The Other Social Science: Three centuries of common heterodoxy

The early development of the Other Social Science

The roots of the Other Social Science run parallel to the formation of the social sciences themselves. The work of Vico comes up periodically in the social sciences – most recently in the 1970s and 1980s; see, for example, Stark (1976) and Mills (1982) – and is sometimes named by political scientists as the founder of constructivism (e.g. Jackson and Sørensen, 2010: 161). His work certainly forms part of a movement against what was in the 18th century becoming an anti-Enlightenment consensus that would eventually form the Romantic movement in Western thought (see Berlin, 2013). His famous verum esse ipsum factum (‘the true is itself made’) argument, first formulated in 1710 (Vico, 2011: 17), rejects many of the early modern epistemological developments, particularly those of Descartes and Newton, whose impact it is difficult to underestimate for the history of Western social science. Descartes’s work played a central role in the foundational distinction between subject and object – intensifying the interiority of that subject and establishing that sensations in the mind are produced in the world ‘out there’, the implication being that the best method for attaining knowledge of that world was through inference (Descartes, 1993: 68). ⁴ Newton’s work, especially his Principia Mathematica of 1687 (1999), was no less than a scientific revolution and, for those interested in what was in the next century mostly referred to as civil society, established the view that the social world, like the physical world Newton described, must consist of orderly patterns and have laws that are discoverable by ontologically separate human intelligence. This ontological split embedded in Newton’s discoveries still resonates in the genetic make-up of the social sciences and is hotly debated today. ⁵

What is of interest for this paper is not Vico’s rejection of modern values – most commentators are sceptical of such a rejection (see Israel, 2006: 521, 523) – but rather his critique of its methods, or more specifically, the ontological principles underlying these methods. As a polymath of the early 18th century, Vico does not present his arguments as specifically ontological, but nevertheless key aspects of his work challenge the ontological assumptions of his day. In his New Science he seeks a science of humanity that attempts to bridge the gap between the mind and the world, rather than working from ontologically distinct first principles. In his section on principles, he writes that:

in the night of thick darkness enveloping the earliest antiquity, so remote from ourselves, there shines the eternal and never failing light of a truth beyond all question: that the world of civil society has certainly been made by men, and that its principles are therefore to be found within the modifications of our own human mind. (1968: 96)

Perhaps the simplest way to understand Vico’s role in the Other Social Science is to glance at the title of his magnum opus, The New Science, and imagine what a thinker writing in the years after Newton’s Principia – in an environment that was at least since Descartes thoroughly post-Scholastic – was up to. Surrounded as he was by the dramatic changes in philosophy and the physical sciences, and by extension the beginning of the social sciences, exactly what newness was he after? After all, what could be more new than the scientific discoveries of the 17th century? The answer is that he already saw his ideas as an alternative to the increasingly established methodological orthodoxy that sought to divide human thought from science. This ontological bifurcation essentially created a whole new set of epistemological problems and solutions that have since dominated philosophy, psychology and the social sciences in their orthodoxy for centuries (see Dewey and Bentley, 1949; Latour, 1993; Rorty, 2009). Vico’s conclusions may not measure up to contemporary standards, nor was he reflexively aware of the role that he would play or the movement to which he would belong, but he was setting out a course away from a Cartesian–Newtonian science of an ordered external world, inducing images in the human mind towards a science that does not draw ontological distinctions between the two.

If some of the first challenges to a burgeoning orthodoxy are apparent in Vico, the major fault line in the development of the social sciences, and sociology in particular, took place at the interstice between Durkheim and Tarde around the turn of the 20th century. Strangely, many social theorists are only vaguely aware of the debate between these two intellectual giants. Through a singular twist of historical circumstances, Durkheim’s eventual triumph essentially erased Tarde from memory and expunged his work from the canon. A quick glance at any sociology textbook or reader emphasizes the totality of Durkheim’s ‘win’. It is as if Tarde’s work is completely insignificant – an irrelevant irritation in the development of social science orthodoxy. This is certainly the case in the English-speaking world where some of Tarde’s work has only recently been translated. Although he is read in France and has some name recognition in Germany, for example, his work is mostly deployed by critical and cultural theorists (see, for instance, Blackman, 2008: 31ff.; Gibbs, 2008). But because of Tarde’s eradication from mainstream social science (at least in the English-speaking world), we tend to forget how much intellectual weight his ideas carried in their day. He was a formidable intellectual of sufficient standing to mount a challenge to Durkheim, his colleague at the Sorbonne. In other words, he was clearly not a lone voice in the intellectual wilderness; the Other Social Science was part of the intellectual climate of the time. It just so happens that that climate changed with the success of Durkheim’s essentially Cartesian–Newtonian thesis. ⁶ In a way, Durkheim helped to seal the end of the romantic era, closing a chapter that included aristocratic rebels (Byron ⁷ ), mysticism (Schopenhauer), duration (Bergson), and will (Nietzsche). Surely Durkheim was not solely responsible for this shift, but after him the burgeoning social sciences became sanitized, earnest, and proto-positivist. Readers in the social sciences may object to the listing of philosophers and poets in the intellectual soup from which the social sciences were cooked up. Orthodoxy is very clear on the founders of sociology, for example Comte, Spencer, and the triumvirate of Durkheim, Weber and Marx. However, if we understand the gravity of the Tarde–Durkheim debate – that is, understand what was at stake – we can see how Tarde is the representative of a very different kind of social science that stands in stark contrast to the ontologically dualistic or equivocal version of which Durkheim is in ways the exemplar.

