The relevance of Luhmann’s systems theory and of Laclau and Mouffe’s discourse analysis to the elaboration of Marx’s state theory

This on-line version is the pre-copyedited, preprint, English version. The published version can be found here:

 ‘Zur Relevanz von Luhmann’s Systemtheorie und von Laclau und Mouffe`s Diskursanalyse für die Weiterentwicklung der marxistischen Staatstheorie’, in J. Hirsch, J. Kannankulam and J. Wissel, eds, Der Staat der bürger-lichen Gesellschaft. zum Staatsverständnis von Karl Marx, Baden-Baden: Nomos, 157-179, 2008.


Neither Luhmann nor Laclau-Mouffe take the state as their principal theoretical object. Luhmann aimed to develop a universal systems theory in which politics was one among several important functional systems; and Laclau-Mouffe have been more interested in developing an ontology of ‘the political’ than in theorizing the state as a distinct ontic structural ensemble. Indeed, although he later modified this view, Luhmann initially presented the state as no more than the self-description of the political system (contrast 1990a with 2000b: 116-118, 244f, 392).; and, although Laclau and Mouffe each referred to the state in different respects before their joint discourse-analytical turn, they subsequently focused on hegemony and radical democracy. Thus we must explore the relevance of their more general arguments to Marx’s state theory.

Marxian State-theoretical Deficits

Marx left us no coherent overall statement of materialist state theory. This failure particularly affects, as the editors have noted, ‘die Konzeptionalisierung des Verhältnisses von Politik und Ökonomie, die Problematik der Basis-Überbau-Metapher und die Frage der Berücksichtigung der von Marx entwickelten Formanalyse’ (this volume). Nonetheless he and Engels did leave many elements for building the required theory in four complementary approaches: (1) a form analytical analysis of the capitalist type of state and its formal adequacy for a social formation organised under the dominance of capital accumulation; (2) analyses of the historical constitution of actually existing states in different periods, including the capitalist epoch, focusing on how different elements were combined to produce a given state form; (3) historical analyses of particular states, their social bases and functional adequacy for particular capitalist interests and/or state projects; and (4) conjunctural analyses of specific political events, stages of political struggle, moments of political crisis, periods of state transformation, revolution, etc. All four approaches are inspired by historical materialism but they tend to deploy different (perhaps eventually consistent) sets of concepts.

Each approach poses important theoretical questions: (1) form analysis involves the issue of how best to understand the relative autonomy of state forms, state apparatuses, and state power; (2) genealogical analysis problematizes the question of different routes to what, despite their ‘motley diversity’ (Marx 1875), one can call the normal form of the capitalist state; (3) genealogical and historical analyses both pose the problem of the utility of the ‘base-superstructure’ metaphor as an interpretive device or, indeed, as an ontological claim about the relation of state forms and state power to the economic foundations of a given social formation; (4) historical analyses invite us to look at the emergence of the prevailing ‘illusory general interest’ in whose name state power is exercised, whether in normal (bourgeois democratic) states or an exceptional regime; and (5) conjunctural analyses pose the problem of how to analyze the relation between the political scene and the broader balance of class forces.

Luhmann’s evolutionary approach to systems theory offers important clues on the first three deficits of form, genealogical, and historical analysis and, in some cases (especially as developed by later autopoietic systems theorists) to questions of government, governance, and societal steering. His work also offers useful insights into the relative autonomy of the state and the forms of contemporary politics. Conversely, Laclau’s and Mouffe’s post-Marxist analyses appear more relevant to historically specific political strategies, state projects, and hegemonic visions and to conjunctural analysis. They also discuss new social movements and the changing forms of political struggles in ways that might be relevant to the fifth deficit. Why they disappoint these expectations will be explained later.

Niklas Luhmann

Luhmann dismissed Marx as a pre-modern thinker on the grounds that he treated the economy as superordinate to other systems when, for social systems theory, all functional systems in a modern society are equally important and irreplaceable. Moreover, since each system has its own codes and programmes and is operationally closed, no other system can control its activities from outside. Luhmann’s analysis of modern societies also excludes a ruling class (in his terms, an Oberschicht) that can govern society as a whole and denies any primacy to class relations and class identities over other social relations or collective identities. These claims do not seem promising for identifying a Luhmannian contribution to Marxist state theory but he also introduced many concepts that could be fruitful if they are reworked to ensure their commensurability with a Marxist approach. Moreover, when he commented on particular functional systems and organizational relations, he often contradicted his foundational claims in ways that moved him closer to Marxism.

The relation of the economic and the political

Autopoiesis offers an apparently novel way of thinking about the ‘relative autonomy’ of the economic, political, legal, and other systems. It is hard for Marxists to reconcile the autonomy of politics and primacy of political struggle with economic determination in the last instance. Luhmann avoids this problem because he insists that all functional systems, including the economy and the polity, display operational (or decisional) independence and material interdependence. This has both structural and strategic consequences that bear on the relation of the economic and political.

First, structurally, no single functional (autopoietic) system could determine societal development ‘in the last instance’. All such systems have absolute (not relative) operational autonomy. For example, the modern economy is a self-perpetuating system of payments; the modern legal system is a self-contained and self-modifying system of legally-binding legal decisions; the science system is a self-perpetuating system of scientific communications coded in terms of true/false; and the political system produces collectively binding decisions that generate further political decisions. Other functional systems explored by Luhmann include religion, art, the family system, health care, and, a late addition, the mass media. Nonetheless any such operational autonomy is limited by a given system’s relation to its external environment and, more specifically, by its material dependence on the performance of other systems that operate according to their own codes and programmes. These constraints can be read as sources of relativization of autopoiesis and encourage the relevant system to construct simplified, selective models of these constraints and integrate these models into its operations. Each system will model these constraints differently, reflecting their observed relevance to its own reproduction. Despite such constraints, however, each system can maintain its operational autonomy insofar as it has its own operating codes and has sufficient time to implement them, faces competing demands so that it can choose which to process, and has the general legitimacy or societal trust needed to operate without having constantly to justify its specific activities on each occasion. Without such conditions, a functional system can lose its operational autonomy. This poses an interesting question, pursued below, about the conditions under which other functional systems might lose some operational autonomy to the economy (or, indeed, vice versa).

