![]()
1 Introduction
Why social ontology?
Practicing social scientists often think that social ontology is not relevant to their daily work aiming at providing explanatory understanding of social phenomena through empirical research. Many of them go even further and consider ontological issues as a serious distraction. This is because ontological discussions easily draw attention away from the methodological questions regarding the uses and limits of specific methods of data acquiring and analysis which pertain to their daily work as empirical researchers. In addition, there is a risk that engagement to highly abstract ontological debates would shift the focus of their research away from the contents of specific models and theories that aim to explain empirical phenomena in their field of study. Many philosophers of social science, too, tend to consider social ontology as somewhat marginal to their enterprise. They accordingly contend that the focus in their field should be on epistemological and methodological questions rather than on social ontological issues. One reason for this view is that ontological debates in philosophy have not only been typically far removed from the empirical research and its methodology but also have a notorious reputation of being inconclusive.
Though I think that these views contain some truth, my contention nevertheless is that we cannot escape ontological issues neither in the explanatory social research nor in the philosophy of the social sciences. This is because specific research practices in social sciences as well as the theories and methods used in these practices always contain ontological assumptions and presuppositions no matter whether the practising social scientists and philosophers of social sciences acknowledge or discuss them. These assumptions and presuppositions concern, for example, the basic ontological categories of which the entities studied belong; the relationships between different kinds of entities studied; between them and those studied in the other social sciences and non-social sciences; and the causal structure of the social world (or the lack of such structure). In addition, ontological assumptions and presuppositions of this kind are not inconsequential in empirical research. Rather, they affect what are considered as proper social phenomena to be explained; what methods are thought to be suitable for studying different types of social phenomena; what are regarded as the sound explanations of these phenomena; and what are considered as possible factors in those explanations. Differences in opinion as to how to answer questions like these are reflected, for example, in the debates between the proponents of various forms of individualism (or microfoundationalism) and collectivism (or holism); and between the advocates of statistical causal modelling, the mechanism-based model of explanation and interpretative methods.
In recent discussions on the philosophy of social science, the importance of ontological issues for explanatory social research has been emphasized, especially by critical realists. The founding father of this movement, Roy Bhaskar (1979, 31), writes for example that ‘it is the nature of objects that determines their cognitive possibilities for us’ rather than the other way round. Another prominent critical realist, Margaret Archer (1995, 17), in turn states that:
the social ontology endorsed does play a powerful regulatory role vis-à-vis the explanatory methodology for the basic reason that it conceptualizes social reality in certain terms, thus identifying what there is to be explained and also ruling out explanations in terms of entities and properties which are deemed non-existent. Conversely, regulation is mutual, for what is held to exist cannot remain immune from what is really, actually or factually found to be the case. Such consistency is a general requirement and it usually requires continuous two-way adjustment between ontology and methodology to achieve and sustain it as such.
I accept Archer’s characterization of the proper relationship between the methodology of social research and the social ontology. Still, as we shall see, I will question the transcendental arguments for the original critical realist social ontology1 as well as some specific tenets of this ontology. The critiques I go on to develop pertain to the issue of ontological naturalism.
Why naturalize critical realist social ontology?
The core argument of this book is that critical realist social ontology should be naturalized. In order to specify what this means, we need to take a look at the position of naturalism in social ontology. In very rough terms, naturalists contend that theories in social ontology should be built by studying (1) the ontological assumptions and presuppositions of the epistemically successful practices of empirical social research (including well-confirmed theories produced in them); and (2) the well-established ontological assumptions advanced in other sciences, including natural sciences. This procedure is needed because naturalists hold that ontological theories cannot be justified by means of philosophical arguments that rely on a priori forms of conceptual analysis and reasoning.2 The epistemic status of the naturalist theories in social ontology is thus regarded exactly the same as the theories produced in empirical social research, even though the level of conceptual abstraction of the former is higher than in the latter. This is not meant to deny, however, that the empirical evaluation of abstract ontological theories is prone to be more indirect than that of less abstract theories and models developed in the context of empirical research. Furthermore, theories in the naturalist social ontology are expected to be compatible with the well-established assumptions and presuppositions of natural sciences since naturalists conceive the social world as being a part of nature.
