Diaspora

In 1991, in the inaugural issue of Diaspora, William Safran presented the following definition of the concept:

This exemplifies what has come to be called the ‘essentialist’ approach: diasporas are treated as concrete entities matching some typological criteria. For the rest of the decade, contrasting typologies abounded. For example, Robin Cohen’s influential 1997 work drops the requirement that dispersal be forced or traumatic to include those who scatter voluntarily, restricts the definition to those to have settled for a relatively long period, highlights the positive impacts of migrants’ lives in regions of destination, and acknowledges that assimilation and integration does occur.Footnote ⁸ He goes on to propose a division of diasporas into five ideal types:

1. 1. Victim diasporas (e.g. those forced into exile, such as the Jewish or Armenian cases)

2. 2. Labour diasporas (e.g. migrations for work and economic opportunity, such as the Turkish or Indian diaspora)

3. 3. Trade diasporas (e.g. those migrating to create trade routes and networks such as the Chinese and Lebanese cases)

4. 4. Imperial diasporas (e.g. migration from the imperial metropole to sustain the empire, such as the British and French cases)

5. 5. Cultural diasporas (e.g. produced through chain migration such as the Caribbean diaspora).

As is clear from Cohen’s five categories, one consequence of this dispute regarding typologies was a gradual expansion of the term to encompass more and more cases. Indeed, by the 2008 edition Cohen shifted to considering diaspora as closer to a Wittgensteinian family resemblance concept:

In 2005, Rogers Brubaker’s now classic article criticised the field for creating what he called ‘the “diaspora” diaspora’: ‘a dispersion of the meanings of the term in semantic, conceptual, and disciplinary space’.Footnote ¹⁰ In contrast, he suggests an anti-essentialist approach which abandons the search for the ‘right’ typology of diaspora in favour of considering diaspora as a ‘stance’ which is adopted by particular communities: scholarship should, he continues, seek to understand the content of that stance, rather than hold up communities in the world to some set of prior criteria in order to ascertain whether or not they are to ‘count’ as diasporic.

In Diaspora Studies it has since become commonplace to suggest that we should not reify ‘diaspora’ as a pre-existing community.Footnote ¹¹ Instead, we must recognise that diasporas are brought into existence through social processes. To take this seriously is to acknowledge that any form of diasporic representation is built upon configurations of power, interests, and ideas, which, differently arranged, would have led to alternative forms of representation. In other words, there is no pre-existing ‘Rwandan diaspora’ or ‘Zimbabwean diaspora’. Where particular forms of transnational political mobilisation have taken place, it has been for someone and for some purpose.Footnote ¹²

This move from reification towards recognising the historical and political contingency of diaspora parallels that which has taken place in the way social science has considered other identity categories. Scholarship on ethnicity and nationalism, for example, has evolved from an essentialist view of these categories as pre-existing and static, to one that historicises and politicises their emergence and use. In broad terms, work on ethnicity and nationalism’s role in conflict has been considered from at least three contrasting perspectives. First, a now largely discredited primordialist view that has considered these identity categories as fixed and essentialised.Footnote ¹³ Second, a constructivist view that has understood that ethnicity and nationalism are socially constructed and historically contingent, mattering only insofar as they are imbued with particular forms of social significance.Footnote ¹⁴ Third, an instrumentalist view that has explored how categories of ethnicity and nationalism are often deployed by individuals and groups for the purpose of economic and political gain.Footnote ¹⁵ Across the existing literature on identity and conflict, a broad consensus position has emerged, discrediting the primordial approach, and recognising that a combination of constructivism and instrumentalism is necessary to explain how and why identity categories come to be politically salient.

Of course, much like ethnicity, just because diasporas are imagined, that does not mean they are not real, or without content. That content is, in general and minimal terms, the organisation of the community around the narratives of (a) being dispersed, (b) resisting assimilation, and (c) retaining an ongoing orientation to the homeland. These identities can become deeply held and culturally rich sources of identity for many communities, but it remains one possible form of transnational mobilisation and one that could have been otherwise – for example, during the Cold War, Russians, Vietnamese, Nicaraguans, and a range of Eastern Europeans were recognised and politically organised by the West as ‘exiles’ or ‘refugees’ as a means to discredit the homeland state.Footnote ¹⁶ What exists sociologically is not a predefined ‘diaspora’ but rather a group of nationals who are outside their state. The forms of political mobilisation they engage in – and the labels used to represent that mobilisation – are politically contingent.

For our purposes, we wish to understand the threshold at which political engagement with the homeland constitutes a diaspora in fairly minimal terms. There may appear to be many self-identified diasporas which engage in little, if any, political activity at all. For example, the work of Oliver Bakewell on Congolese and Senegalese networks across Africa reveals a great variety of social and economic activity, but little that looks, on first glance, like overt politics.Footnote ¹⁷ In this vein, many Diaspora Studies regard diasporas as primarily social entities, which may be, and in many cases are, apolitical. In such a context, they study the cultural life of diaspora more broadly than this work, which adopts a narrow focus on the political. This is in part a reflection of our interests, but it is also driven by our conviction that even relatively thin articulations of diasporic identity have more political content than they are sometimes credited with. As Rogers Brubaker argues,

In this sense, for us, diasporas are contingent social constructions, but are almost necessarily political in content: their asserted existence is an ongoing claim by individuals to have business with the homeland, to a form of kinship, and invariably a narrative of grievance and thwarted entitlement. As such, we focus on the most overtly political forms of diasporic political activity, because we see them as being at the heart of diasporic action more generally. In so doing, this book follows in the footsteps of excellent work on the politics of diasporas. This literature, which could be thought of as beginning with Gabriel Sheffer’s edited volume Modern Diasporas in International Politics,Footnote ¹⁹ has produced a wide-ranging and insightful corpus of work, such as the comparative overview of William Safran,Footnote ²⁰ Jossi Shain’s work on diasporas in the United States,Footnote ²¹ and Eva Østergaard-Nielsen’s study of Kurdish and Turkish communities in Germany,Footnote ²², amongst many others. In 2007 the American Jewish diaspora appeared in a starring role in the work of two of the most prominent ‘realist’ thinkers in International Relations, in John Mearsheimer and Stephen Walt’s The Israel Lobby and US Foreign Policy,Footnote ²³signalling definitively the entry of diasporas into the mainstream of political science. It is not our intention to overturn this body of work, but rather to contribute heuristic tools and build theory which can usefully supplement it.

