The Future Soul of Europe: Nationalism or
A Critique of David Miller's defence
ARENA, University of Oslo
The challenges facing the European Union will require
extensive and mutual trust to ensure compliance. Some
have argued that a shared European-wide national identity
is necessary to ensure such stability of practices and
institutions. The philosopher David Miller is pessimistic
about the prospects of European-wide nationalism of the
appropriate sort. The present paper provides an in-depth
and critical discussion of Miller's argument, and
concludes that Miller's defence of nationalism as the
unique source of trust is unconvincing. Moreover,
Miller's pessimism regarding the European Union is
premature, if relevant at all. Alternative grounds for
supporting common institutions are available, and may be
within reach for the European Union. The shared bases
among citizens need not include a broad range of values
and cultural belonging, but might plausibly be restricted
to Just Patriotism of the kind suggested in the Liberal
One of the most pressing issues within international
ethics today is no doubt the question of national
sovereignty and the persistence of nationalism. This is
of special pertinence and importance to Europeans given
the history of civil wars and wars between nation states
on the European continent. Many see the European Union as
an attempt to keep unrest between nation states in check.
One objective has indeed been to build and maintain trust
and stability with the aid of fair and common
institutions, so that Europe does not once again become
the violent battleground that it became twice in the 20th
century. All in all, there is no doubt that the debate
about the European Union, its shape, future and
preconditions, offers instructive contributions to a
broad range of normative discussions about nation states,
their legitimacy and their future.
The President of the European Commission, Mr. Romani
Prodi, is on a quest for Europe's 'soul'. In an eloquent
speech to the European Parliament in 1999, he insisted
that the further development of Union institutions must
'gradually build up a shared feeling of belonging to
Europe'. The developments he has in mind include
strengthening the European Parliament, regulating the
right of Member State veto in exceptional cases, and
changes in the Commission power. To be sure, the
challenges facing the European Union will require
extensive and mutual trust the prospects of
requiring Europeans to risk their lives in military
action under Union command, and increased reliance on
majority rule to mention but two. But a shared
European-wide national identity need not be part of the
solution, or so I shall argue.
The British political philosopher David Miller might
be thought to support Prodi's diagnosis, by providing a
thought-provoking defence of nationality as the 'cement
of society'. The national community
embodies a real continuity between generations;
and perform a moralising role, by holding up before us
the virtues of our ancestors and encouraging us to live
up to them.
people are held together not merely by physical
necessity, but by a dense web of customs, practices,
implicit understandings, and so forth. ...
in a national community a case can be made out for
unconditional obligations to other members that arise
simply by virtue of the fact that one has been born and
raised in that particular community.
(Miller, 1995: 36, 41, 42)
Miller's philosophical views concerning the need for
national identity provide illuminating insights about the
prospects of legitimate European institutions. His
conclusions are dire for proponents of an ever closer
Union: he is pessimistic about the prospects of
European-wide nationalism of the appropriate sort.
The present paper undertakes an in-depth and critical
discussion of Miller's argument, partly to assessment Mr.
Prodi's diagnosis and prescription for Europe. The upshot
is both bad news and good news. The bad news is that
Miller's defence of nationalism is unconvincing. The good
news, for those who share Mr. Prodi's ends, is that
Miller's pessimism is premature, if relevant at all.
Alternative grounds for supporting common institutions
are available, and may be within reach for the European
After a presentation and elaboration of Miller's
argument in section 1, problems are considered in section
2. We may agree with Miller's claim that trust among
citizens is important. But it remains an open question
whether nationalism is the only source of trust, whether
it indeed is such a source at all and whether what
Miller proposes is nationalism at all. Section 3
considers Miller's reasons for rejecting alternative
bases for trust drawn from 'universalist' theories.
Section 4 defends Liberal Contractualism against such
objections, in support of the role of Constitutional
Patriotism (Habermas, 1992) or a public theory of justice
(Rawls, 1971a, 1993). These alternatives withstand
Miller's criticisms, or so I argue. Thus Miller's
pessimistic prognosis for a fair Europe is premature.
By way of introduction, I first consider the main
motivation that fuels Miller's contribution. He argues
for nationalism as a source of trust required for
redistributive welfare regimes.
Dire Prospects for European Welfare Regimes?
Is a fair Europe likely? Pessimism about where the
European Union is heading is not new, particularly not
among the defenders of redistributive arrangements. Some
critics have noted that efforts at limiting or
compensating for the perceived injustices of the market
have failed as a result of the decision procedures of the
European Union. Co-ordination problems among Member
States render it difficult to maintain the domestic
safety nets, yet prevent the emergence of a European-wide
alternative. Such co-ordination traps may be resolved by
joint action, but the political prospects are dim
(Scharpf, 1988, 1999).
Others note that democratic arrangements at the
European level are missing. Thus we have little reason to
hope for a democratically enacted pan-European welfare
regime to supplement the domestic efforts. Democratic
institutions have yet to be developed, and importantly,
there is no European 'public sphere' of media for
deliberation. In the absence of opportunities for taking
others' weal and woe into consideration through public
discussion and scrutiny, majority rule runs the risk of
tyranny (Grimm, 1995).
These pessimists are prepared to accept that
institutions and media might change, and that these
institutional features and lack of deliberative arenas
may be transient. Thus, careful reform over the long haul
cannot be ruled out. David Miller, on the other hand,
provides a more fundamental argument against European
redistributive policies, on the basis of
thought-provoking claims about the role of nationalism,
and its permanent absence in the European Union 1993
(Miller, 1995, 1993). Over the years, David Miller has
provided sustained and often sound arguments concerning
equality and community, so his present warnings should
give us pause.
Any acceptable normative political theory, on Miller's
view, must justify redistribution beyond self-interest.
