Just What Is Ontological Political Theory Meant to do? The Method and Practice of William E. Connolly

Abstract

This article provides a critical appraisal of the ontological method of political theorizing through an examination of the methodological development of the work of William E. Connolly. Connolly has often been taken as a paradigmatic figure of the ‘ontological turn’. This is not only because of the significance of his work in the field but because he is one of its major methodological articulators. However, there has been no systematic evaluation of that method and its development. This paper rectifies that lacuna by critically illustrating Connolly’s turn from a post-positivistic interpretivism to his much noted ‘onto-political method’. It argues that the latter, while usually thought to be modelled on the work of Friedrich Nietzsche and Michel Foucault, is structured by Heidegger’s understanding of ontological difference. The paper then argues that this leads to several problematic tendencies within Connolly’s model that undermine the critical-explanatory and normative power of his methodology by compromising the critical reflexivity ontology is meant to provide. All of this raises some concerns and criticisms of the use of ontological method of political theorizing, which has escaped sustained methodological analysis and scrutiny.

Introduction

The contemporary ‘methodological turn’ in political theory offers a unique opportunity to re-assess previously-made methodological moves that escaped sufficient critical scrutiny. The ontological turn, associated with the entrance of Continental, post-structuralist political thought into mainstream debates in the 1990s and 2000s, is ripe for such an analysis, having both escaped recent reflection and generated a series of continuing approaches. Often justified through quasi-methodological claims of the necessity of ontological reflection and the presence of unexamined, problematic claims within the dominant traditions of political thought, the ontological turn has rarely been subjected to lengthy analysis as to just what an attention to an ontological approach is meant to do for political theory. That is, what practices of political theorizing is it meant to enable? This oversight is glaring as it has now spawned other approaches (e.g. new materialism) for whom the value of ontology for political thought is very much a given.

Ontological political theory (OPT) can refer to a wide body of theories. Stephen White (2000) in his account of ‘weak ontology’ identified it with William Connolly, Charles Taylor, Judith Butler and George Kateb. The present terrain is more structured around the two major ‘schools’ of OPT that emerged in the interaction between ontological analysis and democratic theory: that of Connolly and Chantal Mouffe. The latter two, in contrast to the former, both explicitly justified their ontological approaches and the use of ontology generally and generated clusters of work informed by their approaches. For example, while Connolly is arguably at the centre of a current American brand of post-structuralist agonistic liberalism, Mouffe (with Ernesto Laclau) produced (largely in the 1990s and 2000s) a school of radical democracy centred in the United Kingdom. ¹ The relation between these groups can be understood in different ways. One common frame is through the ontologies they employ; Connolly’s ontology of abundance is thus opposed to Mouffe’s ontology of lack (Tonder and Thomassen, 2005). However, this distinction ignores how Connolly and Mouffe share a basic Heideggerian understanding of what ontology does that structures their methods, their understanding of politics (and the political) and their enactment of the critical-normative aims of political theory.

These connections make the present analysis of wide relevance to contemporary debates in methodology in political theory and beyond. This discussion hopes to initiate a more substantive reflection on the methods of OPT by, in the first instance, focusing primarily on Connolly, as opposed to Mouffe. This is for the simple reason that the former is much more of a methodologue; he has actively investigated and argued for his ontological method to a level of detail not found in Mouffe’s work. She will however inflect and enrich the argument at several relevant points. ² Nonetheless, this does mean this analysis is and should not be understood as a general account of OPT, but an examination of a major voice there that hopefully will lead to similar analyses elsewhere.

This article proceeds through several sections. First, it examines the ontological turn in political theory and its motivating problems: justification and pluralism. This section argues that it seeks a reflexive brand of critique and normative construction that reverts to the level of ontology in order to dwell within these problems. It also suggests the possibility of a methodological tension within this structure. Second, this analysis turns to Connolly’s ontological method, focusing on its development from interpretivism, to Heideggerian ontology, to new materialism. It argues that the turn from interpretivism to ontology in Connolly’s work involves the introduction of two key arguments, the ontological contingency thesis (OCT) and contestability mechanism (CM). Finally, it connects these to the major criticisms of Connolly’s ontology, illustrating how it reconciles their contrasting accounts. In the third and final section, this analysis critically assesses this method and its two main parts. It argues that the OCT and CM are in tension with each other in a manner that obscures the status of the theory he offers, and undermines his claims as to the critical and normative significance of ontological political thought. Through an examination of Richard Rorty’s critique of ontology, it raises several concerns around the impact of ontological claims on pluralistic political theory. In the process, it also clarifies several recent criticisms of Connolly’s politics by locating those flaws in his ontological method. Together, these various arguments raise important concerns around whether there is a coherent project and set of purposes within Connolly’s method for contemporary political theory.

The Ontological Turn in Political Theory

OPTs occupy the space between mainstream liberal-democratic theory, and post-structuralist critical social theory. Like post-war liberal political thought, they are rooted in a concern with political pluralism. This is the problem of organizing democratic engagement across substantive cultural-political differences where fundamental values, norms and modes of speaking are not shared. It is a problem of power in relation to the goals of equality and popular sovereignty in the Western liberal-democratic tradition (Moon, 2004). OPTs also are rooted in the problem of justification, which has concerned much of critical social theory. This problem stems from the critique of foundationalism that absorbed theoretical discussions in the humanities and social sciences in the latter half of the 20th century. Often articulated in different terms (e.g. as metaphysics/grounds/ presence), these critiques highlight a methodological problem of justification in the realm of socio-political criticism. How do we justify the explicit normative claims and implicit normative assumptions that guide socio-political criticism and reconstruction (Cooke, 2006: 1–8)? These two problems, political and methodological, structure OPT’s critical and normative endeavours, placing it in equal conversation with two projects: the post-foundationalism of ‘the social constructivist turn’; and, the imperative to inclusivism within liberal-democratic theory.

