- Email: [email protected]
Placing splintering urbanism: Introduction
Geoforum 39 (2008) 1815–1820
Contents lists available at ScienceDirect
Geoforum journal homepage: www.elsevier.com/locate/geoforum
Placing splintering urbanism: Introduction Olivier Coutard Université Paris-Est, LATTS (UMR CNRS 8134), ENPC, 6 Avenue Blaise Pascal, F-77455 Marne-la-Vallée Cedex 2, France
a r t i c l e
i n f o
Article history: Received 24 September 2008 Received in revised form 17 October 2008
Keywords: Infrastructures Neoliberalism Splintering Urbanism Europe United States Lower-income countries
a b s t r a c t This paper introduces a collection of case studies aimed at ‘‘Placing Splintering Urbanism”, in reference to the thesis developed by Graham and Marvin [Graham, S., Marvin, S., 2001. Splintering Urbanism. Networked Infractructures, Technoloical Moblilities and the Urban condition. Routledge, London]. Whilst acknowledging the value of the thesis as an analytical framework in opening the way to innovative understandings of contemporary urban dynamics, the paper argues that, taken together, the articles in this themed issue seriously challenge the ‘‘splintering urbanism” thesis theoretically, empirically and methodologically. They question in particular the postulated universality of the ‘‘modern infrastructural ideal” and of ‘‘unbundling” and ‘‘bypass” processes — all of which are key elements in Graham and Marvin’s argument — as well as the assertion that reforms in infrastructure sectors should generally result in more discriminatory, socially regressive patterns of provision of essential services and more splintered urban spaces. Based on these fundamental critiques, the paper concludes that one cannot speak of ‘‘splintering urbanism in general” — i.e., as a global trend — in any meaningful analytical way. Ó 2008 Elsevier Ltd. All rights reserved.
1. Introduction This themed issue brings together a set of papers which empirically explore the ‘‘splintering urbanism” thesis (Graham and Marvin, 2001). Individually and collectively, the papers aim therefore to reflect on the heuristic value as well as on the limits of Graham and Marvin’s argument, both as an analytical framework for researching specific urban contexts and as a wide-ranging thesis for explaining recent urban change. In both respects, the objective is to ‘place’ splintering urbanism – or, as the case may be, to displace it should it prove inadequate to properly account for the processes at play.
2. The splintering urbanism thesis In their influential book Splintering Urbanism, Stephen Graham and Simon Marvin argue that ‘‘a parallel set of processes are under way within which infrastructure networks are being ‘unbundled’ in
ways that help sustain the fragmentation of the social and material fabric of cities” (Graham and Marvin, 2001, p. 33)1. The splintering urbanism thesis articulates four main elements. (1) During approximately a century, until the 1960s, the development and governance of networked systems was embedded in a modern infrastructural ideal that supported the notion of monopolistic, integrated and standardised provision of network service (chapter 2). In particular, ‘‘a set of practices were developed to ensure the rapid roll-out of standardised infrastructure at equal price across national economic space” (p. 80); and ‘‘there are powerful resonances between the (...) modern infrastructural ideal and the colonialist policies shaping the attempted roll-out of infrastructure networks in developing cities” (p. 81). But from the late 1960s, this ideal was progressively undermined by a combination of powerful factors: the urban infrastructure ‘‘crisis”; changing political economies of urban infrastructure development and governance; neoliberalism and the withdrawal of the state; economic integration, urban competition and the imperatives of global–local connectivity; the development of infrastructural consumerism; the collapse
E-mail address: [email protected] In order to understand the thesis defended by Graham and Martin, and the discussion of that thesis here, it is useful to clarify the meaning of the notions at play: inequalities, segregation, differentiation, discrimination, splintering, secession, polarisation, withdrawal from (or collapse of) solidarity. The concept of urban splintering, in the sense in which we employ it, refers to the disintegration of former socio-economic interdependencies and to tendencies towards the withdrawal from (or collapse of) solidarities. This idea is the opposite of urban integration or cohesion, which emphasises the links of interdependence and solidarity that contribute to the cohesive operation of cities and therefore to the fact that cities are ‘society-making’. The impulses of political and fiscal secession that can be witnessed in Los Angeles, or the increasingly polarised settlement patterns in many urban areas, thus appear as symptoms of this fragmentation or splintering. The notion of splintering differs from that of socio-spatial segregation, which draws more on the idea of a clear spatial expression of economic or social inequalities. Clear segregation is completely compatible with, for example, a high degree of economic integration (see May et al., 1998; Jaglin, 2001). When analyzing the patterns of provision of essential infrastructure services, I shall furthermore distinguish between differentiation – i.e., providing different (groups of) users with different services – and discrimination – by which I shall specifically designate socially regressive forms of differentiation, i.e., forms of differentiation that are potentially or actually detrimental to lower-income or otherwise disadvantaged groups. 1
0016-7185/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.geoforum.2008.10.008
1816
O. Coutard / Geoforum 39 (2008) 1815–1820
of the comprehensive ideal in urban planning; new urban landscapes; and ‘‘new structures of feeling” (chapter 3, p. 92 sq.). (2) Concomitantly, within the context created by the resulting ‘‘collapse of the modern infrastructural ideal”, a profound transformation is occurring: ‘‘the economic liberalisation of infrastructure and the development of new [primarily information] technologies have made possible an entirely new infrastructural landscape that radically challenges established assumptions that have underpinned the relations between integrated networks and cities” (p. 139). Working together, they allow powerful coalitions of actors to promote the unbundling of infrastructures, i.e., the segmentation of integrated infrastructure into different network elements and service packages (p. 141). Unbundling in turn allows for bypass strategies, i.e. strategies that seek the connection of ‘‘valued” or ‘‘powerful” users and places, while at the same time bypassing ‘‘non-valued” or ‘‘less powerful” users and places (chapter 4). (3) These bypass strategies contribute to the emergence of socalled premium networked spaces (Graham and Marvin, 2001, p. 249, sq.)2. In particular, elite or higher-income groups are increasingly living in ‘‘secessionary” places/spaces that are ‘‘withdrawn from the wider urban fabric” (Graham and Marvin, 2001, p. 268) in various ways (chapter 6, esp. box 6.4, p. 268–71), yet intensively connected to other, remote premium spaces, thus forming archipelagos of ‘‘global enclaves” (p. 389). This reinforces the ‘‘vicious cycle” of splintering, ‘‘where attempts at socio-technical secession lead to greater fear of mixing, so increasing pressure for further secession, and so on” (Graham and Marvin, 2001, p. 383). The widening gap between connected and unconnected (or disconnected) places and people is all the more worrisome since the world we live in is, increasingly, a network society (Castells, 1996) in which ‘‘the poverty that matters is not so much material poverty but a poverty of connections,” which ‘‘limits a person or group’s ability to extend their influence in time and space” (Graham and Marvin, 2001, p. 288). Central to the argument, therefore, is the notion that ‘‘the diverse political and regulatory regimes that supported the roll-out of power, transport, communications, street and transport networks towards the rhetorical [sic] goal of standardised ubiquity are, in many cities and states being ‘unbundled’ and ‘splintered’ as a result of a widespread movement towards privatisation and liberalisation.” (p. 382) (4) Communities and public authorities in cities around the globe confronted with splintering processes and tendencies may develop forms of ‘‘resistance” (p. 387 sq.). As a result ‘‘a central theme of urban politics and urban social movements in the first decades of the new millennium will therefore centre on the struggle between the ‘global’ forces of attempted, ‘pure’ boundary control and the customisation of premium, commodified network spaces, vs. the imperative of infrastructural, urban and technological democratisation and the need for more egalitarian and democratised practices and principles of development” (p. 405). In this editorial, I wish to briefly discuss the splintering urbanism thesis in light of the arguments and reflections developed in the case-study based papers forming this themed issue3. I will develop my argument in five points, which respectively address: the limits of the modern infrastructural ideal; the politics of infrastructure reforms; the postulated universality of the notions of unbundling and bypass; the urban ‘‘effects” of patterns of service provision; and methodological issues.
2 Economic spaces (e.g., foreign direct investment enclaves or business improvement districts), residential spaces (e.g., gated communities) and ‘‘social life spaces” (such as commercial malls and theme parks). 3 In this editorial, I will signal the articles forming part of the themed issue by writing the names of their author(s) in italics.
