47(13) 2737–2745, November 2010

Globalisation, Networking, Urbanisation: Reflections on the Spatial Dynamics of the Information Age

Manuel Castells

[Paper first received, February 2008; in final form, September 2008]

Abstract The network society is a global society because networks have no boundaries. Spatial transformation is a fundamental dimension of this new social structure. The global process of urbanisation that we are experiencing in the early 21st century is characterised by the formation of a new spatial architecture in our planet, made up of global networks connecting major metropolitan regions and their areas of influence. Since the networking form of territorial arrangements also extends to the intrametropolitan structure, our understanding of contemporary urbanisation should start with the study of these networking dynamics in both the territories that are included in the networks and in the localities excluded from the dominant logic of global spatial integration.

The network society is a global society because networks and in the localities excluded from networks have no boundaries. Spatial trans- the dominant logic of global spatial integration. formation is a fundamental dimension of In this article, I will summarise the main this new social structure. The global process features and underlying causes of the spatial of urbanisation that we are experiencing in dynamics of the global network society on the early 21st century is characterised by the the basis of previous analyses and selected formation of a new spatial architecture in our evidence (Castells, 1989, 1999, 2000, 2004; planet, made up of global networks connecting Castells et al., 2006; Hall and Pain, 2006; Dear, major metropolitan regions and their areas 2005, 2006; Graham, 2005; Sassen, 2006; Lim, of influence. Since the networking form of 1998; Broudehoux, 2004; Kwok, 2005; Lu, territorial arrangements also extends to the 2006; Hackworth, 2005; Wolch et al., 2004; intrametropolitan structure, our understand- Halle, 2003; Graham and Simon, 2001; Abu- ing of contemporary urbanisation should start Lughod, 1999; Scott, 1998; Borja and Castells, with the study of these networking dynamics 1997), in line with the studies presented in in both the territories that are included in the this Special Issue of Urban Studies.

Manuel Castells is in the School of Communication, University of Southern California, Los Angeles, California 90089-0281, USA. E-mail: [email protected].

0042-0980 Print/1360-063X Online © 2010 Urban Studies Journal Limited DOI: 10.1177/0042098010377365 Downloaded from usj.sagepub.com at Univ of Newcastle upon Tyne on October 27, 2016 2738 MANUEL CASTELLS

First of all, a stream of research conducted urban areas, by mid-century it is likely that in the past two decades around the world, around three-quarters of the inhabitants of largely building on the old tradition of human the planet will be urban dwellers. Yet the most ecology and following the path of the pio- important characteristic of this accelerated neering work by Harold Innis, has shown the process of global urbanisation is that we are close interaction between the technological seeing the emergence of a new spatial form, transformation of society and the evolution which is given different names depending of its spatial forms (Scott, 2001; Sanyal, 2003; on diverse analytical perspectives. I call it Graham, 2005; Mattos et al., 2004; Hawley, the metropolitan region to indicate that it is 1950, 1956; Innis, 1950, 1951). We know that metropolitan but it is not a metropolitan area, technology is not the determinant factor of because usually there are several metropolitan this evolution. Nonetheless, microelectron- areas included in this spatial unit. Peter Hall ics-based information and communication and Kathy Pain (2006) call these areas the technologies have been shown to facilitate the polycentric megacity regions on the basis of digital networks that support the diffusion of their empirical study on recent metropoli- the new social structure, as the electrical grid tanisation in western Europe. The polycentric and the electrical engine supported the expan- metropolis, or the metropolitan region, arises sion of the industrial society (Mitchell, 1999; from two intertwined processes: extended Hughes, 1983). We also know that in the age of decentralisation from big cities to adjacent information and communication technolo- areas and interconnection of pre-existing gies, in sharp contrast with the predictions towns whose territories become integrated by of futurologists, we are not witnessing the new communication capabilities. This model end of cities or the annihilation of distance. of urbanisation is at the same time old and Instead, we are in the midst of the largest wave new. In Hall’s and Pain’s words of urbanisation in human history. There is an increasing concentration of population It is a new form, [including] anything between and activities in urban areas and in major ten and fifteen cities and towns, physically metropolitan areas. In 2008, we are expected separated but functionaly networked, to cross the 50 per cent threshold of urban clustered around one or more larger cities, population on the planet, which is 3.3 bil- spatially separate and drawing enormous lion people, according to the United Nations economical strength from a new functional Population Fund (2007), with over 1 billion division of labour. These places exist, both as separate entities in which most residents work living in squatter settlements, particularly in locally and most workers as local residents ... the metropolitan regions (Neuwirth, 2005). and as a functional region that is connected Projections from the same UNPF report by networks of transport and communication estimate that the number of urban residents processing flows of people, goods, services, in 2030 will reach 5 billion, of whom 81 per and information (Hall and Pain, 2006, p. 3). cent are expected to live in developing coun- tries, including one-third of slum dwellers. The transport and digital communication By 2030, the majority of the population on infrastructures, including wireless com- all continents, including Asia and Africa, munication systems, are the nervous system will live in urban areas. South America is of the polycentric metropolis (Rutherford, already 80 per cent urban and Europe and 2004). I would also add that in most cases, North America are approaching 80 per cent. with some exceptions (for example, Toronto Looking ahead, by simple extrapolation and Jakarta), there is no institutional unity in of current trends of population growth in these metropolitan regions, leading to political

