True liberty I am able to imagine only in the form of federation

Lajos Kossuth, 1850 (alas, not 1848!)

Historians are right to describe the 19th century as the age of nationalism. While many also depict the 20th as the triumph of the nation-state, with more justice it could be called the century of its failure, despite the vast proliferation of the form. If collapsing empires brought us the first World War, the new problems of the nation state prepared the ground for the second. In our own century, looking around the world, we encounter countless examples of nation-state failure to solve the problem that brought it into being: the management of plurality and the self-determination of different political identities.

Throughout the crises of empires and of nation-states, the option of federal union was ever-present, promising to solve what neither other political form could ultimately deal with. I define federal unions, following many authors,[i] as non- or post-state polities that preserve independent units, “states,” within political aggregations that have their own identity — these polities representing new subjects in international law. Mostly, federal unions are established by constitutional contracts or treaties, though it is theoretically possible to create them through processes of devolution. In either case, the internal relations of a federal union are domestic, rather than international. Whether formally or not, a federal union has two[ii] types of constitution: of the units and of the union. The relationship between them is heterarchical rather than hierarchical, depending on political legitimacy based on common principles rather than a narrow legality linked to institutions of coercion. While centralized nation-states that control the means of violence also rely on both legality and legitimacy, in the case of federations the balance is different, in favor of legitimacy.[iii] In a federation, both levels have institutions: generally executive, and legislative, and at the very least judicial. A federal union has no sovereignty internally, and externally, depending on the type of federation, either the union or its units could be understood as sovereign, in the modern rather than the classical sense of Bodin and Hobbes.[iv] But as Carl Schmitt thought, and as we will see, the choice may be a matter of ongoing contestation. Finally, a federal union is either a genus with subtypes as Forsythe thought (treaty organization, confederation, federation), or, as in Kelsen’s conception, a spectrum where different degrees and areas of unification allow many forms, most of them “hybrid.” In either case, whether a genus or a spectrum, federal unions are located between two book-ends: a temporary alliance on one side, and a sovereign, federal state on the other.

The history of the form or forms of federal union (e.g. the Swiss federation, the United Provinces and, to an extent, even the earlier United States[v]) is complex. Older than the nation state, federations and federal unions have been often depicted as its successor from a normative point of view. Yet as Latin Americans, Central Europeans, Middle Easterners, and South Asians know, projects of federation fail most of the time, and even when they succeed their instability is notorious. Today only historians are aware of the federalist plans of Simon Bolivar, Frantisek Palacky, Lajos Kossuth, the authors of the Cabinet Mission Plan, Brit Shalom, Leopold Senghor, and Gamal Abdel Nasser, all of which failed, and dramatically. While there are many federal states in the world, the sovereign independence of their units has become mostly a fiction. The United States, Canada and India have become federal states. At this point in space and time only the European Union seems to be a viable federal project, and yet our time (June 23) and place (Budapest) both indicate that this project too is in deep trouble.

Why federate?[vi] The short definition I provided already indicates the combined purpose of forming a relatively large polity while preserving the political independence of the units within it. E plurubus unum. This has been depicted as the desire either to “hold together” or to “come together”.[vii] The dichotomy is slightly misleading, because in the case of the old multi-national empires both motivations are obviously present. The British Government of India Act of 1935 provided for federalism to hold the Raj together, while the Cabinet Mission Plan attempted to bring together several (a very large number if we count the princely states) already politically differentiated Hindu and Muslim communities. Looking at it generally, the deeper reasons for this type of effort pertain to domains of political regimes, geopolitics, economic welfare, as well as cultural differences. The oldest of these has been articulated by Montesquieu: the maintenance of republican form of government given the modern need for a polity with a large territory and population. The Federalist Papers stressed geopolitical reasons, defense against empires and imperialism, as well as maintaining the related desideratum of internal peace. The founders of the European Union were even more concerned with the peace of the subcontinent, and perhaps also the challenge of a new empire, the Soviet Union, but they paid equal attention to creating a single market able to sustain new, economically interventionist socially conscious states. Finally, the explicit aim of Brit Shalom and Hannah Arendt was the management of difference, and the defense of minorities in a region of the world whose cultural make-up they rightly regarded as incompatible with the exclusionary logic of nation states. Thus, speaking normatively, peace, democracy, welfare, and cultural self-determination are the norms justifying the project of federation even if many older federations and especially modern federal states have violated some of these principles.

