The Democratic Challenge

Overwhelming evidence suggests that English is the only realistic contender for a trans-national lingua franca, defined as any language commonly used for communication between people with different mother tongues.Footnote ¹ Philippe Van Parijs's recent text, Lingusitic Justice for Europe and for the World, provides an empirically-based normative account of why we should both welcome and accelerate the spread of English as a lingua franca, first for Europe and ultimately for the world. He puts forward a two-stage argument to make his case. The first step is his ‘ethical contagion’ thesis, which states that the more people communicate from different national backgrounds the more they can accept one another as equal interlocutors and the more quickly they can reach a reflective equilibrium, or well-considered consensus, on some kind of egalitarian justice.Footnote ² A lingua franca, in other words, breaks down national barriers between those living in worse and better off regions by establishing smooth communicative conditions for cross-border identifications that are strong enough to support economic solidarity on a trans-national scale. Van Parijs's next step is to present his ‘political feasibility’ thesis, which is the view that the implementation of a trans-national conception of egalitarian justice requires creating a trans-national demos and, furthermore, that ‘the emergence of such a demos is facilitated, indeed made possible, by the availability of a common language’ (italics mine).Footnote ³ In an earlier essay, Van Parijs refers to this last presupposition as the democratic challenge and neatly sums it up in the phrase ‘No viable democracy without a linguistically unified demos’.Footnote ⁴ He emphasizes that he is referring only to the civic quality of language rather than to its potential ethnic associations. In other words, it is not a common language that represents a single homogeneous culture or ethnos that is essential for democracy but simply a widely understood language that could sustain a well-functioning demos, that is, an arena for deliberation and the mobilization of citizens in struggles for justice.Footnote ⁵ What is fundamentally at stake in Van Parijs's argument is John Stuart Mill's insight that multilingual contexts give rise to distinct public spheres of political discourse, an assumption that receives much empirical support as public spheres are divided along linguistic lines in Belgium (French, Flemish, German), Canada (English and French) and Switzerland (French, German and Italian) with the vast majority of citizens in each country accessing media and engaging politically in their own language. Though there is significant historical evidence to support Van Parijs's democratic challenge, in so far as nation-building in the nineteenth century standardly involved linguistic homogenization in democratizing states across the Western world, it is contradicted to different degrees by the relative success of the aforementioned countries as long-lasting multilingual democracies.

Belgium, Canada and Switzerland have each received fair shares of scholarly attention dedicated to understanding the key to their successes as multilingual democracies. In fact, to the extent that these countries appear to falsify the universality of Van Parijs's democratic challenge, they would seem to serve as the best possible precedent for determining how a multilingual trans-national demos might be possible. Due to space restrictions, but also because I believe it to be the most instructive case, this article focuses on determining whether or not important features of the Swiss case might be informative for considering the possibility of a European demos. Stated briefly, my central argument is that a linguistically unified public sphere, while ideally desirable, is not required for a well-functioning democratic community so long as there is a unified and robust voting space that can (a) serve to symbolically represent ‘the people’ as one regardless of linguistic divisions and (b) provide the same deliberative focus to the public spheres such that their respective discourses are both synchronized and responsive to one another. Before I can make this argument, however, I must first specify the conceptual framework with which I hope to explain the basic logic of Swiss democracy for the benefit of theorizing a European demos. This involves providing an account of political community formation that appreciates the indispensable aesthetic dimension of political discourse and elucidating the idea of a voting space which I take as crucial to understanding the nature of democratic communities. I take up these tasks in the next section. Following this, I define Switzerland as a mono-national rather than a multinational state and then elaborate the nature of its voting space. By doing this, I am in a position to explain what kind of voting space Europe needs if it is to create a multilingual transnational demos. Finally, I counter the objection that the kind of demos I advocate would be insufficient to generate the kind of solidarity required for redistributive justice.

Aesthetic Politics And The Idea Of A Voting Space

My point of departure in elucidating a basic account of political community formation is Nadia Urbinati's perceptive claim that there is an issue contemporary democratic theorists seem reluctant to face: ‘the realignment of the deliberative theory of democracy with the ideological as rhetorical characteristic of the language of politics in the constitutive process of representation’.Footnote ⁶ Here I take Urbinati to be distinguishing between two sides of political representation which have been artificially separated in mainstream democratic theory, namely the deliberative and the aesthetic. Deliberative theorists, with Jürgen Habermas being a prominent example, have largely focused on determining how deliberation can ideally legitimate representative democracy. One of their main aims is to specify standards against which discourse in the public sphere, or what Urbinati calls that ‘special terrain’ in which dialogue between representatives and the represented takes place,Footnote ⁷ can be normatively judged. Such accounts negatively associate rhetoric with stirring the passions rather than appealing to reason and so largely exclude it from any positive normative role in political discourse.

Peculiar to authors who emphasize the aesthetic dimension of political representation is the thesis that representation is constitutive of political communities and therefore of politics itself.Footnote ⁸ On this view, political representation does not mean standing-in for others so much as it involves the representative's creativity in articulating to the represented who they are and what they should stand for. To use Michael Saward's language, political representation is a highly productive symbolic force that operates by way of claim-making, such that representative claims addressed to a wide audience, when consistently well timed and well delivered, have the potential to create new political identities as well as to reinforce, redefine or even undermine the prevailing self-understanding of existing political communities.Footnote ⁹ A helpful way of conceiving this is to employ John R. Searle's notion of ‘we intentionality’, which is one particular way of explaining how collective identity and co-operative activity are possible.Footnote ¹⁰ ‘Intentionality’ is a term he uses to denote mental states that are about some object or state of affairs in the world – mental states like beliefs, desires, promises etc. How an individual is conscious of this table, those people or the current economic situation is just how he is intending them. What happens in collective intentionality is that each individual is conscious of himself as part of a given collective such that he understands certain activities as more or less in line with the group's goals. Effectively, the aesthetic aim of representative claims in politics is to manipulate the collective intentionality of targeted constituents in some way.

