Beyond the Eurocentrism of immigration ethics: Tanzania and pan-African Ujamaa

Abstract

Immigration ethics debates remain deeply Eurocentric in their assumptions and focus. Due to the dominance of a universalising, liberal perspective, the thought and experience of the global south continues to be excluded, except as ‘senders’ or ‘transiters’ of people. Not only does the debate thereby misrepresent the majority of the world, it also necessarily excludes that majority from having anything useful to say about ethical approaches to immigration. In this way, it offers a partial, parochial, local theory that mischaracterises itself as international and universal. By making common cause with decolonising approaches from Latin America, this article seeks to challenge this Eurocentrism by drawing on an example of African immigration ethics: postcolonial Tanzania’s ‘open door’ era. Here, the combination of the OAU’s expanded definition of a refugee, alongside the ‘traditional’ indigenous values of Julius Nyerere’s pan-Africanism and native socialism (ujamaa), made for a generous, if highly restricted welcome for hundreds of thousands of people. This reveals the need for immigration ethics to dispense with the search for ‘universal’ norms that are limiting and exclusionary. Instead, it should explore pluriversality: the importance of local, creative, relational responses to mobile populations that are ongoing in the global south.

It is common to start an intervention on the ethics of immigration by noting that the field is relatively new, only flourishing quite recently by concentrating on the rights of states to restrict entry, the rights of people to move freely and the conditions governing the change from alien to citizen. Excellent surveys of the terrain (e.g. Fine, 2013; Sager, 2016) chart the initial forays by Walzer (1983) and Carens (1987), later leading into a debate between liberal cosmopolitans and liberal nationalists. The dominant frame has therefore become one of liberal and democratic principles and whether they can square the circle of liberalism’s claimed universalism and the apparent necessity of democracy’s particularism and closure (Cole, 2000; Parker and Brassett, 2005). Arguably, the narrowness of this debate has restricted discussion to recent examples of migratory movements. As Cole (2000) points out, for much contemporary liberal philosophy it is as though ‘colonialism and slavery never occurred at all’ (p. 197). Most importantly, it has constrained the spatial scope of debate, with a concentration on the rights (or not) of wealthy, democratic states in the global north to restrict the entry and entitlements of people moving from poorer, post-colonial states in the global south. A variation on this theme includes concentrating on the rights of people from former colonies to move to the metropole (Amighetti and Nuti, 2016).

The narrowing of the immigration ethics debate has been achieved in a variety of ways, but it has resulted in a largely Eurocentric literature which relies on concepts developed by European thinkers and their progenies, excluding and often misrepresenting societies in the global south (Reed-Sandoval and Cepeda, 2022b: 9–10). By implication, no useful ethical principles, values or practices can emerge from regions that are only included as ‘senders’ of migrants. It seems that the immigration policies, practices and principles of Colombia, Turkey or Tanzania can have nothing to add to debates on how we should think of, and deal with, immigration. In contrast, Europe, its settler colonies and its understanding of social justice (equality, liberty and their safeguarding through democratic principles and institutions) appear to have developed in complete isolation. The majority of the world’s peoples – the ‘two-thirds world’ – are thereby written out of a debate that aims to explore global or international political theory.

This article challenges the Eurocentrism of the immigration ethics debate, arguing that it can only offer a partial, parochial theory that ultimately misrepresents itself as international or global. As Reed-Sandoval and Cepeda (2022b: 25–26) point out:

‘Mainstream’ philosophy of immigration has generally been written without reference to Latin American history or philosophy, despite the fact Latin Americans (and particularly Mexicans and Central Americans) are widely regarded as ‘quintessential’ migrants in places like the United States. This is not to discredit such scholarship. . . Still, in neglecting not only ethically relevant aspects of Latin American migrations throughout their history but also Latin American philosophies about these processes, mainstream immigration philosophy delivers an incomplete vision of immigration justice.

Whilst these authors are part of a wider decolonial turn that has recently seen Latin America gain much more attention (see Mendoza, 2022; Reed-Sandoval, 2019; 2022; Reed-Sandoval and Cepeda, 2022a; 2022b; 2022c; Valadez, 2012), African history and philosophy continues to be side-lined.

This article therefore looks to push beyond immigration ethics’ current exclusion of the global south by doing two things. First, it draws out how Eurocentrism operates within the dominant debate, concentrating on three techniques that exclude societies outside Europe and North America. This comprises the first section. Second, the article challenges the implicit justification for this Eurocentrism by drawing on a time and space in which an indigenous African ethics was employed to authorise a policy of impressive, though severely constrained, generosity: Tanzania, from 1964 to the mid-1980s. Despite its poverty and newly created status, Tanzania welcomed hundreds of thousands of immigrants based on an expanded understanding of a ‘refugee’ and a philosophy of pan-African, indigenous socialism (ujamaa). The second section of the article draws out the theory and practice of Tanzania’s immigration ethics during this period, using original essays by Julius Nyerere alongside the work of anthropologists, geographers and sociologists. In the final section, the article asks what we can do with the insights gained from the ethics of non-European immigration spaces. Widening the horizons of discussion beyond abstract, liberal principles is both important in itself (to a genuinely global or international political theory), but also crucial for challenging assumptions and thinking creatively based on a wider set of experiences. Most importantly, ‘pluriversal’ immigration ethics stresses the importance of local responses based in a relational ontology, rather than seeking a new, universalising approach.

