On the fringes of empire? Rethinking suburbs as colonial spaces in early modern South and Southeast Asia

Abstract

This survey challenges conventional perceptions of colonial suburbs in the early modern Indian Ocean world in general, and those under the rule of the Dutch East India Company (VOC) in particular. Recent urban research advocates re-evaluating suburbs as intricate and diverse spaces, yet this shift has had limited impact on historical studies of early modern colonialism. The survey highlights the importance of recognizing suburban regions in eighteenth-century settlements such as Cape Town, Cochin, Colombo or Batavia, where significant population growth resulted from inter-Asian and internal migration. These areas fostered ethnic and cultural diversity, disrupting normative ideas of segregation. By shifting the analytical focus from the core to the periphery and exploring colonial histories from an outside-in perspective, the contribution emphasizes the potential of a more horizontal approach to sub/urbanity for understanding early modern colonial societies, encouraging scholars to delve into the intersection of ‘the imperial’ and ‘the urban’.

For much of the twentieth century, the conceptual and methodological underpinnings of research in urban studies was dominated by inside-out theorization. Rooted in the early twentieth-century Chicago School of urban sociology, this model idealized city centres as hubs of civilization. These physical hotspots were believed to be perfect loci where economies of agglomeration thrived, and fostered spatial clustering and natural spillover effects. In stark contrast to this flywheel of modernization at the heart of the city, the outskirts were viewed as stagnant suburbs, lacking political authority, economic vibrancy or dynamic social interactions. Nothing good or interesting could come out of suburbs, or as former New York City Mayor Ed Koch painted them in a 1982 interview: ‘It’s sterile. It’s nothing. It’s wasting your life.’¹

Since the turn of the century and especially in the past decade, urban research has witnessed a notable paradigm shift.² With suburbanization as the driving force of twenty-first-century urban development, scholars have called for a re-evaluation of the traditional city-centre-focused model.³ Under the impetus of academics such as political scientist Roger Keil, the long-standing ‘myths’ and stereotypes associated with suburbs are gradually being dispelled.⁴ Suburbs are no longer considered mere ‘geographies of nowhere’, lost zones between monolithic conceptual expanses of ‘city’ and ‘countryside’, but rather as extremely heterogeneous as well as complex entanglements built up from historical layers.⁵ They are re-examined as unique physical and abstract spaces, where multiple social identities are shaped, creative economies flourish and layered sovereignties intersect.

This remarkable evolution in contemporary urban research has not yet created the expected ripple effects on historical analysis of urban environments. Admittedly, urban theory and urban history have had different trajectories in the past 50 years.⁶ Historians have followed the paradigm shift to some extent and traced the construction of the suburban myth back to the late nineteenth and early twentieth centuries.⁷ But despite strong calls to action by historians in the past 15 years, comparative studies of suburban environments prior to 1850 remain scarce in urban history as a field.⁸ The aim of this survey is to move these discussions back into the eighteenth century and argues that employing contemporary concepts of suburbanization to analyse cities in the early modern Indian Ocean world and Southeast Asia serves to challenge existing archival limitations and the intellectual complacencies they foster. Using existing historiographies on Dutch colonialism as a feasible vantage point for such an exploration, this survey hopes to inspire other researchers to shift the analytical focus from the urban cores to the peripheries and explore colonial urban histories from an outside-in perspective.

Crowbarring colonial cities under Dutch East Company rule

Examining the historiographies of European urbanization in early modern Asia in general⁹ and Dutch colonialism in particular, one truly wonders why this revisionist literature from urban studies that challenges the urban/rural divide did not have a more significant impact.¹⁰ After all, post-colonial histories of these regions in the last four decades have built a strong track record of abandoning simplistic analytical dichotomies, first and foremost the socio-political binary of colonizer/colonized.¹¹ The traditional consensus among historians has been that the Dutch East Company (VOC), its civil institutions and officials created this attractively simple image of colonial societies in the seventeenth and eighteenth centuries. Perhaps more importantly, the VOC archives perpetuated epistemological inequalities with a hierarchy that put European fortified settlements at the centre of colonial narratives.¹² Scholars such as Stoler advocated discursive methods to try and read against – or even beyond – this archival grain to better understand early modern colonial societies.¹³ More recent contributions to the field stress the need to further disrupt these traditional colonial categorizations not only as epistemological, but also as analytical tools to tackle the skewedness of colonial archives and answer more complex historical questions.¹⁴

