manuel castells - local and global [cities in the network society]

LOCAL AND GLOBAL: CITIES IN THENETWORK SOCIETY 1

MANUEL CASTELLS

Department of City and Regional Planning, University of California, Berkeley, 228 Wurster Hall, Berkeley,California 94720. E-mail: [email protected]

Received: October 2001

Over the past year or so, I have expanded mywork in both areas of my current research.I have explored deeper the interaction be-tween information technology and society,some outcomes of which have recently beenpublished on the internet (Castells 2001).I have also continued my analysis of spatialtransformation through an interdisciplinaryapproach to the problems of cities andregions. This paper summarises my currentideas for research on urban transformation.

INTRODUCTION: THE NETWORKSOCIETY

To begin I explore two sets of relationships –that between the local and the global, and thatbetween certain dimensions of identity and func-tionality as they impinge on spatial forms –and will try to show how they interact in thespatial transformation of the information society.Some people call this the network society –and I do the same myself, for conceptual andanalytical reasons that are explained below.

We are indeed living in a period of historicaltransformation. In my analysis, this processinvolves the interaction of three features that,though distinct, are related to each other. Thefirst is the revolution in information technologythat started in the 1970s and then expandedall over the world. The second is the process ofglobalisation, which incidentally is not onlyeconomic. There has also been globalisationof the media, as well as cultural and politicalglobalisation, etc. The third feature is theemergence of a new form of organisation thatI call networking. This is not just any kind of

networking, but the specific kind of powernetworking that works through informationtechnology. This power networking is chang-ing the way we perceive, organise, manage,produce, consume, fight and counter-fight –embracing practically all dimensions of sociallife. The interaction between the revolution ininformation technology, the process of global-isation, and the emergence of networking asthe predominant social form of organisationconstitutes a new social structure: the networksociety. As we know, industrial society has hadmany different social, cultural and institu-tional manifestations. Likewise, the networksociety has many different manifestations,depending on country, culture, history andinstitutions.

However, some basic commonalities emergewhen we consider specific features of thisnetwork society. This paper focuses on oneof the dimensions of this transformation. Tosome extent, it may be conceptualised underthe notion of the network society. The dimen-sion is the spatial transformation. It is a funda-mental dimension – and always has been, allthrough the world – of the growth process thatwe know as structural change. In that regard,I believe we need a theory of spatial forms andprocesses, which can be adapted to the newsocial, technological and spatial context inwhich we live.

Here, I formulate some elements of thistheory. In the main, I build my theories fromthe bottom up. I try to practice groundedtheory, thus, I try to build analytical frame-works that could be used as tools for empiricalresearch. First, I give a brief overview of what

Tijdschrift voor Economische en Sociale Geografie – 2002, Vol. 93, No. 5, pp. 548–558.# 2002 by the Royal Dutch Geographical Society KNAGPublished by Blackwell Publishers, 108 Cowley Road, Oxford OX4 1JF, UK and 350 Main Street, Malden MA 02148, USA

I see as the emerging spatial trends at thishesitant beginning of the twenty-first century.Then I offer a tentative theoretical interpreta-tion of these spatial trends. Subsequently,I will highlight the main issues arising in citiesand in the theory of cities in the informationage. Particular emphasis is placed on the keytheme of this paper, namely the crisis of citiesand of the city as a socio-spatial system ofcultural communication. I conclude by point-ing out some implications of my analysis forplanning and urban policy.

As usual in my work, this will be verysketchy. I hope that further discussion willprovide further enlightenment both for youand me.

KEY SPATIAL PROCESSES

Let me first identify the key spatial processesof the early twenty-first century. First, I thinkwe are rapidly moving in the direction ofan urbanised world. We are about to cross acritical threshold: half of the population ofthis planet will soon be living in cities. Reliableprojections state that by mid century, betweentwo-thirds and three-quarters of the totalpopulation will be living in ‘some kind’ ofurban agglomeration. And the critical ques-tion is, which kind? Certainly we will not beliving in the countryside as we now know it.We may be living in urbanised villages,though. This is one of the most importantforms of rapid urbanisation, particularly indeveloping countries. This process of urban-isation is concentrated disproportionately –and increasingly so – in metropolitan areasof a new kind. These urban constellationsare scattered throughout huge territorial ex-panses. Gottmann’s megalopolis was some-what reconstructed (Gottman 1961). Today,we have not only metropolitan areas but alsobig ‘metropolitan regions’, and these are veryspecial indeed. They are a mix of cities,countryside, centre, and periphery – they arenot necessarily part of one urban continuity.Some people call them edge cities, others callthem conurbations. I think all these termsbelie the novelty of the process. And thatnovelty lies in the ability to connect function-ally a huge number of people and activitiesthroughout a large expanse of space. That

space is constantly being remodelled andreconstructed by the transformation of thecommunication, transportation and telecom-munications systems.

