The Concept of Authoritarian Governmentality Today

ABSTRACT

The paper examines “authoritarian governmentality”. It argues that there are salient differences between the contemporary intellectual and political context and those of the 1990s when it was first developed. Chief among these is the confidence by which we can approach liberal governing itself as the norm of contemporary governmentality. The first wave of governmentality studies identified authoritarian governmentality in both non-liberal regimes and in the consequences of liberalism’s norm of the self-responsible subject and its governing through civil society. This paper extends these observations: first by proposing an analysis of the kinds of order that different rationalities and technologies invoke, and secondly, by arguing for the linking of an analytics of government to both the practical capacities of sovereignty and the practices through which supreme authority is constituted. While liberal and authoritarian governmentalities are far from mutually exclusive, the absent concept of authority helps clarify what is at issue.

KEYWORDS:

• Government
• authority
• sovereignty
• order
• Foucault

Liberalism and democracy, although compatible, are not the same. The first is concerned with the extent of governmental power, the second with who holds power. The difference is best seen if we consider their opposites: the opposite of liberalism is totalitarianism, while the opposite of democracy is authoritarianism. In consequence, it is at least possible in principle that a democratic government may be totalitarian and that an authoritarian government may act on liberal principles. (Hayek 1967, 161)

Introduction

Improvising on a compliment Michel Foucault gave to Gilles Deleuze, we could say, for better or worse, at least as far as the discussion of government is concerned, this century will be known as Foucauldian. This is due to his elaboration of the idea of governmentality in famous lecture series (Foucault 2007; 2008) and the founding of what has been called “governmentality studies” (Sennelart 2007, 390). The first generation of governmentality scholars sought to address the possibility of an “authoritarian governmentality” (Dean 1999, 131–148). Given that the question of authoritarianism again has an urgency inside and outside liberal democracies, it is timely to return to that concept today.

The term “governmentality” could be taken to mean rationalities or mentalities of government, although not without dissent (Sennelart 2007, 388–389). As an approach, it was understood as a de-centering of the analysis provided by theories of the state to view the state as an effect of multiple rationalities and technologies of governing (Foucault 2007, 117–119). The governing by the state and its various agencies would thus be placed in relation to the governing undertaken by a range of heterogeneous actors and agencies, including businesses, corporations, associations, non-governmental and inter-governmental organisations and ultimately by individuals upon themselves. Government was understood famously as the “conduct of conduct” (192-193) and, despite prevarications, became the fundamental term by which Foucault would approach power relations. As the conduct of conduct, government was defined in Foucault’s (1982, 220–221) most elaborate statement of the concept as a structure of actions on the actions of others that worked through individuals’ agency or freedom and was irreducible to the traditional antinomies of power analysis clustered around consent or coercion.

The current paper re-focuses on the concept of “authoritarian governmentality”. It seeks to complement what was established by the first wave of governmentality studies. That wave made one crucial observation. It proposed that the term “authoritarian governmentality” could be applied not only to regimes that practiced a decidedly non-liberal regime of national government such as contemporary China or Suharto’s Indonesia (Philpott 2000; Sigley 1996), but also to specific practices both within liberal-democracies and in their colonial rule or postcolonial activities. This literature would then distinguish two aspects of authoritarian governmentality within liberal democracies and by them in colonial contexts. The first concerns the “dividing practices” (Foucault 1982, 208) within populations, which are made according to the capacity or potential to act as or become rational, autonomous subjects (Hindess 2001; Valverde 1996); the second, the enfolding of certain of the processes, agencies, regulations and values of civil society conceived as external to government into governing itself. Liberal authoritarianism thus could be approached from the perspective of liberalism as a legal and political order based on a specific kind of autonomous or self-governing subject, on the one hand, and as form of detailed “liberal police” of the population rooted in the knowledge of civil society, on the other (Dean 2002, 41).

Today, the rise of what is called populism, particularly in its hard right-wing versions, in liberal-democracies, and the awareness of diverse forms of autocratic rule throughout the world, has sharpened the concern for authoritarian measures and potentials in the way in which governing occurs. It was one thing to point to the necessity of authoritarian practice within the relatively stable liberal political and economic orders of the late twentieth century; it is quite another to note the rise of opportunistic, messianic, volatile, and apparently anti-rational movements and forces within and outside these liberal orders today (Vasilache 2023). However one diagnoses the present, the last decade has witnessed a series of mutations and disruptive forces in liberal-democracies and in the international order.

This brings us to the stakes of this discussion. The question of authoritarian governmentality of the late 1990s was whether the tools developed to understand what was then called “advanced liberalism” (Rose 1993; 1996) could be extended to analyse certain rationalities and practices of government in various non-liberal settings and practices, and to demonstrate the illiberality of many liberal practices of governing (Dean 2002; Hindess 2001; Valverde 1996). Today it would be how to account for the proliferation of authoritarian forms and forces across these different settings to the extent that it has become increasingly unclear as to what would amount to a liberal governmentality. If we pose the question of authoritarian governmentality again today, it is because we no can longer be certain of what constitutes a liberal, or indeed, an advanced liberal, form of governmentality. At a minimum, we require a greater degree of awareness of the normative and polemical potentials of the use of these and related terms.

