Introduction
It is becoming increasingly commonplace for our world to be imagined first-and-foremost as a world of cities. In an article published in the first issue of the Annals of the Association of American Geographers in the third millennium, Jon Beaverstock and colleagues from the University of Loughborough's Globalization and World Cities (GaWC) research centre described cities and urban regions as the key spatial unit in an emergent global ‘metageography’. This metageography is most clearly visible when looking down at the earth at night: urban regions appear as pin-pricks of light while national boundaries are not visible at all (Beaverstock et al. Reference Beaverstock, Smith and Taylor2000; see also Bunnell and Maringanti Reference Bunnell2010). Urban geographers and scholars of cognate disciplines have had much to say about the implications of a world of cities for national scale framings that have long predominated across the social sciences (Agnew Reference Agnew1994; Brenner Reference Brenner1999; Taylor Reference Taylor1995). Much less attention has been given to implications for regional, areal or continental framings.
Area studies scholars, in contrast, have examined the implications of globalization – and its cross-border flows and connections – for regional partitioning of knowledge (Bentley et al. Reference Bentley, Bridenthal and Yang2005; Kratoska et al. Reference Kratoska, Raben, Nordholt, Kratoska, Raben and Nordholt2005), but have tended to overlook the specifically urban dynamics of alternative geographies. One notable exception here is work by the economic historian of Southeast Asia, Howard Dick. In his contribution to the edited volume Locating Southeast Asia: Geographies of Knowledge and Politics of Space, Dick examines how Southeast Asia might be imagined from a trans-national urban perspective:
“Focused on main cities and their hinterlands, trans-national interactions in the movement of people, goods, money and information define a core region or corridor, which contrasts with dispersed trans-national peripheries in both maritime and mainland Southeast Asia. This approach offers a stimulating and realistic way to re-imagine Southeast Asia without national boundaries in the foreground.” (Dick Reference Dick, Kratoska, Raben and Nordholt2005: 251)
Importance here is given not merely to identifying the region's most economically significant cities – those which shine most brightly on maps of the earth by night – but also to conceptualizing connections between them as forming “contiguous economic space” (Dick Reference Dick, Kratoska, Raben and Nordholt2005: 265). Dick points out that advances in transport and communications technologies mean that in terms of time scales, many cities are ‘closer’ to cities in other national contexts than they are to their own surrounding hinterlands (see also Dick and Rimmer Reference Dick and Rimmer2003; Rimmer and Dick Reference Rimmer and Dick2009). In addition – and as geographers such as Beaverstock and his colleagues also recognize – it is flows, networks and connections between urban centres rather than cities themselves that constitute the new metageography, at least in economic terms.
Geographies of urban networks extending across national borders help to challenge the ingrained methodological nationalism of conventional area studies, including in Southeast Asia. However, to the extent that border-crossing urban corridors continue to be framed, a priori, in regional terms, there is a danger that the ‘territorial trap’ is simply being scaled up from nation-state to region. Although he is clearly aware of this danger, Howard Dick's work provides some evidence of regional-scale methodological territorialism (or methodological regionalism). On the one hand, Dick shows how some of the most significant inter-urban linkages of cities in Southeast Asia extend beyond the conventional boundaries of that region. Defined by Dick as an ‘open system’, Southeast Asia may be conceptualized to include neighbouring urban centres like Hong Kong such that the region is understood to form part of much more geographically extensive urban networks: “Flows of people, goods, money and information reveal that modern Southeast Asia is a network of cities, a subset of broader networks of cities that may be labelled as East Asian, Asian or Asia-Pacific” (Dick Reference Dick, Kratoska, Raben and Nordholt2005: 272). On the other hand, Dick seeks to redefine ‘Southeast Asia’ in terms of urban networks rather than using these to problematize the whole notion of the region. And while reconceptualization of Southeast Asia as an ‘open system’ appears to allow him to avoid any such regional territorial trap, Dick only opens out the network in terms of larger regional framings (East Asian, Asian, or Asia-Pacific).