What distinguishes Tarde from Durkheim is not a rejection of science but of the scientific principles espoused by the dualistic/equivocal ontology typical of Descartes and Newton, specifically the existence of substantive and bounded social objects or ‘facts’ external to human action (Durkheim, 2013: 37). Tarde is deeply interested in developing sociology as a science, but as a science of a kind different than what was becoming orthodox in his time. In his Social Laws: An Outline of Sociology (1899) he presents his own project wherein sociology should seek to develop for itself what he considers to be the three species of laws common to all sciences, namely repetition, opposition and adaptation. The present article-length piece precludes an analysis of this project, but we can note that what brings out Tarde’s rejection of the Durkheimian orthodoxy is his use of Leibniz in his Monadology and Sociology (2012). ⁸ Although the latter is a rather unknown work, only published in English in 2012, Tarde refers to it as his ‘metaphysical theory’ that links his three major works (1899: 211). At the outset it is important to mention that this slim book is not without flaws, which accounts perhaps for its obscurity. Tarde has a suspiciously high regard for the ideas of Spencer, his shrill reliance on scientific discoveries of the day makes the book seem dated and even puerile, and his breezy reading of Leibniz would probably not stand up to scrutiny from today’s scholarship. Nevertheless, for the social sciences in general and sociology in particular the book is revolutionary and, given the strength of social science orthodoxy and the paucity of Tarde scholarship, especially in English, the book is an example of the Other Social Science par excellence.

The title of Tarde’s work directly references Leibniz’s probably most well-known work, Monadology. In it Leibniz presents a parsimonious metaphysics wherein the universe consists of simple substances called monads that have no extension, form or divisibility (1989: 213 ff.). Although ontologically the same, monads differ from each other through their relations, or what Leibniz calls their perceptions, which does not necessarily imply consciousness. All monads perceive in that each has its own specific and relative view of the universe, though Leibniz does distinguish between simpler monads and those that have souls (humans – and angels), the latter endowed with more distinct perceptions and, crucially, memory (1989: 216). ⁹ Tarde uses Leibniz’s monads as a way out of the prevalent and near total fixation of the time on stasis, stability and essences in objects (2012: 10). He seeks instead a sociology of the infinite, a means of understanding the social world not in terms of fixed social entities (with particular characteristics) or facts governed by immutable laws, but rather a social science that is able to deal with infinitesimal differences between social phenomena at all levels. For Tarde, everything – physical objects, individuals, species – is reducible to the infinitesimal ¹⁰ in what today might be called scalar complexity (see Jessop, 2003: 15). No thing – object, individual, species, institution – is static, and that they appear to be so is merely a result of our learned prejudices towards the finite and temporary equilibria, for example, between a species and its environment. One of the consequences of this brand of monism is that everything is a society and every phenomenon a social fact (2012: 28). But if everything begins and ends in the infinite, Tarde argues, we need a social science of the infinite, more like differential calculus and less like those physical sciences that seek finite quantities (2012: 11). Ontologically what he proposes here is a monism that embraces the heterogeneous character of matter and mind and hence does not distinguish between the two (2012: 15).

What this amounts to is a social science of difference. Although identity, he maintains, is usually the point of departure – and for our purposes surely the point of departure for orthodox social science – it is only difference that can account for the infinitesimal in society. He writes:

The example of society is eminently well-suited to promote an awareness of this central fact […] by showing that in this series where identity and difference, the indistinct and the well-characterized each reciprocally make use of the other over and over again, the initial and final term is always difference, the characteristic, the bizarre and inexplicable agitation at the basis of all things, which reappears more clearly and sharply after each successive effacement. (2012: 41)

Years in the wilderness: Philosophy

Having been stymied by Durkheim, the Other Social Science virtually bows out of the social sciences altogether to take refuge amongst philosophers of the infinite and difference, Alfred Whitehead and Gilles Deleuze. Whitehead’s major work, Process and Reality from 1929, is famously dense and treacherous for philosophers; doubly so for social scientists. ¹² Nevertheless, we can draw out a few crucial insights that are relevant to the Other Social Science. The first is that he boldly argues that philosophy has it all wrong, that Western philosophers since Plato ‘were perplexed by the inconsistent presuppositions underlying their inherited modes of expression’ (1978: ix). He points to the centrality of substance–predicate and subject (mind)–predicate, further instilled in Western thought with Descartes’s substance–quality. Thus:

the doctrine of the individual independence of real facts is derived from the notion that the subject–predicate form of statement conveys a truth which is metaphysically ultimate. According to this view, an individual substance with its predicates constitutes the ultimate type of actuality. (1978: 137)

