Though most philosophers of science today consider themselves naturalists in some sense, there still are many social scientists and philosophers of social science who describe themselves as anti-naturalists (or non-naturalists), both in the ontological and the methodological sense. Anti-naturalists often reject naturalism because of the methodological monism and reductionist ontology that are sometimes regarded as essential components of any naturalist philosophy of science. Due to their assumed methodological monism and ontological reductionism, naturalist views are considered problematic since, according to anti-naturalist critics, they impede our understanding of the nature of social research, the humanities, and social ontology. One of the aims of my book, Naturalizing Critical Realist Social Ontology, is to show that while methodological monism and reductionist ontology are indeed problematic, they should not be seen as the defining features of naturalism.
I take “naturalism” to be a philosophy of science that is continuous with the sciences and considers the human mind and social world to be embedded in the causal structure of reality. By contrast, methodological monism and reductionist ontology result from a failure to pay attention to the actual research practices and results of the cognitive and social sciences in the naturalist philosophy of science. In my book, I argue that empirical analysis of the current explanatory sciences does not support a view that the same scientific method is used in all empirical sciences. I also try to show that human agency, intentionality and normativity can be understood in naturalist ontological terms once we pay attention to the actual research practices of the cognitive and social sciences (rather than speculate about what a “complete neuroscience” that does not yet exist might be able to explain in “purely physicalist terms” in some distant future). Indeed, I suggest that the best explanatory practices in these sciences include an assumption that micro-reductive explanations of the systemic (or emergent) properties of cognitive and social systems in terms of their underlying mechanisms do not explain these properties away.
What I think is relatively novel in my arguments in this context is that I reject methodological monism and reductionist ontologies (e.g., eliminative materialism and reductionist physicalism), not because they are naturalistic (or science-oriented), but because they are not naturalistic enough – meaning that they are not compatible with the best explanatory practices and developments in the empirical sciences. The meaning of the term “sciences” is here construed in a broad sense that encompasses the research practices both in the natural sciences and in the “behavioral” and social sciences.
Accordingly, the book contributes to the ongoing discussion on the proper scope of the naturalist philosophy of science. It does this by defending a naturalist program of social ontology that aims to do justice to the best explanatory practices and results of the behavioral and social sciences, and to the relevant parts of the natural sciences. One of the reasons why this task is important is that many recent developments in studies on the cognitive evolution of humans, social neuroscience, and other cognitive sciences support an ontological view that humans are social and cultural beings by nature, even though the conceptualization of culture and social life in these sciences is often relatively underdeveloped from the viewpoint of the social sciences. Although I think that these developments should be taken into account in the social sciences, it is also my view that, in contrast to what appears to be the view by many philosophers engaged in debates on social ontology, a naturalist social ontology cannot be plausibly developed without paying serious attention to the ontological assumptions and presuppositions of the epistemically successful research practices and theories in the social sciences.
Furthermore, unlike some methodological monists, I do not want to restrict my attention to those explanatory practices in the social sciences that use experimental and/or statistical methods that seek to imitate experimentation as far as possible. Though practices of this kind form an important part of the social sciences and cannot be ignored in naturalist social ontology, it should be noted that their ontological assumptions and presuppositions often include remarkable (and sometimes misleading) simplifications and idealizations. This is the reason why I think that ontological studies on these experimentally-oriented practices should be complemented with ontological analysis on the ethnographic and historical studies of human cognition and social life in its naturally occurring settings (in contrast to artificially controlled or idealized environments). In this respect, my usage of the term “naturalism” is somewhat similar to that sometimes found in biology, where this label is earned by ecologists who observe and study plants and animals in the wild (i.e., in their natural ecological environment) rather than in controlled laboratory settings. These issues, regarding the need to incorporate the social sciences into a naturalist philosophy of science, form a part of the broader context of the book.
