3.1 Introduction

This volume is organized around the assumption that apart from Member States – in effect, member governments – there are other ‘public or private institutions’ that often play a part in international organizations.Footnote ¹ In illustrating the range of such participants, Samantha Besson cites ‘NGOs, multinational corporations, employer’s associations, trade unions, or patients’ associations’.Footnote ² And she also makes the point that international organizations (IOs) may themselves constitute non-governmental participants in higher-level organizations.

The focus of my chapter, like that of the present volume, is on how far things can be organized within IOs ‘to ensure sufficient democratic representation of all those they claim to bind legally, either directly or through their Member States’.Footnote ³ Those they bind in this way include most prominently the peoples of Member States. So how far can such peoples be democratically represented in IOs by bodies that form independently of the State or States from which their membership is drawn? I shall describe such agents or agencies as State-independent, or sometimes just as independent, bodies.

Peoples may exercise democratic control over an organization in a variety of ways that do not directly involve representation: for example, by constitutionally or otherwise requiring that it operate under certain constraints, that appointments be made according to strict procedures, that it offer broadly public-interest reasons for its decisions, and that it be subject to monitoring and auditing by an independent body. But the focus in this chapter is on representation only and I shall ignore the democratizing role of other factors like these.

The chapter is in three sections. It begins with an account and critique of two extreme views of how IOs may relate to the peoples they involve, where the difference turns on the part that State-independent actors should be able to play (Section 3.2). In a third, long section, it then presents and defends a series of general principles governing representation, arguing that they undermine each of these two views (Section 3.3). And then in the final section it shows that the principles support a distinct, third model of the role that independent actors might play in representing their peoples within IOs (Section 3.4).Footnote ⁴

3.2 Two Extreme Views

There are two extreme views of how IOs may meet representative constraints on their relationship to the peoples supporting them, one State-centred, the other individual-centred. Each agrees that the people of every Member State may be represented by those State-independent agents or agencies that participate in an IO but they take diverging views of how this can be achieved.

The first, State-centred view holds that it is only States that have the standing to represent their peoples within IOs and that any independent parties that operate there must play a role assigned to them by their respective States. They must be commissioned in that role by State authorities such as the legislature or executive. And they must be required to act on the terms laid down for them by such authorities.Footnote ⁵

The second, individual-centred view argues that while a people may certainly be represented by State appointees or authorities in an IO, it may also be represented, without State appointment or authorization, by independent actors. The idea is that IOs create networks of communication and collaboration across diverse peoples and that those networks should be allowed to vary in how they ensure the representation of peoples. One organization may recruit a people to its network by giving a place to appointees or authorities of their State, another by giving places to independent actors. The ideal projected is one in which States may rule over their peoples domestically – subject, presumably, to constraints of international law and the like – but those peoples may connect and order things in the global sphere within a variety of State-independent, decision-making networks.

These contrasting views may reflect different images of how to understand what a people amounts to: that is, the people of any particular State.Footnote ⁶ Thus, the picture that supports the second individual-centred approach, each people is just the population of those who live – presumably, on a relatively permanent, legally ratified basis – within a political territory. It is a collection of those individuals, considered in their separate identities, with their particular interests, not a body of individuals organized to perform via incorporation in a State, where the State is taken to be a corporate entity like a corporation or church or whatever. This picture of the peoples that IOs would serve makes perfect sense of why they may be represented by a variety of State and other agencies in the different domains of common interest engaged by such organizations.

According to the rival image that supports the State-centred approach to international representation, a people is not just an amalgam of individuals. The people is a set of agents organized within the corporate agency of the State, so that State authorities – executive, legislative, or judicial – speak for them on a range of fronts: in framing and applying the law; in pursuing those policies, including international policies, that are determined in accordance with the law; and in appointing the actors charged with implementing such policies. This picture is embraced on the State-centred view, regardless of whether the State is democratically structured or not. All that counts is that it be a matter of common awareness among the membership – and common acceptance, however reluctant – that the State authorities should play this role. According to the view envisaged, members of a non-democratic State will recognize that in order to enjoy the order and benefits of law-governed relationships among themselves, and with other peoples, they must be organized under a State and have no feasible choice but to live within the established regime.Footnote ⁷

Whatever the merits of these motivating images of the people of a State, neither of the competing views of international representation is particularly appealing. The State-centred view suggests that nothing may be determined by representatives beyond the boundaries of the State that has not been pre-authorized at home or that is not in line with fairly strict pre-authorized instructions. Thus, it would make the forum of an international organization into what Edmund Burke casts as merely ‘a congress of ambassadors’.Footnote ⁸ In such a congress, as Burke envisages it, participants would bargain with each other on the basis of their pre-existing, partisan interests, making only the minimal concessions to one another that are needed to get what they want. This is very different from the sort of organization in which members would focus on common problems and be willing to innovate in their search for solutions; it would fail to resemble anything like a deliberative assembly.Footnote ⁹

While that makes the State-centred view unappealing, quite a different flaw detracts from the appeal of the individual-centred rival. If there is nothing to enforce unity across the commitments of each people, then the corresponding State is liable to find itself obliged to answer to rival and conflicting demands. The State will be the only agency, representative of its people, that operates in different forums and if it is at the mercy of what other representatives of the people agree in one or another domain, there is no guarantee that it will be able to implement all the commitments made in the name of the people. It may be required to pursue a policy on the relaxation of border controls, for example, that is at odds with a policy of international cooperation in policing; or to adopt a commitment to lifting tariffs on certain goods that competes with a domestic commitment to support local industry.

These observations should lead us to think again about how to model the representative role that State-independent members of IOs may play. I think there is a third model that ought to attract us more and in order to motivate it I propose to return to some basic principles of representation – this unfortunately requires a resort to rather abstract theory – and show how they combine to give salience to the possibility that that model represents.

3.3 Some General Principles of Representation

It may be useful to begin by setting out the principles addressed here and to display the argumentative connections between them. We can then discuss each principle in somewhat more detail. The fifth principle, as we shall see, argues against the individual-centred of how a people can be represented in an IO but the sixth supports the claim that this does not commit us necessarily to adopting the State-centred alternative. In the final section, we show that it directs us rather to a third and more appealing model.

The principles I defend are these:

1. i) An agent or agency may represent individuals in a responsive or indicative mode;

2. ii) Whatever the mode, the represented must authorize the representative system;

3. iii) In any domain, the represented can authorize only one representative voice;

4. iv) A people must authorize their State as an all-domains, sovereign spokesperson;

5. v) And so the State must be their only representative, with control over other voices;

6. vi) But the State need not exercise this control actively, only in a standby manner.

3.4 An Alternative View of International Representation

4.1 Introduction

Two myths of statehood are foundational to the international system of representation as it is formally constituted. The first is that only States are international actors. The second is that all States are equal and should be treated accordingly. Both are obviously false. Many non-State actors – private organizations of all types, including non-governmental organizations (NGOs) and corporations, and sub-State and super-State governmental entities – clearly act internationally in important ways, leveraging their authority and power, making decisions and entering into agreements of consequence. And obviously as well States are unequal in any number of respects, including their economic wealth, military power, and technological capabilities. Nevertheless, the rules, processes, and institutions of the international system are designed as if these two assumptions and their corollaries were valid, as if States could be counted on to effectively govern their own territories and represent the diversity of their peoples, as if States alone ruled the world, and as if States participated equally in global governance. It is not surprising that a system built by States and for States would mythologize statehood and strictly and all-but-exclusively incorporate State-centric beliefs into its core formal rules, acting as if States are and should be the sole representatives at the international level.

What is true of the international system generally is true in particular of international organizations (IOs). IOs are created by, and composed of, States and, with rare exceptions (particularly the European Union in its areas of competence and employers and workers in organs of the International Labour Organization) only States. And those States are, with very rare exceptions, treated as equals in the founding documents of organizations. It follows that the rules of IOs, again with rare exceptions (such as voting in the Security Council and international financial institutions (IFIs)), put States on an equal footing when it comes to decision-making: one State, one vote, with all members eligible to vote in the organization’s plenary. It follows as well, in this statist framework, that all States should have an equal opportunity to take part in the organization’s activities: to participate and vote in the organization’s main committees, to be elected to non-plenary bodies within the organization, to have its nationals chosen for high-level positions, and to influence the organization’s agenda and policies. Like the primary myths of international representation, these understandings are untrue. Some States and certain NGOs have heightened influence within organizations that belies the statist paradigm. Nonetheless, it is not surprising that IOs, as the creatures of States, adhere to statist myths and so reserve actual decision-making to States, and regional groups composed of States, and purport on the whole to treat States equally.

Though the myths are well-known as myths, and have been since their creation, that does not diminish their strength in the formal systems of IOs. Indeed, it is a testament to the power and resilience of these myths, and to the control that States still retain, and insist on retaining, at the international level, that their evident falsity has not yet led to any significant changes in blackletter international institutional law and its formal mode of representation.

Even so, the myths are problematic if they are strictly adhered to in practice. Systems cannot operate effectively unless they are recognized as authoritative and unless they can achieve their desired results. The absence of non-State representation and democratic legitimacy in IOs undercuts the authority of organizations because it does not permit the incorporation of a broad range of voices. And the notion that all States are equal undercuts the ability of organizations to achieve results as it fails to recognize the role of power in international society. There is thus a gap – an ‘operational gap’ – between the myth world of IOs and the real world in which IOs exist. It is understood by those who work in, with, and around IOs that there need to be ways to bridge this gap if organizations are to succeed and retain legitimacy and authority.

The operational gap inherent in the formal mode of representation can be, and has been, elided in three distinct ways in order to enhance and achieve both the democratic legitimacy and functionality of organizations. One is through informal adaptations in the operation of organizations – the operational mode – that seek to reconcile the needs for authoritativeness, legitimacy, and effectiveness with a formal system that falls short. A second – the aspirational mode of representation – aims to do the same but through changes in the formal and operational modes. A space for advocacy and critique, not a method of representation itself, the aspirational mode is liminal, for success formalizes or operationalizes the reform’s content. And a third – the alternative mode – closes the operational gap by creating substitutes for IOs, such as informal groupings of States and organizations established under national law, that are not bound by the paradigmatic statist myths of the formal international system and, having been freed of those constraints, provide representation, including forms of democratic representation, in ways that are not fully possible within traditional IOs. Including the formal mode, which is the international system’s default, the practice of representation, as it has evolved, encompasses four distinct forms. From the standpoint of democratic legitimacy, the formal and operational modes are generally weak in their current manifestations, as they most often reflect international law’s State-centric assumptions and the influence of power. Those who advocate for greater democratic legitimacy in IOs therefore work most often within the aspirational frame as they attempt to promote and institute reforms.