Outside these studies, many fail to adequately recognise the political contingency of diaspora. This sense is reinforced by an influential policy discourse on diaspora. Over the last decade, a ‘Diasporas for Development’ literature has emerged, led by the World Bank and others, which posits diasporas as a pre-existing resource that can be drawn upon as a source of remittances.Footnote ²⁴ This has had the effect of reifying and depoliticising diaspora. It masks divergence and obscures the political context in which transnational mobilisation takes places. The ‘Diasporas for Development’ debate has also created a liberal international discourse ripe for selective appropriation.Footnote ²⁵ In particular, it is useful for those with an interest in representing groups of nationals outside the state as diasporic, and therefore depoliticised.

‘Diaspora’ is therefore, for our purposes, an analytically unsound starting point for empirical enquiry. If what we are interested in is the process of transnational political mobilisation, then beginning only with groups that fulfil an arbitrary check-list of criteria or with those that are already widely labelled as ‘diasporic’ is to strip away the possibility of enquiry into the conditions that enable, constrain, or constitute that particular form of mobilisation. As an object of enquiry, it assumes away the very processes we are interested in: the historical and political contingency of variegated forms of transnational political mobilisation.

What we wish to understand is the emergence of diasporic politics from the broader community of nationals outside the state – simply, people who identify as being of a particular nationality but are outside the territory of the associated state – what Myron Wiener calls ‘incipient diasporas’,Footnote ²⁶ and Jossi Shain and Aharon Barth describe as ‘geographically outside the state, but identity-wise perceived (by themselves, the homeland, or others) as “inside the people”’.Footnote ²⁷ This represents one analytical step further upstream from ‘diaspora’. It is a prior category, the raw material from which diasporic identity may or may not be cut. The empirical fact of dispersion alone necessitates only a ‘latent diaspora’.

Of course, the ‘state’ and the ‘nation’ are themselves socially constructed entities. In some cases, these social constructions are the subject of considerably contestation and controversy (e.g. Kurdish national identity). However, in the two cases we look at in this book we are dealing with two relatively strong states with relatively well-defined nations. Neither are weak states, both have clearly defined geographical borders, and in both cases we did not meet anyone in our fieldwork who could not answer the question of their nationality, or sought to contest the self-identification of others. For the purposes of our own research, these categories have sufficient fixity and are the subject of sufficient consensus that we can take nationals as our base category.

Animation

We are interested in explaining how those outside the state but ‘in’ the nation come to be politically mobilised as diaspora. The mobilisation of diasporas has already been fruitfully analysed within the social movements paradigm by Martin Sökefeld.Footnote ²⁸ In his work, Sökefeld draws on the classic typology of McAdam, McCarthy, and Zald:Footnote ²⁹

• Opportunity structures, understood as ‘consistent – but not necessarily formal or permanent – dimensions of the political environment that provide incentives for people to engage in collective action’.Footnote ³⁰

• Mobilising structures and practices understood as ‘collective vehicles, informal as well as formal, through which people mobilise and engage in collective action’.Footnote ³¹ This includes the whole panoply of political parties, advocacy groups, spontaneous protests, violent disruption, and insider lobbying.

• Framing understood as ‘specific ideas that fashion a shared understanding for a social movement by rendering events and conditions meaningful and enable a common framework of interpretation and representation. They are ideas that transform certain conditions into an issue, that help to define grievances and claims, and that legitimise and mobilise action’.Footnote ³²

We wish to characterise the practices of diasporic political mobilisation as ‘animation’, which may be defined as the way in which diasporas (or other identity groups) are brought into existence. The ‘diaspora’ is simply the animated part of that wider group of nationals outside the state: no longer simply a ‘diaspora-in-itself’, but now ‘for-itself’ too. This process is fraught with interests and power. It is also rife with instrumentalisation; the question of ‘cui bono?’ in relation to the emergence of particular manifestations of transnational networks and the strategies that they adopt. Our interest in how transnational communities are instrumentalised parallels the literature on refugee manipulation, for example, which is also focused on explaining the manipulation of nationals outside the state by a range of exogenous actors.Footnote ³³ For us, though, we are implying not just that the diaspora (ie, the people themselves) are instrumentalised but that the very idea of ‘the diaspora’ as a category is constructed and mobilised for political purposes, as has been argued of the label of ‘the refugee’ elsewhere.Footnote ³⁴

Animation is conducted by animators. Animators are simply the actors that promote the mobilisation of an identity group (in this case the diaspora), particularly through the provision of resources. Some established transnational communities have developed self-sustaining sui generis mechanisms for animation, where animators are to be found within the community itself. The Jewish diaspora is one such example. However, many have assumed that this archetype is the norm rather than the exception, and that the model of a self-animated community travels easily to the other transnational communities dubbed ‘diasporas’. Many less established, contemporary diaspora groups are the product of an era in which significant resources from outsiders are available to those who self-represent as diaspora and self-identify in the appropriate way. As with the promulgation of nation or ethnicity, there is a political economy behind the animation of diasporas.

This political economy is usually – at least in Africa, but not necessarily – externally aided and sustained by external actors with a range of political interests behind supporting diasporic representation. Once these external sources of animation wane, the political significance of the diaspora will wane – although there may still be an ongoing, albeit largely hollow, performance, disconnected from wider political impact.

