Natural and Social Sciences: How different are they?

Excerpt from ‘Uncertainty and Social Capital in Outsourcing: A Conceptual Investigation’ PhD Thesis, Surja Datta, 2008

Many (Weber 1978, Hayek 1941, Durkheim 1938) have made the distinction between social and natural sciences. These authors have pointed out to the fact, that social science is distinct from natural science and the way of enquiring into these phenomena should be distinct as well. Yet the field of social sciences has been marked with deployment of research methods that have been the primary methods in physical sciences.

The questions that arise from this observation are as follows:

• If the fields of social science and physical science are fundamentally different as argued by several philosophers, can there be a commonality in research methods for exploring these distinct phenomena?
• Are there certain contexts in social science that are akin to natural sciences and others which are very different?
• What are the research methods that are distinct to social sciences and in which contexts they are most applicable?

This chapter adopts a funnelling approach, in the sense it starts with the broad questions such as the ones articulated above, and then dives into the specific research strategy adopted in this dissertation.

The question of how social science is distinct from physical sciences has been dealt with in different ways by different authors. Searle (1995) makes the ontological distinction between physical and social sciences. According to Searle, the physical world has a ‘reality’ that is ontologically objective, that is, the physical external world is independent in its existence of human observations. The social world, on the other hand, has an ontologically ‘subjective’ mode of existence that is such reality is a product of human mind and without human minds; they do not have a ‘separate’ mode of existence. Making the ontological objective subjective distinction between the physical and social world makes one thing apparent and that is the self referential nature of social reality. The social reality that we experience exists only because collectively we ‘accept’ its reality. Without our collective acceptance, such reality does not exist.

This leads to an important insight on social sciences, as articulated by Hacking (1999): the social reality as experienced by us does not have the sense of inevitability as is in the case with physical reality. A social reality need not have existed, or need not be at as it is. It is not determined by nature of things; it is not inevitable (Hacking 1999 pg 6). Moreover, the labels we give to represent a social object (Hacking 1999) interact with the social object itself. The representation itself is in someway constitutive of the social reality of the object. For example, the author of this dissertation is classified as an overseas student in the University and such classification influences the way in which the author views his relationship with the university and vice versa (overseas student and university are both social constructs and such labelling affects the interaction between the two). In distinction with social objects, the underlying reality of natural objects like quarks do not change, though our representations of them can and does change over time. Behaviour of quarks is independent of the way we represent them.

To explain social ontology, Searle (2005) invokes the concepts of social facts, which he defines as any fact involving collective intentionality of two or more agents. To explain the complex ontology of social reality, Searle distinguishes between social facts and institutional facts; the latter being a special subclass of the former. While the defining feature of social reality is common intentionality, institutional reality has certain additional notions like assignment of function, status function and deontic powers.

Assignment of function relates to the capacity of humans to impose functions on objects that are not intrinsic to the object. For example, a screw driver is recognised as one due to attribution of a certain function onto the object. Humans are not the only species that have such capacity; other species like apes, have also demonstrated such capacity to a limited extent. While a screwdriver can perform its function due to its physical structure, there are certain kinds of assignment of function where the object or the person to whom the function is assigned to cannot perform the function purely due to its physical features.

Such assignment of function works only when there is a collective assignment of a certain status and the object or person can perform the function when there is a collective acceptance by the community that the person or the object is in possession of that status.

So, the prime minister of Britain is expected to perform certain functions that have been collectively assigned to him/her and his/her ability to perform such depends on the collective acceptance of the community that s/he is in possession of the status of prime minister.

This also leads onto the notion of deontic powers, which are certain powers that are acquired by virtue of the assignment of the status function. The power of the prime minister does not arise from his/her physical attributes but rather from the collective acceptance of the community at large that he is in possession of certain powers acquired due to a particular status function. Organisations, money, prime ministers are all examples of institutional reality. Hacking highlights the difference between natural sciences and social sciences succinctly as follows:

‘…a cardinal difference between the traditional natural and social sciences is that the classifications employed in the natural sciences are indifferent kinds, while those employed in the social sciences are mostly interactive kinds. The targets of the natural sciences are stationary. Because of looping effects, the targets of social sciences are on the move.

(Hacking, 1999 pg. 108).

The subjective ontology of social reality, however, does not pre-empt any effort to achieve epistemic objectivity in the field. Indeed as Searle, (1995) argues without the idea of epistemic objectivity, social sciences would not be possible. However, the ontology of social reality does provide certain insights that the researcher should keep in mind while exploring social phenomenon. One such insight is that social facts are subject to change over space and time. Social facts are usually context dependent and it is difficult to make nomological statements in social sciences the way it is possible in physical sciences (Gordon,1991). In other words, it is difficult to generate universal laws for the social world as it is possible in the case of the physical world. This view can and should affect the self awareness of any social science researcher. The insight that the values, moral and beliefs of the researcher have been socially constructed over a period of time should lead to greater consciousness that there is nothing inevitable about such values, moral and beliefs. This in principle, should allow for more reflexivity on the part of the researcher. This view of the researcher is similar to what Rorty (1989) calls an ironist; a person who recognises that even his/her most central beliefs are contingent in nature and there is nothing inevitable about their existence.