Such problems of methodological territorialism, to my mind, also inflect the more recent rise in attention that has been given to Inter-Asia connections and linkages (see Duara Reference Duara2010). I do not doubt the theoretical importance of this work – especially in its more obviously postcolonial forms (Roy and Ong Reference Roy and Ong2011) – in contesting ingrained tendencies to look out or ‘up’ to North America or Western Europe for both aspirational and conceptual models. However, there is also a danger that in the rush to identify and examine linkages and constitutive connections between cities ‘within’ the region – whether that be Southeast Asia, South Asia, East Asia, or simply ‘Asia’ – equally and perhaps even more important urban connections to other parts of the world are obscured. What is more, mapping urban networks as contained within regional formations means that network geographies are not being used to unsettle regional area studies framings in the way they have been applied critically to issues of methodological nationalism. The remainder of this paper consists of two main sections each outlining different perspectives on the continued salience of regional territorial ways of seeing. In the first, I push Howard Dick's conceptualization of Southeast Asia as an open system to its logical conclusion – namely, that cities form part of transnational networks which are global rather than merely regional in scope. In the second section, I turn attention to work on forms of regionalization that have emerged during an era of globalization. In particular, I consider global triad regions, cross-border urban regions, and regional geographies of urban ‘inter-referencing’ (Roy and Ong Reference Roy and Ong2011).
Cities and Systems: Dangers of a Re-Scaled ‘Territorial Trap’
The literature on world and global cities represents an important challenge to nation-state-centred thinking and to bounded territorial presuppositions in general. On the one hand, it was increased trans-national linkages in a late twentieth century era of globalization that enabled urban and regional scholars to see beyond what John Agnew (Reference Agnew1994) termed the ‘territorial trap’. On the other hand, the burgeoning literature on global city networks – and perhaps especially efforts to represent them visually – in itself contributes to reimaging the world in terms of urban-based metageographies. The examination and representation of interconnected urban regions has expanded way beyond the three ‘global cities’ originally examined by Saskia Sassen in her seminal work (Sassen Reference Sassen1991). In addition to the work of the Globalization and World Cities (GaWC)Footnote 1 group, Sassen herself has been among those to bring cities of “the global south and in the mid-range of the global hierarchy” (Sassen 2002: 3) into the literature. It is clear even from this short quotation that the notion of urban hierarchy is strong in this research. While Sassen's ‘mid-range’ was determined in relation to presumably higher and lower range cities, GaWC scholars began by identifying ‘alpha’, ‘beta’ and ‘gamma’ world cities (Beaverstock et al. Reference Beaverstock, Smith and Taylor1999). GaWC tabulates and maps cities according to such hierarchical classifications, rather than on the basis of regional-based typologies. The main point that I wish to elaborate upon in this section is that the challenge that such geographies pose to state-centred framings apply, for the most part, to other more-or-less bounded topological framings, including area studies regions.
When the word ‘region’ does appear in the literature on global or world cities, and indeed in urban studies more generally, it usually connotes sub-national urban territories rather than the kinds of supra-national groupings that constitute area studies regions. Neil Brenner (Reference Brenner1998) was among the first scholars to examine ways in which state power is re-scaled to major urban regions and some of my own earlier work considered such processes in Southeast Asia, particularly focusing on Kuala Lumpur (Bunnell Reference Bunnell2002, Reference Bunnell2004). States increasingly give greater attention and resources to key city-regions as the “motors of the global economy” (Scott et al. Reference Scott, Agnew, Soja, Storper and Scott2001: 15). The geography mapped by these kinds of linkages – global networks of city-regions – are reminiscent of the world before the establishment of state-based political ordering which is conventionally traced back to the Treaty of Westphalia in Europe in the seventeenth century (see e.g. Brenner Reference Brenner1999).
