The Journal of Political Philosophy: Volume 10, Number 2, 2002, pp. 153±174
THEORIES of deliberative democracy consist of a set of principles that
are intended to establish fair terms of political cooperation in a
democratic society. Some theorists believe that the principles should refer only
to the process of making political decisions in government or civil society.1The principles of deliberative democracy, they argue, should not prescribe the
content of the laws, but only the procedures (such as equal suffrage) by which
laws are made and the conditions (such as free political speech) necessary forthe procedures to work fairly. These theorists, whom we call pure
proceduralists, insist that democratic theory should not incorporatesubstantive principles such as individual liberty or equal opportunity beyond
what is necessary for a fair democratic process. They do not deny thatsubstantive principles such as freedom of religion, nondiscrimination or basic
health care are important, but they wish to keep these principles out of theirdemocratic theories.
We argue that thiseffort to keep democratic theory procedurally pure fails,
and that any adequate theory must include substantive as well as procedural
principles. Our own theory, presented in Democracy and Disagreement, offers
one such approach: it includes substantive principles (such as basic liberty andfair opportunity) that extend fairness to persons (for the sake of reciprocity,
mutual respect, or fairness itself). Principles of basic liberty and fair opportunitycan be defended on many substantive grounds; in that book we argue from a
1AsJuÈrgen Habermas writes, ``All contents, no matter how fundamental the action norm involved
may be, must be made to depend on real discourses (or advocatory discourses conducted as substitutes
for them)''. ``Discourse ethics,'' Moral Consciousness and Communicative Action, trans. Christian
Lenhardt and Shierry Weber Nicholsen (Cambridge, Mass.: M.I.T. Press, 1993), p. 94. For comments
and other citations, see our discussion in Amy Gutmann and Dennis Thompson, Democracy and
Disagreement (Cambridge, Mass.: Harvard University Press, 1996), pp. 17±18. Other theorists who
would also be more inclined to limit deliberative democracy to process considerations and are
therefore critical of including substantive principles in its theory include: Jack Knight,
``Constitutionalism and deliberative democracy'' Deliberative Politics, ed. Stephen Macedo (New
York: Oxford University Press, 1999), pp. 159±69; Cass Sunstein, ``Agreement without theory,'' ibid.,
pp. 147±8; and Iris Marion Young, ``Justice, inclusion, and deliberative democracy,'' ibid., pp. 151±8.
For our reply, see Gutmann and Thompson, ``Democratic disagreement,'' ibid., pp. 261±8.
# Blackwell Publishers, 2002, 108 Cowley Road, Oxford OX4 1JF, UK and 350 Main Street,
widely recognized principle of reciprocity or mutual justi®cation among persons
who are bound by the lawsof a democracy.
But our argument here doesnot depend on accepting the whole theory in that
book, or even the speci®c grounds of reciprocity on which we base the principles.
We wish to maintain here that, on a wide range of available grounds, democraticprinciples must be substantive as well as procedural. A democratic theory that
shuns substantive principles for the sake of remaining purely procedural sacri®cesan essential value of democracy itself: its principles cannot claim to treat citizens in
the way that free and equal persons should be treatedÐwhether fairly, reciprocally,or with mutual respectÐin a democratic society in which laws bind all equally.
Pure proceduralists make two kinds of arguments against including
substantive principlesÐone from moral authority and the other from political
authority. The argument from moral authority holdsthat the moral judgment of
democratic citizens, not democratic theorists, should determine the content oflaws. A theory that contains substantive principles improperly pre-empts the
moral authority of citizens. The argument from political authority maintains thatsubstantive principles similarly pre-empt the political sovereignty of citizens,
which should be exercised not through hypothetical theoretical reasoning butthrough actual democratic decision-making. A theory that contains substantive
principles unduly constrains the democratic decision-making process, includingthe process of deliberation itself.
We dispute both of these arguments and defend the inclusion of substantive
principles in a theory of deliberative democracy. We agree with those theoristswho point out that mere procedures such as majority rule cannot justify
outcomes that are unjust according to substantive principles. But these theoristsusually neglect the substantive value in the procedures, and assume that an
outcome is justi®ed if it is just according to their substantive principles.
In any case, our main argument against pure proceduralism is not the same as
the standard objection that procedures can produce unjust outcomes, though weaccept this objection. We also argue for including substantive principles in a
democratic theory for another, generally neglected reason. Such principles shouldbe included so that the theory can explicitly recognize that both substantive and
procedural principles are subject to contestation in similar ways. A critical claim
in our defense of a deliberative democratic theory that is both procedural andsubstantive is that the principles are to be treated as morally and politically
provisional. This provisionality gives deliberation part of its point. Bothprocedural and substantive principles are systematically open to revision in an
ongoing process of moral and political deliberation. If the principles areunderstood in this way, the usual objections against including substantive
principles lose their force. The provisional status of all its principles thusconstitutes a distinctive strength of deliberative democratic theory, and at the
same time offers deliberative democrats an effective response to those who would
exclude substantive principles from democratic theory.
Although we concentrate here on showing the problems with the form of pure
proceduralism that justi®es political outcomes by procedural criteria only, our
general criticisms also apply against any attempt to segregate procedural andsubstantive principles in separate theories. Theorists who judge outcomes partly
by substantive principles of justice are still pure proceduralists (with respect totheir democratic theories) if they assume that the democratic procedures can be
justi®ed without reference to some of the same substantive values expressed bytheir principles of justice. Our argument is intended to show that this kind of
sharp separation between procedural and substantive principles and theories isnot sustainable.
To illustrate some of the major points in the argument for including both
procedural and substantive principles in a deliberative democratic theory, we use
a case involving deliberation about health care in the United Kingdom. In 1999,
the British Government created a new body, the National Institute for ClinicalExcellence (NICE), which is to provide assessments of treatments and clinical
guidelinesfor use by the National Health Service (NHS).2 The impetusfor thenew Institute came from the widespread recognition that the NHS could not fund
care for all health needs, and needed to ®nd a way to make its dif®cult decisionsin a more public and deliberative manner. By creating a deliberative decision-
making body, which includes both expert and lay members, the BritishGovernment may also have hoped that it could defuse some of the controversy
about the hard choices that had to be made. But not surprisingly, shortly after its
creation, NICE itself came under criticism in another deliberative forumÐtheHouse of Commons. Together these moments of deliberationÐthe proceedings
of NICE and the Commonsdebate about NICEÐare more appropriate for ourpurposes than cases from the US. They involve an attempt to institutionalize
nationwide deliberation about health care prioritiesin a way that the US hastriedonly in certain states. Also, the deliberation takes place in a nation in which
principles of justice in health care come closer to being satis®ed than in the US,and therefore poses a greater challenge to our claim that such principles are
necessary in any adequate theory of deliberative democracy. If a theory needssubstantive principles when applied to health care in the UK, then a fortiori it
should need them when applied to similar issues in the US.