Why do federations fail? I haven’t the time to examine the reasons for the failure of particular federal experiments, and wish to raise only the main problems that have emerged since the beginning, down to our day. I see four such issues: the design problem, the boundary problem, the democracy problem and the sovereignty problem already alluded to. All are relevant to the European Union, though I will focus today on the fourth, sovereignty, and to some extent on the third, democracy. As for design problems, the most famous could be also called the Prussia problem, meaning that it is very difficult to maintain the federal form if there is one unit much larger, more populous, and powerful than the rest. The danger, as under the old German Kaisereich, is that one state will extend its internally hierarchical organization to the rest, producing an empire or a nation-state in substance if not in form. It is also obviously difficult to construct a federation from two units. Finally, asymmetric federations, as in the cases of Kashmir in India, Quebec in Canada, and the Kurdistan Regional Government in Iraq, are perpetually torn between secession and recentralization attempts. As to the boundary issue, when relatively successful, federations may try to extend their territorial control, leading to imperialism in many, perhaps all instances. But as Hannah Arendt noted, this leads to a dual organization of metropoles and territories, with the likely result that the imperialist relations of one subvert or at least endanger the republican organization of the other.

The democracy problem is currently the most obvious. It has two dimensions. If democracy is understood as the rule of the majority, based on one person one vote, giving states of different sizes equality in a federation may be seen to violate that principle. The dilemma could be mitigated by pointing out that there is nothing sacred about the larger territory being the fundamental unit, and that a more intensive democracy on the lower level of the states could compensate for losses at the level of the union. Nevertheless, when the union government, understood as incorporating some inequality, has some significant powers over individual citizens, the problem reappears, depending on the nature of these powers. Moreover, the difficulty is exacerbated by the indirect relationship of officials at the union level to voters within the states.

The second dimension of this problem is equally important. Assuming the independence of the units, the states, there are limits regarding the enforceability of a clause concerning the requirement of a democratic form of government, and adherence to fundamental rights. In the EU and the Council of Europe, a concept like the margin of appreciation involves relatively broad limits within which a state can determine its own form of government, for example whether it is parliamentarian, presidential or semi-presidentialist. What about a clearly authoritarian form of what has been called an “executive presidency,” in Africa and now Turkey? What about a monarchy, where the head of state recovers some genuine executive powers? What happens if only some fundamental rights are under threat? And when a constitutional court is packed, with the removal of some of its jurisdiction? Even when forms of federal intervention are provided, it is never clear which types of issues would justify them. Intervention becomes especially difficult when the government of the target country uses what I have called the ”Frankenstein“ argument, meaning that various (partially) authoritarian features, present in almost all democratic constitutions, are selected from different countries, and are cobbled together in a regime much more authoritarian than any of them.

While the three problems of design, boundary and democracy are most relevant to already existing federal unions, their anticipation on the basis of historical experience, as in the case of the American Anti-Federalists’ critique, may lead to a rejection of a close union, or even, when nationalism or other forms of particularism are strong, to resistance to any type of federation beyond a mere alliance. The same is true for the issue I wish to concentrate on, namely sovereignty. The matter of its supposed anonymity or “deferral” was rightly focused on by Carl Schmitt, and it was a bad idea for Arendt, Forsythe and Olivier Beaud to think the question of sovereignty irrelevant to federations.[viii] Schmitt articulated several antinomies of federal unions, all of which in my view are linked to the problem of sovereignty. Without analyzing these antinomies, let me note them: self determination vs. subordination, the possibility of federal intervention vs. its denial; the possibility of secession vs. its denial. Each of these questions could be decided if we were certain about the location of internal sovereignty: but with its deferral (Schmitt) or suspension (Arendt, Forsythe, Beaud) we cannot be. But this is so, as Jean Cohen maintains, only on the theoretical level rather than the level of the actors’ consciousness. Even in theory, with respect to external affairs, sovereignty in the contemporary sense of “sovereign equality” must be present to regulate the relation with other polities, whether states or empires. With the survival of sovereignty, externally, the internal problem cannot be definitively eliminated, whether a clear choice is made or not regarding the locus of its external dimension.