One key problem raised by the discussion of authoritarian governmentality is that it presupposes that the salient distinction is between liberalism and authoritarianism. As Hayek’s epigraph illustrates, the twentieth-century discussion of liberalism was often posed against the background of the experience of another kind of rule, “totalitarianism”, and that, for many, the opposite of authoritarianism is not liberalism, but democracy. From this perspective, liberalism and totalitarianism are located on a continuum defined by degrees of individual freedom, and authoritarianism and democracy on an axis of the sources of legitimacy of sovereign power. This matrix thus complicates the discussion. In this sense, Hayek’s typology suggests that it is possible to have a “liberal authoritarianism” and “totalitarian democracy” as well as a “liberal democracy”. (cf. Chamayou 2021, 340). Given that government is defined in Foucault’s terms (1982, 220–221) as a structure of actions upon the actions of free individuals, distinct from violence and consent, then it would appear to direct us to the liberal/totalitarian axis rather than the democratic/authoritarian one of sovereignty. One view would be that the recrudescence, proliferation, and dispersion of authoritarian forces and forms of rule today – particularly given the intimate relation between populism and democratic discourses – redoubles the urgency to think along both these axes.

At a more basic conceptual level lies the question of the content of the term “authoritarian” itself, and its derivation from its root noun, “authority”. Foucault’s vocabulary of power rarely encompassed or addressed authority itself. With few exceptions, such as discussions of the “authority of expertise” (Rose 1996) and the “enfolding of authority” (Dean 1996) in a key collection, this concern was not addressed in first-wave governmentality literature. And while it would be a mistake to reduce the rule of authority to authoritarian rule (Spillane and Joullié 2023), the question of authoritarian governmentality can no longer bypass that of authority itself.

This article is organised in four parts: first, a critical investigation of the strengths and limitations of studies of governmentality in relation to this question of authoritarianism; second, a brief excursus on the relation of authoritarian governmentality to the concepts of “liberalism” and “totalitarianism”; third, a specification of how we might deploy the notion of authoritarian governmentality today: and fourth, a concluding section. I argue that in addition to the fundaments of the notion of authoritarian governmentality already identified by the first wave that we add both an expansion and a renewed restriction on the concept, both of which introduce the term authority into the governmentality lexicon. The expansion is to move beyond rationalities and technologies of governing to investigate the kinds of authoritative order that these practices of governing seek, invoke, or assume. The restriction is to place the analysis of governmentality in relation to two aspects of sovereignty: its specific practical capacities, on the one hand, and the rituals and practices of its constitution as a supreme authority within a domain, on the other. Both moves allow us to illuminate aspects of contemporary political life and provide us with the four very specific concepts of authoritarian governmentality, and their relation to liberal governmentality, that are detailed in the concluding section.

Critique of governmentality

Foucault (1998, 219) himself used the term, “founder of discursivity”, to refer to one who initiates the rules of formation of a discursive practice. This aptly captures his relation to the field of governmentality studies developed later. What these works indicate is, however, less a theory, method, or a systematic analytical framework, than a definite style of thinking or “thought style” that not only directs ways of seeing and frames the assimilation of knowledge in Ludwik Fleck’s classic definition (1979, 105), but also remains normatively and politically polyvalent and open to linkages with a range of other styles of thought, as Karl Mannheim had earlier noted (1952, 146n). The strengths and weakness of governmentality in its address on authoritarianism grow from this thought style.

Clearly, there has been much to recommend the governmentality framework and the project of developing an understanding beyond narrowly conceived liberal and North Atlantic cases. We can readily identify what both generations of governmentality scholars took to be positive features of the framework for their purposes. In both first and second waves, this includes (a) its analytical acuity regarding problematizations, rationalities, techniques, modes of subjectification, and ends (teloi) of forms of governing, (b) its focus on combinations of heterogeneous state and non-state actors, and (c) its differential treatment of and reordering of populations and multiplicities. More fundamentally, government understood as the “conduct of conduct”, or a deliberate attempt to shape the conduct of the governed, promises to move away from the characteristic oppositions of social and political analysis, particularly those of freedom and power, consent and coercion, and so on.

In its first wave, “governmentality” offers an empirical framework for examining specific ways of thinking and of acting concerned with the shaping of conduct of individuals and collectives for various ends and to various effect, without the assumption that all rationalities and practices have a uniform character in any given polity, state or region. The postulate of heterogeneity overcomes the identification of government with state agents and includes specific constellations of actors in acts of governing, including non-governmental organisations, civil society organisations, corporations and enterprises. The division and composition of populations alerts analysts to key divisions in how individuals and groups are governed and the goals of that governing, and mechanisms of inclusion, exclusion and indeed abandonment.

In the second wave, there are at least two newer elements added to the quest for understanding authoritarian governmentality: first, the focus on the entwinement of the governing of conduct with resistances, revolts of conduct and “counter-conducts” (e.g. Simon 2024) and secondly the potential for a multi-scalar and multi-sectoral analysis (e.g. Heath-Kelly 2024). The focus on counter-conduct and resistance follows from the explicit address on this theme in Foucault’s 1978 lectures (2007), not widely known before they were published, but which arises from the positing of a degree of agency or freedom on the part of the governed we have just noted. The question of scale and sector comes with the movement of the governmentality studies into new disciplines and fields of study including geography, international relations, and security studies, and the promise of governmentality as a way of evading the categorial divisions of the national and international, the domestic and foreign, and the local and the global, state and civil society, security and care.

Nevertheless, from a more critical angle, there are features of this thought style that act as obstacles to the deployment of the concept of “authoritarian governmentality” and may hinder even the already identified positive features. We can identify at least four pertinent ones.

The first is that the narrative of governmentality assumes the unity or at least pertinence of the category of the West and is largely cast from inside it. There is a lack of specificity to the term, as there is to related ones of liberalism and modernity. Thus, while the category of the West sounds a civilizational tone in phrases such as “Western culture” and “Western subjectivity”, it also adopts a geopolitical one when we recall that Foucault is writing and speaking within the post-war Atlanticism of Western Europe, with a particular focus on France, West Germany, and the United States. East Germany is a kind of liminal political space, but clearly not of the West in this sense (e.g. Foucault 2008, 93). The early attempts to construct a history of governmentality, after Foucault, displayed this civilizational and geopolitical fixation.

