Friday, May 4, 2012

Post-Positivism and Social Science.

 In the social sciences, the selection of research design and its constituent elements is an important phase of the research process.  The choice of research design is subject to a number of theoretical and methodological considerations.  Within the discipline of sociology, there exists a high level of theoretical and methodological pluralism with competing approaches to the study of society being pursued simultaneously which often gives rise to contention and contestation over the relative value of approaches. While quantitative research has relatively well-established principles of evaluation, appraisal of qualitative research is highly contentious. Carter and Little (2007) have suggested that consistency of research design is an important criterion via which to evaluate qualitative research:  epistemology, methodology and methods have to be internally consistent in order to form a solid research design. Adoption of a particular epistemological stance can affect researcher’s methodological choices, as some forms of epistemology are inconsistent with certain methodologies.  In view of this problem and to demonstrate the logic of research design, the constituent parts of epistemology, theory, methodology and methods will be given further exposition and organized via the principles of post-positivist epistemology into a consistent model of research design.  

The relationship between research design and epistemology is often left implicit and opaque within the final research report, but the selection of an epistemological position is a significant determinate of an appropriate methodology for a study.  Therefore, clarification of the concept of epistemology is an important first step.  Epistemology is a branch of philosophy that studies the nature of knowledge and justification; it is concerned with questions of ‘what is knowledge?’ and ‘what does it mean to know?’.   In the history of Western philosophy, epistemological preoccupations stretch back to the pre-Socratics.  The notion that objective truth is impossible to attain is ascribed to Protagoras, who is said to have argued that “man is the measure of all things” (Russell, 2006, p. 83).  Thus, in this formulation truth is deemed to be perspectival, a position which is echoed by post-modern and stand-point epistemologies in contemporary sociological theory (Moore, 2009).  However, the notion that truth is perspectival and that objective knowledge is impossible is only one epistemological tradition.  Plato (1987) rejected Protagoras’ relativistic conception of truth and outlined a rationale to differentiate between mere opinion and objective knowledge.  This is encapsulated in Plato’s (1987, p.  249) notion of the ‘divided line’,   on one end of the line we have illusion and unjustified beliefs and on the other end reasoned claims and knowledge.  While the positions of Plato and Protagoras are echoed in contemporary epistemology and sociological theory, important advances have been made in the field of epistemology since antiquity - especially with regards to the post-positivist tradition.        

The post-positivist tradition of epistemology emerged out of the collapse of logical positivism in the 1930s (Moore, 2009).   Kurt Gödel’s “incompleteness theorem” demonstrated that no mathematical system could be both self-validating and consistent, which invalidated the project of logical positivism and members of this tradition sought to reorient their intellectual project (Moore, 2009, p. 77).   Popper (1963) developed a philosophy of science based on the criterion of falsification that broke with the criterion of verification advanced by the logical positivist school.   Science advances by conjectures and refutations: scientific theories are formulated that have predictions that can be falsified (Popper, 1963).  Knowledge is therefore not absolutely certain, but conjectural and the best available understanding of the world – open to the possibility of future refutation.  To exemplify the criterion of falsification, Popper (1963) drew on Albert Einstein’s theory of gravitation.   Einstein’s theory of gravity had specific predictions that if proved false would invalidate the theory; the notion that light was affected by gravitational pull in much the same way as material objects was a claim at odds with scientific expectations of the time.  Through an experiment carried out by Eddington in 1919, the effect of gravitation on light was observed which thereby failed to falsify Einstein’s theory.  Popper (1963) emphasised the risk involved in making such a prediction, because failure to attain those results under observation would have meant  a falsification of the theory.  In terms of social science, the rigor of the criterion of falsification applied to Einstein’s theory of gravity is much harder to achieve due to the problem of reflexivity.   

