Critical Theory and Processual Social Ontology

1 Introduction

At first glance, critical theory and social ontology seem to be incompatible approaches to the social, even when the very notion of social ontology is understood in the broadest sense of the term, that is, as a discussion about the types of entities that compose the social world and the type of being that distinguishes social reality from other types of realities. As a matter of fact, the very notion of critical theory, at least in its Frankfurt School sense, refers to social theoretical projects in which social ontology has never really been taken seriously. In its initial program set out by Horkheimer in the 1930s, critical theory is defined by two theses coming straight from Marx: that of the unity of theory and practice, and what could be called the unity of knowledge and critique. What specifies a critical theory is that it refuses to restrict itself to merely contemplative ends, wanting also to contribute, with theoretical means, to practical efforts toward social transformation. What is required for such a contribution is not only a critique of the social world, but also knowledge of the factors that could foster or impede practical efforts toward social transformation. Hence, two key contrasts with social ontology come to the fore. Firstly, from the point of view of a critical theory, social ontology appears as a merely speculative account of the social world; in other words, using Horkheimer’s famous distinction, social ontology is not a critical but a traditional theory (Horkheimer 1999). Secondly, its account of the social world, being ontological, remains too abstract and undifferentiated; in other words, social ontology remains a fairly classical philosophical approach to the social and not a fully fleshed social theory engaged in a systematic and differentiated knowledge of the social world (Horkheimer 1993). Drawn in this way, this twofold contrast looks fairly sharp, yet it does not mean that a critical theory should necessarily reject any kind of social ontological discussion. This becomes clear when another distinctive feature of critical theory is taken into account, namely the role of self-reflection.^[1] Indeed, what specifies the type of social theory promoted by critical theory is not self-reflection in general but specific types of self-reflection: epistemological self-reflection on the principles and methods of the theory, sociological self-reflection on the social standpoint of the theory, and political self-reflection on the practical consequences of the theory. For instance in Adorno, the notion of social theory clearly denotes an attempt to elaborate an epistemological, sociological and political self-reflection on the principles, methods and practical consequences of the social sciences (Adorno 2008; Renault 2012). But this self-reflection could also very well deal with the ontological presuppositions of the social sciences and of such a social theory.

For such an ontological self-refection, two options are open according to which the very idea of social ontology varies in meaning. One option is to discuss basic concepts of a critical theory of society such as alienation, recognition or immanent critique, and to unfold this discussion from the point of view of a social ontology in the contemporary sense of the term, that is, in the theoretical framework elaborated by analytical philosophers such as Gilbert or Searle (Ikäheimo and Laitinen 2011; Stahl 2013).

The second option consists in taking the notion of social ontology in its broadest sense, that is, as a discussion about the types of realities that constitute the social world and the type of being that specifies social reality, and in trying to make explicit the social ontological presuppositions that are assumed by a critical theory when it works toward its specific goals: elaborating a social theory from the perspective of a social transformation. This second option seems more compatible than the first one with the methodological orientation of the Frankfurt School critical theory. In fact, the project of elaborating a social theory through self-criticism of the social sciences is inspired by Hegel’s project of a philosophy of nature conceived of as a critical systematization of the natural sciences (Renault 2001). Hegel contended that each natural science has its own metaphysics^[2] (Hegel 1970, p. 202), and that a philosophy of nature has to make them explicit and compatible with each other in a general conception of nature. According to such a methodological model, no social ontology could be elaborated in any other way than through a critical analysis of the ontological assumptions of social sciences. And since critical theory is striving toward ends that are not only theoretical (as is Hegel’s philosophy of nature) but also practical, this ontological self-reflection of the social sciences has to be combined with a critical analysis of the ontological assumptions of a project of radical emancipatory social transformation.

