Developing the concept of society: Institutional domains,regimes of inequalities and complex systems in a global era

Abstract

This article develops the concept of society to meet the challenge of cross-border and global processes. Global processes have made visible the inadequacy of interpreting the concept of society as if it were a nation-state, since there is a lack of congruence of institutional domains (economy, polity, civil society, violence) and regimes of inequality (class, gender, ethnicity). The article engages with two strands of intellectual heritage in sociological analysis of society as a macro concept: the differentiation of institutions and the relations of inequality. The concepts of society and societalisation are developed by hybridising these two approaches rather than selecting only one or the other. To achieve this, the concept of system is developed by drawing on complexity science. This enables the simultaneous analysis of differentiated institutional domains (economy, polity, violence, civil society) and multiple regimes of inequality without reductionism. In turn, this facilitates the fluent theorisation of variations in the temporal and spatial reach of social systems.

Introduction

‘Society’ has been a core concept for the discipline of sociology. It offers a unique approach to thinking about large-scale social change: by conceptualising the interconnectedness of social life in which changes in one aspect of social relations change other aspects of social relations. However, the concept of society has been challenged to address the spatiality, re-bordering and re-scaling implied by processes of globalisation, colonisation and Europeanisation and thus to address the tension between cross-border studies and theories of society (Go, 2016; Weiss, 2005, 2017). This requires an answer to the question as to the specificity of the entities between which there are borders. In what way are these entities societies? Are societies best understood as a set of institutional domains or as a set of relations of inequality – or can they be both? How can the varied spatial and temporal reach of these sets of social relations be theorised? How should the concept of society be developed to respond to these challenges? How do new ways of thinking of systems as complex systems assist the rethinking of the concept of society?

Addressing the challenge of theorising the non-alignment of social relations requires addressing both institutional domains and relations of inequality. Although much development of the concept of ‘society’ in sociology draws on the joint classical heritage of Marx, Weber and Durkheim, divergent approaches to society can be distinguished in contemporary sociology. Two key approaches are: institutional differentiation and relations of inequality (Schwinn, 1998), which concern different approaches to social structure: institutional and relational (López and Scott, 2000). Theories of modernity focus on institutional differentiation (Beck, 1992; Eisenstadt, 2002; Giddens, 1990; Luhmann, 1995), building on Durkheim and sometimes Weber. World systems theory focuses on relations of inequality (Arrighi, 2007; Chase-Dunn, 1988; Dunaway, 2014a, 2014b; Wallerstein, 1974), building on Marx and Weber. The first understands variations of modernity linked to development and differentiation (Beck, 1992; Giddens, 1990; Luhmann, 1995), including multiple modernities across space (Eisenstadt, 2002). The second approach articulates the concept of society as a world system (Wallerstein, 1974) or global social formation (Chase-Dunn, 1988), with a global system of relations of inequality, and with position within the world system depending on location at its core or periphery and on the nature of the capitalist hegemon (Arrighi, 2007). Each approach offers something important for the concept of society, but also has limitations. Both have been challenged to address a wider range of inequalities by post-colonial and feminist research (Alatas and Sinha, 2017; Bhambra, 2007; Boatcă, 2015; Go, 2016; Rodriguez et al., 2016).

Ideally, the concept of society should encompass both the traditions of institutional differentiation and relations of inequality. It should combine these two approaches in order to have the theoretical capacity to address change in both institutional configurations and structural inequalities. How this can be achieved is core to this article. The significance of institutional differentiation is developed through the concept of institutional domains, of which four major forms are identified: economy, polity, violence and civil society. The significance of relations of inequality is developed through the concept of regimes of inequality, of which major forms include: class, gender and ethnicity. Both institutional domains and regimes of inequality are needed.

The concept of society is challenged when differentiated institutions and multiple relations of inequalities are not spatiality aligned. The revision of the concept of society needs to address three challenges simultaneously: the differentiation of institutions, relations of inequality, and spatiality.

The way these challenges to the concept of society are addressed is affected by the way the concept of system is deployed. Traditional concepts of system make it hard to address these challenges. The way forward is to develop the concept of system used in sociology by drawing on developments in complexity science. Complexity science offers the conceptual tools needed to rebuild the concept of system in sociology. This article offers a rebuilt concept of systems for this purpose.

The new concept of social system, which draws on complexity science, is deployed to develop the concept of society. This development includes the concept of societalisation. Societalisation is a process of moving towards the greater alignment of social systems. However, the societalisation process rarely reaches the full alignment of institutional domains and regimes of inequality in time and space that is implied by the concept of society.

