The normativity of democracy

Two senses of normativity ¹

According to a widely shared view, a concept is normative when it provides a benchmark against which to assess actions, events, or states of affairs. Stephen Darwall captures this basic understanding when he writes that: ‘[s]omething is said by philosophers to have ‘normativity’ when it entails that some action, attitude or mental state of some other kind is justified, an action one ought to do or a state one ought to be in’ (Darwall, 2001). This conception is generally formulated in terms of norm compliance, where compliance may come by degrees and describes the adequacy of a piece of reality to an external standard fixing, the condition in which that piece of reality should find itself. Its basic intuition is that a normative concept provides a standard against which a portion of reality can be assessed. In the domain of politics, to which the concept of democracy belongs, the concepts of pleasure, utility, equality, freedom, justice and domination are normative in that sense. Indeed, each provides a standard against which the value or appropriateness of an action, event, or state of affairs can be assessed in terms that are politically relevant. For example, when we say that an institution is organized according to criteria of social equality, that a regime is just, that a constitution respects human freedom, or that a social arrangement enables individuals to pursue their utility or their happiness, we are using these concepts in a normative way, that is, we use them to assess an action, event, or state of affairs and to assign it positive value. Democracy is clearly a normative concept in that sense. When we say that a regime, an organization, or a procedure is ‘democratic’ or ‘undemocratic’, what we mean is precisely that this regime, organization, or procedure is politically good or bad, and the concept ‘democracy’ specifies in what sense good or bad. Let us call this first conception ‘norm compliant’.

There is also a second major way in which a political concept can play a normative function. According to this second view, rather than defining a benchmark with limited object validity, a normative concept provides a general framework for interpreting human reality. When used in this way, a concept has the capacity to provide a comprehensive and overarching interpretive framework which accounts for the whole of political reality as well as other major normative concepts. This second sense of normativity cannot be reduced to a merely larger form of norm-compliance. A political concept with paradigmatic normative scope is a concept that aspires to offer a comprehensive account of the largest possible portion of human experience in terms that are politically relevant. When a concept succeeds in playing this normative function, it exercises a specific form of intellectual orientation. More than merely setting a standard against which events or states of affairs can be assessed, it helps to structure a broad domain of experience and organize its overall normative content. Owing to its wider scope, I propose to call this second conception of normativity ‘paradigmatic’.

As I will explain in the next section, a political paradigm normative concept is (a) primitive and (b) has a scope that covers, potentially, the whole domain of politics. As used in this text, the term paradigm is similar to Thomas Kuhn’s use, in which it means a complex and multidimensional interpretative framework through which a large portion of reality is interpreted, and which consists not only of theories but also of explanatory examples, protocols, and methods of empirical inquiry. Throughout the current text, paradigm and norm refer to the status a political concept is given within a theory or discourse. The terms paradigm and norm do not refer to intrinsic properties of concepts but rather to the pragmatic dimension of their use.

If we look at the history of political ideas, we observe that few concepts have historically exercised a paradigmatic normative function. If we look at modern and contemporary western political thought, it seems that only freedom, justice and non-domination have reached this status. ² Other political ideals such as equality and fraternity do not seem ever to have come close to it. My contention is that neither has democracy achieved the status of a paradigmatic normative concept. This fact has negatively affected our understanding of democracy itself.

The main reason for introducing the distinction between norm-compliance and paradigm normativity is to stimulate a reflection on the fact that, while the concept of democracy has been widely used in the last 30 years, even inflationarily, its normative potential in guiding our theoretical and practical reflections upon how to live together has remained overshadowed by the much wider reach of the concepts of justice or non-domination, which have gained a much broader explanatory scope and have generally succeeded in reducing the normative core of the concept of democracy to their own terms, as the works of such as John Rawls and Philip Pettit show. A way of re-describing this empirical fact is to say that, whereas justice and non-domination have largely been acknowledged and used as normative concepts in both senses, the concept of democracy has been de facto conceived and used only as a normative concept in the norm-compliance sense. As a consequence, while the last half century has seen a stunning proliferation of norm-compliance theories of democracy, and while some of them, such as participatory, associative and deliberative conceptions have achieved significant results in developing a wider understanding of the normative scope of the concept of democracy, very little has been done to assign democracy the full status of a paradigm normative concept.

This situation has produced two interrelated and yet distinct orders of consequences which, taken together, have significantly undermined the theoretical potential of the concept of democracy. The first consequence has been the mainstream tendency to explain the normativity of democracy in largely reductionist terms. The second consequence has been either the discouragement of attempts to develop theories of democracy the scope of which would be as wide as that of theories of justice or non-domination or the symmetric development of disheartening accounts. John Dunn has notably recommend that we resort to a more modest, parsimonious and timid reliance on the concept of democracy. As he sorely comments,

The predication of democracy has become overwhelming sedative and disinformative. To block its narcotic impact and restore some clarity, if only at the level of authorization, to what is going on, we would have to relearn our verbal habits and reconstruct our thinking quite fiercely. Under careful scrutiny, that pseudo-democratic provenance rests on nothing more than a feeble pun. […] To block the anesthetic effect of pseudo-democratic authorization of almost everything the United States or any of its contemporary counterparts does as a state, we need to recognize that democracy cannot plausibly be predicated of particular decisions at all except under conditions so restrictive that they can scarcely be inserted into the ongoing life of a state (Dunn, 2014: 147–148).

The upshot of these two tendencies has been that the normative potential of the concept of democracy has been diluted, since it is always reduced to other concepts considered more primitive. In addition, its normative scope has been narrowed down, mostly to the domain of the theory of government (Dunn, 2014: 142).

