Following the separation of the ontological and epistemological CR goes on to characterize the world, for the most part, as an open system from which we can see how the complexity of a multiplicity of generative mechanisms together with the notion of emergence allows these mechanism to interact at and across various levels of reality. Such an open system perspective is in stark contrast to the notion of closed systems, controlled variables and hypothesis testing of scientific experimentation. Bhaskar highlights that the social, unlike the natural, world is permanently open and due to this the social sciences are denied the predictive test for the development of new and existing theories. One of the reasons we can not set up a closed system to test human affairs is that people tend to adapt and act in different ways over and arguably within time. Thus a separate approach to the social sciences to that of the natural (TR) is needed.
By separating the two dimensions of science, and separating the sciences into natural and social, reordering them so that the ontological precedes the epistemological and by perceiving science as a social process of knowledge generation in interaction with the world, Bhaskar creates a stratified reality, the world of three overlapping strata (Bhaskar, 2008).
Sitting between the overlapping real and empirical is the actual. The observed world exists at this level and our interpretation of it forming the empirical. A one-way relationship exists between all strata, with the lowest level shaping the highest, a relationship, which forms the basis for all scientific explanation. What this real stratification highlights is that everything derives from the unseen strata of the real and within that stratum from generative mechanisms. It is with these mechanisms that we are concerned as social scientists as to understand reality we need to come to identify and understand the mechanisms which produce and manifest social phenomena such as marriage, crime, friendship, education etc.
Liberated from the shackles of a closed system the researcher must attend to the multiplicity of generative mechanisms that are at play. Thus prediction at the level of the actual becomes impossible, in the strictest of senses. In an open system, unlike a closed, we do not know if a particular result is due to a particular predetermined variable. What exists in the actual depends on all the mechanisms in effect. The flux of multiple mechanisms acting upon the phenomena is one of the reasons why empirical results cannot be guaranteed in open systems, given that we may have many mechanisms operating in various contexts. The results of social research therefore can never be generalized.
Once emergence is introduced, we include a multiplicity of generative mechanism of different types which are all assumed to interact at different levels of reality, and thus the need for different fields, disciplines, and theories capable of explaining such mechanisms and how they interact at different levels. Depth-inquiries facilitate the process of digging deeper into the complex levels of being through the application of CR explanatory critique.
For the purposes of educational research multiple depth-inquiries weaved together through explanatory critique provide researchers with a potential method, or a process, for undertaking research. Bhaskar further qualifies this process as,
any co-operative enquiry, which includes the frustrated agent(s) concerned, into the structure of some presumed set of mechanisms, constituting for that agent an unwanted source of determination, with a view to initiating, preserving or restoring the agent’s well-being
Such an enquiry within education is co-operative seeking to include agential voice drawn from the educational context (learners, teachers, administrators, officials, parents). It is also suggestive of Action and Practitioner Research, in which a relationship exists between theory and practice in practice, and transformation occurs as part of this process within the agents and when relevant the systems of agential activity (the structures of education and learning). This conceptualization of the parameters and nature of social research is broadly in line with Cultural Historical Activity Theory and ethnographic approaches to educational research, indicating the possibility of brining the three complimentary theories of and approaches to the social together.