Eurocentrism and the ethics of immigration

To say that the immigration ethics debate is ‘Eurocentric’ is to say that it starts and ends with concepts and experiences emerging from Europe and its settler colonies in the global north. It indicates a vision of politics, ethics and culture that treats ‘intra-European phenomena as the starting point of modernity and explains its later development without making recourse to anything outside of Europe’ (Dussel, 2000: 469–470). A Eurocentric analysis therefore often silently assumes that, whilst other cultures were deemed to form and develop through contact with Europe, often via trade and colonial domination, the principles and values of European modernity emerged in isolation and were sufficient unto themselves. Thus, although processes of decolonisation may have removed the formal structure of political domination, Eurocentrism is part of the deeper remaining ‘coloniality’ of Western power. This ‘coloniality’ treats the European experience as ‘an exclusively European product and as a universal paradigm of knowledge and of the relation between humanity and the rest of the world’ (Quijano, 2007: 171–172). In this sense, Eurocentric knowledge does not need to make an explicit claim to universality. The fact that terms such as ‘justice’, ‘freedom’, ‘equality’ are mentioned without placing them in a specific context (e.g. ‘advanced liberal concepts of justice’, or ‘European enlightenment notions of freedom’) demonstrates the way a form of ‘Western localism’ (Mignolo, 2011: 330) silently assumes its own universalism.

We can see the Eurocentric frame operating in a variety of ways throughout the ethics of immigration literature, but I would like to draw out three in particular. The first is through the reliance on ethical and political philosophers that draw exclusively from the European enlightenment tradition, along with their core concepts and ideas – even when those thinkers themselves often do not address the issue of immigration. Carens’ famous article from 1987, for example, adapts the work of John Rawls, Robert Nozick and Utilitarianism despite none having much to say on migration. Walzer (1983) relies less systematically on the writings of Rawls, Thomas Hobbes, Henry Sidgwick and Aristotle amongst others. Liberal cosmopolitans like Benhabib (2004) develop Immanuel Kant and Jacques Derrida on rights and hospitality; Abizadeh (2008) employs a conception of autonomy from Joseph Raz; and important interventions have employed the Republican understanding of non-domination developed by Philip Pettit and Quentin Skinner (see Honohan and Hovdal-Moan, 2014). In contrast, liberal nationalists like Miller (2007, 2012, 2015, 2016) develop their thinking in conversation with the likes of Sidgwick, Kant, Rawls, Walzer and John Locke, whilst Wellman (2008) uses the work of Stuart White and David Gauthier. All thinkers are thoroughly embedded in the tradition of European modernity that appears to have emerged in isolation from the rest of the world. There is no room, for instance, for the liberation philosophy of Enrique Dussel, or the praxis of anti-colonial activist and statesman, Julius Nyerere.

A second way in which Eurocentrism emerges is by the working-through of those ideas only in relation to the experience of immigration in Europe and North America. This experience is used as examples of good and bad practice, and by constituting an audience. Some scholars make this much more explicit than others. For instance, Miller (2008) notes that he is addressing immigration as a ‘feature of political life in Western liberal democracies’ (p. 371). The nature of exclusion should be decided democratically by societies, within certain limits (Miller, 2016: 57 and 106–108). Carens (2013) gives the issue more attention, restricting his focus to advanced liberal states because his interest is in the ‘commitment to democratic principles’ that constrain these states’ behaviour and form ‘broad moral commitments that underlie and justify contemporary political institutions and policies throughout North America and Europe’ (p. 2). ¹ In doing so, he idealises liberal northern states – whose actual practices reveal much about their actual, lived principles – and ignores commitments to such principles in the global south (Bulley, 2023: 28–29). His audience is therefore ‘ordinary men and women’ in these regions who believe in democracy and rights and understand immigration in these terms (Carens, 2013: 3).

Carens briefly demonstrates awareness of this restriction’s incongruity. He notes that authoritarian states in the global south have often been extremely generous in ‘making room for’ millions of refugees, whilst democratic states in the global north have ‘devoted their energies to keeping them out’ (Carens, 2013: 220). Indeed, it is common to refer to a ‘golden age’ of African asylum – from independence in the 1960s up until the mid-1980s – in which most countries adopted ‘open door’ policies that respected refugee rights and awarded immigrants significant freedoms (Rutinwa, 2002: 15–16; see also Crisp, 2000; Milner, 2009). A similar golden age is hard to identify in the global north, making Carens’ Eurocentric idealisation of liberal values and states hard to justify. However, Carens (2013) later clarifies in an appendix that he makes no claims to universalism for his principles; he simply does not have the space or knowledge to investigate democracies or autocracies in the global south (pp. 307–308). However, universalist claims of Eurocentric knowledge continue to operate silently and by implication. Thus, Carens calls his book The Ethics of Immigration. If universality was not averred, An Ethics of Immigration would make more sense.

The reliance on European ideas and experience rarely receives the attention Carens affords it. More commonly, it emerges through the use of ‘ideal theory’ that works with concepts from European modernity but require no empirical confirmation (see summary in Reed-Sandoval, 2016); or it appears through evidence drawn exclusively from states in the global north. ² Other societies and regions may appear, in particular Mexico, Central America or Africa. But here, the global south appears solely as ‘sender’ or transiter of immigrants. For instance, Bauböck (2009: 1) begins an article by referring to ‘African boat people’ who die in ‘failed attempts to reach the shores of Europe’. Abizadeh (2008: 54) ends his important article on democratic theory and border coercion by referring to ‘Africans who risk the treacherous waters between Morocco and Spain’. ³ The reductive, stereotyping reference to ‘African boat people’ is problematic in a range of ways. But as a democratising constitutional monarchy, an increasingly important country of emigration, transition and immigration from Africa and Europe, ⁴ Moroccan experiences are only deemed relevant as a source of South-North migration. There is no room for South-South movement in the immigration ethics literature, despite the fact that most African migration is internal to the continent (IOM, 2019: 54). According to the IOM (2017), 80% of Africans who are considering migration neither wish to leave the continent, nor to move permanently. By writing out African experiences, immigration ethics debates avoid the most common form of immigration: as an everyday reality rather than a problem (Achieng and El Fadil, 2020: 2–4).