The position of ontological naturalism should be distinguished from the naturalist position in the methodology of social research. Bhaskar (1979, 3) defines the latter as a thesis that ‘there is (or can be) an essential unity of method between the natural and the social sciences.’ This means that the issue of naturalism in the context of social ontology concerns mostly the relationship between theories in philosophical social ontology and practices of empirical research (including the models and theories produced in these practices), while the question of naturalism in the methodology of social sciences is about the relationship between the specific methods of social and natural sciences. Nevertheless, as indicated above, a naturalist social ontology also contains an assumption that social phenomena form a part of nature. This assumption, in turn, appears to support the position of methodological naturalism since a denial of the ontological divide between human societies and nature (conceived in the broad sense of the term) appears to imply that there could not be a major difference in the methods used in natural and social sciences either. As will be indicated in detail below, the issue of methodological naturalism is nevertheless more complex than this since closer examination of different sciences has revealed that specific methods used in them are remarkably variable, not only between the natural and the social sciences but also between different disciplines within both of these broad categories of sciences. Hence, while naturalist positions in the social ontology and methodology should not be conflated, they are not totally unrelated to each other. The focus in this book is on naturalism in social ontology although I will discuss, too, on some issues related to methodological naturalism.
Now, one may wonder why I begin with a critique of critical realist social ontology if the ultimate aim is to develop a naturalist social ontology that is relevant to the actual research practices in social sciences. My answer is that the original critical realist social ontology is a promising starting point for this project since it not only recognizes the importance of ontological assumptions for empirical research in social sciences but also contains some important naturalist ontological ideas. The latter include a conception that (in the context of the philosophy of natural science) scientific practices form a necessary starting point of ontological argumentation as well as the ontological doctrines of causal realism and synchronic emergent powers materialism. I especially regard the position of causal realism based on the concept of emergent causal power to be an essential element of a plausible naturalist social ontology. In addition, I think that a naturalist social ontology should be combined with some form of scientific realism, and critical realism is perhaps the most discussed scientific realist position in the current social sciences (I will defend a naturalized version of scientific realism in the next chapter). For these reasons, the original critical realist social ontology forms a suitable starting point for my study even though it is not the only possibility.
In addition to these naturalist elements, the original critical realist ontology nevertheless contains other ingredients that are foreign to the naturalized social ontology that I aim to develop in this study. Though critical realists have recognized that scientific practices are relevant to ontological arguments in the context of natural sciences, it will be indicated below that their specific transcendental arguments for their natural and social ontologies contain certain problematic anti-naturalist features. I will also set out to show that in closer analysis many of the basic concepts of the critical realist social ontology turn out to be ambiguous combinations of the naturalist and the anti-naturalist ontological ideas. The proposed naturalization of the original critical realist social ontology thus means that, in order to avoid these problems and ambiguities, epistemically successful research practices and theories in the social sciences as well as in some related disciplines, such as cognitive and neurosciences, should be taken more seriously in critical realist social ontology. In addition, I will also indicate that some ideas and concepts developed in the recent naturalist philosophy of science could be employed in order to get rid of the conceptual ambiguities in the critical realist social ontology.
In developing a naturalized version of this ontology, I draw selectively on the works of Mario Bunge and William Wimsatt as well as of some other contemporary naturalist philosophers of science. Some aspects of the naturalized critical realist social ontology, especially those related to human action and culture, are also elaborated by taking into account some of the recent developments in cognitive sciences that are directly relevant to the ontological assumptions and presuppositions of social research. As I try to show in the following chapters, the naturalist approach in social ontology enables the clarification of many the critical realist ontological concepts, such as causal power, emergence, agency and social structure. It will be argued, too, that a naturalist perspective elaborated below provides promising prospects for the elaboration of the critical realist methodology of explanatory social research. Hence, this study suggests that the naturalization of the original critical realist social ontology would increase its relevance to research practices in the current social sciences as well as in interdisciplinary studies. In order to further motivate the core argument of the book, I will next specify some of the anti-naturalist elements in the original critical realism.