Therefore, developing a paradigm-normativity conception of democracy is required if we wish to understand the full normative potential this political concept can deliver.

Primitiveness and rival non-reducibility

Normative concepts used as paradigms share a tendency to join with one another in similar relational patterns which can be formulated in terms of primitiveness and rival non-reducibility. By saying that a notion is primitive, it is implied that its normative meaning is not derived from other concepts. To say that a concept is primitive means that it has an intrinsic value, by which it is generally meant that its realization is not pursued for the sake of other, higher, ends, but is an end in itself. To say that a concept is non-reducible means that it is genuinely normative, by which we mean that it cannot be explained in terms of other normative concepts.

Primitiveness is a necessary but not sufficient condition for a concept to exercise paradigmatic normativity. Hence, a concept can be primitive and non-reducible yet lack the capacity to exercise ‘paradigm normativity'. For example, the concept of democracy can be said to be politically primitive and non-reducible, yet its scope is limited to the domain of formal political institutions. Primitiveness is necessary because it avoids reduction to supposedly more fundamental concepts. Yet it is insufficient because paradigmatic functioning also requires width of scope. Hence, a concept that has paradigm normativity (a) should be primitive, and (b) should have a scope that covers potentially the whole domain of politics, and not only a limited portion such as the theory of government, as is the case with many theories of democracy. ³ It is indeed in that sense that we can say that a normative concept operates as a paradigm.

The second feature of this view of normativity concerns relations among paradigm normative concepts. Normative concepts that are considered to possess paradigmatic meaning operate as rival and incompatible frameworks in at least two senses: (1) a paradigm concept is generally presented as having sufficient scope to provide a general framework within which a comprehensive normative theory can be articulated, and (2) a paradigm concept is considered to have the normative resources which are needed to define all other normative concepts in its own terms. ⁴ Interpreted as paradigms, normative concepts are characterized by a high level of abstraction and a potentially very wide domain of application. To say that a concept has paradigmatic normative content means, therefore, endowing it with the power of organizing conceptually a large domain of reality, implicitly disabling other concepts from playing the same function.

Before proceeding in our exploration of the philosophical grammar of normativity, a preliminary clarification concerning the exact aim of this exercise is required. It should be clear by now that this discussion is not intended as a contribution to the ongoing debate on the justification of democracy, nor a fortiori to the question of the comparative advantages of instrumental versus intrinsic justifications of democracy, but rather at clarifying the relation that exists between democracy as a primitive concept and democracy as a paradigm concept. In particular, I wish to emphasize that the adoption of a reductionist attitude toward democracy is the outcome of a preliminary – and generally not justified – decision concerning what should be taken to be the paradigm normative concept. It is precisely such a decision that I intend to challenge.

The normative status of democracy has generally been analysed through the lens of two conceptual distinctions, both of which, I will contend, fail to grasp the meaning conveyed by the distinction between norm-compliance and paradigmatic normativity.

The first distinction opposes instrumental to non-instrumental conceptions of democracy. Instrumentalist arguments conceive of democracy as a mere external means to pursue values whose justification is prior and independent of that of democracy. Hence, by definition they fail to understand the distinctive value of democracy as they reduce it to a means to produce other goods. While remaining reductivist, non-instrumentalist or constitutive arguments conceive of democracy as a constitutive component of a larger and higher value. The main difference between instrumentalist and constitutive arguments is that, whereas for instrumentalists the relation between democracy and the good is external, for constitutivists such a relation is internal (Christiano, 2015; Rostbøll, 2014). Hence, while an instrumentalist should be ready to abandon democracy if other means prove more effective in reaching the relevant normative ideal, constitutivists claim that it is only through democracy that other relevant normative ideals can be attained. It should be noted that, whereas the majority of instrumentalist approaches are developed within the framework of a purely political conception of democracy as regime or procedure, constitutivists are generally more open toward wider accounts of democracy. ⁵ Yet both instrumentalists and constitutivists operate under the assumption that democracy is a normative concept only in the norm-compliance sense.

The second distinction we generally find in the literature opposes reductivist to non-reductivist approaches. Reductivists – who include both instrumentalists and constitutivists – admit the primary status of democracy as a political regime or procedure but explain it in terms of other external values it helps to achieve. On the other hand, anti-reductivists claim that democracy is a primitive concept, which denotes something that has to be pursued for its own sake. Anti-reductivist approaches can be divided into two groups. Group A includes approaches that vindicate democracy as a primitive concept on the basis of purely political arguments. ⁶ While theories within this group overcome the major shortcomings of reductivist ones, they still fall short of explaining the paradigmatic normative content of the concept of democracy, as they confine the scope of the concept to political regimes. They are not concerned with democracy as providing the cornerstone of a much larger and more encompassing normative framework, in the same manner as justice and non-domination. Group B includes approaches that explain the normative primitiveness of democracy by scaling it up from the domain of politics to a larger one, generally that of morality. ⁷ Conceiving democracy as a moral idea is a typical move used to justify it in anti-reductivist terms. It is, however, not clear to what extent a moral or social explanation of democracy is really anti-reductivist, as the moral content of democracy is generally explained in terms of supposedly more primitive moral notions such as respect, equality, or freedom.

What in the end is lost through these distinctions is the basic intuition that democracy, like other normative concepts such as justice and non-domination, has a dual status, that is, at the ordinary level of political theory it provides the basis for a political conception of how formal political institutions should be designed. At the higher paradigm level, it provides a larger normative account of how the whole of social life should be organized, therefore offering normative guidelines for the largest possible array of interpersonal relations. Conceiving democracy as a paradigm normative concept provides a third alternative approach, one that shares with group A the ambition to go beyond the merely political dimension of democracy as regime or procedure, and with group B the ambition to provide a fuller explanation of what makes the qualities and institutional features of political democracy distinctively valuable. Yet it does so by correlating them with properties that have a broader theoretical reach.

