't studies

n the world - estimated to be the twcntieth2003 (United Nations, 2004) - is slippitrg

urbauity, we can be $ure that tlte task ofuch overdue.)n this need to appreciate a divcrsity ofwaysLl benclits of a more cosrropolitan basis fortioD to accounts of the impact of globalisa-

rbal and wotld cities offer an importantof the divisive theoretical manoeuvres thatTerent cities apart wilhin the field o[ urban

tho strong networks that have developed

focus. There is a renerved interest in under-:tiug cities iD quite divergeDt economic andrese al)proachcs have brought into vicw nerv

:s, stretching across a raDge ofdifferent c;ties,

tbey havq perhaps unwittiogly, reinstatedioDs about relationships amongst cities. Inuted to the popularity of a policy narrative,oorest cities need to face the challcng€ ofheir economies in order to become more like

'ntly at the top of a worldwide hicrarchy of

it-colonial critique might allow us to build an

urd ordinary cities, and despite their initialof 'global' and'world' cilies have reinsclibednto the field of urban studies. They havo also

of prescriptiors for all cities based on the

)ite of the strong transnalionalisn of global-

: challenges for a post-colonial urban studies

cities that rests on the expetiences of only a

d ways to approciale the distinctiveness of all

lUorld cities, or o world ofordinorV cities?

INTRODUffION

Globalisation has translormed urban studies. Cities are now routillely viewedas sites fol D'iucli wider social and economic processes, and the focus forundcrstanding urban processcs has shifted to emphasise fiows and networksthat pass through cities ralher thau the teritory of the city itself (Friedman1995a, Sassen 1991, Srnith 2001). The study o[ cities now oommonlyerconpasses the flows ofglobal finance capital, the footloose wanderirgs oftransnational manufacturing lirms, and the diversc mobilities of thc world'sclite alongside diasporic and migrant communities from poorer countries.In many ways, urban studies has becone much more cosrnopolitan in itsouLlook, Globalising featur€s comrnon to rnany cities around the worldencourage more writers to considcr cities lroDr differe[t regions, as well as

wealthicr and poorer cities, within the same field of analysis (Marcuse andvan Kcmpen 2000, Scott 2001, Graham and Marvin 2001). This is c€rtainlygood news for a posGcolonial urban studies, eager to bring different kinds ofcities together in thinking about contemporary urban experiences.

The situation appears more propitious than evet then, for an integrationof urban studies across long-standing divisions of scholarship, especiallybetween Western and othercities, including'Third World'and former socialistcitics, An analytical focus on the transnational global ecouomy could ensurethat such categorisations ofcities will no longer be of any relevance. Indeed,this is a claim made by the key advocates of these approaches (Sassen 1994,

Taylor 2001). Does this mean that urban studies has come to be sensitive tothe diversity of urban experiences, to the wide range of cities across thewolld? Could this be the basis for a post-colonial urban theoly that refus€s toprivilcge thc cxperiences ofsome cities over those of others?

Many studies of globalisation and cities have drawn on the idea of 'world'cities to understand the role of cities ill the widet netq,orks and circulationsassociated with globalisation. Sorne cities outside the usual purview ofWestern urban theory -'Third World cities' - havc been incorporated intothesc studics in so far as they are involved in those globalisiug processes con-sidered relevant to the definition of world cities. This is defirlitely a positive

4

94 llorld cities, or a world of ordinary cities?

development in terms of ambitions to post-colonialise urban studies, to

overcome the entre[ched divisions bet$'€e[ studies of 'Western' and

'Third World' cities. But many cities around the world remain 'off the map'

of this version of urbaD theory (Robinson 2002a). Ar'd despit€ the relative

inclusiveness of the focus on globalisation processes, developmentalism

continues to pervade global- and world-city narratives, consigning poorer

cities to a different theoretical world dominated by the conc€rns of develop-

ment. Although the older categories of First and Third World may have less

purchasq world cities approaches have a strong interest in hierarchies,

and have invented new kinds of catego es to divide up the world of cities

Perhaps most worrying for a post-coloDial urban studies, world-cities

approaches, by placing cities in hierarchical relaiion to one another, implicitlyestablish some cities as exemplars and others as imitators In policy-rclated

versions of these accounts cities either off the \torld-cities map or low

down the supposed hierarchy have an implicit injunction to become more

like those at the top of the hierarchy of cities: they need to climb up the

hierarchy to get a piece of the (gtobal) action. Being one of the top-rankglobal cities can be equally burdensomq though, encouraging a policy

emphasis on only small, successful and globalising segments of the economy

aod neglecting the diversity of urban life and urban economies in these

places.So while there is much to leam from global- and world-cities approaches,

this chapter suggests that there is still considerable work to b€ done to pro-

duce a post-colonial form ofurban studies relevant to a world of cities, mtherthan simply for selected'world cities'. Noting especially the adverse politicalconsequsnces of analyses that emphasise hierarchies and categories and

that still divide the field of urban studies along deYelopmentalist lines, the

chapter presses the impo ance of letting a1l cities be ordinary World citiesapproaches, it will be suggested, operate to limit imaginations of possible

urban futurcs, especially in relation to poorer cities, and the situation ofpoorand marginalised people ir cities around the world. A post-colonial urbanstudi€s needs to move beyond categories and hierarchies and to abandonclaims to represent some cilies as exempla$ for otheN. It needs to be able tobe attentive to the diverse experienc3:s of a world of citi€s. While global-

. and world-cities approaches have much to offer, ultimately they leave these

challenges unmet. Instead, this chapter makes the case that all cities shouldbe viewed as ordinary, both distinctive and part of an interconnecled worldofcities- The last section spells out what this might mean. And, as we will see,

an ordinary-city approach is as important for the wealthiest ('global') cities,

as for the poorest.

groBnt RND llrontD ctTt€s

World ciries are thought to be differcnt from others to the extent that theyplay an important role in a iculating regional, national and international

Wo d cittes, or a world of ordinary cities? 95

economies into a global economy. With the rise since the 1970s of trans-national investment flows and the emergence of a new intemational divisionoflabour based on th€ restructuring of manufactu ng production processe\now involving integrated production processes dispersed across the globe,schola$ have been drawn to rethilk tbe role ofcities in rclation to the globaleconomy. It was suggested that some cities were increasingly serving as theorganising nodes of a global economic system, rather than simply beinglinked into local hinterlands or part ofnational systems ofcities- At the sametimq many populalions were being excluded from these new spaces ofglobalcapitalism, and thus from the field ofworld cities: to these writers some citieswerc becoming 'economically irrelevant' (Knox 1995: 4l).

A corollary, and drawing heavily on wider accounts of uneven develol>ment of the world economy, was the suggestion that these global nodes, orworld cities, could be arranged hierarchically, roughly in accord with theeconomic power they command. A key mechanism in making this a dynamichierarchy, rather tban simply a static given was the existence of competitionbetweeD world cities, which would drive some to se, others to fall. Further-more, external shocks and wider structural processes would shape the respec-tive fortunes of world cities and, to some extent, determine their position inthe hienrchy. It was assumed, then, that cities could rise and fall through thehierarchy, and that their overall position in relation to other cities would bedetermined by the relative balance of global, national and regional iniuencethat they could mobilise (see for example, Hall 1981, 1966, Knox and Taylorr995).