Many years ago, the Dutch invented theRandstad. By now, almost everybody believesin the Randstad, everybody except people inthe Netherlands. Maybe they are starting toexperience a real Randstad now. It may notnecessarily be what it was defined as by spatialplanners. Instead, the Randstad is the articu-lation of this country in a vast agglomerationthat is linked to Germany and all the neigh-bouring countries. It is linked to an entireEuropean network of fast transportation. Thearea contained within two hours transporta-tion time from Amsterdam, Rotterdam, orUtrecht accommodates a huge population.I would not say that everybody in that area isfunctionally linked. But large segments of thepopulation are. Many others are linked partlyto this region and partly to an adjacent one.Interestingly, what the automobile could notentirely accomplish in Europe, the fast train isdoing. Europe is emerging as a set of majormetropolitan regions, which are at the sametime strongly interlinked. The same phenom-enon may be observed all over the world.

The work by Scott and other geographersat the University of California, Los Angeles,shows that there is a new Southern Californian‘metropolis’, as it is called (Scott 1996). Itextends at least 150 miles from north to south,and goes into Mexico. Tijuana is part of it,and so it is a transnational city, although thelargest Mexican city in this conurbation is LosAngeles – LA has four million people who areoriginally from Mexico. The San Francisco BayArea is a different kind of animal with othercharacteristics. In terms of the actual labourmarket, there are 7.5 million people living andworking in that area. San Francisco is nolonger the largest city in the Bay Area. This isSan Jose, with a population of one million,versus 750,000 in San Francisco.

An empirical definition of what a realconurbation is has changed for the USA, atleast. It used to be the telephone network, butnow, with the internet, it has become global soyou do not have any specificity. Now, it is thetelevision market. What the television stationconsiders as their market, that is the city. But

LOCAL AND GLOBAL: CITIES IN THE NETWORK SOCIETY 549

# 2002 by the Royal Dutch Geographical Society KNAG

it is not a city, of course; it is a market link toa connection between residential and workingplaces. In really big cities – not little Europeanconurbations – in Hong Kong, Shenzhen,Macau, Zhuhai, and the other cities of thePearl River delta, all the way to Canton, about65 million people work and live in a highlyinterrelated functional area. Certainly noteverybody does everything in that area, ofcourse. Take Japan – the conurbation ofTokyo, Yokohama and Nagoya (now function-ally extended to include Kobe and Osaka, andKyoto to a large extent) is another huge urbanconstellation.

This is the kind of phenomenon I do notcall a city; my current term is metropolitanregion (which cannot be more than a pro-visional one until we find a serious empiricalinterpretation of what is going on). It is anew kind of urbanised agglomeration that weare generating. These magnets of economic,cultural, political and urban growth are ab-sorbing more and more of their populationand activities in their hinterland. In fact, theybecome nodes in global networks of cities.Indeed, advanced telecommunications, theinternet, and fast computerised transportationsystems (I remind you that planes, trains, andships are all computerised transportationsystems) allow for a simultaneous spatial con-centration in huge areas and thus for decentral-isation. Therefore these systems are introducinga new geography of networks and urban nodesthroughout the world, throughout countries,between metropolitan areas and within metro-politan areas. This is the new urban geography.

CURRENT SOCIAL PROCESSES

On the social side, there is a trend for socialrelationships to be characterised simultaneouslyby two processes: individuation (not individual-isation but the building of meaning vis-a-visthe individual project) and communalism. Bothprocesses use spatial patterning and onlinecommunication. Individuation is both spatialand virtual: physical proximity and onlineconnectivity. The same applies to communal-ism: virtual communities and physical com-munities develop in close interaction. We nowhave enough empirical research to go beyondthese fantasies about virtual communities being

different from physical communities in a worldin which the internet has become a key com-munication mode. We have both online andoff-line social interaction, creating a hybridpattern of sociability. Something else thatshould be emphasised – though not a spatialphenomenon, it does have extraordinary con-sequences for spatial structure and dynamics –is the crisis in the patriarchal family. This hasdifferent manifestations depending on theculture and the level of economic develop-ment. This crisis gradually shifts sociabilityfrom family units, in the traditional sense, tonetworks of individualised units. Most often,these are made up of women and theirchildren in relationship to other women withtheir children, but these units may also consistof all kinds of individualised cohabitation part-nerships. This has extraordinary consequencesfor the uses and forms of housing, neighbour-hoods, public space and transportation systems.