External animators have their own agendas, which are frequently quite different from those of the diasporic individuals they are seeking to assist. This concern with co-option and manipulation connects this book with broader work on ‘unhelpful helpers’: on this view, diasporic support is another outpost of the ‘White Saviour Industrial Complex’, where, in general and somewhat polemical terms, white Western activists treat various subaltern groups, particularly Africans, as a blank space for the projection of their own agendas and fantasies.Footnote ³⁵ In anthropology, Aihwa Ong’s work on neo-liberalism’s mutations in South Asia includes reflections on Chinese Americans in Silicon Valley speaking on behalf of Chinese and Malaysian migrant workers.Footnote ³⁶ In political science, David Zarnett has raised questions about the connections between the Tibetan diaspora and the Western activists of ‘transnational civil society’.Footnote ³⁷ In the Rwandan context, the ways in which ostensibly autonomous grassroots civil society organisations may serve wider agendas is well documented.Footnote ³⁸ In the South African context, there is ongoing debate about the relationship between marginalised groups and ‘zim zims’, middle-class academics and activists who descend on townships talking about neo-liberal capitalism, and so on. This debate is perhaps sharpest in the case of Abahlali baseMjondolo, the shack dwellers’ movement, the largest organisation of the militant poor in post-apartheid South Africa, which got into a protracted disagreement with an NGO called the Centre for Civil Society hosted by the University of KwaZulu-Natal.Footnote ³⁹ As in all these works, we want to raise suspicions about the agendas of external animators, about who speaks for whom, and with what consequences.

Animation is a special case of the classic concept of ‘mobilisation’, much as the animator is a close relative of the ‘diasporic entrepreneur’. Our terminology draws attention to particular aspects of this process. On the one hand, our concept of animator is strongly influenced by the concept of ‘animateur’ in French social theory. Originally theorised in the context of adult education, an animateur is someone who brings to life a new way of thinking, seeing, or interacting by injecting focus and energy into a social group. It is thus narrower than mobilisation, as it refers specifically to that form of mobilisation associated with the creation of an identity group, in contrast to the vast range of activities which fall within the ambit of mobilisation. On the other hand, it draws attention to the way in which animators bring resources into the diaspora. The literature on diasporic mobilisation usually works on the implicit assumption that a diaspora contains resources which the diasporic entrepreneur prepares, organises, and then brings to the field of political action. That sense of mobilising what is already there is distinct from investing new resources into the diaspora. Finally, the literature usually places mobilisers within the movements in question. Our animators may also be external, and in the standard social movements model would be relegated to the role of ‘external opportunity structure’. Much as the transnationalism literature straddles the boundary between inside and outside the state, therefore, our model disrupts the clean distinction between the internal mobilisers and the external opportunity structure.

Potential external animators include (i) the governments and other actors in host states, (ii) third countries, (iii) the global governance establishment, and (iv) self-interested political parties ‘back home’. In the case of the Zimbabweans, the South African political establishment – notably civil society activists plus government bureaucrats and politicians – have played the major role. Third countries – such as the British Foreign and Commonwealth Office (FCO) – were also prepared to pour significant resources into animating the diaspora prior to 2008. Finally, through the trend in global governance relating to ‘Diaspora for Development’, ‘the diaspora’ has become instrumentally and ideationally recognised as important, leading to the establishment of previously unavailable budget lines and resources.

What resources, then, are brought into ‘the diaspora’ by such external ‘animators’ and what motivates them? There are three broad mechanisms of support that appear to bring ‘diasporic’ organisations and activities to life: (i) money (and other material resources), (ii) expertise, and (iii) connections. These resources create the conditions that render transnational groups willing and able to establish a proliferation of organisations and associations in the ‘diasporic’ mode. Inevitably, these inputs are brought in at particular moments and for particular political purposes. In the case of Zimbabweans, the story of motivation is multi-faceted. Globally, the International Relations story emerges from the strategies of agents such as the UK FCO and the US State Department adopted in order to try and shut down Zimbabwe and contest ZANU-PF through supporting the MDC in exile.

The host state’s and society’s relationship with the diaspora, and their ensuing reasons to engage in animation, are likely to be complicated. Motives for supporting the diaspora may include struggle-era solidarities, the desire to protect human rights, or simply to manage migration.Footnote ⁴⁰ Following the transnationalism literature in International Relations,Footnote ⁴¹ we would predict that if the host state were to have a rival relationship with the country of origin, the diaspora might be more likely to become instrumentalised. There would be a strong incentive on the part of the state to support those in exile, thereby supplementing the intergovernmental relationship with the homeland state with a transnational connection to the exiled community. In the case of South Africa, though, Thabo Mbeki’s close relationship to Robert Mugabe before 2008 meant that South Africa was unable to be openly oppositional and so barely investigated this strategy. Nevertheless, the non-unitary nature of the state meant that elements of civil society were able to lobby different sections of the state in different ways.

Not all diasporas will be animated in the same way. Figure 1.1 sets out a typology for considering the different modes of diasporic animation, based on two distinctions: between ‘internal’ and ‘external’ animation, and ‘institutional’ and ‘networked’ animation.

Figure 1.1: Typology showing the relationship between the characteristics of diasporic animators and the predicted effect on the form of diaspora life.

The internal/external distinction relates to whether animation emerges primarily from actors within or external to the extra-territorial community, the community of people outside the state but ‘in’ the nation. Is it self-generated political mobilisation? Do the ideas and the money that enable mobilisation to take place come from self-generated community structures? Or, alternatively, are they the result of input from donor governments, outside human rights activists, or other interested parties? For our purposes, therefore, organisations based in the homeland state, even if they develop extra-territorial outposts operated by co-nationals, are considered external animators.

The institutional/networked distinction relates to the character of the animators. In ordinary speech ‘institutions’ can refer to a wide range of things, including codes, conventions, traditions, rituals, formal organisations, or particular types of organisations. Institutions, for us, are simply ‘enduring patterns of social life’.Footnote ⁴² More precisely, institutions are complex, self-reproducing and stable patterns of human activity such as governments, corporations, universities, (some) families, and so on. So we distinguish institutions from codes, conventions, rules, and rituals, which are the constitutive elements of institutions, and also from cultures, societies, and other macro-sociological entities (such as a diaspora) of which institutions are constitutive elements. Institutions are often, but not always, organisations. Institutionalisation is a spectrum, and at one end we find bureaucratic formalised institutions with the panoply of Weberian resources, roles, and procedures that implies, and at the other we find diffuse networks of individuals linked by personal ties. A simple heuristic test is to ask: Are the organisations of the animators the sorts of things where power inheres in offices (or social roles more generally) or in persons? A diasporic life can be animated by a body which is clearly deeply institutionalised (a classic example would be the Jewish Anti-Defamation League), but it can also be animated by inchoate and temporary networks of individuals.

