Current Debates in Urban Studies - Storper and Scott

Article

Urban Studies2016, Vol. 53(6) 1114–1136� Urban Studies Journal Limited 2016Reprints and permissions:sagepub.co.uk/journalsPermissions.navDOI: 10.1177/0042098016634002usj.sagepub.com

Current debates in urban theory: Acritical assessment

Michael StorperDepartment of Geography and Environment, London School of Economics, UK; Luskin School of

Public Affairs, UCLA, USA; Centre pour la Sociologie des Organisations, Sciences Po/Paris, France

Allen J ScottDepartment of Geography, University of California at Los Angeles, USA

AbstractUrban studies today is marked by many active debates. In an earlier paper, we addressed some ofthese debates by proposing a foundational concept of urbanisation and urban form as a way ofidentifying a common language for urban research. In the present paper we provide a brief recapi-tulation of that framework. We then use this preliminary material as background to a critique ofthree currently influential versions of urban analysis, namely, postcolonial urban theory, assemblagetheoretic approaches and planetary urbanism. We evaluate each of these versions in turn and findthem seriously wanting as statements about urban realities. We criticise (a) postcolonial urban the-ory for its particularism and its insistence on the provincialisation of knowledge, (b) assemblagetheoretic approaches for their indeterminacy and eclecticism and (c) planetary urbanism for itsradical devaluation of the forces of agglomeration and nodality in urban-economic geography.

Keywordsagglomeration theory, assemblage theory, planetary urbanisation theory, post-colonial urbanism,urban theory

Received November 2015; accepted January 2016

Urban challenges and urbantheory in the 21st century

The current period of human history canplausibly be identified not only as a globalbut also as an urban era. This is a period inwhich population, productive activity andwealth are highly and increasingly concen-trated in cities.1 Most cities offer a betterstandard of living for more people than everbefore in human history; even the urban

poor are better off, on average, than therural poor around the world. Cities are pri-mary centres of scientific, cultural and socialinnovation (Glaeser, 2011; Hall, 1998).Cities have also proliferated all over the

Corresponding author:

Michael Storper, Department of Geography and

Environment, London School of Economics, Houghton

Street, London, WC2A 2AE, UK.

Email: [email protected]

at UNIV DE LOS ANDES on April 6, 2016usj.sagepub.comDownloaded from

http://usj.sagepub.com/

globe and have become increasingly interde-pendent so that where once we could speakquite meaningfully of ‘national urban sys-tems’ (most extensively developed in theGlobal North) the current situation is onemarked by an increasingly integrated world-wide network of cities together with anextraordinary surge of urban growth in theGlobal South (McKinsey, 2011). But this erais also in some ways a dark age as marked bygutted-out old industrial cities, concentratedpoverty, slums, ethnic conflict, ecologicalchallenges, unequal access to housing, gentri-fication, homelessness, social isolation, vio-lence and crime and many other problems.There has been a corresponding proliferationof academic and policy-related research oncities and a vigorous revival of debates aboutthe content and theoretical orientation ofurban studies.

In this paper we discuss three currentlyinfluential perspectives on these debates,namely, postcolonial urban analysis, assem-blage theoretic accounts of the city, and thetheory of ‘planetary urbanism’. In their dif-ferent ways, each of these three bodies ofwork attempts to provide bold understand-ings of the empirical trends referred toabove. At the same time, each of them seeksto present an account of the city that posesstrong challenges to much if not mosthitherto existing urban theory. As such,these perspectives are prominent expressionsof a renewed vibrancy and innovativeness inurban studies – reflecting the dramaticallyshifting geographies of urbanisation notedabove – but in ways, as we shall argue, thatoften appear to be highly problematical. Itshould be stressed, at the outset, that thesethree bodies of work have different points ofintellectual origin and different points ofemphasis, though postcolonial andassemblage-theoretic approaches do sharesignificant conceptual common ground,notably their focus on particularity, localism

and difference, and an insistence on theempirical ‘complexity’ of socio-spatialarrangements. Planetary urbanism for itspart concentrates on an attempt to reformu-late the relationship between ‘concentrated’and ‘extended’ forms of human settlement,land use and spatial development by assimi-lating both of them into a theoretical urbanlandscape that is nothing less than global.

We will question these three approachesin a variety of ways. We will argue that eachof them contains major blind spots and ana-lytical distortions and that each has failed tooffer a meaningful concept of urbanisationwith generalisable insights about the logicand dynamics of cities. These weaknesses arenot only regrettable in their own right, butare notably disabling in a field where theneed to frame viable policy advocacies insearch of social justice has become more andmore insistent. In addition, we will arguethat much of the current literature associatedwith these three approaches shares a predi-lection for certain kinds of convoluted philo-sophical and epistemological abstractionsthat actually present barriers to any under-standing of the urban as a concrete socialphenomenon. We begin our discussion bybriefly re-stating ideas developed in an ear-lier paper (Scott and Storper, 2015) wherewe seek to establish a foundational conceptof the urban. On that basis, we claim thatthere are fundamental common genetic fac-tors underlying urban patterns, and a robustset of conceptual categories within whichurbanisation processes and urban experi-ences can be analysed, wherever they mayoccur in the world. We then proceed to dis-cuss in some detail what we take to be themost egregious weaknesses of the three maintargets of our critique. As we work throughthis agenda we also offer a few replies –though less than a complete response – to anumber of critical assessments of our earlierpaper.

Storper and Scott 1115



The nature of cities revisited

Towards an analytical understanding of thecity

In Scott and Storper (2015) we dwelt in parton the high levels of diversity and disagree-ment in urban theory over the last century orso, and we asked if a coherent, stable theoryof the city could be constructed. Such a the-ory, if it were possible, would need (a) toaccount for the genesis of cities in general,(b) to capture the essence of cities as concretesocial phenomena and (c) to make it possibleto shed light on the observable empiricaldiversity of cities over time and space.

Our approach to this theory-constructionchallenge was to build on the observationthat cities are everywhere characterised byagglomeration involving the gravitationalpull of people, economic activities and otherrelata into interlocking, high-density, nodalblocks of land use. The primary, but by nomeans the only mechanism driving this fun-damental tendency, we argued, is the emer-gence of organic divisions of labour in whichsocial and economic life (i.e. the productionof goods and services, but also including cul-tural, religious and governmental pursuits) isorganised and reorganised within networksof specialised but complementary units ofhuman activity. This form of organisationmeans, in turn, that mutual geographicalproximity or agglomeration of these units iscrucial, for otherwise the time and distancecosts of interaction would impede theiroperational effectiveness. In our earlierpaper, we argued at length that all citiesthroughout history are based on this funda-mental process of agglomeration The costsof covering distance were no doubt muchhigher at earlier periods of history, but asthe copious literature on agglomerationdynamics reveals, proximity through co-location is imperative for certain types ofactivities even today (Cooke and Morgan,1998; Fujita and Thisse, 2002; Krugman,

1991; Scott, 2012; Storper, 2013). A furthermajor point must now be made to the effectthat since the interdependent specialisedactivities that constitute the division oflabour (and the residential housing associ-ated with them) cannot all occupy a singlepoint, they must necessarily sort themselvesinto a spatially extensive lattice or patch-work organised around their common centreof gravity and characterised by intricateinternal patterns of geographic differentia-tion. We call any system of this sort an urbanland nexus (cf. Scott, 1980).

These trans-historical and trans-geographical urban processes take on spe-cific concrete attributes that reflect the wider– and ever changing – social, economic andpolitical conditions within which urbanisa-tion is always embedded. We can identifyfive basic variables or forces that shape theprincipal variations of the urban land nexusat different times and places. These can beenumerated as (a) the overall level and modeof economic development, (b) prevailingresource allocation rules, (c) forms of socialstratification, (d) cultural norms and tradi-tions and (e) relations of political authorityand power. We do not have the space hereto work out even a schematic description ofthe empirical diversity that these (and other)contextual variables are capable of generat-ing, but they lead to a great deal of detailedvariation in the urban land nexus from oneinstance to the next. For example, ImperialRome, Xi’an in China, ancient Babylon,Timbuktu in the Empire of Mali,Tenochtitlan in 15th century Aztlan (con-temporary Mexico), Manchester in theindustrial era in Britain and Los Angeles,Mexico City and Hong Kong in the 21stcentury are all quite different from oneanother at one level of analysis, even as theyall share in a common set of fundamentalgenetic forces. In view of the play of thesedifferentiating variables, and notwithstand-ing our theoretical generalisations regarding

1116 Urban Studies 53(6)



the urban land nexus, we disagree with Dickand Rimmer (1998) who state that cities invarious far-flung parts of the world are nowconverging towards a standard template.For the same reason, we also reject theclaims of Roy (2015) when she describes ourearlier paper as an attempt to construct auniversal history whose objective is to oblit-erate ‘historical difference’.