This crisis coincides with changes in thebusiness world. Here, we see the emergenceof the network enterprise as a new form ofeconomic activity, which is a highly decentral-ised yet co-ordinated form of network. At thesame time, we see the emergence of decentral-ised and co-ordinated management patterns.This network enterprise is not a network ofenterprises. Rather, it consists of enterprisesthat are internally organised as networks andthen connected with other networks of otherenterprises. The network enterprise has verysubstantial spatial consequences. The mostimportant is a return to the work-living ar-rangements of the pre-industrial age or of theperiod of industrial craft work. Interestingly,these arrangements for working and living inthe same place often take over the old indu-strial spaces, transforming them into infor-mational production sites. For instance, in SanFrancisco’s multi-media gulch, the city’s lastremaining industrial buildings were trans-formed into spaces for multi-media produc-tion sites. What is multi-media? Manufacturingor services? It is both! It is the productionof dreams, which is the most powerful formof manufacturing in our world. It is a verymaterial production in many ways, but it issoftware, so it is informational production. Itis a different kind of manufacturing. It is aproduction organised in terms of the people

MANUEL CASTELLS550


living there, working there and socialisingthere. And whether we refer to experiencesin London, Tokyo, Beijing, Taipei, Barcelonaor Helsinki, we find exactly the same kind ofwork-living arrangements in the advancedsoftware-based industries.

Urban areas around the world, another keytrend, are becoming increasingly multi-ethnicand multicultural. This is an old theme of theChicago School, but I would say it is nowamplified in terms of its extremely diverseracial composition. (I will come back to theanalytical implications of this matter later.)Another trend is that the global criminaleconomy is local at the same time. It is solidlyrooted in the local urban fabric. The cities arebeing taken over in many ways by this globalcriminal economy. In other words, the globalcriminal economy does not start from local-ities and depressed areas. It is a globalbusiness that penetrates the urban areas indifferent ways. It reaches into the poor ghettosbut at the same time also links up to money-laundering and other activities. The break-down of communication patterns betweenindividuals and between cultures is anothermajor trend. This leads to the emergence ofdefensive spaces, which are in fact at the rootof the formation of sharply segregated areas:gated communities for the rich, territorialturfs for the poor.

THE CHANGING NATURE OF PUBLICSPACE

At the same time, in reaction to the trends ofsuburban sprawl, major metropolisation andindividualisation of residential patterns, urbancentres and public space become critical ex-pressions of local life. As I will show later, thisis in fact a reaction and an interaction. But forthe moment, I am still going down the list ofwhat we can observe on the surface in termsof spatial transformation. In other words,public space is really critical. In most planningprojects everywhere in the world, the revital-isation of urban life and of the city as a com-municative space has become paramount. Infact, it is becoming the most salient sellingdevice for private residential development. Inprinciple, support for the vitality of publicspace is still a major trend. I say only in

principle, because the commercial pressuresand the globalisation of tourism and businesstravel are mimicking urban life in many citiesrather than actually rebuilding urban space.Many public spaces around the world – andthus in your cities too – are also being trans-formed into theme parks, where symbolsrather than experience create a life-size urbanvirtual reality. Ultimately it is the next bestthing to being projected in the media andthen selling the city. In that sense, the LasVegas phenomenon – building all the greatestcities in the world in Las Vegas – can alsobe reproduced, whereby the greatest cities ofthe world become Las Vegas themselves. It isa consequence of the commercialisation ofpublic space, of the massive diffusion, and ofthe suburban and exurban sprawl.

On the other hand, it is a consequence ofthe increasing individualisation, whereby, asGalliano has proposed, consumption itemsbecome individually appropriated. Thus, youhave individualisation of the residential andwork experience, on the one hand, and indi-vidualisation of the consumption of the city,on the other. All in all, the new urban worldseems to be dominated by a double move-ment: inclusion in trans-territorial networksand exclusion by the spatial separation ofplaces. The higher the value of people andplaces, the more they are connected in inter-active networks; the lower their value, thelower their connectivity. In extreme cases,some of the places are by-passed by the newgeography of networks. This is indeed whathappens in depressed rural areas aroundthe world, in declining regions, or in urbanshantytowns. Then the infrastructure of thesenetworks – not only of communication net-works, but also of water, electricity, roads, oradvanced communication systems – reinforcesthis segregation. The work recently publishedby Graham and McMahon on splinteringurbanism clearly shows how these spatial andsocial trends towards splintering spaces are infact materially articulated and reproduced inthe design of telecommunication infrastruc-tures. In this way, the world is not sociallysegregated simply by the market or by peoplemoving or not moving. It is also segregated bythe spatial layering of major communicationinfrastructures – for example, where you have



fast internet access or not, where you have fibre-optic cable or not, where you have advancedtransportation systems or not. In Europe, thelocalities by-passed by the high-speed trainsare being segregated.