3. The deceits of the ‘‘modern infrastructural ideal’’ The notion of a ‘‘modern infrastructural ideal” developed by Graham and Marvin is convincing insofar as it refers to the enduring preference for ‘‘bundled” infrastructure and the related disregard of alternative (whether ‘‘traditional” or ‘‘decentralized”) systems of service provision by the public authorities responsible for this provision. Yet the epochal narrative associated with the modern infrastructural ideal is less convincing. The story goes that, in general, the bundled, monopoly structure of infrastructures sustains (or has sustained) the universal provision of standardised services, consequently reinforcing urban socio-spatial integration. But in-depth analyses of the long-term interactions between network provision and urban dynamics in many cities tell a different story in at least two important respects. They suggest, first, that although the industrial organisation4 of infrastructure sectors certainly matters in the universalisation of basic infrastructure services, it always does so in combination with other factors (see, e.g., Bocquet, Chatzis and Sander, 2008). Three factors seem to be of particular importance here: the will and capacity of government (State or local public authorities) to facilitate, and render solvent, the process of universalisation; public control over land occupation and urban development, as well as housing policies; a small rather than large proportion of households in extreme poverty (Coutard, 2008). In the context of many cities in lower-income countries, for example, these conditions have often not been met. Hence, rather than the ‘‘collapse” of the modern infrastructural ideal, several articles in this themed issue point to its absence (Kooy and Bakker, 2008) or to its failure (Botton and de Gouvello, 2008; Fernández-Maldonado, 2008; Zérah, 2008; see also Jaglin, 2005). As Kooy and Bakker (2008) suggest for Jakarta, many cities in lower-income countries are not splintering, they are and have long been splintered along ethnic or socio-economic lines (see also McFarlane and Rutherford, 2008b, p. 370)5. Second, historical analysis suggests that the relations between universal, homogenous service and urban integration are often ambivalent. In Santiago, for example, the connection of all residential units to the essential networks has, in particular since the 1980s, facilitated and legitimated a strong policy of spatial segregation and functional, social, fiscal and political fragmentation at the local authority level within the urban region as well as processes of urban sprawl, also segregated (Pflieger and Matthieussent, 2008). In Los Angeles, the ‘‘universalisation” of essential services (water, then water and electricity) by a municipally-run entity in the first decades of the twentieth century was primarily designed to serve the interests of the local oligarchy who controlled the municipality and, at the same time, owned the vast tracts of desert land that were open to urbanisation – and the value of which was massively increased – by the provision of those services (MacKillop, 2005). While the control of the water resource and cheap electricity supplies were decisive factors in the political integration of the city of Los Angeles in its current borders6,7, it also favoured (and heavily subsidised) a process of spatial enlargement which, when
4 Single vs. several suppliers; public vs. private ownership; competition vs. planned development. 5 For lack of space, I will not enter here into the (post)colonial dimension of the debate, in spite of its importance and of the fact that several papers in this themed issue participate in this debate. Two recent themed issues address this dimension from complementary perspectives (McFarlane and Rutherford, 2008a; Legg and McFarlane, 2008). 6 Both historically and in the contemporaneous secession debate. 7 Note that political integration between previously or otherwise independent constituencies/local governments through (or for) infrastructure development is frequently observed. Strangely enough, Graham and Marvin do not discuss this often positive correlation.
O. Coutard / Geoforum 39 (2008) 1815–1820
coupled with the historic propensity of the Angelena population to self-segregate amongst themselves, then created a fertile environment for a social and functional splintering of the urban region (growing rejection of social blending, and a functional poly-centrism) and the recent revival of political secession (Boudreau and MacKillop, 2008). In Buenos Aires (Botton and de Gouvello, 2008), a policy strictly conforming to the modern infrastructural ideal resulted in the creation of a premium networked space (in the central part of the capital federal) delimited by the perimeter of the water mains and excluding half the population of the urban region. This apparent failure of the modern infrastructural ideal within a nationally strategic space is interesting in itself. It also raises an important conceptual issue with regard to the integrating or splintering effects of standardised network supplies. Indeed, irrespective of the specific territorial limits of the authority in charge of providing the service, from what spatial scale of universal distribution of a network can one, and should one, consider that its integration effects (within the zone it serves) dominate its fragmentation effects (at its outer edges), i.e. where should we draw a line/boundary for evaluating the dynamics of network provision? Hence while the notion of a ‘‘modern integrated ideal” elaborated by Graham and Marvin is convincing, its articulation with other social, spatial and economic processes and policies is not sufficiently thought through. This is especially regrettable considering its importance in Graham and Marvin’s general argument as the antithesis of splintering urbanism. In particular, the ‘‘collapse” of the modern ideal should not be primarily interpreted as the result of external or externally-determined factors. It should be analysed first as the result of its historical failure, in many contexts, to fulfil its promise of delivering and sustaining universal, affordable, adapted services. In this regard, the evocation of an ‘‘urban infrastructure crisis” deserves to be unpacked and explored in more detail (Graham and Marvin, 2001, pp. 92–94). Furthermore, where the universalisation of essential infrastructure services has been achieved, more attention should have been devoted to its articulation with other policies in the shaping of long-term urban or territorial dynamics. 4. The politics of infrastructure reforms Networked infrastructures and services have been one of the favourite playgrounds of neoliberal economists and their governmental supports worldwide under the ‘‘Washington consensus”. There is no denying the influence in the regulation of utility sectors of a ‘‘liberal convention” (Batley, 1999) characterised by: the ringfencing of individual utility sectors; the commercialisation of services (involving full cost pricing, the shift to a for-profit activity in which profit maximisation has priority over service provision, and the substitution of the economic notion of willingness to pay for the more social notion of ability to pay); the privatisation of companies (and its implications in terms of equity financing); and a change in redistributive principles from social equity to economic equity (via the elimination of cross-subsidies). But much as the modern infrastructural ideal did not deploy its ‘‘bundling logic” universally, the recent wave of reforms that have affected utility sectors have not resulted in a fully ‘‘neoliberal pattern” of service provision, whatever this might mean. The imprint on utility reforms of the liberal convention has systematically been combined with other sociopolitical processes and policies8. As a result, post-reform patterns of provision are, as a rule, quite remote from pure competition and full commodification. This partly reflects
8
5).
This is in line with Castree’s argument on ‘‘neoliberalism plus” (Castree, 2006, p.
1817
the economic characteristics of networked services, with their specific combination of economies of scale and scope, club effects and transactions costs. More importantly, it reflects the deeply political nature of basic infrastructure services, an intrinsic feature of these sectors (Lorrain, 1997; McFarlane and Rutherford, 2008a)9. Many researchers have stressed how ‘‘textbook reforms” have been radically transformed during the sociopolitical process of their design and implementation. For instance, Botton and de Gouvello (2008) analyse how the ‘‘political economy” of the water concession contract in Buenos Aires was radically transformed during the first ten years of its existence to cope with its potential or observed segregation effects. Rutherford (2008) discusses the policy instruments employed in Stockholm that have prevented – so far – the actualisation of the potential regressive effects of utility reforms. Even when public regulation is shaky, as in the case of water and telecommunications in Lima (Fernández-Maldonado, 2008), the outcome of reforms may not be altogether regressive. To a significant extent, the liberalisation of telecommunications in Peru has improved access to internet services in low-income Lima neighbourhoods (see also Fernández-Maldonado, 2005; and for a similar argument on the political nature of infrastructure reforms and the shaping of their ‘‘effects”, see Coing, 2005 on the case of Bogotá). Importantly, this politics of infrastructure reforms does not boil down to a simple dialectic between regressive neoliberal reform and socially progressive ‘‘resistance”. Several papers in this themed issue point to the analytical and heuristic limitations of ‘‘resistance narratives”10. The splintering urbanism thesis implies that people (governments, communities) are faced with overarching dynamics of liberalisation, and that they are left with the simple alternative of either passively accepting these dynamics and their disruptive socio-spatial implications, or outright rejecting them. In many cases, this is not how things happen. For instance, the extensive diffusion of prepayment meters in Cape Town analysed by Jaglin (2008) is not solely a story of resistance by militant groups to the adoption of the intrinsically discriminatory technologies. Rather it is the story of how a particular technology can be appropriated by various actors in various ways according to a diverse set of considerations. These forms of social negotiation around technological change are largely ignored in the splintering urbanism thesis (see also Coutard and Guy, 2007, esp. p. 724)11. The dual interpretation proposed, expressed in terms of generalised neoliberalisation and sporadic manifestations of resistance to this, does not properly reflect the modalities of network service reforms and the complex composition of social interests which shape them and their ‘‘effects”. 5. Questioning the universality of unbundling and bypass Turning to the social and spatial effects of reforms, I would like to question the descriptive, analytical and normative value of the key notions of unbundling and bypass.
9 I obviously do not imply by this that liberalisation per se is apolitical. I just wish to stress that if one wants to form a proper judgment on the causes, processes and effects of reforms in infrastructure sectors, one should take into account the (multidimensional and multiscalar) policies that, together, shape infrastructure reforms. 10 Narratives of resistance to neoliberalism in various domains of social activity abound. See Barnett (2005) and Castree (2006) for an interesting point on resistance narratives as ‘‘consolation for leftist academics”: The vocabulary of neoliberalism, they argue, ‘‘invites us to align our own professional roles with the activities of various actors ‘out there’ who are always framed as engaging in resistance or contestation” (Barnett, 2005, p. 10). 11 Despite Graham and Marvin’s statement, with which I fully agree, that ‘‘the reality of technological innovation, the restructuring of urban landscapes and social and cultural changes in cities, are a great deal more ‘messy’, difficult, contingent and open to contested interpretations and application than [deterministic splintering] scenarios imply” (p. 388).