The core argument of my book, however, is that critical realist social ontology should be naturalized. There are two reasons why I decided to focus on a critical realism that is based on, but not limited to, Roy Bhaskar’s (e.g., 1978, 1979) early philosophy of science. Firstly, it is a relatively popular “metatheory” among social scientists, and it is perhaps the most well-known version of scientific realism in the social sciences. Secondly, even though it is commonly ignored in the naturalist philosophy of science, I think that critical realism provides a useful point of departure for developing a naturalized social ontology because it already contains a number of important naturalist ideas (as is recognized in the title of Bhaskar’s second book, The Possibility of Naturalism). These ideas include an account of science in terms of research practices, a general model of mechanism-based causal explanation that is assumed to fit all special sciences – despite the differences in their specific methods and objects of inquiry – and a view that ontological arguments should begin with an analysis of scientific practices. In addition, critical realists argue for a realist (or generative) theory of causation in terms of causal powers and tendencies as well as for a non-reductionist social ontology that is based upon the concept of emergent causal power. This emergentist ontology seeks to take human agency, social structures and cultural meanings seriously.
Despite the naturalist views it contains, a closer examination of critical realism nevertheless reveals that it also includes certain anti-naturalist ontological arguments and ideas, which are often intertwined with the naturalist ones in a complex manner. In the book, I analytically separate these two kinds of elements and show that the anti-naturalist ingredients in the social ontology of Bhaskar and his followers are sources of serious conceptual ambiguities and methodological problems. My argument then is that because of these ambiguities and problems, the anti-naturalist elements should be replaced by ontological concepts and views (which I outline in the book) that are internally consistent and more compatible with the naturalist position. The book also aims to show that this procedure allows critical realists to formulate ontological theories that are more in tune with many successful explanatory practices in the social sciences and with some of the most promising developments in the cognitive sciences.
More specifically, I argue that Kantian transcendental arguments, utilized by Bhaskar and many other critical realists, cannot be used to justify ontological doctrines without assuming some version of transcendental idealism (including historically and culturally relativized versions of this doctrine) that is incompatible both with ontological naturalism and scientific realism. It is then shown how the problems that pertain to the transcendental arguments of Bhaskar and his followers, such as the failure to provide detailed empirical justification for their premises – as well as the allegedly “transcendentally necessary” status of their ontological conclusions – can be avoided by reformulating these arguments as inferences to the best explanation. The aim of these naturalized ontological arguments is to develop fallible ontological theories that explain the epistemic success of our best practices of empirical research (described on the basis of an empirical analysis of them) and empirically well-confirmed explanatory theories better than the competing ontological views. Since arguments of this kind are also used in the empirical sciences, they are free from those a priori elements that vitiate Bhaskar’s transcendental arguments. Accordingly, I deny that there has to be a specific metaphysical (or transcendental) grounding for social scientific research practices and emphasize that the epistemic status of social-ontological theories is exactly the same (i.e., more or less fallible) as that of the explanatory theories in the social sciences, even though the former are typically more abstract and general than the latter.
In addition to transcendental arguments, the concepts of causal power and ontological emergence are among the key components of critical realism. Though I share some of their basic views pertaining to these concepts, I argue that both of these concepts are used ambiguously by Bhaskar and many other critical realists. For example, I demonstrate that there are sections in Bhaskar’s (e.g., 1978, 1979) books where he tends to detach causal powers from concrete entities by interpreting the former as transcendentally real features of reality. In my reading, this commits him to the view that these “transcendentally real” powers are regarded as some kind of abstract universals and structures that are (or can be) instantiated by (or somehow participate in the generation of) concrete things which, in certain conditions (e.g., when their powers are triggered or exercised), may become the actual objects of our fallible perceptions or causal interventions. Even though he does not consistently advocate this view, it is shown that this kind of conception of causal powers is assumed, for example, in Bhaskar’s (1979) property-dualist philosophy of mind as well as in his conception on social structures in terms of internally related social positions and roles.
My worry is that this view introduces such causal properties to critical realist ontology that cannot be studied using any reliable empirical methods utilized in the explanatory sciences and, hence, it is not compatible with naturalism. By means of a detailed comparison of their views, I further argue that this transcendental realist interpretation of causal powers is clearly incompatible with Rom Harré and Edward Madden’s (1975) account of causal powers in terms of “powerful particulars”. This is because Harré and Madden consistently ascribe causal powers and potentials to concrete entities (in contrast to abstract universals or conceptual abstractions). My suggestion is also that, unlike abstract social structures, concrete and organized social systems – such as particular schools, factories, political parties, governments, and transnational corporations – can be plausibly conceived of as such powerful particulars that may possess empirically tractable (though not necessarily perceivable) emergent causal powers. Accordingly, in contrast to Bhaskar’s relatively ambiguous account of the concept of emergent power, I restrict this concept to non-aggregative dispositional properties of concrete entities.