This chapter reviews the four modes of representation – formal (Section 4.2), operational (Section 4.3), aspirational (Section 4.4), and alternative (Section 4.5). They are separate yet intertwined; they contend with yet depend on each other. Together they form the international system of representation. This is not simply descriptively true. The resort to, and the use of, the four modes is also optimal. All four are necessary, as for any successful system to evolve and maintain equilibrium there needs to be a variety of mechanisms available that allow for the balancing of competing values and the reconciliation of formal rules with the actualities of power and expectations of legitimacy, since alignment will never be perfect.Footnote ¹ Importantly, the four modes allow a range of entry points for democratic legitimacy in the representative practices of IOs, creating the possibility of change going forward (Section 4.6).

4.2 The Formal Mode

The formal mode of representation, with limited exceptions, codifies statist values through rules written into the constitutive treaties and resolutions of IOs and thus represents the greatest challenge to democratic legitimacy.Footnote ²

Representative principles in treaties appear in two ways: (1) in rules regarding the participation of States in decision-making bodies and (2) in rules regarding the election and appointment of individuals to formally non-political bodies, such as secretariats, courts, and expert panels. For the former, the principles largely arrange themselves by subject matter – political organizations, where sovereign equality is mostly paramount, and IFIs, commodities organizations, and similar subject matter specific institutions, where differential responsibilities are highlighted. For the latter, individuals are largely treated equally as to their nationality, but other personal characteristics or qualifications sometimes come into play explicitly, depending on the post.

In the constitutive documents of political organizations, such as the United Nations (UN) and regional institutions like the Organization of American States, the theory of the sovereign equality of States usually dominates. States are considered the electors, each one equal with the other, and all States (or, depending on the position, their individual nominees) are free to run for office. If one reviewed the documents of these organizations, as well as the plenary bodies of many non-political organizations, one would find much evidence to support this representational norm. The UN General Assembly (UNGA), where each Member State has a single vote regardless of the size of its population, territory, military, or economy, is the most obvious example, but there are many others.

Other principles appear in treaty form when the parties strongly agree on their relevance – most often in specialized organizations. For example, under the articles of agreement of IFIs, the members quotas, which are not equally distributed, (roughly) determine the number of votes allocated to each one. International commodities agreements divide votes equally between exporting and importing nations and allocate in the councils of such organizations individual States a subset of votes based upon the volume of their exports or imports. The Montreal Protocol requires that decisions of the Executive Committee of the Multilateral Fund be approved (in the absence of consensus) by two-thirds majorities of both developing and industrial countries.Footnote ³ The Convention on the Intergovernmental Maritime Consultative Organization (now the International Maritime Organization) allocates seats in the Organization’s Council according to the ‘interests’ of its members (for example, international shipping services or seaborne trade),Footnote ⁴ and the Convention on International Civil Aviation sets aside seats in the Council to ‘States of chief importance in air transport’ and States with the ‘largest contribution to the provision of facilities for international civil air navigation’.Footnote ⁵ In the Constitution of the International Labour Organization, ‘members of chief industrial importance’ are given representational preference over other States in the organization’s Governing Body,Footnote ⁶ and the Chemical Weapons Convention allocates seats in the Executive Council by region, designating a certain number for ‘States Parties with the most significant national chemical industry in the region’.Footnote ⁷ Some other agreements, such as the Statute of the International Tribunal of the Law of the Sea, provide regions with minimum representation in certain bodies.Footnote ⁸ Still others encourage, but do not require, electors to ‘give consideration’ to, pay ‘due regard’ to, or ‘bear in mind’ ‘equitable geographic distribution’. In all these organizations, the negotiators of the agreements decided that the representative principle in question (differential responsibilities or regionalism) was essential to the success of the institutions they were building and that the institution itself could not operate unless this principle was incorporated into the organization’s most fundamental document.

Ironically, what all these formal rules have in common is their flexibility. On the surface, they appear rigid – one State, one vote; the allocation of greater power to certain groups of States or regions; the requirement that candidates have certain qualities – but, in fact, these rules are designed as much to allow States to implement, or even avoid, their principles through lower-level decision-making.Footnote ⁹

IOs thus often establish rules regarding representation through resolutions and decisions. These resolutions usually pertain to appointments to other bodies (both subsidiary and coordinate) or to the election of the organization’s officers (such as chairs and presidents of sessions and committees). These rules implement and supplement provisions that appear in constituent documents, filling in their intended gaps and ambiguities.

Resolutions are often used to establish regional representation. Thus, the resolutions of the UNGA rigidly allocate the Assembly’s officers by region, including its presidency, vice presidency, and committee chairs. The General Assembly’s resolutions also divide, by region, positions on subsidiary bodies, funds, and programmes, such as the International Law Commission (ILC), the United Nations Commission on International Trade Law (UNCITRAL), the United Nations Environment Programme (UNEP), the United Nations Children’s Fund (UNICEF), the United Nations Development Programme (UNDP), and the World Food Programme (WFP), as well as positions on other principal UN organs, such as the Security Council and the Economic and Social Council (ECOSOC). ECOSOC resolutions allocate positions by region on its subsidiary bodies, the now defunct Human Rights Commission, the Commission for Social Development, and the Commission on the Status of Women, among others. In the Food and Agriculture Organization (FAO), resolutions distribute council seats by region. The General Conference Rules of Procedure of the United Nations Educational, Scientific and Cultural Organization, divide Member States into groups and give each group a set number of seats on the Executive Board. In the International Criminal Court, a resolution established a requirement that States Parties vote for a minimum number of judicial candidates per region, which number will vary according to judges from the region already sitting or elected.Footnote ¹⁰ A 1994 decision of the Congress of the Universal Postal Union distributed seats in the Council of Administration by region.Footnote ¹¹ And by resolution the States Parties to the UN Convention on the Law of the Sea have allocated seats on the Commission on the Limits of the Continental Shelf and the International Tribunal for the Law of the Sea by region as well.Footnote ¹²

The prevalence of regionalism at this level of decision-making should not surprise. As a representational principle, regionalism has (and has had for some time, particularly since decolonization) a level of legitimacy second only to the notion of sovereign equality. That stems, in part, from regionalism’s guarantee that all (not just some) parts of the world will be represented. But the attractiveness of regionalism results from its effect of decreasing the number of States in the pool for elections and appointments, hence increasing the likelihood of election for States that might not otherwise be able to compete. This dynamic is enhanced by the common practice among regional groups of putting forward agreed slates of candidates that are often based on an informal rotational system among their members. The latter phenomenon points to other features of regionalism: efficiency and compromise. Regionalism ‘reduces the potential for conflict in the selection of Members and gives all groups a more secure sense of representation’.Footnote ¹³ For these reasons, the regionalism principle is relatively easy to adopt in multilateral forums.

Democratic inroads in the formal mode manifest, albeit weakly, through techniques such as observer status that allow the access of non-State actors to IOs without providing them the decision rights accorded to States.Footnote ¹⁴ The UN Charter, for example, authorizes the ‘Economic and Social Council [to] make suitable arrangements for consultation with non-governmental organizations which are concerned with matters within its competence’.Footnote ¹⁵ Accordingly, ECOSOC has established principles for consultative relations, most recently in resolution 1996/31, that outline the criteria for consultative status and the forms of consultation with the Council and its subsidiary organs, including (with some limitations) the ability to propose agenda items, attendance at meetings, written statements, and oral presentations.Footnote ¹⁶ The World Health Organization (WHO) has adopted a ‘Framework of Engagement with Non-State Actors’ that establishes rules for ‘private sector entities’ and ‘international business associations’, and others for ‘nongovernmental organizations’.Footnote ¹⁷ The extent to which such participation can be understood as enhancing the democratic legitimacy of IOs will depend upon how these NGOs are characterized, the access they are provided, and the influence they wield in practice.Footnote ¹⁸

Altogether, the formal mode reflects State-centred international law at its apogee. The limited participation of non-State actors through formal mechanisms demonstrates simultaneously both the recognition of the need for a broader, more democratic representation and the strong impediments that exist for the achievement of that goal.

4.3 The Operational Mode

Formal rules of representation, because they are based on a statist myth, can only rarely recognize the differential power of States and the authority of non-State actors. The operational mode provides informal means that allow IOs to do both – but off the record and only in part. Representational practices in this mode take the form of informal agreements (that give greater say to powerful States), financing mechanisms (that give informal control over IO programmes to wealthy States and private foundations), and consultative processes such as meetings and conferences (that provide voice to non-governmental and sub-State actors).Footnote ¹⁹

Much of international decision-making is done through informal processes, particularly informal agreements, either one-off or long-term. In no area is this more apparent than in the realm of agreements concerning international representation. Because of their unofficial, typically unpublicized, all but hidden, and often evolving nature, such agreements are inherently difficult to identify and ascertain. Unsurprisingly, a search of official records uncovers little, if any, explicit mention of their existence, let alone a comprehensive compilation. But only with an understanding of the entire scope of these informal rules can one comprehend not only international representation as it is currently constructed.

The obvious place to begin an examination of informal agreements concerning representational issues is with IFIs. The most well-known is the agreement between the United States and the Europeans regarding the leadership of the International Monetary Fund (IMF) and the World Bank. But the two flagship IFIs are not alone in allocating posts by custom. The four major regional IFIs exhibit the same pattern of informal delegation of leadership selection to a particular country or countries. Thus, the President of the Asian Development Bank has always been a Japanese national, the presidency of the European Bank for Reconstruction and Development has always alternated between French and German nationals, the President of the Inter-American Development Bank has always been a Latin American national, and the presidency of the African Development Bank has rotated among ‘various [African] linguistic and geographic groups’.Footnote ²⁰ Further, related institutions, such as the Multilateral Investment Guarantee Agency (MIGA), where only Japanese nationals have occupied the highest post, have also informally specified the nationality of their top officials.

Though scholars and the media have focused on the top positions in the IFIs, the allocation of posts goes deeper. For instance, IFI vice presidencies are frequently slotted as well. Adding this layer to the analysis shows that the designation of top posts cannot be seen in isolation – they are often part of a package deal. Thus, while the Managing Director of the IMF is always a European national, the IMF’s First Deputy Managing Director is always a US national, chosen by the United States.

The pattern holds for the regional banks. Here, the designations are usually component parts of informal deals that allocate the top position to one State or group of States from the region the bank serves and the bank’s deputy position or positions to another State or group of States that is located outside the region but contributes significantly to the organization’s coffers. Consistently, whereas the Asian, European, and Inter-American Banks are headed by nationals from their respective regions, a US national has always served as the First Vice President of the European Bank and the Executive Vice President of the Inter-American Bank, and since 1978 has served as a Vice President of the Asian Bank.

