Erik Olin Wright | Compass Points: Towards a Socialist Alternative
Originally published in NLR 41, Sept-Oct 2006
Throughout most of the twentieth century, socialism constituted the central ideological matrix for thinking about alternatives to capitalism and giving direction to anti-capitalist struggles, even where the establishment of a socialist order was not an immediate political goal. If the particular institutional arrangements historically associated with socialism are now seen as incapable of delivering on their promises, many of the traditional socialist criticisms of capitalism seem more appropriate than ever: inequality, economic polarization and job insecurity are worsening; giant corporations dominate the media and cultural production; politics is increasingly run by big money and unresponsive to those without it. The need for a vibrant alternative to capitalism is as great as ever.
My aim here is to propose a way of thinking about a socialist alternative to capitalism that begins from the observation that both social democracy and socialism contain the word ‘social’.  This term is generally invoked to suggest a commitment to the broad welfare of society, rather than the narrow interests of particular elites. In more radical versions, ‘social ownership’ is contrasted with ‘private ownership’; but in practice this has usually been collapsed into state ownership, and the term ‘social’ ends up doing relatively little analytical work. I will argue that the ‘social’ can identify a cluster of principles and visions which differentiate socialism both from the capitalist project and from what can be called a purely statist response to capitalism. These principles revolve around what I will call ‘social empowerment’. In Part One, the problem of rethinking socialism will be located within a broader agenda of emancipatory social theory. Part Two presents a synoptic critique of capitalism, which identifies the problems for which socialism purports to be a solution. Part Three explores the general problem of elaborating credible institutional alternatives to existing structures of power and privilege. Here I will elaborate the idea of social empowerment, and explain what a socialism based upon it would mean. Part Four will then propose a map of pathways to social empowerment which embody the principles of a ‘social’ socialism. Part Five concludes with a discussion of the problem of transformation.
I. TASKS OF EMANCIPATORY SOCIAL SCIENCE
Emancipatory social science, in its broadest terms, seeks to generate knowledge relevant to the collective project of challenging human oppression and creating the conditions in which people can live flourishing lives. To call it a social science, rather than social criticism or philosophy, is to recognize the importance for this task of systematic scientific knowledge about how the world works. To call it emancipatory is to identify its central moral purpose—the elimination of oppression, and the creation of conditions for human flourishing. And to call it social implies a belief that emancipation depends upon the transformation of the social world, not just the inner self. To fulfil its mission, any emancipatory social science faces three basic tasks: first, to elaborate a systematic diagnosis and critique of the world as it exists; second, to envision viable alternatives; and third, to understand the obstacles, possibilities and dilemmas of transformation. In different historical moments one or another of these may be more pressing than others, but all are necessary for a comprehensive emancipatory theory.
Diagnosis and critique
The starting point for an emancipatory social science is not simply to show that there is suffering and inequality in the world, but to demonstrate that the explanation for these ills lies in the specific properties of existing institutions and social structures, and to identify the ways in which they systematically cause harm to people. The first task, therefore, is the diagnosis and critique of the causal processes that generate these harms.
This is often the most systematic and developed aspect of emancipatory social science. In the case of feminism, for example, a great deal of writing centres on diagnosis of the ways in which existing social relations and institutions generate various forms of women’s oppression. The focal point of such research is to show that gender inequalities are not the result of ‘nature’, but are the product of social processes. Studies of labour markets have emphasized such things as sex-segregation of work, evaluation systems which denigrate culturally defined feminine traits, discrimination in promotion, institutional arrangements that put working mothers at a disadvantage. Feminist studies of culture have demonstrated the ways in which a wide range of practices in the media, education, literature and so on have traditionally reinforced gender identities and stereotypes. Feminist analyses of the state have examined the ways in which state structures and policies have systematically entrenched the subordination of women and various forms of gender inequality. A similar set of observations could be made about empirical research inspired by labour-movement traditions, by theories of racial oppression and by radical environmentalism.
Diagnosis and critique are closely connected to questions of social justice and normative theory. To describe a social arrangement as generating ‘harms’ is to infuse the analysis with a moral judgement. Behind every emancipatory theory, there is thus an implicit theory of justice: a conception of what conditions would have to be met before the institutions of a society could be deemed just. A full exploration of the normative theory that underlies the critique of capitalism is beyond the scope of this paper; but put briefly, the analysis which follows is animated by what may be called a radical democratic egalitarian understanding of justice. This rests on two broad normative claims, one concerning the conditions for social justice, the other those for political justice:
- Social justice: in a just society, all people would have broadly equal access to the necessary material and social means to live flourishing lives.
- Political justice: in a politically just society, people should be equally empowered to contribute to the collective control of the conditions and decisions which affect their common fate—a principle of both political equality and collective democratic empowerment.
Taken together, these two claims call for a society that deepens the quality of democracy and enlarges its scope of action, under conditions of radical social and material equality. The problem, of course, is to show how these principles could be put into practice.
The second task of emancipatory social science is to develop a coherent, credible theory of alternatives to existing institutions and social structures that would eliminate, or at least significantly reduce, the harms they generate. Such alternatives can be elaborated and evaluated by three different criteria: desirability, viability and achievability. These are nested in a kind of hierarchy: not all desirable alternatives are viable, and not all viable alternatives are achievable.
The exploration of desirable alternatives, without the constraints of viability or achievability, is the domain of utopian social theory and much normative political philosophy. Typically such discussions are institutionally very thin, the emphasis being on the enunciation of abstract principles rather than actual institutional designs. Thus, for example, the Marxist aphorism describing communism as a classless society governed by the principle ‘to each according to his need, from each according to his ability’ is almost silent on the institutional arrangements which would make this principle operative. Liberal theories of justice similarly elaborate the principles that should be embodied in the institutions of a just society without systematically exploring whether sustainable, robust structures could actually be designed to carry out those principles in the pure form in which they are expressed.  Though discussions of this kind may contribute much to clarifying our values and strengthening our moral commitment to the business of social change, they do little to inform the practical task of institution-building, or add credibility to challenges to existing institutions.
