Cities, Vol. 22, No. 3, p. 189–199, 2005

� 2005 Elsevier Ltd.

*Fax: +441 DespitespecificityRennie S‘‘global urnetworksof an urbathe degree

All rights reserved.

Printed in Great Britain

0264-2751/$ - see front matter

www.elsevier.com/locate/cities

doi:10.1016/j.cities.2005.03.007

Middle East city networks and the‘‘new urbanism’’Bruce Stanley *

London Middle East Institute, UK

Available online 23 May 2005

h

h

Contemporary analysts of Middle East cities have been slow to take up new urbanism�s chal-lenge to perceive the region�s cities via a social network lens. Yet a tenuous body of work doesexist, most of it employing the network-as-metaphor, with some authors delving more deeplyin exciting and promising ways to examine city networks, their implications, and the resultantregional system of cities. This essay highlights a number of the major network concepts avail-able, including connectivity, centrality, black holes, brokers, levels of analysis, city system anddensity. Contributions by a range of regional specialists are presented in the context of theongoing search for understanding of the evolving nature of city networks in the Middle East.The essay concludes with a discussion of insights network analysis might contribute to ques-tions of the distribution of power, community, hierarchy and change across the longue duree.� 2005 Elsevier Ltd. All rights reserved.

Keywords: City networks, city-system, Middle East, new urbanism, social network analysis

2 Even Peter J. Taylor, when employing the ideas of SNA in hiswork, limited his research to only a few of the options which mightbe utilized for conceptualising cities in networks. See any number

Introduction

As opposed to more traditional approaches to cities,‘‘the new urbanism’’ (Amin and Thrift, 2002) drawsheavily on the terminology of networks. From Sassen(2002) to Short (2004b), the discussion of world citiesin globalization borrows its narrative on transnationalurbanism (Friedmann, 2002) or of the ‘‘networkedcity’’ (Castells, 1996) from a perspective of the city asembedded. This creeping terminological shift drawsmuch of its representation and feel from the image ofcities as nodes connected by flows across space: ‘‘Ifthe city is to survive, process must have the final word.In the end the urban truth is in the flow’’ (Kostof, 1992).

However, this literature has tended to employ alimited set of concepts, with terms left undefinedor clarified.1 Only a few authors specifically drawon the ideas of the well-established social network

-20-88811998; e-mail: bruces@huron.ac.uk.the use of the terminology, a low degree of networkis the norm. For example, in his most recent book, Johnort (2004b, p. 16) makes significant reference to theban network of flows’’, and suggests that ‘‘global urbanave existed in the past.’’ There is, however, no definitionn network, and no network concept employed other thanof connectivity.

189

analysis (SNA) literature.2 Recent literature has dis-covered the limits in this use of network-as-meta-phor (Berkowitz, 1982), and is pushing towardsmore concrete understandings of city networks.3

Within the study of Middle East cities, similartrends and processes are also emerging. During thelast thirty years, scattered undeveloped referencesto city networks appeared in the Middle East urbanliterature, while many urban studies made consistentbut un-explained references to significant ‘‘connec-tions’’, exchanges and flows of goods among the re-gion�s cities. Spurred by the work of the newurbanism, however, the network-as-metaphor

of articles by P.J. Taylor contained on the Global and World Citiessite at Loughborough University at www.lboro.ac.uk/gawc.3 Disagreements arise, as expected, on exactly how this should bedone. Some are pursing the post-structuralist track, employingactor-network theory to make their case (Smith, 2003a). Taylor(2004a,b) and others are refining the world city-network conceptusing service-sector data. Other authors are concretising theconcept of city networks by looking at the physical/telecommuni-cations networks that bind cities together (Graham and Marvin,2001; Townsend, 2001); looking at the longue duree in order tostudy ‘‘world-city networks’’ at the core of the world system(Bosworth, 2000; Modelski, 1999; Rennstich, 2003); or evaluatinginterurban networks in the policy discourse (Leitner and Sheppard,2002).

Middle East city networks and the ‘‘new urbanism’’: B Stanley

terminology has recently become more noticeable,and the application of world systems analysis tothe region is also encouraging more careful consider-ation of city networks (Keyder, 1999). Around theedges, authors are borrowing more robust networkapproaches and applying them to the study of citiesin Middle East and North Africa (MENA). Unfortu-nately, few of the diverse concepts available have yetbeen utilized; the general application to the regionlags far behind the study of city networks in otherworld regions; and the insights into Middle Easturbanism tantalizingly offered by a city network ap-proach remain unexplored. Increasingly, authorsare ‘‘talking the talk’’, but as a community of regionalscholars we are not yet ‘‘walking the walk’’.

This paper is an attempt to contribute to a deeperconceptualization and utilization of network con-cepts in the analysis of Middle East cities, stressingthe general benefits of a more conscious and system-atic approach. Within this short commentary, I dis-cuss a range of network possibilities, and suggestvarious benefits that might accrue from their appli-cation to the region.

Moving networks into the core of the newurbanism

The most consistent body of scholarship examiningnetworks is to be found in social network analysis(SNA), which evolved during the 1950s and 60s asan alternative approach to understanding actors ingroups. Today it is a well-established subfieldexploring the nature and implications of networks.4

Four key concepts together craft an organising prin-ciple on city networks: cities are embedded within asystem of cities that can be evaluated across multiplescales for both their relational characteristics andtheir positional implications.

Cities are embedded

Cities have both existence in Cartesian space(bounded territoriality) and as an imagined con-struct of flows.5 Both should be considered. How-ever, most analysts stress a city�s geographicalgroundedness and walled boundedness, to the exclu-

4 For an introduction to SNA and its concerns, see Wassermanand Faust (1994), Wellman and Berkowitz (1988), Knoke (1990),Berkowitz (1982), or Scott (1991).5 Smith (2001) contrasts the material existence of cities with theirmeaning, which is socially constructed. I would argue that when weexperience ‘‘the city’’ today, most of us link that material existencewith geography and thus ground the city in physical space, but thathow we make that link and what its implications are can differ dueto changing meaning-making practices. Those who set out to talkabout the history or chronological development of a particular cityare thus stressing the space/physical grounding, and then buildingtheir meaning-making analysis on that foundation. Those whostress a constructivist approach to the city, and utilize the networkmotif, however, are backgrounding the material existence andprivileging one particular conceptual metaphor.

190

sion of the flows through which it is articulated. Atthe extreme, this ‘‘autonomous’’ view ignores, down-plays or backgrounds the flows, networks and ex-change aspect of cities (Amin and Thrift, 2002).On the other extreme, authors such as Castells(1996) or Smith (2003) privilege the constructivistnature of the city, repressing the geographical refer-ent in order to highlight the city as ‘‘a space offlows’’. Mitchell (2000, p. 3) goes so far as to cele-brate e-topia, the ‘‘new, network-mediated metropo-lis of the digital electronic era’’.

A third set of authors is finding that it needs toarticulate a mix of both views, stressing a city�sgroundedness as well as contextualizing it within aspace of flows (Knox and Taylor, 1995). FernandBraudel (1979, p. 481), often cited as the father ofthe city-networks perspective, put it this way:

‘‘a town is always a town, wherever it is located, intime as well as in space. I do not mean that all townsare alike. But over and above their distinctive and

original features, they all necessarily speak the samebasic language. . . For a town never exists unaccom-panied by other towns: some dominant, others subor-

dinate or even enslaved, all are tied to each otherforming a hierarchy, in Europe, in China, or any-where else.’’