Downloaded from usj.sagepub.com at Univ of Newcastle upon Tyne on October 27, 2016 GLOBALISATION, NETWORKING, URBANISATION 2739 unaccountability and chaotic planning for century the metropolitan regions are a universal these mega human settlements. urban form. In the US in 2005, the Urban The metropolitan region is not just a spa- Land Institute identified 10 megalopolitan tial form of unprecedented size in terms of areas housing 68 per cent of the American concentration of population and activities. population (cited by Hall and Pain, 2006). Yet, It is a new form because it includes in the the largest metropolitan areas in the world are same spatial unit urbanised areas and agri- in Asia. The largest one is a loosely connected cultural land, open space and highly dense region that extends from Hong Kong to residential areas: there are multiple cities in Guangzhou, incorporating the manufacturing a discontinuous countryside. It is a multi- villages of the Pearl River delta, the booming centred metropolis that does not correspond city of Shenzhen, on the Hong Kong border, to the traditional separation between central and the adjacent areas of Zhuhai and Macau, cities and their suburbs. There are nuclei of each one with a distinctive economy and different sizes and functional importance polity, fully interdependent with the other distributed along a vast expanse of territory components of this south China metropolitan following transport lines. Sometimes, as in region, with a population of approximately 50 the European metropolitan areas, but also million people. This pre-figures the megapoli- in California or New York/New Jersey, these tan future of China. During my conversations centres are pre-existing cities incorporated in Beijing in November 2005, planning offi- in the metropolitan region by fast railway cials of the State Council reported their plans and motorway transport networks, supple- to organise China’s metropolitan growth by mented with advanced telecommunication 2020 into 10 major metropolitan regions with networks and computer networks. Sometimes 50 million dwellers each. In fact, the south the central city is still the urban core, like in China region has already reached that size London, Paris and Barcelona. Often, how- and Greater Shanghai in 2007 was home to ever, there are not clearly dominant urban over 30 million people. These metropolitan centres. For instance, the largest city in the regions will constitute the heart of the new, San Francisco Bay Area is not San Francisco increasingly globalised China, the manufac- but San Jose. Yet, San Francisco remains the turing centre of the world in the 21st century. key location for advanced services, while the These ‘cities’ are no longer cities, not only main economics basis of the region (Silicon conceptually but institutionally or cultur- Valley) is neither in San Francisco nor San ally. In fact, they do not even have a name. In Jose, but in between. In other instances, like the place where I live now, Los Angeles, the in Atlanta and in Shanghai, the new centres only people who call it Los Angeles are either (North Atlanta, Pudong) are induced by the visitors or the minority of people inhabiting fast growth of the metropolitan region to the city of Los Angeles (about 3.5 million), host business, services and population that in contrast to the rest of the inhabitants of a gravitate towards the dynamism of these southern California metropolis of about 20 metropolitan magnets. In all cases, the metro- million that stretches from Santa Barbara to politan region is constituted by a multicentred San Diego and Tijuana across the border, in a structure (with different hierarchies between pattern of continuously urbanised landscape the centres), a decentralisation of activities, along the coast, and extends for about 100 residence and services with mixed land uses, miles inland (Wolch et al., 2004). Faced with and an undefined boundary of functionality this troubling namelessness, the southern that extends the territory of this nameless city California media have created a name for this to wherever its networks go. In this early 21st integrated television market, which is used