The study of viable alternatives, by contrast, asks of proposals for transforming existing social structures whether, if implemented, they would actually generate in a sustained manner the emancipatory consequences that motivated their proposal. Perhaps the best known example of this is central planning, the classic form used to implement socialist principles. In lieu of the anarchy of the market, socialists believed that the people’s lot would be improved by a rationally planned economy, implemented through the institutional design of a centralized comprehensive plan. But the ‘perverse’, unintended consequences of central planning subverted its intended goals, with the result that few people today believe it to be a viable emancipatory alternative to capitalism.
The viability of a specific institutional design, of course, may not be an all-or-nothing affair. It may crucially depend upon various kinds of side conditions. For example, a generous unconditional basic income may be viable in a country in which there is a strong, culturally rooted work ethic and sense of collective obligation, but not in a highly atomistic consumerist society. Or, a basic income could be viable in a society that had already developed over a long period of time a generous redistributive welfare state based on a patchwork of targeted programmes, but not in a society with a miserly, limited welfare state. Discussions of viability, therefore, tend also to include the contextual conditions of possibility for particular designs to work well.
The exploration of viable alternatives brackets the question of their practical achievability under existing social conditions. Some have questioned the value of discussing theoretically viable alternatives if they are not strategically achievable. The response to such sceptics would be that there are so many uncertainties and contingencies about what lies ahead that we cannot possibly know now what the limits of achievable alternatives will be in future. Given this uncertainty, there are two reasons why it is important to have clear-headed understandings of the range of viable alternatives. First, developing such understandings now makes it more likely that, if future conditions expand the boundaries of what is possible, social forces committed to emancipatory change will be in a position to formulate practical strategies for implementing an alternative. Second, the actual limits of what is achievable depend in part on beliefs about what sorts of alternatives are viable. This is a crucial sociological point: social limits of possibility are not independent of beliefs about limits. When a physicist argues that there is a limit to the maximum speed at which a thing can travel, this is meant as an objective, untransgressable constraint, operating independently of our beliefs about speed. In the social case, however, beliefs about limits systematically affect what is possible. Developing compelling accounts of viable alternatives, therefore, is one component of the process through which these limits can themselves be changed.
It is no easy matter to make a credible argument that ‘another world is possible’. People are born into societies that are always already made, whose rules they learn and internalize as they grow up. People are preoccupied with the daily tasks of making a living, and coping with life’s pains and pleasures. The idea that the social world could be deliberately changed for the better in some fundamental way strikes them as far-fetched—both because it is hard to envisage some dramatically better yet workable alternative, and because it is hard to imagine successfully challenging the structures of power and privilege in order to create it. Thus even if one accepts the diagnosis and critique of existing institutions, the most natural response is probably a fatalistic sense that not much could be done to really change things.
Such fatalism poses a serious problem for those committed to redressing the injustices of the existing social world. One strategy, of course, is simply not to worry too much about having scientifically credible scenarios for radical social change, but to try instead to create an inspiring vision of a desirable alternative, grounded in anger at the inequities of the world in which we live and infused with hope and passion about human possibilities. At times, such charismatic wishful thinking has been a powerful mobilizing force. But it is unlikely to form an adequate basis for transforming the world so as to produce a sustainable emancipatory alternative. History is filled with heroic victories over existing structures of oppression, followed by the tragic construction of new forms of domination and inequality. The second task of emancipatory social science, therefore, is to develop in as systematic a way as possible a scientifically grounded conception of viable alternative institutions.
Developing coherent theories of achievable alternatives is central to the practical work of strategies for social change. This is a difficult undertaking, not only because assessments of what is achievable are vulnerable to wishful thinking, but also because the future conditions which will affect the prospects of any long-term strategy are highly contingent. As in the case of viability, moreover, achievability does not pose a simple dichotomy: different projects of institutional transformation have different prospects for ever being implemented. The probability that any given viable alternative could at some future date be put into practice depends upon two kinds of process. First, upon the consciously pursued strategies and relative power of the social actors who support or oppose the alternative in question. Second, upon the trajectory over time of a wide range of social structural conditions which affect these strategies’ chances of success. This trajectory is itself partially the product of the cumulative unintended effects of human action, but it is also the result of the conscious strategies of actors to transform the conditions of their own actions. The achievability of an alternative thus depends upon the extent to which coherent, compelling strategies can be formulated which both help to create the conditions for implementing alternatives in the future, and have the potential to mobilize the necessary social forces to support that alternative when such conditions occur.
A theory of transformation
Developing an understanding of these issues is the objective of the third general task of emancipatory social science: the theory of transformation. We can think of emancipatory social science as an account of a journey from the present to a possible future: the critique of society tells us why we want to leave the world in which we live; the theory of alternatives tells us where we want to go; and the theory of transformation tells us how to get from here to there. This involves a number of difficult, interconnected problems: a theory of the mechanisms of social reproduction which sustain existing structures of power and privilege; a theory of the contradictions, limits and gaps in such systems, which can open up space for strategies of social transformation; a theory of the developmental dynamics of the system that will change the conditions for such strategies over time; and, crucially, a theory of the strategies of transformation themselves. I will return to the problem of transformative strategies in the concluding section of this paper. Our central concern in what follows, however, will be the second of the three core tasks identified above: the problem of elaborating viable emancipatory alternatives to capitalism. To set the stage for this discussion, it may be helpful first to sketch the basic elements of a critique of capitalism, laying out the harms that are caused by capitalist processes and which animate the search for an alternative.
II. THE CORE CRITIQUE OF CAPITALISM
Capitalism will be defined here as a form of social organization in which two fundamental properties predominate: first, a class structure characterized by private ownership of the means of production, whereby most people earn their living by selling their labour on a labour market; and second, economic coordination organized through decentralized market exchange. Capitalism is not simply a ‘free market economy’; it is a market economy with a particular form of class relations. Historically, this has proved the most powerful economic system ever developed in terms of generating technological change and a certain kind of economic growth. Nevertheless, serious criticisms can be levelled against capitalism which, if correct, constitute the basis for the search for an alternative. It is beyond the scope of this paper to provide the full evidence and analysis that lie behind them but, from a radical-egalitarian democratic standpoint, the following criticisms would appear especially salient:
- The dynamics of capitalist economic growth systematically generates both increasing concentrations of wealth and privilege and expanding pools of deprivation, marginalization and poverty, both nationally and globally.