A trans-continental world system based on trade connections between cities dating back to the thirteenth century has been identified in the work of historians such as Fernand Braudel (Reference Braudel1992). Janet Abu-Lughod (Reference Abu-Lughod1989) depicted a trans-continental ‘archipelago’ of cities constituted by a series of eight overlapping regional city networks. The term archipelago is used rather than ‘network’ because “there was no direct trade between cities from different ends of the archipelago; rather, middle cities operated as exchange centres, as great entrepôts in this world of merchants” (Taylor Reference Taylor2004: 9). Over the next two centuries, trans-continental connections broke down such that city-based trade connections became more squarely contained within largely ‘economically autonomous’ regions which Braudel (Reference Braudel1992) refers to as ‘world-economies’.
The subsequent economic rise of the world-economy of Europe, initially centred upon the Mediterranean, culminated in the establishment of the modern world-system in the sixteenth century (Wallerstein Reference Wallerstein1979). Two features of this European-dominated world-system are worthy of note. First, trans-continental linkages and connections were extended under European hegemony with the result that Braudel's world-economies became much less autonomous, economically and politically. Long distance connections meant that there were extended commercial networks rather than the archipelago of overlapping regions described by Abu-Lughod for the earlier era. The region, in other words, lost salience. The second feature was the establishment of the system of nation-states which resulted in the world-system becoming subject to inter-state rivalry. Economic domination by the Dutch Republic gave way to British hegemony which in turn gave way to U.S. hegemony. Although each of these successive hegemonic states was home to, and commercially centred upon, world cities – Amsterdam, London, and New York respectively – it was the national rather than city-state scale that formed the basis for territorial organization and which came to be the default unit of analysis in social science research. As Peter Taylor has put it: “By the mid-twentieth century, instead of being viewed as nodes within great worldwide networks, major cities in the world were being seen in territorial terms: either as the capital city within a state, or as a ‘regional centre’ within a state” (Taylor Reference Taylor2004: 15).Footnote 2
This tendency to see cities as operating exclusively within states had profound implications for urban studies. Research in the 1960s and 1970s was dominated by analysis of national urban systems, not just in the West but across the world, including in Southeast Asia (for an example in the context of what became the nation-state of Malaysia, see Lim Reference Lim1978). The taken for granted position of national systemic frameworks was such that few scholars questioned the extent to which national sets of cities do actually form ‘national systems’:
“Within systems thinking, the nearest researchers get to doubting the existence of national urban systems is when the openness of the system is discussed. If one eliminates the possibility of these being closed systems, the question of the degree of open-ness would seem to be of vital importance.” (Taylor Reference Taylor2004: 19)
In hindsight, the fact that the question of openness did not lead to any serious questioning of the validity of national systems thinking seems curious. Taylor considers the case of New York, the city placed number one in the US according to the rank size rule used in systems analyses:
“…might not its prime position be due to its traditional role as gateway between the United States and the rest of the world? If this is so, it is not enough merely to recognize the open-ness of the US national urban system; its leading city is dependent on connections beyond the ‘system’ and therefore problematizes the system's veracity. New York was and is part of a city network that transcends the national territory in which it is located.” (Taylor Reference Taylor2004: 20)
The same may be said of all main cities, in Southeast Asia as much as in North America: “their inter-relations are never respecters of boundaries” (Taylor Reference Taylor2004: 20).
