Critical naturalism
​
In this phase of Bhaskar’s work, he challenged the hegemony present in the scientific world at the time, which assumed that social science was a lesser kind of science, if one could call it a science at all. His challenge consisted of the claim that it is possible for both the naturalist sciences and the so-called anti-naturalist (social) sciences to use the same metatheory to guide their research methodologies, even though they must nevertheless use different research methods because of their different subject matter. This was different from what most academics were saying at the time. They were arguing that social science was not 'objective' - and thus it was not scientific - because social scientists cannot place people into laboratories and do experiments on them and social scientists cannot measure people's ideas to confirm that their ideas and/or what they are about is real. An example of a naturalistic, objectivist approach is positivism , and an example of an anti-naturalistic, subjectivist approach is hermeneutics.
​
To the contrary, Bhaskar showed that good science can be conducted outside laboratories, indeed, it always has been (think here, for example, of astronomy). He argued for his position by demonstrating that natural scientists’ reliance on statistical correlations to determine causal mechanisms is misplaced in the social sciences. Instead, since social science cannot be carried out in a laboratory, confirmation and falsification must depend on something other than predictive correlations. That is, it depends instead on explanation and the judgementally rational process of comparing different, competing theories about a causal mechanism and choosing to run with the theory that explains most of the evidence. He said, “We can have confirmation and falsification of our social theories, provided that these are based on explanatory, rather than formally predictive, criteria” (Bhaskar, Interdisciplinarity and Wellbeing, p. 41).
​
However, the scientific status of the social sciences was not the only problem that Bhaskar’s transcendental realism was able to resolve for the social scientists. He was also able to resolve two longstanding controversies in the field. These controversies centred around the dichotomy between social structure and human agents and the dichotomy between individualism and collectivism, that is, the individual and the whole or the collective.
​
If we consider the dualism between social structure and human agency, we see that it is usually presented as a conflict between the thought of Emil Durkheim [1858-1917], who stressed structure; and Max Weber [1864-1920], who stressed agency. The problem is, if structure governs society the way that, say, gravity governs the material world, then there can be no free will: people must be completely determined. On the other hand, if agents are fully responsible for social structure, they should have a great deal of power to easily change that social structure, which does not match our general experience of the way that social structures limit our actions.
​
Bhaskar resolved this dualism by using the concept of emergent transcendental entities, which he had developed in the first phase of his work. Therefore, society was emergent from the activities of humans, but it could also act back and constrain or enable the activities of those humans. For example, a social structure that we call “ the system of education” has emerged in most countries due to the past and present activities of countless people; and the citizens are constrained and enabled by this educational system, whilst at the same time there are possibilities open to them to change the system (they might, for example, vote for a government leader who plans to reform the education system). Bhaskar said that “it is obvious that society pre-exists us, individual human beings, but at the same time it does not exist independently of the activity of human beings. Therefore, society pre-exists, but is reproduced or transformed by, human beings... One therefore has to think in terms of two levels: a level of agency; and a level of structure. The social structure is what the agent makes use of in carrying out their acts. In virtue of their activity, and the activity of other agents, the social structure is reproduced or transformed.” This concept is called the Transformational Model of Social Activity (TMSA) (Bhaskar, The Possibility of Naturalism, p. 40). Let’s consider the example of the social institution of marriage. We can’t say that humans get married with the objective of maintaining the institution of marriage. They get married, one hopes, because they love each other and want a home together. The institution of marriage enables this dream of theirs. However, by getting married they reproduce the institution of marriage. Furthermore, they can change the institution of marriage, for instance, it is easier to get divorced these days, compared to the past. Thus, marriage pre-exists human agency.
​
Let us look now at the related dualism between the individual and the collective arguments about what constitutes society. It is based on the assumption of “methodological individualists” such as Karl Popper [1902-1994] that anything in society that was not individualist – for example a system of education – was a magical thing created by the actions of huge numbers of individuals. However, this is to misunderstand the subject matter of sociology and social sciences. Researchers in these fields are not interested in the behaviours or actions of huge numbers of individuals. Rather they want to know about the enduring relations between individuals, such as those between citizen and member of parliament; between employer and employee; between husband and wife; or parents and children. These relations are different things from the individuals themselves and therefore we cannot reduce social science to the study of individuals, it is also about the study of the relationships between the individuals. Again, Bhaskar laid the groundwork for this argument in his first transcendental realist phase because there he had argued that things that are not directly measurable - such as "relations" or "social structures" which we cannot "hold in our hands" so to say - are nevertheless real entities.
​
Although Bhaskar argued that both natural science and social science were at base naturalist, he nevertheless fully agreed with the hermeneuticists that there are differences between the subject matter of the natural and social sciences. He stated that these different in terms of:
1. Activity dependence
Social structures are dependent on human activity in a way in which natural structures are not. For instance, gravity does not depend on human beings, but capitalism does.
2. Concept dependence
Social structures cannot operate independently of some conceptualisation by agents, they are concept or belief dependent.
3. Space time dependence
Social structures are more space time dependant in general than natural structures.
4. Internal relationality
As a social scientist, one’s beliefs about the subject matter may themselves be a part of the subject matter. Or looking at it in another way, the social scientist or the philosopher of social science, are part of, are immersed in, society. They are a part of their subject matter and this means that a social scientist has to be reflexive in a way that a natural scientist does not have to be.
​
Figure 1: Transformational Model of Social Activity, Bhaskar, The Possibility of Naturalism, p. 40
​
​
​
