Positivists regarded empirical observation freed of preconceptions as the means by which facts were obtained and explained. This view, however, has been greatly contested since the Vienna Circle’s avid pursuance of it. The main problems include its inability to be checked and criticised by the scientific community members. In other words, they are subjective, fallible and thus unreliable.  It is this initial discontent with positivism, especially with logical positivism which prompted Karl Popper to develop his Theory of Falsifiability, a theory which no longer relies on induction but on deduction, which accepts that truth is not attainable and which casts theories aside which have been refuted by only a single piece of empirical evidence. Falsification is also a demarcation between science and non-science, something which has proved to be very controversial. Thomas Kuhn, perhaps the most well known critic of Popper’s work, does not believe in induction or deduction as methods through which science progresses. Instead, he introduces the concept of normal science, revolutionary science and paradigms. The differences between these two men’s work will be analysed, the implications of each for the conduct of social sciences commented upon and the work of Imre Lakatos, a twentieth century philosopher of mathematics and science, highlighted in order to illustrate just how much both philosophers resonate in the social sciences as a whole.
Karl Popper, Positivism and his Theory of Falsifiability
Karl Popper was first and foremost a philosopher of the natural sciences,  his knowledge of the social sciences being limited basically to economics.  With that in mind, one understands why he agreed with Rudolph Carnap in advocating that philosophy should learn from how the natural sciences operate. He believed scientists should adopt a critical attitude, willing to incessantly test their views with empirical evidence and rational discussion which the Vienna Circle had so avidly promoted. However, Popper was soon to highlight flaws with positivism, especially with logical positivism. These were, in particular, its dedication to the principles of ‘inductivism’ and ‘verificationism’.
Inductivists claim that via induction, one is able to obtain secure scientific knowledge and that the inference is legitimate if a ‘significant number of singular or observational statements are gathered under a wide variety of circumstances’.  In order to maintain the empirical certainty of inferences obtained through the deductive method, ‘the universal law premise must be empirically certain’.  However, as Popper pointed out, one has no assurance that any universal empirical proposition is certain. For Popper, the ‘problem of induction’ was insurmountable, contesting that if science is empirical ‘its […] laws must be treated as tentative hypotheses’. 
Popper accepted the Humean critique of induction, claiming not only that it is never used by scientists but that observation, believed to be an initial step in the formulation of theories, is misguided  . Hume also pointed out that observation is selective and theory-laden and thus one can never make pure or free observations.  Popper, however, disagreed with Hume over whether knowledge could be rationally justified. Hume saw inductively inferred laws as ‘merely an account of habit or custom, (suggesting that) even scientific knowledge is irrational’.  Popper, on the other hand, in order to avoid statements allowing empirical evidence to confirm false theories, believed that induction could be replaced by deduction. Deduction ‘draws inferences about the premises from the observed falsity of the conclusion’.  To justify this, he argued that though even with a body of empirical evidence, one can never be absolutely certain about the validity of a theory, it takes only one empirical rebuttal to determine the falsity of a theory. Popper denominated this ‘the asymmetry between verifiability and falsifiability’  , a difference which became the centrepiece of his philosophy of science.
Scientists begin with universal statements and initial conditions from which they deduce hypotheses which will then be subsequently tested. If they withstand the test, the theory will survive; if falsified, the theory is abandoned. Falsifiability, according to Popper, is the criterion of demarcation between science, or the ’empirical sciences’ and the non-science. There are, however, degrees of falsifiability. The more information a statement contains, the larger its body of observational statements and therefore, the higher its degree of falsification. According to Popper, scientists should aim at highly refutable theories instead of modestly falsifiable ones. It is preferable for the theory to be bold, precise and simple  as their empirical content will be greater and therefore there will be a larger body of potential falsifiers.
Popper’s definite break with logical positivism appears in their search of certainty: the positivists aimed to ‘specify methods that would generate certain knowledge’  whilst in Popper’s view, one can only ‘hope to improve what must always remain imperfect’  as future tests could cast doubt over what was previously thought of as true. If we take Popper’s approach to the search for truth, it would initially appear that there are an endless number of possible true theories.  However, Popper addresses that by explaining his notion of ‘verisimilitude’. The scientific process of trial and error which Popper advocates creates a greater approximation of the truth, or increases the ‘verisimilitude’ of the theory.
