M a n u e l Castells Toward a Sociology of the Network Society Manuel Castells The Call to Sociology The twenty-first century of the Common Era did not necessarily have to usher in a new society. But it did. People around the world feel the winds of multidimensional social change without truly understanding it, let alone feeling a grasp upon the process of change. Thus the challenge to sociology, as the science of study of society. More than ever society needs sociology, but not just any kind of sociology. The sociology that people need is not a normative meta-discipline instructing them, from the authoritative towers of academia, about what is to be done. It is even less a pseudo-sociology made up of empty word games and intellectual narcissism, expressed in terms deliberately incomprehensible for anyone without access to a French-Greek dictionary. Because we need to know, and because people need to know, more than ever we need a sociology rooted in its scientific endeavor. Of course, it must have the specificity of its object of study, and thus of its theories and methods, without mimicking the natural sciences in a futile search for respectability. And it must have a clear purpose of producing objective knowledge (yes! there is such a thing, always in relative terms), brought about by empirical observation, rigorous theorizing, and unequivocal communication. Then we can argue - and we will! - about the best way to proceed with observation, theory building, and formal expression of findings, depending on subject matter and methodological traditions. But without a consensus on sociology as science - indeed, as a specific social science - we sociologists will fail in our professional and intellectual duty at a time when we are needed most. We are needed because, individually and collectively, most people in the world are lost about the meaning of the whirlwind Source: Contemporary Sociology, 29, 5, September 2000: 693-9. we are going through. So they need to know which kind of society we are in, which kind of social processes are emerging, what is structural, and what can be changed through purposive social action. And we are needed because without understanding, people, rightly, will block change, and we may lose the extraordinary potential of creativity embedded into the values and technologies of the Information Age. We are needed because as would-be scientists of society we are positioned better than anyone else to produce knowledge about the new society, and to be credible - or at least more credible than the futurologists and ideologues that litter the interpretation of current historical changes, let alone politicians always jumping on the latest trendy word. So, we are needed, but to do what? Well, to study the processes of constitution, organization, and change of a new society, probably starting with its social structure - what I provisionally call the network society. A New Society Except for a few stubborn academic economists, there is widespread consensus that we have entered a new economy. I contend we are also living in a new society, of which the new economy is only one component. Since this society will unfold, throughout the world, during the twenty-first century, the survival of sociology as a meaningful activity depends on its renewal, in accordance with the new phenomena to be studied and the new analytical issues to be tackled. But what is this new society? Since the focus of this article is on sociology, not society, I have no option but to be schematic and declarative, rather than analytical, taking the liberty to refer the reader to my trilogy on the matter (Castells [ 1996] 2000a). Here are, in my view, the main dimensions of social change that, together and in their Sociology of the Network Society interaction, constitute a new social structure, underlying the "new society." First is a new technological paradigm, based on the deployment of new information technologies and including genetic engineering as the information technology of living matter. I understand technology, following Claude Fischer (1992), as material culture - that is, as a socially embedded process, not as an exogenous factor affecting society. Yet we must take seriously the material transformation of our social fabric, as new information technologies allow the formation of new forms of social organization and social interaction along electronically based information networks. In the same way that the industrial revolution, based upon generation and distribution of energy, could not be separated from the industrial society that characterized the last two centuries, the information technology revolution, still in its early stages, is a powerful component of multidimensional social change. While new information technologies are not causal factors of this social change, they are indispensable means for the actual manifestation of many current processes of social change, such as the emergence of new forms of production and management, of new communication media, or of the globalization of economy and culture. The second dimension of social change is, precisely, globalization, understood as the technological, organizational, and institutional capacity of the core components of a given system (e.g., the economy) to work as a unit in real or chosen time on a planetary scale. This is historically new, in contrast with past forms of advanced internationalization, which could not benefit from information and communication technologies able to handle the current size, complexity, and speed, of the global system, as it has been documented by David Held et al. (1999). The third dimension is the enclosing of dominant cultural manifestations in an interactive, electronic hypertext, which becomes the common frame of reference for symbolic processing from all sources and all messages. The Internet (248 million users currently, in 2000; 700 million projected by the end of 2001; 2 billion by 2007) will link individuals and groups among themselves and to the shared multimedia hypertext. This hypertext constitutes the backbone of a new culture, the culture of real virtuality, in which virtuality becomes a fundamental component of our symbolic environment, and thus of our experience as communicating beings. The fourth axis of change, largely a consequence of the global networks of the economy, communication, and knowledge and information, is the demise of the sovereign nation-state. Not that current nation-states will disappear in their institutional existence, but their existence as power apparatuses is profoundly transformed, as they are either bypassed or rearranged in networks of shared sovereignty formed by national governments, supranational institutions, conational institutions (such as the European Union, NATO, or NAFTA), regional governments, local governments, and NGOs, all interacting in a negotiated process of decision making. As a result, the issue of political representation is redefined as well, since democracy was constituted in the national enclosure. The more key decisions have a global frame of reference, and the more people care about their local experience, the more political representation through the nation-state becomes devoid of meaning other than as a defensive device, a resource of last resort against would-be tyrants or blatantly corrupt politicians. In another axis of structural change, there is a fundamental crisis of patriarchy, brought about by women's insurgency and amplified by gay and lesbian social movements, challenging heterosexuality as a foundation of family. There will be other forms of family, as egalitarian values diffuse by the day, not without struggle and setbacks. But it is difficult to imagine, at least in industrialized societies, the persistence of patriarchal families as the norm. The real issue is how, at which speed, and with which human cost, the crisis