Jump to ContentJump to Main Navigation
Marxism and the City$

Ira Katznelson

Print publication date: 1993

Print ISBN-13: 9780198279242

Published to Oxford Scholarship Online: January 2005

DOI: 10.1093/0198279248.001.0001

Remapping the City

Chapter:
(p.257) 7 Remapping the City
Source:
Marxism and the City
Author(s):

Ira Katznelson (Contributor Webpage)

Publisher:
Oxford University Press
DOI:10.1093/0198279248.003.0007

Abstract and Keywords

Section I of this chapter discusses how, by not embarking on the journey linking city space, capitalist development, and class formation, Marxism denied itself a critical dimension in the material analysis both of the target it wished to confront and of the class it expected to be the agent of this successful engagement. Section II looks at how the separation between the social classes within the new social geography of the capitalist city in the nineteenth century helped assure the residential propinquity of members of the working class, as well as their isolation from other classes. However, with the elaboration of new networks made possible by the nationalization of labour markets, there was a growing sense that working classes shared a fate that transcended given localities, while advances in communications and transportation made the ties between class and space more complicated and tentative. Analyses are included of this break in working‐class history given in the work of Krishan Kumar and Craig Calhoun, and by Olivier Zunz and Richard Oestreicher in their studies of Detroit at the turn of the nineteenth century. Sections III–V show that the relationship of Marxism and the city and urban space now stands on unsure ground, since it is the politics and viability of class itself as the dominant form of collective identity that is currently under challenge; the discussion given here draws on the work of Mark Gottendiener and Eric Hobsbaum within the new urban Marxism.

Keywords:   capitalist cities, capitalist development, cities, class formation, collective identity, Craig Calhoun, Hobsbaum, Krishan Kumar, Mark Gottendiener, Marxism, Olivier Zunz, Richard Oestreicher, separation of social classes, urban Marxism, urban space, viability of class, working classes

We have seen how the making of class categories that shaped the identities, motives, and actions of early working classes proved far more contingent than Engels foresaw,1 and how the ways in which class emerged as a coherent marker and guide for working people may be understood in terms of how working classes mapped the capitalist city. Taking this approach, it is possible to make sense of why in some circumstances, but not all, class came to define the identities of political and social actors beyond the workplace, in the residential community and the home. So, too, an urban‐spatial imagination can assist in addressing other significant puzzles for Marxism with regard to class and class conflict. At no time since the early nineteenth‐century history of working‐class formation have working classes fulfilled their scripted role as agents of fundamental social change in the theory's drama of transition to socialism. In the late twentieth century, moreover, as capitalism is manifestly being restructured on a global scale, the relative significance of class categories themselves is in question. An engagement of Marxism with the city can shed some light on these problems by helping us to understand why militancy in the full Marxist sense has existed only in the realm of theory and by giving us a vantage‐point from which to consider current patterns of class and group formation.

(p.258) I

By not embarking on the journey linking city space, capitalist development, and class formation, Marxism denied to itself a critical dimension in the material analysis both of the target it wished to confront and of the class it expected to be the agent of this successful engagement. This point is best understood if we think again about the content of the second of the four levels of class, that of the social organization of society experienced as ways of life within the various capitalist countries.

There is a fateful compression at the heart of Marxist social theory. Because exploitation, the extraction of surplus value from the proletariat by the bourgeoisie, is the key antagonistic and dynamic feature of class relations and the most important macro‐analytic characteristic of the ‘laws of motion’ of capitalism at the first level of class analysis, that of the structure of capitalist economic development, Marxism has yielded to the temptation at the second level to focus almost exclusively on the capitalist workplace. The most common Marxist approaches conflate these two levels and treat the abstract logic of capitalism and the site of the workplace together as ‘the economy’, the ‘base’ of capitalist social, political, and ideological relations. Whether they work by deduction or induction, virtually all Marxist approaches to social class develop definitions and categories based on an analysis of the technical and social relations of production; such is the case where class is treated abstractly, as in such categories as productive and unproductive labour, or where class is understood more concretely in terms of the attributes of people within the active labour force.

There are many costs to this analytical strategy. One, which I have noted elsewhere, is the elimination ‘in one stroke [of] a series of important questions about the connections between key aspects of capitalist accumulation and national economic histories on one side and the organization of labor markets and workplaces on the other. As any student of capitalist industrialization knows, the growth and expansion of capitalism has proved capable of fostering many different kinds of workplaces (p.259) and work.’2 A second cost, more central to the analysis here, has been the suppression of the off‐work characteristics of social organization. Nineteenth‐ and twentieth‐century industrial capitalism has been in the main an urban capitalism. Cities have concentrated working classes. Even when we define the working class in Marxist terms as a group composed of individuals who work for wages within capitalist workplaces and labour markets and are located in positions of subordination within the hierarchies of businesses and firms (and if we leave aside the puzzle of the class position of those who do not enter a labour market to sell their labour power for a wage), we do well to remember that they also live in particular kinds of residential areas where they are enmeshed in the economic (as well as social, political, and cultural) relations of housing, domesticity, and consumption of both private and public goods. Like sites of work, these places are also the products of the dynamics of capitalist development operating at the first level of class. Although living places and housing markets have not been shaped and determined by capital alone, of course, there have been important features common to cities in capitalism, including those of social geography, because these cities have been organized principally to accommodate the requirements of capitalism as a dynamic system of growth and expansion.

Enmeshed in the social and political relations of cities, working‐class people have experienced the urban face of capitalism. They have done so not only in the structurally and spatially segregated world of work, but at home and in community life in the realm of residence. By engaging with the city to confront the various dimensions of the social organization of class at work and away from it, Marxism can better address its disappointments about the working class. Down this road, Marxism can better understand why its most cherished expectations of revolutionary consciousness have never (p.260) been more than plausible in theory, and, more positively, how today Marxism might be rescued from intellectual obsolescence while it discovers a new modesty.

Marxist social theory has sought to do more than understand day‐to‐day life within capitalism or make sense of determination and of structure and agency as analytical challenges. For Marx himself, the project of social theory capable of analysing specific whole capitalist societies nestled within his ambitious project of comprehending and shaping epochal change. Specifically, Marx sought to provide the means to account for variation, history, and agency because he understood that capitalism's tendencies to crisis would be of historical moment only if they could connect to the self‐conscious behaviour of subjects in specific situations. Thus, social theory about capitalism was meaningful to Marx, and to committed Marxists, because it promised to contribute to the transition from capitalism to socialism.

There have been times in the twentieth century when the structural crises of capitalism seemed to provide Western working classes with opportunities to ditch the system. Nowhere were these possibilities realized. Instead, each such instance has proved to be one of capitalist restructuring marked by the discovery of new paths of accumulation, and by the creation of new social knowledge and public policy capable of reinvigorating the economic system. In this process of capitalist renewal, Western working classes have been reconciled to capitalism and have given allegiance to it. With the recurrent failure of the proletariat to perform its imputed historical task, capitalism has come to be punctuated by crises, not by Crisis.

Classical Marxism did not anticipate this disjunction between structural opportunities and missing agency. Western Marxism, by contrast, has attempted for the past 100 years to make sense of its recurring political defeats. Hankering after the missing proletariat, Marxists have been concerned to understand its painful absence.

In this respect the writings of Marx are of little assistance. Whether because he ran out of time as the fragmentary section at the end of the third volume of Capital suggests, or because he took the various issues of class formation for granted, (p.261) Marx's theoretical work makes no allowance for the existence of a class structure without class struggle in the full sense. Even in his brilliant probes of 1848 and 1871 in France, which are certainly not stories of the triumphant overcoming of capitalism, Marx treats these insurgencies as dress rehearsals for more fundamental changes to come. Like the rest of his corpus, these writings do not anticipate the possibility that once modern working classes accommodated to the existence of capitalism they would forgo efforts to overthrow the mode of production.

Because this issue has proved so fundamental to the politics of the late nineteenth and twentieth centuries, western Marxists have sought to fill in Marx's silences. They have done so either by providing increasingly complex structural elaborations of the two‐class structure that underpins Marx's logical account of capitalist accumulation, as in the scholarship of Erik Olin Wright, or by writing richly textured histories of national working classes in the manner first produced by Edward Thompson.

These important bodies of work can only partially illuminate the core Marxist puzzles of class because they do not fully break with the ‘class in and for itself’ problematic. However softened, it is just this teleological trajectory that is open to question. Assuming good mappings of class structures and good historical understandings of the dispositions and actions of working‐class people, we are still stuck with the fact, using Thompson's language, that the tie between the social being of working classes and their social consciousness has proved not only more contingent than expected with regard to the very existence of class as a global category but capable of supporting a robust consciousness of class without a concomitant ‘revolutionary consciousness’ (even where, as in England, the working class was formed on the basis of holistic understandings of class). Further, the very crises and restructurings of capitalism in the absence of working‐class efforts to create socialism have called into question the durability of class as a basis of naming and representing the core groups of capitalist society. Thus, the absence of coherent and persuasive accounts of the problematic linkages between ‘social being’ and ‘social consciousness’ is not only the weakest (p.262) feature of the treatment of structure and agency in Marxist social theory but is itself a source of political debilitation.

Certainly the history of the West in the past century and a half has been marked by deep conflicts based on class on both sides of the work–home divide. Class has also provided the structural and discursive basis for the organization of working people into political parties and trade unions. Yet, ever since Engels ventured into Manchester in the 1840s to discover the first working class with the potential to become a proletariat, not a single Western working class has fulfilled this role. Nowhere today is capitalism in danger of being supplanted by socialism. There have been such moments, the most recent of which spanned the beginning of the Great Depression to the years following the Second World War, but they passed. Today, class struggle in the full classical usage of the term does not exist, except, perhaps, in working‐class assaults on what little remains of the ‘socialism’ of Marxist‐Leninist‐Stalinist systems.

Equally painful to Western Marxism has been the growing lack of confidence about the capacity of traditional class categories themselves to describe social reality in settings where for some time class did provide the common sense of everyday social and political life. This distress manifests itself in laments for ‘the forward march of labour halted’; in post‐Marxist discourse theory, which makes class relations a contingent and provisional semiotic; and in a shift in considerations of agency from working classes to new social movements.

The twin vexations of the absence of a proletariat willing to act for socialism and challenges to the ability of class categories to chart capitalist politics and society have produced a profound crisis for Marxist scholars. Of course, Marxists have understood the need to come to terms with these analytical challenges, and they have tried to do so: by seeking impediments to class consciousness in gender and patriarchy, labour aristocracies, ethnicity and race, repression and social control; in the bribery that comes with prosperity; in the cultural hegemony of capitalism, and in other such causes. Each of these has been a genuine and important feature of the histories of working classes, but what continues to be defective is the (p.263) undergirding logic which insists that the ‘natural’ process of class formation has been impeded by key mediating factors that have deflected it. A second strategy has focused on the class structure itself, especially the rise of a middle class. This has been the pivotal problem of the most interesting treatments of this kind, and in this way has been a mirror image of the criticisms mounted against Marxism for its obsolescence by non‐Marxist sociology and social theory.3 But if the complications of class structure are now organic to capitalism, can a class struggle based on the idea of struggle between the two basic classes of bourgeoisie and proletariat be countenanced? A third response has been to explain the debilitations of working‐class history in particular times and places as histories of defeat by the dominant classes or as betrayals by the leadership stratum of the working class, but usually without a comparative or theoretically‐constructed frame of reference, and thus without an answer to the puzzle of why it is that this pattern recurs across all the cases. A fourth strategy has been to argue that the organizational, ideological, and programmatic features of the dominant social democratic and reformist politics that has engaged the working classes as a political project within Western Europe and North America contain their own powerful logic of compromise and social integration. Much in this account is persuasive, but much with respect to the problem of proletarian reticence at moments of capitalist crisis is also tautological.