US nationals are not the only ones who benefit from these arrangements. In 1983, a third vice presidency was added at the Asian Development Bank so that the Europeans would have a designated senior official there. And in 2003, a fourth vice presidency was added to give the Chinese representation. In the African Development Bank, vice presidencies are allocated informally as well. From the Bank’s founding until 1985, the four vice presidents were distributed regionally within Africa – one to North Africa, one to Anglophone Africa, another to Francophone Africa, and one to Nigeria. In 1986, a fifth vice presidency was added for a non-African.

The patterns of high-level appointments in IFIs demonstrate a series of interlocking agreements, all off the books, that allocate positions among the institutions’ major contributors and the world’s financial powers – the US, the European Union, and Japan. These three take advantage of their positions as levers of influence within the IFIs, often seeking to staff their posts with individuals who previously served in their respective national finance ministries. Thus, for example, the US nationals who have held regional development bank vice presidencies have frequently been former US Government officials, usually from the Department of the Treasury.

The informal allocation of appointments and positions is also prevalent outside the context of the IFIs, particularly in the UN. Among the foremost beneficiaries in the UN are the P-5 – China, France, Russia, the United Kingdom, and the US. By a rule, nowhere codified and only occasionally breached, each of the members of the P-5 (or one of its nationals) can sit on any UN body, if the country so chooses. The only significant exceptions to this ‘P-5 Convention’ were considerable gaps in China’s representation because of the long-running dispute over whether Taiwan or the People’s Republic properly represented China in the UN.

This informal agreement has applied regardless of the nature of the institution or position, courts included. Thus, until recently, with one notable exception, a judge from each P-5 member, except for China, has had, by ‘tradition’ (the word used in a 1956 Department of State memo),Footnote ²¹ on the International Court of Justice (ICJ) since the Court’s inception, even though its Statute does not provide for such preferences. Similarly, each of the P-5 countries continuously had a judge on one of the two ad hoc criminal tribunals, the International Criminal Tribunal for the Former Yugoslavia or the International Criminal Tribunal for Rwanda.

The P-5’s prerogatives extend to other UN bodies as well. All of the P-5 States, again with the exception of China and some other minor, occasional interruptions, have continuously been members of the ECOSOC and its subsidiary bodies, such as the Commission on Human Rights, the Commission on Narcotic Drugs, the Commission on Population and Development, the Committee for Program and Coordination, the Commission on Social Development, the Commission on Sustainable Development, the Commission on the Status of Women, and the Statistical Commission. With few exceptions, they (or their nationals) have also served, when they have chosen to run, on the subsidiary bodies of the General Assembly – such as the Advisory Committee on Administrative and Budgetary Questions and the ILC – as well as on the executive boards and councils of UN Funds and Program – such as UNDP, UNEP, UNICEF, and the United Nations Human Settlements Programme. Furthermore, they have effectively held permanent seats on the councils and boards of UN specialized agencies, such as the FAO, the International Atomic Energy Agency, the International Civil Aviation Organization, the International Maritime Organization, the International Telecommunications Union, the Organization for the Prohibition of Chemical Weapons, and the World Meteorological Organization.

Finally, the P-5’s preferential treatment extends to the top levels of the UN Secretariat. At least one of its nationals from each member of the P-5 has been appointed to the Under-Secretary-General level (the rank just below the Secretary-General and the Deputy Secretary-General) since the founding of the organization. The purpose is evident. As Marrack Goulding, himself a former British Under-Secretary-General, has written, ‘[a] senior UN official nominated by his or her government was … assumed to be in the Secretariat to do that government’s bidding’.Footnote ²² The P-5 Under-Secretaries-General are assigned the most significant portfolios, each one reflecting the country’s particular interests at that moment. From 1992 until 2007, for instance, the US insisted on the post of Under-Secretary-General for Administration and Management, reflecting the US concern with those topics during this period.

The P-5 Convention is far from the only informal rule operating at the UN in the area of representation. Some positions informally rotate by region. For instance, upon Kofi Annan’s resignation, it was widely said that it was ‘Asia’s turn’ to have one of its own as Secretary-General – and so it was. Predictably, a European national – António Guterres – succeeded Secretary-General Ban Ki-moon. UNCITRAL informally rotates positions in the Commission’s bureau among regional groups, and the ILC does the same for its officers. Further, judgeships at the ICJ are informally allocated by region, and even sub-region. Of the ten ICJ seats not set aside for the P-5, two are from the Western European and Other Group (WEOG), one is from the Group of Eastern European States, two are from the Group of Latin American and Caribbean States, three are from the Group of African States (including one from Arab North Africa, one from Francophone sub-Saharan Africa, and one from Anglophone sub-Saharan Africa), and two are from the Group of Asian States. This distribution mimics the formal distribution of the ten non-permanent seats on the Security Council. Regional groups choose their own candidates for ‘their’ seats, when they can agree, sometimes on the basis of region-specific informal agreements that allocate nominations among sub-regional groups.

Other positions are allocated exclusively to a specific region or country. Thus, the UN Legal Counsel (the Under-Secretary-General for Legal Affairs) is always a WEOG national (from a non-P-5 country). The US has been given the top job at UNICEF and the WFP (since the early 1990s) and the second-ranking job at the FAO. Since the 1970s, Japan, a major funder of the UN, has held a permanent slot at the Under-Secretary-General level and a judgeship on the ICJ for its nationals.

In addition to informal agreements, which privilege powerful States, the operational mode manifests in two other ways: in budgeting practices and informal consultation processes. When an organization depends on voluntary contributions, either for its general budget or for the support of specific programmes, the organization becomes particularly dependent on its donors. And with their contributions, donors expect to exert influence on the operation of the organization. Organizations themselves – and other Member States of the organization – understand this and provide donors with an outsized influence on the organization’s operation. Sometimes, in the case of restricted contributions, this is recognized explicitly by the contracts entered into by organizations and their donors that establish reporting and monitoring frameworks. Sometimes it is done implicitly by giving donors special privileges within the organization.

Organizations that primarily rely on voluntary contributions therefore no longer act as true multilateral institutions with collective principals but rather in significant respects as the delegated agents of multiple principals. In this form of funding, the ‘IO agent is guided by and held accountable through its bilateral relationships with individual donors, rather than through a single relationship with the intergovernmental body’.Footnote ²³ Though the assessed contribution regime is not perfectly multilateral because of the greater sway of high-contributing States, in comparison the funding of IOs through voluntary contributions significantly decreases the role of collective decision-making on the critical issues that pertain to budgetary matters, including the size of the budget, burden sharing among Member States, the organization’s priorities, and its accountability.Footnote ²⁴ Depending on the type of voluntary contribution (restricted or unrestricted), some or all of those issues will be determined not by the organization’s collective principal but by the aggregation of decentralized choices made by individual donors, including non-State actors such as the Bill & Melinda Gates Foundation. Effectively and informally, then, wealthy states and non-profit organizations control IO decision-making through their ability to control the funding of IO budgets.

The operational mode also manifests in informal influence arrangements. As an example, cities, networks of cities, and groups of networks have convened on the margins or sidelines of intergovernmental meetings and diplomatic conferences hosted by IOs, and mayors have testified at committees before and during negotiations. These efforts have had influence. In 2015, the Global Taskforce of Local and Regional Governments, together with others, worked with sympathetic States to ensure the inclusion of Goal 11 Sustainable Cities and Communities in the Sustainable Development Goals.Footnote ²⁵ That same year, city networks were crucial in the run-up to the Paris Agreement and also had influence in the creation of the Sendai Framework for Disaster Risk Reduction 2015–2030. As part of the Habitat III process that generated the New Urban Agenda in 2016, the Global Taskforce advocated through the production of policy papers, the organization of informal hearings for the benefit of the Habitat III Secretariat and the Bureau of the Preparatory Committee, and the coordination of the second World Assembly of Local and Regional Governments.Footnote ²⁶ The New Urban Agenda’s Quito Declaration on Sustainable Cities and Human Settlements for All ‘acknowledge[d] … the contributions of subnational and local governments, in the definition of the New Urban Agenda and [took] note of the second World Assembly of Local and Regional Governments’.Footnote ²⁷ And the Quito Implementation Plan for the New Urban Agenda ‘note[d] the importance of continuing to engage in the follow-up to and review of the New Urban Agenda with subnational and local government associations represented at the World Assembly of Local and Regional Governments’.Footnote ²⁸ Following Quito, the 9th Global Conference on Health Promotion, held in Shanghai, included a Mayors Forum in recognition of the role mayors in urban health. Over 100 mayors adopted the Shanghai Consensus on Healthy Cities. Subsequently, cities had similar influence in the creation of the Global Compact for Safe, Orderly and Regular Migration and the Global Compact on Refugees.

Policy communication between cities and IOs also takes place in forms and spaces specifically designed for that purpose. IOs have organized conferences, such as the World Urban Forum, for the exchange of information and perspectives with representatives of cities and city networks, and IOs have entered into memoranda of understanding and partnership agreements. Since 2014, the Global Partnership for Effective Development Co-operation, a ‘multi-stakeholder vehicle for driving development effectiveness’ that IOs participate in, has included a representative of local governments on its steering committee.Footnote ²⁹ The Global Forum on Migration and Development (GFMD), an intergovernmental venue that works closely with the UN and the International Organization for Migration, has a Mayors Mechanism, which ‘is a means to formalize the relationship between the government-led GFMD process and the local authorities’ and ‘establishes a platform to interact with States … [and] aims to add depth to the GFMD deliberations by bridging existing divides between local realities and global policy discussions, and between migrants, civil society, and States’.Footnote ³⁰ Similarly, local authorities, as a ‘major group’, are provided significant participatory rights in the High-Level Political Forum on Sustainable Development and previously in the Forum’s predecessor, the Commission on Sustainable Development. IOs participate in or have strong links to these dialogic processes where cities have some form of participation.

This is not to exaggerate the role of cities in IOs, which varies significantly by organization and topic, but those in the know appreciate how and when the system works, how it differs from the official narrative (the formal mode), and where the fault lines are located. It is a silent compromise among those involved.

The operational mode works to enhance the influence of powerful actors – States and NGOs – and incorporate the voices of the less influential and those that are not recognized as formal participants at the international level, like cities. There are thus democratic aspects to the operational mode and antidemocratic characteristics in the mode’s current practice.

4.4 The Aspirational Mode

The operational mode incorporates and gives priority to voices and policies that are not countenanced by formal rules of IOs. In the aspirational mode, States, IO officials, politicians, activists, and scholars advocate for changes in the formal and operational modes. The aspirational mode is thus reformist in nature, though it can be revolutionary as well. Either way, it is necessarily a technique of advocacy, not of governance. It is an attempt to change the status quo, its rules and techniques. If successful, the results will be incorporated into the formal and operational modes. Its content, thus, can be transitory.