Social network analysis shares this image, con-trasting its paradigm with the alternative atomisticview that dominates the social sciences (Berkowitz,1982). A network approach, by requiring, at a mini-mum, the specification of the actors (nodes), flows(paths or exchanges) and relationships among them,turns most of social science on its head (Frey, 1978).By shifting the focus in this way, social network anal-ysis seeks to ‘‘reconnect the study of individuals tothe relationships and structures of relationships inwhich they are embedded’’ (Wellman, 1999, p. xiv).

Most urban studies approach cities as autonomousentities, bounded by walls or municipal authority,space to be planned or acted upon, limited in powerand agency, knowable primarily as a geographicpoint in space.6 The MENA city literature is noexception. During the 20th century, this boundedperspective predominated, primarily addressing aparticular city, its history and its evolution as geo-graphically bounded: Beirut as a projected city(Rowe and Sarkis, 1998); the history of Tyre (Flem-ing, 1966); or The Middle East City (Saqqaf, 1987).

Significantly, flows of people (pilgrims into Alep-po), products (dates shipped out of Basra) and ser-vices (hawali) have long been central topics inregional urban analysis. Yet despite such sensitivity,

6 Mitchell (1969, p. 8) addresses these issues in the opening chapterof his seminal edited volume where he contrasts his social networkapproach with the conventional wisdom which saw ‘‘the behaviourof persons . . .largely in terms of their membership in �bounded�groups and their involvement in social institutions.’’ See alsoLaumann et al. (1983).

9 Interestingly, Denoeux�s work (1993) is deeply informed bynetwork analysis, he defines his terms, and is aggressively network-aware. But only within a particular city. There is no considerationof these networks as ‘‘embedded’’ within broader city-networksthat connect the region�s cities, or of these networks as embeddedwithin a global political economy. Thus, in a strange way, thiswork adopts an ‘‘autonomous’’ perspective on cities in the regionwhile still being network savvy.

Middle East city networks and the ‘‘new urbanism’’: B Stanley

authors rarely expressed the city as a set of networksor framed the city as embedded within a broader setof networks or ‘‘space of flows.’’ For example, thework of Andre Raymond (2002), who has alwaysevidenced concern with ‘‘networks’’ and their ef-fects, lacks this reflexive consideration. For Ray-mond, the solidarities and geographic boundednesscome first, and there is little sense of transboundaryties (weak ties) that transcend such communities.Nowhere are ‘‘networks’’ defined or evaluated,although they are, in some mysterious fashion, cru-cial to mobilization.7

Within a system of cities

The new urbanism remains confused over the relatedproblem of the appropriate unit of analysis. So far,proposals have included world city, global city, globalcity-region, world-city system, world city networks,city-states, functional city-systems, and city-systems.We can surely have a debate on where to put the hy-phen, but if we seek a framework for discussing citiesin the global order across the longue duree, then theterm ‘‘city-system’’ seems the more comprehensiveterm. By stressing the accumulated robustness ofmultiple networks among a similar set of nodes acrosstime, city-system is thus a stronger concept than ‘‘citynetworks’’ and incorporates them without implyingtoo much stability. The unit of analysis becomes thearmature emergent from a set of networks that has,over time, taken on the characteristics of a system,making it possible to conceptualize a ‘‘world city sys-tem’’ (Modelski, 1999), ‘‘functional city-system’’ (Loand Yeung, 1997) or ‘‘system of cities.’’ (Skinner,1977; Chase-Dunn and Willard, 1993).8

Some authors writing on the Middle East city haveemployed a form of the city-system perspective.Hodgson (1974, p. 50), for example, conceptualisedthe oikoumene as a ‘‘citied society’’. Hourani (1991,p. 128) offered a region articulated around a ‘‘chainof cities’’ where ‘‘a network of routes ran throughthe world of Islam and beyond it’’ sharing products,ideas, people and culture via mechanisms of ex-change, violence, love and education. These rareexamples adopt ‘‘network-as-metaphor’’ terminol-ogy as well as implying some sense of a broader col-lective within which these urban networks are sited.A few other writers take a similar tack: Denoeux(1993) employed social network analysis to examinethe networks within Middle East cities in his compar-

7 Lapidus (1967), for example, was part of a debate aboutnetworks and associations within the Muslim city and the role ofthe state in controlling urban networks. See also Haneda (1994).8 One author who rigorously sought to apply this approach wasSkinner (1977) in his work on Chinese cities. For Skinner, Chineseurban development prior to the 20th century should be understoodas occurring in eight macroregions, identified primarily on the basisof their physiographic unity. Upon this backcloth, a ‘‘system ofcities’’ emergent from trade flows, becomes the macroregion�sskeletal structure and hierarchy.

ative study of urban unrest.9 Janet Abu-Lughod(1996, p. 187); Abu-Lughod (1991) has long advo-cated expanding our network perspective on the re-gional city, suggesting ‘‘urban studies on the MiddleEast (do not) pay much attention to the way such cit-ies articulate either with the Arab region aroundthem or with the economic and political world systemthat defines their role internationally.’’10 Currently,the most aggressive work on the embeddedness ofMiddle East cities is emerging from those lookingat regional cities in globalisation (Keyder, 1999);those examining networks within cities and how theselink actors transnationally (Huybrechts, 2002); andthose advocating for a city networks or city-systemconceptualization on regional dynamics (Parsa andKeivani, 2002; Stanley, 2005).

What these explorers are doing is the requiredadditional step of locating the city-system, as theunit of analysis or partial network, within its broadercollective or ‘‘total network’’ (Mitchell, 1969). Pro-posed alternative terms for the collective include‘‘mini world systems’’ proposed both by Abu-Lug-hod (1989) and Chase-Dunn and Manning (2001);Hodgson�s (1974) oikoumene; Wilkinson�s (1992),Wilkinson�s (2001), Wilkinson�s (2002) ‘‘civiliza-tions’’; ‘‘worlds’’ as proposed by Alpers (2002);‘‘Braudelian regions’’ of Bin Wong (2003); or Wal-lerstein�s (1974) ‘‘world-system’’. A ‘‘city-system’’fits within all of these as the armature that articu-lates them. Thus, it is possible to specify the MiddleEast as a region not so much on a geographic basis,but as manifestation of the particular city-system(time specific) that emerges from the accumulatednetworks that bind cities.11

That can be evaluated across multiple scales

Centering the network perspective more solidly with-in research on MENA cities requires creative think-ing about the most appropriate scales or levels ofanalysis for assessing urban networks. Unfortunately,

10 Curiously, Abu-Lughod�s latest article on Cairo Abu-Lughod�s(2004) is notable by its lack of a network-aware patina to theanalysis.11 A specification of the ‘‘Middle East and North Africa’’ of 2004seen through the lens of its current city system may have to includeMarseilles and London, given the power and resilience of thenetworks that link populations within these two cities into cities inthe ‘‘region’’; and under certain conceptualizations may not includeTel Aviv. Baghdad and the national city-system in which it sitswere not part of the Middle East for significant periods of the1990s. The city-system of Dar al-Islam as experienced by IbnBattuta incorporated Tangiers, Timbuktu and Zanzibar. See alsoFriedmann (1997).