The article introduces the challenges of the spatial and simultaneous analysis of differentiation and inequality. It sets out how complexity science can offer a new concept of social system. It investigates the work of modernisation and world systems theory and identifies their difficulty in addressing simultaneously institutional differentiation and relations of inequality. It offers an alternative theory of society using a reworked concept of system that builds on developments in complexity science.

The challenges

Society and the spatial

Society is neither contained in a nation-state, nor has it become fully global. How is rescaling to be addressed in social theory?

The nation-state as society is a myth; but a very powerful myth (Walby, 2003). The equation of nation-state and society has a long-standing and resilient presence in sociology (Bruce and Voas, 2004; Giddens, 1990; Mann, 1997; Meyer et al., 1997). This is the notion of a nation-state/society that is a ‘container’ of congruent, neatly overlapping, institutions of economy, polity, violence and civil society. The critique of this position is variously articulated, including by those concerned with the global reach of capital (Hardt and Negri, 2000; Harvey, 2005; Wallerstein, 1974) and the border/security nexus (Anderson, 2013; Anderson and O’Dowd, 1999; Follis, 2012; Vaughan-Williams, 2015). Writers on Europeanisation engage with similar theoretical challenges, since they also address the incomplete and unstable boundedness of societies where there are several types of non-overlapping polities in Europe, from nations and states to empires to organised religions (Zielonka, 2006). The critique is taken further by post-colonial writers who challenge the centring of modernity on Europe (Chakrabarty, 2000); show that the exploitation and inequalities generated by colonialism extend into the present thereby shaping contemporary modernity (Bhambra, 2007); and argue for the importance of voices and social practices in the South as well as the North (Alatas and Sinha, 2017; Rodriguez et al., 2016).

The COVID-19 pandemic further illustrates the varied temporal and spatial scaling of relevant systems: the virus went global; transmission was to persons closer than two metres; health care systems were often national; death was unevenly distributed across systems of inequality; hospitals were overwhelmed for specific periods of time; the crisis cascaded into the real economy in different temporalities.

Traditional theorisations of society that assume boundaries between social entities in which the spatial, political and societal neatly map onto each other are challenged by the empirical evidence that the world is neither fully divided into separate, discrete bounded societies, nor is it fully globalised in all institutions. Institutions and inequalities are not congruent, fully overlapping, in time and space.

The partial overlap of non-nested social systems, which have different spatial and temporal reach and do not saturate their territory, challenges traditional concepts of society that depend on simple concepts of system (Walby, 2009). The concept system needs to be rethought.

Institutional differentiation and relations of inequality

There is a challenge to address the coexistence of multiple forms of modernity, of multiple varieties of capitalism and other regimes of inequality at the same time in the world. This requires the simultaneous theorisation of differentiated institutions and multiple regimes of inequality, which is rarely achieved. As Schwinn (1998) notes, sociology has tended to address one or the other, rather than both simultaneously: one tradition focuses on the differentiation of institutions; the other focuses on inequality. Modernisation theory addresses the differentiation of institutions with nuance and care, but tends to be vaguer on issues of multiple inequalities. World systems addresses inequalities robustly in relation to class, but narrows the focus on institutions to prioritise political economy. Post-colonial theory tends to address global inequalities through the lens of culture.

Both the major approaches to differentiation and inequality are challenged to address multiple intersecting inequalities linked to (post-)colonial relations and gender (Boatcă, 2015; Chakrabarty, 2000; Chatterjee, 1986), let alone address these simultaneously with the differentiation of institutions. Chakrabarty argues for the importance of colonial and thus global relations in thinking about what constitutes society, and for displacing Europe from the central point of reference. The challenge to address multiple intersecting inequalities (Crenshaw, 1991; McCall, 2001) can lead to a focus on the micro and on performances (Butler, 1990) that is insufficient to address issues at a global level. The concerns in both approaches need to be addressed in the conceptualisation of society.

The concept of system

The concept of system underlying these debates on ‘society’ is part of the problem. The traditional concept of system underlying the concept of society is too simple; it makes the theorisation of the global processes, major social changes and multiple inequalities very difficult. The traditional notion of a modern society, as a nation-state, assumes a simple notion of a system in which the parts (economy, polity, violence and civil society) make up the whole and share the same boundaries in time and space (Giddens, 1990; Mann, 1997; Meyer et al., 1997). This is challenged in a global era, since it becomes clear that these boundaries are not the same. How did the notion of society as a social system deal with this challenge?