THE EMERGENCE OF THE NETWORKSTATE

The constitution of these mega-metropolitanregions without a name, without a culture, andwithout institutions weakens the mechanismof political accountability, of citizen participa-tion, and of effective management. In otherwords, there is increasing contradiction be-tween the actual spatial unit and the institutionsof political representation and metropolitanmanagement. On the other hand, however,local governments in the age of globalisationemerge as flexible institutional factors that areable to react, to adapt more quickly to globaltrends. In fact, the dynamics of globalisationdo not eliminate local governments. Rather,globalisation enhances their role and the abilityof local authorities to get closer to the needs oftheir community. In other words, if you cannotcontrol the world, you shrink it to the size ofyour community so you can manage it a littlebit better. Actually, you cannot control it atthe national level either. The rebuilding ofnetworks of co-operation between institutionscan proceed faster and go deeper on the basisof legitimacy. As all surveys show around theworld, whatever is left of political legitimacy,which is not much, is left mainly at the locallevel. So, a new form of state emerges.

More and more, I see this to some extentas a confirmation of the very tentative hypoth-esis posed in my trilogy. That is what I callthe network-state. It does not make thenation-state disappear. Rather, it integratesthe supranational institutions that are madeup of national governments, nation-states, inter-national institutions, regional governments,local governments, and NGOs (which arecitizen representative organisations). In thisparticular network configuration, the network-state becomes the actual institution that ismanaging cities and regions in our context. Inthat sense, local governments become a nodein the chain of institutional representationand management. The local authorities are

able to input the overall process, but withadded value because of their capacity torepresent citizens at a closer range.

However, in any case, this is work inprogress. For the moment, what we observeis an increasing gap between the actual unit ofworking and living in the metropolitan regionon the one hand, and the mechanism ofpolitical representation and public adminis-tration on the other. In this context, urbansocial movements have not disappeared by anymeans; they have merely mutated, essentiallyaround two main lines. The first is the defenceof the local community affirming the right tolive in a particular place and to benefit fromadequate housing and urban services in thatplace. The second, and I would say probablythe most proactive, is the environmental move-ment. It acts on the quality of cities within thebroader goal of achieving a better quality oflife. The environmental movement is not simplya movement for a better life but for a differentlife. In that sense, it is as much a culturalmovement as it is a traditional urban economy-oriented movement.

In my view, these are the main spatialtrends. They are based on pure observationand certainly can be challenged by differentobservations. But this is what I would distilfrom my observation of current changes world-wide in terms of the spatial transformation.Let me try to make sense of what is going onwith the help of some concepts that bring thediscussion to a more analytical level.

AN ANALYTICAL FRAMEWORK

I think that the transformation of cities in thenetwork society can be organised – in terms ofthe building blocks of a new theory – aroundthree bipolar axes. The first relates to func-tion, the second to meaning, and the third toform. Functionally speaking, the networksociety is organised around the oppositionbetween the global and the local. Dominantprocesses in the economy, in technology, themedia and authority are organised largely inglobal networks. But day-to-day work, privatelife, cultural identity, and political participa-tion are essentially local and territorial. Now,cities as communication systems that workthroughout history are supposed to link up

MANUEL CASTELLS552


the global and the local. But that is exactlywhere the problem starts. Cities are in factbeing torn by these two conflicting logics thatdestroy the city as a sociospatial communica-tion system when they try to simultaneouslyrespond to logic. What does this mean? Well,it means that if you organise your spatialplanning to create a competitive city in globalnetworks, maybe you are going to put yourresources – economic, technical, institutional– where they will trickle down to the popula-tion, the neighbourhood. They may sponsorthe expression of cultural identity later on; forthe moment, they can organise a street partyonce a year. Now, if you simply cater to thelocal identity, to the needs of the inhabitants,you will have to ask where the money comesfrom. You have to be competitive and you haveto be productive. So, while we have to be inthe networks, we have to be at the same timerooted in locality and identity. But in ourobservations around the world, we see thatfirst things come first. And the first thing ishow to exist in the global networks. There is atremendous and increasing distance betweenthe locality as an expression of society, on theone hand, and the functionality and theglobality as expressions of competition andproductivity, on the other, whereby they func-tion in the creation and appropriation ofwealth. The second thing is how to exist interms of meaning. As mentioned earlier, oursociety is characterised by the opposing devel-opment of individuation and communalism.Now that we have come to the theoretical part,we can define some things. By individuationI mean the enclosure of meaning in theprojects, interests and representations of theindividual. That is why the concept of indi-vidualisation is not the same as individuation;in the latter, all that matters is enclosed in theindividual. By individual I mean a biologicallyembodied personality system, or – if translatedfrom the French – an individual is a person.Communalism refers to the enclosure of mean-ing in a sheer identity. That enclosure is basedon a system of values and beliefs to which allother sources of identity are subordinated.Society, of course, only exists at the interfaceof individuals and identities mediated byinstitutions. This interface and this mediationare at the source of the network-state.