World-systems theory more generally, which was a strong influence onsome early world-cities theo sts, proposed that count es across the worldare seen to occupy a place withiD the hierarchy of the world economy andpossibly make their way up through the catego es (corg periphery semi-periphery) embedded in the world-economy approach, Following this, theworld-cities approach assumes that cities occupy similar positions andhave a similar capacity to progress up or fall down the ranks. The countrycategorisations of core, periphery and semi-periphery in world-systemstheory were transferred to these initial analyses of world cities, so that thehierarchies of world cities werc determiDed by relative levels of economicdevelopment. This meant that wdters also found it useful to continue touse development-related categorisations to desc be cities' positions in thehierarchy, including the division b€tween First and Third World cities. Acommon suggestion though was that with globalisation First World citieswere assuming some of the characteristics ofThird World cities in terms of agrowing number oflow-wage, sewice-sectorjobs and poor living conditions -and that some Third World citjes were becoming more like First World citiesin terms of the balance of economic activities and the pres€nce of a globalelite (Friedman 1995a, Jones 1998). Knox w tes, for example, that,'justaswecan see the world cityness of rcgional metropoli, so we can see the Third-World-ness ofworld cities' (1995: l5). These categorisations were being used

96 llorld. cities, or a world oJ ordinary cities?

more flexibly then, but remained a very impodant lens with which to view theworld ofc;ties. So, at the same time that world-cities analysts werc extendinstheir focus to include poorer cities, they were reinscribirg developmentalijassumptions about the hierarchical relations amongst cities.

In a prominent contribution to the world-cities lileraturq Saskia Sassen(1991, 1994) coined the term 'global cities'to capture what she suggests is adistinctive feature ofthe current (1980s on) phase ofthe worid economv: theglobal orga[isation and increasingly transnarional structure of-eleme;ts ofthe global economy. He! key take-home poina is that the spatially dispe$edglobal economy requires locally based and integrated orgalisation andthis, she suggests, takes place in global cities. Although many transnationalcompanies no longer keep their headquarters in central areas of thesemajor cities, the specialised firms that they rely on to produce the capabilitiesand innovatioDs necessary for command and control of their globaloperations.have remained or chosen to establish themselv€s therq includingadvanced business and producer services. Iegal and financial servjces. More-ovel it is no longer the lalge transnational corporations that are the centreof these functions, but small parts of a few major cities that play host to andenable the effective functioning through prcximity of a growing number ofthese new producer and business-services firms (Sassen 2001a). A similarargument concerning the benefits of colocation for finance and invest_ment firms suggests that these cuttirg-edge activities are produced in a fewmajor cities. Colocation benefits both these sets of firms as this facilitatesface-to-face interaction and the emergence of trust wjth potential partn€rs,which is crucial in terms of enabling innovation and coping with ihe risk,complexity and speaulative character of many of these activities (Sassen199+ 84).

Both global- and world-cities analyses bdng iDto view the wider processesshaping cities jn a globalising world and economia networks amonlst cities.However, the emphasis has been on a relatjvely small range of economicprocesses with a certain 'global' reach. This has limjted the applicability ofworld-cities approaches, excluding man1, cities from its ;;nsideration_Although status within the world-city hierarchy has traditionallv been basedon a range of criteria, including national standing, location of state andinterstate agencies and cultural functions, the primary determination ofstatus in this frarnework is economic. as Friedmann 1995a notes, ,Theeconomic variable is likely to be decisive for all artempts at explanation,(1995a:317). This has become more, not less, apparent in the world_citiesliterature over time as more recent research has locused on identifying thetransnational business connections that deline the very top rank of worldcities, Sassen's 'global cides' (Beaversloct et al. tlO9, Saisen ZO0ta;.

World-cities approaches have be€n strongly shaped by an interesi in deter_mining the existence of categories of cities and identifying hierarchicalrelations amongst cities. This led John Friedman to ast, in nis 19g6 reviewof 'World City Research: 10 years on,, whether the world_city hypothesis ,is a

World cities, or a world of ordinary ciries? 9'7

h€u stic, a way of asking questions about cities in general, or a statementabout a class of particular cities - world cities _ set apart from other urbalagglomerations by specifiable characte stics?, (1995b: Z3). He suggests rhat itis.both, but that the tendency has been to categorise cities into a'ir-ierarchy inwhich wo d cilies are at the top of the tree of influence. This leaguetableapproach has shaped the ways in which cities around the world have beenrepr€sented - or not represented at all - within the world_cities lit€rature.From the dizzy heights of the diagrammer, certain significant cities are iden-tified, labelled, processed and placed in a hierarchy, with very little attentive_ness to the diyerse experiences of that city or even to extant literature aboutthat plac€. The daDger here is that out-of-datq unsuitable or uueliable dara(Short et al. 1996; although see Beaverstock et al. 2000.) and possibly a lack offamilia ty- with some of the regions being considered can lead to the pro_duction of maps that are simply inaccurate. These images of the world ofimpo ant citi€s have been used again and agail to illustrate the perspectiveof world-cities theorists and leave a strong impression on policy-makers,popularising the idea rhat moving up the hierarchy of cities ii both possible

1nd.a.C3od thing. Peter Taylor (2000: t4) notes with disapproval, ihough,the 'widespread reporting of [. . .] a preliminary ta*ono-yl of *oild cities.However, revised versions of these taxonomies, based on more substantialresearch, dra\{ remarkably similar conclusions, and similar maps, as we willconsrder^in the section 'Extending clobal- and World_Cities ipproaches,,pp- 103-8 below

In contrast to the world-city enthusiasm for categodes, the global_cityanalysis has.a strong €mphasis on procsss. It is the locational dy-namics ofkey, sectors involved in managing the global economy that give rise to theglobal-city label. However, the category of global iity th; is identifiedthrough this subtle analysis depends on the eiperiencei of a minor set ofeconomic activities based in only a small part of these cities. They may con_stitute the more dynamic sectors of these cities, economies, Uut Sassentevidence of declining location quotients for these activities in the 1990s (forexample, 2001a: 134,5) suggesrs that the conc€ntuated growth spurt in ihissector may well be over. And it is important to put the contribution of thesesectors to the wider city economy into p€$pective- In London, for example,where transrational finance and business services are still most dynamic indhighly concentrated, the Londor Development Agency (LDA) suggests thatoDly 'about ?0 per cent of emplolrnent is in firms whose main-irarket isnational rather than international, (LDA 2000: lg). Even this city, routinelyat the top_of world-city hierarchiei is poorly served by a reduciion of itscomplex, diverse social and economic life to the phenomenon of globalisa_tion, and is certainly poorly described as a

,global, ;jty. There h*" tl* -""ycritjcisms of the empirical basis for claims that glob;l cities arc significanflydiflerent from other major centres in terns of the composition of"""orro*i"aclivities, wage levels or social conditions (Abu_Lughod 1995, Short et al.1996, Storper 1997, Smith 2001, Buck et al. 2002).