One response to this challenge to conceptualising society as system was a retreat from the macro-level concept of society as a social system in a poststructuralist and postmodern move (Latour, 2005; Lyotard, 1978) and the development of concepts of ‘network’ (Castells, 1996) and ‘assemblage’ (Deleuze and Guattari, 1987; Latour, 2005) as alternatives. The rejection of the concept of society by many poststructuralist and postmodern writers was bound up with the rejection of the traditional concept of system (Deleuze and Guattari, 1987; Latour, 2005; Lyotard, 1978). Attempts to use mid-level concepts to reach for the global, such as that by Quark (2013), while richly informative, can similarly leave behind the concept of society. The ambivalence towards the concept of social system that is similarly found in the post-colonial literature is attributed by Go (2016) to the influence of the humanities disciplines.

It is argued here that these alternatives to the concept of society, such as network and assemblage, are insufficient to capture the large scale, depth and systematic interconnectedness of social institutions and social inequalities over space. Some other conceptualisation is needed. In this article, the concern is not to reject the concept of society, but rather to revise it. This includes reforming the concept of system that underpins the concept of society.

The cross-disciplinary field of complexity science potentially offers the concepts needed for this reform (Maturana and Varela, 1980; von Bertalanffy, 1968). Some of these advances in the theorisation of systems have been noted and applied in social science (Luhmann, 1995; Pierson, 2000; Urry, 2003; Wallerstein, 1974), but not with sufficient depth on the issues concerning the simultaneous theorisation of institutional domains and regimes of inequality.

The way forward: Rethinking the concept of system underpinning the concept of society

Sociology needs to rethink the concept of system in order to rebuild the concept of society. Complexity science has redeveloped the concept of system in the sciences. It offers a conceptual toolkit with which sociology can rebuild the concept of social system that underpins the concept of society (Capra, 1997; Castellani and Hafferty, 2009; Eldredge, 1986; Maturana and Varela, 1980; von Bertalanffy, 1968). An improved concept of social system can be rebuilt using these innovations in the concept of system, including: system reproduction, a system and its environment, non-saturation, multiple equilibrium points, path dependency, positive feedback, non-linearity, sudden as well as gradual change.

A core notion of systems theory is the autonomous reproduction of systems by themselves (Maturana and Varela, 1980). Systems are self-organising. They do not require additional input to reproduce over time. The term ‘system’ is a generic term, that can encompass a wide range of types of systems. In sociology, ‘institutions’ can be seen to be systems, as are ‘regimes’. The concept of system can be deployed at different levels of scale: a home is a system; a school is a system; a gender regime is a system; employment is a system.

Each system takes all other systems as its environment (von Bertalanffy, 1968). This conceptualisation replaces the notion that ‘parts’ make up a ‘whole’. It provides much greater flexibility for analysis without loss of the concept of system. Social systems can be overlapping, and non-nested, and not saturating any given territory. This approach to the relationship between systems means it is possible to theorise social systems as having different temporal and spatial reach. For example, an economy may have a greater spatial reach than a polity; a polity (e.g. the Catholic Church) may have greater temporal reach than other polities in its territory (e.g. a secular state); finance capital has a different temporal and spatial reach than industrial capital; varieties of gender regime may not map exactly onto varieties of capitalism.

Systems do not necessarily saturate a given space, in complex systems analysis. There can be multiple systems having effects in the same space. This is theoretically possible because in this approach, each system takes all other systems as its environment. This allows for the analytic separation of institutional domains and regimes of inequality in the same space. This enables the theorisation of multiple intersecting regimes of inequalities without identifying or reducing each one to a separate institution (Walby, 2007). For example, multiple regimes of inequality of class, gender and ethnicity can all structure the same institution of employment. This theoretical capacity to distinguish between institutional domains and regimes of inequality is only possible with this new way of thinking about systems. Since each system takes all other systems as its environment, the restrictiveness of the old concept of system is left behind. There is no longer a difficult choice between a theory of society based on institutional differentiation and one based on regimes of inequality: sociology can have both.

There are multiple possible points of equilibrium of a system in complex systems theory. This replaces the old notion that a system has only one point of balance, of equilibrium, to which it will tend. Multiple equilibria mean it is possible to theorise multiple modernities (Eisenstadt, 2002), varieties of capitalism (Hall and Soskice, 2001), and varieties of gender regime (Walby, 2009). It is possible to reject the notion that there is one universal form while retaining the explanatory power of the concept of system. There is no longer a difficult choice between the explanatory power of the concept of system and the downside of a false universalism: sociology can have both.