The social ontological discussion that derives from this model is twofold. On the one hand, it concerns the social ontological assumptions of the various social sciences (and of the various research programs within each social science) that a social theory endeavors to integrate. What is meant by ontological assumptions here is the general conception of the society as a whole that is embedded in the basic concepts and principles of a given social science (and a particular research program). The issue at stake is to make these assumptions explicit and to decide how they could become compatible with a theory that is not only dealing with a specific sector of social reality, but with social reality as a whole. This type of ontological self-reflection is clearly at play when Adorno tries to interconnect the psychological, economical and sociological accounts of the social (Adorno 1976), or when he criticizes certain sociological research programs that think of social reality through the static/dynamic conceptual dichotomy (Adorno 1961). On the other hand, the social ontological discussion at issue concerns the very aim of a critical theory of society: thinking of society as susceptible of undergoing radical social transformations and participating, through theoretical means, in practical efforts oriented toward such transformation.

The claim of this article is that what is required for critical theory is neither a substantial social ontology, nor a relational social ontology, but a processual one. I will vindicate this claim in three steps. The first step will clarify the key distinctions between substantial, relational and processual social ontologies. The second step will discuss the ways in which these distinctions could be used in social ontological discussions. The third step will focus on the various possible social ontological approaches to the issue of social transformation.

2 Substance, Relation and Process

The term “processual ontology”, or “process metaphysics”, is usually associated with Whitehead’s philosophy. But it is also often used as a broader label to denote an ontological option that can be traced back to Leibniz and Hegel, and that is particularly well-illustrated in the works of authors such as Bergson and Dewey (Rescher 1996, ch. 1). This broad use of the term is at issue in the Leibnizian definition of the monad as “appétit” (Leibniz 1991), or the central Hegelian thesis that “die Idee ist wesentlich Prozess” (Hegel 2010, par. 215), or the role given by Bergson to his theory of “élan vital” (Bergson 1998), or the Deweyian contention that “the interaction of organism and environment, (…) is the primary fact, the basic category” (Dewey 1978a, p. 129). In the contemporary literature devoted to process metaphysics, many attempts have been made to spell out its distinctive features. Processual ontology is usually contrasted with substantial ontology, and sometimes identified with relational ontology, as is the case in Rescher (1996, ch. 2; 2000, ch. 2). But it seems more appropriate to consider the concepts of substance, relation and process as defining three rather than only two distinctive ontological frameworks.

In what follows, what will be meant by substantial ontology is the set of theories that assume a primacy of substance over relations and becoming. Aristotle is the most important instantiation of these theories for three main reasons: firstly, he identified being with substance; secondly, he contended that the concepts of relation and becoming presuppose the concept of substance (Aristotle 1974, 6a–8b, 15a–15b); and thirdly, he considered that relations and becoming have to be explained by the essence of substances (Aristotle 1994). Aristotle argued that a relation is either a relation of a substance to something else or a relation inside a substance. In both cases, relation presupposes substance. And in both cases, the nature of a specific relation is always to be explained by the substance, or the substances, to which it is relative. The same is true of becoming: it is always a substance that becomes, and its becoming has to be explained by its essence.

By contrast, the notion of relational ontology can be used to denote theories that assume a primacy of relation over the interrelated terms, and over becoming. The best instantiation of this way of thinking about reality is provided by the idea of a physical law. The laws of nature elaborated by Newton or Galileo are sets of constant formal relations that attribute spatio-temporal properties, that is, relational properties, to the interrelated entities. They explain the physical behavior of these entities by a functional relation between these properties. According to Cassirer, the very project of Newtonian science provided the main incentive for the specific substitution of relational thinking to substantial thinking that took place in Kant (Cassirer 1980). However, it is also true that the distinction between “thing in itself” and “phenomenon” makes it difficult to depict Kant as a full-fledged relational ontologist. In what follows, what will be meant by relational ontology is a position that rejects the classical contention according to which reality in its first sense is substance and relations should be relegated to a secondary position, having only a derived reality. Relational ontology will denote a position that does not only acknowledge the full reality of the relations, but also considers that there is more reality in relations than in the interrelated terms. It would be quite difficult to find such a position in the history of metaphysics. But this is quite easy in the history of social and human sciences. In fact, the various forms of sociological structuralism share such ontological assumptions, from Lévi-Strauss (1974) to Bourdieu (1992).

