Introduction
This article explores the links between patterns of stratification and the form and practice of institutions in international society. International society is characterised by complex, multidimensional patterns of stratification in the sense of a variety of differences and inequalities – for example, in power, in prestige, in wealth, in culture, etc. – which give rise to various forms of ranked orders among states. As the English School emphasises, international society practises two basic approaches to the relation between its stratification and its institutions: The first approach – epitomised by the fundamental principle of sovereign equality – opts to treat the members equally irrespective of stratificatory differences, that is, to delink stratification and institutions. The second approach – epitomised by the primary institution of great powers – opts for a differentiated allocation of rights and duties, deeming some stratificatory differences relevant for the functioning of collective action, and thus interlinks stratification and institutions.Footnote 1 These links are based on different stratificatory understandings and dimensions in different institutions. For instance, while the United Nations Security Council is underpinned by the differentiation between great powers and other (smaller) powers, the World Trade Organization allocates trade privileges to developing and least developed countries and thus applies an alternative, economic interpretation of the patterns of stratification. This article focuses on the level of secondary institutions – that is, international regimes and organisations – to study how distinct understandings of stratification have been mobilised and institutionalised to mould select institutional characteristics of international organisations.Footnote 2
The article aims at contributing to the research on the links between inequalities and institutionsFootnote 3 by proposing an analytical framework for studying these stratificatory interlinkages and by exploring their variety through an explorative investigation of interwar international institutions. It makes two main claims: Firstly, it points to the importance of stratificatory semantics and understandings and in particular the juxtaposition of and competition between different understandings of stratification. This claim ties in with more recent works on shifting prevalent understandings of stratificationFootnote 4 and draws on Bourdieu's notion of struggles over classifications. The complexity and multidimensionality of the patterns of stratification necessitates decisions on how to link institutional characteristics to these patterns. These decisions entail competing interpretations not only of individual ranking positions but also of the most appropriate classification schemes, especially when actors are positioned differently in different stratificatory dimensions. The research on the links between inequalities and institutions still mostly neglects or brackets these struggles. Rather than tracing the different stratificatory interpretations at play, the research usually either concentrates on the tensions between sovereign equality and one stratificatory understanding (for example, grading of powers, gradations of sovereignty)Footnote 5 or brackets the stratificatory understandings of the actors by studying these links via correlations between the distribution of capabilities, often modelled as the number of major powers, and institutional characteristics.Footnote 6
Secondly, the article argues that the variety of stratificatory interlinkages – with respect to the underlying classification schemes, forms, and types of the interlinkages and their legitimacy – has so far been inadequately grasped. The research still predominantly conceptualises the links between inequalities and institutions as the institutionalisation of power – that is the transformation of power differentials into authority relations – and relatedly as the institutional endorsement of positional differences within status or prestige hierarchies.Footnote 7 This leads to a focus on voting rights, on the privileges of the powerful, and on rankings based on power differentials. Yet, while these aspects are certainly highly important, stratificatory interlinkages are much more varied. As the case study will show, the League of Nations used a classification scheme different from the grading of powers to determine the contributions each member had to make to the budget. Stratificatory interlinkages thus serve a variety of purposes and differ both within and from institution to institution. Furthermore, some interlinkages seem to be more legitimate than others, with smaller states even arguing for inequality rather than equality for some institutional characteristics.
Building upon the English School and theories of stratification, the proposed analytical framework centres on stratificatory understandings and interlinkages. A stratificatory interlinkage is established when the relevant constituency of actors tacitly or explicitly agrees to mould certain institutional characteristics in light of some stratificatory considerations and understandings. Debates and negotiations about stratificatory interlinkages comprise two kinds of decisions: firstly, a decision for a coupling – instead of a decoupling – of institutional characteristics to (some aspects of the broader) patterns of stratification and, secondly, a decision for a specific classification scheme and type of interlinkage. These decisions, conceptualised as struggles over classifications, involve both the selection and the modification of existing understandings of stratification and thus constitute a two-way interplay between institutions and patterns of stratification.
The interwar period is especially suited for an explorative study of stratificatory interlinkages mainly for two reasons: Firstly, the privileges of the great powers were for the first time institutionalised in a formal and supposedly global managerial organisation, the League of Nations, while at the same time the Hague Conferences of 1899 and 1907 were interpreted by interwar scholars as the beginning of a new era in which ‘international organization passed in form from oligarchy to democracy’ testifying to a long-term ‘process of equalization’ in international society.Footnote 8 Secondly, interwar international society additionally featured a highly stratified international order that comprised a sphere of sovereign states alongside several colonial empires and that was still shaped by the standard of civilisation. International institutions therefore faced the challenge to accommodate different strata of actors – for example, states, dominions, and colonies – within their institutional frameworks.
The article proceeds in five steps: The first section takes stock of the research on the links between inequalities and institutions. The second section develops and discusses the analytical framework. The third section presents brief case studies of three important stratificatory themes in the League of Nations: namely (a) the special institutional positions for great powers and later middle powers; (b) the treatment of states, dominions, and colonies; and (c) the debates about budget contributions. The fourth section then further develops the analytical framework by proposing four analytical dimensions that make visible the variety of types of stratificatory interlinkages: the perceived composition of the group of actors that is classified, the decision-making procedures that determine the classification scheme, the institutional purposes for which stratificatory interlinkages are used, and the institutional forms of these interlinkages. The variety is illustrated with examples drawn from the League as well as additional interwar institutions. The conclusion finally summarises the main findings and discusses further avenues for research.
Stratification and institutions in international society
This article examines the links between patterns of stratification and institutions. It brackets the broader problematique whether certain patterns of stratification such as a balance of power facilitate the working of international institutions and concentrates on those links created deliberately in order to institutionalise some stratificatory aspects, that is, to align certain institutional structures and norms to some aspects of the broader patterns of stratification. On the whole, the existing research in IR has mainly conceptualised these links as the institutionalisation of power and authority. In recent years, broader and more refined approaches were developed that emphasise the multidimensionality of patterns of stratification and the two-way interplay between these patterns and institutions.
