4.3.1. Representation

In its original ancient Athenian form, democracy was of course literally what its name suggests—the rule (or power, kratos) of the people (the demos). Today known as direct democracy, it was a form of self-government in which citizens participated in making decisions in a political forum without the intermediation of elected representatives. Much subsequent democratic rhetoric has appealed to the same notions of popular rule; for example, Abraham Lincoln’s mantra from the Gettysburg Address describing a “government of the people, by the people, for the people.”

Yet neither in Athens, nor in any subsequent democratic society of large scale, have all the people really participated in government. For example only adult male citizens of Athens and surrounding Attica were permitted to participate in its assembly, which comprised less than a fifth of its resident population.[1] Such a society could hardly be called a democracy today, though by the same token Dahl refuses to grant that appellation to modern liberal states either, preferring to describe them as “polyarchies.”[2]

Dahl’s conception of the appropriate constitution of the demos is more inclusive than that of the Athenians, though he would still exclude transients and those shown to be mentally defective.[3] Alternative formulations focus more on the principle of consent; for example that the people entitled to make a law ought to be those who are subject to it,[4] or those who are significantly affected by it,[5] or all but those who have no legitimate material interest in it.[6] Direct versus representative

There are a few isolated cases in which the demos so constituted does still act directly to some extent today. Examples include the initiative (by which the Parliament can be required by citizens to consider a bill), the referendum (by which the people can pass a bill into law or repeal an existing law), the recall (by which they can remove an official from office) and the plebiscite (by which they can make constitutional changes).[7]

The most significant present-day example of direct (or plebescitary) democracy at the national level is in Switzerland, where a petition signed by 1% of the electorate may call for a referendum on an issue currently before the Parliament, or signed by 2% of voters if the issue is not yet before the Parliament.[8] The referendum itself in most cases passes with a simple majority of those voting, though at the national level there must also be a majority of states in which a majority of votes were in favour of the proposal.

California has had since 1911 a system of a “ballot propositions” similar to the Swiss system, save that the required percentages of registered voters to put forward a proposed statute or constitutional amendment are 5% and 8% respectively, and that only a simple majority is required at referendum in either case. Many other states of the United States have adopted similar procedures.[9] Although there is no equivalent at the national level, at the local government level, by-laws proposed in New England are commonly enacted by a Town Meeting open to all registered voters.[10]

Direct democracy is of course also seen in the private sector and in civil society, although more often in smaller, grass roots organisations. The clearest example of direct democracy in action in the private sector is at general meetings of public companies, which in countries including Australia are required to be open to all shareholders of the company.[11]

In civil society, it is seen at levels ranging from the local bridge club, up to the scale of political parties and trade unions. Sweden is particularly notable here, with political parties at both the local[12] and national levels,[13] as well as a large trade union,[14] organised along direct democratic lines.

However by far the dominant form of democratic rule is representative democracy, which rather than allowing the demos the right to rule directly, grants it the power and responsibility to delegate skilled specialist representatives to rule on its behalf. Seminal liberal scholar John Stuart Mill was amongst those who saw representative democracy as necessary to overcome not only the cost and logistical difficulty of implementing direct democracy on a large scale, but also the disinterest and indeed the incapacity of many citizens to act directly.[15] In more recent times, Dahl has agreed, contending “that representative government by elites is appropriate and that direct (as opposed to indirect through voting) participation is unworkable in the modern bureaucratic state.”[16] Filter versus mirror

There is however a division between the populist conception of representative democracy represented to some extent by Dahl, and the elitist conception of Mill. On the former account, representative democracy is favoured over direct democracy mainly as a concession to practicality. The peoples’ representatives are expected to act as a mirror of the community’s views, reflecting the views they would express in person if given the opportunity.

The alternative conception is of a representative assembly as a form of protection against ill-informed populism, including the tyranny of the majority.[17] Its function is to compensate for the deficiencies of direct democracy, such as the phenomenon whereby broad but shallow interests will always be trumped by narrow and deep interests pushed by organised groups.[18]

This division of opinion has a long heritage; it created a schism during the Constitutional Convention that drafted the United States Constitution,[19] and continues today between those who conceive of government as the agent of the people, and those who see it as separate from the people and potentially acting against its collective wishes.[20]

Whilst it might be assumed that the liberal approach is the former, brief mention of insights from from two quite different theoretical perspectives will suffice to demonstrate that this is not the case, and that the pure direct democratic, and populist representative democratic models are deficient from a liberal standpoint. The root problem from which both theoretical insights stem lies in the difficulty of selecting a voting system which precludes the majority, as determined by the aggregation of individual votes, from tyrannising minorities and thereby compromising the democratic principle.

