The antinomies of the modern imaginary and the double dialectic of control

Introduction

Cornelius Castoriadis was a revolutionary during, what Honneth and Hartmann (2012) term, the ‘social democratic’ period of modern capitalism. That is, the decades after the Second World War in which there was increasing state regulation of the market and the expansion of the welfare state. This phase of ‘organized modernity’ was institutionalized, Peter Wagner (1994) argues, in response to the preceding crisis of capitalist society and the processes of rationalization that resulted in developments like the consolidation of management, systems of mass production, and the growth of state planning. Despite the antagonism between state socialism and capitalist nation-states, there were important features of a shared mentality concerning the rationalization of complex industrial societies and the ‘conventions’ that shaped social practices and policies (Wagner, 1994, 2001). Given the modifications in capitalist societies over recent decades and the rupture with the social democratic mentality, the historical period of Castoriadis’s revolutionary political writings as a member of Socialism or Barbarism appears now as a unique phase and interlude in the history of capitalism. One that would be overwhelmed by the strains and conflicts that had only been rendered latent by state administration and the social rights of citizenship that served to limit the effects of the class structure. The reversals in the progressive transformation of capitalist society should be understood in terms of the practices and dynamics of ‘the dialectic of control’ (Browne, 2017a, 2017b).

The dialectic of control is a notion that derives primarily from the synthesis of Hegelian-Marxist conceptions of the practical constitution of society and the idea of the struggle for recognition, and Max Weber’s interpretation of the processes of rationalization. In particular, the dialectic of control represents an intersubjective theory of social conflict that is concerned with the balance between heteronomous dependency and autonomy in social relations (Giddens, 1979). The dialectic of control involves not only subordinated groups and individuals’ critique, resistance and struggles, but also the counter-resistance and reconfiguration of oppression by dominant groups and institutions. The latter does not necessarily take authoritarian and coercive forms; rather, it can take the more subtle and strategic forms of surveillance, incorporation and concessions that delimit and reorient progressive change. Castoriadis’s revolutionary critique of capitalism was tailored to the contradictions of bureaucratic capitalism. This critique disclosed significant dimensions of modernity as a constellation of social conflict and it provides one of the most important sketches of the dialectic of control during the phase of organized modernity. Castoriadis’s critical theory of the dialectic of control extends to his systematic analysis of the Soviet regime and state socialism, through the reflexive application of the critique of ideology to Marx’s theory and detailing how bureaucratic control constituted a mode of class exploitation (Browne, 2010).

Castoriadis’s critique of the capitalist labour process belongs to the debates of that period, particularly concerning the effects of hierarchical management on labour organization and over changes in the complexion of the capitalist form (Murphy, 2014). It is worth recalling that Marx (1971) considered that the contradiction between increasingly socialized production and the capitalist system of private appropriation significantly enabled the socialist form of organization. Castoriadis’s revolutionary critique amounts to an exploration of the dilemmas posed by this conception and a specification of its theoretical and practical limitations. Besides evaluating this critique, my analysis emphasizes how Castoriadis’s explication of the double dialectic of control resulted from the problem of the relationship of theory and political practice. Extending on Max Weber’s and Robert Michel’s theses about modern organization’s hierarchical and oligarchical tendencies, the double dialectic of control refers to the tendency of oppositional struggles to be subject to forms of internal control by the political parties, trade unions, and social movement organizations that are meant to give expression to these struggles (Weber, 1978; Michels, 1962). Castoriadis counterpoised the normative-political alternative of self-organizing forms of workers’ self-management and autogestion (Castoriadis, 1995, 1997b).

The problematique of the dialectic of control is continued in Castoriadis’s subsequent theory of the social imaginary through its concern with the imaginary of the project of autonomy and the interrogation of the relationship between the instituting imaginary and the instituted imaginary (Castoriadis, 1987, 1991; Singer, 1979, 1980). There is, however, a considerable shift in perspective. The theory of the imaginary institution of society’s concerns are broader than the relational dynamics of social interaction and their institutional mediations. The social imaginary is a ‘world-making’ web of meanings and a magma of significations (Castoriadis, 1987). It is constitutive of cultural horizons of understandings that mobilize and legitimize social practices. From this perspective, the dynamics of control pertain to the general conditions of social institution, reflecting the social-historical deployment of the imaginary. The dialectic of control is then more specifically instituted in and by the predominant modern imaginary, that is, the capitalist imaginary of the unlimited pseudo-rational domination and control of nature and society (Castoriadis, 1991, 1997a).

Castoriadis’s later theory may not amount to a complete break with dialectical thought, but it radically departs from the Hegelian-Marxist conception of contradiction and negation. My analysis shows that the problem of hierarchy is one of the major threads connecting these two phases of Castoriadis’s theory. It will be argued that certain limitations ensue, nevertheless, from the discontinuity between the two phases. The theory of the social imaginary enables a more expansive conception of the social conflicts, but Castoriadis’s later approach is less developed with respect to the practical structuration of conflicts and the metamorphoses of the dialectic of control. The fact that the ‘new spirit of capitalism’, according to Luc Boltanski and Eve Chiapello (2005), derived some of its justification from critiques of bureaucratic capitalism and notions of self-organization, albeit in distorted form, is indicative of the complications of the dialectic of control’s metamorphoses.

I

The historical processes of rationalization that result in the consolidation of bureaucracy were central to Max Weber’s conception of modernity. Weber believed that despite the evident detrimental effects of the formal rationality of bureaucracy, particularly the loss of meaning that ensues from the detachment of formal rationality from substantive value rationality and the loss of freedom with respect to individuals’ capacity for initiative and spontaneity, modern bureaucracy constitutes the most efficient, precise, and rational mode of organization (Weber, 1978). Weber contended that the socialist abolition of private property would extend bureaucratic domination in a negative manner, while the expansion of capitalist business organization would mean a corresponding growth in bureaucracy. Castoriadis broadly agreed with Weber’s prognosis but applied the Marxian concepts of alienation and contradiction to bureaucratic hierarchy. He emphasized the theme of ‘irrationality of rationality’ to a greater extent than Weber did himself, and this thesis formed an important part of his calling into question the institution of bureaucratic hierarchy. The background to Castoriadis’s critique of bureaucratic capitalism was the Trotskyist thesis of the undermining of socialism in the Soviet Union by the development of bureaucracy. Castoriadis similarly argued that the expansion of bureaucracy was incompatible with socialism but rejected the Trotskyist interpretation that ‘bureaucracy was only a parasitic and transitory stratum that maintained itself in power solely as a function of the unstable balance on the world scale between international capitalism, on the one side, and the Revolution on the other’ (Castoriadis, 1997b: 2).