The widespread conceptualisation as the institutionalisation of power and authority arguably derives from the prevalent understanding of hierarchy in International Relations (IR). In this understanding, hierarchy is understood as the opposite of anarchy and as the presence of authority.Footnote 9 The difference between anarchy and hierarchy is then not the existence of inequalities between actors but rather the institutionalisation of these inequalities in form of authority relations. Scholars have accordingly investigated episodes in which powerful states successfully transform their preponderance in (military-political) power into authority relations and institutional privileges such as permanent seats and veto power. This narrow perspective largely informs several variants of the research on the links between inequalities and institutions. A first variant approaches the links as deviations from sovereign equality, which is treated as a foundational principle of international society, and consequently emphasises the tensions between this principle and the principle of inequality underpinning institutionalised hierarchies.Footnote 10 A second variant, often prevalent in statistical studies on institutional design, abstracts from the political debates and analyses the links via correlations between variables relating to the distribution of capabilities and/or number of major powers and variables relating to institutional characteristics.Footnote 11 A third variant, often found in the research on status, analyses the links via the matches and mismatches between status hierarchies and institutionalised hierarchies.Footnote 12 However, all three variants generally analytically privilege one particular stratificatory dimension (political-military power) and one particular institutional characteristic (special privileges and the competences of the powerful). They thus neglect to systematically study how different understandings and dimensions of stratification have been debated and selected as base for institutionalised hierarchies.Footnote 13
While often staying within the above outlined narrow perspective, the English School has nevertheless developed a more nuanced approach to the links between inequalities and institutions. It emphasises that the prevalent understandings of stratification change over time. Notably, the emergence of the institution of great powers in the early nineteenth century was related to the displacement of an ‘old hierarchy of inherited status and precedent’ by a ‘new doctrine of the hierarchy or grading of states’.Footnote 14 As Edward Keene pointed out, this shift from an understanding based on individual ranks of princes to one based on classes of powers represented a crucial condition of possibility for the practices of collective interventions and collective management that are characteristic for great power management.Footnote 15 Stratificatory understandings thus, to a certain extent, both enable and predefine possible institutional forms and practices. Furthermore, the English School has highlighted the salience of another stratificatory understanding, namely the standard of civilisation, that deeply shaped the expansion and globalisation of international society by ascribing full membership rights in international society only to those polities deemed civilised and sovereign.Footnote 16 The English School has, however, mostly studied these links at the more general level of primary institutions and international order rather than on the level of secondary institutions.
The renewed interest in hierarchies in IR has generated broader, often more sociological conceptualisations of hierarchies as forms of social ordering of actors into superior or inferior positions according to some socially important aspects.Footnote 17 These broader conceptualisations stress that social hierarchies are ‘neither immanent nor self-evident, but historically contingent and socially defined’Footnote 18 and thus underscore the importance of stratificatory understandings. For the study of links between inequalities and institutions, two broader themes are particularly relevant: On the one hand, these broader approaches caution against reducing stratification to the distribution of military-political capabilities and point to the multidimensionality and complexity of patterns of stratification in international society. Importantly, different social spheres may be shaped to different degrees by different dimensions of social stratification.Footnote 19 On the other hand, these broader approaches emphasise that institutions play an important role in the interpretation, fixation, and (re)production of patterns of inequalities. They therefore argue for shifting the analytical focus to the institutional debates and mechanisms that reshape understandings about (status) hierarchiesFootnote 20 and contribute to the reproduction of global inequalities.Footnote 21 While these recent approaches establish a productive base for investigating the links between stratification and institutions, they have so far remained mostly on a conceptual level or analysed aspects other than the variety of designed links between patterns of stratification and institutional characteristics of secondary institutions.Footnote 22
Analytical framework: Stratificatory understandings and interlinkages
The analytical framework outlined in this section builds upon English School theorising and sociological theories of stratification. While also being applicable to the level of primary institutions (that is, fundamental sets of practices and rules like diplomacy, the institution of great powers and international law), the analytical framework is primarily developed for the level of secondary institutions (that is, international organisations like the UN and the WTO). The underlying rationale is that the variety of political mobilisations of stratificatory understandings and stratificatory interlinkages is more pronounced and the deliberate decisions more visible on the level of secondary institutions. Debates and decisions on the incorporation of stratificatory aspects in institutional characteristics of secondary institutions are an important site (among others) through which international society produces and changes prevalent understandings about its patterns of stratification and their relevance for the practice of international society.
International society – as do all forms of social systems – displays patterns of inequalities in the distribution of various material and immaterial goods (such as power capabilities, wealth, status, reputation, size, etc.). Patterns of stratification then denote the fabric of social rankings that order actors or groups of actors according to some socially important aspects – for example, certain goods or performances – into superior and inferior social positions.Footnote 23 Patterns of stratification are constituted by a juxtaposition of material and social factors in the sense that they result from intersubjective processes and mechanisms which define certain goods and performances as more valuable/desirable than others and shape the allocation of these goods and the rewards for these performances. As sociologists like Pierre Bourdieu and Max Weber emphasise, these patterns consist of several stratificatory dimensions, which are valued differently in different social situations.Footnote 24 Depending on the social system, both high and low correlations between the positions of actors in different stratificatory dimensions are possible. For instance, in present international society, China is in some situations treated as a developing country, in others as a rising power and in still others as an established power (for example, in the debate about the UN Security Council reform). The decisive analytical question is therefore how the different stratificatory dimensions relate to one another, that is, how they ‘overlap, crosscut, and compete’Footnote 25 – and how, in this interplay, distinct stratificatory dimensions and understandings become more important than others for the self-descriptions, practices, and institutions of international society.
