Change in international order? An institutional analysis

Abstract

Many are now discussing the possible demise of the so called ‘liberal international order’, but how can we know whether any international order is changing? This article argues for understanding order as maintained by institutions of international society and further theorises the role those institutions play in the stability or transformation of international order. To usefully put institutional analysis to work, this article, first, models the stylised evolution of a primary institution. Second, it illustrates this evolution with a discussion of the historical institution of trusteeship in order to historicise adaptation and transformation in international order. Finally, this leads to a generalised idea of how institutional analysis can be employed to study stability and transformation in international order. Beyond making a contribution to the wider debates about the possible demise of the current international order, this piece also fills a gap in English School theory, which is quite silent on the question of when international society furthers transformation, and when it furthers stability. Accepting the view of history that the future is contingent on today’s events, this study suggests possible points where push comes to shove for change and continuity in international order more generally.

When decolonisation gained speed in the 1950s and the European colonial empires started to disintegrate, this was a monumental change in international order. From a system built on racial hierarchy, the moving target of the standard of ‘civilisation’,¹ and a succinct core–periphery dynamic, all peoples became formal equals within only a few decades. Certainly, this did not mean that injustice, discrimination, or economic inequality suddenly vanished, but it did change international order in a very fundamental way: from divided into categories of ‘advanced’ and ‘backwards’ peoples into a system of formally sovereign equal states. Overall order shook in this process but restabilised in a new format. Barry Buzan and George Lawson argue that this transformation marked the shift from modernity’s first phase, ‘Western-colonial international society’, to its second phase, labelled ‘Western-global international society’.² Now, order might be shaking again, with the end of American hegemony, a possible return to multipolarity, and what seems like a major shift in what can be taken for granted in world politics.³

Accordingly, whether the present order is indeed transforming or whether it is rather characterised by overall stability and continuity is an important question. Discussions of whether we witness the end of the so-called liberal international order abound.⁴ Yet, the post–World War II American-led order can be understood as a continuation of previous orders.⁵ Some authors make a point of connecting the post–World War II order with older forms of international order, notably imperial order⁶ and Westphalian order.⁷ In realism, international order is generally treated as static or recurrent;⁸ and power transition theorists argue that systemic changes are predictable and typically accompanied by major war.⁹ This need not be the case, however, as major changes can also be peaceful.¹⁰ Interesting contributions highlight the potentially cyclical nature of readjusting institutional structures;¹¹ the ubiquity of change;¹² and its differing expressions.¹³ Some of these differences build on diverging understandings of order, while others use diverging concepts of change.

Overall, questions of overarching ordering seem to have an increased actuality, as evidenced also by their treatment not only in edited volumes and special issues of prominent journals, but also in outlets such as encyclopaedia entries and podcasts.¹⁴ The implied question is: how can we know whether order is stable, transforming, or even disappearing? In this piece, I will provide a tentative answer to that question by theorising, first, order, and second, stability and transformation therein. The inquiry starts from three assumptions: first, that international order, provisionally understood as a social system of rule-bound processes based on enforcement capabilities as well as acquiescence, can exist without being liberal; second, that order has transformed before and can transform again; and third, that we would do better to understand international order as a dynamic and ongoing condition, continually changing and adapting, rather than as a stable arrangement suddenly coming to an end. In this context, therefore, the much-discussed ‘liberal order’ should be understood as a particular case of order; that is, as an empirical instance of a general condition of ordered relations between political units.¹⁵ To discuss whether that order is stable, transforming, or disappearing, I will draw on English School theory, and specifically on its understanding of primary and secondary institutions as ordering devices in international society. The English School, or international society tradition, has a strong record of theorising how international order is maintained.¹⁶ Yet, as pointed out by Buzan and Lawson, the English School has not said ‘enough about the conditions that lie behind the rise, development and sometimes obsolescence of both primary and secondary institutions’, and it also has not said ‘nearly enough about racism, colonialism and imperialism’.¹⁷ This article seeks to contribute to filling both of those gaps.

The rest of this article is outlined as follows: first, it discusses and refines the English School understanding of international order, before expanding on the institutions of international society and what we already know about their roles for change and continuity in international order. Second, it models the evolution of such an institution, offering a stylised image of what its emergence, evolution, and demise might look like. Third, this stylised evolution is illustrated by the empirical evolution of the institution of trusteeship. Fourth, it discusses the implications of institutional evolution for continuity and change in international order, and finally, it presents the conclusions of this inquiry.

On international order

Since Hedley Bull’s landmark contribution and before, researchers in the English School tradition have argued that current international order is characterised by the presence of international society; and that the substance of that society is its common (primary) institutions, such as war, management by the great powers, international law, diplomacy, and the balance of power.¹⁸ Historically, however, systems of like units (societies of states) are rare,¹⁹ and earlier orders have been organised differently; sometimes as empires and most often as different varieties of hierarchical systems.²⁰ But there, too, any order has been upheld by the relevant entities reproducing common institutions, such as trade, diplomacy, management from the centre, and war.²¹ From this perspective, then, the common institutions of international society at present provide the substance of the current order. What sort of order predominates in any given era thus depends both on the underlying structure of the international system (empire, dominion, hegemony, multiple independences),²² and on what institutions are practised by the participating collectives. Notably, the purpose of an order must also vary with its nature and power relations.²³ Bull’s elementary ‘life, truth and property’ goals for international society – preservation of the society itself, maintaining external sovereignty of the members, a condition of peace rather than war as normal²⁴ – are thus only valid for this international order, and in Bull’s version for other systems of multiple independences. Other orders of other kinds may have different purposes, although Bull held that some version of life, truth, and property would likely apply in any social system.