Therefore, understanding democracy as a normative concept requires that we combine a standard interpretation of normativity as norm-compliance with the idea of normativity as paradigmatic. While defining democracy as a primitive concept accomplishes the first step in this direction, we need also to explain how its scope can meaningfully be extended beyond the domain of political institutions, or political life, to embrace the larger functioning of a society. It is indeed only by conceiving of democracy as a wider overarching framework that we are able to develop a more complete account of the broad social and moral stakes associated with democracy.

One sees immediately that the majority of current theories of political democracy are not suited to this task, as they cannot deliver such a wide view. Historically, the Tocquevillean and the pragmatist traditions are among the few having conceived of democracy as a paradigm normative concept. John Dewey’s conception of democracy as a ‘way of life' and Claude Lefort’s conception of democracy as a form of society are important precursors of the project to which this article contributes.

Before developing at some length this model, in the next section, I will explore the logic of paradigm normative concepts by showing how the concepts of justice and non-domination operate at both normative levels, that is, they provide the normative core of a theory of political institutions, while also identifying positive and highly desirable traits of social phenomena – be they events, situations, entities, or any other ontological types of social existence.

Competing paradigm normative concepts

Introducing the distinction between norm-compliance and paradigm normativity brings to light the apparent paradox according to which, while democracy has achieved unrivaled political standing, its meaning is systematically framed in the theoretical terms of other normative notions such as justice, equality, freedom, or non-domination. Indeed, there is plentiful evidence that political philosophy has rarely considered democracy to be the paradigmatic normative concept assigned to the task of structuring the whole conceptual field of politics. Liberal theory has traditionally assigned this task to the concept of liberty and, since John Rawls, to that of justice. Similarly, republicanism, Marxism and critical theory have tended instead to side with non-domination.

Even if (or when) these traditions are thoroughly committed to democratic ideals and institutions, their allegiance to democracy has been the indirect and derivative result of a prior endorsement either of justice or of non-domination as paradigm normative concepts. As a consequence of this methodological way of proceeding, in all these traditions, the concept of democracy plays only a limited and subordinate role, and tends to be confined to the domain of the theory of government. One can, for example, derive a justification of democracy from Rawls’ theory of justice under the assumption that democracy is the political regime that best fulfils the requirements of a just basic structure of society. Similarly, Pettit has argued that a democratic regime is one that better than any other provides the conditions under which non-domination can thrive. Either way, democracy assumes the status of a derivative, non-primitive normative concept.

Until recently, freedom has provided the mainstream paradigm normative framework for liberal theory. The preservation of freedom, its diffusion, and its custody have been normative goals that the liberal tradition has consistently pursued for more than four centuries. Freedom was (and is) the highest good a political system should deliver, but it is also the presupposition upon which an entire vision of society has been built, that is, providing normative guidance in the domains of personal intimacy, economic exchange, and cultural and spiritual life. From this vantage point, democracy has been seen essentially as a system of government, a regime instrumentally praised for its superior capacity to provide the institutional conditions under which liberty could be achieved in the political domain. ⁸ The appearance of Rawls’ A Theory of Justice significantly changed the situation, determining in particular the fundamental shift from freedom to justice as the paradigm normative concept. Justice has also rapidly acquired the status of paradigm normative category for authors not directly inscribed within the classical liberal tradition, as is the case for critical theory. But scholars such as Ian Shapiro, Michael Walzer, Amartya Sen or Martha Nussbaum have also come to accept justice as the most appropriate paradigm normative concept, significantly extending its reach to all domains of social life. According to these and other authors, justice refers to the ways in which a large variety of social goods including education, health and the quality of the environment are distributed within a society. In an even deeper way, recognition-based conceptions of justice, which have initially stemmed from Axel Honneth’s path-breaking work (Honneth, 1995), have contributed to consolidate justice as a paradigm normative concept which expands the normative reach of the concept of justice to the whole of social life, from intimate relations within the family to those more formal relations made possible by the legal system. Theories of epistemic justice sparked by Miranda Fricker’s seminal work (Fricker, 2007) can be seen as a variant of the recognitional approach to justice, insofar as they similarly insist upon defining justice in terms of human dignity, that is to say, a good that is incompatible with the logic of distribution.

The concept of non-domination has undergone a similar trajectory, progressively reaching the status of a paradigmatic normative concept whose scope is as wide as that of justice. Indeed, for a large plurality of traditions in political philosophy including Marxism, republicanism and critical theory, non-domination provides the normative standard against which to assess the largest possible array of social phenomena. Like justice, non-domination too can in principle be applied to all patterns of human interaction, in the public as well as the private sphere. As Pettit among others has made sufficiently clear, while republicanism originated essentially as a political theory of government, non-domination, ‘as a condition under which a person is more or less immune, and more or less saliently immune, to interference on an arbitrary basis’ (Pettit, 1997: viii), has significantly broadened its scope, as it applies to any form of interaction among individuals. As such, domination and non-domination describe relations that can take place at all levels of social life. They are, in that sense, paradigm normative concepts.