The first theoretical perspective is Dworkin’s observation that the preferences of voters expressed in a democratic forum will tend to include preferences as to distributions of benefits that other voters should receive. These so-called “external” preferences are not entitled to be given any weight in a democratic calculus which attempts to aggregate individual preferences in order to maximize overall welfare.[21] An example of this problem is that the community may express its preferences through the ballot box as to the rights that should be accorded to (or withheld from) homosexual couples, whereas it is only the preferences of the couples themselves that should be counted on that issue.[22]

The second perspective comes from economist Kenneth Arrow, who has demonstrated that it is impossible even in theory to construct a democratic voting system that simultaneously satisfies all the criteria that might be considered necessary to produce an outcome fairly representing the voters’ collective best interests.[23]

Amartya Sen extended Arrow’s impossibility theorem to demonstrate a further and even more fundamental shortcoming of democracy for the liberal, which has come to be known as the liberal paradox: that it is impossible to devise a voting system that both upholds individual rights, and results in a Pareto optimal outcome. Put another way, if individual liberty is to be upheld, then the outcome of any voting system will always be inferior to some alternative that all voters would have been happier with.[24]

These observations of Dworkin, Arrow and Sen illustrate just some of the significant shortcomings with all democratic voting systems in expressing the true preferences of the demos (and there are further practical shortcomings not even mentioned, such as those of strategic voting and institutional effects).[25] Whilst they do not detract from the democratic principle, they do demonstrate that its implications extend well beyond a system of voting.

They also give licence to elected representatives to pursue the democratic ideal beyond the expressed preferences of their constituents. Even to state this fact immediately heralds its risks. Whilst further thoughts will be offered in the succeeding section on transparency and accountability, for now it can be foreshadowed that any representative democratic authority must “be held in check by institutional constraints that guarantee turnover in government and provide dissenters with many opportunities for political veto.”[26] Procedural versus substantive

Another perspective on the observed procedural shortcomings of representative democratic government (or governance) comes from the proponents of what can be called substantive democracy. What distinguishes procedural democracy from substantive democracy is a difference in focus: an institutional focus in the former case, and a social focus in the latter. Where the procedural democrat may be satisfied that democratic institutions and procedures are in place, the substantive democrat looks behind them to examine outcomes: does a particular system of governance actually reflect the aggregated preferences of the electorate, and if not, why not?

In practice, what this means is that procedural democracy is concerned mainly with the establishment and maintenance of a free and fair electoral system, whereas the concern of substantive democracy is broader and encompasses the maintenance of a society in which electors are, and remain, free to form and change their preferences.[27] The liberal, therefore, is a substantive democrat, who requires of the state not only a fair electoral system, but also to maintain the bedrock upon which that system stands. Such a liberal scholar is Dahl, who posits five criteria that a democratic polity should satisfy (most of which will fall, or have fallen, for separate discussion within the four slightly broader headings of this section on democracy):

The satisfaction of these criteria in turn requires the state to ensure that their preconditions are safeguarded even against infringement by the state itself (as in a protective democracy). The first of two related ways in which this can be done is by recognising these preconditions as rights, and there are in turn two main categories into which such rights could be classed.

The first category of rights required to meet the criteria of a democratic society are the basic civil and political rights of individual autonomy common to liberal scholars from Kant to Nozick, that Habermas describes as the “right to the greatest possible measure of equal individual liberties,”[29] and with which he includes the right to freedom of political association and due process. These are the rights necessary for even procedural democracy to function as designed, and which international law recognises through the International Covenant on Civil and Political Rights.

In the second category, and more peculiar to substantive democrats, are what Dahl refers to as rights to support the agency of individuals,[30] and that Habermas describes as rights to the basic living conditions that are necessary in order for the citizens to exercise their other rights.[31] These rights, which include the right to a basic standard of living and to an elementary education,[32] are amongst those found in the International Covenant on Economic, Social and Cultural Rights.[33]

Deontological liberalism has long recognised such fundamental pre-legal rights (at least of the first category) as being sacrosanct against the state, though such priority is more difficult to invoke in terms of teleological liberalism (and does raise difficult questions about how conflicting rights claims are to be resolved).[34]

A second and complementary way in which to safeguard the preconditions of democratic governance against infringement by the state is to elevate them to constitutional level. This is the manner in which the political institutions of representative democracy are generally enshrined in law, including the separation of powers and the rule of law, both of which will be discussed in more detail in the section on transparency and accountability below.