OPT is thus defined by a complex mixing of the critical and the normative from within the conditions of the problems of justification and pluralism. This has resulted in several broad imperatives. First, OPTs are anti-authoritarian, both epistemically and ethically. They cannot establish themselves as the top-rung of a ladder of either epistemic access or ethical value. Interestingly, this constraint crosses the major division in this literature between context-dependent and context-independent approaches. Second, OPTs are generally committed to reflexivity. They reject any assumptions of a neutral perspective outside history and particularity, and recognize the contestability of the theorist’s perspective. Consequently, they don’t seek more true or rational models of social criticism or inter-normative engagement, but more reflexive forms of political theorizing and practical interaction. These include reciprocal logics of interaction, exchange and learning which can increase the responsiveness of dialogical partners. They are thus part of what James Tully has termed the ‘dialogical turn’ (Tully, 2004).

While other critical social theories also pursue anti-authoritarianism and reflexivity, OPT is distinctive in its recourse to ontological forms of analysis. ³ The key difference is OPT’s attempt to use specifically ontological reflection to enable reflexive, non-authoritarian forms of critical analysis and normative prescription. For OPTs, ontology is both fundamental and contestable, both unavoidable and ungrounded. This paradox necessitates a twofold approach. OPTs attempt what Connolly calls a ‘double-entry orientation’ that both critically uncovers covert (and problematic) ontological assumptions within the practices and norms of contemporary liberal-democratic life (and their pernicious effects), that is, its critical task, and reconstructs those norms informed by alternative ontological figurations they argue have validity, that is, its normative task (Connolly, 2004, 2008: chap. 3). Thus, OPTs call for a deep engagement with contingency, that mixes assessments of current assumptions with alternative projections of ontologically-informed norms (Marchart, 2007: 31–32). This does not produce neutral norms, whether as a common ‘political’ conception or as a reflection of universal structures of validity within language-use. Rather:

[t]he fundamental conceptualizations such an ontology provides can, at most, prefigure practical insight or judgment, in the sense of providing broad cognitive and affective orientation. Practice draws sustenance from an ontology in the sense of both a reflective bearing upon possibilities for action and a mobilizing of motivational force (White, 2000: 11).

While ontological political thought rejects the project of making political forms flow from ontological analyses, it is concerned to make positive normative claims based on its ontologies.

This emphasis on normative construction links ontological approaches with a general desire in contemporary political theory to achieve ‘normativity without foundations’; to make political claims without reliance upon contentious philosophical foundations; to be critical and prescriptive without absolute grounds. This project spans Rawlsian ‘political’ liberalism, Habermasian discourse-ethics and Continental post-foundationalism. However, while this framing illuminates considerable crossover, specific differences remain in how OPTs (along with post-foundationalism) pursue normativity. To balance anti-authoritarianism and normative affirmation, they understand all claims as contestable. However, contestability is not merely stated but integrated into their ontologies themselves. Ontological (i.e. necessary) pluralism structures their analyses and norms. Contestability must ‘in some sense fold back upon itself, disrupting its own smooth constitution of a unity. In a way, its contestability will thus be enacted rather than just announced’ (White, 2000: 8). OPT resorts to the level of ontological analysis in order to integrate contestability into its presuppositions. Such a level subsists below conscious dialogue, conditioning our interactions, requiring a specialized form of theoretical access.

This raises a potential weakness that will manifest in Connolly in this account. OPTs, by making norms flow from ontological assumptions, risk establishing themselves in a relation of authority with their interlocutors through an implicit methodological claim of privileged access. By resorting to an external theoretical condition, they can assume a natural order to reasons with ontology/affect/sensibility/faith at the top and argument/dialogue/reasons at the bottom. As others have observed, when ontology/contingency are necessary, say in the priority of a distinct ontological realm of politics, political theorizing risks becoming a ‘form of political ontology as prima philosophia’ (Marchart, 2007: 166) and the original reflexivity for inter-normative engagement is lost. In this manner, ontological normativity can become rigid and deterministic.

Connolly’s Methodological Development

This section begins to illustrate at least one way this weakness manifests within a OPT. It does this by illuminating the development of Connolly’s ontological method. It treats him as an illustrative figure of OPT for several reasons. First and principally, the depth of his reflections on the use of ontology is unmatched in other thinkers. Connolly is a methodologue who has not only engaged in in-depth methodological discussion of ontology but has also compared his approach with others. Specifically, in the development of his work Connolly’s turn to ontology is situated in an ongoing discussion of the dominant methodological frameworks in the political sciences, with a particular concern to how emerging challenges (e.g. interpretivist, social constructionist, communitarian, liberal, etc.) require radically rethinking our descriptive and normative capacities. Second, the growing role of ontology in his work is particularly illustrative of both the reasons given to employ ontology, and the problematic tendencies of an ontology-centred approach. Third, Connolly is at the centre of debates that go beyond OPT and agonism. There was a surge of interest in his work following the mainstream debates on ontology confirming this (Campbell and Schoolman, 2008; Connolly Symposium, 2008; Finlayson, 2010). As a result, he is a privileged analytical object to consider the potential strengths and pitfalls of ontology. ⁴

The Early Connolly. From Interpretivism to Ontology

Connolly’s early methodological work occurred during the high period of debates between empiricism and interpretivism. In his Terms of Political Discourse (1994), Connolly offers an understanding of concepts as ‘essentially contestable’ that is interpretive and naturalistic. He argues for the theory-embedded and contestable character of political concepts, the primacy of interpretation and the resulting paradoxical condition of normative and descriptive argument: that practices of rationality are necessary and insufficient to ground criteria. These undermine the law-like model of the social sciences and the assumption that language is neutral medium for representation and prescription. Instead, Terms’s contextualism forefronts the linguistic dimension of political practice in how conceptual relations affect the politics of a community (Connolly, 1994: 39; see also Connolly, 1981: 28). This contextualism raises the possibility of a constitutive definition of politics that privileges language in some way (ontologically, epistemically, metaphysically) as generating political life. As discussed below, this is the overt result of Connolly’s ontological turn. However, it is not a step undertaken in Terms. In the final chapter, he examines the dominant Anglo-American and continental challenges to the essential contestability thesis. For continentals, it does not sufficiently undermine universal standards of reason or politicize the use of concepts because it remains committed to a ‘conception of the self as agent and the agent as a centre of responsibility’. The norms of agency and responsibility which flow from these conceptions continue to be a ‘guide and limit to critical contestation’ (Connolly, 1994: 233).