However, it is not completely clear what the concept of international order entails.²⁵ In English School theory, where the current order is supposedly characterised by international society, those two concepts often blend, and their mutual relations are not precisely stipulated.²⁶ Here, I argue that order is present when there is an (any) international society, and that when there is no society, there is no international order.²⁷ Order can thus be anarchical or hierarchical, and – as is currently the case – both of those things at the same time.²⁸ In historical ‘international’ systems, where relations were often more hierarchical than anarchical, order was similarly present when interaction between polities was rule-bound and predictable.²⁹ For this article, this means that order is not conditioned on American primacy,³⁰ on liberal values,³¹ or on the current framework of international organisations.³² Rather, order is a cross-temporal concept which may be applied to any dynamic system with social and rule-bound processes and practices, producing some degree of predictability in interaction. Although generally driven by the powerful, order requires at least temporary acquiescence by other members, as having an order is most of the time seen as preferable to not having one. This understanding draws on that of Evelyn Goh, who defines international order as a social system ‘in which shared norms, rules, and expectations constitute, regulate, and make predictable international life’.³³ In a later piece, she adds: ‘Creating these shared understandings involves contestation and negotiation, with consensus neither easily achieved nor automatically sustained.’³⁴ This means that order does not require consensus, peace, or mutual agreement. It can harbour quite strong tensions, as long as the baseline rules are respected sufficiently often to preserve some predictability in the interaction between units.

Since order is understood here as a dynamic state of affairs, where processes and practices evolve so that order remains stable, some preliminary discussion of when change means adaptation and when it means transformation is required. Adam Watson’s overarching idea of raison de système, the belief that it pays to make the system work, offers a way to conceptualise a stable order.³⁵ It also implies that when the relevant agents of an international order no longer believe that it pays to make the system work, the order erodes.³⁶ As addressed by Andrew Phillips, the underlying cause of this erosion can be a combination of changes in material factors – such as new technologies for warfare – and ideational changes undermining belief in that order or its institutions.³⁷ This can happen in a revolutionary or evolutionary manner.³⁸ Although the idea of a drastically punctuated equilibrium does not work well with the dynamic and processual understanding of order proposed here, erosion of order can nonetheless be sudden or gradual. Another way of seeing this, as will be clear in the next section, is that agents ‘withdraw their consent’,³⁹ which may happen quickly (if a critical mass of them do it simultaneously) or slowly (if agents reconsider their stance over a long period of time). Since institutions are pillars of order in the English School understanding, it follows that transformation (or breakdown) of order happens when those pillars collapse. When enough agents withdraw their consent, by acting in opposition to the institution or by abandoning it, it can no longer contribute to upholding order. Being able to identify such fundamental turning points in international relations is arguably important for studying and interpreting both historical transformations and current events.

International society and its institutions

There is no generally accepted definition of an institution of international society, but it is clear that they are of a broader and more sociological character than the international regimes or organisations which International Relations (IR) scholars sometimes think about when faced with the term.⁴⁰ I call these institutions the ‘primary’ institutions of international society, to distinguish them from ‘secondary’ institutions, on which more below.⁴¹ Primary institutions are defined in this article as reproductive international practices, connected to discursive legitimation consisting of norms, beliefs, and expectations defining the roles and relations of international society’s actors.⁴² As such, they maintain international society and function as pillars of the current international order. A practice might thus exist as a practice before being tied to the relevant discursive legitimation.⁴³ Importantly, for this article, it should also be possible to disconnect the practice from the discursive legitimation. In this way, it is possible for a practice to revert from being a pillar of order in international society to being ‘simply’ a practice.

Tonny Brems Knudsen has usefully distinguished between a ‘narrow’ set of primary institutions, and a ‘broad’ set.⁴⁴ The narrow set consists of institutions which are ‘the very preconditions of international society’, while the broad set include institutions which are ‘constitutive of meaningful or legitimate interaction more broadly’.⁴⁵ This article employs the broad approach rather than the narrow approach, thus taking seriously the possibility that the discontinuation of one institution may change the character of what is considered to be meaningful and legitimate interaction, rather than to order as a whole disappearing. Institutional evolution might thus change the internal balance of institutions in international society (for instance: more trade, less war), thus changing the character of international order rather than undermining it.

Primary institutions can also be formalised into secondary institutions, defined by Kilian Spandler as ‘sets of discursively formulated expectations … refer[ring] to temporally and spatially discrete sections of international reality and apply[ing] to a clearly defined set of actors’.⁴⁶ This allows for a systematic analysis of how the institutions of international society relate to international organisations and other negotiated and designed, rather than evolved, tenets of international order.⁴⁷ The clearest indication of the presence of a secondary institutionalisation of a primary institution is an international practice that has been equipped with some type of organisational superstructure.⁴⁸ It should also be duly noted that this is not some structural magic happening by itself, but of agents acting for what they consider to be the best within the limits of the framework in which they see themselves.⁴⁹ The outcomes are the result of their (sometimes) skilful political manoeuvring, and of the ensuing compromises.

The role of institutions in maintaining order

The common institutions of international society are the substance of that society, much like horse riding is the substance of a pony club.⁵⁰ If states were to add new institutions, this would change the character of international society, which is an oft-rehearsed theme in the discussion of whether international society should be conceptualised as being of a minimalist pluralist kind, or rather a solidarist association of increasingly shared values.⁵¹ In this sense, the selection of common institutions, which states practise, give international society its character. Meanwhile, the presence of any institutions which states practise is how order comes about. New routines, beliefs, and expectations do not have to be invented ad hoc at every international event, but there is a catalogue of relevant practices with adherent ‘constitutive principles’,⁵² or discursive legitimations, to refer to in each situation, which makes international affairs at least somewhat predictable and orderly.