These questions can be asked of particular animators, but can also be asked of animators in general, in which case the question is of the preponderance of institutionalisation (or lack thereof), and what proportion of the animators are external or internal. As such, these are not dichotomous variables, but vary on a scale. We are not suggesting that the characteristics of the animators are the only factors that determine the life cycle of diaspora. Most obviously animation only takes place when it is in the interest of the animators that it do so, and that is determined by a wider geo-political context. Our contention is that, where animators exist, their characteristics do much to determine the form of diaspora which is animated.

The properties of the quadrants are the characteristics of the diaspora which has been animated. We suggest two dimensions of variation: stability and contingency. Stability denotes the degree to which diasporic political mobilisation exhibits lasting and unchanging patterns of association (throughout its life cycle). Contingency denotes the degree to which diasporic political mobilisation is likely to survive changes in the external environment (of the sort that would generate interest divergence between the diaspora and external animators).

Our claim is that, all other things being equal, the extent to which the animators are institutionalised is directly related to the extent to which diasporic political action is stable, and the extent to which the animators are internal is inversely related to the extent to which diasporic political action is contingent. This creates the four broad possible cases indicated in the four quadrants above.

The top-right quadrant, of ‘stable and locked-in’ diasporic animation, is most clearly represented by the Jewish archetype. The political life of the Jewish diaspora has proved resilient to dramatic changes over the course of the twentieth century and, further to that, exhibited a great continuity of institutional forms. The Anti-Defamation League, for example, is over a century old. The Board of Deputies of British Jews was founded in 1760. Although of course the Jewish diaspora has also exhibited great change, this is, in contrast to other diasporas, really very stable. This can be seen most clearly by contrasting it with the bottom-right quadrant, that of ‘unstable and locked-in’ diasporic animation. Here the animators are internal, and so unlikely to leave at a moment’s notice, but, for whatever reason, prove unable or chose not to develop thick institutionalised patterns of association. A clear example of this case would be the current Eritrean diaspora, formed largely of exiles from the EPLF one-party state. Eritrean political activity has bubbled away at a relatively consistent temperature, sustained largely without external support by a hard core of Eritreans, but it exhibits a bewildering degree of fragmentation and near continual turnover of institutional forms as organisations are founded, split, and reformed, and fall into abeyance nearly constantly.

The other side of the typology is cases of external animation. In the top-left quadrant we find ‘stable but contingent’ mobilisation. In this case, political activity is characterised by relatively thick institutionalised patterns of association, but because the roots of this animation are external, they may disappear at any time. This is probably the most unusual case, but one case might be the anti-Castro organisations of the early 1960s, particularly the Cuban Revolutionary Council. These organisations were resourced and organised, but their dependence on their privileged access to the American Presidency was revealed when, following the Missile Crisis of 1962, the US government abandoned all attempts to unseat Castro violently, and therefore dropped these organisations, which rapidly decayed and quietly disintegrated by the close of 1963. The bottom-left quadrant, ‘unstable and contingent’, denotes the most ephemeral and volatile form of mobilisation. Here the preponderance of animation is external, and the animators work through weakly institutionalised networks and fleeting personalised connections. One potential case might be the ‘white émigrés’, who fled from Imperial Russia to inter-war Western Europe following the 1917 Revolution. Many received intermittent and ad hoc support from sympathetic individuals, particularly in Britain and Germany, but largely failed to secure the sustained backing of institutions. As such, their organisations, such as the Monarchist Mladorossi, the Eurasian Evraziitsi, and the accommodationist Smenovekhovtsi, exhibited huge degrees of volatility, and ultimately did not survive the vicissitudes of inter-war Europe (except, notably, in the form of the Russian Orthodox Church Outside Russia, which survives to the present day and is internally driven and strongly institutionalised).

Life Cycles and Other Trajectories

Our key claim is that much more of diasporic political activity can be explained by the positioning of animators on this graph than is usually assumed. The conventional picture of diasporas as unproblematic bounded entities was destabilised by our research on Zimbabweans in South Africa. The ‘Zimbabwean diaspora’ – putatively one of the most visible and active in Africa – appeared fragmented, exhausted, and of limited relevance to the politics back home by the time of the 2013 Zimbabwean elections. Yet this was not to say that there had not been a prior ‘diasporic moment’. Around the 2008 elections, a series of international animators had enabled the transnational political mobilisation of Zimbabweans in South Africa and the UK. There had been an interest in investing money, ideas, and network capital into those nationals outside the state. However, at a certain point, this moment passed. The animators left, and the viability, organisation, and relevance of the Zimbabwean diaspora waned. The money stopped, relevant political actors went home, and only a residual and largely powerless group remained.

To address this observation conceptually and to rehistoricise the process of animation, we developed the concept of a diasporic ‘life cycle’. The life cycle idea is used elsewhere in the social sciences. One prominent example used in International Relations is a norm life cycle, which Finnemore and SikkinkFootnote ⁴³ use to explore how ‘appropriate behaviours for a given actor’ emerge and evolve. They examine three stages of the norm life cycle: norm emergence, norm cascade, and internalisation, highlighting the different actors, motives, and mechanisms that are relevant at each stage. Others have built on this work to discuss ‘norm death’ and ‘norm regress’.Footnote ⁴⁴

We use a similar idea to capture the idea of a diaspora life cycle. In a life cycle, diasporas are born, become established, and may die, just as norms in the work of Finnemore and Sikkink. However, much as with norms in the international system, not all diasporas follow a linear life cycle, and a given diaspora may move back and forth, waning but then being reanimated. Diasporas may seem to be dead, but in fact enter a state akin to ‘suspended animation’ – a slowing or stopping of processes without their termination – only to undergo what appear to be phoenix-like resurrections. For example, the Armenian diaspora in France, long thought to be moribund, has recently sprung back to life.Footnote ⁴⁵ The life cycle works in tandem with our idea of animation: it is our contention that externally animated diasporas are more likely, all else being equal, to exhibit the form of a classic linear life cycle. Diasporas are brought to life for someone and for some purpose. Built out of the raw material of pre-existing nationals outside the state, they are mobilised through the animators who introduce money, ideas, or networks in order to stimulate political mobilisation. Importantly, though, ongoing animation is a pre-condition for continued diasporic activism. Without animation the activities, relevance, and impact of a transnational migrant network is likely to wane and dissipate.