1818
O. Coutard / Geoforum 39 (2008) 1815–1820
The notion of unbundling, which lies at the heart of the splintering urbanism thesis, suggests that the provision of basic services was previously ‘‘bundled.” In cities in lower-income countries, at least, this has usually not been the case, and the standard pattern of service provision during the ‘‘monopoly era” was very much already an unbundled pattern, with diverse suppliers and diverse forms of supply12. Hence the notion of unbundling is deceptive when it comes to studying the cities (especially in lower-income countries) where recent evolution does not involve a passage from an integrated system to an unbundled one, but rather a passage from one more or less unbundled system to another. In these urban contexts, the dualistic framework of bundled/unbundling infrastructure does not allow us to analyse if, in what measure and under what conditions the evolutions observed are more conducive or, conversely, harmful to the generalisation of access to the service in question (Jaglin, 2008; Kooy and Bakker, 2008; Botton and de Gouvello, 2008; see also Flux, 2004). Likewise, the nature as well as the socio-spatial and sociopolitical meaning of bypass processes varies too greatly from one context to the other to be usefully and heuristically apprehended as variations of the same overarching phenomenon. In many cities in Western Europe, for example, the combined effect of the reforms and the introduction of new information and communications technologies have effectively led to an increase in the differentiation of service offers. These differentiations have often led to a better value for money for most – and sometimes for all – customers. They have also frequently given rise to discrimination and increased inequalities in terms of access to services. Nevertheless, the scale of these inequalities and discriminations has remained limited (problems of affordability, debts, and temporary disconnection). It does seem, in effect, socially inconceivable and politically unacceptable that a large number of households be durably (much less permanently) denied access to essential services because they are insolvent (see Rutherford, 2008; see also Coutard et al., 2008). On the other hand, in most of the cities in lower-income countries examined in this themed issue, the relationship between solvency and effective access to services (in the zones served) is stronger, largely irrespective of the public or private status of the operator. In Cape Town, insolvent households risk being permanently disconnected from their (municipality-run) electricity supply, or even being evicted from their homes (Jaglin, 2008). Although the proportion of households affected is small, the fragmentation logic is obvious. In Jakarta, the enduring exclusion for the majority of Jakarta’s poor from piped water supply also reflects fragmentation patterns, although not primarily rooted in recent reforms (Kooy and Bakker, 2008). In both cases (and in many others), the social meaning of ‘‘bypass” is very different from that of the forms of ‘‘soft” socio-spatial differentiations observed in other cities in the ‘‘global North”. In this perspective, the case of Santiago, Chile (Pflieger and Matthieussent, 2008), an emerging city marked by very strong socioeconomic disparities, but equipped with basic universal networks, stands as an exception in the sample of cities examined here. In Santiago the modalities of supplying essential services appear to be more integrationist: certainly a small proportion of households is excluded from the welfare system for failing to follow the budgetary discipline imposed by social services; but the analysis suggests that, once a network is universally accessible, the public authorities are obliged to guarantee the effective access of the greatest number of people to the essential service thus provided. This suggests a positive correlation (although probably not a one 12 e.g., the combination of (usually restricted) residential tap water supply, standpipes and water carriers of all sorts typical of water provision in many cities in lower-income countries.
way causality) between the universal coverage of a network infrastructure and the ‘‘right” to access the basic services it provides. 6. The urban effects of infrastructure services: toward integration by differentiation? The postulated socio-spatial effects of infrastructure unbundling and bypass are captured in particular in the key notion of premium networked spaces. Throughout their book, Graham and Marvin indifferently use the notions of ‘‘premium network spaces” (see also Graham, 2000) and ‘‘premium networked spaces”, defined as ‘‘new or retrofitted transport, telecommunications, power or water infrastructures that are customized precisely to the needs of powerful users and spaces, while bypassing less powerful users and spaces” (Graham, 2000, p. 185). This definition is intriguing as it seems to postulate a univocal relation between urban spaces and infrastructures (or infrastructure spaces). More importantly, the notion of premium network spaces suggests that, in general, the quality of a given space is determined by the quality of network services supplied to it. In contrast, I would support the view that, in most instances, the notion of ‘‘premium networked spaces” is more appropriate, because it refers more explicitly to social or economic/fiscal premium spaces supplied with tailored network services (see also Coutard, 2002). But we can question whether it is always the case, as the splintering urbanism thesis implies, that increased service differentiation aggravates socio-spatial discrimination. Several papers in this themed issue offer counter-perspectives. For example, water supply service in Cape Town (Jaglin, 2008) is marked by a strong social regulation producing important solidarities between the city’s rich and poor zones/populations (at the price, financing, technical and management levels). However what appears as a possible factor of splintering is the insufficient differentiation of the service, because it incites both the wealthier and the poorer to leave the network, in the former case because the service does not suit them, in the latter case because they lack the means to pay. In the context of a Cape Town type situation where the networks are incomplete, the most appropriate public intervention may well be, at least during a long intermediate phase, to regulate the alternative supply offers to the network (quality of service, price, business profit margins) (see also Zérah, 2008; and for a similar argument in the context of Delhi, see Llorente and Zérah, 1998). The question whether, in the longer term, this creates more or less irreversible discriminations between spaces within urban regions is a crucial but open one. Symmetrically, the forms and degree of what could turn out to be a ‘‘desirable” differentiation are very much dependent upon the context and deserve careful examination. Both issues ultimately have to do with the regulatory and coordinating capacities of public authorities (see Bénit-Gbaffou, 2008, for a similar point with regard to the ‘‘pluralisation” of policing and security services in Johannesburg). In sum, reforms that involve an increased differentiation in services supplied do not necessarily lead to increased segregation (inequality) or fragmentation (removal of solidarity) in terms of access to essential services. The social and urban impact/signification of these reforms – and of different regimes of service provision more generally – can only be appreciated on the basis of an in-depth analysis of the various interactions, over a period of time, between these reforms and the wider context within which they take place.
7. Placing splintering urbanism: from universal patterns to contextual diversity, and back My final point is primarily methodological: it has to do with the issue of evidence. The splintering urbanism thesis rests on numer-
O. Coutard / Geoforum 39 (2008) 1815–1820
ous anecdotes (see in particular the use made of text boxes) taken from very different ‘‘contexts” that all point to splintering processes and hence support the general thesis. But the accumulation of splintering stories is no demonstration. This approach leaves little room for counterfactual analysis. It leaves the authors open to the suspicion of having silently passed over developments every bit as significant as the ones they cite, but which do not support the arguments they seek to advance. Hence the quite different approach used in all the articles forming this themed issue: a monographic approach which seeks to first account for the coevolution of urban infrastructures and the urban fabric in context (as a result of various historical and contemporary policies and processes) in order to then be able to address the relevance of the splintering urbanism thesis in the case at hand within a more ‘‘symmetric” framework, i.e. an analytical framework allowing the possibility that socio-spatial discrimination processes are actually not at work, or that they are so much intertwined with other processes that one cannot significantly refer to a universal, overarching ‘‘splintering” trend. What, then, do the case studies assembled in this themed issue add to the knowledge and understanding of the interactions between the provision of infrastructure services and urban dynamics? The response to this question closely resonates with Castree’s discussion of the ‘‘epistemology of particulars” (Castree, 2005). As Castree convincingly argues, case study research should not just aim at documenting context-related differences between otherwise supposedly analogous manifestations of an overarching phenomenon (in our case: liberalisation, unbundling, bypass, premium networked spaces, splintering and so on). Rather it should aim at identifying substantive similarities or differences between processes in the various contexts studied, in order to shed light on the important question of whether or not the postulated generic phenomenon exists, in a substantive sense of the word. As he further argues in his commentary on recent geographical research on the ‘‘neoliberalisation of nature” (Castree, 2006), ‘‘If, in terms of causal processes and outcomes, neoliberal policies turn out to have highly specific, even unique, environmental impacts from situation to situation, then this idiographic finding should not be glossed in the desire to identify ‘general’ patterns that can then be used to condemn a nonexistent ‘neoliberalism in general’” (page 6). What the case studies gathered in this themed issue strongly suggest is that one cannot speak of ‘‘splintering urbanism in general” in any meaningful analytical way. Processes of infrastructure reforms and transformations are too diverse from one context to the other and from one sector to the other. More importantly, the social significance and implications of these processes are different not only in degree but also in kind between, for example, contexts in which access to basic services is universal and contexts in which it is not. In this regard, the parallel established or suggested between cities in higher-income (long industrialised) countries and in lower-income ones is misleading. Theoretically, the key point is that such case studies invite us to shift the focus from abstract notions, such as the ‘‘modern integrated ideal” or ‘‘bypass”, to the identification and understanding of the combination of values (ideals?), interests and instruments actually mobilised by the actors involved in the planning, development, management or use of infrastructure networks (for a similar argument, see McFarlane and Rutherford, 2008b, p. 370). Placebased case studies need to articulate many dimensions of the infrastructure policies and of the context in which they are deployed, such as: the legal status of services provided, the subscription options available, the structure of real estate property ownership, etc. As a consequence, ‘‘the process of universalization, far from being the inexorable and linear outcome of the action of a small number of ‘major historical forces’, was the upshot of the actions of a whole range of actors acting in a specific geographical, social, economic
1819
and technical context”. (Bocquet, Chatzis and Sander, 2008). This focus on micro processes or variables does not imply, however, that no general lessons can be drawn from such studies. Indeed, as Bocquet et al. (2008) argue, micro analyses allow to ‘‘point up (micro) ‘processes’ and ‘variables’ that may be identified individually in a whole range of domains” and contexts, although they may well be, and usually are, diversely articulated in different contexts. Last but not least, in addition to its methodological and theoretical value, this case-study based approach also has the advantage of reopening the debate on urban futures beyond the dispiriting vision of more or less helpless local communities seeking to (transitorily) resist the inexorable advance of neoliberalism (see also Coutard and Guy, 2007; and for a similar argument Latham, 2006). This would be in line with the objectives outlined in Graham and Marvin’s ‘‘manifesto for a progressive networked urbanism” (p. 404 sq.) resting upon a ‘‘spatial imagination for the splintered metropolis” (p. 405), a perspective which is strangely and regrettably absent from their book. In sum, interpreted as a thesis, the splintering urbanism narrative invites serious theoretical and methodological critiques. First, the absence of any real historical analysis of the phenomena at play gives rise to a certain doubt as to the strength of the conclusions which contrast the current period with a former period characterised by an ‘integrated ideal’. Given the absence of any convincing interpretation as to the reasons why, historically, certain essential networks (water/sewage, electricity, telephone) were made universally available in certain contexts and not in others, the authors struggle to define the specificity of today’s ‘‘neoliberal” era. The clash between the modern infrastructural ideal and the contemporary era appears to be largely unfounded. To a large extent, it seems rather to rest on an artificial rapprochement between, on the one hand, a historical model largely confined to the higher-income countries of Western Europe and North America and, on the other, contemporary processes generally observed in cities in lower-income countries. This thesis also rests on the idea that changes in industrial organisation and the regulation of networked services conform to some universal neoliberal dynamic and that the urban context within which these changes occur has a limited impact on either their modalities or their effects. This universalist vision is impregnated with a soft economic determinism – hardly relieved by the mention of more or less sporadic, local forms of resistance to this neoliberal logic – which the results of the place-based case studies in this themed issue clearly question. Finally the splintering urbanism thesis suggests a positive connection between an integrated offer of networked services and urban integration (and in a symmetric manner between unbundled networks and urban splintering) without the authors ever offering themselves the means to validate (or invalidate) this hypothesis. Indeed, on the methodological level, the demonstration as a whole suffers from an overly superficial empiric validation, advancing by means of an accumulation of rapidly examined examples rather than through detailed empirical works. In closing, however, I would like to emphasise that, putting aside the thesis, Splintering Urbanism offers a fruitful and valuable analytical framework13 highlighting how an infrastructure perspective is key to the understanding of contemporary urban dynamics (and, for that matter, of territorial dynamics on all geographical scales). In this sense, the narrative developed by Graham and Marvin should be interpreted as prefigurative social theory rather than as 13 For the use of splintering urbanism as an analytical framework, see Zérah, 2008; Rutherford, 2008; and Kooy and Bakker, 2008. I thank S. Jaglin for her help in clarifying the distinction between the two interpretations.