98 WotA c ies,otawo dof odinaD) cities?

Nonetheless, the global-city hypothesis has had a powerful discursive effect

in both academic and policy circles. The pithy identification of the 'globalcity' as a category of cities which, it is claimed, are powerful in terms of theglobal economy, has had widespread appeal.r However, this has depended oncontinuing, indeed strcngthening, the world-city emphasis on a limited range

of economic activities with a ce ain global reach as defining f€atures of the

global city. This has the effect of hiding most of the activities within global

cities frorn view, while at the same time also dropping most cities in the worldfrom its vision. The insights of tbe global-city analysis are very important tounderstanding the way in which some aspects of some cities are functioningwithin the global economy. But perhaps it would be more appropriate iftheseprocesses were described as an example of an 'industrial' district- They couldbe called new industrial districts of tmnsnational management and control.The core understaDding about these novel processes would remain imPodantboth theoretically and in policy terms, But perhaps some of the moreunfo unate consequences for cities of the global- and world-cities labels,

rvhich we will consider further below, could be avoided.World-cities resealch, then, has moved on from the time of Friedmann's

influential mid-1980s review of the field and, especially in the wake of Saskia

Sassen's study, The Global C,tl, it has adopted a strong and intenselyresearched empirical focus on transnational business and finance n€tworks(see, for example, Beave$tock et al. 1999, Taylor 2004). In some ways, thefocus of attention has narrowed, although there has been a concerted effortto focus on processes and to track connections amongst cities rather thansimply to map city attribut€s (Beavemtock e! al. 2000). However, cities stillend up categorised in boxes or in diagrammatic maps and assigned a place inrelation to a priori anall4ical hierarchies. A view of the world of citiesemerges where some cities come to be seen as the pinnacl€ of achievement,setting up sometimes impossible ambitions lbr other cities. This also suggests

to the most powerful cjties that they need to emphasise those aspects oftheircities that conform to the global and world cities account, with som€timesdetrimental effects on other kinds ofactivities and on the wider social life ofthe city (see, for example, Markusen and Gwiasda 1994, Sites 2003). Global-and world-cities approaches expose aD analytical tension between assessing

the characteristics and potential of cities on the basis of the proc€sses thatmatter from within their diverse dynamic social and economic worlds or onthe basis of criteria determined by the external theoretical construct of theworld or global economy (see also Varsanyi 2000). This is at the heart of howa world-cities approach can limit imaginations about the fuaures ofcities andwhy I propose instead to think about a world ofcities, all quite ordinary

If global- and world-cities approaches offer only a limited window ontothose cities that make it into the league iables, and even those at the top, weshould also be concemed about the effect of these hierarchies and leaguetables on those cities that are quite literall)' off the maps of the global- andworld-cities theorisls- Millions of peopl€ and hundreds of cities are dropped

World cities, or a world oJ ordinary citiesT 99

off the map of rnuch research in urban studies to service the very restrictedview that the global and world cities analyses encourage rcgarding the signifi-cance or (ir)relevance ofcities in relation to certain rather narrow s€ctions ofthe global economy. For the purposes of developing a post-colodal urbanstudies relevant for a world ofcities, global and world cities approaches havesome serious limitations.

'FlttlNG lN THC VOIDS": OFF TH€ UJOR[D-(|T|€S mRP

In his account of cities across the world, King provocatively noted that'all cities today are "world cities" ' (1990: 82). Unfortuoately, research andwriting within the rubric of th€ world-cities approach or hypothesis hasgenerally not chos€n to build on this observation. Because the analysis ofglobal- and world-cities theorists has come to rely on identifying the signifi-cance ofcities to only certain elements ofthe global economy, cities that arepoorer, marginal to key globalising economic sectors ot as Manuel Castells(1983) puts it, 'structurally irrelevaDt' receive very little attention in thisapproach. However, precisely because the geography of these globalisingeconomic sectors is changing, world-cities analysts have been druwn toexplore those cities in poorer contexts that arc increasingly assuming whatSaskia Sassen calls global-city fuDctions.

According to Sassen, some of the functions of command and control ofthe global economy also take place in what were once peripheral cities. Firmsin these places are increasingly drawn to play a role in coordinating globalinvestments, as well as financial and business sewices regionally- Some ofthis, in Latin Ame ca for example, has to do with the strong privatisation ofstate-owned businesses and services that has drawn more pdvate-sectorinvestment and has created greater demand for financial and business servic.es(Schifrer 2002). Nonetheless, in her view, these processes signify theemergence of a new geography to the pe phery - a select $oup of cities,some in poorer counLries, are now deemed to have 'global city functions'although they fall short of being first-order global cities. She menrionsToronto, Sao Paulo, Mexico City, Miami and Sydney- This signals somethingofthe'end ofthe Third World'(Harris 1986) as a category in urbar studies.Nonetheless, Sassen acknowledges that her approach 'cannot account forthe cases ofmaDy cities that may not have experienced any of these develop-ments' (1994: 7)-

So, Sassenjoins others in consigning substantial cities around the globe tothe theoretical void because of their apparent structural irrelevance: 'signifi-cant parts ofAfrica and Latin America became unhinged from their hithertostrong ties with world markets in commodities and mw materials' (1994: 27),and, 'Alongside these new global and regional hierarchies of cities is a vastterritory that has becorne increasingly peripheral, increasingly excluded fromthe major economic processes that fuel economic growth in the new globaleconomy' (1994: 4).

100 World cities, or a world of ordinarJ cities?

'Knox goes even further to suggest that 'the mega cities of the peripherywill fare no better than the catatonic agradan societies that have fuelledtheir (demographic) growth, and in which both will lapse decisively andiretdevably into a "slow" economic time zone' (1995: 15).

There are obviously important ways in which the changing geography

of the ilternational economy has impacted negatively on cities io poorcrcountries. As Sassen suggests, Western investment in poorer countries hasdeclined precipitously since the time in the 1970s when substantial oilsurpluses were .ecycled through poor count es as debt (1991: 83). Thisgeography has been very uneven, though: since the 1970s Latin America hasbeen overtaken by south-east Asia as the top destination for foreign invest-ment in manufacturing, and rnany poor count es hay€ became net exportersof capital through debt repayments (1991: 63). Nonetheless, the global-citiesanalysis alerts us to the role of the international financial centres of manycountries in performing important 'gateway functions'for inte.national flowsof finance and the provision ofglobal business seryices (1991: 173). Even so,

the focus of global- and more rec€nt world-cities work remains on a limitedset of seNice-sector economic activities. And although global-city functionsare assuming an incrcasingly tmnsnational form, relatively few cjties canhope to participate in them (Taylor 2004).

The'end ofthe Third World'(Harris 1986) is perhaps an accurate assess-

ment ofchanges over the past three to four decades in places like Hong Kong,Singapore, Taiwan, South Korea and even Malaysia, and the appearanceof city-states and some major urban c€ntres in roste$ of first- and second-order global cities refl€cts this. But in parts of the world where global citieshave not been identified the 'ioids' of world- and global-city analyses theexperience ofmany countries and cities in relation to the global economy hasbeen much more uneven than global- and world-cities analyses suggest.For many, the 1980s and 1990s have been long decades of little growth aDdgrowing inequality. It is, however, inaccurate to caricature even the poorestregions as excluded from the global economy or doomed to occupy a slowzone of the world economy. Africa, frequently written off in total in theseglobal analyses, has in fact had a very uneven growth record. As the AfricanDevelopment Bank (2000) notes:

Wlile the continent has, in overall terms, lagged behind other regions, afew countries have produced remarkable economic results, even by worldstandards [. . .] In an encouraging development, as many as 12 countriesarc estimated to have rccorded real GDP growth rates above 5 per centwhile close lo 30 countries had positive real GDP p€r capita growth.