The problem here is not the usual reproach of Eurocentricism but the elision between “liberal” and “Western” so that the analyst is immediately having to deal with not only the question of liberal/non-liberal governmentality but also with that of Western/non-Western concepts of government themselves, and the relation between these two binaries, the poles of which have multiple ambiguities. This is the case when an analyst approaches a clearly non-Western case outside the protocols of liberal-democracy, such as the one-child policy in China (Sigley 1996), or the emergence of a “neo-socialist governmentality” in China more recently (Palmer and Winiger 2019).

The second feature that acts as an obstacle to understanding authoritarian governmentality is that this narrative often has a teleological character that is driven by what it posits as the features of liberal (or neoliberal) governmentality as a form of governing that seeks to maximise self-governing. This has been observed for a long time, particularly in relation to Foucault’s earliest published lectures on the history of governmentality (Hunter 1998).

Here, the narrative of governmentality can be read as the gradual emergence, through the successive and perhaps even dialectical, critique of various “transcendent” obstacles or blockages, of an ever more immanent form of government – of a liberal or neoliberal government rooted in the rationality of the governed. These obstacles, overcome during the narrative, include the hierarchical and obedience-demanding character of the Christian pastorate, the “theological-cosmological” worldview of early Christianity, princely rule, sovereignty, the doctrine of reason of state, the familial and patriarchal model of the economy, and even the disciplinary organisation of institutions. These transcendent blockages of the art of government are overcome by specific historical tendencies such as the counter-conducts of the Christian groups we have just mentioned, the “atheism” of reason of state (and latter, of Adam Smith’s notion of the market), the critique of government characteristic of the European Enlightenment, and, more broadly, the critical ethos of liberalism. Clearly Foucault and his followers do not intend to reduce “governmentality” to its liberal and neoliberal manifestations, but such a teleological narrative makes it difficult to avoid this elision. Given that there are no historical universals in Foucault’s “genealogies”, asserting the generality of the term governmentality does not magically banish this problem but places one in open contradiction with its methodological framework.

From this perspective, authoritarian governmentalities appear as atavistic remnants or revivals of earlier transcendent forms – a return to an outmoded sovereignty, or a manifestation of the police state. If the Occidental focus leads to a kind of geo-spatial privilege, this teleology of government leads to a kind of temporal privilege in which effectively coterminous kinds of government are arranged on a continuum so that such forms are manifestations of the past in the present.

The most striking feature of this in Foucault’s narrative is that it has very little to say about non-liberal forms of governmentality during the nineteenth and twentieth centuries. Indeed, he mentions only two in passing: the “governmentalities of the police state” (Foucault 2008, 93), which clearly refers to his analysis of the early modern Polizeiwissenchaft; and the “governmentality of the party” (191). Non-liberal governmentality would appear to reactivate early modern doctrines and modes of government which are associated with reason of state and police science, or, because it uses the central and totalising control by a political organ, the features of twentieth-century “totalitarianism”.

Together with his repeated insistence that “there is no autonomous socialist governmentality” (2008, 92, 93), Foucault’s own narrative thus provides us with few analytical indications of how we move beyond liberal rationalities in the study of governmentality. We can reasonably infer that the history of governmentality contains a kind of normative position that equates governmentality, at least in its contemporary and hence most complete forms (given the teleological character of his narrative), with liberal and neoliberal forms of it. Liberalism and neoliberalism would appear to be innovative forms of governmentality from which any other form, including not only variants of socialism but also dictatorships and autocracies, must borrow.

The third problem concerns the relation of governmentality to other forms, kinds and relations of power. Partially due to this teleological narrative of Western political and power forms, Foucault (and the governmentality thought-style) assumes either that government, sometimes closely related to or sometimes opposed to biopolitics, displaces earlier forms and relations of power including discipline and, most importantly, sovereignty. This is the long-noted binary character of his power narrative. In a more complex version, Foucault (2007, 107–108) speaks of a “triangle” of government, discipline and law in modern governmentality, but this merely replaces a linear narrative of successive forms of power with a narrative in which the newly emerging power relations become dominant over the previously dominant others. Again, this does not make the successive displacement problem in Foucault magically disappear; it reiterates it at the level of what becomes dominant. This is allied with his resolution of forms of power into dispositifs or economies of power. Here, Foucault (2007, 8) posits a “system of correlation” and “dominant characteristic” across multiple dispositifs, but in which the dispositif of security, defining of modern governmentality, ultimately re-inscribes law, discipline and other dispositifs. From an analytic perspective, however, there is no reason to subscribe to the a priori dominance of one form, economy or dispositif of power. There are also a number of other reasons to be wary of the figure of the dispositif itself as a way of thinking about power or government. Above all, the reduction of sovereignty, classically understood as the claim to a supreme authority within a domain, to merely one of several jostling dispositifs is to remove its unique character and thus effectively to deny the term its commonly understood meaning in political discourse.

The question of sovereignty and its relation to governmentality haunts Foucault’s work, lends it considerable ambiguity and unclarity, and comes to be a defining characteristic of the thought-style of governmentality studies. It is not overcome by simply pointing to the persistence of sovereignty because it requires us to rethink completely the concept outside the conceptual binary between sovereign and modern powers. This problem is amplified when we approach authoritarian governmentality which requires a much more precise specification of sovereign governmental instruments and the constitution of a supreme authority within a given domain.

