内容提要：关系方法在政治地理学和岛屿研究领域的运用日益增多，但往往不能充分反映其自身的关系定位。本文以科摩罗群岛为个案，探讨学者如何将地方置于关系的一般话语中，使地方变得易懂和易用。以关系方法为导向的相关研究通常将科摩罗构建为一个间隙岛屿、边缘岛屿或岛屿微观世界。基于科摩罗已有研究成果的解读，本文从不同关系视角阐述科摩罗群岛的历史，剖析这些视角在服务岛民需求和表征岛民经验方面的局限性。本文认为，学界须认识到在面对复杂的地理现实时，已构建的关系叙事有失偏颇，且存在简单化和政治化的倾向。
Abstract：Relational approaches are gaining strength in political geography and the field of island studies but are often insufficiently reflexive of their own relational positioning. This paper uses the case of the Comoro Islands to explore how scholars render places comprehensible and useful by setting them in general discourses of relation. In this case, relational approaches frequently construct Comoro as an island in-between, island on the edge, or island as microcosm. Based on a reading of English-language scholarship concerning the Comoros, the paper narrates the history of the Comoro Islands from these various relational perspectives before analysing their limitations in terms of serving islanders’ needs and representing islanders’ experiences. The paper ultimately argues for the necessity of scholars to acknowledge that the relational narratives they construct are necessarily partial, oversimplified, and politicised representations of a much more complex geographic reality.
1. Introduction: Familiarising the Comoro Islands
Every writer of geography is confronted by the question of where to begin.
Geographical knowledge is, after all, relationally produced, and the places about which geographers write must always be located and observed from somewhere. Researchers and readers who grapple with the profound complexity of the world are often tempted to render places comprehensible and useful by setting them in generic and naturalised discourses of relation.
The present paper, for instance, analyses the relational construction of the Comoro Islands through the lens of political geography and island studies theory. There are many ways in which the Comoro Islands could conceivably be approached, yet any scholar undertaking political geography research concerning this archipelago must start with the understanding that, generally speaking, very few of their peers will know anything at all about Comoro.
It is thus that writing about the Comoros typically begins by situating the islands in relational space. For example, Bourde (1965, p. 91) indulges in a classic island romantic style at the start of his article ‘The Comoro Islands: Problems of a microcosm’:
Far out in the Indian Ocean, stretched like a string of pearls between the Grand île of Madagascar and the coast of East Africa, lies one of the world’s most beautiful groups of islands, one of its most intriguing civilisations, and one of its most puzzling territories: the Comoro Islands. The archipelago is altogether one-quarter the size of Corsica. […] Tiny as they are, apparently unaffected by the wind of change, and isolated from the main tide of the world events, the islands still present a microcosm of the problems encountered by the developing countries on the mainland.
If Bourde’s poetic air of mystery sounds familiar, it is because it is designed to be familiarising. It indulges, in fact, in what we identify as three relational tropes commonly used for understanding islands and archipelagos, presenting Comoro as an island in-between, island on the edge, and island as microcosm.
Such a method is extremely effective. Bakar (1988, p. 181), who is himself from the Comoro island of Ngazidja, begins his own study more prosaically but no less relationally:
Situated in the south-east of the African continent in the Mozambique Channel, the Comoro Islands cover a surface of 2235 km² . The archipelago’s national territory is smaller than that of the Reunion islands (2512 km² ) and a little bigger than that of Mauritius (2210 km² ), but almost five times bigger than that of the Seychelles (444 km² ). The four islands—Ngazidja (Great Comoro), Ndzuwani (Anjuan), Mawore (Mayotte) and Mwali (Moheli) are clustered halfway between the Malagasy Republic (Madagascar) and East Africa. Ngazidja, the island situated more to the West, is 400 km from the Mozambique Coast and 700 km from Dar-es-Salam (Tanzania), while Mawore, the more easterly of the islands, is also 300 km from the West Coast of Madagascar.
This information helps situate readers with scant knowledge about the archipelago. It may nevertheless be asked: What are the political implications of such geographical situations, of the placement of objects of geographical study within readymade relational frameworks? The present paper ultimately argues for the necessity of scholars to acknowledge that the relational narratives they construct are necessarily partial, oversimplified, and politicised representations of a much more complex geographic reality. Relational thinking may be natural and inevitable, but it should not be unconscious or applied uncritically.
The Comoro Islands provide a useful case for analysing these pro- cesses precisely because they have played so insignificant a role in international English-language social sciences research that they scarcely exist in the scholarly consciousness. The vast variety of names applied to each of the archipelago’s constituent islands, reflecting differences in languages and in transliteration, further add to the Comoros’ obscurity in the scholarly and popular literature. Due to our more general preference to avoid colonial names where widely accepted Indigenous names are available, in this paper, we call the four main islands of the archipelago by the currently most widely accepted transliterations of their local, Comorian names (Ngazidja, Ndzuwani, Mwali, and Maore) rather than by their French names.
The tendency to situate the Comoros in relation to other islands and mainlands is about more than just instilling essential introductory knowledge. Alpers (2001, p. 74) asserts that “There can be no meaningful appreciation of the history and culture of the Comoros in isolation from the human currents of the Mozambique Channel and the wider world of the western Indian Ocean.” Commerce, conflicts, and connections between societies both in the Indian Ocean region and from farther afield have made the Comoro Islands what they are today.