It was a mistake, according to Castoriadis, to presume that bureaucracy would simply disappear with the overcoming of social conflicts of capitalism and that it was therefore wrong to attribute bureaucracy to the persistence of capitalism. Rather, the bureaucracy of state socialist societies was effectively a class with its own interests. It enacted a mode of exploitation that served the reproduction of this class and its domination over society. Bureaucratic class domination in socialist societies was anchored in state control. The bureaucracy’s control in state socialist societies encompassed the social relations of production and therefore constituted a dominant class (Castoriadis, 1997b). Although this system of property relations demarcated state socialist societies from the private ownership of capitalist societies, Castoriadis emphasized the continuities in the two systems’ broad cultural outlook. In fact, he argued that Marx imported into the workers’ movement significations that derived from the capitalist worldview, or what he later termed the capitalist imaginary (Castoriadis, 1987). Notably, Marx proposed that the rationalization of the forces of production and economic accumulation were preconditions of emancipation. These notions served to justify the state and Communist Party’s centralized control in the Soviet Union.

The early Castoriadis’s analysis of the dialectic of control cannot be divorced from the development of his extended critique of Marx’s theory. In fact, his critique is not only of particular details of Marx’s theory; it concerns the entire conception of theory to which Marx adhered. Further, Marx’s positing of the developmental preconditions of socialism exemplifies the problems of the hierarchical relationship of theory to practice that conditioned Marxist politics. Marx originally contested the hierarchical relationship that set theory above practice, but he would come to reassert it. The hierarchy of theory to practice constituted a specific instantiation of control in Marxist party apparatuses and a justification of this control. This form of control is operative in both the communist and social democratic parties inspired by Marxism, although there are significant differences between them in the manner and degree to which it conditions the operation of power in the political organization. Marx’s theory served as a means of political legitimation in communist parties and the double dialectic of control unfolded in relation to it (Castoriadis, 1987, 1997b).

For Castoriadis, the privileging of theory over practice is one of the reasons why bureaucratic hierarchy is an institutional expression of the organizational form implied by Marx’s theory, rather than the distortion of it. Marx’s theoretical project had originally understood itself to emanate from the practices and struggles of the workers’ movement; however, it incorporated a range of assumptions that undermine this intention and restore the hierarchy of theory over practice. These assumptions range from Marx’s claim to know more about the historical mission of the proletariat than ‘this or that proletarian’, through the formulation of historical laws of development, to the idea of a rigorous theory of the logic of the capitalist economy system (Castoriadis, 1987). Marx’s theory initiated a form of the double dialectic of control that displaced the creativity of the working class and served to subordinate practice to theory. Marx, to be sure, always emphasized the inventiveness of the workers’ movement, and the problem of the practical institutionalization of a theory or a broad cultural outlook is far from unique to Marxism. For instance, Weber’s notion of political-religious routinization of charisma points to the rudiments of bureaucratization in processes of institutionalization and social reproduction (Weber, 1978).

The fact that Marxism participates in the Weberian problematique of institutionalization is significant (Browne, 2017a). Castoriadis’s assessment is, however, too unequivocal. The original connection of his own theory and political practice to Marxism counts against this assessment and there are other examples that might qualify it. Heterodox currents of Marxism have appealed to Marx’s theory in order to contest communist parties’ institutionalized practices, especially during the brief periods of hopes in the possible internal reform of state socialist societies. It could be argued, nevertheless, that these considerations reinforce rather than qualify Castoriadis’s assessment: heterodoxy is by definition a subordinate tendency; it is orthodoxy that is the medium and outcome of institutionalization. The normative orientation of the notion of the dialectic of control clarifies the difference between the practices of an autonomous collective that institutes equal participation and open public deliberation, and the heteronomous practices that institute the hierarchical dependence of subordinates upon the dominant, particularly through the control of material resources, knowledge, and the means of administration.

The tendencies of Marxist theory and practice that were consolidated were exactly those that fitted best with the developing capitalist reality, particularly the concentration of the forces of production and the centralization of authority. These continuities led Castoriadis to concern himself with the wider sense of control that derives from the social imaginary’s capacity to frame understandings and to institute reality. The institution of the social imaginary is a more diffuse form of control, and this is evident in the limitations to the power of the instituted imaginary. Castoriadis distinguishes between the instituted social imaginary and the instituting radical social imaginary, drawing attention to the challenge that the latter poses to the control of the former (Castoriadis, 1987). The social imaginary is a type of infrastructural power that underpins and overarches the more explicit power to sanction, administer, and coerce.

The notion of the social imaginary further contrasts with the trajectory of Marx’s theory, insofar as the imaginary evades the logical-rational complexion of theory. Rather, the social imaginary is an instituting condition of rationality and is not reducible to the logical and rational (Castoriadis, 1987). The social imaginary perspective diverges from the Hegelian dialectical logic that Marx incorporated into his theoretical perspective and to which Castoriadis’s (1995: 197) early analysis of capitalism was indebted. The imaginary involves a different sense of dialectics to that of the ‘determinate negation’ that conditions Marxian critique of capitalism and its delineation of the logical possibilities for socialism present in the capitalist mode of production. The partly psychoanalytic inspiration of the notion of the imaginary elucidates how Marx combined some otherwise irreconcilable general contentions and how the horizon of capitalist significations overhung opposition to this system of production. Marx both imported capitalist significations into the workers’ movement and drew on the prior workers movement’s justified critique of capitalism, like those of exploitation and the contradictions of socialized production (Castoriadis, 1987, 1995). Marx’s theory, in effect, distorted the workers movement’s critique and constituted a division between theory and practice.

Castoriadis’s critique of bureaucratic hierarchy originally follows the Marxian perspective of the philosophy of praxis and explicates a critical dimension of the dialectic of control. Namely, it explicates the centrality of knowledge and information to control. In a bureaucratic hierarchy, Castoriadis (1995) argues, knowledge ascends to the top and does not descend downwards again and recirculate. Hierarchical control derives from this closure of information, and the possession of knowledge legitimizes managerial direction and oversight. This analysis is similar in its equating control with the monopolizing of knowledge and information to that of Michels’s arguments regarding the tendency towards oligarchy of party-political organizations (Michels, 1962). Castoriadis, however, contends that bureaucratic control is limited. The effective functioning of production depends on the transcendence in practice of the division between knowledge and doing, that is, the separation between managerial direction and workers’ execution of tasks (Castoriadis, 1995). The practical reasoning of workers continuously rectifies capitalist management’s limited knowledge. Further, production regularly depends on workers’ disobeying managerial orders and organizational rules; an important aspect of this disobedience is the recirculation of information in opposition to bureaucratic organizations’ formal rules (Castoriadis, 1995).

Given the significance of factors like labour’s commodification, the scale of industrial production, and the centrality of the regulation of time to capitalism, the social form of this dependency of capital on wage labour differs from the types of interdependency of premodern feudal and slave relationships. Nevertheless, there is an important underlying similarity of the dialectic of control in capitalist and pre-capitalist class societies, on the one hand, that is misunderstood and necessarily denied by dominant classes, and, on the other hand, there is a radical difference between these social relations:

Exploiting societies persist because those whom they exploit help them to survive. Slave-owning and feudal societies perpetuated themselves because ancient slaves and medieval serfs worked according to the norms set by the masters and lords of those societies. The proletariat enables capitalism to continue by acting against the system. Here we find the origin of the historical crisis of capitalism. (Castoriadis, 1997b: 54)

The dialectic of control is driven by the resistance of labour to this process of turning workers into ‘objects’ and workers’ ensuing demand for autonomy. The objectifying aim of bureaucratic management can never be completed. Its completion would undermine an organization’s operation entirely, because the total objectification of labouring subjects would be counter-productive from the standpoint of capitalist production. This limitation would be subsequently recognized by 20th-century management theory, particularly concerning the failings of preceding forms of management, like ‘scientific management’ and Fordism (Boltanski and Chiapello, 2005; Mathews, 1989). The new ideology of capitalist management partly develops in response to this critical assessment, but the change in managerial perspective was more of a retrospective response to contestation. From Castoriadis’s perspective, the transformation in managerial practices is the outcome of workers’ struggles and resistance, rather than modifications in managerial objectives. The institution of capitalist management expands as a result of the primary contradiction of capitalism.