Now, the distinctive feature of a processual ontology is the fact that the relationship between the relation and the interrelated elements is internalized and conceived of in dynamic terms. On the one hand, the interrelated elements exist nowhere else than in their interrelations so that the elements are no longer external to their relations (as in substantial ontologies). On the other hand, their interrelation is nothing else than the development of their own activity so that the relation does not have any kind of ontological priority over the elements (as in relational ontologies). Thought so, the idea of process denotes the fact that the mutual activity has the power to modify the properties of the elements as well as the form of relation that shapes this mutual activity. The two best philosophical instantiations of this type of ontology are probably to be found in Hegel and Dewey. In Hegel, the most satisfying definition of reality is provided by the concept of “Wirklichkeit” understood as “energy” and dynamic overcoming of the external relation between internal and external reality (or “reciprocal action”) (Hegel 2010, par. 142). In Dewey, “interaction” and “transaction” are considered as the main philosophical categories, and they are conceived of in terms of an internal relation as well as a dynamics of readjustment between the interrelated elements (Dewey and Bentley 1985, p. 97–110). It is sometimes considered that Hegel’s ontology is a relational ontology, but his criticism of the dualism between relation and related terms, his specific conception of internal relations, as well as his dynamic conception of “Wirklichkeit” as self-transformative relations, all of this amounts to a processual ontology rather than to a relational ontology, at least in the sense given here to these notions. And it is likely that Dewey’s processual ontology is part of what he termed the “permanent Hegelian deposit in his own thinking” (Dewey 1985, p. 154).

An important point to note is that substantial and relational ontologies share a Platonist presupposition. They identify being with self-identity and permanence, be this self-identity and permanence of substances or of relations. A substance remains self-identical and is what remains permanent in the becoming. A law is a set of relations that enable subsuming a variety of phenomena under a single formula, and this set of relations is what remains permanent in the becoming. By contrast, processual ontologies consider difference and becoming as primary features of reality, whether this is to be understood in terms of self-differentiation (in Leibniz), in terms of contradictions and sublation of contradictions (in Hegel), or in terms of a dialectic between tendencies and obstacles (in Bergson or Dewey). This does not mean that the issue of becoming could not play any role in substantial of relational ontologies; on the contrary. It is clear enough that Aristotle’s substantial ontology is intended to contend, against Plato, that becoming is real and not only apparent, and that it deserves a specific explanation. It is all the more clear that the very idea of a physical law presupposes the reality of the motion and that it intends to explain it. Moreover, in such ontologies, becoming can also play an explanatory role. Aristotle’s substantial ontology is associated with a genetic account of natural substances (Aristotle 2008, books I and II), and structuralism can give room to history, as in Lévi-Strauss (1974) or Althusser (2009). The fact that becoming presupposes substance is fully compatible with the fact that substances become what they essentially are in a process of generation, before losing their essential attributes in a process of corruption. Similarly, the fact that structures constitute the core of the social reality does not preclude that they could have a historical genesis. What specifies a processual ontology is not the claim that substances or sets of relations have a genesis, but that this genesis does not have less reality than its results, be they substances or sets of relations, and that these results are only moments of a becoming.

To sum up, what is characteristic of processual ontologies is that they give full reality to relations and becoming. This does not mean that they give only a secondary reality to substantial and relational properties. On the contrary, substantial and relational properties are moments (in Hegelian terms) or phases (in Deweyan terms) of processes. Therefore, processual ontologies are more integrative than substantial and relational ontologies, or, in Hegel’s terms, they sublate these ontologies (it is also in this sense that the concept of “actuality” is the sublation of the whole series of ontological concepts that have been previously considered by the Science of Logic). But it does not follow from this integrative dimension that processual ontologies are compatible with substantial and relational ontologies. On the contrary, it is incompatible with them since processual ontologies give full reality to relations and becoming whereas substantial ontologies refuse to give full reality to relations and becoming, and relational ontologies refuse to give full reality to becoming.

