Nevertheless, there are two points that I would like to comment on. I hope that these remarks may also clarify some issues that pertain to the critical evaluation of the critical realist ontology.
First, you write that: “The naturalistic argument consistently replaces ‘reasoning derived from transcendental necessity’ by ‘reasoning within the general framework of what we know about the world’, but leaves the deductive flow of the argument unchanged.”
According to my interpretation, Bhaskar’s transcendental arguments are not best described as strictly deductive arguments (in the sense that their inferential structure would consists of the steps that strictly follow the rules of deductive logic). Though there are sections in his early work where he describes transcendental arguments as a kind of deductive arguments (these were cited by Mervyn Hartwig in an earlier post), I nevertheless think that this construal of transcendental arguments tends to trivialize them. From this viewpoint, the whole debate would be about the justification of the premises of Bhaskar’s transcendental arguments, not about their inferential form. In my book, I argue instead that both the epistemic status (or justification) of the premises and the inferential structure of Bhaskar’s transcendental are problematic.
There are at least two observations that speak against the strictly deductive interpretation of Bhaskar’s transcendental arguments. First, in his later works, Bhaskar (e.g. 1986, 11; 1993, 405) suggests that his transcendental arguments are instances of retroductive arguments rather than deductive arguments. In this context, he also insists (as doesHartwig) that transcendental arguments are used in empirical sciences. In my book, however, I try to show that this is not the case unless transcendental arguments are identified with inferences to the best explanation. The problem with this latter interpretation in turn is that transcendental arguments lose their status as specifically philosophical arguments, and, therefore, this interpretation would question Bhaskar’s strict distinction between philosophical ontology and scientific ontologies.
Second, the Kantian language used by Bhaskar suggests that his arguments are rather “transcendental deductions” in a Kantian sense than deductive arguments in the sense of deductive logic. This was also one of the reasons why I did not formulated them as deductive arguments in my book. I would accordingly argue that Kant´s transcendental deduction of the pure categories of understanding is not best described as a deductive argument in the above sense. Though I am not a Kant scholar, I nevertheless think that interpretations of this kind are historically inadequate and lack textual evidence. I also argue in the book that Bhaskar’s (e.g. 1978, 259) employment of Kantian terminology creates tensions to his philosophical position: He wants to defend transcendental realism by using Kantian transcendental arguments while these arguments are tightly connected to the doctrine of transcendental idealism (at least in Kant’s Critique of Pure Reason). In addition, transcendental realism was the position that was firmly rejected by Kant. Hence, to say the least, I think that Bhaskar owes us an explanation of what exactly is the purpose (and meaning) of the Kantian terms like “pure reason” and “synthetic a priori truth” in his philosophy of science. Anyway, it is these Kantian inspired transcendental arguments that I reject. Though I admit that there are some social theorists that still utilize Kantian transcendental arguments, I would claim that these arguments are not used in the best practices of empirical social research.
In addition, as you mention in the post, the inferential form of my naturalist arguments can be understood in terms of inference to the best explanation. This makes them inductive rather than deductive arguments.
For these reasons, I would say that your point about “the deductive flow of the argument” is not exactly right. This was my first point.
In your interesting discussion with Merwyn Hartwig, you write that: “critical realism naturalized (as Kaidesoja advocates) would eliminate the philosophical apriorism of CR while preserving many (all) of the ontological conclusions”.
I just want to note that, in addition to providing a critique of Bhaskar’s transcendental arguments and outlining a naturalist alternative to them, my book also contains a series of critiques about the core concepts and doctrines of the critical realist ontology (including social ontology). For example, it is argued that Bhaskar’s accounts of the concepts of causal power and emergence are not only ambiguous, but also contain problematic anti-naturalist assumptions. These assumptions are related to Bhaskar’s tendency to detach causal powers from the concretely existing powerful particulars in Harré and Madden’s (1975) sense. I especially argue that this view is involved in his conception of the“transcendentally real” nature of causal powers, including the emergent causal powers of minds and social structures.
Furthermore, I try to argue that a non-transcendental realist (or naturalized) interpretations of the concepts of causal power and emergent property (that can be justified by considering the ontological assumptions and presuppositions of the epistemically successful scientific practices and theories) allows the development of the naturalized (in the broad sense of the term) version of critical realist social ontology. These hypothetical ontological views imply, among other things, that causal powers are properties of concrete entities that may, but need not, be their essential properties in the sense that they would fix the membership of the concrete entity in a natural kind consisting of the collection of entities with identical essences. They also entail that emergent properties of concrete entities can often be reductively explained in terms of their underlying mechanisms though reductive explanations of this kind do not eliminate emergent properties from the scientific ontology nor do they require conceptual reductions. I also suggest that Mario Bunge’s (e.g. 2003, 35-36) CESM (Composition, Environment, Structure, Mechanisms) model of concrete social systems and William Wimsatt’s (e.g. 2007, chap. 12) notions of emergence in terms of failure of aggregativity and mechanism-based reductive explanation can be used in specifying the above views. In addition, the book also discusses how the theories of action and culture may be naturalized by employing the ideas developed in recent cognitive science, including the perspectives of embodied, situated and distributed cognition.
For these reasons, critical realism naturalized would not only eliminate the philosophical apriorism of CR. It will also suggest major revisions to the original critical realist ontology developed by Bhaskar and others while still preserving many of its core ideas in an elaborated form.
Bhaskar, Roy (1978) A Realist Theory of Science. 2nd edition. Brighton: Harvester Press.
Bhaskar, Roy (1986) Scientific Realism and Human Emancipation. London and New York: Verso.
Bhaskar, Roy (1993) Dialectics: The Pulse of Freedom. London and New York: Verso.
Bunge, Mario (2003) Emergence and Convergence: Qualitative Novelty and the Unity of Knowledge. Toronto: Toronto University Press.
Harré, Rom & Madden, Edward (1975) Causal Powers: A Theory of Natural Necessity. Oxford: Basil Blackwell.
Wimsatt, William (2007) Re-Engineering Philosophy for Limited Beings: Piecewise Approximations to Reality. Cambridge MA: Harvard University Press.