The ontological method, introduced above in his double-entry orientation to onto-political analysis, replaced this earlier post-positivism. This shift principally occurred between two works: Appearance and Reality in Politics (1981) and Politics and Ambiguity (1987). In Appearance and Reality in Politics, Connolly argues that the emphasis in interpretivist frameworks on exposing the antecedent conditions (i.e. the present linguistic conceptual frame) offers a particular reflexivity that enables the freedom and action of political agents. Illuminating the intersubjective dimension of social reality can enhance freedom and reason, constructed though they be (Connolly, 1981: 28). In contrast, methods like deconstruction and genealogy totalize the project of politicization that is pursued more tentatively in the essential contestability thesis. The important result is that this thereby undermines any normative standards that could flow from them (contextualist or not) (Connolly, 1994: 234). In the final pages of Terms, he closes the book claiming that the critique of the subject does not require deconstructing it in a way that removes any normative value from the conception.

This reveals an early struggle in Connolly’s work over the normative consequences of social constructivism (whether interpretivist or ontological) and the resulting relationship between political theory and practice. Connolly is ambivalent about totalizing politicization. He grappled with how to combine a social constructivist approach to political analysis that could assess political language, practice or institutions critically, while retaining a normative capacity that for him seemed lacking in Continental approaches like deconstruction and genealogy. Furthermore, he did so from within a project of reflexivity that would enable political actors while questioning dominant frameworks. Importantly, this involved rejecting constitutive accounts of language that undermine this reflexive normativity.

Despite Connolly’s initial scepticism of constitutive and totalizing approaches, the next phase of his work argues that an ontological lens for social constructivism can balance critical reflexivity and normative agency. This shift begins in Politics and Ambiguity (1987) and his Foucault essays (Connolly, 1985, 1993a), and is firmly consolidated in the books of the 1990s (Connolly, 1991, 1993b, 1993c, 1995). My claim is not that there is a fundamental division, many common themes and ideas carry over (e.g. contestability, the role of interpretation in politics). Rather, there is a shift in the sources of Connolly’s analysis and thus in his approach, and that this is generally ignored in the burgeoning literature on Connolly, which only focuses on his work produced after his ontological turn.

Connolly’s ontological turn was prompted by a reading of Martin Heidegger. In his early work, Heidegger argued for the necessity of questioning Being through the examination of the situated nature of human being (Dasein), while his later work illustrates the impoverishment of the history of metaphysics, including modernity, as the attempt to ‘grasp’ Being. ⁵ This second aspect is particularly important, as it directly feeds into Connolly and OPT’s ontological approach. Heidegger expands the critique of metaphysics to a critique of modernity’s general ontological orientation and develops an ontology of language that serves as a model for a general post-foundationalist ontological orientation that is seen to be both critical and normatively reflexive. Connolly takes on both of these imperatives in his framing of ontological analysis. However, while he enacts the former situated project through a reception of twentieth century French philosophy (particularly Michel Foucault), ⁶ his understanding of modern thought and its conditions is decidedly Heideggerian. Furthermore, in his discussions of ontology in general and modernity’s relation to it he most frequently references Heidegger and, as I argue below, Connolly’s later work on Foucault and the question of human being are read through that lens. ⁷

In Politics and Ambiguity, Connolly shifts his understanding of language from social constructivism to ontology, through a discussion of the relation between articulation and the world. He follows Heidegger’s view that words are essential to the being of things, that ‘where the word breaks off no thing may be’ (Connolly, 1987: 144). This demonstrates the paradox of articulation. ‘For things to be, they must be brought into a web of articulations which gives them boundaries, specificity, complexity; but any particular web of discourse fixes things in particular ways and closes out other possible modes of being’ (Connolly, 1987: 145). Language is poetic. It calls forth things into being. However, by speaking we determine a thing, isolating it within one level of articulation; this ultimately excludes its other possible modes. This is the structure of Being: it is always limited in a particular determination and thus any eventing (instantiation) of Being is both an unconcealing and a concealing. In this, Connolly takes a constitutive approach identifying a fundamental ontological structure of limitation within language. He expands the social constructivism of the essential contestability thesis, politicizing language by revealing how it confines socio-political life.

The ontological account of language led Connolly to a wider criticism of modernity. He follows Heidegger arguing that modernity is defined by a veiling of this structure of ontology and language. Specifically, it obscures this dynamic through its prioritization of knowledge and assumption that all language can be measured along the truth/falsity axis. This conceals the presence of truth’s other: a fundamental untruth that exceeds the current determination and constitutes it:

Modern ‘truth’ is a mode of revealing that enables judgments of correctness and incorrectness within its frame. But every historical regime of revealing also conceals. It conceals possibilities of being that cannot be brought into a particular way of life without confounding its basic principles of organization. ‘Untruth’ is deeper than truth and falsity, then: untruth is that which cannot achieve sufficient standing within the terms of discourse of a time without stretching contemporary standards of plausibility and coherence to their limits of tolerance (Connolly, 1995: 5; Heidegger, 1998).

For Connolly, Heidegger’s ontological understanding of truth draws that concept out of the epistemological framework of modernity undermining the truth/falsity pairing and exposing the ontological structure it obscures. This pairing conceals the ‘untruth’ that exceeds it. Truth (Heidegger and Connolly’s) is not epistemology but the attempt to understand how truth was reduced to epistemology. ⁸ Connolly follows Heidegger in arguing that modernity’s central flaw is obscuring the presence and necessity of this ontological veil: that the thinking of Being will always have gone too far. The imperative of both their ontological projects is to ‘let the veil appear as what veils’ (Connolly, 1991: 31; Heidegger, 1977: 16). ⁹

Connolly’s consequent imperative is twofold, to (1) explore the conditions which allowed truth to be reduced to ‘reliable criteria for knowledge’ and (2) to engage thinking that can exceed the current determination of thought. That is, to offer a method that dwell in this paradox and the conditions it places on socio-political theorizing and to offer alternative determinations of reality (i.e. ontologies). The former is addressed to close this section while the latter occupies the next.