The replacement of the notion of a single point of balance with that of multiple points of balance is linked to the notion of multiple paths. Rather than a single path of development, there may be several path-dependent trajectories of change of a social system (Pierson, 2000). This means that the conditions at the starting point are particularly important; for example, that the path-dependent development of capitalism in different locations is shaped by events in its early history; or utilisation of systems of enslavement in early years has consequences for the ethnic structuring of employment relations in later years. There is no longer a difficult choice between the explanatory power of the concept of system and the historical nuance of different paths of development: sociology can have both.

Sudden as well as gradual change can be theorised in complex systems analysis. Social systems can have positive as well as negative feedback loops (Arthur, 1994). While a negative feedback loop can restore a system to equilibrium, positive feedback loops move it further away from equilibrium. This addition means that it is possible to abandon the notion that social systems must return to equilibrium without abandoning the concept of system. Thus, this problematic feature of ‘functionalism’ is overcome. It facilitates the theorisation of economic and financial instability (Minsky, 1986) and of crisis (Walby, 2015), where positive feedback loops can drive the financial system away from equilibrium. There is no longer a difficult choice between the explanatory power of the concept of system and the recognition of instability: sociology can have both. Non-linear as well as linear change is possible in a system. A small change may have a large effect; cause and effect need not be proportionate in a direct linear fashion. As systems interact, they mutually adapt; this can produce non-linear changes that are more complicated than proportionate unidirectional change (Urry, 2003). For example, the nature of human-produced climate change is not linear. Changes in systems can be sudden as well as gradual (Eldredge, 1986). There may be tipping points (crises, revolutions). Change may occur in waves, rather than in linear forms of gradual accumulation (Walby, 2009). Both feminist and far-right projects have taken the form of global waves, with varied consequences as they engage with societies. There is no longer a difficult choice between the explanatory power of the concept of system and the recognition of sudden and uneven change: sociology can have both.

The variable inclusion of these developments in systems theory has implications for the concept and theory of society, as discussed below.

Existing approaches to society and system

Existing approaches to society and system are reviewed below in order to show the implications of the variable inclusion of developments in systems theory for the concept and theory of society. The problems generated by the old notion of system are contrasted with the ease of theorising using the new complex systems approach.

Differentiation and modernities

In modernisation theory, there has often been an assumption of bounded societies, in which social structures are mapped onto each other in the same space and time. As societies develop, these institutions differentiate and change accordingly. Some accounts theorised a trajectory of change over time, along which most societies were expected to develop. The classic models were developed from Durkheim – with recent divisions in the tradition into several strands.

Modernity is a key theme in the work of Giddens (1990) and Beck (1992) who, significantly influenced by Weber, developed accounts of changes in the form of modernity over time, distinguishing between late, second and reflexive varieties. In many texts on modernisation, there is an implicit or explicit tendency to assumptions of universalism, with references to spatial specificity quite rare. Sometimes this is explicit, as in the work of Domingues (2012), who writes of modernity as a global process, even though he also provides empirical accounts of regional variations. Most often the institutions considered to be central to modernisation are limited to either the economy or culture; but this is debated. Mann’s (1986) work is an advance in adding the military to the trilogy of economy, state and ideology as institutional sources of social power. However, despite his (Mann, 1986) analysis of empires in earlier eras, Mann (1997) defends the conceptualisation of society as a nation-state in the current one, and considers the lack of alignment of military and economic power in the US as incoherent and exceptional (Mann, 2003). While inequality is noted in this body of work, the differences between forms of social inequality are little specified, with a tendency to understand inequality through the lens of softly defined class relations. There is little explicit discussion or development of the concept of system underpinning this concept of society.

A second strand, in which Luhmann (1995) is central, developed a fine-grained account of the complex differentiation of subsystems of the social system, with a focus on the logic of communication. Luhmann’s work developing the concept of society as a system was informed by the developments in systems theory complexity science, especially by Maturana and Varela (1980). This enables Luhmann to avoid the assumption that systems are nested and congruent. Potentially this approach to systems allows for greater flexibility and nuance in the theorisation of society over space; but the level of abstraction at which his theory is pitched means that he does not utilise this advantage. The analysis of inequality is largely absent. The narrow focus on the logic of systems through communication leads to a focus on ideational issues and a neglect of power, especially in its economic forms.