The continuing relevance Marxism has as a guide to late twentieth‐century capitalism rests on its capacity to address its disappointments about the working class more persuasively than it has to date. In Chapter 6 I tried to show how a respatialized and permeable Marxism could discern reasons for partial as opposed to holistic patterns of early working‐class formation. In this chapter, I tackle the problem for Marxism of the historical absence of class struggle in the full Marxist (p.264) sense, and I examine the contingent relationship linking the recession of working‐class identities as ways of organizing, understanding, discussing, and acting on contemporary capitalism and recent changes in the urban built‐environment.

II

The drawing of lines in space between the social classes within the new social geography of the capitalist city in the nineteenth century helped assure the residential propinquity of members of the working class as well as their isolation from other classes. From the later years of the century to the First World War, the patterning of space in these cities crystallized in an unambiguously modern form characterized by a more stark division of work and home (whose distances grew with the accessibility of public transportation to the working classes),4 segregation of city space by class, and a growing (p.265) differentiation of housing as a result of the operations of autonomous housing markets.5 This solidification of patterns which had only achieved a partially modern format in earlier decades had a contradictory relationship to working‐class formation. On the one hand, it corroborated the different specific mappings of the city crafted earlier by working classes making sense of their new social and spatial realities. Both the partial and the holistic understandings of class were reaffirmed in this new milieu by the ways existing patterns of talk and organization had already made sense of the new divisions in space. Materiality and meaning proved mutually confirming.6 On the other hand, with the elaboration of new networks made possible by the nationalization of labour markets, the (p.266) growing sense that working classes shared a community of fate that transcended given localities, and significant advances in communications and transportation, the ties between class and space grew more complicated and tentative.

It is by coming to grips with this double process that I think we can best readdress the haunting puzzle of how to account for the fact that nowhere—in spite of the elaboration of strong union and political party organizations that challenged both the terms of life in capitalism and rhetorically announced for socialism—have working classes, whether characterized by a divided or holistic consciousness, become politicized in the revolutionary manner classical Marxism took for granted. Once the fundamental struggles about capitalist control had been settled to the detriment of artisans experiencing a devaluation of their labour, social conflict ceased to be about the most fundamental question in Marxist social and political theory, the transformation of the mode of production. From a Marxist perspective, the battles between the social classes became secondary: no longer about the existence of capitalism as such, but about conditions of life within the capitalist epoch.

In a cautionary essay, Krishan Kumar argues persuasively that this break in working‐class history must call into question facile evolutionary accounts that assume a high degree of continuity between the working classes of artisans at the moment of the birth of class and the working classes of unskilled and semi‐skilled factory labourers at the end of the nineteenth century. The most radical battles were fought by the first; a reformist pattern of integration characterized the second: ‘a politicised working class, comparable to that which existed before 1850, has never again been seen in the kingdom.’7

But why should the shift from an artisanal to a proletarian (p.267) character necessarily defang the working class? Kumar argues that the new proletariat could not get very far without forging alliances across class lines, but his own evidence that artisans also fought their struggles with important allies undercuts this point.

A start toward a better explanation is provided by Craig Calhoun's account of ‘Class, Place, and the Industrial Revolution’. The premiss of this paper is a critique of the sharp divide in Marx's work between his treatment of class struggle in his logical account of capital accumulation and in his delineation of social theory. The former sees such struggle as inherent in the abstract relations of capitalism at work; the latter assumes a translation of this imperative into specific times and places. Missing in each, Calhoun avers, is both a spatially specific treatment of the city and an organizational analysis that insists that the mechanisms and scale necessary for the working class to confront capitalism must be more than a series of local systems. Without such a capacity, he insists, class struggle in the full Marxist sense is not possible.

Thus, the centrepiece of Calhoun's essay is the joining of an organizational to a spatial analysis of class formation. Unlike treatments of working‐class formation that leave place considerations out, Calhoun underscores the significance of local community‐based relationships for the development of nationally distinctive features of classness during the first industrial revolution. The direct, face‐to‐face, social relationships within segregated working‐class residential areas facilitated trust and solidarity that in some circumstances came to be based on class understandings. Yet, he argues, even where a global classness on the English model defined working‐class readings of the city and capitalism, the localism of class understandings hindered the capacities of workers to engage with capital effectively at just the moment when workers were most willing to challenge the fundamental premiss of the capitalist mode of production. ‘To be salient in the class struggle engendered by capitalism,’ Calhoun insists, ‘classes—bourgeoisie, proletariat—must be organized at the same level as capital accumulation.’ At this early moment, such translocal organization was achieved more readily by members of the bourgeoisie, who were smaller in number and possessed more (p.268) resources, than by the dispersed urban working classes who, enclosed within their factories and residential communities, were inhibited from confronting capitalism as a system that transcended the experiences of this or that local area.8 While the local spatial environment provided workers with affective bases of support, unlike the workplace the residential community did not include ‘the enemy to be confronted nor was it composed of a sufficiently broad network of relationships to reach all those concerned’.9 Hence a first historical paradox: just when beleaguered artisan communities were most willing to challenge the prerogatives of capital they lacked the transportation and communications infrastructure to do so in a manner that confronted the indirect, translocal features of capitalism as a mode of production; ‘in the early nineteenth‐century, class struggle, at least the struggle of bourgeoisie against proletariat, which Marx proposed, was impossible.’10

From the perspective of the organizational problems of co‐ordination and communications, class struggle in the full Marxist sense did become possible only in the late nineteenth century. Just as classical Marxism paid little attention to actual working‐class people outside their places of work, so, too, Marx and Engels had little to say about the organizational dimensions of class outside specific workplaces. ‘One result of this is that as classes are deduced from the economic theory, their collective action is presumed to follow simply from rational recognition of common interests.’ Calhoun stresses, instead, how important it is to see that ‘classes become important social bases for collective action when society is knit together through large‐scale systems of indirect relationships’.11 In the late nineteenth century, trade unions with a (p.269) national scope and political parties contesting for power at the centre became just such organizational vehicles. As such, these institutions cannot be understood simply as the products of class formation but as mechanisms for the making and remarking of class (and, of course, of other social categories).12 Direct face‐to‐face relationships, whether in factories or in residential communities, Calhoun stresses, cannot by themselves ‘give the class collective agency’. He insists that

For these collectivities to provide the basis for sustained, effective insurgencies their members must be linked to each other through some mediating agency. Trade unions work in this way for their members, and are thus in direct line of development of class struggle (as Marx thought) and not necessarily to be distinguished from a more revolutionary class consciousness (as Lenin suggested). Trade unions and working‐class political parties do vary in the extent to which they represent loosely organized constituents, or organize those constituents for direct participation in action (the latter comes much closer to Marx's conceptualization of class struggle). In either case, this sort of mobilization differs significantly from that which is based on direct relations such as those of the local community. Moreover, (p.270) it depends on a level of communications and transportation infrastructure which had not been developed prior to 1840.13

With the radical diminution in the costs of transaction between dispersed working‐class communities by the elaboration of national railway and telegraph networks, the institutions of national trade unions and political parties became possible. But with this development there emerged a second historical paradox: at just the moment when class struggle in the full Marxist sense achieved its necessary infrastrucure, it came to be shackled by the organizational imperatives of the new working‐class institutions of union and party. As formal organizations enmeshed in the possibilities for gains made feasible by the permissive rules of democratic political regimes, unions and parties pursued class interests in ways that were inherently short term and limited, whatever their rhetorical commitments to more fundamental social change. If ‘these organizations are necessary to the co‐ordination of action at the same level at which capital and political power are centralized’, they are ‘distinct from the classes they represent’. Parties and unions engage these classes only in partial and role‐specific ways. They have their own reproduction imperatives. Apart from the activists who staff and run them, they ‘appear as non‐essential consumer goods rather than an essential part of life’. Participation in such organizations lacks the vibrant intensity of social movements based on direct face‐to‐face relationships. Indirect class struggle is thus ‘a part of capitalism—or at least capitalist democracy’.14

I find this analysis suggestive but incomplete. While Calhoun, like Kumar, focuses on the disjuncture between the working classes of the early and late nineteenth century, and thus on the distinction between localized and direct versus national and indirect patterns of collective action, he considerably understates the integral connections between the local and national, the direct and indirect. Nor does he closely examine the ties between early nineteenth‐century mappings of the city, with their resulting alternatives of partial or global classness, and later ‘modern’ patterns of class organization. (p.271) Missing, too, is any comparative sense of the differing potentials of the new forms of organization depending on their linkage to older patterns of reading the city.15

Once again, consider Chartism. Calhoun treats this massive class‐based movement as a transitional one, because of the way it joined the residential communities of workers, the union organizations that were embedded within them, and an emergent national infrastrucure made possible by improved road transport. By transitional, he means a pattern coming to an end, not one heralding the future. For, in his view, the national‐organizational level was soon to overwhelm the local.

This is not persuasive. There was no simple transition from the local urban setting of class politics to national organizations and networks. If capitalism, as Calhoun stresses, was translocal and abstract, and thus needed to be confronted on comparable terms by working‐class contestants, it also remained deeply and inherently experiential and local; it is impossible to conceive of cross‐class confrontations outside the lived worlds of workers and their families.

Not surprisingly, the city level has remained a constitutive aspect of modern working‐class politics. It was certainly integral to the new unions and political parties Calhoun wrongly counterposes to locality‐based relations. While their national scope was a genuine innovation, it is important not to lose sight of the fact that their very existence as organizations depended on their federated character. Unions are based in specific workplaces, and parties in constituencies, and these are community‐specific. Unions and parties interpret the daily lives of working people where they work and live. Indeed, if union and party organizations had not become integral to these urban experiences they would have had no capacity to recruit and mobilize followers. Just as Calhoun is right to (p.272) reject the notion of class as an abstract rather than an organizational matter, so it is necessary to insist that the organizational aspects of class have been grounded in the materiality of the city, thus in the ways cities have been mapped and remapped by working classes under the spatial and organizational conditions that confront them.

But, given earlier nineteenth‐century mappings of the city, workers could not respond to these challenges just as they pleased. The available repertoire of late nineteenth‐ and early twentieth‐century working‐class collective action in different countries was informed and limited by the patterns of working‐class formation that had developed earlier in the nineteenth century. The scope of union aspirations; their ties to political parties; the very existence of various labour, social democratic, and socialist political parties; their representations of capitalism, and their capacities to mobilize followers all depended in considerable measure on the outcomes of the first period of working‐class formation. The choices available to English working‐class organizations in the later years of the century were different from those open to unions and parties in the United States. There the earlier split consciousness of the working class, reproduced in the ordinary politics of the early and middle years of the century and by the intensification of the divisions in space between work and home in urban life, directed unions to the path of a restricted, if militant, workplace labourism that eschewed party politics; at the same time it directed parties to the path of mobilizing workers on the basis of nonclass identities. In this context, the foundation of a political party like the Labour Party, if not excluded a priori, was made very difficult. American efforts along these lines proved remarkably unsuccessful in spite of the intensification of capitalist exploitation and the existence of advanced forms of the infrastructural conditions stressed by Calhoun. The content of the organizational revolution, in short, was in considerable measure determined by the terms in which the early separations of work and community had been talked about and fought about.

The existing differences between the United States and England became even more pronounced as a result of the demographic transformation of the American working class by mass immigration from southern and eastern Europe after the (p.273) 1870s. Entering an urban world in which the place of residence was constructed in linguistic and organizational terms as separate from the place of work, the newcomers rapidly gave a more pronounced ethnic twist to city neighbourhoods while they joined their lot to transethnic labour—management struggles in the factories.