Downloaded from usj.sagepub.com at Univ of Newcastle upon Tyne on October 27, 2016 2740 MANUEL CASTELLS at the beginning of evening news broadcasts: of specialised functions. Why so? What are “Your local news. From the Southland”. The the reasons for the formation of these met- Southland (south of where?) is this undefined ropolitan regions? metropolitan region where 20 million people Peter Hall and Kathy Pain propose a major work, live, commute and communicate by hypothesis which is one of the keys to unlock- using a network of freeways, media coverage, ing this mystery, but not the only one. In cable networks and wireline and wireless tele- the knowledge economy, advanced services communication networks, while retrenching are the dynamo of urban growth, wealth in the polity of a fragmented territory’s locali- and power. Advanced services are globally ties and identifying their diverse cultures in organised. So the globalisation of advanced terms of ethnicity, age and self-defined social services is at the source of concentration in networks. Thus, the Southland lacks a defini- some areas of the world that are the pivotal tion of institutional, cultural or geographical nodes of the networked management capacity boundaries, but has a strong functional and in our society. These advanced services act as economic unity. a driver of urban centrality, because they are In Europe, Peter Hall and Kathy Pain (2006) concentrated in old or new centres of our have identified the dynamics of the polycen- major cities. These high-level service centres tric metropolis in the eight major regions of are located in places that are well connected Europe they studied. What they found is the in terms of transport and telecommunication persistance of urban centrality at the core of and possess a strong basis in terms of knowl- the region, in spite of the articulation between edge generation and professional labour. various urban centres. In other words: there This is clearly a major reason for the phe- is a hierarchical specialisation of functions nomenon of metropolitan concentration, between different urban centres. The overall but there are others. I will start with the spatial structure is polycentric and hierarchi- proposition that the key spatial feature of cal at the same time. Yet there is no sprawl. the network society is the networked con- In fact, the traditional residential suburban nection between the local and the global. sprawl observed by American urban studies The global architecture of global networks in the 1960s and 1970s is no longer the pre- connects places selectively, according to their dominant pattern in American metropolitan relative value for the network. The research areas. The residential settlement process has presented in this Special Issue, particularly extended to exurbia, while many suburbs have the studies by Peter Taylor and his collabora- become dense areas, sometimes dominated by tors, Denise Pumain, Céline Rozenblat and high-rise buildings, and economic activities others, demonstrates the importance of the have decentralised along transport lines, so global networking logic for the concentration that there is a mix of activities in the outly- of activities and population in the metropoli- ing areas, together with the diversification tan regions. This is not only to say that these of urban centrality functions. The notion of metropolitan regions are connected globally, residential suburban sprawl as a predominant but that the global networks, and the value urban form is outdated. Nowadays, we observe that they process, need to operate from nodes a distributed centrality and a multifunctional in the network. It is not the financial centres spatial decentralisation process. The key in London, Tokyo and New York that have features are the diffusion and networking of produced a global financial market made population and activities in the metropolitan of telecommunicated computer networks region, together with the growth of different and information systems. It is the global centres interconnected according to a hierarchy financial market that has restructured and