The network-state relates to the grassroots,to the people themselves. Remember, sinceGramsci onwards, civil society was always seenin relationship to the state. That’s what makescivil society interesting. That’s what organisesan autonomous transition from the organis-ation of people in society to the institutionsof the state. So, civil society is not the contraryof a state. It is the complement of a state. Itis the bridge towards the state, in the originalGramscian theory and as it has been devel-oped by others from there. But again, ingeneral, people tend to think the contrary.So, in a more modern formulation: the civilsociety in fact is developed in a sheer publicsphere a la Habermas. Now, what we observein the formative stage of the network societyindicates the increasing tension and distancebetween personality and culture, betweenindividuals and communes. In other words,there are two logics. And in between theinstitutions of civil society, the political institu-tions and the public sphere – as legitimateinstitutions of communication and represen-tation – seem to fade away. And then we areconfronted with the logic of individuals, onthe one hand, and of communes, on the other.Therefore, this split between personality andcommonality puts extraordinary stress uponthe social system of cities as communicativeinstitutionalising devices. In this sense, theproblematic of social integration again be-comes paramount, as it was in the origins ofurban sociology and urban studies, but now incompletely new circumstances and in termsradically different from those of early indus-trial cities. Why? Well, social integration is nowproblematic because of the urban transfor-mation represented by a third and major axisof opposing trends, this one concerning spatialforms.

FLOWS AND PLACES

We have dealt with function, and we havedealt with meaning. Let us now look at theissue as a question of form. In terms of form,the major bipolar opposition is between what Icall the space of flows and the space of places.In the space of flows, separate locations arelinked up electronically in an interactivenetwork that connects people and activities



in different geographical contexts. Now, thespatial flows – let us say, the financial net-works, the international production networks,and the media networks – are not a-territorial.They consist of territories which are distant,which are linked to different geographichinterlands. But they are electronically con-nected; their function and their meaning comefrom their connections. Thus, they do not existseparately. In that sense, they are not purelyelectronic networks. The electronic networkslink up the specific places, and it is this hybridspace that is the space of flows. The spaceof places organises experience and activityaround the confines of locality. What is criticalin our society is that cities are structured andrestructured simultaneously by the competinglogics of the space of flows and the space ofplaces. Cities do not disappear into the virtualnetworks. Rather, they are transformed in theinterface between electronic communicationand physical interaction. They are trans-formed by the combination in practice ofcities, networks, and places but without fullyintegrating them.

Let me give two examples of this ratherabstract problematic. One concerns urbanstructure, the other urban experience. Inurban structure, the example is the notion ofglobal cities. The global city is not a medalof honour given to certain cities that havebecome important, which would imply thatthere are also semi-global cities, upcomingglobal cities, etc. We already had a concept forthat – an old one that Friedmann reinvented,the ‘world city’, which is a hierarchical con-cept (Friedmann 1986). If the global city hasany meaning other than that it is possible tosubstitute one by another, it is a differentmatter. It is the notion that there is a global cityin certain dimensions. For instance, financialnetworks – which is an easy example to under-stand – are made up of bits and pieces ofdifferent cities across the globe. The financialdistricts of New York, London and Tokyo areall part of the same city. They work symbiotic-ally. They connect with each other but alsowith Frankfurt and Amsterdam and so on. Andto a large extent even La Paz, Bolivia, is part ofit. A little bit of La Paz is in that global citybecause that is how lots of money (they dosome good trading) circulates in these global