Second, strategically, modern societies are so highly differentiated and polycentric that no single system, central decision-making body, or ruling class could ever coordinate their diverse interactions, organizations, and institutions and ensure their harmonious cooperation toward a common end. Once systems reach ‘autopoietic take-off’, they only respond to problems defined in their own terms. External demands stated in other codes and/or in terms of more general ‘noise’ from the everyday ‘life-world’ will be dismissed as irrelevant or else handled as an irritation to be avoided or overcome in whatever way the perturbed system itself thinks fit.

Base and Superstructure

Luhmann took six concepts from evolutionary theory to analyze inter-systemic relations in an ecology of self-organizing systems. The same concepts could assist historical materialism explore causality in base-superstructure relations, enabling it to escape the twin traps of ‘the common undialectical conception of cause and effect as rigidly opposite poles, the total disregarding of interaction’ (Engels to F. Mehring in Berlin 14 July 1893) and, conversely, of an ‘internal relations’ approach in which all influences are symmetrical as well as reciprocal (see the critique in Rubin 1977: 23).

The six concepts are: operational coupling, co-evolution, interpenetration, structural coupling, structural drift, and ecological dominance. Each identifies a different type of dialectical causal relation whose combination permits us to theorize an open-ended, non-deterministic dialectical interaction that nonetheless displays a non-teleological, long-term directionality. These are precisely the emergent properties that typify base-superstructure relations and that cannot therefore be reduced to a single type of linear causation between distinct entities. Transferred to historical materialism, the first four notions suggest the necessary reciprocity between base and superstructure and the last two provide the means to think about contingent asymmetries in these reciprocal causal relations. Together these concepts enable us to re-interpret ‘determination in the last instance’ as a special type of empirically testable causal relation within a broader set of modes of causal and non-causal determination.

  1. For Luhmann, every functional system is operationally coupled to its environment whenever it responds to perceived irritations and perturbations that it attributes to that environment. The key point about such coupling is that, while it is purely momentary, it can lead to unintended path-dependent effects.
  1. Extending this argument to the self-organizing ecology of operationally autonomous functional systems, the evolution of world society always occurs through blind co-evolution. This process is nonetheless shaped by the interdependencies among functional systems insofar as they constitute mutually relevant environments with the result that the evolution of each is conditioned by ‘disturbances’ that are created by other systems and to which they react and adapt, if at all, in terms of their respective codes and programmes (Luhmann 1986). Those variations will thereby get co-selected that interfere least with the autopoiesis of the interacting systems and will then get co-retained as these selections become suitably sedimented in the programmes, organizational intelligence, strategic capacities and moral economies of the various co-existing systems (Teubner 1989: 78-9).
  1. Autopoietic systems may also be related through interpenetration. This occurs when an autopoietic system presupposes the complex achievements of another system and can treat them as parts of its own system operations. Luhmann illustrates this from the economic, legal, and political systems: for the economy depends on the securing of its juridical and political conditions of possibility; and the legal and political systems depend in turn on the performance of the economy for revenues, legal cases, and political legitimation. Organizations are an important site of interpenetration – indeed Luhmann describes them as ‘Treffräume für Funktionssysteme’ (2000a: 398).
  1. Structural coupling occurs where systems evolve structures that connect systems and facilitate mutual responsiveness in the application of their respective codes.[1] There are few such structures at the system level and they allow only a limited range of adaptations. For example, Luhmann identifies three structures that facilitate coupling among the economic, legal, and political systems: property, contract, and constitution. These structures have different meanings in each system and trigger different sets of operations in each; but they also connect the systems (Luhmann 1997: 787). Luhmann also observed a particularly tight structural coupling among the economic, legal, political, and science systems compared to the mutual irritations between these and other functional systems in world society (2000b: 382).
  1. Structural drift: Where systems are regularly subject to continuing irritations of a specific type, routine forms of dealing with them may develop. These routines are reinforced through the continuing autopoiesis of the coupled systems and become hard to alter (Luhmann 1990b: 408; 1995: 494-5; 1995: 32f; 2000b: 391-2)). The sensitivity of organizations to their environment make them especially susceptible to structural drift through their couplings to various functional systems, leading to various forms of self-limitation (Luhmann 2000a: 397-8).
  1. Ecological dominance originally refers to the relative dominance of a given species in the overall development of its ecosystem. Luhmann applied it only twice (to my knowledge) to explore possible relations among social systems, once to describe the dominance of bureaucratic organization within the political system and once more generally to describe the relation among functional systems. Consistent with his argument that no single system can control other functional systems in world society, Luhmann cites Edgar Morin (1980: 44) on ecological dominance, i.e., an ecological relation wherein some systems may be dominant but where none dominates (Luhmann 1987: 109-110; 1990a: 147-8). This idea was already implicit in some earlier remarks on the world economy (e.g., 1974); and many other Luhmannian analyses suggest to an interested observer the need for such a concept to provide a more complex account of the relations among autopoietic systems. The result of ecological dominance operating in the context of the other five types of reciprocal relation could be that one system in a self-organising ecology of self-organising systems imprints its developmental logic on other systems’ operations more than any of the latter can impose their respective logics on that system (Jessop 2002). In short, even if all functional systems are equal, some may be more equal than others. The concept can also be fruitfully applied, as the field of organizational ecology indicates, to inter-organizational relations. Moreover, because a social ecosystem comprises different types of autopoietic system (including various types of organization as well as various functional systems and subsystems), the relative dominance of specific organizations and subsystems in a given functional system will impact the overall dynamics of ecological dominance.

The obvious question to ask here is whether one functional system is more likely to be ecologically dominant than the others. Luhmann provides several direct and indirect hints that the international market economy tends to be the ecologically dominant system but, to my knowledge, he never argued this explicitly. One of the most interesting texts on the relative significance of functional systems is also one of his most accessible: Ecological Communication (1988). This indicates quite clearly that, although each functional system has its specific codes and programmes, there is considerable inequality in their capacity to structure the environments of other systems. The following section draws on Luhmann, other system theorists, and evolutionary and institutional economists to suggest how ‘ecological dominance’ can help to resolve some key problems in historical materialism.