Such a theory must support two considered judgements.
Individuals unable to provide for themselves must be
protected and provided for, and there are special
obligations holding between compatriots: 'obligations
that we owe to one another obligations of the sort
that are manifested in social security schemes and public
provision for citizens' needs' (Miller, 1995: 178). These
commitments to solidarity make Miller wary of theories
that regard community membership as a 'cultural
supermarket', since they fail 'to address the question
what holds a society together, and what is the source of
the obligations that we owe to one another' (Miller,
1995: 178). Miller holds that arguments of 'reciprocity'
based on self-interest are insufficient. Instead, he has
sought to rehabilitate the notion of nationalism, of some
disrepute in recent political philosophy. Shared
nationality provides the answer to the question as to who
has responsibility for those unable to provide for
themselves. It is only
because we have prior obligations of nationality
that include obligations to provide for needs that arise
in this way that the practice of citizenship properly
includes redistributive elements of the kind that we
commonly find in contemporary states.
(Miller, 1995: 72)
His version of nationalism provides one justification
for redistribution, and no other accounts appear to fit
the tasks he identifies:
Nationality has to be credited with much that is
valuable in our political life, especially, as I noted at
the beginning of the paper, those social-democratic
institutions that in European states have served to
counteract the economic polarisation and social
fragmentation that a market economy tends to produce.
(Miller, 1995: 81)
Global markets threaten this solidarity. Duties
towards the vulnerable may corrode due to the 'polarising
effects of the global market' (Miller, 1995: 187). When
domestic national identities dissolve, political elites
will not be stopped from dismantling the institutions
that protect the vulnerable.
For the purposes of this article, let us accept such
solidarity norms of redistributive obligations and hence
lay aside libertarian conceptions of justice such as
those of David Gauthier (Gauthier, 1986, 1990). The issue
to be addressed is Miller's defence of the nation state
as the largest circle of such solidarity. Miller holds
that his account and apparently no other
will allow 'that redistributive elements can be built in
which go beyond what the rational self-interest of each
participant would dictate' (Miller, 1995: 75). I shall
suggest that Miller dismisses such alternative theories
Miller seeks to argue for nationalism on the basis of
trust, and this renders him pessimistic about the
prospects of redistributive arrangements in Europe. This
argument is flawed. The upshot is that the need for
redistributive arrangements based on trust need not draw
on nationalism of the kind Miller suggests. In
consequence, the prospects of redistributive arrangements
in Europe need not be as bleak as Miller suggests. The
quest for Europe-wide trust is important, but a European
national identity of the sort Miller supports may not be
Nationalism to Maintain Trust and Stability
One reason why Miller's argument is of obvious
interest for the study of international relations and the
future of the European Union is his explicit concern for
stability. General compliance with existing policies and
institutions is a paramount condition for a political
order worthy of the name. One - of several - mechanisms
for securing the requisite expectations of reciprocal
compliance builds on shared values. Miller's main
argument for nationalism provides a thought-provoking
exposition of this argument:
a viable political community requires mutual
trust, trust depends on communal ties, and nationality is
uniquely appropriate here as a form of common identity.
(Miller, 1994: 143)
A charitable interpretation of this argument from
stability takes it to proceed in four steps.
1. Compliance with redistributive arrangements
typical of welfare state regimes requires trust
The state is more stable if based on a national
community, since compliance with costly policies requires
trust. The population must believe that need-based
provisions will in fact be distributed fairly according
to need (Miller, 1994: 142).
The reasons why trust is important for welfare
arrangements appear to stem from problems of partial
compliance: the suspicion either that the contributors or
the beneficiaries of redistribution are failing to do
their part, to the detriment of others. The participants
in redistributive welfare arrangements will want to know
that others contribute when they contribute themselves.
Partial compliance may make already costly burdens
excessively costly for those who do their share. Even the
suspicion of such partial compliance can wreak havoc with
co-operative arrangements. This is because the motives
for compliance are often complex.
Three different motivational set-ups may be
distinguished. Firstly, people may comply unconditionally
regardless of whether others do their share.
For this segment of the population, the expected
behaviour of others is irrelevant.
People may instead have a preference for not doing
their part, even when others co-operate but they
prefer general compliance to general defection. This
second kind of motivation the Free Rider
gives rise to dilemmas. Each of the participants
may prefer all others to comply, without themselves
complying. The result may be the end of collective
action. A collective solution to this conundrum is
coercive compliance, sanctioning unilateral defection for
the sake of general co-operation. In our setting,
sanctions can ensure that citizens of Europe contribute
to redistributive schemes benefiting non-nationals, or
punish those Europeans who seek benefits beyond their due
The third preference set-up is known as an 'Assurance
Game': each prefers general compliance, and prefers to
comply, if all (or most) others comply but each
prefers to defect if all others defect (Sen, 1967). 'I am
prepared to do my part, but only if I am convinced that
others will do theirs.'
2. Trust is important to remove problems of
Trust in the behaviour of others can serve to reduce
the problems generated by the suspicion that others are
free riders and by conditional compliers. Such trust can
be secured by sanctions, which shift individuals'
preferences, so that even free riders will prefer to
comply if the risk of being caught is high enough.
Sanctions also promote compliance by conditional
compliers, since sanctions provide more reliable
expectations about the compliance of others. But for a
community of conditional compliers, information may
suffice for trust. Sanctions can therefore play an
important role for fostering trust in the compliance by
others, particularly when individuals do not know whether
the others are free riders or conditional compliers.