Two important studies of Detroit at the turn of the century, with their different emphases, capture this reassertion of older patterns of class formation in an age of new working‐class organizational capacity, and show quite clearly the character of the new urban situation and the challenges they posed for working‐class re‐readings of the city. The first, Olivier Zunz's The Changing Face of Inequality, by focusing on neighbourhoods, homes, and housing markets, argues that the evergrowing separation between the factories and the residences of the working class made it possible for ethnic neighbourhoods to become ‘the spatial anchor of the communities, the primary colors of the urban quilt’; immigrant areas thus became relatively autonomous environments. The working class, fragmented by ethnicity and by place, resisted the homogenizing forces of the second industrial revolution. ‘Because ethnic bonds remained so strong during the industrialization process, the American social structure remained free from the threat of organized class conflicts despite extreme inequality of conditions.’16 The immigrant neighbourhoods were not just the product of the desire of the newcomers for propinquity, but of migration chains that pulled kin and acquaintances together, as well as common occupations and income levels. Moreover, as Kathleen Conzen observes, once an area secured an ethnic character it became less attractive to non‐group members with equal or higher status. ‘Thus the locations of jobs and cheap housing, combined with the degree of heterogeneity of the group, strongly influenced the extent of residential concentration.’17 The extent of this concentration was determined by (p.274) the character of journeys to work. The more independent the location of the workplace from that of the residential neighbourhoods, the more likely it was that cross‐ethnic workplaces could draw their work‐forces from different, ethnically distinctive areas.18

The second study, Richard Oestreicher's Solidarity and Fragmentation, examines the same immigrant working class only to find a robust pattern of working‐class consciousness and collective action, especially in the 1880s, in spite of the important differences of ethnicity and neighbourhood stressed by Zunz. Working‐class solidarity competed with these fragmenting features of the city's life, and, from time to time, proved the dominant shaper of working‐class dispositions. ‘In a contradictory social environment, there was always an inner tension in which the extent of class consciousness was a function of concrete opportunities for its expression. Where such vehicles, e.g., the Knights of Labor, thrived, class consciousness increased; when they did not, it declined.’19

These very different accounts of working‐class Detroit tell two aspects of the same tale. Whereas Zunz draws our attention to the ways in which a pre‐existing pattern of class formation provided a crucible for the forging of a keen localized ethnic identity and sociality, Oestreicher shows how the same individuals engaged the labour side of the divide between work and home. If ethnicity, as Zunz rightly insists, mitigated class consciousness in the demanding Marxist sense, and provided the main basis for participation in political party life, it did not prove incompatible with a quite enthusiastic participation in the American labour movement. Just as in England in this period, labour organizations utilized heretofore unavailable transport and communications technologies to build unions and federations of unions that joined together widely dispersed union locals. And, just as in England, there was a good deal of tension within the labour movement about the desirability of building on the base of workplace affiliations to enter the (p.275) domain of party politics as labour. However, whereas in England this problem was resolved in keeping with older patterns and experiences in favour of the founding of a new kind of political party grounded in the union movement, in the United States, after the rapid collapse of the Knights of Labor in the 1880s, the bread‐and‐butter unionism of the American Federation of Labor (AFL), which refused to commit the labour movement to political engagement, emerged as the hegemonic form of trade unionism by the end of the century. Because the AFL was convinced of the futility of socialist and other third‐party political activism, it sought to protect its sphere of action by taking on only those conflicts it thought it could win.20 In this way, not only were the older split readings of the city reinforced, but so too were the most integrative and least militant features of working‐class life at work and at home. The labour movement focused on the most skilled and on battles for better wages and terms of work; in the neighbourhoods, the differentiation of workers by ability to rent and purchase housing was interpenetrated by ethnic definitions of space and party politics. The result, Oestreicher argues, was a mixed picture of high levels of intra‐class ethnic antagonisms, which coexisted with impressive cross‐ethnic solidarities:

the workers who fought police on behalf of streetcar workers, the stove workers who refused to patronize a saloon selling scab bread, the Italian stonecutters who honored the stonecutters' picket lines, the Polish street laborers who called for equal division of available street work, the working‐class theater goers who refused to patronize boycotted theaters in 1899 all demonstrated that the concept of solidarity had been effectively communicated to workers of many trades and nationalities.21

It is noteworthy that each of these instances of class‐conscious collective action cited by Oestreicher was work‐related. Continuous with the mid‐nineteenth‐century readings of the city, (p.276) the American working class continued to be formed as a class, but class understood only as labour.

In spite of enduring differences in patterns of class formation, there were, nevertheless, important shifts in the character of urban relations that were shared by the English and American working classes at the turn of the century which challenged them both to re‐read the city, and to do so in ways that reduced significantly the prospects of militancy in the Marxist sense. These changes concerned the structuring of city space and a new role for the state on the urban scene.

Capitalist cities continued to grow dramatically and to become increasingly intertwined with each other as points of accumulation and exchange. As they grew, the ambiguities of form that have recently produced debates between social geographers about the extent of the modernity of the early industrial city gave way to an intensification of the various spatial separations that had earlier interlaced the cross‐class and mixed land‐use patterns inherited from household‐based production arrangements. This clarification had a contradictory aspect in both countries, carrying with it powerful impulses that on one side sharpened the immediacy and intensity of class and that on the other significantly diminished the prospects of militancy.

One of the deepening features of city space concerned the spatial divisions between working‐class and non‐working‐class residential areas. With this working out of by now longstanding trends came the increasing autonomy of working‐class neighbourhoods and institutions and thus ‘the mutual insulation and repulsion of the classes’,22 their organizations, and their ways of life. The divisions between ‘us’ and ‘them’ that had also grown more prominent with the creation of largescale factories and their hierarchical organization of production now became more visible away from work as well.

Concurrently, the operation of housing markets, over time, came to divide the working class into distinguishable subgroups that occupied quite different parts of the city, characterized by diverse standards of housing and amenities. (p.277) This intra‐class stratification established new bases of difference and meaning within the working class.

In the United States, this pattern of working‐class spatial differentiation was complicated by the much more varied ethnic heterogeneity, but both countries—indeed all the industrialized Western capitalist countries—shared in this reorganization of city space. The English case is a reminder that ethnic divisions in American cities reinforced this common spatially grounded differentiation between working‐class groups by their ability to purchase residential environments. Indeed, as Zunz carefully demonstrates, the steady working out of this market and status‐driven process restructured the character of American ethnic neighbourhoods to produce something of a convergence with English arrangements. Whereas in Detroit in the 1880s ethnicity as a principle of spatial segmentation was sufficiently strong to bring together in ethnic enclaves members of ethnic groups who had quite different market attributes, by the 1920s, except for blacks who remained pressed together by racial segregation, the socio‐ethnic geography of the city had altered. Members of ethnic groups now sought housing appropriate to their incomes and styles of life. As a result, neighbourhoods took on more of a cross‐ethnic aspect even as they became more homogeneous in market terms.23 Poles, Irish, Italians, Jews, and other white ethnic groups increasingly came to live side by side in areas stratified by the housing market. With these changes in working‐class communities, the link between residential patterns and ethnic identity grew less direct. Initially, Conzen notes, the emergence of ethnicity as a central category of meaning and action in the United States depended on common residence. This concentration was necessary at this first stage to support the development of locality‐based ethnic organizations. But with shifts generated by the continuing operation of autonomous housing markets, ethnicity became despatialized as it steadily lost its place‐specific underpinnings. (p.278) With this change, ethnicity began to be transmuted into something more cultural and symbolic; which, in turn, later in the twentieth century, has crystallized into the formal solidarities of national interest groups.24

These changes on the ground in cities occurred at exactly the same time as the nationalization of the infrastructures of class emphasized by Calhoun. The organizations of union and party were able to tap into a spatially rooted, thus magnified, consciousness of class that helped them overcome barriers to solidarity and collective action. Yet, the new spatial divisions within the working class helped limit their political horizons by deflecting them into new areas of concern. As a result, at the very moment when the capacities of the working class were enhanced by their new translocal institutions, though limited in goals by their organizational imperatives, working‐class resources were also tempered and channelled by intraclass spatial divisons.

As new housing markets functioned, geographer David Ward observes, ‘the increased internal differentiation of the residential patterns of the city’ divided the working class

into smaller sub‐cultures each with its own small‐scale versions of success . . . Expectations were formulated in terms of minor gains in income and status rather than as a reorganisation of society as a whole . . . As older bases of status disappeared and as income discrepancies between the skilled and unskilled diminished, the more prosperous and secure manual workers sought a more emphatic definition of their place in society by moving to residential quarters housing people with similar occupations and incomes. . .social differences were increasingly calibrated on a spatial order rather than one of external appearance . . . For the upwardly mobile segments of the working class, a new address with generally recognised social associations, provided a new but emphatic basis of their rising places in society.25

There is something of a resemblance between this perspective and the most familiar account by English Marxists, that of the ‘labour aristocracy’, for the failure of the working class to be more militant. Work in this tradition has examined (p.279) wages and conditions of work (including the divide between craft and factory workers), and the distinctions in world‐views between leaders and followers in labour unions and political parties, as well as urban neighbourhood differentiation—the focus I find most congenial—to find a fault‐line dividing the working class. The analytical pay‐off is an understanding of how the ‘aristocracy’ sold out a working class otherwise prepared to do battle with capital. As Henry Pelling has observed, ‘It is an essential feature of the Marxist theory of the labour aristocracy that this supposedly small section of the working class was conservative in politics and imposed its conservatism upon working class institutions, thereby concealing, but by no means eliminating the underlying militancy of the mass of workers.’26

The labour aristocracy approach has elusive qualities. As a number of critics have noted, micro‐level treatments of relations of craft and manual workers in specific industrial settings and of workers in residential areas indicate just how inexact the term ‘labour aristocracy’ is and how difficult it is to specify locations where aristocrats of labour actually confronted and contained a more radical mass of workers. Contrary to expectations of the theory, empirical evidence, especially for the late nineteenth century, suggests that ‘it was the more prosperous workers who were the more politically militant and radical, while the lower ranks displayed either apathy or conservatism’.27 Correct though this criticism by Pelling may be, it (p.280) misses the point that it is the undergirding theoretical apparatus of the ‘labour aristocracy’ position that is its deepest flaw. While right on the mark in looking at both organizational and spatial determinants of a lack of militancy, the refusal of this approach to break with the view that the working class is immanently revolutionary produces a rather idealist imputation of blame. Instead, with Calhoun and with Ward, we would do better to focus on how the organizational and spatial features of working‐class life entirely undercut the possibilities for classically Marxist class consciousness and militancy. The organizational imperatives enmeshed leaders and followers, and the new spatial design of cities established new issues, concerns, and family mobility strategies that divided workers from each other in a finely graded way, not a small privileged group from a seething mass that it managed to restrain and sell out.

This portrait of the working class is an ideological construction, rather different, if in some ways drawing on, the complex interplay between inherited systems of meaning and collective action from an earlier moment of class formation, the new translocal features of working‐class unions and party organizations, and the deepening segregation of city space in conjunction with the differentiation of neighbourhoods by subclasses with differential access to housing. It was this combination that produced an enhancement of the capacity of working classes to confront capital at work and away from work, to join their senses of grievance in both domains, and to mobilize resources of organization and solidarity in both spheres in pursuit of their aims; but it also produced a situation that limited the horizons and aspirations of the working classes as wage workers, as citizens, and as members of a class‐based civil society. In this way, the indirect and face‐to‐face dimensions of class worked in tandem to minimize the prospects that the working class would act as a proletariat that confronted capitalism as a whole, with the goal of supplanting it with a new, socialist mode of production.

If important shifts in social geography provoked new readings of class and capitalism, so, too, did the fact that the shaping of city space was no longer the privilege of capital working with a high degree of independence. The manifestly (p.281) more important role of the state in shaping urban environments impelled members of the working class to address issues of work and home not only as workers confronting capitalists in labour and housing markets, but as citizens confronting the state and emergent political issues about land, the urban environment, and municipal services.