If this was the case for New York in the 1960s and 1970s, what more for this and other major cities in an era of economic globalization? Economic restructuring from the 1970s was characterized by a ‘global shift’ in the activities of multinational corporations (Dicken Reference Dicken1998). The new international division of labour (Frobel et al. Reference Frobel, Heinrichs and Kreye1979) through which firms relocated parts of their production in order to take advantage of low wages in Third World countries meant that MNC networks transcended individual states and regions. This was followed in the 1980s and 1990s by widespread deregulation of financial services which extended possibilities for footloose financial firms to establish direct operations in dispersed urban centres rather than serving them from a distance. The countervailing tendencies of spatial dispersal and urban agglomeration of economic activities formed the foundation of Sassen's identification of a new kind of city: the global city. And while this work was focused on a relatively narrow set of financially-related economic activities – so-called advanced producer services – the global cities literature formed part of a wider framing of social and political analysis in ‘global’ terms. In part, it was increased connections and linkages that compelled scholars to look beyond national framings. However, this is of course not to say that trans-national and/or trans-regional linkages are necessarily new to an era of globalization. As the case of New York above demonstrates, certainly such ‘external’ connections existed earlier, albeit in diluted form. Nevertheless, it took the rapid increase in connections associated with late twentieth century processes of economic globalization to unsettle the nationalization of social thought (see also Taylor Reference Taylor1995). It might be suggested that global networks have themselves now become taken for granted spatial imaginings of researchFootnote 3 such that the arrangement of groups of cities contained within nation-states (or other territorializations) is increasingly difficult to imagine. In recent years, geographers have been among scholars demonstrating global dimensions of even small cities that do not feature in the literature on world or global cities, or rank in the GaWC list (e.g. McCann Reference McCann2004).
What are the implications of these developments for area studies supra-national regions such as Southeast Asia? Here, I return to Howard Dick's conception of urban corridors in Southeast Asia as a regional system. While Dick recognizes that the system is of course ‘open’, how open does a regional system have to be before the whole notion of a regional framing becomes untenable? The main example of openness considered by Dick (Reference Dick, Kratoska, Raben and Nordholt2005) is extension of the system to Hong Kong, a city located beyond the conventional boundaries of Southeast Asia. This forms part of a wider understanding that cities in Southeast Asia comprise a subset of broader networks of cities that may be labelled variously as East Asian, Asian, or Asia-Pacific. However, this begs the question as to whether linkages between, for example Kuala Lumpur and London or Jakarta and Amsterdam are as significant as those between either Kuala Lumpur or Jakarta and Hong Kong. Also, even within the conventional cartographic limits of Southeast Asia, a similar question could be raised about the significance of links between Kuala Lumpur or Jakarta and Manila compared to linkages with former colonial metropoles in Europe.
These questions are raised here in the absence of empirical evidence of the actual links between cities ‘within’ and ‘beyond’ the region. However, in light of the significant challenge posed by conceptions of global networks to previously taken-for-granted national systems and, by extension, bounded territorial presuppositions in general, the burden of proof lies with those who assert the existence of regional systems. Failure to do so may leave scholars open to the charge of methodological regionalism – escaping from national territorial traps only to become ensnared in a regional one. At the same time, it is important not to over-extend network framings so as to imply that all territorial frameworks simply melt away. Scholars such as Neil Brenner (whose work highlights the limits of national territorial framings) are not positing complete deterritorialization in the space of flows (Brenner Reference Brenner1999). Rather, the rise or return to prominence of cities and city-regions is understood to take the form of rescaled territorializations of political economic power.Footnote 4 There are also ways in which reterritorialization is occurring at the level of supra-national regions such as ASEAN, although, as I seek to show in the next section, these take very different forms from conventional area studies regions such as Southeast Asia.
Regions in an Era of Globalization
Focusing on city and city-regional territorializations and associated cross-border linkages, I am not directly concerned with forms of regionalization that take nation-states as their building blocks. Clearly groupings such as ASEAN play an important role in fomenting collective senses of regional belonging (Thompson Reference Thompson2006; Thompson and Thainthai Reference Thompson2008) and often put in place policies that enable inter-city connections. However, it is important to emphasize that linkages between cities are not confined to, nor are necessarily at their strongest, within regions. Extended network spatialities problematize regional framings such as ‘Southeast Asia’. Nonetheless, there are also ways in which alternative regional form(ul)ations have risen to prominence in an era of globalization. In what follows, I identify and elaborate three such examples: 1) Asia as a global triad region; 2) Cross-border meso-scale regions; and 3) Inter-Asia urban referencing.