However, sanctions are a less than perfect mechanism
for ensuring compliance. Sanctions can be costly and
cumbersome, and suspicion of other's low risk defection
can threaten complex co-operative ventures such as
Thus, unconditional or conditional compliers are to be
preferred. When all participants are known to prefer such
compliance, suspicion of defection is reduced and
hence the defection by conditional compliers is reduced
(Ostrom, 1991). This makes for more stable redistributive
Trust that others are conditional or unconditional
compliers with redistributive arrangements, rather than
free riders, is necessary for general compliance and
3. Nationality provides the common identity
appropriate for redistributive arrangements
Miller then claims that communal ties and common
identity, in the sense of shared interests, are central
for ensuring the requisite trust. Clearly, individuals
who prefer conditional or unconditional compliance, and
who are known to have these preferences, will secure
stable redistribution better than individuals who are
free riders. Thus, a shared, public commitment to such
arrangements seems beneficial to redistribution.
Miller regards nationality as a particular collective
source of personal identity:
These five elements together a community
(1) constituted by shared belief and mutual commitment,
(2) extended in history, (3) active in character, (4)
connected to a particular territory, and (5) marked off
from other communities by its distinct public culture
serve to distinguish nationality from other
collective sources of personal identity.
(Miller, 1995: 27)
The need for shared beliefs and mutual commitment is
presumably due to the role of nationality in removing
problems of partial compliance through trust.
4. Nationalism provides the unique common
identity appropriate for redistributive arrangements
Miller then appears to hold that nationalism provides
the uniquely appropriate common identity
appropriate for redistributive arrangements. It is not
clear how this follows, except as an argument by
exclusion that there are no alternative accounts
left, once libertarianism is rejected.
Miller's Argument Considered
What are we to make of Miller's argument for
nationality from trust?
Let us agree with Miller's observation that trust
plays an important role for stability, and that correct
stable expectations about others' behaviour is important
for maintaining redistributive arrangements.
However, it is not clear that Miller's conception of nationality
is required to solve the problems of partial compliance.
a) What must participants be committed to?
In light of the sketch offered about why trust is
needed, it is not clear that those maintaining a practice
need to share a broad or fundamental set of beliefs or
commitments beyond a commitment to upholding the
redistributive arrangements. An important, and
unresolved, issue is thus what shared values and 'shared
public culture' Miller can have in mind and
whether this is reasonably called 'nationality'. For the
issue of redistributive arrangements in the European
Union, the relevant questions are parallel: What shared
values and culture is needed to sustain such solidarity
Three alternatives may be indicated.
Firstly, there may only be agreement on complying with
certain principles of action. Thus, famously, Jacques
Maritain argued for human rights as constituting a very
thin overlapping consensus:
It is not reasonably possible to hope for more
than the convergence in practice in the enumeration of
articles jointly agreed. The reconciling of theories and
a philosophic synthesis in the true sense are only
conceivable after an immense amount of investigation and
elucidation of fundamentals, requiring a high degree of
insight, a new systematisation and authoritative
correction of a number of errors and confusions of
(Maritain, 1949: 11-12)
However, it is not clear that mere agreement on
certain principles or practices suffice. Rules of
practices must be interpreted, revised and applied; and
suspicion of defection will flourish unless the reasons
others have for complying are visible for others.
Secondly, a 'thick' set of values, if shared by all,
can have a substantive content confirming the
redistributive practices as expressing the proper
responsibility of each member of society for one another.
This approach faces the challenge of value pluralism. How
can these institutions and such comprehensive values be
imposed on all members of a society, even when citizens
disagree about the good life? This is one fundamental
challenge facing Miller's view insofar as it is taken to
support 'nationalism'. Furthermore, of course, only some
such substantive values will endorse redistribution.
Miller must thus provide some criteria for determining
which beliefs, commitments and cultures are within the
acceptable range for legitimate nationalism
domestically and in the European Union.
A third strategy requires agreement on certain
practices, and on their immediate justification
sufficient to ensure shared interpretation, application
and stable compliance, all the while seeking to avoid
controversial metaphysical assumptions or contested views
about the proper ends of humans. I take John Rawls,
J�rgen Habermas, and other liberal theorists to
contribute to such projects. Indeed, Rawls underscores
the concern for stability, and claims that
the problem of political liberalism is: How is it
possible that there may exist over time a stable and just
society of free and equal citizens profoundly divided by
reasonable though incompatible religious, philosophical,
and moral doctrines?
(Rawls, 1993: xviii)
It would seem possible for Miller to pursue the third
strategy. However, he hesitates about the alleged
'neutrality' imposed by such views. This brings us to the
second issue pertaining to Miller's position: what is the
content of the common political culture shared by all
b) What common political culture must be
An important issue concerns the content of the 'common
political culture'. Miller regards this as
a set of understandings about the nature of a
political community, its principles and institutions, its
social norms, and so forth, [as opposed to] ... a private
culture is all those beliefs, ideas, tastes, and
preferences that may be unique to an individual, or more
likely shared within a family, a social stratum, an
ethnic group, or what has been called a 'lifestyle
(Miller, 1995: 158, referring to Bellah
et al., 1985)
However, Miller's account fails to explain in a
convincing way what is and should be part
of the political culture. The problem is partly one of
contested demarcations. This can be illustrated by his
own view that Scots, the Welsh, and the Northern Irish
are not separate nations. Britain should not be thought
of as 'a multinational state in which common political
institutions hold together communities with separate
identities' (Miller, 1995: 173), but rather as one
nation. This surely raises the issue of what are
necessary or sufficient elements of a common political
The distinction between private and public culture
becomes even more important insofar as Miller wants to
distinguish his nationalist position from liberalism. He
notes that his nationalists,
though perhaps favouring neutrality on some
cultural questions, are committed to non-neutrality where
the national culture itself is at stake. In other words,
where some cultural feature a landscape, a musical
tradition, a language has become a component part
of national identity, it is justifiable to discriminate
in its favour if the need arises.