Not only does our analysis provide uswith the tools for distinguishing between thegeneral and the particular in urban out-comes, but also for separating out that whichis distinctively and inherently urban from therest of social reality. In particular, we mustdistinguish between phenomena that occurin cities but are not generated by urbanisa-tion processes as such, and phenomena thatare legitimately elements of cities in the sensethat they play an active role in defining theshape and logic of urban outcomes. Thus, ahospital located in an urban area will usuallyplay an important role as an element of theurban land nexus, both as a specific kind ofservice provider and as a catchment pointfor those who use its services, but its internaladministrative arrangements are not likely tobe of much relevance to any understandingof the city. Similarly, the interest rate, ideol-ogies of imperialism or the price of sugar arenot intrinsically urban; or rather, they can besaid to have urban significance only insofaras they can be shown to play some role inthe dynamics of the urban land nexus. A fur-ther illustration of these remarks is offeredby the phenomenon of poverty, which hasimportant urban dimensions but also hasmany substantive and relational manifesta-tions that are not generated by the urban assuch. To state this in another way, measure-ments of inequality or poverty in cities arenot equivalent to the claim that inequality orpoverty are basically engendered by cities(Sampson, 2012). In capitalist or marketeconomies especially, poverty is not funda-mentally caused by urban processes, but by

the complex forces that shape income distri-bution in an economy marked by privateproperty, competitive markets and wagelabour. Equally, although researchers oftenuse urban entities as units of observation invarious kinds of statistical exercises (just aswe use counties, states or countries for thesame purpose), this alone does not endowthese exercises with intrinsically urban mean-ing. The claim that any phenomenon occur-ring in a city is urban by nature is – withoutfurther specification – liable to the error ofecological fallacy. Political outcomes in thecity, too, need to be carefully scrutinised inorder to distinguish the specifically urbanfrom what is merely contingently so. In par-ticular, the urban land nexus is by its verynature subject to peculiar and endemic formsof politicisation. The tensions created bycompetition for land uses, the urge to secureaccess to positive externalities and to avoidthe effects of negative externalities, the rent-seeking behaviour of property owners andthe need to protect or enhance certain kindsof urban commons (such as agglomerationeconomies), among other frictions, all createconstantly shifting circles of urban social col-lisions. Urban governance arrangements,too, or what Molotch (1976) called the urban‘growth machine’, are in significant wayscaught up in these frictions through theirfunctions as suppliers of public goods andservices and their role as mediators of urbanconflicts.

From these comments it follows (and eventhough we have affirmed that we live in anurban era in the sense that cities formallyrepresent the principal geographic containerswithin which contemporary human societyunfolds) that not all aspects of life, perhapsnot even most aspects, can be understood asbeing necessarily (that is, ‘ontologically’)urban phenomena in the very specific mean-ing as identified here. For these reasons, too,we are reluctant to accept Lefebvre’s (1970)proposition that we are evolving in the




direction of a full-blown ‘urban society’ withits implied sub-text to the effect that societyand the city are becoming one and the samething. Similarly, the remark by Taylor (2013)that cities constitute the essential motors ofall human society, politics and economythroughout history, and that hence all socialscience must become ‘city-centric’, is clearlyexaggerated. With a conscientiously delim-ited and focused concept of the city it is pos-sible to identify how the urban generatesspecific kinds of social phenomena and setsthem apart from non-urban phenomena.This is what provides a distinctive place forurban analysis in the academic division oflabour and what, together with an appropri-ate analytical machinery, endows it with acentral mission.

Some practical and theoretical implications

The urban land nexus emerges in the firstinstance out of dynamics of agglomerationand accompanying processes of land-usesorting, thus generating a complex lattice oflocational activities over a shared gravita-tional field. In capitalist systems, significantparts of the urban land nexus are subject tothe rule of private property and are hencecommodified. In other types of social sys-tems, land use decisions are apt to bedirected by different kinds of mechanismsinvolving, say, limited or non-existent indi-vidual property rights or communal regimesof ownership (such as ethnic or clan rule).

Whatever the system, individual units ofland ownership always have more than apurely private, atomised dimension. Morespecifically, agglomeration, proximity anddensity result in many different kinds ofexternalities (positive and negative) that cir-culate through the urban land nexus so thatland use at one location invariably hasimpacts on other locations. Positive out-comes from agglomeration include processesof sharing (e.g. the joint usage of large-scale

infrastructural artifacts), matching (e.g. thelocal availability of many alternative choicesto purchasers and sellers of goods, servicesor labour) and learning (e.g. the rapid diffu-sion of cultural or technological informa-tion), which in part account for thedynamism we typically associate with citiesthroughout history and especially in capital-ism (cf. Duranton and Puga, 2004).Negative outcomes may include the conges-tion, land use incompatibilities, incentives tocrime, segregation and inequality, socialconflicts and other undesirable consequencesthat arise out of the dense coexistence ofhighly differentiated social and economicactivities in a relatively restricted spatialorbit. The importance and pervasiveness ofthese effects means, as already suggested,that some form of collective, non-individualcontrol is necessary if the city is both toavoid internal blockage and if the individu-als, households and firms that it containsare to seize jointly on strategic developmen-tal opportunities. This explains in largedegree why the individual decision-makingand behavioural mechanisms of the urbanland nexus are virtually everywhere regu-lated by collective governance arrangementsdesigned to safeguard cities against implo-sion and stagnation (Roweis and Scott,1977). Within the city, interrelated units ofeconomic production typically form distinc-tive clusters interpenetrated by swaths ofresidential activity. Areas outside the cityare sources of the food, resources and mate-rials that are not produced internally; andthey offer, in addition, markets for the city’stradable, specialised products. These areasare represented both by the immediate hin-terland of the city and other cities andregions at more distant locations. Even inancient times, long-distance trade was char-acteristic of many cities, as exemplified mostdramatically by the case of Classical Rome.In the 21st century, cities interact with oneanother in a globally-integrated system of




trade and information exchange as expressedin an emerging global mosaic of cities andcity-regions.

In the light of these remarks, we can nowstate that the city represents a very specificscale of economic and social interaction gener-ated by agglomeration processes and focusedon the imperative of proximity, and almostalways endowed with governance arrange-ments that attempt to deal with the problema-tical effects of density and propinquity. At thesame time, the city is always embedded in afar-flung spatial economy that sustains itwithout compromising its integrity as a dis-tinctive social phenomenon (Fujita et al.,1999). Accordingly, as we shall argue inmore detail later, it cannot simply be dis-solved away by fiat into a sort of overarch-ing global plasma as theorists of ‘planetaryurbanization’ proclaim (e.g. Angelo andWachsmuth, 2015; Brenner and Schmid,2015). Our argument thus goes stronglyagainst the grain of the main theses of pla-netary urbanists or those who, like Aminand Thrift (2002), claim that ‘the city iseverywhere and nowhere’. A fortiori westand in opposition to those urbanists whostate that the idea of the city is purelyideological.

This is also an appropriate moment toallude to some of the criticism that has beenmade of our earlier analysis on the groundsthat it is ‘economistic’ (Mould, 2015; Roy,2015). Given the primary role that weascribe to economic forces in the genesis ofthe urban land nexus, this line of critique isentirely predictable but essentially misin-formed. We assuredly do propose that theorigins of the urban land nexus reside in theeconomic tensions engendered by the divi-sion of labour and agglomeration (and weoffer strong justifications for this position),but our claim is very far indeed fromany argument to the effect that cities areexclusively or monocausally structured by

economic variables. Indeed, we have expli-citly suggested that diverse other social, cul-tural and political forces are also at work inshaping the urban land nexus. Accordingly,our response to the charge of economism istwofold. On the one hand, we invite our crit-ics to identify exactly what it is that theymean by ‘economism’ (a term that is almostalways vacuous in actual usage2). Our ownsuggestion here is that the most demandingsense in which the term can be used is toreserve it strictly for situations where thatwhich is not economic is erroneously pro-claimed to be economic (e.g. claims to theeffect that the level of economic develop-ment determines the form of sociability inurban neighbourhoods or that the city isnothing but an economic phenomenon). Onthe other hand, we challenge our critics’attempts to characterise our work as econ-omistic by asking them to go beyondpurely gestural allegations and to demon-strate in disciplined critical detail how ourformulations about the analytical originsof cities might actually be wrong and howthey can be corrected. In fact, a close read-ing of our text should make it abundantlyclear that our theory of the urban landnexus remains open to an enormous diver-sity of non-economic elaborations andhybridisations, and, indeed, to any numberof complex reflexive relations between theeconomic and the social, political and cul-tural dimensions of urban life. Moreover,although this point surely should not needto be made explicit, the urban land nexusobviously is a fundamental key to under-standing the city as a locational matrix ofbuilt forms and associated symbolic assets,which, according to the views of Walker(2016), are culpably absent from our ownanalysis.

This brief exposition of our theoreticalviews now serves as a point of referenceagainst which we will review and criticise a




number of currently fashionable theories ofurbanisation that we take to offer seriouslyflawed accounts of both the scientific andpolitical challenges posed by cities today.

Postcolonial urbanism:Cosmopolitan but provincial

Much contemporary postcolonial researchoriginated in cultural and historical studieswhere it has functioned as a critique ofnumerous blind spots in Northern traditionsof theoretical analysis. Above all, postcolo-nial thinking, as represented, for example,by Said (1978) and Spivak (2008), demon-strates how diverse intellectual legacies ofcolonialism (ethnocentric biases and preju-dices in particular) enter unconsciously intoscholarly writings about the Global South.Postcolonial scholars (such as Comaroff andComaroff (2012)) are also, and correctly,intent on showing that the claims of univers-ality that Euro-American theory has oftenarrogated to itself are sometimes demonstra-bly false. These same lines of thinking andcritique have recently become strongly influ-ential in urban studies. Robinson (2006,2011) and Roy (2009, 2011), among manyothers (for example, Edensor and Jayne,2012; Myers, 2014; Ong and Roy, 2011;Patel, 2014; Sheppard et al., 2013), havebeen notably vocal in this regard, and havebeen especially outspoken in decrying theapplication of urban theories constructed inEurope and North America to cities in theGlobal South.