networks. So London is not a global city, if youunderstand global city to mean that the wholeor the majority of London is integrated in aglobal network. No, London is very local andvery parochial. If you go around Hampstead,you seem to be in an upper-class village of thegentry. And the same thing applies to everycity. Take New York – New York is very local.Queens is a very local area, these days usuallylocal from the point of view of all kinds of im-migrants from around the world. The globalcity, therefore, is a network of financial spaceswhen the global city is defined in terms offinancial networks. It is a network of the ad-vertising or media industry when it is definedin those terms. It is a network of high-techspaces – along with Silicon Valley, Helsinkiand Munich – when defined in those terms.So, there are many global cities. But the manyglobal cities are not London, Zurich, NewYork and Frankfurt etc. There are many dif-ferent dimensions of globalisation, of urbanactivities, which are connected functionally. Allcities, to very different degrees, are to someextent under the stress of the connection ofeach key centre, of each key activity in thisglobal network, while at the same time most ofthe city is engaged in a very local life. It is thetension between the two activities that iscritical. From the point of view of the urbanexperience, we are entering a built environ-ment that is increasingly incorporating elec-tronic communication devices everywhere. Infact, our urban life, as Bill Mitchell from MIThas pointed out, is becoming what he calls an‘e-topia’ (Mitchell 1995). That is, a new urbanform in which we constantly interact, eitherdeliberately or automatically, with online in-formation systems, which increasingly will bein the wireless mode. So, materially speaking,the space of flows in terms of the experienceis folded into the space of places. But at thesame time, there is tension between what youdo and moving, with your head-set, on theinternet all the time. This tension leads togreat difficulty in terms of the social inte-gration of cities as communication devices.

Let me now turn to the issue of social inte-gration, which is really at the forefront of thetheory of cities in this network society. Thenotion here is that we have a fragmentation ofmeaning in areas that are functionally inte-

MANUEL CASTELLS554


grated. While an urban agglomeration is afunctional unit, at the same time it creates adisparity of cultures, systems of representation,and systems of meaning. Thus, an agglom-eration is multicultural but at the same timemulti-meaning. Concretely, if we live in a worldof individuals and communes, each communehas its own set of values and each individualhas his/her project. Therefore, it is extremelyimportant to see how this multiplicity ofmeanings and of cultural sets can interactand communicate. The traditional problem-atic of social integration – of the origins ofcities in the industrial area – was a differentone. There was a dominant urban culture intowhich rural migrants or migrants from otherparts of the world had to be integrated, assim-ilated. There is no way to assimilate anythingnow because there is no dominant culture.The dominant culture, if there is one through-out the world, is represented by the massmedia and by the hypertexts that this mediacontains. But this dominant culture is, in fact,very malleable, because it is a market-orientedculture. It follows. It is not a culture with valuesthat everybody has to believe – that was the oldindustrial culture. It is a culture that followswhatever happens in the market; it identifiesniches. Well, so rap is the thing. I create MTVand we will rap. This is not the dominantculture; it is a market-oriented culture. Ratherthan unifying the diversity of cultural messages,it amplifies this diversity by transforming dif-ferent cultures and sets of mind into marketniches, thereby enhancing and deepening thefragmentation. So, we have a fragmentation ofthe spatial configuration of the metropolis, wehave an individualisation of communication,and we have a constellation of cultural subsets.Under such conditions, the notion of publicsphere disappears. The traditional notion couldonly be reconstructed by institutions – throughfundamentally political institutions and in ageneral crisis of legitimacy. Take, for instance,California. These days 39% of people in Cali-fornia think (at least a year ago they did) thatthe governor as well as the political represen-tatives in their state and in the USA are crooks– yes, the word used in the survey is crooks.And 70% believe they are not being repre-sented. According to Kofi Annan’s survey lastyear for the United Nations, two-thirds of the

people on the planet think they are notrepresented by their governments. Withoutpinpointing specific countries, Kofi Annanadded that this is also the case for the mostestablished and advanced democracies in NorthAmerica and Europe. Two-thirds of the peopleon the planet!

Under such conditions, we have an individual-isation of work, an individualisation of themetropolis in terms of spaces, and an individual-isation and communalism in terms of thecultural sets. Communalism is collective indi-vidualisation vis-a-vis the rest of the society. Youcan individualise as an individual; that is to say,‘Me and my group, and my culture, and I donot know anything about the rest.’ Multiplefragmentation creates a crisis of the city as acommunicative device, which is in fact theoriginal and historical function of the city. Thisis not the same thing as the traditional crises –those of integration of migrants and the urbananomie of the early industrial age. It is a frag-mentation that reproduces itself at the spatiallevel, the work level, the cultural level and thepolitical level. In that sense, we could be livingin the paradox of an urbanised world withoutcities. The key challenge is how to live together.It is as simple as that: how to live together if wedo not share communication codes, not only ifwe do not agree – it is a matter of being able tospeak some kind of language to each other. Ifthe working class opposes capital and fights,that is class struggle, and that can be bloody.That is a language; they know what they aretalking about. Communication can be conflict-ive communication, as it has been throughouthistory. This is different. This is fragmentationand alienation. You are an alien. I cannot talkto you, I do not understand what you are saying,and it is not a matter of language. It is a matterof the values, of the value set. Therefore, I getclose to myself, my family, my group, my project,and we split. That is the notion of com-munication in cities in which the dominantculture has been irreversibly suppressed. In-stead, there is coexistence and a multiplicity ofsources of meaning and expression. The keychallenge for the new urban civilisation is torestore communication. To restore communi-cation means the building and development ofcommunication protocols. This is not a meta-phor; it is a concept from information tech-



nology theory. Which kind of communicationprotocols? Let me present three of them.