191

14 It is important to note that taking a functional perspective on apartial network of linked cities can also be extremely useful for

Middle East city networks and the ‘‘new urbanism’’: B Stanley

the theoretical literature tends to reduce our optionsto a dichotomy of local versus global, or componentparts within a city versus a world urban system. Somenetwork-aware authors, however, are offering a moresophisticated representation. For example, Fried-mann (2002) argues for a global view, tempered bya sense of scales including local, national and global.Smith (2003), however, in order to reject structural-ist arguments, goes so far as to suggest that scalesstop being important when we start thinking aboutnetworks and cities.

I disagree. A fuller conceptualisation of scales iscrucial to how we act and think about implications,causality and policy. Levels of analysis are vital heu-ristic devices for both simplifying the chaos of theworld (a total network) and disarticulating it in man-ageable chunks (bounded or partial networks).When we take a network approach to the urban inthe Middle East, we need to situate the city withinan issue and time specific set of nested scales andpossible levels of analysis (Smith, 2001, p. 5).12

The following is an example of possible scalesdrawn from the new urbanism. Although these arenot the only levels that could be identified, they haveconceptual utility as a starting point for analysis.13

� Global urban world (Clark, 1996). A world city-

system plus all the attitudes of urbanism, cognitive

aspects and culture, which makes it operate at

times like a single seamless urban entity.� World city-system (Bosworth, 2000; Knox and

Taylor, 1995). A city-system which incorporates

all the cities on the globe; or it could be delimited

by type or position as incorporating just the ‘‘glo-

bal city’’ system, the world-city system, or the glo-

bal city-region system.

� Regional city-systems. A geographically based des-

ignation of city-systems using recognizable regio-nal designations: a MENA city-system (Stanley,

2003a); the Indian Ocean world city-system of

Chaudhuri (1985); or the Arabian Sea�s port sys-

tem, those ‘‘brides of the sea’’ during the 17th cen-

tury (Barendse, 2002).

12 Fawaz and Bayly (2002, p. 20) make a similar point when theysay ‘‘By forcing us to think beyond the region, it also made us cometo terms with places where the histories of the regions convergedand diverged. It made clear that regional histories could only beone way of writing world history, and that radical histories couldonly be one way of writing world history, and that radicalrelativism, the emphasis on the fragment, the decontextualiseddiscourse, or the shallowly researched representation could neverbe enough.’’13 This schema does tie network flows to an underlying territorialbase to create the boundedness of the scale. Other criteria forconstructing scales are possible.

192

� Sub-regional city-systems (Skinner, 1977). A tem-

porarily bounded and articulated city-systemwhichoperates within a proscribed geographical area,

often determined by flows shaped by geographical

features rather than those shaped by political regu-

latory arrangements: Red Sea city-system; trans-

Saharan city-system; Nile River city-system.14

� National city-systems (Jacobs, 1984). This is the pre-eminent traditional construct, emergent from our

state-centric metageography, for bounding a city-systemusing the current regulatory edgesof the state.

� Local city-systems. Networks of cities that operate

within a smaller, proscribed geographical area,

showing a degree of complementarity and

exchange. For example, the cities of the Fergana

Valley; Najd market towns; Nabataean trading cit-

ies; an Uruk world system (Algaze, 1993).

� Dyads or triads. SNA suggests that it is conceptu-ally fruitful to pay attention to dyads and triads

within networks (Wasserman and Faust, 1994;

Frey, 1978). Thus we might look at two or three

cities linked by multiple exchanges. This could

mean geographically separate units, or merging

conurbations: Cairo – Alexandria; Sijilmassa –

Timbuktu; Hodeida – Cairo Istanbul15; Eilat –

Aqaba – Taba; Jerusalem – Jaffa.� City-region (Scott, 2001). A single ‘‘city’’ and its

hinterland, tightly linked by multiple networks.

A related term is greater metropolitan area[GMA], a collection of administrative units that

often act as a single unitary actor. A political or

regulatory base to the city-region is not required;

the concept of a city�s ecological ‘‘footprint’’, for

example, or its immediate production reach,16 isalso possible: Greater Beirut; the al-Ain/Buraimi

conurbation; Greater Jerusalem; Jabal Nablus17.

analysis: the commodity chains of ‘‘drug cities’’ in the Middle Eastor the textile cities of Irbid and Tulkarem manufacturing via Haifafor Wal Mart; or the functional city chains radiating through theregion inferred by terms such as the Silk Road or Fur Route.15 Linked by the coffee flotillas during the 1650s. Barendse (2002,p. 173).16 ‘‘The directors of the Company (VOC) . . .like spiders in thecentre of a large web of runners and regular correspon-dents. . .(were) linked (to) the factory with a large number of out-and up-country stations and shops for the procurement of goods.’’(Barendse, 2002, p. 168).17 Beshara Doumani (1995, p. 1) puts it very nicely in his‘‘Introduction’’, when he says: ‘‘During the eighteenth and mostof the nineteenth centuries, the city of Nablus was Palestine�sprincipal trade and manufacturing centre. It also anchored dozensof villages located in the middle of the hill regions which stretchednorth-south from the Galilee to Hebron and which were home tothe largest and most stable peasant settlements in Palestine sinceancient times. . .(we want to explore) how merchants built andreproduced local and regional networks. . .’’.

Middle East city networks and the ‘‘new urbanism’’: B Stanley

� ‘‘The city’’ (Mumford, 1961). Bounded by any

number of possible variables or concepts. Histori-cally, the city was defined by its walls.18 Today, in

the way the ‘‘city’’ actually works, analysts are

finding that the geographical referent is no longer

sufficient, although the power of the regulatory/

territorial construct remains strong, as does our

socialization into the concept.19

� ‘‘Communities’’. Wellman�s (1999) emphasis on

loosely bounded, sparsely knit, specialized socialnetworks is very different than the majority of

the literature on MENA neighbourhoods or quar-

ters. Traditional literature employs strongly terri-

torially-based concepts, reifying solidarities based

on attributes. New studies on MENA urban com-

munities demonstrate greater network awareness,

starting with various types of linkages and follow-

ing them where they lead (Al-Ali, 2000; Spellman,2004).

� ‘‘Ego networks’’ (Knoke, 1990). Here the analyst

examines particular firms, NGOs, individuals, or

cities as the focal node from which to trace net-

works outward. Thus, an analysis of the way a

particular bank in Beirut ‘‘attends’’ to the redevel-

opment of Beirut�s world city network evaluates

the connections and implications of their net-works: i.e. its centrality, density, extensiveness,

range, etc (Beaverstock et al., 2002; Huybrechts,

2002). The reach of Cairo under the Fatimids;

the Armenian role in Aleppo�s power; the Mayor

of Tehran�s role in Metropolis would be examples.

For both the characteristics of their relations and theirpositional implications

A half-century of SNA literature offers those inter-ested in adopting a network-informed approach awide range of concepts from which to draw. Alterna-tives fall into two primary categories: those con-cerned with characteristics of the relationshipsamong the actors within the network; and those thatattempt to summarize the nature of those relation-ships into a set of structural or positional types basedon the pattern of their ties.Relational concerns look for the nature of the ties

that bind nodes into networks. One of the most sig-nificant concepts is connectivity, since it determinesboth a system�s ‘‘structure and dynamical proper-ties’’ (Watts, 1999, p. 240). Whether it is through‘‘weak ties’’(Granovetter, 1973) or ‘‘bridges’’ (Wass-erman and Faust, 1994, p. 114), nodes and sub-

18 The Sumarian word for city and wall were the same.19 Janet Abu-Lughod holds the view that the concept of the ‘‘city’’may no longer be relevant. Personal discussion with the author.Also see her discussion of ‘‘activity patterns’’ in Abu-Lughod(1991).

networks are knitted together. Recent work on‘‘small worlds’’ has investigated the ‘‘short cuts’’ or‘‘very long-range global edges, which connect other-wise distant parts’’ (Watts, 1999, p. 241). Conversely,a lack of connectivity can leave nodes isolated andtalking only to their immediate neighbours (Smithand Timberlake, 1995, 2001, 2002). Recent studiesby Taylor (2001) and Shin and Timberlake (2000)demonstrate that Middle East cities are generallypoorly linked among themselves, and the links whichdo exist, by tying cities like Tel Aviv, Dubai andIstanbul more to the European or Asian city-systemsthan to each other, distort regional interaction andintegration.