Despite the challenges posed by these colonial archives, crowbarring this simplistic colonizer/colonized divide from the mid-1980s onwards resulted in more intricate pictures of the multi-ethnic urban melting pots of overseas settlements in the seventeenth and eighteenth centuries under VOC rule.¹⁵ But although this wave of social urban histories at the turn of the twenty-first century clearly discarded the binary colonizer/colonized – even beyond the lacklustre integration of a mestizo middling group – they adhered to the city-centre-focused model that dominated historiography in urban research for such a long time. The fortified settlements – often on the seashore – remained literally and metaphorically at the core of most analyses of early modern Dutch urban settlements in present-day South Africa, India, Sri Lanka and Indonesia.¹⁶ Mesmerized by the colonial gaze of the archive, concentric histories were constructed looking outwards from the fortified European settlement over the direct suburbs, the surrounding areas (e.g. Ommelanden or Four Gravets) and the further-off rural border areas.¹⁷

Parallel to these localized histories of colonial ‘Dutch’ cities, other VOC historians inscribed themselves onto the burgeoning field of global history. In a lengthy 2004 essay on Dutch settlements along the Indian Ocean, Vink laid out five spatial approaches that would allow narratives between the local and the global to interact with world history. He differentiated between (1) port city ‘cum umland’, hinterland and foreland networks, (2) frontiers/borderlands studies, (3) ocean- or sea basin-centric and macroregional analysis (4) area or civilizational studies and (5) world-systems and dependency school.¹⁸ Although each of these spatial approaches have been explored since Vink’s 2004 study, numbers 1 and 3 have particularly stood the test of time. First, within the network model, port cities have been analysed as ‘European’ gateways into an ‘indigenous’ hinterland.¹⁹ Although initially used mainly for writing economic histories of the Indian Ocean, Ward opened up the model in her 2009 seminal work to include the (forced) movement of people between the different nodes of the Dutch imperial network.²⁰ Because this approach gravitated toward the interconnectedness of port cities, it inadvertently coagulated suburban areas, mixed urban/rural areas, borderlands and even overseas settlements into a conglomerate of ‘territorial and oceanic hinterlands’.²¹

Other literature scaled up from the local to the global and absorbed a transoceanic approach to Dutch colonialism, close to Pearson’s concept of littoral societies.²² That historiography tended to portray the Dutch as an archetypical ‘alongshore empire’, where an archipelago of cities were ‘connected more by water than by land’.²³ Worden for example wrote about Cape Town as a perfect example of a Dutch settlement that had more in common with other settlements across the ocean, than with its direct hinterland.²⁴ Even more so than in the port city/hinterland model, the transoceanic gaze failed to adequately address the specificity of suburban areas. This is certainly the case for early phases of the ‘colonial transition’ in the eighteenth century, when dwindling profit margins in maritime trade increasingly forced the attention of local VOC colonial governments across the Indian Ocean towards increasing direct and indirect revenues from their ‘hinterlands’.²⁵ However, in both the port city/hinterland approach and the ocean- or sea basin-centric approach, ‘European’ city centres remained at the analytical heart, with a conceptual focus on Europe and the Europeanized colonial town.