Connolly’s ontological account of language (as limitation) and critique of modernity (as veiling) led to a general account of the contingency that pervades human socio-political life. The concept of contingency often suggests an interpretive pluralism similar to the essential contestability argument. However, Connolly does not posit contingency as a condition of language-use but as a condition of Being as such, as an ontological condition. This is a constitutive claim that makes contingency about general reality and led to an attempt to articulate an overall conception of contingency that affects not only human language but ontological reality. If all determinations are a limitation then contingency is necessary. Consider two key passages:

By contrast to the necessary and the universal, it [contingency] means that which is changeable and particular; by contrast to the certain and the constant, it means that which is uncertain and variable . . . by contrast to the expected and the regular, it means that which is unexpected and irregular; and by contrast to the safe and reassuring, it means that which is dangerous, unruly, and obdurate in its danger (Connolly, 1991: 28).

Suppose internal and external nature contains, because it is neither designed by a god nor neatly susceptible to organization by human design, elements of stubborn opacity to human knowledge, recalcitrance to human projects, resistance to any model of normal individuality and harmonious community (Connolly, 1991: 31). ¹⁰

Let’s call this argument the OCT. In Connolly’s ontology, contingency is not simply an acknowledgement of the historicity, fallibleness and socially situated nature of humanity, as in other account of that concept (Rorty, 1989). It is something active in the world, that not only limits human knowledge and language but is beyond or resistant to control. He is quick to caution that does not mean contingency is knowable; ‘resistance is not the resistance emanating from a world knowable in itself. It resides in gaps, disjunctures, anomalies, discordances and uncertainties which constantly plague every system of knowledge’ (Connolly, 1993b: 145). In fact, the fullness of this contingency means that social and political thought confronts a paradoxical field of enquiry in attempting critical explanation and normative reconstruction. As Connolly notes, the:

most persistent issue facing critical interpretation today is the ironic relation it assumes to its own ontopolitical projections . . . The sense that this ambiguous condition sets the terms within which thought necessarily proceeds today constitutes our reverence and spur simultaneously (Connolly, 1995: 38).

Human social life is similarly paradoxical at its core in that the drives to explain and prescribe always run up against resistance to those designs (Connolly, 1993b: 138–139, 2011: chap. 4).

The OCT is the core of Connolly’s method and how it contests modern ontologies. Modernity has traditionally responded to this contingency by seeking to master it, something represented in the critiques of concord and mastery that Connolly deploys against a variety of other approaches in this period (Connolly, 1987: chap. 3, 1993b, 1995: chap. 1). I argue however that it creates a deep tension in his thought that undermines his critical and normative aims. Presently however, the discussion will briefly address how it develops in his more recent method, to demonstrate its lasting significance and the terms of that tension.

Connolly’s Mature Method: Ontological Articulation

These dynamics are not isolated within this earlier period of Connolly’s work. Since the late 1990s he has developed the OCT, his positive ontological account, which has served as the basis of his critiques of existential resentment and his theory of agonistic democracy. In terms of its ontological sources and figurations, these have moved through several stages. From an articulation of the ‘protean vitality of being’, he turned to a ‘deep rhizomatic pluralistic universe’, and subsequently an ‘onto-cosmology of becoming and fragility’. This series of figurations have held one common terminology: his materialist ontology of immanent naturalism. ¹¹ This ‘depicts the world as a layered immanent field of “infrasensible forces”’ (Wenman, 2013, 126). From readings of Spinoza, Nietzsche and Deleuze, Connolly takes an image of the world as a ‘common plane of immanence’ in which all bodies, structures, minds and individuals are structured. Deleuze’s inorganic vitalism is an important source, figuring life as a constantly emerging multiplicity of forces, periodically coalescing into assemblages before dissolving, without inherent end or purpose. To return to the previous Heideggerian language, this is a figuration of Being, not only as partial/concealed, but as actively resisting articulation and the necessary social limitation that results. It is an elementary flux that means traditionally distinct categories (e.g. nature and culture, mind and body, etc.) are mixed immanent processes. Connolly has developed this ontology in a several themes (e.g. temporality, embodiment, complexity, becoming and fragility) to emphasize the creative, self-organizing and emergent nature of the world and human life.

This framework is overtly ontological, intended to modify descriptive and normative analysis. It offers a detailed model of the complexity and interdependencies of reality and explains, ontologically rather than epistemologically, why conceptual systems always fail to understand the complexity of the world. For example, this ontology led Connolly to reformulate traditional social scientific concepts, such as agency and causality, both of which he figures in emergent and distributive ways. Similar to ‘new materialisms’ it is an application of the ontological turn to the material world, an extension of OPT’s original social constructivist framework which, as I argue below, stems from the ontological method itself. More importantly, it contributes to his methodological reflections on the consequences of these ontological reflections for critical and normative political theorizing.

Connolly’s main methodological statements of OPT occur from the mid-1990s to the mid-2000s. During this period he offered a model of ‘ontopolitical interpretation’ or ‘ontology’, which acknowledges that while an ontological method cannot offer incontestable understandings of reality, it can argue for the inevitability of ontological presuppositions and the need for affirmative declaration of ontological assumptions (Connolly, 1993c, 1995, 2004, 2008: chap. 3). For him, every political interpretation invokes a set of fundaments about the nature of the world and the necessities/possibilities of human being. These assumptions structure politics and political activity. Initially, he applied these claims to the contemporary matrix of socio-political theory, exhuming hidden assumptions of mastery and attunement, before applying this broadly to contemporary American politics. The intent behind his ontological method is critical-normative destabilization. It aims to destabilize these positions and their assumption of a world predisposed to humanity and intelligibility. For this, he employs strategies of ontological detachment and attachment. Critically, this amounts to deconstructive and genealogical techniques intended to expose covert, structuring assumptions within our political theories, public discourses and broader practices. However, these are attached to an equally important positive normative project. Ontological articulation also destabilizes the hegemony of current ontological assumptions, while offering new contenders for conversion into public debate. These strategies of attachment to alternative ontological figurations, which Connolly refers to as ‘positive onto-political interpretation’, involves projecting explicit presumptions into one’s interpretations of actuality; that is, they require making controversial ontological claims as key to the task of socio-political criticism. Together detachment and attachment are intended as a critical method with a normative impetus.