Deleuze was self-admittedly interested in the (dry, dusty) history of philosophy, and yet his work informs the Other Social Science a great deal. In what he himself sees as his central work, Difference and Repetition, he objects to representation in philosophy – the idea that difference must be subordinated to identity within the concept. Like Tarde, Deleuze is interested in repetition amongst elements, arguing that true repetition is ‘difference without a concept’ (2004a: 26). This means that difference does not inhere between a repetition of objects represented by the same concept but is rather without a concept, literally the difference between differences. In Deleuze difference is a dynamic process of movement – it creates dynamic space and time (actualizations) which corresponds to immanent, pre-singular relations, a repetition of difference which accounts for eternal emergence and newness. In a way difference (or differences differing) is what pushes the universe forward; it is that which ushers in the new. Clearly this stands in sharp contrast to social sciences based on discreet external objects, sui generis social facts, or discoverable (objective) social laws.

What makes Deleuze interesting for the Other Social Science is his univocal ontology of the virtual-actual. In Tarde, and to a certain extent in Whitehead, it is difficult to clearly see the dynamics of the infinitesimal and difference. How precisely do we talk about the relationship between substance and flux, or being and becoming? Deleuze explores this thoroughly, and his solution effectively divides reality in two: the virtual, characterized by intensive qualities, immanence and duration; and the actual, characterized by extensive quantity, external relation and chronological time. Real things (actual entities – individuals, species, groups, ideas) are as if stretched on this continuum between the virtual and the actual, on the one hand existing in infinite, qualitative relations with the entire world (flux or becoming), and on the other hand caught in finite, quantitative relations with specific parts of the world (substance or being). It is important to bear in mind that Deleuze is not talking about two separate realms of reality. Both the virtual and the actual are ontologically equivalent. It is as if everything is split into two halves (Deleuze, 2004a: 261). In his under-appreciated book on Leibniz, The Fold: Leibniz and the Baroque, Deleuze, like Tarde, seeks to understand the relation between elements as an essential envelopedness – that is, at once enveloped and enveloping (see Deleuze, 2004a: 314–17) – forming in effect the point of view of the monad. But unlike Tarde he sees the fold rather than the point as the best description of the infinitesimal (2006a: 6). Thus the world is an infinite series of folds within folds. The how of the folding is the virtual nature of elements in the world, the unfolding their actualizations. Because of the focus on forces and materiel, Deleuze’s work helped inform research programmes that are collectively known as the new materialism. ¹³

We can see what an ontology of folds and infinite envelopedness means practically for the social sciences by considering an illustrative example provided by Deleuze and Félix Guattari in A Thousand Plateaus (2002: 352–3). Crucial here is the notion of illustrativeness. Nowhere do Deleuze and Guattari claim that this shows the difference between the virtual and the actual. But for social scientists the following outlines the contours of what a science of pure relation might look like. The example concerns two board games, chess and Go. ¹⁴ The former is defined by structural rules governing distinct and finite pieces (a player only has so many pawns, knights and so on) that have inherent characteristics or capabilities. Indeed, they have these characteristics abstractly, even when they are not in play. The strategy of chess is linear – to capture the king. Moreover, the relatively small board is a bounded, finite surface. In Go the stones are functionally the same – they have no inherent characteristics. The character that they take during play is derived solely from their relations to other stones. When they are not at play, sitting in their little jars, they really have no characteristics at all. The strategy is fluid and highly intuitive: players must develop relations amongst stones in such a way as to block and surround the opponent or flee from attack – to create space. And while a Go board is also technically finite, the relational nature of the stones makes the entire playing space much more relative and thus infinite, in a sense. Now Go stones are not exactly folds, as Deleuze would have the smallest point be (though their relations come closer), but understanding how Go works and how it is different from chess is a rough approximation of what a social science of difference, the infinitesimal and the relational, might look like. Of course few orthodox social scientists argue that there is a one-to-one relation between social facts and statements about them, but through baroque approximations many feel justified in presuming that their finite, linear and fixed representations are close enough to underwrite objective statements and then laws. In other words it is not a matter of perfect captures of objective data, but rough takes that allow for productive statements. A more Go-inspired Other Social Science would disavow any claims to objective approximations, focusing rather on relational, often chaotic, aspects that in fact constitute fluid phenomena.

Recent renaissance

Throughout the first part of the 20th century, by and large there were no social scientific approaches that explicitly challenged the prevailing Durkheimian orthodoxy of subject–object and observational method on a purely ontological basis. One of the first instances where the Other Social Science formally shows itself in the literature is in chaos theory, related to, and sometimes used interchangeably with, complexity theory. ¹⁵ The main argument of complexity studies states that if in the physical sciences we have disavowed flawed models of finite relations and linear change, why do we persist in using those flawed models in the social sciences? What most social scientists take from complexity theory is its resistance towards linearity and determinism, but for the purposes of this paper the ontological and methodological implications are more significant. What complexity really demands is a social science of the infinite, relations and emergence wherein there is no ontological difference between element and system.