The present article proposes the analytical concept of ‘stratificatory interlinkages’ to grasp deliberate links between the patterns of stratification and institutions. An interlinkage attempt occurs when an actor, trying to alter an institution according to his or her interests and aims, promotes certain stratificatory considerations and arguments as relevant for the form and practice of the institution. An interlinkage attempt may be countered by alternative interlinkage attempts (emphasising different stratificatory considerations and understandings) as well as contestations of such interlinkages. A stratificatory interlinkage is established when the relevant constituency of actors tacitly or explicitly agrees to mould the form and practice of the institution in light of some stratificatory considerations and understandings.Footnote 26 A stratificatory interlinkage consists of two kinds of decisions that are empirically generally intertwined:
(1) Coupling vs decoupling: The first decision is to choose a coupling of the institution to patterns of stratification over a decoupling. In an English School perspective both of these contrasting logics are at work in international society. The first logic, decoupling, is epitomised by the principle of sovereign equality. It stipulates that rights and duties are to be allocated (equally) without regard to the various material and immaterial inequalities between the members of international society. This amounts to a decoupling of allocations and institutions from patterns of stratification. In contrast, the second logic maintains that the self-management of international society, at least for some purposes, requires a differential allocation of rights and duties. In this sense, allocations of rights, duties, and roles are coupled to some valued stratificatory aspects. The prime example is the allocation of special rights and duties to the great powers. As a caveat, it should be noted that, depending on the reference group, one and the same arrangement may be identified both as decoupling and coupling. For instance, within the state membership of international society sovereign equality functions as a logic of decoupling whereas seen from the broader standard of civilisation the principle of sovereign equality functions as a form of social closure, linking full membership rights to a certain stratum (that of sovereign states).Footnote 27
(2) Selection and elaboration of stratificatory classification scheme(s): The second decision revolves around the selection of some stratificatory aspects and understandings as more relevant than others for the institution and the elaboration of a classification scheme specifying how the institution is coupled to patterns of stratification. As stratificatory understandings are often at least to some degree vague, and exact classifications of at least some actors contested, this act of decision involves more than a simple selection of pre-existing classification schemes. Rather, while drawing on pre-existing shared understandings, it is likely to be characterised by ‘struggle[s] over classifications’,Footnote 28 both in the sense of the appropriate (combination of) classification scheme(s) and in the sense of the exact classification of individual actors. Stratificatory interlinkages may adopt institutional classification schemes that only partially mirror the stratificatory understandings to which they are linked (for example, the bipartite distinction between great powers and non-great powers, which is informed by the tripartite status hierarchy of great, middle, and small powers). To a considerable degree, coupling decisions thus (re)interpret, ascertain, and/or alter shared stratificatory understandings.
The variety of stratificatory interlinkages – as well as the choice for forms of decoupling and coupling more broadly – varies both across institutions and within (some) institutions. To give an example: The UN General Assembly embodies the logic of decoupling while the UN Security Council epitomises the logic of coupling. Insomuch as the various stratificatory interlinkages (re)interpret and refine stratificatory understandings, the interplay between stratificatory understandings and institutions is a two-way interrelation: Institutional debates are both influenced by stratificatory understandings and contribute to the further development of these understandings. The stratificatory interlinkages of key institutions – like the League of Nations – are likely to play a more influential role than those of less important institutions for the prevalence of particular stratificatory understandings in international society.Footnote 29
Several factors are likely to influence the debates and decisions on stratificatory interlinkages. Among these factors are especially: A first factor is the situational configuration of power as actors with more favourable positions within existing patterns of stratification and power are probably more successful in promoting their preferred interpretations of stratification and their preferred forms of stratificatory interlinkages than actors with less favourable positions. This factor accentuates the role of power (inequalities) and conflicts in attempts of actors to influence and alter the fabric of institutions and the existing patterns of control over societal functions.Footnote 30 A second factor are the prevalent legitimacy principles in international society. As the English School emphasises, the institutions of international society are underpinned by (temporary) political agreements regarding appropriate and acceptable forms and practices of institutions.Footnote 31 Sovereign equality and the principle that multilateral treaties require the consent of all participants are thereby important principles that pre-structure debates and decisions on stratificatory interlinkages. A third factor is the set of prevalent understandings of stratification. The debates and negotiations take place in social contexts that are already shaped by understandings of stratification, some of which are more prevalent and important than others and therefore more likely to be accepted as a suitable interpretation of patterns of stratification if mobilised by an actor. Analytically, the relation between these factors – as well as the potential role of additional factors – is treated as an open empirical question and as a relation that might change over time.
Two further aspects merit attention: Firstly, the proposed analytical framework accentuates the relevance of stratificatory considerations and understandings. Nevertheless, stratificatory arguments are but one kind of possible arguments. Examples of other kinds of arguments are prevalent legitimacy principles (such as sovereign equality), practical considerations (for example, about the feasibility and effectivity of certain institutional arrangements) and alternative logics of social organisation and differentiation (for instance, an allocation of seats not to strata but regions).Footnote 32 For that reason, these debates and negotiations do, secondly, not necessarily take place between groups defined in stratificatory terms. Rather, a conflict constellation that opposes ‘great powers’ and ‘smaller powers’ is itself already the result of prior / ongoing processes in which distinct understandings of stratification have become more important for conflict constellations within international society than other stratificatory and non-stratificatory understandings (such as ‘East’ vs ‘West’).
Stratificatory considerations and interlinkages in the League of NationsFootnote 33
This section presents brief case studies of three institutional characteristics to illustrate the variety of stratificatory understandings and stratificatory interlinkages incorporated in arguably the most important secondary institution of the time, the League of Nations.Footnote 34 The first characteristic is the differentiated membership of the Council that was informed by understandings about the grading of powers. The second characteristic is the treatment of the status differences between states, dominions, and colonies. The Covenant created a certain equalising trend by formally granting all three groups equal treatment in the League. At the same time, the selective admission of colonies as well as the mandates system remained clearly influenced by and further institutionalised prevalent understandings about the standard of civilisation. The third characteristic relates to the budget and the apportionment of contributions to the individual members of the League. The initially chosen classification scheme was soon criticised as being unjust vis-à-vis the smaller members, which resulted in a reformed classification scheme featuring a more unequal apportionment of contributions.