The new capitalism of the second industrial revolution no longer accorded to capital the autonomy that had characterized the first. Laissez‐faire as prescribed by the political economists proved insupportable, for the sphere of democratic politics engendered movements to tame the cycle of boom and bust and to mitigate the harsh operations of the market‐place. Further, as capitalism grew in scale and reach at both the domestic and international levels, the state moved to organize the very terms of the functioning of markets themselves—through regulation, tariff and tax policy, and imperial assertion. As integral aspects of this changing tapestry of the relations of state and capitalism, governments in cities moved to shape and reshape space within the field of tension established by the conflicting claims of accumulation and citizenship. City growth became less and less a matter of the direct imposition of the logic of capital and more and more the product of planned state interventions. Road‐building, slum clearance, the provision of sanitation and water infrastructures, and the like became politicized, and objects of attention for all the various social classes.28 The dominant classes understood the explosive potentialities of a concentrated (p.282) and disaffected urban working class, and they were reminded forcefully of that potential not only by such dramatic events as the Paris Commune but by recurring urban protests, demonstrations, and riots rooted in the networks and solidarities of residential communities. In these circumstances the class‐segregated city became a quite terrifying object for the middle and upper classes, who were uncertain about what the ‘dangerous classes’ would do next. As a product of a ‘mixture of fear, bad conscience and goodwill . . . the conditions of the working classes, especially in big cities, [became] a major subject of concern in the mid‐Victorian years’.29

Though the relationship between these urban issues and other union and party concerns varied from time to time and from place to place, nowhere could the new organizations of the working class successfully mobilize workers without attention to these questions; they became integral to the new class understandings. More generally, the connections between the more intimately local and the more experience‐distant national‐organizational levels defined the terms and content of working‐class political life. The city did not cease (p.283) to be significant, as Calhoun implies, only different in its importance.

The failure of Western working classes to act as revolutionary proletariats during the most profound crisis of modern capitalism in the Great Depression was not the result of mediating factors that interposed themselves between a class‐conscious working class and the collective expressions of their militancy. Rather, the very organizations workers required to confront capital limited their politics, and the very spatial patterns that divided workers from the dominant classes fragmented their life experiences and aspirations. In the interconnections between working‐class organizations and local relations, the barriers to a revolutionary Marxist project became insurmountable. Workers were thus capable of an assertive class perspective during the 1930s and 1940s, but seen through the lenses of Marxist expectation this politics of class was a politics lacking in class struggle.30

III

The relationship of Marxism and the city stands today on unsure ground. Now it is the politics of class itself that is in question. The class identities and actors so central to Marxism have receded as the basis of individual dispositions and strategic group activity in the cities that have been the most important loci for class formation under capitalism. Cities themselves, moreover, no longer seem very well defined. Clear boundaries between city and country are hard to find in the face of a formless, sprawling urban pattern. It is reasonable to think these developments are connected, but how?

In discussing this question, I draw on and assess a lively discussion within the new urban Marxism on the persistence (p.284) and character of the city and on arguments about the troubled status of class and class struggle in today's capitalism. Given the vast expanse of these subjects, it is not my intention to offer more than a sketch of an approach that joins these issues together, overcoming the old, unwelcome division, and that shows that some of the more persuasive answers can still be found in an engagement of Marxism with the city.

I proceed by examining two of the most significant attempts to show how changes to cities and urban space affect contemporary patterns of class formation, those of Mark Gottdiener and Eric Hobsbawm. They show how recent complex transformations of urban social geography go a long way towards helping us account for why it is that the class categories of traditional Marxism are losing their grip at the centre of the capitalist world; but they do so, I argue, in ways that are not entirely satisfactory.

Their work is grounded in ideas about changes to cities that they share with a large number of urban‐oriented Marxists who have been attempting to make sense of the restructuring of capitalism in the closing decades of this century. If the issues of domestic capital accumulation and local social movements filled the pages of the initial renewal of interest by Marxists in the city in the 1960s and 1970s, today the overriding topic in the relevant journals and books is the reorganization of capitalism, and with it the transformation, some think the elimination, of the city as a distinctive kind of space. And if Engels was the key initiator of Marxist work on cities, as well as its missed chances, Lefebvre has re‐emerged as the progenitor of appropriate work on the new capitalism because of his insistence that we shift attention from the city as a determined and bounded place to a more fluid conception of space.31

If there is no single version of capitalist, and urban, restructuring in these largely (but not, we shall see, entirely) persuasive discussions, there are common threads.32 Capitalism has (p.285) become more globalized and interdependent. The internationalization of production and finance have gone hand in hand. Large‐scale ‘Fordist’ manufacturing has declined while services, design‐intensive industries, and high‐technology production have become increasingly significant both in structuring the economy and in shaping its social relations. In this dual process of deindustrialization and reindustrialization capital and labour have become more specialized and mobile. The circuits of capital move more swiftly and are less anchored in specific places. A profound technological revolution in microelectronics has transformed communications, and has made information, not goods, the key commodity of advanced capitalism. Production is more oriented to process than to product. The disarticulation of work from place has grown so stark that an American insurance company serving customers in the United States can process their claims in a remote Irish village.33 Decentralization and flexibility in labour as well as capital markets are new hallmarks of the economy. Governments underwrite these developments by subsidies of defence budgets, corporate research and development, transportation (p.286) infrastructures, and competitive bidding for the location of new industries and plants. In the most ambitious interpretations, these trends are thought to constitute an entirely new stage in the development of capitalism.34

The new capitalism is transforming the organization of territory and the spatial division of labour. The great metropolises—London, New York, Tokyo, Paris, Berlin—are in the midst of a shift from centres of manufacturing to centres of knowledge. Outside these giant cities, large manufacturing centres are declining in scale and significance. Space is being reshaped. New configurations of work and home are literally plonked in the middle of nowhere, located at key points of communication and transport created by high‐speed motorways and airports, as the real‐estate industry converts the rural landscape to office, production, and living uses. Areas come to be known by such signifiers as Route 128 in Massachusetts or the M4 corridor west of London. Places with existing built‐forms are put at a competitive disadvantage, as they are burdened by the fixed arrangements of older forms of manufacturing (p.287) and by the costs of unionized labour. With the exceptions of those at the nodes of international finance, services, and trade, cities wither from disinvestment and abandonment. Capitalism now grows either in ‘enclaves within older manufacturing regions, or . . . [in] a series of areas that hitherto largely coincided with the extensive geographical margins of capitalist industrialization’.35

Grounded in Lefebvre's conceptions of space, the most theoretically ambitious and self‐consciously Marxist elaboration of these themes is Mark Gottdiener's The Social Production of Urban Space. Much as Engels in the 1840s sought out the most advanced spatial example of the new industrial capitalism in Manchester, so Gottdiener premisses his book on today's equivalent of the ‘shock city’: the dispersed spatial arrangements of the United States in the 1970s and 1980s. Gottdiener's premiss is that the traditional city, whose hallmark has been the concentration of capital, production, people, and power, has ceased to be a generative force:

Urban life has become portable and, thus, so has the ‘city’. In place of the compact city form which once represented a historical process years in the making, we now have a metropolitan population distributed and organized in ever expanding regional areas that are amorphous in form, massive in scope, and hierarchical in their scale of social organization.

This new spatial morphology represents not a continuation of older patterns, but a radical break, with profound significance (p.288) for all aspects of economics, politics, culture, and society, not least for the character and meaning of class and class struggle. Above all, the new polynucleated growth is characterized by deconcentration, ‘the massive regional dispersal of people, commerce, industry, and administration along with the contemporary restructuring of such regions into multicentered realms—sprawling for miles and miles and located everywhere in the country, especially in those areas once thought immune from urban development’.36

Gottdiener upbraids both mainstream ecological and recent Marxist approaches to urban development for not treating this new spatial configuration head‐on and for not recognizing that the city can no longer be considered an object of analysis, but only a part of the larger process of spatial reorganization.37 Urging a return to the production of space perspective of Lefebvre, Gottdiener seeks to develop theory appropriate to capitalism's new spatial situation. He asserts that recent spatial changes must be understood as informed by the structural nexus of the emergence of the global corporation, the interventionist state, and the acceleration of technical innovation. These developments have a distinctly capitalist twist in ‘the progressively more footloose nature of industry’, the more fluid flow to the circuits of capital, the fractionization of capital and labour, and the uneven development that ‘represents both a process of capital accumulation and a competitive relation between different fractions of capital’. Within ‘late capitalism’ the key agency undergirding these processes is provided by interests and conflicts that revolve around the distribution of land and property. This sector is now ‘the (p.289) leading edge of capitalist restructuring in space’. No longer entirely or even principally private, the property sector often combines ‘public–private coalitions which also include elements of organized labor and which support themselves through local bureaucracies deeply dependent on growth’. These growth networks—their interests, activities, politics—produce the new spatial relations by connecting the characteristics of the political economy and the volitions of actors capable of changing the organization of space. In the United States, this shift has entailed the deconcentration of housing and workplaces in existing metropolitan areas and the shift of people and production to the sunbelt of the South and Southwest. Gottdiener concludes that ‘the production of space has occurred in the main not because of economic processes alone but, more specifically, because of a joint state–real estate sector articulation which forms the leading edge of spatial transformations’.38

Gottdiener connects these dramatic changes to issues of class formation. Under these new conditions, he argues, what is at issue is not how working classes map the city, but the absence of a city to map and the absence of a working class to map it. In his view, the new decentralized organization of space has transformed social relations on the ground to the profound detriment of class solidarities or, indeed, any single, coherent vision of the social structure.

He sees a great paradox at work. The acceleration of communication and the dissemination of information and the dramatic extension of transportation routes, especially roads (car ownership has penetrated quite deeply into the various Western working classes), have more radically severed the relationship between work and home than at any time in the past—thus divorcing class based on labour force experiences from class based on locality. At the place of residence, these changes have produced an ever‐finer segregation of the population for whom the determination of where to live is shaped mainly by the operation of highly autonomous housing markets. The result has been a contraction of social and political horizons. ‘The spatial segregation of social groups’, Gottdiener (p.290) writes, ‘has liberated the vast majority of the population from responsibility for the less advantaged, because the former no longer live in close proximity to the latter.’ Even more important, he argues, local space, bereft of streets and public areas, has contracted primarily to isolated single‐family homes. In such circumstances, in effect, the residential community disappears. ‘Neighbors become increasingly estranged through a lack of common experiences, despite the superficial appearance of civility between them, as the personalized network of commuters replaces the localized community of the past, with its once dense social relations.’ Space becomes abstract. It loses its tangible markers. Place‐specific social relations and cultures attentuate, even disappear. ‘Local communities are transformed into privatized domains devoid of street interaction, with limited services and limited use of public space.’39 If the nineteenth‐century revolution in settlement space provided a material underpinning for working‐class fraternal, instrumental, and political organizations, the new spatial patterns, by changing the experiences of everyday life, have reduced possibilities for class dispositions and collective action at just the moment when the infrastructural barriers to class formation identified by Calhoun, and discussed earlier in the chapter, have been overcome decisively.

Though Gottdiener's avowed aim is the writing of a new Marxist urban theory appropriate to the novel conditions of abstract space and the shattering of familiar urban forms, his Lefebvrian path takes him to the point of abandoning the project of Marxism and the city. Hobsbawm, by contrast, reaches conclusions rather similar to those of Gottdiener with regard to the fate of the working class from within far more traditional Marxist ways of working. His approach reasserts the structural vitality of Marxist analysis and extends it through an analysis of city space to account for the indeterminate patterns of class and group formation that currently obtain. For Hobsbawm, the current crisis of class is grounded in basic shifts in the experienced worlds of capitalist societies that have eroded the material and spatial bases of working‐class dispositions and collective action.

(p.291) Hobsbawm first attempted to account for the crisis of class within Marxism without any reference to cities and space. His 1978 Marx Memorial Lecture, ‘The Forward March of Labour Halted?’, took the form of an elegy for the British labour movement. Except for a voluntaristic coda, the talk was something of a reluctant and deeply regretful farewell to the predictable, if limited, labourist politics of the cloth‐cap, mainly male, manufacturing working class. The Left now had to face the reality of the collapse of working‐class agency despite the increasingly dismal structural situation of capitalism. ‘My conclusion is that the development of the working class in the past generation has been such as to raise a number of very serious questions about its future and the future of its movement. What makes this all the more tragic’, he continued, ‘is that we are today in a period of world crisis for capitalism.’40

Although Hobsbawm upbraided the Labour Party for choices badly made, the weight of his argument was that the march had been halted less by developments in the realm of subjectivity and organization than in the realm of class as an objective way of life. The ‘common style’ of working‐class life that had developed in the late nineteenth century and which flourished until the middle of the twentieth century could not survive technological and sectoral transformations in the organization of work, rising standards of living, and demographic changes in the work‐force that introduced divisions based on gender and nationality. It was the challenge of these changes that the trade unions and the Labour Party had failed to meet.