The Ecological Dominance of the Economy

Ecological dominance is a contingent emergent relationship between two or more systems rather than a naturally necessary property of a single system. Thus a given functional system can be more or less ecologically dominant, its dominance may vary across different systems in its environment and/or with changing circumstances, and the continuation of any dominance will depend on the development of the ecosystem as a whole.[2] So there is no ‘last instance’ in relations of ecological dominance. But, given that the capitalist economy is structurally coupled to other operationally autonomous systems and to the lifeworld (and these to each other too), we can ask which, if any, of them could become ecologically dominant.

There are seven analytically distinct, but empirically interrelated, aspects of the social (as opposed to biological) world that affect a system’s potential in this regard (see Table 1). Considered in these terms, the capitalist economy, with its distinctive, self-valorizing logic, tends to have just those properties that favour ecological dominance.


 First, as it gets increasingly disembedded from other systems, internal competition to reduce socially necessary labour-time and socially necessary turnover time becomes an ever more powerful force driving accumulation. Extra-economic pressures are thereby translated into competition among capitals to find new opportunities for profit and/or to exit from particular markets in order to preserve capital. Different degrees of liquidity, flexibility, and fungibility mean that capitals vary in their ability to respond. Finance capital controls the most liquid, abstract, and generalized resource and therefore has the most capacity to respond opportunities for profit and external perturbations (Luhmann 1996).

Second, the capitalist economy is internally complex and flexible because of the decentralized, anarchic nature of market forces and the dual role of the price mechanism as a flexible mechanism for allocating capital to different economic activities and a stimulus to second-order observation, learning and self-reflection. One of the aspects contributing to ecological dominance in the natural world is a superior capacity to tolerate environmental disturbances (Keddy 1989: 18-19) and this capacity is well-developed in the economy because of its greater internal complexity (multiplicity and heterogeneity of elements), the looser coupling among these elements, and the high degree of reflexive capacity (self-monitoring) in the market economy (Baraldi et al., 1998: 151).

Third, capital has developed strong capacities to extend its operations in time and space (time-space distantiation) and/or to compress them (time-space compression). The mutual reinforcement of time-space distantiation and time-space compression facilitates real-time integration in the world market and makes it easier to maintain its self-expansionary logic in response to perturbations. These capacities are related to the formal, procedural rationality of the market, its highly developed abstract and technical code, the requisite variety of its internal operations, and its reliance on the symbolic medium of money – all of which increase its ‘resonance capacity’ to react to internal and external conditions (Luhmann 1988: 37-41). The greater is this capacity relative to other systems, the greater the scope for capital’s ecological dominance.

Fourth, through these and other mechanisms, capital develops its chances of avoiding the structural constraints of other systems and their attempts at control, thereby increasing its ‘indifference’ to the environment (cf. Luhmann 1988; Lohmann 1991). This is especially true of the only fully global subsystem in the economy: international finance (Luhmann 1996). This does not mean that either finance or the economy more generally can escape from its overall dependence on other functional systems’ general contribution to its operations or, of course, from the crisis-tendencies associated with its own contradictions and dilemmas. Attempts to escape particular constraints and particular attempts at control can nonetheless occur through its own internal operations in time (discounting, insurance, risk management, futures, derivatives, etc.) or space (capital flight, relocation, extra-territoriality, etc.), through the colonization of organizations central to the operation of other functional systems by the logic of exchange value, or through simple personal corruption.

Fifth, in contrast to natural evolution, where species must adapt to or exit from their environment, social evolution may involve reflexive self-organization and attempts to redesign the environment (cf. Marx on the distinction between bees and architects, Capital I, 1965: 284). This capacity may even extend to attempts to change the mode of social evolution (Willke 1997). This does not mean that the evolution of particular functional systems, let alone the evolution of world society, can be fully controlled but nor does it exclude attempts at shaping the path of co-evolution among organizations, systems, and, eventually, world society.  Where different organizations and systems seek to adapt to and/or to change their environment, ‘the logic of evolutionary progress is toward ecosystems which sustain only the dominant, environment-controlling species, and its symbionts and parasites’ (Bateson 1972: 451). This poses the question of the relative capacity of different organizations and systems to change their environment rather than being forced to adapt to changes in their respective environments (see also point seven).

Sixth, the primacy of accumulation over other principles of societalization can be explored in terms of the relative influence of the self-descriptions of different functional systems and the role of the mass media. Self-descriptions may vary in their importance within generalized societal communication in relation to: (a) alternative Vergesellschaftungsmodi; (b) secondary coding in the programmes of each functional system; (c) the decision premises of organizations; (d) the relative weight of different interests in negative coordination among organizations with different functional primacies (ensuring that the application of their respective codes does not lead to mutual blockages), and (e) the changing character of public opinion. The mass media also have a crucial role in offering information to functional systems, organizations, and interactions, especially where there is a tendency for those who control the means of production also to control the means of mental production and thereby shape news values. The struggle for hegemony in this context will be easier where a functional system is internally organized, like the world economy, on centre-periphery lines rather than on the basis of stratification or, least favourably, segmentation (Luhmann 1996; Simsa 2002). Hegemonic struggle will also be easier where social forces emerge that cross-cut functional systems and seek to harmonize (through positive or negative coordination) their operations. A power bloc organized through parallel power networks provides an important mechanism of system and social integration in this regard (Poulantzas 1978; cf. Baecker 2001, 2006). None of this implies that a hegemonic vision could adequately represent the identity of world society as repraesentatio identitatis any more than this would be possible from the viewpoint of a single system. But the function of hegemony is not to represent the whole of society but to represent a set of particular interests as the interests of society (cf. German Ideology; Gramsci 1971).