A concern with connectivity also encourages a fo-cus on vulnerability and the ‘‘critical nodes’’ and‘‘critical paths’’ that stabilize a group. Some citiesand linkages within those cities may carry more ofthe responsibility for linking sub-networks togetherin the Middle East than others, and if they are‘‘lost’’, mini-circuits are cut off. For example, duringthe 1990�s, the Baghdad city-system (Mosul, Tikrit,Basra, etc.) was only connected into the world polit-ical economy through three critical paths: Baghdadto Aqaba; the desert route to Damascus; and Mosulto Istanbul. Without those links, oil, medicines,funds did not flow. When Beirut suffered its civilwar, nodes in its city-system had to find other routes,via Damascus or Latakia, in order to acquire goodsand services. Some cities even went independent(Juniyah) and forged new ties to Cyprus-Europeon their own. Beginning with ancient Ugarit, MiddleEast urban history is replete with the destruction ofcertain cities or sets of cities whose demise broughtdown their entire first order network. The physicalloss of certain nodes or paths among cities due toearthquake (Bam or Atlantis) or the intentionalattacking of cities in war (the Israeli closure policyon Palestinian cities, or Churchill�s attack on Galli-poli to close Istanbul) represent similar strategiesof attacking critical nodes or shutting down criticalpaths (Stanley, 2003b).

One of the more interesting approaches to con-nectivity is Burt�s (1992) concept of ‘‘structuralholes.’’ Burt argues that social networks, as a resultof being shaped by who is connected to whom, willcontain ‘‘structural holes’’: situations of disconnec-tions or gaps in the armature. Such chasms can sep-arate, but when bridged, are value added to actorsand networks, since actors who broker across thesegaps gain ‘‘social capital.’’ In the region, many citieshave made great advantage over their sister cities viatheir connectivity: Tyre won out over the otherPhoenician cities due to its stronger connections toKition. Az-Zubair�s combination of autonomy andfar-flung linkages allowed it to withstand Ottomanintrusions for 60 years (Fattah, 1997). Mecca cameto dominate the pre-Islamic cycle of markets aroundthe Arabian Peninsula. Looking to the future, whichMiddle East cities are filling in structural holes along

193

Middle East city networks and the ‘‘new urbanism’’: B Stanley

the new global ‘‘digital trade route’’ of the 21st cen-tury? (Rennstich, 2003, p. 17) In fact, a similar ques-tion underlies the work of Keivani et al. (2003) incomparing Singapore and the cities of the UAE.

The concept may also enlighten our understand-ing of Iraqi urban dynamics. The cities of Iraq havebecome such zones of insecurity and chaos that thecapital of Iraq is actually Amman: both the Iraqigovernment and civil society in reality function outof the Jordanian capital.20 Without an articulatedconcept of structural holes, that examines the evolv-ing nature of the post-1990 sanctions regime, thedestruction of Baghdad�s networks, the shift to Am-man, and the power of Washington, how can wehope to understand, much less improve, the every-day reality of urban life in the region? Our analysisand praxis are left to flounder in the misperceptionof Baghdad as ‘‘the capital of Iraq’’, all the whilemissing the reality of its evolved isolation withinthe regional urban network.21

There are many other relational concepts thatmight be usefully employed in the study of theMENA city; I will cite two as illustrative. Centralityrefers to an actor�s prominence or importance withina network (Wasserman and Faust, 1994). Funda-mentally, Sassen (1991) and Friedmann�s (1986)ideas of C2 (command and control) housed in afew global cities rests on the concept of centrality(Sassen, 1995). Taylor�s hierarchies Taylor�s(2004a) indicate that Middle East cities have a lowdegree of centrality in the world city network.Improving centrality appears to be one of the moti-vations driving city elites to attract the Olympics toIstanbul or to host the NAM in Tripoli. It certainlymotivates Dubai to establish a Media City, an Inter-net City and to host the second largest arms fair inthe world. Djibouti, Cairo and Amman are all con-cerned ‘‘not to lose out’’ in the race for regionalIT centrality. Such ‘‘globalising projects’’(Short,2004b, p. 7) seek to increase centrality along certainspecialized circuits assumed to be critical forenhancing broader connectivity into the world citysystem, while often ignoring the importance of localnetworks for the city�s vitality.22

The relative number of actual connections amongactors in a network compared to the possible total(density) has long been of interest to urban scholars.Massey (1999, p. 160) argues that density gives citiestheir ability to foster creativity, and the level of den-

20 The global hotels of Amman are so full of diplomatic personnel,government ministers, INGO representatives, and private securityconsultants that there are no rooms available; security trainingtakes place not in Baghdad, but in Amman; consultations amongIraqi domestic political leaders occur in the hotel suites of Amman;business decisions concerning importation into Mosul or Basra aretaken over breakfast in the restaurants of Amman hotels.21 Janet Abu-Lughod (1991) implies some of this, albeit at adifferent scale, in her discussion of paths through physical space.22 Beirut is a good example of this dilemma. See Kassab (1997),Rowe and Sarkis (1998) and Trawi (2003).

194

sity effects network cohesion and stability, thusimproving resistance to attack.23 It would be inter-esting to see how this works in mini-circuits suchas the Philistine cities or the Phoenician city-systemwhere high density may have facilitated the overallstability of the system. We could postulate that thedysfunction and conflict in the current Persian Gulfcity-system or between Palestine and Israel mayhave something to do with low number of linksamong their cities, and that increasing such linkscould change the conflict spiral.24 Certainly, thearguments made by American policy makers duringthe post-Oslo period for a Gulf of Aqaba Develop-ment Zone were based on the belief that simple den-sity among cities would facilitate peace.25

From the earliest writings on SNA, the belief thatpositions have consequences has been seminal, gen-erating a concern with the deeper topography ormorphology of structures, as manifested both in par-ticular structural positions (stars, hubs, dead ends,brokers, bridges, structural holes, etc.) as well aswith the implications of the relative distribution ofsuch positions (a lack of brokers; too many powercentres; small worlds; etc: Frey, 1978; Wassermanand Faust, 1994).