To determine what kind of principlesbelong in a deliberative democratic theory,
we need ®rst to consider the meaning and implications of the fundamentalprinciple of reciprocity. Reciprocity iswidely recognized asa core principle of
2See statements by NICE's newly appointed director Michael Rawlins: Richard Horton, ``NICE: a
step forward in the quality of NHS care,'' The Lancet, 353 (March 27, 1999), 1028±9, and Gavin
Yamey, ``Chairman of NICE admitsthat itsjudgmentsare hard to defend,'' British Medical Journal,
democracy in its many moral variationsÐliberal, constitutional, procedural, and
deliberativeÐbut most theories do not give it the central role that deliberative
democracy does. Reciprocity holds that citizens owe one another justi®cationsfor the mutually binding lawsand public policiesthey collectively enact. The aim
of a theory that takes reciprocity seriously is to help people seek politicalagreement on the basis of principles that can be justi®ed to others who share the
Mutual justi®cation means not merely offering reasons to other people, or even
offering reasons that they happen to accept (for example, because they are ina weak bargaining position). It means providing reasons that constitute a
justi®cation for imposing binding laws on them. What reasons count as such ajusti®cation is inescapably a substantive question. Merely formal standards for
mutual justi®cationÐsuch as a requirement that the maxims implied by laws be
generalizeableÐare not suf®cient. If the maxim happens to be ``maximize self- orgroup-interest,'' generalizing it does not ensure that justi®cation is mutual.
Something similar could be said about all other conceivable candidates for formalstandards. Mutual justi®cation requires reference to substantive values.
We can see more clearly why mutual justi®cation cannot proceed without
relying on substantive values by imagining any set of reasons that would deny
persons basic opportunities, such as equal suffrage and essential health care. Evenif the reasons satis®ed formal standards, they could not constitute a mutual
justi®cation because those deprived of the basic opportunities could reasonably
reject them. Denying some persons suffrage is a procedural deprivation that isinconsistent with reciprocity: we cannot justify coercive laws to persons who had
no share in making them. Similarly, denying persons essential health care is asubstantive deprivation that cannot be justi®ed to the individuals who need it.
That such denials are unacceptable shows that the mutual justi®cation is neitherpurely formal nor purely procedural.
Because such denials of basic opportunities cannot be mutually justi®able, the
principles of a democratic theory must be both procedural and substantive. A
democratic theory whose principles would permit some persons to beunnecessarily deprived of a basic opportunity like health care does not take
seriously the value of mutual justi®cation implied by the principle of reciprocity.
Furthermore, it does not treat persons as free and equal beings. While we arguefrom the fundamental principle of reciprocity, thisprinciple convergesin its
implications with the ideal of free and equal personhood, which is the basis ofmany democratic theories, not only deliberative ones.
The principlesof our deliberative democratic theory s
cooperation that satisfy reciprocity. Such terms are similar to what John Rawls
calls ``fair terms of social cooperation.'' But the procedural and substantivecontent of fair termsof social cooperation will vary with different interpretations
of what reciprocity requires. A theory is ``deliberative'' if the fair terms of social
cooperation include the requirement that citizensor their representativesactually
seek to give one another mutually acceptable reasons to justify the laws they
adopt. The reasons, as we have seen, refer to substantive values no less than to
Although reciprocity isa foundational value in deliberative democracy, it
does not play the same role that ®rst principles, such as utility or liberty, play intheories such as utilitarianism or libertarianism. These theories derive all of their
other principles from their ®rst principles. Reciprocity is not a ®rst principlefrom which the rest of justice is derived, but rather a regulatory principle that
serves two different roles. First, it guides thinking in the ongoing process inwhich citizens as well as theorists consider what justice requires in the case of
particular laws in speci®c contexts. Second, it shows the need for otherprinciplesto ®ll out the content of a deliberative democratic theory. Reciprocity
points to the need to develop such principles as publicity, accountability, basic
liberty, basic opportunity, and fair opportunity, which are necessary for themutual justi®cation of laws. As the ®rst role of reciprocity suggests, such
principles should be developed in an actual ongoing process of mutualjusti®cation.
An important implication of reciprocity isthat democratic deliberationÐ
the processof mutual reason-givingÐisnot equivalent to the hypothetical
justi®cations proposed by some social contract theories. Such justi®cations mayconstitute part of the moral reasoning to which some citizens appeal, but the
reasoning must survive the test of actual deliberation if it is to ground laws that
actually bind all citizens. This deliberation should take place not only in theprivate homes of citizens or the studies of philosophers but in public political
forums. In this respect, deliberative theory proposes a political ideal that isprocess-dependent, even if its content is not exclusively process-oriented.
The requirement that actual deliberation take place isnot simply a matter of
trying to ensure that citizens feel that their views were taken into account even
when they disagree with the outcome. Actual political deliberation at some timeisrequired to justify the law for thissociety at thistime. The reason-giving
process is necessary for declaring a law to be not only legitimate but also just.
The process is necessary to give assurance that (substantive or procedural)
principlesthat may be right in general are right in the particular case or rightly
applied to this particular case. No amount of hypothetical reasoning is likely tobring out all the complexitiesthat are relevant to determining whether a law is
justi®ed at a particular time in any given society. It would be dif®cult to decide onthe basis of any general principle of basic opportunity whether, for example,
NICE wasjusti®ed in denying coverage for a new anti-¯u drug (zanamivir)marketed asRelenza by the pharmaceutical company Glaxo±Wellcome.3 What
would be missing is not simply factual information but the weighing of facts and
3See ``NICE appraisal of Zanamivir (Relenza)'' posted at www.nice.org.uk. For some of the
reaction, see Stephen D. Moore, ``U.K. rebuffs Glaxo on new ¯u drug,'' Wall Street Journal (October
the balancing of valuesin the context of other health care and related decisions
that of®cialsaswell ascitizensneed to make.
It may be helpful to think of the requirement of actual deliberation as
analogous to a feature of scienti®c inquiry. Reciprocity is to justice in political
ethics what replication is to truth in scienti®c ethics. A ®nding of truth in sciencerequiresreplicability, which callsfor public demonstration. A ®nding of justice in
political ethicsrequiresreciprocity, which callsfor public deliberation.
Deliberation is not suf®cient to establish justice, but deliberation at some point
in history is necessary. Just as repeated replication is unnecessary once the truthof a ®nding (such as the law of gravity) has been amply con®rmed, so repeated
deliberation is unnecessary once a precept of justice (such as equal protectionunder the laws) has been extensively deliberated. Deliberation may still be
desirable, of course, even when justice does not demand it.
The practice of actual deliberationÐgiving justifying reasons for mutually
binding lawsto one'sfellow citizensÐitself both exempli®esand promotesthe
value of reciprocity. Citizenswho have effective opportunitiesto deliberate treatone another not merely asobjectswho are to be judged by theoretical principles
but also as subjects who can accept or reject the reasons given for the laws thatmutually bind them. The reasons are not to be regarded as binding unless they are
presented to citizens who have the chance to consider and reject them eitherdirectly or indirectly through their accountable representatives in a public forum.