These and other analysts have sought tocorrect what they see as imbalances and mis-representations in Northern urban theoriesby means of two overlapping strategies. Oneis to call for more cosmopolitan forms ofurban theory (what Ong and Roy (2011)refer to as ‘worlding’) that take seriously theexperiences of the cities of the Global South.The other is to insist on the irreducible coreof idiosyncrasy that marks every city and to

focus on the resulting play of empirical ‘dif-ference’ and ‘complexity’. A further impor-tant point of departure for postcolonialurban scholars resides in the notion of the‘ordinary city’ developed by Amin andGraham (1997) to the effect that cities areall equally distinctive and unique and thatnone can be claimed to function as a privi-leged archetype or exemplar relative to theothers. Robinson (2006), in particular, hasappealed to this notion by way of assertingthe equivalent standing of all urban centresacross the North-South divide, as well as byway of proclaiming that any meaningfulproblematic of the urban must focus intentlyon the essential character of cities as sites ofdifference. In a more radical vein, Roy(2009: 820) has advocated sweeping much ofextant urban theory away with the peremp-tory injunction that ‘the center of theory-making must move to the Global South’.However, as Peck (2015) points out, there isan apparently unresolved tension in postco-lonial studies between constant calls for aworlding of urban analysis on the one sideand the equally constant affirmation of aNorth/South binary on the other, and even,in some cases, as we shall see, a tendency tofavour a wholesale ‘provincialization’ ofurban theory (Ren and Luger, 2015).

Postcolonial commentators are especiallydissatisfied – not always incorrectly but fre-quently without appropriate nuance – withwhat they allege to be the pervasive moder-nist and developmentalist biases of urbantheory as elaborated in the Global North.One of the most baleful cases of this kind ofbias, in the view of these commentators, isrepresented by the Chicago School of UrbanSociology. A particular point of contentionis the Chicago School’s notion of the folk-urban continuum comprising primitive, non-urban social formations on the one side, andadvanced, urbanised social formations onthe other, and the extension of this notion inthe work of Wirth (1938). Postcolonial




urban theorists criticise modernism-developmentalism as a discourse that con-signs the cities (and societies) of the GlobalSouth to the status of underdevelopmentand backwardness, an outlook that is mani-fest, according to Roy (2011: 224) in ‘apoca-lyptic and dystopian narratives of the slum’.She herself sees the poverty, informality,marginalisation and extensive slums ofSouthern cities as a mode of urbanisation(Roy, 2005; emphasis in the original). Quitewhat this phrase might mean is difficult todetermine, but it presumably functions as agesture intended to eliminate the allegedlypejorative implications of Northern theory.Modernism-developmentalism is further cri-ticised by postcolonial scholars for its pro-motion of a teleological concept of cities inthe Global South in which growth andchange are alleged to be subject to evolu-tionary stages involving shifts from less tomore modern and developed. The more spe-cific claim here is that it is unreasonable toexpect any linear movement from less formalto more formal arrangements in regard tosettlement-building and property rights inthe cities of the Global South (Roy, 2005)

The critique of postcolonial urban studies

Obviously, cities of the Global South havebeen severely overlooked in past researchefforts;3 obviously we must be careful to payattention to the specificities of these cities;and obviously we need to acknowledge thaturban theory must now range over the entireworld for its sources of data and evidencewhile remaining fully open to new concep-tual insights generated out of the experiencesof the cities of the Global South. Equallyobviously, we must beware of the dangers ofEurocentrism, by which we mean theoreticaloverreach based on limited evidence derivedfrom Northern cities, but that is inappropri-ate or irrelevant with respect to Southern cit-ies. Where postcolonial urban theory errs,

we argue, is in its own peculiar forms of crit-ical overreach and its overall commitment towhat we have called a ‘new particularism’(Scott and Storper, 2015). In what now fol-lows, we address what we take to be threemajor failures of postcolonial urban theory,namely, its exaggerated complaints regard-ing Euro-American epistemological bias incontemporary urban analysis, its highlyselective critique of modernism-develop-mentalism and its strong methodologicalcommitment to theoretically-unstructuredcomparativism. Note that all of these themesare essentially branches of a single meta-claim, that of a set of incommensurabilities:in point of view, in development and inrepresentativeness.

Eurocentrism and the provincialisation ofknowledge. To begin, then, postcolonialurban studies are broadly motivated by theclaim that theory produced in the GlobalNorth is inescapably unable to account forempirical situations in the Global South (seePeck (2015) for an analogous characterisa-tion of postcolonial theory). Roy (2009)adds the further damning claim that Euro-American urban theory ‘. keep(s) alive theneo-orientalist tendencies that interpretThird World cities as the heart of darkness,the Other’ (though we can think of no scho-larly paper on cities published in at least thelast half-century that would bear this asser-tion out). In harmony with these judgments,many urban theorists with a postcolonialbent (notably Sheppard, 2014; Sheppardet al., 2013) state that theories must necessa-rily be local and confined in their empiricalreach to specific segments of geographicreality. As Leitner and Sheppard write(2015), ‘Our position, then, is that there canbe no single urban theory of ubiquitousremit’. Even though the authors fail todefine what they mean by ‘different con-texts’ and how we might identify them, theythen go on to call for self-conscious




‘provincialization’ of urban theory as a vir-tue in itself and as a way of delegitimisingwhat they see as the pervasive pretensions touniversalism of European and Americanurban theory.

A first direct and simple answer to this callto provincialise theory is to ask for a clearand direct demonstration of the fundamentalincommensurability of urban phenomena indifferent parts of the world, above andbeyond assertions about empirical diversity.A second is to propose a counter-argument,as we have done (and which we offer for dis-confirmation), to the effect that there areindeed theoretically generalisable features ofurbanisation as a whole. Of course, we knowfrom the work of Livingstone (2014) and oth-ers that theoretical work very often doesunconsciously reproduce geographical andideological biases reflecting the circumstancesin which it arises, and urban theory is noexception to this observation. Moreover, var-ious streams of philosophy and historiogra-phy, most especially since the middle of the20th century, clearly recognise the social con-structivist character of all intellectual activity(Haraway, 1988; Kuhn, 1962; Mannheim,1952). This work points not only to the essen-tial social and historical foundations of allforms of discourse, but also to the absence ofany Archimedean point from which knowl-edge claims can be fully and finally adjudi-cated. These comments signify thatknowledge is always provisional and moti-vated by human interests (Habermas, 1971),and in some cases (e.g. imperialist accountsof dominated peoples) can be grotesquelydistorted representations. So far so good. Itby no means follows, however, that ideas cannever attain to universal value, or, more sim-ply that an idea developed at place a mustinvariably fail when transferred to place b.This is a matter for step-by-step judgment,not for a blanket diktat. By contrast, commit-ment to the notion that theories must be pro-vincialised as prescribed by Leitner and

Sheppard (2015) calls for a clear identifica-tion of what constitutes a meaningful ‘prov-ince’, and in the absence of any operationalguidelines in this matter (as in the workunder evaluation here) amounts to little morethan an arbitrary and self-defeating prefer-ence for intellectual parochialism at theexpense of more searching theoretical gener-alisation. At the same time, and as a correc-tive to the one-dimensional critique ofNorthern theory that is offered by postcolo-nial urban scholars, many of these sameissues of bias and ethnocentricity are onesthat theorists have struggled with since theEnlightenment, above all in regard to thequestion as to what constitutes the commonor universal features of humanity and whatin different contexts represents essential dif-ferences in human behaviours and aspira-tions (Pagden, 2013). The tensions in thisduality were especially prominent inEuropean debates over the 18th and 19thcenturies (and even as far back as the 16thcentury if we consider Montaigne) about thenature of distant ‘others’.

We can see in more detail why the critiqueof ‘Northern’ theory by postcolonial urbanscholars is unduly one-sided by examininghow these scholars deal with modernism anddevelopmentalism.

Modernism-developmentalism. There can be lit-tle doubt that some versions of modernist-developmentalist theory impose misguidedconcepts not only on the cities of the GlobalSouth but also on those of the GlobalNorth. The implausible mechanical model ofstages of growth is one such theory. Thesame can be said for the organic-ecologicalmodel of neighbourhood succession asdeveloped by the Chicago School, which isespecially suspect given its Darwinian under-tones and its association with the concept ofthe folk-urban continuum (Robinson, 2006).

