The Logic of Scientific Discoveries - Thesis Example

? Critical review of Karl Poppers’ The Logic of Scientific Discoveries Introduction Karl Popper is characterizes as “the most prominent” proponent of “logical decuctionism” in the 20th century (Turner 2006, 343). Popper’s The Logic of Scientific Discovery first published in 1934 has emerged as the 20th century’s model for deductionism (Turner 2006, 343). Essentially, Popper’s deductive theory sets out to draw distinctions between theoretical assumptions and empirical analyses. In short, Popper rejects the view that empirical analyses are used to prove or disprove a theory, but rather puts forth the view that empirical analyses are used to validate or justify a theory. The scientific truth is revealed when a theory is proven false under Popper’s famous falsificationism doctrine. Drawing on Popper’s deductive theory and falsificationism doctrine, Simon (1973) argues that Popper is essentially stating that “scientific discovery has no logic” (471). In other words, despite the Popper’s title, The Logic of Scientific Discovery, Popper argues quite the opposite proposition throughout his book. The title of the book implies that there is logic associated with scientific discovery. However, a close reading of the text reveals that there is little or no logic to scientific discovery. Scientific discovery commences with random suppositions which are tested so that those that can withstand rigorous tests to refute these suppositions are the end result of scientific discovery. As Shah (2008) puts it: When Popper refers to the logic of scientific discovery, he uses the term ‘discovery’ as a success word, implying that a discovery is something that has already survived critical refutation (303). This paper provides a critical analysis of Popper’s the Logic of Scientific Discovery and focuses more intently on the concept that the book’s title is misleading. A brief summary of Popper’s theory is presented and this is followed by a critical analysis of Popper’s theories and assertions relative to Popper’s approach to scientific truths in the context of theory and testing. The idea is to demonstrate that Popper essentially holds fast to the idea that logic has no place in scientific discovery and how that proposition is logical. I. Brief Summary Essentially, Kopper’s The Logic of Scientific Discovery tackles the problem of induction. Kopper argues that it is not possible to prove scientific theories, but rather those theories can only be tested and then corroborated. What distinguishes inquiries into scientific theories is the fact that they are capable of being tested or in Popper’s own words, scientific theories can be investigated by reference to the falsifiability implicit in their theories (Popper 2002, 57). The distinction therefore arises since an unfalsifiable theory is not scientific because it is not capable of being tested (Popper 2002, ch. 4). Popper’s The Logic of Scientific Discovery is therefore a methodological theory. It sets out Popper’s theory on the form that scientific discovery takes. For Popper this form is not a naturalistic or sociological methodology. What amounts to science is ultimately tradition or decisive. Empirically, tradition or decisiveness is a product of experience and is not capable of naturalistic explanation (Popper 2002). Therefore, while Popper is in fact discussing a scientific philosophy, although science is a natural phenomenon, Popper manages nonetheless, to separate it from conventions associated with positivist or empiricist convention. The empiricist or the positivist, unlike Popper, do not emphasise the significance of decisiveness or tradition. Cumulatively, Popper’s theory of scientific discovered culminates in the view that science is the result of a collection of knowledge founded on falsifiability and is perpetually vulnerable to modification (Popper 2002). Just how Popper presents and rationalizes this theory of scientific discovery is analysed below. II. Analysis of Popper’s The Logic of Scientific Discovery A. The Problem Associated with Induction Early on, Popper (2002) warns his reading audience that he intends to oppose the “widely accepted view” that “the empirical sciences can be characterized by the fact that they use inductive methods” (3). Popper (2002) goes on to explain: According to his view, the logic of scientific discovery would be identical with inductive logic, i.e. with the logical analysis of these inductive methods (3). In other words, empirical research studies in the sciences amongst the positivists approaches, according to Popper represented the conventional methods. To this end, the sciences were dependent on a vague notion, otherwise referred to by Popper (2002) as “inductive logic” (3). However, the opposition to inductive logic as raised by Hume required some attention. Ultimately, Hume argued that inductive logic erroneously required the substantiation of on fact by reliance on general truths (Popper 2002, 4). Popper (2002) explains that in general an inference is referred to as inductive when it transforms a sole assumption of fact like the outcome of “observations or experiments”, to theories (3-4). Logically, drawing an inference of universal import from a single observation or experiment could end up being misleading. As Popper explains: No matter how many instances of white swans we may have observed, this does not justify the conclusion that all swans are white (4). The problem of induction is therefore, inquiring whether or not an inductive inference can be justified and determining the circumstances in which those inductive references can be justified. Put another way, the induction difficulty asks about the veracity or validity of an all-inclusive assumption grounded in experience, “such