In this respect, the creation of NICE supported the value of reciprocity by
providing citizenswith an example of deliberation in action in which they couldassess the justi®cations their representatives give for policies that will affect their
wellbeing in important ways. The possibility of continuing debate in Parliamentabout the deliberative practicesaswell asthe decisionsof NICE further helpsto
The processof deliberation also hasepistemic value. Decisionsare more likely
to be morally justi®able if decision-makers are required to offer justi®cations forpoliciesto other people, including those who are both well informed and
representative of the citizens who will be most affected by the decisions. Theepistemic value of deliberation is especially great when the justi®cation for a
decision must combine factual and evaluative matters, as is the case with most
health care decisions, including the kind that NICE makes. While experts may bethe best judges of scienti®c evidence, they have no special claim to ®nding the
right answer about priorities when degrees of risk and trade-offs of costs andbene®tsare involved.
II. WHY RECIPROCITY REQUIRES SUBSTANTIVE PRINCIPLES
The practice of deliberation isan ongoing activity of reciprocal reason-giving,punctuated by collectively binding decisions. It is a process of reaching mutually
binding decisions on the basis of mutually justi®able reasons. Because the reasons
have to be mutually justi®able, the process presupposes some principles with
substantive content. It is possible, and sometimes desirable, to distinguish
procedural and substantive aspects of principles and theories, but to turn thesedistinctions into separate principles or distinct theories is to distort both the
theory and practice of (deliberative) democracy. Although for convenience werefer to principles and theories as procedural and substantive, strictly speaking
democratic principles and theories have both procedural and substantivedimensions, and approaches that force a sharp division are misleading.
The principle of reciprocity itself expresses neither purely procedural nor
purely substantive values. A reciprocal perspective is both procedural and
substantive because mutual justi®cation cannot proceed without appealing toreasons that refer to both procedures of government and substance of laws, often
at the same time. Even philosophers like Stuart Hampshire who seek to exclude
substantive justice completely from their procedural political theoriesacknowledge the need for some substantive valuesÐsuch as ``common
decency''Ðin the very concept of justice.4 Hampshire says justice is ``primarilyprocedural''Ðnot entirely so.5 Like other philosophers who want to be pure
proceduralists, he never says what constitutes the correct set of proceduralprinciples, and why people who remain subject to tyrannical rule should settle for
only procedural principlesif they permit tyranny.
At a minimum, no one would seriously dispute that justi®cations should
recognize some values expressed by substantive principles such as liberty and
opportunity. It would hardly be suf®cient for NICE to justify a decision to denyprescription drugs to West Indian immigrants on the grounds that they are not
white. EvenÐor especiallyÐif a large majority of British citizens would acceptsuch reasoning, the justi®cation would not satisfy any adequate standard of
reciprocity. Nor would it be any more acceptable to deny prescription drugs to adisadvantaged minority on the grounds that they agreed with the conclusion.
They might have agreed simply because they had less power than the groups thatprevailed and had no better alternative in a bargaining situation.
To see more clearly why reciprocity requires substantive principles, we might
further imagine a situation in which the process of decision-making itself was fair
in the sense that the bargaining power of the parties was equitable, but in which
the reasoning of the decision-makers was prejudiced (or could only be reasonablyinterpreted as based on prejudice) against West Indian immigrants or another
disadvantaged minority group. The prejudiced reasoning then yields anoutcomeÐsupported by the vast majorityÐthat denies critical health care to
the disadvantaged minority. This outcome could not be justi®ed on grounds ofreciprocity, even if the proceduresby which it wasreached were otherwise
completely fair. The justi®cation for the outcome does not treat members of the
4Stuart Hampshire, Innocence and Experience (Cambridge, Mass.: Harvard University Press,
minority group as worthy of a justi®cation that they could reasonably accept.
Alternatively, one might say that the prejudiced reasoning denies members of the
minority group the status of free and equal persons. Given the nature of thereasoning, this would be so no matter how fair the process of decision-making
We can see the principle of reciprocity in action, and the mixture of procedural
and substantive values it implies, in the debate about NICE in the House ofCommons. The debate had hardly begun when an MP (who is also a physician)
challenged the idea that NICE or anyone else has the moral or political authorityto ration health care. Another MP responded, saying that rationing was necessary
and therefore justi®able: ``sometimes some treatments are not available whenthey would bene®t patients or populations, because there simply are not the
resources to provide all those treatments on the NHS.'' Although the debate at
®rst seemed to turn on issues about the legitimacy of the process (who has theauthority to decide), most critics (as well as most defenders of the Government)
agreed that NICE represented an improvement asfar asprocesswasconcerned.
Most recognized that the new decision-making process is preferable to the old,
and much superior to the less deliberative process that prevails in the US.
The challenge instead was directed against the substance of NICE's decision in
its ®rst review of a drug. NICE had recommended against the NHS's funding thenew anti-¯u drug Relenza.6 The criticsworried that thisdecision would be a
precedent that would justify NICE's recommending against funding of other
more expensive and effective new drugs, such as beta interferon (which treats thesymptoms of multiple sclerosis). The critics argued that decisions denying
coverage are likely to deprive less advantaged patients of life-enhancing and life-saving treatments that more advantaged patients receive, and that this unequal
opportunity cannot be justi®ed. It leaves the less fortunate without the healthcare and the life chancesthat, if any citizensenjoy, then all should be entitled to.7
They appealed to substantive principles, not simply to a claim that the processwasunfair, or even that it wasnot deliberative.
Defenders of NICE's decision rightly realized that they needed to justify the
substance of the decision because the deliberative process in which NICE had
engaged (and in which they were engaging in the Commonsdebate) could not in
itself be a suf®cient justi®cation of the decision. They explicitly invokedsubstantive standards to defend NICE's decision. They argued, for example,
6See ``NICE appraisal of Zanamivir (Relenza).'' For some of the reaction, see Moore, ``U.K.
rebuffs Glaxo on new ¯u drug.'' The Food and Drug Administration approved Relenza for use in
the US despite a 13±4 vote of an outside panel of experts recommending against approval. Some
critics believe that the drug has been overprescribed during the current ¯u season. See Sheryl Gay
Stolberg, ``F.D.A. warns of overuse of 2 new drugs against ¯u,'' New York Times (Jan. 13, 2000),
7Asone MP put it in the Commonsdebate: ``When we talk about rationing of NHS treatments, we
aren't saying no one in the UK has them. What we are saying is that they aren't available to poor
people. The rich and those who can afford it can get these treatments privately'' (House of Commons
that the decision not to fund Relenza would not adversely affect the basic life
chancesof any citizen, not even patientswho are at high risk of complications
from in¯uenza. They called for more research on the effects of Relenza on high-risk patients, and suggested that if there were evidence of Relenza's bene®t in
reducing the serious secondary complications of in¯uenza in such patients, thenthey would support NHS funding. Their arguments, whether correct on the
merits, were entirely in order, and if correct they were also necessary to justifytheir conclusion. That they were necessary cannot readily be accommodated in a
democratic theory that limits itself to procedural considerations only.