3 Two Types of Social Ontological Approaches to the Social

Before considering how these distinctions could be used in social ontological discussions, it is important to stress the fact that the very notion of social ontology could be understood in various ways. On the one hand, as already noted, the term “social ontology” refers notably to the ways in which analytical philosophers such as Gilbert and Searle have framed the discussion about the social. In these discussions, the question of the meaning of “ontology” as understood in “social ontology” is not a real matter of interest. Nor is the question of the type of “ontology” that is supported or presupposed. On the other hand, the contemporary discussions about the ontology of the social are not only taking place in this analytical framework and are open enough to allow various definitions of what an ontological approach to the social is or should be. In what follows, I will simply take for granted that there are two possible types of ontological interrogations about social reality, without assuming that this distinction should be accepted by all of those who think of themselves as doing social ontology. The first of these two ontological interrogations deals with the types of entities that compose or constitute the social world. The second approach deals with the type of being that is proper to social reality.

In the first case, the core issue is: what are the types of entities that compose the social world? Various social ontological positions are defined by the various possible answers to this question. These social ontological positions can consist either in articulated theses and arguments or in implicit assumptions, and they can be elaborated in philosophical theories as well as in the social sciences. In order to map out these various positions, three distinct social entities can be distinguished: individuals, relations and institutions. Relations can be conceived of either as relations between individuals, at the micro level, or as relations between institutions or groups of individuals, at the macro level.^[3]

In the social sciences, as well in social theory conceived of as a global and differentiated knowledge of the social world, an articulated social ontology is usually thought to be required only with regard to this first type of interrogation (what are the types of entities that compose the social world?). As a matter of fact, in social sciences and in social theory, the question at stake is not so much that of the type of being that specifies the social world, but rather the role played by individuals, relations and institutions in the social functioning and in social transformations. Weber’s Economy and Society provides the best illustration of a systematic attempt to articulate a social ontology in this first sense: it consists mainly in spelling out the nature of the basic entities that are constitutive of the subject matters of the social sciences (Weber 1978).

But it is also possible to engage in the second type of ontological interrogation about the social and to wonder which type of conception of being is most appropriate in a theory of the social world. This latter issue can again either be considered as such, or indirectly, as when one wonders which conceptions of being are assumed by the main paradigms in social sciences. This latter, indirect, approach to ontological issue corresponds precisely to what a social theory in its Adornian sense can try to do. It is when we adopt this approach that the distinction between substantial, relational and processual social ontologies becomes a convenient means for mapping out different orientations.

There are two different ways in which social sciences or social theories may assume a substantial social ontology. The first focuses on the conceptual opposition of individual versus social relations. It reframes the Aristotelian argument of the primacy of substance over relation, by contending that individuals have primacy over their interrelations. This type of ontological assumption is characteristic of all forms of methodological individualism that are based on atomistic premises. A paradigmatic example here is neoclassical economics (Menger 1976) or imitation theories in sociology (Tarde 1907). A second type of theory assumes substantial premises when it locates social reality in institutions rather than in individuals, and conceives of the relations between institutions and individuals through the Aristotelian model of the anteriority of the whole over the parts. This type of ontological assumption is characteristic of the Durkheimian definition of institutions as a reality having stability and authority over individuals, and it is precisely this definition that leads to the idea that the social should be studied as a “thing” (Durkheim 1965), that is as a substance.