Seventh, the ecologically dominant system will be the most important source of external adaptive pressure on other systems. In general, any increase in the complexity of one functional system increases the complexity of the environment of other systems and forces them to increase their own internal complexity in order to maintain their capacity for autopoiesis (Baraldi et al., 1998: 96). For the first four factors given above increasing internal complexity is most likely, in the context of an emerging world society, to characterize the international market economy. Indeed, for Wagner, it is the system with the highest Versagensquote that will gain primacy (Primat) or, in current terminology, ecological dominance (2006: 8). This is especially likely because the organizations that are so important for the realization of other systems’ activities must secure the revenues to support their operations from the economy, either directly or indirectly (cf. Lange 2003: 233). This enhances the capacity of the profit-oriented, market-mediated economy to colonize other functional systems and the life-world through the logic of commodification and the adoption of net revenues as the major secondary code. Pressures on individual territorial states in this regard have been increased through globalization (Stichweh 2000: 195f), leading to permanent irritation by economic problems (Wagner 2006: 7).

This all suggests that ‘ecological dominance’ could be used productively to re-interpret the classical Marxist idea of ‘economic determination in the last instance’ and the Gramscian idea of the ‘historical bloc’. The former was always problematic because the capitalist mode of production lacks the autonomy (as a cause without cause) to be fully determinant in the first, medium, or last instance. But a theory of internal relations cannot explain the asymmetry entailed in the Marxist claim about the primacy of economic relations. An alternative is to suggest that capital is ecologically dominant insofar as the logic of accumulation tends to cause more problems for other systems than they cause for the expanded reproduction of capital. This does not exclude reciprocal influence from other systems as their operations and dynamic disturb, irritate, or disrupt the circuit of capital and thereby influence in turn its profit-oriented, market-mediated evolution. The resulting co-evolutionary structural drift in the shadow of ecological dominance can in turn explain the nature of the ‘historical bloc’ as a pattern of structured coherence between base and superstructure (Gramsci 1971). The core to the formation of such a bloc is the nexus of the economic, juridical, and political systems, which, even in Luhmann’s analysis, tend to be more tightly coupled than any other set of systems (1988; cf. 2000: 51, 181-2, 243). More generally he notes that

The working together of function systems is also necessary in practically all cases. For example, scientific research has made the construction of nuclear power plants economically possible through a political decision about legal liability limitations. The world is not just constituted so that events generally fit within the framework of one function alone (Luhmann 1988: 49-50; cf. 90).

Taking these arguments further one could argue that the ecological dominance of capitalism is closely related to the extent to which its degrees of freedom, opportunities for self-reorganization, scope for time-space distantiation and compression, externalization of problems, and hegemonic capacities can be freed from confinement within limited ecological spaces policed by another system (such as a political system segmented into mutually exclusive sovereign territories). This is where globalization, especially in its neo-liberal form, promotes the relative ecological dominance of the capitalist economic system. For it reinforces the dominance of the exchange-value moment of the various forms of the capital relation and frees money capital as the most abstract expression of the capital relation to move at will within the world market to maximize opportunities for profit (cf. Jessop 2002).

In conclusion, I suggest that Marx and Engels themselves operated implicitly with the concept of ecological dominance. For they seem to have developed ideas about blind co-evolution and structural coupling avant la lettre as they elaborated the materialistic conception of history and addressed the relationship between different social forms. In the absence of a developed concept of ‘ecological dominance’ they resorted to a wide range of metaphors to express it in pre-theoretical terms (see, for example, the German Ideology, the 1857 Introduction, and Engels’ letters on historical materialism). As a general thought experiment (Gedankenexperiment), it would be interesting to substitute the concept of structural coupling for terms such as ‘derivation’, ‘correspondence’, etc., and the concept of ecological dominance for phrases such as ‘ultimately determining’, ‘determinant in the last instance’, ‘determinant in the final analysis’, ‘finally asserts itself’, and the like. I suggest that in each case ecological dominance provides a more precise and testable account of what is at stake for Marx and Engels in these arguments. And, because the argument is both theoretically grounded and empirically open – there is no guarantee in practice of the emergence and reproduction of the conditions for ecological dominance by the capitalist market economy, this analysis must attend not only to structural relations but also to the role of social forces in mediating these relations.

Luhmann and the State

So far I have focused primarily on the first two issues raised by the editors because this is where, I believe, Luhmann’s work is most relevant to historical materialism. But he also provides a wealth of interesting – often positively ‘irritating’ — observations and insights, which cannot be fully explored here. With all due recognition of the brevity of these remarks, therefore, I simply highlight eight points among many possible issues for further exploration:

(a)   The historical semantics of political and state discourses and the implications for the development of the state (see also Lange 2003; Wimmer 1996).

(b)   The connection between Luhmann’s emphasis on selection at different levels of analysis from functional system to interactions and its implications for a strategic-relational approach to the state and the political system (cf. Jessop 1990). This also includes his analysis of micro-diversity and self-organization at the macro-level (see Luhmann 1997b)

(c)   The significance of the transition to democratic politics for the autopoiesis of the political system – it is only then that politics becomes non-hierarchical, symmetrical, capable of self-observation, and recursive. Paradoxically, this could have interesting implications for the bourgeois democratic republic as ‘the best possible political shell for capital’ (cf. Poulantzas 1973).

(d)   The orientation of political decision-making and action to internal political environments and political codes (including ‘government-opposition’, ‘progressive-conservative’, ‘interventionist-restrictive’) and the importance of these three reference points in political calculation – public opinion (as defined by the mass media and manipulated by political parties, pressure groups, and governments), persons as links between government and bureaucracy, and law as the link between administration and people as subjects (2000b). This has major implications for understanding the flexibility of the normal capitalist state, the mediatisation of politics, and the recurrent tendencies to state failure – including in the field of the expanded reproduction of capitalism.

(e)   The state’s strong dependence on law and money for securing its collectively binding decisions – which implies a tight coupling among the political, legal, and economic systems, even though each sees this coupling from its own distinctive viewpoint (2000b). This has important implications for base-superstructure relations and the contingent co-evolution of historical blocs.

(f)    The function of the discourse of the state (as a self-description of the political system) in focusing, unifying, and aggregating political action (2000b). This requires the development of constitutional law to resolve the resulting paradoxes of self-reference and to provide some basis for its apparatus unity and accountability and also highlights the importance of state projects as one dimension of hegemony.

(g)   The impossibility of any functional system being able to steer another and/of the development of world society and the possibility of more limited steering through mutual irritation through codes, programmes, interface structure, and multi-functional organizations have interesting implications for the changing forms in which the economic and political are articulated in different stages and varieties of capitalism.