(2000:1)

It is hard to disagree that some countdes and cities have lost many of thetrading and investment links that charaDterised an earlier era of globaleconomic relations. A country like Zambia. for example, now one ofthe most

llotld cities, or a world of ordinary cities? 101

heavily iodebted nations in the world and ce ainly one of the poorest, hasseen the value of its primary export, coppel plummet on the world marketsince the 1970s. Its position within an older intemational division of labouris no longer econornically viable, and it has yet to find a successful path forfuture economic growth. En route it has suffered the consequ€nces of one ofthe World Bank/IMF's most ruthless Structural Adjustment Programmes(Young 1988, Clark 1989, BoDnick 1997). However, Zambia is also one ofthemosl urbanised countries on th€ African continent, and its capital cityLusaka is a testimony to the modemist dreams of both the form€r colonialpowers and the post-independ€nce government (Hans€n 1997)- Today,though, with over 70 per cent of the population in Lusaka dependent onearnings from the iDformal sector (goverDment bureaucrats arc know! toearn less than some street traders, see Moser and Holland 1997), the oIIc€bright economic and social future of this city must feel itself like a dream -albeit one which lvas for a time very rcal to many people (Ferguson 1999).

Lusaka3 is certainly not a player in the 'major economic processes that fueleconomic growth in the new global economy' (Sassen 1994: 198). But copperis still exported, as ar€ agdcultural goods and oppo unities for investment as

state assets are privatised. Despite the lack of foreign currency (and some-times because of it) all sorts of links and connections to the global economypersist. From the World Bank, to aid agencies, intemational politicalorganisations and trade in second-hand clothing and other goods and ser-vices, Lusaka is still constituted and reproduced through its relations withother parts of the country other cities and othq parts of the region andglobe (see, for example, Hansen I994, 1997). The city continues to perform itsfunctions of national and regional centrality in rclation to political andfinancial services, and operates as a significant market (and occasionally pro-duction site) for goods and services from across the country and the world.

Il is one thing, though, to agree that global links are changing and thatpower rclations, inequalities and povedy shape the quality ofthose links. It isquite another to suggest that poor cities and countries are irelevant to theglobal economy. When looked at from th€ point of view of these places thatare dlegedly'offthe map', the global economy is ofenormous significance inshaping the futures and fo unes of cities around the world. For many poor,'structurally irrelevant' citi€s, the significance offlows of ideas, practices andresources beyond and into the city concerned from around the world standsin stark contrast to these claims of irrelevance. As Gavin Shatkin writesabout Phnom Penh, 'In order to arive at a proper understanding of theprocess of urbanisation in LDCs fless developed countri€s], it is nec€ssary toexamine the ways in which countdes interface with the global economy, aswell as the social, cultural and historical legacies that each country cariesinto the era of globalisation' (1998: 381).

The historical legacies of these cities, it is clear from his account, are alsoproducts of earlier global encounters. Even the pooresa cities haye longhistories of intemctions and contacts with other places and have, over timq

102 World cities, or aworld of ordinaD; cities?

been drawn iDto the global economy in differ€nt roles, for tmde, production,extraction or cultural exchaDge. These connections, perhaps transformed, canremain vital components ofcontemporary urban dynamics-

Viewed from offthe (world-cities) map, some of the earliest versions oftheworld-city hypolhesis are more relevant to poorer cities than latet moreeconomistic accoun!$ Earlier wdters suggested a rang€ of criteria by whichto assess the role and functions of differcnt world cities- They determinedwhether cities were world cities with reference to economic, cultunl andpolitical processes (Friedmann and coetz 1982, Friedmann 1995a). Thismeans that the wider functions ofmany more cities can be brought into view.In these accounts, too, it is not only global processes that are rclevant: thespatial reach ofa city's iDfluenc€ is understood to vary and there is scope forthinking about the wider role of cities ir relation to their hinterland andnation, as well as to the global economy- Many more world cities are broughtinto view as significant provincial centrcs, political or slrnbolic centres, orperhaps as important transport and production hubs in national and regionaleconomies (Sirnon 1995). Guarding against economic reductionism andmoving beyond the limitations of the global scale of transnational activitieswould ensure that the range of cities of concern to world-cities theo sts isless exclusive (Varsanyi 2000).

But therc is still King's claim that'all cities are world cities', which we needto consider, and the fact that the world-cilies literaturq even in its mostnuanced and extensive form, persists in defining some cities out of thegamg as 'excluded lrom global capitilism' and therefore as irrelevant to theirtheoretical reflections Writers on ciLies in Africa, for example, asked to con-sider world cities in their region, conclude dismally that there ar€ no worldcities on the continent although they pojnt to Cairo and Johannesburga aspotentials (Rakodi 1997). Scholars of orher peripheral places such as LatinAmerica wonder about the usefulness ofthese categodes in,analysing what isoccurring' (Gilbert 1998: 174), and they have been thought to have lirtlerelevance to places in the Middle East or north Africa. As Stanley writes,'cities in this rcgion are not on the world lcities] map, (2001: 8).

If the category of world city is not applicable to a wide range of cities(Simon 1995), are there other ways in which the world-city hypothesis mightbe mobilised in these 'irrelevant' cities? A stronger focus on process thancategories could lead oDe to think about how ,global' economic processesaffect all cities - as Marcuse and van Kempen (2000) frame it, this leads to afocus on 'globalising cities', since 'globalisation [. . -] is a process that affectsall cities in the world, if to varying degrees and varying ways, not only thoseat the top of the global hierarchy' (2OOO: xvii). This formulation still leavesthe enthusiasm about hierarchies and categories in place though and retainsan emphasis on economic activities with a ,global' reach, but at least itidentifies a research agenda applicable to a wider range ofcities.

Most importantly, perhaps, but seldom mentioned, the particular ,globaleconomy' that is being used as the ground and foundation for identifying

Iyorld cities, or a world of ordinary cities? lO3

both place in hierarchy and relevant social and economic processes is onlyone of many forms of global economic connection.5 The crite a for globalsignificance might well look very different were the map-makers to relocatethemselves and review significant transnational networks in places such asJakarta or Kuala Lumpur, where ties to Islamic forms of global economicand political activity might result in a very different list of powerful cities(\4/hite 1998, Allen 1999, Firman 1999, Simone 2004). Similarty, rhe trans-national activities of agencies like the World Bank and the InternationalMon€tary Fund (IMF) who drive the circulation of knowledge and the dis-ciplining power involved in recovering old bank and continuing bilateral andmultilateml debt from the poorest countries in the world - debt, it shouldbe pointed out, which, in aa earlier phasg these agencies rccommended topoor countdes - would draw another crucial gaph of global financial andeconomic connections shaping (or devastating) city life.

Global- and world-cities analyses have done much to enhance under-standings of cities in a globalising economy, focusing on flows and networksamongst cities, and including some poorer cities within the same freld ofanalysis as the most w€althy- In some ways, then, these analyses hav€ con-tributed to postcolonialising urban studies, underrnining inherited assump-tions about the differences amongst cities. But at the same time they hav€reinscribed these assumptions, through a llarrow focus on only a small rangeofglobal economic activities, even leaving some cities off the theoretical mapaltogelher. Some of these points have been noted by the leading theorists ofglobal and world cities. And there have been some new initjatives to expandthe field of analysis to include more cities, and to respond to concems aboutthe division ofthe world ofcities along developmentalist lines. The followingsection explores whether these new trcnds in global- and world-citiesapproaches can produce an analysis relevant for a world of cities.

€XT€NDING GTOBRI.. RND II'OBTD-(ITI€S RPPfiOR(H€S?