A third strand differentiates multiple modernities thereby allowing for variation in the form of modern societies over space (as well as over time) (Eisenstadt, 2002). The multiple modernities approach is part of a wider post-colonial or de-colonial tradition that displaces Europe from the centre of analysis of the world and modernity (Boatcă, 2015; Chakrabarty, 2000; Chatterjee, 1986). Multiple modernities are entangled with complex effects on each other, not least due to colonial and post-colonial forms of inequality. Chakrabarty (2000) is not against universalism, but rather argues that what counts as the universal is unstable, always already inflected by its actual location and the histories linked to this. Potentially, the multiple modernities approach offers a framework that allows for differences in relations of inequalities and in differences in institutional domains at the same time. However, in practice much analysis is focused rather narrowly on values and culture, leaving out of focus the significance of economy, polity and violence. Despite the interest in new forms of inequalities of the post-colonial era, this rarely extends to the full range of inequalities.

In conclusion, the differentiation approach only partially meets the challenge of border studies and globalisation to the concept of society. This is largely because of the difficulty of addressing differentiated institutions and multiple inequalities at the same time. The texts that are the strongest on institutional differentiation are the weakest on inequalities. The differentiation of institutions is the basis of societal distinction, ranging from a soft political economy inflected with science (Giddens, Beck) to strong culturalism (Luhmann, Eisenstadt). The differentiation of institutions is most developed in the systems theory of Luhmann. The analysis of the full range of inequalities is underdeveloped, either generic or soft class (Giddens, Beck), or almost absent (Luhmann, Eisenstadt).

Modernity theory deploys a range of approaches to the concept of system. The simple old concept of system, which assumes that the whole is made up of parts, is largely adopted by Giddens, Beck and Mann, limiting their capacity to address both institutional differentiation and multiple inequalities. Eisenstadt adopts one innovation in systems theory by using the notion of multiple equilibrium points to support his analysis of multiple coexisting modernities, but does not take it much further. Luhmann adopts a wider range of complex systems concepts, but stops short of applying them to the analysis of inequalities. Potentially, Luhmann had the conceptual tools in his systems theory to go further, but he stayed within the tradition of institutional differentiation rather than meet the challenge to analyse multiple relations of inequality as well.

The weak conceptualisation of system in the first and third strands makes the theorisation of both differentiated institutions and multiple inequalities difficult to achieve. While the second strand, Luhmann, uses a concept of system that has the potential to enable both to be addressed, he does not.

Relations of inequality and world systems

Relations of inequality are at the centre of the world systems approach to society. The world systems approach to the challenge of globalisation to the concept of society starts from the opposite direction to the modernity theorists discussed above. It assumes that society, or rather social formation, is already global. The frontiers between countries are not borders between societies, merely lines that divide between some forms of regulations. World systems theory addresses how the parts of a single world social formation are connected, rather than conceiving this as the relations between multiple different societies. In this approach, the analysis of inequalities drives the analysis of institutions and the analysis of space.

World systems theory, centred on political economy, is developed by Wallerstein (1974), building on the heritage of Marx, Weber and Braudel, and developed further by Chase-Dunn (1988) and Arrighi (2007). In world systems theory, class is the most important relation of inequality. Processes of capital accumulation shape the main social relations, which are defined by class positions. Capital accumulation is a global process, not centred in individual countries. Value is extracted during production and moves along a chain from the periphery towards the core. At each link of the chain of commodity production, local class relations structure this process and assist the appropriation of value. Profit occurs when specific forms of capital gain a monopolistic position, giving it power to shape markets in its interests.

World systems analysis struggles to fully address relations of inequality other than class in its theory, despite repeated empirical recognition of their existence. This is because of a narrow interpretation of the process of value extraction, tending to limit this to the point of production under market relations; and the tendency to attribute the central dynamics of change to processes of value extraction. Nonetheless, there are significant attempts within this framework to address challenges concerning gender and colonial inequalities (Boatcă, 2015; Dunaway, 2014b; Moghadam, 1993).

For Wallerstein (1974), racism and sexism are located in ideology, not political economy. On ethnicity, although the exploitation of the periphery by the core is fundamental to relations of ethnic inequality on a world scale, he understands this appropriation through the concept of class rather than ethnicity. This is in contrast to Rex (1973), who understood the political economy of empire as constructing ‘race’. On gender, although Wallerstein notes gender inequality as significant, he considers that households ‘pool’ their income from various sources rather than being sites of gendered exploitation (Wallerstein and Smith, 1992). He does not conceptualise gender relations in the household as ones within which there is an appropriation of value.