(h)  The political system is necessarily segmented into territorial states and that these are also ordered according to centre-periphery relations opens interesting ways to explore international relations, imperialism, and questions of global governance (for Luhmannian approaches to international relations, see Albert and Hilkermeier 2004).

Ernesto Laclau and Chantal Mouffe

Laclau-Mouffe offer a self-proclaimed ‘post-Marxist’ account of social practices in terms inspired by linguistics and discourse analysis. Their key text is a much-discussed intervention into political theory, namely, Hegemony and Socialist Strategy (1985). This elaborated two main arguments: first, that the real scope of contingency in social relations is far wider than classical Marxism has admitted in theory or practice and, indeed, has become even more extensive in late capitalism; and, second, that the resulting theoretical and strategic deficits in classical Marxism can be resolved through a rigorous and systematic development of a post-structuralist, discourse-analytical approach to the nature of politics and hegemony and through the development of a radical and plural democracy in which socialism is at best primus inter pares among many interconnected demands and aspirations. In later work, Laclau has continued to explore the implications of their discursive turn for identity formation, the hegemonic logic of the social, and the nature of political struggles. His analysis of populist reason is especially relevant for our purposes as it argues that populism provides the key to deciphering the general nature of politics (2005). Mouffe in turn has explored other themes from their co-authored book, especially the nature of radical and plural democracy, and has focused on the liberal democracy, agonism and antagonism, and right-wing extremism.

Laclau-Mouffe claimed that the social world (as a world of relational meaning) is inherently discursive and, indeed, is constituted only in and through discourse. All social relations are endowed with meaning and are differentially articulated in various ways to each other in an all-encompassing discursive ensemble. In short, their work involves an unambiguously ontological discursive turn. Although their equation of the social and discourse is a foundational claim, it is presented as anti-foundational and anti-essentialist. They critique the tendencies in the Second International – which they also attribute to Marx – to economic and class reductionism and to a statist view of both reform and revolution and, in doing so, claim to reveal the theoretical and political space in which contingency can and must operate. Focusing on meaning production discloses the inherently discursive aspects of the forces of production, the social relations of production, and class relations. This vastly expands the scope of historical contingency and therefore for political strategies to make a difference by altering identities, building alliances, and developing a radical and plural democracy. But this brings with it another form of essentialism – the reduction of the social to politics such that every social space is either politically contested or, although ‘sedimented’ (i.e., stabilized, naturalized), can be re-politicized. This goes beyond any claim about the primacy of the political (which always depends on the existence of extra-political regions or spheres) to a pan-political ontology of the social in which every social relation is either actually or potentially political (cf. Laclau 2005: 154).

While Luhmann starts from the claim that social systems are constituted exclusively by flows of communications, Laclau and Mouffe begin from the claim that all social systems involve the production of meaning. They conclude from this that an adequate social explanation must refer to signifying relations rather to any type of physical or material causality (1985; cf. Bernans 1999: 20). This suggests that their work is based on an ‘anti-determinist acausalism’ (cf. Bunge 1961: 29). They deny lawful links among events and qualities in the social world in favour of their purely contingent discursive articulation. This argument ignores the need, long ago noted by Max Weber (1949), for explanations that are adequate at the levels of causality and meaning. While Luhmann is also suspicious of causal explanation, his operative constructionism does allow a regulative role for the real world beyond communication. Laclau-Mouffe hint at this (e.g., 1987) but Laclau affirms it in his analysis of populism, in which external reality expresses itself via negation, i.e., by providing a ‘reality check’ that limits the resonance of alternative political projects, making some more plausible and appealing than others (2005: 89, 91-96, 190-1, 201ff).

This leaves Laclau-Mouffe with three relatively underdeveloped phenomenological and psychoanalytic concepts to study structure: sedimentation, sutures, and nodal points. First, ‘sedimentation’ refers to the naturalization and institutionalization of social relations so that they come to be reproduced through dull repetition rather than deliberate articulation (Laclau 2005: 154; cf. Torfing 1999: 69-71). But this can be reversed simply through a new hegemonic articulation that deconstructs and re-politicizes sedimented relations. Second, the concept of suture refers to purely temporary attempts to bind different elements and relations together, despite their differences and distinctions. Thus, consistent with its metaphorical connotations, a suture is a short-term fix that is bound to dissolve. Third, nodal points are provisional and unstable centres that emerge from the primordial flux of social relations to provide temporary points of reference for the contingent articulation of social relations and attempts to suture them into relatively stable, sedimented ensembles. Given their pan-politicist approach, Laclau-Mouffe insist that power cannot be localized in the state or any other single nodal point (or power centre) but occurs across the whole field of discursivity. Thus hegemony is ‘free-floating’ and must be articulated everywhere and in all directions (1985: 139). Moreover, because there is always a plurality of power centres, any one of them will be limited in its effectiveness by other nodal points (1985: 139, 142-3). Moreover, in line with their rejection of Marxist ‘statist’ assumptions, they do not even claim, as others have done, that the state is more privileged than other sites of political struggle.

The economic and the political

Notwithstanding earlier rigorously argued work on capitalism (Laclau 1971), the economic and the political (Laclau 1975), the role of populism in mobilizing popular opposition to the power bloc (Laclau 1976), and hegemony and the historical bloc (Mouffe 1981), Hegemony and Socialist Strategy denies the difference between the economic and the political because hegemonic (i.e., political) articulations are crucial even inside the conventionally defined economy (1985: 77ff, 120-1, 140, 180). In emptying the economic space of all traces of essentialism, however, they also relieve it of theoretical content (Diskin & Sandler 1993: 30). Just as their ontologization of the political leads them to examine hegemony in an institutional vacuum, they also turn the economy into an ontic void. They then ‘fill [this] in an ad hoc fashion with unexamined economic concepts and relationships which, ironically, retain their essentialist underpinnings. In the latter half of HSS [Hegemony and Socialist Strategy], the economy is a blank space, with a marker (‘the economy’) and sign posts (“commodification”), inscribed upon its surface. … There are economic concepts in HSS but no concept of the economy’ (Diskin & Sandler 1993: 30).