With the broader analysis of contingency, this method leads Connolly to argue for the necessity of ontology. These two conditions, the inevitable presence of ontopolitical assumptions within all political thought and the necessity of contingency, require political theorists to both examine the ontological assumptions behind various contemporary theories and explicitly articulate the ontology that motivates our interpretations. However, these claims overturn Connolly’s earlier interpretivism and thesis of essential contestability by making the structure of ontology (determination as limitation) constitutive of language and the current socio-political matrix. This is a totalizing criticism which, on Connolly’s own terms, politicizes all concepts without a means of normative assessment. This threatens to compromise the reflexivity I argued he and OPTs seek in critical-normative analysis.

Connolly offers a CM to moderate the danger of ontological methods becoming totalizing and exclusionary in a way similar to ontologically minimalist approaches within his analysis. ¹² This is a modified, reduced version of the essential contestability argument which does not offer a detailed justification for contestability along the lines of an analysis of the nature of concepts or normativity. Rather, it is an ethical claim that one’s presumptions exceed both intention and ability to demonstrate their truth. This acknowledgement of contestability challenges closure in the field by both critically assessing other approaches and refusing to establish another absolute space. For Connolly, this dual movement means his method is a ‘double-entry orientation’ to politics. Critical detachments and affirmative attachments are treated as incontestable initially, but are then withdrawn in an acknowledgement of contestability. For him, as a result of its paradoxical nature, enquiry must proceed in the median space between these gestures (Connolly, 2004: 344; see also: (Howarth, 2010: 22–23)). Thus, ontological affirmations are not merely strategic gestures but comprise ontological assumptions constitutive of both the critiques and the alternatives they present. They are ‘vague essentialisms’ and ‘happy posit-ivisms’. The key point is that these projections are neither merely provisional nor prediscursive realities beyond question. Rather, they are both affirmative (like prediscursive ontologies) and contestable (like a provisional ontology would be). As discussed above, this paradoxical pairing of opposites for Connolly characterizes interpretation in general. The double-entry orientation to interpretation attempts to do justice to the inadequacy and contestability of ontology, the inevitability of ontological presuppositions, and the ambiguous nature of human enquiry. All of these conditions undermine our capacity to make critical (explanatory) and normative claims within socio-political theorizing. Ontology is thus his way of acknowledging and negotiating these limitations.

Operationally, the CM manifests in a series of caveats Connolly makes about the status of his ontology. These abound in his work but usually take a similar form. His ontology is a ‘thought experiment’ through which one detects and considers implicit assumptions in ones own or another’s thinking (Connolly, 1993b: 134). This goes to its role in critical analyses of the view of others: political theorists and/or political groups/ideologies. For him, the:

presumption of resistance carries no more or less credibility now than the presumption of creation and eternal life did previously. It is merely a possibility. But it is a valuable presumption for thought which seeks to revise the text in which it is now inscribed; it is constructive for thought which wishes to think unthought presumptions in its previous thinking; and it is disturbing and unsettling for thought which has floated unconsciously on the surface of a world in distress. This possibility shakes and transforms us when applied relentlessly to contemporary experience. It is thus a tool in the service of thought (Connolly, 1993b: 147). ¹³

The internal importance of contestability to Connolly’s ontological method is often ignored. Pairing this mechanism with the OCT is his principle claim to maintain a self-reflexive approach that still has analytic and normative potential. Essentially, ontology services the claim to expose a relevant and ignored dimension of linguistic and political practice that has real consequences for the normative status of its object, while contestability ensures that this approach is not subject to accusations of totalizing ‘foundationalism’ that have accompanied similar arguments.

Criticizing Connolly: Too Strong or Too Weak

Interest in Connolly’s approach to political theory has ballooned in recent decades. However, despite the centrality of ontology within his method, few critics have systematically examined his method for political theory. Those that do often ignore the development of his method (or only focus on one or two phases). In this, they ignore the problems, around justification and pluralism, that originally motivated his work, as well as the question of whether ontology may be problematic for these ends. ¹⁴ Those that do examine his ontology tend to ignore the distinctness of key structures of his ontological method, the OCT and CM, subsuming his account either in his theory of contingency or his account of contestability As a result they tend to read him in one of two ways, illuminating part of the concerns with his ontology.

Stephen White has notably read Connolly (and others) as offering weak ontological methods along the lines discussed above. For White, what is distinct about Connolly’s ontological method is exactly its weakness. Connolly has enacted an affirmative turn to ontology, that requires that as one criticizes existing ontological assumptions in some area, one attaches oneself to an explicit and articulated set of ontological presumptions. The latter then serve as a basis for normative arguments about the sensibility and ethos that should govern a more generous pluralism. In this way, Connolly offers an ontological method that has contestability at its core. For White (2000: 108), his theory’s distinctiveness is to ‘not simply declare its contestability, but rather enacts it in some way, in the sense of turning the unifying momentum of its concepts back upon itself’. However, this also exposes Connolly’s key weakness. For White, Connolly’s ontologies ‘inadequately prefigure’ his ethico-political norms. When framed within contestability, ontologies of abundance and pluralism can both be ignored by other theorist, or connect to normative ends that Connolly would reject. There is a resulting critical-normative gap between his ontology and politics, undermining the power and relevance of his work to contemporary democratic theory (White, 2000: 127–129).

On the other hand, several critics have argued that Connolly’s ontological methods revert to strong ontological claims that diminish his qualification of contestability. ¹⁵ Mark Wenman has offered one of the most challenging versions of this criticism. For him, Connolly’s politics of becoming and immanent naturalisms are ‘ontologically too weighty’ (Wenman, 2013: 131). By seeing contingency, flux, resistance and becoming as necessary, he assumes that social and political process are determined by ontological dynamics. As a result, his diagnosis of contemporary political problems always reverts back to whether the groups or institutions in question dynamics (e.g. the resentment of right-wing of American politics) can accept the plural conditions of socio-political life he identities in the constitutive nature of his pluralistic (Wenman, 2013: 126–127). Lois McNay similarly claims that Connolly often relies on a simplistic assumption that the contingency of socio-political forms matches the contingency of ontology. This is why Connolly’s method results in a ‘socially weightless’ form of political thinking. In focusing on the essence of politics, as contingency, and reifying that essence into the nature of the political, Connolly’s model is so removed from the actual languages and practices of social life that its analytic and normative relevance is unclear (McNay, 2014:83). Even Connolly’s supporters like David Howard recognize that at his method’s core, there is a ‘fidelity to radical contingency itself’ (Howarth, 2010: 39). Thus, on this second account, the key critical weakness of Conn’s ontology is the tendency to over-determine his critical accounts of contemporary socio-political dynamics.