Perhaps because it was written from quite traditional Marxist premisses about the privileged place of production and of the manufacturing working class, Hobsbawm's lecture elicited howls of anguish from the Left for its negative reading of the historical role of the working class and its immense scepticism about the capacities of Marxism as social and political theory (p.292) to cope with shifts in the social organization of contemporary capitalist societies. His talk, moreover, antedated the assertive ideological conservatism of the Thatcher governments at home, the widespread disarray of left‐wing political parties and trade union movements elsewhere in the West, and the collapse of avowedly Marxist regimes in East and Central Europe. The persuasive power of Marxism has eroded even further under the pressure of these developments. If, earlier in the twentieth century, Marxism had to face the situation of class without class struggle, now it has to confront the prospect of capitalism without class.

With the exception of a brief reference to the nineteenth‐century working class as a collection of separate localized communities, Hobsbawm's talk devoted no attention to cities and their social geography. Likewise, the many rejoinders to the lecture had a good deal to say about the character of the labour force, specific industries, and the sociology of work, but nothing about the experience of the working class away from work. The debate thus had something of a partial quality, as if members of the working class lived and experienced capitalism, and found ways to resist integration within its embrace, either without reference to place or exclusively in the realm of work.

It was Hobsbawm himself who shifted the debate about the decomposition of class in late twentieth‐century capitalism to the theme of ‘Labour in the Great City’ in his 1986 Herbert Gutman Memorial Lecture, virtually the only attempt by someone working outside the framework of the new Marxist urban studies, but within Marxism, to connect issues of city space to contemporary working‐class formation. In this second formulation of the halting of the working‐class march, capitalist cities numbering in the hundreds of thousands and millions—‘a giant concentration of workers’—were regarded as conditions of working‐class consciousness and movements. ‘Clearly, without the existence of class‐specific residential concentrations such organizational triumphs’ as working‐class mobilizations and massive voting for left‐wing political parties ‘would have been much more difficult, or even impossible’. It was the segregation of the working class, carried to extremes in large‐scale tenement blocks, that concentrated activists and (p.293) followers in a distinctively proletarian milieu. In these places, housing and rents joined workplaces and wages as key issues of working‐class political consciousness. ‘Actual social movements were particularly likely to flare up in the “towns within cities”, where wages, rent, community feeling and class organization coincided.’41

I turn later to what I think are shortcomings in this argument which reduce the pay‐off of this engagement with the city for Marxism. None the less, Hobsbawm's essay is important for the manner in which it treats the city and its social–spatial relations as necessary, if not sufficient, conditions for class‐based dispositions and patterns of collective action. By bringing these matters to the fore, Hobsbawm challenged Marxism—theoretically and politically—to confront the implications of late twentieth‐century spatial change for the very existence of class as a meaningful political category within advanced capitalism. In the nineteenth century it was plausible for Engels to think that the urbanization that concentrated capitalist power and production by also concentrating the working class would help organize the agents of capitalism's transformation.42 Changes in the social geography of capitalist cities at the end of the twentieth century, Hobsbawm insisted, have dashed even diluted expectations of a class conscious proletariat.

Hobsbawm argued that the separation between work and home has grown to such an extent that there is today an insufficient basis in city space to sustain an integrated working‐class politics at work and away from it. Suburbanization has decentralized cities, and they have become deindustrialized as manufacturing and service work have found new locales outside the city. The commuting patterns of the working class have been extended, snapping ‘the links between day and night, or between the places where people live and those where they work, with substantial effects on the potential of labour organisation’. Inner cities have come to be (p.294) peopled increasingly by ‘the miscellaneous poor, the unskilled, the socially marginal and problematic, the ethnic and other minorities of whatever kind’. Further, urban renewal, road‐building, and central business district redevelopment have broken up the traditional working‐class neighbourhoods. ‘The effect of all this on labour movements in the great city has been to deprive them of their former cohesion.’ Now that ‘the labouring population of the great city is returning from the status and consciousness of a “proletariat” to that of the pre‐industrial and miscellaneous “labouring poor” . . . labour movements are losing, or have lost, . . . much of that class identification of working people as such which used to give them force, and a sense of collective power.’43 The solidaristic politics of class has been replaced by the much more fragmented politics of minorities seeking coalitions.

IV

Gottdiener and Hobsbawm suggestively connect up issues that too frequently in the history of Marxism have been considered separately. Each in his own way, however, does so in a manner that insufficiently apprehends the complex layering of changes to the city now under way and that links spatial to class analysis in too unmediated and unidirectional a way.

Both are impressed by the range and intensity of urban–spatial transformations; rightly so. None the less, there is something unsettling about Hobsbawm's generalizations about the impact of alterations to the cityscape on class formation from the experience of a small number of ‘great’ cities, Gottdiener's confident reasoning about the end of the city, and the now hegemonic portrait of city change within urban Marxism, with their shared assurance about the timing of change and the way they simplify current trends to the material and spatial base of urban life.

Surely some scepticism is in order with regard to the urban literature's account of the abruptness of transition in the regime of accumulation and the character of cities. After all, (p.295) not many people share the privilege of living at a moment said to be as pivotal as ours, and even fewer have so quickly understood the profound qualities of such transformations just as they were happening. In accounts such as Gottdiener's of transition to a new stage in the history of capitalism and its organization of space, the crisis of spatially agglomerated ‘Fordist’ capitalism is said to have become manifest, with great suddenness, in the early to middle 1970s. The various elements of ‘post‐Fordism’ are presented as having come together in a sufficiently robust way to define a new regime of accumulation only since the oil shock of 1973 and the recession of 1974–5.

‘Such profound transitions to new forms of economy, society and culture’, Nigel Thrift usefully reminds us, ‘usually take longer, are more complex, and are more conservative than commentators of the time, who fix on the new, tend to realise.’ The watershed shift to industrial capitalism in nineteenth‐century England—the most dramatic such stepwise change within a compressed time and place yet to occur—was no simple matter of the new instantly supplanting the old. ‘Victorian Britian,’ Thrift observes, ‘the paradigm of the shift to industrial capitalism, has been revealed by social and economic historians as the heyday of the horse as well as steam power, of hand power as well as mechanical power, and of a services‐oriented South East England as well as an industrial North.’ If older forms of production and space are now being supplanted, they will continue to count, if in a new context and in new ways. It makes little analytical or political sense simply to shift attention to the newer forms of social being as if they did not maintain dense relationships with the older ones.44 Likewise, ways of understanding the new situation (p.296) necessarily draw on traditional languages and tools of comprehension. The formation of group and class dispositions under new circumstances can hardly happen as if the past did not exist.

The literature on recent changes to urban spatial structures stresses discontinuities, but, as Norman and Susan Fainstein argue, their most important qualities are at least as much characterized by substantial continuities with earlier moments of the development of capitalist cities. The Fainsteins do not question the post‐1973 production and spatial restructurings of Western capitalism, but they sensibly underscore both that such reorganizations in response to impasses and bottlenecks are endemic to the history of capitalism, and that the mechanisms identified by analysts of recent changes to account for industrial and spatial restructuring—‘the devalorization of fixed capital; geographical disaggregation of the integrated firm; the flight of industry to areas with lower factor costs; absentee ownership; the entry of new states into the capitalist core; new international divisions of labor; and the increased importance of service‐and‐control employment within the occupational structure’—are long‐familiar hallmarks of capitalist development.45

Indeed, in our examination of the role of cities in early modern Europe, we have seen how the search for economic locations on the peripheries of established spaces and regulations (p.297) has been a recurring and fundamental feature of modern Western economies. Decentralization and the search for a greater scope for innovation in relatively peripheral places, marked by a leapfrogging in technology, transportation, organization of the firm, and financial practices, were certainly the hallmarks of the nineteenth‐century industrial revolution. Now, as then, Dodgshon observes, existing industrial complexes and regions have something of ‘the nature of a cul‐de‐sac, a trajectory of commitment that offers finite possibilities and diminishing returns’.46 Entrepreneurial capital, seeking enhanced opportunities, tries to find them in new, less bounded places. From this vantage‐point, the tensions between fixed and circulating capital—between the build‐up of concentrated areas to capture specific advantages of location and the tendency to find new, less‐fettered possibilities—provide crucial mechanisms for capitalism to maintain its revolutionary character. Further, this movement of capital in space in search of advantages continues to have the effects we have already observed in nineteenth‐ and early twentieth‐century cities: transformations in the connections between the worlds of work and home, the division of cities into working and residential zones, and ever‐finer distributions of space by income and housing quality.

Gottdiener has created a new teleology of space, in which the linear trend from concentration to deconcentration is inexorable. There are two main problems with this view. Why assume that a given reconfiguration of production and space will necessarily extend its reach everywhere, or, when it does diffuse, that it will have equal effects in each location on space, civil society, culture, politics, and the formation of collective identities and behaviour? Such patently was not the case with the industrial revolution; why now? The significant changes under way in the organization of economic activities and urban space in the capitalist countries of the West are hardly uniform or one‐dimensional. The older (p.298) Fordist industrial areas are not simply disappearing in a headlong rush to a flexible regime of accumulation; far more typical are attempts by firms to take advantage of the shift in class forces between capital and labour to preserve Fordist patterns in a more globalized and competitive climate. While there is a new diversity in forms of production and labour markets, in the older industrial regions these have not added up to a new regime of accumulation. A multi‐layered complexity, not an emergent homogeneity, characterizes the current situation.47

Another difficulty in Gottdiener's argument is that his portrayal of current changes is deliberately overdrawn. If the dynamic reorganization and layering of space defines what may in fact be a critical branching moment, modern capitalism has remained decidedly urban. Of the host of new centres of living and production that have sprung up, the vast majority concentrate living and working, even if in privatized, sprawling ways that are jarring to residents of older, more dense cities. Cities in the familiar form have not disappeared, nor, on recent trends, are they likely to, even if they are undergoing significant alterations in form and function. The trend of deconcentration, in short, has been accompanied by the counter‐trend of urban concentration.48 The nodal points of the American polynucleated landscape stressed by Gottdiener are rarely situated ‘nowhere’, but either within, near, or in functional relationship with, city settlements. Jacksonville, Houston, Denver, Los Angeles, San Diego, Seattle, to name just a few of the most robust centres of growth in the United States, are cities that, like their older counterparts, concentrate people, culture, capital, and power, even if their agglomerations, (p.299) especially on their outer fringes, have many features that make them appear unrooted in historical space. In these cities and their metropolitan areas, as well as in Boston, New York, Chicago, and other urban centres in the East and Mid‐west, there can be found a complex and dense patchwork of land uses, housing patterns, types of workplace, and linkages to regional, national, and global economies. High technology intersects older patterns of labour‐intensive manufacturing and services. Housing markets appear as ordered and as chaotic. Diversity and predictability compete.49

Overall, there is no evidence of a rush away from cities as loci of living. On the contrary, if the industrial revolution entailed a ruralization of the countryside, by stripping it of non‐agricultural functions, the combination of new communications and transportation technologies, as well as the immense growth in the productivity of agriculture, has now made it possible for the countryside to urbanize. If we examine trends in urbanization since the Second World War in North America and Western Europe, we find a steady increase in the population living in towns and cities side by side with a slowdown in the rate of urbanization as compared to the nineteenth‐century period of city‐building. Bairoch argues that this trend is less a flight from the city as a locus of living and working than a diffusion of city living and ‘a certain harmonization of levels of urbanization’.50 The average size of cities has not grown, but the number of people living in cities, or in metropolitan areas with integral and interdependent ties with central cities, has continued to grow (whether suburban land is included in legal definitions of cities is more a juridical matter than a substantive one of spatial arrangements or ways of life). In spite of recent examples of dispersion, the twentieth century has continued to intensify as an urban century. As Bairoch observes: (p.300)