Another problem in the concept of causal power in critical realism is that it is connected to the essence-based notion of natural kind both in Bhaskar’s early works and in Harré and Madden’s book. For example, Bhaskar (1978, p. 171) claims that there is “the necessity implicit in the concept of a thing’s real essence, i.e., those properties and powers, which are most basic in an explanatory sense, without which it would not be the kind of thing it is, i.e., which constitute its identity or fix its membership in its kind.” I argue, by contrast, that we should disconnect the concepts of causal power and essence-based natural kind because the latter does not fit, for example, the current understanding of biological species in evolutionary biology, nor does it do justice to those psychological and social kinds that can be considered interactive (or looping) kinds in Ian Hacking’s (1999) sense. It is further suggested that Richard Boyd’s (e.g., 1991) Homeostatic Property Cluster theory of kinds appears to provide useful ideas for the ontological analysis of the “non-essence-based” kinds studied in the sciences.
The general conclusion of my book is that the original formulation of critical realism contains these and some other unresolved ambiguities and problems, which result from the attempt to combine naturalist and anti-naturalist ontological views that do not fit together. I further point out that these tensions have practical consequences as to how critical realism is utilized as a “meta-theory” in empirical social research since, at the methodological level, critical realism appears to give rise to an unjustified bias towards abstract theorizing and the favoring of interpretative methods over the “post-positivist” uses of statistical methods and theoretical modeling in explanatory social research. The message of my book is that, despite of the fact that Bhaskar and other critical realists should be credited for developing compelling critiques of the Humean regularity theory of causality (including the “positivist” research practices in the social sciences that are based on it) and extreme forms of social constructionism (including hermeneutical anti-naturalism), critical realism has not yet fully realized its potential as an underlabourer to the social sciences due to the problematic anti-naturalist elements it contains.
Although I have so far mostly concentrated on describing my critical assessment of the original critical realism, the book also elaborates a naturalized version of critical realist social ontology. The latter includes a non-transcendental-realist and non-essentialist concept of causal power that reconnects causal powers to what Harré and Madden (1975) call “powerful particulars” without assuming that these particulars should have real essences that fix their memberships in essence-based natural kinds. Furthermore, I combine the understanding of causality in terms of powers with an ontology of concrete social systems that selectively and critically draws on recent realist and naturalist philosophy of science, including Mario Bunge’s (e.g., 1998, 2003) systemic social ontology and William Wimsatt’s (e.g., 2007) notion of ontological emergence in terms of failure of aggregativity. Ontologies of human agency, cognition and culture are in turn expounded by examining the assumptions and implications of the perspectives of embodied, situated and distributed cognition that are in certain respects incompatible with Bhaskar’s philosophy of mind and social ontology. Finally, I demonstrate the fruitfulness of the basic ideas of Bunge’s systemic social ontology by using them to clarify the ontological assumptions related to some of the basic concepts of social theory. Nevertheless, I readily admit that I was only able to sketch a naturalized critical realism in broad outline. Although I regard the sketch as highly promising, I admit that it needs more work.
Bhaskar, Roy (1978): A Realist Theory of Science. 2nd edition. Sussex: The Harvester Press.
Bhaskar, Roy (1979): The Possibility of Naturalism: A Philosophical Critique of the Contemporary Human Sciences. Brighton: The Harvester Press.
Boyd, Richard (1991): “Realism, Anti-Foundationalism and the Enthusiasm for Natural Kinds”. In: Philosophical Studies 61, p. 127–148. [Crossref]
Bunge, Mario (1998): Social Science under Debate: A Philosophical Perspective. Toronto: University of Toronto Press.
Bunge, Mario (2003): Emergence and Convergence: Qualitative Novelty and the Unity of Knowledge. Toronto: University of Toronto Press.
Hacking, Ian (1999): The Social Construction of What? Cambridge MA: Harvard University Press.
Harré, Rom and Edward Madden (1975): Causal Powers: A Theory of Natural Necessity. Oxford: Basil Blackwell.
Wimsatt, William C. (2007): Re-Engineering Philosophy for Limited Beings: Piecewise Approximations to Reality. Cambridge MA & London: Harvard University Press.