The first stage of a classical life cycle is the birth and life of the diaspora. During this phase, diasporic consciousness emerges, a related network of political activists forms, developing shared goals and strategies. It is also likely to result in the creation of formal organisations that connect to the wider network, sometimes with connections to political parties. This phase will result from animation. Actors will contribute money, ideas, and their own networks in ways that give rise to coherent objectives, strategies, activities, and organisational structures. Such animators may include governments, transnational activists, or non-governmental entities such as foundations and think-tanks. They may be using animation as a means to directly or indirectly influence transnational politics, including to exert influence on the domestic politics of the country of origin.

The second stage is the death of a diaspora. It is what happens what the animators pack up and leave. It arises when the money, ideas, and networks that sustained the networks, organisations, activities, and strategies that made diasporic activity meaningful cease to exist or become marginal to the point of irrelevance. This may be attributable to a stark moment of change – such as regime transition in the country of origin, the results of an election, or a change of geopolitical alignments – or it may be a long, drawn-out process of decline.

The third stage relates to what we call the afterlife. It is the part of the process that struck us most starkly in fieldwork. When animators are not present or they are weak, and diasporic death occurs, there may nevertheless continue to be a residual set of actors who remain, assuming the mantle of the diaspora, despite lacking the capacity to engage in meaningful substantive activity. Aspirational but tragic political figures, community leaders seeking pay-outs from international funders who have ceased to care, and heroic humanitarians may all continue to ‘play the diasporic card’, even after the meaningful activity has ceased. Some of these people and organisations are ‘briefcase activists’, engaging in empty rhetorical games in pursuit of funding and legitimacy. Others continue, against the odds, to perform important community roles. In neither case, though, can they sustain meaningful political activity in the absence of animators.

Our theoretical framework suggests that variation along our two key axes – external/internal and institutional/networked – should predict variation in the life cycle of diasporic political activity. Of course, these are merely two of many variables. The nature of the host state and its regime type affect this process, the reason for emigration undoubtedly matters. We would also aver that geographical factors such as whether where the diaspora resides is contiguous to the homeland is of importance, as is the size of the diaspora and the timing and manner of emigration. Many of these factors are already treated in depth in the literature on social movements, particularly that sub-set dealing with transnational mobilisation, most successfully in Sidney Tarrow’s The New Transnational Activism,Footnote ⁴⁶ Terrence Lyons and Peter Mandaville’s Politics from Afar,Footnote ⁴⁷ and Gabriel Sheffer’s Diaspora Politics: At Home Abroad.Footnote ⁴⁸ Of course, these factors (and others like them) are of great importance, and where appropriate we refer to them in the body of this work. It is only our intention to argue that the character of the animators also has important effects. Concretely, the following hypotheses emerged from our cases:

• Hypothesis 1: Where animators are internal, we would expect diasporic life cycles to be less contingent. Where animators are external, we should expect them to be more contingent.

• Hypothesis 2: Where animators are institutionalised, we would expect diaspora life cycles to be more stable. Where animators are networked, we should expect them to be less stable.

The causal mechanisms underlying each of these claims are simple. In relation to the external/internal distinction, the mechanism works through interest convergence. Put simply, if animators are external to the community, they can leave at any time. Their involvement in mobilising is necessarily contingent on holding a set of interests that converge with those of the community. Once they have got what they want, they can end their involvement. In relation to the networks/institutions distinction, the mechanism works through the type of resources. Institutions are generally based on offices or roles while networks are generally based on persons. Institutions possess buildings, legal status, salaries, and recognition, which, on average, give them a greater capacity to engage in animation.

Applying the Theory to the Cases

These hypotheses were built out of a close study of the recent history of our two cases. We focus on the case of the political mobilisation of the Zimbabwean and Rwandan communities. Both relate to competitive authoritarian regimes, and both have been highlighted as putatively significant diasporas, in terms of both size and impact. In both cases we trace the life cycle of diasporic mobilisation for the purpose of contesting a particular regime, from birth to death to afterlife. Based on multi-sited fieldwork conducted across the main transnational nodes in the networks, we look at Zimbabwean mobilisation between 2000 and 2015 and Rwandan mobilisation between 1998 and 2015, reflecting the life cycle of transnational mobilisation to contest to regimes of Robert Mugabe and Paul Kagame, respectively.

Although different, the two cases offer parallel stories. They go through a period of political animation, in which diasporic consciousness is brought into existence to serve particular political purposes. Once that purpose has been served, and the interest or influence of the animators dissipates, so the political relevance of the diaspora wanes. However, crucially, even once this process has occurred, an ‘afterlife’ emerges, in which the façade of meaningful activity continues, a range of actors continue to try to mobilise resources, but the substantive political relevance of the diaspora to the contestation of political authority in the homeland has largely disappeared.

We tell this story across three parallel chapters for each case: (1) the animation process, (2) the performative afterlife of that first period of animation, and (3) the parallel interaction between diasporic politics and humanitarian action, which we use to draw attention to the gaps between elite animation and grassroots community concerns. In the first of these we document the rise and fall of diasporic activity to contest competitive authoritarianism back home, highlighting the centrality of external elite actors (‘animation’). In the second, we document what happened next. In the Zimbabwean case animators ceased to support the diaspora, and its significant political activities waned. In the Rwandan case, oppositional animation was supplanted by the pro-government diasporic animation of the homeland state (‘performance’). In the third, we show that away from the political parties, a range of actors continue to work to support their communities – sometimes heroically and sometimes problematically – within the residual diasporic space (‘humanitarianism’).

