Reply to Habermas. Journal of Philosophy 92: 13280.
. 1999. The Idea of Public Reason Revisited. In J. Rawls, Collected Papers.
Cambridge, MA: Harvard University Press.
Sabel, Charles, Archon Fung, and Brad Karkkainen. 2000. Environmental Democracy.
Ed. J. Cohen, and J. Rogers. Boston, MA: Beacon Press (forthcoming).
Scanlon, Thomas M. 1998. What We Owe To Each Other. Cambridge, MA: Harvard
University Press.

COHEN - Reflections On Habermas On Democracy PDF

Uploaded by

Copyright:

Available Formats

COHEN - Reflections On Habermas On Democracy PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

COHEN - Reflections On Habermas On Democracy PDF

Uploaded by

Copyright:

Available Formats

Ratio Juris. Vol. 12 No.

4 December 1999 (385416)

3) What relevance might a radical-democratic outlook have for contemporary democracies?

human reason, communicative action, and argumentationbecause I think

fundamental (co-original): Indeed, each is required to explain one another.

debate among citizens proceeds in splendid isolation from the exercise of

II. Co-Originality and Private Autonomy

Blackwell Publishers Ltd 1999.

to changing circumstances (Habermas 1996a, 125). Such exercisedemocratic

elaboration, internalized in the system of rights (Habermas 1996a, 122). So

The discussion that follows draws on Cohen 1998.

assumptions appear to be normatively more substantive than the notions of

defining condition of participation or a test of the acceptability of arguments

to advance considerations that one takes to be true or compelling. For such

III. Discursive Democracy

In achieving such reasonable results, the two tracks of deliberative politics

Formal political processeselections, legislatures, agencies, and courts

Thus, Habermas interprets popular sovereignty procedurally, as the possible

Blackwell Publishers Ltd 1999.

deliberative democracy institutionalizes the concern for justifiability to

to others whose conduct is governed by those choices, and who themselves

matter, a person can believe that the exercise of power is well-justified

way to determine what would be agreed to by all in idealized discourse is to

political participation in a democratic polity; and views, founded on religious

the deliberative idea requires. No similar result can be expected if we assign

IV. Problem of Possibility6

This section draws substantially from Cohen and Sabel 1997.

Blackwell Publishers Ltd 1999.

the abstract ideal of a self-organizing community of free and equal

Blackwell Publishers Ltd 1999.

deliberation that encompasses fundamental political values; and (here we go

relevant decision-making bodies act deliberatively, notso far as possible

Blackwell Publishers Ltd 1999.

what modifications of those projects might be needed under local

Blackwell Publishers Ltd 1999.

You might also like