The primary contradiction of capitalism is not located in the irrationality of the market, although the capitalist market system is contradictory in many respects, especially with respect to the misallocation of resources and the failure to meet needs. Rather, the core contradiction is located in the generative processes of production and the dynamics of the dialectic of control. Capitalism is dependent upon the creativity and agency of workers; however, its heteronomous organization excludes them from effective control. In other words, capitalism has to simultaneously include in order to ‘lay hold’ of the self-conscious activity of wage labourers in the process of production while excluding workers from effective decision-making and control (Castoriadis,1987). Bureaucratic capitalist hierarchy is a manifestation of this contradiction and subject to it. Capitalism is dependent on the motivations and capacities of workers that it is unable to generate itself (Castoriadis, 1995; Boltanski and Chiapello, 2005). The relationship of wage labour to capitalist production is that of ‘dependent participation’ (Touraine, 1977); participation is conditional on employment and the participatory practices are not autonomous. Worker autonomy would imply the direct democracy of self-management and ‘the elimination of all externally imposed norms, methods, and patterns of organization’ (Castoriadis, 1997b: 54).

In Castoriadis’s opinion, the demand for autonomy takes precedence over the struggle for distributive justice in the market. The latter follows the format of the struggles for autonomy and the level of control. A major component of Castoriadis’s critique of Marx’s political economy is of its alleged subordination of these struggles. Castoriadis argues that workers’ struggles are a key factor determining the rate of exploitation, and these struggles’ significance contradicts Marx’s notion of the extraction of value commensurate with the social necessary labour time and the presumptions that underpin it, such as that the rate of labour productivity follows from the control generated by the labour contract and the technical division of labour (Castoriadis, 1997b). Marx certainly recognizes, to be sure, workers’ struggles but downplays their significance in attempting to formulate a rigorous theory of political economy. Marx’s theory of value relies on the idea of the market as a mechanism generating equivalence, rather than explaining the market as an institution generated by the instituting practices of social labour (Castoriadis, 1984; Browne, 2016).

It could be argued that Castoriadis is correct that in Marx’s critique of political economy the dynamics of struggle appear less central than may be expected and that struggles relating to the labour process are sometimes compartmentalized, for instance, in relation to the length of the working day. Yet, Castoriadis is certainly wrong to the extent that he is denying the more abstract conception of the class struggle that is the intention of Marx’s critique of political economy. That is, Marx sought to disclose how class struggle is present in the more mediated, complex and invisible form of the general mechanisms of the capitalist system of production and that capitalist exploitation therefore has a systematic character. In other words, class struggle is present not just in overt instances of contestation but throughout the entire mode of capitalist production. For Castoriadis, Marx’s deluded commitment to rigorous theory results in struggle becoming simply another factor taken-into-account in determining the value of production. Marx’s theory does not account for the different intensity of the dialectic of control and diminishes the transformative potential of workers’ struggles (Castoriadis, 1997b). It is clear nonetheless that Marx had a more elaborated conception of the structuration of capitalist production and of the mediation of conflict through the constituents of an economy than Castoriadis’s representation of it. Even so, Castoriadis was in a position to perceive major economic developments after Marx, like mass consumption, technological change’s contradictory effects, and the enhanced state regulation of the market. These developments certainly challenged Marx’s historical projections, and their implications threw into relief some of the general framing assumptions of Marx’s conception of an economy (see Browne, 2016; Castoriadis, 1984).

There are a number of ways in which capitalism’s primary contradiction finds expression in modes of conflict, and the enactment of dialectics of control is conditioned by the latitude of agency that workers and subordinates possess. Castoriadis shows that it is a mistake to consider workers’ struggles to be simply reactive to capitalist exploitation and domination. Rather, workers’ practices of contestation entail dimensions of creativity and contingency (Castoriadis, 1987). Capitalism’s objectification of subjectivity generates workplace resistance, like sabotage and non-cooperation, that undermine production. These struggles can sometimes take the spontaneous form of ‘wild cat’ strikes, and this spontaneous action discloses aspects of the double dialectic of control. It is worth keeping in mind that the structural contexts of the dialectic of control are both enabling and constraining of agency (Giddens, 1984). This feature of the dialectic of control extends to the workers’ movement and its industrial organization. Namely, trade unions enable workers’ struggle and contestation, but they are constraining and limiting in practice with respect to struggles, for instance, regularly opposing ‘wild cat strikes’ and their implied autonomous agency.

The workers’ movement has contested capital and pursued the necessary objective of improved conditions, but the predominant strands of the union movement have come to function as a vehicle of integration. In this sense, they can be part of the dialectic of control and the mediator of conflicts. Castoriadis’s analysis implies that union organizations have engaged in a variant of the double dialectic of control that enables progressive reforms and blocks radical transformation and democratization. This tendency is, of course, a cause and consequence of the overall orientation of the workers’ movement. The broad integration of the working class into capitalism over the course of the 20th century has many sources, including the role of the state, the extension of citizenship rights, competing ideologies like nationalism, the rise of the ‘culture industry’, and the expansion of mass consumerism. The integration of the working class should not be principally attributed to unions in spite of the evident double dialectic of control.

Despite the historical tendency towards integration, the struggles of trade unions and the working class politically are central to Castoriadis’s distinction between the ‘fragmented bureaucratic capitalism’ of the West and the ‘total bureaucratic capitalism’ of the Soviet Union and state socialism (Castoriadis, 1997b). This assessment’s parallels with the Frankfurt School’s diagnosis during the post-war period of the consolidation of the ‘totally administered society’ are too self-evident to be almost worth noting. There is a similar conception of the market’s subordination and displacement by administration and how this ensues from the development of the capitalist form of organization, including the growth in the size of bureaucratic management, the concentration of power and resources, and monopolistic control (Adorno and Horkheirmer, 1979; Marcuse, 1964).