An obvious but no less important virtue of a theory that does not limit itself to
procedural principles is that it has no problem with asserting that what themajority decides, even after full deliberation, is wrong. Within a deliberative
theory, one should be able to condemn majority tyranny on substantive grounds:
one should be able to say that a majority acts wrongly if it violates basic libertyby denying health care on grounds of race, gender, or poverty. Or suppose that
the majority, following perfectly deliberative procedures, decides to institute apractice of compulsory organ donation. On a purely procedural conception of
deliberative democracy, thislaw would be justi®ed. If a deliberative theoryincludes substantive principles such as basic liberty which protect bodily
integrity, democratswould be able to object to s
abandoning their commitment to deliberative democracy.
Democrats of course may be mistaken when they assert claims based on
substantive principles either because they draw incorrect implications from acorrect principle or because they rely on an indefensible principle. Perhaps
compulsory organ donation does not violate basic liberty, or perhaps thisparticular principle of basic liberty is ¯awed. Our argument for including
substantive principlesÐbased on reciprocityÐnot only allows for both kinds ofmistakes; it also incorporates into the theory itself the insight that democratic
theorists and citizens may be mistaken about both procedural and substantiveprinciples. Deliberation explicitly deals with the likelihood of mistaken views
about principlesand their implicationsby considering the principlesof a theoryto be provisional, and therefore subject to ongoing deliberation. To point out the
possibility of being mistaken about substantive principles is therefore not an
argument against including such principles within a deliberative democratictheory.
The conclusions of purely procedural theories sometimes converge with the
claims of the substantive standards that reciprocity requires. For example, a
procedural theory of democracy may say that racial discrimination in voting isnot justi®ed because it excludes a class of human beings from citizenship, and this
violatesthe procedural requirementsof democracy, which demand theenfranchisement of all adult persons. This procedural reason is ®ne as far as it
goes. But it does not go far enough in establishing why such discrimination is not
justi®ed. Democratic theorists should be able to object that racial discrimination
(for example, in the provision of health care by a for-pro®t Health Maintenance
Organization) is not justi®ed even if democratic citizenship or no other process
values are at stake. Majority tyranny is objectionable on substantive, not onlyprocedural, grounds.
Moreover, thiskind of objection should be capable of being made from within
a deliberative democratic theory. After all, democracy hasnever meant merely
majority rule. Denying basic liberties and opportunities by raciallydiscriminatory policies is either the result of state action or can be remedied by
state action, and any such action or inaction requires a justi®cation that couldreasonably be accepted by those whose liberties and opportunities are denied.
requirement to give such a justi®cationÐto invoke substantive principles in the
public forum to justify a mutually binding law or policyÐis therefore not an
incidental feature of deliberative democracy. The substantive principles areintegral to the deliberative process itself.
To say that the principles are integral to the process is not to deny that they
may be justi®able outside of that process. Like any theorist of justice (or citizen
making a claim about justice), deliberative democrats may put forward principlesfor consideration which they regard as justi®ableÐand which indeed may be
correct, but simply not yet justi®ed as laws. Deliberative theorists try to justifytheir substantive principles in a number of familiar ways, some just like those
used by any theorist. We justify the substantive principles such as basic liberty in
Democracy and Disagreement, ®rst and foremost, on their own termsÐbyidentifying core values, convictions, and paradigmatic cases where no one would
reasonably deny that they were violated (for example, discrimination on groundsof race). Then by analogy and other formsof reasoning, we try to thicken and
extend the principles to apply to more controversial cases. This is also how muchof actual political deliberation proceeds.
Certainly, these substantive principles might be rejected, and perhaps even
reasonably rejected, in a deliberative process that satis®es the procedural
conditionsof deliberative democracy. But a precisely parallel argument can bemade about procedural principles. Procedural principles may also be rejected by
a deliberative democracy (and so may a purely procedural conception of
deliberative democracy). Pure proceduralists do not have access to some moralbasis, which our conception lacks, on which to claim that the procedural
constraints that they recommend for a constitutional deliberative democracy arecorrect or authoritative.
Some critics who object to including substantive principles in a deliberative
democratic theory are themselves not pure proceduralists with respect to justice.
They agree that justice requires the protection of basic liberties andopportunities, including perhaps even access to adequate health care. But they
still insist that the subject matter of democratic theories should be kept distinct
from questions of distributive justice. They are pure proceduralists with respect
to democracy, but not justice. Democracy, they imply, is supposed to tell us how
to decide when we do not agree on what is just; we should not confuse matters by
combining principles of justice with the procedures for deciding disputes aboutthose principles.
This argument is not so much substantive as it is de®nitional: democracy
(including deliberative democracy) means fair procedures, not right outcomes.
The critics cannot rely on ordinary usage or the history of modern democratictheory, because representative democracy has rarely been characterized as
exclusively procedural. Ordinary usage of a concept as complex as democracy isenormously varied, as are the conceptions of democracy found in modern
democratic theory. And democratic practice itself is full of debates aboutsubstantive principles. Why then strain so hard to exclude them from the
The reason cannot be that democratic theory is somehow internally
inconsistent if it contains substantive as well as procedural principles. To be
sure, the more principles a theory contains, the more likely there are to becon¯icts among them. And including both substantive and procedural principles
certainly increases the potential for con¯ict. But democratic politics itself is rifewith con¯ict among principles, and a democratic theory that tries to insulate
itself from that con¯ict by limiting the range of principles it includes is likely to beless relevant for recognizing and resolving the disagreements that democracies
typically confront. When the disagreements mix substantive and procedural
values as so many do in actual democratic practice, theorists who arti®ciallysegregate substance and procedure in separate theories of justice and democracy
are prone to distort the role of both.
Some pure proceduralists may wish to keep out substantive principles because
they are contestable, and democracy is supposed to be a means of resolvingdisagreement among contestable principles such as basic liberty. But the content
accountability, are also contestable. A purely procedural theory does not avoid
fundamental disagreement: con¯icts among procedural principles are no lesssevere than among substantive principles. For example, in the debate in
Commonsabout NICE'sdecision to deny coverage for beta interferon, the MP
from North Wilshire implicitly raised a basic procedural questionÐto whatextent doesdemocratic control require local autonomyÐwhen he argued that his
constituents should have access to the drug. He objected thatÐbecause of therelative autonomy of regionsÐsome citizens in other parts of the country could
get beta interferon from the NHS while hisconstituentscould not. Thisis``aterrible tragedy for constituents such as mine, who could be prescribed beta
interferon if they lived in Bath or Oxford, but not in Wiltshire.''8
8House of Commons Debate, November 10, 1999. Also see Jo Lenaghan, ``The rationing debate:
Central government should have a greater role in rationing decisions,'' British Medical Journal, 314
The political debatesover health care rationing that are occurring not only in
the UK but also in almost every contemporary democracy clearly reveal the need
to consider both procedures and outcomes in judging democratic justice. At stakeare both the conditions under which these decisions are made and their content.