1820
O. Coutard / Geoforum 39 (2008) 1815–1820
proven hypothesis14. It rightly points to the context-shaped, and context-shaping, dynamic interactions between infrastructure provision and urban socio-spatial configurations and functioning – even though it does not thoroughly investigate these complex interactions. And it opens the way15 to innovative understandings of contemporary urban processes and policies, thereby sustaining the exciting research agenda set forth in the postscript of the book (p. 404). Acknowledgments The majority of the papers forming this theme issue (all except Kooy and Bakker, Bénit-Gbaffou, and Fernández-Maldonado) originate in a research programme I coordinated from 2002 to 2005. Preliminary versions of all papers were discussed at the Placing Splintering Urbanism international workshop in Autun (France) in June of 2005. I acknowledge support from several French institutions: the Centre National de la Recherche Scientifique (CNRS); the Agence Nationale de la Recherche (ANR Grant No. ANR-05BLAN-0344); the Ministry of Sustainable Development (MEEDDAT); and the Ecole nationale des ponts et chaussées (ENPC). I wish to thank Sylvy Jaglin and Jonathan Rutherford, both at Latts, as well as Dominique Lorrain (CNRS) for the wealth of stimulating conversations we have had over the past few years on splintering urbanism; and Sylvy, Jon and Karen Bakker for their very helpful comments on a draft version of this paper. Finally I would like to thank Neil O’Brien who translated from the French several articles in this themed issue. References Barnett, A., 2005. The consolation of ‘‘neoliberalism”. Geoforum 36, 7–12. Batley, R., 1999. The Role of Government in Adjusting Economies: An Overview of Findings. The International Development Department, University of Birmingham. Benit-Gbaffou, C., 2008. Unbundled security services and urban fragmentation in post-apartheid Johannesburg. Geoforum 39 (6). Bocquet, D., Chatzis, K., Sander, A., 2008. From free good to commodity: universalizing the provision of water in Paris (1830–1930). Geoforum 39 (6). Botton, S., de Gouvello, B., 2008. Water and sanitation in the buenos aires metropolitan region: fragmented markets, splintering effects. Geoforum 39 (6). Castells, M., 1996. The Rise of the Network Society (The Information Age: Economy, Society and Culture Vol. I). Blackwell, Cambridge, MA, Oxford, UK. Castree, N., 2005. The epistemology of particulars: human geography, case studies and ‘‘context”. Geoforum 36, 541–544. Castree, N., 2006. Commentary. From neoliberalism to neoliberalisation: consolations, confusions, and necessary illusions. Environment and Planning A 38, 1–6. Coing, H., 2005. Public utilities in Bogotá: the impact of reforms on territorial and social cohesion. In: Paper Presented at the International Seminar Placing Splintering Urbanism: Changing Network Service Provision and Urban Dynamics in Cross-National Perspective, June 2005, Autun, France, pp. 22–24. Coutard, O., 2002. Premium network spaces: a comment. International Journal of Urban and Regional Research 26 (1), 166–174.
14
I thank Karen Bakker for this point. It does so dialectically, as this line of research is supported by the need/desire to move beyond the splintering urbanism narrative. 15
Coutard, O., Guy, S., 2007. STS and the city: politics and practices of hope. Science, Technology and Human Values 32 (6), 713–734. Coutard, O., 2008. The urban impacts of reforms in the networks sector. In: Bamme, A., Getzinger, G., Wieser, B. (Eds.), Yearbook 2007 of the Institute for Advanced Studies on Science, Technology and Society. Profil, Munich and Vienna, pp. 135– 150. Coutard, O., Jaglin, S., Rutherford, J., 2008. Splintering Urbanism à la française? Les leçons d’un détour international. In: Aubertel, P., Ménard, F. (Eds.), La ville pour tous un enjeu pour les services publics. La Documentation française, Paris, pp. 117–134. with the collaboration of F. Boucher-Hedenström. Fernández-Maldonado, A.M., 2008. Expanding networks for the urban poor: water and telecommunications services in Lima, Peru. Geoforum 39 (6). Fernández-Maldonado, A.M., 2005. The diffusion of information and communication technologies in lower income groups. In: Coutard, O., Hanley, R., Zimmerman, R. (Eds.), Sustaining Urban Networks. The Social Diffusion of Large Technical Systems. Routledge, London and New York. Flux, 2004. Themed issue services en réseaux services sans réseaux dans les villes du sud. In: Jaglin, S., (Ed.), Flux, pp. 56–57. Graham, S., 2000. Constructing premium network spaces: reflections on infrastructure networks and contemporary urban development. International Journal of Urban and Regional Research 24 (1), 183–200. Graham, S., Marvin, S., 2001. Splintering Urbanism. Networked Infrastructures, Technological Mobilities and the Urban Condition. Routledge, London. Jaglin, S., 2008. Differentiating networked services in Cape Town: echoes of splintering urbanism? Geoforum 39 (6). Jaglin, S., 2001. Villes disloquées? Ségrégations et fragmentation urbaine en Afrique australe. Annales de Géographie 619, 243–265. Jaglin, S., 2005. Services d’eau en Afrique sub-saharienne: la fragmentation urbaine en question. CNRS Editions, Paris. Latham, A., 2006. Anglophone urban studies and the European city: some comments on interpreting Berlin. European Urban and Regional Studies 13 (1), 88–92. Kooy, M., Bakker, K., 2008. Splintered networks: the colonial and contemporary waters of Jakarta. Geoforum 39 (6). Legg, S., McFarlane, C. (Eds), 2008. Theme issue on ‘‘ordinary urban spaces: between postcolonialism and development”. Environment and Planning A 40, 6–130. Llorente, M., Zerah, M.-H., 1998. La distribution d’eau dans les villes indiennes: quels réseaux pour quels services?’. Flux 31–32, 83–89. Lorrain, D., 1997. Le politique à tous les étages (la construction des modèles de services urbains). In: Bagnasco, A., et Le Galès, P. (Eds.). Les villes en Europe, La Découverte, Paris, pp. 201–229. MacKillop, F., 2005. The Los Angeles ‘‘oligarchy” and the governance of water and power networks: the making of a municipal utility based on market principles (1902–1930). Flux 60–61, 23–34. MacKillop, F., Boudreau, J.-A., 2008. Water and power networks and urban fragmentation in Los Angeles: rethinking assumed mechanisms. Geoforum 39 (6). May, N., Veltz, P., Landrieu, J., Spector, T. (Eds.), 1998. La ville éclatée. Editions de l’Aube, La Tour d’Aigues, France. McFarlane, C., Rutherford, J. (Eds.), 2008a. Symposium on ‘‘Political infrastructures: Governing and experiencing the fabric of the city”. International Journal of Urban and Regional Research 32(2), 363–451. McFarlane, C., Rutherford, J., 2008b. Political infrastructures: governing and experiencing the fabric of the city. International Journal of Urban and Regional Research 32 (2), 363–374. Pflieger, G., Matthieussent, S., 2008. Water and power in Santiago de Chile: sociospatial segregation through network integration. Geoforum 39 (6). Rutherford, J., 2008. Unbundling Stockholm: the networks, planning and social welfare nexus beyond the unitary city. Geoforum 39 (6). Zérah, M.-H., 2008. Splintering urbanism in Mumbai: contrasting trends in a multilayered society. Geoforum 39 (6).