Although ethnicity and gender are left to culture and to the margins in most world systems analysis, this is not an inevitable feature of world systems analysis, merely of its dominant texts (Boatcă, 2015). For example, Dunaway (2014a, 2014b) analyses the transfer of value from women in her gendering of commodity chain analysis, deftly showing how this is a significant source of value for global capital and transferred from one location to another. This is part of a long tradition of analysis of political economy of gender at a global scale (Mies, 1986; Moghadam, 1993; Schreiner, 1911), but it remains marginalised within world systems theory.

World systems theory has a focus on political economy, tending to leave civil society and violence out of focus. Civil society, culture and ideology are treated as if they are simultaneously marginal and the main site of generation of gender relations. Violence and war are largely treated as epiphenomenal, being generated by moments of crisis in the political economy. States are primarily understood as nodes in the capitalist system.

Engagement with and within the field of world systems analysis on the issue of the relative importance of institutional domains other than political economy has implications for the notion that there is a single world system. The post/de-colonial critique reframes issues of culture (Boatcă, 2015), simultaneously raising its importance within social theory and questioning whether there is a single world system. Debate on the conditions of generation of counter-hegemonic movements reframes issues of civil society since this potentially has the power to transform the system (Chase-Dunn, 1988). The regulation school of capitalist development identifies a range of institutions that establish the conditions for a regime of accumulation (Aglietta, 1979), which can then potentially vary between regions. The outcome of processes of contestation over global standards for the regulation of specific industries is shaped by multiple processes (Quark, 2013). The concept of competing hegemons challenges the notion that there is a single global hegemon for a single world system (Arrighi, 2007; Pieterse, 2018). The notion that more than one variety of capitalism can coexist in time in different spatial locations challenges the notion of a single world system. The assumption of the congruence over time and space of political economy with other social institutions is strained to breaking point and beyond.

The challenge of cross-border studies to traditional concepts of bounded societies is met in world systems analysis by downgrading the concept of border from a boundary between societies to a line that separates regulatory regimes of minor significance. In world systems analysis, the world system is the unit, not individual countries. The connections are a consequence of processes of capital accumulation, linked in detail by commodity chains that straddle the world. There are spatial differences, not least between core, semi-periphery and periphery; countries may move between these levels.

But when the focus of attention moves beyond a narrow focus on capital accumulation, the single global unit is challenged. Competition between hegemons implies a plurality of regimes of accumulation. The search for counter-hegemonic movements is obliged to engage with varied civil societies. Wallerstein’s version of world systems theory overstates the extent to which political economy saturates its global space. The conclusion drawn here is that while some aspects of capital accumulation are near-global, other institutions in the social formation are not.

While Wallerstein has welcomed developments in the concept of system that are informed by complexity science, and uses some concepts as metaphors in his work, including on crisis, his application of these new concepts of system are uneven and not used to their full potential.

World systems theory mistakenly theorises political economy as spatially and temporally congruent with systems of civil society and of violence. Yet, complexity concepts of system would permit the theorisation of the different temporal and spatial reach of different institutions, since it does not require nesting or congruence of systems. World systems theory has not adequately theorised the multiple inequalities within political economy, despite noting this empirically, relegating race and gender to institutions of culture and ideology. The assumption is that the institutional space of political economy is saturated by class relations and cannot allow relations of ethnic, colonial and gender inequality determining power. Yet complexity concepts of systems do not insist on such notions of saturation, and permit theorisation of multiple intersecting inequalities.

In conclusion, world systems theory meets the challenge of cross-border studies to conventional distinctions between societies by treating the world system as one society, thereby demoting the significance of border to a minor status. Yet, while finance capital probably is global, industrial capital retains regional and national roots, political institutions are positioned at national levels as well as the global, and violence and civil society have multiple spatial levels. There is varied temporal and spatial reach to these different institutional domains. While capital accumulation is a strong driver of global social relations, it is not the only one. World systems theory meets the challenge of treating relations of inequality as a core aspect of its theory of society, but it is limited in its relation of non-class relations of inequality to a narrow set of institutional domains that are regarded as of minor importance in driving the system forward. The utilisation of complexity conceptions of system could facilitate the incorporation of these issues into the core of its theoretical framework.

A new approach to society and system

The new approach to conceptualising society proposed here integrates the analysis of differentiated institutional domains and multiple regimes of inequality. It allows for the different spatial and temporal reach of each of these institutional domains and regimes of inequality, while still retaining the concept of society as a system. It draws on the development of the concept of system in complexity science. It introduces the concept of societalisation.