Asia as a Global Triad Region
The term ‘global triad’ was coined by Japanese management writer Kenichi Ohmae (Reference Ohmae1985) to refer to the three regions of the world that are home to the global economy's major markets, ‘competitive threat’ and new technologies. For Ohmae, writing in the 1980s, these three ‘regions’ were the United States, Europe, and Japan. The economic rise of other nation-states in Asia besides Japan, plus the development of supra-national political and trade groupings, meant that subsequent writers drawing upon Ohmae's triadic formation defined the regions differently (Peter Dicken in Global Shift, for example, refers to the European Union, North America (NAFTA), and South/Southeast Asia). In 1994, those three ‘macroregions’ together accounted for 87 per cent of total world manufacturing and 80 per cent of world merchandise exports; and those proportions appeared to be continuing to increase (Dicken Reference Dicken1998). According to Dicken, “These trends imply that the global triad is, in effect, ‘sucking in’ more and more of the world's productive activity, trade and direct investment. The triad sits astride the global economy like a modern three-legged Colussus” (Dicken Reference Dicken1998: 61). Such tripolar global economic imaginings have been extremely influential but, as Dicken notes, whether the regions amount to anything more than a statistical artefact – whether the triad represents “a functional reality (actual or potential) with internally oriented production and trade systems” (Dicken Reference Dicken1998: 62) – remains very much open to debate. Irrespective of the functional validity of a global triad region in Asia, its very discursive existence has had powerful effects.
Ohmae's ideas in the 1980s and work that built upon it considered nation-states as the fundamental building blocks of each of the global triad regions (albeit with highly variegated degrees of political organization into formal trading blocks). Nonetheless, a related global economic imaginary informs early GaWC work that formed part of attempts to decentre the nation-state from social science research. Beaverstock et al. (1999) note that the 55 world cities on their roster are overwhelmingly geographically concentrated in northern America, western Europe, and Pacific Asia. These concentrations are termed ‘major globalization arenas’ (Beaverstock et al. Reference Beaverstock, Smith and Taylor1999: 457). In later work too, Peter Taylor asserts that “globalization is a very regional process” (Taylor Reference Taylor2004: 175). However, in keeping with some of the concerns that Peter Dicken has about the functional reality of triad regions, Taylor's assertion appears to be based on observation of shared characteristics or emphases of cities within particular regions – he notes, in particular, the ‘bias’ of Pacific Asian city globalization to the banking and finance sector – rather than on any inter-city relations or linkages. The key point here is that Southeast Asia, when it is mentioned at all in work on global triads regions, appears as part of a larger collection of nation-states (Ohmae Reference Ohmae1985) or world cities (Beaverstock et al. Reference Beaverstock, Smith and Taylor1999). In either case, Southeast Asia is conceived of as merely a cartographic description of an area which is understood to form part of a larger regional economic formation.
Cross-Border Meso-Scale Regions
As scholarship on globalization burgeoned in the 1990s, there was growing interest in regions that were not comprised of multiple national territorial units. This is reflected in shifts in Kenichi Ohmae's thinking, particularly in his book, The End of the Nation State, which was published in 1995. In that book, Ohmae argues that “traditional nation states have become unnatural, even impossible, business units in a global economy” (Ohmae Reference Ohmae1995: 5), and casts groupings such as ASEAN, NAFTA, and the EU as “creatures of a nation-state defined and -funded universe” (Ohmae Reference Ohmae1995: 2). According to Ohmae, it is ‘region states’ that form ‘natural business units’ in a globalized world:
“They may lie entirely within or across the borders of a nation state. This does not matter. It is the irrelevant result of historical accident. What defines them is not the location of their political borders but the fact that they are the right size and scale to be the true, natural business units in today's global economy. Theirs are the borders – and the connections – that matter in a borderless world.” (Ohmae Reference Ohmae1995: 5)
In academic human geography and related disciplines, Ohmae's notion of the end of the nation-state has been widely cited as exemplifying work which overstates the diminution of state power in what remains a far-from-borderless world (e.g. Yeung Reference Yeung1998). However, it is worth pointing out that the subtitle of Ohmae's book is ‘The Rise of Regional Economies’ – something that is referred to far less frequently than his much-critiqued notion of a ‘borderless world’.