This vast literature of the past 30 years has undoubtedly pushed the historiographical boundaries of Dutch colonialism in the era of the VOC. The traditional trade-oriented focus was abandoned to look beyond commercial endeavours and study different ways in which Dutch institutions and officials tried to shape colonial societies.²⁶ Although this historiography was continuously infused by breakthroughs from parallel disciplines, the paradigmatic shift in contemporary urban studies towards suburbanization hardly broke through. In the above-mentioned spatial models, the maritime character of Dutch port cities across the Indian Ocean and in Southeast Asia was foregrounded, often reducing the direct hinterland and further off interior to merely serving the economic and labour needs of the port city. When urban experiences could be studied in more localized historiographies, urbanity seemed to be confined to the fortified settlements and their relatively small ‘Europeanized’ communities, in stark contrast to the ‘indigenous’ rural areas beyond those walls. In sum, the abandoned inside-out theorization of twentieth-century urban studies still simmered in the twenty-first-century urban histories of the colonial Indian Ocean and Southeast Asia.

In the wake of recent critiques of both the maritime gaze and the network model for studying global and colonial histories of the Indian Ocean, this survey reconsiders the suburban as a socio-spatial vantage point.²⁷ All too easy dichotomies between the ‘European’ city and ‘Asian’ hinterland should be abandoned. But in defence of colleagues who wrote about colonial settlements in the past, it is not an easy feat to establish spatial boundaries or define the precise characteristics of colonial suburban spaces – or any historical suburb for that matter.²⁸ Early European visitors to ‘Portuguese’ or ‘Dutch’ cities in maritime Asia expressed their amazement at how green, open, spatially scattered and village-like these settlements were.²⁹ To avoid getting stuck in the ‘skewed and misleading Anglo-Saxon historical trajectory as a general, linear model’³⁰ of suburbanization, it might be better to think of the suburban areas adjacent to colonial settlements as ‘disjunct urban fragments’, a contemporary concept launched by Roger Keil, with roots in the work of Henri Lefebvre. In essence, Keil advocates a more horizontal approach to urban forms with multiple centralities, and discards the idea of suburbanization as dependent on one urban centre. Sub-urbanization and the more recent peri-urbanization (unplanned suburban settlements in the post-colonial Global South³¹) are essentially Western academic concepts, implying a Western grasp on urban space that is better understood not only from the outside in, but also, and perhaps primarily, from the vantage point of the local population itself.³² How did non-Europeans view these settlements (both the Europeanized centre and the disjunct fragments surrounding it), their growth and their place within local urban networks, of which European settlers and traders were not always a part? When applied to areas surrounding fortified ‘European’ or other imperial settlements around the Indian Ocean and further into Southeast Asia, these zones could be considered hybrid and open-ended landscapes where early modern urban/rural land use and commercial functions mingled in a fragmentary fashion.³³ Throughout the above-mentioned histories of VOC settlements, the potential of a horizontal approach to suburbanization to rethink and reframe existing tropes in historiographies already shines through. When reconsidering urbanity from the outside in, we can discover the emergence of heterogeneous societies beyond the walls of Cape Town, Cochin, Colombo or Batavia, which were perhaps more characteristic of the unique intersection of colonialism and urbanity than the worlds within those walls.

From the fringe to the fore: disjunct urban fragments under VOC rule

From the time the Dutch set up their first settlements across the Indian Ocean and Southeast Asia in the early seventeenth century, the small group of European migrants, their families and enslaved living within the forts was outnumbered by the suburban population. To protect social homogeneity and inspired by colonial anxiety, Dutch local governments promulgated settlement laws and pursued a certain degree of ethnic segregation.³⁴ In seventeenth-century Batavia for example, the Javanese were banned from living in the walled city, and forced into ethnically inspired quarters or kampongs in surrounding areas, as were the Balinese, Bugi, Madurese, Ambonese and Chinese.³⁵ The social composition of these settlements became more or less fixed.³⁶ In addition to lawmaking of this kind, market forces such as high rental prices for stone houses also further drove Asian or poorer mixed-descent inhabitants out into the suburbs.³⁷