In the academic literature, such as it is, however, the Comoros themselves have the curious habit of remaining vague. They take up space, are situated between other places, but focus often rests on their in-between or composite character. Or sometimes on how peripheral to and different they are from other places. Or perhaps on their usefulness as a microcosm, as an illustration or reflection of realities elsewhere. It is as though we must pin the Comoros to the map in order to conceive of them, to make them familiar. If the Comoros are defined primarily in relation to places that are discursively constructed as more real or graspable, perhaps as a consequence of their being perceived as more important, then how can we understand the Comoros themselves?
It is just as well to admit upfront that the present paper never arrives at an answer as to how we can gain complete knowledge of the Comoros. The paper does, however, trace out various generic ways in which the Comoros are narrated in a relational manner. Rather than asserting that the Comoros are useful for this or that kind of relational thinking, we describe and analyse the limitations of such thinking beyond its ability to make the Comoros, that is, to fit these islands into preestablished generic relational roles. This paper is, at its heart, a study of how political geography is practiced.
Because we are interested in the scholarly production of place, the empirical sections of this paper are grounded primarily in readings of English-language scholarship concerning the Comoros. Our findings would no doubt have differed had we included research in other lan- guages and had we given more attention to popular writing.
The remainder of the article is structured as follows: First, we review the literature on relational political geography and island studies. Then, in the empirical section, we narrate the history of the Comoro Islands, taking in turn three relational perspectives that are productive of the generic island: island in-between, island on the edge, and island as microcosm. We follow this with an analysis of the limitations and omissions of such generic relational approaches relative to the Comoros specifically but also political geography research more generally. We end the paper with a brief conclusion.
2. Knowing in relation
The scopes, approaches, and outlooks of geography and the related fields within area studies have always been bound up with societal developments in the centres of scholarly production (Power & Sidaway, 2017; Sidaway, 2013). As Tuathail (1996, p. 2) writes at the start of his seminal book Critical geopolitics, “Geography is about power. […] The geography of the world is not a product of nature but a product of histories of struggle between competing authorities over the power to organize, occupy, and administer space.” There has been growing attentiveness in political geography to the necessity of questioning the bases of our geographical representations and the knowledge that arises from them.
This has encouraged relational approaches aimed at providing new or enriched understandings of the objects of geography (e.g. Anderson et al., 2012; Bregazzi & Jackson, 2018; Dittmer, 2014; Ettlinger & Hartmann, 2015; Pugh, 2016; Veal, 2021). Often, however, discussions of the so-called ‘relational turn’ in geography seem to presume a shared understanding of ‘relations’ and ‘relationality’ that may not in fact exist. Anderson et al. (2012, p. 172) argue that relational perspectives “may become too familiar. Rather than a provocation that forces us to think again, relational thought risks becoming a routine to be mastered and repeated.”
Critical geographers are increasingly highlighting the how unreflexively applied relational approaches may reproduce, rather than reveal, the harmful effects of the relations they are used to study. Naylor et al. (2018, p. 199) argue that even efforts at critical geopolitics have tended to overlook “geopolitics of knowledge production, as well as the scale of empire, which creates multiple and competing peripheries and signals the need for a reframing or retheorizing that is attentive to multiple and diverse ways of knowing and understanding the world.” The purportedly “‘universal’ assemblages of sovereignty, state, citizenship and territory” can be challenged through “site-specific relational configurations of the materialization of western power” (Radcliffe & Radhuber, 2020), but such challenges cannot be taken for granted.
Our own approach to relationality follows that of Qin Yaqing, who makes relation per se the central concept in his study of world politics. Qin (2018, p. xii) holds “that self-existence is simultaneous with coexistence; that self-identity is formed in and through social relations; and that self-interest is shared with other-interest and collective interest.” From such a perspective, relationality is not just a tool that scholars can use to achieve certain intellectual ends; it is one of a number of fundamental ways of perceiving the world. If all things exist in relation, then all theory is located somewhere and imagines space from a particular location within a set of relations (Nadarajah et al., 2022). This is a relational extension of Cox’s (1986) dictum that “Theory is always for someone and for some purpose” (qtd. in Acharya & Buzan, 2009, p. 289). In other words, we share Anderson et al.’s (2012) concern that relational thinking is frequently routinised, while also underlining that all theory—even theory that is critical of ostensibly uncritical theory—itself exists in relation and must be understood as both situated and selective (Grydehøj & Su, 2022).
If, following Qin (2018), all actors exist in and are constituted by relation, then a political geography study of the Comoros—or indeed any place that one chooses to conceive of as place—requires first and foremost an understanding of the islands’ positioning in complex relational space. Such an approach is, at first glance, now typical of research within island studies, a field that has developed a tendency to challenge traditional and clich´ed representations of islands as isolated and disconnected entities. Islands are inevitably relational, and far from being merely being acted upon by the (larger) world, islands and islanders are deeply implicated in global processes. Processes of migration, militarisation, trade, security, and climate change and environmental degradation are furthermore increasingly difficult to disentangle from one another (Giuffr`e, 2021; Agius et al., 2022; Hastrup & Brichet, 2022; Kelman & Stojanov, 2020; Nadarajah, 2021; Nimführ & Otto, 2021; Perdikaris et al., 2021; Stratford et al., 2023).