Downloaded from usj.sagepub.com at Univ of Newcastle upon Tyne on October 27, 2016 GLOBALISATION, NETWORKING, URBANISATION 2741 strengthened the places, old and new, from Johannesburg. Every country has its major where global capital flows are managed. They node(s) that connect(s) the country to stra- are not global cities, but global networks that tegic global networks. These nodes underlie structure and change specific areas of some the formation of metropolitan regions that cities through their connections. After all, determine the local/global spatial structure of much of New York (for example, Queens), each country through their internal, multilay- Tokyo (for example, Kunitachi) or London ered networking. Outside the landing places (be it Hampstead or Brixton) is very local, of networked value creation lay the spaces of except for their immigrant populations. The exclusion, or ‘landscapes of despair’ (Dear global functions of some areas of some cities and Wolch), either intrametropolitan or rural. are determined by their connection to the Why do these global networks linked global networks of value making, financial through nodes need to land in some specific transactions, managerial functions or other- metropolitan regions? Why is the processing wise. And from these nodal landing places, of their highly abstract operations unable through the operation of advanced services, to free itself from spatial constraints? Here, expands the economic and infrastructural we can use traditional models of explana- foundation of the metropolitan region. So the tion (Castells, 1989; Sassen, 1991). What changing dynamics of networks, and of each is important in the location of advanced specific network, explain the connection to services is the micro network of the high-level certain places, rather than the places explain- decision-making process, based on face-to- ing the evolution of the networks. The points face relationships, linked to a macro network of connection in this global architecture of of decision implementation, which is based networks are the points that attract wealth, on electronic communication networks. power, culture, innovation and people, inno- In other words, meeting face-to-face to do vative or not, to these places. financial deals or political deals is still indis- For these places to become nodes of the pensable, particularly where there is a need global networks, they need to rely on a mul- for absolute discretion in the case of deci- tidimensional infrastructure of connectivity: sions that provide a competitive edge. In the multimodal transport on air, land and sea; locational decisions of the managerial func- telecommunication networks; computer tions of business corporations, the intangible networks; advanced information systems; and factor is still about having access to the micro the whole infrastructure of ancillary services networks located in certain selective places, in (from accounting and security to hotels and what I named ‘milieus’ (Castells, 1989). They entertainment) required for the functioning can be financial milieus (for example, New of the node (Kiyoshi et al.ter, 2006). Every York, London, Tokyo; Sassen, 1991) but also one of these infrastructures needs to be technological, like in Silicon Valley (Saxenian, served by highly skilled personnel, whose 1994) or other centres of technological inno- needs have to be catered to by service work- vation in the world (Castells and Hall, 1994), ers. These are the ingredients for the growth or media-related, as in Los Angeles and New of the metropolitan region. Knowlege sites York (Abrahamson, 2004). The key innova- and communication networks are the spatial tion and decision-making processes occur attractors for the information economy as the during face-to-face contact, and they still sites of natural resources and the networks require a shared space. of power distribution determined the geog- What is fundamentally new is that these raphy of the industrial economy. And this nodes interact globally, instantly or in chosen is valid for London, Mumbai, São Paulo or times throughout the planet. So the network