Contents lists available at ScienceDirect
Geoforum journal homepage: www.elsevier.com/locate/geoforum
Placing splintering urbanism: Introduction Olivier Coutard Université Paris-Est, LATTS (UMR CNRS 8134), ENPC, 6 Avenue Blaise Pascal, F-77455 Marne-la-Vallée Cedex 2, France
a r t i c l e
i n f o
Article history: Received 24 September 2008 Received in revised form 17 October 2008
Keywords: Infrastructures Neoliberalism Splintering Urbanism Europe United States Lower-income countries
a b s t r a c t This paper introduces a collection of case studies aimed at ‘‘Placing Splintering Urbanism”, in reference to the thesis developed by Graham and Marvin [Graham, S., Marvin, S., 2001. Splintering Urbanism. Networked Infractructures, Technoloical Moblilities and the Urban condition. Routledge, London]. Whilst acknowledging the value of the thesis as an analytical framework in opening the way to innovative understandings of contemporary urban dynamics, the paper argues that, taken together, the articles in this themed issue seriously challenge the ‘‘splintering urbanism” thesis theoretically, empirically and methodologically. They question in particular the postulated universality of the ‘‘modern infrastructural ideal” and of ‘‘unbundling” and ‘‘bypass” processes — all of which are key elements in Graham and Marvin’s argument — as well as the assertion that reforms in infrastructure sectors should generally result in more discriminatory, socially regressive patterns of provision of essential services and more splintered urban spaces. Based on these fundamental critiques, the paper concludes that one cannot speak of ‘‘splintering urbanism in general” — i.e., as a global trend — in any meaningful analytical way. Ó 2008 Elsevier Ltd. All rights reserved.
1. Introduction This themed issue brings together a set of papers which empirically explore the ‘‘splintering urbanism” thesis (Graham and Marvin, 2001). Individually and collectively, the papers aim therefore to reflect on the heuristic value as well as on the limits of Graham and Marvin’s argument, both as an analytical framework for researching specific urban contexts and as a wide-ranging thesis for explaining recent urban change. In both respects, the objective is to ‘place’ splintering urbanism – or, as the case may be, to displace it should it prove inadequate to properly account for the processes at play.
2. The splintering urbanism thesis In their influential book Splintering Urbanism, Stephen Graham and Simon Marvin argue that ‘‘a parallel set of processes are under way within which infrastructure networks are being ‘unbundled’ in
ways that help sustain the fragmentation of the social and material fabric of cities” (Graham and Marvin, 2001, p. 33)1. The splintering urbanism thesis articulates four main elements. (1) During approximately a century, until the 1960s, the development and governance of networked systems was embedded in a modern infrastructural ideal that supported the notion of monopolistic, integrated and standardised provision of network service (chapter 2). In particular, ‘‘a set of practices were developed to ensure the rapid roll-out of standardised infrastructure at equal price across national economic space” (p. 80); and ‘‘there are powerful resonances between the (...) modern infrastructural ideal and the colonialist policies shaping the attempted roll-out of infrastructure networks in developing cities” (p. 81). But from the late 1960s, this ideal was progressively undermined by a combination of powerful factors: the urban infrastructure ‘‘crisis”; changing political economies of urban infrastructure development and governance; neoliberalism and the withdrawal of the state; economic integration, urban competition and the imperatives of global–local connectivity; the development of infrastructural consumerism; the collapse
E-mail address: [email protected] In order to understand the thesis defended by Graham and Martin, and the discussion of that thesis here, it is useful to clarify the meaning of the notions at play: inequalities, segregation, differentiation, discrimination, splintering, secession, polarisation, withdrawal from (or collapse of) solidarity. The concept of urban splintering, in the sense in which we employ it, refers to the disintegration of former socio-economic interdependencies and to tendencies towards the withdrawal from (or collapse of) solidarities. This idea is the opposite of urban integration or cohesion, which emphasises the links of interdependence and solidarity that contribute to the cohesive operation of cities and therefore to the fact that cities are ‘society-making’. The impulses of political and fiscal secession that can be witnessed in Los Angeles, or the increasingly polarised settlement patterns in many urban areas, thus appear as symptoms of this fragmentation or splintering. The notion of splintering differs from that of socio-spatial segregation, which draws more on the idea of a clear spatial expression of economic or social inequalities. Clear segregation is completely compatible with, for example, a high degree of economic integration (see May et al., 1998; Jaglin, 2001). When analyzing the patterns of provision of essential infrastructure services, I shall furthermore distinguish between differentiation – i.e., providing different (groups of) users with different services – and discrimination – by which I shall specifically designate socially regressive forms of differentiation, i.e., forms of differentiation that are potentially or actually detrimental to lower-income or otherwise disadvantaged groups. 1
0016-7185/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.geoforum.2008.10.008
1816
O. Coutard / Geoforum 39 (2008) 1815–1820
of the comprehensive ideal in urban planning; new urban landscapes; and ‘‘new structures of feeling” (chapter 3, p. 92 sq.). (2) Concomitantly, within the context created by the resulting ‘‘collapse of the modern infrastructural ideal”, a profound transformation is occurring: ‘‘the economic liberalisation of infrastructure and the development of new [primarily information] technologies have made possible an entirely new infrastructural landscape that radically challenges established assumptions that have underpinned the relations between integrated networks and cities” (p. 139). Working together, they allow powerful coalitions of actors to promote the unbundling of infrastructures, i.e., the segmentation of integrated infrastructure into different network elements and service packages (p. 141). Unbundling in turn allows for bypass strategies, i.e. strategies that seek the connection of ‘‘valued” or ‘‘powerful” users and places, while at the same time bypassing ‘‘non-valued” or ‘‘less powerful” users and places (chapter 4). (3) These bypass strategies contribute to the emergence of socalled premium networked spaces (Graham and Marvin, 2001, p. 249, sq.)2. In particular, elite or higher-income groups are increasingly living in ‘‘secessionary” places/spaces that are ‘‘withdrawn from the wider urban fabric” (Graham and Marvin, 2001, p. 268) in various ways (chapter 6, esp. box 6.4, p. 268–71), yet intensively connected to other, remote premium spaces, thus forming archipelagos of ‘‘global enclaves” (p. 389). This reinforces the ‘‘vicious cycle” of splintering, ‘‘where attempts at socio-technical secession lead to greater fear of mixing, so increasing pressure for further secession, and so on” (Graham and Marvin, 2001, p. 383). The widening gap between connected and unconnected (or disconnected) places and people is all the more worrisome since the world we live in is, increasingly, a network society (Castells, 1996) in which ‘‘the poverty that matters is not so much material poverty but a poverty of connections,” which ‘‘limits a person or group’s ability to extend their influence in time and space” (Graham and Marvin, 2001, p. 288). Central to the argument, therefore, is the notion that ‘‘the diverse political and regulatory regimes that supported the roll-out of power, transport, communications, street and transport networks towards the rhetorical [sic] goal of standardised ubiquity are, in many cities and states being ‘unbundled’ and ‘splintered’ as a result of a widespread movement towards privatisation and liberalisation.” (p. 382) (4) Communities and public authorities in cities around the globe confronted with splintering processes and tendencies may develop forms of ‘‘resistance” (p. 387 sq.). As a result ‘‘a central theme of urban politics and urban social movements in the first decades of the new millennium will therefore centre on the struggle between the ‘global’ forces of attempted, ‘pure’ boundary control and the customisation of premium, commodified network spaces, vs. the imperative of infrastructural, urban and technological democratisation and the need for more egalitarian and democratised practices and principles of development” (p. 405). In this editorial, I wish to briefly discuss the splintering urbanism thesis in light of the arguments and reflections developed in the case-study based papers forming this themed issue3. I will develop my argument in five points, which respectively address: the limits of the modern infrastructural ideal; the politics of infrastructure reforms; the postulated universality of the notions of unbundling and bypass; the urban ‘‘effects” of patterns of service provision; and methodological issues.
2 Economic spaces (e.g., foreign direct investment enclaves or business improvement districts), residential spaces (e.g., gated communities) and ‘‘social life spaces” (such as commercial malls and theme parks). 3 In this editorial, I will signal the articles forming part of the themed issue by writing the names of their author(s) in italics.