Popper, Marxism and Pseudo-Science
The young Popper had been attracted to the apparent strength of theories such as Freud’s psychoanalysis and Adler’s ‘individual psychology’. These theories were regarded as capable of explaining virtually everything related to human behaviour as verifications were found to justify every advancement. Popper, however, was soon to discover a major flaw in them: they could not be refuted. Freud was therefore severely criticised by Popper for producing immunised theories against falsification. A theory unable to be falsified belongs, in Popper’s view, to a non-science. His drastic approach towards pseudo-science was also extended to Marxism, especially the Marxism that Neurath had brought to the Vienna Circle.
Neurath interpreted Marxian ‘materialism’ as epsitemically equivalent to his own ‘physicalism’  and hailed Engels and Marx as having produced the foundations of a truly scientific study of society. In Popper’s opinion, this could not have been further from the truth. The problem with Marx was not only that he was considered a historicist, but that he was a utopian too.
Marxism, initially considered as a science because of its predictive nature, was soon re-classified as fundamentally non-scientific. The predictions Marx had made had not been borne out and in order to save it from falsification and refutation, ad hoc hypotheses were added, making the theory compatible with facts. These factors prompted Popper to adopt falsifiability as his criterion for demarcation between science and non-science. If a theory, according to Popper, is capable of being falsified or, in other words, is incompatible with empirical evidence, it is considered as scientific. If, on the other hand, a theory is compatible with all observations and is capable of explaining virtually everything be it because, as with the case of Marxism, it has been modified to accommodate newly made observations or, because, as in the case of psychoanalysis, it is indeed consistent with all observations made and to be made in the future, it is categorised as unscientific.  It is this criterion which characterises Popper’s theory of falsifiability and which was soon criticised.
Implications for the Social Sciences
Popper helps demolish one of the notions positivism embraces, namely that science progresses from the observation of data by means of experiments. These experiments are verified when repeated allowing general laws about the nature of reality to be inferred. Popper, therefore, shows that progress is made not by verifying facts, but by attempts of falsifying the results of other theories.  The theories of science, he argues, are conjectures to solve problems and cannot be verified by empirical evidence.  The switch from induction to deduction also means that rather than proceeding from the particular to the universal, science originates from the ‘universal (i.e. scientific hypotheses) to the particular’. 
Thomas Kuhn’s Theory of Paradigms
Thomas Kuhn began his career as a physicist and then turned his attention towards the history of science where his preconceptions about natural history were shattered  . His Structure of Scientific Revolutions (1962) was developed as an attempt ‘to give a theory more in keeping with the historical situation as (Kuhn) saw it’  . Unlike Popper, his main aim was not to provide guidelines to scientists about how to proceed or to develop a normative philosophy of science. The central concern of his thesis was to characterise the way in which science historically develops and to explain why scientists have operated in such a way.
Kuhn’s Structure of Scientific Revolutions has been ‘one of the most provocative (pieces of work) to appear in the last fifteen years’  , offering ‘the most sophisticated alternative to Popper’.  Science, in his opinion, does not progress inductively as positivists would maintain nor by falsification as Popper would argue. Alternatively, Kuhn places focus on the revolutionary character of scientific process, where a revolution involves the abandonment of one theoretical structure and its replacement by another, incompatible one.
Kuhn’s approach to the way science progresses can be summarised by the following open-ended scheme:
pre-science – normal science – crisis – revolution – new normal science – new crisis 
According to Kuhn, the pre-science stage is a disorganised and diverse activity preceding the formation of science. It eventually becomes structured, directed and channelled when a single paradigm emerges and is adhered to by the scientific community. As will be discussed below, the concept of paradigms itself has been subjected to heavy criticism, not least because of its ambiguous nature. However, vaguely, one can postulate that paradigms contain ‘some very general methodological prescriptions’  to guide scientific work. Paradigms also ‘serve a regulative function in directing future research.’  Workers within a specific paradigm whether it be Newtonian mechanics or wave optics practise what Kuhn denominates normal science.
As professed in Structure of Scientific Revolutions, normal science is:
research firmly based upon one or more past scientific achievements that some particular scientific community acknowledges for a time as supplying the foundation for its further practice. 
Normal science is inextricably characterised by a dominant paradigm, something that Popper quickly picked up upon as irrational and superficial.  In normal science the scientist’s work is ‘devoted to the articulation and wider application of the accepted paradigm’.  In other words, their main aim is to ‘fill out what is suggested by the accepted paradigm.’  It is therefore clear that little emphasis is placed upon normal science and research to produce major novelties as a primary aim.