All these paradigm concepts are interrelated, so that each can, at least partially, accommodate the normative requirement set by the others. However, because each one operates as an hegemonic paradigm normative framework, it will also tend to subordinate the other normative categories to itself. While adopting a paradigm normative framework does not require or imply rejection of the others, it will, as a general rule, imply their theoretical subordination. This fact has obvious intended and unintended consequences. With reference to democracy, it is indeed evident that, while it is true that justice and non-domination in their paradigmatic normative function acknowledge the relevance of democracy, the place they assign it is inevitably narrower than the position that a democracy-based paradigm normative framework would allow.

Democracy as a paradigm normative concept

By and large, the western tradition in political philosophy has understood democracy as a political category that describes a regime or a form of government. This conception has dominated political philosophy from Aristotle’s Politics to our times. In the terms I have introduced, the western tradition has denied the concept of democracy paradigm normative status. As a consequence, the concept of democracy has generally been articulated in the terms set by other paradigm normative concepts, be it freedom, justice, or non-domination. Under these conditions, the very idea that democracy could describe a final state desirable in itself has appeared ill-conceived. Contrary to this mainstream understanding, I wish to suggest that democracy can be entitled the same paradigm normative status traditionally endowed upon justice and non-domination, provided it is conceptualized in appropriate ways.

As I have indicated above, a definition of democracy which is suited to fulfil the paradigm normative role demands a high level of abstraction to avoid identification with too narrow cultural or historical content. It also requires a broad range of application so as to fulfil the function of normative guidance that is typical of paradigmatic normative concepts. To function as a paradigmatic normative concept, democracy needs to have a scope that cannot be limited to a sub-set of social phenomena such as the functioning of political institutions or procedures.

A higher level of abstraction and a broader scope can be achieved in two ways. The first consists in identifying democracy with one or more political requirements (principles or procedures) and extending their reach beyond the realm of politics. The second approach proceeds in the reverse order and begins by defining democracy as a social category, one that describes broader societal phenomena of which political democracy is but a specification. This is a more radical and less followed route, one that is, however, more promising as it takes its starting point a dimension of experience that is prior to its articulation in political institutions and procedures. This is the route I propose to follow and will introduce in the next section.

Before examining the social strategy, let me explain why I consider the first alternative to be second rate. Two major strategies have been followed to provide a wider account of democracy based on the extension of its political core. The first focuses upon democracy as a procedure for decision-making – in the manner pursued notably by deliberative and participative theorists – while the second emphasizes democracy as a method for managing power. The participatory conceptions of democracy articulated by Carole Pateman and Jane Mansbridge are an example of the first approach (Mansbridge, 1983; Pateman, 1970). Participatory and deliberative theorists begin by identifying democracy with inclusive and deliberative forms of decision-making, and then proceed to extend the scope of the concept of democracy to all those institutions and situations within which decisions are taken according to democratic procedures. The second strategy, exemplified by Ian Shapiro’s theory of democratic justice (Shapiro, 1999) proceed in a similar way but singles out horizontal management of power relations as the identifying principles whose normative scope should be extended to the whole society. As is known, he notably proposes to analyse family relations and workplace interaction from the perspective of their compliance with an egalitarian norm of power management. Both strategies begin with a political concept of democracy and proceed to extend its reach beyond the realm of formal political institutions to the functioning of the whole society.

From the perspective I’m developing, both strategies are promising and yet still incomplete because unilateral. While indeed participation as well as horizontal power relations belong to the idea of democracy, neither they nor their combination exhaust its content. Theorists following this strategy insist upon the instrumental value of democracy as a means for achieving other ends, in being a subordinate good necessary only insofar as it is the best means to achieve superordinate goods. As Shapiro writes: ‘Collective goals are better pursued democratically than not, and better pursued more democratically than less democratically. In this sense doing things democratically should be thought of as inherently valuable’ (Shapiro, 1999: 24). In the case of Shapiro’s substantial account of democratic justice, the legal or negative dimension of freedom is constantly over-emphasized in a way that clearly reflects a liberal understanding of politics based on the idea of an autonomous and self-determined individual striving to pursue their interests and fulfilling their preferences. This should come, however, as no surprise, as the very method consisting in first taking a political definition of democracy and then extending it to the whole social body inevitably ends up finding in society what had previously been included in the political definition of democracy itself.

While I concur with Shapiro’s claim that deliberation or participation should not achieve the status of final ends, from the vantage point of a social account of democracy his strategy is plagued by the same mistake of participatory accounts, which consists in reducing the social dimension of democracy to the domination of a single political factor, be it participation, deliberation, or managing power.

A similar fate plagues those conceptions of democracy which attempts its theoretical reduction to one of its three constitutive principles, be it freedom, equality, or solidarity. The reason for this is simple, although demonstrating it would require much more space: it is that each of these three principles expresses a distinct feature of the normative core of this concept. While the case of liberty seems more evident in light of the debate having opposed liberalism to democracy – or liberal democracy to social democracy – the recent resurgence of egalitarianism has given new impetus to this reductive move. As I show in the following section, the idea of equality – and hence egalitarianism – accounts only for one of the three principles defining democracy as a paradigmatic concept. One should also consider that those attempt that come closer to reducing democracy to either freedom or equality rely in fact on a much broadened understanding of these concepts, one that in fact is so broadened as to include the other concept. ⁹

As I have pointed out, an alternative and more promising strategy for developing a concept of democracy capable of operating as a paradigmatic normative concept consists in proceeding from a social conception of democracy. This is a less followed route, one that is, however, more promising in that it takes as its starting point a dimension of experience that precedes political institutions and procedures. In following this route, I can rely upon a few historical precursors, among which I single out prominent representatives in the political philosophical traditions of Tocquevilleanism and pragmatism. Indeed, as 20th century champions of these traditions such as Claude Lefort and John Dewey have contended, democracy does not refer primarily to a set of political institutions or procedures, but rather to what Lefort calls ‘a form of society’ and Dewey calls ‘a way of life’. ¹⁰

My assumption is, therefore, not only that there is room for developing a paradigmatic normative theory of democracy, but also that the best chances of success are obtained when we begin from a social rather than a political notion of democracy. I contend, therefore, that rather than starting with and confining oneself to a thin conceptual definition in the same manner in which the notions of justice and non-domination are currently defined, we need to take a social-theoretical route. This will imply in turn relying upon a social-theoretic account of the social dimension of human action and of the basic forms of social cooperation.