The text that introduced chaos and complexity to a wider audience and inspired social scientists was Ilya Prigogine and Isabelle Stenger’s Order Out of Chaos, originally published in French in 1979. The book is concerned with the physical sciences primarily (à la Mandelbrot), but they draw on a broad literature to make their arguments, freely quoting philosophers like Bergson, and perhaps more significantly for the purposes of this paper, Whitehead. Central to their critique is what they refer to as the image of nature. They argue that there have been quantitative changes as science has advanced (for example, observing the very big and very small), but that there are also qualitative changes which extend beyond what we consider to be the physical sciences that alter our very image of nature. Prigogine and Stengers point to how the founders of Western science were focused on understanding the universe and its eternal natural laws that corresponded to their own idea of rationality. They sought immutable laws whereby everything can be logically or causally interconnected. And yet over the centuries little evidence of any grand unifying theory along these lines has been found. In a tone very reminiscent of Tarde’s Monadology and Sociology they write: ‘Wherever we look we find evolution, diversification, and instabilities. Curiously, this is true on all levels, in the field of elementary particles, in biology, and in astrophysics, with the expanding universe and the formation of black holes’ (1984: 1–2). They make the significant point that in the physical sciences, classical, reversible, deterministic phenomena only really occur in closed, experimental conditions, in effect by blocking the chaos and flux of the world outside. The fact that this model is applied to orthodox social science – that is, blocking out the fundamentally chaotic nature of the world – is for them only more troubling (1984: 9).

The ontological implications of this for the social sciences are brought to the fore in David Byrne’s classic 1998 introductory text, Complexity Theory and the Social Sciences, where he asserts the foundationalist aspect of complexity theory, but without it being reductionist or positivist. In other words, it rejects epistemological critiques of social science orthodoxy: Byrne is saying that we can have our cake and eat it too. We can have what this essay calls an Other Social Science, one that ‘provides a way of relating the macro and the micro which is not inherently aggregative and reductionist and […] provides a way of describing the relationship between agency and structure’ (1998: 35). Again, what is crucial here is that he presents his project as realist; that he is making substantive claims about the status of reality (this is ontology, as he rightly points out). Byrne argues that such an approach is perfectly consistent with the ‘autonomous logic of sociological theory’ in the sense that it favours the emergent and non-reducible nature of the social (1998: 47). Thus it qualifies as an important part of the fabric of the Other Social Science in its appeal to a monism where the physical and the social are ontologically equivalent.

To break the historical thread slightly here, Niklas Luhmann presents a very technical though perhaps somewhat sterile ontology with his systems theory. In a number of respects it shares a great many of the premises of complexity theory, but its critique of social science orthodoxy remains rather difficult to deploy, not least because it effectively severs his social science from many of the humanistic concerns on which others focus, relegating the mental life of humans to psychic systems (1995: 255). Many are uncomfortable with this. But if Luhmann lacks in human connection he makes up for it in virtuosity, painstakingly describing society as (in non-Luhmannian jargon) a complex network of relations tout court, with no theoretical left-overs sticking out the edges. Luhmann is included as part of the Other Social Science for his important work exploring the notion of the system itself at its logical and methodological limits, thereby delineating their promise (and dangers) when developed in the social sciences. To be more precise, his ontological formulation must be considered univocal or ‘flat’. Systems are autopoietic: nothing is privileged or transcendent; nothing comes from without. Communication is simply an ongoing relation of elements rather than an ontic mode – a ‘self-referential process’ (Luhmann, 1995: 143). Thus, his relegation of the human to psychological systems and his penchant for cybernetics is suggestive as to what a truly anti- or even post-humanistic social science would look like. In any case he follows the line of a material, non-human aspect similar to Latour and Deleuze, breaking down the differences between the human and the non-human, again establishing a univocal ontology. Luhmann of course rejects injecting any kind of agent-centred or progressive social posture back into his work, for which he is criticized (see King and Thornhill, 2003: 204). Clearly for most social scientists – and most of the authors here including Whitehead, Deleuze and Latour – such human-centred problems are of the highest importance and should be the goal of productive thinking. But what the Other Social Science offers is a way of talking about human concerns without a basic, culturally-induced humanism forming the foundation of thought. Following Luhmann’s impulse to treat human concerns not as ontologically distinct concepts but as effects of systems is important because it shows us what such a non-human foundation might look like.

Ultimately, complexity studies never found a comfortable ground between the suggestive and the technical. Many proponents (for example Urry, 2005) deferred sketching out what an actual research paradigm would look like, and attempts at the deployment of complexity theory in social science research were often caught up with the allure of computer modelling, which to this day, despite Deepmind’s recent advances (see for example Strogatz, 2019), requires a logic based on points and therefore on units – in effect discreet substances. Thus the models, for all their vastness, were more studies about complicatedness rather than complexity, for true complexity must jettison the notion of units and move to pure relation (as in Go) or Deleuze’s folds, where researchers track not only how units respond to chaotic interaction, but how the qualitative changes derived from the complex system transform the ‘units’ altogether. In this sense Luhmann’s systems theory, with its emphasis on the relations and connection between at least inert ‘carriers’ of code, is preferable. In any case, complexity theory presents a robust way of showing how the world is irreducible to fixed entities with properties in a deterministic, linear development; that we need a way of understanding flux and infinite relations.