Contemporarily, broad calls for ‘democratic representation within international organizations’ dominate the aspirational mode,Footnote ³¹ as do more narrow proposals to change IMF quota formulas and decision-making rulesFootnote ³² and ‘calls for merit-based, open and transparent … selection process[es]’ for high-level IO staff appointments, such as the World Bank president.Footnote ³³ These critiques of the formal and operational modes of representation seek to change the rules to bring IOs more in line with the representational principles of democratic legitimacy propounded by their advocates.

In this way, the aspirational mode operates outside of the rules of IOs as they are currently constituted, though its policies can be advocated both inside and outside organizations by their supporters. The aspirational mode, therefore, is a position from which to reform IOs. It serves, in this way, as a great value for organizations, since it provides an opportunity to think through, suggest, and lobby for change and improvement. IOs need to maintain their democratic legitimacy, and therefore organizations require reformers to push for change. The aspirational mode serves that function by providing a space for bringing IOs into line with new and evolving expectations.

4.5 The Alternative Mode

The aspirational mode can lead to changes in IOs and their formal or operational modes, but sometimes the desired change will not be possible or will not occur soon enough. Under these circumstances, actors may resort to alternatives to IOs as a means of allowing representational practices that the formal and operational modes do not permit. In the alternative mode, States and non-State actors establish either groups or NGOs that are not constituted within international law’s formalist framework, thereby allowing them to create representational systems that are not available under current international law.

Perhaps the best examples of IO alternatives are the Global Fund to Fight AIDS, Tuberculosis and Malaria and Gavi, the Vaccine Alliance, both organized under Swiss domestic law. The Global Fund’s board includes twenty voting and eight non-voting members.Footnote ³⁴ Of the voting members, seven represent developing countries (with regional distribution among them), eight represent donors, and five represent civil society and the private sector (including one person living with HIV/AIDS or from a community living with tuberculosis or malaria). The non-voting members represent additional constituencies or Fund officials (the Board Chair and Vice-Chair, the Executive Director). Similarly, the Gavi Alliance board is divided into ‘Representative Board Members’ (representatives from partner institutions and stakeholders, including permanent seats for UNICEF, WHO, the World Bank, and the Bill & Melinda Gates Foundation) and ‘Unaffiliated Board Members’ (individuals appointed in their personal capacity on the basis of their skills and networks).Footnote ³⁵ Both organizations are established as foundations under the Swiss Civil Code and are provided a significant degree of independence that they generally would not have under most countries’ domestic law. Further, under its Host State Act, and previously under other domestic authority, Switzerland has entered into headquarters agreements with both organizations because they ‘carry[] out tasks which are normally the responsibility of those intergovernmental organizations, international institutions or States; [and] play[] a key role in an important area of international relations’.Footnote ³⁶ Such agreements grant them privileges and immunities, including tax exemptions, distinct from those that would otherwise apply under Swiss law.Footnote ³⁷ Other States have also provided privileges and immunities to the Global Fund by amending their domestic lawFootnote ³⁸ or entering into a multilateral privileges and immunities treaty for their benefit.Footnote ³⁹ Effectively, these States have recognized these NGOs (foundations) as IOs.

Groups, like the G7 and G20, are another alternative to IOs. These meetings allow States to meet together and with representatives of IOs, such as the African Union, the European Union, the IMF, and the Word Bank. Recently, for example, the heads of the IMF, World Bank Group, Organisation for Economic Co-operation and Development, and Financial Stability Board met with the G7 finance ministers and central bank governors.Footnote ⁴⁰ In September 2023, the African Union was made a permanent member of the G20.Footnote ⁴¹ These types of State-IO interactions would generally be difficult within the formal confines of IOs.

The alternative mode provides opportunities for the representational assumptions of the formal mode to be bypassed. It is not a perfect solution to the challenge of democratic legitimacy, however. There is nothing inherent in the alternative mode that leads to that outcome. The alternative mode is used primarily by States when it is convenient for them to do so: when States find that they want to operate outside of the formal and operational modes because those modes do not allow them to achieve their desired goals. The alternative mode therefore is an opportunity to achieve democratic legitimacy, but as with change to the other modes, there must be political commitment and an openness to innovative thinking to realize that goal.

4.6 A Second-Best System of Representation and Democratic Legitimacy

Each of the four modes of representation reflects different inclinations: the positivist, the functionalist, the principled, and the practical. The positivist prefers the formal mode; the functionalist employs the operational mode; the principled advocates through the aspirational mode; and the practical looks to the alternative mode. All four have their advantages, and all four have their limits.

Where an actor (a State, a sub-State actor, an NGO, a corporation) falls along the representational spectrum will depend on particular circumstances, not who they are. For some issues, a State will act within the formal and operational modes, while at other times and for other issues they will use the alternative mode. Similarly, an NGO may be satisfied with the formal mode in some instances but resort to the alternative mode in other situations. Many actors thus work in multiple modes simultaneously, often doubling up with the aspirational mode as they operate within established rules and attempt to reform them at the same time. The four modes are different techniques of representation, used by participants to achieve desired goals within the State-centric framework of international law and IOs.

From the perspective of minimum public order, the co-existence of all four is essential. Each mode does separate work. Together, they allow for change and provide processes to achieve an equilibrium among competing values.Footnote ⁴² Within the statist paradigm that defines the international system, a system of multiple modes of representation is thus not only descriptively accurate, it is also necessary because of the challenge of democratic legitimacy that is most prominently manifested in the aspirational mode. Optimally, the formal system would fully adapt to that challenge. But there are limits to the formal system’s ability to change because of the unwillingness of certain actors to accede to that change. In the meantime, there need to be methods and spaces to advocate and implement change. The international representational system of the four modes is thus second best. Yet, second best is better than any of the possible alternatives while the process of change slowly proceeds.

5.1 Introduction

This chapter proposes demoicratic representation as a subtype of representation in international organizational practice. While the regime-neutral representation of States remains prudent for legitimate international organization of States with different and conflicting regime types, demoicratic representation ought to guide international law- and policy-making among democratic Peoples.

The tenets of the chapter’s argument are the following:

1. P1: The People is constituted by rules that create a set of interconnected statuses, institutions, and procedures which jointly exercise highest political authority, i.e. govern sovereignly.

2. P2: The People is composed of citizens. The democratic People is composed of citizens who hold specific deontic powers that enable them to govern via the institutions of the People, directly and by representatives whom they authorize and hold accountable.

3. P3: Political representing is either standing for the People as a whole or its institutions, or it is the acting of representatives in the stead of and for the citizens in governmental institutions of the People.

4. P4: The People is represented only by all the different types of representative persons who act within different types of governmental institutions and procedures of the People. These types of governmental institutions are determined by essentially different and jointly complete types of governmental actions, such as law-giving, executive government, and jurisprudence.

5. P5: Democratic Peoples are accountable to each other as Peoples and to each other’s citizens.

6. P6: Under any decision rule, even consent, there is a possible consent deficit about the decision rule even in case of unanimity on a substantive issue.

7. C1: Towards each other, the Peoples are adequately represented and held accountable only via the systemic interaction of all the different types of highest governmental institutions.

8. C2: Demoicratic representation needs to be designed in common institutions of the Peoples that unite its highest governmental intuitions, i.e. in the Peoples’ councils of parliaments, the Peoples’ councils of executives, the Peoples’ councils of high courts.

9. C3: Demoicratic representation ought not to be understood as working exclusively under the principle of Peoples consent (unanimity rule). Rather it is the representational space in which the consent deficit about the decision rule of inter-People relations is addressed and calibration sought by seeking a balance between different decision rules.

10. C4: Political international organizations (IOs) ought to conform to the principle that all institutional manifestations of the People be represented in the corresponding common governmental institutions. They need to embody demoicratic representation within their organization. The less comprehensive, economic, legal, technical intergovernmental organizations (IGOs) ought to be part of a system of mutual accountability and thereby assure demoicratic representation by IOs.

The goal of this chapter is to sketch out a conception, justification, and some basic guidelines of an institutional design of demoicratic representation.Footnote ¹ It aims to contribute to further joining the fields of philosophy of democracy and philosophy of international law.Footnote ² Given the plurality and interdependence of Peoples, philosophy of democracy ought not to remain fixated on the monadic demos and exclude the dimension of the relation among governing Peoples from the core of its normative and empirical inquiries. Alternatively, it ought not to dissolve the demos by conceiving democracy as ‘subject-less’ and ‘unbounded’ deliberationFootnote ³ as this disregards the capacity of Peoples to act effectively and responsibly which is necessary for any meaningful realization of democracy.Footnote ⁴ Philosophy of international law, for its part, ought not to remain largely a philosophy of States, of will and consent of States, and of law of States, the legal possibility of treaty-making by IOs in their area of competence notwithstanding.Footnote ⁵ Important ground has been broken by the thesis of the exception of democratic State consent in international law at large.Footnote ⁶ This chapter proposes that the forging of consent and addressing of reasonable disagreement among democratic States in international relations and in IOs ought to be realized by demoicratic representation. It thereby goes beyond but not against the thesis of exception of democratic State consent.