- Since capitalism systematically denies the conditions for free human flourishing and development to large sections of the world population, even within the most advanced economies, universal emancipation is impossible under capitalist rule. ‘Freedom of choice’, claimed by its defenders as capitalism’s central moral virtue, can only be partial under capitalism, since the inequalities that it generates entail limitations to the ‘real freedom’—the effective capacity to act on their life plans, to implement the choices which matter to them—of so many. 
- Capitalism’s profit-driven dynamic towards increasing production and expanding markets—with its bias towards a consumption-oriented society and the creation of artificial consumer ‘needs’—has inordinate environmental costs. Profit-maximizing creates incentives for capitalist firms to dump waste into the environment, and leads to the underproduction of non-profit-making public goods (clean air, public health, general education). The failure of capitalist markets to integrate the long-term costs of non-renewable natural resources leads to their under-pricing and over-exploitation.
- Capitalism’s expansionist drive threatens to subject ever wider domains of human activity to market forces: the commodification of health, childcare, elder care, human reproduction and so on raises significant normative issues.
- Antagonisms of interest within capitalist class relations undermine the sense of collective destiny and mutual generosity. In an exploitative relation, the exploiting category has active interests in maintaining the vulnerability and deprivations of the exploited.
- Capitalism imposes important constraints on democracy; the private ownership of the means of production removes significant domains of economic activity from collective decision-making.
It is important to be clear about the character of these criticisms. The central claim in each is that the harms described are generated by mechanisms that are intrinsic to capitalism as such. This does not mean that in a capitalist society—a society with a capitalist economic structure—there is nothing that can be done to counteract these harms. But it does imply that in order for this to happen, non-capitalist mechanisms must be introduced to counteract the effects of capitalism itself. This leaves open the question of how far one can go in mitigating these harms without cumulatively introducing so many counter-capitalist mechanisms as to transform the capitalist character of the economic structure itself. This, as we shall see in Part Three is a central issue in the problem of envisioning alternatives to capitalism.
III. THINKING ABOUT ALTERNATIVES
Historically, the most influential approach to thinking about alternatives to capitalism is that developed by Karl Marx. His solution to the problem of specifying a credible alternative to capitalism, if ultimately unsatisfactory, was intellectually brilliant. Rather than develop a systematic theoretical model which could demonstrate the possibility of a viable emancipatory alternative, he proposed a theory of the long-term ‘impossibility’ of capitalism. His arguments are familiar: because of its inner dynamics and contradictions, capitalism destroys its own conditions of existence. This is a deterministic theory: in the long run capitalism will become an impossible social order, so some alternative will of necessity have to occur. The trick is then to make a credible case that a democratic egalitarian organization of economy and society is a plausible form for such an alternative. Here Marx’s theory is especially elegant, for the contradictions which propel capitalism along its trajectory of self-erosion also create the historical agent—the working class—with both an interest in a democratic egalitarian society and an increasing capacity to translate its interests into action. Given these elements, Marx’s actual theory of socialism could be deemed a pragmatist one, centring on the problem-solving capacity of creative solidaristic workers: as capitalism moves towards long-term, intensifying crisis and decline, the working class develops the collective political organization needed to seize state power, create a rupture with capitalism and experimentally construct a socialist alternative. In a sense, then, Marx combines a highly deterministic theory of the demise of capitalism—whose laws of motion will ultimately make it unsustainable—with a largely voluntaristic theory of the construction of the alternative.
That theory was an extraordinary intellectual achievement, animating radical social and political movements for over a century. However, in certain crucial respects it is flawed and cannot, I believe, serve as the basis for the ongoing egalitarian project of challenging capitalism. I will confine myself here to noting four central problems. First, the classical Marxist arguments for the pivotal thesis of the theory—that capitalism necessarily destroys itself and will therefore necessarily be replaced by some alternative—are unsatisfactory. This prediction depends on the claim that capitalism is not simply prone to periodic crises, but that there is a systematic tendency for crises to intensify over time—a claim for which there are questionable theoretical grounds.
Second, the classical Marxist prognosis of transformations in the capitalist class structure towards increasingly homogeneous proletarianization has proved inaccurate. While it is certainly true that the course of capitalist development has incorporated a growing proportion of the labour force into capitalist employment relations, at least in the developed capitalist world this has not resulted in a process of intensified proletarianization and class homogenization, but rather in increasingly complex class structures. Contradictory locations within class relations have proliferated, self-employment has steadily increased in most developed capitalist countries since the mid-1970s, significant proportions of the working class own some stock through ESOPs and pension funds, households have become more heterogeneous in class terms as more married women entered the labour force, and career trajectories introduce temporal uncertainty into class locations.  None of these forms of complexity in class relations mean that class is of declining importance in people’s lives, or that class structures are becoming less capitalist in any fundamental way; rather, they show the ways in which structural transformations of capitalism have undermined the thesis that the working class has an increasingly homogeneous relationship to capitalism.
Third, the collective class capacities of potential challengers to capitalism have not systematically strengthened with capitalist development. In part this is because of the heterogeneity of interests within the broadly defined working class, but it is also because of the robustness of various forms of class compromise which undermine the capacity for challenging the system. Finally, the theory of ruptural transformation is not a plausible basis for constructing a democratic egalitarian transcendence of capitalism. While there have been revolutionary challenges to capitalism, the historical examples of ruptural transformation have never been able to sustain an extended process of democratic experimentalist institution-building. The voluntarist theory of constructing alternative, emancipatory institutions depends upon the active, creative and empowered participation of ordinary people in a process of deliberation and decision-making. There have been brief episodes of such participation within attempts at the revolutionary transformation of capitalism, but they have always been short-lived and relatively isolated. It is, of course, a complex matter to diagnose the reasons for these failures, but it is likely that the concentrated forms of political power and organization needed to produce a successful revolutionary rupture with capitalist institutions are themselves incompatible with the participatory practices needed for democratic experimentalism. Revolutionary parties may be effective ‘organizational weapons’ to topple capitalist states in certain circumstances, but they appear to be extremely ineffective means for constructing a democratic egalitarian alternative. As a result, the empirical cases we have of ruptures with capitalism have resulted in state-bureaucratic forms of economic organization rather than anything approaching a democratic egalitarian alternative to capitalism.