Finally, and even more fundamentally, there is no major reflection on the question of authority and thus what would constitute a form of governing grounded in authority in either Foucault or the literature that followed. Is the question of authority subsumed under the concept of government or does it belong to the theoretically displaced “juridical-political discourse of sovereignty”? If the former, then would it not be redundant to speak of an authoritarian government? If the latter, then does that not confirm (as per Hayek’s quote) that the key opposition is between liberal and totalitarian government, and that the problem of authoritarian government belongs to the (outmoded or at least not relevant) question of the foundations of sovereignty?

Excursus: liberal governmentality and totalitarianism

Let us then try to clarify the relation between authoritarian governmentality and liberal governmentality, on the one hand, and the notion of totalitarianism on the other.

To do this, then, we need to suspend both the temporal narrative that identifies the development of the art of government with modern/liberal/Western societies and a classificatory scheme that categorically distinguishes between liberal and other forms of governmentality, whether totalitarian, authoritarian, or of the police state or the party.

This suspension of the temporal narrative enables us to understand that what are generally thought of as pre-liberal or pre-modern forms of rule are capable of governmental innovation that seeks to induce people to accept regulation, including through the use and facilitation of markets, and by deploying tactics, reflections and forms of calculation. Meloni and Bashirov (2023) have surveyed three distinct formations of rule that disrupt the view of the art of government becoming dominant only in eighteenth-century Western Europe. The first is the existence of a state pastorate in ancient Greece and Rome. The second is the widespread reliance in medieval Islam on a system of market inspectors called muhtasib. The third is the well-developed urban government that existed in twelfth and thirteenth-century Italian urban government. Moreover, a governmental rationality in the form of a reflection on the art of government is found in ancient Greek and Roman culture and discussed as the very object of the science of politics in Book IV of Aristotle’s Politics. According to Pocock (1975) and Skinner (1978) this governmental rationality enters early modern Europe through the Italian republican thinkers of the Renaissance, including one to whom Foucault (2007, 91–92) assigned the role of representing the outmoded and displaced sovereign model, Machiavelli.

More importantly, for our purposes, we should allow the possibility that forms of governing in what are held to be non-liberal, authoritarian or dictatorial regimes often display features akin to those of liberal ones. For example, such regimes might work through the voluntary capacities of individuals to enable them to govern themselves in the general interest of society, as was the case of the German Democratic Republic.¹ Non-liberal regimes might also use markets as governmental devices in setting the limits to governmental action and seeking long-term national and political goals (China from the last decades of the twentieth century).

So how do we distinguish authoritarian forms of government from government as the conduct of conduct more generally? As we have already suggested, one step rests on a (re)introduction of the notion of authority. Kender (2022, 115) argues that authority is a subset of the Foucauldian notion of government in that it refers to the practices for the guidance of conduct which are made binding either by the personal qualities of the leader, including access to knowledge and capacity for argumentation, or through the impersonal necessity or force of entities such as the market, the price mechanism, economic and population data, algorithms, and audit cultures with their rituals of verification through ranking systems. Hannah Arendt (1961, 106) offers the exacting and paradoxical specification in respect to ancient Rome that “authority implies obedience in which men retain their freedom”. Government or governmentality becomes authoritarian, we might say, to the degree to which subjects no longer choose to govern themselves in a certain way but are bound to obey some form of personal or impersonal authority.

Turning to the term totalitarianism, Arendt (1956, 406) argues that identification of totalitarian rule with authoritarianism is a liberal one stemming from the misunderstanding of the term authority. The category of a totalitarian governmentality, nevertheless, introduces another range of problems. The tradition of classical post-WWII work on “totalitarianism” (Arendt 1973; Friedrich and Brzezinski 1965) conjures up the image of a political system that obliterates the realm of a sphere of personal or private life or autonomy altogether – whether through mass indoctrination or sheer terror. This would rule out anything that could be described as a form of “government” in Foucault’s sense as a power that acts through self-governing capacities of individuals. There would be total subjection but little “subjectivation”, complete inner terror and mind control, not the working on and through the capacities, attributes and actions of subjects. This view has been contested at least since Goldhagen’s (1996) claim of a high degree of voluntary participation of the population in the actions of the Nazi regime. Beyond this, the differences, for example, between the GDR and Nazi Germany are sufficiently great for it to be problematic to group the two together despite what these theorists during the Cold War would contend, and the category of “governmentality of the party” also does (Clowes Huneke 2020).

One resource that allows us to resolve this problem would be to follow the early genealogy of the notion of “totalitarianism” in German debates of the early 1930s, which we will return to in the next section.

Analytical directions

We are now able to specify the uses of the term, authoritarian governmentality, by both extending and restricting the notion of governmentality itself.

Let us start with an extension of the analysis of governmentality. In the classical understanding of practices of governmentality, the salient features (as we noted above) are the rationalities and techniques or technologies of government. These practices might be arranged on an illiberal/liberal continuum according to the degree of self-government they seek to work through and the kind of rationalities/technologies they use from more coercive and disciplinary to therapeutic and self-fashioning ones. Moreover, context is always important. For example, therapists of various sorts might advise us to keep a journal of our aspirations and our actions to become more self-governing. However, when a welfare officer demands that a welfare recipient do so to plan their pathway to economic autonomy as a condition of receiving benefits, then the everyday illiberality of liberal governing becomes evident. When the governmental practice becomes binding for whatever reason, in this case because the economic costs of not following the welfare officer’s advice are too high for those with limited means of subsistence, then we have crossed into an authoritarian government. Not only disciplinary practices but also education, social care and health systems all contain this kind of authoritarian potential. More recently it has been convincingly demonstrated that this binding authority can also be used covertly, together with intelligence and security services, in multiagency counter-terrorism surveillance and intervention (Heath-Kelly 2024).