More specifically, a social account of democracy should begin from the baseline of social interactions and proceed to identify social interactional patterns of which the concept of ‘democracy’ can be predicated and take this as the building block for a wider theory of democracy. Such a theory has two stages. The first consists in a theory of patterns of social interaction which should explain in what sense social interactions may be qualified as democratic. The second stage consists in a social ontology that, building on these social-interactionist micro-foundations, expands the scope of the concept of democracy to the largest possible domains of social life, from basic face-to-face interactions to formal political institutions.

Such a theory needs to explain in what conditions the democratic properties of these basic social interactions are preserved and expanded beyond circles of primary relations, and which are the enabling conditions which prompt this process.

I have stated that, like justice and non-domination, democracy as a paradigm normative concept has a direct normative impact on our understanding of the appropriate ways of organizing social life. Indeed, its scope includes a vision of how social relations should be shaped and social and political institutions organized. This vision cuts much deeper into the social body than competing paradigmatic normative conceptions of justice and non-domination, precisely because a social view of human life is embedded in its very concept.

Social interactionism as the micro-foundation of democratic theory

Several avenues are available to those desiring to explore the consequences of a social-theoretical approach to the concept of democracy. Two are most prominent. The first is anthropological and seeks to identify universal traits of human nature that are assumed to correlate positively with forms of social and political organization to which the term ‘democracy’ may significantly apply. Recent theories of human development advanced in particular by Ronald Inglehart and his associates (Inglehart and Welzel, 2005) follow this path.

A second, and sociologically more refined strategy, begins from the baseline of patterns of social interaction, identifying interactional qualities on which the concept of democracy can be predicated and take this as the building block of a wider theory of democracy. The idea that social interaction is the basis of all social life has been defended by a plurality of sociological schools that I will group under the label of ‘social interactionists’, and which includes the Chicago school of sociology, pragmatism, symbolic interactionism and ethnomethodology. ¹¹ These approaches share the idea of a micro-foundation of social phenomena in elementary patterns of social interaction. This idea can be taken into the political realm and translated into that of a micro-foundation of democracy.

Social interactionists generally subscribe to the following assumptions: (a) individual identity is shaped through social interactions in which individuals take part; (b) social interactions are the flesh of associated life and compose the constitutive order of social life; (c) properties of social interactions can be inherited by more complex social aggregates such as organizations and institutions, and (d) normative orders are a mix of formal and informal patterns of social interaction and their capacity to steer social life depends upon the combined effects of these different dimensions.

It follows that normative expectations and requirements emerge out of social interactions. The normative relevance of social interactions is directly tied to its constitutive function in determining the quality of human life by shaping individual identity as well as chances of social realization. Social interactionists suggest that democratic patterns of social interaction are characterized by the following traits: (a) a prima facie preference for horizontal and symmetrical patterns of interaction (hierarchy and asymmetry always need to be justified to those on whom they are imposed); (b) promotion of cooperative relations which acknowledge the reality of interdependence; (c) equal participation in practices of inquiry and decision-making; (d) privilege of deliberation over other methods of taking decisions; (e) effective integration of all participants in social practices and (f) successful integration between the functional dimension of problem-solving and the expressive dimension of self-realization. ¹²

These traits can be more synthetically refined down to three major principles, whose political relevance appears immediately evident: (1) relational parity; (2) inclusive authority and (3) social involvement. While such a task cannot be undertaken here, these principles can in turn be given more specific content so as to make more explicit the meaning of ‘democratic’ as applied to patterns of social interaction.

Relational parity refers to the status individuals achieve in social interactions. Relational parity obtains when each individual in a relation is treated in ways that do not depend upon his or her social status. In particular, it requires that stigma and disadvantage are not imposed on the basis of status, and that religion, gender, race, ethnicity, class and other social markers do not affect our status within social interactions. The idea that democracy requires that interactions among individuals are not status-dependent – the Tocquevillean idea of democracy as the “society of equals” – has recently received fresh impetus under the banner of relational egalitarianism. Opposing traditional views of redistributive justice, relational egalitarianism proposes the idea that: ‘in an egalitarian society people should relate to one another as equals or should enjoy the same fundamental status (and also perhaps the same rank and power)’ (Arneson, 2013). According to Elizabeth Anderson, a leading relational egalitarian, the goal of egalitarian justice is: ‘to create a community in which people stand in relations of equality to others’ (Anderson, 1999: 289).