An alternative approach to alternatives
The classical Marxist theory of alternatives to capitalism is deeply anchored in a deterministic theory of key properties of capitalism’s trajectory: by predicting the basic contours of its future, Marx hoped to contribute to the realization of an emancipatory alternative. In the absence of such a theory, the task of making a credible case that there is a viable emancipatory alternative to capitalism is more difficult. One strategy, of course, would be to try to develop a comprehensive blueprint of socialist institutions, demonstrate that these institutions would function effectively, and then elaborate a road map indicating the possible routes from the world as we know it to this destination. With a road map in hand, our main task would be devising the right kind of vehicle for making the trip.
No existing social theory is sufficiently powerful to even begin to construct such a comprehensive chart of possible social destinations beyond capitalism. It may well be that such a theory is impossible even in principle—social change is too complex and too deeply affected by contingent concatenations of causal processes to be represented in plan form. In any case, no map is available. And yet we want to leave the place where we are because of its harms and injustices. What is to be done?
Instead of the metaphor of a road map guiding us to a known destination, we could think of the project of emancipatory social change as more like a voyage of exploration. We leave the familiar world equipped with navigational devices that tell us the direction in which we are moving and how far from our point of departure we have travelled, but without a map laying out the entire route from origin to endpoint. This has perils, of course: we may encounter unforeseen obstacles which force us to move in a direction we had not planned; we may have to backtrack and try a new route. Perhaps with technologies we invent along the way we can create some artificial high ground and see somewhat into the distance. In the end, we may discover that there are absolute limits to how far we can go; but we can at least know if we are moving in the right direction.
This approach to thinking about emancipatory alternatives retains a strong normative vision of life beyond capitalism, while acknowledging the limits of our knowledge about the real possibilities of transcending the capitalist system. This is not to embrace the false certainty that there are untransgressable limits for constructing a democratic egalitarian alternative: the absence of solid scientific knowledge about the limits of possibility applies not only to the prospects for radical alternatives but also to the durability of capitalism. The key to embarking on such a journey of exploration is the usefulness of our navigational device. We need, then, to construct what might be called a socialist compass: the principles which tell us whether we are moving in the right direction.
Conceptions of socialism
Most discussions of socialism build the concept in terms of a binary contrast with capitalism. The standard strategy is to begin with a discussion of different ways of organizing production, and from this to define capitalism as an economic structure within which production is oriented towards profit maximization through exchange on the market, the means of production are privately owned, and workers do not own their means of production and thus must sell their labour power in order to obtain their livelihoods. Socialism is then defined in terms of the negation of one or more of these conditions. Since the pivot of the concept of capitalism is the private ownership of means of production, generally this has meant that socialism is understood as public ownership in one form or another, most typically through the institutional device of state ownership. Here I will elaborate an alternative approach to specifying the concept of socialism in which it is contrasted to two alternative forms of economic structure, not just one: capitalism and statism.
Capitalism, statism, and socialism can be thought of as alternative ways of organizing the power relations through which economic resources are allocated, controlled and used. As a first approximation we can define the power dimension of these concepts as follows:
- Capitalism is an economic structure within which the means of production are privately owned, and thus the allocation and use of resources for different social purposes is accomplished through the exercise of economic power. Investments and the control of production are the result of the exercise of economic power by owners of capital.
- Statism is an economic structure within which the means of production are owned by the state, and thus the allocation and use of resources for different social purposes is accomplished through the exercise of state power. State officials control the investment process and production through some form of state-administrative mechanism.
- Socialism, by contrast, is an economic structure within which the means of production are owned collectively by the entire society and thus the allocation and use of resources for different social purposes is accomplished through the exercise of what can be termed ‘social power’. Social power is rooted in the capacity to mobilize people for cooperative, voluntary collective actions of various sorts in civil society. This implies that civil society should not be viewed simply as an arena of activity, sociability and communication, but also of real power. Social power is contrasted to economic power, based on the ownership and control of economic resources, and state power, based on the control of rule-making and rule-enforcing capacity over a given territory. Democracy, in these terms, can be thought of as a specific way of linking social power and state power: in the ideal of democracy, state power is fully subordinated and accountable to social power. Democracy is thus, inherently, a deeply socialist principle. If ‘democracy’ is the label for the subordination of state power to social power, ‘socialism’ is the term for the subordination of economic power to social power. In socialism control over investment and production is organized through some mechanism of social empowerment.
This idea of a socialism rooted in social power is not the conventional way of understanding socialism. Indeed many people use the term ‘socialism’ to describe what I am here calling statism. This reconceptualization, however, does capture a central moral idea about socialism: it is an economy organized in such a way as to serve the needs and aspirations of ordinary people, not elites, and to do this the economy must in some way or another be controlled by ordinary people—that is, subordinated to social power.
It is important to be clear about the conceptual field being mapped here: these are all types of economic structures, but only in capitalism is it the case that economically based power plays the predominant role in determining the use of economic resources. In statism and socialism a form of power distinct from the economy itself plays the dominant role in allocating economic resources for alternative uses. It is still the case, of course, that in capitalism state power and social power exist, but they do not play a central role in the direct allocation and use of economic resources.
For each of these three ideal types one can imagine an extreme form, in which only one sort of power is involved in controlling economic resources. In these terms, totalitarianism can be viewed as a form of hyper-statism in which state power is not simply the primary form of power over economic processes, but in which economic power and associational power largely disappear. In a pure libertarian capitalism the state atrophies to a mere ‘night watchman’, serving only to enforce property rights, and commercial activities penetrate into all areas of civil society, commodifying everything. The exercise of economic power would almost fully explain the allocation and use of resources; citizens are atomized consumers who make individual choices in a market but exercise no collective power over the economy through association in civil society. Communism, as classically understood in Marxism, is a form of society in which the state has withered away and the economy is absorbed into civil society as the free, cooperative activity of associated individuals.