To specify the nature of a political community then, the question that needs answering concerns the kind of collective intentionality that is required for its effective operation.Footnote ¹¹ Determining this, however, will always be an imperfect and crude process given the unique cultural context and fluctuating heterogeneity of socio-political attitudes and identifications that characterize modern political systems. Nevertheless, what we can hope to identify with some precision is the kind of collective intentionality that serves to make a political community democratic. Clearly, if we are to accept the constitutive and creative role of representation, the point is not to overcome the aesthetic nature of politics but to understand normative concerns in its light. While there are few limits, beyond better or worse public norms, on the manner in which political discourse is carried out we can agree with deliberative theory that representative claims should ideally be based on rational argumentation and judged by constituents primarily on those merits.

The main distinguishing feature of every democratic community is what I shall call its voting space, defined as the totality of procedural devices employed by a democratic community that formally open up the main institutions of political decision making to the input of individual citizens through the ballot. A voting space is not reducible to any particular vote or election, but is the constitutionally embedded framework that specifies the conditions under which particular votes can and must take place. For the purposes of this article, two important features of a voting space should be highlighted. First, as the basis of democratic representation, a voting space is the very heart of a democratic public sphere. This is not to say that all public debate is about who or what to vote for, but rather that in a democratic context all discourse in the public sphere is permanently haunted by the spectre of the next election or common vote. Indeed, it is the freedom of citizens to cast an equal ballot in the always forthcoming vote that gives public discourse its basic democratic credentials. As Urbinati correctly recognizes, the popular vote is the kernel of democratic representation not simply because it is the basic act of authorization, but rather because it structures the public sphere in such a way as to instigate a participatory process of opinion formation between representatives and the represented.Footnote ¹² The centrality of a voting space to the public sphere manifests itself most clearly around the time of an actual vote for it is during this time that key political issues tend to emerge in their most salient and urgent fashion and when contact between representatives and their constituents is at its most intense. Nevertheless, as Jane Mansbridge illustrates in her account of anticipatory and promissory models of representation, the incumbent's desire to anticipate his constituents’ preferences in the next election and the constituents’ desire to make sure that the representative is keeping his election promises, motivates various forms of discursive interaction that help keep representatives and the represented deliberatively engaged between voting periods.Footnote ¹³

Secondly, a voting space encourages the symbolic interpretation of all particular votes as the expression of ‘the people's will’ such that they are represented to themselves, not necessarily as homogeneous or in full agreement, but as one political community. This is at least in part made possible by the fact that, except in extreme circumstances, the losers of a given vote will not exit or violently disrupt the political community but rather continue to support it in the hope of better outcomes for the future. There are, of course, many ways in which a voting space can be constitutionally articulated and procedurally arranged that will in turn have deep implications for how the public sphere functions. As I attempt to demonstrate in this article, what allows Switzerland to undermine Van Parijs's democratic challenge is the fact that its voting space is set up to be both unified and robust.

Mono-National Switzerland

Capturing the basic logic underlying the operation of Swiss democracy first requires a brief skeleton survey of its socio-political landscape. As we shall see, the difficulty is not in describing this landscape but in conceptualizing it as a mono-national or a multinational state. Switzerland is one of the world's oldest federal states, which, since its constitutional inception in 1848, has developed a complex legal-political framework that today accommodates four national languages and twenty-six cantons, which, in turn, are subdivided into a total of roughly 2,800 communes. German, though verbally fragmented by several dialects, is the majority language spoken by around 64 per cent of the population; 20 per cent are Francophone or Swiss Romands; 7 per cent are Italian speakers, often referred to as Ticinesi since they are almost exclusively located in the canton of Ticino; 0.5 per cent speak one of the five Rhaeto-Romansch dialects; and 9 per cent speak another first language.Footnote ¹⁴ Though the communes are the fundamental building blocks of Switzerland's political system, the cantons are the main federal sub-units with residual competences, most notably in policy areas like education and taxation. Cantons work off what Van Parijs favourably refers to as a ‘territorially differentiated coercive regime’, which means that there is an official language of administration and public discourse in each region.Footnote ¹⁵ Some cantons are multilingual in their coercive regimes, however, breaking up the evenness of borders that mark off the linguistic communities. The national parliament or Federal Assembly consists of two chambers: the National Council and the Council of States. The former is composed of 200 elected politicians and represents the nation as a whole. Despite the dramatically varying population and geographic size of the cantons, their citizens are equally represented at the federal level by two members who are elected to the Council of States.Footnote ¹⁶ Every four years the Federal Assembly elects seven Swiss citizens (usually elected politicians) to the executive branch, the Federal Council, with each member taking charge of a federal administrative unit.