Much as in the wider human geography, however, such relational investigations of islandness usually do not devote much attention to examining the nature of relation as such. It remains common for island studies researchers to deductively analyse island cases in order to make the kinds of arguments about islands that the academic field sees as appropriate and worthwhile (Grydehøj, 2017), while work that deconstructs generic island tropes (e.g., Burgos Martinez, 2021; Davis et al., 2020; Farbotko, 2010; Fuggle, 2022; Knoll, 2021; Tran, 2022) coexists uneasily with work that builds these tropes back up again.
Stratford et al. (2011, p. 114) make an early attempt at nuancing island relationalities within island studies by situating their study of archipelagic relations in contrast to:
two somewhat overworked topological relations of islands. The first predominant relation presents a clear focus on islands’ singularity, unique histories and cultures, crafted and inscribed by the border between land and sea. The second distinguishes islands from mainlands/continents, and dwells on their differences from these larger settings. […] Such rhetorics can be conscious and unconscious acts to reduce and manage complexity for convenience, benign indifference, or malevolent resolve; they produce dominant discourses about and on islands and islanders rather than with, from or for them.
This is an effort to trouble oversimplified and naturalised representations of islands, to draw attention to the way islandness is constructed through thinking about islands in relation to other places.
Drawing upon Glissant (1997) and Morton (2013), Chandler and Pugh (2020) go a step further by advocating for seeing islands within the vast, humbling relationality of the Anthropocene. Whereas Stratford et al. (2011) are interested in shifting the terms of island geography away from hegemonic discourses so as to reflect a diversity of islander experiences, Chandler and Pugh’s (2020, 2021) central interest is in interrogating the uses of islands within a world of immense spatiotemporal complexity.
Another consciously relational approach is taken by Zhang and Grydehøj (2021), who use the concept of the interstitial island to analyse the distinction between understanding islands as powerless sites at the nexus of and illustrative of global processes on the one hand and island societies as agents that actively engage with global processes on the other. Zhang and Grydehøj (2021) argue that the ‘edginess’ of islands—beyond and distinct from the city and the continent—paradoxically encourages social actors to use islands as tools for reflecting back upon and providing definition to centres elsewhere but that islanders are active in negotiating these roles.
Yet as islands come be taken as symbols of or signals within complex global-scale dynamics, it remains easy to lose sight of specific, individual islands and to instead slip into generic notions of the relational island: as an in-between place buffeted by processes beyond its control, as an outlier or place at the edge that sets processes elsewhere in relief, or as a microcosm that renders visible wider-scale processes. The generic island encompasses not only distinction and uniqueness (people believe islands to differ from mainlands and other islands) but also essentialisation and obscurity (people believe some pieces of land surrounded by water to possess stronger characteristics of islandness than others). One consequence is that scholars frequently discount, delegitimise, or overwrite island experiences that diverge too far from the generic island of our imagination (Nadarajah et al., 2022).
That is, relational movements within human geography in general and island studies in particular have the potential to create new narratives regarding a place’s spatial orientations, but if the relational analyses are themselves generalised, they will produce generic accounts of relation. In the words of Tuathail (1996, p. 34), “In shedding light upon the scenes of international politics, the geopolitical gaze is also forcing them into particular prisons of meaning. The geopolitical envisioning of the global scene is inseparable from the desire to use the displayed scene for one’s own purposes.” The imposition of “a normative system of control and intelligibility upon” the world (Tuathail, 1996, p. 34) is characteristic of all kinds of political geography (critical or otherwise) inasmuch as theory is inevitably situated and relational (Grydehøj & Su, 2022).
In the history of the Comoros that follows, we illustrate how these islands have been observed from varying—and at times, mutually contradictory—relational perspectives. Despite their frequently explicit desire to perceive the Comoros in counterhegemonic ways, relational approaches often place the Comoros within preexisting and geopolitically familiar narratives of islandness. Our descriptions of Comoro are thus intended to denaturalise relational tellings of this island place: The Comoro Islands do exist in relation, but different perspectives on this relationality are productive of different geopolitics.
3. Relating with the Comoro Islands
3.1. Comoro in-between
The Comoro Islands are often narrated as the history of a place inbetween other places. In such a telling, the Comoros become a place that is significant on account of its relations with places that are more frequently centres of scholarly attention.
The Comoros were settled as part of the wider wave of Swahili expansion that began in the 6th Century CE. Swahili culture emerged through the creation of new kinds of coastal urban settlements (LaViolette & Wynne-Jones, 2018, p. 7), particularly on small coastal or riverine islands, such as Pate Island, Manda Island, and Lamu Island in the Lamu Archipelago; and Unguja Island, Mafia Island, and Pemba Island in the Zanzibar Archipelago; Kilwa Kisiwani Island and Songa Mnara Island in the Kilwa Archipelago; Sofala; and the Island of Mozambique. That is, the Comoros were settled by a culture with roots in a different kind of island geography, the near-shore islands of the Swahili Coast.
In the 8th Century, the Comoros were likely also settled by immigrants coming from Sumatra in present-day Indonesia, then part of the Srivijaya empire (Boivin & Crowther, 2018). These settlers from the opposite end of the Indian Ocean brought with them crops and practices that would have a lasting impact on Comorian culture. Subsequent centuries saw the islands’ integration into the wider Indian Ocean trade network, the arrival of Islam, and the formation of an agricultural economy dependent on slave labour. The periodic arrival of settlers from sites across the Swahili Coast and West Asia contributed to a hybrid cultural and political system, combining Swahili, inland Bantu, Malagasy, Arab, and Southeast Asian characteristics (Walker, 2019). The Comoros “played a vital role in the commercial and religious history of the south-west Indian Ocean and as a human bridge between the African continent and Madagascar” (Alpers, 2001, p. 73).