Global- and world-cities approaches have been around for some time andhave come in for a lot of criticism even as they have dominated studiesof cities around the world. As c ticisms have mounted, the advocates ofth€se approaches have slowly extended their databases and entered a rangeof caveats into their accounts. Thes€ adde[da and caveats address some ofthe concerns I've noted above, especially regarding the importance ofpayingattention to a wider range of cities than initially considercd. Together, theyindicate a strong movement beyond the odginal assumptions of global- andworld-cjties approaches, but also suggest the limits of this productive era inurban studies. Alternative ways of analysing cities around the world areemerging which, in my view, offer a more convincing starting point for a post_colonial urbau studies committed to a cosmopolitan form of theoriiing.Building oa the appreciation of the role of globalisatioq in shaping citylif€, accounts of'ordinary cities', howevet emphasise a diversity ofeconomic

L

104 World cities, or aworld of ordinary cities?

and social neiworks, overlapping to produce dynamic and complex urban

societies. We will return to consider the case for ordinary cities below The

path to this alternative approach, though, is paved with the criticisms and

caveats acknowledged by global- and world-cities theorists themselves

A major data-construction exercise by Beavemtock et al. has ofered an

oppo unity to assess some of the claims of the world-cities analyses (see

Beaverstock et al. 1999, 2000). Building on Sassen's identification ofadvanced producer sewices (APS) as the key distinguishing feature ofthe top

tier of global cities, these researchers 'scavenged' (their word) data on majorfinancial, legal, accounting, advertising and banking firms with a presence

'in at least fifteen different cities, including one or more cities in each of the

p me globalization arenas: northern America (USA and Canada), Western

Europe and Pacific Asia' (Taylor 2004: 65). As we noted above, this strongly

skews the data set to networks of Western firms.6 Using this data set, allddifferently from earlier approaches, the researchers were eager not to assume

the existence of hierarchies amongst cities. Taylor notes that 'hierarchical

structures are to be found in frrms not in the cities'(2004: l8), and insists thatwhether there is a hierarchy amongst world cities must remaiu an empiricalquestion (2004: 3l), For him, power in networks ofcities can take two forms:

what he calls 'networked power', and the more familiar power to 'command

and control' the global economy, which remains of interest to Sassen- As she

notes, '[t]he concept of the global city introduces a far stronger emphasis

on strategic components of the global economy, and hence on questions ofpower' (2001b: 80).

It is this concern with the relative power of cities that sees the persistent

identification of hierarchies in even these nuanced responses to their critics.

Cities that are very low down the ranking ofconnectivity to what Taylor calls

the 'world city network' - in reality, a network of selected APS firms - are

construed as having 'connectivity-through-subordination' (Taylor 2004: 93).

Mile this means that even the most slightly connected global cities are

acknowledged to have networked capacities in relation to APS growth, Taylorobseryes tha!'the fact that some power has been diffused is totally swamped

by [the] Iarger economic forces that have steered contemporary globalization'(Taylor 2004: 199). Pursuing an assessment of where power is located withinworld-cities networks, Taylor concludes that 'Globalization may be a world-wide phenomenon but its command centres are most certainly not so dis-tdbuted' (2004: 89) and that 'globalization begios to looft very "Western" as

soon as dircct expressions ofpower are investigated' (20M: 91).It is profoundly important to remember, along with these researchers

themselves, lhat 'Of course, th€se hierarchical processes are generated in thefirms themselves' (Taylor 2004: 175)- But they assume: or insist, that'theycarry through into ar ordering of world cities in terms of high connectivitiesof world and rcgjonal headquarters cities' (Taylor 2004: 175)- It is thisslippage, from the networked power relations of firms to the ascription of ahierarchical placing of cities in relation to one another that is th€ source of

World cities, or a world of ordinary cities? 105

ongoing concern about the impact of global- and world-cities analyses onthe world ofcities. It bears paying closer attention to the data on which theseconclusions are based.

Despite a strong argum€nt about the evidential gap within global- andworld-cities research, namely the absence ofevidence concerning actual flowsand connections between cities as opposed to locational or attribute datafrom within cities (Short et al. 1996), the data that sustains the latest round ofglobal- and world-cities analyses remains profoundly locational. The locationof major producer-services firms in cities across the globe was recorded, andrelations amongst different branches of these firms inferred from the relativesize of different operations and the 'extra local functions' of a firm's office ina city, such as headquartem or r€giotral offices (Taylor 2004: 66). No qualita-tive information was collected conceming the nature or meaning of theserelationships. There is no direct evidence concerning th€ networking rclatiolsamongst the flrms, let alone the consequenc€s of these relationships forunderstanding relations amongst cities Thus, it is not possible to draw robustconclusions about the powe! relations amongst cities or city stalus in theglobal economy. The evidential gap in world-cities research, unfortunately,remalns.

A further problem with much of the evidenc€ gathered in this study is itsvery narrow focus (Hall 2001): 'the research is very big geographically -global - but very narrow in topic' (Taylor 2004: 3). Thus it still remaiosfocused on the very small advanced producer-services sector. Recognisingthis, the world-cities research team collated some evidence on other forms ofglobalisation, including non-governmental organisation (NGO) Iocation indiffercnt cities, for example. Nairobi emerged top of the list of global citiesdefined along these criteria. A wide range of cities previously unrecognisedwithin global- and world-cities approaches come into view in this way, whichis to be welcomed. However, they still failed to capture the diverse and robustconnectedness of many large cities. My own home town, Durban, forexamplq is labelled a near-isolate according to these cdteria. Certailly it haspractically no headquarters function within the national economy and it hasfew branches of major multinational accounting, management consultingand banking firms (although it has some)- But this still misses the fact thatthis city of almost 3 million people is a major trading port of Africa and thesecond manufacturing city of a significant middle-income economy - clearlyit is not an isolated place! Indeed, Durban has many different economicconnections through the continent and the world and has strong connectionswith JohanDesburg-based companies for the provision of a wide range ofservices, as witressed by the very busy airljnk connecting the two cities. ThiskiDd of result illustrates how a continuing investment in global- and world-cities approaches directs our attention away from the dive6e and dynamiceconomic worlds of cities-

This example reioforces the conclusion that by relying on such narrow,location-bound indices to map world-city networks, global- aIId world-cities

106 World cities, or a world of ordinary cities?

analyses cannot offer any useful assessmenl of the economic significanc€ or'worldliness' ofmost cities. Moreovet and quite ironically, by only measuringplace-based evidence, they are unable to capture the economia importanceand potential ofa city's diverse con[eclions. The evidence they do gather ona wider range ofconnections - iDcluding advertising, legal firms and NGOs -is quickly passed over- Taylor asserts that even though '[c]learly cities inglobalization involves more than financial and business services [...] theiatter are the dominant networkers and I continue to focus on them'(2004:100). However, his own evidence sugg€sts that a proper accounting for thediversity of globalisations that shape cities is now in order, including manu-facturing, trade and NGOs and also informal networks- As Sassen notes inrelation to the three cities (New York, London and Tolcyo) she identifies as

the top tier of global cities, 'B€sides.the vast sei of activities that make uptheir economic base, many typical to all cities, these global cities have apafliculat component in their economic base [. . .] that gives them a specificrole in the current phase of the world economy' (Sassen 2001a: 127). It is,

I suggest, time to tum from these highly concentrated and specialist service-sector activities and pay proper attention to the vast range of activitiesmany also strongly shaped by globalisation - that make up the actualeconomies of cities.