Numerous commentators have referred to Switzerland as a multinational state, some drawing sub-state national lines around the linguistic communitiesFootnote ¹⁷ and others at the cantonal level,Footnote ¹⁸ while there is no serious proposal to consider the communes as nations. David Miller perceptively points out, however, that the idea of a multinational state implies a purely civic (state) identity that holds together several ethnically peculiar nations. Suspicious of the civic/ethnic dichotomy, which implies that political institutions can somehow be developed extra-culturally, he suggests that we would do better to jettison the idea of a multinational state in favour of ‘nested nationalities’, a concept that appreciates the cultural embeddedness of the state. This term, he believes, should be used to describe cases where citizens can be said to have national identities at both the state and sub-state level.Footnote ¹⁹ More accurately then, the question is whether Switzerland is mono-national or a case of nested nationalities.Footnote ²⁰

Part of the problem in answering this question has to do with disagreements over what in fact counts as a nation. Minimally, we can say that a nation is always an identity with a political relationship to the state where the issue of autonomy is central. To borrow Brubaker's terminology: a nation can be ‘state-framed’, which means that it is coterminus with the state and, therefore, a fully autonomous nation, or it can be ‘counter-state’, such that it either has some degree of constitutionalized autonomy within the state or is making a credible demand for such autonomy by struggling for the devolution of powers or secession.Footnote ²¹ Counter-state nations or national minorities, in other words, are rightfully so-called only if they exist as what Will Kymlicka terms ‘distinct and potentially self-governing societies’.Footnote ²² A prominent indicator of nationhood strongly associated with this definition is that there must be widespread use of the terms ‘nation’ and ‘national’ with reference to the community.

Let us begin by assessing the nationhood of the linguistic communities. According to Jan Erk, the Swiss national public broadcasting system was the first federal institution to recognize the existence not just of different languages but of distinct language communities by devolving itself to these very communities in 1964.Footnote ²³ Embracing linguistically distinct public spheres in this way, he believes, has only served to strengthen the identity of the linguistic communities, which is marked by increasing cross-cantonal co-operation along linguistic lines.Footnote ²⁴ Yet we can hardly think of the linguistic communities as nations. Most obviously, they have no political autonomy whatsoever nor is there any detectable appetite for or credible effort to achieve such autonomy.Footnote ²⁵ Furthermore, the language of nationalism is hardly ever used when speaking about the linguistic communities.Footnote ²⁶

Cantons seem to be far more promising candidates to qualify for the status of nations. Not only do they exercise an impressive degree of political autonomy, they also carry typical signs of nationhood such as a shared history and territory, flags and sometimes an anthem, as well as their own Constitutions. Indeed, along with their communes, the cantons have jealously guarded their role in granting Swiss citizenship which maintains a certain degree of homogeneity within their borders by requiring that applicants for citizenship renounce their original cultural identity and assimilate into the cantonal culture. We have good reason to believe, however, that the cantons should not qualify as nations, since the language of nationalism is hardly ever used to describe them in either federal or cantonal discourse.Footnote ²⁷ On close inspection, we find that the infiltration of Swiss identity into cantonal identity has hollowed out whatever claims to national identity the cantons may have had in the past.Footnote ²⁸ Unlike Canada and Britain, where the Québécois and the Scottish may reasonably be thought of as having distinct national identities strong enough to extricate themselves from the larger national identity within which they are nested to become self-governing societies, the cantons are thoroughly Swiss. Put differently, cantonal identities have developed with and melted into the wider Swiss identity such that being a proud Zürcher or genevois Footnote ²⁹ is just one way of being a proud Swiss. Lacking an identity distinct from or separate to the wider Swiss identity, the cantons cannot be viewed as potentially self-governing. This is partly explained by the fact that there is no real cantonal cleavage. According to Dardanelli, citizens have rarely had to mobilize along cantonal lines since the autonomy of the cantons and their various linguistic make-ups have been respected from the outset, both vertically and horizontally, across the society's political institutions.Footnote ³⁰ From this perspective, it is unsurprising that secession has never seriously threatened the Swiss federal state.Footnote ³¹ Summarily then, Switzerland is not a case of nested nationalities because its citizens are not conscious of themselves as having stacked nationalities, which is reflected by the fact that we do not find representative claims in the public sphere that could hope to generate this kind of collective intentionality.Footnote ³²

So what kind of collective intentionality does characterize mono-national Switzerland? Switzerland's long history is a gradual coming together of distinct political units, a contingent process sometimes driven primarily by the desire for mutual gain and at other times by the need to counter an external or internal threat. As a result of this process, and especially due to the explicit nation-building of the nineteenth century, a sovereign Swiss consciousness has developed which is largely defined by an attachment to its own historically developed and idiosyncratic political features. A unique interpretation of multi-level citizenship and the institutions of federalism and direct democracy, not to mention a foreign policy of neutrality, are some of the dominant features marking off a Swiss identity. Contributing to the particularly Swiss flavour of these interpretations, according to Hanspeter Kriesi and Alexander H. Trechsel, are heroic myths about how ‘the Swiss way’ was defended from outside aggressors.Footnote ³³ Important for the continued reinforcement of this identity is arguably the desire the linguistic communities share to differentiate themselves from their respective neighbouring linguistic counterparts: Germany, France and Italy.

The collective intentionality constituting Swiss nationhood should not, however, be confused with some kind of deep homogeneity. As Antoine Chollet points out, employing the insights of Eric Hobsbawm, the distinctiveness of cantonal cultures makes visible the uneven development of national consciousness that fractures all nation-states. What we call national consciousness, he rightly claims, basically refers to the existence of a common object of identification around which different images can and do form across any given nation according to socially relevant differences between regions.Footnote ³⁴ Crucially important for the sustainability of a national demos is that this object of intentionality has a persistent and significant presence in the minds of citizens and, at the very least, that the diversity of images attached to it do not conflict to the point at which the object itself is seriously called into question. More can be said about the nature of Switzerland's collective intentionality once we turn to the question of how it is possible in the face of linguistically distinct public spheres.