Bruno Latour, probably the most well-known proponent of actor-network theory (ANT), neatly provides a concise definition of what this article has been calling 20th-century orthodoxy or ‘mainstream’ social science. He identifies it as simply the sociology of the social, by which he means one that posits the rule to explain the instance or activity, contrasting it with what he calls a sociology of association:

[F]or sociologists of the social, the rule is order while decay, change, or creation are the exceptions. For the sociologists of associations, the rule is performance and what has to be explained, the troubling exceptions, are any type of stability over the long term and on a larger scale. It is as if, in the two schools, background and foreground were reversed. (2005: 35)

A novel intervention that illuminates the typical tensions between social science orthodoxy and the Other Social Science is Alexander Wendt’s work on panpsychism in Quantum Mind and Social Science: Unifying Physical and Social Ontology (2015). The title itself, of course, is suggestive, signalling not only social science’s jettisoning of the anchor of finite, linear science but an overthrowing of the idea – one sprung from the persistent failure to find laws governing society similar to Newton’s laws of motion – that the social world is somehow qualitatively different from the physical world and so other rules must forcibly apply. When looked at in this light it is a wonder that no one had attempted this earlier, for, as has been already mentioned, it has been clear for more than a century that the physical world of Newton is much less discreet, eternal and linear than anyone imagined (for social science orthodoxy our concepts of mental life, on the other hand, seem not to have progressed much past Descartes’s autonomous, singular and transcendent res cogito ¹⁹ ).

In the context of the present paper what is impressive about Wendt’s text is that he is clearly going about these questions in a different way, drawing on twentieth century physics and the conceptual framework of philosophy of mind. For social scientist interested in the questions with which this paper deals (ontology, orthodoxy), what is most striking is Wendt’s complete avoidance of the historical arc described in this paper. His project has nothing to do with Tarde or Whitehead, Deleuze or complexity. ²⁰ What it does not ignore – and what makes it particularly useful for the purposes of this paper – is the question of ontology. Wendt is steadfast in his determination to develop his grand unifying theory along ontological lines. Like Byrne above, Wendt is interested in the ontic furnishings of the universe. But rather than start with the big problem of the infinite and the relational (which Tarde observed all around him), Wendt starts with the (very Cartesian) problem of mind–body duality and the (philosophy of mind) problem of consciousness: How can we reconcile the quantum nature of the universe with our apparent consciousness and will? His answer is that the human mind shares the same quantum nature as the rest of the universe, that it is both particle and wave, depending on how it is observed. The book, although massive in scope, is troubling for its omissions, and one cannot help but feel that Wendt is solving problems that perhaps did not need solving (there is a strong sense in which Whitehead, for example, ‘solved’ these problems over a century ago). Put another way, Wendt does not take the step to treat the problems associated with Western science (Descartes, Newton as described here) as historically and geographically contingent (Rorty, 2009: 45 ff.), tending to see them as quasi-psychological facts of nature. But, on the other hand, the book is an exemplary crash-course on quantum physics for social scientists and offers a sort of bridge between quantum physics/philosophy of mind and the Other Social Science that this paper seeks to illuminate.

An approach that stems from the sociology proper of recent decades is the loosely defined area of relational sociology, which for now is more of an internal debate in sociology rather than a movement in the social sciences. Nevertheless, the terrain on which it sets itself is relevant to the Other Social Science. In the simplest terms, relational sociology is an attempt to side-step the problems associated with structure–agency dualism by focusing on the relations and interactions amongst elements rather than the units or, on the other hand (and perhaps especially) the structures. These relations – to various degrees – furnish the social and for certain authors locate the agents and define the contexts in which these agents derive meaning. It is important to highlight that in some renditions of relational sociology there is a latent dualism, which, that while emphasizing relations as primary, is at the same time unwilling to surrender the fundamental centrality of human actors. We can see this most clearly in the writings of Archer and Donati. Although Archer writes that ‘[r]elational sociology has a deeper social ontology because the very existence of the social order is itself relational’ (2010: 201), a position that would seem to point to a ontological monism as described in this paper, in general she is committed to a social ontology that upholds a modern sense of agency (Donati, 2006), in other words, that subjects are able to intervene in the flow of events to make them as they otherwise would not be, a position consistent with her commitment to critical realism (Archer et al., 1998).

This is not the place to consider what a social science without modern agents would look like, but for the purposes here it is important to note that relational sociologists such as Archer and Donati can only be tangentially seen as part of the Other Social Science, in so far as they focus on relation or what others (Whitehead, complexity theorists) call process. Likewise Crossley’s research (2010) focuses on relations and interaction, but he does not clarify the relative ontological status of individuals and wholes. An approach more aligned with the Other Social Science would be Emirbayer’s classic 1997 paper which rejects the whole notion of pre-given units (1997: 287), a position that has a great deal in common ontologically with the univocal approaches described throughout this paper. Rather than the typical point of intervention in relational sociology of agency and structure, he casts the fundamental dilemma in sociology as ‘whether to conceive of the social world as consisting primarily in substances or in processes, in static “things” or in dynamic, unfolding relations’ (1997: 281). From the perspective of the Other Social Science it comes down to two fundamental versions of relational sociology. Many, it seems, want to hold on to a dualist position that does take as a methodological focus the relations between elements, but nevertheless posits ontologically distinct agents. ²¹ A ‘harder’ relational sociology would belong more to the Other Social Science and assert that what we call individual selves (subjects, agents) are in fact the effects of emergent, processual social relations which in themselves constitute the social field.