David Miller exhibits a third form of Eurocentrism in the immigration ethics literature: through forms of inclusive exclusion, using concepts that appear to expand the frame but effectively do the opposite. For Miller (2015), this means restricting his analysis of justice in immigration to legitimate states. As summarised in his most recent book, he sees a state’s legitimate right to control its borders as based in three elements: it must maintain social order and human rights; it must represent the inhabitants of its territory, in the sense of forming a legitimate authority over them; and those inhabitants must have ‘the right to occupy the territory in question’ (Miller, 2016: 59–60). Illegitimate states that do not fulfil his criteria by implication can have no relevance to questions of ethics and justice in immigration; no examples of legitimacy are offered from the global south.

Each criterion is problematic, but the final one – the right to occupy the territory – requires unpacking to see how the exclusion operates. In earlier work, Miller makes clear that the right to legitimately occupy a territory requires two elements. First, it needs a ‘transhistorical agent’, whose identity ‘can be transmitted across time’ – nations and indigenous groups, for example, that share a common identity and set of social rules (Miller, 2012: 258). Second, possession of the territory is insufficient – the agent must ‘establish a transformative relationship’ to the land, defined by adding ‘material’ and ‘symbolic’ value (Miller, 2012: 258–260). In a Lockean frame, adding material value includes cultivation and infrastructure; symbolic value is a question of adding meaning to land and buildings by connecting them to myths, ‘rituals and practices’ that build national memory. While the material value argument has been used to justify colonisers’ displacement of indigenous groups, ‘it need not be understood in this way’ (Miller, 2012: 259). Indeed, Miller (2012) adds the symbolic value argument in part to ‘do justice’ to the claims of indigenous peoples who ‘may relate to their land in a different way’ (p. 262). By maintaining the material element, with its Lockean history, a Eurocentric understanding of property and its dispossession nonetheless remains central. Once again, no instances are offered of postcolonial territories as ‘legitimate’ states.

Despite the examples, Miller’s legitimacy argument appears to open the ethics of immigration to include democratic and non-democratic states of the global south. After all, he specifies that a legitimate authority can be conferred by religion and inheritance as well as democracy (Miller, 2016: 60). And the inclusion of symbolic value adds at least some cover to the importance of material cultivation. But stressing the maintenance of social order and protection of human rights as a hallmark of legitimacy sets Miller’s conception within a broader tradition of colonial and neo-colonial interventionism that extends from the genocide and dispossession of South America to the idea of Africa as full of ‘quasi-states’ (see Doty, 1996). Meanwhile, what is to be made of the need for a people to be explicable as a ‘transhistorical agent’, with a common identity that extends across time? Does this reflect only indigenous groups in now Latin America? Or does it include ‘nations’ such as Brazil, created from the extractivist, genocidal, settler colonialism of the Portuguese, the legacy of which continued with immigration policies based on racialised hierarchies (see Nascimento and Griesse, 2022: 100–112)? How does it apply to the decolonised states of Africa, created by ‘senseless’ lines on maps or policies of divide and rule that suited only the European colonists (Nyerere, 1966)?

Silently then, and by implication, spaces such as postcolonial Tanzania are excluded from the discussion of immigration justice. After all, there was no Tanzanian transhistorical agent to be awarded legitimacy: as a state, Tanzania was created in 1964 from the union of Tanganyika (formerly part of German East Africa – along with present-day Burundi and Rwanda – and achieving independence from Britain in 1961) and the island of Zanzibar (a British protectorate from 1890 following a complex colonial history, until this status was terminated in 1963), as a direct response to Cold War rivalries after Zanzibar’s anti-Arab revolution (Shivji, 2012: 106). The territory has over 120 ethnic groups with different cultures, symbols and traditions. And this is without counting the changing population of hundreds of thousands of refugees that are usually resident. Even the name, ‘Tanzania’, was a brand-new creation – a combination of Tanganyika and Zanzibar – unheard of in any of Tanzania’s 125 different languages. Regardless of its precarious construction, Tanzania’s record on social order and rights protection (Miller’s first principle of legitimacy) is impressive – it has never experienced serious civil unrest, unlike its neighbours in the Great Lakes region, such as Uganda, Kenya, Burundi, the Democratic Republic of Congo (DRC) and Rwanda.

However, Tanzania’s ‘legitimacy’ as a representative authority in this period is much harder to determine (see Miller’s second criteria). After all, rather than a democracy, from its formation in 1964 up until the 1990s Tanzania could best be described as a socialist autocracy. It had one party, the Tanganyika African National Union (TANU) which, under its first President, Julius Kambarage Nyerere, governed continuously until Nyerere voluntarily stepped aside in 1985 to make way for multi-party elections. But Nyerere’s rarely questioned authority was also not based on religion or inheritance, so it is difficult to fit with Miller’s notion of legitimacy. And yet, despite all this, Tanzania quickly established a reputation for itself as ‘one of the most hospitable countries of asylum in East Africa, if not the world’ (Milner, 2009: 108; see also, Veney, 2007: 63). Indeed, the UNHCR praised Tanzania’s ‘exemplary record’ whilst awarding Nyerere the Nansen Medal for outstanding service to the refugee cause in 1983 (Chaulia, 2003: 147). It is to the Tanzanian case, then, that we now turn to challenge the Eurocentrism of immigration ethics.