(Miller, 1995: 195)
Moreover, it is unclear why Miller requires a 'thick'
political culture in order to maintain trust in shared
institutions beyond that suggested by liberal
c) Why political and territorial aspirations?
If trust and a narrow political culture are required,
we are led to wonder why Miller insists that the nation
must have political and territorial aspirations. Here he
departs from other defenders of moderate nationalism,
such as Yael Tamir (). Miller claims that 'it is an
essential part of having the [national] identity that you
should permanently occupy that place ... A nation ...
must have a homeland.' (Miller, 1995: 24) This 'helps to
explain why a national community must be (in aspiration
if not yet in fact) a political community' (ibid., 24).
However, making this into a defining criterion of
nationality does not explain why general compliance with
institutions requires such aims.
It is not clear that considerations of trust support
the restriction to groups with political and territorial
claims. To be sure, he notes that traditional sovereignty
need not always be needed, but still the nation itself
must decide on what to claim, and what counts as
constitutive of their culture:
The guiding ideal here is that of a people
reproducing their national identity and settling matters
that are collectively important to them through
democratic deliberation. To achieve that, they need a
political unit with authority of the relevant scope, but
what that scope must be will depend on the particular
identity of the group in question, and on the aims and
goals that they are attempting to pursue. ... It is
therefore going to be difficult to set apriori limits to
the proper scope of sovereignty from this perspective.
Moreover, we cannot tell in advance which particular
features of a society's way of life will come to assume
importance as markers of national identity. ... In this
area /national currency/, a collective belief that
something is essential to national identity comes very
close to making it so. Once you combine the principle of
national self-determination with the proposition that
what counts for the purposes of national identity is what
the nation in question takes to be essential to that
identity, it follows that nothing in principle lies
beyond the scope of sovereignty.
(Miller, 1995: 100-1)
One reason for insisting that nations must have
political autonomy is to protect the common culture,
which in turn is regarded as a condition 'for a person's
having an identity and being able to make choices in the
first place' (Miller, 1994: 154). This defence seems
unsatisfactory. It would support legal immunity not only
for groups we intuitively may grant are nations, but also
for a wide variety of groups who share values and seek to
pursue them. At the same time, this defence stops far
short of justifying sovereign statehood. What seems
required is systematic argument, in sector after sector,
on the basis of a variety of interests, about what sorts
of legal powers a group needs to secure the legitimate
elements of their common culture. Thus this argument
might support the claims of any set of individuals should
enjoy whatever powers they need to ensure the flourishing
of their own 'culture'. The result would be a much
broader range of social groups enjoying a broad variety
of legal powers and immunities than what is traditionally
meant by 'sovereign nation states'. In response, Miller
might pursue two further reasons for the concern with
political and territorial claims, relating to trust.
This definition of nationality may make sense when
shared territory is a good indicator of shared values or
interests. This territorial focus also identifies one set
of comprehensive and mutually exclusive cultures,
avoiding the myriad of conflicting claims that would
otherwise arise for individual with multiple loyalties
However, neither of these two is decisive in today's
world. Firstly, in states with freedom of religion and
speech, territorial borders do not identify a population
with a broad range of shared values and a common,
dominant identity. Political borders may delineate the
reach of common institutions and sanctioning mechanisms,
but these cannot uncontroversially be used in the pursuit
on common values. Nor, I have suggested, are shared thick
values needed. Secondly, the territorial focus removes
conflicts only if contested territorial claims can be
adjudicated. However, this is notoriously not the case in
the present world order. Furthermore, even though
undisputed borders may eliminate conflicting claims,
other solutions are also available for instance, a
distribution of authorities in a federal or
consociational arrangement according to private-public
distinctions, or social functions, or by appeals to
subsidiarity. Miller is right to point out that such
arrangements often entail less redistribution across
sub-units (1995: 84-85), but this fact does not entail
that less redistribution is wrong. The force of this
observation depends on whether we agree with Miller that
redistributive claims are completely independent of the
extent of mutual interdependence in collaborative
To conclude, it is not clear that Miller's defence of
nationality on the basis of the need for trust succeeds.
What is justified is a shared set of practices with some
public, common value platform, but this requires neither
a 'thick' public political culture nor a set of
individuals who have political or territorial
aspirations. This modest shared normative platform would
seem a poor explication of the term 'nationality'.
Miller's Objections to Universalist Theories of Justice
Miller raises several concerns
against 'Ethical Universalist' normative theories,
apparently including those associated with Habermas and
Rawls.  He holds
that the personal commitments such theories focus on are
not sufficient for the tasks performed by nationality. In
the European context this is relevant, as Habermas for
one has stressed the need for a shared political culture
focussed on the constitution or its equivalent. Such a
constitutional patriotism must suffice (Habermas, 1992:
1997). Miller disagrees:
the national identities that support common
citizenship must be thicker than 'constitutional
patriotism' implies. If we are attempting to reform
national identity so that it becomes accessible to all
citizens, we do this not by discarding everything except
constitutional principles, but by adapting the inherited
culture to make room for minority communities.
(Miller, 1995: 189)
Miller offers two reasons why these identities must be
'thick', summarised by his claim that
Subscribing to them [the constitutional
principles] marks you out as a liberal rather than a
fascist or an anarchist, but it does not provide the kind
of political identity that nationality provides.