The network aware literature on Middle East cit-ies has generally failed to take up the challenge ofpositional analysis. There are a few exceptions: Tay-lor (2001) and Felsenstein (2001) are notable, as isKipnis� (2001) application of dead ends to the studyof Tel Aviv. These are nodes at the end of paths thathave high input/output into the broader world citynetwork, but have low transboundary linkages with-in their own region: a �cul-de-sac� for world traffic.The implications of low local connectivity includean inability to reach others in their neighbourhoodto accomplish their goals, and difficulty in craftinga future for themselves as a local �anchor�. Tel Aviv�selites had intentionally crafted close ties into globalhigh-tech networks but as a dead end, they were sig-nificantly damaged by the dot.com meltdown (Kip-nis, 2001).Brokers inhabit the space between partial net-

works, bridging gaps in return for fees (Wassermanand Faust, 1994). The regional urban literature haslong argued that port cities do the heavy lifting inthe Middle East for linking diverse circuits of cities(Reimer, 1991); unfortunately few authors have sys-tematically deconstructed their observations aboutthis phenomenon (McPherson, 2002). We havemany possible examples (Basra for Baghdad; Istan-

23 Massey (1999, p. 160) says ‘‘What makes �spatial configurationsgenerative� are the intense social effects resulting from �densenetworks of interaction� within them.’’24 Baruch Kipnis has argued this point in a conversation with theauthor.25 Unfortunately, simply having multiple contacts does not sayanything about the quality of those links nor about what flows overthem; hence density is actually a crude measure of a network�sintegration, and should be used with caution.

Middle East city networks and the ‘‘new urbanism’’: B Stanley

bul for Ankara; Adulis for Aksum) of geographi-cally close brokers and their clients; we shouldremember as well that cities may broker far outsidetheir immediate vicinity (Venice for Constantinople;Zanzibar for Muscat). Other actors may also brokerfor city networks: such ‘‘attendants’’ may play a keyrole in maintaining connections among world citiesduring crisis (Beaverstock et al., 2002).

The positional concept of black holes offersintriguing scope for analysis. The term was first pro-posed by Short (2004a) who argued that there arenumerous city nodes that we would expect, due tosize, location or conventional wisdom, to be ‘‘worldcities’’ but which are not. They are black holes onthe global connectivity circuit. Short postulates anumber of reasons for this: poverty, economic orpolitical collapse, exclusion and resistance. He endshis argument by calling for greater attention to the‘‘silences and voids’’ in city networks and for a pushto understand case studies of less-connected cities(Short, 2004b, p. 47). Among the black holes dis-cussed are Baghdad, directly excluded from the glo-bal political economy, and Tehran, which hasintentionally resisted tying in.26 Within the litera-ture, one finds other alternative understandings ofthe idea: as a node with evil or malevolent inten-tions27; or as protected, rogue sub-networks, inwhich certain flows or exchanges are intentionallyhidden (Ardelean, 2002). It is this concept of theblack hole that drives much of the anti-terrorismwork in and around Middle East cities, as well asfeeds conspiracy theories.

Black holes may also be created by removing crit-ical nodes or by constructing alternative paths. Plac-ing a city under siege or destroying it has profoundcity-network implications: Roman development ofan alternative sea route down the Red Sea turnedMeroe�s Nile River city-system into a black hole;the French Mandate cut Aleppo�s extensive ties toAnatolia, creating a black hole in northern Syria thattook the city (and the Syrian city-system) years toovercome. Perhaps the whole MENA regional city-system is a ‘‘black hole’’? Considering the poor flowsof FDI, disconnects in transport and key infrastruc-tural linkages, or the flows of intraregional asopposed to interregional trade, the MENA city-system might be understood as a black hole, poorlyconnected into the global political economy, lackingglobal cities other than Istanbul to articulate its net-works, and locked into dependent or trans-regional

26 Short misses a key point about Tehran and its extensiveparticipation in global urban organisations like Metropolis. Thisdemonstrates that Tehran has selectively tied in rather than simplyresisted tying into the global urban networks. See Stanley (2004).27 Recent work in the protection of computer networks fromattack has taken this definition. One recent study focused onprotecting wireless ad hoc networks such as those used by the USmilitary to give them field C3 (command, control and communi-cation) advantage from attack by ‘‘cooperative black hole attack.’’(Ramaswamy et al., 2003).

(asymmetric) relationships. The MENA city-systemis thus ‘‘off the map’’, doing little better than Africacity circuits, in global connectivity or centrality. Thismay be a key regional challenge of the 21st century:to build a better urban structural armature throughwhich economic and political development mayflourish (Stanley, 2003b; Taylor, 2004a).

Observing Middle East cities through the socialnetwork lens

When we approach the Middle East city from a net-work perspective, what issues can be usefullyre-examined in new ways? Out of any number ofoptions, let me present three: the question of power;that of community; and changes over the longueduree.

How is power distributed and applied?

As students of the Middle East city, we are drawn toissues of power. We find that we are concerned withwhat it is, who has it, how actors employ it, and whatgoals they are attempting to achieve. A network ap-proach enhances such analysis for four key reasons.First, because network analysis sees power as a rela-tionship and not as a thing, power is inherent in theinteraction of actors embedded within relations,rather than in attributes, hardware, or ‘‘capabili-ties’’. This means that power is issue and time spe-cific, forcing us to look for the ‘‘paradoxes’’ ofpower (Baldwin, 1989) that appear due to resistance,new capabilities, new social capital, etc. Thus, as op-posed to traditional analysis that stresses bounded-ness, essentialist solidarities, and eternal powercapabilities, the agency inherent in bridging struc-tural holes, constructing transnational networks(Smith, 2001), or in ‘‘going global’’ (Stanley,2003a) is privileged. The paradoxes of power be-come exposed,28 and the agency of new actors a fo-cus of interest (Ossman, 1994).

Secondly, it becomes possible to witness a cate-gory of events we might term ‘‘wars of control’’: con-flict among different actors to dominate resourcesand revenues, set regulation, or determine security.Sometimes such �space wars� involve the struggle toincorporate resistant cities into an evolving territo-rial state or empire (Taylor, 2000): the Saudi stateproject vs. Jeddah or al-�Asa; the revolt of Tyreagainst the Assyrians; or President Asad crushingresistance in Hamah. Other times, they are conflictsfor control over critical paths or nodes (Bosworth,2000): the struggle between the Byzantines and the

28 For example, Najaf�s citizens rose up against the OttomanTurkish soldiers within their city, kicked them out, raised the Arabflag and declared themselves a free city in 1914; they retained theirautonomy until 1917. Such historical events, particularly in light ofthe August 2004 uprising in the city, take on a new meaning, andpush us to ask questions about why certain cities have a history ofresistance while others do not.

195

Middle East city networks and the ‘‘new urbanism’’: B Stanley

Safavids over the Yemeni city-system; the projectionof Moroccan power towards Timbuktu in 1591; orEthiopia�s decision to ship its products via Port Su-dan rather than via Massawa. And sometimes thesespace wars are about global capital and its neo-liberal restructuring of the cities of the region(Owen, 1998; Oncu and Weyland, 1997); about therevanchist city struggling to impose social controlover those excluded by the new growth policies of‘‘the competitive city’’ (Smith, 1996); or concerningthe nature of meaning inherent in symbolic buildingsor quarters of the city (Yacobi and Fenster, 2004).Being sensitive to flows and patterns of structurealso allows us to examine overall distributions andhierarchies that emerge over time. This is what Tay-lor (2004a) is stressing in his work on the world citynetwork, as is Massey�s (1993) focus on ‘‘power-geometry.’’29

Finally, a concern with the distribution of powermay lead us to the ‘‘power law’’, or 80/20 Rule(Watts, 1999), which suggests that, due to ‘‘preferen-tial attachment’’ (Shirky, 2003; Klemm and Eguiluz,2002), power will evolve to those few cities which al-ready have power (Adamic et al., 2002). This ten-dency toward unequal distribution may helpexplain why certain ‘‘phoenix’’ cities (Ashdod; Nip-pur; Damascus) reappear within key circuits forthousands of years. And it raises the research ques-tion as to why other cities (Ugarit; Ebla; Memphis;Sijilmassa; Adulis) die, never to be reborn.30

What is the basis of community?