The new data has ce ainly brought into view a much wider range of citiesthat global- and world-cities researchers are happy to call,'global cities'.Taylor notes, for example, that there is no such thing as a non-global city(Taylor 2004: 42), as all cities have the capacity to link in to the global net-work when they choose to, or when they are able to mobilise the resources todo so. More strongly, he suggests that 'the world city network is not consti-tuted as an exclusionary club of the major cities but has numerous linkagesinto regions beyond world cities' (2004: 78). Sassen wishes to sketch a'decentred map of the global economy' (Sassen 2002: 30) in which citiesplace themselves in multiple circuits of globalisation and choose to createnetworked relations to n€ighbouring and distant places that are not centml interms of Western economies. As a result, for both writers, many more citiesare now considered within the rubric of the global- and world-citiesapproaches. Sassen notes, for example, that in relation to producer servicesthrough the 1990s therc has beeD'both consolidation in fewer major centresacross and within countries and a sharp grcwth in the number ofcentres thatbecame part of the global network as countries deregulate thejr economies'(2001a: I l8).

In the context of this wider view of global and world cities, some writershave made a stronger attempt to move beyond using the category of'Thid World city'. However, this has been replaced by a reinscription of thequalities of this calegory onto a certain rurlge of cities, esp€cially Africancities. Taylor (2004) borrows from Jane Jacobs the distinction betwe€n staticand dynamic cities, and draws on the analogy of entrepot cities in colonialcontexts which, he suggests, were parasitic, draining their hinterlands.T

World cities, or a world of ordinary cities? l0'/

Howevet in the contemporary period, such port cities are amongst the largestand most connected ofAfrica's cities and have often been the sites for diversemanufacturing development under both import-substitution arld export-processing economic development regimes. Once again, paying attention tothe diversity of activities and connections that shape cities would be morefruitful than simplistic assessments based on pre-given catego es. It isimportant to recognise that ao absence of outposts of large advanced pro-ducer services firms does not spell the end of the economic road or theabsence of a global function for cities around the world, small or large.

Global- and world-cities approaches, however, have neglected both thewider economy and the role of local and national contexts in shapingthe proc€ss of globalisation. Not only service-sector firms, but also localgovemments, national-policy frameworks, oppositional movements andthe politics of land use all determine the potential for globalisation of cityeconomies.s The complex politics of making global cities bas been the focusof many other wdters (for examplg Abu-Lughod 1999, Hill and Kim 2000,Jessop and Sum 2000, Sites 2003). But the continuing attention paid byglobal- and world-cities authors to APS means that very often the policyimplications oftheir work remain potentially detrimental to the well-being ofboth citizens and wider city economies. Taylot for example, observes thatthere arc different policy consequences for two diFerent types of'wannabeworld cities' that he identified through the data set he was using: inner (mostlysecond cities in European count es) and out€r (mostly 'what used to becalled Third World cities' arld east Europ€an cities):

These are two distinative policy worlds. For Out€r Wannabees, rising upth€ runks of world cities is prima ly a 'development' issug attractingglobal capital to become more central in the world city network. ForInDer Wannabees, rising in world city status is about changing the natureof national city hierarchies in order to come out of the shadow of adominant local world city.

(2004: 160)

Although he and his co-researchers are eager to insist that there is 'no simplehierarchy of world cities' (2004: 165), these policy conclusions operationalisea hierarchical penpective for city competition. However, this is based onassumptions about the consequ€nces of intra-firm relatio[s for inter-citycompetition. The consequences of this anal'.tical slippage for cities are two-fold. First, it reinforces a competitive approach to city development. Whercascross-city collaboration may be the political strategy appropriate to firms thatinvest in many different cities around the world, city managers have only thewell-being of their own city to consider. Attempts to move up the world-cityhierarchy, if understood as the ambitioD of cities rather than as a lesultpurely of the locational strategies of firms, can have severe consequences forthe welfare of the residents of the city concerned. We will explore this further

108 Wo d cities, or a world of orditnry cittes?

below. For those cities on th€ edges of rhe APS nerworks, Peter Hall (2001)suggests that being able to identify other forms of global connectednessthat are important to the city can help to focus development strategies. Butwhile this offers a broader development vision, it still privileges the intercstsand activities of those elements of the city economy that are involved inspecifically global networks - a strategy which, as we will se€ both belowand in the following chapter, can be severely damaging especially for poorercrtles.

In sum, new initiatives to map world and global cities contjnue to entrain aconcern with hierarchy, an emphasis on a small sector ofthe global economy,and to propose (without evidence) a close relationship between intra- andinter-firm nelworks and the identification of hierarchical relations amongstcities. But at the same tim€, both global- and world-cities advocates haveameliomted some of their earlier claims, paying closer attention to a muchwider range ofcities around the globe (but still exhibiting a prejudice againstmany important cities because of the dgid focus on ApS), and entertainingthe possibility of alternative forms of globalisation. However, extendingglobal and world cities analyses in new directions has not solved some of itsfundamental difficulties. In response to the cdtiques and caveats ofworld-cityapproaches, some other writers have thought they might well begin theiranalysis of citjes somewhere else altogether. Amin and Graham (1997)specifically propose that we consid€r the'ordinary city,. Or, as I will suggestrather more shongly, that we allow all cities to be thought of as ordinary.This move is especially important if urban scholars are to move towardsa post-colonial urban theory and to geDerate accounts of cifies and citydevelopment that are relevant to a world of cities, rather than to the limitedprocesses that charactedse 'world cities'.

TH€ (RS€ FOB ORDINRRY (ITI€5

So far in this book I have advanced a number of arguments foi ordinarycities. I have identified the impoftaDce of thinking about cities withoutprivileging the experiences of only certain kinds of cities in our analyses; thevalue of learning how to think difrerently about cities by exploring differentways of life in other cities; and the benefits of a cosmopolitan approach tocities, including attending to the wider circulations and flows that ihape themin order to appreciate the potential creativity and dynamism of all aities- Anumber oftactics - dislocating ethnocentric accounts, deploying comparativeand cosmopolitan approaches have been drawn on to move us towards apost-colonial form of urban theorising. At the same timq they have broughtinto view the ordinary city. Instead of seeing some cities as more advaDcedor dynamic than others, or assuming that some cities display the futuresof othe6, or dividing cities into incommensurable groupings through hier_archising catego es, I have proposed the value ofseeing all cities as oidinary,part of the same field of analysis. The consequence of this is to bring into

World cities, or a world of ordinary cities? 109

view different aspects ofcities than those which are highlighted in global andlvorld cities analyses.

Fi$t, ordinary cities can be understood as unique assemblages of widerprocesses - they are all distinctive, in a category of one. Of course there aredifferences amongsr cities, but I have suggested that these are best thought ofas distributed promiscuously across cities, rather than neatly allocatedaccording to pregiven categories- And even when there are vast differences,between very wealthy and very poor cities, for example, I have suggested thatscholars of these cities have much to learn from one another. This will bediscussed in more detail through Chaptem 5 and 6.

Second, and learning much from global- and world-cities approaches,ordinary cities exist within a world of interactions and flows. However, inplace of the global- and world-cities approaches that focus on a small rangeofeconomic and political activities within the restrictive frame of the global,ordinary cities briDg together a vast arlay of networks and circulations ofvarying spatial reach and assemble many different kinds of social, economicand political processes. Ordinary cities are diverse, complex and intemallydifferertiated.

The consequences of thinking of cities as ordinary are substantial, withimplications fo! the direction of urban policy and for our assessment of thepotential futurcs of all sorts of different cities. Amin and Graham (1997),setting out their account of 'The Ordinary City' suggest that thinking aboutcities as distinctive combinations of overlapping networks of interactionleads very quickly to an account of the capacity of cities to foster creativity-In Western policy circles, they Dote, there has been a rediscovery of'thepowers of agglomeration', aDd an excitement about cities as crcative centres.