Downloaded from usj.sagepub.com at Univ of Newcastle upon Tyne on October 27, 2016 2742 MANUEL CASTELLS of decision implementation is a global macro which are the most important nowadays. electronic network. Meanwhile, the network Spatial economies of synergy mean that being of decision-making and generation of initia- in a place where there is potential interaction tives, ideas and innovation is a micro network with valuable partners creates the possibility operated by face-to-face communication of adding value as a result of the innovation concentrated in certain places. This spatial generated by this interaction. Economies of architecture simultaneously explains the scale can be transformed by information and concentration of some metropolitan places communication technologies in their spatial and the diffusion in terms of networks: the logic. Electronic networks allow for the for- space of places and the space of flows. Once mation of global assembly lines. Software this mechanism is identified, everything else production can be spatially distributed and can be explained: concentration of ancillary co-ordinated by communication networks. services, infrastructure in communication On the other hand, economies of synergy still that develops in one site and not in others, require the spatial concentration of interper- attraction of talent, satisfactory living condi- sonal interaction because communication tions for the creators of value, etc. operates on a much broader bandwidth than Communication infrastructures are decisive digital communication at a distance. This is components of the process of mega metro- why scientific research is still concentrated in politanisation, but they are not the origin of campuses around the world at the same time the process. Infrastructure of communication that these campuses cannot operate without develops because there is something to com- being networked with the world wide web municate. It is this functional need that calls of science. for the development of infrastructures. The In the age of information, and innovation, value-making locales offer greater opportuni- cities remain more than ever the sites of ties and better services, and this offer attracts generation of value and the material basis talented and innovative professionals. And of power, cultural production and social because there is money, there is a thriving selection (Hall, 1998). Quality of life has market and there are better cultural amenities, nothing to do with it. Quality of life is an educational facilities and health services, and entirely subjective notion. Green Silicon therefore jobs which are still the main source Valley suburbs are boring places to live in of urban growth. Since job opportunities are from the point of view of a hard-core New globally appealing, these metropolitan regions Yorker or an unreconstructed Parisian. Yet also become the hubs for immigration. They that is where the most advanced technological develop as multi-ethnic places and establish innovation happens and where every wave of global connections not only at the level of major technological innovation in the past 50 functional and economic interactions, but years has occurred. Silicon Valley engineers also at the level of interpersonal relations— do not often go to San Francisco’s bars. They the networks and people, conceptualised as barely have time to go to their suburban ‘transnationalism from below’ by Michael bars in Silicon Valley. Why are they there Smith and Luis Guarnizo (1998). then? Because of the quality of life? Because At the source of the process of metropoli- of night life? No, because they are excited by tanisation, there is the ability to concentrate their work, they are fascinated by their own production of services, finance, technol- creativity and they cherish the possibility of ogy, markets and people. And this creates being close to other creators. Cities become economies of scale, as in previous forms of trendy only when they have the power and urbanisation, as well as economies of synergy money to launch the trends.

Downloaded from usj.sagepub.com at Univ of Newcastle upon Tyne on October 27, 2016 GLOBALISATION, NETWORKING, URBANISATION 2743