Once that choice is made, internal sovereignty claims tend to follow in spite of the organizational form that is supposed to exclude it, either to make consistent or to balance the external sovereignty. [ix] Thus, on the level of the relevant actors, performatively, sovereign claims are made on both sides, as we can see in the American debate between J.C. Calhoun and Daniel Webster, Jefferson Davis and Abraham Lincoln. Accordingly, from one point of view, the states’ full self-determination must be affirmed, federal intervention must be banned, and secession must be possible. From the other perspective, that of the union, the self-determination of the units is enhanced through its limitation, federal intervention must be possible when the republican clause of the Constitution (Article 4 sec 4 Clause 1) is under threat, and secession cannot be allowed in a “perpetual union.” Whatever the theoretical structure of the union, the antinomies can lead to open conflict, and unilateral intervention and/or secession.

Presumably, in principle at least, both the elimination of internal sovereignty, and the choice of the external sovereign can be regulated in a constitutional treaty. So can many of the rights associated with full sovereignty, as the ius belli and “exit rights.” For example, the right of secession can be entrenched, as it was in the old USSR and it is now in the Lisbon Treaty of the EU (Article 50), though the right is not entirely unilateral. Yet even assigning only the powers of external sovereignty to the states in a constitution, like the ius belli, creates the temptation to claim its internal powers as well. A right of secession too can lead to the demand of more internal powers (and just did in the case of the UK). If however the overall constitutional arrangement seems to benefit the union, by implying its supremacy, defenders of states’ rights can appeal to the constituent power they originally possessed as transcending the constitution itself. This is especially the case when the amendment rule requires unanimity, as in the EU, preserving the form of origins of the union. While on the contrary there may and should be an amendment rule, giving the constitution-making power to some combination of the organs of the union and states, without unanimity on either level, as in the US Article V, the states can still claim, as did Calhoun, that they possess something more fundamental, namely the original as against the derived constituent power, based on the contract of independent units. The answer that a status contract has changed the identity and the relation of the original participants most likely will not alleviate the conflict. It certainly did not in the US until the military defeat of the Southern States.

I think it is well known that before the Civil War, the Southern and Northern states of the US had dramatically different forms of life and economic organization, including forms of government and rights, if we consider the material rather than the formal constitutions. Contrary to J. Jay’s claim in the Federalist, the states were in many respects hetero- rather than homogenous. Heterogeneity only increased with religious toleration, slave emancipation, territorial expansion and immigration. It is this example among others that led to Schmitt’s belief that only “homogeneity” among the peoples of the units, and between them and the people of the union, can eliminate or mitigate the antinomies, or the deeper antinomy of sovereignty leading to violent conflicts. I would like to abstractly support this claim, while rejecting what Schmitt thought was desirable and possible, namely homogeneity based on ethnicity, language and/or religion. I don’t doubt that heterogeneity on these three levels has often led to deep internal conflicts in federations. Yet, so does the attempt to establish homogeneity by relying on any of these three factors. The replacement Cohen and I first proposed is homogeneity of political form, based on establishment of the same republican or democratic form of government and fundamental rights throughout the federation, all entrenched preferably in both state and union constitutions. We assumed that a civic notion of citizenship would follow from this institutional framework, if supported in the public sphere and the education system. I still think that this is a basic condition of a normatively defensible form of integration by homogeneity, but now believe it is insufficient. Even a civic notion of citizenship in the states has a tendency unfortunately to transform into an ethnic or religious one, as we see every day. While this cannot happen with respect to the citizenship of the union, which is fated to be civic, unfortunately in a multi-ethnic, -lingual and -religious federation, solidarity resources are too thin

The point of the insufficiency of merely political homogeneity can be well grasped if we consider the reasons why Schmitt introduced homogeneity as the solution of the antinomies in the first place. On grounds derived from Rousseau, he implied that there should be no conflict among ultimate claimants of sovereignty (individuals in Rousseau, polities in Schmitt) if they have the same view of the world, opinions and interests. While we have separate particular wills, the general will we all do share, making us relatively homogeneous. All actors are thus supposed to be able to easily identify with the actions and the decisions of all the others. Thus for example, if either a union or a unit goes to war or decides to suppress an insurrection, all other units as well as the union will maintain their solidarity and, if needed, generate support. This argument may or may not work for individuals in a small polity as Rousseau suggested, or for Schmitt’s religious, ethnic or national homogeneity, but certainly does not work for democratic regimes. Consider again the issue of going to war. If democratic assemblies on the level of the union, and in the states, debate commencing hostilities, and if there is genuine debate, the outcomes in each assembly are unpredictable. The harmony of their majorities is by no means guaranteed. This maybe a good thing, if it could stop a stupid war like the one in 2003. Yet as we have seen even then that the war cannot easily be stopped, and the formation of the macabre “coalition of the willing,” of which both Britain and Hungary were members, had deeply corrosive effects for the EU.