Conclusions

If this article has succeeded in some way in weaving together the strands of a social science that presents a viable alternative to 20th-century orthodoxy, we should now trace this science’s contours. What kind of social science are we talking about and what are the practical implications for actually doing this science? At its foundation the impetus for identifying this broad arc of thought over several centuries rests on an ontological question. That is to say that there are implications to how we understand how things be that determine the kinds of problems social scientists are interested in addressing and subsequently how they go about addressing them. As explored in this article, the fundamental univocal monism of the Other Social Science sees no difference in how all things (individuals, institutions, thoughts, objects, relations, etc.) are. Thus this view supports no assumptions about the fundamental furnishings of the social world or specific, dualistic methodological points of departure (structure–agent, subject–object, mind–body). For example, one might begin research with an interest in the effects of such-and-such a social phenomenon on individuals, but only with the full recognition that to ‘stop’ the eye of research on either is to ignore their enveloping/envelopment in each other – a methodological move analogous to how the physical sciences shut out chaos through controlled experiments. Thus one may not assume the central relevance of any level of analysis or starting point. This certainly pushes back on the current sensibility towards ‘rescuing’ the individual in social science research, or the tendency to reify macro flows into structures. As we saw with Tarde, and this is brought out in Deleuze to a great extent (1995: 86), we must ‘start in the middle’ of any given series and work our way along. This approach can be inscribed by various positions: immanence, monism, pure relation or simply a ‘flat’ ontology; they all make the same basic ontological argument. This brings us back to the notion of ‘mainstream’ social science with which the paper began, which simply means the opposite of this ontological stance, namely transcendence, dualism, substance, subject–object. All of these posit by one means or another that things are in different ways.

It likely may not be immediately clear which approaches belong to the Other Social Science. Most social scientists do not set out with flawless ontological principles in mind – probably none do. Moreover, adhering strictly to such ontological commitments may not be plausible due to study design or the cultural assumptions of researchers who have most likely trained in an environment (or indeed a society) that, for example, tacitly accepts Cartesian notions of the subject or Newtonian notions of society. Or perhaps the human mind may ultimately be incapable of fully grasping the infinite or virtual. This brings us back to Tarde’s infelicitous notion of ‘great individuals.’” Andrew Barry argues that Tarde was a Spencerian elitist who believed that social superiors create, inferiors imitate (2010: 188). Deleuze takes a much more sympathetic view, ²² for which it must be said there is evidence. Clearly Tarde did not have a consistent view on this; only future research will solve this question. Likewise with the analysis of relational sociology above. There are considerable differences – crucial ontological differences – between Archer’s agent-centred relations and Emirbayer’s pure relation. Despite these clear and important differences, however, what is important in the foregoing analysis is the tendency towards the purely relational: the infinite, the immanent, a univocal ontology that rejects transcendent elements such as reified sui generis structures, autonomous agents, and essentialist objects.

Using this tendency towards the univocal or pure relation as a criterion answers many questions about who is to be included in the Other Social Science. There are many thinkers who appear to be part of the club in their rejection, inter alia, of mainstream positivism. Goffman’s (1971) inductive approach, for example, with its focus on localized, situational emergence, seems to suggest a kind of relationalism, yet without definitive ontological statements it is difficult to tell. Surely Goffman is not endorsing the kind of Durkheimian dualism of social facts and individuals as presented earlier in the paper. However, for Goffman there are implicit ontological cleavages between the self and other as well as between the social and the physical, the latter of which is not taken up in his matrix of causality. In any case Goffman’s focus on agents would seem to exclude him from the Other Social Science: even if he insists on a performative subject, the latter still has interiority, and is thus ontologically different. As Butler (1988: 528) points out, Goffman simply posits the self. This likewise applies to approaches in the social sciences which focus on sense-making or meaning amongst actors. Social network analysis (White, 2008) and ethnomethodology (Garfinkel, 1994) thus can rightly be seen as a rejection of positivism, and yet their reliance on the interiority of agents implies an equivocal dualism that is at odds with the ontological tenets of the Other Social Science. Weber is similarly interested in the social action of human individuals (1968: 88) which implies actorhood; he inevitably falls back on personal motivations (see Guy, 2020: 95–6), which is inconsistent with the ontological criteria presented in this paper such as immanence, process, or difference. In short, if it seems that there are commitments to a univocal monism in these authors, it is probably more apparent than real. Meaning- and action-centred approaches simply retain subjects and external social facts as ontologically distinct, ²³ and so in the absence of explicit ontological claims to the contrary, cannot be seen to be part of the Other Social Science. This is not to say that the Other Social Science is the only or preferable alternative to mainstream orthodoxy; the purpose of the paper is merely to illuminate connections that remain obscured without the ontological approach offered here.