3. The deceits of the ‘‘modern infrastructural ideal’’ The notion of a ‘‘modern infrastructural ideal” developed by Graham and Marvin is convincing insofar as it refers to the enduring preference for ‘‘bundled” infrastructure and the related disregard of alternative (whether ‘‘traditional” or ‘‘decentralized”) systems of service provision by the public authorities responsible for this provision. Yet the epochal narrative associated with the modern infrastructural ideal is less convincing. The story goes that, in general, the bundled, monopoly structure of infrastructures sustains (or has sustained) the universal provision of standardised services, consequently reinforcing urban socio-spatial integration. But in-depth analyses of the long-term interactions between network provision and urban dynamics in many cities tell a different story in at least two important respects. They suggest, first, that although the industrial organisation4 of infrastructure sectors certainly matters in the universalisation of basic infrastructure services, it always does so in combination with other factors (see, e.g., Bocquet, Chatzis and Sander, 2008). Three factors seem to be of particular importance here: the will and capacity of government (State or local public authorities) to facilitate, and render solvent, the process of universalisation; public control over land occupation and urban development, as well as housing policies; a small rather than large proportion of households in extreme poverty (Coutard, 2008). In the context of many cities in lower-income countries, for example, these conditions have often not been met. Hence, rather than the ‘‘collapse” of the modern infrastructural ideal, several articles in this themed issue point to its absence (Kooy and Bakker, 2008) or to its failure (Botton and de Gouvello, 2008; Fernández-Maldonado, 2008; Zérah, 2008; see also Jaglin, 2005). As Kooy and Bakker (2008) suggest for Jakarta, many cities in lower-income countries are not splintering, they are and have long been splintered along ethnic or socio-economic lines (see also McFarlane and Rutherford, 2008b, p. 370)5. Second, historical analysis suggests that the relations between universal, homogenous service and urban integration are often ambivalent. In Santiago, for example, the connection of all residential units to the essential networks has, in particular since the 1980s, facilitated and legitimated a strong policy of spatial segregation and functional, social, fiscal and political fragmentation at the local authority level within the urban region as well as processes of urban sprawl, also segregated (Pflieger and Matthieussent, 2008). In Los Angeles, the ‘‘universalisation” of essential services (water, then water and electricity) by a municipally-run entity in the first decades of the twentieth century was primarily designed to serve the interests of the local oligarchy who controlled the municipality and, at the same time, owned the vast tracts of desert land that were open to urbanisation – and the value of which was massively increased – by the provision of those services (MacKillop, 2005). While the control of the water resource and cheap electricity supplies were decisive factors in the political integration of the city of Los Angeles in its current borders6,7, it also favoured (and heavily subsidised) a process of spatial enlargement which, when
4 Single vs. several suppliers; public vs. private ownership; competition vs. planned development. 5 For lack of space, I will not enter here into the (post)colonial dimension of the debate, in spite of its importance and of the fact that several papers in this themed issue participate in this debate. Two recent themed issues address this dimension from complementary perspectives (McFarlane and Rutherford, 2008a; Legg and McFarlane, 2008). 6 Both historically and in the contemporaneous secession debate. 7 Note that political integration between previously or otherwise independent constituencies/local governments through (or for) infrastructure development is frequently observed. Strangely enough, Graham and Marvin do not discuss this often positive correlation.
O. Coutard / Geoforum 39 (2008) 1815–1820
coupled with the historic propensity of the Angelena population to self-segregate amongst themselves, then created a fertile environment for a social and functional splintering of the urban region (growing rejection of social blending, and a functional poly-centrism) and the recent revival of political secession (Boudreau and MacKillop, 2008). In Buenos Aires (Botton and de Gouvello, 2008), a policy strictly conforming to the modern infrastructural ideal resulted in the creation of a premium networked space (in the central part of the capital federal) delimited by the perimeter of the water mains and excluding half the population of the urban region. This apparent failure of the modern infrastructural ideal within a nationally strategic space is interesting in itself. It also raises an important conceptual issue with regard to the integrating or splintering effects of standardised network supplies. Indeed, irrespective of the specific territorial limits of the authority in charge of providing the service, from what spatial scale of universal distribution of a network can one, and should one, consider that its integration effects (within the zone it serves) dominate its fragmentation effects (at its outer edges), i.e. where should we draw a line/boundary for evaluating the dynamics of network provision? Hence while the notion of a ‘‘modern integrated ideal” elaborated by Graham and Marvin is convincing, its articulation with other social, spatial and economic processes and policies is not sufficiently thought through. This is especially regrettable considering its importance in Graham and Marvin’s general argument as the antithesis of splintering urbanism. In particular, the ‘‘collapse” of the modern ideal should not be primarily interpreted as the result of external or externally-determined factors. It should be analysed first as the result of its historical failure, in many contexts, to fulfil its promise of delivering and sustaining universal, affordable, adapted services. In this regard, the evocation of an ‘‘urban infrastructure crisis” deserves to be unpacked and explored in more detail (Graham and Marvin, 2001, pp. 92–94). Furthermore, where the universalisation of essential infrastructure services has been achieved, more attention should have been devoted to its articulation with other policies in the shaping of long-term urban or territorial dynamics. 4. The politics of infrastructure reforms Networked infrastructures and services have been one of the favourite playgrounds of neoliberal economists and their governmental supports worldwide under the ‘‘Washington consensus”. There is no denying the influence in the regulation of utility sectors of a ‘‘liberal convention” (Batley, 1999) characterised by: the ringfencing of individual utility sectors; the commercialisation of services (involving full cost pricing, the shift to a for-profit activity in which profit maximisation has priority over service provision, and the substitution of the economic notion of willingness to pay for the more social notion of ability to pay); the privatisation of companies (and its implications in terms of equity financing); and a change in redistributive principles from social equity to economic equity (via the elimination of cross-subsidies). But much as the modern infrastructural ideal did not deploy its ‘‘bundling logic” universally, the recent wave of reforms that have affected utility sectors have not resulted in a fully ‘‘neoliberal pattern” of service provision, whatever this might mean. The imprint on utility reforms of the liberal convention has systematically been combined with other sociopolitical processes and policies8. As a result, post-reform patterns of provision are, as a rule, quite remote from pure competition and full commodification. This partly reflects
8
5).
This is in line with Castree’s argument on ‘‘neoliberalism plus” (Castree, 2006, p.
1817
the economic characteristics of networked services, with their specific combination of economies of scale and scope, club effects and transactions costs. More importantly, it reflects the deeply political nature of basic infrastructure services, an intrinsic feature of these sectors (Lorrain, 1997; McFarlane and Rutherford, 2008a)9. Many researchers have stressed how ‘‘textbook reforms” have been radically transformed during the sociopolitical process of their design and implementation. For instance, Botton and de Gouvello (2008) analyse how the ‘‘political economy” of the water concession contract in Buenos Aires was radically transformed during the first ten years of its existence to cope with its potential or observed segregation effects. Rutherford (2008) discusses the policy instruments employed in Stockholm that have prevented – so far – the actualisation of the potential regressive effects of utility reforms. Even when public regulation is shaky, as in the case of water and telecommunications in Lima (Fernández-Maldonado, 2008), the outcome of reforms may not be altogether regressive. To a significant extent, the liberalisation of telecommunications in Peru has improved access to internet services in low-income Lima neighbourhoods (see also Fernández-Maldonado, 2005; and for a similar argument on the political nature of infrastructure reforms and the shaping of their ‘‘effects”, see Coing, 2005 on the case of Bogotá). Importantly, this politics of infrastructure reforms does not boil down to a simple dialectic between regressive neoliberal reform and socially progressive ‘‘resistance”. Several papers in this themed issue point to the analytical and heuristic limitations of ‘‘resistance narratives”10. The splintering urbanism thesis implies that people (governments, communities) are faced with overarching dynamics of liberalisation, and that they are left with the simple alternative of either passively accepting these dynamics and their disruptive socio-spatial implications, or outright rejecting them. In many cases, this is not how things happen. For instance, the extensive diffusion of prepayment meters in Cape Town analysed by Jaglin (2008) is not solely a story of resistance by militant groups to the adoption of the intrinsically discriminatory technologies. Rather it is the story of how a particular technology can be appropriated by various actors in various ways according to a diverse set of considerations. These forms of social negotiation around technological change are largely ignored in the splintering urbanism thesis (see also Coutard and Guy, 2007, esp. p. 724)11. The dual interpretation proposed, expressed in terms of generalised neoliberalisation and sporadic manifestations of resistance to this, does not properly reflect the modalities of network service reforms and the complex composition of social interests which shape them and their ‘‘effects”. 5. Questioning the universality of unbundling and bypass Turning to the social and spatial effects of reforms, I would like to question the descriptive, analytical and normative value of the key notions of unbundling and bypass.
9 I obviously do not imply by this that liberalisation per se is apolitical. I just wish to stress that if one wants to form a proper judgment on the causes, processes and effects of reforms in infrastructure sectors, one should take into account the (multidimensional and multiscalar) policies that, together, shape infrastructure reforms. 10 Narratives of resistance to neoliberalism in various domains of social activity abound. See Barnett (2005) and Castree (2006) for an interesting point on resistance narratives as ‘‘consolation for leftist academics”: The vocabulary of neoliberalism, they argue, ‘‘invites us to align our own professional roles with the activities of various actors ‘out there’ who are always framed as engaging in resistance or contestation” (Barnett, 2005, p. 10). 11 Despite Graham and Marvin’s statement, with which I fully agree, that ‘‘the reality of technological innovation, the restructuring of urban landscapes and social and cultural changes in cities, are a great deal more ‘messy’, difficult, contingent and open to contested interpretations and application than [deterministic splintering] scenarios imply” (p. 388).