While the above gives us a sense of why social sciences are different to natural sciences, it does not answer the questions articulated earlier: Are the research methods of natural sciences fundamentally ‘unsuitable’ for social sciences? Or are there contexts in social sciences that are more suitable to the application of methods from natural sciences.

Some philosophers like Taylor (1985) take the view that such commensuration of research methods between natural and social sciences is not possible as the epistemological foundation of physical sciences is an unsatisfactory base for enquiry into social sciences. The main reason, behind such dissatisfaction is that methods of natural science fail to capture the social meanings that are embedded in interactions between individuals in society. Beliefs, motives, intentions of individuals are central to the creation of social reality and any epistemic method that does not capture the interpretive nature of human interactions should have limited applicability in social sciences.

These concerns are quite valid and indeed an attempt to apply methods of natural science to social science necessitates an account of how such concerns are alleviated. Phillips (1992) makes the point that not all social science requires strong hermeneutics (or in other words not all social science needs that we understand the beliefs, motives, intentions of individual actors in great depth). He goes on to say that there a number of ways in which social sciences can be held to be similar to the natural sciences. Popper (1961) for example points out that much of the current social institutions are not consciously designed but have evolved as the un-designed results of human action. An example of such unintended consequence of human action can be the case of rise in house prices. When a buyer purchases a house, he does not necessarily intend that the market for house prices go up as a result of his purchase. Nevertheless, an unintended consequence of his action can be that prices of houses in the market indeed rise. A study of real estate prices and how much the price of a commodity will change as the number of individuals in the market changes can be cited as an example of research in social sciences that requires relatively less hermeneutics.

Even in this example, a research question that requires explaining the motives of individual buyers would require a more interpretive stance on behalf of the researcher than a research question that looks at the phenomenon at a more aggregate level.

The requirement of a strong interpretive stance for exploring certain research questions in social sciences also raises the issue whether it is possible to achieve epistemic objectivity in social sciences.

The issue of epistemic objectivity has sometimes been equated with the idea of attaining absolute truth. It has been suggested that as social sciences in many occasions need a hermeneutic approach, it is not possible to strive for truth and hence epistemic objectivity, when used as a synonym for truth, is an unattainable goal in social sciences. Popper argues against this position by stating epistemic objectivity and truth are two separate issues. While we are never sure about the ‘truth’, we can and indeed should strive to be ‘objective’ in our research approach. The following two excerpts from Popper articulate his viewpoint on epistemic objectivity quite clear:

What may be described as scientific discovery is based solely upon a critical tradition which, despite resistance, often makes it possible to criticise a dominant dogma. To put it another way, the objectivity of science is not a matter of individual scientists but rather the social result of their mutual criticism, of the friendly-hostile division of labour among scientists, of their co-operation and also of their competition.

(Popper, 1976, pg. 95)

So, epistemic objectivity of sciences (social science included) rises from the particular research process that researchers follow and which needs to meet certain criteria that are not arbitrary. So, the peer review process in academic publishing and the process of having Doctorate (PhD) dissertations reviewed by external examiners are examples of the effort of the academic community to achieve epistemic objectivity of their work. It is important to note that to attain objectivity in research does not mean that the researcher adopts a ‘value free’ approach to the phenomenon being explored.

It has been shown that it is difficult and indeed may not be desirable for the researcher to adopt a value free approach (Popper 1976; Phillips 1992) as the values are intertwined with the consciousness of the researcher. However, precisely because of the value laden perspective of the researcher, it is important that the researcher him/herself and the academic community at large adopt a critical attitude towards the research and look for assertions made in the research that are ‘falsifiable’.

‘Truth’ on the other hand, is an altogether separate issue according to Popper. He uses the metaphor of a mountain peak almost permanently wrapped in clouds to drive home the message that it may indeed be impossible to achieve ‘absolute truth’ through the enterprise of science. The status of ‘truth’ in the objective sense, as correspondence to the facts, and its role as a regulative principle, may be compared to that of a mountain peak which is permanently, or almost permanently, wrapped in clouds.

“The climber may not be merely having difficulties in getting there- he may not know when he gets there, because he may be unable to distinguish, in the clouds, between the main summit and some subsidiary peak. Yet this does not affect the objective existence of the summit…The very idea of error, or of doubt…implies the idea of an objective truth which we may fail to reach”

(Popper 1976, pg 226).

The above metaphor is manifested in Popper’s idea of falsification, where the focus of scientific enterprise is to ‘falsify’ rather than verify existing scientific theories. To conclude this section which attempts to articulate the ontological and epistemological assumptions made in this research, the following statements can be made. The researcher recognises the subjective nature of social ontology and thus the inherent contingency of social reality. It is also the view of this researcher that recognition of such contingency does not preclude epistemic objectivity of social science research. Following from Popper, it is felt that epistemic objectivity rises from a particular process of enquiry followed in science rather from a goal to achieve an ‘objective truth’. The researcher also recognises the limitation of using research methods of natural sciences in social sciences.

Such limitations do not lead to an exclusion of research methods of natural sciences being applied in social sciences but rather to the recognition that application of such methods will be contingent on the nature of research questions being asked and the context in which the research is carried out.