However a constitution can also be used to enshrine rights, as Kant for example observed,[35] (which avoids the difficulty of the recognition of pre-legal rights and duties in teleological liberal theory, since a constitution is still law, though its most fundamental example). Rawls identifies the purpose of constitutional rights as preserving or mirroring the justice of the original position, a position of equality and fair representation from which common principles are identified that are to the mutual benefit of all.[36] For Dworkin on the other hand, the purpose of rights is to compensate for the deficiencies of procedural democracy described above.[37]

Habermas, too, sees the place of rights as being constitutional principles, generated through a public discourse that constitutes its citizens as legal subjects, whereafter they are constrained from determining discursively the form that their discursive creation of other law will take. Thus for Habermas, “the requirement of legally institutionalizing self-legislation can be fulfilled only with the help of a code that simultaneously implies the guarantee of actionable individual liberties.”[38]

In summary of the last three subsections, then, it has been found that liberal democratic governance requires the establishment of accountable and transparent institutions that allow for the public to effectively participate as authors of the laws to which they are subject, subject to various superadded guarantees such as the protection of civil and political, and probably some degree of economic and social rights of individuals, through constitutional and international law. Transnational democracy

Unfortunately however, even this is not enough in the transnationally interconnected world of the new medieval era, since democracy might well be protected by such measures within the state, and yet still be impeded by forces from outside the state such as other states or transnational institutions (including government networks, civil society and the private sector),[39] which take roles in governance without the concomitant constitutional and international legal responsibilities of states.[40] Since decisions made in such fora may affect individuals across diverse territorial boundaries, it becomes difficult to determine whose consent to them should be required, and how that consent could be measured.[41]

The study of transnational democracy is directed towards this dilemma, and puts the normative position that it is necessary to extend democratic guarantees of autonomy on a transnational basis at all applicable levels of governance. There are however three broad approaches to achieving this end.

The first of these is liberal institutionalism (of which regime theory is a special case).[42] For the liberal institutionalist, transnational democracy is furthered through pluralism of transnational governance and by increasing the transparency of the operations of each of the actors, which approximates to the position put forward at Section 3.5. Fukuyama has describes it as a vision of “a world populated by a large number of overlapping and sometimes competitive international institutions, what can be labelled multi-multilateralism.”[43] Civil society has a central role to play here both instrumentally (in facilitating this process) and normatively (as a participant in it).[44]

A limitation of the liberal institutional approach is that its normative programme does not include any mechanism to ensure that there is an overall balance of views represented amongst the various actors participating in transnational public policy governance. For example, an issue area such as IPR may be dominated by private sector rights holders, governments beholden to them, civil society organisations representing them, and WIPO, leaving alternate perspectives under-represented.

This is one of the insights that may be drawn from the second approach to transnational democracy, which can be termed radical democratic pluralism[45] or radical plural democracy.[46] The radical democratic pluralist is critical of the ability of powerful interests to capture pluralistic governance processes, and in response urges direct action by grass roots social movements to achieve transnational governance outcomes through a variety of fora at different levels.[47]

This position is antagonistic to the dominance of the existing liberal world order even in its pluralistic form, so for example the radical democratic pluralist would likely not seek to work within the IGF, but to subvert it. From a liberal perspective this is the main shortcoming of the approach, along with the fact that although it aims to develop a structure of checks and balances to hold the hegemony of powerful interests to account, the legitimacy, transparency and accountability of these grass roots endeavours themselves remain, if anything, even more obscure than those of a governance network structured within the liberal paradigm.

The third transnational democratic approach is cosmopolitan democracy, which builds on liberal institutionalism in its vision of multi-layered governance, but has a more ambitious normative program. It proposes, in the words of its foremost scholar David Held, “the establishment of an international community of democratic states and societies committed to upholding a democratic public law both within and across their own boundaries: a cosmopolitan democratic community.”[48] This requires all transnational governance institutions, including private regimes such as the new law merchant, to be held to an overarching cosmopolitan legal framework that upholds accountability to liberal democratic norms such as the rule of law.[49]

In the long term, this will mean nothing less than the creation of a global parliament and an internationalised global legal system.[50] The short-term objectives of the cosmopolitan programme are not much less ambitious; including the creation of a second chamber of the UN, the reform of the UN Security Council, enhanced political regionalisation on the model of the EC, the creation of a new international Human Rights Court with compulsory jurisdiction, and the establishment of an international military force.[51]