5.2 Demoicratic Theory

European law theory has been the primary meeting ground of philosophy of democracy and of international law.Footnote ⁷ Kalypso Nicolaïdis, who coined the term demoicracyFootnote ⁸ succinctly defines it as ‘a Union of peoples who govern together, but not as one’.Footnote ⁹ In the version of Pavlos Eleftheriadis, the European Union as ‘Union of Peoples’ is a dual system, based on reciprocity and equality of Peoples.Footnote ¹⁰ The assumption is that a multitude of democratically governed Peoples relate to each other horizontally based on shared principles. A demoicratic union is a plurality of concordant members. A democratic unit, on the other hand, is a whole without vetoes, opt-outs, or exit rights for the subunits.Footnote ¹¹ The debate on demoicracy has gained momentum.Footnote ¹² The recent publications remain confined to EU studies, but the normative core includes claims that go beyond contingent voluntary association: mutual non-domination and mutual recognition of Peoples.Footnote ¹³

The concept and theory of demoicracy have been criticized in three ways: (1) Demoicracy as the finalité of European integration.Footnote ¹⁴ (2) Demoicracy as justification of the actual EU.Footnote ¹⁵ (3) Miriam Ronzoni criticized Nicolaïdis’ ‘third way’-conception of demoicracy, stating that it conceptually collapses either into intergovernmentalism or statist federalism.Footnote ¹⁶ Ad (1) and (2): The present chapter is committed to an ideal-typical and general justificatory theory of demoicracy, not a philosophy of history about the perfect end-state of the EU, and not a legitimation of the actual EU. The legitimation of the instantiation of demoicracy in a concrete political entity such as the EU cannot be directly deduced from a general concept and justification of demoicracy. Ad (3): First, Ronzoni’s position neglects the possibility of the existence of a rule of recognition among demoi who want to act together but not as one. The existence of such a rule would mark a qualitative difference between demoicracy and intergovernmentalism as well as State federalism. Second, if demoi, as popular sovereigns, are understood to have legal disabilities, multiple institutional modes of being, and guarantee transnational fundamental rights, the reduction of demoicracy to either statist federalism or executive intergovernmentalism is inadequate. Third, intergovernmental rule-making limited to executives violates the separation of powers principle and sidelines representational bodies through which citizens can authorize and control legislation.Footnote ¹⁷ Federalism, for its part, is a theory that predates liberal democracy.Footnote ¹⁸

5.3 The Social Ontology of the Demos

At the basis of a theory of demoicratic representation lies the question how the citizen, the People, and a group of Peoples can be understood ontologically as social entities, and what modes of representation are possible and normatively necessary from the democratic standpoint. I avoid the dichotomy between a substantialist and interactionalist ontology of democracyFootnote ¹⁹ by assuming that social beings are constituted by social speech and document acts.Footnote ²⁰ If we assume that (1) ‘people’ denotes the same social entity in all three descriptions of democracy as government of, by, and for the people; if we further assume that (2) the people are the sovereign person in democracy, then there occurs:

1. (i) a circularity problem concerning the people as pouvoir constituant (‘of the people’) and pouvoir constitué (‘by the people’; demos problem);

2. (ii) a paradox of popular sovereignty: the sovereign and the subject are identical (transitivity problem).

We must be able to show that ‘people’ does not refer to the same entity in the three nominal descriptions of democracy. The People as origin or ground of the exercise of governmental powers (pouvoir constituant) cannot be the same as the People that rules via institutions (pouvoir constitué). Furthermore, the People that rules cannot be the same as the People that is ruled. I will not address the demos problem. As I have tried to show, demoicracy might be the closest shot we can get at solving the demos problem.Footnote ²¹ I will also not further pursue the question what the people is, for which rule is exercised (population). I assume it is the group of agents under the obligation to follow the laws and orders of the demos and the group of agents to whom the governing People is supposed to give priority in serving. It is important to understand the ontological difference between the People as pouvoir constituant, the People as pouvoir constitué, and the population to be served. To understand the term in the middle that is of interest here, the People that rules, I follow and complement the Ciceronian, jural position, and circumscribe the People in the following terms.

The democratic People is a group that is constituted by the common recognition of a coherent set of primary and secondary rules, which establish statuses, institutions, and procedures through which status holders, first and foremost citizens, exercise highest lawmaking and law-enforcing authority (deontic power) in view of providing the service of government, that is to say, of public goods and public services within a territory, to a population.Footnote ²²

The democratic People has ownership of and priority over the State and not vice versa. A State controlled by the People deserves the name ‘res publica’, as opposed to a State that is the instrument of self-serving rulers. It is important to determine what constitutes a People (demos) as a specific social entity in distinction to what constitutes a State, and how the two are related. The ontology of the State is beyond the remit of this chapter. I consider the State an apparatus of epistemic and coercive administrative instruments and procedures that are purely rational from an ideal-typical point of view (as opposed to reasonable). In a democracy, the State apparatus is controlled by the democratic People. In virtue of this control and constraint of the State by the People via reasonable debate on the yielding of rational State power, the State becomes res publica. ‘Est res publica … res populi’ as Cicero wrote. Whoever officiates in the State or represents the State as res publica, must act under the control of the People in the service of the Population. The representation of the People vis-à-vis other Peoples might, and most likely must pass through the State, but the latter has only an instrumental value for the People. For demoicratic IOs this means that the core social entity that is to be represented is the People, not the State.Footnote ²³ From a democratic point of view, international relations are inter-People relations. The term ius gentium can be reinterpreted in this jural and republican sense, that is, as ius populorum. To conceive international law as a law of States can be accepted as a shorthand that indicates that only Peoples disposing of certain State-like capacities and instruments to give credible assurance to other Peoples can honour justice among Peoples.Footnote ²⁴ Second, international representation as representation of States can be understood as compromise that one must consider to avoid severe conflict with and to allow for constructive peaceful relations with non-democratic States. But among governing populi, the system of representation needs to be conceived and designed as a representation of Peoples.

I cannot go through the full set of rules that constitute a governing People. Nor can I fully explain my approach to social ontology, which draws on Adolf Reinach, Herbert L. A. Hart, John Searle, and Barry Smith. However, I can aim for completeness regarding the types of secondary rules that constitute a democratic People: (1) a fundamental act of recognition of a union claim of the People as constitutional unit under one system of law. This enactment of the claim to constitute a coherent system of law is the rule of recognition;Footnote ²⁵ (2) constitutive and regulative rules that establish statuses, institutions, and procedures and regulate them.Footnote ²⁶ The regulative rules include rules of change and adjudication; (3) rules conferring deontic rights, powers, and duties to office holders and institutions, but first and foremost to citizens.Footnote ²⁷

Here, I skip over the rule type (1). It is identical in content for democratic and non-democratic peoples. Peoples under any type of government require a rule of recognition, a recognized union claim, to exist as a constituted people or union of Peoples. To (2) and (3): a democratic People is distinguished from other legally constituted non-democratic Peoples by the content of specific secondary, formalized rules. Also by informal rules and social norms, but these cannot be the object of my analysis here and require a historical and empirical method of analysis. Furthermore, I cannot do justice to the concrete instantiations of the formal rules. I hold that the distinctive formal rules of a democratic People concern mainly the status (office) and deontic powers of citizenship, and the procedures that citizens are involved in when making authoritative decisions themselves or via representatives, when governing together as People.

These rules, first, determine freedom rights that are necessary elements of democratic procedures.Footnote ²⁸ These freedoms do not conceptually depend on the status of citizenship. Freedom of speech and association, essential conditions of the deliberative demos in any democracy, are not specifically citizen-freedoms, they are fundamental freedoms of humans. It necessarily follows that the deliberative People (as type) and the formally governing People (as type) are not constituted by the same set of statuses of individuals, and they are potentially composed by different individuals de re.Footnote ²⁹ This means that liberal democratic Peoples, in which citizens and non-citizens have freedom of expression of opinion and association, are essentially open towards each other as deliberative Peoples. In this sense, deliberative demoicracy is a fact as soon as contiguous liberal democratic Peoples exist. The fact that domestic publics are stronger than transnational publics and the individual ability and structural conditions to participate in deliberation are different empiricallyFootnote ³⁰ does not change this ontological fact of reciprocal openness of liberal democratic demoi. The existence of ‘echo chambers’ and ‘filter bubbles’ in cyber space would reinforce the point, because they do not only fragment the deliberative demos within the borders of the formal demos but also cut across formal demoi. The formal governing People as demos type is composed by members constituted by rules that define the status of citizenship and regulate accession and exit to this status. Following Searle, the status is a term in a rule holding: X counts as Y in C. Meaning, X counts as citizen in C. The C-term refers to the entity established by the recognition of the union claim. The status-term, the Y-term (the status of citizen in our example), is established by a rule that enacts something like ‘members of C have the status of citizens of C’. This is then further substantiated by deontic powers given to individuals with this status, and by the rules of accession to and exit from the status, as well as the rules how to document identification of a person as holder of a status, and so on. The conceivable deontic powers of the democratic citizen are the power to vote and stand in elections of political representatives, to vote in issue-specific votes (referenda or initiatives) as well as to be active in bringing them about. The status of citizens and of office holders of the institutions of the People also comes with a right over duties, duties that only the status-holder can claim to be under an obligation to fulfil and that persons who do not hold the status are not allowed to fulfil. Examples are jury duties, military service, and civil service.

The status rules just mentioned are not sufficient to constitute the People. When citizens sleep in bed, gather informally, gather to protest, and so on, they are, and they might even be acting specifically as citizens, but they are not the People that rules with democratic authority. Their acts (e.g. of protesting, of making representative claims, deliberating, etc.) are not acts that directly yield the deontic powers of the People. The ruling demos needs what Jean-Jacques Rousseau called ‘lois fondamentales’. Minimally, they consist in rules of assembly and decision-making procedure.Footnote ³¹ But more are necessary, because the People need executive and juridical institutions, and a functional legislative assembly in most cases is an assembly of representatives, not of all citizens. Direct democracy is not assembly democracy, it is the existence of direct democratic devices within fundamentally representative democratic systems.Footnote ³² The People is thus further constituted by rules setting up an ensemble of governmental institutions and procedures with office holders who are social persons: representatives; and by rules conferring deontic powers to the citizens and office holders. Democratic authority is yielded by the citizens and representatives just in case they act within specified institutional procedures.

As ruling collective, the People need to perform many different types of acts at the same time and hence delegate these tasks to various institutions. This seems clear, independent of any (inadequate) conception of direct versus representative democracy and independent of any conception of group agency one might subscribe to.Footnote ³³ The people will delegate control to institutions, who control each other. The consequence is that vertical accountability is complemented by horizontal accountability and that the realization of democracy is best understood not by looking at the individual institutions of the People in isolation but as a democratic system solving certain problems.Footnote ³⁴ Any form of horizontal accountability necessarily implies representation in some form and shape.Footnote ³⁵ Overall, the exercise of the deontic powers of citizens happens via institutions which necessarily demand the existence of the status of representative in different institutional instantiations. It is these institutions through which the People is to be represented toward other Peoples in demoicratic representation.