 Popper (1963) was critical of social theory on that basis that any number divergent events could be interpreted as consistent with a theory and therefore validation of a social theory advanced by confirmation rather than attempted falsification. This is perhaps true with regards to “grand-theory” that attempt to conceptualize social systems in their totality and are therefore subject to high degrees of abstraction, but not “theories of the middle range” or working hypotheses used to explain specific social fields and empirical questions (Merton, 1967, p. 39).  Merton (1967) argues that middle-range theories involve less abstraction and are subject to empirical substantiation in regards to their specific field of application and therefore are not consistent with a wide range of empirical results. The specific sociological and psychological theories Popper (1963) addressed himself too were of a particular stripe, third-international era Marxism, Freud’s psychoanalysis and Adler’s individual psychology, which invoked defence mechanisms - false consciousness, repression and ego-threat, respectively – that saved the theory from failures of empirical falsification. Nevertheless, strict adherence to Popper’s criterion of falsification would be problematic within the social sciences. Bourdieu (1991, p. 32) argues that unlike the natural sciences, the ‘scientific’ credentials of social science should be judged like Kant’s categorical imperative:  according to intention and means as opposed to the result alone. Perhaps most importantly, as Alexander (1995, p. 91) argues, post-positivism is the solution to an “epistemological dilemma” that grips contemporary sociological theory between notions of absolute truth and relativism.   In a sense, post-positivism navigates a path between Plato and Protagoras and the problem of epistemological absolutism and relativism by conceptualizing knowledge as best available theory subject to further empirical investigation that might falsify the original hypothesis.  From this discussion it follows, that while epistemology is often implicit in the research design, it is an important component of the research strategy and determines the kind of knowledge that the study aims to attain and therefore exerts an influence over the selection of methodology and methods. 

In the design of a research project, the methodology and theory components are often more explicit than the epistemological stance.   Methodology is often used interchangeably with method; however some social theorists and researchers maintain that the terms should be differentiated for each other to denote specific aspects of research design.   Carter and Little (2007, pp. 1317-1318), building on Keplan’s distinction between “logics-in-use” and “reconstructed logic”, maintain that methodology should be defined as an explicit justification of research strategy and selection of methods,  methods defined in turn as research action.   Keplan (1964) defined logics-in-use as the procedures and strategies employed by researchers to produce knowledge of the social world and reconstructed logics as explicit attempts to formalize, analyse and justify logics-in-use.   Thus, reconstructed logics are commensurate to methodologies:  “the study – the description, the explanation and the justification of – the methods, and not the methods themselves” (Keplan, 1964, p. 18).   From this definition of methodology, the significance of epistemology as the study of knowledge and justification to research design becomes obvious as there is a clear line of inference between attempts to justify method as means to gain knowledge and formal theories of knowledge.   In research design, post-positivist epistemology has implications for both the selection of methodology and theory, and moreover the value placed upon both.

The implication of post-positivism for research design becomes evident when contrasted with post-modern and relativistic epistemologies that favour different methodological stances and place a different value on the knowledge produced (Alexander, 1995). In reference to the most basic division of research method, between quantitative and qualitative approaches, divergent epistemologies affect selection of methods and their justification in methodology. From the position of post-modern and relativistic epistemology, the adoption of a quantitative methods concerned with statistical generalizations based upon representative samples of a target population is less consistent with its underlying epistemological assumptions than a set of qualitative research methods that explore the subjective construction of meaning by participants (Bryman, 2008; Carter and Little, 2007). Quantitative methods are logically consistent with post-positivist epistemology, and moreover when appropriate the ability to formulate empirical hypotheses with statistically tuned predictions allows for a more faithful application of the principle of falsification.  Research designs informed by post-modern and relativistic epistemology are more likely to adopt qualitative methods and have distinctive methodological justifications (Carter and Little, 2007). In this vein, Geertz (1973, p. 3) called for a sociology of “thick description” against the abstract grand theory of structural-factionalism that developed in the immediate post-war period.  