Societalisation is the process during which institutional domains and regimes of inequality tend towards alignment. There are competing projects to societalise around a preferred set of social relations. If societalisation were to be completed, then there would be a fully-formed society. Societalisation is rarely completed, so there are rarely fully-formed societies. The concept of society is an important benchmark for social theory, even though it is rarely realised. Societalisation is a project. It is often taken forward as a ‘national’ project. There are also non-national societalisation projects, including those of the European Union, universal human rights and socialism. A societalisation project may capture a state as a vehicle to advance its claims. It engages in a process of attempting to reorganise and align further social institutions within its vision.

The concept of society developed here includes both institutional domains and relations of inequality. The tendency to separate these two kinds of social structure into two separate theoretical schools of thought is a detriment to the development of sociological theory. The two schools need to be hybridised. The way forward is to understand that both institutional domains and regimes of inequality are systems, though of different types, and both can be included within the concepts of social system, society and societalisation.

It is necessary to theorise simultaneously both multiple inequalities and multiple institutional domains. Both the major approaches to society are challenged by this requirement. They address it by segregating each form of inequality into a different institution. Typically, class is seen as based in the economy, and gender in culture or ideology. The post-colonial critique tends to have a narrow focus on culture, even when its ambitions are larger. When the implicit concept of system adopted is the old and simple one, this is not a surprising outcome. However, the assumption that each set of social relations of inequality (class, gender, ethnicity) is lodged in its own institution is unsustainable when the significance of multiple institutions for each regime of inequality and multiple regimes of inequality for each institution is realised. Rejecting this unsustainable assumption challenges the ‘container’ notion of a society as made up of congruent systems of inequalities. Gender inequalities do not map directly onto class inequalities: variations in gender relations do not exactly follow variations in class relations, although they intersect (Crenshaw, 1991; Walby et al., 2012).

It is necessary to conceptualise both the range of institutional domains (economy, polity, violence, civil society) and the range of relations of inequality (class, gender, ethnicity and more) simultaneously while not reducing one to the other (Walby, 2009). It is necessary to allow for the conceptualisation of institutional domains (economy, polity, violence, civil society) and regimes of inequality (class, gender, ethnicity) that are partially but not completely overlapping, non-nested, non-congruent, with different spatial and temporal reach, while allowing for the analysis of their interconnections. This requires a new approach to the conceptualisation of ‘society’ through a revised concept of social ‘system’, as developed here.

The analysis of the non-congruence of institutions (that is, systems) of economy, polity, civil society and violence in space and time is better addressed within a theoretical framework in which any one system takes all other systems as its environment, rather than within the framework in which parts of a system are considered to make up a whole system. This means that each of the systems of economy, polity, civil society and violence can be overlapping, and non-nested, and not saturating any given territory, with different temporal and spatial reach.

The analysis of the non-congruence of regimes (that is, systems) of inequality including those concerning class, gender and ethnicity is also better addressed within this theoretical framework in which one system takes all other systems as its environment; where systems can be overlapping, non-nested and not saturating of their space, and have different temporal and spatial reach.

There are rarely if ever ‘societies’ in which there is congruence (full alignment) of institutional domains (economy, polity, violence, civil society) and regimes of inequality (class, gender, ethnicity, nation). However, there are processes of societalisation in which there are tendencies towards such a congruence, which, in practice, are always interrupted. There are multiple attempts to produce such alignment, as for example in nationalist projects that seek a state of their own, and the current project of Europeanisation. There are greater and lesser degrees of alignment of institutional domains and regimes of inequality. While there are some tendencies towards societalisation in which social processes move towards sharing the same boundary, other contesting processes, including rival projects, mean that full alignment hardly ever occurs. There is currently no actually existing ‘society’ in which all social processes share the same boundary in time and space.

A project is a set of practices oriented towards change in society. While usually rooted in civil society they can have ambitions for any or all institutional domains. A project will have a set of values around which society should be organised and engage in practices to attempt to realise them. Examples of projects include nationalist projects that seek a state of their own to realise their goals. There are feminist projects that attempt to reshape gender relations and anti-racist projects that attempt to reshape society. There is a European project centred on the EU, which has sought peace in Europe. Projects may be local or national or global in scope. The practice of attempting to take control of institutions such as the state is a route to obtain the power to pursue the reshaping of society. It is rare that one project successfully monopolises all levers of power. Multiple projects coexist and compete. The process towards societalisation led by one project is often interrupted by another before it is complete.