A similar analysis must be applied to approaches based in phenomenology, which cannot be properly said to belong to the Other Social Science. Phenomenology at its foundation posits an ontological extra – whether in the form of consciousness (Husserl) or Dasein (Heidegger) – and is thus dualist or equivocal. In a way, the Other Social Science can be seen as a distancing from or a solution to the intentionality inherent in phenomenology (see Agamben, 1999: 225; Badiou, 1997: 21) and its largely French movement of the middle of the 20th century. Thus on this ontological ground Emmanuel Levinas (1996) cannot be seen to have a strong tendency towards the Other Social Science. Likewise, although thinkers such as Cornelius Castoriadis sometimes appear to have such tendencies, they must be put in the phenomenological column (see Adams, 2011: 2) and thus not in the Other Social Science proper.

Of course there are many other ways of differentiating movements within the social sciences. The ontological approach presented in this paper is only one way of doing this, and by no means the most important way. But this fundamental ontological tendency has clear ramifications for the social sciences, some of which this paper has touched on tangentially. Clearly the autonomous and originary nature of human subjects is pushed to the background or, more radically, eliminated altogether. This has clear implications for notions of free will and determinism. It must be noted that complexity studies have made considerable progress in coming to terms with this apparent dichotomy, and Deleuze (2004b: 63) overcomes it with notion of the Event – a dynamic differentiation operating at the edge of chaos. Another implication of the Other Social Science is the emphasis on becoming over being (or process over substance). This means that whatever it studies, social science must view variables or elements as inherently mobile, meaning that they cannot be apprehended from a fixed position or a ground; it must seek becoming in itself and not in relation to some other fixed thing or point of view. And thus every additional variable we consider introduces only another order of mobility, leaving us to deal (somehow) with mobility squared, cubed, and so on. This explains why many of the authors in this paper are interested in the infinite and differentials. Using the discussion of chess and Go from above, the Other Social Science clearly entails new concepts and methodologies: the 20th century delivered an orthodox social science of chess (things, individuals, laws, strategy); the Other Social Science suggests the development of the social science of Go (difference, emergence, relations, intuition). ²⁴ Such a science can illuminate and hopefully simplify (in terms of their formation, not necessarily in terms of answers) the most pressing fundamental questions facing social science today. Again, this entails moving away from the emphasis on enduring things, individuals and structures to a study of (infinite) mobile relations. In other words we are no longer going to be tackling problems with the strategy of maintaining the stability or integrity of ‘fixed’ entities (individuals, societies) and institutions (organizations, nation-states), but rather by considering vectors and tangents of change and transformation, intimately related to other equally mobile elements in evolving systems of relations. This implies that the Other Social Science is a micro sociology. Tarde, Deleuze, and perhaps most significantly Latour are specific on this point.

Finally, the material and the non-human must forcibly play a role in social scientific analysis. In simple terms this would include all aspects of material – architecture, the built environment, climactic systems, geography, technical artefacts – in a mutually shaping system (what Deleuze and Guattari call assemblages) of the material and the social. Ultimately the goal of analysing the Other Social Science is not merely to untangle abstract theory but to use or deploy it. This deployment will not involve defining, classifying, searching for social structures or measuring human reactions, but mapping assemblages – understanding but also explaining how social machines work. Many of the fields discussed above have rigorous empirical research programmes. These include the considerable number of ANT studies, the burgeoning and broader field of applied relational sociology (see Dépelteau, 2013), and assemblage theory analysis (see Legg, 2009; Lenco, 2014; Youdell and McGimpsey, 2015).

The limitations of this paper are many. It never directly addresses structure and agency. As mentioned above, this is an important implication of the Other Social Science, although there is not nearly enough room for a discussion of it here. Suffice to say that for a truly univocal ontology the ‘illusion of agency’ is merely the result of the isolation or a stop in the density of infinite folds of which the social world consists. This is not necessarily to relegate the importance of individuals (though this would perhaps be logical and productive), but is rather to render them, variously, as the elements of a subsystem of the social whole (Luhmann), points of monadic perception and memory (Deleuze), societies in themselves (Tarde and Whitehead), or nodes in complex networks (complexity theory). From the perspective of the Other Social Science, the problem of structure–agency comes from a (most certainly historically and geographically contingent) fixation on agents as autonomous subjects. What all the perspectives dealt with in this paper have in common is that they more or less do away with the notion of autonomous, self-same, and especially transcendent subjects altogether. This is not to say there is no such thing as individual humans or that it is impossible to imagine the contemporary world without the transcendent subject. It merely necessitates rethinking many aspects of contemporary society (see Hirst and Woolley, 1982: 131 ff.).