In the 1400s, the Comoros became a site for the transshipment of Malagasy slaves to West Asia (Vernet, 2009). As Portugal conquered Swahili Coast towns in the early 16th Century, the Comoros began supplying both slaves and agricultural products to Portuguese Mozambique as well (Alpers, 2001, pp. 75–76). Wright (2017, p. 522) contrasts the Comoros’ success with the Swahili Coast colonies’ struggle “to maintain control of the all-important continental-oceanic interface:” The Comoros’ location between Madagascar and East Africa allowed the archipelago to serve the needs of traders based in India, West Asia, East Africa, and later also Southern Africa and China. Periodic and devastating raids by Malagasy pirates began in the late-18th Century, motivated in part by the market for slaves in the French plantation economies of R´eunion and Mauritius. The Portuguese Swahili Coast colonies had been conquered by the Omani empire over the course of the 17th and 18th Centuries and were subsequently conquered by the British, reducing the Comoros’ appeal as a supply point for British vessels (Dubins, 1969).
In the early 1800s, increasing French influence in Madagascar deepened links between the French empire and the various rulers in the Comoros. The French took possession of Maore in 1843, and the Comoros gradually once again became a source of unfree labour for other French colonies (Walker, 2019). At the Berlin Conference of 1885, the European powers assigned France authority over all the Comoros (Wilkinson, 1996), and France gradually gained practical control over the entire archipelago, putting an end to the system in which each island had its own ruler, power structure, and system of governance.
In the above telling of Comoro’s precolonial history, the Comoros are defined primarily through their situatedness in-between other places, serving as an interface for interaction between, for example, Swahili city-states, Indian Ocean islands, European empires, and West Asian powers. Although the Comoros are presented as playing a significant role in the history of the region, this is largely through their presence as a third party in other places’ histories of relation.
3.2. Comoro on the edge
The coming of colonialism to the Comoros tends to mark a shift in how the islands are narrated: They go from being a place in-between multiple more important places to instead being a place on the edge of a particular more important place (e.g., France, South Africa, China). This is partly because, within the colonial framing, the African and Indian Ocean island places between which the Comoros are said to sit (e. g., Zanzibar, Mozambique, Madagascar, R´eunion) no longer appear as power centres. Even studies concerning destructive aspects of Comoro’s colonial history tend to peripheralise the archipelago, push it out to the edge and turn it into an outlier that can reflect back upon conditions elsewhere.
Under French colonial rule, the Comoros’ economy shifted to an extraction-oriented plantation system familiar from other French colonies in the region (Walker, 2019). Maore was given over to sugar production, yet economic activity across the entire archipelago was eventually devoted to high-value but labour-intensive crops used primarily in perfume manufacturing (e.g., ylang-ylang, vanilla, cloves), which remain the Comoros’ largest export. Shortly before Comorian independence, Ostheimer (1973,p.493) wrote that “Today, the Comoros could certainly never be used to bolster an argument for the benefits of empire,” complaining that reliance on cash crops subject to massive price fluctuations mean the islanders “are pitifully poor, there is virtually no press and only a single secondary school. Per capita income is surely among the lowest in the world.”
The French colonial project began in Maore but ultimately sought to bring the entire archipelago under a single administration, replicating practices and personnel deployed in Madagascar (Walker,2019, p. 593). The colonial administrators constructed a structurally complex governmental system without fostering political participation, organisation, or communication (Ostheimer,1973,p.497;Bourde,1965; Sanches et al., 2022). As French focus shifted gradually from Maore to Ngazidja, a gap grew not only between Maore and the rest of the archipelago but also between Ndzuwani (formerly the wealthiest island) and the rest of the archipelago. Although the “four islands are quite homogeneous in racial, linguistic and historical terms,” it was noted even in the pre-independence era that they “also have differing traditions that have helped to produce strong rivalries that are potentially seriously disruptive” (Ostheimer, 1973, pp. 491–493).
Comoro’s post-World War II decolonisation was complicated by the lack of cohesion between the archipelago’s islands. There were tensions between Ngazidja and Ndzuwani, while many in Maore—the capital of the colonial administration up until its relocation to the town of Moroni on Ngazidja in 1958 (Dobler, 2019)—feared the loss of their privileged status. A referendum in 1974 resulted in an overwhelming archipelago-wide majority in favour of independence from France. However, 65% of voters on Maore voted against independence, and the French government went back on its earlier statements and insisted that Maore remain part of France (Daou, 2017; Massey, 2015). The government in Ngazidja made a unilateral declaration of independence, producing a jurisdictional irregularity that still exists today. Whereas the United Nations, African Union, and other international bodies regard the Union of the Comoros as consisting of all four main islands in the archipelago (Ngazidja, Ndzuwani, Mwali, and Maore), Maore continues to be administered by France, first as a departmental collectivity and from 2011 as an overseas department.