These seemingly contradictory accounts, that Connolly’s ontology is too weak and too strong ontologically, represent the most in-depth accounts of his ontological method. And yet, they come to seemingly opposite conclusions. This disparity is clarified by the fact that neither White nor Wenman/McNay consider the separate nature of Connolly’s arguments for contingency and contestability. Parsing them out as this account has done allows their inherent tension to come to light, and its consequences examined below.

Ontological Tensions: Connolly’s Two Problems

With this view of the development, motivation, method and criticism of Connolly’s ontological approach, this section assesses these moves. It examines the consequences for the analytic and normative capacities of his theory, arguing that the ‘double-entry’ structure, coupled with the OCT and CM, mean that Connolly’s framework is simultaneously too strong and too weak. This develops the existing criticism around Connolly’s ontological method: there is a continual tension between strong ontological and weak contestable claims in his work that obscures the model of political theory he is offering. The underlying conditions of this tension are found in the imbalance between the OCT, as an ontological thesis, and the resulting CM as an ethical norm. Connolly’s separation and pairing of these two creates a disjuncture between the two main aspects of his theory that undermines the coherence of the model he offers to political thought. This section examines this tension in Connolly’s work and its consequences in his political theory. To do so, it both turns to several compelling criticisms of that politics, connecting them to their roots in his ontological method, as well as to the pragmatic critique of ontology from the work of Richard Rorty. All of this illustrates how ontology is problematic for the reflexive justification Connolly attempts.

The OCT is the central claim motivating Connolly’s ontological method. Human knowledge, and the social and political systems that govern our interactions, all require us to determine and thus limit being, in order to speak and act in concert. For Connolly, it is the human condition that we must limit ourselves and our world, even as they resist that limitation in a more primary dynamic of becoming, complexity, fragility and pluralism that continually erupts. ‘Under these conditions, each worthy design [intellectual, social, political] we enact will subjugate some characteristics while releasing others, create new resistances while dissolving previous ones, and engender new contingencies while taming old ones’ (Connolly, 1993b: 132). The language here is important. Social, political and intellectual life for Connolly occurs ‘under these conditions’ of contingency (later: pluralism, becoming, fragility, complexity). It is in this sense that contingency creates resistances, disruptions and other disturbances within a unity or settled state of affairs. On a strong reading, this could lead some to charge Connolly with an anthropomorphism: of treating reality broadly understood as something with intention and agency. It not only lacks susceptibility to human intentions but actively resists them. However, even on a weaker reading (which I assume here) this is a problematic claim as it seems to create a hierarchy of conditions with ontological contingency at the top, affecting other entities, and human socio-political life at the bottom, being affected.

This tendency to the priority of ontology in Connolly’s method can be illuminated through a central distinction he shares with other ontological approaches. Connolly is linked by his Heideggarian assumptions to what Oliver Marchart has called a post-foundational ‘Heiddegarianism of the Left’ (Marchart, 2007: chap. 1). Like Connolly, post-foundationalists grapple with the necessity and contingency of foundations in political philosophy. Foundations cannot simply be negated; determination as limitation and the necessity of contingency entail that our thinking will always fall short of full presence and clarity, while not being able to pursue any other goal. This ‘quasi-transcendental’ move depends upon Heidegger’s notion of ontological difference, a distinction between the everyday ontic world of beings, and ontological Being which determines the former through the structure of determination as limitation.

Importantly, for both Connolly and post-foundationalists, ontological difference highlights the necessity of distinguishing between politics and the political in socio-political theorizing. This distinction has traditionally been ignored in political theory, as part of the minimalism which excluded ontology as irrelevant to socio-political discussion. However, it is central to many agonist perspectives, and is strongly associated with the work of Chantal Mouffe. Like Connolly’s use of ontology, the distinction between politics and the political is not solely a point about language or the social, but a claim about reality. Politics is the ontic realm of everyday political practice and institutions. That is, it is the current determination of the political in structures, institutions and discourse. The political concerns ‘the essence of politics’, its necessary character. Much like Heidegger’s characterization of Being, as both concealing and unconcealing, the political is the constitutive dimension of political life that structures it through a fundamental limitation. For Mouffe, it ‘concerns the very way in which society is instituted’ through ‘the dimension of antagonism which I take to be constitutive of human societies’. Politics is always a limitation of this underlying dimension, a particular way to order and organize ‘conflictuality’ in order to domesticate it (Mouffe, 2005: 9).

The fundamental conflictuality of the political sits alongside Mouffe’s analysis, in her work with Ernesto Laclau, of the ontological category of ‘discourse’ and the conditions it places on linguistic claims about the world. Drawing on Wittgenstein’s notion of language games, Saussure’s understanding of linguistic structure and Derrida’s argument for différence as the condition of impossibility of full presence in language, Mouffe and Laclau offer an understanding of the pluralistic nature of language rooted in an account that meaning and use always exceed any attempt to lock particular uses of language into determinate meanings. This leads to one of the most revealing concepts in her use of ontology, the notion of a ‘constitutive outside’. This idea denotes an external (ontological) condition on both language use and political life which means any attempt to achieve full meaning, identity or legitimacy/consensus through language-use will always be disrupted by an external element that is excluded. The constitutive outside is a pluralistic claim that there is always much more and that this more is constitutive of the everyday determined reality. It stems both from the antagonistic reading of the political and the pluralistic reading of language (Mouffe and Laclau, 1990: 103–105).