A$eeing that many accounts of cities highlight only ce ain elements of thecity (financ€ services, information flows) or c€rtain parts of the city bothleading to a problem of synecdoche they mther descdbe (all) cities as 'theco-prcsence of multiple spaces, multiple times and multiple webs of relations,tying local sites, subjects and fragments into globalising networks of eco-nomic, social aDd cultural change. [. . .] as a set ofspaces where diverse rangesof relational webs coalesce, intelconnect and fragm€nt' (Amin and Graham1991: 411-18).

It is the overlapping networks ofinteraction within the city - networks thatstretch beyond the physical form of the city and place it within a mng€ ofconnections to other places in the world - which, for Arnin and Graham(1997), arc a source of potential dynamism and change- The range ofpotential int€mational or transnational connections is substantial: cultural,politicai, urban design, urban planning, informal trading, religious influ-ences, financial, institutional, interyovernmental and so on (Smith 2001). Tothe extent that it is a form of economic rcductioDism (and reductionism toonly a small segment ofeconomic activity) that sustains the regulating fictionof the global city, this spatialised account of the multiple webs of socialrelations that produce ordinary cities could help to displac€ some of the

110 World cities, or a world of ordinary cities?

hierarchising and excluding effects of this approach. For with so many dif-ferent processes shaping cities aIId so many potential interactions amongstthem, it would be dimcult to decide agairrst which cdteria to raise a judge-ment about rank.

Categorising cities and carving up the reaim of urban studies has hadsubstantial efects on how cities around the world are understood and hasplayed a role in Jimiting the scope of imagination about possible futures forcities. This is as true for cities declared 'global' as for those that have fallen offthe map of urban studies. The global-cities hypothesis has described citiessuch as New York and London as'dual cities', with the global functionsdrawing iD not only a highly professional and well-paid skilled labour force,but also reJying on an unskilled, very poorly paid and often immigrant work-force to service the global companies (Sassen 1991, Allen and Henry 1995),These two extremes by no means capture ths range of employment opportun-ities or social circumstances in these cities (Fainstein et aI. 1992, Buck et al.2002). It is possible that these cities, allegedly at the top of the global hier-archy, could also benefit frcm being imagined as 'ordinary'. The multiplicityof economic, social and cultural networks that make up these cities couldthen be drawn on to imagine possible paths to improving living conditionsand enhancing economic growth across the whole city.

In this regard, Michael Storper (1997) has focused on the economiccreativity of urban agglomerations in his description of the 'reflexive city,.Whereas global city analyses explore the impo ance of colocation ofspecialised activities involved in the control and command of the globaleconomy in small areas of the city, Storper generalises the need for ,proxim-ity' in economic interactions to cement relations of trust amongst complexorganisations and between both individuals and organisations. Storper seesthe city generally as providing a key context for these social interactions, socrucial to the 'untradeable' and 'tacit' elements of economic life. However,mther thar being limited to a focus on the workings of single industry prc-duction complexes or production chains, social interaction or reflexivity is agenenlised possibility in ciry life. He suggests then that we think of ,the

economies of big cities [. . -] as sets ofpartially overlapping spheres ofreflex-ive economic action [. . ,] (including) their conventional and relational struc-tures of co-ordination and coherence' (1997 245). Citi€s, then, remainattractive locations for business activity across a range ofsectors and offer anenvironmenL that enables economic producrion and innovation. This is tomake a case for the broad economic potential of all cities. (Chapt€r 6 willexplore the consequences of an ordinary cities appro4ch for understandtngurban economic development in some detail.)

Ordinary cities, then - and that means all cities - are understood to bediverse, creative, modern and distinctive with th€ possibility to imagine(withilr the not-inconsiderable conslraints ofcontestations and uneven powerrelations) their own futures and distinctive forms of cityness. And theie areimportant consequ€nces of this approach for ways of thinking about the

Wo d cities, or a world of ordinary citips? lllfutures of cities. That alterlative approaches to urban policy are neededis evid€nt when the political consequences of global- and world,citiesapproaches ar€ conside.ed. These are especially onerous for the many poorcities that do not qualify for global- or world-city status. Often such cities arecaught within a very limited set of views of urban development:e betweeDfinding a way to fit into globalisation, emulating the appareDt success€s of asmall runge of citi€s and, as we will see in the following chaptef embarkirlgon dev€lopmentalist initiatives to redress poverty, maintain infrastructureand ensure basic sewice delivery- Neither the costly imperative to go globalnor developmentalist interentions that build towards a certain vision ofcityness and that focus attention on the failures of cities arc very chresources for city planners and managers who are confronted with thedemands of a complex, diverse, ordinary city. Global- and world-citiesapproaches, once translated into a policy or political terain, can have someserious, perhaps unforeseen consequenc€s.

In policy terms, the bierarchies and categories embedded in the global- andworld-cities approaches suggest that if cities are not to remain incon-sequential, marginalised and impoverished or to trade economic growth forexpansion in population, they need to aim for the top! Global City as aconcept becomes a regulating fiction. It offers an authorised image of citysuccess (so people can buy into it) that also establishes an end point ofdevelopment for ambitious cities. There are demands, from Ista[bul (Robinsand Askoy 1996) to Bombay (HaIIis 1995) to be global. As Douglass (1998:lll) writes, 'world cities are the new shibboleth of global achievement forgovemments in Pacific Asia' (see also Douglass 2000, Olds and Yeung 2004)_But, as a number of authors have noted, calculated atterDpts at world-o! global-city formation can have devastating cons€quences for most peoplein the city, especially the poorest, in terms of service provision, equalityof access and redistribution (Bemer and Korff 1995, Robins and Askoy1996, Douglass 1998, Firman 1999). Global- and world-cities approachesencoumge an emphasis on promoting economia relations with a globalreach and pdo tising certain prominent sectols of the global econorny lordevelopmeDt and investment. Alternatively, the policy advice is for citiesto assume and work towards achieving their allocated 'place' within thehierarchy ofworld cities (Taylor 2001).

Most cities in poorer couotries would find it hard to reasonably aspire tooff'ering a home for the global economy's command and contrcl functionswhich Sassen identifies as concentrated in certain global cities Although,as Tyner (2000) argqes, differ€nt aspects of the global economy requirecoordination and organising and some of these activities are concentrated incities that arc not usually labelled as global. Manila, for examplg has a con-centration of agencies and institutions that facilitate the movement of low-paid migrant labour to wealthier countries. More feasible for many poorercities is to focus on some of the other 'global functions, Sassen associateswith global cities. These ilclude promoting attractive 'global, tourisr

ll2 World cities, or a world of ordinary cities?

environments, even though these have nothing of the locational dynamics ofcommand-and-control global-city functions. Disconnected from the concen-tration of arts and culture associated with employment of highly skilledprofessionals in global cities, the impulse to become global in purcly touristterms can place a city at the opposite end of power relations in the globaleconomy, while substantiaily undermining provision ofbasic services to localpeople. (Robins and Askoy (1996) discuss this in relatioo to Istanbul.) Inaddition, expo -processing zones (EPZs) may be 'global' in the sense thatthey are 'transnational spaces within a national territory' (Sassen 1994: 1),but they too involve placing the city concemed in a relatively powerlessposition within the global economy, which is unlikely to be the city's only orbest option for future growth and development (Kelly 2000). These are notplaces from where the global economy is controlled: they are at quite theother end of the command-and-control continuum of global-city functions.More than.that, the reasons for colocation would rot involve being able toconduct face-lo-face meetings to foster trust and co-operation in an innova-tive ellvironment, Rathet they are to ellsut€ participation in the relaxationof labour and environmental laws which are on offer in that prescribedarea of the city. Cities and national governments often have to pay a highpric€ to attract these kinds of activities to their territory. Valorising ,global,

economic activities as a path to city success - often the conclusion ofa policyreversioning ofworld-cities theory - can have adygl5g qarraauences for localecoDomies jn poor countdes.