It is to criticism of these high-flown ideals that cosmopolitan democracy is most vulnerable, with Keohane and Nye dramatically pointing out that “a cosmopolitan view that treats the globe as one constituency, implies the existence of a political community in which citizens of 198 states would be willing to be continually outvoted by a billion Chinese and a billion Indians.”[52]

On this account, cosmopolitanism falls victim to its own ambition. It encounters the same problems discussed at Section as to the extent to which diverse communities can form a single normative regime that also has adequate content to be of any value.[53] Keane, for example, doubts that even the principles of procedural democracy, that the cosmopolitan programme seeks to instil in all transnational governance institutions, are culturally neutral.[54]

These criticisms lose some of their cogency when the principles of cosmopolitan democracy are applied to a governance network such as the IGF rather than to the international system as a whole. In this narrower context, the preparedness of affected stakeholders to accept the democratic principle should be widespread enough that few will be excluded (and for those who are, the IGF exercises only soft power anyway, and is unable to enforce its output upon them).

On the other hand it is true that the range of substantive issues upon which it would then be possible for stakeholders to go on to decide by democratic means may be limited: cultural or ideological differences may either prevent decisions from being made at all, or if the decision-making method employed (such as a voting system) allows the forum to push through a majority decision, it may yet be defeated by the minority’s ability to boycott it. Granted that this provides an incentive for the forum to aim towards consensus, but if the only decisions that can thus be made are amorphous platitudes, it would still be as well if no decisions were made at all.

Whilst this criticism is relevant and will be addressed in detail later,[55] it does not foil the cosmopolitan democratic programme. Such cases are simply examples of those in which it is not appropriate for the decision in question to be dealt with at such a high level, and where it should fall back to be dealt with at a lower layer of governance in accordance with the principle of subsidiarity.

Held has developed tests of extensity, intensity, and comparative efficiency to determine whether it is most appropriate for governance to be exercised at a grass roots level (the corporation or city, for instance), the national level, or a higher, supranational or transnational level.[56] These three tests respectively examine how extensive is the range of people affected within and across borders, how intensely each particular group of people is affected, and whether a lower governance level would likely be ineffective, requiring it to be dealt with at a higher level.

The appropriate constituency in any given case is to be “defined according to the nature and scope of controversial transnational issues.”[57] In some cases it may be unavoidable that the closest fit is not a perfect one, and in others the best outcome may be for an issue to be dealt with in overlapping and competing ways at multiple layers of governance. This equates to Fukuyama’s “multi-multilateral” model, and illustrates the commonality of the core of the cosmopolitan democratic approach with that of liberal institutionalism.

In fact, leaving aside the most overreaching ambitions of cosmopolitan democracy and the postmodern excesses of the radical democratic pluralist programme, all three approaches affirm the importance of the democratic principle not just at the level of the state, but at all levels of transnational ordering, public and private,[58] extending beyond the state-centric demos, to the cosmos; that is, to all humankind.[59]



Ober, Josiah, Mass and Elite in Democratic Athens: Rhetoric, Ideology, and the Power of the People (1991), 128


Dahl, Robert, Who Governs?: Democracy and Power in the American City (1961)


Dahl, Robert, Democracy and its Critics (1989), 129


See Graham, Gordon, The Internet: A Philosophical Enquiry (1999), 72


Held, David & McGrew, Anthony, The Global Transformations Reader: An Introduction to the Globalization Debate (2000)


Gastil, John, Democracy in Small Groups (1993), 19


Clarke, Roger, The Internet and Democracy (2004)


Kobach, Kris W, The Referendum: Direct Democracy In Switzerland (1993).


Zimmerman, Joseph F, The Initiative: Citizen Law-Making (1999)


For example, this is required for any town of Massachusetts with a population of up to 6000, and is optional in larger towns. See Constitution of the Commonwealth of Massachusetts, 25 Oct 1780, article LXXXIX.


Corporations Act 2001 (Cth), part 2G.2


Demoex; see http://demoex.net/en/.


Aktiv Demokrati; see http://aktivdemokrati.se/.


Sveriges Arbetares Centralorganisation (Central Organization of the Workers of Sweden); see http://www.sac.se/en/.