TYPES OF URBAN INTERACTION

The first kind is the physical protocol of com-munication. How do you restore communica-tion in a fragmented sprawl? Well, you have tointroduce new forms of symbolic nodality thatwill identify places in this endless sprawl. Whatkind of symbolic nodality? We do not have tothink of fancy solutions. I am not postmodern,so I always want to give an example. I do notconstruct and deconstruct; I am trying toanalyse. Let us consider the Barcelona modelof planning and design, which has been highlycommended around the world. How was theperiphery of the working-class district (whichwas much worse than the Parisian peripheryof the grandes ensembles) marked and redefinedby the Barcelona planners? Well, they startedto construct horrible monuments, statues andsquares of very doubtful taste. But that doesnot matter, really. Some people like it – notmany – but it does not matter. When you getcompletely lost, and you ask anyone livingthere: ‘How can I get to that HLM number134 in the second town?’, the person will tellyou, ‘Well, continue here, then you will findan absolutely horrible statue in a shabbysquare, then you turn right and you are there.’In other words, symbolic nodality reconstructsspatial meaning in the city. That is why archi-tecture again becomes very important. Archi-tecture always had been about the marking ofplaces. Urban design has always been aboutthe marking of urban forms in relation toculture and meaning.

The second level of urban interaction refersto social communication patterns. That is, itconcerns how people can start being together,sharing cities without being able to speak toeach other and without going through thepublic institutions. How can people be publicin the post-Habermas society, in which onlyHabermas thinks that there is public legiti-macy in the institutions? How can people dothat? Well, remember the old child psychol-ogy, explaining how children learn communi-cation. They learn it by doing things – not bytalking but by doing things and sharing things.If you share something with others in your

spatial practice, if you say: ‘That is my spaceand also the space of this other guy’, and thenif we share the space and we do not kill eachother the first week, then maybe we can startpopulating this particular space together. So,rather than recreating the public sphere, weare moving toward public places as an alter-native model, so away from Habermas andcloser to Kevin Lynch. And in that sense, thespontaneous social interactions in public placesare the communicative devices of our society.Meanwhile, formal political institutions havebecome a specialised domain that hardly affectsthe private lives of people who do not want tobe professional politicians. So, in the practiceof the city, the answer lies in public spaces,including what I call the social exchangers orcommunication nodes. These are the stations,airports, all those places where people haveto bump into each other because they have tochange trains or planes or buses. And theseare the squares, which have some kind ofsocial activity. These spaces are in fact thedevices with which to reconstruct sharingcommunication and therefore city life. I callthis level of urban interaction the sociability ofpublic spaces in the individualised metropolis.

The third level of urban interaction refersto the new combination of electronic communi-cation and physical face-to-face communicationas new forms of sociability. These days, we knowthrough rigorous empirical research that it iscompletely false that the internet alienates andisolates, etc. That proposition is based on justtwo studies, and they can be easily criticised.All the other studies show the contrary. Peoplewho engage in sociability on the internet havemore sociability, more friends, more activities,more everything, controlling for level of edu-cation. What we also know is that this sociabilityis not separate – not a virtual reality versusthe real reality. No, it is a different domainof reality. The communities that exist on theinternet are different kinds of communitiesthan the ones that exist in different forms.In fact, they are not communities. They areindividual networks of socialisation, but theywork together to induce forms of face-to-facesociability. So, virtual communities as net-works of individuals connect to face-to-facesociability, thereby recreating some form ofsociability. The analysis of code-sharing in the

MANUEL CASTELLS556


new urban world requires the study of theinterface between physical layout, social organ-isation, and electronic networks of communica-tion. In this sense, the analysis of the newnetwork of spatial mobility in the mobilephone era is a critical frontier for the newtheory of urbanism. The young people in oursocieties are building their sociability on themobile phone. But they do not spend theirtime talking on the mobile phone to theperson they will meet when they get to theirappointment. While they are walking towardstheir appointment, they are anticipating theirappointment. Some empirical research hasbeen done on the use of the mobile phone toactually build a multi-spatial system, includinga spatial back-up system at home, perhaps.While providing a connection with theirfriends, the mobile phone also transcendsthe boundaries of space. They use it to reachultimately the closest discotheque or the closestbar or the closest excursion to the mountains.So, the places of the space of flows – that is,the corridors and the halls that connect placesaround the world – will have to be understoodas exchangers and social refuges, as homes onthe run as much as offices on the run.