In the two decades following independence, the country was rocked by political instability and coups, at times with foreign involvement from France and Apartheid South Africa (Massey & Baker, 2009; Terrill, 1986). Under the ‘mercenary regime’ (Daou, 2017) of Nduani-born President Ahmed Abdallah, propped up by the French solider-of-fortune Bob Denard, the Comoros became a key ally of Apartheid South Africa in the early 1980s. The islands provided South Africa with a base for military, diplomatic, economic, and transport activities, allowing the country to circumvent embargoes and boycotts. Beyond the material benefits accrued by the Comoros’ civilian and military leaders, this arrangement granted the archipelago a degree of security (Terrill, 1986).
In the following period, Ndzuwani (1997 and 2007) and Mwali (1996 and 2006) (Hassan, 2009) both attempted secession. France rebuffed the 1997 request by Ndzuwani to rejoin its former coloniser, though the French state provided support for secessionist leaders (Dobler, 2019; Massey, 2015). Following negotiations between the various island leaders, a 2001 constitutional referendum led to the introduction of a federal system of government. Between 2002 and 2016, the federal presidency rotated among the archipelago’s constituent islands (excluding Maore), with five-year terms. Each of the islands had its own elected president and was largely autonomous (Daou, 2017).
In 2018, the then- and current federal president, Ngazidja-born Azali Assoumani, pushed through a new constitutional referendum in the face of a boycott by opposition parties. The new constitution removed the rotation of presidencies and permitted presidents to serve two consecutive five-year terms. Power has since consolidated in the federal government (Massey, 2021; Freedom in the world 2021: Comoro, 2021). Although the Comoros are often said to have slipped into authoritarianism, Sanches et al. (2022, p. 211) suggest that this notion is misguided: “While it is true that the Comoros do not perform as well as the other [African] islands in democracy indexes, this is mainly due to persistent political instability and inter-island rivalries rather than authoritarian excess.” Perceptions of the Comoros’ status as a democratic thus state depend on the precise relational perspective one takes.
Maore is included in the country’s federal structure by law but is excluded from it in practice. Dobler (2019, pp. 157–158) notes the difficulty in defining the process by which Maore and the other islands separated: Because Comoro was not yet a country at the time Maore declared its intent to transition from part of the colony to part of the coloniser, Maore’s departure was more a case of irredentism than secession, while France’s own actions to acquire an island that was in the process of being decolonised amount to territorial expansion. France’s claim to Maore is also used to justify its claim over the Glorioso Islands, a small uninhabited archipelago to the northeast of Maore.
At a purely practical level, the Comoros seem to be trapped on the edges of other places’ stories. Writing over three decades ago, Mukonoweshuro (1990, p. 555) describes the Comoros in a manner that would not appear out of place today:
The country has never had a viable economy, with the result that occasionally the Comorian state has had to ‘auction’ itself to the rival interests of several bidders, such as France, South Africa, the Gulf States, Mexico, Morocco and even the tiny enclave state of Gabon. This need to establish materially rewarding clientelist ties with external partners has given rise to a bizarre brand of politics, in the context of which it has proved to be virtually impossible to identify and distinguish sovereign Comorian ‘national interests’ from those of the patronizing states. Day to day governance has been so severely compromised that it is virtually synonymous with the individual private pursuits of the politically powerful.
The Comoros’ pursuit of international patronage has developed as the global system itself has developed. The country has cultivated mutually beneficial relations first with Iran and then with Saudi Arabia (Cafiero, 2016; Ali, 2018; Farrar-Wellman, 2010); has, like many other small island states before it, engaged in passport sales (Cafiero, 2016; cf. Van Fossen, 2018); has made itself amenable to flows of aid and humanitarian assistance; and has entered into bilateral collaborations with China, including as part of the 21st Century Maritime Silk Road project (Mohan, 2022; Zhong & Wu, 2020; Zhu et al., 2020, pp. 159–164).
The Comoros have experienced rapid population growth over the past decades, being home to around 191,000 people in 1960 and 870,000 people in 2020 (World Bank, 2022b). In common with many other colonised island societies, the Comoros have a large diaspora population, which has a complex relationship with the homeland. In the 19th Century, many people moved from their islands in response to warfare, poverty, or colonial taxation and land seizure, with emigrants from conflict-prone and densely populated Ndzuani often settling on other islands in the archipelago and with emigrants from Ngazidja moving to East Africa, especially Zanzibar, which became a major source of remittances (Walker, 2019, pp. 223–226). Comorian communities in East Africa furthermore nurtured early pro-independence thinking and activism through their contact with national liberation movements on the continent (Ostheimer, 1973, p. 495; Roberts, 2021).
In the 20th-Century, distinct Comorian immigrant communities formed in what was initially the French colony of Madagascar (Walker, 2019, pp. 226–228), though anti-Comorian riots in Mahajanga in 1976 led to tens of thousands of Comorians leaving now-independent Madagascar and settling in Comoros or moving overseas (Bakar, 1988). Comorians who settled in France in the late-colonial period provided a means of family or ‘chain’ migration to the colonial metropole, and there are today hundreds of thousands of Comorian immigrants and descendants of immigrants living in France (Walker, 2019, pp. 228–229; Englert, 2018). Remittances sent back to the Comoros from family members living overseas account for an estimated 13% of GDP (World Bank, 2022a). These remittances flow primarily to Ngazidja, which is the source of most of France’s Comorian immigrants (Da Cruz et al., 2004).