Whatever the failures of these particulartheories may be, scholars in both the Global




South and the Global North are in practicefaced with the evident empirical fact of themarked differences in levels of economicdevelopment and income exhibited by citiesin different parts of the world and the effectsof these differences on urban outcomes (cf.Smith, 2013). Acknowledgement of the pow-erful role of economic forces in shaping theurban milieu is not to advocate any sort ofteleology of urban history, with all citieseverywhere eventually converging to a stateof achieved ‘modernity’. To the contrary, werecognise that the empirical trajectories ofdevelopment followed by individual citiesvary markedly, both within the GlobalSouth and North as well as within singlecountries. Over time, some cities grow at anaccelerated pace; some grow rapidly andthen decline; some remain in a proto-capitalist state of development; some areprosperous while others are impoverished;some specialise in manufacturing while oth-ers are more given to service provision; somehave dependent branch-plant economieswhile others become centres of innovativehigh-income entrepreneurialism; and so onthrough any number of possible variations.Throughout all of this diversity, however,there remains the burning issue of how spe-cific forms and levels of economic develop-ment shape specific variants ofagglomeration and high-density land use –in other words, the urban land nexus – andhow this in turn feeds back upon those sameforms and levels.

Postcolonial scholars’ fixation on the sup-posed exceptionalism of the Global South isparticularly evident in their treatment ofsuch favoured themes as poverty, slums,informal labour markets, vulnerable prop-erty rights, inadequate infrastructure andlack of sanitation (Roy, 2005). These themesare frequently dealt with as though they hadno family resemblance to similar issues inthe Global North. Yet we only need think ofCharles Dickens’ London, Emile Zola’s

Paris and Sinclair Lewis’ Chicago, or morerecent cases of deprivation and spatial exclu-sion in Europe and America revealed in thestudies of Chetty et al. (2014), Sampson(2012), Standing (2011) and Wilson (1987),to recognise that there is much in commonbetween the cities of the Global North andthe Global South in regard to poverty, andthat examination of the former has much tooffer to scholars of the latter, and vice versa.These remarks lead on to consideration ofpostcolonial scholars’ approach to develop-mental issues generally, and in particular, asChibber (2013) points out, their claimsabout developmental theories in the GlobalNorth being simple and linear as comparedwith the experience of the Global Southwhere development is said to be complexand non-linear. Of course, as we havealready pointed out, these claims aboutNorthern development theories are a misre-presentation. Many different formulationsof the diverse Northern routes to develop-ment have long constituted one of the princi-pal axes of debate within Northern historicalresearch (Allen, 2009; Aston and Philpin,1987). Even so, post-colonial scholars con-tinue to assert these claims as background totheir view that urban development of theGlobal South is so unique as to defy any the-oretical description that might establishcommonalities with cities elsewhere. In otherinstances, post-colonial scholars (such asRobinson, 2011) effectively shift questionsabout the interrelations between economicdevelopment and urbanisation into the dis-tant background as nothing but Northerntheoretical fantasies irrevocably marred byEurocentric parochialism, reductionism andteleological thinking. In fact, in both theNorth and the South, despite many empiricaldifferences of history and geography, theshifting forces of capitalism and markets andtheir expression in production, trade andemployment pose a consistent set of concep-tual problems. These include the ways in




which capitalist and non-capitalist systemsarticulate with one another, as in the case ofthe co-agglomeration of producers in theinformal and formal sectors in India asdescribed by Mukim (2015; see also Reyet al., 1971) or the dynamics of informality inAmerican cities offered by Mukhija andLoukaitou-Sideris (2014). Certainly, there ismuch in the way of difference and idiosyn-crasy to investigate in cities around theworld, but theory is required for this sort ofinvestigation to take on any wider meaning.We have posed a generalised theoreticalframework of the urban land nexus inter-twined with five crucial processes shapingthe specifics of urbanisation in differenttimes and places (level of development;resource allocation rules; forms and levelsof social stratification; cultural norms andtraditions; authority and power). Thisframework precisely addresses the need toacknowledge diversity, but without fallinginto the sophism of particularism andthereby losing sight of the forces that affectall cities.

Comparativism and its limits. One of the waysin which postcolonial scholars seek to com-pensate for their deep scepticism aboutmuch current urban theory is by means ofthe ‘comparative gesture’ stressing ‘thinkingacross differences’ in ways that are ‘poten-tially open to the experiences of all cities’(Robinson, 2014: 57). A comparativeapproach is especially congenial to postcolo-nial studies because it is assumed to reducethe dangers of aprioristic thinking about cit-ies and the inappropriate imposition of alienconcepts on given empirical situations.Myers (2014), for example, offers what isintended to be an exemplary comparativeaccount of multi-racial policy approaches tourban transportation policy in Capetownand Nairobi on the one hand and Hartford,CT on the other. This account turns on theproposition that the experiences of these

African cities in the matter of communitydevelopment could have usefully informedthe design of the Hartford-New Britaintransportation corridor, and it is no doubtinteresting and pertinent within its own lim-ited terms of reference. Yet like so muchother work in this comparative genre it sig-nally fails ‘to transform existing conceptuali-zations’ as optimistically promised byproponents of the genre (Robinson andRoy, 2015: 3). Our own argument is that thewell-travelled but narrow road representedby comparative and classificatory methodscertainly adds a number of legitimate proce-dures to the social scientist’s toolbox.However, if comparisons are to be effective,they can never proceed on the basis of theo-retically uninformed choices about cases forcomparison or the specific variables that areisolated for examination. Prior conceptuallabour about these matters is essential if com-parative methodologies are to produce – otherthan by accident or good luck – significantresults. This means specifically that we needto have a degree of conceptual clarity orintuition about the issues under examinationin order for comparison to proceed in a waythat reveals consequential insights when dif-ferent empirical situations are brought intoconjunction with one another.

To be sure, the comparative gesture canbe useful and interesting, but our point isthat a more theoretically self-conscious pool-ing of data, experiences and investigativeresults is essential if urban investigations areto progress beyond localism, difference andthe celebration of empirical complexity forits own sake. For this reason, there is muchto be said in favour of identifying theoreti-cally meaningful categories of cities thatcontain multiple cases of similar but notidentical cities. This point is recognised byRobinson (2011), though her argument isstrongly in favour of inductive analysis ofthe data. Other postcolonial urbanistsappear to be rather more ambivalent about




this issue. For example, Bunnell andMaringanti (2010) and Roy (2009) raisearguments against categories like world cit-ies, international financial centres and city-regions, not so much because they may bemisidentified but because they are said,somehow or other, to relegate excluded citiesto secondary status while supposedly divert-ing our attention away from the full diver-sity of urban forms and experiences that theworld has to offer. Here again we come faceto face with the new particularism and therelegation of all cities to the status of ‘ordi-nariness’. This insistence on difference andidiosyncrasy within ordinary cities is espe-cially evident in the case of another favouritetarget of postcolonial critique, i.e. the notionof the representative or exemplary city, and,notably, the writings of the so-called LASchool about the ‘paradigmatic’ status ofLos Angeles towards the end of the 20thcentury. We would be the first to acknowl-edge the incautiousness of many of the LASchool’s theses, and yet it is important torecord that Los Angeles, in its pioneeringstatus as a globalising post-fordist centre offlexible specialisation, disorganised labour,growing social inequality and polycentricitydid indeed turn out to be an early and pow-erful expression of several incipient world-wide trends (Soja and Scott, 1986).Accordingly, the LA School called attentionat an early stage to a developmental path-way that many other cities all over the worldhave subsequently followed, and from thisperspective it was most certainly exemplary.

McFarlane’s railway ticket

Over the last few decades, ‘assemblage the-ory’ has emerged as a major genre of work inurban studies, as in the social sciences in gen-eral (see, for example, DeLanda, 2002;Latour, 2005). This theory, which has manyaffinities with postcolonial urbanism, has fil-tered into urban studies from the work of

continental post-structuralist philosophers(in particular, Deleuze and Guattari, 1972).Assemblage theory is of considerable com-plexity in its philosophical representations,the understanding of which is not made anyeasier by the langue de bois favoured by itschief protagonists and the sharply conflictinginterpretations of their work by secondarycommentators. For our purposes, however,focused as they mostly are on abridged‘applications’ of this theory to urban analy-sis, a few essentials will suffice to motivateour critique.

Assemblage theory is first and foremostan ontological view of the world conceivedas a mass of rhizomatic networks or finely-grained relationships constituting the funda-mental character of reality. These networksbind together unique human and non-human objects within fluid, hybrid mosaicsforming more or less temporarily stabilisedsystems of interconnections representing thecurrent state of the observable world.Assemblages become stabilised by ‘territor-ialization’ (as opposed to destabilising deter-ritorialisation) when they are anchored toparticular tracts of geographical space.Importantly, any state of reality in this the-ory is taken to be ‘flat’ in the sense that anyperceived hierarchical or scalar ordering(from a top to a bottom) decomposes backagain into the kaleidoscopic, rhizomatic andhorizontal relations that are said to consti-tute it (DeLanda, 2002; see also Marstonet al., 2005). This point is largely sharedbetween assemblage theory and postcolonialtheory, via the latter’s emphasis on differ-ence and its focus on the incommensurableuniqueness of cities.

There are several variants of assemblagetheory, but one of the most influential isactor-network theory, a body of ideas asso-ciated above all with the work of Latour(2005). This is, again, a way of exploring themultiple relationships that tie human andnon-human objects together, but with the




additional claim that all of these objects areconstituted as actants, i.e. capable of agencyin the sense that they exert effects on otheractants. In a number of methodological andtheoretical publications, Farıas (e.g. 2010,2011) has outlined the main implications ofassemblage and actor-network theory forurban studies. He writes that the city is ‘.an object which is relentlessly being assem-bled at concrete sites of urban practice, or,to put it differently, as a multiplicity of pro-cesses of becoming, affixing sociotechnicalnetworks, hybrid collectives and alternativetopologies’ (Farıas, 2010: 2).