In the case of Zimbabwe, external animation supported the initial political mobilisation of the diaspora up until a peak point in 2008. This was driven by a combination of governments putting money in, but also activists and private individuals, notably an organisation called the Southern African Liaison Office (SALO), providing leadership. With the creation of the Government of National Unity in 2008, the most relevant politics shifted away from the diaspora back to Harare, animation waned and the political relevance of the diaspora dissipated. Yet there was an afterlife. Political parties – such as the regional branches of MDC in South Africa and the MDC veterans’ movement – continued to engage in a parochial politics sidelined by international donors and largely ignored by the MDC back home. A group of community-based humanitarian actors, such as those under the umbrella of the Global Zimbabwe Forum, also continued to work within the diaspora to support Zimbabwean refugees, doing important work with limited means, and struggling for vestiges of international support.

In the case of Rwanda, after initial tolerance in the late 1990s, the government began to actively dismantle the opposition abroad using a combination of threats, violence, and espionage. In its stead, it animated a pro-government diaspora creating community associations, development funds (e.g. the Agaciro Development Fund), and structured return visits to build national consciousness (Itorero Ry’Igihugu). Yet despite being effectively smashed, the dismantled opposition diaspora had an afterlife. Although ineffective, the political parties – notably FDU-Inkingi and RNC – continued to ‘wait for the winds to change’ but were continually thwarted by a combination of their own limitations and the ruthlessly effective strategy of the Rwandan government. On the humanitarian side, a range of human rights activists continued to work with the Rwandan diaspora to support the rights of Rwandans in exile. We illustratively focus on the campaign against the invocation of the Cessation Clause for Rwandan refugees between 2011 and 2013 to show how such work was only possible insofar as it was animated and in many ways hijacked by external actors.

Overall, the cases depict the tragic story of two opposition diasporas, both given life insofar as they were externally animated. Without animation, the diasporas have dissipated and waned, being left to engage in parochial political struggles of limited wider consequence. Although both diasporas have sometimes engaged in important humanitarian and human rights work, this has been either heroic but desperately under-funded or hijacked by outsiders with interests divergent to those they purported to be helping.

The Zimbabwean Case

Zimbabwean migration within and beyond Southern Africa has a long history. However a wave of mass migration took place from 2000. It was triggered by the repression unleashed following the land invasions known as the Third Chimurenga and the MDC’s successful campaign against ZANU-PF’s proposed constitution. The movements were driven by a messy mix of political, economic, and security-related push factors. Estimates of their numbers during this period range from a conservative eight hundred thousand to a preposterous four million. Dubbed ‘Zimbabwe’s New Diaspora’Footnote ⁴⁹ many of these migrants politically mobilised from abroad, notably in South Africa but also in Botswana and the United Kingdom.

A plethora of organisations was founded in South Africa between 2000 and 2008. These included platforms such as the Global Zimbabwe Forum, which at its height claimed to represent more than fifty separate organisations of the Zimbabwean diaspora; political parties (mostly, but not exclusively, the Movement for Democratic Change); associations walking a line between trauma support and overt political action; groups of lawyers; and professional associations for teachers, nurses, and others.

Within South Africa, the period from 2003 to 2008 presents a crowded social calendar of meetings, panel discussions, toyi-toyis,Footnote ⁵⁰ and other forms of social activism. Johannesburg became the switchboard for much Zimbabwean journalism, as information about the situation in the homeland was funnelled out to the world and back into Zimbabwe by a small set of journalists who also occupied key positions in the political organisations of the diaspora.

A small number of elite ‘representatives’ of the diaspora played a central role in mobilising the diaspora. Crucial to mobilisation, though, was the role played by external animators. One South African civil society organisation, the SALO (originally the Zimbabwe Liaison Office), led by Joan Brickhill, played a particularly crucial role in coordinating these elites and using personal networks to influence the South African government’s policy stance towards Zimbabwe. It worked closely with other elite organisations in order to mobilise funds. Indeed, up until 2008, a lot of money was channelled into diaspora organisations and activities by several embassies and a couple of other funders. There were notable successes in the build-up to the elections, including the involvement of the diaspora in judicial activism directed against ZANU-PF politicians guilty of torture, and in the Durban arms shipment blockade that prevented Chinese weapons being delivered to Zimbabwe.

The 2008 Zimbabwean elections, though, were the apogee of the diaspora. The Global Peace Agreement (GPA) of September 2008 led to the Government of National Unity being formed in Zimbabwe in February 2009. This led to a relocation of the most relevant politics for Zimbabwe shifting back to Harare. Activities in the diaspora became considerably quieter thereafter. Toyi-toyis became less regular, public meetings became smaller and less frequent, and the ‘general meetings’ which the constitutions of all these organisations specify must happen annually did not happen at all.

In essence, after 2008, the money and the source of external animation dried up. What follows is the afterlife of the diaspora. Obviously everything did not shut up overnight (although some organisations do appear to have died more or less instantly after the signing of the GPA). A few organisations remain genuinely and substantively active: our interviews suggest that a few isolated organisations continued to engage in real and prominent political activity after what we have set as the death date.

By the time of the subsequent round of Zimbabwean elections in 2013, what remained of the Zimbabwean diaspora was extremely limited. It essentially bifurcated into two parts: the political parties and the humanitarians. In the former case, a small group of MDC activists remain but they seem marginal to the party back home – left disillusioned that the party did not even bother to push to include the diaspora vote. An organisation called MDC-VAAFootnote ⁵¹ is active in Johannesburg but was sidelined by the mainstream of the party, and a core group of diasporic activtists had either gone home to contest the election or, in many cases, remained in South Africa to engage in a parochial politics with more focus on gaining power and recognition ‘within the diaspora’ than having any impact back home. In the latter case, a small group of under-funded community-based organisations continued to provide basic services and support to Zimbabwean refugees and torture victims, doing work of great significance, but marginalised from politics and funding.