Switzerland's Voting Space: Undermining The Democratic Challenge

Notwithstanding the above-mentioned elements that have helped to create and maintain a Swiss identity, I wish to highlight the importance of Switzerland's voting space. Structured according to both indirect and direct democratic institutions, I consider this to be one of the primary explanatory factors behind this mono-national state's sustained success as a multilingual democratic community. The first thing to note is that Switzerland has a national party system, unlike Belgium where the party system is organized along linguistic lines. In the Belgian case, two sets of political parties fight elections out in their own public spheres until votes are cast and the victorious Flemish party (it is to be hoped) negotiates a federal government with the successful Walloon party.Footnote ³⁵ Despite its national structure, the double fragmentation of Switzerland's party system is well known: (a) more than a dozen parties tend to occupy seats in the National Council and (b) the decentralized nature of the individual parties themselves is such that the situation of the same party varies from canton to canton, effectively weakening the vertical and horizontal links between the parties’ politicians. Counterbalancing this fragmentation, however, are the facts that only some parties are consistently considered for government formation,Footnote ³⁶ and that the parties are organized on ideological rather than linguistic or cantonal lines so that the right/left cleavage is primary (often resonating with the rural/urban cleavage).Footnote ³⁷

A second thing to consider is Switzerland's long-standing direct democratic tradition. Globally, to date, more than a third of all referendums held at the state level and more than half of those held at the sub-state level have taken place in Switzerland. From 1960–2011, there were 373 referendums at the national level averaging seven votes per year.Footnote ³⁸ Usually, on about three scheduled dates each year, voters will be asked to decide on several issues at once. The volume of Swiss national referendums is chiefly due to the fact that the Swiss style of direct democracy has both top-down and bottom-up dimensions, respectively referred to as the popular initiative and the popular referendum.Footnote ³⁹ The popular initiative takes place at the beginning of the legislative procedure and is the right of citizens to make a constitutional proposal to the electorate, provided a show of support in the form of at least 100,000 signatures can be presented less than eighteen months since the initiative was registered and published in the official Federal Gazette. If the initiative is deemed valid by Parliament then the latter can either put the proposal directly to the electorate or respond by adding a counter-proposal for the voter's consideration. The popular referendum, by contrast, occurs at the end of the legislative procedure and is the right of citizens to accept or reject legislation proposed by the federal government. Depending on the issue popular referendums can be either mandatory or optional, the former taking place automatically and the latter requiring the signatures of 50,000 citizens within a hundred days of the date on which the legislation was officially published if it is to transpire. To date, six out of ten popular votes in Switzerland have been the result of either popular initiatives or optional referendums,Footnote ⁴⁰ proving that its direct democracy has a strong bottom-up dimension.

With this outline in place, I am now in a position to explain how Switzerland's voting space is both unified and robust (see Table 1). Fundamentally, what makes the Swiss voting space unified is the fact that citizens together elect representatives to the Federal Assembly and make decisions on federal legislation through direct democratic choice. The citizens’ necessary collective object of intentionality when voting on such occasions is Switzerland's future as a nation. Nationally held votes, in other words, are performative activities, where participating citizens are forced to explicitly recognize themselves as part of a national project. A unified voting space essentially presupposes the existence of ‘the people’ for whose input it is asking and, through the performance of particular votes, serves to reinforce the people's collective intentionality as Swiss. We saw in the Belgian case, however, that a unified voting space in the sense elaborated so far is not unproblematic. Conceiving of political parties as the main articulators of political preferences and attitudes, the representative claims of the dominant parties are crucial to how the citizens are conscious of themselves in relation to the state. On this basis we can affirm that the existence of a national party system, primarily framed in ideological terms, is what makes the unity of Switzerland's voting space sufficiently deep to avoid splitting the national consciousness.

Table 1

Main Features of Switzerland's Voting Space

In criticizing the idea of Switzerland as a mono-national state, Donald Ipperciel claims that ‘[e]ven in cases of state-wide referenda, it is misleading to speak in a meaningful way of the people, in the singular, expressing its will’.Footnote ⁴¹ His argument is that since political debate takes place in different public spheres the result of a referendum is simply the aggregation of multiple public spaces irrespective of any ‘unitary sentiment’ that might follow. Such an objection, I believe, comes from an implicit commitment to what Frank Ankersmit has referred to as the ‘mimetic’ view of political representation.Footnote ⁴² On the mimetic account, political representation means reflecting as accurately as possible what is already there, which, in this case, are distinct public spheres. Rejecting this view along with Ankersmit, and endorsing the aesthetic theory that recognizes the constitutive role of representation in political community formation, I propose that it is through artifices like the state-wide referendum and general election that the people are constituted as a political community. The symbolic effect of a vote is not simply to reflect what is there, but to create something new on the basis of what is there. Thus, the unitary sentiment following a referendum is far too important to warrant dismissal for it is a strong indicator that the referendum has been successful in translating the votes of individuals from different public spheres into a decision for which a single national polity is responsible. Taking this position, however, should not make us blind to the grain of truth implicit in Ippericel's claim, namely that the sense of a unified people will tend to be weaker and more easily splintered in Switzerland than in those states where there is a single public sphere.