In turn, this means that if the institutional framework changes, the character of international order changes. If institutions are understood as practices, which are discursively tied to norms, beliefs, and expectations and continuously reproduced so that they appear to be stable, this means that institutions change if they are either practised differently or legitimised differently.⁵³ This sort of change, or gradual adaptation, is ongoing. Furthermore, revolutionary change can also happen, in which institutions may collapse if they are no longer practised (states withdraw from, or act in opposition to, the established practice) or no longer legitimised (states no longer mention it, or argue against it). If institutions are replaced by other institutions, the character of international order changes accordingly. If institutions are not replaced by other institutions, order is no longer maintained and hence disappears. In the next section, a more detailed discussion of how institutional evolution may play out is presented. In the historical part of this article, an example of this is discussed with the collapse of the institution of trusteeship. The legitimation of inequality was replaced with the legitimation of equality, and the character of international order changed from imperial to liberal.

The evolution of a primary institution

There is a case for assuming primary institutions to be in some important respects similar to each other, thus making them a meaningful analytical category. If a primary institution, by definition, consists of a reproductive practice tied to a discursive legitimation made up of norms, beliefs, and expectations defining the roles and relations of international society’s actors,⁵⁴ there are some stages that logically will have to happen to them (see Figure 1). First, a practice must be united with norms, beliefs, and expectations about that practice, suggesting, if rarely clearly defining, the what, the who, and the how of the practice. Second, this nexus of practice and discursive legitimation must be acknowledged at the international level through states’ conduct and the words of their representatives.⁵⁵ This is what makes a primary institution. Those two stages are therefore by definition necessary to the emergence of any primary institution.

Figure 1. The stylised evolution of a primary institution.

Third, a primary institution may very well be institutionalised in a secondary institution; that is, formalised in an international agreement or organisation. Although this stage is not logically necessary, it is presently likely, given how international society has been organised in the last century.⁵⁶ If this happens, we should see the primary institution being expressed in a treaty or charter, and an administrative machinery attached to it. Given the definition of primary institutions as reproductive practices tied to discursive legitimations, one can assume that international organisations would be an arena both for the practices to be reproduced and for the discursive legitimation to be verbally reinforced. International organisations have become a first-choice forum for states to voice their concerns and argue for their causes.

Fourth, if a primary institution wanes, we should see states withdraw their consent from it, in words as in actions.⁵⁷ If enough states cease to reproduce the practice and to verbally reinforce the discursive legitimation, the institution is no longer reproduced either in practice or discursively. If there is a secondary institution attached to the primary institution, it is probable that states’ withdrawal from the primary institution will play out at the secondary institution. The withdrawal will then be argued not so much in terms of the primary institution, but in terms of its secondary institutional manifestation.

Finally, when states have withdrawn their consent from the primary institution, the secondary institution becomes outdated. This should be visible in treaty or charter amendments, or at least in changes in the way the formal rules are re-enacted in the organisation. Importantly, though, this does not necessarily mean that the practice itself disappears. A sixth step of evolution could therefore be added, in which the practice is ongoing but justified in an ad hoc manner and no longer performs an ordering function for international society.

Importantly, there are self-reinforcing effects to all these steps, which make it probable that primary institutions will follow similar trajectories, although the time frames and specific circumstances surrounding their various stages of institutionalisation and deinstitutionalisation vary. At the same time, there is a relational mechanism involved, as primary institutions are constitutive of the players.⁵⁸ Those self-reinforcing effects should therefore not be understood as performed by set entities, but the entities themselves are being shaped in the process. In this way, the ongoing recreation of primary institutions creates not only a class of processes but also defines who can practise those processes. Without assuming law-like regularity and the possibility of neo-positivist generalisations, it is both possible and probable that those effects make primary institutions over time go through these steps,⁵⁹ although the same general dynamics may still produce different outcomes in different contextual settings.⁶⁰ Thus, accepting the view of history that the future is contingent on today’s events, we may use institutional analysis to pinpoint possible crossroads where the balance between change and continuity in international order is particularly vulnerable.⁶¹

With secondary institutionalisation, step three above, the boundaries of which processes are included in the practice and its legitimation get harder and less porous, thus making the nature of the institution more fixed. The same goes for the shaping of entities, which become more set as secondary institutional reproduction is connected to membership in the international organisation and to being a subject of international law. Yet the practice underlying all of this keeps evolving, so that it may over time come into direct conflict with its secondary institutionalisation. This leads to a break between the practice and the secondary institution, where the secondary institution will end up no longer reflecting the practice. Once this happens, one might reasonably expect the secondary institution to be dismissed, but this is hard, precisely because of the less flexible nature of secondary institutions, including the formal difficulty of reopening negotiations of treaties and international covenants.

This process of institutional evolution might look different for different institutions, and in different times, which is what motivates this schematic depiction. Yet it is worth the while to also attempt to exemplify this evolution empirically. In the next section, therefore, the evolution of a primary institution will be illustrated by the institution of trusteeship.

The evolution of an institution: Trusteeship

Trusteeship, that is the idea that it was the responsibility of ‘civilised’ colonial powers to care for and educate ‘uncivilised’ natives of colonised territories, was for some time an extremely important phenomenon, colouring and contributing to the constitution of international society via its division between ‘advanced’ and ‘backwards’ peoples. If trusteeship nowadays sounds old and non-applicable, that is what makes it a fitting illustrative example: trusteeship has played out its role as a primary institution. For some time, it filled the function of a pillar of order in international society, both as a practice in which some political communities (with varying degrees of sincerity and success) acted as trustees for others, and as a discursive legitimation of the practice, manifest in norms, beliefs, and expectations. It was a creature of its time, and of the specific character of (‘Western-colonial’) international society during that era, but it also, importantly, contributed to shaping the self-understandings of the actors of that era. Its influence on world order during its time was important, notably in its effects in defining the entities involved and what their agents could do, and in structuring the relations between them. The study of trusteeship thus provides a useful prism through which to observe how self-reinforcing practices contributed to upholding, and changing, world order during that time.