(Miller, 1995: 162)
a) Universalist principles fail to provide the
political identity offered by nationality
Talk of citizenship should not replace nationality. To
get a grasp of Miller's argument it is worth quoting in
full. Citizenship cannot solely be
understood in terms of subscription to a set of
political principles: tolerance, respect for law, belief
in the procedures of parliamentary democracy, and so
forth. These principles should undoubtedly feature
centrally in any story about what it means to be British
today, and as I shall suggest shortly it would be very
helpful to have them formally inscribed in a
constitutional document. It does not, however, seem that
these principles, which after all are the common currency
of liberal democracies everywhere, can by themselves bear
the load that would otherwise be carried by a national
identity. As I have already argued, a national identity
helps to locate us in the world; it must tell us who we
are, where we have come from, what we have done. It must
then involve an essentially historical understanding in
which the present generation are seen as heirs to a
tradition which they then pass on to their successors.
(Miller, 1995: 175)
It seems correct that a set of such principles
provides little in the way of identifying the proper role
of a people in the world, or the appropriate story of its
past. However, two comments are appropriate. It remains
to be seen whether these theories only allow
shared principles of this general form we turn to
this in section 4. Moreover, and against Miller's claim,
it is unclear why there must be a 'thick' and shared set
of values and practices that should serve as the
justification for political power within a nation
state or in the European Union. As discussed in section
2, the arguments from trust do not support the need for a
shared 'thick' platform.
b) Constitutional principles do not mark
unique features of the nation
The principles Miller lists would be similar and
shared among citizens of many or most legitimate states.
This creates a problem for theories focusing on such
principles, or so Miller argues. Nothing distinguishes a
nation from others on this view. For Miller this is a
flaw. He holds that
a national identity requires that the people who
share it should have something in common, a set of
characteristics that in the past was often referred to as
a 'national character', but which I prefer to describe as
a common public culture.
(Miller, 1995: 25)
National divisions must be conceived as natural
ones; they must correspond to what are taken to be real
differences between peoples.
(Miller, 1994: 140)
As discussed above, the reasons for nationality based
on trust do not require that the shared values are unique
to the participants: The point is to ensure general
compliance among themselves, not to set them apart or
exclude anyone else.
Against this, Miller might respond that the problem
remains because these principles do not bind individuals
to the practices and people of their own state, rather
than to any other just state. Thus, universalist theories
seem unable to account for 'political obligations'
the special ties that hold between individuals and their
state. I consider this worry in the next section.
For these purposes, the normative tradition of
'liberal contractualism' include writers such as Brian
Barry (Barry, 1995), John Rawls (Rawls, 1971, 1993) and
T. M. Scanlon (Scanlon, 1982, 1998) as well as
several elements of J�rgen Habermas' contributions
(Habermas, 1996, 1992). The aim here is to sketch some
elements consistent with these approaches, in order to
indicate how Liberal Contractualism avoids the problems
identified by David Miller.
Liberal Contractualism addresses the conditions under
which citizens have reason to accept institutions and
cultures as normatively legitimate and binding on their
conduct. The set of social institutions as a whole should
secure the relevant interests of all affected parties to
an acceptable degree. Institutions are legitimate in this
sense only if they satisfy principles that can be
justified by arguments in the form of a social contract
of a particular kind. The principles of legitimacy to
which we should hold institutions, are those that the
persons affected would unanimously consent to under
conditions which secure and recognise their status as
appropriately free and equal.
Note that, unlike the contractualist tradition of
Thomas Hobbes and David Gauthier, Liberal Contractualism
does not seek to justify morality from a premise of
self-centred individuals. Rather, it assumes that
individuals generally have an interest in being 'able to
justify one's actions to others on grounds they could not
reasonably reject' (Scanlon, 1982: 116). They 'desire to
act in accordance with principles that could not
reasonably be rejected by people seeking an agreement
with others under conditions free from morally irrelevant
bargaining advantages and disadvantages' (Barry, 1989:
8). This commitment to give reasons is an expression of
our belief in and respect for the reasonableness of
others (Macedo, 1990). The aim of Liberal Contractualism
is thus not to justify morality, but rather to bring this
commitment to justice to bear on our rules and practices
be they domestic or European.
The limited significance of consent
One contribution of Liberal Contractualism is to
delineate some limits to the morally binding rules and
practices we find ourselves surrounded by, regardless of
actual consent. Every individual's interests must be
secured and furthered by the social institutions as a
whole (Dworkin, 1978). This commitment is honed by the
notion of possible consent, allowing us to bring the
vague ideals of equal dignity to bear on pressing
questions of legitimacy and institutional design.
Our moral obligation to obey the law of the land is
justified in part by the claim that this social order
could have been the subject of consent among all affected
parties. But this does not entail that such hypothetical
consent creates the moral obligation or duty in
the same way as free and adequately informed consent
binds those who so consent. Instead, Liberal
Contractualism serves to delineate the limits to these
duties that hold regardless of actual consent, including
what Rawls' theory of Justice as Fairness calls the
natural duty of justice, 'to support and to comply with
just institutions that exist and apply to us' (Rawls,
1971a: 115; cf.Klosko, 1994).
Thus, Liberal Contractualism does not entail that
individuals are only bound by obligations voluntarily
taken on. The existing legitimate institutions are not
binding on us because we actually consent, or participate
in a daily tacit plebiscite (Renan, 1992). To be sure, we
usually act according to the practices we find ourselves
part of (Walzer, 1997: 54), but we do not have, and have
never had, a real freedom with regards to the social
institutions. Indeed, ordinarily we cannot choose to
reject them, and not even the act of voting expresses a
morally binding tacit consent to be governed.