At this point, Dave Sinardet would rightly claim that how the results of a referendum are publicly interpreted is crucial to its impact on community formation.Footnote ⁴³ Speaking in the Belgian context, where secessionist rhetoric has made the national-linguistic cleavage between Wallonia and Flanders quite pervasive, he insists that the results of any referendum can be used to drive a wedge between different public spheres just as easily as it can unite them. Even a referendum with a high level of state-wide consensus, he thinks, can be used divisively.Footnote ⁴⁴ This is a formidable objection that can be rebutted by reflecting on the robustness of Switzerland's voting space which consists in both the frequency and diversity of direct democratic votes. As both Nicholas Schmitt and Nenad Stojanovic recognize, Swiss referendums are (a) held so frequently and (b) cover such a wide range of topics that it becomes implausible for representative claims to highlight the same cleavage as important on each occasion.Footnote ⁴⁵ To clarify this, on any given issue the political tension is likely to be characterized by a left/right, liberal/non-liberal, religious/secular, urban/rural, or one of several other kinds of cleavage. This creates multiple majorities whose regular visibility makes it difficult to concentrate for a sustained period on any particular tension as definitive of Swiss politics.Footnote ⁴⁶

Democratic community formation is an ongoing aesthetic practice taking place discursively in the public sphere that finds its basic structure in the arrangement of the community's voting space. As the foregoing account illustrates, the symbolic constitution of the Swiss people takes place to a significant degree through representative claims in the lead up to and after the outcome of popular votes. While Richard Bellamy affirms the deliberative power of electoral politics, given the heightened activities of parties in engaging the electorate on a wide range of issues through their campaigns for a greater share of power, he is much less enthusiastic about the deliberative potential of referendums: ‘there is nothing particularly deliberative about referenda – its simplistic alternatives between “yes” and “no” on a question set by elites on the basis of partial consultation lies in stark contrast to the deliberative process of electoral politics’.Footnote ⁴⁷ To this we can say three things from the Swiss point of view.

First, just like general election campaigns, public discourse surrounding direct democratic votes is largely driven by Switzerland's national party system. According to Kriesi, the salience of a vote has a substantial impact on voter turnout and this, in turn, is itself greatly influenced by the intensity of the campaign leading up to the vote.Footnote ⁴⁸ As suggested from voter turnout, averaging at 42 per cent, referendum campaigns can deliberatively engage between 30 and 79 per cent of the electorate.Footnote ⁴⁹ Secondly, the robustness of a voting space makes a great deal of difference to the deliberative quality of referendums. While we may be unimpressed with the discursive impact of referendums in the Irish context, for example, which are relatively rare and in recent times most often about issues of European integration, the large number and diversity of national direct democratic votes each year in Switzerland makes the cumulative intensity and range of deliberation involved in referendums much more formidable. Thirdly, as already mentioned, more than half of Swiss direct democratic votes have a strong bottom-up dimension. On the one hand, the procedure for collecting signatures generates deliberative ripples in the public sphere by raising awareness of the issue among a significant number of citizens who must consider whether or not to add their name as a signatory.Footnote ⁵⁰ Iwan Barankay et al. convincingly argue for a strong positive correlation between the number of signatures that needs to be collected and citizen awareness, which ultimately translates to higher voter turnout.Footnote ⁵¹ Though their data is based on bottom-up referendums held at the cantonal level, there is good reason to believe that the correlation also holds for the national stage. On the other hand, according to Leonard Neidhart, Switzerland's bottom-up variety of direct democracy hangs like a sword of Damocles over the political system which has served to turn Switzerland into a negotiation democracy.Footnote ⁵² That is to say, the ‘partial consultation’ referred to by Bellamy is particularly noteworthy in Switzerland, because all those who could credibly garner the support required for a referendum on a particular issue are invited to participate in the pre-parliamentary stages of the decision-making process so that the vote might be avoided.

At this point, it is finally possible to see how multilingual Switzerland undermines the universality of Van Parijs's democratic challenge. Crucially, the collective intentionality of Swiss nationhood forged in the minds of citizens with the help of general elections, referendums and the national party system has the effect of promoting a multifaceted phenomenon commonly referred to in studies of public spheres as discursive integration. This term refers to the reality that the greater the comparability of discourse between public spheres the less distance there is between them in terms of what they can be said to have in common. There are both vertical and horizontal aspects to discursive integration, the former referring to the interaction between federal politics and the respective public spheres and the latter concerning the responsiveness or openness of public spheres to one another's discourse.

Considering the vertical dimension, what gives Switzerland's public spheres a great deal in common and prevents them from drifting too far apart is that their discourses are relatively synchronized by the national voting space, in the sense that citizens belonging to distinct linguistic communities are asked to consider and decide upon the same issues at the same time. The national structure of the party system, dominated by four traditional ruling parties, helps to ensure that the way in which the issues are articulated do not radically differ from one public sphere to another. As Anke Tresch has demonstrated in her comparative media analysis of debate in the German-speaking and French-speaking public spheres on referendums held in 2001 and 2002, there are few systematic differences in the discourse of the two linguistic communities, so that they can be said to operate within ‘a shared system of meaning’.Footnote ⁵³ Given the national frame of important political issues, it is unsurprising that the different public spheres are somewhat responsive to one another. This horizontal engagement, however, only marginally contributes to discursive integration, according to Tresch, since it is largely limited to actors from the different public spheres critiquing the national government rather than engaging each other in dialogue.Footnote ⁵⁴

In short, Switzerland falsifies the universality of Van Parijs's democratic challenge primarily because its unified and robust voting space helps to sustain and reinforce a mono-national consciousness that keeps the public spheres together in one complex system of co-operation. Certainly Switzerland does not measure up to the ideal-type ‘demos’, which we might have in mind when using this term, but for this we should not exclude it from such a categorization since it constitutes its own unique arena of political deliberation and mobilization. Yet the democratic challenge may still hold true trans-nationally. In what follows, I elucidate the extent to which the European Union (EU) could in principle follow the Swiss example if it were to become a demos capable of redistributive justice.

Must Europe Be Swiss?

