Advantages and Disadvantages of Positivism
Q. Discuss the advantages, strengths, disadvantages and weaknesses of a positivist approach to the social sciences.
The profusion of use and multifariousness of meaning of the word positivism results in a need for any essay on the subject to first give its own precise definition for its use of the term, distinguishing its particular context from its use in other contexts. The term positivism, first coined by the philosopher Auguste Comte in the nineteenth-century, was first originally confined to the boundaries of philosophy and natural science; by the present, the term has spread its meaning to cover fields as diverse as law, political theory, the social sciences, philosophy and even literature. In all of these fields the dictionary definition of positivism as ‘. . . a system recognizing only that which can be scientifically verified or logically proved, and therefore rejecting metaphysics and theism’ (Oxford, 1989: pp. 385-386) remains broadly true of most of its uses, though it does little to reveal the subtle distinctions of use of the word positivism in each of these disciplines. For instance, legal positivism is ‘. . . a view which, in contrast to the natural law view, claims that a legal system can be defined independently of evaluative terms or propositions is the view that in law’ (Hugh-Jones, S. & Laidlaw, J, 2000: p88); in literature positivism refers to a specific period of Polish literature where writers were inspired by the nascent achievements of science and technology; and in philosophy the term logical positivism meant the scientific investigation of the philosophy of language — as in writers such as Wittgenstein. All in all then, the term positivism has an umbrella use designated by the dictionary definition, but then has several further and more individualistic uses depending upon the context in which it appears.
‘Positivism is the view that serious scientific inquiry should not search
for ultimate causes deriving from some outside source but must confine
itself to the study of relations existing between facts which are directly
accessible to observation’
(Hugh-Jones, S. & Laidlaw, J: 2000: p.3)
The definition of positivism chosen for use in this essay, its particular domain being the social sciences, is that stated above by Hugh-Jones and Laidlaw. According to this version of positivism, data gathered from sense perceptions is the only possible data that may be used as a foundation for knowledge and thought. Hence, all data and phenomena taken from beyond sense perceptions or the properties of observable things is banished — thuds a priori metaphysics and theology dismissed in toto. Science alone sets the perimeters for human knowledge, and, accordingly, positivism maintains the expectation that science will ultimately attain to solve all human problems. As such, a social scientific definition of positivism regards the research of social scientists as identical in importance to that of natural scientists; that is, social scientists, like natural scientists, employ theories and explanations for phenomena, inferred from sense data for the purpose of social benefit. With respect to political science as a social science Popper thus says ‘We get the particular definition of one of the social sciences — political science — which tries to separate the subject from the values we apply to it, and argues that it is possible to develop value-free knowledge’ (Popper, 1983: p. 75). This quotation shows the extent to which one particular social science’s use of the term positivism has mutated from its general umbrella use.
For the purposes of this essay, positivism will be regarded as having four essential characteristics (King, 1994: p. 204). (1) It is concrned with the search for the unification of scientific method, that is, with the notion that logic and inquiry are universal principles extending across all scientific domains. (2) That the ultimate end of scientific inquiry is to gives explanations of social phenomenon and to make predictions about their behaviour as according to discernable laws of society. Thus positivism in the social sciences seeks also to develop a ‘general law of social understanding’, by discovering necessary and sufficient conditions for any phenomenon. (3) Positivism maintains that social scientific knowledge must always be subject to proof through empirical experimentation. All subjects of reaseach and investigation in the social sciences should be based upon observations derived from sense-perceptions. (4) Social sciences must seek to free themselves of value-judgements as far as possible, and of moral, political, and religion ideas that might contaminate their research. Thus, in short: social sciences must seek to dicover universal conditions behind social phenomena;all social scientific empirical statements must be asolute truthes which are true at all times and true in all places; finally, research can proved only by empirical experimentation.
– – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – –
In There Is More Than One Way To Do Political Science Marsh & Smith (2001), while debating whether the social sciences might legitimately have both a positivist and realist approach to science, argue that one of the principal strengths of positivism is that it is ‘foundationalist’: that is ‘. . . in ontological terms it argues that there is a ‘‘real world’’ out there, that it is independent of an agent’s knowledge of it’ and that ‘. . . it is possible, using the proper ‘‘research methods’’ for an observer to discover these real relationships between social phenomenon’ (Marsh & Smith, 2001: p. 529). Thus the great strength and advantage of a positivist approach to the social sciences is that it grounds anthropology, sociology, political science and so on upon a hard and definite ‘foundation’ of empirically testable data, and makes theories out of this data from which absolute laws of social behaviour may be attained. A second distinct advantage then of positivism is that it permits an analysis of the causal relationships between phenomena. Positivism thus allows the social sciences to make certain predictions about the phenomenal world. Thus Dowding states ‘. . . all good political scientists produce models with definite predictions . . . which they can then test one way or another against data gathered from the actual world’ (Dowding, 2001: p. 92). A chief strength then of a positivistic approach, is that it brings to the social sciences the desire to emulate the excellence of the natural sciences in respect of their rigorous experimentation, precisely stated hypotheses, definite laws, and thus prediction of behaviour. By approaching its investigations thus, social scientists attain a high level of accuracy in their results and in their predictions, and thus come closer to a total description of the behaviour of social phenomenon. By approaching the social sciences from a positivist position, social scientists are able to cut away from existing ‘knowledge’ many prejudices, suppositions, superstitions and other non-scientific opinions that have gathered about these social phenomena (Marsh & Smith, 2001). In other words, positivism, by declaring valid only those things which conform to its vigorous standards of investigation, strips social phenomenon of their perceived nature and reveals them as they really are.
A second key advantage of taking a positivist approach to the social sciences is that such a move solidly roots the social sciences in the accomplishments of the natural sciences over the past four hundred years. Early positivists like Comte, Spencer and Saint-Simon understood their theory and work as something growing directly out of the experimental and theoretical achievements of the great natural scientists like Newton, Spinoza, Darwin and others. Comte knew that the natural sciences and natural scientists, were essentially positivist: that is, they appealed to the perception and measurement of objective sense-data from which to make experiments, analyze results and make theory, predictions and laws. Comte and the other early positivists thus understood their work as an act of ‘making explicit’ the theory which natural scientists had adhered to for centuries. When, in the twentieth-century, social positivists like Ernst Laas, Friedrich Jodl and Eugen Duhring began to establish the theoretical and experimental parameters of the social sciences, they also understood their work as a branch of the natural sciences and as a continuation of its discoveries. Anthropologists, sociologists, social scientists of the early twentieth-century faced a choice: they could orientate their subjects within the sphere of natural science and its immense harvest of the past two decades, or they could orientate it in the sphere of theology and the liberal arts which had dominated all human history before the advent of natural science. Laas, Jodl, Duhring and later Marsh, Smith and others have all agreed that the social sciences must be built upon the platform established by the natural sciences. These sciences have been the predominant intellectual authority for Western Europe for nearly four hundred years, and social scientists think that the positivist approach to the natural sciences offers greater objectivity, certainty of prediction, and deeper insight into their subjects than could achieved by any other method of inquiry.
Further, the allegiance of the social sciences to the natural sciences, through a shared conviction in the positivist philosophy, means that the social sciences can constantly draw upon the fund of new empirical material daily unearthed by these natural sciences. In other words: if the social sciences have an exchange of knowledge between themselves and the natural sciences, then every refinement of experimental method, theory, or analysis achieved by the natural sciences may be immediately seized upon and utilized by the social sciences also. And, vice-versa, this interchange allows the social sciences to more freely disseminate their discoveries within the world of the natural sciences. Moreover, by sharing a positivist philosophy with the natural sciences, the social sciences may draw from its authority in the presentation of their results to the wider scientific and academic community. That is, the employment of positivism by the social sciences, dispels and neutralizes the accusations from some quarters of the scientific and outside world, for instance those of Karl Popper, that such sciences are ‘pseudo-sciences’. This claim can hold no weight if it is seen that the natural and social sciences share alike the same methodology and principles of operation. Nonetheless, it should be made clear that whilst the social sciences derive authority and knowledge from the natural sciences, that they do not depend upon it exclusively for authority. Indeed, the social sciences have made their own refinements to positivism, and thus their methods of experimentation and analysis, quite independently of those achieved in the natural sciences. The social sciences have adapted the positivism they received from the social sciences to conform to their own empirical material and the idiosyncratic and diverse domains encountered in societies and the human world. In short, the social sciences have moulded positivism to the world of empirical human affairs, thus entering a territory that the natural sciences had previously not trodden.
– – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – –
Historically, perhaps the greatest weakness and hence disadvantage of positivism generally, and with respect to the social sciences in particular, has been its insistence upon methodological absoluteness. Since the time of positivism’s foundation in the philosophy of Auguste Comte, positivists have persistently sought to use its scientific methods to explain every conceivable aspect of social phenomenon; that is, they have wanted to observe an object in its totality, tracing its entire phenomenological casuistry, its material composition, and thus produce a absolute theory of knowledge about that phenomenon. According to this scientific philosophy positivism must produce absolute laws to describe the behaviour and nature of phenomenal objects. The naivety of this search for the perfection of methodology and absoluteness of social scientific laws was exposed in the second half of the twentieth century, firstly by the advent of post-modernism (Popper, 1989: p.109-128), which showed the epistemological difficulties — impossibilities? — of extending science to such extreme levels; secondly, positivism’s applicability in all instances was increasingly undermined by the new theories of social scientists themselves. The various discoveries of anthropology, sociology, political science and other social sciences led researchers to an ever clearer conclusion: the phenomena of social science are far too sophisticated and involve the intimate interaction of too many separate objects, people and processes to be scientifically observed in their totality.
Sociologists for instance, in their investigations into the mechanisms of the smallest of social units, the family, soon realized that no absolute and all-encompassing laws could be applied to the behaviour of these units (Gerrad, 1969: pp. 201-212); the great complexity coming from the need for the axioms and paradigms which are true of one family unit must, according to pure positivism, be shown to be true of all family units in all places and at all times. Pure positivism states that the laws of social science are of the same type and significance as the laws of physics, biology and chemistry; but for these laws to attain this equality, the laws of social science must be easily expressible and as rigorously testable as those of the natural sciences. The difficulty of attaining such equality is easily demonstrated by Gerrard’s (Gerrard, 1969) experiments, where he discusses the complexity of social issues involved in a four member family unit in America, and then postulates the near impossibility of scientifically demonstrating that family units in Northern France, in Thailand, in Hawaii and in all other places can be shown to obey the same exact rules as those affecting the family in America. Thus social scientists from the 1950’s onwards, confronted with the sheer vastness of ethnic, racial and community diversity, began to question the possibility of producing social laws that would be universally and ubiquitously binding. And in 2006 when even natural scientists have no certainties even about the exact behaviour and nature of a single atom; how can social scientists hope to prove laws for something as complex as a city?
Another weakness of extreme positivism has been its inability to accurately prove its hypotheses through empirical experiments (Popper, 1983: p. 12 & also: Dowding, 1995: p. 138). Whereas experimentation in the natural sciences usually involves the investigation of inanimate or relatively simple objects such as metals, stars, chemicals and so, these having the same properties constantly, in contrast, social phenomenon — people, communities, organizations etc., — are animate and are compositions of vast complexly intertwining feelings, emotions, thoughts, volitions, passions, motives, associations and so on. Thus, to undertake a social experiment, a social scientist has to be sure that he can separate the single mental or behavioural element, say ‘a criminal tendency’ that he wants to investigate, and then to exclude or control the influence of the other mental and social factors that will otherwise affect the accuracy of the experiment. In many instances such exclusion is nearly impossible to the degree of purity demanded by extreme positivists; a human being cannot be put in a test-tube or a vacuum and so shielded from external influences in the way that magnesium or atoms can. Thus social scientists have become ever more conscious that a major limitation of the positivist approach in respect to their discipline is its insistence upon perfect conditions for experimentation and for the accuracy of hypotheses and predictions (Dowding, 1995).
Further, other discoveries in the social sciences have begun to place an ever greater emphasis upon the life of the individual and upon subjective experiences as vital factors in the constituency of societies (Marsh & Furlong, 2002). The hermeneutic or ‘interpretive’ approach has come to assume ever greater importance within the social sciences, setting up for itself an area of investigation of phenomenon quite different from positivism, and therefore undermining the legitimacy of positivism’s claims to describe the totality of social phenomenon. Positivism is, according to this view, the outcome of a particular culture and particular history (Western European); what legitimacy then does it have to proclaim its results as of universal validity, as it must, to meet its own standards of scientific investigation? Moreover, social scientists themselves bring to their experiments their own subjective experiences, their own thoughts, volitions, prejudices etc., and these all affect experimentation and thus the security of results — just as surely do these things in the subjects of analysis. Thus David Marsh and Martin Smith have stated, in their powerful metaphor derived from Marsh’s earlier article, that ‘In the social sciences . . . subjective ontological and epistemological positions should not be treated like a pullover that can be ‘‘put on’’ when we are addressing such philosophical issues and ‘‘taken off’’ when we are doing research’ (Marsh & Smith, 2005: p.531). That is, they should not be treated as a ‘pullover’, as temporary measure, as they have been by positivists to date.
– – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – –
In the final analysis, it seems clear that neither the extreme positivism once advocated in the wake of Auguste Comte’s first philosophical writings, nor extreme anti-positivism nor anti-foundationalist positions as have recently been taken by some hermeneutists and realists, can lead to significant future progress in the social sciences. The chief strength and advantage of a positivist approach is the vigorous process of setting hypotheses, of empirical experimentation to test these hypotheses, of deep analysis to measure the results, and then the ability to codify the results in a set of laws and predictions. Claiming for themselves, in this sense, a parallel certainty of laws and predictions as and laws demanded by the natural sciences, positivism reveals to the social sciences phenomenal objects as they really are — as they are when stripped of superstitions, fallacious theories, prejudice and so on. Positivism demands a definite residue of facts and ‘truths’ that are universally applicable to social groups and communities irregardless of time, place or environment. In striving so vigorously for such ideals, positivism gives the social sciences a high degree of authority and respectability within the wider scientific and academic community as a whole. Further, a positivist approach in the social sciences affords a ready means of comparison and exchange of knowledge between other disciplines such law, philosophy, literature and so that employ positivism also. Indeed, in seminal respects, such is the importance of positivism for the social sciences that it is difficult to see how they could justify being ‘sciences’ without it.
The two principal disadvantages of a positivist application to the social sciences are these: firstly, that its search for ideal and perfect standards of scientific methodology and analysis are too unrealistic when set beside the extreme complexity of social phenomenon; the second weakness, is positivism’s lack of empathy and consideration of the subjective, individual and hermeneutic aspects of social phenomenon. Dealing with the first objection, critics of positivism argue that it cannot — working as it does in the outside world, in cities and in companies, in villages and mass organizations — attain the same standards of empirical excellence, either in experimentation or in verification of results, as can natural scientists working in the controlled conditions of a laboratory and deriving principles mostly from inanimate matter of slighter sophistication than human beings. Moreover, social scientists have a nearly insuperable difficulty in codifying laws of social phenomena with the precision that physics or chemistry allow for material phenomena. Thus positivism in the social sciences attains a lower level of prediction and accuracy with respect to the phenomenon it observes, than do the natural sciences. The second major weakness of a positivist application is its failure to take sufficient account of the subjectivity of individual life and to interpret the meaning of that phenomenon for the subject and the community of the subject. On these matters positivism has nearly nothing to say, and thus it is barred from a whole hemisphere of human social experience.
As the first sentence of this conclusion suggested: neither an extreme positivist not an extreme subjective or hermeneutic attitude can dominate the future of the social sciences. Rather, social scientists must learn to join positivism with subjectivism, thus fusing the two halves of social phenomenal experience. If positivism can be brought into union with the subjective in the social sciences, and if positivists can learn to tolerate something less than perfection in their methodological approach, then positivism must still be said to have a large contribution to make to the future of social science. In might be said then, in our final words, that positivism is simultaneously an advantage and disadvantage for the social sciences; whether one or other of these qualities is dominant remains to be seen.
– – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – –
BIBLIOGRAPHY
— Dowding, K. (2001). ‘There Must Be An End To Confusion: Policy Networks,
Intellectual Fatigue, and the Need for Political Science Methods Courses in British
Universities, in Political Studies, Vol 1., pp. 89-105.
— Dowding, K. (1995). Model or Metaphor? A Critical Review of the Policy of
Network Approach. Political Studies, Vol. 45, Issue. 1, pp. 136-158.
— Green, D. P. & Shapiro, I. (1994). Pathologies of Rational Choice Theory : A
Critique of Applications in Political Science, pp. 89-95. New Haven, London.
— Gerrard, James. (1969). The Sociology of the Family, pp. 303-316. Ford Press,
Pittsburgh.
— King, G. (et al.). (1994). Designing Social Enquiry: Scientific Inference in
Qualitative Research, pp 201-208. Princeton University Press, Princeton.
— Hugh-Jones, Steven & Laidlaw, James. (2000). The Essential Edmund Leach,
p163. New Haven, London.
— Marsh, David & Smith, Martin. (2001). ‘There Is More Than One Way To Do
Social Science: On Different Ways To Study Political Networks’ in Volume 49,
Number 3, pp. 528-541.
— Marsh, David & Furlong, Paul. (2002). ‘A Skin Not a Sweater: Ontology and
Epistemology in Political Science’ in Marsh, David and Stoker, Jerry (Eds.).
Epistemology in Political Science, pp. 17-41. Palgrave, Basingstoke.
— Popper, Karl R. (1983). Realism and the Aim of Science, pp 1-13. Routledge,
London.
— Popper, Karl R. (1989). Conjectures and Refutations: the Growth of Scientific
Knowledge, 69-76. Routledge, London.
— Quirk, Randolph (et al.) (Eds.). (1989). The Oxford English Dictionary. Oxford
University Press, Oxford.