Nevertheless, we have many duties that we have not
explicitly or tacitly consented to. Actual, tacit or
hypothetical consent is not the source of moral
obligation to comply. The idea of possible consent in the
contractualist tradition does not provide the source
of moral duty, but is an expression of one important
condition for such duties. Obedience is required only
when power is distributed fairly. Appeals to consent thus
serves to recognise legitimate authority, but consent is
not held to generate the moral authority of institutions
Mode of arguments: consistency vs deduction,
Of some relevance to the issues at hand, we should
note that the contractualist approach leads us to search
for principles against which no reasonable objections can
be made. Principles are presented for such assessment,
and the process of checking whether objections can be
made yields a set of permissible principles. Note two
There is no sense in which such principles of justice,
or particular institutions, are deduced or generated
by the process of checking whether equal respect is
secured. The procedure checks for consistency, and does
not offer a deductive path. Moreover, the process can in
principle allow several alternative principles. Thus, the
set of principles may be underdetermined, in the sense
that alternative principles may all be unobjectionable.
Furthermore, the same set of principles for legitimacy
may allow a variety of sets of institutional
arrangements, each of which satisfies the distributive
requirements of Liberal Contractualism. The moral unity
provided by such a theory is not one, therefore, of
deduction, but of an analysis of the institutions which
show that they are consistent with, and can be regarded
as an expression of, a view of individuals as enjoying
In light of these comments, it should be clear that
this tradition might allow for different just
institutional arrangements. Blueprints of institutional
design are out of reach. And instantiations of just
institutions may be different, yet possibly equally just
for instance within different European states.
The interest in culture
Contractualist theories are said
to deny the intrinsic value of community, and ignoring
the 'embedded' nature of human beings.  Instead, society is
exclusively regarded as instrumental for benefiting the
interests of 'atomistic' individuals. However, we may
agree with Taylor's conception of the self as embedded,
and with MacIntyre's claim that our duties can only be
determined by reflecting on the roles we are born into,
which expand and constrain our choices (Taylor, 1985;
MacIntyre, 1981: 216).
Liberal Contractualism recognises that the basic
structure shapes our expectations and aspirations in
fundamental and inescapable ways. Social institutions
have a pervasive impact on the development and
satisfaction of our interests by framing our
expectations. We are concerned with the legitimacy of
social institutions precisely because they exercise a
strong influence on us, or our life plans and our
expectations (Rawls, 1978). With changes in values,
norms, institutions, history and language, new options
for life choices appear, while others disappear. Members
have an interest in revising their plans as options and
consequences change. Liberal Contractualism can thus
accommodate the communitarian concern for constitutive
attachments and commitments, found within the traditions
and roles we take part in, and not chosen by the
individual. The satisfaction of legitimate expectations
is an important interest, and stable social institutions
are crucial for making and pursuing life plans. We thus
have good reason to maintain social institutions, insofar
as it is only under fairly stable institutions that
expectations can be created and met (Follesdal, 1996).
Challenge 1: The significance of particular
The relationship between principles of justice and
participation in particular institutional practices must
be made more clear in order determine how the 'sense of
justice' expresses itself among compatriots. Recall that
a 'sense of justice' in Rawls' sense does not only mean a
commitment to two very general principles of justice, or
to abstract constitutional principles. Rather, the sense
of justice is 'an effective desire to comply with the
existing rules and to give one another that to which they
are entitled' (Rawls, 1971a: 312). Thus acting on a sense
of justice entails interacting in our day-to-day lives
with other individuals in accordance with the legitimate
expectations they have about our behaviour, honouring
their trust in our responses.
Miller, as well as some communitarians, hold that the
fact that we inhabit certain roles and are members of
certain communities provides us with a sufficient
justification for acting in a certain way. Thus the fact
that an institution is ours is taken to be morally
relevant when justifying the institutions and our
compliance with its rules. The contractualist
perspective, on the other hand, is criticised for leaving
little if any justificatory role to the common life of an
existing community, appealing instead to abstract
But this objection misunderstands the role of
principles of justice, which address the institutional
issue of what standards a set of institutions should
satisfy. For this task, it is irrelevant whether I or we
live under those institutions. It does not matter for
this topic, 'from the moral point of view,' that this
is the society that has shaped me into who I am. But of
course, within any set of institutions,
individuals justify their actions by appeal to the rules
of the set of institutions in place (Rawls, 1971b). So
one's culture may well spell out the content of these
special duties that are said to bind us, but these rules
and practices do not give a complete justification
of the duties, since we must then offer a justification
of the existing culture as well. And for this purpose it
does not matter that this particular set of
institutions has shaped me into who I am.
Liberal Contractualism thus grants a justificatory
role to existing practices one is part of (cf. : 134). To
illustrate where this account differs from Miller's, note
that he holds that
the relevant question to ask, if some question
arises about what I owe to my compatriots, is not just
how I feel about them at present, but what flows from my
(normally unchosen) membership of a constitutive
(Miller, 1997: 74)
Similarly, Stuart Hampshire says that institutions are
sometimes justified 'historically' by showing that 'they
have become an essential element in the subject's way of
life' (Hammpshire, 1983: 5), instead of by
reference to a general principle of justice.
In comparison, Liberal Contractualism offers an
account of justification that accepts such a sociological
or historical account as a part of, but only one part
of, a full justification of the institution. The
justification of my duty to abide by the rules of a
specific social institution must consist of 1) an
argument that it is just (i.e. part of a just set of
institutions); 2) an argument that this institution is in
effect in my society, and 3) that its rules apply to me.
The fact that other institutions existing elsewhere might
be somewhat more just does not by itself generate a duty
on my part to abide by them, because the second and third
steps are missing.
Thus, a complete justification of an existing and just
set of institution must refer to our shared history, even
on the contractualist view. When we justify compliance
with the existing institutions (or sanction
non-compliance with it, as the case may be) we must do
two things: 1) we must show that they satisfy (or
alternatively violate) the relevant principles of
justice. 2) We must also show that this particular set of
institutions, and not another one, does in fact exist in
our society: these rules, and not another set of rules,
are publicly known and generally complied with. Thus a
thoroughgoing justification of our institutions over
other just ones must refer in part to our shared history,
the general acceptance of these rules etc. On this view,
then, we may agree with Miller about granting some
normative importance to
the historical identity of the community, the
links that bind present-day politics to decisions made
and actions performed in the past.