The differentiated membership of the Council of the League of Nations
The membership of the Council was tensely debated during the drafting of the Covenant.Footnote 35 While most accounts focus on the tensions between sovereign equality and the claims of the great powers for special privileges, these debates also featured a second layer of controversies relating to the question of how the grading of powers was to be adapted to the differentiation of the membership in the Council. Notably, some of US President Woodrow Wilson's drafts translated the grading of powers into a tripartite differentiation allocating distinct seats to the groups of great, middle, and small powers. According to one draft, the Council
shall consist of the representatives of the Great Powers, together with representatives drawn in annual rotation from two panels, one of which shall be made up of the representatives of the States ranking next after the Great Powers and the other of the representatives of the minor States.Footnote 36
In his second draft, Wilson had already noted, that ‘[a]ll would now agree that there are five great powers’ but that the further classification of states in middle and small powers was ‘likely to cause discord’ and was therefore better to be delegated to the future Assembly.Footnote 37 Wilson's drafts drew upon the plan of Jan Smuts, a South African military leader participating in the British preparations, who had proposed the same design: a Council of nine members, drawn from three ranks with a small majority of one vote for the five great powers United States, Great Britain, France, Italy, and Japan.Footnote 38 This scheme, Smuts reasoned, would secure the great powers from ‘being swamped by the multiplicity of small states’ and at the same time give the middle and small powers ‘a very substantial representation’.Footnote 39
In the joint British-American preparations for the Paris Peace Conference, however, Wilson finally agreed to the plan of Lord Cecil to propose a Council composed of solely the five great powers to which other powers could be invited depending on the issue at hand. The resulting draft, the so-called Hurst-Miller draft (of 3 February 1919), was strongly contested by the smaller states, which were discontent with their minor role both in the Paris Peace Conference and the draft covenants. Their protests resulted in an enlarged League commission and an eventual return to Wilson's and Smuts's model of a council of five great powers and four lesser powers. The non-great powers principal ‘equality’ claim centred more on the 5:4 ratio between great powers and non-great powers in the Council rather than on the critique of great power privileges as such. Their aim was to secure a fair representation and to forfeit what the Belgian representative Hymans called a new ‘Holy Alliance’:Footnote 40 a council in which the great powers alone decided on all important matters. The League commission thus followed Wilson's idea of a council composed of both great and small powers, but devised a two-tier classification scheme (distinguishing between great and small powers) instead of Wilson's proposed three-tier classification scheme (that had effectively distinguished between great, middle, and small powers).
Although finally dismissed, two group-related arguments pervaded the discussions on the election mode for the non-permanent Council members. On the one hand, several non-great powers complained that the great powers would choose the permanent members (namely themselves) and additionally via the Assembly be involved in the selection of the non-permanent members. They consequently demanded that the group of non-great powers, and not the whole Assembly, should constitute the electorate. After remarks by Hymans and the Serbian delegate Vesnitch that more suitable candidates would be selected if the great powers participated in the election, the commission adopted that the whole Assembly would choose the non-permanent Council members.Footnote 41 On the other hand, some countries followed Wilson's original ideas insomuch as they proposed classifications within the group of non-great powers to safeguard that all states have comparable chances to be elected to the council. While Belgium thought of a seat-allocation to the groups of middle and small powers respectively, Sweden talked about ‘classes or groups’ and proposed the idea of an election system allocating non-permanent seats to regional groups.Footnote 42
The Covenant was unanimously adopted as part of the Versailles Treaty on 29 April 1919. While the membership in the Council was shaped by a stratificatory interlinkage with a two-tier understanding of the grading of powers,Footnote 43 the Assembly was informed by a purposive decoupling that accorded all members the same rights in the Assembly. Both organs normally decided on the basis of unanimity, which was seen as a symbol of sovereign equality. Although the Covenant did not stipulate any criteria apart from annual elections by the Assembly, the first Assembly's First Committee eventually decided in November 1920 to elect the non-permanent members by majority vote based on regional groups, that is, based on segmentary rather than stratificatory considerations.Footnote 44
The Council Crisis of 1926 then led to an important revision of the Council design that introduced a new category of semi-permanent seats for select middle powers.Footnote 45 Germany applied to enter the League as a permanent member of the Council, which was welcomed as an important opportunity for overcoming the League's image as the club of the victorious powers of the last great power war. Yet, Germany's application sparked additional bids for permanent membership by several states, including Brazil, Spain, and Poland. In reaction to the non-recognition of its bid, Brazil vetoed Germany's admission, thus causing an institutional crisis. A Commission on the Composition of the Council was subsequently convened to resolve the crisis. The commission granted permanent membership only to Germany, reasoning that permanent membership was linked to great power status, which only Germany was recognised to have. It thus reaffirmed the stratificatory interlinkage established at the founding of the League. At the same time, the commission sought to meet both the demands of the additional candidates for permanent membership as well as the demands of smaller powers for better representation by reworking the system of elected seats on the Council. It replaced the existing re-eligible seats (six since 1922) with two new categories of non-permanent Council seats: three re-eligible, semi-permanent seats intended for Brazil, Spain, and Poland and six non re-eligible seats intended to provide for the rotational representation of smaller powers. However, discontent with the non-recognition of their bids, Brazil and Spain announced their withdrawals from the League (though Spain eventually remained a League member). Nonetheless, the Council reform thus effectively expanded the classification scheme underpinning the distribution of Council seats from a two-tier differentiation (between great and smaller powers) to a de facto three-tier differentiation (between great, middle, and small powers).