1818
O. Coutard / Geoforum 39 (2008) 1815–1820
The notion of unbundling, which lies at the heart of the splintering urbanism thesis, suggests that the provision of basic services was previously ‘‘bundled.” In cities in lower-income countries, at least, this has usually not been the case, and the standard pattern of service provision during the ‘‘monopoly era” was very much already an unbundled pattern, with diverse suppliers and diverse forms of supply12. Hence the notion of unbundling is deceptive when it comes to studying the cities (especially in lower-income countries) where recent evolution does not involve a passage from an integrated system to an unbundled one, but rather a passage from one more or less unbundled system to another. In these urban contexts, the dualistic framework of bundled/unbundling infrastructure does not allow us to analyse if, in what measure and under what conditions the evolutions observed are more conducive or, conversely, harmful to the generalisation of access to the service in question (Jaglin, 2008; Kooy and Bakker, 2008; Botton and de Gouvello, 2008; see also Flux, 2004). Likewise, the nature as well as the socio-spatial and sociopolitical meaning of bypass processes varies too greatly from one context to the other to be usefully and heuristically apprehended as variations of the same overarching phenomenon. In many cities in Western Europe, for example, the combined effect of the reforms and the introduction of new information and communications technologies have effectively led to an increase in the differentiation of service offers. These differentiations have often led to a better value for money for most – and sometimes for all – customers. They have also frequently given rise to discrimination and increased inequalities in terms of access to services. Nevertheless, the scale of these inequalities and discriminations has remained limited (problems of affordability, debts, and temporary disconnection). It does seem, in effect, socially inconceivable and politically unacceptable that a large number of households be durably (much less permanently) denied access to essential services because they are insolvent (see Rutherford, 2008; see also Coutard et al., 2008). On the other hand, in most of the cities in lower-income countries examined in this themed issue, the relationship between solvency and effective access to services (in the zones served) is stronger, largely irrespective of the public or private status of the operator. In Cape Town, insolvent households risk being permanently disconnected from their (municipality-run) electricity supply, or even being evicted from their homes (Jaglin, 2008). Although the proportion of households affected is small, the fragmentation logic is obvious. In Jakarta, the enduring exclusion for the majority of Jakarta’s poor from piped water supply also reflects fragmentation patterns, although not primarily rooted in recent reforms (Kooy and Bakker, 2008). In both cases (and in many others), the social meaning of ‘‘bypass” is very different from that of the forms of ‘‘soft” socio-spatial differentiations observed in other cities in the ‘‘global North”. In this perspective, the case of Santiago, Chile (Pflieger and Matthieussent, 2008), an emerging city marked by very strong socioeconomic disparities, but equipped with basic universal networks, stands as an exception in the sample of cities examined here. In Santiago the modalities of supplying essential services appear to be more integrationist: certainly a small proportion of households is excluded from the welfare system for failing to follow the budgetary discipline imposed by social services; but the analysis suggests that, once a network is universally accessible, the public authorities are obliged to guarantee the effective access of the greatest number of people to the essential service thus provided. This suggests a positive correlation (although probably not a one 12 e.g., the combination of (usually restricted) residential tap water supply, standpipes and water carriers of all sorts typical of water provision in many cities in lower-income countries.
way causality) between the universal coverage of a network infrastructure and the ‘‘right” to access the basic services it provides. 6. The urban effects of infrastructure services: toward integration by differentiation? The postulated socio-spatial effects of infrastructure unbundling and bypass are captured in particular in the key notion of premium networked spaces. Throughout their book, Graham and Marvin indifferently use the notions of ‘‘premium network spaces” (see also Graham, 2000) and ‘‘premium networked spaces”, defined as ‘‘new or retrofitted transport, telecommunications, power or water infrastructures that are customized precisely to the needs of powerful users and spaces, while bypassing less powerful users and spaces” (Graham, 2000, p. 185). This definition is intriguing as it seems to postulate a univocal relation between urban spaces and infrastructures (or infrastructure spaces). More importantly, the notion of premium network spaces suggests that, in general, the quality of a given space is determined by the quality of network services supplied to it. In contrast, I would support the view that, in most instances, the notion of ‘‘premium networked spaces” is more appropriate, because it refers more explicitly to social or economic/fiscal premium spaces supplied with tailored network services (see also Coutard, 2002). But we can question whether it is always the case, as the splintering urbanism thesis implies, that increased service differentiation aggravates socio-spatial discrimination. Several papers in this themed issue offer counter-perspectives. For example, water supply service in Cape Town (Jaglin, 2008) is marked by a strong social regulation producing important solidarities between the city’s rich and poor zones/populations (at the price, financing, technical and management levels). However what appears as a possible factor of splintering is the insufficient differentiation of the service, because it incites both the wealthier and the poorer to leave the network, in the former case because the service does not suit them, in the latter case because they lack the means to pay. In the context of a Cape Town type situation where the networks are incomplete, the most appropriate public intervention may well be, at least during a long intermediate phase, to regulate the alternative supply offers to the network (quality of service, price, business profit margins) (see also Zérah, 2008; and for a similar argument in the context of Delhi, see Llorente and Zérah, 1998). The question whether, in the longer term, this creates more or less irreversible discriminations between spaces within urban regions is a crucial but open one. Symmetrically, the forms and degree of what could turn out to be a ‘‘desirable” differentiation are very much dependent upon the context and deserve careful examination. Both issues ultimately have to do with the regulatory and coordinating capacities of public authorities (see Bénit-Gbaffou, 2008, for a similar point with regard to the ‘‘pluralisation” of policing and security services in Johannesburg). In sum, reforms that involve an increased differentiation in services supplied do not necessarily lead to increased segregation (inequality) or fragmentation (removal of solidarity) in terms of access to essential services. The social and urban impact/signification of these reforms – and of different regimes of service provision more generally – can only be appreciated on the basis of an in-depth analysis of the various interactions, over a period of time, between these reforms and the wider context within which they take place.
7. Placing splintering urbanism: from universal patterns to contextual diversity, and back My final point is primarily methodological: it has to do with the issue of evidence. The splintering urbanism thesis rests on numer-
O. Coutard / Geoforum 39 (2008) 1815–1820
ous anecdotes (see in particular the use made of text boxes) taken from very different ‘‘contexts” that all point to splintering processes and hence support the general thesis. But the accumulation of splintering stories is no demonstration. This approach leaves little room for counterfactual analysis. It leaves the authors open to the suspicion of having silently passed over developments every bit as significant as the ones they cite, but which do not support the arguments they seek to advance. Hence the quite different approach used in all the articles forming this themed issue: a monographic approach which seeks to first account for the coevolution of urban infrastructures and the urban fabric in context (as a result of various historical and contemporary policies and processes) in order to then be able to address the relevance of the splintering urbanism thesis in the case at hand within a more ‘‘symmetric” framework, i.e. an analytical framework allowing the possibility that socio-spatial discrimination processes are actually not at work, or that they are so much intertwined with other processes that one cannot significantly refer to a universal, overarching ‘‘splintering” trend. What, then, do the case studies assembled in this themed issue add to the knowledge and understanding of the interactions between the provision of infrastructure services and urban dynamics? The response to this question closely resonates with Castree’s discussion of the ‘‘epistemology of particulars” (Castree, 2005). As Castree convincingly argues, case study research should not just aim at documenting context-related differences between otherwise supposedly analogous manifestations of an overarching phenomenon (in our case: liberalisation, unbundling, bypass, premium networked spaces, splintering and so on). Rather it should aim at identifying substantive similarities or differences between processes in the various contexts studied, in order to shed light on the important question of whether or not the postulated generic phenomenon exists, in a substantive sense of the word. As he further argues in his commentary on recent geographical research on the ‘‘neoliberalisation of nature” (Castree, 2006), ‘‘If, in terms of causal processes and outcomes, neoliberal policies turn out to have highly specific, even unique, environmental impacts from situation to situation, then this idiographic finding should not be glossed in the desire to identify ‘general’ patterns that can then be used to condemn a nonexistent ‘neoliberalism in general’” (page 6). What the case studies gathered in this themed issue strongly suggest is that one cannot speak of ‘‘splintering urbanism in general” in any meaningful analytical way. Processes of infrastructure reforms and transformations are too diverse from one context to the other and from one sector to the other. More importantly, the social significance and implications of these processes are different not only in degree but also in kind between, for example, contexts in which access to basic services is universal and contexts in which it is not. In this regard, the parallel established or suggested between cities in higher-income (long industrialised) countries and in lower-income ones is misleading. Theoretically, the key point is that such case studies invite us to shift the focus from abstract notions, such as the ‘‘modern integrated ideal” or ‘‘bypass”, to the identification and understanding of the combination of values (ideals?), interests and instruments actually mobilised by the actors involved in the planning, development, management or use of infrastructure networks (for a similar argument, see McFarlane and Rutherford, 2008b, p. 370). Placebased case studies need to articulate many dimensions of the infrastructure policies and of the context in which they are deployed, such as: the legal status of services provided, the subscription options available, the structure of real estate property ownership, etc. As a consequence, ‘‘the process of universalization, far from being the inexorable and linear outcome of the action of a small number of ‘major historical forces’, was the upshot of the actions of a whole range of actors acting in a specific geographical, social, economic
1819
and technical context”. (Bocquet, Chatzis and Sander, 2008). This focus on micro processes or variables does not imply, however, that no general lessons can be drawn from such studies. Indeed, as Bocquet et al. (2008) argue, micro analyses allow to ‘‘point up (micro) ‘processes’ and ‘variables’ that may be identified individually in a whole range of domains” and contexts, although they may well be, and usually are, diversely articulated in different contexts. Last but not least, in addition to its methodological and theoretical value, this case-study based approach also has the advantage of reopening the debate on urban futures beyond the dispiriting vision of more or less helpless local communities seeking to (transitorily) resist the inexorable advance of neoliberalism (see also Coutard and Guy, 2007; and for a similar argument Latham, 2006). This would be in line with the objectives outlined in Graham and Marvin’s ‘‘manifesto for a progressive networked urbanism” (p. 404 sq.) resting upon a ‘‘spatial imagination for the splintered metropolis” (p. 405), a perspective which is strangely and regrettably absent from their book. In sum, interpreted as a thesis, the splintering urbanism narrative invites serious theoretical and methodological critiques. First, the absence of any real historical analysis of the phenomena at play gives rise to a certain doubt as to the strength of the conclusions which contrast the current period with a former period characterised by an ‘integrated ideal’. Given the absence of any convincing interpretation as to the reasons why, historically, certain essential networks (water/sewage, electricity, telephone) were made universally available in certain contexts and not in others, the authors struggle to define the specificity of today’s ‘‘neoliberal” era. The clash between the modern infrastructural ideal and the contemporary era appears to be largely unfounded. To a large extent, it seems rather to rest on an artificial rapprochement between, on the one hand, a historical model largely confined to the higher-income countries of Western Europe and North America and, on the other, contemporary processes generally observed in cities in lower-income countries. This thesis also rests on the idea that changes in industrial organisation and the regulation of networked services conform to some universal neoliberal dynamic and that the urban context within which these changes occur has a limited impact on either their modalities or their effects. This universalist vision is impregnated with a soft economic determinism – hardly relieved by the mention of more or less sporadic, local forms of resistance to this neoliberal logic – which the results of the place-based case studies in this themed issue clearly question. Finally the splintering urbanism thesis suggests a positive connection between an integrated offer of networked services and urban integration (and in a symmetric manner between unbundled networks and urban splintering) without the authors ever offering themselves the means to validate (or invalidate) this hypothesis. Indeed, on the methodological level, the demonstration as a whole suffers from an overly superficial empiric validation, advancing by means of an accumulation of rapidly examined examples rather than through detailed empirical works. In closing, however, I would like to emphasise that, putting aside the thesis, Splintering Urbanism offers a fruitful and valuable analytical framework13 highlighting how an infrastructure perspective is key to the understanding of contemporary urban dynamics (and, for that matter, of territorial dynamics on all geographical scales). In this sense, the narrative developed by Graham and Marvin should be interpreted as prefigurative social theory rather than as 13 For the use of splintering urbanism as an analytical framework, see Zérah, 2008; Rutherford, 2008; and Kooy and Bakker, 2008. I thank S. Jaglin for her help in clarifying the distinction between the two interpretations.