5.4 Representation

Political representation is a broad concept, widely debated, multidimensional, and complex in different contexts.Footnote ³⁶ Nevertheless, I hold that the analytical structure of political representation is simple, and it is the following. Political representation of x is either standing for x or acting for x.Footnote ³⁷ The former can be attributed to many sorts of beings including agents, the latter is enacted only by social persons, which I call representatives. ‘X is a social person, if and only if X is the subject of social acts’.Footnote ³⁸ Social acts are acts that a person addresses to another person and that have as their success-condition that the other person grasps it. Political representation is a social act. If A (or a group A_1-n)Footnote ³⁹ represents B (or a group B_1-n) in the mere sense of standing for, then A embodies, depicts, or symbolizes B. It can also be that A carries essential traits of B (descriptive representation). Examples are a head of State who stands for the union of the People under a constitution. The fish represents the Christian community in coded form.Footnote ⁴⁰ The half-moon stands for Islamic community. The star of David stands for Judaism. A flag symbolizes the nation, and so on. These entities (fish-image, half-moon image, flags, etc.) are not social persons. Unlike promising, commanding, requesting, agreeing with, symbolizing is not a social act. A ‘mini-public’, created by semi-random sortition, depicts the people as image of the public at large, that is why it is called a mini-public in distinction to an elected assembly, which is not an image of the people but an elite or group of interest representatives under some criterion of choice. An image does not perform social acts, a ‘mini-public’ as such is not a social person. Women can be understood to represent women in the sense that they share their essential descriptive features and life-perspectives (descriptive representation). A descriptive representative as such is not a social person. An attorney defending a serial killer sharing no descriptive traits with the serial killer is a social person, they act for the serial killer in court. The representational action for can only be tied to descriptive representation under the empirical assumption that people who share features of other people can be expected to act in their interest because the descriptive features have led them to certain experiences and perspectives which will allow them to act for others. They thus draw trust in view of acting for. But this is not a necessary dependence. The lawyer representing the serial killer does not have to share the experience of this practice to be a good representative of the serial killer at court. Pro tanto, B does nothing for A or vis-à-vis A by simply sharing features of or symbolizing A. This aspect merits a lot more analysis, but I will leave it at this because I think democratic legitimacy depends on social action much more than symbolic action. It is a feature of fascist theory of representation to misrepresent the symbolic standing for the People as a whole of one leader as the acting for the People as a whole.Footnote ⁴¹ Authorization and accountability that are part of an a priori theory of social acts of representation are cut out.

If A represents B in the sense of acting for, this comes in two forms: (1) B puts an obligation or a rule on A to carry out an act on its behalf or in its stead (delegation); (2) B creates representative powers in A, that is, powers to act freely in their name and best interest and thereby commit B to certain positions, or even to put B (and themselves de re) under rules and obligations. Who puts a rule and obligation on whom is the opposite in the two figures of representation: (1) B puts an obligation on A; (2) A puts an obligation on B (or on B and A). Both cannot be the case at the same time regarding the same content of obligation. One collides with the other. Furthermore, both figures can be conditional or unconditional. Ad (1): B can give A a conditional command (obligation) with unconditional content (B commands A to do y just in case x occurs) or an unconditional command with conditional content (B commands A to do y or z in case x occurs).Footnote ⁴²

The democratic representative government adopts both counter-directional forms of representation (1) and (2), and this in respect to the same legislative or executive body of representatives. The system of democratic representation is like a two-way street with no clear rule on which side to drive on. Representatives have a mandate in the sense that they are elected by the represented based on promises and programmes of what they will do on their behalf or interest, or in what they see as the public interest, with different conditionalities. The election is the acceptance of a promise by the represented that puts an obligation on the representatives. When they get elected, there is a sense in which the represented rightly see the representatives as being under the obligation to carry out the programme for which they were sent to parliament, the executive, or the judiciary. The representative promises ‘if elected, I will do x’.

The counterargument to this thesis would be to say that the representatives put the obligation on themselves by making the electoral promises. There is thus no bidirectionality. But that is not what happens in an election, and it does not cover the election of the representatives in the judiciary. The promise, as any social act, is ‘vernehmungsbedürftig’.Footnote ⁴³ Some act of grasping, reception, or acceptance is necessarily part of its success conditions. If you promise to help me move next weekend that promise does not put any obligation on you if I do not acknowledge your promise. Representatives make all sorts of promises before the election. By casting a vote, the voter chooses between different promises of different representatives and parties. The winner of the election is under a pro tanto obligation to carry out a promise only because their promises of the promisor have been accepted by the promisees in the electorate and their votes have been aggregated according to the rules. It is thus the represented who choose (under a decision rule) which promises count and thereby confirm a mandate of the representatives to do x in the representative assembly. Besides, offices also come with preset duties ex ante. There is always a non-promissory mandate implied in the office of a representative.

However, in many cases representative government implies that the political representatives impose obligations on the represented in a way not foreseen in the mandate they voted for. Representatives have a mandate and their promissory obligations are only pro tanto obligation. Additionally, they have representative powers to put binding obligations on all in ways their voters did not foresee or in an interpretation of the public interest that is not necessarily shared by their constituents. In all this, the representatives still claim to act for the represented, but not in a narrow sense of responsiveness to mandates. The represented, on the other hand, still expect the representatives they voted for to stay as close to the mandate as possible, or to have a reasonable explanation for why they moved away from the mandate, or even did something completely not foreseen in the mandate. One should not misconceive this situation with a simple citizen-bourgeois dichotomy ascribing the mandate to narrow interests of a constituency and the going past the mandate to a citizen perspective of common interest or best overall solutions. The mandate can be understood in terms of best overall solution and the compromise going beyond the mandate can be the result of interest bargaining. Be that as it may, it is inconceivable to ascribe only a mandate to representatives who have powers to put obligations on all (legislative, executive orders). They operate in a normative space that is characterized by the countervailing normative figures of mandate and representative power. The creative granting of deontic powers authorizes the representative to act in the name of the represented when putting obligations upon them. The election is the clearest and most certain form of authorization, albeit not the only one.Footnote ⁴⁴ On the other hand, it is also a priori possible that A act in the interest of B and according to the intention of B without representing B and even without making a claim to represent B.Footnote ⁴⁵

The ‘puzzle’ of the two necessary, but counter-directional figures of democratic representation has no definite solution in purely representative systems. It is only manageable over time in a cycle of authorization and accountability. One should not expect this to change in demoicratic representation in and by IOs. Elections are moments of authorization of representatives with both a mandate and with representative powers for cases in which it is not possible or not reasonable to narrowly adhere to the mandate. What is possible and reasonable must be judged by the represented and the representatives and re-negotiated at each election. By the formal periodic repetition of elections (and informal critique and debate), the electoral cycle also assures accountability. Representatives explain their choices, or the choices of their predecessors in the same party in case they are not incumbents. Voters judge if the representatives they voted for kept their promises and programmes, for which the represented gave them a mandate via their vote. They also judge whether the representatives have used their representative power in a reasonable way given their original programme and promises. The right to vote is a normative power, an instrument of authorization and accountability, meaning also of sanction or confirmation of the representatives or their party, in the hand of the represented, the citizens. Be that as it may, the tension and possible confusion between delegation of a mandate and representative power always remains present in democratic systems.Footnote ⁴⁶ The representative democratic system that includes the facultative referendum is a close shot at solving the ‘puzzle’ of democratic representation. For one, it puts additional normative powers of accountability in the hands of citizens (and non-elected representatives such as action committees) and increases deliberative qualities.Footnote ⁴⁷ This make for a reasonable justification of facultative referenda.Footnote ⁴⁸ But here I mean something more fundamental. The facultative referendum implies that representative powers are given to the representatives conditionally, and that the direction of command can be reversed by the represented in the framework of a democratic procedure (e.g. collecting a certain amount of signatures that trigger a vote the result of which is legally binding and can annul an obligation the representatives wanted to put on the represented). A representative democratic system with the facultative referendum has an instrument that allows for the possibility to avoid bidirectionality in case the countervailing forces of the latter cause an issue-specific gap between responsiveness to preferences of the represented and the decisions of the representatives, and in case this gap is judged too large by a majority of the represented. This is especially important in international demoicratic representation where the long chains of representation can make the gap between the represented and the representatives particularly large. A representative system with facultative referendum is a bidirectional system of obligation-imposition that can switch back to a unidirectional system of command with the possibility to avoid negative or confusing effects of bidirectionality. I say closest shot to solving the problem and not total solution to the problem because in the facultative referendum the command remains one of editing and not authoring the obligation (law). The facultative referendum gives citizens the right to refuse an obligation put upon them by representatives in the exercise of their representative powers. It does not give them the right to statute, to be the authors of the laws.

5.5 Demoicratic Representation

The People as governing entity is a semi-abstract entity and as such a composite of real human individuals under documents and representations that constitute offices and institutional entities and institutional procedures by document acts. This is important to understand democratic and demoicratic representation. It might be the case that a single institution of the demos alone is understood to stand for the People as if it were the whole or as if it were to encapsulate the whole, for example, the flag, the head of State, the citizen assembly. But this should not be misconceived as representative acting for the People. No single human or social status person, no single party, no single institution of the People alone, no single branch of government, not even the assembly of all the citizens, can act solely for the People as a whole. The People can only act via the institutional constellation of all interconnected offices, institutions, and procedures. Hence, if the People is to be represented adequately vis-à-vis other Peoples, all these institutions need to be involved and several different IOs together might be understood to represent the Peoples. The representative acting for needs to be performed by the corresponding highest institution of the People, that is, of lawmaking, law enforcement, and jurisprudence. Consequently, demoicratic representation ought to be understood systemically, as horizontal and vertical interplay of checks and balances of several institutions (IOs) commonly created and upheld by the Peoples.

In what follows, I try to show that the democratic representation is unfinished if there is no demoicratic representation and that demoicratic representation needs to be designed as representation of the Peoples via all their highest governing institutions, that is, legislative, executive, and judicial. In other words, the law of Peoples should be drafted by lawmaking representatives of the Peoples, the execution of the law overseen by representatives from the executive, the judiciary of Peoples by representative from the highest judiciary powers of the Peoples, and these institutions ought to respond to each other in a representative system. Part of this are also referenda of the Peoples.

(1) Only in a demoicracy, the governing People is accountable to other governing Peoples. In international relations in general, the People is accountable to other States or IOs of States overseeing international law, but not to other Peoples. Within a democracy, the popular sovereign might have restricted domain via legal disabilities, but in such a setting the popular sovereign is ultimately accountable to no one. It does not matter whether popular sovereignty is enacted as parliamentary sovereignty, in a system of checks and balances with constitutional jurisprudence, or direct democracy, or any other mode or combination of the above. The People, as a whole, is accountable to no one if we conceive democracy as an enclosed system of a singular demos. If the place of the sovereign is considered empty and the democratic sovereign replaced by the constant struggle for hegemonyFootnote ⁴⁹ we can no longer distinguish the legitimate electorate or legitimate voice of the people from usurpation. But in a democracy, the place of the sovereign is not empty. It is occupied by a set of interconnected institutions that represent the legitimate expression and authoritative voice of the People. A group of demonstrators is a legitimate form of voicing demands in democratic politics, but it is not an authoritatively binding voice of the People that replaces or empties the seat of sovereign People. The first raison d’être of demoicratic representation is the necessity of accountability of the Peoples towards each other. In international law, States are accountable to each other and that is of course a good thing and not to be replaced but complemented by the mutual accountability of Peoples. If we agree that the State, as a rational administrative and coercive apparatus is only an instrument of the People under its control in democracy, and if we hold that many States are not controlled by the People via corresponding institutions, then if follows that a democratic People is accountable to other democratic Peoples, not States. Second, all the democratic Peoples together cannot form a sovereign democratic People because this People would be an entity that is again accountable to no one.Footnote ⁵⁰ The overarching union of democratic PeoplesFootnote ⁵¹ is a union of sovereigns, not a sovereign unit.