Do the decision-making bodies bring together representatives of all the peoplewho are most affected by the decisions? Are the representatives accountable to all
their constituents? These procedural questions cannot be answered in the contextof these debates without also asking: to what extent is the substance of the
decisions justi®able to all the people who are bound by them? To excludesubstantive criteriaÐsuch as liberty and opportunityÐthat judge the justice of
decisions would be morally arbitrary and incomplete according to deliberativedemocracy's own premise of reciprocity. (To exclude substantive criteria would
also be morally arbitrary and incomplete according to other premises that are
often identi®ed asfundamental to deliberative democracy, such asfree and equalpersonhood or mutual respect.)
To af®rm that a democratic theory should include substantive principles does
not of course commit one to any particular set of principles. In Democracy and
Disagreement, we propose a set of principles that are both substantive andprocedural, and present arguments for their inclusion as part of the constitution
of a deliberative democracy.9 The argumentswe present are intended to be partof a deliberative process itself, and in fact include fragments from actual
deliberations. For example, we argue that laws or policies that deprive
individuals of the basic opportunities necessary for making choices amonggood lives cannot be mutually justi®ed as a principle of reciprocity requires. The
basic opportunities typically include adequate health care, education, security,work, and income, and are necessary for living a decent life and having the
ability to make choicesamong good lives. We therefore would include aprinciple of basic opportunity as part of any adequate theory of deliberative
Criticswho object that thisprinciple isnot mutually justi®able or that other
principlesof equality are more mutually justi®able are effectively accepting theidea that democratic theory should include substantive principles. Even while
challenging the content of the principles, they are nevertheless accepting that
the termsof the argument should be reciprocal. Such challengesare welcomeby the terms of the theory itself, which asks for reasons that can be publicly
assessed by all those who will be bound by them.10 Thiskind of challenge canthen become part of the continuing deliberative process. The reason that such
9Gutmann and Thompson, Democracy and Disagreement, pp. 199±229.
10Not so welcome are other criticsÐthose who reject the aim of giving substantive content to
the claimsof reciprocity or who reject the very standard of reciprocity. But neither are their
claims cogent. Having rejected the idea of mutual justi®cation, they are hard-pressed to explain
how they can justify at all imposing coercive laws and policies on citizens who morally disagree
with them. See section III below, and Gutmann and Thompson, Democracy and Disagreement,
a challenge ®tswithin the termsof a deliberative theory itself isthat the
principles of the theory per se have a morally and politically provisional
III. WHY THE PRINCIPLES SHOULD BE MORALLY PROVISIONAL
How is it possible for a theory to propose substantive principles to assess laws
while regarding citizensasthe ®nal moral judgesof the lawsthey make? The keyto deliberative democracy's answer lies in the provisional status of its
principles.11 The principlesof deliberative democracy have a different statusin
deliberative democracy than they do in most moral and political theories. Theyare morally and politically provisional in ways that leave them more open to
challenge and therefore more amenable to democratic discretion. The moral basisof the provisional status of deliberative principles comes from the value of
reciprocity. Giving reasonsthat otherscould reasonably accept impliesacceptingreasons that others give in this same spirit. At least for a certain range of views
they oppose, citizens should acknowledge the possibility that the rejected viewmay be shown to be correct in the future.12 Thisacknowledgement has
implications not only for the way citizens should treat their opponents but also
for the way they regard their own views.
The process of mutual reason-giving further implies that each participant
involved take seriously new evidence and arguments, new interpretations of oldevidence and arguments, including moral reasons offered by those who oppose
their decisions, and reasons they may have rejected in the past. ``Takingseriously'' means not only cultivating personal dispositions (such as open-
mindedness and mutual respect) but also promoting institutional changes (suchas open forums and sunset provisions) that encourage reconsideration of laws
and their justi®cations. One implication is that citizens and their accountablerepresentatives should continue to test their own political views, seeking forums
in which their views can be challenged, and keeping open the possibility of their
Deliberative democracy thusexpressesa dynamic conception of political
justi®cation, in which provisionalityÐopenness to change over timeÐis anessential feature of any justi®able principles. Provisionality takes two general
forms. The principles are morally provisional in the sense that they are subject to
11The discussion here of moral and political provisionality draws on our analysis in ``Why
deliberative democracy isdifferent,'' Democracy, ed. Ellen Frankel Paul et al. (Cambridge: Cambridge
University Press, 2000), pp. 161±80.
12The range is determined by what we call ``deliberative disagreements,'' which are those in
which citizens continue to differ about basic moral principles even though they seek a resolution
that is mutually justi®able. The dispute over abortion is an example of a deliberative disagreement
because both sides can justify their views within a reciprocal perspective. A dispute about racial
segregation is an example of a nondeliberative disagreement because one side can be reasonably
rejected within a reciprocal perspective. See Gutmann and Thompson, Democracy and
change through further moral argument; and they are politically provisional in
the sense that they are subject to change through further political argument.
Morally provisional principles are presented as claims that can be challenged
and changed over time in response to new philosophical insights, empirical
evidence, or interpretations of both the insights and evidence. They are justi®edonly when they are so presented. Many theories endorse something like this
general outlookÐfor example, by adopting some form of fallabilism, or moresimply by expressing general approval of moral and intellectual open-
mindedness. But the provisional stance that deliberative democracy takestoward itsown claimsisdistinctive in being integral to the theory. Deliberative
democracy supports the means for fundamental change in the content of thetheory itself. Deliberative democracy subjects its own principles, as well as other
moral principles, to critical scrutiny over time. If as a consequence of such
scrutiny, its fundamental principles substantially changeÐsay, from a moreegalitarian to a more libertarian orientation (or vice versa)Ðthe theory is
appropriately seen as undergoing revision rather than rejection.
Not all principlescan be challenged at the s
deliberative democratic theory, but any single principle (or even severalprinciples) may be challenged at a particular time by other principles in the
theory. Citizens and accountable of®cials can revise one principle in a sequentialprocess in which the other principles are held constant. They can alter their
understanding of all the principles by applying them in a different context or at a
different time. For example, when the NICE Board decided against fundingRelenza, it implicitly made the provisional status of its decision clear by limiting
its decision to one in¯uenza season only. NICE also recommended thatadditional trials be conducted and further data be obtained so that its decision
could be reassessed in the next in¯uenza season. Particular attention should bepaid, the Board said, to ®nding out whether Relenza has positive effects on
reducing serious secondary complications of in¯uenza in high-risk patients. Themoral basis for a decision against funding RelenzaÐthe claim that the use of
Relenza does not signi®cantly affect anyone's basic opportunity in lifeÐwouldno longer be defensible if evidence came to light showing that Relenza could
signi®cantly reduce serious complications of in¯uenza in high-risk patients. In
this way, both the process and the substance of the decision kept open thepossibility of revising the recommendation in the future.