Tanzania’s ‘open door’: Pan-Africanism, Ujamaa and Burundian refugees

The first two decades of Tanzania’s independence makes a fascinating case study for immigration ethics in part because its situation is so deeply unpromising: unlike wealthy liberal democracies of the global north, postcolonial Tanzania was a recent creation, produced from the ashes of German and British extractivist colonialism; it was also poor, predominantly agricultural and peaceful, but surrounded by unstable neighbours reckoning with their own colonial legacies. In particular, civil wars and periodic genocide in Burundi and Rwanda, but also Mozambique, alongside harsh dictatorships in Congo, Malawi and later Uganda, often made Tanzania a destination or first port of call for mobile people. Whilst Tanzania also sought to resettle South African and Rwandan refugees that had arrived in Botswana, Lesotho, Swaziland and Congo (Rutinwa, 2002: 18), most arrivals at its borders were spontaneous. Already by 1962, Tanganyika was receiving up to 100 refugees per day, as around 130,000 Tutsis fled Rwanda (Milner, 2009: 111). These numbers were dwarfed 10 years on, when Tanzania experienced its biggest population movement into the country during its ‘open door’ era. Over 150,000 mainly Hutu refugees ultimately arrived in the country after fleeing from a selective Tutsi genocide in Burundi that left at least 100,000 dead (Kuch, 2018: 109).

But if liberal egalitarian principles of human rights and free movement did not justify its ‘open door’, how did the Tanzanian government under Julius Nyerere understand its responsibilities in relation to those arriving? It is important to stress that such openness was not that unusual (Rutinwa, 2002). It built on long histories of movement and entanglement amongst the peoples and territories of the Great Lakes. After all, Rwanda, Burundi and Tanzania were previously yoked together as ‘German East Africa’. The colonial context was crucial. Nonetheless, Tanzania’s response remained ‘exemplary’. This section will draw out two elements of Tanzania’s ethos from this time – pan-Africanism and ujamaa – both of which were pioneered by Nyerere, the baba wa taifa (father of the nation), who came to be known as Mwalimu (the teacher) (Fouéré, 2014: 2; Milner, 2009: 110). It will then examine how the ethics of pan-African ujamaa played out practically in Tanzania’s treatment of immigrants.

Pan-Africanism

The Tanzanian national ethos under Nyerere was certainly one of committed pan-Africanism – the movement to encourage solidarity and unity between Africa’s peoples, which has its origins in opposition to colonialism and the slave trade. Having been an original signatory to the Organisation for African Unity (OAU) in 1963, Nyerere noted that the fight for decolonisation and independence was one for black Africa as a whole: ‘it was as Africans that we dreamed of freedom. . . Our real ambition was African freedom and African government. The fact that we fought area by area was merely a tactical necessity. . . each local colonial government had to be dealt with separately’ (Nyerere, 1966). Therefore, cooperation to achieve the liberation of Africa as a whole, as well as its movement from poverty to prosperity, formed a key part of the Arusha Declaration of 1967, in which Nyerere set out the core elements of African socialism (Tanganyika African National Union [TANU], 1967). On this basis, Tanzania supported a wide range of African liberation movements in Mozambique, Namibia, Angola, Rhodesia/Zimbabwe and South Africa (see Brankamp and Daley, 2020: 125; Chaulia, 2003: 155).

A key part of supporting black African rule was welcoming freedom fighters who were seeking shelter from colonial, oppressive, and white minority regimes. In place of an abstract, universalising responsibility for human rights and democratic freedoms, Tanzania’s immigration ethics was partly dependent upon a particular responsibility born out of the common African experience of white, colonial domination. Tanzania was therefore pivotal in supporting the OAU’s expanded definition of a refugee along the lines of pan-African liberation (Chaulia, 2003: 155). After all, such ‘freedom fighters’ would not necessarily qualify as refugees under the 1951 Refugee Convention which restricted its understanding to the ‘well-founded fear of being persecuted’ for specific reasons (Article 1.2). The OAU’s Refugee Convention of 1969 ⁵ broadened the definition to include ‘every person who, owing to external aggression, occupation, foreign domination or events seriously disturbing public order in either part or the whole of his country of origin or nationality’, has had to leave that country (Article 1.2). What would often be called ‘irregular’, ‘illegal’ or ‘economic’ immigrants in Europe and North America became refugees in postcolonial Africa.

Because of this stress on pan-African liberation and unity, all refugees were not considered equal in Tanzania. Representatives of independence struggles were treated differently, suffered fewer constraints and were welcomed by both the state and wider society ( Brankamp and Daley, 2020: 125; Rutinwa, 2002: 31). Crucially, these freedom fighters were not expected to stay long: the largesse of their welcome was partly dependent upon their returning to the fight for independence (Chaulia, 2003: 156). Tanzania’s door may have been especially open to these groups, but it was a revolving door.

Nonetheless, it is important to remember the context, relative to the states favoured by the Eurocentric immigration ethics literature. At a time when liberal democracies, particularly the US and UK, were demonstrating their respect for democratic rights, freedoms and equality by maintaining diplomatic and economic ties with racist, oppressive, white minority rule in Rhodesia and South Africa, Tanzania was using immigration policies to directly oppose these regimes. However, with the increase in people arriving from decolonised, independent African territories such as Rwanda and Burundi, values beyond pan-Africanism were required to justify the ‘open door’ of immigration policy (Chaulia, 2003: 156).