Anti-naturalist elements in the original critical realism
Critical realism, initially developed in Roy Bhaskar’s (e.g. 1978, 1979, 1986, 1989) early philosophical works, started as a research programme that vindicated ontological questions in the context of philosophy of science. Its main contribution to the philosophy of science at the time was an analysis and ontological critique of the empiricist (or positivist) accounts of science and an introduction of the abstract philosophical ontology that was meant to provide a basis for a critical realist and naturalist epistemology and methodology. Explicit focus on ontological issues was surely a radical and fruitful idea in those days when discussions on ontological questions in the mainstream philosophy of science were almost non-existent. Bhaskar (1978, 16, 40) made his case by arguing that in ‘empirical realism’, advocated by the most empiricists (or positivist) philosophers of science, ontological questions about being are illegitimately reduced to questions about our knowledge of being, which amounts committing an epistemic fallacy.
However, Bhaskar not only brought ontological questions back in to the philosophy of science. He also drew a sharp distinction between his transcendental realist philosophical ontology and scientific ontologies of specific scientific theories and research programmes. He writes that ‘a philosophical ontology is developed by reflection upon what must be the case for science to be possible; and this is independent of any actual scientific knowledge’ (Bhaskar 1978, 39). Hence, an adequate philosophical ontology ‘must avoid any commitment to the content of specific [scientific – T.K.] theories and recognize the conditional nature of all its results’ (Bhaskar 1979, 6). In this conception, philosophical ontologies are conditional in the sense that they assume the existence of science, or certain generally recognized scientific practices, but they are independent of the ontological assumptions made in specific scientific theories and in the uses of these theories in the practices of empirical research. Bhaskar (e.g. 1978, 36; 1986, 13–14), however, admits that his transcendental realist philosophical ontology should deal with the same world studied by empirical sciences and that, in the long run, it must be consistent with the findings of sciences.
What is important for the ensuing arguments is that this relatively sharp distinction between the transcendental realist philosophical ontology and the scientific ontologies is not compatible with the naturalist approach to ontology advanced in this book. This is because ontological naturalism not only emphasizes that the current scientific and naturalist philosophical ontologies are continuous with each other. It also requires that the epistemological status of the claims (or assumptions) made in them both should be considered as fallible exactly in the same sense though the naturalist philosophical ontologies are typically developed by using more abstract terms. This means that all naturalist ontological theories should be understood as knowledge a posteriori which is always hypothetical, because, as will be later argued, there is no specifically philosophical or transcendental (as distinct from empirical) warrant for any philosophical ontology.
In keeping with his distinction between philosophical ontology and scientific ontologies, Bhaskar, by contrast, assumes that his transcendental realist ontology should be given a specific philosophical justification. This appears to be the reason why he invokes Kantian transcendental arguments in order to defend the transcendental realist ontology developed in A Realist Theory of Science (RTS, 1978). In the context of natural sciences, Bhaskar’s transcendental arguments take as their premises generally recognized natural scientific practices (e.g. experimentation). Intentional action and social practices as conceptualized in everyday experience, in turn, form the premises of his transcendental arguments for his social ontology (Bhaskar 1979). It should be emphasized here that these latter arguments omit detailed references to the social scientific theories and practices of empirical social research. Bhaskar (1978, 1979) nevertheless incorporates immanent critiques of other philosophical and methodological positions to his transcendental arguments. The importance of immanent critiques is also emphasized in his later reflections on the method of transcendental argumentation (e.g. Bhaskar 1986, 14; 1989, 182). Nevertheless, it will be argued in Chapter 4 that transcendental arguments are incompatible with the naturalistic view on ontological argumentation, since they contain certain problematic a priori elements. I will also contend that the previous point applies to Bhaskar’s qualified version of the method of transcendental argumentation developed in his later texts.