The language of constitution reveals the particularly problematic nature of ontological accounts of politics. Specifically, it reveals that ontology has a tendency to assume its own priority, creating a hierarchy of entities with ontological ones at the top, and everyday institutions, actors and events at the bottom. Once its sense is distinguished from the usual physical meaning (e.g. where buildings are constituted by various materials), there is no clear sense in which language or politics are constituted by pluralism/antagonism. Does this entail that these dynamics are more determinative of their respective objects than other attributes we might ascribe them? What kind of priority is being argued for pluralism/antagonism over say the agreement-seeking aspects of language or the cooperative aspects of politics and how is it known? The tangible sense this claim has is that, as language-users and political actors, we couldn’t get very far if we didn’t admit that meanings change in language and politics always contains the possibility of descent into conflict and/or violence. However, this is simply the pragmatic claim of fallibility, that our ideals of transmitting meaning and overcoming antagonism are never predetermined to be successful, which is why language use and politics will continue without end. For this reason, Richard Rorty argued that the claim that something constitutes something else (i.e. that it is more essential to it than any other aspect) is as empty, as ‘drawing a line around a vacant place in the middle of the web of words, and then claiming that there is something there rather than nothing’ (Rorty, 1982: xxxvi). It is a rhetorical sleight-of-hand to argue that our favourite things have metaphysical priority over all other aspects and all the ways they could be articulated. This is because the idea of constitution, or of essence, assumes an inaccessible (and unconfirmable) and hierarchical condition that is posited as outside of and determinative of this everyday world. Such hierarchies always precipitate the claim that these structural features require a distinct method of study, entail a distinct object of enquiry and require a distinct discipline of enquirers.

Rorty’s work on ontology is especially useful for developing this line of criticism. For him, ontological explanations of social life always fall into a claim of privileged access: that there is some method with a unique ability to understand and articulate the world. Whether it is Kantian categories, Platonic forms or contingent resistance, whenever you posit governing entities removed from everyday experience that condition other, available entities, you create two theoretical problems. First, you create a hierarchy of conditions with those on the top, seemingly immutable, and those on the bottom, subject to change and development. Second, you create a self-referential problem of explaining your theoretical access to those immutable conditions:

all unexplained explainers, are in the same situation as a transcendent Deity. If we are entitled to believe in them without relating them to something which conditions their existence or knowability or describability, then we have falsified our initial claim that availability requires being related by something other than the relata themselves. We have opened up the question of why we ever thought that there was a problem about availability in the first place. We have thereby questioned the need for philosophy, insofar as philosophy is thought of as the study of conditions of availability (Rorty, 1991: 54).

For Rorty, theoretical forms of reflection that necessitate, rather than just argue for the usefulness of, ontology assume that ‘ontology precedes cultural politics’, that reality trumps cultural practice in determining what language a community uses and what objects they discuss through this vocabulary (Rorty, 2007: 3–5). In opposition, he follows Robert Brandom in arguing that the only non-hierarchical way to understand the presence of authority in language is as the practice of certain groups treating certain types of claims as authoritative. As a result, he argues for the ‘ontological priority of the social’. He is concerned not with disproving the existence of ontological conditions but with arguing for the priority of social and cultural categories over ontological ones in social practices. By referring to the presence of an authority (God, empirical reality, contingency, etc.), ontology attempts to take a privileged stance it has no real basis for. ‘The reason why quarrels among metaphysicians about the nature of Reality seem so ludicrous is that each of them feels free to pick a few of her favorite things and to claim ontological privilege for them’ (Rorty, 2007: 106). In social and political theory, for Rorty, we need to remove all external forms of normative and explanatory authority by dropping the assumption that vocabularies like ontology, epistemology and metaphysics precede the social practices of claiming and giving authority to sets of linguistic claims. ¹⁶

The Rortian critique of ontology is especially powerful because Connolly takes this ontology to have normative force. Despite any caveat Connolly can make to contestability, the impetus within his theory to open social and political systems to articulations (whether natural or social) of resistance stems from the claim that these emerge out of something beneath and more primary than the existing socio-political system. Rather than rejecting the whole transcendental mechanism of external criteria, as Rorty does, Connolly ‘naturalizes the transcendental’ (Howarth, 2010: 27). External normative and explanatory authority remains in this model; it is just radically immanent to the world in a series of dynamics (e.g. self-organizing processes, systems in disequilibrium, etc.) that Connolly identifies and then applies back onto sociopolitical questions. The OCT injects a basic normative commitment into Connolly’s work that motivates his critical account of contemporary politics, and serves as the basis of his model of negotiating pluralism. The necessity of contingency as a structure of the limitation of Being through articulation leads to a basic normative commitment to the pluralism that always exceeds any current determination, linguistic or material. This means that Connolly places political theory under a general ‘responsibility to otherness’ that normatively values the alternative potentialities of current forms over their present manifestation (White, 1991: 20–27). This, as a result, leads to a consequent devaluation of the situated present and whatever social forms we currently employ, solely on the basis of the ontological possibility of them being-otherwise.

This tendency to devalue the present is most dramatically revealed in his ontological reading of Foucault on freedom, where he attempted to provide a consequent ethical sensibility for contemporary democratic politics. For Connolly, Foucault’s genealogies ‘support an ontological thesis with political implications’ (Connolly, 1985: 365). Specifically, Foucault claims that discourse, linguistic practice with its inevitable norms and exclusions, is ‘a violence we do to things’ (Connolly, 1985: 368). Foucault’s accounts of various social practices intend to incite the experience of discord and pull us out of established frames of meaning and practice. The motivation behind this is ontological; ‘every such settlement [social order] involves imposition even while it may enable life to be in particular ways’. The problem is in where this logic leads. If the norms of social life are always a sedimentation of a more primary, constitutive flux, then every set of norms is imposed and one can legitimately assume that every set of norms must be resisted. ‘Freedom, in this perspective, is not reducible to the freedom of subjects; it is at least partly the release of that which does not fit into the moulds of subjectivity and normalization’ (Connolly, 1985: 371). As above, otherness, as an undifferentiated bloc, is normatively prioritized over the present in an unreflective and highly non-specific way as bearing some sort of authority on the present. Thus, the type of hierarchy established within an ontological account of politics violates the very pluralism and anti-authoritarianism it seeks by determining the pluralistic conditions of political life and then drawing norms out from that very process which have a claim to be situated in ‘how the world really is’, and thus have an authenticity based in their constitutive nature from which to make normative claims. ¹⁷