Parallel to the discarding of the dominant dual city model in literature on modern Asia,³⁸ historiographies of early modern Asia in the last decades have also abandoned this rather normative view on ethno-diversity and spatial segregation. They have turned instead to everyday cohabitation in these colonial societies that took place regardless of settlement laws.³⁹ Although socio-spatial boundaries were installed by colonial authorities that distinguished intra-urban neighborhoods as well as cities from suburban zones, the various social and ethnic groups crossed paths on a daily basis in and around Dutch settlements.⁴⁰ Muslims and Chinese in Batavia were constantly moving around between areas of the city and the suburbs, although the law forbade them to do so.⁴¹ When the population numbers in the fortified settlements of Batavia, Colombo, Cochin and Malacca stagnated from the late seventeenth century, the merchant elite moved to the surrounding areas.⁴² As a result, the suburbs of VOC cities around 1700 were no longer a mere conglomeration of scattered and disconnected village-like settlements, but an integral part of the urban experience in the colonial Indian Ocean and Southeast Asia.

In 2019, Vink gathered the available census-like data from the colonial archives to measure the growing Dutch littoral footprint across VOC territories.⁴³ Despite potential Stoler-inspired questions about the limitations of the colonial archive and the inherent colonizer’s gaze, his analysis indicates a remarkable growth of the VOC subject population from about half a million individuals at the end of the seventeenth century to 2.5 or even 3 million a hundred years later. Over the course of a century, Batavia, the largest hub in Dutch Asia, grew to a city of over 100,000 inhabitants, making it one of the largest European power centres in Asia next to Portuguese Macao and Goa or Spanish Manila.⁴⁴ This rapid population growth across VOC territories was not so much caused by European immigration or forced migration of African and Asian enslaved or exiled, but instead was to a large extent the result of inter-Asian migration and internal migration from the countryside to suburban areas. As the boundaries of fortified settlements were more or less fixed by walls, it was the suburban areas that absorbed this exponential population growth. Batavia, for example, witnessed a dwindling number of inhabitants in the inner city in the second quarter of the eighteenth century, with a significant population growth in the surrounding area (Ommelanden).⁴⁵ The fastest-growing ethnic group around Batavia after 1750 were people with a Chinese background, with over 10,000 new inhabitants by 1759.⁴⁶ These demographic changes must have influenced socio-economic realities in suburbs across Dutch Asia. In the 1760s, the real estate market in the suburbs of Colombo (Four Gravets) in Sri Lanka underwent significant changes; when land became scarce, the number of transactions grew and prices increased.⁴⁷ After the lifting of a century-old Dutch ban on their land transactions in 1746, Muslims became the largest group of landowners in the Four Gravets in barely a generation.⁴⁸ Questions about how the significant increase in foreign migration, a (supposed) building boom and increased pedestrian traffic influenced daily life under Dutch colonial rule in these suburban areas are still shrouded in mist.

Indeed, while this population growth has been established in previous scholarship, the city-centre-focused model of urban colonial histories kept most of the colonial gaze limited to the stagnating urban life within the city walls.⁴⁹ Remarkable eighteenth-century transformations in Dutch colonial rule across the board have been studied from the inside out, from the colonial power hub stretching out over the surrounding hinterlands. The explanation given by this research was the need for local governments to increase their grip on – again – the ‘hinterland’ to raise taxes on land and agricultural produce in order to supplement a significant decline in the income from maritime trade. While acknowledging the expanding geographical and social reach of colonial institutions such as courts of law,⁵⁰ registration and land taxation offices⁵¹ or the Reformed Church,⁵² this was unintentionally framed as European institutions in need of expanding jurisdictions that reached further and further from the (port) city into the hinterland of Jaffna or Galle, the Bay of Bengal or the Ommelanden of Batavia.

But at the same time, and well beyond the control of European institutions, inter-Asian migration and internal migration from the rural interior to the cities ensured a growing ethnic, cultural and religious diversity in the suburban areas. They became highly diverse zones, both socio-economically and culturally. Moreover, recent scholarship in slavery studies indicates that manumitted individuals who legally left behind their Dutch or European bondage settled in these zones as well, with anecdotical evidence from Cape Town and Zanzibar to Cochin, Colombo and Galle.⁵³ Given that the ethnic background of (former) slaves went well beyond East Africa, with forced migrants from all over Asia, this heterogeneous group further contributed to the increasingly rich socio-economic and cultural landscape of the suburbs.⁵⁴ If these strikingly parallel demographic and socio-economic changes across suburban zones of Dutch VOC cities in the eighteenth century led to a gradual dissolving of immigrant identities into a hybrid society, as Raben has argued, is still very much up for debate.⁵⁵