Uniting practice with norm, beliefs, and expectations⁶²

Trusteeship as an idea began in late 18th-century Britain,⁶³ where the dubious rule of the East India Company in British India was debated.⁶⁴ The understanding that ruler and subject were ‘joined by a sacred trust’, and that government must benefit the Indians, became established. ‘Recognition of the [East India] Company’s claim to rule now depended on the performance of duties pertaining to the general well-being of people on behalf of whom the government of India was obliged to act as a faithful trustee.’⁶⁵ Alas, India, in British discourse, went from being seen as a polished and advanced culture, different from European cultures but nonetheless worthy of respect, to being perceived as clearly inferior to them by the late 18th century.⁶⁶ Anthony Anghie shows how positivist international lawyers formulated the central distinction between civilised and non-civilised in order to justify how previously recognised entities were no longer deemed to be equal to European states.⁶⁷ In this way, the discursive construction of trusteeship was a solution to the ideas of equality and freedom: the administration of colonial territories was rendered permissive by the distinction between ‘adult’, ‘mature’ peoples who were considered competent to govern themselves, and ‘child-like’, ‘immature’ peoples who were said to need the tutelage of others because of their own temporary incapacity.⁶⁸

At this stage, trusteeship was merely a norm, which emerged in response to misrule in India. According to some, trusteeship ought to guide the actual practice of all British rule over other peoples, but implementing this view was a process. At this stage, actual practice was not necessarily in accordance with what the norm prescribed. The practice of colonial rule was still in the process of being discursively tied to norms, beliefs, and expectations about trusteeship; trusteeship at this stage was rather a patchily held ambition to rule India according to the logic of trusteeship than an ongoing practice.

Acknowledging the institution at the international level

Over time, and especially in connection with ‘the scramble for Africa’, the language distinguishing between ‘adult’, ‘mature’, ‘civilised’ peoples, competent to act as trustees for other ‘child-like’, ‘immature’, ‘uncivilised’ peoples came to take hold among all the European imperial powers.⁶⁹ In his opening remarks to the Berlin Conference of 1884–5, Bismarck stated that ‘all the Governments invited share the wish to bring the native of Africa within the pale of civilization’.⁷⁰ Peter Lyon remarks that this, however, applied especially to the colonial administrations of Britain, the Netherlands, and, partially, France; while the Belgian, Spanish, and Portuguese representatives paid mere lip service to the ideal of trusteeship.⁷¹ With respect to the Russian Empire, Filippo Costa Buranelli has pointed out how it saw its own ‘civilizing mission’ in Central Asia in rather instrumental terms, as something that would (but ultimately failed to) earn it its own spot at the centre of European international society.⁷² Lacy Pejcinovic makes a similar point about the westwards expansion of the United States, which was framed as aiming for the ‘pacification’ of the ‘savage’ Native Americans.⁷³

In William Bain’s account, the Berlin and Brussels conferences of 1884–5 and 1890, respectively, ‘internationalised’ the idea of trusteeship. It then went from being a British idea of how to rule India to a collective idea of how to rule Africa, shared, in words if not necessarily sincerely, by the then members of international society.⁷⁴ This can be understood as the end point of a process during which one set of norms, beliefs, and expectations came to subsume a variety of rival ideas of what the European powers did, and should do, in other parts of the world.⁷⁵ British dominance at the time was presumably also important for making the idea central to international society. Bain writes that ‘members of European international society internationalized the idea of trusteeship by establishing in international law obligations that explicitly repudiated relations based on dominion and exploitation; and, in doing so, they accorded international legitimacy to the principle that the strong should rule on behalf of the weak’.⁷⁶ In the Berlin Act, this is expressed in Article VI, stating that participants should ‘watch over the preservation of the native tribes, and to care for the improvement of the conditions of their moral and material well-being’.

Yet, despite Europeans’ emerging consensus on the norms, beliefs, and expectations of trusteeship, there was no guaranteed adherence to those values in practice. Notwithstanding the condescending and paternalistic but well-meaning language used, mistreatment of the Indigenous populations of Africa was widespread, and the worst case was the Congo Free State. It was created on behalf of King Leopold II of Belgium and was well received by the parties at the Berlin Conference as a mission to bring trade and civilisation to Africa.⁷⁷ By the standards of the institution of trusteeship, however, the Congo Free State turned out to be a disaster.⁷⁸ In spite of its humanitarian tropes, and in the face of the emphasis of the Berlin Act on stopping slavery and the slave trade, the territory was quickly turned into the king’s personal commercial estate, built on a system of forced labour. Abuse, torture, starvation, and murder underpinned this system, and there was no rule of law to protect the interests of the Indigenous population.⁷⁹ As the humanitarian catastrophes in the Congo became known in Europe, the British government protested, referring to the undertakings stipulated in the Berlin Act to care for the Indigenous populations of Africa.⁸⁰ ‘The British government’s repeated references to the obligations imposed by Article VI [of the Berlin Act] reaffirmed, and, indeed, vindicated, the legitimacy of the idea of trusteeship.’⁸¹ After a drawn-out diplomatic struggle, where the British position received some support, Belgium eventually took the Congo Free State out of the hands of King Leopold by annexing it, and, under continued British pressure, finally bowed to the agreements of the Berlin Act and committed to living up to its obligations.⁸² Here, again, it seems important that it was Britain, from its dominant position, that upheld the rules and enforced Belgium’s eventual compliance.