The same point holds for their class analysis. For Laclau-Mouffe effectively equate Ricardo and Marx by incorrectly accusing the latter of defining labour as a commodity like any other, of ignoring the role of power tout court in shaping the forces and relations of production, and of arguing that there is a necessary contradiction between capital and labour in their encounter as commodity-owners in the labour market. It is easy for them to reject all three positions (as Marx himself had, of course, done before them). It follows that the capital relation is a purely contingent political relation and that anti-capitalist resistance, if any, can arise from any identity or set of identities that real workers may have and not merely from their position as a wage-earner in the labour market. These claims eliminate any understanding of the historical specificity of capitalism, of the distinctive form of capitalist exploitation (which is actually rooted in formally free and equal exchange in the labour market and ‘factory despotism’ in the labour process), and of the material grounding of class identities in the relations of production rather than the relations of exchange.

These difficulties in addressing the specificity of the economic in capitalism are the joint product of their pan-political ontology and their associated emphasis on the contingency of all social relations. It follows that they cannot have a clear account of the specificity of the political in capitalism (contrast Laclau 1975). Thus Laclau recently argued that ‘all struggles are, by definition, political … There is no room for a distinction between economic and political struggles’ (2005: 154). This may explain the failure to offer any concepts for analyzing state structures, state capacities, and the specificity of state power. Instead Laclau-Mouffe use conventional terminology when they do discuss the state and its policies (as well as capitalism, its economic dynamics and crisis-tendencies, and social repercussions) and this makes their empirical examples appear as ad hoc Plausibilisierung rather than rigorous analyses.

Base and superstructure

Laclau and Mouffe totally reject the notion of base and superstructure. They take this metaphor literally and conclude that it posits total determination of the superstructure by the base and implies that the economic base is wholly self-sufficient sui causa (Laclau & Mouffe 1985: 120-1, 142; Laclau 1990: 6-14, 55; Laclau 2005: 250). This is sufficient grounds for them to ignore other meanings of the metaphor and refuse to consider whether there might be ways to reinscribe the idea into post-Marxist analysis. On the contrary, post-Marxism is a radically new response to ‘the increasingly desperate contortions which took place around notions such as “determination in the last instance” and “relative autonomy”’ (Laclau & Mouffe 2001: viii; Laclau 2005: 250). Even when he concedes that base-superstructure arguments have ‘varying degrees of sophistication’ (1990: 202), Laclau disavows them all as ‘unappealing’ as if this were simply a matter of taste. Laclau-Mouffe also reject the relative autonomy of the state as philosophically meaningless because autonomy is an absolute rather than relative term (1985: 139-40). Indeed, their whole approach lacks concepts to think the articulation of the economic and the political.

Form analysis

Given their general approach and their desire to deconstruct Marxist concepts, form could only mean the relative fixity of sedimented structures, which are always open to re-politicization. Indeed the whole thrust of their work is to either to reject Marxist concepts as essentialist (form would be a prime candidate for this) or to empty them of their original content and redefine them in ways consistent with linguistic analysis and/or their own preference for radical and plural democracy.[3]


Luhmann and Laclau-Mouffe share several starting points in their approach to social analysis. They are ontological and epistemological anti-foundationalists and develop arguments that abandon any claim to be grounded in the nature of the social world. For different reasons they also reject ‘meta-narratives’ and insist on the contingency of the social world (cf. Stäheli 2000). Luhmann’s analysis is oriented to the distinction between system and environment and adopts modern systems theory as its entrypoint to analyse such distinctions and their structural and semantic effects. In contrast, Laclau and Mouffe are concerned with the distinction between the self and the other and adopt relational linguistics and (for Laclau) Lacanian psycho-analysis as their entry points to speculate about the resulting antagonisms and struggles for hegemony that arise over individual and collective will formation. All three agree on the key role of the ‘constitutive outside’ in analysing their chosen theoretical objects.[4]

Let us now return to our three key questions and see what answers our authors can offer. First, in regard to the economic and political, Luhmann initially treats them as distinct autopoietic systems but later provides a whole series of concepts to explore their structural coupling, interpenetration, blind co-evolution, and even the ecological dominance of the logic of capital accumulation. In contrast, Laclau and Mouffe dissolve the distinction between the economic and political on the grounds that these are the product of semantic distinctions and their boundaries are consequently unstable. In consequence, in contrast to Luhmann, who has a rich set of concepts for analyzing the specificity of the economic and political systems, Laclau-Mouffe effectively ignore the specificities of both systems and their different forms of articulation in favour of a pan-politicist ontology.

Second, Luhmann enables us to rethink economic determination and the historical bloc. This is probably his most useful contribution to a reinvigoration of Marxist political economy – especially given the current dominance of neo-liberalism in world society. Laclau-Mouffe reject the very idea of base-superstructure as essentialist.  All they offer in its place is the contingent articulation of social relations around several nodal points to produce an inherently unstable capitalist order. Yet they have no doubt about the importance of capitalism, of the plurality of anti-capitalist struggles that it provokes, and, indeed, ‘globalized capitalism’ is now regarded by Laclau as the principal source of social ills (Laclau 2005: 230-231, 242). They also once regarded the abolition of capitalism as the sine qua non of radical and plural democracy.

Third, regarding the question of state form, while none of our theorists take the state as a primary object of analysis, Luhmann still has the most to offer a reconstruction of materialist state theory. Within his own theoretical framework, he offers many provocative insights. In contrast, the state figures minimally in the post-Marxist writings of Laclau-Mouffe, is discussed in conventional terms when it does appear, generally lacks specific institutional forms and organizational architectures, and is typically considered as one among several addressees or targets of democratic and populist demands rather than as a crucial nodal point in the overall articulation and reproduction of social power. This reflects their view that hegemony is a de-centred principle of articulation concerned to produce a relatively unified political subject (or collective will) with its connection to the state left underdetermined and, perhaps, undecidable. They argue for radical democracy rather than critique the state, paying more attention to the heterogeneity of civil society than to attempts to unify the state apparatus. And even in this respect they focus one-sidedly on hegemony to the detriment of other Gramscian categories such as passive revolution, force-fraud-corruption, and open class war.