(Miller, 1995: 163)
I thus submit that, in contrast to the theories
criticised by Miller, Liberal Contractualism can
accommodate and justify special duties and the value of
community, based on an ideal of the individual as deeply
embedded in social relations.
Challenge 2: Whence Special Duties?
We often have special duties towards particular
others, both one's family and compatriots. Miller
stresses special duties, and holds that they are
incompatible with universalistic theories. Again, on this
issue Liberal Contractualism agrees with Miller
but offers a better justification of special duties.
To be sure, special duties such as patriotism are not
provided or generated by contractualism, but they are
permitted, duly pruned, as compatible with principles of
justice. Thus, John Rawls appeals to the natural duty of
justice 'to support and to comply with just institutions
that exist and apply to us' (Rawls, 1971: 115). For
instance, citizens have special patriotic duties to their
own country insofar as the nation-state, including
the office of citizenship, firstly is a just set of
institutions and secondly, applies to the citizens.
Special duties, then, are ultimately supported by an
'impersonal' theory of justice: The special duties an
institution imposes are morally binding insofar as the
institution is permissible, i.e., part of a society which
satisfies contractualist principles of justice.
A stable and just peaceful order among states requires
converging expectations about mutual compliance with
practices and institutions. The pervasive influence of
nationalism has traditionally been regarded as a major
hindrance to stability in this sense. David Miller's
arguments in defence of nationality is therefore a
striking contribution, particularly since he defends
national identity for reasons of trust. I have argued
that Miller's argument yields flawed conclusions on the
basis of sound premises. He points to an important
insight concerning the need for trust in ensuring stable
institutions and practices. However, this basis does not
offer support for nationalism of the kind he sketches.
Moreover, other accounts, including Liberal
Contractualism, appear to be equally appropriate
solutions to the problem he identifies. Miller's
substantive concerns appear to be compatible with Liberal
Contractualism as I have sketched it here. Liberal
Contractualism may also avoid some of the problems of
'universalist' theories Miller is concerned with
including the need to acknowledge the value of culture
and shared practices in the right way.
Miller's dire expectations for Europe are too hasty,
particularly given his claim that even national
identities can be moulded. Historically, allegiances have
proven fluid, as has been the case in Ireland. And he
holds it important to 'purge the national identity of
elements that are repugnant to a particular minority
group (Miller, 1994: 156):
If we wish to be a self-determining nation, and if
we share our territory with people who are like us in
respects A, B and C but unlike us in respect D, it would
be perverse to insist on D as a condition of membership.
In real cases groups may choose the perverse option; but
where a common national identity already exists, it can
always potentially at least be extended to embrace all
those who inhabit a geographical area.
(Miller, 1994: 143)
Nevertheless, Miller would warn against a deliberate
political strategy for fostering a European nation:
A national identity depends upon a pre-reflective
sense that one belongs within a certain
historic group, and it would be absurd to propose to the
subjects of state X that, because things would go better
for them if they adopt a shared national identity, they
should therefore conjure one up.
(Miller, 1995: 143, my emphasis)
We can agree with Miller that the development of a
European nation in his sense would be a bad idea.
Nevertheless, we might look more favourably on Habermas'
our task is less to reassure ourselves of our
common origins in the European Middle Ages than to
develop a new political self-confidence commensurate with
the role of Europe in the world of the twenty-first
(Habermas, 1992: 12)
We might conclude, then, that Miller's pessimism about
the prospects of redistributive arrangements in the
European Union is unwarranted as of yet. This is not to
say that time may not prove him right, for instance due
to the co-ordination problems discussed by Fritz Scharpf
(Scharpf, 1988, 1998). But the grounds Miller offers for
his pessimism are insufficient. Mr. Prodi may continue
developing European institutions that require trust among
Europeans. A European national identity may be beyond
reach, but just European patriotism may suffice.
Barry, Brian, 1989. Theories of Justice. A
Treatise on Social Justice, 1. Berkeley: University of
Barry, Brian, 1995. Justice As Impartiality.
Oxford: Oxford University Press.
Bellah, Richard et al., 1985. Habits of the Heart:
Individualism and Commitment in American Life.
Berkeley: University of California Press.
Buchanan, Allen E., 1989. 'Assessing the Communitarian
Critique of Liberalism'. Ethics 99(4): 852-82.
Caney, Simon, 1992. 'Liberalism and Communitarianism:
a Misconceived Debate'. Political Studies 40:
Cohen, Joshua, 1986. 'Review of Walzer's Spheres of
Justice'. Journal of Philosophy 83: 457-68.
Dworkin, Ronald, 1978. 'Liberalism.' Public and
Private Morality. Stuart Hampshire, editor, 113-43.
Cambridge: Cambridge University Press.
Fishkin, James, 1986. 'Theories of Justice and
International Relations: the Limits of Liberal Theory',
in Anthony Ellis, ed., Ethics and International
Relations. Manchester: Manchester University Press
Follesdal, Andreas, 1996. 'Minority Rights: A Liberal
Contractualist Case', in Juha Raikka, ed., Do We Need
Minority Rights? Conceptual Issues. The
Hague/London/Boston: Kluwer Academic Publisher/Kluwer Law
Fried, Charles, 1978. Right and Wrong.
Cambridge, Mass.: Harvard University Press.
Gauthier, David, 1986. Morals by Agreement.
Oxford: Clarendon Press.