of patriarchy will extend, with its own specific forms, into other areas around the world. The crisis of patriarchy, of course, redefines sexuality, socialization, and ultimately personality formation. Because the crisis of the state and of the family, in a world dominated by markets and networks, is creating an institutional void, there are (and increasingly will be) collective affirmations of primary identity around the key themes of religion, nation, ethnicity, locality, which will tend to break up societies based on negotiated institutions, in favor of value-founded communes. Last, but not least, progress in scientific knowledge, and the use of science to correct its own one-sided M a n u e l Castells development, are redefining the relationship between culture and nature that characterized the industrial era. A deep ecological consciousness is permeating the human mind and affecting the way we live, produce, consume, and perceive ourselves. We are just at the beginning of a most extraordinary cultural transformation that is reversing the course of thought that has prevailed among the world's dominant groups since the Enlightenment. different levels of institutional decision making linked by information networks. And the most dynamic social movements are connected via the Internet across the city, the country, and the world. The Network Society: the Social Structure of the Information Age Networks are, however, a very old form of social organization. But throughout history, networks had major advantages and a major problem. Their advantages are flexibility and adaptability, characteristics essential for managing tasks in a world as volatile and mutable as ours. The problem was the embedded inability of networks to manage complexity beyond a critical size. Networks were historically useful for personal interaction, for solidarity, for reciprocal support. But they were bad performers in mobilizing resources and focusing these resources on the execution of a given task. Large, centralized apparatuses usually outperformed networks in the conduct of war, in the exercise of power, in symbolic domination, and in the organization of standardized, mass production. Yet this substantial limitation of networks' competitive capacity was overcome with the development of new information/communication technologies, epitomized by the Internet. Electronic communication systems give networks the capacity to decentralize and adapt the execution of tasks, while coordinating purpose and decision making. Therefore, flexibility can be achieved without sacrificing performance. Because of their superior performing capacity, networks, through competition, are gradually eliminating centered, hierarchical forms of organization in their specific realm of activity. The new society is made up of networks. Global financial markets are built on electronic networks that process financial transactions in real time. The Internet is a network of computer networks. The electronic hypertext, linking different media in global/local connection, is made up of networks of communication - production studios, newsrooms, computerized information systems, mobile transmission units, and increasingly interactive senders and receivers. The global economy is a network of financial transactions, production sites, markets, and labor pools, powered by money, information, and business organization. The network enterprise, as a new form of business organization, is made of networks of firms or subunits of firms organized around the performance of a business project. Governance relies on the articulation among A network is a set of interconnected nodes. Networks are flexible, adaptive structures that, powered by information technology, can perform any task that has been programmed in the network. They can expand indefinitely, incorporating any new node by simply reconfiguring themselves, on the condition that these new nodes do not represent an obstacle to fulfilling key instructions in their program. For instance, all regions in the world may be linked into the global economy, but only to the point where they add value to the valuemaking function of this economy, by their contribution in human resources, markets, raw materials, or other components of production and distribution. If a region is not valuable to such a network, it will not be linked up; or if it ceases to be valuable, it will be switched off, without the network as a whole suffering major inconvenience. Naturally, networks based on This new society was produced during the last quarter of the twentieth century, through the interaction among three independent processes that happened to coincide in time: the revolution in information technology; the socioeconomic restructuring of both capitalism and statism (with different fates for these antagonistic modes of production); and the cultural social movements that emerged in the 1960s in the United States and Western Europe. While this multidimensional social change induces a variety of social and cultural expressions in each specific institutional context, I propose the notion that there is some commonality in the outcome, if not in the process, at the level where new social forms are constituted - that is, in the social structure. At the roots of the new society, in all its diversity, is a new social structure, the network society. Sociology of the Network Society alternative values also exist, and their social morphology is similar to that of dominant networks, so that social conflicts take the shape of network-based struggles to reprogram opposite networks from the outside. How? By scripting new codes (new values, for instance) in the goals organizing the performance of the network. This is why the main social struggles of the information age lie in the redefinition of cultural codes in the human mind. The prevalence of networks in organizing social practice redefines social structure in our societies. By social structure I mean the organizational arrangements of humans in relationships of production/consumption, experience, and power, as expressed in meaningful interaction framed by culture. In the Information Age, these specific organizational arrangements are based on information networks powered by microelectronicsbased information technologies (and in the near future by biologically based information technologies). Under the conditions of this new, emerging social structure, sociology must address several conceptual and methodological issues in order to be equipped to analyze core processes of social organization and social practice. Theorizing Social Structure as Interactive Information Networks The study of social networks is well established in sociological research, spearheaded in contemporary American sociology by Wellman (e.g., 1999), Fischer (e.g., 1992), and Granovetter (e.g., 1985). There is also an international association for the study of social networks, which constitutes a fruitful milieu of research. It can provide concepts and methods that will foster understanding of social networks as specific forms of organization and relationship, including electronic communication networks. Yet, while building on this tradition, I advance the notion that twenty-first-century sociology will have to expand the network-based perspective to the analysis of the entire social structure, in accordance with current trends of social evolution. This implies more than analyzing social networks. It will require reconceptualizing many social processes and institutions as expressions of networks, moving away from conceptual frameworks organized around the notion of centers and hierarchies. For the sake of communication, I will use two illustrations to make my case, taking them from two different and very traditional sociological fields: industrial sociology and urban sociology. I will then draw some general theoretical