1820
O. Coutard / Geoforum 39 (2008) 1815–1820
proven hypothesis14. It rightly points to the context-shaped, and context-shaping, dynamic interactions between infrastructure provision and urban socio-spatial configurations and functioning – even though it does not thoroughly investigate these complex interactions. And it opens the way15 to innovative understandings of contemporary urban processes and policies, thereby sustaining the exciting research agenda set forth in the postscript of the book (p. 404). Acknowledgments The majority of the papers forming this theme issue (all except Kooy and Bakker, Bénit-Gbaffou, and Fernández-Maldonado) originate in a research programme I coordinated from 2002 to 2005. Preliminary versions of all papers were discussed at the Placing Splintering Urbanism international workshop in Autun (France) in June of 2005. I acknowledge support from several French institutions: the Centre National de la Recherche Scientifique (CNRS); the Agence Nationale de la Recherche (ANR Grant No. ANR-05BLAN-0344); the Ministry of Sustainable Development (MEEDDAT); and the Ecole nationale des ponts et chaussées (ENPC). I wish to thank Sylvy Jaglin and Jonathan Rutherford, both at Latts, as well as Dominique Lorrain (CNRS) for the wealth of stimulating conversations we have had over the past few years on splintering urbanism; and Sylvy, Jon and Karen Bakker for their very helpful comments on a draft version of this paper. Finally I would like to thank Neil O’Brien who translated from the French several articles in this themed issue. References Barnett, A., 2005. The consolation of ‘‘neoliberalism”. Geoforum 36, 7–12. Batley, R., 1999. The Role of Government in Adjusting Economies: An Overview of Findings. The International Development Department, University of Birmingham. Benit-Gbaffou, C., 2008. Unbundled security services and urban fragmentation in post-apartheid Johannesburg. Geoforum 39 (6). Bocquet, D., Chatzis, K., Sander, A., 2008. From free good to commodity: universalizing the provision of water in Paris (1830–1930). Geoforum 39 (6). Botton, S., de Gouvello, B., 2008. Water and sanitation in the buenos aires metropolitan region: fragmented markets, splintering effects. Geoforum 39 (6). Castells, M., 1996. The Rise of the Network Society (The Information Age: Economy, Society and Culture Vol. I). Blackwell, Cambridge, MA, Oxford, UK. Castree, N., 2005. The epistemology of particulars: human geography, case studies and ‘‘context”. Geoforum 36, 541–544. Castree, N., 2006. Commentary. From neoliberalism to neoliberalisation: consolations, confusions, and necessary illusions. Environment and Planning A 38, 1–6. Coing, H., 2005. Public utilities in Bogotá: the impact of reforms on territorial and social cohesion. In: Paper Presented at the International Seminar Placing Splintering Urbanism: Changing Network Service Provision and Urban Dynamics in Cross-National Perspective, June 2005, Autun, France, pp. 22–24. Coutard, O., 2002. Premium network spaces: a comment. International Journal of Urban and Regional Research 26 (1), 166–174.
14
I thank Karen Bakker for this point. It does so dialectically, as this line of research is supported by the need/desire to move beyond the splintering urbanism narrative. 15
Coutard, O., Guy, S., 2007. STS and the city: politics and practices of hope. Science, Technology and Human Values 32 (6), 713–734. Coutard, O., 2008. The urban impacts of reforms in the networks sector. In: Bamme, A., Getzinger, G., Wieser, B. (Eds.), Yearbook 2007 of the Institute for Advanced Studies on Science, Technology and Society. Profil, Munich and Vienna, pp. 135– 150. Coutard, O., Jaglin, S., Rutherford, J., 2008. Splintering Urbanism à la française? Les leçons d’un détour international. In: Aubertel, P., Ménard, F. (Eds.), La ville pour tous un enjeu pour les services publics. La Documentation française, Paris, pp. 117–134. with the collaboration of F. Boucher-Hedenström. Fernández-Maldonado, A.M., 2008. Expanding networks for the urban poor: water and telecommunications services in Lima, Peru. Geoforum 39 (6). Fernández-Maldonado, A.M., 2005. The diffusion of information and communication technologies in lower income groups. In: Coutard, O., Hanley, R., Zimmerman, R. (Eds.), Sustaining Urban Networks. The Social Diffusion of Large Technical Systems. Routledge, London and New York. Flux, 2004. Themed issue services en réseaux services sans réseaux dans les villes du sud. In: Jaglin, S., (Ed.), Flux, pp. 56–57. Graham, S., 2000. Constructing premium network spaces: reflections on infrastructure networks and contemporary urban development. International Journal of Urban and Regional Research 24 (1), 183–200. Graham, S., Marvin, S., 2001. Splintering Urbanism. Networked Infrastructures, Technological Mobilities and the Urban Condition. Routledge, London. Jaglin, S., 2008. Differentiating networked services in Cape Town: echoes of splintering urbanism? Geoforum 39 (6). Jaglin, S., 2001. Villes disloquées? Ségrégations et fragmentation urbaine en Afrique australe. Annales de Géographie 619, 243–265. Jaglin, S., 2005. Services d’eau en Afrique sub-saharienne: la fragmentation urbaine en question. CNRS Editions, Paris. Latham, A., 2006. Anglophone urban studies and the European city: some comments on interpreting Berlin. European Urban and Regional Studies 13 (1), 88–92. Kooy, M., Bakker, K., 2008. Splintered networks: the colonial and contemporary waters of Jakarta. Geoforum 39 (6). Legg, S., McFarlane, C. (Eds), 2008. Theme issue on ‘‘ordinary urban spaces: between postcolonialism and development”. Environment and Planning A 40, 6–130. Llorente, M., Zerah, M.-H., 1998. La distribution d’eau dans les villes indiennes: quels réseaux pour quels services?’. Flux 31–32, 83–89. Lorrain, D., 1997. Le politique à tous les étages (la construction des modèles de services urbains). In: Bagnasco, A., et Le Galès, P. (Eds.). Les villes en Europe, La Découverte, Paris, pp. 201–229. MacKillop, F., 2005. The Los Angeles ‘‘oligarchy” and the governance of water and power networks: the making of a municipal utility based on market principles (1902–1930). Flux 60–61, 23–34. MacKillop, F., Boudreau, J.-A., 2008. Water and power networks and urban fragmentation in Los Angeles: rethinking assumed mechanisms. Geoforum 39 (6). May, N., Veltz, P., Landrieu, J., Spector, T. (Eds.), 1998. La ville éclatée. Editions de l’Aube, La Tour d’Aigues, France. McFarlane, C., Rutherford, J. (Eds.), 2008a. Symposium on ‘‘Political infrastructures: Governing and experiencing the fabric of the city”. International Journal of Urban and Regional Research 32(2), 363–451. McFarlane, C., Rutherford, J., 2008b. Political infrastructures: governing and experiencing the fabric of the city. International Journal of Urban and Regional Research 32 (2), 363–374. Pflieger, G., Matthieussent, S., 2008. Water and power in Santiago de Chile: sociospatial segregation through network integration. Geoforum 39 (6). Rutherford, J., 2008. Unbundling Stockholm: the networks, planning and social welfare nexus beyond the unitary city. Geoforum 39 (6). Zérah, M.-H., 2008. Splintering urbanism in Mumbai: contrasting trends in a multilayered society. Geoforum 39 (6).