Gauthier, David, 1990. Moral Dealing: Contract,
Ethics, and Reason. Ithaca: Cornell University Press.
Grimm, Dieter, 1995. 'Does Europe Need a
Constitution?' European Law Journal 1(3): 282-302.
Gutmann, Amy, 1985. 'Communitarian Critics of
Liberalism'. Philosophy and Public Affairs 14(3):
Habermas, J�rgen, 1992. 'Citizenship and National
Identity: Some Reflections on the Future of Europe'. Praxis
International 12(1): 1-19.
Habermas, J�rgen, 1996 . Between Facts and
Norms [Original: Faktizit�t und Geltung],
William Rehg, trans. Cambridge, Mass.: MIT Press.
Habermas, J�rgen, 1997. 'The European Nation-State:
on the Past and Future of Sovereignty and Citizenship',
in The Inclusion of the Other: Studies in Political
Theory. Cambridge: MIT Press (106-127).
Hampshire, Stuart, 1983. 'Morality and Conflict', in Morality
and Conflict. Cambridge, Mass.: Harvard University
Klosko, George, 1994. 'Political Obligation and the
Natural Duties of Justice'. Philosophy and Public
Affairs 23, 3: 251-70.
Macedo, Stephen, 1990. Liberal Virtues:
Citizenship, Virtue, and Community in Liberal
Constitutionalism. Oxford: Clarendon Press.
MacIntyre, Alasdair, 1981. After Virtue: a Study in
Moral Theory. Notre Dame, Indiana: Notre Dame
Maritain, Jacques, 1949. Human Rights: Comments and
Interpretation. London: Allan Wingate.
Miller, David, 1993. 'In Defense of Nationality'. Journal
of Applied Philosophy 10(1): 3-16.
Miller, David, 1994 'The Nation-State: a Modest
Defence', in Chris Brown, ed., Political Restructuring
in Europe: Ethical Perspectives. London: Routledge
Miller, David, 1995. On Nationality. Oxford:
Oxford University Press.
Miller, David, 1997. 'Nationality: Some Replies'. Journal
of Applied Philosophy 14, 1: 69-82.
Mulhall, Stephen, 1994. 'Liberalism, Morality and
Rationality: MacIntyre, Rawls and Cavell', in John Horton
& Susan Mendus, eds., After Macintyre: Critical
Perspectives on the Work of Alasdair Macintyre.
Cambridge: Polity (205-224).
Mulhall, Stephen & Adam Swift, 1996. Liberals
and Communitarians. Revised version. Oxford:
Murphy, Mark C., 1994. 'Acceptance of Authority and
the Duty to Comply With Just Institutions: a Comment on
Waldron'. Philosophy and Public Affairs 23(3):
O'Neill, Onora, 1975. Acting on Principle: As Essay
on Kantian Ethics. New York: Columbia University
Rawls, John, 1971a. A Theory of Justice.
Cambridge, Mass.: Harvard University Press.
Rawls, John, 1971b . 'Two Concepts of Rules', in
Samuel Gorowitz, ed.,Utilitarianism: John Stuart Mill,
With Critical Essays. Reprint from Philosophical
Review 64:3-32. Indianapolis: Bobbs-Merrill
Rawls, John, 1978. 'The Basic Structure As Subject',
in Alvin I. Goldman & Jaegwon Kim, eds., Values
and Morals: Essays in Honor of William Frankena, Charles
Stevenson, and Richard B. Brandt. Dordrecht, Holland:
D. Reidel Publishing Company (47-71).
Rawls, John, 1993. Political Liberalism. New
York: Columbia University Press.
Renan, Ernest, 1992. Qu'est-ce qu'une nation?
Paris: Presses Pocket.
Scanlon, Thomas M., 1982. 'Contractualism and
Utilitarianism', in Amartya K. Sen & Bernard
Williams, eds., Utilitarianism and Beyond.
Cambridge: Cambridge University Press (103-128).
Scanlon, Thomas M., 1998. What We Owe to Each Other.
Cambridge, Mass.: Harvard University Press.
Scharpf, Fritz W., 1988. 'The Joint Decision Trap:
Lessons From German Federalism and European Integration'.
Public Administration 66(3): 239-78.
Scharpf, Fritz W., 1999. Governing in Europe:
Effective and Democratic? Oxford: Oxford University
Sen, Amartya K., 1967. 'Isolation, Assurance and the
Social Rate of Discount'. Quarterly Journal of
Economics 81: 112-24.
Tamir, Yael, 1993. Liberal Nationalism.
Princeton, New Jersey: Princeton University Press.
Taylor, Charles, 1985. 'The Nature and Scope of
Distributive Justice', in Philosophical Papers.
Cambridge: Cambridge University Press (289-317).
Walzer, Michael, 1977. Just and Unjust Wars: a
Moral Argument With Historical Illustrations. New
research for this paper ha been funded by ARENA, a
research program on the Europeanisation of the
Nation-State, under the auspices of the Research Council
of Norway and by a TSER grant SOE2-CT97-3056-EURCIT. The
author is grateful for this support, and for constructive
comments from two referees and Henrik Syse
ANDREAS F�LLESDAL, b. 1957, PhD in Philosophy (Harvard
University 1991); Associate Professor at the University
of Oslo and Senior Researcher at ARENA. Main interest:
political philosophy of international relations. Most
recent book i English: Democracy in the European
Union (SPRINGER, 1998).
He also appears to hold that these theories aspire to
offer deductions of all principles and rules from
abstract premises, and challenges the plausibility of
this strategy , 71). I share his misgivings about such
projects, but fail to find evidence that this is a
correct description of the theories offered by Habermas
different sorts of criticisms, and defences on other
grounds: , , , , , .
[Date of publication in the ARENA
Working Paper series: 15.01.1998]