Ujamaa

States such as Tanzania, and their pan-Africanist leaders such as Nyerere, faced a difficult conundrum in advocating African freedom and unity alongside creating and promoting a cohesive national identity. Tanzanian-ness needed to be literally invented, without it conflicting with a wider African identity and solidarity. Nyerere was aware of this as a problem and contradiction (Nyerere, 1966; Shivji, 2012) and sought to bridge the gap through a broad political and economic morality. Nyerere named this ujamaa, translating the Kiswahili term as ‘familyhood’, the end and means of specifically indigenous, African socialism. Introducing the idea, it is worth quoting at length from his original 1962 essay, reprinted here in a 1968 collection:

The foundation, and the objective, of African socialism is the extended family. The true African socialist does not look on one class of men as his brethren and another as his natural enemies. . . He rather regards all men as his brethren – as members of his ever extending family. That is why the first article of TANU’s Creed is: ‘Binadamu woteni ndugu zangu, na Afrika ni moja’. If this had been originally put in English, it could have been: ‘I believe in Human Brotherhood and the Unity of Africa’.

‘Ujamaa’, then, or ‘Familyhood’, describes our socialism. . . We, in Africa, have no more need of being ‘converted’ to socialism than we have of being ‘taught’ democracy. Both are rooted in our past – in the traditional society which produced us. Modern African socialism can draw from its traditional heritage the recognition of ‘society’ as an extension of the basic family unit. But it can no longer confine the idea of the social family within the limits of the tribe, nor, indeed, of the nation. For no true African socialist can look at a line drawn on a map and say, ‘The people on this side of that line are my brothers, but those who happen to live on the other side of it can have no claim on me’; every individual on this continent is his brother. (Nyerere, 1968: 11–12)

Completely rejecting class struggle (no ‘natural enemies’) as the basis of European socialism, Nyerere stresses a mythical, traditional African family system and relied on indigenous values of brotherhood/friendship (ndugu), humanity (utu), compassion and hospitality mixed with Catholicism, Fabian and Maoist socialism (Chaulia, 2003: 154; Ibhawoh and Dibua, 2003: 62–63). Whilst the ‘African way of life’ was praised, its hierarchies, gender inequality and poverty needed reform (Nyerere, 1968: 108–110). The practical policies of ujamaa constituted the nationalisation of key industries, housing and land (TANU, 1967); independence as ‘self-reliance’, to be achieved through a stress on agricultural production and import substitution (Nyerere, 1968: 23–27); the provision of key public goods such as healthcare and education that taught African values and history (Nyerere, 1968: 44–75); and the famous ‘villagisation’ policy for rural development (Nyerere, 1968: 120–136). The latter scheme (ujamaa vijijini) was essentially the outworking of ujamaa’s ideals: accepting the agricultural economy of Tanzania and organising its development through the creation of self-reliant communal villages that cooperatively owned and worked the land whilst providing for its members’ needs.

Just as ujamaa sought to tie together pan-Africanism and the building of a Tanzanian identity, so it inevitably had implications for immigration policies as well as domestic political economy. If everyone in Africa is a Tanzanian ndugu (brother/comrade), then all Africans have ‘a claim’ on each other. Ujamaa necessitates a relational sense of responsibility, regardless of whether that ndugu is actively participating in pan-Africanism, anticolonialism and national independence. This was not an unconditional responsibility, however. Nyerere’s summary of the ‘traditional African way of life’ includes the ‘tradition of hospitality’, the universality of which allows all to feel secure (Nyerere, 1968: 5). But he stresses the Kiswahili saying, ‘Mgeni siku mbili; siku ya tatu mpe jembe’ (‘Treat your guest as a guest for two days; on the third day give him a hoe!’).

As a philosophy of development and independence, as well as a set of practical policies implemented between 1964 and 1975, ujamaa continues to be the subject of much debate and controversy – especially as the villagisation of ujamaa communities became less about persuasion and more about compulsion from 1973 onwards. ⁶ However, alongside the pan-Africanism it also sustained, ujamaa did provide the moral basis for a very porous border regime.

Ambivalent hospitality: Settlements and closing doors

Tanzania’s ‘open door’ for non-freedom fighting refugees was an ambivalent combination of generosity and heavy conditionality. Whilst the outer door of Tanzania may have been open, its internal borders were often closed (see Bulley, 2023: 102–108). As noted, Tanzania maintained a strong commitment to the key principle of non-refoulement, as well as the ‘burden sharing’ of the OAU’s Refugee Convention, even as hundreds of thousands of refugees entered the country from all over the continent. These new arrivals were offered a generous provision of land within specified ‘settlements’, access to the job market, social welfare, education, and health services – former refugees went on to achieve positions of power in the civil and diplomatic service of Tanzania (Rutinwa, 2002: 19). At the same time, Bonaventure Rutinwa describes the 1965 Refugees Act, directed at the 1964 influx of Rwandan refugees, as one of the most ‘oppressive’ laws in Tanzania – it included measures allowing the government to impound and confiscate refugee property, detain and deport them on flimsy pretexts, and restrict their movement to specific areas (Rutinwa, 1996: 292). Yet, in 1980, over 35,000 Rwandan refugees were also simply naturalised as a group (Milner, 2014: 558).