Conclusion: suburbs as socio-spatial colonial categories between global and local

For a VOC historiography currently concerned with further deconstructing analytical binaries and rethinking socio-political categorizations, the revisionist urban research with its increased focus on suburbanization could offer valuable conceptual and methodological tools. Too often, suburbs in early modern colonial histories are literally and metaphorically mentioned in passing as mere transitional zones between the dichotomies urban/culture and rural/nature. But seldom are they studied in their own right as socio-spatial entities. When global urban historians of VOC territories – and hopefully beyond – finally discard the twentieth-century inside-out theorization and redefine suburban zones as open-ended colonial spaces, that new historiography from an outside-in perspective will open up further research avenues for reconsidering political, economic and social interactions in colonial settlements.

This approach will also challenge more complex but equally arbitrary divisions between the global and local, European and indigenous, and even modern and traditional. Suburbs as socio-spatial entities possess the unique capacity to simultaneously embody all of these characteristics. As such, they could offer an alternative spatial approach to compare early modern cities around the Indian Ocean and further into colonial Asia with both ‘European’ and other imperial cores. It would move research beyond the ‘tyranny of the particular’, and find ways to integrate separate localized urban histories into a more holistic narrative.⁵⁶ If such a horizontal conceptualization was adhered to, we could perhaps pick up the gauntlet recently thrown down by Biedermann in his search for a ‘more fruitful approach’ in ‘connecting, comparing and disconnecting cities in creative new ways’.⁵⁷ Similar to Subrahmanyam’s idea of connected histories, adhering to a horizontal approach of colonial disjunct urban fragments – which by their nature defy Euro-centric and proto-nationalistic approaches – could even be a path beyond the ‘national’ as a category of analysis. This would shift (urban) colonial histories away from the dominant national European histories they are still subjected to.⁵⁸ After all, reversing the colonizer’s gaze could and should be more than a discursive reading against or beyond the grain. If we rethink the colonial city from the outside in and attempt to shift analysis from the core to the periphery, we could offer new avenues for exploring the underexposed intersections of ‘the imperial’ and ‘the urban’.⁵⁹

Such a historiography would help us unpack the black box that colonial suburbs still exist in. The goal of this intervention is to show that they existed empirically and that they matter conceptually for how we think about early modern colonial histories. Unfortunately, the limited nature of literature on suburbanization in early modern Europe does not yet offer models that could be transferred to the Global South in the seventeenth and eighteenth centuries, or to reflect on (dis)continuities between the European and colonial worlds.⁶⁰ But perhaps that is a blessing in disguise that allows global historians to truly work from a tabula rasa, to set their own conceptual parameters, differentiate between suburban forms and reflect on different typologies. Rather than following in the footsteps of European urban histories, global urban historiographies could herald an overturn of the inside-out model.

If we move beyond the spatial gaze perpetuated by imperial archives and city-centre-focused theorization, we could write even more nuanced and balanced narratives, and look differently at, for example, the presumed expansion and bureaucratization of colonial rule across VOC territories. Rather than looking at these phenomena as part of colonial encroachments of the urban centres on the rural hinterlands and a parallel shift towards land-based income to counter dwindling maritime trade, all these separate histories could be reconsidered as a much broader surge of colonial civic self-awareness. They could be seen as signs of a growing suburban willingness to interact with colonial authorities, to become an integral part of the colonial urban world and beyond and to showcase their individual and collective need for socio-political participation with the colonial powers. Dwellers from all walks of life, with geographical origins throughout West Africa and Asia, and living in expanding colonial suburbs perhaps encroached upon urban centres rather than the reverse. When rewriting colonial histories outside in, we would discover that everyday life in eighteenth-century suburban areas was anything but sterile, and that lives there were lived to the fullest, and not wasted in suburban nothingness.