Mill, John S, Considerations on Representative Government: Remarks on Bentham’s Philosophy (2000)


Dahl, Robert, Democracy and its Critics (1989)


Walton, D, Appeal to Popular Opinion (1999)


Olsen, M, The Logic of Collective Action: Public Goods and the Theory of Groups (1965)


Fishkin, J S, Virtual Democratic Possibilities: Prospects for Internet Democracy (2000), 2–10


Holden, B, Understanding Liberal Democracy (1998), 20


Dworkin, R, Taking Rights Seriously (1977), 277


As will be seen, Dworkin’s answer to this paradox is that it is the purpose of human rights to prevent most such external preferences from being given expression in a democratic political system.


Arrow, Kenneth J, Social Choice and Individual Values (1963)


Sen, Amartya, Liberty and Social Choice (1983)


Cox, Gary W, Making Votes Count: Strategic Coordination in the World’s Electoral Systems (1997)


Immergut, Ellen M, The Theoretical Core of New Institutionalism (1988), 14


Fox, Gregory H & Nolte, Georg, Intolerant Democracies (2000), 400–401, and see Section


Dahl, Robert, Democracy and its Critics (1989)


Habermas, Jürgen, Between Facts and Norms: Contributions to a Discourse Theory of Law and Democracy (1996), 122 (emphasis in original)


Dahl, Robert, Can International Organizations be Democratic?: A Skeptic’s View (1999); Dahl, Robert, Democracy and its Critics (1989) , 170


Habermas, Jürgen, Between Facts and Norms: Contributions to a Discourse Theory of Law and Democracy (1996), 123


See General Assembly of the United Nations, Universal Declaration of Human Rights (1948), Articles 25 and 26.


This second category of rights, which are in the nature of entitlements rather than freedoms, are more controversial than the first because their fulfilment is likely to require the redistribution of property (normally through taxation) in violation of the property rights of others: Nozick, Robert, Anarchy, State and Utopia (1974). But by the same token, the status of the right to private property, and the free market system which arises from it, is itself controversial. Some liberals point out that the right to private property may be distinguished from other fundamental rights in that its exercise is rival: Weale, Albert, The Limits of Democracy (1989) , 42. It is unnecessary for present purposes to enter this debate.


Held, David, Democracy and the Global Order: From the Modern State to Cosmopolitan Governance (1995), 217


Kant, Immanuel & Reiss, Hans, Kant: Political Writings (1991), 191


Rawls, John, A Theory of Justice (1971), 221–222


Dworkin, R, Taking Rights Seriously (1977), 277


Habermas, Jürgen, Between Facts and Norms: Contributions to a Discourse Theory of Law and Democracy (1996), 455 (emphasis in original).


Held, David, Democracy and the Global Order: From the Modern State to Cosmopolitan Governance (1995), 226; Held, David, Democracy and the New International Order (1995) , 99–100


Picciotto, Sol, Democratizing Globalism (2001), 342


Held, David, Democracy and the New International Order (1995), 102


McGrew, Anthony, Transnational Democracy (2002)


Fukuyama, Francis, America at the Crossroads: Democracy, Power, and the Neoconservative Legacy (2006), 158


McDougal, Myres S, Lasswell, Harold D, & Reisman, W M, The World Constitutive Process of Authoritative Decision (1981), cited in Charnovitz, Steve, Two Centuries of Participation, NGOs and International Governance (1997) , 271–273


McGrew, Anthony, Transnational Democracy (2002)


Hekman, Susan, Radical Plural Democracy: A New Theory For the Left? (1996)


McGrew, Anthony, Transnational Democracy (2002), 274


Held, David, Democracy and the Global Order: From the Modern State to Cosmopolitan Governance (1995), 229


Held, David, Democracy and the Global Order: From the Modern State to Cosmopolitan Governance (1995), 234; Held, David, Law of States, Law of Peoples (2002)


Held, David, Democracy and the New International Order (1995), 113


Held, David, Democracy and the New International Order (1995), 111; Held, David, Democracy and the Global Order: From the Modern State to Cosmopolitan Governance (1995) , 279–280


Keohane, Robert O & Nye, Joseph S, The Club Model of Multilateral Cooperation and Problems of Democratic Legitimacy (2000), 33.


McGrew, Anthony, Transnational Democracy (2002), 276


Keane, John, Global Civil Society? (2003), 124–125


To some extent in the following subsection, and again at Section


Held, David, Democracy and the Global Order: From the Modern State to Cosmopolitan Governance (1995), 234


Held, David, Models of Democracy (1996), 355


Held, David, Democracy and the Global Order: From the Modern State to Cosmopolitan Governance (1995), 252


Held, David, Principles of Cosmopolitan Order (2005)