The possibility of revision applies not only to substantive principles but also to
a principled defense of the practice of deliberation itself. This is why it is
misleading to claim that substantive principles should have no place in the theorybecause they are merely philosophical proposals. Substantive principles are no
more or less provisionalÐand no more or less philosophical proposalsÐthan thecase for deliberation itself. It is as possible to question, from within deliberative
theory, whether deliberation is justi®ableÐand what it entailsÐas it is to
question whether basic liberty is justi®ableÐand what it entails.
Consider the deliberation in Commons about whether NICE itselfÐalso a
deliberative forumÐis the justi®able way to make health care decisions. This part
of the debate began when several MPs objected that letting NICE makerecommendations is ``to shield the Government from the very dif®cult decisions
that have to be taken.'' Should NICE recommend beta interferon, which costsabout £10,000 per patient per year and hasbeen judged ``marginally effective''?
(It treats incurable multiple sclerosis ``by reducing the exacerbation rate inpatients who have relapsing±remitting disease without important disability.''13)
Should NICE recommend the new taxane drugsfor chemotherapy, which do notcure but, asone MP put it, ``can add yearsto life at a cost of about £10,000 per
year''? If NICE recommends against prescribing expensive new drugs that canprovide some health care bene®ts to patients, will it thereby be shielding the
Government from pressure to increase the total NHS budget? If NICE
recommends in favor of the NHS prescribing these drugs, will it thereby beforcing the NHS not to fund some other existing and highly valuable treatments
(or pressuring the Government to increase funding for the NHS)? The answer tothese substantive questions thus depends on taking a position on what the process
Not even the deliberative principle that callsfor giving moral reasonsin
politics is beyond reasonable disagreement. Some critics of deliberation arguethat bargaining isnot only more common but also preferable asa way of
resolving moral disagreements in politics. The claim that self-interested (or
group-interested) bargaining processes are better than deliberative ones relies onthe premise that interest-based politics is more morally desirable and mutually
justi®able than a deliberative politics. Whether political bargaining satis®esreciprocity (or any other moral standards) depends in part on the actual
consequences of political bargaining in a particular social context. If thoseconsequences can be shown to be mutually justi®able to the people who are
bound by them, or more mutually justi®able than the consequences ofdeliberative processes, then to this extent substituting bargaining for
deliberation would satisfy the fundamental aim of deliberative democracy. Atleast one claim that deliberative democrats often make about the general
superiority of deliberation over bargaining would need to be revisedÐbut revised
in order to satisfy the demands of deliberative theory itself.
In any actual political context, a general defense of bargaining is not likely to
be plausible. The main problem with bargaining as a general substitute fordeliberation is that it accepts the current distribution of resources and power as a
baseline, the place to begin the negotiations. On the face of it this is not the bestsite for a moral defense of democratic procedures or outcomes.14 It issigni®cant
that no one defending NICE'srejection of Relenza attempted to justify the
13E. Rous et al., ``A purchase experience of managing new expensive drugs: interferon beta,''
British Medical Journal, 313 (November 9, 1996), 1195±6.)
14Gutmann and Thompson, Democracy and Disagreement, pp. 57±8.
decision as the outcome of bargaining. Nor did anyone in Commons suggest that
bargaining should play any role in the process, as they might propose for a labor±
management dispute, or a controversy about tax policy.
Isthere no limit to what deliberative democratscan treat asprovisional? They
can encourage reinterpretationsof the meaning and implicationsof deliberativeprinciples, even the guiding principle of reciprocity, but they cannot
accommodate the wholesale rejection of the moral justi®cation required notonly by reciprocity but many other morally based democratic theories.
Deliberative democrats can welcome criticism of any of their principles,including reciprocity, but they cannot accept a general rejection of the
requirement that binding political decisions must be justi®ed by moral reasons.
The refusal to give up that requirement is not peculiar to deliberative democracy.
To reject the idea of moral reasoning in politics tout court isto abandon not only
deliberative democracy, but also any form of democracy that would claim that itslawsare justi®ed to the citizenswho are bound by them. Although criticsof
deliberative democracy sometimes write as if they reject moral reasoning inpolitics, they rarely face up to what such a rejection would entail either in
What such rejection would meanÐeven in a partial form in a particular caseÐ
can be illustrated by imagining what would have happened if NICE had made itsdecision about Relenza on the basis of considerations of bargaining power. The
single most powerful agent in this case, the one who stood to gain the most from
NICE's decision, was Relenza's manufacturer and distributor, the Glaxo±Wellcome pharmaceutical company. When Glaxo executiveslearned of the
possibility that NICE might recommend against NHS funding of Relenza, theythreatened to abandon Glaxo's operations in the UK. They also said they would
encourage other pharmaceutical companiesto boycott the British economy. Asitturned out, NICE stood its ground and Glaxo backed down from its threat.
Deliberation and justice coincided in this case, and both prevailed. That wouldhave not been the outcome, given the baseline distribution of power, had NICE
Deliberative democrats thus rejectÐand not just provisionallyÐany theory
that denies the need for moral justi®cation, and therefore also any theory that
bases politics only on power. Deliberative democrats are committedÐand notjust provisionallyÐto mutually justi®able ways of judging the distribution of
power. Deliberative democracy acceptsthe provisionality of itsprinciplesbutrejects the provisionality of moral reasoning itself as a way of assessing politics.
Theorists who claim that politics is only about power must reject far more thanthe moral termsand the adjudicative meansof deliberative democracy. They
must also reject criticism of any current distribution of power, however unjust itmay be. Or if they criticize it, they must do so in terms thatÐon their own viewÐ
the persons who would be constrained by the power have no moral reason to
accept. If they decline to search for political principles and practices that can be
mutually justi®ed, who should listen to them? Certainly no one who is motivated
to ®nd fair termsof social cooperation. Their audience can be only people who
have themselves already given up on ®nding mutually defensible reasons. Thosewho would renounce reciprocal reasons are therefore trying either to persuade
the already converted or to reach the unreasonable. In the ®rst case their audiencehas no need and in the second case no reason to listen.15
IV. WHY THE PRINCIPLES SHOULD BE POLITICALLY PROVISIONAL
We are now in a better position to address the second objection against includingsubstantive principles in deliberative democratic theoryÐthat their inclusion
usurps the political authority of democratic citizens. A democratic theory thatincludes substantive principles can declare that a law citizens make is unjust,
however correct the proceduresby which they make it. It isno comfort to thedefender of the authority of citizens to be told that the substantive principles are
morally provisional. Even morally provisional principlesÐif they are the most
theoretically justi®able at the timeÐcarry the implication that they should bepolitically enacted. Acting on thisimplication deniesdemocratic citizensthe
authority to determine through a deliberative process what should be politicallyenacted and why. To respond to this objection, deliberative democracy relies on
the second kind of conditional statusÐpolitical provisionality.