(2) In a constellation of interdependence, democratic Peoples face (minimally) the following challenge: It is possible that People P₁ take a domestic decision d, that would not be accepted by either all or a majority of the citizens of Peoples P_2-n, a majority of the citizens P_2-n, a minority of the citizens P_2-n, or any combination of the three possibilities, i.e. a majority of P2, a majority of P3, a minority in P4, and so forth. The unacceptability of expected consequences C_exp in case they would affect P1 constitute the reason why the majority of P₁ take decision d. The citizens of P1 thus take a decision they would possibly have reason to reject were they the citizens of P2, and so on, but they have no idea if it is really the case and if it were the case how they could justify their decision to the citizens of other peoples or their corresponding representatives. Suppose that to create accountability among democratic Peoples, international institutions are created as we know them. In the classic form this happens by giving the executives representative powers in common international institutions. Suppose this institution of P_1-n discusses decision d. This can give the citizens of P₁ important guidance. They will know how many and which States do (not) accept their decision d, and for what reasons. But the citizens of P₁ still have at least the following epistemic problems:

1. (i) The citizens of P₁ have no way of knowing if the executives of P_2-n represented the majority or minority will of their citizens regarding decision d;

2. (ii) The citizens of P₁ have no way of knowing if decision d would be accepted by a majority of all citizens of P_1-n, albeit not by all the executives of the States, due to the fact that majorities and minorities will occur in Ps of unequal size;

3. (iii) The citizens of P₁ have no way of knowing the reasons why the majorities and minorities of the citizens of P_2-n reject (accept) their decision d.

Demoicratic representative institutions, in which the domestic political constituencies are represented vis-à-vis other Peoples have an epistemic and normative function. They can produce the knowledge about what other Peoples want when confronted with decisions of common concern. ‘People’ here refers to all the essential institutions that constitute the People as a collective actor, as opposed to only the executive. Such common institutions create a knowledge that central international bureaucracies, State executives, or domestic institutions that only debate internally cannot produce. Common demoicratic institutions in form of IOs can further serve the normative function to address mutual accountability issues of popular sovereigns in a manner that fulfils normative principles of the relation of sovereign Peoples.Footnote ⁵² If the citizens of P1 or any P_2-n assume that being a citizen is accepting accountability to citizens acting in legitimate democratic institutions, they will accept demoicratic institutions that serve an epistemic function of knowing how other majorities and minorities of other Peoples view decisions of common concern, and that serve the minimal normative function to calibrate decisions while taking into account the information produced by common institutions.

(3) In a democracy, the decision process is based on the majority rule in some form and shape. In a demoicracy, it is assumed that decisions are based on unanimity and that individual Peoples have veto rights and exit options. I will show that, fundamentally, demoicracy at the core is about dealing with reasonable disagreement among democratic Peoples about the adequate decision rule.Footnote ⁵³

As opposed to legal representation where one to one representation is possible, in political representation a representative represents not only one individual, always a group’s interests, opinions, perspectives. That is the whole point of political representation. This means that no individual’s particular interests are represented entirely by one single representative, as is the case when an attorney represents a client. The largest group in a constitutional unit are the group of the represented (all the citizens) and the group of the representatives (the government in various institutional forms and instantiations). The collective action of the group becomes possible by a collective decision rule. The decision so taken is considered as the decision of the group in virtue of the correct application of the decision rule (and further fairness conditions). In that case, the representative assembly (as group) acts in the name of the represented (as group), the disagreement of individual representatives and represented notwithstanding.

The consent of the group as such is established by a collective decision rule and only applies to the group.Footnote ⁵⁴ If the group adopts a unanimity rule, the difference between the number of vetoes and the majority votes is called a majority deficit.Footnote ⁵⁵ If a decision is rejected because three out of five persons vote against, the majority deficit is two, meaning that the majority of two who would have decided the vote under a majority rule are disregarded under the unanimity rule. If the group adopts a majority rule, the number of minority votes can be called the consent deficit. The consent deficit is the absolute number of minority votes under a majority rule, not to be confused with the minority–majority quota. If four out of five vote against a decision, the consent deficit is four meaning that the decision was adopted against four who would have been necessary to change their vote to reach consent, or who would have each been decisive in rejecting the decision under the unanimity rule. In this case the majority-superavit is one as an absolute number and the minority–majority quota (mi/maj) is 80 per cent. The mi/maj is one in case of a tie (a non-decision) and decreases the higher the number of the minority–majority difference gets. The higher the mi/maj, the less voters are disregarded in proportion to the total number of voters under a majority rule. The mi/maj thus puts the consent deficit in relation to the number of voters the veto-rule would disregard.

If the mi/maj ratio is repeatedly very low under the unanimity rule, the unanimity rule gets charged with increasing burdens of justification because a minority is constantly vetoing decisions. If the mi/maj ratio is repeatedly high under a majority rule, the majority rule gets charged with increasing burdens of justification because only a small majority is deciding the outcome. It is thus important to notice that the decision about the decision rule is subject to the same formal relations as the decision on material issues. It is not the case that under a unanimity rule everybody gets their way due to everybody’s veto position. There can be a second-order consent deficit, including a very large one, about the application of the unanimity rule even if the decisions that are taken under this rule are unanimous, that is, even when there is first-order unanimity about content. The application of the unanimity rule is in that case ‘pseudo-libertarian’, because under full libertarianism there would be no second-order consent deficit about the unanimity rule. But regarding the majority rule as well there can be a large consent deficit regarding its application even when the min/maj ratio is very low.

What does this mean for international relations’ principle of State consent in relation to the idea of demoicratic representation? First, international relations, based on free State consent, are possibly ‘pseudo-libertarian’ until proven otherwise. In the Vienna Convention on the Law of Treaties the ‘principle of free consent’ is voiced as ‘universally recognized’ but applied only to States, not citizens.Footnote ⁵⁶ Granted, all States that sign the Vienna Convention on the Law of Treaties agree to the consent-rule. There is thus no second-order consent deficit regarding the general application of this rule as far as the States that signed the convention are concerned. Now, in international representation, the States are represented by their executive governments. In some States, freely and equitably elected parliaments have the right to refuse binding obligations by the right to ratification of treaties. If that is the case, we could say that the principle of State consent is realized as Peoples’ consent in so far as the highest legislative institution of the People consents to an obligation. In the case of decision-making in the representative assembly of the People (parliament), the individual represented can assume that their representatives weigh in on the collective decision in deliberation and decision-making. If the assembly is elected by proportional representation that possible weight increases for minority positions. In international representation as it is currently practiced, the individual represented (citizen) is no longer represented by a representative of their constituency in an assembly, but only by a representative chosen by the assembly by majority rule or by direct election by majority rule, or by usurpation of the ruler. In other words, international representation means moving up the steps of a multilevel representation system whereby the possible consent deficit increases at each step, State consent notwithstanding. Furthermore, there is no voice in agenda-setting. In international relations where States have veto powers, there is no consent deficit regarding new decisions taken, but there is a possible large consent deficit among States regarding this decision procedure when it comes to certain issues, and this consent deficit will not be noticed from the inside of the system of State representation. What is more, from the point of view of the individual citizens, there might be a large first-order consent deficit regarding the decisions taken by State consent and regarding the unanimity rule of binding common State action. In brief, the problem of international representation is not only that international law is a set of primary rules and that IO lacks a single legislator adapting the law under a rule of change.Footnote ⁵⁷ Arguably, by endorsing the universal principles of free consent, of good faith, and of pacta sunt servanda (this principle is grounded on the a priori rule that promises create obligations), international law is based on a rule of recognition: the States form a union by recognizing as law all law that is enacted under these principles. As Hart mentioned,Footnote ⁵⁸ this is increasingly changing.Footnote ⁵⁹

The argument here is thus not about making consent and unanimity a direct source of international law by demoicratic representation of Peoples with veto powers. The point is that demoicratic representation can be understood as a way of institutionalizing a more adequate management of disagreement of international lawmaking, which, as Samantha Besson has pointed out, is arguably more pervasive in international than in domestic lawmaking, consent as constitutive element of international law notwithstanding.Footnote ⁶⁰ If we assume that international representation ought to be the representation of Peoples at least among democratic Peoples, meaning of citizens governing via common representative institutions, the possible first-order consent deficit of decisions taken by States and the possible second-order consent deficit of citizens of Peoples acting together regarding the principle of State consent cannot be ignored.

For a representation of democratic Peoples, this means that first-order State consent in international lawmaking is too reductive. From the point of view of the citizens, State consent to international law is given with a possible (and highly probable) first-order consent deficit, but one that can be assumed to be accepted within a democratic constitutional order accepting majority decision rule. But citizens can arguably assume that:

1. (i) first-order State consent masks large first-order majority dissent in case the population-strong States’ numerically large minorities who are discarded in the decision outnumber population-weak States’ numerically weak majorities who are decisive in the decision;

2. (ii) there is a large second-order consent deficit of citizens regarding the principle of consent, at least on some issues.

Demoicratic representation therefore minimally implies a principle of free consent of Peoples as opposed to States. The free consent of the People needs to be given to international obligations (treaties) by the People’s highest law-giving body, which can be a freely and equitably elected parliament or a direct vote of the citizens. But that is not where demoicratic representation can stop. Demoicratic representation needs common institutions in which representatives of Peoples address the second-order consent deficit about the principle of People-consent in relation to the positioning of the citizens regarding this decision rule when applied to specific laws and issues. Demoicratic representation cannot be solely committed to the principle of State or People-consent, nor can it simply abandon the principle of People-consent in favour of a majority decision rule across the board. The latter is democratic representation. Demoicratic representation needs to account for the citizens’ possible second-order consent deficit regarding the principle of People-consent in common institutions, in which the decision which decision rule ought to be adapted for what decision is debated, settled, and constantly re-adapted. In other words, demoicratic representation institutionalizes a rule of change that allows for switching from the decision rules of People-consent to majority voting in common institutions of the Peoples in international lawmaking.