The Frankfurt School differed from Castoriadis in their stronger belief in the decline of opposition in the totally administered society; they considered the dialectic of class struggle to be in large part inoperative and that it no longer found direct expression in late capitalist society. In other words, the Frankfurt School held a stronger version of the thesis of the integration of the working class into capitalist society, and they were undoubtedly sceptical about the prospects of radical opposition, whereas Castoriadis’s distinction between ‘fragmented’ and ‘total’ bureaucratic capitalism implies greater opposition and struggle against bureaucratic capitalist domination. He considered the primary contradiction of the capitalist dialectic of control to be still operative, although the trajectory of his theory was similar to that of the Frankfurt School. He would move even further away from the Marxian paradigm of production and describe the later decades of the 20th century as an ‘era of generalized conformism’ (Castoriadis, 1997a) – a diagnosis of the times that is certainly consistent with those developed by the Frankfurt School

The primary contradiction of capitalism’s dialectic of control is present in these two variants of bureaucratic capitalism. The implications of the ‘total’ control in the ‘Soviet regime’ are precisely those counterproductive tendencies that are lessened in ‘fragmented bureaucratic capitalism’ by the configuring of interdependency in the labour market and workers’ greater capacity to resist. The hierarchical command of ‘total bureaucratic capitalism’ precludes alternative information and its mechanisms of control lead to the dissipation of motivations. The critical difference between ‘fragmented’ and ‘total’ bureaucratic capitalism is not so much the result of struggles in the process of production; rather, it is the broader overall contestation of bourgeois society and the opportunities it contains for class resistance:

Yet, the fate of the worker, and of the population in general, outside production, is not an additional characteristic but, rather, an essential component of the worker’s situation. Deprived of political, civic, and trade-union rights, forcibly enrolled in ‘unions’ that are mere appendages of the State, the Party, and the KGB, subject to permanent police control […] The situation has no analogy in the ‘classical’ capitalist countries, where very early on the working class was able to wrest some civic, political, and trade-union rights and contest explicitly and overtly the existing social order – doing so at the same time as it was constantly exerting decisive pressure upon the evolution of the system, this pressure ultimately being the principal factor in imposing limitations on the irrationality of the system. (Castoriadis, 1997b: 222)

Finally, the question of organization was an unavoidable dilemma for Socialism or Barbarism, and it was the major source of schisms. Unless one adhered to the myth of spontaneous agreement then the problem of control is intrinsic to the practices of a political group. One of the conundrums that confronted Socialism or Barbarism was that of how to sustain control without reproducing the dominance of bureaucratic organization and hierarchical leadership. In some respects, the problem of control was mapped onto the intellectual divisions that emerged with the evolution of Castoriadis’s critique of Marxist theory, such as through the different positions of the two editorial groups of Socialism or Barbarism’s publications – one consistent with the evolution of Castoriadis’s critique and the other adhering to revisionist Marxist interpretations (see Castoriadis, 1997b). Castoriadis rejected the position that a political organization could avoid the conundrum of authority through limiting itself to the coordination and dissemination of information.

Socialism or Barbarism confronted the problem of the autonomous participation of its members. This intention contradicted much of its members’ prior experience, especially that of the working class. Strong political commitments did not enable the majority of members to fully overcome the dispositions towards passivity and dependency that had been instilled by socialization and work experience (Castoriadis, 1997b). Castoriadis’s political practice and theoretical innovations of the time were unable to overcome the conundrum of leadership generating dependency. Despite these issues forming part of the reasons for terminating the group Socialism or Barbarism, the resonances of its project of autonomous self-organization are evident from the events of May 68 shortly after its cessation and the parallels with more recent instances of democratic self-organization by anti-austerity protests (Browne, 2017a; Browne and Susen, 2014).

II

Castoriadis provides the notion of the dialectic of control with strong normative and political grounding through analyses of how modern revolutionary initiatives instantiated forms of self-governing collectives. Initiatives like the original Soviets, workers’ self-management, and public assemblies represented attempts to ‘self-organize’ through equal participation, and they opposed the divisions that characterize traditional political forms, like the principle of the delegation of authority and subordination to an external sovereign power (Castoriadis, 1995). Instead, these initiatives sought to exercise self-control and to empower individuals’ creativity. In Castoriadis’s opinion, democratic self-organization was commensurate with forms of collective ownership, as opposed to existing socialism’s bureaucratic hierarchy. Unlike the double dialectic of control of heteronomous political organizations, radical democratic initiatives emerge from individuals’ practices rather than imposed structures. It is too linear to claim, as Honneth (1995) suggested, that Castoriadis’s theory of the social imaginary seeks to sustain the revolutionary project by means of an ontology. Yet, there are certainly affinities between the practices of self-generating autonomy and the ‘autopoetic’ quality of the social imaginary in its capacity for the self-creation of forms, figures, representations, and affects (see Castoriadis, 1997b: 371; 1987).

These instances of social self-organization sought to overcome the mediations that are the condition and outcome of dialectics of control and the regulation of them. In particular, their efforts to transcend hierarchy, heteronomy and inequality illuminated features of the dialectic of control – notably, how it develops by way of separations, like that between political representation and the represented, how counter-resistance operates as a tendency (and counter-tendency) in emancipatory movements as well as in opposition to them – and these exemplars of self-organization pose the dilemma of how to mobilize and sustain the revolution, given that it is entirely contingent on individual and social autonomy under these conditions. Significantly, this dilemma discloses the ideological overhang of established social imaginaries and how they undermine practices that are opposed to elements of the existing institution. The notion of the imaginary constitutes a shift in Castoriadis’s theory towards meaning and the elucidation of broader dimensions of control that inflects institutionalization. Castoriadis’s later works highlight this in various ways, notably, through emphasizing the indeterminacy of the imaginary, the ancient Greek appreciation of the chaotic character of the world, the sense in which Freud’s theory of subjectivity implies a lucid rather than closed and repressive relation to oneself and others, and the civilizational dimensions of the irrationality of rationality (Castoriadis, 1984, 1987, 1991, 2007). Further, Castoriadis throws into relief the gradations of control through defining democracy as a regime of self-limitation (Castoriadis, 1991, 1997a).

In Castoriadis’s later theory, the dialectic of control is set within the larger parameters of the social imaginary (Browne, 2017a). It is open to diverse forms of institutionalization, and the signification of control is associated with a distinctive orientation to the world. The background to its enactment is, on the one hand, the primary antinomy of the imaginary institution of society in general, and, on the other hand, the specific social-historical institution of the distinction between heteronomy and autonomy. The primary antinomy is between the instituted social imaginary and the radical instituting social imaginary. That is, the sense of order and stability that derive from the instituted imaginary are threatened by the radical creativity of the instituting imaginary and the indeterminacy of imaginary significations, since they refer to other meanings than logical or canonical definitions and involve additional associations and connotations. The dominant mode of control of heteronomous societies is the occlusion of the institution as deriving from the practices of social instituting, that is, that society is its own creation. Instead, the institution is attributed to some extra-social source, like the will of God, the metaphysics of the Great Chain of Being, or the teleology of reason (Castoriadis, 1987, 1991).

The project of autonomy derives from the recognition that society is an outcome of instituting practices, because this creates the possibility of being self-determining and self-legislating. It emerges originally with the ancient Greek institution of democracy and then re-emerges in Western Europe around the 12th century with the founding of self-governing city-states (Castoriadis, 1991). The project of autonomy puts the extant institution of society into question and involves the creation of a certain latitude of accepted contestation. Notably, there is the genesis of a public space for reasoned, though not rationalistic, discussion. The project of autonomy enables a rupture with the closure of instituted society and its enacting leads to the critical disclosure of the limitations of the justifications of heteronomy. It shows how these justifications are contingent on the extant institution and the denial of social instituting, such as through the attribution of language to the power of the divine (Castoriadis, 1997a).