Tanzania’s was a complex, changing, non-uniform welcome. And yet, faced with the largest crisis it had seen until that point, with around 150,000 Burundian Hutus fleeing genocide in 1972, Tanzania maintained its ‘open door’ to safely welcome the majority of this population (Kuch, 2018: 112). Nyerere’s government swiftly finalised a tripartite agreement with the UNHCR and the Tanganyika Christian Refugee Service (TCRS), an NGO operated by the Lutheran World Federation (LWF). Under this arrangement, refugees were removed from the border and placed in rural ‘settlements’ – first of all in Ulyankulu, Tabora region, in 1972, followed by Katumba (in 1973) and eventually the remote Mishamo, Rukwa region, from 1979 (Malkki, 1995: 38–39). So-called ‘self-settlement’ in towns and cities was strongly discouraged, though not banned (Miletzki, 2020: 260). These rustic settlements, which the refugees themselves described as 'camps’, effectively became a durable solution that lasted until the mass naturalisation of 162,000 Burundian refugees in 2014 (Daley et al., 2018: 22).

One of the major differences between these Tanzanian settlements and UNHCR-style refugee camps, however, was the generous provision of land for cultivation. Each refugee was offered 50–100,000 square metres of farmland, allowing them to live independently with the possibility of generating a trading surplus (Chaulia, 2003: 157). This contrasts starkly with UNHCR recommendations for refugees to be given an average of 30 square metres (which would include space for shelters, roads, fire breaks, schools, and so on), plus 15 square metres each for ‘small vegetable gardens’ (UNHCR, 2007: 210). Crucially, these settlements in Ulyankulu, Katumba and Mishamo were enrolled within, rather than separate from, the development policies of ujamaa. The aim was for these settlements to clear and cultivate virgin forest and bush, particularly in Mishamo; they were to be self-sufficient and self-reliant, allowing the potential of a long-term stay from the beginning. They essentially formed ujamaa villages for refugees (Milner, 2009: 110–111) – a way of enacting a responsibility for non-Tanzanian, African ndugu that was not dissimilar to the native population. Indeed, these refugee settlements quickly sustained themselves, with surpluses traded in the local area; in 1985, the tripartite agreement effectively came to an end, with the UNHCR handing sole management of the settlements to the Tanzanian government (Miletzki, 2020: 260). In-keeping with the principles of ujamaa, villages and refugee settlements alike became self-reliant, which in turn was meant to allow for national, Tanzanian self-reliance instead of dependency on the UNHCR and NGOs (see Nyerere, 1968: 129).

The creation of ujamaa villages always contained more compulsion than Nyerere and TANU claimed, and this coercion became explicit from 1975 onwards (see Lal, 2010). But there was also a qualitative difference between these two fabricated communities. Unlike ujamaa villages, refugee ‘settlements’ sought to maintain firm control of Burundians’ movement within Tanzania. Leave passes, granted by the Commandant and Village Chairman, were required to exit the settlement as well as bribes to ensure police would respect such passes (Kuch, 2016: 477; Malkki, 1995: 138). Meanwhile, the rural remoteness of Mishamo, as well as its surroundings of thick forest, swamps, bush land and tsetse flies, ensured that movement was terribly difficult (Malkki, 1995: 39–40).

Liisa Malkki’s famous anthropological study of Burundi Hutus who settled in either Mishamo or the town of Kigoma, nearly 200 km away, notes that these forms of control had unexpected effects. Whilst town refugees effectively shook off their identities, allowing them to blend in and avoid the rising restrictions on refugee movements that emerged in the 1990s (Malkki, 1995: 153–183), ‘camp’ refugees did the exact opposite. Stasis, the need for self-reliance, isolation from Tanzanian society, and the spatial organisation of the settlement meant that residents of Mishamo interpreted every aspect of their existence as an attack on their identity – in response, a Burundian Hutu purity was created, an identity which involved not integrating, with naturalisation seen as a threat rather than a solution, and repatriation the only option (Malkki, 1995: 230).

As Tanzania descended into political and economic turmoil in the 1980s, Nyerere voluntarily stepped down and ujamaa was abandoned in favour of Structural Adjustment Programmes, internationally imposed liberalisation and IMF/World Bank loan conditionality (see Milner, 2009). The ‘golden age’ of Tanzanian and African asylum passed (Rutinwa, 2002), replaced with a Europeanisation of immigration policies that blamed Nyerere and immigrants for Tanzania’s continued poverty (see Bulley, 2023: 110–113). Nonetheless the boundaries that separated refugee settlements from Tanzanian society were largely dismantled during this era. Scholars who have since studied Mishamo (Hovil, 2018) and Ulyankulu (Daley et al., 2018; Kuch, 2016; Miletzki, 2020), observe the way forms of incorporation have occurred through what Janna Miletzki calls a ‘pragmatics of belonging’. Living together, trading, business links, intermarrying and speaking fluent Kiswahili produced a de facto integration (Kuch, 2016: 476; Miletzki, 2020: 266; Whitaker, 2002: 344), even before naturalisation was offered in 2007 and finally delivered in 2014. Belonging has been negotiated, creating ‘new forms of attachment – which, in turn, challenges notion of “guest” and “host”’ (Hovil, 2018: 30).

That said, the passing of Africa and Tanzania’s golden age of asylum has also brought major new restrictions and closed doors for refugees (Miletzki, 2020: 262–266). Rwandans and different Burundian ‘case loads’ were deported en masse during the 1990s; Tanzania’s immigration policies were markedly variable across the time and place of different arrivals (see Whitaker, 2002: 349–355). But the specific practices of welcoming and naturalising immigrants, as well as the particular indigenous philosophy and values on which it was based, are nevertheless historically important and surely offer something significant to immigration ethics. But what is this ‘something’?