Some of the multiple ways in which theurban might be assembled are thenenumerated:

. as a transport system, as a playground forskateboarders and free-runners (parkour), as alandscape of power, as a public stage for polit-ical action and demonstration, as a no-goarea, as a festival, as a surveillance area, as asocialization space, as a private memory, as acreative milieu, as a jurisdiction, etc. (Farıas,2010: 14)

This conception then leads to a descriptive,anecdotal and notably indiscriminateapproach to urban investigation. Farıas(2011: 367) with apparent faith in the powersof inductive empiricism goes so far as to saythat ‘we don’t know what we are looking foruntil we find it’. Little wonder, then, thatBrenner et al. (2011) characterise this line ofresearch as ‘naıve objectivism’ and point toits failure to distinguish between the signifi-cant and the insignificant in urban analysis.Certainly, the assemblage approach is poten-tially of positive value in certain kinds ofethnographic and narrative accounts of thecity such as those offered by de Boek andPlissart (2004), Mbembe and Nuttall (2004)or Simone (2014); and as Geertz (1973) hasshown, thick description of social practicesand material forms in cities or elsewhere canoften provide sensitive depictions of the

ways in which social lives are woventogether. One example might be the complexmanner in which we build high-rise down-town environments in major cities and theconnections of this process to the construc-tion industry, architectural practices andbuilding norms, competing demands forspace, visual conceptions of the built envi-ronment and office employment in the city.Our critique of assemblage theory thereforedoes not deny the possibility of certainimportant feedbacks between non-humanobjects and human society and it is emphati-cally not intended to repudiate the reflexiverelations between technology, urban spaceand social life (cf. Graham and Marvin,2001). However, we certainly do have strongreservations about the capability of inani-mate objects to ‘act’ as if whatever causal orgenerative powers they may possess wereontologically equivalent to sentient, purpo-sive human behaviour.

Assemblage theory radically privileges theactivity of assemblage itself, seeing no widerforces that might determine what assem-blages are possible or not possible; rather, itadvocates a methodology of building the ele-ments of social organisation a posteriorifrom the ground up (Bender, 2010) andfocusing on specific sites of daily life(Simone, 2011). The result is a largely inde-terminate concept of the city as a complex,variegated, multifarious, open-ended, fluid,unique, hybrid, unruly, nonlinear, etc., etc.aggregate of disparate phenomena tiedtogether in a haphazard mix of causal andcontingent relationships. This concept, likeRobinson’s (2011: 13) (postcolonial) view ofthe city as ‘a site of assemblage, multiplicity,and connectivity’ is at one level of observa-tion certainly correct, but at another levelinterposes mere empirical convolution as asubstitute for a deeper and more systematiclevel of (theoretical) comprehension. Thisnaıve objectivism frequently also results inmarkedly indiscriminate bodies of




information being packaged into empiricalassemblages, perhaps especially where, inthe words Acuto (2011: 553), those ‘missingmasses of non-human actors often degradedto the role of mundane artefacts’ arebrought into concatenation with human life.In other words, there are no theoreticalguideposts in assemblage theory for tellingus how tease out significant relationships orto distinguish between the trivial and theimportant.

We may further pin down these remarksby reference to the work of McFarlane(2011a, 2011b), another prominent spokes-person for assemblage theory. In his livelydefense of this theory, McFarlane (2011c:216) draws on his research on poverty andinformal housing in Mumbai. He insists thatany attempt to understand ‘the everydaylives and hardships faced by the poor’requires us to pay attention to an eclecticcollection of ‘urban materialities’, whichinclude in this instance such disparateobjects as sackcloth, corrugated iron, brick,breezeblock, hydroform and infrastructuresof drainage, sanitation, water and electricity.These elements are then organised into adescription of poverty in Mumbai, but criti-cally the account – which has the analyti-cally ‘flat’ quality prescribed by assemblagetheory – is essentially devoid of useful expla-natory ideas. The same can be said for theanalogous work of Dovey (2012) who, likeSimone (2011), puts forth an extended dis-play of deleuzoguattarian4 jargon in anattempt to illuminate descriptions of urbaninformality that nevertheless remain unin-formative about the basic logic of social andeconomic marginalisation.

The fetishisation of inanimate objects asinstruments of agency is dramatically high-lighted in McFarlane’s (2011c: 217) discus-sion of the work of political activists incombatting poverty in Mumbai. He pointsto the way in which these activists discov-ered that they could make free telephone

calls by inserting a railway ticket into areceiver, and by this means greatly extendtheir outreach. McFarlane then states thatthis is all part of the ‘experience and possibi-lities of urban life’. Our point, by the way, isnot to dismiss this kind of narrative asmeaningless in principle. A good story is agood story, after all. Our point is rather thatin the case of McFarlane’s railway ticket atrivial contingency is in all seriousnessoffered as a link within a chain of agencythat is supposed to function as a way ofunderstanding urban poverty and as aninformed account of the struggles that peo-ple engage in to escape from it. This picturesharply contrasts to the more analyticallycontrolled realism about obstacles in theway of the poor that is painted by suchdiverse authors as Aw (2013), Boo (2012),Caldeira (2000) and Cole (2014). As Brenneret al. (2011: 233) write of actor-network the-ory: ‘This mode of analysis presupposes thatthe ‘facts’—in this case, those of intercon-nection among human and nonhumanactants—speak for themselves rather thanrequiring mediation or at least animationthrough theoretical assumptions and inter-pretive schemata’. Thus, in the flattenedworld of assemblage theory there is a peri-lous tendency to fail to distinguish betweenthe inanimate character of material objectsand the intentionality of humans, and tocompound this oversight by undertheorisedpresentations of social interconnectivity (cf.Tonkiss, 2011). This flattening of the worldalso evacuates any meaningful political con-tent from assemblage theory since every-thing is equally important (or equally trivialand unimportant).

The principal problems of assemblagetheory as discussed in this section of thepaper – the notion of reality as mere rhizo-matic entanglements without underlyingprocesses of structuration, the indiscriminateattribution of agency to things and theabsence of concepts of human action – make




this theory unable to detect urban dynamics,movement, change and causality in mean-ingful ways. Critical realism has long pro-vided a way out of this kind of dead-end byinsisting on the importance of necessary rela-tionships, causal powers and theoreticalabstraction as fundamental to the identifica-tion of the central properties and conditionsof existence of social phenomena (Sayer,2004). One searches in vain in assemblagetheory and urban research based on it toknow what larger difference assemblagesmake, which assemblages are important andwhich are insignificant and fleeting, whichare empowering and which are disempower-ing and what kinds of policy interventionsare most likely to bring about desired formsof social change.

Planetary perplexities

Some urban analysts today, most notablyBrenner and Schmid (2015), suggest that inthe 21st century a radical blurring of thecategory of the urban versus everything elsehas come about, and that what were for-merly identified as urban areas can no longerbe distinguished from the rest of geographi-cal space, conceptually or empirically. Theseare the central doctrines of ‘planetary urban-ism’. As Brenner and Schmid (2014: 750)write:

It is clear that settlement-based understandingsof the urban condition have now become obso-

lete. The urban cannot be plausibly under-stood as a bounded, enclosed site of socialrelations that is to be contrasted with non-urban zones or conditions. It is time, therefore,to explode our inherited assumptions regard-ing the morphologies, territorializations andsociospatial dynamics of the urban condition.

Given the geographically intensive andextensive development of global capitalism,the authors are doubtless correct to refer toan integrated planet-wide socio-economic

system. They are also right to claim that thenotion of a purely ‘rural’ realm occupyingthe interstitial spaces between cities isarchaic and misleading. The notion hasnever, in any case, been entirely satisfactorygiven the diversity of these territories(deserts, forests, mountain ranges, sites ofpeasant farming, expanses of industrial agri-culture, spaces of resource extraction, touristregions, etc.). But we are at a loss to under-stand how these facts can lead to a claimthat the idea of the city only ‘persists as anideological framing’ (Brenner and Schmid,2015: 152), a phrase that is echoed byMerrifield’s (2013) characterisation of thesame idea as a ‘pseudo-concept’. Above all,as we show below, Brenner and Schmid donot conclusively demonstrate that the cityfades away as an identifiable geographicentity and scale of socio-economic interac-tion within planetary space, or that any dis-tinction between the urban and the rest ofgeographical space (what they misleadinglyinsist on calling ‘the rural’) must now beabandoned; and they are merely bafflingwhen they write about the full extent of pla-netary space as being ‘urbanized’, especiallywhen this includes ‘rainforests, deserts,alpine regions, polar zones, and oceans andeven the atmosphere’ (Brenner and Schmid,2015: 152–153).5

Brenner and Schmid hedge their betsrather clumsily by saying that there is some-thing called ‘concentrated’ urbanisation, orwhat we usually call cities, and somethingcalled ‘extended’ urbanisation, which moreor less corresponds to everything else. Thepuzzle is why they want to introduce thesemantic confusion that ensues from apply-ing the term ‘urban’ with all its familiar city-centric connotations to everything else whennumerous other descriptive terms are quiteconceivable.6 We shall argue that not only isthere no conceptual (or what they relent-lessly call ‘epistemological’) gain by thismanoeuvre, but considerable theoretical




loss. Here, Angelo and Wachsmuth (2015)enter the fray with their commentary onsomething that they allude to as ‘methodolo-gical cityism’. They identify this forbiddingsin with the error of ascribing to the circum-scribed geographical structure of the cityprocesses that they say are more properly tobe analysed within the wider framework ofBrenner and Schmid’s ‘planetary urbaniza-tion’. The plot thickens when Brenner andSchmid point out (correctly as it happens)that there are usually no simple orintuitively-identifiable boundaries betweenthe city (concentrated space) and the rest ofthe world (extended space) so that the conti-nuity between the two appears to be unbro-ken. This is a familiar problem that hasalways perplexed urban analysts, butBrenner and Schmid are wrong to think thatthe issue goes away by assimilating thewhole of geographical space into an urbanproblematic. There is in fact a more satisfac-tory way of approaching this problem.