The concept of border draws attention to the difference in content between two systems. While traditionally the concept of border necessarily implies something spatial, that is not the intention here (Appadurai, 1990). There are borders between institutional domains of economy, polity, violence and civil society; and between regimes of inequality of class, gender and ethnicity. All these types of border matter, not only the traditional type of border as a political border that maps onto physical territory. The spatial is only one type of border. The concept of space is ambiguous as to whether it necessarily always has a physical referent. Differences may be spatialised or not. There is restructuring of the relationship between social categories and physical space. Sometimes the concept of border contains a partial and contested spatial component. During Europeanisation, political borders have been partially de-territorialised, for example, in the changing EU regulations applied to migrants, their movement and their rights (Follis, 2012; Vaughan-Williams, 2015). The intention here is not to deny the significance of physical space for some matters, while dislodging any definitive and absolute alignment between social and spatial categories. The priority concern is with social differences not spatial differences.

The concept of ‘society’ is useful as a benchmark, even if full alignment between social systems is rare. The process of ‘societalisation’ is common and important. It can change in different directions at different times.

The traditional concept of society assumes that all types of border are congruent; that variations in the form of society affect each of these institutional domains and regimes of inequality simultaneously in time and space. Empirically, full alignment is rarely if ever completed. Rival projects interrupt projects of societalisation. This traditional static concept of society needs to be replaced by this more fluid understanding of the processes of societalisation.

One of the key problems in reconceptualising ‘society’ to meet the challenges of globalisation and multiple inequalities lies in the concept of system deployed, implicitly or explicitly. Rethinking the concept of ‘system’ is critical to rebuilding the concept of ‘society’ in sociology and other social sciences to theorise contemporary social change. There are challenges to theorise the range of institutional domains and the range of regimes of inequality without reducing one to the other and to address variations in the temporal and spatial reach of each of these. This requires a more nuanced concept of ‘system’ than has traditionally been used. The problem with the old concept of system is that it implies the nesting of the parts of a system within a whole, and a self-equilibrating system with feedback loops that return a system to stability after a disturbance.

Drawing on this reinvigorated concept of system, it is possible to develop the concept and theory of society that meets the challenges of space, differentiated institutions and relations of inequality. These developments in systems theory enable the theorisation of a wider range of social processes than are possible with the simple traditional concept of system. They enable more nuanced treatment of the relationship of the social and spatial when the assumption of ‘saturation’ is removed. There is experimentation with the implications of some or all of these theoretical innovations in systems thinking across a range of social sciences including economics (Arthur, 1994), political science (Pierson, 2000) and sociology (Luhmann, 1995; Urry, 2003; Walby, 2007, 2009, 2015).

Conclusion

This article has developed the concept of society by applying the new ways of thinking about systems found in complexity science to the concept of social system. This theoretical development is needed to address the challenges posed by globalisation and Europeanisation, which are indicated in the tension between ‘theories of society’ and ‘cross-border studies’. This requires the analysis of both multiple institutional domains and multiple relations of inequality simultaneously rather than only one. The multiplicity of institutional domains is needed to avoid simple reductionism to one dominant institution, such as culture. The multiplicity of relations of inequality is needed to avoid reduction to one dominant relation of inequality, such as class. The empirical evidence shows that these do not map onto each other in time and space. Rather they have different temporal and spatial reach. This entails hybridising the two sociological traditions that have focused on differentiated institutions and on relations of inequality, rather than treating them as alternatives. The more nuanced and flexible concept of system available in complexity science is needed for this development.

Including both multiple institutional domains and multiple relations of inequality in the concept of society while recognising that they do not map onto each other in time and space stretches the old concept of social system beyond breaking point. However, the response should not be the abandonment of the concepts of system and of society; rather it should be the development of the concept of system and of society. The conceptual toolbox for the development of the concept of system is available in complexity science.

Complexity science offers a series of developments in the concept of system that enable the development of the concept of society. A key development is the notion that a system takes all other systems as its environment, rather than a system being made up of parts. This conceptual device enables both flexibility and nuance, while retaining the capacity to theorise. Other developments include concepts of multiple points of equilibrium (rather than one), path dependency (rather than one universal pathway), positive feedback loops that drive a system further from equilibrium (rather than only negative feedback loops that restore equilibrium), which facilitate the theorisation of the development of multiple coexisting varieties of modernity.

An important addition to the theoretical vocabulary is that of societalisation, a process of moving towards the alignment of institutions and relations of inequality, although the reverse may also occur. Nation-state building, Europeanisation and globalisation are all examples of processes of societalisation, in which there is building of differentiated institutions and relations of inequality that advance towards congruency. However, full alignment is rarely achieved. And processes can also move away from such alignment. Retaining but rethinking the concept of society is important for the analysis of contested developments in borders. With these theoretical developments, it is possible to retain the concept of society for productive analytic work in sociology and other social science disciplines.