None of these extreme forms could exist as a stable, reproducible form of social organization. Totalitarianism never completely eliminated informal social networks as a basis for cooperative social interaction beyond the direct control of the state, and the practical functioning of economic institutions was never fully subordinated to centralized command-and-control planning. Capitalism would be an unreproducible and chaotic social order if the state played the minimalist role specified in the libertarian fantasy, but it would also, as Polanyi argued, function much more erratically if civil society was absorbed into the economy as a fully commodified and atomized arena of social life. Pure communism is also a utopian fantasy, since it is hard to imagine a complex society without some sort of authoritative means of making and enforcing rules (a ‘state’). Feasible, sustainable forms of large-scale social organization, therefore, always involve some kind of reciprocal relations among these three forms of power.
Within this general conceptualization, capitalism, statism and socialism should be thought of not simply as discrete ideal types but also as variables. The more the decisions made by actors exercising economic power based on private ownership determine the allocation and use of productive resources, the more capitalist the economic structure. The more that power exercised through the state determines the allocation and use of resources, the more the society is statist. And the more power rooted in civil society determines such allocations and use, the more the society is socialist. There are thus all sorts of complex mixed cases and hybrids—in which, for example, a society is capitalist in certain respects and statist or socialist in others.
The idea of economic structures being hybrids of different power relations is fundamental to the idea of transforming these structures. All existing capitalist societies contain significant elements of statism, since states everywhere allocate part of the social surplus for investment in areas such as public infrastructure, defence and education. Furthermore, in all capitalist societies the state removes certain powers from the private owners of the means of production, for example when capitalist states impose health and safety regulations on firms. State power, rather than economic power, controls those specific aspects of production, and in these ways that aspect of ownership has been transferred to the state. Capitalist societies also always contain at least some socialist elements, if only through the indirect influence of collective actors in civil society on the allocation of economic resources through their efforts to sway the state (on legislation) and capitalist corporations (on pay and conditions). The use of the simple, unmodified expression ‘capitalism’ to describe an empirical case is thus a shorthand for something like ‘an economic structure within which capitalism is the predominant way of organizing economic activity’.
IV. PATHWAYS TO SOCIAL EMPOWERMENT
To recapitulate the conceptual proposal: socialism can be contrasted to capitalism and statism in terms of the principal form of power that shapes economic activity—the production and distribution of goods and services. Specifically, the greater the degree and forms of social empowerment over use and control of economic resources and activities, the more we can describe an economy as socialist. What does this actually mean in terms of institutional designs? For capitalism and statism, because of the rich examples of historically existing societies, we have a reasonably clear idea of the institutional arrangements which make these forms of economic structure possible. But what about socialism? What sorts of institutional designs would enable power rooted in voluntary association within civil society to control the production and distribution of goods and services? What does it mean to move in the direction of a society within which social empowerment is the central organizing principle of the economy? What does it mean institutionally to say that the means of production are collectively owned by everyone in a society but not by the state?
Our task here is not so much to propose blueprints for the realization of social empowerment over economic activity, but rather to elaborate a set of principles that would tell us when we are moving in the right direction. This is the task of a socialist compass. Such a compass will need to indicate three main directions, anchored in the forms of power we have been discussing:
- social empowerment over the way state power affects economic activity;
- social empowerment over the way economic power shapes economic activity;
- social empowerment directly over economic activity.
These three directions of social empowerment yield an array of pathways through which social power can be translated into power over the allocation of resources and control of production and distribution, as illustrated in Figure 1. 
Five pathways are especially important. Within each of these pathways we can think of an array of specific institutional proposals which would move us in the direction of greater social empowerment. I call the formulation of such institutional proposals ‘envisioning real utopias’: utopias because they embody emancipatory ideals; real because they attempt to formulate viable institutional designs. In what follows we will briefly discuss the character of each of these five pathways to social empowerment.
In traditional socialist theory, the essential route by which popular power—power rooted in the associational activity of civil society—was translated into control over production and distribution was through the state. It is for this reason that those visions can be described as models of statist socialism. The basic idea ran as follows: political parties are associations formed in civil society with the goal of influencing states. Their members join in pursuit of certain objectives, and their power depends upon their capacities to mobilize for collective actions of various sorts. If a socialist party was deeply embedded in working-class social networks and communities, and democratically accountable through an open political process by means of which it represented the working class (or some broader coalition), and if that party then controlled the state, and the state the economy, one could argue that an empowered civil society controlled the economic system of production and distribution. This classic model of statist socialism is diagrammed in Figure 2. Here, economic power as such is marginalized: it is not by virtue of actors’ direct ownership of assets that they have power to organize production, but rather through their collective political organization in civil society and their exercise of state power.
Statist socialism of this sort was at the heart of traditional Marxist ideas of revolutionary socialism. The scenario—at least on paper—was that the party would be organically connected to the working class and effectively accountable to associated workers, and thus its control over the state would be a mechanism for civil society (understood in class terms) to control the state. Furthermore, revolutionary socialism envisioned a radical reorganization of the institutions of the state and economy—through organizational forms of participatory councils, called ‘soviets’ in the case of revolutionary Russia—that would directly involve workers’ associations in the exercise of power in both the state and production. These councils, if fully empowered in democratic ways and rooted in an autonomous civil society, could be thought of as a mechanism for institutionalizing the ascendancy of associational power. Again, the party was seen as pivotal to this process, since it would provide the leadership (the ‘vanguard’) for such an associational translation of civil society into effective social power.
This is not, of course, how things turned out (see Figure 3). Whether because of inherent tendencies of revolutionary party organizations to concentrate power at the top, or because of the terrible constraints imposed by the historical circumstances of the Russian Revolution and its aftermath, whatever potential existed for the Bolshevik Party to be subordinated to an autonomous civil society was destroyed in the decade after 1917. By the time the new Soviet state had fully consolidated power and launched its concerted efforts at transforming the economy, the Party had become a mechanism of state domination, a vehicle for penetrating civil society and controlling economic organizations. The Soviet Union, therefore, eventually became the archetype of authoritarian statism under the ideological banner of socialism, but not of socialism itself. While certain socialist elements may have remained in the hybrid character of this economic structure, its core principle of organization was statism, not social empowerment. Subsequent successful revolutionary socialist parties, for all their differences, followed a broadly similar path, creating various forms of statism, but never a socialism based on an empowered civil society.