That crisis is now almost forgotten. Europeans unfortunately did not follow (Obama’s Chief of Staff) Rahm Emanuel’s dictum of never letting any serious crisis go to waste.[x] The two current ones, of the economy and of immigration, present an opportunity to learn, perhaps the last such opportunity. Each is a crisis of redistribution. The economic one, because the wealthier states, led by Germany, have refused to use the full mechanisms of redistribution (collective taxation, issuing European bonds, launching new employment-producing projects) to keep the weakest members viable. On the contrary, in the immigration crisis, Germany became the victim. Now other states led by the UK refuse to redistribute the burdens of large population inflows to the great detriment of some very strong as well as some of the weakest members (Germany, and again Greece). Neither refusal will work, because Greece will never pay its debt, and with Erdogan allowed to turn the faucet on and off, the immigrants will keep on coming, first to Greece and then to Germany as well as the states in the North that refuse “quotas”.

The corrosive effects of these twin crises should be obvious to the participants of this conference. If the voters of the UK vote for exit, the crisis of redistribution and the immigration crisis will have played a significant role. Voters resent even the paltry redistribution within the Union[xi], which they think is to their detriment, and, even more, fear the uncontrolled inflow of new immigrants. Closer to where we are today, increasingly authoritarian regimes in the East gain more and more support from those who rightly think the redistribution is too small to justify the level of regulative intervention[xii], and wrongly imagine that they are targets of settlement by many immigrants.

What would it mean not letting these twin crises go to waste? Together, in my view, they point to the level where greater homogeneity could be achieved, presupposing and reinforcing the same democratic forms of government and structure of rights in all the states. Federations, as defined here, are not based on equality among states, and the very independence of the units could become a spur to greater inequality. While, among the goals, federating economic welfare plays an important role, the result is not spontaneously achieved by the magic of an even greater internal marketplace. For many member states and population sectors, the contrary happens, especially when neoliberal ideologies replace ideas like the “social Europe” of the previous generation of Brandt, Schmidt, Delors and Mitterand. Ever increasing inequality produces, however, the kind of heterogeneity among members that becomes the reason for leaving, for both the weak and the strong, as the terms Grexit and Brexit indicate.

As G.A. Toth kindly pointed out to me, equalization through redistribution could of course lead to the same type of conflict I indicated in relation to the example war. This is indeed so, if the types and extent of redistribution are to be determined on the policy level. The wealthy, whether in London, Barcelona or Milan, will undoubtedly object to supporting the consumption of the supposedly profligate and undeserving, whether in Athens, Seville, or Palermo. They ought to be more willing of course if redistribution would favor production and productivity[xiii] rather than mere consumption, but perhaps it would be wrong to presuppose too much rationality to be expressed in polls and votes, both being influenced by tabloids. Thus redistribution as a way of producing greater homogeneity cannot work on the level of ever changing policy considerations; and therefore it must be constitutionalized. Yet, only a more democratic constitution and process of constitution-making could make such a constitutionalization of redistribution legitimate. The democratic constitutionalization of redistribution would not only be an opportunity to address the highly burdensome if slowly decreasing democracy deficit of the union, which has been much talked about. It would also provide an even more needed opportunity to address the rapidly increasing democracy deficit within some member states by for example tying any new level of redistribution benefit for any given country to adhering to strict political and fundamental rights standards.