Approaching the region via networks also directs usto the ‘‘Community Question’’, which stands ‘‘at acrucial nexus between societal and interpersonal so-cial systems.’’ (Wellman, 1999, p. 2) Traditional ur-ban literature often identifies ‘‘communitysolidarity in terms of shared values and social inte-gration’’ where it does not exist; misses translocalties; stresses order and stability; and is wrapped upin normative integration and the persistence of con-sensus. Network approaches, on the other hand,allow ‘‘the discovery of other forms of community–perhaps sparsely knit and spatially dispersed—andother forms of organization—perhaps loosely cou-pled or virtual’’ (Wellman, 1999, p. 17). This para-digm can make a contribution to our search for the‘‘solidarity’’ which holds the Middle East togetheras a region. At its heart, the question ‘‘what is theMiddle East?’’ is a community question. Caught be-tween an image of a region of bounded states, and

29 Numerous authors have addressed the distribution of powerwithin cities of the region, although few have consciously employedthe network terminology (Keyder, 1999). Abu-Lughod (1989)touched on the question of regional city hierarchy, but to myknowledge, no one has examined shifting hierarchies over time justfor the Middle East.30 Both Peter J. Taylor and Janet Abu-Lughod suggested this pointin separate discussions with the author.

196

the mosaic of essentialist religious solidarities, weneed alternatives. A network perspective with citiesas nodes transmutes ‘‘the region’’ into a city-system,which frees us to see solidarities as nested, evolving,and contingent. This may empower the rejection ofthe simplistic dichotomy between the global andthe local that underlies both so much of the discus-sion about cities and globalization, as well as thatabout the neighbourhood vs. the state.

How do city-systems and city hierarchies evolve?

Network analysis warns us that both the structuralmorphology of networks evolves over time, as doesthe hierarchy of various nodes. Thus, conclusionsabout cities and city-systems must be informed by alonger-term view of network evolution, what Braudel(1979) called the longue duree. Abu-Lughod (1989)was one of the first to take up this challenge in study-ing the region, arguing that analysis must take ‘‘intosufficient account the changing shape of the worldsystem over the long historical time’’ (Abu-Lughod,2001, p. 3). Such an attitude also informs the recentwork of Fawaz and Bayly (2002, p. 2):

The decision to take into consideration the longuedure reflects the conviction that such an approachwould clarify continuity and change from the seven-

teenth to the nineteenth centuries, particularly as thecities of the Red Sea region did not undergo the samedegree of change as the cities of the Mediterranean

and Indian Ocean.

We find a similar approach in Eldem, Goffmanand Masters� study of Ottoman cities Eldem et al.(1999, p. 212) and in Rennstich�s placing of thePhoenician Trade Network System squarely withinthe longue duree (Rennstich, 2003, p. 7). Modelski(1999) pushes the argument back to 4,000 BC, look-ing for the centrality of Middle East city networkswithin the nascent world system. Others make theUruk trading network the heart of the first ‘‘world-system’’(Algaze, 1993), or consider the evolving citysystems of the Indian Ocean (Chaudhuri, 1985) orthe Persian Gulf (Stanley, 2005). Analysis alongthese lines opens up fundamental questions of regio-nal power, continuity and structural transformation.

Empowering a network perspective on middleeast cities

The Middle East/North Africa has the longest his-tory of urbanization of all the world�s regions. Itscultures and religions are urban based, and the re-gion�s political entities, whether empires or states,have always been grounded in cities as the primarysource of wealth. Its cities are sites for violenceand war, but are also nodes in webs of resistanceand conflict resolution. New approaches to theregion itself (civil society, Islamist movements, gen-der) plus new paradigms emerging within social sci-

Middle East city networks and the ‘‘new urbanism’’: B Stanley

ence, suggest that a fresh perspective on the region�scities, their networks and their place in the worldsystem would be a useful (re)conceptualization.Likewise, within the region, forces such as urbaniza-tion, globalisation, migration, democracy andregionalism are reshaping political dynamics withprofound implications for the embedded city.

The region�s city system is experiencing greatdifficulties and challenges. It is poorly integrated,having more interregional connections than intrare-gionally. Across a range of sectors, the MENA city-system plays little role in the urban world. There aremany black holes and deadends, too little brokerage,not enough global cities, few weak ties, low density,poor reachability, limited isomorphism, strong pri-macy, and hierarchy. There is a high degree of com-petition among cities to move up the regional andglobal hierarchy, and preferential attachment isincreasing the centrality of the few. Cities, municipalauthorities and city-systems are consistently over-shadowed by the state, and there is limited decen-tralization. Reciprocity is low within, among andthrough cities for communities, or for cities withthe state, and political space is tightly controlled.Local, national and international actors share ametageography that disempowers MENA citiesand city-systems, excluding the urban from deci-sion-making. In sum, the MENA city-system doesnot play its potential role as an engine for change,and the region suffers as a result.

Despite such difficulties, there is much aboutwhich to be positive: municipal foreign policy is onthe increase (Magnusson, 1996), both in conjunctionwith and as resistance to state policy; global trendsappear to be opening up options for cities as actorsand sites for agency; MENA cities are beginning towork in greater coordination among themselvesand with other cities globally; new types of cross-border connections such as growth corridors areemerging that suggest dynamic alternatives; the‘‘fierce state’’ is beginning to work with cities as partof its developmental agenda as it ‘‘rearticulates andreterritorialises’’ itself in relation to the worldly cit-ies in its midst (Brenner, 1998).

Analytically, the network-as-metaphor approachto understanding Middle East and North African cit-ies has proved to be a useful first stage in the re-con-ceptualisation of issues, trajectory and dynamics.The task ahead is to become more sophisticated inour use and application of network concepts acrossmultiple scales and levels of analysis. Analytically,we must be more comparative, both across the lon-gue duree and across sector, function and sub-region.

Ultimately, we need to rediscover what CharlesTilly (2002, p. 72) termed relational realism, the doc-trine which believes that ‘‘transactions, interactions,social ties, and conversations constitute the centralstuff of social life’’. When we do, we may then beable to re-craft the MENA city-system as a keyarmature for regional empowerment, and give pur-

pose to the discussion of networks within the newurbanism.

Acknowledgements

The author would like to thank Relli Shechter andHaim Yacobi for their helpful comments on thedraft, and for their organization of the conferencefrom which this paper emerged.

References

Abu-Lughod, J (1989) Before European Hegemony. Oxford Uni-versity Press, London.

Abu-Lughod, J (1991) Changing Cities. Harper Collins, NewYork.

Abu-Lughod, J (1996) Urbanization in the arab world and theinternational system. In The Urban Transformation of theDeveloping World, (ed.) J Guger. Oxford University Press,Oxford.

Abu-Lughod, J (2001) Middle East-Asian Connections: Before,During and After Western Hegemony, Talk to the Center forMiddle East Studies, UC Santa Barbara, March, 2001,www.cmes.ucsb.edu/meastsasia/janetl.html (accessed 3/02).

Abu-Lughod, J (2004) Cairo: too many people, not enough land,too few resources. In World Cities Beyond the West, (ed.) JGugler. Cambridge University Press, Cambridge.