The treatment of the differences between states, dominions, and colonies
The grading of powers was not the only stratificatory understanding that shaped the debates about membership and membership rights in the League of Nations. The distinction between states, dominions, and colonies and the underlying standard of civilisation likewise played an important role. This distinction became relevant in several contexts: firstly regarding the nature and scope of the political entities that could join the League, secondly regarding the possible limitation of Council membership to states and thirdly through the establishment of the mandates system of the League.
The founders of the League of Nations decided to not limit membership to states but to admit any ‘fully self-governing State, Dominion or Colony’ (Article 1) as a member.Footnote 46 While this formulation reaffirmed the prevalent classification scheme differentiating between types of polities (for example, states and colonies), the provision nevertheless levelled the differences between these types of polities by giving every member of the League full membership rights. Membership rights were thus decoupled from the political and legal status of the members. This decoupling undercut the principle of sovereign equality insomuch as equality was not limited to ‘sovereign states’ as a stratum and insomuch as the Covenant envisaged a regime of equality which, at least formally, treated colonies in the same way as states within the League of Nations.
In practice, however, this regime of equality was strongly qualified by political considerations and prevalent understandings about the standard of civilisation (which, inter alia, surfaced in the notion of ‘fully self-governing’). In fact, the ‘states, dominions, colonies’ membership clause more or less constituted a special provision for the admission of British dominions and colonies. Four of the then five British dominions (Australia, Canada, South Africa, and New Zealand) and one British colony (India) became founding members of the League,Footnote 47 but no colony from another colonial power. This led to criticism that the Covenant privileged Great Britain, which formally entered the League as the ‘British Empire’, by giving it six votes. This criticism was also voiced in the US Senate debates about the ratification of the Versailles Treaty.Footnote 48 Tellingly, while all other original members were listed alphabetically at the end of the publication of the Covenant in the League's Official Journal, the four dominions and one colony were treated differently and listed as an indentation directly under the ‘British Empire’.Footnote 49
Over the course of the League's history, only one other British dominion (the Irish Free State in 1923) and no further colonies were admitted as new members. The membership of the dominions and India were therefore considered as anomalies – an American legal adviser later referred to India's League membership as ‘an anomaly among anomalies’Footnote 50 – and did not lead to a more general opening up of the League membership to colonies. That said, despite the peculiarities of its implementation, the ‘states, dominions, colonies’ membership clause of the League followed the example of several nineteenth-century international organisations, which had not limited their membership to states but also admitted colonies. While adding the differentiation between dominions and colonies, the ‘states, dominions, colonies’ membership clause was thus otherwise part of a broader trend of institutional arrangements that attempted to accommodate both states and colonies as members creating distinct problematiques such as the question of voting rights for colonies (which will be discussed in more detail in the next section).
The equal treatment was, though, not envisaged in all drafts of the Covenant. Notably, some drafts limited the eligibility for non-permanent seats on the Council to ‘states’. The British dominions protested that such a provision would give them fewer rights than the ‘state’ members of the League. Great Britain, whose own proposals had included the ‘states’ wording, consequently requested to replace ‘states’ with ‘members of the League’ to which the US agreed on 23 April.Footnote 51 In 1927, Canada was elected as the first dominion to the Council and thereafter ‘the dominions were in effect accepted as constituting an electoral group’,Footnote 52 which guaranteed one non-permanent seat for the group of dominions. All in all, therefore, the dominions eventually became the only stratificatorily defined electoral group in the otherwise region-based system of electoral groups for non-permanent seats. In contrast to the dominions, India was never elected as a non-permanent member of the CouncilFootnote 53 so that only two of the three categories of the membership clause (states and dominions but not colonies) were actually represented on the Council during the League's history.
The standard of civilisation was arguably most strongly institutionalised in the mandates system of the League of Nations.Footnote 54 The mandates system was devised in the negotiations on how to deal with the former German and Ottoman colonies as an institutional compromise that allowed the occupying powers to de facto keep these colonies and at the same time moved beyond traditional forms of colonialism by incorporating liberal internationalist ideas of the transformation of colonies to independent states.Footnote 55 Importantly, the mandates system implied that the former German and Ottoman colonies were not considered as annexed by the new occupiers but as being temporarily entrusted to them by the League as mandate territories (which were ranked into three categories according to their ability to govern themselves). The group of League members administrating the mandate territories included both states and dominions (namely Australia, New Zealand, and South Africa). The Covenant imbued the mandates system with a civilising mission by speaking of a ‘tutelage’, which advanced nations offered to polities deemed ‘not yet able to stand by themselves under the strenuous conditions of the modern world’.Footnote 56 While intended to perpetuate colonial domination by many imperialists, the mandates system in this sense also showcased the idea of the higher social strata in international society helping the lower social strata to socially move upwards. This idea – to help the mandate territories to acquire the capacity for self-governance and eventually become a member of the League – remained an ideal that was often not adhered to by the mandate powers but to which they had to pay at least lip service in their regular reports. The only mandate territory to actually become a member of the League of Nations was Iraq in 1932.Footnote 57
The apportionment of the contributions to the budget
During the negotiations on the Covenant, there was no substantial debate on the classification scheme to be used for allocating the expenses of the League budget to its members. In Article 6, the Covenant followed the widespread practice to adopt the budgetary classification scheme of the Universal Postal Union (UPU). The UPU classification scheme, developed in 1878, divided the UPU members into seven classes depending on their population, territory, and importance for the postal traffic.Footnote 58 The UPU's classification scheme thus combined stratificatory aspects relating to the general size of a country (population, territory) with a more issue-specific ranking factor (postal traffic). This classification scheme was afterwards adopted by several other institutions in the late nineteenth century, including the Permanent Court of Arbitration established by the Hague Peace Conferences of 1899 and 1907. By the Paris Peace Conference, it had become the model classification scheme for budgetary issues.Footnote 59
However, the League soon encountered various problems of implementing the UPU classification scheme, which sparked demands for a ‘fairer basis of apportionment’ (as the Secretary-General put it 1919).Footnote 60 Importantly, the stakes were much higher as the League had to apportion six million gold francs among 32 states while the UPU only assigned 125,000 francs among 81 members. According to the UPU scheme, each member of the first class had to shoulder 25 units and each member of the seventh class 1 unit of the overall budget. Besides classificatory problems – as not all League members were UPU members, and vice versa – the scheme was strongly contested, especially by the smaller powers, for two reasons: Firstly, according to the UPU formula, three of the dominions (Australia, Canada, South Africa), the only colony in the League (India) as well as China and Poland had to shoulder the same budget share as the five great powers, as they were all listed in the first class. Secondly, several smaller powers were assigned budget shares bringing them into financial difficulties or exceeding their financial capacities. In 1920, both the Council and the Assembly therefore decided to design a better classification scheme.