Consider, to begin with, certain kinds ofphenomena that exist at the intra-urbanlevel, such as neighbourhoods, slums, indus-trial quarters, central business districts andsuburbs. Each of these phenomena repre-sents a distinctive and multifaceted type ofsocio-spatial outcome within a wider urbanspace (the urban land nexus) and none isdivided from the rest of the city by a clearline of demarcation. Yet each appears to usas an ontologically distinctive scale of urbanspace not only because of its empirical char-acter but also because each poses uniquelyproblematical scientific and political questionsderiving from its mode(s) of operation.Sampson (2012), for example, has shownthat there are many and sundry ‘neighbour-hood effects’ on people who live in poorcommunities, and Chetty et al. (2014) haveshown how these effects also have an impacton intergenerational poverty rates. Similarly,the vast literature on local economic

development reveals that intra-urban clus-ters of production units are marked by pow-erful spatial dynamics that are uniquelyproblematical as objects of inquiry. All ofthese phenomena are embedded in andmarked by all manner of continuities withthe urban land nexus, but in no case is ituseful or meaningful simply to dismiss themas ideological constructions. Two relatedpoints now need to be made.

First, the city is a composite social, politi-cal, cultural and economic phenomenon(anchored and integrated by the urban landnexus) that is very much greater than thesum of its parts, signifying, in turn, that ithas a potent collective presence. In particu-lar, the city is a site of joint dynamics with ajoint identity (e.g. ‘the San Francisco BayArea’, ‘Rio de Janeiro’) deriving from itscharacter as an agglomerated land nexus.This state of affairs means that the widerpolitical interests of the individual firms,households and other behavioural units thatmake up the urban sphere become entangledwith a concrete set of political interests spe-cific to the city (including those forms ofconflict, coalition, exclusion and deprivationpeculiar to the urban land nexus). Thesepolitical interests are partly mobilised in col-lective action and are almost always associ-ated with formal institutions (especiallygovernmental institutions) that endow citieswith powers of taxation, managerial regula-tion and the capacity to make substantialpublic investments). Among the moreimportant concerns of these institutions isthe performance of the city as a centre ofemployment, earnings and quality of life,again rooted in the urban land nexus(Molotch, 1976). For all of these reasons,the city at large – especially given its founda-tions in agglomeration and its dense institu-tional and political overlay – poses questionsthat are quite specific to the urban arenaboth as an object of scientific enquiry and as




a scale of human political and economic life.This is why proponents of so-called metho-dological cityism are mistaken in their char-acterisation of the city as nothing more thanan ideological mirage.

Second, just as neighbourhoods, slums,industrial quarters, etc. are distinctive andidiosyncratic socio-spatial articulations(albeit within the urban land nexus), so theurban land nexus itself is a distinctive socio-spatial articulation (within wider global orplanetary space). The city, in a nutshell, is inimportant ways an irreducible collectivityand, as we argued earlier, its peculiar char-acter derives from its properties as a locus ofagglomeration, gravitation and density aswell as from its specific daily and weeklyrhythms of life. These rhythms are embodiedmost notably in its local labour markets andits regular patterns of commuting (Cheshireand Hay, 1989; Kerr and Kominers, 2015).To state this latter point in another way, cit-ies concern us because distance is not dead,and substantial elements of our lives areanchored in these spatially-and temporally-constrained urban systems. The day whenwe can move with no cost in time or effortfrom one place to another (i.e. a world of‘magic carpets’) is the day when can say thatthe city is dead. But the overwhelming situa-tion in the contemporary world is one inwhich – despite the growth of long-distancelinkages – proximity and density remain cri-tically important as arrangements that facili-tate the still expanding volumes of detailed,small-scale, intimate and ever-changinginteractions that lie at the heart of humanrelationships within the urban land nexus.

There remains an unanswered question.Even given the above discussion, where andhow do we draw the dividing line betweenthe city and the rest of geographic space?We have argued, with specific reference tothe city, that in spite of the continuity/indivi-sibility of geographic space (or of reality as a

whole for that matter) there are differingscales and articulations of empirical phe-nomena, underlying processes and politicalinterests that make it imperative to distin-guish specific units and levels of interactionwithin the totality of planetary space as awhole. Moreover, there is no rigid line thatseparates the urban land nexus definitivelyfrom the rest of geographic space, but rathera series of spatial gradations in which wemove from the one to the other. This doesnot mean that the urban land nexus and itsdynamics as identified above are illusions,just as neighbourhoods, slums, industrialquarters, etc. do not dissolve away into anurban totality, and just as the fact that theseasons fade gradually and unevenly intoone another does not mean that they do notexist as identifiable phenomena in their ownright. The evident deduction from theseremarks is that we almost always have con-siderable leeway in practice as to how wedemarcate the spatial extent of the urbanland nexus, but that the best bet is to defineit in any given instance in a way that opti-mises our ability to deal with whatever givenquestion(s) we may have in hand (e.g. eco-nomic development, public transport, ethnicconflict, neighbourhood blight, urban politi-cal strategy and so on) while eliminatingfrom consideration as much irrelevant terri-tory as possible. In practice, we have littleoption but to follow the pragmatic rule ofthumb that has always been adopted by geo-graphers and to locate the line of division insome more or less workable way relative toavailable data.

One possible objection to these lines ofreasoning is that cities have diverse func-tional connections to other places in manydifferent parts of the world. Indeed Brennerand Schmid (2014, 2015), among others (e.g.Amin and Thrift, 2002), make the explicitclaim that the identity of the city as a spatialunit is deeply compromised by the widening




external relations that form its so-called‘constitutive outside’. Our response here issimple. These relations are capable of indu-cing certain kinds of changes in cities, suchas bursts of growth in central business dis-tricts or changes in given population cate-gories, but their effects are virtually alwaysassimilated into the urban land nexus as suchwithout destroying its integrity as a complexsocial unit. For example, the New Yorkfinancial district has strong external connec-tions to far-flung customers and sources ofcapital, resulting in the growth of local firmswith diverse impacts on land use patterns inManhattan and on workers’ residentialbehaviours. Whatever the effects of the con-stitutive outside of the city may be, however,these in no way undermine the theoreticalnotion of the urban land nexus as the criticalconstitutive inside of the city. Indeed, theurban land nexus gains in terms of its inter-nal complexity even as these effects intensifyand multiply. Equally, and despite the factthat in the world system of the 21st centuryspatial interconnections have attained unpre-cedented levels of volume and geographicextension, the need for proximity and localinteraction has in many ways been bolsteredwithin the urban land nexus (Duranton andStorper, 2008; Hummels, 2007). Andersonand van Wincoop (2004), for example, havedemonstrated that trade costs remain soimportant in today’s world that they fre-quently reinforce a distinctively local scaleof interaction. We should point out in anycase that long-distance interconnectionsbetween cities have always been a feature ofurban life, beginning in Jericho 6500 yearsago, and, as we have already stated, theyare typically a crucial condition of contin-ued urban viability. As such, they do notrepresent the negation of the identity of thecity but one of the conditions that havemade the existence of cities possiblethroughout history.

Summing-up: A challenge tourban theory and research

We have tried in all of the above to blast

open a number of theoretical geographies of

cities and urbanisation processes, and we

have criticised in particular certain recent

trends that for one reason or another deform

or mischaracterise or conceal the essential

functions and identity of the urban. At the

same time, we have offered as background

to our critique a concept of the city as a tan-

gible phenomenon, distinct from but con-

tained within society as a whole, and with

specific genetic roots and unique internal

organisational dynamics. This concept

allows us to distinguish what is authentically

urban from the merely contingently urban

and hence to bring a degree of disciplined

focus to the investigation of urban matters.

We should add that precision of ideas in this

respect is especially important in policy-

relevant research (see Scott and Storper,

2015).Against the backdrop of our own propo-

sitions about the nature of cities, we have

examined three influential alternative views

on urban matters, namely postcolonial the-

ory, assemblage-theoretic approaches and

planetary urbanism, and found them want-

ing. Postcolonial commentators argue for an

approach to urban studies that is simultane-

ously provincial, comparativist and focused

on difference, which in practice means parti-

cularity. While they invoke ambiguous

notions of ‘worlding’ they reject as a matter

of principle the transfer of analytical results

from cities of the Global North to cities of

the Global South and by the same token any

generalised theoretical concept of the urban,

and presumably (at least for purists) any

trans-provincial fertilisation of ideas.Assemblage-theoretic approaches have

much in common with these features, but inaddition are intent on portraying social




outcomes as relational kaleidoscopes inever-changing combinatorial arrangementsthat offer few or no insights as to the genet-ics of indurated spatial and institutionalarrangements. Not only are assemblage andactor-network approaches to the city nota-bly weak in grasping fundamental social andeconomic processes, but they compound thisweakness by suggesting that purely passivethings lacking in intentionality and socialdiscretion, like the door hinge mentioned byAcuto (2011), or the scallops studied byCallon (1984), or the railway ticket thatenters into the account of poverty byMcFarlane (2011c) are endowed with powersof agency akin to those of human subjects.