Inclusive authority requires that individuals are the authors of the decisions whose consequences they will suffer rather that their passive recipients. Whereas relational parity refers to the consequence of social status in interpersonal interactions, authority concerns the power one has of influencing others’ decisions. Authority can be exercised in egalitarian or in hierarchical ways, and it is not specific to political settings but pervades the entire society. Prima facie the notion of inclusive authority refers to patterns of authority in which those who are affected by the consequences of a decision are included in the decision process. Not all human relations can, however, be organized in ways that avoid hierarchical patterns of authority. Harry Eckstein, for example, identifies politics as the domain of hierarchical relations, so that a pattern of authority is defined as: ‘a set of asymmetric relations among hierarchically ordered members of a social unit that involves the direction of the unit’ (Eckstein and Gurr, 1975: 22). Obvious examples of hierarchical authority relations that fall within Eckstein’s definition are those between parents and children, teachers and students, managers and the rank and file in workplaces. In all these social settings, hierarchy seems to a certain extent unavoidable. In these cases, the democratic quality of hierarchical patterns of authority depends on the degree to which subordinates are involved in decision processes. Eckstein suggests, for example, that hierarchical relations have a democratic quality when superordinates exercise only limited direction and are highly responsive to the claims and influence of subordinates, and when subordinates in turn are entitled to a high degree of participation and comply on the basis of perceived legitimacy. Similarly, Shapiro claims that a prima facie preference for horizontal forms of power relations implies that the burden for justifying hierarchical authority must fall upon those who impose them upon others (parents upon children, employers upon employee, political elites upon citizens). According to this view, the democratization of power relations rather than their abolition is the normative expectation consistent with democracy.

The criterion of social involvement refers to a social unit’s capacity to involve its members in a plurality of practices. The notion of involvement reminds us that this requirement is stronger than the more standard condition of inclusion in decision-making processes which is implied by the notion of authority. Social interactionism assumes that, besides the protective function of inclusion in decision-making processes, participation in social intercourse is an activity with intrinsic value, because, through these interactions, individuals shape their own identity and at the same time cooperate in solving the problems that affect their social world. This reciprocal interdependence is realized once and at the same time functionally and expressively: It generates solutions to shared problems and contributes to self-development. The major implication of this assumption is a rather wide understanding of participation that the old-fashioned notion of ‘partaking’ expresses better than that of participation. Partaking means being socially and morally included in the concrete activities of a community. It certainly includes participation in decision-making – the enlightenment ideal of autonomy – but also and in a more basic way, unrestrained access to social practices and spaces, integration within the workplace, in the neighbourhood, and in the school system. Cynthia Estlund’s analyses of the democratic potential of cooperation among workers in the workplace (Estlund, 2003) provide vivid and compelling arguments and empirical evidence in support of the claim that inclusion in social practices rather than mere participation in decision-making is the appropriate normative standard for democracy.

So conceived, democratic patterns of social interaction can therefore be found in all dimensions of social life in the family, the groups of friends, the workplace, in learning environments, in voluntary associations, in all types of organizations, up to the level of the constitutional architecture. On the one hand, they structure informal relationships among acquaintances and among strangers. On the other hand, they are also embodied in organizational routines and in forms of institutional design. In both ways, they are not mere ‘enablers’ of political democracy (situations in which democratic habits are learned) but constitutive components of democracy.

Such an account of democracy is primitive in the sense stated above because democratic patterns of interaction have the status of ends in themselves. Not that we engage in these interactions for their own sake, but we appreciate their democratic quality as contrasted to other patterns. It is part of the normative content of democracy that democratic patterns of social interaction are preferable to nondemocratic one in any walk of life. Saying that democracy is a primitive concept means precisely this: that a form of society in which all interactions are organized and arranged according to the three principled stated above provides a normative ideal we cherish and pursue for its own sake. In other words, that we assign democracy paradigmatic normative scope means that we desire to live in an environment structured by horizontal, egalitarian and inclusive relations, and this desire has a final value, it is not a means we pursue for the sake of higher goals.

Such an account is also, inevitably, larger in scope than more conventional theories of democracy. One obvious and first implication of this account of democracy is indeed that opportunities for active involvement on an equal basis should be maximized across society. Because patterns of interaction happens everywhere in society and not only among individuals considered in their capacity as citizens or rulers, the idea of democratic patterns of social interaction provides guidance for normative considerations that apply also outside the domain of political relations. The existence of opportunities should first of all be understood in the sense of the undermining of formal and informal barriers that prevent participation. Another, perhaps less obvious, implication is that expanding the notion of democracy to include the public sphere and civil society is a necessary but insufficient move to achieve a full-blown paradigmatic conception of democracy. The relevance for democracy of patterns of social interaction implies that how individuals remain together in all social spheres and institutions has a decisive impact on democracy, one that notions of civil society, social capital and public sphere fail to capture. As Estlund notes, working together provides as much opportunity for democratic improvement as bowling together (Estlund, 2003: 5).

A social conception of democracy is compatible and indeed requires that the democratic contribution of formal politics be specified so as to be distinguished from that of other types of institutions. A paradigmatic conception of social democracy, in other words, does not eliminate the need for a full blown account of democratic political institutions. Such an account will have to explain how formal political institutions can promote the democratization of a society, extending the reach of democratic patterns of interaction. This task cannot be pursued here. What needs to be stated is that this task has two major components. On the one hand, politics defines and regulates relations among individuals in their capacity as citizens. A democratic polity is one which construes these relations in terms that are consistent with the three principles of (1) relational parity, (2) inclusive authority and (3) social involvement. On the one hand, formal political institutions can promote and sustain the spread and consolidation of democratic patterns of social interaction in the other spheres of social life. Legal regulations, economic incentives and institutional actions are the most common leverages through which such goals can be achieved. ¹³

A paradigmatic concept of democracy has momentous consequences also when used to examine cases of failed democratization. Indeed, it requires us to acknowledge that undemocratic societies are not only those lacking democratic formal political institutions or that are inadequately embedded in a public sphere and a civic society (Merkel, 2004). They are also societies in which access to social spaces and practices is determined by group identity, societies in which status determines patterns of authority, in which someone’s social standing, gender, age, ethnicity, determines their position in a social interaction. These norms of interaction determine entitlements, obligations, rights within the family, in interactions with partners, or in the workplace. But they also determine the places they can access and those from which they are formally or informally excluded. Formal and informal practices of racial zoning and gated communities are two examples of this trend. The overall implication of this approach should be clear. A country may be fully democratic according to international surveys such as those of Freedom House, and yet, according to a wider paradigm normative concept of democracy, it will have to be considered as undemocratic.