Towards a pluriversal immigration ethics

There is an irony in relying on the ideas and practices of Europe and its settler colonies – some of the most anti-immigrant states in the world – to ground an ethical approach to immigration. In contrast, Mendoza (2022) suggests that we can start to reach beyond the Eurocentrism and coloniality of existing immigration ethics by ‘making a concerted effort to work from and build on the frameworks developed by non-Western radical thinkers’ (pp. 72–73). This article has sought to do precisely that, undermining the Eurocentrism of immigration ethics by starting from the complex historical case of Nyerere’s postcolonial Tanzania. Though never a dictator, Nyerere was certainly a controversial and problematic character in many respects (see Shivji, 2012). Indeed, the radicalism and indigeneity of his thinking have both been challenged, with ujamaa relying on a developmentalism that reduces rural Africans to agency-less victims (Ibhawoh and Dibua, 2003: 61), and producing villages that were far from reflective of African traditions (Lal, 2010: 2–3). There is also a danger of romanticising Tanzania’s ‘open door’ which, as noted above, was never as uniformly open as sometimes implied. However, Nyerere was a genuinely radical thinker when it came to constructing a national identity on the basis of openness to, and responsibility for, outsiders. This section situates the Tanzanian contribution to immigration ethics by placing it within decolonial opposition to universality in favour of the pluriversality of politics and values. Whilst decolonial pluriversality has prompted important debates in international/global ethics (Dunford, 2017; Fitzgerald, 2024; Hutchings, 2019), these have not yet made their way into immigration ethics.

Decolonial critique begins by debunking the ‘one world’ ontology of Eurocentric modernity that justified colonial practices and continue to undergird the coloniality of Western power (Hutchings, 2019: 116). European modernity’s ‘one world’ ontology ‘dictates a belief in one reality, one world, one truth, and ultimately one very blinkered vision of the possible’ (Escobar, 2020: 21). Though many cultures and beliefs exist in this ‘one world’ vision, these ultimately give way to the universal, scientific truth emerging from European culture. The violent spread of this universal way of knowing and being continues after formal decolonisation. We see this persistent coloniality in the devaluing and erasure of non-Western ways of life; the hierarchy imposed between different cultures; and a linear understanding of time and development in which Western truth and being will gradually triumph over backward indigenous beliefs (Escobar, 2020: 13–15). Coloniality assumes the universality of one world, with its limits of possibility characterised by capitalist economics, representative democratic politics and individualising, liberal ethics. One small example was outlined above in the Eurocentrism of immigration ethics literature, whose ‘one world’ ontology outlines the ethics and justice of immigration.

Decolonial critique seeks to ‘disconnect’ or ‘delink’ from this Western ontology, epistemology and teleological temporality, in order to ‘relink’ with the practices of thinking, living and doing that are erased by the globalisation of European capitalist modernity (Mignolo, 2011: 235; Mignolo and Walsh, 2018: 147). This process is captured by the decolonial stress on the pluriversal. In direct opposition to the universal vision of ‘One World made from one world’, Escobar (2020: 26) and Mignolo (2011: 273) characterise the pluriverse via the Zapatista’s dictum: ‘a world in which many worlds might fit’ or ‘coexist’. Adopting a decolonial, pluriversal perspective allows us to see and value multiple worlds: indigeneous ways of being, knowing and relating that have been systematically erased by the universalisation of European thought and culture. Pan-Africanism, ujamaa and Tanzania’s ‘open door’ era are just one world of possibility. Exploring such local and particular worlds allows us to provincialise European thought and experience (Chakrabarty, 2000; Dussel, 2000). Eurocentric perspectives thereby become a form of ‘Western localism’ with a ‘global design that became synonymous with universalism’ (Mignolo, 2011: 330). From a pluriversal perspective, the coexistence of and cooperation between compatible worlds is both possible and something that can be promoted – but conflict and antagonism is guaranteed with those worlds that continue to insist on their own universality (Mignolo and Walsh, 2018: 175, fn. 18).

In order to consider the value of Tanzania’s ambivalent hospitality, we must first understand the ethics of receiving immigrants as pluriversal rather than universal. Once we relink with Tanzania’s postcolonial experience, the multiplicity or pluriversality of worlds becomes a fact rather than an alternative perspective. The necessity of pluraversalism is therefore a major contribution to the field of immigration ethics. Once this is accepted, however, a range of further insights emerge from this ‘open door’ era that could be drawn on by scholars and activists that look to exceed the possibilities imposed by the dominant form of Eurocentrism. I will concentrate on three here.

First, in line with decolonial thought, Tanzania’s ‘open door’ era demonstrates the importance of a relational ontology for opening up creative possibilities in response to population movements. Liberal immigration ethics assumes a socially atomised ontology, whether that be in the form of autonomous individuals (and their right to free movement) or states (and their right to self-determination): the separation of people and their territories is assumed and naturalised, with rights held over or against others (see Bulley, 2023: 39–46). In contrast, pan-African ujamaa was based in an ontology of fundamental interdependence emerging from connected histories of racialised colonial domination, the fight for independence and indigenous values associated with family and community. Such interdependence was summarised by Nyerere’s catchphrase: ‘I am because you are’ (in Chaulia, 2003: 154). This is what Escobar (2020) refers to as a form of ‘radical relationality’ common to pluriversal indigenous philosophies, in which the entities constituting the world are ‘so deeply interrelated that they have no intrinsic, separate existence by themselves’ (p. xiii), producing a sense of profound responsibility for others.