A more sophisticated version of this framework would examine the aims, goals or ends that such practices seek (the teloi of governing) and thus place these practices within a more general framework. A specific telos, such as a society of “life-long learning”, can be more liberal in that it encourages the self-development of capacities and interests beyond the years of formal education or it can be more authoritarian to the extent that it becomes a binding obligation for those in precarious labour-market positions requiring urgent retraining as a condition, again, of accessing social support and benefits. Given these are not opposites, it can be both.

Yet this move still fails to exhaust the element of governing that requires the extension we seek here. This extension would ask what it is that the practices of government in question presuppose as the general order according to which they seek to govern and which dictates what they hope to achieve. What are the different cosmologies, Weltanschauungen, economic and political theologies, and, more generally, forms of order, that one seeks to govern in the name of and to serve. A key component of governing is not simply the different practices that seek to shape the conduct of individuals and collectives – or even the goals they seek to achieve – but also the general principles or forms of order that are invoked or presupposed in relation to various practices of governing and that themselves help define the ends or goals of such practices.

One could thus speak of a genealogy of order that follows the trajectory of this “transcendent” aspect of ways of governing (Dean 2016). Such a genealogy historically has included naturalised racial hierarchies, patriarchy, heteronormativity, and so forth. It includes conceptions of divine law, providence, natural law, civil society, globalisation, complexity and network thinking, spontaneous social orders and, of course, the self-regulating or constructed order of the market. In this respect, both the Ordoliberals and the Austrian neoliberalism of Mises and Hayek, could be said to be defined by their different conceptions of how we govern to ensure the order of a competitive market, including the preservation of private property and individual choice. What defines variants of economic liberalism is that they govern through or in the name of various conceptions of market and other orders. Broadly, one variant assumes the existence or evolution of spontaneous orders that constrain governing and present limits to knowledge (Hayek, von Mises), while the other seeks to govern in such a way to allow a certain competitive order to emerge (Ordoliberals). Other forms of governing, both socialist and social democratic, invoke the general order of society, its welfare and its interests.

Following Arendt (1956, 406), we could say that when a governing agency seeks to bind its own conduct or the conduct of the governed according to an order that exerts “a force external and superior to its own power”, it becomes authoritarian. Authoritarian governmentality can be based not only on the personal authority of, for example, the monarch or charismatic leader, but also, pace Arendt, on the impersonal authority of, for instance, a providential order, the market or globalisation. An authoritative form of governing such as the management of a university, might be based on giving good reasons for decisions, such as the need to improve its place in global rankings system, but such reasons themselves assume an underlying and imperative order that provides the framework for validity, e.g. that higher education is, or should properly organised as, a national or international competitive system or market for students and researchers. To be more exact than Arendt, we can add the proviso that it is when this order becomes unquestionable then it becomes authoritarian.

Outside liberal-democracies and in contemporary populism, contemporary authoritarian governing is conducted in the name of very different orders. In contemporary Russia, for example, one might observe a cultural-religious sovereignty based on Russianness and a kind of security pastorate of the Russian flock (Vasilache 2023, 253). Trump’s campaign slogan of MAGA (Make American Great Again) was a component of an America-first political order that displaced earlier conceptions of American-led international order. Nevertheless, as Marcuse bluntly noted in 1934 (2009, 7), while many of these forms of authoritarian rule present themselves as antiliberal, they “never attack the economic functions of the bourgeois in the capitalist production process” and “the social order intended by liberalism is left largely intact”.

The debate in Germany in the early 1930s indeed provides an excellent example of how an explicitly politically antiliberal rationality could be marshalled in favour of supporting a liberal social and economic order. Then, thinkers such as Marcuse (2009) and Herman Heller (2015) responded to Carl Schmitt’s advocacy of a liberal “qualitative total state” in 1932 (Cristi 1998; da Silva 2021; Schmitt 1998). They focused on his claim of a need to restore a classical liberal economic order based on the clear separation of state and society, which putatively had been undermined by the growth of welfare interventions in the Weimar republic. Schmitt’s argument can be seen as part of the attempt to establish an authoritarian liberal state under the then chancellor of the Reich, Franz von Papen, which we will not go into here (Cristi 1998, 212n; Kender 2022, 46–48).

Schmitt (1998, 216–219) had argued that the Weimar Republic was a weak “quantitative total state” which had undermined the strict division of state and society acknowledged by classical liberalism. This undermining had occurred by the colonisation of the state by particular interests through the system of representative government in the development of a welfare state. What was needed to produce a “sound economy” was a strong state that would intervene to ensure a strict separation of state and society, particularly the economy, and limited intervention into the economy by an “autonomous economic administration” itself separate from the state. The order of the economy would be secured by the “qualitative total state” that only intervened to secure its autonomy from these special interests.

In 1933, Heller famously called this “authoritarian liberalism”, which, in a caustic response to Schmitt, he characterised as a:

… retreat of the “authoritarian” state from social policy, liberalisation (Entstaatlichung) of the economy and dictatorial control by the state of politico-intellectual functions. According to Schmitt’s quite credible reassurances, such a state has to be strong and “authoritarian,” for only a state of this type is able to sever the “excessive” connections between the state and the economy. Of course, the German people would not tolerate for long this neoliberal state (deutsche Volk diesen neoliberalen Staat nicht lange ertragen) if it ruled in democratic forms. (Heller 2015, 300)²

Authoritarian liberalism, for Heller, was a way of drawing a sharp line between the economy, which would be managed separately from the state by an independent administration to preserve its autonomy, and state control of repressive apparatuses and the means of mass manipulation. The strengthening of the state would be a condition of removing the interventions of the Weimar welfare state and recreating the separation and distinction of state and economy.