The situation on Maore is complex in other ways. In 2012, the Comoros had a GDP per capita of US$858, compared with Maore’s GDP per capita of US$10,270 (Massey, 2015, p. 109). Although Maore is far and away the least developed overseas department of France, individuals from other islands are drawn to Maore by the prospect of better economic conditions, public services, family connections, business opportunities, and allowing one’s future children to obtain French citizenship by virtue of being born on French soil (Massey, 2015). It is estimated that between 60,000 and 100,000 irregular migrants live on Maore, with the vast majority coming from other islands in the archipelago, particularly Ndzuani, and with many thousands of migrants drowning at sea while attempting the crossing to Maore since the introduction of the current restrictive immigration regime in 1995 (Massey, 2015). Immigrants to Maore also include individuals from Madagascar, Mozambique, Kenya, and Tanzania. Recent years have seen rising tensions between various communities on Maore as well as between the French and Comorian governments (Carver, 2018). In Daou’s (2017, p. 37) analysis, there has been a shift in public opinion on Maore: Whereas concern formerly focused on “fear of suppression” by Ngazidja, concern now focuses on “fear of being overwhelmed” by immigrants from Ndzuani. Irregular immigrants, living at the edge of or beyond the law, rely on the informal economy to make their way (Trenchard, 2019).
In the above telling, the Comoros are still understood in relation to other places, but instead of serving as a site for interaction between multiple other places, Comoro is cast as the weaker partner in one-to-one relations, as a place at the edge from a continental perspective. Colonial and postcolonial processes have effects on the Comoros, but the Comoros are not seen as actively affecting other places. The Comoros play a minor role in the stories of larger powers.
3.3. Comoro as microcosm
Both narratives of Comoro in-between and edgy narratives are apt to slip into using the archipelago as a kind of laboratory or world in miniature, in which wider (usually problematic) social, economic, environmental, and political practices play out.
Microcosmic imaginings of the archipelago often arrive with reference to its economy. Comoro is frequently regarded as among the world’s least-developed countries (Daffay & Mohamed, 2021; Ibouroi et al., 2021). 57.4% of the population engages in agricultural production, and around 90% of the country’s export revenues come from production of cloves, vanilla, and ylang-ylang. The dominance of cash crops means that the Comoros currently imports much of its food, at least on the basis of formal indicators (Comoros annual country report 2019: Country strategic plan, 2020). The islands’ informal economy and substantial subsistence agricultural production tends to be overlooked in the scholarship (exceptions include Taylor et al., 2019; Hauzer et al., 2013; Mikuˇs, 2009, pp. 2008–2009). In 2019, Cyclone Kenneth devastated much infrastructure in the Comoros and severely damaged the agricultural sector. This confluence of factors makes the islands useful for presenting a range of conditions in today’s world, from capitalist exploitation, to purported democratic retreat, to globalised industry, to climate change vulnerability, to neocolonial depredation.
Early in this paper, we saw how Bourde (1965, p. 91) grants the Comoros microcosmic significance, in the manner of so many islands, which are used to illustrate processes that occur elsewhere, perhaps serving as warnings to the world: “Tiny as they are, apparently unaffected by the wind of change, and isolated from the main tide of the world events, the islands still present a microcosm of the problems encountered by the developing countries on the mainland.” More than half a century later, Trenchard (2019) continues this narrative, updated now for the planetary threats of a new era:
The longer I stayed in the archipelago, the more I began to think of it as a microcosm of a planet nearing its breaking point. It became impossible not to see the islands’ increasingly familiar constellation of problems—a rapidly growing population, the unsustainable exploitation of resources, climate change, economic inequality, and the ensuing mass migration from the poor world to the rich—as a blueprint for many crises to come.
Such writing turns the Comoros into just another ‘correlational machine’ (Chandler & Pugh, 2021, p. 210): “The figure of the island has emerged as the emblematic canary in the coalmine […] a way of registering, sensing or revealing processes of anthropogenic influence which often otherwise would go unseen.” A microcosmic Comoro is one that lends itself to imagination as the mirror image, reversed representation, or prophesised dystopic future of the wider world. This is a Comoro that must be seen in relation to the entire world at once.
4. The political implications of relational perspectives
We may return to our introductory efforts to pin the Comoros to the map. Bourde (1965, p. 91) describes the islands as “stretched like a string of pearls between the Grand île of Madagascar and the coast of East Africa” and as “one-quarter the size of Corsica.” Bakar (1988, p. 181) is more prosaic but says much the same thing, comparing, however, Comoro’s size with that of other Indian Ocean island states (R´eunion, Mauritius, and the Seychelles).
Why is it so tempting for scholars to present the Comoros in this manner? Bourde (1965, p. 91) places the archipelago “far out in the Indian Ocean,” but such a positioning is complex: ‘Far out’ from where? ‘Far out’ from Madagascar and East Africa? ‘Far out’ from Bourde’s desk at L’Universit´e d’Aix-Marseille? It is not Bourde’s continental perspective that is the problem; the problem is his naturalisation of this perspective. This remains the case in most academic production today, even within the more critical strands of political geography, which often engage in universalising discourse and fail to acknowledge their epistemic positioning (Grydehøj et al., 2021; Naylor et al., 2018).