Now, the most strategically important nodes for the entire global system, connecting observation for an analysis in terms of spatial various networks. London and New York are networks is that these global networks do not typical cases of this multiple nodal advantage. have the same geography, and they usually do Boston does not reach the same level because, not share the same nodes. The network of even if it is probably the dominant node in innovation in information and communica- academic research and an important node tion technology—i.e. Silicon Valley—is not in technological innovation (particularly in the same as the network of finance, except biotechnology), it is only a secondary node in for the network of venture capital usually financial networks and is subsidiary to other originated from inside the high-technology nodes in a number of important dimensions industry. Political agencies, nationally and of wealth and power. This is also another internationally, build their own spatial sites reason why in China there is a clear differen- and networks of power. The global network tiation between Beijing and Shanghai in terms of scientific research does not overlap with the of the nodes and the distinct role they play in networks of technological innovation. That the global architecture: Beijing specialises in is why so many are surprised by the failures the political, financial, scientific and techno- of projects aimed at developing new Silicon logical; while Shanghai specialises in financial Valleys around a new university. Artistic cre- networks and global trade. ativity also has its own network, which shifts These mega nodes are not global cities. constantly, depending on fields of art and They are simply the urban dimension of movements of fashion. The global criminal multilayered global networks, which is a dif- economy (accounting for 5 per cent of global ferent matter. In other words, to understand GDP) is built on its own specific networks the dynamics and meaning of the node, we with nodes that do not generally coincide with must start with the analysis of each network those of finance or technological innovation. and their interaction as facilitated by their The management of drugs traffic features spatial convergence. However, each mega places such as Medellin, Bogota, Mexico, node becomes an attractor of capital, labour Miami, Bangkok, Kabul and Amsterdam, and innovation. This is where the contradic- most of them secondary nodes for other tions arise. A mega node attracts resources major networks. and accumulates opportunities to increase Therefore, there is a multilayering of global wealth and power. At the same time, because networks in the key strategic activities that it rarely has institutional existence or the structure and deconstruct the planet. When political capacity of autonomous decision- these multilayered networks overlap in some making as a metropolitan region, it can hardly node, when there is a node that belongs to implement redistributive policies on behalf different networks, two major consequences of the needs of the local. In the absence of follow. First, economies of synergy between active social demands and social movements, these different networks take place in that the mega node imposes the logic of the global node—i.e. between financial markets and over the local. The net result of this process is media businesses, or between academic the co-existence of metropolitan dynamism research and technology development and with metropolitan marginality, expressed in innovation, or between politics and media. In the dramatic growth of squatter settlements addition, because these multilayered networks around the world and in the persistence of land on particular places, and many networks urban squalour in the banlieus of Paris and share a node in such places, these localities in the inner cities of America. There is an become mega nodes: they become switching increasing contradiction between the space

Downloaded from usj.sagepub.com at Univ of Newcastle upon Tyne on October 27, 2016 2744 MANUEL CASTELLS of flows and the space of places. These mega Castells, M. (2004) The Power of Identity, 2nd nodes concentrate more and more wealth, edn. Oxford: Blackwell. power and innovation in the planet. At the Castells, M. and Hall, P. (1994) Technopoles of the same time, few people in the world (actually World. London: Routledge. Castells, M., Fernandez-Ardevol, M., Qiu, J. L. 13 per cent according to the World Values and Sey, A. (2006) Mobile Communication Survey) identify with the global, cosmopoli- and Society: A Global Perspective. Cambridge, tan culture that populates the global networks MA: MIT Press. and is worshipped by the mega nodes’ élites. Dear, M. (2005) Comparative urbanism, Urban In contrast, 47 per cent of people feel a strong Geography, 26(3), pp. 247–251. regional or local identity. Thus global net- Dear, M. (2006) Critical responses to the Los works integrate certain dimensions of human Angeles school of urbanism. Paper presented at life and exclude other dimensions regardless the 2006 meeting of the Association of American of what the intentions of the actors are. The Geographers, Chicago, IL. contradictory relationship between meaning Graham, S. (Ed.) (2005) Cybercities: A Reader. London: Routledge. and power is manifested by a growing disas- Graham, S. and Simon, M. (2001) Splintering sociation between the space of flows and the Urbanism: Networked Infrastructure, space of places. There are places in the space Technological Mobilities and the Urban of flows and flows in the space of places but Condition. London: Routledge. the meaning is defined in place terms, while Hackworth, J. (2005) Emergent urban forms or the functionality, the wealth and the power emergent post-modernisms? A comparison of are defined in terms of flows. And this is the large US metropolitan areas, Urban Geography, most fundamental contradiction emerging in 26(6), pp. 484–519. our globalised, urbanised, networked world. Hall, P. (1998) Cities and Civilization. New York: Pantheon Books. Hall, P. and Pain, K. (2006) The Polycentric References Metropolis. London: Earthscan. Halle, D. (Ed.) (2003) New York and Los Angeles: Abrahamson, M. (2004) Global Cities. Oxford: Politics, Society and Culture. Chicago, IL: Oxford University Press. University of Chicago Press. Abu-Lughod, J. (1999) New York, Chicago, Los Hawley, A. (1950) Human Ecology. New York: Angeles: America’s Global Cities. Minneapolis, Free Press. MN: University of Minnesota Press. Hawley, A. (1956) The Changing Shape of Borja, J. and Castells, M. (1997) Local and Global: Metropolitan America. New York: Free Press. The Management of Cities in the Information Hughes, T. (1983) Networks of Power: Electrification Age. London: Earthscan. in Western Society, 1880–1930. Baltimore, MD: Broudehoux, A. M. (2004) The Making and Selling The Johns Hopkins University Press. of Post-Mao Beijing. London: Routledge. Innis, H. (1950) Empire and Communication. Castells, M. (1989) The Informational City. Oxford: Oxford: Oxford University Press. Blackwell. Innis, H. (1951) The Bias of Communication. Castells, M. (1999) The culture of cities in the Toronto: University of Toronto Press. information age. Paper presented at the the US Kiyoshi, K., Lakshmanan, T. R. and Anderson, W. P. Library of Congress Conference on Frontiers (2006) Structural Changes in Transportation in the Mind in the Twenty-first Century, and Communication in the Knowledge Society. Washington, DC, June [published in I. Susser Northampton, MA: Edward Elgar. (Ed.) (2002) The Castells Reader on Cities and Kwok, R. Y. W. (2005) Globalizing Taipei. London: Social Theory, pp. 367–389. Oxford: Blackwell]. Routledge. Castells, M. (2000) The Rise of the Network Society, Lim, W. (1998) Asian New Urbanism. Singapore: 2nd edn. Oxford: Blackwell. Select Books.