What followed was a protracted search for a more suitable classification scheme. The League had either to amend its Covenant or to persuade the UPU to change its classification scheme. As the latter proved unsuccessful, Article 6 was revised to give the Assembly the competence to determine the classification scheme in 1921. The amendment only came into effect in 1924. In the meantime, the negotiations continued and, lacking an agreement, the UPU scheme was reluctantly applied. The revised classification scheme increased the ratio between the highest and lowest unit share to 90:1. The guiding idea was to base the classification on a formula capturing the relative ‘ability to pay’ of each member, measured in terms of population and revenues corrected by certain additional factors. Accordingly, the composition of the groups was changed, placing only Great Britain and France in the highest group and China, India, Italy, and Japan in the second highest group. The Assembly eventually converged on a further reworked classification scheme with a range from 1 to 95 units. This scheme was adopted without dissenting voice in 1923 after additional allowances were made for war damages (which, inter alia, lowered the relative share of France, Italy, Romania, and Belgium) and other difficulties. After August 1924, the Assembly adjusted and fixed the classification scheme for each new budget interval on proposition of an allocation committee.Footnote 61
In this sense, the allocation of the expenses followed a different logic of stratificatory interlinkage than the differentiation of membership rights in the Council. Remarkably, the eventually rejected UPU scheme pooled in the highest class 11 League members that belonged to distinct strata in the other two dominant stratificatory understandings (grading of powers, standard of civilisation). The revision of the classification scheme shifted the form of the classification from seven clearly demarcated groups to a scheme still in principle ordered in form of groups but practically, due to the many correcting factors, resembling a scheme with individual ranks. This classification scheme provided for a differential treatment of the great powers. In other words, the allocation of special rights followed a different differentiation than the allocation of special (financial) duties. Perhaps more importantly, while in the first two cases, the smaller powers lobbied against unequal allocations, in the case of the League budget demands for a fairer and more equitable allocation were demands for an increased inequality between the smallest and the biggest contributors.
Four analytical dimensions of the variety of stratificatory interlinkages
This section further unpacks the variety of stratificatory interlinkages firstly by developing a framework of four analytical dimensions as a first step towards a more systematic mapping of this variety and secondly by illustrating this framework with examples from the League of Nations and select additional international institutions of the interwar period (chosen because they typify and/or best display certain manifestations within the four analytical dimensions).Footnote 62 The section underscores that the variety of stratificatory interlinkages has both intra-organisational and inter-organisational aspects. The framework represents a tentative approach to further refine the analytical tools based on empirical examples and insights from a particular historical period rather than an exhaustive list of all possible types of stratificatory interlinkages. Four analytical dimensions are distinguished: (1) the composition of the reference group that is classified for certain institutional purposes; (2) the decision-making process that assigns actors to different positions within a classification; (3) the links between classification schemes and institutional purposes / functions; and (4) the institutional form of stratificatory interlinkages. The framework is summarised in Table 1.
Four dimensions of the variety of types of stratificatory interlinkages.
Dimension 1: Composition of the reference group
The reference group – in the sense of the constituency of actors to be classified – may either be composed of actors of the same kind (that is, be homogenous) or be populated by actors of different kinds (that is, be heterogeneous). One decisive difference between the League of Nations and its successor, the UN, was that the former admitted actors of different kinds according to its Covenant (namely states, dominions, and colonies, see Article 1), while the latter envisaged only one kind of member in its Charter (namely states, see Articles 3–4). In its practice, though, the UN featured some deviations from this states-only principle.Footnote 63 The International Labour Organisation (ILO), founded in 1919, created an innovative new institutional structure based on a tripartite constituency consisting of state delegates, employer delegates, and trade union delegates with each group controlling one third of the vote.Footnote 64 While the ILO thus paved the way for the integration of non-state actors, the main interwar problematique regarding to the composition of the membership of international organisations related to the integration of dominions and colonies. The following two examples of the UPU and the International Telecommunications Union (ITU) show that: (a) the classification of actors as being of the same kind (postal unions) or of different kinds (states, colonies) was itself part of the debates and negotiations; and that (b) status levelling – that is, the decision to treat states, dominions and colonies equally – could in combination with majority voting actually result in increased inequalities in voting power in favour of colonial powers.