This is dangerous because it impedes scholarly awareness of processes that construct particular peoples and places in generic ways, making them more amenable to existing academic theories and narratives. For example, although Bourde (1965, p. 91) presents the Comoros as utterly unique, his generic wording suggests otherwise: If the Comoros are a “string of pearls,” they are in good company, for the ‘pearl’ metaphor is applied to islands and archipelagos the world over, from Bornholm in the Baltic (Hofmo, 2020) to Zhoushan in the East China Sea (China National Tourist Office, 2022) to Haiti in the Caribbean (S´eraphin & Butcher, 2018) to Samoa in the Pacific (Barradale, 1907). The Indian Ocean itself is replete with ‘pearls’, including Maldives, Seychelles, and Mauritius (Out of Office, 2022).
Walker (2019, p. 2) suggests that one reason Comoro has received such little attention from researchers is precisely that it fits uncomfortably within existing fields of area studies: “The Comoros remain obscure, peripheral to historical analysis and current events […]: not sufficiently African for the Africanists, too African for scholars of Madagascar, not really Arab but not creole like the other Indian Ocean islands either.” Studying Comoro in relation allows the islands to be placed within more familiar scholarly narratives, but Walker (2007, p. 590) also recognises the limitations to envisioning the islands as an essentialised and hybridised island society, as isolated from one another, as naturally drawn together, or as passively relational:
These movements and counter-movements ensure a constant circulation of individuals, practices and ideas both within and between islands (and beyond) and a subsequent socio-cultural cohesion that is dynamic and constantly (re-)negotiated. There are differences and similarities both within islands and between islands [yet] there is no explicit recourse to cultural symbols as national symbols: the ubiquity of the symbolic is recognized but not exploited; the differences are highlighted, perhaps in an acknowledgement of the dangers of excessive proximity; similarities with elsewhere—Madagascar, France, Zanzibar, Yemen—are promoted over local belonging. There is no explicit construction of national unity.
The peoples of the Comoros have a history of engaging in a variety of relations. Individual island identities are forged through the highlighting of similarities and differences but not in a single, easily generalisable way.
Is this really because the Comoros are uniquely complex in their relational construction? What does it mean to study the Comoros as a place? The notion of the Comoros as a unified archipelagic territory, as a set of islands that ought to be bound together in close relation, is a colonial creation, first formalised in 1946 (Dobler, 2019; Hassan, 2009). Prior to the introduction of the French colonial administration, there was neither reason, inclination, nor potential to treat the islands as an archipelago, as a single governable and comprehensible unit. Despite the four main islands’ cultural similarities, their distinct political and economic development, distinct relations with other islands and mainlands (e.g., Maore’s stronger links with Madagascar and Ngazidja’s stronger links with the Swahili Coast island cities), and simple environmental differences (which influenced potentials within agriculture, port services, and fishing) combined to create a sense of difference even before the French came on the scene. The Comoros were thus born into postcolonial statehood as simultaneously a failed economy (relative to formal, international measures) and as a failed newly independent state (relative to expectations for postcolonial nationbuilding). Daou (2017, p. 3065) shows that acknowledgement and understanding of the divisions between the islands in the archipelago vary between islands. The Comoros lack a shared perception not just of what brings the archipelago together but also of what pulls it apart. Yet even research grounded in decolonial scholarship (such as the present paper) is liable to seek to understand Comoro as an archipelago (indeed, as an incompletely decolonised archipelago), notwithstanding all the evidence suggesting that a great many people in the Comoros would prefer an individual island perspective. As Dowler and Sharp (2001, p. 170) note, “The experiences of the marginalized are used as the raw materials for postcolonial theories but this does not require an opening of the process of theorizing to the knowledges and wisdom of the marginalized.” Performing scholarship in routine and habitual ways produces routine and habitual results.
Earlier in this paper, we followed Qin (2018) in stating that the world itself is relational and must be understood relationally. At the same time, we stated that the outcomes of relational thinking depend strongly on the placements and perspectives of the people doing the thinking (Nadarajah et al., 2022). Our above description of the Comoros took on three different ways of seeing these islands in relation: as a place in-between place, as a place on the edge, and as a microcosm. Ronstrom ¨(2013, p. 153) highlights “the contrast between the immense diversity of islands in the world on the one hand, and the standardized genre of ‘the island’ on the other,” arguing that even those who set out to study an island or archipelago in its particularity often end up reinforcing the generic ‘island’ metaphor. Many efforts to understand the Comoros in relation take their points of departure in the argument that the Comoros are different, yet this difference ultimately speaks back to generic sets of politically charged geographic relations. Narratives of the Comoros often end up showing them to be different in much the same way that other island geographies are shown to be different. Difference rooted in being situated in-between, difference rooted in being on the edge or at the periphery of power structures, and difference rooted in being a smaller and more intense version of something else all rely on generic notions of spatial relation, notions that are seen as useful for understanding space beyond place. The harder we try to see the Comoros in relation-in-between, on the edge, as microcosm—the more these specific islands recede from view.
There is a “rupture between island experience and island representation” (Hong, 2020, p. 238). All too often, scholarship repairs this rupture by working to describe island experience in a way that better accords with island representation. The island is created through the interplay of both the island as an imagined space and the island as a tangible piece of land surrounded by water (Lin & Su, 2022). The island is inevitably located at the intersection of physical and mental geographies. This inevitability should be confronted not by efforts to theorise relational place and space from a position outside relation but instead by a recognition that the vastness and complexity of relation precludes any full account—and thus that simplified relational accounts are only a partial and politicised truth. To regard the Comoros as a place that gives meaning to or teaches lessons to other places is to perceive the Comoros as useful, though not necessarily as useful to Comorians themselves. The island that is a correlational machine (Chandler & Pugh, 2021) is not an island that serves islanders. The island that is a canary in a coalmine is an island that lives and dies in service to others (Grydehøj & Kelman, 2017).