This develops one of the most challenging criticisms of Connolly’s politics of ethos made in recent years. Inder Marwah and Alex Livingston have both expressed concerns about Connolly’s affective politics of ethos (Livingston, 2012; Marwah, 2015). By levelling rational reasoning and visceral affect, Connolly politicizes ethos in a way that pushes the boundaries of democratic legitimacy. In effect, in responding to what Connolly sees as a manipulative politics of affect on the Right, which encourages politics of resentment, he supports an equally manipulative politics of generosity. For both Marwah and Livingston, the issue is that he responds to a covert politics with another covert politics, just because he reduces political behaviour to affective conditions that exist below the conscious level of citizenship. For Livingston, this produces just the sort of tension I identify in Connolly’s ontology, at the political level:

Torn between the adversarial and manipulative politics of war machines on one side and an ethically sensitive negotiation that seeks understandings and mutual respect on the other, Connolly’s recipe for a democratic reinvigoration of the left oscillates back and forth between a worrisome politics that would subvert popular reflection and a dialogical art that would provoke it (Livingston, 2012: 289).

Connolly is implicitly aware of ontology’s tendency to colonize his critical and normative accounts. Thus, his second methodological move imposes the CM. The role of this mechanism is especially clear in his later methodological discussions that extend the double-entry orientation. For him, every ontology, what he calls an ‘existential faith’, is not only assumed by our beliefs and enquiries, but layered into our affective dispositions, habits and institutional priorities. They are embodied, lived and structured in the materiality of our socio-political lives. Consequently, these existential faiths have deep connections to the methods of enquiry and political problems identified in any approach (Connolly, 2004, 2008: chap. 3). The continued hierarchical structure is important. The CM relativizes, attempting to save ontology from authoritarianism, while still claiming that each framework’s ontological assumptions are determinative of its methods, problems and politics. That is, it assumes a hierarchy of reasons with ontology at the top and the others flowing from it.

As a result, Connolly takes the CM one step further. For him, the connections between method, problem and faith cannot be schematized, as there is no ‘general formula’ for how they are connected. He rejects the idea of a method of adjudicating between such competing frameworks as this assumes a super-problematic capable of resolving inter-methodological disputes (Connolly, 2004: 338–339). However, this raises a problem around inter-methodological relations. In the absence of such neutral arbitration, how are we to speak to one another across ontological, methodological and political difference? This goes to the question of how political theory in general can productively engage with OPT and its claims. Similarly, it has important consequences for engagement across cultural-political difference in contemporary democracy. Connolly’s answer resembles his noted model of agonistic democracy: inter-methodological engagement should be governed by an ethos of engagement:

The first thing is to articulate comparatively the problematic you support with as much energy and skill as you can, pressing others to take it seriously as a possibility to respect . . . The second thing is to come to terms publicly with the persistent element of contestability in the eyes of others of the faith infusing your problematic (Connolly, 2008: 90).

Beyond the incredible broadness of such a framework for engagement, which Connolly gives few further details about, this ethos places all the work of normative engagement on self-examination and declarations of our methods, problems and faiths (i.e. ontologies) that motivate us. Thus, it requires us to examine our often hidden ontological assumptions, connect them to our approach to politics and understand how this raises and excludes certain political problems. As such, all the weight of this model is placed on ontological examination and articulation as the main resource for, and condition of, making normative claims across significant differences.

As a growing list of scholars have argued, this presents a significantly reduced task for political theory that potentially compromises any disciplinary claim to distinct approaches or tasks different from any socio-cultural production. After White’s first account of weak ontology, Elizabeth Wingrove wondered whether there wasn’t a dangerous move towards theology in ontological frameworks. The injection of acknowledged faith-based judgements into political theory, in her view, stands under an ambiguity in who exactly is doing this affirming. While it may be quite right and positive for political theorists to justify political actors making various sorts of ontological claims in the ebb and flow of politics, especially in terms of how those have been delegitimized classically in liberal theory, it is quite different for the task of political theory to be ontological articulation (Wingrove, 2005: 88–90). Similarly, Colin Koopman has wondered if Connolly’s use of James’ pragmatism to justify ontological analysis misses the point of a key distinction James makes between acts of judgement and the work of critique. The former is key to the everyday life of political actors who must make controversial judgements, declare them and act. However, the latter, where theorists attempt to offer new concepts, descriptions and possibilities for political life is the second-order task before the discipline of political theory (Koopman, submitted). The point is not that there is any hierarchy of legitimacy between these two activities but that they are focused on different ends with different resources (at least partially). Ontology as a vocabulary for articulating controversial claims about the broad nature of reality may not, on this view, be an appropriate forum for political theorists to do the work of critical and normative innovation.

Conclusion

These tendencies to overly strong ontology and too weak contestability within Connolly’s ontological method raise serious questions about the role of ontology in political theory. On this account, there is a persistent tension in Connolly’s method between the OCT and the CM. The former frames contingency as an ordering condition on socio-political life, re-creating a hierarchy to reasons with ontology at the top. The latter, in contrast, insists this is simply a contestable judgement, a faith like any other and all political theories, potentially undermining the idea of any task before that literature. This tension leads to, at the very least, a real unclarity around the status and use of Connolly’s method and the role he lays out for political theory. While there is great nuance behind his project and method, and many of his analyses of other theories and problems are highly insightful, this article has argued that this tension raises important questions for the contemporary methodological turn in political theory. There are of course other ontological approaches that require similar concern and assessment, as they construct and enact their methods within different terms.

My sense here is that the problem with Connolly’s method is one of indeterminacy. That the median space he carves out for political thinking is normatively suggestive without being normatively productive. In the context of deep disagreement (both in political theory and democratic politics), which will extend to the presence and use of speaking ‘ontology’ itself, it frames differences in highly abstract and impenetrable terms as metaphysical differences. This does not provide resources for reflective engagement or the construction of possible dialogue that could lead either to productive disagreement or new areas of agreement. In these ways, Connolly’s theory ends up being deeply in tension with itself, either too strong in its claims about contingency or too weak in its self-understanding of the status of its claims. Importantly, this leaves us without a good reason to speak ontologese.