Before we can establish what Europe might be able to learn from the Swiss case, it is necessary to define the precise sense in which they can be sensibly compared. Following Stein Rokkan's enthusiastic affirmation for the validity of the comparison, I submit that Switzerland can in many ways be seen as a ‘microcosm of Europe’.Footnote ⁵⁵ Most obviously, both Switzerland and the EU are multilevel governing systems with claims to democratic legitimacy that must deal with the problem of linguistic diversity. Furthermore, they are both highly decentralized in the implementation of legislation enacted by the centre. In each case a big reason for sustained decentralization is the need to respect the sovereign identities of sub-units – cantons in Switzerland and nation-states in Europe.Footnote ⁵⁶

Rainer Bauböck warns us, however, that even if we could define a general condition for democratic federal states we could not simply apply the same criteria to a supranational entity that is not itself a federal state.Footnote ⁵⁷ Perhaps the most significant difference between the cases is that Switzerland's twenty-six sub-units are not nations but cantons whose sovereignty is constitutionally conditioned by the federal state, whereas the EU can be more accurately conceived as a pooling of twenty-seven national sovereignties such that final authority ultimately remains with the member states. Another key difference to note in the present context is that, with twenty-three official languages, the EU has to manage a far greater measure of linguistic diversity than Switzerland. Despite these differences, and there are certainly more, we must not conclude that the EU cannot be compared to or learn from the Swiss case. According to Peter Mair, while Europe is not a federal state it is a ‘political system’ and it is at this level of generality that we can meaningfully compare it with nation-states.Footnote ⁵⁸ Erik Oddvar Eriksen and John Erik Fossum follow just such an approach in conceptually distinguishing between the ideas of government and state where the former is a system of legitimation principles for political power and the latter a particular historic organizational form in which these principles have sometimes been effectively instantiated.Footnote ⁵⁹ A model of comparison along these lines, where principles of representation and democracy are at stake independently of their institutional form, should be sufficient to avoid the pitfalls of methodological nationalism.

Without going into detail Van Parijs simply states that the only way in which Europe's democratic deficit can be overcome is through the adoption of English as a lingua franca.Footnote ⁶⁰ From other comments we can infer his position: the emergence of a European demos, made possible by a lingua franca, will give rise to the kinds of deliberation and mobilization required not only to bring about redistributive justice but also to improve democratic conditions in the EU. Contrary to Van Parijs, who believes that the democratic challenge must be met if Europe's legitimacy problems are to be resolved, my contention is that Europe can address its democratic deficit by following Switzerland in undermining the democratic challenge. In fact, the desire among citizens for and the political feasibility of extending redistributive powers to the EU would seem to depend significantly upon the prior establishment of its democratic legitimacy. This is not to deny that the EU has developed some important democratic qualities. Crucially, it has a parliament that is elected by the people of Europe every five years, which, in a kind of bicameral arrangement analogous to the workings of the Swiss Federal Assembly, mostly operates in conjunction with the Council of Ministers that consists of rotating representatives who hold ministerial positions in the member states. Similar to the election of the Swiss Federal Council by the Federal Assembly, both the European Parliament (EP) and the Council elect the European Commission, which is the main executive arm of the EU, composed of as many Commissioners as there are Member States.Footnote ⁶¹

Europe's democratic deficit is often characterized by the fact that (a) the EP has little control over the European Commission once the latter's members are elected, (b) aside from well spaced-out popular elections to the EP, there is almost no other participation by the citizens of Europe in EU decision making, and (c), in so far as there is a European public sphere, it is not yet sufficiently developed to make political debate visible to or inclusive of European citizens. Those who would deny the seriousness of Europe's democratic deficit, thus characterized, have argued that Europe's legitimacy should not be based on democratic inputs from citizens but rather on the effectiveness and fairness of legislative outputs,Footnote ⁶² which have to do not with redistributive but regulative issues that can be effectively decided only by experts capable of understanding the logic of good regulation.Footnote ⁶³ Adding to this argument, Andrew Moravcsik suggests that it is hardly necessary to hold the EU any more democratically accountable since the member states do not demand such standards of their own regulatory institutions.Footnote ⁶⁴

Bellamy rightly objects to these positions by maintaining that regulation is not normatively neutral and so cannot simply be entrusted to technical experts, whose deliberative quality is always under threat from their susceptibility to group thinking that may lead to an uncritical ‘technocratic consensus’.Footnote ⁶⁵ The normative aspect of regulative decisions is all the more difficult to deny in the European context when we consider the impact that regulation designed to deal with the euro currency crisis has had on national redistributive regimes since 2010. Bellamy also points to the fact that, in domestic politics, regulative bodies are entrenched in majoritarian electoral systems so that they cannot escape the influence and monitoring of elected politicians and the media. Europe, he claims, lacks the electoral and media structures required to keep its institutions in check.Footnote ⁶⁶ Accepting the reality of Europe's democratic deficit as formulated above, the question is what Europe's voting space would need to look like if it were to establish itself as a legitimate political system in the absence of a lingua franca, one that could make the development of trans-national redistributive capacities a political possibility.

Given popular elections to the EP, we can speak in a very weak sense of a unified European voting space. It is weak because this is the only outlet for European voters and there are no European-wide parties that could serve to reinforce this unity. Luciano Bardi et al. insist that so long as there is no direct competition for executive office in European elections there is no incentive for a European party system to form. Maintaining the status quo, they aver, means that parties operating at the European level will continue primarily to identify with and to frame their pattern of competition within their own national contexts.Footnote ⁶⁷ The problem, in other words, is that without executive competition European parties do not require the endorsement of voters outside their own member state and so have no reason to appeal to other European electorates.Footnote ⁶⁸ On this basis we can conclude that it is only by incentivizing the formation of a European party system, making the European Commission fully accountable to the Parliament, similar to the way in which Switzerland's Federal Council is answerable to the Federal Assembly, that the unity of Europe's voting space can be strengthened and the first aspect of the democratic deficit solved.