One consequence of many in-between, edgy, and microcosmic perspectives on the Comoros is that the islands appear as vulnerable, helpless, hapless, put-upon, incapable, dependent, or disempowered. The islands are easily converted into ‘development space’ (Sidaway, 2007). Such representations are connected to ‘facts on the ground’, but they are also convenient for established ways of performing geopolitics. Small island states and territories in particular tend to be drawn into discourses of dependency, in which economic and political relations—processes through which all states and territories are constructed—are cast as evidence of a lack of agency (Grydehøj, 2020, p. 104).
The Comoros are relational actors with agency of their own. When the federal government seeks out aid flows from international organisations, enters into collaborations with other governments, or plans on remittances serving as an important plank in the economy, it is as a political actor relating with other political actors in a manner conditioned by its particular form of island and archipelagic geography. When individual Comorians move overseas and send remittances back to their families on the islands, all the people involved are actively participating in a transnational network of economic and social exchange: The system of remittances, grounded in kinship (Akanle & Adesina, 2017), is indicative of a kind of social robustness and expansiveness, in which Comoro is a node toward which various diasporas are directed. When ordinary Comorians grow vegetables or keep animals for household or village consumption, it is not evidence that the formal economy has failed them but that Comorian society contains informal economic capacities that allow communities to thrive.
Even explicitly anticolonial focus on the continuing effects of the Comoros’ colonial history at times risks concealing the country’s South-South relations. Showing how Comoro has been pushed away from and peripheralised by the continent is important, but so too is showing how these islands engage outside hegemonic relations. It perhaps remains easier for political geography to conceptualise relations between coloniser and colonised and between the developed and developing worlds than to perceive of relations that do not directly involve the Global North (Kragelund, 2019; Jackson, 2012). This is the case for the Comoros’ relations with other states in Africa and Asia, and it is also the case for the Comoro Islands’ intra-archipelagic relations. The Comoros’ political history has indeed been turbulent, but the appearance of this turbulence varies depending on whether one views the Comoros as a country relating to France and South Africa, as a part of the postcolonial Indian Ocean realm, as a Small Island Developing State, as an archipelago of islands relating with one another, or even perhaps as a group of islands that are each capable of determining their own relations.
All these choices and approaches are entangled in geographies of power. Our choice to pin the Comoros to this or that map and see it from this or that perspective inevitably brings certain sets of relations into focus while obscuring others. Are our perceptions of the Comoros rooted in processes of colonial map-making and map-breaking (Grydehøj et al., 2020)? Are they driven by an impulse toward conceiving of unified island and archipelagic territories (Baldacchino, 2013)? Are they perhaps focused on one selected timescale or period, to the exclusion of other histories of relation (Amundson, 2021; Farinelli, 2021)? Are they at all attentive to mythic or traditional understandings of spatial relations, which often diverge from those with which political geography is most comfortable (Chatterjee, 2022; Johnson, 2021; Schneider, 2020)?
5. Conclusion
The Comoros are relational, and they are relational in more than one way. It is impossible to take all perspectives, to envision all potential maps simultaneously. However, it is important that, when we consider islands in relation, we are aware of the consequences of our choices to do so.
Some relational perspectives are more productive of local agency than others. Some foreground an island’s resistance to the generic island role that has been projected upon it. Others foreground islands’ failure to cohere into an archipelago in the manner desired first by colonial authorities and later by the international political system. Some may assert the necessity of listening to claims of autonomy on individual islands and in individual communities, while others may assert the federal government’s need to engage in postcolonial development at a truly national scale. Those who relate the Comoros to African states may perceive an unusually democratic political system, while those who relate the Comoros to small island states may perceive an unusually authoritarian political system. To look at a map of the archipelago itself is to see Maore as a place of wealth and opportunity, while to look at a map of France is to see Maore as a site of extreme poverty and to see Ngazidja as a site of strong transnational connection. Microcosmic imaginations, meanwhile, rarely serve local needs, even when well intentioned: They emerge as yet another form of colonial extraction, always directed elsewhere and suggesting that other stakes matter more than those of islanders themselves.
There is no better solution than to approach islands relationally. But relational approaches should be critical of how and why islands are being placed in-between, on the edge, or transformed into microcosms. These approaches should be reflexive of their own positioning, frequently on a continent, from a Western epistemic starting point, and steeped in disciplinary traditions. The trouble with relational thinking is not that there is too much of it but instead that it too often situates itself outside relation while simultaneously failing to acknowledge the limits to human understanding of relational complexity.
（参考文献与注释略）
英文标题
Troubling the politics of island relation: The Comoros in between, on the edge, and as microcosm
作者简介
朱献珑：华南理工大学印度洋岛国研究中心研究员（第一作者）；
葛陆海（Adam Grydehøj)：华南理工大学印度洋岛国研究中心研究员（通讯作者）。
文章来源
Political Geography（《政治地理学》）
Volume 101, 2023, 102838
https://doi.org/10.1016/j.polgeo.2023.102838