We end with a paradox. While Luhmann simply dismissed Marx as a pre-modern theorist and then ignored him, Laclau and Mouffe immodestly presented their work as the culmination-cum-reinvention of Marxism for the complex modern world and the challenges of contemporary socialism. Yet Luhmann’s work has more to offer – albeit unintentionally – for the development of historical materialism and state theory. In contrast, Laclau-Mouffe have abandoned Marxist theorizing and dialogue with Marxist theorists in favour of critical engagement with other philosophical and theoretical traditions. There is considerable work to be done to integrate Luhmannian concepts into historical materialism but I believe that they were prefigured in the work of Marx and Engels and that undertaking this task would strengthen some of their key insights and arguments. There is also considerable work to be done to eliminate the contribution of Laclau-Mouffe to the weakening of Marxism through their failure to engage constructively with the work of its founding fathers and leading theoreticians. The positive elements in their analysis are also prefigured in Marx’s work because he was deeply concerned with an historical, materialist, and critical understanding of language in all its forms (cf. Fairclough and Graham 2002) and they have been further developed in Gramsci’s ‘vernacular materialism’ (cf. Ives 2005) and similar currents in Marxist linguistic analysis. Marxist scholars have also developed alternative approaches to critical discourse analysis. These approaches also follow Marx in developing a more or less adequate set of concepts for the analysis of the structural moments of social action. Thus, while I accept that there Laclau and Mouffe offer some interesting, provocative, but by no means unparalleled ideas, the cost of integrating them in the strong Laclavian programme[5] (given the necessary relational nature of all arguments in their research programme) is too high to warrant taking them as a starting point for the reconstruction of Marxist state theory and historical materialism more generally. This may reflect the fact that, whereas Luhmann has been accused of trying to solve philosophical problems through sociological means (Hondrich 1973: 89ff), Laclau-Mouffe have tried to solve political questions through philosophical means.


Albert, M. and L. Hilkermeier, eds (2004) Observing International Relations. Niklas Luhmann and World Politics, London: Routledge.

Andersen, N.A. (2003) Discursive analytical strategies: understanding Foucault, Koselleck, Laclau and Luhmann, Bristol: Policy Press.

Baecker, D. (2001) ‘Managing corporations in networks’, Thesis Eleven, 66, 80-98.

Baecker, D. (2006) ‘Network Society‘, in Niels Overgaard Lehmann, Lars Qvortup, Bo Kampmann Walter, eds, The Concept of the Network Society: Post-Ontological Reflections, Copenhagen: Samfundslitteratur.

Baraldi, C., Corsi, G., Esposito, E. (1998) GLU. Glossar zu Niklas Luhmann’s Theorie sozialer Systeme, Frankfurt: Suhrkamp.

Bateson, G. (1972) Steps to an Ecology of Mind: Collected Essays in Anthropology, Psychiatry, Evolution, and Epistemology, London. Intertexts

Bernans, D. (1999) ‘Historical materialism and ordinary language: grammatical peculiarities of the class struggle “Language Game”’, Rethinking Marxism, 11 (2), 18-37.

Bunge, M. (1963) Causality, Cleveland: Meridian.

DIskin, J. and Sandler, B. (1993) ‘Essentialism and the economy in the post-Marxist imaginary: reopening the sutures’, Rethinking Marxism, 6 (3), 28-48.

Fairclough, Norman and Phil Graham (2002) ‘Marx as a Critical Discourse Analyst: The genesis of a critical method and its relevance to the critique of global capital’, Journal Estudios de Sociolinguistica, 3 (1), 185-229.

Gramsci, Antonio (1971) Selections from the Prison Notebooks, London: Lawrence & Wishart.

Hondrich, K-O. (1973) ‘Systemtheorie als Instrument der Gesellschaftsanalyse’, in F. Maciejewski, Ed., Theorie der Gesellschaft oder Sozialtechnologie: Beiträge zur Habermas-Luhmann-Diskussion, Frankfurt: Suhrkamp, 88-114.

Ives, P. (2005) ‘Language, agency and hegemony: a Gramscian response to post-Marxism’, Cambridge International Review of Social and Political Philosophy, 8 (4), 455-468.

Jessop, B. (1990) State Theory: Putting Capitalist States in their Place, Cambridge.

Keddy, P.A. (1989) Competition, London: Chapman and Hall.

Kiss, G. (1986) Grundzüge und Entwicklung der Luhmannschen Systemtheorie, Stuttgart: Ferdinand Enke.

Laclau, E. (1971) ‘Feudalism and Capitalism in Latin America’, New Left Review, 67, 19-38.

Laclau, E. (1975) ‘The specificity of the political: the Poulantzas-Miliband debate’, Economy and Society, 4 (1), 87-110.

Laclau, E. (1977) Politics and Ideology in Marxist Theory: Capitalism; Fascism; Populism, London: New Left Books.

Laclau, E. (1990) New Reflections on the Revolution of our Times, London: Verso.

Laclau, E. (2005) On Populist Reason, London: Verso.

Laclau, E. and C. Mouffe (1985) Hegemony and Socialist Strategy, London: Verso.

Laclau, E. and C. Mouffe (1987) ‘Post-Marxism without apologies’, New Left Review, 166, 79-106.

Laclau, E. and C. Mouffe (2001) ‘Preface to the Second Edition’, in idem, Hegemony and Socialist Strategy, London: Verso, vii-xix.

Lange, S. (2003) Niklas Luhmanns Theorie der Politik. Eine Abklärung der Staatsgesellschaft, Opladen: Westdeutscher Verlag.

Lohmann, G. (1991) Indifferenz und Gesellschaft. Eine kritische Auseinander-setzung mit Marx, Frankfurt: Suhrkamp.

Luhmann, N. (1983) ’Der Wohlfahrtsstaat zwischen Evolution und Rationalität, in Soziologische Aufklärung, 4, P. Koslowski, P. Kreuzer, and R. Löw, eds, Chancen und Grenzen des Sozialstaats: Staatstheorie – Politische Ökonomie – Politik, Tübingen: J.C.B. Mohr (Paul Siebeck), 26-40.