Kuhn, in effect, reduces Popper’s falsification theory to problem solving within the confines of normal science. According to Kuhn, science is merely a set of puzzles whose solutions are to be found within the operating paradigm  . Normal scientists do not actively look for anomalies which the content of their paradigm will be hard-pressed to solve. However, when a recurrent anomaly does arise which the paradigm is unable to resolve, crisis will break out.
During such a crisis, ‘extraordinary science’ occurs characterised by a plurality of views and a challenge to the fundamentals of the paradigm. The crisis will then be resolved when a completely new paradigm emerges which has the capacity to resolve the previous, problematic anomalies and, in doing so, attract the allegiance of a growing scientific community until eventually the paradigm posing the problem is abandoned. Therefore, the new paradigm not only has to be able to resolve the anomaly, it also has to be subsequently accepted as normal science, thus establishing a new consensus. A scientific revolution according to Kuhn is constituted by ‘discontinuous change’  as the newly adopted paradigm will be confronted with problems it is unable to resolve and thus the never-ending cycle continues.
Kuhn and Popper
The Structure of Scientific Revolutions soon became problematic to reconcile with Popper’s theory of falsification as Kuhn’s historical account about how scientists operate came into conflict with Popper’s work. The emphasis Kuhn placed on scientific communities, their rules and expectations, was used to explain why scientists were not always willing to refute and actively search for falsifications of their theories. Unlike what Popper claimed, the scientific communities would not question the paradigm they work within until a particular anomaly was repeated. Instead, they might question their own calculations or instruments implemented, but never the broader framework they operate within. 
Popper’s reaction to Kuhn’s severe criticism was veritably weak. He simple asserted that Kuhn’s accurate historical account of science ‘clashes with the facts as I see them.’  According to Kuhn, falsification has not been in use in the past for the reasons highlighted above. Popper’s rebuttal to this was that he, unlike Kuhn, had not focused on providing a historical account but on providing guidelines for future scientists. He also criticises Kuhn for producing a highly selected theory, one which ‘disregarded large chunks of ‘normal science.’ ‘ 
Popper also criticised Kuhn for ‘paving the way for irrationalism and relativism,’  the reason for this lying in two of Kuhn’s statements. Firstly, the fact that Kuhn equated the switch in paradigms to a ‘gestalt switch’ or a religious conversion because he believed in a ‘holistic theory of meaning’  means that it is very difficult to compare scientific theories. Secondly, because of Kuhn’s cynical approach to ‘verisimilitude’ and his belief that we never get closer to the truth, his explanation on how science progresses seems ill-founded. In these contexts, Popper criticises Kuhn of adhering to the ‘myth of framework’ which presupposes that rational and critical discussions can only take place if fundamentals are agreed upon. Popper strongly disagrees with this concept, as with the belief that science will not progress across paradigms and argues that different frameworks always have enough in common to allow the scientific community to compare and judge them, triggering progress.
Other Criticisms of Kuhn’s Work
Popper has not been alone in criticising aspects of Kuhn’s Structure of Scientific Revolutions. Bernstein succinctly puts the majority of the criticisms in his The Restructuring of Social and Political Theories  . The ambiguous nature of paradigms and the irrationality of the paradigm shift have been discussed above. Critics have also pinpointed Kuhn’s misrepresentation of the history of science  , the inaccurate description of normal science  and the exaggerated distinction between normal and revolutionary science  . In order to further show the relevance of Kuhn’s work to the social sciences, the vagueness of paradigms will be discussed, as the irrationality of paradigms has been explained above.
When first introduced, Kuhn claimed paradigms were ‘universally recognised scientific achievements that for a time provide model problems and solutions to a community of practitioners.’  However, the ‘elusive’ and ‘slippery’ concept  of paradigm was shown when Kuhn acknowledges that he had been using the term paradigm in many ways  , citing Masterman who listed at least twenty two senses in which the term was used in the book.  To tackle the confusion created by his multiple use of paradigms, Kuhn proposes to replace it with a disciplinary matrix.  A disciplinary matrix includes the ‘shared commitments of the community of scholars, the shared symbolic generalizations and […] the shared problems and solutions in the discipline’. 
Other Implications for Social Scientists.
Even though Thomas Kuhn has been almost exclusively concerned with the natural sciences, social scientists have repeatedly claimed his work ‘offers fresh illumination for understanding social sciences and theory.’  Part of the reason for this is, as Kuhn himself pointed out, the fact that his work is ‘regretfully […] too nearly all things to all people’. 