Adamic, L, Lukose, R, Puniyani, A and Huberman, B (2002)Search in power-law networks. Physical Review E 64,046135.

Al-Ali, N (2000) Secularism, Gender and the State in the MiddleEast. Cambridge University Press, Cambridge.

Algaze, G (1993) The Uruk World System. The University ofChicago Press, Chicago.

Alpers, E (2002) Imagining the Indian Ocean World, OpeningAddress to the International Conference on Cultural Exchangeand Transformation in the Indian Ocean World, UCLA.

Amin, A and Thrift, N (2002) Cities: Reimagining the Urban.Polity Press, Malden, MA.

Ardelean, D (2002). Tracking routing black holes with the RIS,Available from <http://scss.cs.utt.ro/2002/1.pdf>.

Baldwin, D (1989) Paradoxes of Power. Basil Blackwell, NewYork.

Barendse, R J (2002) The Arabian Seas. M.E. Sharpe, London.Beaverstock, J V, Doel, M A, Hubbard, P J and Taylor, P J (2002)

Attending to the world: competition/co-operation and co-efficiency in the world city network. Global Networks 2, 2.

Berkowitz, S (1982) An Introduction to Structural Analysis.Butterworths, Toronto.

Bin Wong, R (2003) Between Nation and World: BraudelianRegions in Asia. Review Fernand Braudel Center 26, 1.

Bosworth, A (2000) The Evolution of the World-City System: 3000BCE to AD 2000. In World System History, (eds.) RDenemark, J Friedman, B Gills, G Modelski. Routledge,New York.

Braudel, F (1979) The Structures of Everyday Life. In Civilizationand Capitalism. Civilization and Capitalism, vol. 1. Harper &Row, New York.

Brenner, N (1998) Global cities, glocal states: global city formationand state territorial restructuring in contemporary Europe.Review of International Political Economy 5, 1.

Burt, R (1992) Structural Holes. Harvard University Press,Cambridge, MA.

Castells, M (1996) The Rise of the Network Society. In TheInformation Age: Economy, Society and Culture. The Infor-mation Age: Economy, Society and Culture. Blackwell, Cam-bridge, MA.

Chase-Dunn, C and Willard, A (1993) Systems of Cities andWorld-Systems: Settlement Size Hierarchies and Cycles ofPolitical Centralization, 2000 BC–1988 AD. Working Paper

197

Middle East city networks and the ‘‘new urbanism’’: B Stanley

17, www.jhu.edu:80/~soc/pcid/papers/17/pcidpap17.html (acce-ssed 1/00).

Chase-Dunn, C and Manning, S (2001) City systems and worldsystems: four millennia of city growth and decline. InternationalStudies Quarterly.

Chaudhuri, K N (1985) Trade and Civilisation in the Indian Ocean.Cambridge University Press, Cambridge.

Clark, D (1996) Urban World/Global City. Routledge, London.Denoeux, G (1993) Urban Unrest in the Middle East: A Compar-

ative Study of Informal Networks in Egypt, Iran and Lebanon.State University of New York Press, Albany, NY.

Doumani, B (1995) Rediscovering Palestine. University of Califor-nia Press, Berkeley.

Eldem, E, Goffman, D and Masters, B (1999) The Ottoman Citybetween East and West. Cambridge University Press,Cambridge.

Fattah, H (1997) The Politics of Regional Trade in Iraq, Arabia, andthe Gulf 1745–1900. State University of New York Press,Albany.

Fawaz, L , Bayly, C A (eds.) (2002) Modernity and Culture.Columbia University Press, New York.

Felsenstein, D (2001) New agglomerations of technologicalactivity—anchors or enclaves. Paper presented at the AnnualResidential Conference, IGU Commission on the Dynamics ofEconomic Spaces, Turin, Italy.

Fleming, W (1966) The History of Tyre. AMS Press, New York.Frey, F (1978) The Analysis of Social Structure. Updated version

of paper presented at World Congress of Sociology, Uppsula,Sweden.

Friedmann, J (1986) The world city hypothesis. Development andChange 17.

Friedmann, J (1997) Cities Unbound: The Intercity Network inthe Asia-Pacific Region, Discussion Paper Series 23, www.une-sco.org/most/friedman.htm (accessed 11/00).

Friedmann, J (2002) The Prospect of Cities. University ofMinnesota Press, Minneapolis.

Graham, S and Marvin, S (2001) Splintering Urbanism: NetworkedInfrastructures, Technological Mobilities and the Urban Condi-tion. Routledge, London.

Granovetter, M (1973) The strength of weak ties. AmericanJournal of Sociology 78.

Haneda, M (1994) Introduction: an interpretation of the conceptof the ‘‘Islamic City. In Islamic Urban Studies: HistoricalReview and Perspectives, (eds.) M Haneda, T Miura. KeganPaul International, London.

Hodgson, M (1974) The classical age of Islam. In The Venture ofIslam. The Venture of Islam, vol. 1. University of ChicagoPress, Chicago.

Hourani, A (1991) A History of the Arab Peoples. Warner Books,New York.

Huybrechts, E (2002) Beirut: building regional circuits. In GlobalNetworks, Linked Cities, (ed.) S Sassen. Routledge, New York.

Jacobs, J (1984) Cities and the Wealth of Nations. Random House,New York.

Kassab, S (1997) On two conceptions of globalisation: the debatearound the reconstruction of Beirut. In Space, Culture andPower, (eds.) A Oncu, P Weyland. Zed, London.

Keivani, R, Parsa, A, Sim, L L, Ong, S E and Younis, B(2003) Going Global: a comparison of Singapore and thecities of UAE. GaWC Research Bulletin 110, Availablefrom <www.lboro.ac.uk/gawc/rb/rb110.html>.

Keyder, C (ed.) (1999) Istanbul: Between the Global and theLocal. Roman and Littlefield, Lanham.

Kipnis, B (2001) Tel Aviv, Israel – a world city in evolution: urbandevelopment at a deadend of the global economy. GaWCResearch Bulletin 57, Available from <www.lboro.ac.uk/gawc/rb/rb57.html>.

Klemm, K and Eguiluz, V (2002) Highly clustered scale-freenetworks. Physical Review E 65, 036123.

Knoke, D (1990) Political Networks. Cambridge, CambridgeUniversity Press.

Knox, P , Taylor, P J (eds.) (1995) World Cities in a World System.Cambridge University Press, Cambridge.

198

Kostof, S (1992) The City Assembled: The Elements of Urban Formthrough History. Thames and Hudson, London.

Lapidus, I (1967) Muslim Cities in the Later Middle Ages. HarvardUniversity Press, Cambridge, MA.

Laumann, E O, Marsden, P and Prensky, D (1983) The boundaryspecification problem in network analysis. In Applied NetworkAnalysis, (eds.) R Burt, M Minor. Sage Publications, BeverlyHills, CA.

Leitner, H and Sheppard, E (2002) �The City is Dead, Long Livethe Net�: harnessing European interurban networks for aneoliberal agenda. Antipode, 495–518.

Lo, F C and Yeung, Y M (1997) Emerging World Cities in PacificAsia. United Nations University Press, Tokyo. Available from<www.unu.edu/unupress/unupbooks/uu11ee/uu11ee03.html>(accessed 1/00).

McPherson, K (2002) Port cities as nodal points of change. InModernity and Culture, (eds.) L Fawaz, C A Bayly. ColumbiaUniversity Press, New York.