As a germ, autonomy emerges when explicit and unlimited interrogation explodes on the scene – an interrogation that has bearing not on ‘facts’ but on the social imaginary significations and their possible grounding. This is a moment of creation, and it ushers in a new type of society and a new type of individual. I am speaking intentionally of a germ, for autonomy, social as well as individual, is a project. The rise of unlimited interrogation creates a new social-historical eidos: reflectiveness in the full sense of the term, or self-reflectiveness, as well as the individual and the institutions which embody it. (Castoriadis, 1991: 163)

The first order institution is that of the world-constitutive horizon of meaning of the social imaginary. The social imaginary creation of figures, forms, and representations instaurates a primordial form of control. This is the creation of a sense that the world is not an abyss or a void but is inherently meaningful and that to belong to the institution of society is to be engaged with this meaning (Castoriadis, 1997a). Social instituting is the work of society as an anonymous collective, or, in other words, a power of the entire social-historical field. It extends beyond the dimensions of explicit power, since it concerns the establishment of dimensions and relations, like that between the part and the whole, or more concretely, for instance, the explicit power of the state and the family. The social imaginary generates the ontological sense of control that is enabled by the confidence that words have meaning, that figures stand for something, that a sign and referent are connected, and that what happens is not without meaning and significance. The other side of this institution is that the imaginary creations are in some respects arbitrary and specific to a particular institution of society: the same sign need not refer to the same thing in another society, the idols of another tribe are meaningless, and a sense of meaningful purpose is something that is imposed in the vast majority of heteronomous societies, rather than open to genuine deliberation. The constitution of social institutions in terms of the instituted social imaginary, including particularly the institution or fabrication of the individual, represents a powerful infrastructural and invisible mode of general control.

The instituted social imaginary is controlling without necessarily involving the explicit exercise of control. It is not driven by the imperative of control in the way that might be suggested by the systems theory notion of the reduction of complexity (Luhmann, 1979, 1990). The latter could be described as the interpellation of the imaginary signification of control as the basis of order, organization, administration, and systems. The reduction of complexity is, or would be, for instance, a specific social-historical institution; one that exhibits the strong orientation to control of the modern capitalist imaginary. Types of explicit control converge in their institution with the social imaginary, but the imaginary’s constitution of forms of control through establishing orientations and generative matrixes of meaning has historically taken a variety of social-historical forms, such as through privileging the imaginary significations of the sacred and the pure (Durkheim, 1995; Dumont, 1972).

The infrastructural forms of control can be seen in various dimensions of the imaginary institution. In his major work, Castoriadis emphasizes the social-historical configuration of the institution and the imaginary creation of distinctive temporalities, the socialization of individuals that generates an attachment on their part to the institution that has fabricated them and involves the rather fraught developmental process of breaking down the pre-social radical imaginary of the individual psyche, and the linguistic and technical-instrumental proto-instituting that is principally achieved through identifying, classifying, organizing and techniques (Castoriadis, 1987). All societies involve some processes of classification and arranging into groups; however, the composition of sets, systems, and structures, as well as instrumental and technical achievements, are all oriented by the social imaginary and its projection of meaning. For this reason, the implication of Castoriadis’s theory of the social imaginary is that interpretations of regimes of control that are limited to the workings of anonymous systems, the disciplining effects of technical-instrumental arrangements, and the patterning of structures, are insufficient. It is necessary to elucidate the conjunctures between organizations, systems, power-regimes and so on with the deployment of the social imaginary. The failure to do so involves an unrecognized complicity with the modern capitalist imaginary. That is, the extent of the valourising of formal rationality, in Weber’s sense, is a distinctive feature of the capitalist imaginary (Weber, 1948). One that can serve to conceal this condition of the extensive and intensive consolidation of control.

The capitalist imaginary that develops originally in Europe is exceptional in the extent of its valourising the technical-instrumental dimensions of the institution of society, and the modern signification of control is strongly bound up with this world-orientation. It is a signification that forms part of a horizon of meaning that inflects the institutions of capitalism and modern state formation, although the intention of rational mastery can be applied to different objectives, and institutions possess different modalities of control. The capitalist imaginary is unique in its appeal to rationality, and its secular signification of the ‘unlimited’ is integral to modern notions of progress and capitalist expansion, such as the extension of rational accounting, the technical dominance of nature, and the impetus towards the ‘commodification of everything’ (Castoriadis, 1991; Wallerstein, 1983; Browne, 2016). The power of the capitalist imaginary derives most of all from its encompassing the primordial form of the institution. Further, the capitalist imaginary’s orientation and societal objectives can be seen to constitute the cultural horizon that shapes the ‘institutional clusters’, according to Anthony Giddens, of modernity: capitalism, industrialism, surveillance, and military violence (Giddens, 1985, 1990).

Giddens contends that each of these institutional domains involves dialectics of control: industrialism is contested by the ecological movement, the workers’ movement struggles against capitalist domination, surveillance is contested by civil rights movements, and peace movements oppose and seek to regulate military violence (Giddens, 1985, 1990). However, Giddens did not specify in detail the cultural sources of these institutions’ orientations to control, and it could be argued that Giddens treats this contestation as an adjunct to the historical expansion in each of these capacities for social domination, for instance, through his considering civil society to be a product of modern state formation (Giddens, 1985). It is clear then that Castoriadis considers that control develops as an overarching signification with the institution of the capitalist imaginary and that it extends to the development of the modern state (see Browne, 2005). Johann Arnason has effectively elucidated this articulation of the modern capitalist imaginary and some of the principal reasons for the transmission of its core signification across modern social institutions:

As a societal goal, it becomes a trans-functional point of reference for the functional determination and differentiation of institutions and activities; but it can also appear as a counterweight to the fragmentation of modern society and a temptation to overcome it by subjecting social life to complete rational control. In this capacity, the central signification of modern capitalism is one of the sources (not the only ingredient) of the totalitarian project. And on the level of more concrete value-orientations, the imaginary transfiguration of the effort to maximise control – more precisely: the imaginary extension, unification and deproblematization of new strategies for the conquest of nature – can be combined with – and given new direction to the utilitarian emphasis on the primacy of human needs as well as the expressivist ideal of a free development of human capacity. (Arnason, 1989: 330)

With the possible exception of his reflections on science, this juxtaposition meant that there is no extended exploration by Castoriadis of the interpenetration of the two modern imaginaries. Rather, he believes that the ‘totalizing’ aspirations of the capitalist imaginary make any institutionalized combination of the two imaginaries inherently unstable. As a consequence, the political-legal rationalization of the dialectic of control, such as through legislative recognition of the collective bargaining of labour and rights to arbitration, are from this standpoint the product of struggles connected to the project of autonomy and bureaucratically instituted delimitations of workers’ control. The capitalist imaginary’s remobilization, which has seen the reversal of many of the labour and social rights instituted during the ‘social democratic’ era of organized modernity, lends considerable justification to this position. Even so, this background to Castoriadis’s juxtaposition of the two imaginaries meant that the innovative revisions made by other theoretical diagnoses of the period were largely closed to him. Notably, it did not lead to a reassessment of the broadly understood potentials of liberal democracy and liberalism that were undertaken, respectively, by his former Socialism or Barbarism colleague Claude Lefort and by Jürgen Habermas in terms of revising his Frankfurt School Critical Theory predecessors’ social diagnoses (Lefort, 1988; Habermas, 1996). Lefort’s argument that the political imaginary of modern democracy involves the internalization of conflict is undoubtedly a significant alternative perspective on the dialectics of control’s rationalization.