Political provisionality means that deliberative principles and the laws they
justify must not only be subject to actual deliberation at some time, but also thatthey be open to actual reconsideration and revision at a future time. Like the
rationale for treating principles as morally provisional, the justi®cation for
regarding principles as politically provisional rests on the value of reciprocity.
From the perspective of reciprocity, persons should be treated not merely as
objects of legislation or as passive subjects to be ruled. They should be treated aspolitical agentswho take part in governance, directly or through their
accountable representatives, by presenting and responding to reasons thatwould justify the laws under which they must live together. We showed earlier (in
Section I) why reciprocity requiresactual not merely hypothetical deliberation.
Because deliberative principles must be justi®ed in an actual deliberative process
in which citizensor their accountable representativestake part, the politicalauthority of democratic citizens is to a signi®cant degree respected.
But it still may be objected that once a principle or law is justi®ed in this
process, it acts as a constraint on other new laws that citizens may wish to make.
Because the body of citizens and their representatives who deliberate about the
new laws are never exactly the same as those who enacted the old laws, thedemocratic authority of citizensat any particular time isheld hostage to
principles justi®ed at a previous time. Providing such constraints is of course
what constitutions are supposed to do, and that may be why some deliberative
democratsare wary about regarding their principlesaspart of a constitution. But
if all the principles of deliberative democracy are seen as provisional in the senseof being open to revision over time, these constitutional constraints that embody
substantive principles such as basic liberty and opportunity are not so threateningto the political authority of citizens.
Deliberative democratsdo not favor continual deliberation, but they are
committed not only to deliberation about laws at some time but also to the
possibility of actually reconsidering them in the future. The extent to which anylaw is subject to actual reconsideration should depend on the strength of the
available moral reasons and empirical evidence (and any other morally relevantconsiderations) supporting it, which often change over time. In the case of
Relenza, part of the reason that NICE's decision was justi®ed is that it called for
review after a year'stime, an appropriate interval in light of the signi®cantchanges that may occur over the course of even a single ¯u season. A decision by
any of®cial body not to fund Relenza even if otherwise well supported would notbe justi®ed without its providing for the possibility of continuing deliberation in
the future. To be sure, at some point a decision is reached, as it was in this case,and it is justi®ably enforced. But deliberative theory emphasizes, more than other
democratic theories, what happens before the decision andÐeven more to thepoint of provisionalityÐwhat happens after.
Political provisionality thus goes further than its moral counterpart. It implies
that principlesshould be open to challenge over time in an actual politicalprocess that not only permits but encourages revision. Even when a law is rightly
enacted today, the practices and institutions of deliberative democracy shouldensure that it is subject to the regular reconsideration that is necessary to its
justi®cation over time. Deliberative democrats should therefore be especiallysuspicious of practices that routinely defer to the ``intent of the framers'' or that
make constitutional amendments almost impossible even when the inheritedreasons for supporting the laws in question are not compelling. Deliberative
democrats should be favorably disposed toward practices that attach sunsetprovisions to everyday laws and procedures, and require administrators to issue
periodic ``impact statements'' describing the effects of these laws and the
regulationsthat enforce them. We noted earlier that the moral justi®cation forNICE's decision against funding RelenzaÐthe claim that the use of Relenza does
not signi®cantly affect anyone's basic opportunity in lifeÐdepends in part onevidence that the drug does not signi®cantly reduce complications in high-risk
patients. In its deliberations, NICE demonstrated its respect for politicalprovisionality by taking speci®c institutional measures: limiting their decision
to one ¯u season, providing for continuing review, and recommending furtherresearch.
The objection that the presence of substantive principles in a deliberative
theory pre-emptsdemocratic authority, it should now be clear, either provestoo
much or too little. It proves too much if the mere inclusion of substantive
principles is taken to imply that these principles must therefore be politically
binding on citizens. This objection would apply equally to the inclusion ofprocedural principles, which may be no less reasonably contestable than
substantive principles. If the objection were accepted, it would requiredemocratic theory to exclude all reasonably contestable principles, procedural
aswell assubstantive. The objection provestoo little if the complaint isonlyagainst making provisional judgments (however substantive) that challenge laws
enacted by proper procedural methods. Even the critics of deliberativedemocratic theory could hardly fault it for rendering provisional judgments of
this kind. If political theory were disbarred from offering such judgments, itwould have little relevance to the democratic politicsit purportsto criticize.
V. WHEN MORAL AND POLITICAL JUDGMENTS CONFLICT
Ideally, the moral and political judgmentsthat deliberative democracy renders
coincide. What deliberative politics decides will satisfy deliberative morality.
Indeed, this happy conjunction occurs more often than is usually assumed. In the
case of NICE, the decision against funding Relenza, made in a process that wasmore deliberative than that in which such decisions were made in the past, seems
to be morally justi®ed (at least provisionally). The most reliable studies, asreported by NICE, show that the only bene®t of Relenza is a one-day reduction of
in¯uenza symptoms in the median patient at a direct ®nancial cost to the public
of as much as 10 million pounds annually. NICE also took into account the costof the expected increase in visits to doctors by typical in¯uenza patients (most of
whom are otherwise healthy), and the risk that the increase would overload thehealth care system for a very minor health bene®t. Furthermore, NICE found no
evidence that Relenza reduces any of the serious, life-threatening effects ofin¯uenza on high-risk patients.
Nevertheless, there were some critics who raised reasonable moral objections
to NICE's decision. In the Commons debate, the physician±MP mentioned earlier
complained that the decision not to cover Relenza discriminated against poor
people. ``When we talk about rationing of NHS treatments, we aren't saying noone in the UK hasthem. What we are saying isthat they aren't available to poor
people. The rich and those who can afford it can get these treatments privately.''The money saved for all patients would be at the expense of the poor because
more af¯uent citizens could obtain Relenza by means of postcode prescribing.
The decision by NICE against funding Relenza would therefore at least appear to
be sacri®cing the welfare of some poorer citizens in order to save taxpayermoney.
Whatever the meritsof thisMP'scriticism, all partiesto the dispute over
Relenza should be able to agree that he has raised a serious moral question about
a decision that was reached by politically justi®able (deliberative) process. Even
if, under current conditions, the NHS cannot do anything about this differential
access to drugs such as Relenza, defenders of the decision (and other similarly
hard choices) should be prepared to acknowledge the moral costs inherent in asituation in which rich people tend to live in richer districts that provide better
health care and they also can buy any health care treatment on the market whilepoor people are completely dependent on the NHS for funding those treatments
that may be cost-effective for society as a whole.