Downloaded from usj.sagepub.com at Univ of Newcastle upon Tyne on October 27, 2016 GLOBALISATION, NETWORKING, URBANISATION 2745

Lu, D. (2006) Remaking the Chinese Urban Form. Saxenian, A. L. (1994) Regional Advantage. London: Routledge. Cambridge, MA: Harvard University Press. Mattos, C. de, Figueroa, O., Bannen, P. and Campos, D. Schneider, J. and Susser, I. (Eds) (2003) Wounded (Eds) (2004) Santiago en EURE: Huellas de Cities. New York: Berg. una metamorfosis metropolitana, 1970/2000. Scott, A. (1998) Regions and the World Economy. Santiago de Chile: Instituto de Estudios Urbanos Oxford: Oxford University Press. y Territoriales, Universidad Catolica de Chile. Scott, A. (Ed.) (2001) Global City-regions: Trends, Mitchell, W. (1999) E-topia. Cambridge, MA: Theory, Policy. Oxford: Oxford University MIT Press. Press. Neuwirth, R. (2005) Shadow City: A Billion Squatters: Smith, M. P. and Guarnizo, L. E. (Eds) (1998) A New Urban World. New York: Routledge. Transnationalism from Below. New Brunswick, Rutherford, J. (2004) A Tale of Two Global NJ: Transaction Publishers. Cities: Comparing the Territorialities of Suketu, M. (2004) Maximum City: Bombay Lost Telecommunications Developments in Paris and Found. Delhi: Penguin. and London. Aldershot: Ashgate. UNFPA (United Nations Population Fund) (2007) Sanyal, B. (Ed.) (2003) Comparative Planning State of World Population 2007: Unleashing the Cultures. Cambridge, MA: MIT Press. Potential of Urban Growth. New York: United Sassen, S. (1991) The Global City: London, New York, Nations. Tokyo. Princeton, NJ: Princeton University Press. Wolch, J., Pastor, M. and Dreier, P. (Eds) (2004) Sassen, S. (2006) Territory, Authority, Rights: From Up Against Sprawl: Public Policy in the Making Medieval to Global Assemblages. Princeton, NJ: of Southern California. Minneapolis, MN: Princeton University Press. University of Minnesota Press.

Downloaded from usj.sagepub.com at Univ of Newcastle upon Tyne on October 27, 2016