It should be noted that while postcolonialand assemblage-theoretic commentatorshave strong views about the conduct ofurban research, none of them offers anycoherent concept of the urban as such.Planetary urbanists for their part makestrong claims about the deliquescence of thecity as commonly understood and the assim-ilation of the urban into a world-wide space-economy. They provide little in the way ofconceptual value-added by this manoeuvrewhile gratuitously deforming the usuallyaccepted meaning of the term ‘urbanisation’pointing, as it does, to agglomeration, den-sity and nodality and, by extension, to dis-tinctive political, social, economic andidentity-forming processes at the urbanscale. Our own propositions regarding thematerial and relational structures of theurban land nexus suggest that the claims ofplanetary urbanists about the supposed way-wardness of what they call methodologicalcityism and about the purely ideological sta-tus of the concept of the city are in the endseriously mistaken.

At least some of these aberrant tendenciesin contemporary urban theory can be tracedback to a remarkably uncritical faith amongmany contemporary analysts in the ability ofabstracted philosophical ideas to orchestrate

the shape and form of concrete investiga-tions of cities. We are not opposed to incur-sions of philosophical ideas into the work ofurban theorists; far from it. We are only tooaware of how necessary philosophically-based criticality and clarity are to viablesocial analysis. Our concern here is focusedprimarily on what we take to be the unfortu-nate influence of post-structuralist philoso-phy in urban studies. We are referring here,first, to the semantically-inflated jargon thatmars so much of the literature today. Moreimportantly, and second, we also point tothe overblown interpretative schemas thatpost-structuralism licenses and their ten-dency to crowd out analytically-orientedforms of social (and especially economic)enquiry in favour of a conceptually barrensearch for difference, particularity and local-ism. The ontologies of flatness favoured bypost-structural theory are equally damagingto the vibrancy of urban studies especially intheir denial of scalar dimensions to space ina manner that effectively dissolves the cityaway as a structured socio-geographic entity,and this encourages in turn a rampant eclec-ticism so that the city as such tends to shiftpersistently out of focus. Planetary urbanistsare also at pains to secure this same dissolu-tion, but this time on the basis of an enig-matic ‘epistemology’ that in practice standsin for some rather unexceptional and, in ouropinion, imperfectly digested observationalstatements. To repeat the message of ouropening line, the current period of historycan most certainly be characterised as anurban era, in the sense that more and moreof humanity lives in distinctively urban set-tlements. If we are to come to some sort ofunderstanding of the new and daunting chal-lenges posed by this state of affairs (includ-ing a clear understanding of what is and isnot ascribable to urban processes in modernlife), we need an urban theory that is fullyup to the task. We have tried to clear awaysome of the obstacles that we argue stand in




the way of the accomplishment of this goal,and to propose some essential groundworkfor building more effective theories of theurban and the urbanisation process.

Finally, we strongly advocate abandon-ment of the classification of cities in terms ofa Global North and a Global South with itscurious echo of First and Third Worlds.Postcolonial theorists, of course, have theirown reasons for hewing to this terminology,and we ourselves certainly have no intentionof suggesting that colonialism, even today,has not left deep traces on many differentparts of the world and in many domains ofhuman enquiry. That said, and in view ofthe prevailing, many-sided patchwork ofspatial outcomes exhibiting many differentempirical varieties of economic and politicaldevelopment in today’s world, this schematicbinary is quite definitely inadequate as anorganisational framework for huge swathsof contemporary social investigation, andnowhere more so than in the case of urbanstudies.

Acknowledgements

We dedicate this paper to our late friend and col-league Edward Soja in the full knowledge that hewould have been unable to resist expressing alter-native perspectives on almost everything we say.

Funding

This research received no specific grant from anyfunding agency in the public, commercial, or not-for-profit sectors.

Notes

1. In this paper, the term ‘cities’ will generallybe used to cover both small and large urbanforms, including metropolitan areas and city-regions.

2. Much the same can be said for the term‘determinism’ that Mould (2015) invokes incriticism of our earlier paper. Any self-respecting determinist is likely to insist at a

minimum that a deterministic approachinvolves the suppression of free will in favourof purely material or structural-functionalcausalities. Mould mobilises no reasoning orevidence as to how or why our theory of theurban land nexus involves any conception ofthis sort.

3. And now that opportunities for research onthe cities of the Global South are expandingapace throughout the North and the Souththis relative neglect on the part of urban scho-lars will presumably fade rapidly away.

4. We use the term satirically.5. A detailed response to Brenner and Schmid

has been offered by Walker (2015). His mainlines of critique differ from ours, but areequally adamant about the integrity of thecity as an object of theoretical enquiry.

6. For example, among the many possible alter-native terminologies are: ‘global space-econ-omy’, ‘planetary capitalism’ or the‘geographical anatomy of global society’.These terminologies capture the spirit of whatBrenner and Schmid seem to be saying, with-out obliterating the commonly receivedmeaning of the term ‘urban’.

References

Acuto M (2011) Putting ANTs into the mille-

feuille. City 15: 552–562.Allen RC (2009) The British Industrial Revolution

in Historical Perspective. Cambridge: Cam-

bridge University Press.Amin A and Graham S (1997) The ordinary city.

Transactions of the Institute of British Geogra-

phers 22: 411–429.Amin A and Thrift N (2002) Cities: Reimagining

the Urban. Cambridge: Polity.Anderson JE and van Wincoop E (2004) Trade

costs. Journal of Economic Literature 42:

691–751.Angelo H and Wachsmuth D (2015) Urbanizing

urban political economy: A critique of metho-

dological cityism. International Journal of

Urban and Regional Research 39: 16–27.Aston T and Philpin CHE (1987) The Brenner

Debate: Agrarian Class Structure and Eco-

nomic Development in Pre-Industrial Europe.

Cambridge: Cambridge University Press.




Aw T (2013) Five Star Billionaire. New York:

Random House.Bender T (2010) Postscript: Reassembling the

city: Networks and urban imaginaries. In: Far-

ıas I and Bender T (eds) Urban Assemblages:

How Actor-Network Theory Changes Urban

Theory. London: Routledge, pp. 303–323.Boo K (2012) Behind the Beautiful Forevers: Life,

Death and Hope in a Mumbai Undercity. New

York: Random House.Brenner N and Schmid C (2014) The urban age

in question. International Journal of Urban and

Regional Research 38: 731–755.Brenner N and Schmid C (2015) Towards a new

epistemology of the urban. City 19: 151–182.Brenner N, Madden DJ and Wachsmuth D (2011)

Assemblage urbanism and the challenges of

critical urban theory. City 15: 225–240.Bunnell T and Maringanti A (2010) Practising

urban and regional research beyond metrocen-

tricity. International Journal of Urban and

Regional Research 34: 415–420.Caldeira T (2000) City of Walls: Crime, Segrega-

tion and Citizenship in Sao Paulo. Berkeley

and Los Angeles: University of California

Press.Callon M (1984) Some elements of a sociology of

translation: Domestication of the scallops and

the fishermen of St Brieuc Bay. Sociological

Review Special Supplement S1: 196–233.Cheshire P and Hay D (1989) Urban Problems in

Western Europe: An Economic Analysis. Lon-

don: Hyman.Chetty R, Hendren N, Kline P, et al. (2014)

Where is the land of opportunity? The geogra-

phy of intergenerational mobility in the United

States. Quarterly Journal of Economics 129:

1553–1623.Chibber V (2013) Postcolonial Theory and the

Spectre of Capitalism. London: Verso.Cole T (2014) Every Day is for the Thief. New

York: Random House.Comaroff J and Comaroff JL (2012) Theory from

the South: How Euro-America is Evolving

toward Africa. Boulder, CO: Paradigm.Cooke P and Morgan K (1998) The Associational

Economy: Firms, Regions, and Innovation.

Oxford: Oxford University Press.de Boek F and Plissart M-F (2004) Kinshasa:

Tales of the Invisible City. Antwerp: Ludion.

DeLanda M (2002) Intensive Science and Virtual

Philosophy. London: Continuum.Deleuze G and Guattari F (1972) Capitalisme et

Schizophrenie. Paris: Editions de Minuit.Dick HW and Rimmer PJ (1998) Beyond the

Third World city: The new urban geography

of South-East Asia. Urban Studies 35:

2303–2321.Dovey K (2012) Informal urbanism and complex

adaptive assemblage. International Develop-

ment Planning Review 34: 349–367.

Duranton G and Puga D (2004) Micro founda-

tions of urban agglomeration economies. In:

Henderson JV and Thisse JF (eds) Handbook

of Regional and Urban Economics, Vol. 4.