The region states or regional economies that Ohmae identifies include examples from North America (e.g. San Diego/Tijuana and the Silicon Valley/Bay Area in California), Europe (e.g. northern Italy and Baden-Wurtenberg) as well as in Asia. Three ‘natural economic zones’ for a borderless world are identified in Southeast Asia: the Growth Triangle comprising Singapore, Johor (Malaysia), and Riau islands (Indonesia); the Malaysian island of Penang; and what is described as “the newly emerging Greater Growth Triangle, unveiled in 1992 across the Strait of Malacca, connecting Penang, Medan (an Indonesian city in Sumatra), and Phuket in Thailand” (Ohmae Reference Ohmae1995: 80). During the time since the publication of Ohmae's book, the ability of these three areas to function globally as economic regions has remained very much circumscribed by the continuing power of nation-states. In the case of Penang, economic development has in part been limited through being an ethnic-Chinese dominated state in an ethnic-Malay dominated federation (Teo Reference Teo2003) and associated federal government privileging of greater Kuala Lumpur as the locus of Malaysia's globally-oriented economic ambition (Bunnell Reference Bunnell2002, Reference Bunnell2004). The Indonesia-Malaysia-Thailand growth triangle, meanwhile has remained little more than an idea, not least because of conflict in the northern Sumatra province of Aceh (Miller Reference Miller2009), although the secession of conflict in Aceh and a wider decentralization of power and resources in Indonesia after the fall of President Suharto mean that international donors such as the German Organisation for Technical Cooperation (GTZ) continue to talk of potential cross-border complementarities, especially in the tourism sector (Phelps et al. Reference Phelps, Bunnell and Miller2011).Footnote 5
The Indonesia-Malaysia-Singapore growth triangle (IMS-GT), in contrast, does show evidence of extended cross-border economic activity, particularly in the form of projects such as Batamindo Industrial Park, Bintan Industrial Estate, Bintan Beach International Resort, and the Karimun marine and industrial complex on the Riau islands of Batam, Bintan and Karimun (see Grundy-Warr et al. Reference Grundy-Warr, Peachey and Perry1999). Critical scholarship on these developments has contested official discourses of complementarity and cooperation, with evidence of competition and even conflict. In contrast to Ohmae's conception of a looming borderless world, the triangle is transected by all kinds of divides and disjunctures that may be understood as new forms of boundary drawing, within as well as between the component national territories (Ford and Lyons Reference Ford and Lyons2006; Sparke et al. Reference Sparke, Sidaway, Bunnell and Grundy-Warr2004). The cross-border IMS-GT region was previously called SIJORI (Singapore, Johor, Riau), and it is worth noting that the new naming is composed of national units, rather than sub-national regions, while the alphabetical ordering of IMS was intended to decentre Singapore. Despite this symbolic move, to the extent that this region has functioned economically across borders, there is no doubt that it has been driven by capital from global city Singapore and its demands for labour and space (Sparke et al. Reference Sparke, Sidaway, Bunnell and Grundy-Warr2004). There has been very little in the way of cross-border economic activity between the Malaysian and Indonesian parts of the triangle, which for the most part function as extended hinterlands for tourist and manufacturing activities from Singapore (Bunnell et al. Reference Bunnell, Muzaini and Sidaway2006; Hampton Reference Hampton2010; Rimmer and Dick Reference Rimmer and Dick2009). Nonetheless, despite these limitations and global city specificities, the IMS-GT continues to be hyped and even viewed as a model, most recently for the Greater Mekong Sub-Region in a report commissioned by the Asian Development Bank (Weimer 2009).Footnote 6
Inter-Referencing Regionalization