Bhaskar’s transcendental arguments are not only incompatible with the naturalist philosophy of science. They are also the source of certain problems in the critical realist social ontology. One of these problems concerns the dualistic nature of Bhaskar’s early social ontology. By taking our everyday conceptualizations of the features of intentional action and our social environment as the premise of his social ontological arguments, he ends up separating social reality quite strictly from nature. For example, he identifies the three much-discussed ontological differences between the natural and the social structures (i.e. the activity-and concept dependency as well as the more restricted space-time endurance of the social structures), and argues that it is precisely due to these ontological differences that a realist social science is possible (Bhaskar 1979, Chapter 2). In addition, as will later be later indicated in detail, Bhaskar’s (ibid., Chapter 3) theory of human agency is largely dualistic in that it fails to explicate the relation of mental states and powers that constitute human agency (i.e. capacity for intentional action) to the material powers and processes related to human bodies and embodied social interactions. It will be argued that, despite some recent attempts to address this relationship (e.g. Archer 2000; Elder-Vass 2010, Chapter 5), critical realists have not yet been able to solve this problem convincingly.
Furthermore, Ted Benton (1981; see also Collier 1994, 242–248) has argued quite convincingly that, notwithstanding Bhaskar’s claims for critical methodological naturalism, his early conceptions of the methods of social sciences and social ontology may be regarded as versions of qualified anti-naturalism rather than naturalism. According to Benton (1981), this is because Bhaskar emphasizes the differences between the natural and the social sciences over their common features. Bhaskar, for example, takes for granted some aspects of the hermeneutic and neo-Kantian traditions in the social sciences, such as those related to the meaningful and intentional nature of human agency, while rejecting their hostility to causal explanations. This is evident in his claim that ‘any adequate account of social science must accept that social causation depends upon the identification by the agents concerned of conceptual connections (that is that social causation is conceptually and linguistically mediated)’ (Bhaskar 1979, 173) which is taken to imply that ‘social science depends upon [the method of – T.K.] verstehen’ (ibid.; see also Sayer 1992). Bhaskar (1979, 173) nevertheless qualifies these statements by contending that the concepts (or meanings) of agents concerned may not be adequate with respect to the social practices and structures in which they are involved nor are the latter exhausted by the agent’s concepts. As a result Bhaskar (1979) ends up maintaining that, just like natural sciences, social sciences should aim at delivering causal explanations of empirically observed phenomena, even though the methods used in the social sciences should be tailored to appreciate the open systemic and hermeneutical nature of their subject matter.
Benton (1981) notes, too, that Bhaskar’s conception of the natural sciences in his early works is far too restricted and that he therefore ends up underestimating the methodological differences between different natural sciences. This is problematic because the methodological differences between physics, geology, ecology and evolutionary biology may well be as vast as between, say, evolutionary biology and sociology. In particular, by focusing on experimental physics and chemistry:
Bhaskar […] seem to neglect a range of natural sciences in which experimental closure is not an available means of empirical control on theory. Historical natural sciences such as geology and evolutionary biology explain phenomena in terms of the interaction of pluralities of mechanisms in open systems. In each of these sciences techniques have been developed […] for including an element of empirical control into theory-production and theory-correction.
(Benton 1981, 19–20)
So, according to Benton’s argument, we should not postulate a methodological gulf between the natural and social sciences, but rather to appreciate that ‘methodologically […] sciences display a “family resemblance”, of cross-cutting and overlapping differences and similarities of method’ (ibid., 21). Indeed, Bhaskar (e.g. 1998a, Postscript) himself has also later acknowledged the importance of these methodological differences w...