Let us consider ontological assumptions of the relational type. Here one has to distinguish again between different ways in which social sciences or social theories may assume a relational social ontology. Various types of social research programs assume that individual behaviors as well as institutional functioning, are structured by sets of social relations existing either at the micro level of interactions, or at a macro level irreducible to this micro level. Weber’s definition of the “social relation” as a form of interaction that can be anticipated by its various participants (Weber 1978, ch. 1, § 3–4) provides a good illustration of the first type of relational ontological assumptions. The second type of relational ontological assumptions is characteristic of structural social theories. Bourdieu’s definition of the “social classes” through their location in a social “field”, that is through the various hierarchical relations that structure the social world (Bourdieu 1991), is an interesting example since he endorses explicitly a relational ontology by reference to Cassirer (Bourdieu 1990, p. 40).

There are also different ways in which social sciences or social theories may assume a processual social ontology. At the micro level, it can be highlighted that even if individuals never exist outside of social interactions and through social roles that are imposed upon them by specific logics of interaction, since they actively contribute to the permanence of these logics of interactions, they also have the power to transform them within the process of the social interaction. This is one of the main theses of Goffman’s theory of “frames of interaction” (Goffman 1974). The fact that the Chicago school of sociology has been deeply influenced by Dewey’s philosophy (Joas 1993) gives another reason to think of sociological interactionism in terms of processual social ontology.

A processual social ontology can also be assumed at the macro level when institutions are conceived of as existing in a network of internal relations, and as being involved in a process that transform them as well as this network of relations. Marx’s theory of capitalism as “an organism capable of change and constantly engaged in a process of change” (Marx 1990, p. 93) provides an illustration of this type of approach to the social. Here, capitalism is not only defined as a social structure, or as a set of macro level “social relations of production” that shape the whole social world. It is also defined by “tendencies”^[4] rooted in the functional relations between institutions. These tendencies have the power not only to transform the basic institutions (transforming for instance “manufacture” into “large scale industries”), but also the very “social relations of production”, giving rise to various forms of “socialization of the production”, be it through share holder companies, or through cooperatives. It is worth noting that Marx gives much weight to the very concept of “process”, defining his Capital as a theory of the “process of the capitalist production”.^[5] His use of the concept of “process” should be traced back to Hegel’s processual ontology, as becomes clear when Marx explains what he means by process in a note in the French edition: “a development considered in all of its real conditions” (Marx 1978, p. 181).^[6] It is precisely this type of social ontological premises that critical theory has inherited. It is a distinctive feature of social theories premised upon the project of a critical theory to think of the social world from the point of view of the tendencies^[7] that transform it, in order to elaborate accurate social diagnosis and political positions adjusted to the present situation.

In discussions about the types of social ontologies that are presupposed by social sciences, the distinction between substantial, relational and processual social ontologies should only be considered as a distinction between ideal-types. Since social sciences do not aim to elaborate social ontologies, there is no reason why their social ontological presuppositions would be necessarily univocal and mutually compatible. If the three social ontologies define three distinct theoretical orientations as far as ontological discussions are concerned, they are best thought of as ideal-typical distinctions in so far as what is at stake is not to specify social reality as such, but to describe and explain social phenomena. Indeed, certain research programs or theories in social sciences may instantiate only one type of social ontology, whereas others may instantiate two simultaneously. A good illustration of such mixed social ontological assumptions is provided by Althusser who, on the one hand, elaborates a “structuralist” interpretation of Marx’s Capital that leads him to conceive of “social formations” as “structures in dominance” (Althusser 1970), and, on the other hand, conceives of history as a “process without subject”, relating his interpretation of Marx’s concept of “process” with Hegel (Althusser 2004).