The revolutionary transformation in the use of observation as a means leading to theory has also had an impact on social scientists. Kuhn’s starting point for the formulation of theories is not reality but construction.  Kuhn ‘contributed to demolishing […] positivism’  not only by admitting revolutions in science involve the intrusion of non-scientific elements such as habits, customs or cultural values, but also in casting doubt over the possibility of attaining perfect knowledge and over the established idea that progress in science is cumulative.
One of the most important consequences Kuhn’s work had for the social sciences was the significance he attributed to the role played by the sociological characteristics of scientific communities. Kuhn turned away from the search for an ideal methodology to the study of science by scientific means and, in doing so, invigorated the empirical study of science.  Finally, Kuhn may have hastened the demise of positivism by prompting and then influencing the naturalisation of epistemology, a movement which has become prominent through, for example, a conventionalistic and naturalistic study of science.
Imre Lakatos: A Middle Man?
By the late 1960s a great deal of the debate on the philosophy of science had come to focus on the difference between Kuhn’s paradigms and Popper’s revision of positivism.
Numerous epistemic doctrines entered this debate and different interpretations of Popper and Kuhn’s works emerged, reflecting the impact they had on their contemporary critics and their effect on the conduct of social science as a whole. Lakatos is one of the most prominent critics of their works, his critique generally considered as ‘the most important attempt to place the post-empiricist theory of science somewhere between Popper and Kuhn’. 
Imre Lakatos at the outset appears to be a supporter of Popper’s falsification theory. He strongly criticised Kuhn for his ‘irrationalist and too general’  concept of a revolution and his notion of a single, dominating paradigm. Lakatos defends Popper against the charge of ‘naive falsificationism’, the immediate discarding of a theory as soon as contradictory evidence is exposed. However, he goes beyond Popper in claiming that science progresses by ‘sophisticated falsification’ which focuses on the comparative evaluation of whole research programs. 
Sophisticated falsificationists realise that the ‘conditions that a hypothesis should satisfy in order to be worthy of a scientist’s consideration […] alone are insufficient’  and that the need for a hypothesis to be more falsifiable than the other it will replace is necessary for scientific progress. Thus, it is not single theories which are falsified but entire programs, embodying the notion of ‘refutation […] not automatically lead(ing) to rejection’.  Such an epistemic theory strikingly resembles Kuhn’s theory of paradigms. The difference between them only appears when closely examining Lakatos’ notion of ‘research programs’.
According to Lakatos, every scientific research program has a ‘hard core’, a ‘set of propositions that are immune from empirical tests’  because they are surrounded by a ‘protective belt’ of ‘assumptions or conditions’.  Though research programs and paradigms have been equated, Lakatos proposes that normal science be considered more as a research program for ‘reasons of its general acceptability’  and does not attribute the general status to it that Kuhnian paradigms have. Furthermore, the transition from one research program to another is the product of ‘rational exploration of rival methodologies’  and not, as Kuhn implied, a ‘mystical conversion’  to a new ontology. Cultural values, historical events and other external factors are far less important in Lakatos’ eyes and play little part in particular scientific theories or the choice of general research programs, levelling Kuhn’s theory ‘down to critical rationalism’. 
Lakatos’ MSRP has not emerged without enticing criticisms. Firstly, he seems to have physics exclusively in mind when he developed his theory and when referring to ‘science’. Other natural sciences ‘cannot as easily be accommodated to the Procrustean bed of the MSRP’  and it is only economics which seems to offer the possibility of an easy fit.  As a model for the history of science, MSRP ‘fails to meet the empirical test of general acceptability’  . It is also limited in explaining how science works, failing to formulate the criteria needed to be employed for it to work. However, as Gordon highlights, the fact that Lakatos was flexible in not regarding former scientists as misguided in adopting theories that now would be considered irrational is a ‘significant point of merit in Lakatos’ epistemic stance’.  The MSRP model allows the possibility of ‘gaining knowledge by using theories that are subsequently regarded as, in the absolute sense, false.’ 
As Lakatos claimed,
The clash between Popper and Kuhn is not about a mere technical point in epistemology. It concerns our central intellectual values, and has implications not only for theoretical physics but also for the underdeveloped social sciences and even moral and political philosophy. 
As seen with Sander’s account, Popper has greatly influenced the political sciences, contributing to xxxx. Kuhn’s work, on the other hand, as Mark Smith rightly points out, has had a deep impact on the conduct of social sciences because of the vagueness and therefore adaptability of the term paradigms.  Despite their distinct approaches, however, both men have met with severe criticism, not only from each other, but from scientific colleagues and both have apparently failed to address these adequately.  It is therefore not surprising that xxxxxx