In the terms employed in this article, the Berlin Conference marks the primary institutionalisation of trusteeship.⁸³ Cornelia Navari has suggested that for (the discursive legitimation of) a primary institution to count as such, rather than ‘just’ a norm, it must be genuinely international and not come from some domestic constituency.⁸⁴ At the time for the Berlin and Brussels conferences, trusteeship had been raised to being a concern for all of international society (at that time still rather modest in membership, as indicated by the sparse attendance at the Berlin and Brussels conferences, of 14 and 17 signatories, respectively). The fact that Britain’s critique of the atrocities taking place in the Congo Free State was eventually taken up by the United States, Italy, and the Ottoman Empire is another sign of the institutionalisation of trusteeship: it shows that members could be held accountable and sanctioned for failing to abide in practice by the norms and expectations dictated by trusteeship. Both the British critique and the Belgian responses play out in the context of what was agreed to in the Berlin Act. The British victory in this duel manifests in its recognition in 1913 of Belgium’s annexation of the Congo Free State, thus bestowing legitimacy on Belgium’s eventual commitment to trusteeship. The exchange shows that trusteeship was established and accepted as an institution of international society at that time.

Secondary institutionalisation: The League of Nations

At the end of World War I, trusteeship was given an international administrative machinery, thus undergoing its secondary institutionalisation, in the Mandates System of the League of Nations. This secondary institutionalisation was the outcome of a debate between those members of international society who argued for annexation of the territories lost by the losing powers in the war, and those arguing for some form of international trusteeship over these territories.⁸⁵ The victorious powers needed, primarily, to reach an understanding about the imploding Russian, Ottoman, and Austrian empires in Europe.⁸⁶ Meanwhile, both Germany and the Ottoman Empire had overseas possessions, which were deemed unprepared for independence.⁸⁷ According to the logic of trusteeship, then, some other state needed to take on the task of administrating these territories. The United States under the internationalist Wilson administration preferred to create a mechanism for international oversight, but the other great powers of the time argued that they were competent and qualified to act as trustees for those territories.⁸⁸ Notably, the British saw the resulting mandates system as a continuation of the British Empire, while the Americans saw it as a new invention.⁸⁹ The compromise reached was largely effected by the South African representative, Jan Smuts, who suggested a system of trustees acting under graduated international oversight. In Bain’s words, ‘the mandates system represented the middle way between annexation and anarchy; that is, between aggrandizement and the chaos Smuts believed would surely follow a premature granting of political independence’.⁹⁰ The ensuing compromise dictated that the territories in question should be administered by mandatory powers but with international accountability to the League of Nations.⁹¹

In the vocabulary employed in the present article, the mandates system of the League of Nations was the secondary stage of the institutionalisation of trusteeship. This is the process whereby some practices become formalised in a way that ties them very closely to a more rigid set of norms, beliefs, and expectations, notably by being written down in binding treaties and surrounded by organisational frameworks of buildings, staff, and rules of procedure.⁹² This process also contributes to constituting the actors involved; in the League of Nations, it was very clear who was a trustee and who was put under tutelage. The formal international rule is contained in Article 22 of the League Covenant, which very clearly expresses the idea of trusteeship: ‘those colonies and territories … inhabited by peoples not yet able to stand by themselves under the strenuous conditions of the modern world. … [T]he tutelage of such peoples should be entrusted to advanced nations who by reason of their resources, their experience or their geographical position can best undertake this responsibility.’ The very agreement on both the practice of trusteeship and its accompanying norms, beliefs, and expectations as a pillar of order in international society comes through in the choice of including it in formal terms in the League Covenant. In Bain’s words, ‘the founders of the League of Nations took a momentous step beyond the idea of trusteeship that the European powers had enshrined in the Berlin and Brussels Acts. Not only did they affirm that the well-being of subject peoples constituted a legitimate subject of international scrutiny, but they also erected what was often referred to as “international machinery” to ensure the faithful performance of the obligations of trusteeship.’⁹³

Secondary institutionalisation: The United Nations

At the end of World War II, the administration of non-self-governing territories had to be renegotiated once more.⁹⁴ The reorganisation of territories previously claimed by Germany and Japan exposed a rift between especially the British and the American views. While the United States under the F. D. Roosevelt administration aimed for independence for those territories as for all colonies, the British were wary of the implication for their own empire and vouched instead for self-government within the empire.⁹⁵ In the UN Charter, two chapters relate to trusteeship: Chapter XI, the Declaration Regarding Non-Self-Governing Territories; and Chapter XII, the United Nations Trusteeship System. Bain argues that this reflects the discord between the United States and the European colonial empires. While the colonial powers accepted the universality of trusteeship, they did not mean to subject their own dependent territories to a system of international supervision. The French deemed trusteeship as being applicable only to existing mandates and to territories detached from enemy states as a consequence of the war. … the British delegation drew a very clear distinction between ‘the principle of trusteeship which should guide Colonial Powers in the administration of their dependent territories (and should therefore be of universal application) and the creation of a special system of international machinery, to apply to certain specified territories’.⁹⁶ In Bain’s interpretation, Chapter XI is thus about trusteeship as imperial trusteeship, and as such closely resembles the Berlin Act, while Chapter XII is more faithful to the American preference of international administration and oversight of non-self-governing territories.⁹⁷ Neta Crawford points out how the British, in accepting Chapter XII, ‘were pushed by a change in international political capabilities … to do, if not an about face, a right angle turn’.⁹⁸ This development can thus be interpreted as a part of the process of the United States taking over the agenda-setting position which was previously held by Britain.