Within social sciences, many authors have been interested in making the presuppositions of their theory explicit and have therefore been led to some kind of social ontological discussion. This is the case when Bourdieu mentions the polarity between relational and substantial theories (with reference to Cassirer, as already noted) in order to explain the nature of his structuralism (Bourdieu 1990, p. 40). In such ontological elaboration, each social science may be led to support a different ontology. This might be thought to suggest that there is not only one but many types of conceptions of being that apply to social reality considered as a whole, and that in each of the specific domains studied by the social sciences the generic traits of reality are different. But if a general social theory is to be possible, it has to elaborate a conception of society in general, and it necessarily has to support one among the three social ontologies. It is of course possible to respond to this by insisting that social reality is too diverse and fragmented to be subjected to a unique ontological description. But critical theory rejects such a position. It assumes that social reality is unified by certain structuring processes that have to be transformed, and that a global knowledge of the social world is required for thinking of the possibility of a radical social transformation and providing theoretical tools for practical efforts of bringing it about.

Adorno provides an illustration of a type of general social theory that tries to integrate the knowledge and methods elaborated by various social sciences and to make explicit their ontological presuppositions as well as its own ontological presuppositions. In order to achieve this ontological articulation, Adorno employs a method thought of as a continuation of the efforts made by social sciences to make their presuppositions explicit, that is a method of self-reflection. This method is also conceived of as an attempt to think the social sciences from the perspective of the transformation of the social world. In Adorno’s view, thinking of the social world from this perspective involves criticizing various types of Platonic assumptions at play in the social sciences. In his article “‘Static’ and ‘Dynamic’ as sociological categories” (Adorno 1961), he rejects two types of identification of reality with permanence on the one hand, and becoming with appearance or lesser degree of reality on the other. The first type, attributed to Comte, from whom Durkheim inherits his substantial social ontology, considers society as a social organism that is the subject of its becoming. The second one, attributed to Weber, identifies implicitly social reality with a set of constant relations that a theoretical construction of ideal-types is intended to describe. Even if Adorno is not using these words, he is clearly criticizing substantial and relational presuppositions and is contending that only a processual social ontology that considers the social in terms of contradictions and tendencies is able to pay due attention to different kinds of ongoing social transformations and to the role of social action in these transformations.^[8]

4 Different Conceptions of Social Transformation

What are the implications of the substantial, relational and processual social ontological assumptions for the issue of social transformation? The first thing to note is that due to their identification of reality with identity and permanence, following the old Platonist metaphysical prejudice, the first two approaches cannot give full reality to social transformation and are led to consider it as a secondary feature of the social world. This is precisely what Adorno rejects. Nevertheless, it is not impossible for these first two approaches to give genetic accounts of substantial or relational realities that are identified with the core of the social.

The fact that social sciences and social theories assuming substantial or relational ontological premises will usually be reluctant to give due consideration to processes of social transformation finds an interesting illustration in Searle’s The Construction of Social Reality, in his discussion of the “the secret of (…) the continued existence of institutional facts” (Searle 1995, p. 117). That the basic trait of social reality is on Searle’s account permanence rather than transformation, means that he assumes a substantial ontology. But since his explanation of this continuity of existence is provided by a theory of collective intentionality, The Construction of Social Reality does nevertheless combine substantial ontological assumptions with a genetic account of social reality. Of course, giving a genetic account of permanence is still a far cry from thinking of permanence as a phase of processuality. But what does Searle then have to say about social transformation? He contends that institutional changes, such as the destruction of the Soviet bloc, are explained by the fact that the system of functions and statuses has ceased to be accepted (Searle 1995, p. 91–92). In other words, social transformation is analyzed in genetic terms as a loss of social reality: the necessary condition of full social reality (collective intentionality) has been undermined. In this account of a particular social transformation, two issues are disputable. One is specifically ontological: should it be assumed that there is more reality in the “continued existence” of institutions than in their transformations, or is this a questionable assumption? The other issue is explanatory: is it helpful to explain a particular social transformation in terms of loss of a necessary condition (collective intentionality) of full social reality? In so far as a critical theory wants to locate tendencies and contradictions at the core of social reality, it will have to refuse the assumption that there is more reality in the permanence of institutions than in their transformations. As we noticed from the outset, a critical theory will also point out that a particular social transformation cannot be sufficiently explained from the abstract point of view of a social ontology, and that a differentiated social theory is required. Why is it that an institution is at a particular point in time no longer accepted, and how and when does the shift in collective intentionality lead to institutional changes rather than a new permanence that could for instance be described in terms of “institutional lag” following Dewey (1978b), or as “world alienation” following Arendt (1958)? These questions are not articulated by Searle because they belong to a social theory rather than a social ontology, but they are crucial to a genuine explanation of any institutional transformation.