Today, few socialists believe that comprehensive statist central planning is a viable structure for realizing socialist goals. Nevertheless, statist socialism remains a component of any likely process of social empowerment. The state will remain central to the provision of a wide range of public goods, from health to education to public transportation, and in spite of the record of central planning in the command economies, it could also be the case that efficient and democratic forms of central planning over certain kinds of goods production may be viable at some point in the future, under altered historical conditions. The crucial question for socialists, then, is the extent to which these aspects of state provision can be placed under the effective control of a democratically empowered civil society. In capitalist societies, typically, public goods provision by the state is only weakly subordinated to social power through the institutions of representative democracy. Because of the enormous influence of capitalist economic power on state policies, often such public goods are more geared to the requirements of capital accumulation than social needs. Deepening the democratic quality of the state is thus the pivotal problem that will have to be solved in order for direct state provision of goods and services to become a genuine pathway to social empowerment.
Many will be sceptical about the possibility of achieving this. The failure of command-and-control bureaucracies in both state-socialist and capitalist economies has fuelled calls for the privatization of state services, not for their democratization. Yet a range of innovative designs provide reason to believe that more energetically participatory forms are possible, especially at the local and regional level, and that these can enhance both the effectiveness of public goods provision and the accountability of democratic institutions.  In Brazil, the system of participatory budgeting developed during the 1990s in Porto Alegre involved large numbers of ordinary citizens and secondary associations in real decision-making over city budgets, and especially over state production of local public goods. While it lasted, the participatory budget contributed to a significant redirection of infrastructure investment by the local state towards the needs of the poor and working class.
Social democratic regulation
The second pathway for potential social empowerment centres on the ways in which the state constrains and regulates economic power (see Figure 4). Even in the present period of deregulation and the triumph of free-market ideologies, the state is deeply implicated in the regulation of production and distribution in ways that impinge on capitalist economic power. This includes a wide range of interventions: pollution control, workplace health and safety rules, product safety standards, providing skill credentials in labour markets, minimum wages and so on. All of these involve state power restricting certain powers of capital owners, and thereby affecting economic activities. To the extent that these forms of affirmative state intervention are themselves effectively subordinated to social power, this then becomes a pathway to social empowerment.
Statist regulation of capitalist economic power, however, need not imply significant social empowerment. Again, the issue here is the extent and depth to which the regulatory activities of the state are genuine expressions of democratic empowerment. In actual capitalist societies, much economic regulation is in fact more responsive to the needs and power of capital than to those of civil society. The result is a configuration more like Figure 5 than Figure 4: state power regulates capital, but in ways that are systematically responsive to the power of capital itself. The question, then, is the degree to which it is possible within capitalist society to democratize state regulatory processes in ways which undercut the power of capital and enhance social power. One way of doing this is through what is sometimes called ‘associative democracy.’
Associative democracy—the third pathway—encompasses a wide range of institutional devices through which collective associations in civil society directly participate in various kinds of governance activities, usually along with state agencies.  It can be interpreted as involving the joint effects of social power and state power on economic power (see Figure 6). The most familiar form of this is probably the tripartite neo-corporatist arrangements in some social democratic societies, in which organized labour, employers’ associations and the state meet together to bargain over economic regulations, especially those affecting the labour market and employment. Associative democracy could be extended to many other domains, for example watershed councils which bring together civic associations, environmental groups, developers and state agencies to regulate ecosystems; or health councils, through which medical associations, community organizations and public-health officials plan aspects of healthcare. To the extent that the associations involved are internally democratic and representative of interests in civil society, and the decision-making process in which they are engaged is open and deliberative, rather than heavily manipulated by elites and the state, then associative democracy can contribute to social empowerment.
Economic power is rooted in the private ownership and control over the allocation, organization and use of capital of various sorts. Secondary associations of civil society can, through a variety of mechanisms, directly affect the way such economic power is used (Figure 7). For example, unions often control large pension funds. These are generally governed by rules of fiduciary responsibility which severely limit the potential use of those funds for purposes other than providing secure pensions for the beneficiaries. But those rules could be changed, and unions could potentially exert power over corporations through the management of such funds. More ambitiously, Robin Blackburn has proposed a new kind of pension fund, financed by a share-levy on corporations, which would enable a broader array of secondary associations in civil society to exert significant influence on the patterns of capital accumulation. In Canada today, the union movement has created a venture capital fund, controlled by labour, to provide equity to start-up firms that satisfy certain social criteria.
Social capitalism is not limited to capital funds controlled by associations in civil society. Other proposals which could enhance the power of civil-society associations to constrain economic power centre on the collective involvement of workers in aspects of workplace management. The co-determination rules in Germany, which mandate worker representation on boards of directors of firms over a certain size, are a limited example of this. Proposals to replace shareholder councils with stakeholder councils for the control of corporate boards of directors would be a more radical version. Or consider the regulation of workplace health and safety, usually performed by a government regulatory agency which sends inspectors to workplaces to monitor compliance with rules; an alternative would be to empower workers’ councils within the workplace to monitor and enforce health and safety conditions. The latter is an example of enhancing social over economic power. Social movements putting consumer-oriented pressure on corporations would also be a form of civil society empowerment. This would include such things as the anti-sweatshop and labour-standards movements based on university campuses, and organized boycotts of corporations selling products that do not conform to some socially salient standard.
The social economy
The final route to empowering civil society implies the direct involvement of secondary associations in civil society in organizing various aspects of economic activity, not simply shaping the deployment of economic power (Figure 8). The ‘social economy’ constitutes an alternative way of directly organizing economic activity that is distinct from capitalist market production, state-organized production or household production. Its hallmark is production organized by collectivities directly to satisfy human needs, and not subject to the discipline of profit-maximization or state-technocratic rationality.  It includes such things as community organized healthcare clinics and daycare providers, but also bodies such as NGOs that facilitate fair trade in products from the global south by eliminating exploitative middlemen, and community land trusts that remove land from the market in order to create affordable housing.