At San Francisco, the United States, possibly already suspicious of the Soviet Union, suddenly stopped arguing for independence for all dependent peoples and thereby joined forces with the colonial empires in instead emphasising self-government.⁹⁹ This meant that the United States now was less insistent on the end of empires, and that the idea of trusteeship was no longer kept separate from the idea of colonialism, as it had been hitherto in its American version. For colonies, the disappearance of the distinction between international trusteeship (argued by the United States) and imperial trusteeship (argued by Britain and other European empires) clearly signalled that trusteeship equalled colonialism, and that both would have to go.¹⁰⁰ Crawford points out how ‘the trusteeship system became the model for decolonization of non-trust colonies’.¹⁰¹

In terms of primary and secondary institutions, the San Francisco compromise on trusteeship was mainly a continuation of the process that had already played out at Versailles, but, crucially, with adaptations to account for the ongoing renegotiations of racial hierarchy. In the framework employed here, one might imagine that the primary institution, in both its practice and its discursive legitimation, had evolved since 1919, and that some adaptations of the secondary institution were made in the UN Charter to account for it. The new British policy towards its colonies, in which the goal of colonies was their progressive development, is one such sign.¹⁰² Another is the admission of trust territories as strategic or non-strategic, thereby playing a role in the trustee’s politics of security. Yet, as the continuing story tells us, these adaptations were far from enough to preserve trusteeship as a pillar of order. Both the practice and its discursive legitimation were soon to implode under their own emerging inconsistencies as the distinction between ‘mature’ and ‘immature’ peoples collapsed.

States withdraw their consent from the primary institution

In his discussion of the end of empire, Bain brings out two arguments in particular: the violation of trust and the right to self-determination.¹⁰³ The trust argument was that the trustees did not safeguard the interest of peoples put under their tutelage – in short, they were doing the wrong things, exploiting instead of assisting. Expressed in terms of trusteeship as a primary institution, this means that the practice went wrong and did not accurately reflect the norms, beliefs, and expectations of its discursive legitimation. There was thus a rift between the practice and its discursive legitimation, which was pointed out by leaders of dependent peoples: For the Africans or Asians who wished to escape a condition of servitude and tutelage, the charge that European colonial rulers failed to fulfil their obligations as trustees of civilization provided a powerful argument in support of their claim of independence. But this argument did not discredit the idea of trusteeship itself; it merely undermined the justification of the means by which colonial powers attempted to carry out their obligations.¹⁰⁴

The other argument was about how self-determination reflected the idea that all peoples should have the right to govern themselves, and that the division into mature and immature peoples based on race was in itself flawed.¹⁰⁵ Once ‘humanity and at least theoretical equality’ was granted to all, colonialism became increasingly incoherent and difficult to justify.¹⁰⁶ Equality of all human beings cannot coexist with basic inequality, and the emergent norm of equality of peoples clashed with the division of peoples into ‘mature’ and ‘immature’ that underpinned trusteeship. This argument of equality invalidated the institution of trusteeship itself, as it questions both the practice (that some should be ruled by others) and the discursive legitimation of the practice (the norm, belief, and expectation that the practice is for the best of the ruled, because of the inferior stage at which they allegedly found themselves).

It is conventional to date the end of trusteeship as a primary institution around 1960. Among the key developments in support of that conclusion are the Declaration of the Granting of Independence to Colonial Countries and Peoples in General Assembly Resolution 1514 (1960); and the General Assembly Resolution 2145 (1966), in which South Africa’s mandatory rights to South West Africa (Namibia) were revoked.¹⁰⁷ This was motivated by South Africa’s failure ‘to fulfil its obligations in respect of the administration of the Mandated Territory and to ensure the moral and material well-being and security of the indigenous inhabitants of South West Africa’.¹⁰⁸ Although the terminology here is different, those instances should indeed signify the end of trusteeship as a primary institution. Resolution 1514 formally makes nonsense of the idea of inequality, which is a necessary underpinning for the idea that some should act as trustees for others. The context of resolution 2145 can be understood as a test of the primary institution, the outcome of which was that it no longer applies: rather, it was rejected by the General Assembly, understood in this context as international society’s main plenary forum. The implementation of this change, however, took another two decades, as the South African occupation of South West Africa (since 1968 recognised as Namibia) did not cease until 1989.¹⁰⁹ Yet Resolution 2145 signals that the discursive legitimation of trusteeship, that is, the idea, belief, and expectation that more mature peoples should rule on behalf of less mature peoples, was unequivocally rejected by the General Assembly.

The fall from grace of trusteeship as a primary institution arguably caused a stir in world order. This is visible primarily in the secondary institutions contained in the United Nations (UN), the membership of which expanded with the creation of newly independent states. The agenda of the organisation was quickly affected by the interests of the new General Assembly majority, and the winning coalition from World War II saw their power over the organisation quickly dilute.¹¹⁰ This coincided with the beginning of Cold War order (‘Western-global’ international society, in Buzan’s and Lawson’s terms), which divided international society’s membership into the categories of first, second, and third worlds.¹¹¹