This example demonstrates that an authoritarian governmental rationality can be based on a kind of liberal order. Here, it is a form of governing grounded on a separation between state and society established by classical liberalism now threatened by features of welfarism, the representative system, and by political rights of assembly, speech and free media. Because such a governmentality seeks to preserve economic liberties such as property rights and the autonomy of the market economy, it is also a kind of liberal governmentality. It is both a liberal governmentality within the market and an authoritarian one in its own sphere. As Benjamin Constant put it, “the government outside its sphere must have no power; in its sphere, it cannot have too much” (quoted by Cristi 1998, 168).

This is not to suggest that authoritarian governmentality is always based on liberal conceptions of social or political order. However, it is also to indicate the possibility of an affinity between an economic liberalism which privileges safeguarding individual economic liberty and property rights and an authoritarian government. This kind of authoritarian liberal governmentality is an oft-noted feature of contemporary neoliberalism and it was identified from at least the time of the beginning of the Thatcher government in the UK (e.g. Gamble 1979) and is what Hayek and his colleagues (Fischer 2009) imagined they found in Pinochet’s Chile. It cannot pass without note that Heller would use the term “neoliberal” to describe this formation almost a century ago.

This debate also highlights the terms, total and totalitarianism, and their tensions with authoritarianism. While other conservative thinkers would advocate an authoritarianism of the state to curtail the “total mobilization of the people by the state” (Heinz O. Ziegler, quoted by Kender 2022, 106), Schmitt would link the legitimacy of the authoritarian state to a “qualitative total state”, which he calls “total in the sense of quality and energy” and immediately references the Italian fascist state as “stato totalitario” (1998, 217, original emphasis). In this debate then one could speak, as did Marcuse in 1934 (2009, 1), of a “total-authoritarian” state in which authority was ultimately grounded in the mobilisation of the people expressed through plebiscitary democracy. Alternatively, with Ziegler, one could also imagine a conservative authoritarianism to protect the state from precisely such a totalitarian movement.

Thus, states, politics and movements might be described in these terms as anti-authoritarian totalitarian (e.g. a populist mobilisation against liberal elites, the deep state, etc.), authoritarian anti-totalitarian (e.g. military regimes suppressing communism), and as “total-authoritarian”, which might even seek to enforce a liberal economic order, as in Schmitt. The point here is not to suggest a definite typology but to indicate that under given conditions it might be possible to distinguish different kinds of authoritarian governmentality with very different relations to liberalism and totalitarianism.

Let us now turn to what we have characterised as a restriction of the scope of the concept of governmentality. According to Richard Tuck (2015), since at least Bodin and Rousseau, political thought has been clear that the state and its constitution, whether democratic or not, cannot be viewed as merely derived from different ways of governing either in the restricted sense of the bureaucratic, legislative and executive bodies within the state or the more extended Foucauldian one of the heterogeneous forms of the conduct of conduct. This is possibly best expressed by Rousseau’s claim about those who have examined the democratic constitution: “None of them has sufficiently distinguished the Sovereign from the Government” (quoted by Tuck 2015, 1). Translating this observation to our own problem, government as the “conduct of conduct” is best thought of not as a replacement for a monstrous theory of power or a conception of the state, but a relatively delimited component of relations of power and of the functioning of agencies and organisations within and across states, that needs to be supplemented by a developed, and not simply residual, concept of sovereignty.

Since Bodin in the sixteenth century (1955, 40–49), there are definite “marks” of sovereignty. In the political theory of the twentieth century such marks were key elements which are used to define the state: the claim to the monopoly of violence (Weber), the claim to the monopoly of final decision (Schmitt), the right of war (ius belli), the maintaining of public order, the levying and collecting taxes, the making and enforcing of laws, the conduct of international relations, the contracting of public debt, and so on and so forth. Without trying to defend a definite list or account of state capacities, which is not our task here, we could make a couple of elementary points. Some political entities – states in the modern sense – have greater or lesser capacities to do these things than others. Others can delegate parts of these to other agencies, some formally private, and some “below” and some “above” the territorial state. And most importantly, these capacities are dependent upon but irreducible to government, much of which, even in the expansive Foucauldian sense, is dependent on at least some of them. Thus, it would be hard to imagine the proliferation of disciplinary and biopolitical governing in education and health care and their extension to national populations without the development of a modern taxation system, or an agency that successfully defends a preferred version of public order and security within a territory. Moreover, it has been long observed in the international sphere that the relation of governmentality and sovereignty is not of one replacing the other but of each acting as a condition of the other. The international system of nominally independent states is a condition of forcing open the geopolitical spaces in which the arts of governing populations can occur (Dillon 1995), and the existence of an international governmentality in the form of a set of supranational agreements and the assignment of populations to states is a condition for a world inhabited by these sovereign states (Hindess 1998a).

The analytics of government that provides a way of examining different modes of the conduct of conduct thus meets what could be called an analytics of sovereignty that studies the formation, existence and effectiveness of given state capacities (Dean 2016). Sovereignty, from this perspective, is a “vicarious” form of power, one that can be delegated onto all sorts of agencies from bodies constituted by international agreements, such as the European Union, to the legal officers and state bureaux of national states, to members of the armed and police forces, to corporations and private companies, medical professionals and, in some cases, ordinary citizens. One of the minor virtues of the analytics of sovereignty is that it reintroduces the question of violence and contestation over violence into the analysis that is occluded in the very definition of government and power in the Foucauldian framework (Foucault 1982, 220). So, a second way of thinking about authoritarian governmentality would be to identify and to follow its relations with the kind of state capacities and actions I have just mentioned, and the delegation, abrogation and derogation of sovereign right (Dean 2007, 176 ff.) Given that these capacities include the use of violence, compulsion, and authoritative decision, such an analysis begins to provide an understanding of how governing in the Foucauldian sense becomes intertwined with types of rule that discharge various kinds of formal and often delegated authority that act not simply through direction or persuasion but through coercive means.