In this case there was some reasonable disagreement about whether the
political judgment conformed to the moral judgment, but in many other casesthere may be no doubt that the (procedurally correct) political judgment con¯icts
with the (carefully considered) moral judgment. A deliberative process thatdeliberative democratsrecommend may yield an outcome that runscontrary to
one or more of the substantive principles of justice that deliberative democrats
also wish to defend. This kind of con¯ict does not seem to be a serious problemfor a purely substantive theory, which simply declares the outcome of the process
unjust. Similarly, a purely procedural theory faces no serious problem here either;it declaresthe outcome just, aslong asthe procedureswere proper. But, aswe
have seen, a deliberative democratic theory should include both procedural andsubstantive principles because the pure approaches at best neglect and at worst
deny the moral complexity of democratic politics. A democratic theory thatrecognizes what is morally at stake in political decision-making must contain
principles that are both substantive and procedural. Moreover, as we have also
seen, the basic premise of deliberative democratic theoryÐreciprocity that callsfor mutual justi®cation among free and equal personsÐsupports both substantive
Thus, compared to the purer theories, deliberative democracy more fully faces
up to the potential con¯ictsbetween moral and political deliberation. It doesnotprovide a simple resolution, but instead relies on deliberation itself to deal with
the con¯icts as they arise. But then the question persists: how can deliberativedemocrats af®rm substantive conclusions about politics and still support the
value of actual deliberation, which may or may not produce those sameconclusions? Political philosophers, including deliberative democrats like us,
reach substantive conclusions (including conclusions about what procedures are
most justi®able) without engaging in any actual political deliberation. This seemsto ¯y in the face of the commitment to actual rather than hypothetical
adjudication of political disagreements.
Some criticsof deliberative democracy pose thisproblem asa paradox of
deliberative democracy.16 The criticsargue that if, on the one hand, they acceptthe arguments and conclusions of a substantive deliberative theory (such as the
one we present in Democracy and Disagreement), they need not bother calling
16Frederick Schauer, ``Talking asa decision procedure,'' Deliberative Politics, ed. Macedo,
for actual political deliberation. The substantive theory provides all the reasons
anyone who accepts it needs for making sound political judgments, and without
the aid of any actual deliberation. If, on the other hand, criticsreject thearguments and conclusions of the substantive theory, then they should also reject
the deliberation that it recommends. Either way, deliberative theory that includessubstantive principles seems to eliminate the need for the practice of deliberation.
As our earlier discussion of moral and political provisionality should indicate,
substantive principles (even the political conclusions) that deliberativedemocratsdefend do not pre-empt actual deliberation. According to the very
terms of the theory, the (substantive and procedural) principles and conclusionsneed to be subjected to the rigors of actual deliberation over time; that is part of
what it means to treat them as politically provisional. (Moral provisionality also
bene®ts indirectly from political provisionality, because individuals thinking inthe privacy of their own homes or of®ces often draw on ideas, arguments, and
perspectivesÐor respond to challengesÐthat public deliberations bring to theirattention.) Deliberative democratsoffer their principlesand conclusionsnot as
authoritative philosophical constraints on democratic politics but as provisionalcontributionsto democratic deliberation.
The conclusions that deliberative democrats reach about substantive and
procedural principles should be understood as normative hypotheses about
political morality. Given certain assumptions about reciprocity, for example,
certain principles are the most mutually justi®able. The hypotheses are normativebecause simply showing that some people, even a majority, in fact reject the
principlesor their policy implicationsdoesnot refute them. Restricting coveragefor experimental drugsto only those who are willing to participate in random
clinical trials, for example, may be the best policy even if a majority of citizensreject it.
But the principles and policies recommended by deliberative theorists are still
hypotheses because they may be refuted or re®ned by showing that there are
better arguments for competing principles or conclusions in the same context.
And they are hypotheses about political moralityÐnot morality in generalÐ
because their con®rmation, refutation, or revision calls for public deliberation in
the democratic process. Whether a normative hypothesis is con®rmed, refuted, orre®ned, this kind of criticism can succeed only by subjecting rival arguments to
the rigors of actual deliberation. Deliberative theorists should of course take intoaccount the imperfections from which any such process suffers in practice. But so
should they take account of the imperfections in their own process of reasoning,less obvious though these imperfections may be (to them if not to other theorists).
The problem of the con¯ict between moral and political judgment is
misconceived if it is understood as requiring a choice in general or in advance
between substantive and procedural principles. Both kinds of principles are
subject to deliberation and are equally provisional in the way we have described.
The choice between substantive and procedural principlesÐwhen they con¯ictÐ
is similarly subject to deliberation and should accordingly be regarded as
provisional. Neither substantive nor procedural principles have priority eventhough citizens sometimes justi®ably choose one over the other at some
In the debate about NICE in Commons, both the critics and the defenders
appealed to both substantive and procedural principles. They addressed thejustice of the decision (does it violate equal opportunity by hurting poor people?),
and the process (does it insulate the government from demands to increaseexpendituresfor health care?). Even while agreeing that the processwasbetter
than in the past, some critics challenged the substance of the decision. And evenwhile agreeing that the substance of the decision was correct, some critics
questioned the process. Some criticized and some defended both substance and
process. But no one tried to argue that as a general rule one had priority over theother, or that the disagreement should be resolved by deciding once and for all
whether substantive or procedural principles should prevail. In this respect thedebate in Commonsillustrated the nature of public deliberation in many of the
best democratic forums. The debate also captured the moral complexity ofdemocratic politics far better than do theories that seek to resolve the con¯ict of
substance and procedure by excluding one or the other, or declaring one ratherthan the other trump.
Deliberative democratic theory can and should go beyond process. It can
consistently incorporate both substantive and procedural principles. It should gobeyond process for many reasons that we have suggested, but above all because
its core principleÐreciprocityÐrequires substantive as well as proceduralprinciples. Reciprocity is widely accepted as a core principle of democracy, but
even those democrats who do not emphasize this principle argue from ideals suchas free and equal personhood, mutual respect or avoidance of majority tyranny,
which like reciprocity require both substantive and procedural principles tojustify the laws that democracies adopt.
Deliberative democratic theory isbetter prepared to deal with the range of
moral and political challengesof a robust democratic politicsif it includesboth
substantive and procedural principles. It is well equipped to cope with the con¯ict
between substantive and procedural principles because its principles are tovarying degreesmorally and politically provisional. Deliberative democratic
theory can avoid usurping the moral or political authority of democraticcitizensÐand yet still make substantive judgments about the laws they enactÐ
because it claims neither more, nor less, than provisional status for the principlesit defends.
City Way Medical Practice Vasectomy Information Leaflet General Information Vasectomy is a safe, simple and very effective method of contraception for men. It should also be considered permanent . You will need to have two appointments. The first appointment will be for counselling and at this appointment you will be given a date for your forthcoming operation so please bring y
Alco_1893007162_6p_01_r5.qxd 4/4/03 11:17 AM Page 388 This lawyer tried psychiatrists, biofeedback, relax- ation exercises, and a host of other techniques to con-trol her drinking. She finally found a solution,uniquely tailored, in the Twelve Steps. W hen i wasa newly minted lawyer starting outin the practice of criminal law, there were fiveof us in our law office. My favorite lawyer was the