For sure, it remains possible also for social sciences or social theories relying on substantial or relational ontologies to take processes of social transformation into consideration. This is not only possible but also necessary since social transformation is one of the most traditional subject-matters of the social sciences: social sciences are expected to elaborate convincing and accurate explanations of processes of social transformation. But again, substantial or relational ontological premises may lead social sciences and social theories to denying the full reality of these processes. This is quite clear in Durkheim’s theory where social processes are mainly articulated as pathological problems: “social pathologies” or “anomy” (Durkheim 1951, books II and III; 1997, part. III) are characterized by a lack of reality. The practical aim of the theory is then to find the means for stoping these processes and returning to permanence and stasis. Here, the contrast with critical theory is clear and in no need of further elaboration: for critical theory, the study of social processes should be a first priority and the practical aim should not be to stop the processes but to give them their best possible orientation.

The second thing to note is that the three different ontological assumptions can lead to distinct conceptions of social transformation. Firstly, from the point of view of a substantial ontology, social transformations can be understood either in terms of transition to a lower level of reality, as in the Durkheimian conception of social pathologies, or in terms of destruction of reality, as in Searle when he explains social transformation by a shift of collective intentionality and understands the shift as undermining social reality. Secondly, from the point of view of a relational social ontology, social transformations tend to be understood as the emergence of properties that are not compatible within the existing social relations, as something that breaks with the current logic of social relationships. These transformations are not considered as loss of reality but as a rupture, or as an event. It is surely not by chance that French post-structuralist thinkers have put so much weight on the concept of “event”, both form a theoretical and political perspective: they consider social transformations as events rather than as historical processes,^[9] and radical politics in terms of participation in revolutionary events rather than in processes of social transformation.^[10] By contrast, what is specific with the processual approach is that social transformation is conceived of neither in terms of loss of reality, nor in terms of the emergence of the new that breaks with the past, but in terms of a new that comes from the past and that transforms the present. On the one hand, since the new comes from the past, it has to be explained by existing structures, contradictions, tendencies and obstacles. On the other hand, the new transforms the present since it involves a rupture with the past that initiates a process that impacts its actual conditions. In other words, the reality of the new cannot be located in itself, as is suggested by the very notion of event, but it spills over to the past and the future. This idea can again be traced back to Hegel, namely to his conception of the present as the process unifying the past and the future, an idea that has been taken up in Dewey’s presentist account of the relationships between present, past and future.^[11]

As a conclusion, it should be admitted that that social sciences and social theories can no doubt elaborate robust explanations of social transformation whatever their ontological assumptions. Yet, it makes a difference whether an explanation assumes, explicitly or implicitly, a conception of the new as a loss of reality, as an event, or as a moment of a process, and it is only in the third case that it accords full reality to the dynamics of social transformation. This also means that for a social theory that thinks of politics in terms of practices rooted in tendencies and contractions, and that thinks of radical transformations in terms of dynamics of social transformation, only the processual ontology is a consistent option. For sure, even a fully articulated processual social ontology could not all by itself provide a direct contribution to the study of the dynamics of social transformation and to practical efforts of transforming the world. What it could do, is to help critical theories in clarifying their principles and to offer conceptual tools for critical reflection on the social sciences as well as for bridging the gap between social theory and practical efforts toward social transformation. In other words, such a social ontology could be useful for the particular kind of self-reflection associated with the very idea of critical theory.