The three principles of (1) relational parity, (2) inclusive authority and (3) social involvement aptly convey this idea of social democracy, provided we do not understand them only in terms of self-government. To begin with, we should emphasize the very fact of having free and unhampered access to social processes and spaces in order to be given the chance to interact on a par with the highest number of individuals in the largest range of social situations. This neutralization of the social consequences of status is a major historical achievement of democracy as a principle of social organization. That is, individuals can join social groups, occupy social spaces, interact with others without considering that their social, economic, religious, racial, or political status defines the core of the social revolution that brought the ancien régime to a close and marked the advent of modern democratic societies. From a socio-historical perspective, it is a unicum in a world that has been shaped by the opposite principles of hierarchy and social exclusion.

We should, therefore, pay particular attention to the social and not merely legal or political meaning of patterns of democratic interaction, for reasons that are, however, not only historical. Indeed, tendencies toward exclusion of the underprivileged and the seclusion of the privileged are constantly at play at all levels of social life, and relational parity is regularly undermined by the rise of new forms of asymmetry. The very idea of the democratic project implies that this task is incomplete by definition, and that no formal guarantee of political equality will suffice to render all interactions consistent with the democratic idea.

The social ontology of democracy

In itself, social interactionism provides a necessary but insufficient basis for understanding democracy as a paradigmatic normative concept. This requires a theory that explains how this normative principle operates in different dimensions of social life. This function is fulfilled by a social ontology of democracy, which I conceive as a theory that describes the basic layers that compose social life. Being built on the micro-sociological foundations provided by a theory of social interaction, it will be a social interactionist ontology of democracy.

Of course, the use of social ontology in political theory is not completely new. Recall, for example, the works of such diverse thinkers such as Carol G. Gould and Charles Taylor. ¹⁴ More generally, the revival of Hegel’s scholarship since the late 1970s has played a major role in renewing the prospects of social ontology as a basis for democratic theory. Social ontology has generally been conceived of as a countermeasure against mainstream methodological individualism, as a way to emphasize the social prerequisites upon which political regimes rely. In fact, political ontologies never attempted to provide a micro-foundation of politics. Similarly, no attempt has been made to flesh out a full social ontology in terms of the constitutive layers composing social reality. Philosophers such as Crawford B. Macpherson, Pateman and Gould have relied upon social ontological arguments to extend democratic practices to non-political institutions such as the workplace. Yet their argument lacks a systematic social-theoretical foundation of the kind attempted here.

The baseline of my social interactionist approach to social ontology is that, for an ontology to provide the building block of a paradigmatic normative theory of democracy, the normative properties displayed by basic patterns of social interaction should be inherited by more complex and higher layers of social ontology. This should also hold for that specific normative property we call ‘democracy’. Hence the property that specifies under what conditions social interactions are democratic provides the content upon which we can generalize the normative use of the concept of democracy in ways consistent with its use as a paradigm normative framework. Indeed, democratic interaction at different social sites may, and in fact does, assume different forms. For example, democratic interactions at the level of primary groups take place within the framework of trust-based relations, whereas the workplace is better described in terms of the obligation to achieve social cooperation under the pressure of external constraints in an environment where we cannot choose with whom we cooperate.

One of the basic implications of this social interactionist micro-foundation is that active participation in a large plurality of forms of social practices has to be seen not as a prerequisite in the Tocquevillean manner, but rather as a constitutive feature of a democratic form of life. The usual argument about the necessity to embed democratic political institutions within democratic social practices should be replaced with the idea that democratic social practices are a constitutive and indispensable dimension of democracy, and that for democracy to exist social interactions at all the major ontological layers should be organized according to the three normative principles of (1) relational parity, (2) inclusive authority and (3) social involvement.

The social ontology of democracy should, therefore, explain under what conditions the democratic properties of these basic social interactions are preserved and expanded beyond the circles of primary relations. Social interactionists see society as dynamic and characterized by ongoing processes of structuration and destructuration of group life (Giddens, 1984). Hence, rather than emphasizing permanent structures, they emphasize dynamics of organizational formation and dissolution. This activity of grouping (Follett, 1919), this ‘ongoing concern’ (Hughes, 1984), denotes a fluid activity that displays indefinite degrees of variation. Hence, it is only by accepting a gross simplification that we can identify a finite set of conventional ontological layers.

It should be noted that, while properties of interaction can be inherited by organizational settings which host complex patterns of social interaction, inheriting is a process fraught with difficulty and it usually requires complex social arrangements. In other terms, the social construction of democracy requires a sophisticated sociological imagination and the capacity to innovate.

The argument about inheritance is different from the usual arguments about congruence and spillover. The congruence thesis states that lack of congruence among patterns of interaction in different spheres tends to produce maladjustment and instability. Hence, so the argument goes, political democracy can flourish only in societies in which democratic patterns of authority are also diffused beyond the political domain. ¹⁵ The spillover thesis states that democratic interactions within specific social spheres such as voluntary associations, the workplace, or the family generate democratic skills that carry over into the rest of civic life, also spilling over into the political sphere. ¹⁶ The inheritance argument differs from both, as it states that an analytical concept of democracy should provide the common denominator that unifies diverse patterns of democratic interaction across a plurality of different social spheres. While congruence and spillover require empirical evidence to be defended, inheritance is conceived here in analytical terms.