Nevertheless, global cities have become the aspintion of many citiesaround the world, sprawling and poor mega-cities the dangerous abyss intowhich they might fall should they lack the redeemirg qualities of citynessfound elsewhere. This may not have been the intention of urban theorists,robut ideas have a habit of circulating beyond our control. It is my contentionthat urban theory should be encouraged to search for alternative formula-tions of cityness which don't privilege only certain cities, plac€d at the topof a hierarchy. Ideas about what cities are and what they might become needto draw their inspiration from a much wider range ofurban contexts.

A question that writers about cities in peripheral areas pose, looking at thistheory from off the map, is how to distinguish cities they know from thosethat can be identified as 'world' cities. This leads quite quickly to asking howcities get to be world cities, as Alan Gilbert (1998: 178) puts it, ,So whattransforms an ordinary city into a world city?' But as Mike Douglass (199g,2000) writes, and Olds and Yeung (2004) concur, there is little explanarion inthis literatur€ for 'world city formation' - or for how cities become worldcities. Douglass (1998) reminds us that this is a highly contested process withprofound consequences for the built environment of cities and for the well-being of citizens. Whereas the emphasis of the world-cities approach hasbeen on und€rstanding the 'structural' positions of cities, taking a vi€wfrom off the map of this approach draws attention to political actors andinstitutions as active agents making the world cityness of cities (Machirnura

Wotld cities, or a world of ordinary cities? II3

1998, Abu-Lughod 1999, Douglass 2000, Varsanyi 2000, Lipietz 2OO5). Theseprocesses of world-city formation are perhaps more relevant to cities definedoff the map of world cities, but eager to make their way onto it. And theprocesses involved in making world cities are usually not very progressiveprocesses and are often at odds with promoting cities that are good to live itr.They have been much discussed elsewhere in urban studies and include place-marketing, tourist promotion, subsidies to attract productive enterprises,costly remaking of the urban environment, all relying on often destructiveforms of competition between cities and the emergence ofcopy-cat forms ofurban entrepreneurialism (Logan and Molotch 1987, Haryey 1989, Bernerand Korff 1995, Hall and Hubbard 1998, Jessop ard Sum 2000, Beauregardand Pierrc 2000).

Cdtically evaluating these world-city-making processes and incorporatiflgthem into their explanatory frameworks and empirical research they arenotably absent from the key studies withir the field, for example Sassen(1994, 2001a) and Taylor (2004) - could help to sustain the critical edge oftheworld-cities approach and also to ensure that it remains a 'heuristic' ratherthan categorising device (Friedman 1995b)- A greater emphasis on processrather than assigning cities to a category would certainly enable the world-cities approach to be more applicable to cities currently left offits maps. Butit might also lead us to dismiss the activity of categorising cities and thecategory ofworld cities altogeth€r. Instead, we might be encouraged to widerthe range of processes considered rclevant to understandiog the futureofcities, both geographically and functionatly (Smith and Timberlake 1995).As we have seen here, the motivatiors for doing so are not simply int€llec-tual accuncy or elegance; there are serious political consequences tointerventions iNpired by global- altd world-cities analyses.

To aim to be a'global city' in the formulaic sense may well be the ruin ofmost cities, Policy-rnakers need to be offered altemative ways of imaginingcities, their distinctiveness and their possible futures. A stronger focus on thepolitics of urban development initiatives, as suggested by scholars of citiesoff the world-cities map, would expose the range of interests that find ituseful to harness the global- and world-cities analyses to their ambitions. Itwould also bring into view the diveffity of interests which are available tocontest and shape the future ofcities. In ordinary citiel it is this diversity - ofpolitical interestq social relations and economic activities - that can formthe basis for an alternative view of cities and thet futures. So, rather thandevelop a regulatory fiction of the powerful global city, an ordinary cityperspective will start from the assumption that all cities can be thought of asdiverse and distinctive with the possibility to imagine (within the constraintsofcontestations and uneven power relations) their own futures and their owndistinctive forms of cityness.

174 Wotld cities, or a wo d of oftlinarJ cities?

CONCTUSION

Global- and world-cities analyses have been enormously productive inrefocusing urban studies on the wider processes and networls that shapecities; and they have a[nou[ced a new, more inclusive geography of therole of cities in globalisatior. But rhey have left intact earliei aisumprionsabout hierarchical rclations amongst cities, with potentially damagingconsequences especially, but not only, for poorer cities. They have, in fact,consigned a large number of cities around the world to theoretical iffele_vance. Building on global- and world-cities approaches, but mindful of thesecriticisms, other write$ have turned to the ordinary city _ diversg contested,distjnctive as a better starting point lor understanding a world ol cities.Ordinary cities also emerge from a post-colonial critiqui of urban studiesand signal a new era for urban studies research charicterised by a morecosmopolita! approach to understanding cityness and city futur€s. This canunderpin a field of study that encompasses all cities and ihat distributes thedifferences amongst citi€s as diversity rather than as hierarchical categories.It is the ordinary city, then, that comes into view within a postcolonllsedurban studies.

More than this, the overlapping and mulriple networks highlighted in theordinary city approaches can be drawn on to inspire altern;tiv; models ofurban development. These would be approaches that see the potential forproductive connections supporting the diverse range of econo;ic activitieswith varying spatial reaches that come together in cities. Approaches thatexplore the diversiry of economic acrivjties presenr in any iordinaryt ciry(Ja€obs 1965: 180-81) and that emphasise the general creative potential ofallcities could help to counter those that encourage policy-makirs to supoortone (global) seclor ro rhe derriment oforher{. The^ following two chapters explore some of the arguments and founda_

tions for development ilterventions and theoretical pirp."tiu", tfr"i*lgl,tbe more appropriate to ordinary cities with diverse economies. Cfr"p,". Stlrns to a body of research and policy_making inspired by the problim ofdeveloping poorer cities. In some ways, the mirror i-ug" of ffoUul_

-uoa

world-cities approacheg developmentalist approaches to- citl"s-irrspi."- anemphasis on the poorest parts of cities, thosl lacking in much of th'" bas,"infrastructure taken for graDted in wealthy urban co-ntexts una if,or" *,tf,little in-the way of cutting-edge formal global business. To d;lo;;post_colonial account ofordinary ci!ies, the divide between gtotut-

"rrJlriorta'_crtyanalyses and developm€ntalist urban approaches needl to be breach"J; inediversiry and complexity of ordinary ciries needs ro be brough, i";;;i;;

Both of the_remaining chapters in this book explore howichofui, ,oiti"gon poorer and wealthier cities have much to learn from each other- I exflicittyset out to_inspire a more cosmopoliran forrn of theorising: one thalir;;ksacross different kinds of cides and that, like the invendie una innou"t,u"urbanisms explored in Chapter 3, learns promiscuousfy f.o_ u- *.rg""of

Wotld cilies, or a world of ordinary cities? 115

contexts. Thus I also insist that imagiling the futures of ordinary citiesrequires more than a developmentalist approach, just as it requires more thana global- or world-cities approach. These currently divided areas of urbanscholarship need to be brought together, in both theory and policy, if urbanstudies is to find the resources to addr€ss the challenges of an urban future;the challenges ofa world of(ordinary) cities.