Magnusson, W (1996) The Search for Political Space. University ofToronto Press, Toronto.

Massey, D (1993) Power geometry and a progressive sense of place.In Mapping the Futures: Local Cultures, Global Change, (ed.) JBird, et al. Routledge, London.

Massey, D (1999) Cities in the World. In City Worlds, (eds.) DMassey, J Allen, S Pile. Routledge, London.

Mitchell, J C (1969) The concept and use of social networks. InSocial Networks in Urban Situations, (ed.) J C Mitchell.Manchester University Press, Manchester.

Mitchell, W (2000) e-topia. MIT Press, Cambridge, MA.Modelski, G (1999) Ancient World Cities 4000–1000 BC:

Centre/Hinterland in the World System. Global Society13(4).

Mumford, L (1961) The City in History. Harcourt, San Diego.Oncu, A , Weyland, P (eds.) (1997) Space, Culture and Power.

Zed Press, London.Ossman, S (1994) Picturing Casablanca. University of California

Press, Berkeley.Owen, R (1998) The Metamorphosis of Cairo�s Medan al-Tahrir

as Public Space: 1870–1970. Harvard Middle Eastern andIslamic Review 4, 1–2.

Parsa, A and Keivani, R (2002) The Hormuz Corridor: building across-border region between Iran and the United Arab Emir-ates. In Global Networks, Linked Cities, (ed.) S Sassen.Routledge, New York.

Ramaswamy, S, et al. (2003) Prevention of cooperative black holeattack in wireless ad hoc networks, www.cs.ndsu.nodak.edu/~ramaswam/BlackHole.pdf (10/28/03).

Raymond, A (2002) Urban networks and popular movements: InCairo and Aleppo. In Arab Cities In the Ottoman Period:Cairo, Syria and the Maghreb, (ed.) A Raymond. Ashgate,Burlington, VT.

Reimer, M (1991) Ottoman-Arab seaports in the 19th century:social change in Alexandria, Beirut and Tunis. In Cities in theWorld-System, (ed.) R Kasaba. Greenwood Press, Westport,CT.

Rennstich, J (2003) Naval Ports and Virtual Nodes: the HistoricalEvolution of World-City Networks in the Global WorldSystem. Paper presented at ISA, Portland OR,www.rennstich.com.

Rowe, P , Sarkis, H (eds.) (1998) Projecting Beirut. Prestel,London.

Saqqaf, A (ed.) (1987) The Middle East City. Paragon House, NewYork.

Sassen, S (1991) The Global City: New York, London, Tokyo.Princeton University Press, Princeton.

Sassen, S (1995) On concentration and centrality in the global city.In World Cities in a World System, (eds.) P Knox, P J Taylor.Cambridge University Press, Cambridge.

Sassen, S (ed.) (2002) Global Networks, Linked Cities. Routledge,New York.

Scott, A J (2001) Global City-Regions. Oxford University Press,Oxford.

Scott, J (1991) Social Network Analysis. Sage, London.

Middle East city networks and the ‘‘new urbanism’’: B Stanley

Shin, K and Timberlake, M (2000) World cities in Asia: cliques,centrality and connectedness. Urban Studies 37(12).

Shirky, C (2003). Power Laws, Weblogs and Inequality 65.Available from <www.shirky.com/writings/powerlaw_web-log.html> (accessed 3/24/03).

Short, J R (2004a) Black holes and loose connections in theglobal urban network. GaWC Research Bulletin 76,Available from <www.lboro.ac.uk/gawc/rb/rb76.html>.

Short, J R (2004b) Global Metropolitan. Routledge, London.Skinner, G W (1977) The City in Late Imperial China. Stanford

University Press, Stanford.Smith, D A and Timberlake, M (1995) Conceptualising and

Mapping the Structure of the World System�s City System.Urban Studies 32.

Smith, D A and Timberlake, M (2001) World city networks andhierarchies, 1977–1997: an empirical analysis of global air travellinks. American Behavioral Scientist 44(10).

Smith, D A and Timberlake, M (2002) Hierarchies of dominanceamong world cities: a network approach. In Global Networks,Linked Cities, (ed.) S Sassen. Routledge, New York.

Smith, M P (2001) Transnational Urbanism. Blackwell, Oxford.Smith, N (1996) The New Urban Frontier: Gentrification and the

Revanchist City. Routledge, London.Smith, RG (2003a) Actant Cities, GaWC Research Bulletin 117,

www.lboro.ac.uk/gawc/rb/rb117.html.Smith, R G (2003b) World city topologies. Progress in Human

Geography 27(5).Spellman, K (2004) Religion and Nation: Iranian Local and

Transnational Networks in London. Berhahn Press, Oxford.Stanley, B (2003a) Going global: wannabe world cities in the

Middle East. In Emerging Issues in the 21st Century World-SystemDunaway, W (ed.) (Ed).. Emerging Issues in the 21stCentury World-System, vol. 1. Praeger, Westport, CT.

Stanley, B (2003b) City wars or cities of peace: (Re)Integrating theUrban into conflict resolution. GaWC Research Bulletin 122,Available from <www.lboro.edu/gawc/rb/rb122.html>.

Stanley, B (2005) Back to the developmental future: (re)empow-ering the Gulf�s city-system. Gulf Papers (March 2005), GulfResearch Center, Dubai. www.grc.ae.

Stanley, B (2004) C2C Cooperation and Middle East Cities.Presented at the BRISMES Annual Conference.

Taylor, P J (2000) World cities and territorial states underconditions of contemporary globalization. Political Geography19(1).

Taylor, P J (2001) West Asian/North African cities in the worldcity network: a global analysis of dependence, integration andautonomy. Paper prepared for the Arab World GeographerConference, Malta, October.

Taylor, P J (2004a) Homo Geographicus: a geohistorical manifestofor cities. Review (Fernand Braudel Center) 27(1).

Taylor, P J (2004b) World City Network. Routledge, London.Tilly, C (2003) Stories, Identities and Political Change. Rowman

and Littlefield, London.Townsend, A (2001) Network cities and the global structure

of the internet. American Behavioural Scientist 44(10),1697–1716.

Trawi, A (2003) Beirut�s Memory. Anis Commercial PrintingPress, Beirut.

Wallerstein, I (1974) The Modern World System. Academic Press,New York.

Wasserman, S and Faust, K (1994) Social Network Analysis:Methods and Applications. Cambridge University Press,Cambridge.

Watts, D (1999) Small Worlds: The Dynamics of Networks betweenOrder and Randomness. Princeton University Press, Princeton.

Wellman, B (1999) Networks in the Global Village. Westview,Boulder, CO.

Wellman, B , Berkowitz, S D (eds.) (1988) Social Structures.Cambridge University Press, Cambridge.

Wilkinson, D (1992) Cities, Civilizations and Oikumenes: 1.Comparative Civilizations Review, 27.

Wilkinson, D (2001) The status of the far eastern civilization/worldsystem: evidence from city data. Journal of World-SystemsResearch 8(3), Available from <http://jwsr.ucr.edu>.

Wilkinson, D (2002) The polarity structure of the central worldsystem/civilization, 1500–700 BC. Paper presented at the ISAAnnual Meeting, www.longtermchange.net/article.php?sid+40(accessed 11/03).

Yacobi, H and Fenster, T (2004) Whose right to the city? Urbanplanning and local knowledge in globalizing Tel Aviv-Jaffa.Paper presented at Cities in the Middle East Conference, Ben-Gurion University.

199