The Zimbabwean diaspora had a life cycle: it was constructed, animated, ‘forged’ in the particular configuration suited to the times. When the exogenous currents which brought it into being dissipated, parts of it staggered on, struggling for resources, whether to support parochical political activity or genuine humanitarian assistance.

The Rwandan Case

Rwanda has a long history of migration and diaspora political movements. For example, Rwanda’s current ruling party, the Rwandan Patriotic Front (RPF), was founded in the refugee settlements of southern Uganda in 1986, and recruited heavily from Rwandan diaspora communities there. However, the genocide of 1994 ushered in a new phase of movement that has given rise to the contemporary diaspora.

The civil war, genocide, and its aftermath generated what amounts to a ‘swap’ of giant proportions: some half a million mainly Tutsi Rwandans returned home in the final months of 1994, but were more than replaced by those leaving. By the end of August, the United Nations High Commissioner for Refugees (UNHCR) estimated that there were 2.1 million Rwandan refugees in neighbouring countries located in thirty-five camps.

This first incarnation of the post-genocidal diaspora was abruptly and violently dismantled in 1996. The remains of the former genocidal government in eastern Zaire were re-arming, supporting an armed insurgency in Rwanda’s northwest, and openly planning an armed return to Rwanda. This led the RPF to arm several Congolese groups and back the AFDL rebellion to smash the genocidal state in exile, forcibly clear the camps, and ultimately remove Zairean President Mobutu from office, in what became the First Congo War.

What survived was the rump of the former government of Rwanda and their allies fighting an increasingly lonely and desperate fight in eastern DRC, the Rwandan refugee population in Uganda (some 130,000 of them), and the assorted ex-genocidaires, politicians, businessmen, and renegade priests propping up bars in (francophone) European cities. These individuals organised in an extremely diverse array of locations: near-stateless parts of eastern DRC, the refugee settlements of Uganda (and elsewhere, but mainly Uganda), and cities of European states willing to give shelter to refugees with political aspirations, most prominently Paris, Antwerp, and Amsterdam.

These dispersed groups mobilised politically and fragmented in a variety of ways. But core to their objectives was a return to Rwanda and to be able to contest the RPF. By the time of the 2003 election, the Rwandan political opposition in exile was composed of numerous fragmented coalitions, each with different backgrounds and degrees of violent and non-violent strategies for contesting politics back home. Yet they struggled to develop a coherent electoral strategy, leaving the main opposition candidate in those elections, Faustin Twagiramungu, largely isolated and acquiring just 3.6 per cent of the vote against an official vote for Paul Kagame of 95.1 per cent.

In the aftermath of the election, the Rwandan government developed a more directed strategy of actively dismantling the diaspora, not just within the Great Lakes region but across the continent and in Europe. This approach used a combination of espionage, threats, and violence. The Rwandan state kept up an unremitting campaign against its politically active diaspora on a variety of fronts: overt military action against them, the arming of proxies in eastern DRC to destroy them, the alleged deployment of intelligence agents to conduct extra-judicial hits anywhere in the world (those most commonly alleged would include Seth Sendashonga in Nairobi in 1999, Charles Ingabire in Kampala in 2011, Patrick Karegeya in Johannesburg in 2014, and a failed attempt against Jonathan Musonera and Rene Mugenzi in London in 2010), and legal activism aimed at getting any such organisations banned, entered onto terror lists, or officially classified as hate groups (as when the US government agreed to classify the FDLR as a terrorist organisation in 2005). This campaign of political dismantling has been unremitting and extremely effective. It has guaranteed that oppositional diaspora activity is decentralised, unfunded, bereft of political or organisational expertise, and confined to online activity and the occasional ineffective protest outside wherever Kagame is picking up his latest honorary degree.

From 2007, the strategy of diaspora demobilisation on the part of the Rwandan state has been supplemented by a parallel strategy of mobilisation, but this is the state-directed mobilisation of an unthreatening, non-political diaspora directed towards fundraising for the homeland, innocuous cultural activities, and the running of events congenial to the present regime’s vision of Rwanda. These organisations are supported by the Rwandan Diaspora General Directorate (inaugurated in 2007).

Such associations, dotted across the world (e.g. in the United Kingdom, in Coventry, Birmingham, Reading, and London), raise money for inoffensive causes and apolitical development projects, run football leagues or genocide commemoration days, and thereby promote a vision of Rwanda as normal, peaceful, and likeable, and the government of Rwanda as similarly bland and unobjectionable. They fund weekends for Rwandan youth in the diaspora to ‘learn about Rwanda’s culture’ (usually a potted history lesson totally in key with the narrative promoted by the government), organise docile crowds for visiting dignitaries, and so on.

Although this new diaspora is presented as apolitical, it exerts political effects by providing PR for the new Rwanda, and delegitimising opposition to the regime. In so doing, the Rwandan state is reaching out both to control ‘its’ diaspora, and in order to do that, it had to construct the diaspora itself. Yet this is not to say that there has been no ‘afterlife’ following the dismantling and reanimation by the government of Rwanda. The opposition parties in exile have tried to regroup through the formation of two transitionally mobilised opposition political parties: FDU-Inkingi, which had been pieced together from the more substantial Hutu populist parties, and the RNC, based mainly on RPF defectors.

The removal of substantive military options for contesting political authority has led to a shift in strategy, attempting to field and support candidates for the presidential election in 2010, for example. However, they have been left weak and ineffective. Their candidates, Victoire Ingabire and Déo Mushayidi, were both imprisoned upon their return to Rwanda, and diasporic political parties proved too weak and disunited to face a hegemonic, wealthy, well-organised RPF with the will and ability to repress their efforts.