Readers may feel that the historical scope of the analysis here is unwarranted, but looking back now we can see with greater clarity how thinkers like Vico, Whitehead and Latour, for example, share ontological commonality. Vico prefigures Deleuze in reducing the ontological difference between humans and human society, and by starting in the middle: focusing on the things in themselves rather than a reductionist something called civil society or objective social structures (Newton via Durkheim), or the centrality of the mind (Descartes via Locke: the res cogito, the I, the ego). Likewise each thinker overviewed here is an integral part of a long-term movement, though there are clearly many omissions in the Other Social Science as presented here. Spinoza’s development of the notion of immanence is certainly significant. Bergson’s duration is a key step in the development of a univocal ontology. And surely Nietzsche, especially in his influence on Deleuze and Foucault, must be seen as a part of the Other Social Science. In the sense argued above these authors must be built into this historic arc as well. ²⁵

Staying at the philosophical level, this univocal monism of the Other Social Science can be understood in contrast to the rational monism of Hegel. In one sense Hegel can be read as a counter to – and in many ways predicted – the neo-Kantianism that would permeate sociology (Rose, 2009), an argument analogous to the one in the present paper. However, the Other Social Science differs in one crucial respect, which goes a long way to explaining why Hegel is not brought up in the literature examined here, and in fact is only critiqued or excluded (see for example Deleuze and Guattari, 1994: 12; Whitehead, 1978: 210): In Hegel phenomena emerge in their relation to absolute Idea – in the Other Social Science phenomena are actualised through a pure emergence of difference, of relation. As Elias notes, Spirit for Hegel becomes the order of human society (1978: 444), and any such an ordering principle must be ontologically prior and thus distinct, or in other words equivocal. Hegel presents a monism of a unifying principle, not the univocal monism of the Other Social Science as presented here. Thus in The Science of Logic we are told that the absolute Idea is the only reality, which seems to imply the parts are somehow less real than this whole: ‘First of all, this advance [from content to content] is determined as beginning from simple determinatenesses, the succeeding ones becoming ever richer and more concrete. For the result contains its beginning and its course has enriched it by a fresh determinateness’ (2010: 840). Hegel may profess to monism, but his is an equivocal ontology. Hegel presents difference towards rational unity. What unites the authors and perspectives in this paper is difference within substantive unity. To put it another way, Hegel’s philosophy has an arrow of purpose which is clearly not the case in the Other Social Science. In a like manner Marx’s dialectical (though material) monism (Smith, 2009) differs from the univocal monism presented in this paper. Marx, of course, must be read as an Enlightenment thinker par excellence, and his legacy to orthodoxy is found in his objectivity and Hegelian notion of linear progress, as exemplified in his commitment to historical materialism. On the other hand his interest in the objective (gegenständlich) nature of social relations does point to a focus on genuine materiality and thus univocity. A more detailed discussion of this slight nod to the Other Social Science is, however, beyond the scope of this paper.

The Other Social Science is not a ‘turn’ or a graduate student fad. It has engaged with its counterpart as long as there has been social science. This is important, not least because it de-legitimizes the social science orthodoxy that has dominated the 20th century and dominates still. The task for future research will be to shore up common aspects to form an alternative to orthodoxy. As a corollary, the Other Social Science should be normalized as a social science. It is not some sort of sporadic, ad hoc critique, nor is it an impenetrable, jargon-laden distraction; rather it is a connection of intellectual and philosophical commitments with a distinguished pedigree that is, most of all, scientific. This means, at the very least, an approach that focuses on empirical questions, has clear methodological principles, an adherence to logic, is comprehensive yet strives for parsimony, and has a developed notion of the thinker and what thinking is. It is empiricist without being positivist, post-structuralist but materialist. The status of such a science has been clouded by the positivist versus anti-explanatory debate, which distracts from the presentation of the Other Social Science as one more-or-less coherent, three-centuries long movement in the social sciences committed to certain ontological principles. This movement can serve as a map and a resource for researchers to cross-pollinate and sharpen their ontological arguments in order to bolster their research agendas.

In using the lens of univocal ontology, this paper follows an admittedly Deleuzian sort of analysis, but I think that this can be justified and ultimately useful. Deleuze is a consummate and very detailed ontologist and a perspective such as his can illustrate how although all the strands and movements over-viewed in this paper are not the same, and perhaps sometimes not even similar, they have the same tendency towards univocal monism with a corresponding focus on difference, the virtual, the infinite, relation over substance, etc. Indeed, Deleuze is making inroads of applicability, for some time in political studies but increasingly in sociology (see for example Keller, 2019; Tonkonoff, 2017), and his work suggests many productive lines of research. For example, his notion of the fold can say something about the nature of relations themselves. Points, even infinitesimal ones, connote elements whereas folds could be well understood as relations (how society unfolds). Likewise Leibniz is one of the usual suspects throughout much of the research presented above. Tarde (2012) evokes him considerably in his most ontological work, and Deleuze (2006a) also looks to Leibniz in an effort to formulate a science of the infinite. As a counterpoint to Wendt’s line of research, a deeper appreciation of Leibniz’s work for the social sciences seems in order. This paper has tried to map out the historical arc of the Other Social Science, but also to sketch out an arrow of direction for the future. It sees the last century or so of social science research as an ontologically dualistic diversion from a more monistic or ontologically univocal approach which the social sciences seem slowly to be embracing. Surely, the great research agendas of the 20th century are not likely to fade overnight, but by identifying the Other Social Science we can see a very real shift, and if we can appreciate the commonality of approaches, this can only bolster ’heterodox’ research agendas in the future.