Around 1800 less than 3% of the total population of developed countries lived in cities with populations of more than 100,000. By 1900 the proportion had risen to 14%, approaching 50% in 1980. In the most highly developed countries, the proportion is even in the order of 60%, and in 1980 there were in the developed world as a whole some 110 cities with populations of more than 1 million.51

If it is not the demise of cities that characterizes the present moment, what does seem to distinguish it is a multiplicity of urban forms, a speed‐up of change, and a reduction in the share of the landscape that is not urban in some extended sense of the term. This new urbanization is highly contradictory in character, mixing centralization and decentralization, a more decisive break between work and residence and new ways of joining these spheres together, a growing homogeneity of space and an enhanced importance of place‐identities and competition between places for economic resources. The conclusion of Susan and Norman Fainstein is apt: ‘In general, we may say that these forces are reducing the technical necessity for a particular form of urban development, rather than imposing an emergent type.’52 Reinvigorated traditional cities, suburban sprawl, and new agglomerations in formerly peripheral areas are developing simultaneously. There has been a change in the new division of labour between cities, with some, usually where major universities are located, becoming centres of research and development, others providing concentrations of headquarters and associated legal and banking services, others manufacturing goods with new technologies, others specializing in white‐collar information processing, others as border entrepôts, yet others as retirement centres.53 In the giant cities like New York and Los Angeles, all of these activities may take place, but in most there is a layering of functions within a hierarchy of economic activities. Urban business people and government officials search for ways to make their specific locales successful ‘growth machines’, and the strategies they (p.301) adopt in pursuit of this goal (and in pursuit of the effort to make growth seem value‐free) have a formative role in shaping the character of local politics, social movements, and conflicts.54

From this perspective, we are not witnessing the end of the city, but a proliferation and diversification of city‐types. The new contingencies of economic and social organization in space certainly include the outcome identified by Gottdiener, but not as the exclusive or inevitable end‐point of current technological, public policy, and investment trends. The new complicated and highly variable patterning of space in late twentieth‐century capitalism has thus created a situation in which for some individuals and families ‘space in toto has been socialized through the annihilation of space by time’,55 and by the replacement of the city as a physical entity by an abstract spatial landscape in which propinquity counts for very little; but for others, still the very large majority, diverse kinds of physical cities remain the location of the pressing features of everyday life.

V

In the face of challenges to the viability of class as the dominant form of collective identity and action, Hobsbawm and Gottdiener sought to find alternative bases of challenge to advanced capitalism. Hobsbawm ended his 1978 lecture on a hortatory note. While the barriers to the resuscitation of working‐class parties and movements are formidable, ‘there is no reason for automatic pessimism’, for even if men and women must act in ‘circumstances that history has provided for them and within its limits . . . it is they who make their history’. But, he counselled, (p.302)

if the labour and socialist movement is to recover its soul, its dynamism, and its historical initiative, we, as Marxists, must do what Marx would certainly have done: to recognize the novel situation in which we find ourselves, to analyse it realistically, and concretely, to analyse the reasons, historical and otherwise, for the failures as well as the successes of the labour movement, and to formulate not only what we would want to do, but what can be done.56

The second lecture provided content to these goals by advising the supercession of old‐fashioned political appeals based on class by a return to what Edward Thompson, in his discussion of the eighteenth century,57 labelled class struggle without class: ‘Now that we are in some ways reverting to a version of the “city crowd” politics of the pre‐industrial days of the “lower orders” and the “labouring poor”,’ Hobsbawm asks, ‘would it not be logical to revert also to the populist politics of those days?’58 In this, Hobsbawm counsels a fundamental rethinking of the practical and discursive bases of collective identities and activities. Similarly, Gottdiener proposes to refocus socialist strategy on struggles to democratize the various features of lived experience, but his farewell to class is even more emphatic. ‘The new political agenda organized around what can be called the sociospatial praxis’, he argues, ‘can make progress only through a clean break with past notions which focus on some abstraction called the working class.’ Moreover, this new politics will require a break with place‐specific organization and understanding in favour of building networks of people to counteract the currently dominant growth strategies. To be viable analytically and politically, Marxism, he counsels, must now leave the city and the working class as relics of times past.59

Hobsbawm's project of re‐reading the city in pre‐industrial fashion and Gottdiener's prescription for a post‐class politics follow in rather too assured a way from their underlying analyses of the city, both of which are only partial accounts of (p.303) changes in particular, not necessarily representative, kinds of urban space. Further, Gottdiener and Hobsbawm arrived at quite similar strategic conclusions by very different means. The resemblance in their calls for a politics that differs from one of class and class struggle should not obscure the irreconcilable features of their analyses. Gottdiener severed while Hobsbawm tightened the ties binding place, social geography, dispositions, and collective action. Neither approach is quite satisfactory. Each is worth a closer look.

Because of the complex layering of time and space in the urban situation of older city forms that refuse to expire and new patterns of urbanization whose contours have yet to be established, and because the pace of change seems ever more rapid, the relevant distances more encompassing, and the line between city, suburb, and country out of focus, the relationship between space and place seems out of kilter to residents and to analysts of the city. Not since the late nineteenth century has urban life been so bewildering, even threatening. The contingent qualities of spatial change, the emergent jumble of mixed forms of old and new, homogeneous and marginally differentiated, cities, and the compression of time and space that has taken place under the impact of massive changes in technology, transport, and communications make it difficult for people to get a coherent grip on urban reality. Cycles of building and rebuilding occur much more rapidly. Familiar landscapes alter. Boundaries between city and country blur. Information diffuses quickly. The pace of life quickens. Ephemerality, fragmentation, and plasticity have become hallmarks of the age.

We have seen how, at a similar moment of bewildering transformations to time and space at the turn of the century, differentiation theory sought to apprehend the new features of the city but in fact redescribed them without locating urban change within a social process framework. A parallel impulse to make plasticity internal to social theory is at work today. Once again it is tempting to mimic the bewilderment of the situation on the ground.

By the time Gottdiener concludes his provocative analysis of the effect of large‐scale processes on cities and space he arrives at a place very similar to Laclau and Mouffe, whose (p.304) Hegemony and Socialist Strategy makes sense of the erosion of class dispositions and behaviour and the new plurality of social movements by setting aside the social‐theoretical tensions of structure and subject in favour of a virtually unanchored, free‐wheeling agency. Recoiling from essentialism, they abjure strong structural‐causal accounts (leaving some room for weaker ones, acknowledging that ‘between total determination and partial limitation there is a whole range of intermediate possibilities’), especially Marxist ones which ascribe a privileged position to a working class with objective interests in anti‐capitalist struggles:

What is now in crisis is a whole conception of socialism which rests upon the ontological centrality of the working class, upon the role of Revolution, with a capital ‘r’, as the founding moment in the transition from one type of society to another, and upon the illusory prospect of a perfectly unitary and homogeneous collective will that will render pointless the moment of politics.60

Their solution is surgical: the excision of the leading position of the working class, and of class structures and categories, from the project of socialism.61 In their rendering, actual (p.305) historical working classes are understood not as having been formed by multi‐levelled processes of materiality and meaning, but exclusively in terms of meaning and the discursive construction of collective identities, almost as if this were wholly a voluntary matter. The result, as Przeworski has put it, is ‘a radical indeterminism in which everything is possible and hence the success of political projects is exclusively a matter of will’.62 Gottdiener differs from Laclau and Mouffe in his acceptance of a special role for the working class within capitalism, but only before recent transformations to production and space. Now that spatial relations have become abstract and generalized, he too thinks the choice of social identities and strategies has become a matter of open selection from competing possible discourses. Gottdiener's materialism becomes an anti‐materialism; it is precisely the dissolution of place in favour of more generalized and abstract space in the (p.306) manner of Lefebvre that effectively makes this unbounded voluntarism possible.

If Gottdiener now opts for unconstrained agency, Hobsbawm's lectures overtilt the other way, as people appear as the bearers of social, structural, and spatial properties. Welcome as Hobsbawm's turn to the city was, his analysis treats the connection between urban social geography and class consciousness in a manner that withdraws independence and autonomy from agency in favour of structural and spatial determinations of working‐class dispositions and patterns of collective action. We have seen, however, that capitalist cities of the nineteenth century sharing in key spatial features were mapped by working classes in quite diverse ways. Why should we think only one possible mapping is currently available?

In a critique of Laclau and Mouffe marred by tendentious personal attack, Norman Geras asserts that their discursive theory that ‘turns its back on Marxism will quickly reach its limits, limits continually recreated by capitalism and class’,63 rather than open up to alternative identities with the capacity to become bases of systemic challenge. These are not mutually exclusive options.

As the social geographies of cities change and diversify, older mappings of the city prove insufficient. Because of the new range and complexity of the urban‐spatial situation, not because discursive choices are commanding and open in principle, issues of class (and group) formation have come to the fore once again with a plurality of solutions. These are not indeterminate, however. They are limited and shaped both by characteristics of urban life, now much more complex and varied than before, and by the imperative that patterns of group and class formation must still map actual urban spaces in ways that make sense of them. There is no a priori reason why class and non‐class categories cannot be put to use in such mappings. In some settings, class identities may recede in importance; in other very similar ones they may grow in significance. It is this variability in relationship to the social and spatial organization of cities that must now be confronted.

We do well to recall, moreover, that the process of making (p.307) sense of cities is not only constrained, as Hobsbawm reminds us, by such material and spatial characteristics as the relationship of work and home and demographic settlement patterns, but that it is also informed by existing urban mappings. These, like all representations of social reality, possess the capacity to outlive the particular conditions of their creation. For this reason, holistic class mappings of the current situation in English cities remain more likely than in equivalently structured American ones.

Thus, both Gottdiener and Hobsbawm can be faulted for eliding some vital issues that pivot on the contingent but not unconstrained relationship of city space and class formation. Because the spatial conditions underpinning working ‘classness’ have altered considerably, the formation of social‐structural identities is once again at a formative state with prospects that cannot be determined directly either by treating space as the unmediated outcome of capitalist development or by treating the dispositions of city residents as the unmediated outcome of spatial configurations. Urban mappings are uncertain because the material world they map is complex, because alternative coherent ways of reading the situation on the ground make a good deal of sense, and because past urban and class representations can still be adapted to fit the new spatial realities. Consider just a limited number of possible alternatives: given the formative power of capitalism in shaping urban space, it is still plausible to map the city in holistic class terms. Given the heightened importance of fragmented housing markets and political jurisdictions, it makes sense to map the city in Weberian class terms as divided between areas distinguished by residents' fine‐grained capacities to consume in the market‐place. And given the ever more stark divide between work and home, a divided pattern of class formation on the traditional American pattern provides yet another possible way to map the new spatial situation. If the new spatial complexity has ushered in a new battle of urban representations, we have learned from past equivalent moments that the situation on the ground alone will not decide it.

If Hobsbawm shares in the limitations of a Marxism that moves too reflexively from material conditions to their representation, (p.308) Gottdiener's attempt to transcend the materiality of city space is at least as problematical as the direction taken by the differentiation theorists of a century ago. An alternative approach of the kind sketched in earlier chapters, that seeks to understand the causal grounding of urban‐spatial change and the ways these have been mapped, continues to commend itself. The large‐scale processes of capitalism and national states continue to be the fundamental determinants of urban‐spatial arrangements. These are structural matters grounded in systemic relationships that have an objective existence independent of any claims to know or to speak of them. In turn, individuals and collectivities continue to experience these processes in actual places of residence and work. How they do so, and with what effect on the composition of structures and spaces, remain central and pressing questions.