implications from this change of perspective. The prevailing form of business organization emerging in advanced societies and diffusing throughout the global economy is the network enterprise, which I define, in sociological terms, as the specific form of enterprise whose system of means is constituted by the intersection of segments of autonomous systems of goals. It follows a complete transformation of relationships of production and management, and thus of the occupational structure on which social structure is largely based. How can we conceptualize the role of producers of information in their differential position along an interactive network? How can we conceptualize the variable geometry of new industrial organizations, based on firms' permeable boundaries, bringing together workers, capital, and knowledge in specific projects that form, dissolve, and reform under a different configuration? Yes, work, workers, exploitation, cooperation, conflict, and negotiation do not disappear, but the ensuing individualization of the relationship between management and labor and the ephemeral character of project-based, industrial organizations require a new conceptual apparatus, focusing on networked relationships rather than on vertical hierarchies. In this perspective, I propose to conceptualize the new occupational structure around the interaction among three dimensions of production relationships: value making, relation making, and decision making. For value making, in an information-based production process, we may differentiate various structural positions: the commanders (or strategists), the researchers, the designers, the integrators, the operators, and the human terminals. Relation making defines another set of positions: the networkers, the networked, and the switched-off. And the relative positioning in decision making differentiates among the deciders, the participants, and the executors. The three dimensions are analytically independent. Thus, the empirical observation of the various arrangements among different positions in the three dimensions built around the performance of a given project may yield some clues on the emergence of new social relationships of production, at the source of new social structure. M a n u e l Castells A second example: the transformation of spatial structure, a classic theme of urban sociology. With the diffusion of electronically based communication technologies, territorial contiguity ceases to be a precondition for the simultaneity of interactive social practices. But "the death of distance" is not the end of the spatial dimension of society. First, the "space of places," based in meaningful physical proximity, continues to be a major source of experience and function for many people and in many circumstances. And second, distant, interactive communication does not eliminate space; it transforms it. A new form of space emerges "the space of flows." It is made of electronic circuits and information systems, but it is also made of territories, physical places, whose functional or symbolic meaning depends on their connection to a network, rather than on its specific characteristics as localities. can develop models to analyze the new spatial forms constituted around interterritorial networks, and then examine their differential relationship to their surrounding, local environments. Thus, it is the connection between local and global, rather than the "end of geography" in the age of globalization, that becomes the appropriate perspective for the new urban sociology (Borja and Castells 1997). Networks of discontiguous places in interaction with a diverse range of localities are the components of the new sociospatial structure. The central analytical question then becomes how shared social meaning is produced out of disjointed spatial units reunited in a purely instrumental, global logic (Castells 2000b). By redefining spatial structure on the basis of a networking logic, we open up a new frontier for one of the oldest sociological traditions, urban sociology. The space of flows is made of bits and pieces of places, connected by telecommunications, fast transportation, and information systems, and marked by symbols and spaces of intermediation (such as airports, international hotels, business centers, symbolized by de-localized architecture). For instance, in recent years there has been considerable debate about the emergence of "the global city." The global city is not just a major metropolitan center that ranks high in the worldwide geography of management of wealth and information. For such cities (New York, London, Tokyo, Paris, or Sao Paulo) we already had the descriptive notion of "world city," proposed 20 years ago. The global city, in the strict analytical sense, is not any particular city. And empirically it extends to spaces located in many cities around the world, some extralarge, others large, and still others not so large. The global city is made of territories that in different cities ensure the management of the global economy and of global information networks. Thus, a few blocks in Manhattan are part of the global city, but most of New York, in fact most of Manhattan, is very local, not global. These globalized segments of Manhattan are linked to other spaces around the world, which are connected in networks of global management, while being loosely connected to their territorial hinterlands. The analysis of social structures as a multidimensional, evolving system of dynamic networks may help explain social evolution in the Information Age. Indeed, networks are dynamic, self-evolving structures, which, powered by information technology and communicating with the same digital language, can grow, and include all social expressions, compatible with each network's goals. Networks increase their value exponentially as they add nodes. In formal terms, as proposed years ago by computer scientist and Internet entrepreneur Bob Metcalfe, the value of a net increases as the square of the number of nodes on the net. (The precise formula is V = n ~ \ where V* is the value of the network and n the number of nodes). Thus, a networked social structure is an open system that can expand indefinitely, as long as the networks included in the meta-network are compatible. So the global city is a network of noncontiguous territories, reunited around the task of managing globalism by networks that transcend locality (Graham and Simon 2000). From this theoretical perspective we {n l The issue arises, then, of the contradictions among networks, which lead to conflicts and social change. In fact, network theory could help solve one of the greatest difficulties in the explanation of social change. The history of sociology is dominated by the juxtaposition of and lack of integration between the analysis of social structure and the analysis of social change. Structuralism and subjectivism have rarely been integrated in the same theoretical framework. A perspective based on interactive networks as the common basis for social structure and social action may yield some theoretical results by ensuring the communication, within the same logic, between these two planes Sociology of t he N e t w o r k Society of human practice. A social structure made up of networks is an interactive system, constantly on the move. Social actors constituted as networks add and subtract components, which bring with them into the acting network new values and interests defined in terms of their matrix in the changing social structure. Structures make practices, and practices enact and change structures following the same networking logic and dealing in similar terms with the programming and reprogramming of networks' goals, by setting up these goals on the basis of cultural codes. A theory based on the concept of a social structure built on dynamic networks breaks with the two reductionist