The broader phenomenon of colonial voting had already surfaced before the First World War when a number of colonies became members of technical international conferences and organisations.Footnote 65 Examples of interwar international technical organisations admitting colonies and protectorates as members were the Universal Postal Union, the International Telecommunications Union, and the International Institute of Agriculture.Footnote 66 The issue of colonial voting became a matter of conflict in several of these organisations. In the UPU, several states – among them the Soviet Union, Argentina, and Uruguay – criticised the practice of colonial votes as unjust privilege of the colonial powers. Yet, in 1929 a majority voted for retaining the voting system on the grounds that the UPU had postal unions and not states or colonies as members.Footnote 67
In the ITU, colonial votes were likewise contested. When the ITU was established in 1932, states such as the US and Italy proposed to limit membership to self-governing entities entitled to become League members, whereas several colonial powers, and some colonies such as, notably, the Dutch East Indies, pushed for voting rights also for colonies. The resulting compromise effectively granted each colonial power two votes (one for itself, one for the group of its colonies), awarded Germany and the Soviet Union an additional compensatory vote (to account for their ‘importance’), and additionally gave voting rights to the British dominions, India, the Dutch East Indies, as well as Morocco and Tunisia.Footnote 68 The 1938 ITU conference again featured debates between the colonial powers and proponents of a limitation of voting rights to sovereign political entities.Footnote 69 In this sense, the institutional debates responded to the unequal effects of the integration of colonies and dominions into international organisations by creating new regimes of equality and by pleas for limiting full membership rights to the group of sovereign states.
Dimension 2: Decision-making regarding classification of members
The decision of how exactly actors are ranked in a classification scheme is often an essential part of the struggles over classifications shaping the creation of stratificatory interlinkages. This dimension distinguishes between two basic modes through which actors are assigned their specific places within a chosen classification scheme: firstly by the decision of an authoritative body (authoritative classification) or secondly by the open choice for actors to choose their own rank (self-classification).
The authoritative body may either be the founding conference or a competent body designed by this conference. For instance, the permanent members of the Council of the League of Nations were designated by the Paris Peace Conferences, as were the first eight ‘states of chief industrial importance’ sitting in the Governing Body of the ILO. Subsequent revisions of the list of these eight states of chief industrial importance were thereafter made initially by the Council of the League and then by the Governing Body itself. The first eight permanent members were Belgium, Denmark, France, Germany, UK, Italy, Japan, and Switzerland. In the interwar period, the list was changed in 1922, 1935, and 1940. These changes temporarily resulted in permanent seats for a dominion (Canada) and a colony (India) and served to accommodate the entries of the US and the Soviet Union.Footnote 70 Further examples are international organisations, which implement abstract formula for the allocation of budget contributions and thereby calculate the quota for each member (such as the League of Nations). The ILO example – as well as the initial coupling of the allocation of budget contributions to the UPU scheme – furthermore illustrate that the competent body was in some cases an organ of another international organisation.
With self-classification are meant, in contrast, institutional settings that provide for a set of classes but give each member the choice to choose its own class. Such an arrangement was for instance practiced by the International Union for the Protection of Literary and Artistic Works.Footnote 71
Dimension 3: Classification schemes and institutional purposes
The third dimension relates to the relation between the selected classification schemes and the institutional purposes and functions of the institution (or the organ in question). On a basic level, one can distinguish between institutions employing different classification schemes for different institutional characteristics (intra-institutional variance) and institutions employing the same classification scheme for several institutional characteristics (intra-institutional congruity). As the case study in the previous section has shown, the League of Nations is an example for intra-institutional variance. It employed different classification schemes for the differentiation of membership rights for the Council and for the allocation of budget contributions. In contrast, several interwar institutions were characterised by an intra-institutional congruity insomuch as they used the same classification scheme for weighted voting and the allocation of budget contributions. An example is the International Institute of Agriculture, which used the same classification scheme, comprising five classes, for voting rights (giving 1 to 5 votes to the different classes) and for budget contributions (assigning 1, 2, 4, 8, and 16 units to the different classes).Footnote 72
In addition, the correlation between classification schemes and institutional purposes may also be studied through the comparison between different organisations. The relevant criterion is then whether the organisation employs classification schemes specifically tailored to its institutional purposes (purpose-specific schemes) or whether it uses classification schemes based on general criteria (general schemes). Purpose-specific schemes for instance underpinned the Sugar Regime (weighted voting and budget contributions depending on relative importance in sugar trade), the International Union for the Publication of Customs Tariffs (budget contributions linked to relative importance in international trade), and the ILO (states of chief industrial importance’ for permanent membership in Governing Body).Footnote 73 The UPU scheme, in turn, mixed general factors (size of population and territory) with a specific factor (volume of postal traffic).
The UPU scheme is also an example for an initially purpose-specific scheme that was later adopted by other organisations – notably the League of Nations, the International Bureau for the Protection of Industrial Property, and the International Office of Public Health – and thus in some sense became a general scheme.Footnote 74 The most important – because most prevalent – general scheme, however, was arguably the distinction between states and colonies, which informed debates about colonial voting (see dimension 1 above) and decisions to admit only states as members, that is, to make sovereign statehood the prime unit for classification schemes.
Dimension 4: Institutional forms of stratificatory interlinkages
This dimension concerns the institutional forms and means through which the interlinkage between patterns of stratification and institutional characteristics is established. One can broadly distinguish between fix links, which tie the institutional characteristics to a specific interpretation of the patterns of stratification and flexible links, which allow for reinterpretations and readjustments.
An example for a fix link between institutional characteristics and a given pattern of stratification is permanent membership in the Council of the League of Nations. While permanent membership in the Council was framed in the negotiations as a privilege of the great powers, the Covenant did not stipulate an abstract formula interlinking permanent membership and great power status but instead listed the names of the great powers at the time of the negotiations. In this sense, the institutional privileges were coupled to the grading of powers at the time of the Paris Peace Conferences and thereby decoupled from further changes in the stratification of international society.Footnote 75 The Covenant, though, allowed for the addition of new permanent members and – as discussed above – Germany was accordingly admitted as permanent member in 1926 (and later the Soviet Union in 1934).