In Castoriadis’s later work, there are a range of complementary explorations of dialectics of control. While only a few details of these analyses can be sketched here, their broad implications are shaped by a distinctive antinomy. On the one hand, the elucidation of the imaginary discloses the limits to the institution of control and how the instituted form and organization of society are shaped by the meanings created by the instituting imaginary. On the other hand, despite its extension into dimensions of social relations that have long been hierarchically structured, the project of autonomy is under considerable challenge in the contemporary period. There is the risk of its eclipse due to a lack of genuine and reflexive commitment to the project of autonomy and the consolidation of opposition to it, both within Western societies and outside of them (Castoriadis, 1997a). This assessment is exemplified by Castoriadis’s claim that Western capitalist societies are ‘liberal oligarchic’ regimes and that the decline of the project of autonomy is evident in the failure to properly recognize and contest this system of domination and control (Castoriadis, 1997a, 1997b).

The fragility of the project of autonomy is a constant theme of Castoriadis’s later writings, especially given that autonomy is founded upon itself. Even where this was most deeply recognized there emerged in ancient Athens the counter-discourse of Plato, which sought to subject democracy to a system of political control external to it (Castoriadis, 1991). Similarly, psychoanalysis, according to Castoriadis, aims to unlock aspects of repression in order to reveal the subject’s potential for autonomy. Yet, this is countered by the phantasy of omnipotence that Freud identified with the earliest ‘monadic core’ of the psyche (Castoriadis, 1987). The monadic core can readily cathect with significations of mastery. In this way, psychoanalysis involves the peculiar dialectic of control of working both with and against the radical imaginary of the psyche, as well as involving the interactional dynamic of transference and counter-transference between patient and analyst (Castoriadis, 1984, 1997a). The questions that psychoanalysis encounters are of broader salience to the project of autonomy. Castoriadis argues that individual autonomy entails the autonomy of all. The conundrum of how to counter resistance to autonomy that is not just based on liberal-individualistic misunderstandings of freedom but also individuals’ attachments to the instituted order and refusals of autonomy, including of its facilitation by others, finds parallels in the practices of psychoanalysis and pedagogy.

For Castoriadis, the integration of individuals into the institution that has fabricated them is the deepest form of control and the basis of the occlusion and denial of heteronomy. It is one of the reasons why the ‘asocial’ radical imaginary of the psyche is relevant to the project of autonomy. It can posit forms, significations and affective states that differ from those of the instituted society and which may in modified forms, to be sure, infuse the project of autonomy. The radical imaginary can generate in individuals a sense of possibility and an attachment to the signification of autonomy insofar as this has been enabled by the broad processes of socialization. This is particularly significant because social autonomy would never have been truly experienced in a heteronomous society and yet it is the project of autonomy that instantiates the dialectic of control. In any event, psychoanalysis shows that individuals’ cathecting of autonomy is equally contrary to the ‘primary narcissism’ of the deep structure of the psyche and that the transition from dependency to autonomy is a fraught and tension-ridden process; one that can never be entirely controlled by the individual because it is contingent on the other. The typical course of socialization is an individual’s acceptance of heteronomous societies’ instituted closure of meaning. Freud by and large, Castoriadis (1997a) observes, identified normal development with adaptation to the established institution of society.

For all of the insights into autonomy that Castoriadis derives from psychoanalysis, it is worth noting that the interactional dynamics of dialectics of control have been more fully explored by other traditions of psychoanalysis, especially those of object relations and the intersubjective theory of Jessica Benjamin (Benjamin, 1988; Winnicott, 1991). The major reason why is that Castoriadis is principally concerned with the basic disjuncture between the a-logical character of the individual psyche and the institution of society. It is an antithesis that the individual’s participation in society modifies but which nevertheless persists in terms of the unconscious and the indirect expression of its workings by the radical imaginary, such as in dreams. Even so, this perspective reveals his later work’s subordinating the dynamics of practical interaction to their framing preconditions, and this has substantial implications for Castoriadis’s conception of the configuration of society.

The tension between the two dominant modern imaginaries of the project of autonomy and the capitalist imaginary of rational mastery has shaped modern scientific thought. On the one hand, science develops from the questioning of instituted representations of the world and the extension of open inquiry. In this respect, it draws on aspects of the project of autonomy and contributes to its institution. On the other hand, the equation of thought with the intention of control has been a basic supposition of modern science and a source of its integrity with the capitalist imaginary. Descartes, Castoriadis notes, proposed that the aim of knowledge was to make man the ‘masters and possessors of Nature’ (Castoriadis, 1991: 184). The extension of this modern project of unlimited mastery has generated techno-science (Castoriadis, 1991: 272). Techno-science’s orientation to control has superseded the suppositions of scientific thought that owe to the project of autonomy (Castoriadis, 1991, 1997a). Indeed, techno-science undermines some of the contrary implications of historical innovations in scientific understanding. Innovations, like those in quantum physics and mathematics in the 20th century, led to appreciations of contingency and indeterminacy. In this way, science itself constituted a challenge to elements of the notion of rational control and Castoriadis claimed that his explication of the imaginary was consistent with these scientific innovations (Castoriadis, 1984, 1987). Techno-science, by contrast, renders the innovations in different scientific disciplines consistent with the modern imaginary of rational mastery.

Techno-science is a complex form of control that is applied to a contingent environment. It aspires to be self-regulating, such as through the application of ‘expert systems’ that automatically make decisions (Castoriadis, 1991: 273). The implication of the institution of techno-science is a kind of ‘meta-control’ that seeks to control through anticipating the dialectic of control. In particular, it deploys a higher level of reflexivity concerning processes of contestation and, in Castoriadis’s opinion, it extends the dehumanization that is present in the formal rationality of bureaucratic capitalism, hence his arguments in opposition to this dehumanizing for the recognition of human mortality and for an ethics appropriate to it (Castoriadis, 1991: 273–5; 1997a). There were undoubtedly speculative elements to the techno-science thesis at the time of Castoriadis’s critical assessment, but these features have been rapidly implemented and diffused without necessarily being recognized as such. The occlusion is partly due to techno-science applications’ appeal to discourses of individual autonomy, self-regulation, and efficiency for legitimacy, such as with respect to information technologies and medicine. The paradox is that what appears like the escalation of control by techno-science is in another way actually its reverse. By no means does this amount to techno-science’s subordination to the lucid deliberation of a democratic collective; rather, this reversal is a product of the very extension of its objectifying tendencies and impersonal quality:

As soon as rationality was seen to be perfectly ‘objectifiable’ (which quickly came to mean capable of being put in to the form of an algorithm and susceptible to impersonal mastery), ‘rational mastery’ came to imply, and in truth requires, impersonal mastery. But impersonal mastery, when extended to everything, obviously becomes the mastery of Nobody, of outis –and it becomes, thereby, complete non-mastery, impotence. (Castoriadis, 1991: 273)

III

Castoriadis’s theoretical and political programs constitute a distinctive and significant explication and interrogation of the dialectic of control. Castoriadis poses a challenge to other perspectives on the dialectic of control and the manner of this challenge alters in line with the modifications in his theory. As we have seen, his early work is principally concerned with the contradictions of capitalist production and labour relations. These revolutionary texts concluded, however, that developments contradicted the dialectical model of the proletariat taking control and that one could ‘no longer maintain the idea that the “proletariat” is “the” depository of the revolutionary project’ (Castoriadis, 1997b: 27). Bureaucratic capitalism’s objectification of subjectivity was now opposed by a variety of social groups in diverse spheres, particularly by students and women. His later work on the social imaginary and interrelated themes elucidates broad dimensions of control and the complications that bear upon it. Social imaginaries establish primordial forms of control and are threatened by the radical instituting imaginary; instituting is a process of social-historical creation and the imaginary is an indeterminate and never fully controllable ‘magma’ of significations, forms, and figures. From this perspective, the antinomies, tensions, and conflicts between the two dominant modern social imaginaries of the project of individual and social autonomy and the capitalist imaginary of the (pseudo-)rational unlimited domination and control of nature and society shape the specific institution of the dialectic of control in modernity.

The project of autonomy underpins the normative and emancipatory implications of the dialectic of control. This contention informs Castoriadis’s opposition to approaches that define resistance primarily as a counterpart of power. In Castoriadis’s opinion, this view veils the proper meaning of autonomy. These conceptions consider resistance to be caught within the matrix of power and are accordingly sceptical about emancipation. The best-known version of this scepticism is Foucault’s critique of the enhancement of control in modern disciplinary society and which was influenced by Bataille’s preceding theory of transgression being dependent on the interdiction which it violates (Foucault, 1978; Bataille, 1997). While these conceptions of resistance often have a genealogy that can be traced to the relational account of the dialectic of control in Hegel, there are other significant influences upon them, especially Nietzsche and structuralism. These latter sources contribute to their undermining the normative basis of the significations of autonomy and emancipation. For these approaches, the opposition to power is merely counter-power and therefore an extension of powers (Castoriadis, 1997a: 51). Castoriadis’s conception of the imaginary of the project of autonomy represents then a constructive alternative to notions of resistance that are delimited by instituted society. The project of autonomy entails the lucid and reflexive deliberation of a politics that opposes the heteronomous institution of society and that intends the autonomy of all members of society.

The notion of the social imaginary was developed to illuminate the persistence of hierarchy in social relations (Browne, 2019). Its insights into the occlusion of instituting is relevant to discerning the heteronomous aspects of intersubjective relations involving complex dialectics of control, like those of cooperation, mutuality, and psychoanalytic therapeutic practice. From the standpoint of the imaginary of the project of autonomy, these practices are not reducible in their normative intentions to power and control. The complexity of the dialectic of control in these intersubjective relations derives from their association with the significations of the project of autonomy and the intention of overcoming heteronomy. Yet, the reliance of any social practice upon the significations of instituted society means that there is always the potential for regression in relation to the project of autonomy. The project of autonomy entails higher levels of indeterminacy than the heteronomous institution’s prioritizing of the instituted over the instituting, hence Castoriadis’s repeated assertion that the project of autonomy implies a new relationship of instituting and instituted society (Castoriadis, 1991, 1997a, 1997b). It takes a major reorientation of the social imaginary for the controlling facets of social dependency to be overcome. However, autonomy is not simply the freedom to do whatever an individual and collective wants; rather, it is a regime of ‘self-limitation’. For this reason, an autonomous society entails a radically different mode of control to that of hierarchical social relations.

The contrast that Castoriadis drew between his traditional conception of autonomy as self-legislation and Francesco Varela’s term ‘biological autonomy’ provides some insight into democratic control through the description of its antonym. According to Castoriadis, Varela refers to a living being’s ‘operational, informational and cognitive closure’ and this formation should be defined instead as ‘self-constitution’ (Castoriadis, 1997b: 308). It fabricates an entirely ‘controlled system’ and operates like an automaton in the Aristotelian sense of something that is ‘self-moving’. Its projection to the social level would be an entirely ‘closed and rigid institution’, limited to the technical and mechanical principles of identity-ensembling logic, and driven by an imperative of self-preservation (Castoriadis, 1997b: 309). Although the aspiration of complete control is delusional, techno-science owes a great deal to this vision, particularly – as we can see today – through the intended applications of algorithmic predetermination, automated machine-learning, and encoded self-referential communication. By contrast, democratic control involves explicit and justified value commitments, the responsibility of individual and collective agency, an acceptance of the contingency of the world and human practice, an openness to creativity, and the replacement of external control by self-limitation. Of course, these stipulations can only be realized in a society that facilitates the sharing and participation of all members of society in its institutions of ‘explicit power’ and that is committed to social equality and autonomy (Castoriadis, 1991: 173).

Democratic control then involves lucid deliberations and public disagreements over the collective good, something Castoriadis believes can be facilitated by the institution of a ‘public-public’ sphere, following Athenian democracy, that is distinct from the private sphere and the representation of private interests in a public realm of social relations (Castoriadis,1997b: 407–11). ‘The first condition’, he argues, ‘for the existence of an autonomous society – of a democratic society – is that the public/public sphere become effectively public, become an ekklësia and not an object of private appropriation by particular groups’ (Castoriadis, 1997b: 407). Consequently, democratic control derives from the practical instantiating of the imaginary of autonomy, rather than any specific instituted form, like a constitution or bill of rights. In Castoriadis’s opinion, it is a mistake of modern political theory to presume that these apparatuses of explicit power are sufficient to prevent democracy’s degeneration, however valuable they may be in other respects. It is only the people in an autonomous society that can control themselves. Critics of Castoriadis regularly reject this position as inadequate with respect to the potential negative consequences of collective self-determination. For instance, Agnes Heller considers that it does not take into account the dangers of majoritarian and ‘totalitarian’ democracy (Heller and Feher, 1991). Lefort’s arguments about modern democracy involving the internalization of opposition and the separation of power and knowledge amounts to an extended response to Castoriadis’s position (Lefort, 1988).