(4) For IGOs the argument here presented has different normative implications depending on the type of IO. There is no room here for an in-depth typology of IOs. My normative argument proposes that the IOs of democratic Peoples ought to form a demoicratic system of representation. Not all democratic IOs are as or need to be as comprehensive as the EU, that is, engage in lawmaking, executive transnational government, and jurisprudence.Footnote ⁶¹ Ideally, such a comprehensive political IO ought to embody demoicratic representation within, that is to say, contain all the representative institutions of the Peoples. Less comprehensive IOs are not under this imperative of separation of power and completeness of representation. But they ought not to stand alone. They ought to be a complementary part of a representational system and thereby contribute to assuring demoicratic representation by IOs. As I propose to understand democracy and demoicratic representation systemically, there is thus no need that all IOs encompass all the institutions of the Peoples in one organization. Demoicratic representation can be realized if IOs are integrated in a coherent and transparent institutional system of separation of power and checks and balances upheld by the member Peoples.

6.1 Introduction

Both the object and the approach of this chapter are thought-provoking and in need of clarifications, not only with regard to the general theory of democracy but also, and especially, with regard to their own meaning in international institutional law.Footnote ¹ Applying the notions of ‘democratic representation’ and ‘parliamentarization’ to international organizations (IOs) is no small undertaking and carries its share of weighted connotations. To be precise, ‘democratic representation’ is not ‘representative democracy’, and, consequently, ‘parliamentarization’ is not a variation of democratic representation in the way that ‘parliamentarism’ is a variation of representative democracy.Footnote ² One must be wary of false friends and false pretences.

First, democratic representation is not the same as representative democracy. Unlike what the latter notion seems to suggest, democratic representation is not about apprehending IOs as political systems or forms of government. Nor is it self-evident to associate democratic representation with the governance (rather than government) of IOs since governance refers to the way in which international (normative) power is exercised in or through IOs. In any event, this association deserves further evaluation, and it is one of this chapter’s objectives. The European Union (EU) is an exception worth mentioning: its functioning is uniquely ‘founded on representative democracy’,Footnote ³ even if it does not escape duality. In the EU, representative democracy is – at the risk of falling sometimes into oxymoron, sometimes into pleonasm – direct in the European Parliament, and indirect in the Council – through governments ‘themselves democratically accountable, either to their national parliaments, or to their citizens’.Footnote ⁴

As for democratic representation in relation to IOs, this notion appears in the discourses on the democratic deficit of which IOs are accused under the combined effect of two phenomena. On the one hand, the end of the Cold War led to the universalization of democracy as a form of government and political system suitable for States. While the promotion of democracy in the States by IOs is not the object of this contribution, the fact remains that the generalization and idealization of democracy in States has spread to the institutions they have created in the international order.Footnote ⁵ On the other hand, and correlatively, contemporary globalization has been accompanied by an internationalization of the exercise of public power. In other words, more and more decisions affecting individuals are now taken jointly by States within IOs and the IOs participate in the exercise of international public authority and establish themselves as international governance bodies.Footnote ⁶ As IOs have become the seat of a true international (normative) power (understood as the elaboration process of international norms, including soft norms as they can affect individuals), they are subjected to democratic requirements. Without pretending to entrust the exercise of such power to an untraceable international people, the ‘democratization’ of IOs aims at bringing the human component of States closer to international governance bodies so that individuals may ‘understand’ themselves as the authors of the international norms that affect them.Footnote ⁷ In that sense, international democracy is ‘understood as an institutional form providing accountable channels of representation between people and governments, and producing policies based on this interaction’.Footnote ⁸ Democratic representation is one of the tools for the ‘democratization’ of IOs. It is also one of the most classic, even if its modalities have evolved and taken different shapes.

Applied to IOs in the context of globalization, democratic representation can be defined as the legal mechanisms and techniques that make the voice of individuals heard and make individuals present in the functioning of IOs, particularly in the elaboration process of international norms (including soft norms) of which IOs are the crucibles. As a result, democratic representation is a driver in the movement to reform the law of IOs into a new law of international institutions.

Indeed, if IOs are ‘ultimately, nothing more than the “personification of a representative system”’, that system is inherently ‘intergovernmental’.Footnote ⁹ Traditionally, IOs make (Member or non-Member) States present in their functioning – including in decision-making processes – through their governments. The fact that some IOs have extended membership to non-State entities has altered the principle of the governmental representation of States, but it did not call it into question, even if this principle has known some long-standing exceptions.Footnote ¹⁰ Traditionally, the governmental representation of States assured the legitimacy of the power exercised by IOs, including a form of indirect democratic legitimacy when State governments are democratically elected. This perception remains relevant, as evidenced by a report of the Swiss Federal Council on the ‘democratization of the United Nations’ and according to which ‘[d]emocratic UN member States, such as Switzerland, have a democratically legitimate government, and therefore also a democratically legitimate representation at the UN’.Footnote ¹¹ After noting that the United Nations (UN) Charter leaves no room for parliaments in the UN structure, the Standing Committee on United Nations Affairs of the Inter-Parliamentary Union (IPU) took a qualitative leap and concluded that the ‘“Peoples of the United Nations” are represented by governments, and not by the representatives they have elected’.Footnote ¹² Beyond the governmental representation of States, there would also be democratic representation by governments.

In addition to the looseness of the representative link with individuals,Footnote ¹³ not all State governments enjoy democratic legitimacy. By introducing parliaments – ‘the cornerstone of functioning democracies’Footnote ¹⁴ – into the machinery of IOs, parliamentarization aims precisely at correcting the democratic limits of the governmental representation of States … without necessarily coinciding with democratic representation.

Next, the parliamentarization of IOs claims to involve national parliaments to a greater extent, whether as such, through their members, or through the institutions that convene them, first and foremost the IPU. The plural character of this parliamentarization is apparent from the outset, at the level of both its actors and its modalities, given that this ‘involvement’ covers a whole range of mechanisms, instruments, and techniques whose nature varies: representative, consultative, participatory. Understood like this, the parliamentarization of IOs gets closer to related notions without being confused with them.

First, the parliamentarization of IOs is distinct from the parliamentarism of political theory, which, as a form of representative government, is based on the elective principle. Because of the ‘particularly great distance separating individuals from international governance institutions … the electoral technique allowing the governed to directly choose their rulers remains logically marginal at the universal level, even if it is carried out for the appointment of MEPs [Members of the European Parliament] who share the exercise of legislative power with the EU Council’.Footnote ¹⁵ The parliamentarization of IOs largely ignores the elective principle, including when it rests upon the representative technique.

Second, the parliamentarization of IOs differs from parliamentary diplomacy by its modalities and actors, even if it also contributes to bridging the democratic deficit of international relations.Footnote ¹⁶ On the one hand, the parliamentarization of IOs only considers the involvement of parliaments within IOs, and not the role that these parliaments may play, at the national level, on their governments’ international action. On the other hand, the parliamentarization of IOs encompasses the parliamentary world lato sensu, that is to say, beyond national parliaments. In addition to national parliaments and their members, the parliamentarization of IOs also concerns international parliamentary institutions, which include international parliamentary bodies – still known as international parliaments – and interparliamentary assemblies.Footnote ¹⁷ The former are organs of an IO, especially at the regional level; the latter are created by agreement outside any IO, but remain established in connection with and/or under the aegis of an IO, when they are not themselves constitutive of an IO. The IPU is an ‘international organization sui generis, that is, it is an international parliamentary, political and representative organization’.Footnote ¹⁸ International parliamentary institutions appear to be a decisive instrument for the parliamentarization of IOs because they bring about the (sometimes personified) institutionalization of international parliamentary representation. Yet, with such definitions, they mostly feed the democratic ambivalence of the parliamentarization of IOs and the confusion between parliamentary presence and democratic representation.

Finally, although parliamentarization seeks the democratization of IOs, the extent to which parliamentarization advances democratic representation within these IOs deserves further discussion. In other words, while the parliamentarization of IOs may contribute to reducing the distance separating individuals from international governance bodies, it is not necessarily by making individuals or their interests present in the exercise of the international normative power of which such bodies are the seat. From this perspective, in the era of globalization, the parliamentarization of IOs feeds into the democratic illusion – ‘to qualify this form of representation without reality’Footnote ¹⁹ – which ultimately excludes individuals from decision-making. In particular, the parliamentarization of representation does not necessarily mean the democratization of the representative link in IOs. False friends and false pretences reappear.

To better identify these false friends and false pretences, it is necessary to separate, on the one hand, the question of who represents and how, and, on the other, the question of what is represented, and why. Do the parliamentary modalities of representation (who represents and how) in IOs give them a democratic object (who is represented and why)? The plurality of the former precludes a clear-cut answer. The duality of international parliamentary institutions and the diversity of their relations with IOs are worthy of special attention and call for distinguishing what is represented in and by these different international parliamentary institutions and their members. The measure of democratic representation found in IOs is clearly not the same in these two hypotheses.

On the one hand, representation in international parliamentary institutions reveals that the parliamentary representative can be a false friend of the democratic representative in IOs (Section 6.2). On the other hand, representation by international parliamentary institutions or their members is often a mere false pretence of democratic representation in IOs despite some actual democratic virtues for their functioning (Section 6.3).

6.2 The Realities Represented in International Parliamentary Institutions: The Parliamentary Representative, a False Friend of the Democratic Representative?

This section focuses on what international parliamentary institutions members ut singuli make present within them. From this viewpoint, the very composition of international parliamentary institutions suggests that representation is double, if not dual. It stems from the statutory duality of their members due to their quality, but also to their designation modalities. In this respect, States’ human component is obviously present in these institutions if their members are elected by direct universal suffrage. But in practice, this configuration remains exceptional for international parliamentary organs, and is never realized for interparliamentary assemblies, which can also admit parliaments as members (Section 6.2.1). Thus, the most frequent hypothesis remains that of the ‘dual mandate’ of the members of international parliamentary institutions, which raises questions about the entity on whose behalf they speak (Section 6.2.2).

6.3 The Realities Represented by International Parliamentary Institutions: False Pretences of Democratic Representation in IOs?

On whose behalf do IPIs speak? And, consequently, whom do they make present in the IOs to which they are functionally linked, when they are not an organ thereof? The answer to this question overlaps, at least in part, with that of representation in the international parliamentary institution when the international parliamentary institution is a component of the organic apparatus of an IO.Footnote ⁶⁰ But it is not perfectly identical, a fortiori when the relationship with the IO is only functional. In any event, the representative link with individuals is significantly loosened, and the gap widens even further in the hypothesis of interparliamentary assemblies jointly established by some of these organizations. The IPU and its various relationships with intergovernmental organizations (first among them the UN) illustrate the distention and the multiplication of the relationship with individuals, a relationship which then has only the appearances of a representative link (Section 6.3.1). It therefore seems that parliamentary involvement in IOs allows for the adjustment of democratic principles to these organizations, rather than a true democratic representation. In other words, IPIs encourage and promote a more inclusive, transparent, and accountable functioning of IOs, but without necessarily making the voice of individuals heard (Section 6.3.2).