In contrast, the allocation of budget contributions in the League of Nations represents an example for a flexible link. The allocation followed a classification scheme that was periodically readjusted to variations in the different members’ ability to pay. While this type of interlinkage holds the abstract formula constant (a classification based on the ability to pay), it allows for variations and adjustments in the exact classification and thus for a flexible adaptation of the interlinkage to shifting patterns of stratification. Likewise, a flexible link was employed for permanent membership in the Governing Body of the ILO. While Great Britain had initially planned to list the names of the permanent members, the ILO constitution reserved eight of the then twelve state delegation seats in the Governing Body for the ‘states of chief industrial importance’.Footnote 76 As already mentioned, the revision of these states of chief industrial importance was then entrusted to a competent body, which several times changed the composition of the group of states of chief industrial importance.
The Council Crisis of 1926 also produced an interesting type of interlinkage that enabled the Assembly to periodically switch between the logics of coupling and decoupling. The League of Nations created three re-electable semi-permanent seats for the middle powers and made all other non-permanent seats in the Council non re-electable. Although designed for three particular states (Brazil, Poland, and Spain) these special seats provided for the possibility of periodic revisions of the ranks of middle powers. At the same time, the Assembly had each time to decide whether the seats were re-eligible or not which enabled the Assembly to periodically review the coupling of these seats to patterns of stratification and, possibly, to suspend this coupling for some time.
Conclusion
This article approached institutionalised inequalities in international society as interlinkages between patterns of stratification and institutional characteristics. It showed that these interlinkages are characterised by a variety of types that has not yet been systematically addressed by the English School and IR more broadly, which still tend to equate institutionalised inequalities foremost with great power privileges, political authority relations, and politics of prestige.Footnote 77 In order to better grasp the variety of interlinkages, the article developed the analytical concept of ‘stratificatory interlinkages’. Stratificatory interlinkages were modelled as a twofold decision: firstly for a coupling – instead of a decoupling – of institutions to patterns of stratification and secondly for a particular stratificatory understanding, classification scheme, and type of interlinkage. Studies that focus on the tension between sovereign equality and institutionalised inequalities often tend to pay too little attention to the second of these decisions.
By highlighting the variety of stratificatory interlinkages, the article contributes to the unpacking of the ‘black box’Footnote 78 of institutional mechanisms of inequality (re)production in international society. The example of the League of Nations underscores that one and the same institution may employ different stratificatory understandings and types of stratificatory interlinkages for different institutional characteristics and purposes. The seats on the League's Council were allocated based on the classification of states in great, middle, and small powers. The League's membership principle was informed by the differentiation between states, dominions, and colonies. The allocation of budget contributions, in turn, was in the beginning linked to the UPU's classification scheme until the League devised its own classification scheme based on the ability to pay of its members. This economically framed stratificatory understanding cut across the other two stratificatory understandings, with India (formally still a colony) as well as some dominions paying more than many ‘state’ members, including some great powers.
As the empirical examples illustrate, the variety of types of stratificatory interlinkages has implications for the legitimacy and effects of institutionalised inequalities. Some stratificatory interlinkages are apparently considered to be more legitimate than others. In particular, the smaller states demanded a more – and not a less – unequal treatment of the League members in the budget negotiations. This observation also suggests that stratificatory interlinkages seem to be more readily accepted as legitimate – and even be demanded – by smaller states when they pertain to the allocation of duties (as opposed to the allocation of special rights).Footnote 79 Moreover, the effect of ‘setting status into stone’, which has been diagnosed for the UN Security Council,Footnote 80 is an effect of a specific type of stratificatory interlinkage rather than of all stratificatory interlinkages. The stratificatory interlinkages underpinning the ILO's Governing Body as well as the League's budget are examples of stratificatory interlinkages that allowed for periodical revisions and updates of the classification schemes coupling the respective institutional characteristics to the patterns of stratification.
Reconstructing the evolution of the variety of stratificatory understandings and interlinkages furthermore helps to identify and analyse overarching patterns and developments in the history of international institutions. The empirical analysis touched upon processes of diffusion of institutional arrangements for interlinking stratification and institutional characteristics. The UPU's scheme for budget contributions for instance served as model for other institutions, including initially also the League of Nations. Furthermore, the empirical analysis showed that the ‘states, dominions, colonies’ membership clause of the League was not simply an exception for the British Empire but part of a broader trend of admitting colonies as members to international organisations that had already begun in the late nineteenth century and was still a prominent – and disputed – problematique of institutional design in the interwar period.
Three further avenues for research emerge from these arguments and findings. Firstly, this article explored the variety of stratificatory interlinkages only within one historical period. In order to generate a more comprehensive account of how stratificatory interlinkages change and vary over time, the research on stratificatory interlinkages has to be expanded to other historical periods. Secondly, this expanded investigation can then serve as base for a more sophisticated analysis of the factors that account for the variances within and between different international organisations. The article identified several factors that played a role in the interwar period, notably the configuration of power, the prevalent legitimacy principles, the prevalent understandings of stratification, and functional considerations. Thirdly, the mapping of the evolution of stratificatory interlinkages provides insights into the historical evolution of the salience of stratificatory understandings and dimensions in international society. Do, for instance, stratificatory concepts such as ‘developing countries’ and ‘rising powers’ point to a gradual shift from politico-military to politico-economic interpretations of the patterns of stratification in present-day international society?
Acknowledgements
Earlier versions of this article were presented at the workshop ‘Inequalities and Global Institutional Order: The Case for a New Research Agenda’ in January 2015 as well as the EWIS workshop ‘Institutionalized Inequalities: How International Organizations Shape Global Order’ in April 2016, both of which were organised by Caroline Fehl and Katja Freistein. I would like to thank the two organisers as well as Edward Keene, Lora Anne Viola, and the other participants of the two workshops for their helpful and constructive comments. I am likewise grateful to Mathias Albert, Bernd Bucher, Martin Koch, and the anonymous reviewers of the Review of International Studies for their valuable feedback. I gratefully appreciate the funding provided by the German National Academic Foundation (Studienstiftung des deutschen Volkes) for my PhD research, upon which this article is based.