Much like that of Switzerland, however, a European-wide party system would inevitably be fragmented by the number of parties in Parliament and the decentralization of the parties themselves.Footnote ⁶⁹ Mair and Jacques Thomassen claim that while there is no European-wide party system, the left/right cleavage dominant in most EU member states has replicated itself as the chief cleavage on the European level.Footnote ⁷⁰ If this is correct, then the development of a European-wide party system would already seem to be in a good position to help undermine the democratic challenge for it would be based not on national or linguistic tensions that divide member states but on ideological cleavages that unite them. A European-wide party system, directly connected to the locus of power, is necessary for opening up the space in which credible representative claims about ‘the people of Europe’ can be made. To clarify this, trans-national parties and their representatives will be required to claim to represent the people of Europe convincingly if they are to receive votes and support from across the continent. This is likely to have the effect of strengthening the collective intentionality of European citizens by confronting them with various articulations of the European project on which they have to vote as part of that very project.

There may be a variety of ways in which citizen participation in EU decision making can be improved in order to address the second aspect of the democratic deficit. One way is the introduction of European-wide referendums. Instruments of direct democracy would further open up European politics to citizens while potentially creating, or at least emphasizing, multiple majorities such that divisive national or linguistic cleavages do not emerge as pervasive or in an especially egregious manner. Forms of bottom-up direct democracy would seem important for sustaining the volume of votes, since, as the Swiss case indicates, governing bodies will tend to avoid referendums where possible so that the larger share will probably need to be called by citizen mandate.Footnote ⁷¹ As my analysis of the Swiss case would suggest, regular referendums would be expected to contribute towards forging a European consciousness by making people frequently aware of themselves as participating in an exercise of collective will formation. Also worth noting is the potential for the use of direct democracy to become a marker of European identity that further distinguishes it from ‘others’, such as the United States and China, where direct democracy is absent from national politics. As already noted, national folklore and a sense of shared history contributes to Swiss identity formation. Such appeals to a common heritage are rarely invoked with regard to Europe and appear to have low resonance among citizens. Plausibly, however, as EU politics becomes more participatory and contributes to forming a deeper European identity, citizens will be more receptive to those who would follow Winston Churchill and loosely appeal to a ‘common inheritance’ distinctly European.Footnote ⁷²

Considering the magnitude of Europe's linguistic diversity and its associated nationalisms, could a unified and robust voting space realistically produce results that are at least structurally similar to those we find in the Swiss example? While Kriesi and Trechsel sing the praises of Switzerland as a potential model for the federal development of the EU, they declare that it falls short as a viable model for dealing with Europe's linguistic diversity. In less radical terms than Van Parijs, they advocate the need for a strong degree of linguistic homogenization in Europe.Footnote ⁷³ Contrary to this line of thought, Habermas contends that a supra-national public sphere united by a single language is not necessary for the viability of European democracy. My concentration on the Swiss case can be seen as an attempt to work out and expand upon his idea that a common European discourse must consist in the trans-nationalization of existing linguistically distinct public spheres such that they focus upon the same topics at the same time and ultimately become more responsive to one another.Footnote ⁷⁴

Several cross-country studies of media content have demonstrated that there is in fact already a significant European communicative space. According to Hans-Jörg Trenz's analysis, one out of every five articles in high quality European newspapers reports on a European issue, making the EU somewhat visible to individual citizens.Footnote ⁷⁵ Stefanie Sifft et al. nuance this point, reporting that Europe most often becomes the focus of media attention only after decisions have been made and when they begin to affect domestic affairs.Footnote ⁷⁶ In other words, the visibility problem is not so much with recognizing European issues and institutions but rather the escape of EU decision-making processes from journalistic monitoring. Far more serious than visibility, claims Paul Statham, is the issue of inclusion. Using data collected as part of the Europub project, he indicates that the vast majority of European media reporting concerns claims made by national and EU executives, to the regular exclusion of claims voiced by other political actors and especially civil society.Footnote ⁷⁷ Despite problems of visibility and inclusion, there are encouraging levels of vertical and horizontal communication in Europe. This is particularly true for vertical communication which has doubled from 1982 to 2003, with less impressive levels of horizontal communication remaining stable over this period.Footnote ⁷⁸ The question these analyses throw up for our current investigation is how could a voting space that is unified and robust strengthen vertical and horizontal communication to the degree that it would address the third aspect of the democratic deficit by making European politics more visible and inclusive?

Consider the contention that political power and issue salience are the main determinants of public discourse, in the sense that those who can exercise the most political power on what is considered to be an important issue by a political community will receive most attention from that community. Vertically, we find support for this in the above-mentioned observation that national and EU executives dominate European discourse in national public spheres, since it is these actors who tend to possess the most power in European affairs. Horizontally, we must consider that political power is roughly proportional to population size in the EU and, on this basis, we can say that the most powerful public spheres will generally be the main object of focus of all the others. Indeed, media analysis corroborates this point by showing that France, Germany and Britain receive by far the most attention from less powerful public spheres.Footnote ⁷⁹ Lacking a lingua franca, we should not expect a unified and robust voting space to forge a common European discourse, at least not in a way that approaches the level of commonality we often find in unilingual federal states. Rather, what such a voting space stands to contribute greatly towards achieving is the progressive maturation of both vertical and horizontal trends of communication.