In capitalist societies, the primary way that production in the social economy is financed is through charitable donations; this is one of the reasons why such activities are often organized by religious institutions. An alternative would be for the state, through its capacity to tax, to provide funding for a wide range of socially organized non-market production. This is already common in the performing arts: many arts organizations are run on a non-profit basis, are designed to satisfy a particular kind of human need, and receive significant state subsidies in order to be insulated from the pressures of the market. In Quebec there is an extensive system of home care for the elderly organized through producer co-ops, and childcare offered by parent-provider co-ops. The Provincial Government heavily subsidizes these co-ops and establishes rules that essentially block the entry of profit-oriented firms into the subsidized social economy sector. These co-operatives grew out of a movement for a social economy and are coordinated by a civil-society association, the Chantier de l’économie sociale.  The Canadian single-payer healthcare system also has an important element of social economy: the state funds virtually all healthcare and regulates standards, but it generally does not directly organize its provision, as in the British National Health Service. Rather, hospitals, clinics and medical practices are run by all sorts of entities in civil society, including community-based organizations. This creates a space within which a social economy in health care, operating on the basis of grassroots participatory co-operatives, could potentially play a larger role.
One of the major obstacles to a dramatic expansion of the social economy is the problem of providing an adequate standard of living for people who work within it. One way of overcoming this obstacle would be the implementation of an unconditional basic income. The latter is generally defended on the grounds of egalitarian principles of social justice. But it can also be seen as a strategy for transferring part of the social surplus from capital accumulation to what might be called social accumulation, by reducing the pressure on collective associations in the social economy to provide for the entire standard of living of producers. In this way social empowerment along the pathway of the social economy would be accelerated.
Problems of social power
These five pathways constitute the principal ways in which social power can theoretically be translated into effective control over economic production and distribution. To the extent that social empowerment along these pathways increases, then we are moving in the direction of socialism. There are, of course, good reasons to be sceptical as to how power rooted in civil society could be organized to control the allocation of resources for production and distribution. Two problems are especially troubling. First, a vibrant civil society is precisely one with a multitude of heterogeneous associations, networks and communities, built around different goals, based on different solidarities; this does not seem a promising basis for the kind of coherent power needed to control a complex economic system. Second, the voluntary associations that comprise civil society include many based on narrow exclusionary interests and the preservation of privilege: the KKK as well as the NAACP. Why should we believe that empowering such associations would contribute to ameliorating the harms of capitalism, let alone to human emancipation?
The first of these two objections is one of the reasons why the conception of socialism proposed here is not the same as anarchism. An anarchist vision of transcending capitalism imagines a world in which the voluntarily coordinated collective action of civil society can spontaneously achieve sufficient coherence to provide for social order and social reproduction without the need for a state. Socialism, by contrast, requires a state—and one with real power to create rules of the game and mechanisms of coordination without which the collective power of civil society would be unable to achieve the necessary integration to control either state or economy. Just as a capitalist economy requires a capitalist state to ensure the conditions for the sustainable exercise of capitalist economic power, and a statist economy requires a ‘statist state’ to ensure its minimal coherence, so any sustainable process of social empowerment over the economy would require a socialist state.
The second objection—that civil society contains many associations inconsistent with egalitarian emancipatory ideals—is more troubling, for it raises the spectre of a socialism rooted in exclusion and oppression. There is no guarantee that a society within which power rooted in civil society predominated would uphold radical, democratic, egalitarian ideals. This, however, is not a unique problem for socialism; it is a characteristic of democratic institutions in general. As conservatives often point out, inherent in democracy is the potential for the tyranny of the majority, and yet in practice liberal democracies have been fairly successful in creating institutions that protect both individual rights and the interests of minorities. A socialist democracy would face similar challenges: how to devise an institutional framework for democratic deepening and associational empowerment which would foster radical, democratic, egalitarian conceptions of emancipation. My contention here is not that a socialism of social empowerment will inevitably meet this challenge successfully, but that moving along the pathways of social empowerment will provide more favourable terrain on which to struggle for these ideals than either capitalism or statism.
The set of real utopian proposals along these five pathways of social empowerment do not constitute a comprehensive blueprint for a society beyond capitalism. Many other institutional transformations would be necessary for the construction of a robust, socialist economy fully embodying the ideal of collective ownership of the means of production, let alone for a society within which democratic egalitarian emancipatory ideals could be realized outside of economic structures as well. Nevertheless, if the full array of institutional proposals we have examined were implemented in a serious and thoroughgoing way, capitalism would be profoundly transformed. While some of the proposals, taken individually, might be considered only modest movements along a particular pathway of social empowerment, taken collectively they would constitute a fundamental transformation of capitalism’s class relations and the structures of power and privilege rooted in them. Capitalism might still remain a component in the hybrid configuration of power relations governing economic activity, but it would be a subordinated capitalism heavily constrained within limits set by the deepened democratization of both state and economy.
This inventory of real utopian designs is not achievable under existing social and political conditions. They constitute desirable and viable but not, as a package, currently achievable alternatives. Achievability depends upon the powers of social actors and the circumstances in which they act. Developing an understanding of these issues is the objective of the third general task of emancipatory social science: the theory of transformation.
The central question of a theory of transformation is this: given the obstacles and opportunities for emancipatory transformation generated by the process of social reproduction, the gaps in that process, and the uncertain future trajectory of social change, what sort of collective strategies will help us move in the direction of social emancipation? Struggles for democratic, egalitarian, emancipatory ideals have historically clustered around three basic modes of transformation through which new institutions of social empowerment might be built: ruptural, interstitial and symbiotic.
Ruptural transformations envision creating new institutions of social empowerment by a sharp break with existing forms and social structures. The core idea is that direct confrontation and political struggle will create a radical disjuncture in which existing institutions are destroyed and new ones built, within a short space of time. A revolutionary scenario for the transition to socialism is the iconic version of this: a decisive, encompassing victory of popular forces resulting in the rapid transformation of underlying economic structures. However, ruptural transformations are not confined to revolutions. They may involve clusters of institutions rather than the foundations of a social system; they may also be partial rather than total. The unifying idea is of sharp discontinuity and rapid change, rather than metamorphosis over an extended period of time.