This move puts governing in relation to and tension with another, heterogeneous, set of relations of power, that include claims over and contestation for a monopoly of violence.. It also shows how everyday forms of coercion – the compulsory nature of taxation, drivers’ licenses and passports, for instance – are imbricated within everyday practices of governing such as regulating flows of traffic and travellers. In the case of practices of deportation, for example, liberal governing includes the use of corporeal measures including physical force and violence that seeks to remove individuals with minimum harm and involves not simply the shaping of conduct through various means but the use of bodies and affects in a myriad of ways by agents exercising delegated sovereign authority (Borrelli and Walters 2024). These examples problematise or at least weaken generalisations that claim that governing is defined as the shaping of freedom or the shaping of conduct.

In general terms, then, practices of government are entwined with capacities and practices of sovereignty. In this sense they proceed to govern with different degrees of self-government, autonomy, consent, coercion, compulsion or domination, and their combination. With some care, it might be possible to distinguish more liberal and more authoritarian practices of government. While in general, one would clearly want to argue that Denmark relies on a more liberal set of practices of government than, say, Putin’s Russia with its propensity for the repression of political opponents, this does not preclude the discovery of practices of authoritarian governmentality in the Scandinavian country. One example would be the passing of legislation, and its implementation, that forcibly removes the homes and relocates segments of the Danish population to better integrate ethnic groups within the Danish community and prevent the formation of what were initially called “ghettos” and more recently, “parallel societies” (Bubola 2023).

There is, however, another way in which we can identify forms of governmentality as authoritarian and this entails turning our attention from the practical side of sovereignty to the question of how sovereign power gains legitimate authority. This is the sphere of the constitution of a supreme authority within a domain and includes the symbols, rituals, liturgies, and ceremonies of political power, including those associated with different forms of democracy such as representative democracy, and what might be called “plebiscitary democracy” (Weber 1979, 268–269) or the “acclamatory model of democracy” (McCormick 2004, xl), as well as classical notions of monarchic and aristocratic rule. In the words of classical political theory, such as Sieyès (1963), this is the realm of constituent power rather than constituted power. In Weberian terms, we can speak of the different ways in which authority is constituted and the forms that it takes: rational-legal, charismatic and traditional. This is the core terrain of a political archaeology of glory, to borrow a term from Giorgio Agamben (2011). Here, one can begin to talk about the way various forms of authoritarian or democratic forms of sovereignty are constituted while taking care not to over-unify such categories. The theocratic sovereignty constituted by the Iranian uprising under the charismatic authority of the Ayatollah Khomeini, was clearly different from that of military dictatorship in Chile under the decidedly non-charismatic General Pinochet in the same period, both of which are distinct from the recent right-wing “populist” administrations in representative democracies from Trump to Georgia Meloni, in which their leaders are held to personify the “sovereign will of the people” by virtue of special personal qualities (Vasilache 2023, 245). More broadly, constitutional monarchies and republican democracies also differ from one another other and among themselves, as do federal and centralised systems of government, in the way sovereign authority is constituted and distributed. If we relax the Foucauldian distinction between governmentality and sovereignty, we can ask how one form of governing, namely binding authority, is constituted, and thus arrive at a more nuanced understanding of authoritarian governmentality.

Four senses of authoritarian governmentality

How then can we speak of authoritarian governmentality today? There are at least four quite precise ways of using the term. I shall try to draw a distinction in each case between liberal and authoritarian forms of governmentality given the caveat that we need to avoid a categorical opposition between the terms.

1. Authoritarian governmentality is above all based on authority. It is understood as a subset of governing understood as the direction of conduct, but one that does that by force of binding obligations. This authority could be personal, based on charisma, the sacred properties of individual office-holder, their capacities for argumentation and even the giving of reasons in certain instances, or peculiar attributes of age, gender, ethnicity, and social status. It could be impersonal, such as those neoliberal governmentalities based on the authority of the market, or those based on tradition and on religion. And, while with Arendt, we need to clearly distinguish it from totalitarian rule (1961, 97–100), it could become totalitarian to the extent that such governing requires a mobilisation of the population through “plebiscitary leadership and democracy” in which the people or public is formed through its physical or virtual presence in acclaiming and publicly ratifying the leader, decisions or policies (Weber 1979, 268-269). Generally, authoritarian governmentality tends to elevate and exploit the liturgical aspects of political rituals and ceremonies that are constitutive of a supreme authority while liberalism tends to view these as atavistic remnants or simply “symbolic” or “traditional” that can be overcome by rational and calculable technologies of government. This does not mean that liberal-democratic polities evade the need to constitute supreme political authority, as the rituals of the United States party conventions and presidential inaugurations so insistently remind us (Dean 2017). Nor does it rule out the possibility that liberal governing relies on a form of authoritarian governmentality that elevates forms of impersonal supreme authority, e.g. of some form or other of the market. These impersonal forms are often related to the question of order.

2. Authoritarian governmentality seeks its foundation in different orders. This governing through binding obligations can be based on what is deemed necessary or compelled by particular but totalising social, political and economic orders external to the exercise of power by the governing agency. Thus, a liberal order of economic globalisation mandated a kind of total reformation of institutional and individual conduct in the neoliberal discourses of the late twentieth century (Hindess 1998b). Similarly, the attempt to repair the damage done by the Weimar Republic to a classical liberal social order consisting of the separation of state and society was identified by Schmitt (1998, 217) as requiring a “qualitative total state” with “modern weaponry” and the “new technical means of mass control, mass suggestion and the formation of public opinion”.