To the extent that Marxist social theory wishes to contribute to an understanding of the current kaleidoscope of urban classes, groups, and social movements, it can best do so by learning from the shortcomings of the episodic engagement of Marxism with the city, and from the ways a more open and respatialized Marxism has been able to address such basic puzzles as the transition from feudalism to a world of capitalism and states, divergent patterns of working‐class formation, and the acquiescence of Western working classes in capitalism. In so doing, two questions which classical and most Western Marxism have long considered resolved would be approached agnostically: the degree to which economic determinations as understood by Marxist theory possess the privilege of primacy, and the extent to which phenomena in specific historical settings that lie outside the confines of the process of capitalist accumulation are superstructural in G. A. Cohen's demanding sense of the term. ‘After so many failed prophecies, is it not in the interest of [Marxism] to embrace complexity, be it at some sacrifice of its claim to predictive power?’64

Notes:

(1) ‘The ideological struggle is a struggle about class before it is a struggle among classes.’ Adam Przeworski, ‘Proletariat Into a Class: The Process of Class Formation from Karl Kautsky's The Class Struggle to Recent Controversies’, Politics and Society, 10/1 (1980), 371.

(2) Ira Katznelson. ‘Working Class Formation: Constructing Cases and Comparisons’, in Ira Katznelson and Aristide Zolberg (eds.), Working‐Class Formation: Nineteenth Century Patterns in Western Europe and the United States. Princeton, NJ: Princeton University Press, 1986, 16.

(3) A significant variant of this strategy is the location of sources of working‐class consent within production itself, understood as an ideological and political, as well as economic, process. See Michael Burawoy, Manufacturing Consent. Chicago: University of Chicago Press, 1979; and Burawoy, The Politics of Production. London: Verso Books, 1985.

(4) In some accounts, it was the revolution in transportation technology that was the primary driving force behind the reorganization of city space in the 19th century, rather than the dynamics of capitalist accumulation and the tandem operation of newly distinctive housing and labour markets. I find this unpersuasive. As we saw in Chapter 6, the dynamics of separation predated the late 19th‐century accessibility of public transport to all social classes. Indeed, as Dennis stresses, against a simple technological determinism, investments in transportation making more distant journeys to work possible only made sense once the process of separation of work and home was already under way: ‘Once intra‐urban transport exists, and especially once there is competition between rival railway companies, or between independent bus, tram, and waggonette operators, it is reasonable to argue that innovations are adopted in anticipation of greater profits, or to prevent the erosion of profits by competitors. But to explain initial investment in public transport we have to understand why investors put their money into something new, rather than something proven.’ Noting that alterations to urban structure preceded changes in transportation, Dennis continues: ‘The implication is that transport services facilitated urban growth and change, they permitted suburbanisation, segregation and the separation of residence and workplace beyond walking distance, but they did not initiate change.’ Richard Dennis, English Industrial Cities of the Nineteenth Century: A Social Geography. Cambridge: Cambridge University Press, 1984, 110–11. Where the new technologies of transportation, especially railways, had the greatest impact was less in creating a new urban social geography than in demanding quite radical transformations to accommodate their routes, central city terminals, and sites adjacent to them. In terms of transportation as such, the greatest impact of railways was in reducing barriers to trade and communications between cities, rather than in shifting patterns of commutation within cities and their suburbs. See also John Kellett, The Impact of Railways on Victorian Cities. London: Routledge and Kegan Paul, 1969; Harold Pollins, ‘Transport Lines and Social Divisions’, in Ruth Glass et al., London: Aspects of Change. London: McGibbon and Kee, 1964; and esp. John P. McKay, Tramways and Trolleys: The Rise of Urban Mass Transportation in Europe. Princeton, NJ: Princeton University Press, 1976. McKay stresses how ‘the scope and content of change in European urban transport long lagged behind burgeoning industrialization and urbanization’, and the importance of decisions by public authorities as well as private investment in the making and extension of public transport in cities.

(5) Even for those social geographers who wish to stress the premodern character of mid‐19th‐century cities, this period unambiguously marks the hegemony of these spatial patterns. See three essays by David Ward: ‘Victorian Cities: How Modern?’, Journal of Historical Geography, 1/2 (1975); ‘Environs and Neighbours in the “Two Nations”: Residential Differentiation in Mid‐Nineteenth Century Leeds’, Journal of Historical Geography, 6/2 (1980); and ‘The Place of Victorian Cities in Development Approaches to Urbanization’, in John Patten (ed.), The Expanding City. London: Academic Press, 1983.

(6) I take up this theme in Ira Katznelson, City Trenches: Urban Politics and the Patterning of Class in the United States. New York: Pantheon Books, 1981, ch. 3.

(7) Krishan Kumar, ‘Class and Political Action in Nineteenth‐Century England: Theoretical and Comparative Perspectives’, Archives Européennes de Sociologie, 24/1 (1983), 18. For a discussion of the break between early and late 19th‐century working classes, see also Eric Hobsbawm. ‘The Making of the Working Class 1870–1914’, in Hobsbawm, Worlds of Labour: Further Studies in the History of Labour. London: Weidenfeld and Nicolson, 1984.

(8) Craig Calhoun, ‘Class, Place and Industrial Revolution’, in Nigel Thrift and Peter Williams (eds.), Class and Space: The Making of Urban Society. London: Routledge and Kegan Paul, 1987, 51.

(9) Thus, Calhoun argues, since classes ‘are too large and widely dispersed to be mobilized on the basis of direct interpersonal relationships . . . Large‐scale organization of indirect relationships becomes essential.’ Ibid., 53, 59–60.

(10) Ibid., 54.

(11) Ibid., 56. In a suggestive discussion, Calhoun argues that the modern state provides the archetype for the development of indirect social‐organizational ties between members of the working class. ‘Over a period of hundreds of years,’ he writes, ‘the development of absolutist and eventually parliamentary states reduced the role of personal control and co‐ordination in favour of formal organizational structures. The direct, personal relations of domination characteristic of both feudalism and the cities which grew in late medieval Europe were replaced by the indirect relations of bureaucracy’, and, one might add, of citizenship. The state, moreover, not only was a model for the indirect manifestations of class in organizations but was a key ‘part of the process that produced them. It not only made a broader organization of markets possible, but it sundered the autonomy and unity city life had maintained in both economic and political spheres. Aside from differences in content, this made possible a transformation of the scale of state functioning. States became simultaneously more permanent, more efficient and more powerful . . . Whatever the reasons, though Marxists have debated the relationship between state and party at length, they have not considered that it might be much like that between state and citizenry.’ Ibid., 58–9.

(12) This theme has been the centrepiece of Przeworski's many writings on this theme. See esp. his Capitalism and Social Democracy. Cambridge: Cambridge University Press, 1985.

(13) Calhoun, ‘Class, Place and Industrial Revolution’, 60.

(14) Ibid., 64–7.

(15) To be sure, Calhoun is not entirely silent about these questions. He is sensitive to the enduring significance of face‐to‐face relationships but tends to dichotomize between conflicts based on these community‐based relationships, which he sees mainly in terms of a vulnerable populism, and conflicts based on translocal class organizations. The most satisfactory of his discussions linking the local and the national is that of Chartism, which I comment on in the discussion that follows immediately.

(16) Olivier Zunz, The Changing Face of Inequality: Urbanization, Industrial Development, and Immigrants in Detroit, 1880–1920. Chicago: University of Chicago Press, 1982, 178, 5.

(17) Kathleen Neils Conzen, ‘Immigrants, Immigrant Neighborhoods, and Ethnic Identity: Historical Issues’, Journal of American History, 66 (Dec. 1979), 610.

(18) Zunz, The Changing Face of Inequality, 178–83.

(19) Richard Jules Oestreicher, Solidarity and Fragmentation: Working People and Class Consciousness in Detroit, 1875–1900. Urbana, Ill. University of Illinois Press, 1986, p. xviii.

(20) For a sympathetic discussion of the choices made by the AFL, see Richard Oestreicher, ‘Urban Working‐Class Political Behavior and Theories of American Electoral Politics, 1870–1940’, Journal of American History, 74 (Mar. 1988).

(21) Oestreicher, Solidarity and Fragmentation, 229.

(22) Kumar, ‘Class and Political Action’, 41.

(23) See the discussion of the new social and spatial cleavages of early 20th‐century Detroit in Zunz, ch. 13. He speculates that these changes in living space made the development of a new industrial class consciousness less difficult to achieve.

(24) Conzen, ‘Immigrants, Immigrant Neighborhoods’, 612 ff.

(25) David Ward, ‘Victorian Cities: How Modern?’, 148. Italics added.

(26) Henry Pelling, ‘The Concept of the Labour Aristocracy’, in Pelling, Popular Politics and Society in Late Victorian Britain. London: Macmillan, 1968, 41. Rooted in an article by Engels in 1885 that sought to explain why his expectations of the 1840s had not been borne out, and in Lenin's treatment of imperialism, the concept of a labour aristocracy was introduced into contemporary historical scholarship by Eric Hobsbawm. For the development of the concept, see Friedrich Engels, ‘England in 1845 and 1885’, in the collection of works by Marx and Engels, On Britain. Moscow: International Publishers, 1934, 67, 99; and E. J. Hobsbawm, ‘The Labour Aristocracy’, in Hobsbawm, Labouring Men. London: Weidenfeld and Nicolson, 1964.

(27) Pelling, ‘The Concept of the Labour Aristocracy’, 56. See also the telling criticisms made by H. F. Moorhouse in ‘The Marxist Theory of the Labour Aristocracy’, Social History, 3 (Jan. 1978), and ‘The Significance of the Labour Aristocracy’, Social History, 6 (May 1981).

(28) For a narrowly focused, but revealing discussion see Colin G. Pooley, ‘Housing for the Poorest Poor: Slum‐Clearance and Rehousing in Liverpool, 1890–1918’. Journal of Historical Geography, 11 (Jan. 1985). For different vantage‐points on this set of issues, see the discussion of tax abatements and tax policy in the rebuilding of Vienna in Carl E. Schorske, Fin‐de‐Siècle Vienna: Politics and Culture. New York: Knopf, 1980, ch. 2; for the organization of the politics of property see Avner Offer, Property and Politics, 1870–1914: Londownership, Law, Ideology, and Urban Development in England. Cambridge: Cambridge University Press, 1981; and for an elaboration of urban planning as a discourse and a set of practices see Anthony Sutcliffe, Towards the Planned City. Oxford: Basil Blackwell, 1981, and Richard Fogelsong, Planning the Capitalist City. Princeton, NJ: Princeton University Press, 1986.

(29) Mark Girouard, Cities and People: A Social and Architectural History. New Haven, Conn.: Yale University Press, 1985, 356. In part as a consequence of the work of the 1884–5 Royal Commission on the Housing of the Working Classes, which found that the private sector, even when augmented by building trusts and philanthropic associations, was unable to meet the pressing need for decent working‐class housing, Parliament in 1890 passed the Housing of the Working Classes Act, which enlarged the capacities of local governments to construct housing. This and other such national policy developments went hand in hand with the modernization of local government, represented most dramatically by the creation of the London County Council to replace the mish‐mash of private vestries and the Cities of London and Westminster that had previously been responsible for local government. The result was a new politicization of local issues, the contesting of seats by Fabian socialists and social Liberals at municipal elections, the emergence of municipal socialism as a feature of working‐class mobilization, and the creation of new local instrumentalities to affect the daily lives of workers, especially in the area of housing. Public housing also became a major focus of attention on the Continent, where cities also became the sites, as in the well‐known case of Vienna, of an assertive and successful socialist party politics.

(30) Capitalism did not survive this moment unchanged. Under the impact of a working‐class politics the transactions between the state and the capitalist economy and between the state and its citizens altered significantly to provide for a more assertive role for the state in organizing capitalist markets and, through the welfare state and redistributive public policies, to mitigate the market apportionment of wealth, income, and services.

(31) Mark Gottdiener, ‘Space as a Social Force of Production: Contribution to the Debate on Realism, Capitalism, and Space’, International Journal of Urban and Regional Research, 11 (Sept. 1987).