The end of the secondary institution

Trusteeship finally came to a halt as a secondary institution with the suspension of the UN Trusteeship Council in 1994, around three decades after its end as a primary institution. It is unclear if or under what circumstances it might resume its activities, and in the reform agenda of 2005, proposed by Kofi Annan, a complete abolishment of the Trusteeship Council was included as a part of a larger reform package.¹¹² What is a secondary institution without a primary institution? Not very long-lived or active, as a look at the protocols of the Trusteeship Council in the last years before its closure will affirm. Ralph Wilde describes how, in connection with the acceptance by the General Assembly of Resolution 1514, ‘the eleven Trust Territories exercised their right to self‐determination by either becoming independent states or associating themselves with other states or territories to form new states’.¹¹³ After 1975, only the Trust Territory of the Pacific Islands, administered by the United States, remained. These territories achieved independence (Micronesia, Marshall Islands, and Palau) or self-government (Northern Mariana Islands) between 1990 and 1994.¹¹⁴ The Trust Territory of the Pacific Islands was, notably, designated as a strategic area, and its trusteeship was therefore terminated by the Security Council rather than by the General Assembly (UNSC res. 683 and 956). In the terms of this article, the Trusteeship Council was an empty shell of a secondary institution already by 1975.

The decoupled practice

A practice that is decoupled from its discursive legitimation might be the end of a primary institution, but it does not terminate the practice itself, as illustrated in the case of South West Africa/Namibia. Nor did, at the secondary level, the suspension of activities of the Trusteeship Council mean that the practice of trusteeship discontinued.¹¹⁵ Rather, the practice might be ongoing but called something else and lacking the discursive legitimation that once made it a primary institution. Wilde describes how different versions of international administration of territories, sometimes in a manner very similar to trusteeship, has been ongoing throughout the 1990s and early 2000s: Trusteeship through foreign administration did take place, but on an ad hoc basis without a dedicated, institutionalized system of international organization involvement in constitution and oversight. … This perhaps reflected a reluctance on the part of member states, especially former colonial states, to legitimize international trusteeship in this manner, even if they were prepared to see it reintroduced in particular places on an ad hoc basis when administration was conducted by the UN.¹¹⁶

In a similar vein, Bain discusses several cases of present-day international administration of territories. He settles on the experience of UNMIK, the United Nations Mission to Kosovo, as being the example of international administration most reminiscent of trusteeship as it used to be practised in an earlier era. Writing in the first years of the 21st century, he remarks: if we were to express Kosovo’s present status in the paternal language of empire, we would say that it is analogous to a political child that is being prepared, under the United Nations’ watchful eye, for the adulthood that comes with constitutional independence. Indeed, UNMIK represents not merely the resurrection of trusteeship, but the resurrection of the nineteenth century practice of making dependencies out of territories whose domestic arrangements and practices fall outside the bounds of something called ‘civilization’.¹¹⁷

Yet an international practice which lacks the discursive legitimation which makes of it a primary institution of international society is solely a practice. It is not a pillar of order and does not play the advanced role in the maintenance of overall international order that we ascribe to a primary institution. UNMIK might have constituted Kosovo as immature, but it does not constitute anyone in particular as its trustee and in this sense does not contribute to ordering relations in international society as trusteeship once did. Moreover, there are important differences between trusteeship in the 19th and 20th centuries and present cases of international administration, residing notably in their justifications. If, in the colonial age, the peoples of a territory could be judged immature, barbarian, or savage, simply on account of race or of custom that was unfamiliar and seemed impenetrable or counter-intuitive to Europeans, nowadays international administration is applied to post-conflict territories where institutions have been severely damaged or literally wiped out. Interestingly, the very fact that the administration of those territories has been ad hoc rather than following a predefined logic shifts the organisational responsibility in the UN from the General Assembly (which elected the Trusteeship Council and to which it reported) to the Security Council (which is responsible for peacekeeping operations). Nowadays, all territories under international administration are thus a priori strategic.

There is a lively scholarly debate about post-conflict international administration that mirrors these differences.¹¹⁸ For the purposes of the present article, the point is that although the practice of trusteeship is arguably still around, there is no consensus, or agreement, on what norms, beliefs, and expectations this practice entails. As such, it is not a pillar of international order at the present time. Moreover, it lacks secondary institutionalisation, as its placement, actual as well as potential, within the UN system is debated and its application ad hoc. Trusteeship, as a primary and secondary institution, is, and remains, dead.

Change and continuity in international order

In the analysis of trusteeship, the starting point was a time when international law was being constructed, by Europeans, to account for racial hierarchy, and in a way which would guarantee the European empires the upper hand. Trusteeship thus interacted with other institutions to configure international order; notably war, great power management, international law, sovereignty, and trade. All these institutions were closely interwoven during the colonial era, and it is arguably because the other institutions were continually reproduced that international order adapted rather than collapsed when trusteeship was deinstitutionalised. The practice and legitimation of international law and sovereignty were adjusted to include newly independent states on formally equal terms; great powers kept managing international society, although the balance of power necessarily shifted across the Atlantic; and war and trade arguably continued much as they had before, only among newly recognised parties, too, rather than simply among the old ones. For war, instances of fighting or coercion which had previously not qualified as war ‘in the strict sense’¹¹⁹ now involved recognised members of international society and were therefore more easily recognised as war.

The main question in this inquiry was how to know whether order is stable, transforming, or disappearing. The argument presented is that we can study the primary and secondary institutions that maintain order, to see if they are robust in their functions as pillars of order, or if they are breaking down. As discussed in general terms, and illustrated by the discussion of the evolution of trusteeship, there are ways in which we may distinguish whether institutions are being recreated and legitimised in an ongoing manner, thereby looking stable, or whether states are withdrawing their consent from them, in words as in actions. Although trusteeship was not necessarily a typical or representative primary institution, this way of studying institutions could, in principle, be applied to any institution. For the current order to be disappearing, several institutions should be breaking down simultaneously. For order to be transforming, however, it is enough that one institution is deinstitutionalised or that its practice or legitimation are transformed.