What was left is discursive and symbolic politics conducted outside Rwanda with a view to undermining the RPF’s ‘donor darling’ reputation, contesting their right to speak for Rwandans and monopolise narratives of genocidal grievance and patriotic struggle. However, at each turn the political parties have been left effectively ‘waiting for the winds to change’, seeking opportunities for the RPF to become discredited with the international community. Yet each opportunity – such as the UN Mapping Report in 2010 or the UN Panel of Experts in 2012, which have implicated the RPF in crimes against humanity in eastern DRC – has led to only short-term international opprobrium without dramatic consequences for the RPF. Even high-profile assassinations such as that of Patrick Karegeya in January 2014 led to only short-term criticism, leaving FDU-Inkingi and the RNC marginalised.

Similarly to the Zimbabwe case, the other aspect of transnational mobilisation has been the promotion of human rights and humanitarian causes. One example, which we unpack, is the case of the campaign against the invocation of the Cessation Clause for Rwandan refugees. In 2009, UNHCR agreed to support the invocation of the ‘Cessation Clause’ of the 1951 Refugee Convention for Rwandan refugees, which allows for refugee status to be withdrawn when the circumstances requiring international protection have changed. It immediately led to a Tripartite Agreement between Rwanda, Uganda, and UNHCR, creating fear that Rwandans would be forcibly returned to face persecution.

The campaign was interesting because it involved transnational political mobilisation within the Rwandan diaspora. However, as the campaign evolved, the network behind the campaign and their core objectives shifted, and a breakaway campaign formed which was hijacked by a set of agendas that ultimately had little to do with refugee rights. Two individuals came to assume extremely powerful positions in the breakaway campaign, effectively coming to represent the legitimate voice of Rwandan refugees. First, Barbara Harrell-Bond, Director of the Fahamu Refugee Programme and in her early eighties, provided the main source of convening power and the moral authority that came from her background as founding Director of the Refugee Studies Centre at the University of Oxford. Second, though, the underlying agenda was set by one Rwandan refugee, Manzi Mutuyimana (known as ‘Manzi’), the self-styled ‘President’ of Rwandan refugees in Uganda, who used his position as a comprador between Rwandan refugees in the field and the global network to enhance his own power base and to imprint his own political agenda on the campaign. Overall, the story is one of a campaign captured by an elite group of external animators, largely exogenous to the local context, purporting to represent the rights and interests of Rwandans abroad.

Conclusion

The two cases provide chronological accounts of the process of external animation that has defined the life cycle of the Zimabweans and Rwandan diasporas. Reconceiving diaspora as a mode of political representation, contingently shaped by external actors, has profound implications for how we think about the very concept of diaspora. Nevertheless, the two cases examined here exhibit interesting similarities and differences in the mechanisms of animation, and the consequences this has for the diasporic life cycle in both cases.

First, and most obviously, both diasporas exhibit a huge degree of superficial fragmentation: there is a vast roster of acronyms, associations, and different organisations in both cases. However, only in the Rwandan case does this represent genuine ideological fragmentation – the Zimbabwean diaspora is, ideologically speaking, pretty coherent. By contrast, the high levels of demographic stratification (to do with location, cohort, ethnicity, legal status, class identity, etc.) are real in both cases and represent a significant barrier to a genuine mass politics. In neither case are there genuine mass organisations of the diaspora, with the possible exception of the FDLR.

Second, neither diaspora can be said to be very successful, at least in their own terms. The Zimbabwean GPA represented a high point of diaspora activism, but also one during which the MDC became convinced that the diaspora were not a relevant political force – hence their widespread marginalisation in later Zimbabwean politics and exclusion from subsequent deals. In the Rwandan case, the early attempts at armed return were violently smashed in eastern DRC, as were the later attempts at organising a legitimate civilian opposition in exile. Insofar as current Rwandan diaspora groups are successful, it is in part because their aims are too inoffensive and unambitious to possibly worry anyone.

Third, both diasporas are externally animated – which is to say, a large part of the explanation for their current political landscape, organisational structure, and roster of activities is directly attributable to non-diasporic animators (the South African politicians and activists of the first case, and the Rwandan state in the second). The key differences here then are to do with the character of the animators, which has huge effects on the shape of the diaspora which is subsequently imagined, organised, and (occasionally violently) beaten into shape. This is because the different animators had vastly different sets of resources to bring to bear, a different internal structure (a network of civil society activists, as opposed to a sovereign state), and totally different agendas. In one case, the reification of an oppositional entity with a legitimate stake in the future of Zimbabwe, a constant moral reproach to ZANU-PF, and a visible wedge with which to pressure the government of South Africa (amongst others) and the human rights establishment. In the other, the creation of a politically neutered and docile diaspora blandly serving a normalising and developmental role. This means that currently both diasporas have façade-like characters: they are invoked by others to serve political ends, rather than being independent agents in their own right. They are not in any sense substantive sociologically real entities in the way that they have to be ideologically presented, imagined, and discursively deployed – instead they have served as hollow vehicles for the agendas of others.

Fourth, the post-1980 Zimbabwean diaspora never seriously considered a violent strategy of armed return, which points to two key variables of interest: the role of territory and the role of other states. Although it is relevant that the Rwandans in Zaire between 1994 and 1996 used the resources of the state they had looted in order to fund their rebellion, Zimbabwean exiles (taken collectively) can hardly be considered poor, and so merely pointing to low levels of resources is inaccurate. Instead, the circumstances in which these resources may be deployed must be considered. Simply put, the Zimbabwean diaspora never held any territory within which they had the degree of operational independence (or tolerance from the host government) to contemplate a violent strategy, whereas in eastern DRC, the FDLR was able to strategically instrumentalise the lawlessness of the region to protect themselves as they organised, but also enlist the support of Mobutu as a willing patron for his own political games in the east. By contrast, the Rwandan diaspora in East Africa never had access to the kinds of legal and activist communities which would have made the strategies of the Zimbabwean diaspora successful: the Rwandans who tried to resist the implementation of the Cessation Clause in southern Uganda using this strategy didn’t even manage to get UNHCR to respond to their letters. This difference points to a broader continuity: both these diasporas are prisoners of their fate, and their paths were mapped out for them by their social, economic, political, and legal exigencies.