(32) The literature on the restructuring of capitalism is immense. Some of the best of this work has been provided by Manuel Castells and David Harvey. Castells has been particularly concerned with rapid shifts in communications and with technological transformations. Harvey, within a framework that treats recent changes as evidence confirming Marx's statement in the Communist Manifesto that it is a hallmark of capitalist change that ‘all that is solid melts into air’, adopts the language of the French Regulation school to chronicle a shift from the Fordist regime of post‐war accumulation to a new regime of flexible accumulation. Harvey is particularly concerned to understand the connections between these changes to capitalism's base and the bewildering shifts of experienced time and space and their cultural representations in post‐modernism. I return to these themes below. See Manuel Castells, The Informational City. Oxford: Basil Blackwell, 1990; Castells, ‘High Technology, Economic Restructuring, and the Urban–Regional Process in the United States’, in Castells (ed.), High Technology, Space, and Society. Beverly Hills, Calif.: Sage Publications, 1985; David Harvey, The Condition of Postmodernity. Oxford: Basil Blackwell, 1989.

(33) Sandra Barwick, ‘Sleepy Castleisland Enjoys Home Comforts in the Global Office: A US Insurance Company has Set Up a Claims Office in an Irish Market Town’, Independent, 30 Aug. 1988, 4.

(34) For discussions and representative examples of this literature, see Saskia Sassen, The Mobility of Labor and Capital: A Study in International Investment and Labor Flow. Cambridge: Cambridge University Press, 1988; Doreen Massey, Spatial Divisions of Labour. London: Macmillan, 1984; Erica Schoenberger, ‘From Fordism to Flexible Accumulation: Technology, Competitive Strategies and International Location’, paper presented at the Conference on Technology, Restructuring and Urban/Regional Development, International Sociological Association, Dubrovnik, Yugoslavia, June 1987; Alan J. Scott, ‘Flexible Production Systems and Regional Development: The Rise of New Industrial Spaces in North America and Western Europe’, International Journal of Urban and Regional Research, 12 (June 1988); Peter Hall and Ann Markusen (eds.), Silicon Landscapes. Boston: Allen and Unwin, 1985; N. S. Dorfman, ‘Route 128: The Development of a Regional High Technology Economy’, Research Policy, 12 (1983); Michael Piore and Charles Sabel, The Second Industrial Divide. New York: Basic Books, 1984; John Urry, ‘Class, Space and Disorganized Capitalism’, in Keith Hoggart and Eleonore Kofman, Politics, Geography and Social Stratification. London: Croom Helm, 1986; Scott Lash and John Urry, The End of Organized Capitalism. Cambridge: Polity Press, 1987; and the excellent collection of essays in Alan J. Scott and Michael Storper (eds.), Production, Work, Territory: The Geographical Anatomy of Industrial Capitalism. Boston: Allen and Unwin, 1986.

(35) Scott, ‘Flexible Production Systems’, 179. For discussions, see Thierry Noyelle and Thomas Stanback, The Economic Transformation of American Cities. Totowa, NJ: Rownan and Allanheld, 1984; Peter Hall ‘Technology, Space, and Society in Contemporary Britain’ and Annalee Saxanian, ‘Silicon Valley and Route 128: Regional Prototypes or Historic Exceptions?’, in Castells (ed.), High Technology; Manuel Castells, ‘High Technology and Urban Dynamics in the United States’, and John D. Kasarda, ‘Economic Restructuring and America's Urban Dilemma’, in Mattei Dogan and John D. Kasarda (eds.), A World of Giant Cities: The Metropolis Era. Beverly Hills, Calif.: Sage Publications, 1988; and Christopher B. Leinberger and Charles Lockwood, ‘How Business is Reshaping America’, Atlantic, 258 (Oct. 1986); and Michael Peter Smith, City, State, and Market: The Political Economy of Urban Society. Oxford: Basil Blackwell, 1988.

(36) Mark Gottdiener, The Social Production of Urban Space. Austin, Tex.: University of Texas Press, 1985, 4, 5, 9.

(37) More fundamentally, Gottdiener rejects Marxist treatments of spatial forms as containers of social processes and as manifesting more fundamental material relations in favour of a perspective that gives to space itself a constitutive status at the intersection of a multiplicity of social processes during the ‘late’ phases of capitalist development including, but not limited to, the accumulation process. Space is socially produced at the intersection of economic, ideological, administrative, and political processes, and thus has a multiple ontological status.

(38) Gottdiener, Social Production, 199, 200, 209, 222, 227–8, 241.

(39) Ibid., 285–6, 289–91.

(40) Eric Hobsbawm, ‘The Forward March of Labour Halted?’, and commentaries by Raymond Williams, Robin Blackburn, and Tony Benn, among others, in Martin Jacques and Francis Mulhern (eds.), The Forward March of Labour Halted? London: Verso Books (in association with Marxism Today), 1981, 18.

(41) Eric Hobsbawm, ‘Labour in the Great City’, New Left Review, 166 (Nov.–Dec. 1987), 43, 47.

(42) This is the theme of a neglected article by Nigel Harris, ‘Urban England’, Economy and Society, 3 (Aug. 1974).

(43) Hobsbawm, ‘Labour in the Great City’, 48, 49.

(44) N. J. Thrift, ‘New Times and Spaces? The Perils of Transition Models’, Society and Space, 7 (June 1989), 127–8. Thrift also warns against some other dangers in this mode of analysis. One pitfall is that ‘of simply making the wrong choice about the key elements of the model, leading to a blinkered approach about what is going on. Thus some commentators have fixed upon flexible production systems as the touchstone of current change. Of course, flexible production has led to important changes in certain industries . . . but it is certainly legitimate to ask if it is the most important process at work. What if the assumption implicit in many commentators' accounts, that the important trigger of change must be found in the productive sphere, is wrong? Changes during the 1980s can perhaps be described equally well as the result of the growth of fictitious capital combined with restructuring of financial and commercial capital, the redistribution of income, a massive consumer binge fuelled by easy credit, and an associated engineering of class and consumption.’ An ambitious attempt to develop a more varied, supple, cautious, and contingent model of restructuring is A. Warde, ‘Industrial Restructuring, Local Politics and the Reproduction of Labour Power: Some Theoretical Considerations’, Society and Space, 6 (Mar. 1988).

(45) Susan S. Fainstein and Norman I. Fainstein, ‘Technology, the New International Division of Labor, and Location: Continuities and Disjunctures’, in Robert Beauregard (ed.), Economic Restructuring and Political Response. Urban Affairs Annual, xxxiv. Newbury Park, Calif.: Sage Publications, 1989, 24.

(46) R. A. Dodgshon, ‘Geographical Changes: A Study in Marching Time or the March of Time?’, Society and Space, 5 (June 1987), 190. See also Dodgshon, The European Past: Social Evolution and Spatial Order. London: Macmillan, 1987, chs. 9 and 10.

(47) I have been impressed by the careful review of R. Hudson, ‘Labour Market Changes and New Forms of Work in Old Industrial Regions: Maybe Flexibility for Some but not Flexible Accumulation’, Society and Space, 7 (Mar. 1989).

(48) ‘A suburban office boom is matched by a downtown office boom, and the expansion of “planned communities” in the suburbs is paralleled by the pullulation of fashionable residential and recreational playgrounds downtown.’ Neil Smith, review of Mark Gottdiener's Social Production, American Journal of Sociology, 92 (Sept. 1986).

(49) For a discussion of contemporary Los Angeles along these lines, see Mike Davis, ‘Chinatown, Part Two? The “Internationalization” of Downtown Los Angeles’, New Left Review, 164 (July–Aug. 1989).

(50) Paul Bairoch, Cities and Economic Development: From the Dawn of History to the Present. Chicago: University of Chicago Press, 1988, 220.

(51) Ibid., 226.

(52) Fainstein and Fainstein, ‘Technology’, 25.

(53) This classification of emergent cities is drawn from Harvey Molotch and John R. Logan, ‘Urban Dependencies: New Forms of Use and Exchange in U.S. Cities’, Urban Affairs Quarterly, 21 (Dec. 1985).

(54) The treatment of the city as a ‘growth machine’ is the centrepiece of the important work on urban development and property relations by John R. Logan and Harvey L. Molotch, Urban Fortunes: The Political Economy of Place. Berkeley, Calif.: University of California Press, 1987.

(55) Kevin Cox, ‘Review Essay’, Political Geography Quarterly, 3 (Jan. 1984), 81.

(56) Hobsbawm, ‘The Forward March’, 18, 19.

(57) E. P. Thompson, ‘Eighteenth Century English Society: Class Struggle Without Class?’, Social History, 3 (May 1978).

(58) Hobsbawm, ‘Labour in the Great City’, 51.

(59) Gottdiener, Social Production, 272.

(60) Ernesto Laclau and Chantal Mouffe, Hegemony and Socialist Strategy: Towards a Radical Democratic Politics. London: Verso Books, 1985, 2. See also Laclau and Mouffe, ‘Post‐Marxism Without Apologies’, New Left Review, 166 (Nov.–Dec. 1987), 94, 96, 104.

(61) The disappointingly orthodox rejoinder to Laclau and Mouffe by Ellen Wood reasserted the strong determinism of traditional Marxist base–superstructure social theory, and, in this way, confirmed Laclau and Mouffe's portrait of the baleful features of Marxist analysis as requiring the simplification and reduction of agency to structural necessity. Ellen Meiksins Wood, The Retreat from Class: A New ‘True’ Socialism. London: Verso Books, 1986. See also Norman Geras, ‘Post‐Marxism?’, New Left Review, 163 (May–June 1987); Laclau and Mouffe, ‘Post‐Marxism Without Apologies’; Norman Geras, ‘Ex‐Marxism Without Substance: Being a Real Reply to Laclau and Mouffe’, New Left Review, 169 (May–June 1988). In the first of his essays, Geras upbraids Laclau and Mouffe for giving ‘a caricatured and impoverishing account of what Marxism is . . . In a nutshell, Marxism is defined by Laclau and Mouffe in the most uncompromisingly necessitarian or determinist, most rigidly economistic, and—if one must—most simplifyingly “essentialist” terms; and then dismissed for being determinist, economist, “essentialist” ’ (48). There is some merit to this charge, but, in the circumstances, it is beside the point because it misses the radical quality of the critque of Laclau and Mouffe. For them—and this is a perspective that demands to be taken seriously—the achievements of Marxist writing have been gained in spite of the strong tendencies of the theoretical apparatus of Marxism to reductionism and determinism. The best Marxist work modifies or disposes of some of the core assumptions of the work of Marx. My own approach to their critique is to insist on distinctions between different projects in the work of Marx, and to distinguish aspects of his work where the system of analysis is logically closed and those aspects that are better viewed as open and speculative hypotheses. I am fully in accord with Laclau and Mouffe when they rail against the hubris of a totalizing Marxism; I am in disagreement with their solution, which amounts to a departure from any kind of materialist or realist perspective, whatever its provenance. I think it should be a matter of some regret that Geras chose to pursue his own critique in an incredibly nasty fashion, impugning the personal motivations of Laclau and Mouffe as opportunists seeking after fashion and fame. Not only is this ridiculous and offensive, but it has shifted the character of debate about their work away from issues that matter. Wood's work, while also polemical, is responsible. The problem with her book is its wooden reiteration of old truths as if their restatement amounted to a refutation of Laclau and Mouffe's position or to a coherent way to confront difficulties in Marxist social theory and class analysis.

(62) Adam Przeworski, ‘Class, Production and Politics: A Reply to Burawoy’, Socialist Review, 19 (Apr.–June 1989), 96.

(63) Geras, ‘Post‐Marxism?’, 80–1.

(64) Albert O. Hirschman, ‘Rival Interpretations of Market Society: Civilizing, Destructive, or Feeble?’ Journal of Economic Literature, 20 (Dec. 1982), 1483. Where I have inserted ‘Marxism’ in brackets, Hirschman's text says ‘social science’. I think he will forgive this liberty.