But of course, all this is relevant only for those who choose to remain in the Union. Paradoxically, it is the threat or the actuality of British secession that should put the constitutionalization of redistribution in a more democratic constitution again on the agenda. If the UK leaves, it is obvious that Europe must be reconstructed it if it is to survive. But the same would be true if the UK stays. Europeans are now getting the third big scare after the first two (the economic and emigration crises), in the form of a secession crisis. We can easily imagine that if one important state leaves, others could follow. Admittedly, any attempt to create or re-create a social Europe may itself inspire defections or secession efforts. But leaving things as they are would certainly have that result. Thus the issue should be faced squarely, unlike in 2004 when the constant British threat led to formulas eventually rejected by two much more loyal European communities.[xiv] A new constitutional convention, as provided for even by the Lisbon treaty, could work, but obviously not if again put into an intergovernmental frame of negotiation and approval. As already proposed in 2004, a new convention should therefore give itself a new ratification rule. I would model it on the highly innovative example of the old Article VII of the U.S. Constitution, which allowed a smooth transition between a treaty and a constitution.[xv] According to that provision, instead of unanimity, only a qualified majority was needed to ratify. Such a constitution then can have a rule of change tied to qualified majorities on both levels, as the U.S. Article V. Contrary to its desirable amendment rule, which would bind all, however, the new constitution would apply only to those who do ratify. Admittedly, under such a scenario some states could be lost, though interestingly none were in the US in 1787-8. In Europe the ones lost should be lost, but not on ethnic or geographical or economic-developmental grounds, but rather because of the likely unwillingness of some states to accept any form that would work, unlike the present one. Right now, it is the federation as a whole that is in danger, and only new, decisive action can save it in the face of challenges that are not going to disappear: the challenge of economic coordination, the challenge of immigration, and the challenge of inequality.




[i] Especially Jean Cohen who synthesizes the views of Carl Schmitt, Murray Forsythe, and Olivier Beaud, and tries to go beyond them. Globalization and Sovereignty (Cambridge: CUP, 2012)

[ii] Kelsen thought there were three constitutions: of the units, of the union, and of the combined system (the latter having a material constitution regulating the relations of the other two). But a constitution has internally hierarchical relations, and in the case of the federation the whole does not. See General Theory of Law and the State (Cambridge: Harvard, 1945)

[iii] Federations represent a field in between international law dependent to a great extent on legitimacy, as well as law in the sense of non-coercive secondary rules, and domestic law capable of relying on legality in the traditional sense, involving primary rules enforced through hierarchical and centralized coercion. See T. Frank The Power of Legitmacy Among Nations p.23 who sees the contrast as between legitimacy and legality even when comparing a federation like the US in 1787 and a national state. Of course all three domains rely on both legality and legitimacy, primary and secondary rules. But the balance between these normative forms is different in each case.

[iv] I prefer this formulation to Cohen’s attribution of sovereignty to the units.

[v] M. Forsythe   Unions of states : the theory and practice of confederation (1981)

[vi] I follow questions raised by D. Elazar Exploring Federalism (1987), but only part of his answers. I think the tradition inaugurated by Schmitt (as in f.n. 1) has probed deeper.

[vii] J. Linz, A. Stepan, Y. Yadav Crafting State Nations (2011)

[viii] O. Beaud. “Fédéralisme et souveraineté: notes pour une théorie constitutionnelle de la Fédération.” Revue du droit public et de la science politique en France et a l’etranger 1 (1998): 83-122. See also: Cohen op.cit.; Arato and Cohen “Banishing the Sovereign? Internal and External Sovereignty in Arendt.” Constellations 16.2 (2009): 307-330.

[ix] See Arato-Cohen op. cit.

[x] “You never let a serious crisis go to waste. And what I mean by that it’s an opportunity to do things you think you could not do before.”

[xi] At the moment the EU budget is astonishingly small, about 1% of the combined GDP of the member countries. (114 billion Euros in 2013; 155 billion for 2016). In contrast, the national European budgets together represent 48% of their combined GDP. The US Federal yearly budget is 41% of GDP.

[xii] Paradoxically, with all the demands of Law and Justice to restore Polish sovereignty,  Poland is and has been one of the top beneficiaries of EU redistribution such as it is European Commission adopts ‘Partnership Agreement’ with Poland on using EU Structural and Investment Funds for growth and jobs in 2014-2020 (European Commission, 2014) Hungary has been only slightly less fortunate, if we take into account its much smaller population. “Structural Funds for Hungary”

[xiii] As it is supposed to in the so-called partnership agreements of the EU.

[xiv] P. Norman. The accidental constitution: the making of Europe’s constitutional treaty. EuroComment, 2005.

[xv] The legality of this move has been long contested. On this, see Ackerman We the People I and II with their different answers, as well as my Post Sovereign Constitution Making: Learning and Legitimacy (Oxford: OUP, 2016) chapter 1, that points out the duality of interpretations.