As the case of Heller’s notion of “authoritarian liberalism” shows, liberal economic government can often be aligned with an understanding of an economic order that produces binding and hence authoritative obligations – and hence can also be or become a form of authoritarian governmentality. Other forms would subsume governing in the name of a liberal economic order to the service of national, racial, ethnic, cultural, religious or patriarchal orders, or seek to reinstate a now presumed lost, or endangered, liberal order, as in the case of Schmitt’s arguments. The presence or absence of authoritarian governmentality in this sense might be understood in terms of the degree of reliance on authoritative orders to justify, explain and bind the conduct of individuals, collectives and institutions. Broadly, however, while liberal governing is grounded in orders that work through providentially beneficial outcomes accruing through individual free conduct, authoritarian governing works though orders that leaves no alternative for individuals and collectives other than to be bound by them. But as Margaret Thatcher’s TINA (There is No Alternative) principle shows the advocacy of a liberal or neoliberal governmentality can easily be translated into an authoritarian and apparently binding imperative.

3.

Authoritarian governmentality applies to and emphasises specific sectors of populations. Here, authoritarian governmentality as a governing of conduct which is binding is practised by agencies and actors within liberal democracies and non-liberal polities on specific sectors of the population. First-wave governmentality studies provided inventories of the differential treatment of subjects within liberal democracies using coercive and disciplinary instruments on those incapable or not yet capable of acting with the attributes of responsible freedom or conforming to norms and regulations derived from civil society (Hindess 2001; Dean 2002). The source of the need for binding obligation on these types of subjects is that they are insufficiently mature or simply unwilling to recognise the necessary lessons to be drawn from authoritative social orders, to follow common-sense social norms in organising their lives, or to respect the advice of agents and forms of authoritative expertise. Liberal rationalities of governing will tend to emphasise the existence and potential existence of rational, self-governing subjects rather than the tacit order that defines the parameters of that self-government; while authoritarian rationales will explicitly view the existence of forms of authority as necessary in the constitution of such subjects. The free subject of liberal mentalities of government is a self-governing one that navigates its own social and economic destiny within a specific order; the free subject of authoritarian governmentality is a freely obedient one who serves an order within its definite hierarchy.

4.

Authoritarian governmentality and the exercise of sovereignty. Authoritarian governmentality as a form of governing that seeks to govern through binding obligations might be indicated by the extent of the use of the coercive dimension of the exercise of sovereignty imbricated within forms of governing to ensure those obligations are followed. A clear example would be measures against the COVID-19 pandemic such as work and school closures, home confinement, vaccine mandates, and so forth, justified by the authority of public-health experts, governed through extra-legal or extraordinary measures. These measures made possible counter-conducts precisely in the name of self-governing individuals, such as those of the German Querdenker movement, that would criticise such measures as authoritarian or liken digital passes to National Socialist measures to badge Jews with yellow stars (e.g. Agamben 2021). The difference here between liberal and authoritarian government is relative: the first stresses the necessity of coercive intervention and regulation only in exceptional circumstances and in relation to particular populations, and foreshadows a (often always deferred) return to normal governing through freedom, while the second can claim that these exceptional measures are continuous with use of binding obligation in the usual course of governing, such as taxation and constabulary policing. Liberal government uses instruments of coercion to protect and foster what it posits as requisitely self-governing subjects; authoritarian governmentality uses them to support and enforce the binding obligations it governs through.

Authoritarian governmentality can be identified and hence analysed in four distinct ways: the form and source of authority that governs through binding obligations and the requirement of obedience; the presence and mobilisation of social, political and economic orders that are external to both governors and governed and serve as the foundation and binding justification of governing; the identification of populations that require governing through binding obligation rather than through their own responsible and mature active self-government; and the recognition that coercive techniques of sovereignty not only are used to solve certain kinds of extraordinary and urgent problems but also as a matter of course to strengthen and ensure the authoritative government of individuals and collectives. As one can see from such an inventory of features, there can be no categorical distinction between authoritarian and liberal government.

This paper, then, has sought to revisit and repurpose the concept of authoritarian governmentality to today’s very different intellectual and political environment characterised by forces and trajectories that call into question earlier diagnoses of our present in terms of liberal or advanced liberal governmentality. This is certainly not easy, particularly given the lack of discussion of terms such as authority or order within the governmentality literature and the attempted displacement of the “juridical-political discourse of sovereignty”. The central concern of the paper is to specify the valid uses of the concept of authoritarian governmentality itself. The paper has not attempted to apply systematically that repurposed concept to a diagnosis or history of the present, however these might be considered. That will form a whole new investigation.

Acknowledgements

This paper was written while I was a Mercator Fellow in the “Voluntariness” research project headed by Professor Jürgen Martschukat at the University of Erfurt in the Spring of 2023, and funded by the German Research Council. It benefited in particular from long discussions and comments by Pia Herzan. I also deeply thank Julia Simon for editing this special issue and organising a prior workshop that provided inspiration to resume a reflection of this problem.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 This is one component of the recent debate in Germany about the GDR engendered by Katja Hoyer’s book (2023). Qualitative research into voluntariness in an East German factory is currently being conducted at the University of Erfurt: Elena M.E. Kiesel, “Ein Neuerer sein. Freiwilliges Mitmachen als Selbst-Technik” (“To be a Neuerer. Voluntariness and subjectivation”), 27th Transatlantic Doctoral Seminar in German History: Nineteenth- and Twentieth-Century German History, June 14–17, 2022, German Historical Institute in Washington DC.

2 German original in Heller (1992, 665). I am grateful to Pia Herzan for finding the original German quote, and for her thoughtful comments on an early version of this paper.