Alexander (1995, pp. 100-101), places Geertz’s notion of “thick description” within a general movement that emphasised a “return to the concrete” against nomothetic discourses in the 1960s and 1970s.  At this time, there was a proliferation of methodological and theoretical approaches that focused on microscopic analysis of social interaction and subjective experience and were critical of grand theory, a trend exemplified by Geertz’s (1973, p. 3)  ethnography steeped  in “thick description”,  Lyotard’s (1984, p. 3) “incredulity towards meta-narratives” and Garfinkel’s  (1967)  ethnomethodology .  Geertz’s explicitly justified his methodological choice of ethnology and hermeneutics on the basis that knowledge of the social world is localized and subjective as opposed to generalizable, objective and universal (Alexander, 1995).  Thus, in the research design adopted by Geertz there is a clear line of inference between epistemological considerations and the selection and justification of research methodologies.  Alexander (1995, p. 91) argues that Geertz’s set of theoretical oppositions fell into the trap of an “epistemological dilemma”.   From this epistemological stance, the importance of sociological theory is reduced and the possibility of objective knowledge of the social world is abandoned.  Greetz (1983) cited the work of Michel Foucault as an influence on his conception of epistemology and social theory.  Foucault (1980, p. 93) defined his intellectual problem as “what rules of right are implemented by the relations of power in the production of the discourses of truth?”  He came to the conclusion that knowledge was a function of power:  
  “There can be no possible exercise of power without a certain economy of discourses of truth which operates through and on the basis of this association.  We are subjected to the production of truth through power and we cannot exercise power except through the production of truth” (Foucault, 1980, p. 93.)
 Thus, Foucault inverted François Bacon’s dictum “knowledge is Power” (Russel, 2006, p. 498) and concluded that power is knowledge.  The epistemological implication of this reduction of knowledge to power is that truth is relative to power-relations and not an assertion that something is or isn’t the case.  Foucault (1980, p. 131) draws this logical conclusion from his intellectual problem:
“Truth is a thing of this world: it is produced only by virtue of multiple forms of constraint. And it induces regular effects of power. Each society has its regime of truth, its ‘general politics’ of truth: that is, the types of discourse which it accepts and makes function as true”.   
Foucault’s conception of power and knowledge provides no means to assess the relative value of epistemological claims or prioritise between discourses of ‘truth’.  In fact, Foucault (1980) was critical of the centralising nature of scientific discourse on the basis that it demarcated between the legitimate and the illegitimate leading to the subjugation of certain types of knowledge.   Foucault’s analysis of knowledge and power, despite it short-comings, highlights some of the problems that plagued traditional positivist epistemology – most importantly Foucault (1980) demonstrated that knowledge is produced by historical actors, temporally situated and enmeshed in social relations of power. However, as Bourdieu (1991, p. 32) argues, the ‘scientific’ credentials of social science should be judged like Kant’s categorical imperative:  according to intention and means as opposed to the result alone.  The failure of epistemological absolutism should not give way to epistemological relativism. Moore (2009) argues that relativists are disappointed absolutists, but this does not provide social science with a realistic and useful approach to knowledge production.  Knowledge should be pursued in terms of a rational tradition, which advances by conjecture and refutation whilst always maintaining its provisional and tentative nature (Popper, 1963).   

Alexander (1995) maintains that post-positivism provides a framework to develop a useful social science; which he argues can even integrate methodological approaches developed to accommodate epistemological relativism.  Geertz’s explicitly justified his methodological choice of ethnology and hermeneutics on the basis that knowledge of the social world is localized and subjective as opposed to generalizable, objective and universal (Alexander, 1995).  However, the information gained from these social research practices can be evaluated via the normative principles of the post-positivist tradition and integrated into the body of social science knowledge.  Even if only at the level of exploratory study, or justified as means to produce new hypotheses post-positivist epistemology provides an appropriate means to logically connect various components of the research design.  Thus, both deductive and inductive research is integrated logically via post-positivism.   Post-positivist epistemology allows for consistent research design on that basis that it provides a framework to accommodate and differentiate between the relative value and merit of a methodological approach based upon the nature of the research question undertaken.    