Tanzania and Tanzanian identity were, as outlined above, postcolonial constructions that both enabled responsibility-taking, especially for African immigrants, but also endangered this relationality through separation and ‘self-reliance’. Whilst this may seem entirely sui generis, what it highlights is a truism of international political theory: all states, societies and identities are unnatural constructs, the products of entangled histories of relation and separation. This contains a lesson for societies everywhere: adopting a relational ontology opens our horizons of possibility for creative responses to immigrants. For European moderns at least, this requires a ‘journey of deindividualization and towards recommunalization’, in which we learn to accept obligations that Eurocentric thought has taught us to deny (Escobar, 2020: xvii). ⁷ But without this journey towards a relational ontology there is no possibility of learning from Tanzania’s ambivalent hospitality: our ‘worlds’ of immigration ethics would remain incompatible.

Second, the Tanzanian case demonstrates the importance of developing immigration ethics from local narratives, values and histories. The intention has not been to argue that ujamaa and ‘traditional’ African hospitality be universalised or simply transposed wholesale onto other societies. I am not suggesting that European states should enact a form of traditional pan-African socialism in response to its recurring ‘migrant crises’. Staying with pluriversalism, Tanzania’s ‘open door’ shows what is possible when building from local, irreplicable worlds. Instead of looking to this time and space to tell us what a just immigration system would look like, it directs our attention to the way regional relationality and responsibility (pan-Africanism) were negotiated alongside a particular, national identity project and vision of justice (ujamaa). It draws our attention to the way that ethics, of immigration and otherwise, are ‘about finding new ways of relating to ourselves and to each other in our pursuit of whatever we think of as justice’ (Hutchings, 2019: 121). Local contexts must look to their own relational ontologies, histories and values to build creative responses. Whilst this may prompt charges of moral relativism, such criticism assumes the priority and possibility of moral universalism, something that is already ruled out by pluriversality. Moral universalism is the problem, not the solution. The point is not that pan-Africanism, ujamaa and the uncertain welcome they allowed were right or just; rather, the creativity of welcoming hundreds of thousands whilst providing land in self-sufficient, severely restrictive settlements emerged from choices between ethical priorities ‘within complex and power-laden situations’ (Hutchings, 2019: 123).

The final insight can only be posed as a question due to space constraints. Are inventive responses to immigration possible under capitalist conditions? Tanzania’s indigenous socialism was formed in direct opposition to Western capitalism, which Nyerere (1968) saw as individualist and competitive, based in the ‘exploitation of man by man’ (p. 12), and preventing the development of African socialist society (TANU, 1967) outlined above. The spread of European capitalism and its individualist ethos of exploitation is likewise a central plank of modern coloniality, wiping out indigenous forms of relationality (Quijano, 2007). It is significant that the nationalisation of Tanzania’s economy was immediately obstructed and undermined by three large British banks, aiming to ‘destroy international confidence in Tanzania’s export economy’ (Ibhawoh and Dibua, 2003: 65). Meanwhile, Tanzania’s ‘open door’ era formally ended with the failure of the ujamaa experiment, prompted by the global recession and debt crisis of the 1980s (Milner, 2009: 114–115), the expensive war in Uganda to depose Idi Amin (Shivji, 2012: 111), and the subsequent acceptance of loans from the World Bank and the IMF on the basis of denationalisation, massive cuts in spending and the embracing of free market capitalism (Veney, 2007: 66–67).

However, ujamaa was equally opposed to ‘European socialism’ and its basis in ‘class war’, seeing it as a ‘philosophy of inevitable conflict between man and man’ (Nyerere, 1968: 12). These oppositions are echoed by Mignolo’s (2011) decolonial critique, in which both capitalism and communism operate within the same ‘one world’ ontology that seeks to universalise European experience (pp. 325–326). In this sense, it becomes clear that immigration ethics needs to pay more attention to the impact of political economy in forming horizons of possibility for creative responses to population movements. More generally, the question to be posed from the Tanzanian experience is whether imaginative, local responses to immigration are even possible in the global north, given the widespread acceptance of neoliberal capitalism and its non-relational ontology?

Conclusion

Whilst many other areas of politics and IR have been seeking to challenge their Eurocentric assumptions and decolonise their knowledge production, the immigration ethics debate has remained relatively immune. Due to the dominance of a universalising, liberal perspective, the thought and experience of the global south continues to be excluded, except as ‘senders’ or ‘transiters’ of people moving towards Europe and North America. Not only does the debate thereby misrepresent most of the world, it also necessarily excludes that majority from having anything useful or legitimate to say about ethical approaches to immigration. In this way, it offers a partial, parochial, local theory that mischaracterises itself as genuinely international or global.

By making common cause with decolonising approaches from Latin America, this article has sought to challenge the Eurocentrism of the literature by drawing attention to a particular example of African immigration ethics: postcolonial Tanzania’s ‘open door’ era. Here, the combination of the OAU’s expanded notion of a refugee, alongside Julius Nyerere’s pan-Africanism and native socialism (ujamaa), made for a generous, if highly restricted welcome for hundreds of thousands of people. Many other examples could have been chosen, including more recent responses to the exodus of people from Syria and Venezuela, using local interpretations of Islam (Turkey) and friendship (Colombia). ⁸ The aim has not, however, been to generalise or universalise from the example of Tanzania. Rather, by stressing ‘pluriversality’ the objective has been to show how pan-Africanism and ujamaa challenges assumptions, demonstrates the possibilities of creative thinking, and underlines the importance of thinking in local terms, from a relational ontology that questions capitalism, rather than creating new universal norms and imperatives that can work for all societies. To move beyond Eurocentrism and the limitations it imposes, immigration ethics needs to step into the pluriverse.