Luhmann, N. (1974) MS Bielefeld (cited in Kiss 1986)

Luhmann, N. (1982) ‘Autopoiesis, Handlung und kommunikative  Verständigung’,  Zeitschrift für Soziologie, 11, 366-79.

Luhmann, N. (1987) ’Der Wohlfahrtsstaat zwischen Evolution und Rationalität, in Soziologische Aufklärung 4: Beiträge zur funktionalen Differenzierung der Gesellschaft, Opladen: Westdeutscher Verlag, 104-116.

Luhmann, N. (1987) Soziale Systeme: Grundriß einer allgemeinen Theorie, Frankfurt.

Luhmann, N. (1988) Ecological Communication, Cambridge: Polity.

Luhmann, N. (1988) Die Wirtschaft der Gesellschaft, Frankfurt: Suhrkamp.

Luhmann, N. (1990a) ’State and politics: towards a semantics of the self-description of political systems’, in idem, Political Theory in the Welfare State, Berlin: de Gruyter, 117-154 (see Luhmann 1984).

Luhmann, N. (1990b) Die Wissenschaft der Wirtschaft, Frankfurt: Suhrkamp.

Luhmann, N. (1994a) ‘Die Gesellschaft und ihre Organisationen’, in H.-U. Derlien, U. Gehrhardt, and F.W. Scharpf, eds, Systemrationalität und Partialinteresse: Baden-Baden: Nomos, 189 – 201.

Luhmann, N. (1994b) ’Warum Systemtheorie?’, in A. Boronoev, ed., Probleme der theoretischen Soziologie, St. Petersburg: St Petersburg State University, 25-42., accessed 21.04.2007

Luhmann, N. (1996) ’Politics and Economics’, Thesis Eleven, 53, 1-9.

Luhmann, N. (1997) Die Gesellschaft der Gesellschaft, Frankfurt: Suhrkamp

Luhmann, N. (1997b) Selbstorganisation und Mikrodiversität: Zur Wissenssoziologie des neuzeitlichen Individualismus, Soziale Systeme, 3 (1), 23-32.

Luhmann, N. (2000a) Organisation und Entscheidung, Opladen: Westdeutscher Verlag.

Luhmann, N. (2000b) Die Politik der Gesellschaft, Frankfurt: Suhrkamp

Luhmann, N. (2002) Einführung in die Systemtheorie, Opladen: Westdeutscher Verlag.

Marx, K. (1867) Capital 1.

Marx, K. (1875) Kritik des Gotha Programms

Morin, Edgar (1980) La Méthode: La nature de la nature, vol 1, Paris: Seuil

Mouffe, C. (1981) ‘Hegemony and the integral state in Gramsci: towards a new concept of politics’, in G. Bridges and R. Brunt, eds, Silver Linings: some Strategies for the Eighties, London: Lawrence & Wishart, 167-186.

Poulantzas, N. (1973) Political Power and Social Classes, London: NLB.

Poulantzas, N. (1978) State, Power, Socialism, London: Verso.

Ruben, D.-H. (1977) Marxism and Materialism: a Study in Marxist theory of Knowledge, Brighton: Harvester.

Schimank, U. (2005) ‘Funktionale Differenzierung und gesellschaftsweiter Primat von Teilsystemen – offene Fragen bei Parsons und Luhmann’, Soziale Systeme, 11 (2), 395-414.

Simsa, R. (2002) ‘Strukturelle Kopplung: die Antwort der Theorie auf der Geschlossenheit sozialer Systeme und ihre Bedeutung für die Politik’, in K.U. Hellmann and R. Schmalz-Bruns, eds, Theorie der Politik. Niklas Luhmanns politische Soziologie, FrankfurtL Suhrkamp, 149-170.

Stäheli, U. (2000) Sinnzusammenbrüche. Eine dekonstruktive Lektüre von Niklas Luhmann’s Systemtheorie, Wielerswist: Velbrück.

Teubner, G. (1989) Recht als autopoietisches System, Frankfurt: Suhrkamp.

Torfing, J. (1999) New Theories of Discourse: Laclau, Mouffe and Žižek, Oxford: Blackwell.

Townshend, J. (2003) ‘Discourse theory and political analysis: a new paradigm from the Essex school?’, British Journal of Politics and International Relations, 5 (1), 129-142.

Wagner, T. (2006) Funktionale Differenzierung und ein ökonomischer Primat? Paper available at, last accessed 19.12. 2006

Weber, M. (1949) The Methodology of the Social Sciences, Glencoe: Free Press.

Willke, H. (1997) Der Supervisionsstaat, Frankfurt: Suhrkamp.

Wimmer, H. (1996) Evolution der Politik: von der Stammesgesellschaft zur modernen Demokratie, Wien: Wien Universitäts Verlag.



[1] Interpenetration is a special case of structural coupling.

[2] Similar points apply to the ecological dominance of organizations.

[3] An exception is Laclau’s advice to Aletta Norval that ‘the basic categories of Marxism must be presented as specific historical forms within a wider universe of possible articulations’ (1990: 166)

[4] Stäheli argues that Luhmann’s systems theory would benefit from insights from deconstructive theory and discourse analysis, especially regarding its attempt to impose closure on the communication of meaning to avoid facing the real possibility of the collapse of meaning (Sinnzusammenbrüche). In particular, he calls for ‘‘the reconfiguration of three system-theoretical problems to enable the inscription of ‘the political’ into system theory: surpluses of meaning lead to a non-algorithmic undecidability in the form of dislocations; self-descriptions have a system-constitutive role in the de-paradoxification of systems that are continually overrun by unstructured complexity; and the relation of openness and closure engenders an ‘ungrounded’ interdiscursive site that does not function beyond figures of closure but is actually sustained by the ‘failure’ to secure closure (2000: 310-11, my translation). This does not affect my own arguments about Luhmann’s work in the first part of this contribution.

[5] Townshend (2004) draws an important contrast between the ‘strong’ programme of the Essex School founded by Laclau-Mouffe, with commitments to an acausal notion of contingency and Lacanian psychoanalysis, and the ‘weak’ programme, which allows for material and causal explanation and does not ground antagonism in Lacanian lack or desire.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s