Under these conditions, a dominant trendemerges towards the disintegration of citiesas communicative devices. The beginning ofgeneral urbanisation could be at the sametime the end of urban civilisation, which isbased on communication and sharing, evensharing in a conflictive manner. This is thecurrent situation, but at the same time there is acounter-offensive. Again, these are not my ideas;I always look at what is happening, and thenI say, ‘Aha, good, counter-offensive; here we go.’

CONCLUSION: THE INTEGRATION OFURBAN LIFE

Throughout the world, a number of people incities – politicians and, particularly at the locallevel, planners and citizen groups – are tryingto reconstruct urban life. Urban life is to beunderstood here not as the traditional historiccentral city, not like a world of Amsterdams.Rather, it should be seen as a world of socialinteraction and meaning operating on thebasis of the appropriation of a space bysociability and by the society that goes beyond

the functionality of integration in the globalnetworks. In that sense, the process of recon-structing urban life is the process of recon-struction of the city as a communicationsystem in its multi-dimensional sense. Restor-ing functional communication through metro-politan planning, providing spatial meaningthrough a new symbolic nodality created byinnovative spatial projects, and reinstating thecity in its urban form through the practice ofurban design, focused on the preservation,restoration and construction of public space –these are the critical issues in the new type ofurbanism. Conversely, there is what I wouldcall the defensive battle of nostalgic recon-struction of the old city in the suburbs throughnew traditionalism and the new urbanism.Well, this is really giving up. It amounts tosaying, I am going to build a suburb that lookslike a city. But the important thing is howpeople interact. And people can interact inhorrible places. If they interact in horribleplaces, then they can reconstruct this spaceand make it meaningful. So, that level of com-munication seems to be the critical one. Thereare a number of cities around the world wherewe see just that. You can observe it on all con-tinents, from Portland to Curitiba to Barcelona.There are a number of very good examplesthat always combine an emphasis on publicspace, competitiveness in the global networks,a strong emphasis on local governments andcitizen participation, and the ability to reinte-grate symbolic nodality, symbolic represen-tation in the reconstruction of space. But inthe end, none of these efforts by people, byplanners, and by urban designers can functionwithout a transformation of the urban policy,and that depends in turn on the transforma-tion of urban polities. Ultimately, the meaningof cities depends on the governance of cities.How can we introduce this notion of gov-ernance of cities in a situation of increasingbureaucratisation and alienation of institutionsvis-a-vis their citizens? That is a fundamentalquestion. In the absence of any empiricalevidence we cannot accept the idea that somegreat urbanists or great architects or greatpolicy-makers would be able to make any signi-ficant advances without transformation of urbanpolicy on the basis of urban polities. So, ulti-mately it is a political problem in the traditional



sense, in the sense of the polis, the challenge isto reconstruct society through the practice ofliving together, in the process of communicat-ing with each other in the urban civilisation.

Note

1. This essay is a revised version of the Alexandervon Humboldt lecture presented by the authoron 8 October 2001 at the University of Nijmegen,The Netherlands, which was sponsored by theRoyal Dutch Geographical Society, the Universityof Nijmegen, The Netherlands Graduate Schoolof Housing and Urban Research (NETHUR),and the Department of Human Geography ofthe University of Nijmegen. The evening wasorganised by Professor H. Ernste and ProfessorJ. Terwindt. The editoral board thanks Ms LiekeStelling for her transcription of the text from arecording of the lecture

REFERENCES

CASTELLS, M. (1996) The Information Age: Economy,

Society and Culture (three Volumes). Oxford:Blackwell

CASTELLS, M. (2001) The Internet Galaxy. Reflections on

the Internet, Business and Society. Oxford: OxfordUniversity Press

FRIEDMANN, J. (1986) The World City Hypothesis.Development and Change 17 (1) pp. 69 – 84.

GOTTMANN, J. (1961) Megalopolis. The Urbanized North-

eastern Seaboard of the United States. New York, NY:Twentieth Century Fund.

MITCHELL, W.J. (1995) City of Bits: Space, Place and the

Infobahn. Cambridge, MA: MIT Press.SCOTT, A.J. (1996) The City: Los Angeles and Urban

Theory at the end of the Twentieth Century.Berkeley CA: University of California Press.

MANUEL CASTELLS558


manuel castells - local and global [cities in the network society]

Documents