The notion of IMS-GT as a model leads neatly onto my third category of region formation. This is one which I sketch tentatively, drawing upon fast-proliferating work on inter-city referencing, policy mobilities, and urban modelling practices. ‘Inter-referencing’ is taken from a recently-published book edited by Ananya Roy and Aihwa Ong (Reference Ong, Roy and Ong2011), who take the term to mean ways in which urban development in Asian cities increasingly references other cities in Asia. In some of my own recent work (Bunnell and Das Reference Bunnell and Das2010), I have examined the way in which Kuala Lumpur inspired and was adopted as a blueprint for high-tech urban development in Hyderabad, India.Footnote 7 Similar examples abound, both in the Roy and Ong volume (which includes a chapter on ‘Singapore a model’ – Chua Reference Chua, Roy and Ong2011) and in papers presented at the workshop on Global Urban Frontiers: Asian Cities in Theory, Practice and Imagination that was held in Singapore in September 2010.Footnote 8 According to Ong, “As cities in Asia begin to look to each other for achievable models and norms of city building and living, a distinctive system of images shapes an ecology of urban significance” (Ong Reference Ong, Roy and Ong2011: 20). At one level, to think of this in terms of regionalization evokes Howard Dick's notion of systems of interconnected cities (and associated critiques) noted above. Certainly, cities within Asia continue to look and refer to cities beyond the region, and especially in North America and Western Europe. However, there is surely some significance to the observation that “Beijing, Shanghai, Hong Kong and Singapore instead of New York, London, or Paris have become centres to be envied and emulated” (Ong Reference Ong, Roy and Ong2011: 19).
There is an important distinction to be made between: (a) locating networks of urban emulation and inter-referencing within a priori regional frameworks (such as Southeast Asia); and (b) allowing actually existing linkages and relations to form the basis for new regional mappings. While my coverage of the world and global cities literature so far has tended to focus on the contribution that this has made to unsettling methodological nationalism (and regionalism), another contribution has to do with promoting understandings of cities in relational terms. The GaWC group, in particular, have collected vast data sets on urban connectivity which have given rise to powerful new ways of seeing and understanding the world. Among the problems with such literature to date is the fact that it has focused overwhelmingly on economic inter-linkages and, indeed, for the most part, on a financial services sub-sector of relational economic geographies.Footnote 9 Without denying the importance of financial services linkages between cities, it is clear that these are only one among a range of activities and practices that may form the basis for meaningful configurations of space. Inter-referencing is a potentially region-making set of practices that may range from the visions and idealizations of policy elites (as in the growing literature on policy transfer and mobilities; see McCann Reference McCann2011; as well as Roy and Ong Reference Roy and Ong2011) to the dreams and aspirations of individual city dwellers and migrants (Bunnell and Goh Reference Bunnell and Goh2012). We might also consider the activities of civil society organizations that develop network solidarities across national borders, such as the case of squatters resisting eviction from the Surabaya riverbank, who drew upon the experiences of neighbourhood activism in Thailand and India through the Asian Coalition for Housing Rights (see Idawati Reference Idawati2009).Footnote 10
Finally, and more closely related to the economic sphere which has formed the basis for existing work on urban relationality, we might consider the region-making possibilities of real estate investment (Percival and Waley Reference Percival and Waley2010), tourist practices (Winter Reference Winter2007), and the models of mobile architects (McNeill Reference McNeill2009). Scholars who are considering the methodological implications of such relational studies note the need for forms of mobile ethnography (McCann Reference McCann2011; following Burowoy 2000). At one level, the geographies concerned run across area studies partitions – looking, for example, at how urban centres in Southeast Asia connect with South Asia or West Asia, or even Europe – in a way that some scholars see as a threat to area studies (Kratoska et al. Reference Kratoska, Raben, Nordholt, Kratoska, Raben and Nordholt2005). However, to the extent that relational framings of the region studied through mobile ethnographies continue to necessitate learning local languages and other cultural skills – perhaps in more than one areal region – such training might best be carried out in/through existing area studies programmes.Footnote 11
Conclusion