metaphors on which sociology was based historically: the mechanical view of society as a machine made up of institutions and organizations; and the organicist view of society as a body, integrated with organs with specific bodily functions. Instead, if we need a new metaphor, the sociology of the network society would be built on the self-generating processes discovered by molecular biology, as cells evolve and develop through their interaction in a network of networks, within the body and with their environment. Interactive networks are the components of social structure, as well as the agencies of social change. The sociology of the network society may be able to bridge structure and practice in the same analytical grasp. A New Methodology? The renewal of the study of society cannot proceed just on theoretical grounds. Sociology is an empirical science, within all the limits inherent to the constraints of observation under non-experimental conditions. Thus, new issues, new concepts, new perspectives require new tools. The emergence of interactive information networks as the backbone of social structure makes even more acute the need to take up the greatest methodological challenge for empirical research in sociology. While most of our analytical tools are based on linear relationships, most social phenomena - even more so in the network society - are characterized by nonlinear dynamics. But in the last two decades, we have witnessed the development of numerous research tools able to deal with nonlinear relationships. On one hand, we have an expanding field of the new mathematics of complexity based on notions such as fractals, emergent properties, autopoietic networks, and the like (Capra 1996). Most of these mathematical discoveries remain confined to formal exercises with slight relationship to empirical research. But they are tools ready to be used, transformed, and perfected by able researchers with both the knowledge of the tools and the substantive knowledge to make sense of this formal language. On the other hand, enhanced power of computers, and new, flexible computer programming languages, enable us to handle the complexity of an interactive network structure in precise terms. Computer-based system analysis of dynamic networks may constitute a fruitful approach through which observation and theory can be reconciled without excessive social reductionism. Simulation models in the social sciences got off to a bad start in the 1960s because their underlying theories were utterly simplistic, and computer programs were technically constrained by their set of rigid assumptions. But new computing capacity, in dynamic interaction of alternative assumptions processed at high speed, may change everything - as is already happening in biological research. In this sense, computational literacy (that is, knowing how to interact with computers, rather than just run statistical programs) may be a fundamental learning requirement for the current generation of young sociologists those who will analyze the network society. In doing so, they will be fortunate enough to have access to a huge pool of information via the Internet. Given knowledge of languages (or automated translation programs), access to global sources may liberate sociology from the embedded ethnocentrism of its observation. Each study may be comparative or crosscultural in its approach, by contrasting observation generated ex novo in a particular study to the accumulated knowledge on the matter from global sources. Naturally, critique of sources as well as problems of methodological integration of diverse data will be necessary requisites for use of this wealth of information. The practice of meta-analysis, in full development in other sciences, particularly economics, may become a standard tool of sociological research. This would also require proper training and methodological guidance for sociologists to benefit from Peter Marcuse expanded possibilities of information without being overwhelmed by it. Overall, sociology should, and will, overcome the sterile, artificial opposition between quantitative and qualitative research, and between theory and empirical study. In the perspective of computational literacy, and with the formal integration of observations in a theory that conceives social structure as a network of interactive networks, it does not really matter what comes from statistics or from ethnography. What matters is the accuracy of the observation, and its meaning. Thus, formal models scripted in the computer programs must be theoretically informed, yet able to be given information apt to answer the questions raised in the theory. The sociology of the network society will develop through synergy among relevant theorizing, computational literacy, and sociological imagination. REFERENCES Borja, Jordi and Manuel Castells. 1997. Local and Global: The Management of Cities in the Information Age. London: Earthscan. Capra, Fritjof. 1996. The Web of Life: A New Scientific Understanding of Living Systems. New York: Doubleday. Carnoy, Martin. 2000. Work, Family, and Community in the Information Age. Cambridge, MA: Harvard University Press. Castells, Manuel. [1996] 2000a. The Information Age: Economy, Society, and Culture. 3 vols. 2d Ed. Oxford and Maiden, MA: Blackwell. . 2000b. "The Culture of Cities in the Information Age." Presented at conference on The Library of Congress, "Frontiers of the Mind in the 21st Century." Forthcoming (2001) in The Castells Reader on Cities and Social Theory, edited by Ida Susser. Oxford and Maiden, MA: Blackwell. Fischer, Claude. 1992. America Calling. Berkeley: University of California Press. Graham, Stephen and Marvin Simon. 2000. Splintering Networks. London: Routledge. Granovetter, Mark. 1985. "Economic Action and Social Structure: The Problem of Embeddedness." American Journal of Sociology 19:481-510. Held, David, Anthony McGrew, David Goldblatt, and Jonathan Perraton. 1999. Global Transformations. Stanford, CA: Stanford University Press. Wellman, Barry, ed. 1999. Networks in the Global Village: Life in Contemporary Communities. Boulder, CO: Westview Press. Depoliticizing Globalization: From Neo-Marxism to the Network Society of Manuel Castells Peter Marcuse [...] It is precisely the shift of focus away from the nature of, and the relationships among, social groups that marks Castells's trajectory. It is a move that suppresses the political, in the broad sense of the dynamic between the exercise of power and the resistance to it, and moves toward a determinism that undermines the relevance of political action. Power and conflicts over power disappear from view; classes, when they appear, have a very subordinate role. Capitalism is conflated with globalization, but in an ambiguous and ahistorical fashion; technology, the media, demographic changes, the state appear as homogeneous, autonomous entities, actors themselves, behind whom actual actors are not to be seen. It is a classic case of reification, making the relations among human beings appear as a relationship among things, the relationships of social and economic position appear as relationships to or against Depoliticizing technology, to or against the ascendance of "information." In place of the tensions, the contradictions, the conflicts among human actors and groups as the motor of change, there is a march of technology, of organizational forms, of their own accord, inexorably, globally. Human actors only react to these developments (some benefit from them, but not much attention is paid to them, and they are not seen as more than passive participants in the march). The critique of globalization implicit and often explicit in the books concludes with an appeal to "us" to understand, communicate, become aware, together; any drawing of policy conclusions or indications for action is deliberately rejected. The discussion