Amsterdam: Elsevier, pp. 2065–2118.Duranton G and Storper M (2008) Rising trade

costs? Agglomeration and trade with endogen-

ous transaction costs. Canadian Journal of

Economics 41: 292–319.Edensor T and Jayne M (2012) Introduction:

Urban theory beyond the West. In: Edensor T

and Jayne M (eds) Urban Theory Beyond the

West: A World of Cities. London: Routledge,

pp. 1–27.Farıas I (2010) Introduction: Decentering the

object of urban studies. In: Farıas I and

Bender T (eds) Urban Assemblages: How

Actor-Network Theory Changes Urban Theory.

London: Routledge, pp. 1–24.Farıas I (2011) The politics of urban assemblages.

City 15: 365–374.Fujita M and Thisse J-F (2002) Economics of

Agglomeration: Cities, Industrial Location, and

Regional Growth. Cambridge: Cambridge Uni-

versity Press.Fujita M, Krugman P and Venables AJ (1999)

The Spatial Economy: Cities, Regions and

International Trade. Cambridge, MA: MIT

Press.Geertz C (1973) Interpretation of Cultures. New

York: Basic Books.Glaeser EL (2011) Triumph of the City. London:

Macmillan.Graham S and Marvin S (2001) Splintering

Urbanism: Networked Infrastructures, Techno-

logical Mobilities, and the Urban Condition.

London: Routledge.Habermas J (1971) Knowledge and Human Inter-

ests. Boston, MA: Beacon.




Hall P (1998) Cities in Civilization. New York:

Pantheon.Haraway D (1988) Situated knowledges. Feminist

Studies 14: 575–599.Hummels D (2007) Transportation costs and

international trade in the second era of globa-

lization. Journal of Economic Perspectives 21:

131–154.Kerr W and Kominers SD (2015) Agglomerative

forces and cluster shapes. Journal of Economics

and Statistics 97(4): 877–899.Krugman P (1991) Geography and Trade. Leuven:

Leuven University Press.Kuhn TS (1962) The Structure of Scientific Revo-

lutions. Chicago, IL: University of Chicago

Press.Latour B (2005) Reassembling the Social: An

Introduction to Actor-Network Theory. Oxford:

Oxford University Press.Lefebvre H (1970) La Revolution Urbaine. Paris:

Gallimard.Leitner H and Sheppard E (2015) Provincializing

critical urban theory: Extending the ecosystem

of possibilities. International Journal of Urban

and Regional Research. Epub ahead of print.

DOI: 10.1111/1468-2427.12277.Livingstone DN (2014) Dealing with Darwin:

Place, Politics, and Rhetoric in Religious

Engagements with Evolution. Baltimore, MD:

Johns Hopkins University Press.Mannheim K (1952) Essays in the Sociology of

Knowledge. Henley-on-Thames: Routledge

and Kegan Paul.Marston SA, Jones JP and Woodward K (2005)

Human geography without scale. Transactions

of the Institute of British Geographers 30:

416–432.Mbembe A and Nuttall S (2004) Writing the

world from an African metropolis. Public Cul-

ture 16: 347–372.

McFarlane C (2011a) Assemblage and critical

urban theory. City 15: 204–224.McFarlane C (2011b) Learning the City: Knowl-

edge and Translocal Assemblage. Oxford:

Wiley-Blackwell.McFarlane C (2011c) On context: Assemblage,

political economy and structure. City 15:

375–388.McKinsey & Company (2011) Urban World.

San Francisco: McKinsey Global Institute.

Available at: http://www.mckinsey.com/global-

themes/urbanization/urban-world-mapping-the-

economic-power-of-cities (accessed 19 February

2016).Merrifield A (2013) The urban question under

planetary capitalism. International Journal of

Urban and Regional Research 37: 909–922.Molotch H (1976) The city as a growth machine:

The political economy of place. American

Journal of Sociology 82: 309–332.

Mould O (2015) A limitless urban theory? A

response to Scott and Storper’s ‘The nature of

cities: the scope and limits of urban theory’.

International Journal of Urban and Regional

Research. Epub ahead of print. DOI: 10.1111/

1468-2427.12288.Mukhija V and Loukaitou-Sideris A (eds) (2014)

The Informal American City. Cambridge, MA:

MIT Press.Mukim M (2015) Coagglomeration of formal and

informal industry: Evidence from India. Jour-

nal of Economic Geography 15: 329–351.Myers G (2014) From expected to unexpected

comparisons. Singapore Journal of Tropical

Geography 35: 104–118.Ong A and Roy A (eds) (2011) Worlding Cities:

Asian Experiments in the Art of Being Global.

Oxford: Wiley-Blackwell.Pagden A (2013) The Enlightenment and Why it

Still Matters. New York: Random House.Patel S (2014) Is there a ‘south’ perspective to

urban studies? In: Parnell S and Oldfield S

(eds) The Routledge Handbook on the Cities of the

Global South. London: Routledge, pp. 37–53.Peck J (2015) Cities beyond compare. Regional

Studies 49: 160–182.Ren J and Luger J (2015) Comparative urbanism

and the Asian city: Implications for research

and theory. International Journal of Urban and

Regional Research 39: 145–156.Rey P, Duroux Y and Bettelheim C (1971) Sur

l’Articulation des Modes de Production. Paris:

Ecole Pratique des Hautes Etudes, Centre

d’Etudes de Planification Socialiste.Robinson J (2006) The Ordinary City: Between

Modernity and Development. London:

Routledge.Robinson J (2011) Cities in a world of cities: The

comparative gesture. International Journal of

Urban and Regional Research 35: 1–23.




Robinson J (2014) New geographies of theorizingthe urban: Putting comparison to work forglobal urban studies. In: Parnell S and Old-field S (eds) The Routledge Handbook on Cities

of the Global South. London: Routledge, pp.57–70.

Robinson J and Roy A (2015) Global urbanismsand the nature of urban theory. InternationalJournal of Urban and Regional Research. Epubahead of print. DOI: 10.1111/1468-2427.12272.

Roweis ST and Scott AJ (1977) Urban planning

in theory and practice: A re-appraisal. Environ-ment and Planning 9: 1097–1119.

Roy A (2005) Urban informality: Toward an epis-temology of planning. Journal of the American

Planning Association 71: 147–158.Roy A (2009) The 21st-century metropolis: New

geographies of theory. Regional Studies 43:819–830.

Roy A (2011) Slumdog cities: Rethinking subal-tern urbanism. International Journal of Urban

and Regional Research 35: 223–238.Roy A (2015) Who’s afraid of postcolonial the-

ory? International Journal of Urban and

Regional Research. Epub ahead of print. DOI:10.1111/1468-2427.12274.

Said EW (1978) Orientalism. New York:Pantheon.

Sampson RJ (2012) Great American City: Chi-

cago and the Enduring Neighborhood Effect.Chicago, IL: University of Chicago Press.

Sayer A (2004) Foreword: Why critical realism?In: Fleetwood S and Ackroyd S (eds) CriticalRealist applications in Organisation and Man-

agement Studies. London: Routledge, pp. 6–20.Scott AJ (1980) The Urban Land Nexus and the

State. London: Pion.Scott AJ (2012) A World in Emergence: Cities and

Regions in the 21st Century. Cheltenham:Edward Elgar.

Scott AJ and Storper M (2015) The nature of cit-ies: The scope and limits of urban theory. Inter-national Journal of Urban Theory 39: 1–15.

Sheppard E (2014) Globalizing capitalism andsouthern urbanization. In: Parnell S and Old-field S (eds) The Routledge Handbook on Cities

of the Global South. London: Routledge, pp.143–154.

Sheppard E, Leitner H and Maringanti A (2013)Provincializing global urbanism: A manifesto.Urban Geography 34: 893–900.

Simone A (2011) The surfacing of modern life.City 15: 355–364.

Simone A (2014) Jakarta, Drawing the City Near.Minneapolis, MN: University of Minnesota

Press.Smith RG (2013) The ordinary city trap. Environ-

ment and Planning A 45: 2290–2304.Soja E and Scott AJ (1986) Los Angeles: Capital

of the late twentieth century. Environment and

Planning D: Society and Space 4: 249–254.Spivak GC (2008) Can the subaltern speak? In:

Nelson C and Grossberg L (eds) Marxism and

the Interpretation of Culture. Basingstoke:Macmillan, pp. 271–313.

Standing G (2011) The Precariat: The New Dan-

gerous Class. London: Bloomsbury.Storper M (2013) Keys to the City: How Econom-

ics, Institutions Social Interaction and Politics

Shape Developnent. Princeton, NJ: PrincetonUniversity Press.

Taylor PJ (2013) Extraordinary Cities: Millennia

of Moral Syndromes World-Systems and City/

State Relations. Cheltenham: Elgar.Tonkiss F (2011) Template urbanism: Four points

about assemblage. City 15: 584–588.Walker R (2015) Building a better theory of the

urban: A response to ‘Towards a New Episte-mology of the Urban?’ City 19: 183–191.

Walker R (2016) Why cities? A response. Interna-tional Journal of Urban and Regional Research

(forthcoming).Wilson WJ (1987) The Truly Disadvantaged: The

Inner City, the Underclass, and Public Policy.Chicago, IL: University of Chicago Press.

Wirth L (1938) Urbanism as a way of life.American Journal of Sociology 44: 1–24.




Documents

Current Debates in Urban Studies - Storper and Scott