In explaining the democratic relevance of social practices, two dimensions have to be taken into account. The first is what I have called ‘inheritance’, i.e. the idea that, notwithstanding the diversity among social practices, they can always be organized according to the three democratic principles of social interaction, that is, relational parity, inclusive authority and social involvement, so that the quality ‘democratic’ can in principle be predicated on any social practice which is consistent with them. The second is the dimension of heterogeneity whereby each type of social practice contributes to the democratic quality of society in its distinctive way. As has been noted by several commentators, involvement in voluntary associations has a significant impact in the development of civic competences (Putnam, 1994), whereas for example, cooperation in the workplace can have more significant effects on our ability to tolerate difference and develop more inclusive attitudes (Estlund, 2003). Similarly, the ‘bonding’ type and the ‘bridging’ type of association promote different aspects of social attitudes and therefore contribute to the democratic quality of a society and to the realization of individual life in different ways (Putnam, 2000).

A social ontology of democracy should explain under what conditions the democratic properties of these basic social interactions are developed, and how requisite conditions vary across different layers of social life. For the sake of simplicity, we can break social reality down into a finite set of ontological layers. These layers differ in their organizational properties rather than in their scale.

Individual habits are the deposits of social learning, and in that sense they define the basic layer of a democratic social ontology. Habits themselves have an individual and a social dimension. On the one hand, there are individual habits such as the capacity to engage in cooperative discussion and communicative competences. On the other hand, there are those tacit and shared collective habits which shape informal interaction, including interaction among strangers. Democratic civilities are exemplary of this social layer.

The second layer is composed of primary groups such as the family, a group of friends and other forms of spontaneous grouping which take place on an everyday basis and whose basis is essentially affective. Informality, trust, affective bonds qualify patterns of interaction at this level. Here too democracy qualifies a form of interaction based on equality of relations and cooperative forms of exchange.

A third social ontological layer is composed of informal associations such as networks of solidarity and communities of peers. Trust and informality qualify these social units, although elements of functional differentiation and asymmetry begin to appear. At this level too, the informality of horizontal inclusive interactions provides the core of the normative argument.

A fourth layer consists of those voluntary associations in which membership is free and the level of obligation generally low. Charities, NGOs, neighbourhood associations, urban centres belong to this category. While informality, trust and affective bonds continue to play an important role, the functioning of these social units also requires a significant level of formal organization and role-definition. Scholars of social capital have traditionally emphasized the function of these associations for the democratic life of a society since Tocqueville. Here democracy as a normative standard refers to day-to-day interactions but also to the explicit procedures and rules presiding over assignments to roles, relations between officers and members, the capacity to actively involve the community of those concerned etc. Yet at the basis of all these dimensions, we continue to find relational parity, inclusive authority and social involvement.

At a higher level we find more structured forms of interaction which take the form of stable organizations, with rules, entry and exit conditions, and which impose higher constraints on an individual’s behaviour. At this level, compliance with the three normative principles of democracy requires more sophisticated forms of organizational engineering as social relations tend to be more complex.

At a still higher level of abstraction, we find institutions such as the education system, the army, the church, the state and the public administration. Here the realization of democratic conditions of associated living requires complex institutional efforts. The various traditions of industrial and workplace democracy have shown to what extent democracy provides an inspiring, demanding and revolutionary standard for shaping patterns of social interaction in the workplace.

At an even higher level, we find the formal political institution that composes the constitutional architecture of a political unit. Formal political institutions have two main and distinct functions in preserving and promoting democracy. They guarantee that political decisions are taken in ways that do not violate the three democratic principles and they promote the diffusion of democratic patterns of social interaction throughout all spheres of social life.

At each of these levels, the idea of democracy plays a normative role that may differ from case to case. While the requirements that each of these social aggregates should satisfy in order to be democratic are different, reference to relational parity, inclusive authority and social involvement unifies them all, giving democracy the strength, coherence and breadth of scope required to function as a paradigmatic normative concept.

Conclusion

As I have indicated above, a definition of democracy suited to play a paradigm normative role requires a high level of generalization, so as to avoid identification with too narrow cultural or historical content. It also requires a broad range of applications so as to fulfil the function of normative guidance that is typical of paradigmatic normative concepts.

If the concept of democracy, like that of justice and non-domination, is to be emancipated from its narrowly political-institutional use, then we should find an appropriate strategy to move from the level of a political theory of democracy to that of a paradigmatic normative theory of democracy. My proposed strategy takes its starting point from a social rather than in a political notion of democracy. I have assumed that a sociological interactionist account of social life provides an appropriate theoretical framework, as it enables us to identify social objects – patterns of social interaction – of which the concept of democracy can be predicated and take this as the building block of a wider theory of democracy. I have then integrated this element with a social ontology that, building on these social-interactionist micro-foundations, expands the scope of the concept of democracy to the whole of social life, from basic interactions to formal political institutions.

Such an approach has several advantages, the most important being its hermeneutical contribution to a better understanding of who we are and who we want to become, and its conceptual contribution to a richer understanding of what democracy promises to deliver in terms of social as well as political goals.

Understanding democracy as the unfinished project of achieving a form of society committed to values of relational parity, inclusive authority and social involvement provides us with a fruitful orientation to engage not only with intellectual projects of social and political critique, but also with practical attempts to design new institutions better capable of realizing our normative goals. As pragmatist philosophers discovered long ago, it is only through experimentalist practices that the democratic project can be advanced. Hence, democratic experimentalism is another name for that form of society we call ‘democracy’.