The transition from the critique of bureaucratic capitalism to the elucidation of the social imaginary as the leading motif of Castoriadis’s work discloses how the notion of the dialectic of control has tended to be oriented towards power at the expense of culture. This orientation is certainly the case for Giddens’s structuration theory conceptualization, and the qualities attributed by Castoriadis to the imaginary shows how the signification of control involves some compression of culture (Giddens, 1979). The explicit valourising of control undoubtedly belongs to a specific social-historical form of the institution and, in this respect, it is a meta-category of Western modernity. One that animates the ‘second-order’ institutions and regularly distorts oppositions to heteronomy through the equation of control with hierarchy. Modern notions of reason and rationality incorporate the social imaginary signification of control, according to Castoriadis (1987), through commitments to applying the logic of identity and the ontological principle of determination. Modern reason’s orientation to rational mastery has led to limited conceptions of dialectics. The Hegelian-Marxist notion of dialectics is unable to encompass the magma of social imaginary significations (Castoriadis, 1987, 1997a). The imaginary’s elucidating requires additional senses of social creativity to that stimulated by contradictions and the Marxian conceptions of production, whether as a system or the expression of a subject (Murphy, 2012). In a similar vein, it could be argued that the notion of the dialectic of control, particularly as it develops from Hegel’s account of the struggle for recognition in the relationship of master and slave, is limited to an intersubjective perspective rather than the properly trans-subjective one required for elucidating the social imaginary (Adams, 2011; Hegel, 1977).

The notion of the imaginary enables the elucidation of the deeper suppositions of the perspective of the dialectic of control and the implications of its institution in modernity. At the same time, the notion of the dialectic of control provides a means of rectifying deficiencies in Castoriadis’s theoretical proposals and their practical sequels. The most important rectification concerns the main limitation of Castoriadis’s later theory compared to his earlier work. Despite its much more limited critique of the epistemological foundations of political economy compared to the later challenge that the notion of the social imaginary poses to the entire idea of an economy and the need to conform to its imperatives, Castoriadis’s earlier critique of bureaucratic capitalism contained a more detailed account of the practical processes of social reproduction and the structuration of social antagonisms. In his later work, the ongoing social struggles are broached in more general terms and the re-conceptualization of social practices is a never realized intention of his later theoretical program. This omission is of major significance because the dialectic of control is something that is contingent on its practical enactment. Castoriadis’s categorical statements about the contemporary decline of critique and social struggles obscure this practical enactment just as much as the omission of social practices from his extended theoretical approach to the social imaginary. The deficient conceptualization of the dialectic of control’s structuration is compounded, I argued, by the fact that his earlier critique of bureaucratic capitalism continued to condition his later interpretation of the capitalist imaginary and its juxtaposition to the imaginary of the project of autonomy.

Castoriadis describes the historical period commencing in the last quarter of the 20th century as one of social-historical regression. This is evident, he argues, in the waning of modernist culture’s creativity, the ‘generalized conformism’ that is a product of consumerism, the decline in social critique, and the dominance of a ‘liberal oligarchic’ regime that has stifled democracy (Castoriadis, 1991: 219–42; 1997a: 32–44). There has been a ‘counter-offensive’ against the economic regulation of capitalism and progressive changes that ensued from the struggles of the previous century, pursued particularly by the workers’ movement (Castoriadis, 2007). For this reason, Castoriadis’s later work does point in a rudimentary way to the unfolding structuration of the dialectic of control. Yet, these analyses are undoubtedly incomplete and the reasons for the waning of critique that expressed and informed opposition to capitalist social relations are more complicated than Castoriadis’s assessment. The unintended contribution of the originally progressive ideals of the critique of alienation and inauthenticity to the social regression ensuing from the resurgence of the capitalist imaginary is one of the central theses of Boltanski and Chiapello’s analysis of the new spirit of capitalism (Boltanski and Chiapello, 2005).

Boltanski and Chiapello (2005) highlight capitalist management’s appropriation of the significations of the critique of bureaucratic hierarchy that were developed by Socialism or Barbarism and other radical groups of the period. This appropriation and distortion were driven by the high levels of contestation over capitalism during the later phase of ‘organized capitalism’ and the need for ideological justifications that veiled the primary conflicts of capitalism. The new ideology’s justifications contributed to the genesis of the motivations and legitimations needed by the capitalist system (Boltanski and Chiapello, 2005). In one sense, this analysis does not pose any challenges to Castoriadis’s theory and it is even described by Boltanski and Chiapello (2005) as an updating of his early critique of capitalism’s contradictions. The notion of the social imaginary can readily account for the distorted institution of emancipatory ideals. Similarly, for Castoriadis, the ancient Greeks already well-understood human projects’ paradoxical outcomes. Nevertheless, the new spirit of capitalism thesis draws attention to the weaknesses of his later work’s lesser engagement with practices, since the dynamics of this change in capitalism have only been explicated by Castoriadis in general terms from the social imaginaries’ standpoint.

Similarly, Castoriadis’s theory incorporates many of the categories of, what Boltanski and Chiapello (2005) term, the artistic critique of capitalism, like imagination and creation. These categories were originally considered to be set against capitalism in a way that is more difficult to hold to today, because of the appropriation of them by the new capitalist ideology (see Browne, 2014b; Boltanski and Chiapello, 2005). In other words, it is necessary to revisit the tensions and contradictions that shape the dialectic of control and to deepen explanations of how capitalist management functions in the present through the ‘pseudo-autonomy’ of workers’ self-control. The current alienation of compelled but thwarted participation will undoubtedly generate novel forms of struggle and contestation (see Browne, 2017a). The notion of the imaginary enables the elucidation of the innovative features of struggles; however, Castoriadis’s over-integrated model of society has difficulty explaining the fractures and institutional modifications that ensue from the practical instantiating of the dialectic of control. This limitation is evident in his failure to explore the implications of his claim that the modern project of autonomy has generated the democratizing contesting of heteronomous social relations in domains beyond the realm of traditional political democracy (Browne, 2014a; Castoriadis, 1991). Its exploration would clarify the practices animated by the imaginary of autonomy and their outcomes, as well as the background to the capitalist imaginary’s remobilization and the relative solidity of its current institution.

In conclusion, although Castoriadis’s brilliant and pioneering conception of the social imaginary opens up an original perspective on the dialectic of control, his early work remains one of the most significant extensions of Marxist interpretations of its capitalist dynamics. These dynamics, Castoriadis argued, resulted in the institution of bureaucratic capitalism, the distortion of emancipatory projects by organizations intended to realize them, and the dominance of the bureaucratic class in the Soviet regime. The early Castoriadis addressed the problems of the dialectic of control at the level of theory and practice. Despite the heightened reflexivity ensuing from his revolutionary political engagement and the significance of Castoriadis’s commitment to self-management and self-organization, Socialism or Barbarism never resolved the problem of leadership creating dependency. Nevertheless, Castoriadis’s later elucidating of the social imaginary deepened his critique of hierarchy and clarified the cultural sources of the modern aspiration to rational domination and control. It disclosed the primordial form of control that is present in the institution of society and how the overturning of heteronomous social relations involves the project of autonomy’s rupture with the closure of meaning. Democratic control can only flow from individual and social autonomy; however, Castoriadis’s juxtaposition of the dominant modern imaginaries foregrounds the general framing of the dialectic of control, rather than the practical dynamics of its structuration and resultant metamorphoses. The latter is essential to addressing the question of whether contestation informed by the project of autonomy can effectively challenge techno-science’s extension of control and the broader contemporary renewal of the capitalist imaginary.