I have sought in this paper to bring urban studies and area studies into critical conversation with each other. Work in Anglophone urban and regional studies over the past decade or so has been influenced by and in turn shaped imaginings of the world in terms of networks of interconnected cities. Work in Southeast Asian studies, meanwhile, has taken on board the challenge that these kinds of imaginings pose to methodological nationalism but largely continues to work within existing regional areal framings. As such, although the main target of work on cross-border urban linkages has been national scale framings which have dominated the social sciences, network spatialities challenge territorial framings at other scales too, including that of area studies regions. The key question that emerges from this is: what kinds of regional geographical formations make sense in a world of mostly urban-based interconnections? I have considered three broad regional formations that have risen to prominence in globalization-related research. The first two of these – Asia as a ‘global triad’ region and cross-border meso-scale regions – are well established in the published literature. It is the third regional formation or, rather, form of region-making which is perhaps the most novel (and thereby difficult to pin down), and this is the one which I have termed, following Roy and Ong (Reference Roy and Ong2011), ‘inter-referencing’ regionalization. Three points arise from this form of region-making: (1) it should not take Southeast Asia, Asia or any other territorialized region as the starting point for research but should conceptualize regions based on actual connections and linkages of various kinds; (2) to the extent that such connections are contained within (Southeast) Asia, this does not mean that they should necessarily be labelled as such (some sub-areal territorial unit may be more appropriate, or even non-territorially based ‘regional’ spatialities); and (3) actually existing connections and linkages extending across areal regions also demand alternative, non-contiguous regional framings.
In a recent article in the Journal of Asian Studies, Prasenjit Duara (Reference Duara2010) considered what it means to conceptualize the region for our times. In what is a very wide-ranging article, Duara points out forces for regional integration that include but also extend beyond examples that I have covered in this paper: growing financial interdependence especially after the late 1990s Asian financial crisis; the development of regional supply chain production networks; extended and shared ecological and environmental challenges; growing cultural interest in Asia (including Asian art) among Asians; migration and sojourning with associated transfers of cultural values, styles, and remittance monies; and knowledge-based service sector circulations. As a historian, Duara pays attention to continuities and disconnections between these and earlier Asian regionalisms. In doing so, he notes in particular that “Interdependence…is being managed by ad hoc arrangements and specialized transnational institutions with little possibilities of large-scale state-like coordination and control' (Duara Reference Duara2010: 24). As such, Asian regionalism is ‘multipath’ and ‘pluralistic’, and, unlike Europe, does not seek to homogenize itself from within. Rather, according to Duara, the non-congruence between state and culture is reminiscent of earlier Asian maritime networks. However, as has been highlighted by work on global and world cities – which as we have seen also makes connections back to eras before the state-dominated world system – today's Asian cities are much more-than-regionally interconnected.Footnote 12 To speak of Asia Redux – of bringing Asia back – is to imply the salience of existing or formerly-existing frameworks, while I have sought to show that forms of ‘regional’ territorialisation and topologies need to be charted anew following the very diverse forms of interdependence that Duara has highlighted.
Acknowledgements
This paper was first presented at the conference Beyond Boundaries: Southeast Asian History, Culture and Society which was held at Sogang University in October 2010. I am grateful to Professor Shin Yoon Hwan, Lee Sang Kook, Kim Jee Hun and other members of The Institute for East Asian Studies at Sogang University for their kind invitation and hospitality. The paper is not associated with any specific research project or grant. Rather, it emerges from conversations and collaborations of various kinds at the National University of Singapore over the past decade. I am grateful, in particular, for the ferment of ideas arising from interactions with Diganta Das, Daniel Goh, Carl Grundy-Warr, Anant Maringanti, Michelle Miller, James D. Sidaway and Eric Thompson. Michelle kindly provided comments on an earlier draft of the paper and I am grateful for feedback from participants of the Beyond Boundaries conference. Helpful comments from two anonymous referees, and Lindsay Lloyd-Smith's editorial attention to detail, are also gratefully acknowledged. I alone, of course, am responsible for any remaining errors or shortcomings.