Interstitial transformations seek to build new forms of social empowerment in the niches, spaces and margins of capitalist society, often where they do not seem to pose any immediate threat to dominant classes and elites. This is the strategy that is most deeply embedded in civil society and often falls below the radar of radical critics of capitalism. While interstitial strategies are at the centre of some anarchist approaches to social change and play a large practical role in the efforts of many community activists, revolutionary socialists have often disparaged such efforts, seeing them as palliative or merely symbolic, offering little prospect of serious challenge to the status quo. Yet, cumulatively, such developments can not only make a real difference in people’s lives, but potentially constitute a key component of enlarging the transformative scope for social empowerment in the society as a whole.
Symbiotic transformations involve strategies in which extending and deepening the institutional forms of popular social empowerment also solves certain practical problems faced by dominant classes and elites. The democratization of the capitalist state, for instance, was the result of concerted pressures and struggles from below which were initially seen as a serious threat to the stability of capitalist dominance. The increase in social empowerment was real, not illusory, but it also helped to solve problems in ways that served the interests of capitalists and other elites, contributing to the stability of capitalism. Symbiotic transformations thus have a contradictory character to them, often taking advantage of a tension between short- and long-term effects of institutional change: in the short term, symbiotic forms of social empowerment are in the interests of elites and dominant classes; in the long term they can shift the balance of power towards broader social empowerment.
These three modes of transformation suggest very different postures towards the politics of transformation. Ruptural transformation, at least in its more radical forms (‘Smash the state’), assumes that the core institutions of social reproduction cannot be effectively used for emancipatory purposes; they must be destroyed and replaced with something qualitatively new and different. Interstitial transformation (‘Ignore the state’) aims to get on with the business of building an alternative world inside the old from the bottom up. Perhaps there are moments when established institutions can be harnessed to facilitate this process, but interstitial transformation mostly sidesteps centres of power. Symbiotic transformation (‘Use the state’) looks for ways in which emancipatory changes can be embodied in the core institutions of social reproduction, especially the state. The hope is to forge new hybrid forms which have a ratchet-like character, moving us in the direction of enlarged scope for emancipatory social empowerment.
None of these strategies is unproblematic. None of them guarantees success. All of them contain risks and dilemmas. In different times and places, one or another may be the most effective, but typically none of them is sufficient by itself. It often happens that activists become deeply committed to one or another of these strategic visions, seeing them as universally valid. As a result, considerable energy is expended fighting against the rejected models. A long-term project with any prospects for success must grapple with the messy problem of combining these strategies, even if the combination inevitably means that struggles often operate at cross-purposes.
It is easy, at the beginning of the 21st century, to be pessimistic about the future prospects of a socialism of social empowerment. But it is important to remember that around the world many of these proposals are being tried. Experiments exist, new institutions are continually being built (and alas destroyed) in the interstices of capitalist societies, and from time to time political victories occur in which the state can be enlisted in the process of social innovation. New forms of social empowerment continually arise. We do not know what the limits of such partial and piecemeal experimentation and innovation are within capitalism: social empowerment may ultimately be restricted to the margins, or there may be much more room to manoeuvre. But what is certain is that we have not reached those limits yet.
Thinking systematically about emancipatory alternatives is one element in the process by which limits of the possible can expand. What for now remain only visions for viable change can potentially become coherent political projects. By embarking on the journey of social empowerment within capitalism we may reach a world of empowerment over it—and eventually, perhaps, a destiny beyond it.
Originally published in New Left Review.
 In conventional political terminology, ‘social democracy’ refers to a reformist project inspired by socialist ideals which accepts the constraints of accommodating to capitalism, whereas ‘socialism’ refers to a project of social transformation beyond capitalism. In practice, many socialist parties pursue strictly social-democratic agendas, and some left-wing social democrats remain firmly committed to a more anti-capitalist transformative vision. In the present context I will treat both forms as occupying a position within a broad spectrum of democratic egalitarian challenges to capitalism.
 The term ‘science’ has been taken to imply privileged access to truth by experts who are willing to impose their vision on ordinary people. But while claims to ‘science’ are sometimes used in this way, I see science as a deeply democratic principle, since it rejects all claims to absolute certainty and insists on open, undominated dialogue as the basis for correcting errors and advancing knowledge.
 Moral philosophers generally argue that ought implies can—that there is no moral imperative to do the impossible—and thus, at least implicitly, that arguments about what would constitute a desirable alternative to the present world require that viable institutions could in theory be constructed to actualize those principles. In practice, however, very little attention is given to these issues in political philosophy. John Rawls, for example, argues that his ‘liberty principle’ is lexically prior to his ‘difference principle’ without ever asking if this is possible in real institutions.
 See respectively Milton and Rose Friedman, Free to Choose, San Diego 1990; Philippe Van Parijs, Real Freedom forAll, Oxford 1995.
 For an extended discussion see Wright, Class Counts: comparative studies in class analysis, Cambridge 1997.
 Arrows represent the predominance of one domain over another; thus the arrow from social to state power means that power rooted in civil society directly shapes the exercise of state power. This figure only illustrates the pathways through which social power operates; it is not meant to be a comprehensive map of all power relations over economic activity. A similar sort of map could be drawn for the pathways to statism, and the pathways of capitalist economic power.
 Archon Fung and Erik Olin Wright, Deepening Democracy: institutional innovations in empowered participatory governance, London 2003.
 For an extended discussion, see Joshua Cohen and Joel Rogers, Associations and Democracy, London 1995.
 In Figure 8 there is no arrow from the state to production and distribution since the state itself is not directly involved in these activities. The state is, of course, still important in setting up the legal parameters within which the social economy functions.
 An excellent discussion of innovative initiatives in the Quebec social economy can be found in Marguerite Mendell, ‘The Social Economy in Québec: Discourses and Strategies’, in Abigail Bakan and Eleanor MacDonald, eds, Critical Political Studies: Debates From the Left, Kingston 2002, pp. 319–43; and Nancy Neamtan, ‘The Social Economy: finding a way between the market and the state’, Policy Options, July/August 2005, pp. 71–6.