The most important ordering function of a primary institution is arguably to constitute the players and the relationships between them. This came through clearly from the example of trusteeship, as its role for defining who was ‘advanced’ and who was ‘backwards’, and the relationships between those two groups, was central. Yet it also regulated the relationships within the group of trustees, insofar as they all recognised each other as ‘civilised’ and insiders of international society.¹²⁰ This was clearly signalled in the attendance of the various international conferences where trusteeship was regulated (Berlin, Brussels, Versailles, San Francisco), as well as in the diplomatic controversy between Britain and Belgium over the Congo Free State. However, as soon as the peoples under tutelage started to recognise each other as such and establish relationships within the group, this led to them organising against the division of peoples into a racial hierarchy, and eventually to the demise of both the discursive legitimation and the practice. This historical experience thus provides a clear hint as to when primary institutions start to come apart: when the roles and relationships that they prescribe are rejected by enough of the players involved.

Another hint may reside in the timing of secondary institutionalisation of institutions of international society. For two reasons, it might well mark the peak of the primary institution by being the beginning of the end of its role as a pillar of international order. First, states may formally agree on secondary institutionalisation in order to prevent an (already-suspected) future decline; that is, as a conscious measure of preservation or a way to ‘lock in’ perceived achievements.¹²¹ Suppose that international society’s then ‘insiders’, when agreeing on the League Covenant, wanted to protect their acquis communautaire; that is, safeguard a system of common rules – including understandings of who was ‘mature’ and who was not – which possible future members would have to buy as a package deal. In such a case, the formalisation of trusteeship as a secondary institution was one of the parts included in the package, and it was there to be protected against current outsiders, who might take a very different view on trusteeship, and who might one day make it to international society’s inside.

Second, there is a theoretical possibility that secondary institutionalisation in itself might be the beginning of future irrelevance. This could happen if the formal agreement on rules by which to regulate an established practice creates a split between the rules and the (continuously evolving) practice. As long as the practice is a primary institution, and as such associated with norms, beliefs, and expectations, there is certainly a possibility for those norms, beliefs, and expectations to shift, evolve, and adapt to changes in the (reproductive) practice over time. However, as soon as those norms, beliefs, and expectations are formally codified (that is, in a secondary institution), they are stabilised, or frozen, and agents’ abilities to adapt them to changes in the practice are thereby weakened.¹²² Illustrative of this, there is a possible rift between the secondary institutionalisation of great power management in the Security Council, and the practice of great power management at large. In some ways, they harmonise, but in others, important decisions are taken elsewhere (G7, bilateral deals) and including other players than the P5 (the European Union, Germany, Japan, India, Brazil, South Africa). If the secondary institution does not accurately reflect the practised primary institution, this affects the stability of both the primary and the secondary institution.

Should several primary institutions collapse at once, thus no longer contributing to the maintenance of order, there would be serious repercussions for international order. If only one primary institution is shaking, it is probable that the continued reproduction of others could compensate for the turbulence. In the case of great power management and sovereign equality, there is an ongoing tension between them, as they are logically incongruent. This means that in a multipolar situation, where great power management is spread out more thinly among more regional great powers rather than among just the P5, it is probable that sovereign equality will be practised more consistently in its place (for instance by General Assembly resolutions overriding Security Council inertia). This, in turn, would probably lead to a more hands-off order, where it would be more difficult for the West – which is as privileged by the post–World War II great powers arrangements as it was during the earlier era of trusteeship, but which holds a minority position in the General Assembly – to enforce its preferred course of action consistently. The order thus created would consequently be less West-centric and probably less liberal, but not necessarily less ordered.

Conclusions

The aim of this article was to use the history of trusteeship to model the evolution of primary institutions, thus contributing to theorising change and stability in international order, by answering the question of how to know whether order is stable, transforming, or disappearing. The overall argument has been that, assuming that order changes and adapts in an ongoing fashion, it is the emergence or decline of primary institutions that gives rise to more significant transformations in order. Primary institutions over time emerge, evolve, peak in their ordering function, and decline, with very tangible consequences for international order, notably the constitution of the implied parties and the relations between them. In general terms, primary institutions are important for continuity, in the sense of maintaining international order. As long as they are routinely reproduced in practice as in discursive legitimation, they contribute to stability and continuity by maintaining a recognisable order. This happens in large part by the function they perform in defining the involved players and the relations between them.

Conversely, primary institutions are also important for change in international order. Most often, this change takes the shape of gradual, contingent evolution. The ongoing reproduction of practice, as well as the continuous construction of its discursive legitimation, makes the institutions seem stable over time, but this is largely due to their malleability and adaptability to gradual and incremental change. Change may also come as quicker transformation, as was the case with decolonisation, if the ordering functions of primary institutions suddenly cease. If or when primary institutions are formalised into secondary institutions, notably within international organisations, this very institutionalisation reduces their flexibility and may thereby affect their durability.

In conclusion, the current order can usefully be understood as based on a handful of primary institutions defining the relevant players and the relationship between them. As long as those institutions are recreated, in practice and in discursive legitimation, international order remains stable, albeit continuously evolving. If one of them is collapsing, and its ordering function is undermined by states withdrawing their consent from it, in practice and in discourse, as happened during decolonisation, this may challenge and transform overall order. However, it does not lead to a substitution of order with general disorder understood as unpredictability, as long as order is continually reproduced by means of other primary institutions. It is not until several primary institutions collapse simultaneously that overall international order risks disappearing. Therefore, when we talk about the demise of the current order, what we actually talk about might in fact be the transformation of order, from American/European leadership, rather than a demise of order itself.