On this view, the vertical dimension is at the heart of discursive integration in so far as it regularly synchronizes national public spheres by providing them with a common object of discourse. We already see a similar phenomenon occurring in weak fashion as every now and then a salient issue, usually dictated by the EU policy agenda, will be addressed at the same time by national media systems.Footnote ⁸⁰ The kind of voting space I am advocating for Europe synchronizes the public spheres in a much more substantive way by placing a greater share of power in the hands of citizens than they currently hold. Wielding this power would be likely to have the double impact of generating a much greater appetite among citizens for European politics, not just when decisions begin to affect the domestic realm but as the decisions are made, while motivating European-wide parties in competition for executive office to actively engage and inform citizens. In short, citizen power in a context of ideologically-based party competition should not only make European political debate more inclusive but also increase the visibility of European institutions within each public sphere, entrenching them in those kinds of electoral and media structures that Bellamy considers to be currently lacking. Moreover, much as they do in Switzerland, referendums with a bottom-up dimension would be likely to open up the political system and make public debate even more inclusive by empowering a range of alternative political and civil society actors to launch signature-collecting campaigns that would mandate referendums on particular issues.

If the responsiveness of Switzerland's public spheres is anything to go by we can confidently assume, slightly dashing Habermas's high hopes,Footnote ⁸¹ that the horizontal dimension would make only modest contributions towards discursive integration in Europe. While the most powerful public spheres will receive greatest attention, we can reasonably hypothesize that there would be many departures from this norm. First, due to institutional arrangements or the contingency of political events, there will probably be occasions on which different member states punch above their weight by holding an unusual amount of political power at a certain time and on a specific issue.Footnote ⁸² Secondly, perceived issue salience would be likely to differ between national contexts so that the responsiveness of one public sphere to another would depend on how they stood in relation to each other on an issue of mutual concern.Footnote ⁸³ Thirdly, because of factors like geographic, cultural or linguistic proximity, certain public spheres would be more likely to pay attention to one another's discourse even if political power or issue salience were not major factors. In the light of these considerations, we can see that Europe's linguistic diversity brings a certain complexity to the interaction between public spheres. Effectively, a unified and robust voting space would mean that different public spheres would be disproportionately and intermittently responsive to one another. What is important, however, is that a European-wide party system would be capable of managing both vertical and horizontal aspects of discursive integration, such that distinct public spheres could approach a shared system of meaning around a common political object.

By synchronizing the public spheres vertically and motivating them horizontally in differential fashion to take an active interest in each other's discourse, it seems that the deliberation and mobilization required for a well-functioning democratic demos need not be lacking, where we find a voting space that is unified and robust.Footnote ⁸⁴ In saying this, we must remember that forms of deliberation and mobilization are usually generated by the representative claims of political elites or leaders of social movements. In a trans-national context a lingua franca becomes more important for efficient co-operation between these crucial political actors. Yet this admission does not entail that there is anything unworkable about a highly multilingual EP or administration much like that which is in operation today. Indeed, Van Parijs considers such arrangements to be not only workable but also necessary so long as they are required to make the political process understandable to citizens from different linguistic backgrounds.Footnote ⁸⁵ What would be difficult to justify, and here I agree, are the fiscal and efficiency costs of a multilingual European political system when a lingua franca had fully emerged, so that it was in fact available for trans-national deliberation and mobilization.

Towards A Redistributive Europe

How a unified and robust voting space might come about and ultimately work in the European context could certainly do with greater elaboration than I have been able to give it here. Nevertheless, this article should not close without dealing with the objection that, however far my account goes towards solving Europe's democratic deficit, it does not generate the kind of solidarity required for redistributive justice. Recall that Van Parijs's argument for a lingua franca has a first ethical contagion step after which follows a second political feasibility step that amounts to the democratic challenge (see the introduction to this article). He believes that before a lingua franca can make a demos capable of redistributive justice possible it must be given time to forge a trans-national reflective equilibrium on principles of justice by allowing for relatively fluid cross-border communication. His account relies on two questionable assumptions: that the people who are worse off in one country will interact with the better off in another and that trans-national interlocutors will be in some way morally predisposed towards extending their sense of justice towards one another's nations. On the one hand, there needs to be a reason for communication and there is little to suggest that the better and worse off from different national cultures will normally interact. On the other hand, like any contagion, there will be people who are more or less susceptible to infection so that how widely and deeply ethical contagion spreads throughout Europe will be contingent upon the attitudes with which the average adult approaches situations of international engagement. It is unclear, on Van Parijs's account, why we should put faith in the openness of European citizens to one another's moral demands.Footnote ⁸⁶

My conception of a unified and robust voting space for Europe adds a prior stage to Van Parijs's account and thereby makes his two-step process towards trans-national redistributive justice more plausible. On my account, the EU must first make great strides in resolving its current democratic deficit. A likely consequence of doing this in the way I have suggested is a stronger European identity forged through the regular act of voting in common and the synchronicity and responsiveness of national public spheres to one another. Only through the creation of such a European identity, I suggest, will individuals become more predisposed to the moral demands of their fellow European citizens so that ethical contagion has a better chance of taking hold. Highly differentiated power relations, not to mention the length of EU membership and differences in national cultures, means that the collective intentionality uniting the citizens of Europe will develop even more unevenly than it has in Switzerland and other nation-states. Nevertheless, if I have characterized the basic logic of a unified and robust voting space correctly, there is every possibility that this political consciousness can become integrated and deep enough to make the transfer of significant redistributive competences to the EU politically feasible. Certainly, a lingua franca would make the creation of a stronger European voting space easier and would be a valuable tool in deepening cross-border solidarity. But from this we cannot conclude that trans-national justice or the demos from which it must emerge are dependent upon the achievement of English as a lingua franca.