becomes depoliticized, both in its analysis and in its stance toward prescription: in Castells's words, "the power of flows takes precedence over the flows of power." To be clear: by "depoliticized" I do not mean that Castells, or any other author, has an obligation to draw political conclusions and/or present political prescriptions as part of his or her work, although it may be desirable that more extend their work in these directions than now do. I mean rather that the political content present in the world Castells is analyzing is suppressed, played down, becomes incidental, in contrast to its role in reality. I take the political to be centered on relations of power among social actors; these play at best a secondary role in Castells's analysis, where they appear at all. The criticism is not that Castells fails to introduce a political analysis into the material he examines, but that he does not adequately deal with the content that is in fact in his material; not that he should politicize material that is nonpolitical, but that he has depoliticized material that is itself heavily political. The problem is symbolized and encapsulated by the very title of Castells's magnum opus: The Information Age. What is central in the analysis is a technical development (and a somewhat mystified one at that [ . . . ] ) , not a social one. It makes the tools of production, rather than the relations of production, the characteristic of the age: thus the sequence might be: Stone Age, Bronze Age, Iron Age, Steam Age, Information Age, rather than Imperial Age, Feudal Age, Capitalist Age, Imperialist Age, Fordist Age, followed perhaps by various attempts at a further definition: Neo-Imperialist, Post-Fordist. The point is not the accuracy of any of these classification schemes, but what it is that is at the center of Globalization them, what is taken as the indicative classificatory criterion. Even in traditional sociology and traditional economics, and certainly in Marx, it is the relations among and characteristics of groups within each society that are its defining characteristics. Not here. The depoliticization of what would be, underneath it all, a sharp analysis of events can be traced in a number of areas. The language used systematically undermines the substance of the analysis and robs it of a political force it might otherwise have. A few examples highlight the issues here raised. The Eradication of Human Agency A key aspect of depoliticization is to make everything that happens anonymous, actor-less. It is not merely the old agency versus structure argument within Marxism, for in those discussions both sides always assume that structure refers to the pattern of relations among actors, among classes, and the issues invoke scale, proportion, relative weight, scope of human agency within structure. With Castells, agency vanishes, actors disappear from sight. Both the language and the content of what he writes lead in this direction. Castells does at times deal with the question of agency: "who are the capitalists?" he asks. He points out that there is no simple answer, that they are a "colorful array" of characters, and seems to open the door to a deeper discussion of class composition in advanced industrial societies and their global linkages. But then he proceeds: "above a diversity of human-flesh capitalists and capitalist groups there is a faceless collective capitalist, made up of financial flows operated by electronic networks." Important points do need to be made here as to the autonomy of individual capitalists, the difference between a conspiracy and a class, how power is exercised, and so on. But the discussion does not go in this direction. Instead, the conclusion is the flat statement that "there is not [...] such a thing as a global capitalist class." Rather, "capitalist classes are [...] appendixes to a mighty whirlwind." "Who are the owners, who the producers, who the managers, and who the servants, becomes increasingly blurred." Maybe to Castells, but not to the majority of the world's peoples, I would guess. This is depoliticization with a vengeance: not power relations, but a "mighty Peter Marcuse whirlwind governs our actions [...] Power [...] is no longer concentrated in institutions (the state), organizations (capitalist firms), or symbolic controllers (corporate media, churches). It is diffused in global networks of wealth, power, information, and images [...] The new power lies in the codes of information and in the images of representation around which societies organize their institutions [...] The sites of this power are people's minds." If power should be challenged, then, the entity responsible is the "society" which does the organizing; it does no good to criticize the state, or firms, or the media. The "realpolitik" of domination, to which Castells also refers elsewhere, is not the logical description: a "first level" which "concerns the holders of property rights." The "second level [...] refers to the managerial class." But here the reference to class ends; we get no closer than this to the flesh and blood of real actors. For "the third level [...] [has to do with] the nature of global financial markets. Global financial markets, and their networks of management, are the actual collective capitalist [...] global financial networks are the nerve center of informational capitalism." So, in the end, the capitalists are not a "who" but a market; not those networking, but the network itself. The Passive Voice The Excluded Without the Excluders In general there is much detail on those who are excluded, but not on those who exclude them. The process of exclusion is faceless, a world-historical process at the "end of millennium," not one for which any single group or class can be held accountable. In the substantial discussion of the exclusion of "the majority of the African population in the newest international division of labor," Castells concludes "that structural irrelevance (from the systems point of view) is a more threatening condition than dependency"; "a considerable number of humans [...] are irrelevant [...] from the perspective of the system's logic." Irrelevance is from "the system's" point of view, not from the point of view of those who can make no profit from the lives of the excluded. Some are excluded, but no one does the excluding. Actors disappear entirely in the blanket laid down by the language of sweeping phrases: "social forms and processes induced by the current process of historical change." (And one might raise the question of whether the excluded are really excluded from the system, or whether they are in fact quite useful for it but simply excluded from its benefits . . . ) In the conclusion to the third volume, Castells deals most explicitly with the question of who is responsible to the new informational/global economy. "The rule is still production for the sake of profit, and for the private appropriation of profit, on the basis of property rights - which is of the essence of capitalism. But [...] [w]ho are the capitalists?" The discussion then begins with a Castells uses the passive voice constantly, where an active grammar would raise the question of exactly who is responsible, or, if simple agency is not adequate to explain structural patterns, what forces, what relationships of power, what institutions or practices are involved and should be held accountable. The problem occurs from the opening to the closing of the three volumes. In the first chapter, "global networks of instrumental exchanges selectively switch on and off individuals, groups, regions, and even countries, according to their relevance in fulfilling the goals processed in the network, in a relentless flow of strategic decisions [...] Our societies are increasingly structured around a bipolar opposition between the Net and the Self." "The" Net (capitalized?) and "the" Self (capitalized?). Just what does that mean? Networks among some groups are indeed in opposition to the selfdevelopment of other groups; there is "opposition" in the patterns Castells describes, but not conflict. In fact, it is not "global networks of instrumental exchanges" but networks of specific corporations, power blocs, states that "switch on and off" very specific individuals, groups, regions, and countries - and not any random individuals, countries, etc., all characterized by their concern with the "Self," but poor and working people, Third World countries, women. In the last chapter, the passive voice continues to color the discussion of the transformations the three volumes describe. "Relations o/production have been transformed." "[L]abor is redefined in its role as produced, and sharply differentiated according to workers' characteristics," and "generic labor is assigned a given Depoliticizing G l o b a l i z a t i o n task." "[C]apital is as transformed as labor is in this new economy," just as Castells elsewhere gives ample evidence of who benefits and who is hurt. But the presentation shifts the focus away from any person's or group's responsibility and on to the tools, the instruments, the "networks of instrumental exchanges" used by some to achieve their results at the cost of others. on that issue would seem vital, since if it is "technology that transforms," little can be done about it, absent Luddite initiatives, but if socioeconomic forces are involved, they can indeed be addressed, and with them the uses to which technology is put. Globalization as "Actor," Ail-Powerful The Imputation of Agency to Things This is, in a sense, the mirror image of the disappearance of real actors from view: processes and relationships become reified, become actors themselves, autonomously, independently of human agency. Real actors disappear, and things become actors. Technology becomes an independent actor, an autonomous force. We read sentences like: "technology has transformed the political role of the media." Not that political actors have taken advantage of technological developments to use media in a new role; the technology itself achieves the transformation. The new "techno-economic paradigm [...] is based primarily on [...] cheap inputs of information." The role of the media is indeed analyzed perceptively, even with an undertone of moral condemnation, but, since technology is to blame, there is no suggestion that calls for different ownership or control of the media would make a difference. The kind of media analysis undertaken by writers such as Herbert Gans, Noam Chomsky, or Douglas Kellner is not mentioned. The opposite view is also to be found in Castells, with the contradictions unresolved. For instance, elsewhere Castells explicitly abjures technological determinism; he could hardly have said it more bluntly: "The Information Technology Revolution DID NOT create the network society." Yet, as is frequently the case, the language of the discussion constantly contradicts the broad theoretical statement. Technology is an independent process, independent both of economics and culture. At the same time, "Information technology bec[omes] the indispensable tool for the effective implementation of processes of socio-economic restructuring." The ambivalence as to the explanatory role of technology vis-à-vis socioeconomic restructuring runs throughout the discussion. For any analysis of the politics of the developments he describes, clarity It is treated as an entity, an active force; indeed, if the whirlwind has a name, it is globalization. Yet the precise meaning of globalization remains fuzzy. In volume 1 it appears primarily as a globalization of the economy, coupled necessarily with "informationalism," as a "historical discontinuity" from the past. In volume 2 its sweep is broader, and it assumes cultural and social forms as well. The issue of its newness "does not concern my inquiry." Yet we read that "globalization [...] dissolves the autonomy of institutions, organizations, and communication systems." If that is the case, just what globalization is, whether it is a new phenomenon or not, becomes critical, despite Castells's claim to the contrary. The picture suggests that not specific actors, not multinational corporations overriding national boundaries, not capital moving without effective restraint to and from wherever it wishes are at work, but the anonymous process of globalization. If globalization is not new, then we might well ask whether it is not capitalism as such, perhaps simply in a further advanced form, which is responsible for the developments Castells accurately describes. And if it is indeed capitalism, then capitalists might also bear some responsibility, and the political content of the conceptualization becomes clear. With the shift of focus to globalization, that political content disappears. Nowhere is there an intimation that globalization is a process that can be altered or stopped, that really existing globalization is not the only form globalization might take. Globalization is presented as whirlwind, sweeping everything in its path. Conflict Is Bypassed or Suppressed The second volume, titled The Power of Identity, focuses on social movements, which are defined "as being: Peter Marcuse purposive collective actions whose outcome, in victory as in defeat, transforms the values and institutions of society." The implication here is that conflict, victory or defeat, is the essence of what social movements are about, with those who support and represent the "values and institutions of society" as their clear antagonists. Conflict might thus be expected to be a critical element in the discussion of social movements, now discussed under the rubric of "identity." But in what follows "social actors [...] excluded from [...] the individualization of identity [...] in the global networks of power and wealth" are not engaged in conflict with those who have excluded them (nameless; see below), but rather these social actors are engaged in a search "for the construction of meaning." Their organizations, social movements, are not movements defined by conflict with those who have deprived them of meaning (and, presumably, of key material resources for living a decent life - the term "exploitation" does not feature in any of the three volumes). They are "cultural communes," "organized around a specific set of values [...] marked by specific codes of selfidentification." As elsewhere, Castells has it both ways. In the end, there need not be conflict; ultimately, the solution is for "all urban [sic] agents [to develop] a city project which impregnates civic culture and manages to achieve broad consensus." The earlier centrality of conflict has given way to the anticipation of consensus. Identity (Social Movements) Becomes a Reactive Phenomenon What identities react to, and indeed the definition of identity, is unclear. A formal definition is provided: "I mean by identity the process by which an actor [...] constructs meaning primarily on the basis of a given cultural attribute [...] to the exclusion of a broader reference to other social structures." Why an identity thus constructed cannot also have reference to other social structures is uncertain, and indeed in many examples in volume 2 they clearly do, for example, the feminist movement or the civil rights movement. And within a few pages fundamentalism, clearly taken as an identity movement, is put forward as a reaction to the exclusion of large segments of societies, presumably a "reference to other social structures." And why is a working-class identity not an identity? And to what are "identities" reacting? In one place it is to "the logic of apparatuses and markets," in other words, to social circumstances; in other places it is to globalization; in others, to "excluders"; in another, to "the crisis of patriarchalism"; in still another, to "the unpredictability of the unknown." Granted that identities are indeed very diverse, in what sense can one then use the category as a meaningful single concept? And yet, in the discussion, the functional differences among identities in the end disappear; all identities are treated as reactions, and reactions against generalized processes. Enemies do not appear; processes operate without operators or subjects. Although there is detailed and perceptive discussion of resistance movements in volume 2, the resistance is not against any one or any group in particular: Religious fundamentalism, cultural nationalism, territorial communes are [...] defensive reactions. Reactions against three fundamental threats [...] Reaction against globalization [...] Reaction against networking and flexibility [...] And reaction against the crisis of the patriarchal family [...] When the world becomes too large to be controlled [...] When networks dissolve time and space [...] when the patriarchal sustainment of personality breaks down [...] [people react.] The reaction is not by people to other people doing things to them, but to faceless processes. True enough, people often do not see who is doing what to whom, and the descriptions Castells provides are often graphic and trenchant. But then is it not precisely the obligation of analysis to clarify who and what is involved, and are not formulations like those above in fact concealing what is happening, disarming more targeted resistance? In presenting identity movements as against faceless and actor-less processes, the movements themselves become similarly "soft"; they are not defined by their own interests, their own capacities, their own understandings, but only by that "process" which they are up against. In fact, Castells also includes a much more analytic and political discussion of identities, differentiating between legitimizing identities, those which are Depoliticizing G l o b a l i z a t i o n introduced by dominant institutions and reinforce domination, resistance identities, those generated by the dominated to creates trenches of resistance, and project identities, those seeking to redefine positions in society and the transformation of the overall social structure. It is a useful categorization, harking back to the discussions of the 1960s as to the nature of social movements and their radical or systemmaintaining roles. But it is a tool not then consistently carried forward in a discussion in which religious fundamentalism, the Zapatistas, the Patriot Movement in the United States, Japan's Aum Shinrikyo, the environmental movement, the women's movement, and the Lesbian and Gay Liberation movements are more or less given equal treatment under the uniform heading of "identity" movements. Are there in fact any "project identities"? John Friedmann points out that the category of "project identity" into which Castells puts movements that "seek the transformation of overall social structure" is empty. Castells is a little ambiguous on the issue; at one point, he suggests that project identities may be involved in efforts at liberating women "through the realization of women's identity," or in movements, "under the guidance of God's law, be it Allah or Jesus." At another point he says that from "cultural communes" "new subjects [...] may emerge, thus constructing new meaning around project identity." And in the concluding chapter of the volume entitled The Power of Identity, he speaks merely of "project identities potentially emerging from these spaces [of resistance]." Identity, social movements built around identity, are not then today agents of political action; identity is not very powerful, according to Castells, despite the book's title. The Independence of Key Phenomena This is a part of the picture. At various times and places, Castells suggests the connections among the various phenomena he includes together under the various umbrella terms that frequently appear: the "information age," the "network society," the "global era." While these phenomena are discussed separately in the three volumes, Castells brings them together in a summary article: "The Information Technology Revolution [...] The restructuring of capitalism [...] The cultural social movements." And he is explicit about the connection: "The network society [...] resulted from the historical convergence of [these] three independent processes, from whose interaction emerged the network society." The language is slippery: are they independent if they interact? To what extent does their interaction determine their nature and direction? Is the "historical convergence" just an accident? The detailed discussion of each suggests that they are indeed independent forces, each with an independent shape. Technological development, appearing independent, moves by its own laws, outside of political control, and social movements are not presented as efforts to control, redirect, or prevent the restructuring of capitalism. That a coherent set of actors is involved in each of the three phenomena drops out of sight. The evidence that "capitalist restructuring" molded the direction, extent, and nature of technological change, coming into conflict with, exacerbating, and highlighting cultural and social movements, is not taken up. "The Depoliticization of Space" This is a somewhat unexpected aspect of Castells's presentation. Castells has made a major contribution to the contemporary discussion of space in his evocation of the duality of the space of places and the space of flows; the terms have become an accepted part of the social science vocabulary. The space of places refers to that space to which some people are bound: perhaps unskilled workers, those without the means or the legal status for mobility, those to whom a particular location, city, territory, is a fundamental part of their identity, those who are tied to a particular space/place. The space of flows, by contrast, is used by those with unrestricted mobility and is the space in which capital moves, in which high-level financial transactions occur, in which decisions are made and control exercised, the space which the dominant networks of the advanced network society occupy. There is real meat here: the worlds of those who are locationbound and those with unrestricted mobility, both in their personal lives and in their transactions, are two different worlds; although, as Michael Peter Smith points out [ . . . ] , to set the two up as a binary opposition Peter Marcuse hardly reflects their complex and overlapping nature: the users of the "space of flows" are also place-bound in many aspects of their activities, and many denizens of the "space of places" frequently move large distances and across borders, in increasingly frequent transnational patterns. Is it useful to convert the differences between these two worlds into a difference originating in/characterized by their use of space, rather than looking at the differences in the use of space as the outcome of differences in wealth, power, resources? Is the space of flows in any meaningful way really a space, or is it not rather a freedom from spatial constraints? Is the space of places really not also made up of flows as well as localities? What needs analysis, for political evaluation, is the extent to which those who use the "space of flows," the dominant groups in the global society, are or are not free of locational bounds. The difference between the occupants of the space of places and the users of the space of flows is a class difference, reflected in their relationship to space, reinforced but not created by it. Examining differences in the use of space without examining the differences in class, power, and wealth which produce those spatial differences is stripping social science analysis of its political relevance: depoliticization. Worse, space itself becomes an actor, affirmatively displacing real persons and interests: "Function and power [...] are organized in the space of flows [...] the structural domination of its logic [...] alters the meaning and dynamic of places [...] a structural schizophrenia between two spatial logics [...] threatens to break down communication channels [...] a horizon of networked, ahistorical space of flows, aiming at imposing its logic over scattered, segmented places [...] Unless cultural and physical bridges are deliberately built between these two forms of space, we may be heading toward life in parallel universes whose times cannot meet." The logic of space becomes the cause, not the consequence, of social change. Just how do you build a "physical" bridge to a space of flows? An interesting conceptualization, with which Castells does not play; perhaps just an errant use of words. In any case, the insight has moved from a potentially striking and politically meaningful one into a play of metaphors, in which it is the "logic of space" that needs to be dealt with, not the relations among people using space. It hardly helps to get a grip on industrial relations in a global age to be told that "the very notion of industrial location [has been transformed] from factory sites to manufacturing flows [...] [by] the logic of information technology manufacturing [and] the new spatial logic." Playing with Time As with the treatment of space, this is insightful and provocative in Castells's handling, but depoliticized; he fails to pursue his real insight to its logical conclusion. He points out, and illustrates, the differences in the "time-boundedness" of different actors and activities. To some extent the differentiation parallels longstanding Marxist and classical economists' distinction between those paid hourly wages and those on longer-term salary bases or making profit without regard to time spent, a distinction that then feeds into definitions of class and class relations. Castells deepens the differentiation: it is not just between those paid hourly and those paid in other ways, but between those for whom time itself is an important factor in determining the way their lives are lived and those independent of it, living in "timeless time." Time is thus a constraint on some much more than on others; it "means" different things to different people. Fine. But to different classes? No, the analysis does not go in that direction; it rather plays with the catchy phrase "timeless time" as characteristic of a type of person and activity, jetsetters, instant communicators and instant manipulators of capital, and instant and constant (time-independent) exercise of control. The truth is that some control the time of others but are free to determine their own time, while the time of others is controlled despite their will. Just as with "space of flows," the metaphor reflects a real truth, but the emphasis on the metaphor conceals the very real class differentiation it in fact only reflects. "Selected functions and individuals" do not "transcend time"; they simply have the power to control their own use of time, and that of others. The Autonomy of the State This is a complex subject. The intellectual and political tradition from which Castells comes had a central Depoliticizing G l o b a l i z a t i o n concern with the role of the state. Marx's classic formulation of the state as the "executive committee of the ruling class" was widely seen not as wrong but as incomplete. To explain contemporary developments, Castells's close friend Nicos Poulantzas produced a complex analysis of the subject that was at the heart of the intellectual ferment in which Castells first worked. But Castells opens his chapter on the state with a repudiation of Poulantzas's description as no longer applicable. Little of the earlier rich discussion survives, except as an echo. Instead, the state becomes an actor: "the state's effort to restore legitimacy," "the state's attempt to reassert its power." And there are sweeping statements such as "the nation-state [...] seems to be losing its power, although, and this is essential, not its influence." Or elsewhere: "the state does not disappear. It transforms itself. This transformation is induced not only by globalization, but by structural changes in the work process, and in the relationship between knowledge and power." There are outside pressures, but the state itself acts to transform itself. What does that mean? Castells never returns to the formulation, but at the end of the chapter says that "in the 1990s, nation-states have been transformed from sovereign subjects into strategic actors." It is a muddled discussion. One possible interpretation might be that the nation-state remains important in the development of technology and in the support of "its" multinationals. Indeed, Castells emphasizes both points in various contexts in all three volumes. But why is that not a continuing source of power? The "nation-state" is used as a synonym for "state" in the global era, but the distinction between nation and state is never explored in the analysis; the capacity of the nationstate "is decisively undermined by globalization" but not by any specific actions of any specific actors, even though as a result multinationals can operate freely disregarding national borders. The nation-state has a "commitment to provide social benefits," although why that commitment should exist is not clear. There is a "destabilization of national states" through the globalization of crime and a "crisis of legitimacy" that is equally applicable to the Mexican and the United States state, although both countries seem remarkably stable in almost every regard. Such an interpretation simply avoids the question of what the state is. Throwaway lines like "states are the expression of societies, not of economies" do nothing to help. Furthermore, Castells describes the state's activities as if it was or had been an independent, autonomous actor - precisely the conception that has been so systematically questioned in critical sociology over the last century and more. Yet there is also, in passing, the comment that "each nation-state continues to act on behalf of its own interests, or [sic] of the interests of constituencies it values most." That latter comment might be the beginning of a discussion of where power in and over states actually lies, a discussion opening up the political questions that are so little regarded in the books. But it is not a comment that is pursued. And its very formulation is already misleading: the question is posed as who "the state" autonomously values, the state as actor, the constituency as passive beneficiary, rather than as what active "constituencies" control or put pressure on the state. Remarkably, little of the current discussion about the state "losing control" ever specifies who is winning control. And so we end with what appears a most ambiguous comment in the post-Seattle world: "the International Monetary Fund experts do not act under the guidance of governments [...] but as self-righteous surgeons skillfully removing the remnants of political controls over market forces." Of course, the International Monetary Fund and its related international bodies are deeply concerned with regulating, using the political power of governments and international transactions, and are critically dependent on governments for all of their activities - and particularly the one most powerful government in a one-superpower world. And in so doing they hardly act as independent experts or surgeons but are directly serving identifiable and very specific interests. Their actions are the subject of heated political discussion in countries around the world. Yet any discussion of those politics, however, is avoided. [...] 
Purchase answer to see full attachment