Consistent research design is an important criterion for social science research; this requires that there is a clear line of inference between epistemology, methodology, theory and method (Carter and Little, 2007).   Epistemology is often left implicit in research design; however it is an important component   that often determines the type of methodology a researcher pursues to justify their research strategy and methods.  Within the history of philosophy, there has been a conflict between forms of epistemological absolutism, represented by Plato, and epistemological relativism, represented by Protagoras.   Both epistemological traditions are echoed in contemporary sociological theory.   Positivism can be said to derive from the traditions established by Plato while post-modern and stand-point epistemologies can be said to derive from the tradition established by Protagoras.  Both traditions constitute an “epistemological dilemma” for the social sciences and a third tradition, post-positivist epistemology proves a means overcome these short-falls by conceptualizing knowledge as best available theory subject to further empirical investigation that might falsify the original hypothesis. (Alexander, 1995; Popper, 1963).  Thus, the certainty required of knowledge in the epistemological absolutism is reduced opening knowledge up to the possibility of falsification and fallibility.    Given that epistemology determines the justification of knowledge, this has clear implications for the justification of methodologies (Carter and Little, 2007).   Geertz’s (1973; 1983) example demonstrates this:  relativistic epistemology led to the selection of ethnographic and hermeneutic research methodologies.  However, relativism poses serious problems for the epistemological status and worth of sociological theory and post-positivist epistemology provides a better framework to integrate different methodologies and methods into a coherent body of theory.   Theories at different levels of abstraction: from grand-theory, theories of the middle range and working hypotheses (Merton, 1967) can be integrated and evaluated by the normative standards of post-positivism.  Despite the problems that plague the traditional epistemology, post-positivist epistemology is an important normative method to integrate social theory and forge a variety of useful methodologies and research methods into a consistent research design.  

Written by Mathew Toll.


Alexander, J.C. (1995),   Fin De Siecle Social Theory:  Relativism, Reductionism and The Problem of Reason, London; Verso. 

Bourdieu, P. (1991), Language and Symbolic Power, Trans.  G. Raymond and M. Adamson, Ed.  J.B. Thompson, Cambridge; Polity Press. 

Bryman, A. (2008), Social Research Methods, 3rd Ed, Oxford; Oxford University Press.

Carter, S.M. and Little, M. (2007),   “Justifying Knowledge, Justifying Method, Taking Action:  Epistemologies, methodologies, and Methods in Qualitative Research”,   Qualitative Health Research, Vol. 17, pp. 1316-1338.

Foucault, M. (1980), “Two Lectures”, Power/Knowledge:  Selected Interviews and Other Writings 1972-1977, trans.  C.  Gordon, L. Marshall, J. Mepham, K. Soper,  Ed. C. Gordon, Brighton ; The Harvester Press, pp. 78-133. 

Garfinkel, H. (1967), Studies in Ethnomethodology, New Jersey; Prentice-Hall. 

Geertz, C. (1973), The Interpretation of Culture, New York; Basic Books. 

Geertz, C., (1983), Local Knowledge, New York; Basic Books.

Keplan,  A. (1964),  The Conduct of Inquiry: Methodology for Behavioural Science,  San Francisco,  Chandler. 

Lyotard, J-F. (1984), The Postmodern Condition: A Report on Knowledge, Minneapolis; University of Minnesota Press. 

Moore, R. (2009), Towards the Sociology of Truth, London; Continuum. 

Merton, R.K., (1967), On Theoretical Sociology: Five Essays, Old and New, New York; Free Press. 

Russell, B. (2006), History of Western Philosophy, London; Routledge.

Plato, (1987), The Republic, Trans.  D.  Lee,   2nd Ed, London; Penguin Books. 

Popper, K. (1963), Conjectures and Refutations:  The Growth of Scientific Knowledge, London; Routledge.

1 comment:

Mathew Toll said...

This post has been cited here:

Sardamov, I., 2015. Out of Touch: The Analytic Misconstrual of Social Knowledge. Critical Review, 27(1), pp.89-126.