as the hypotheses and theoretical systems of the empirical sciences” (Popper 2002, 4). Put another way, Popper defines the problem of induction as erroneously purporting to reduce collective observations or experience into a single statement of fact. The problem however, arises because although it can be concluded that A is followed by B, there is no guarantee that the next time a similar experience or observation is conducted, that the result will be the same. Regardless of how frequently, our observations or experiences produce the same result, there is always the possibility that a different outcome can be produced later on. There is no denying the merits of Popper’s position with respect to the problem of inductivism. Certainly, if we repeatedly see swans and from our repeated observations and experiences, swans are white, there is no guarantee that all swans are white. There may very well be swans of another colour out there. However, in the absence of evidence to the contrary, are we not entitled to assume that all swans are white? Or are we justified in rejecting a generalization on the off-chance that swans are indeed of different colours and we simply have not been able to observe them? Popper’s identification of the problem with induction appears to suggest, that scientific discovery regardless of logical processes of elimination, are flawed since they do not deliver precise and irrefutable outcomes. However, Popper’s argument is flawed as it borders on inviting the consideration of a mere possibility to refute what amounts to an entirely strong probability. Moreover, Popper is ignoring the significant role that history, society and culture plays in the building and decimation of knowledge and information (Willig 2002, 4). The fact is, history, society and culture cultivates and decimates information and knowledge upon which generations accept without question. If we were to demand irrefutable proof of this knowledge and information, experience and observations would be reduced to purely personal events with no educational value. Observations and experiences from history, culture and society are accepted without question because mankind necessarily accepts that not every assumption is an exact science. The same assumption is applied to scientific discovery or experiences. Scientists generally accept what is capable of observation and measurement. That which cannot be or has not been observed is not taken into account. The fact is, anything is possible. The real test is determining what is probable and improbable. If mere possibilities were to direct our acceptance and rejection of facts, we would be left with very few accepted facts and assumptions. This is the basis of empirical research studies. We observe and measure and then we formulate logical hypotheses based on these experiences of fact or deliberate experimentation (Willig 2002, 4). Nevertheless, Popper insist that basing empirical studies on induction is problematic as it does not sufficiently account for rational scientific information. The truth may be grounded in generalizations deduced from the truth attached to particular experiences and observations. Popper, fortified in this belief, suggests that there is an alternative method for overcoming the difficulties he perceives with induction. Instead looking for firmly established inductive evidence, it is important to realize that for the purposes of demarcation and finding truths, falsification would substantiate that which inductive verifications cannot. B. Falsifiability as a Solution to Inductivism Popper (2002) explains that his primary reason for rejecting induction is based on the fact that it: Does not provide a suitable distinguishing mark of the empirical, non-metaphysical character of a theoretical system; or in other words, that it does not provide a suitable ‘criterion of demarcation’ (11). According to Popper, the problem of demarcation is therefore the difficulty with identifying a model for enabling a distinction among empiricism and mathematic and logic and the metaphysical. For Popper, falsifiability is the solution to the demarcation question because is makes provision for a deductive outcome and explaining scientific process in a “non-jusficationist” basis (Garcia 2006, 9). Popper notes however, that his proposal to resort to falsifiability is destined to be objected to (Popper 2002, 57). Indeed philosophical scholars have expressed the view that Popper did not in fact solve the problem of induction, but merely reinserted it in a back-handed manner (Garcia 2006, 9). Others suggest that Popper’s dismissal of induction is flawed because he ignores the fact that predictions made in science have enjoyed significant success (Garcia 2006, 9). In order to appreciate these criticism and Popper’s own anticipation that these criticism would come, a closer examination of Popper’s falsifiability proposal is warranted. Popper asserted that any hypothesis or theoretical assumption takes on a scientific nature when it is capable of falsification. When a statement cannot be falsified, it is not scientific in nature, although it could have some relevance. Essentially, Popper provides several definitions of falsifiability in his book The Logic of Scientific Discovery. Essentially, falsification is presented as material by Poppers. It has the potential to categorize theories that might coincide with specific circumstances. When theories coincide with circumstances, the theories are rendered falsified. Falsifiability therefore conveys the information contained in a theory. The higher the degree of falsifiability in a theory, the more valuable it is as a source of information and the more suitable it is for formulating theories in the context of empirical studies in science (Popper 2002, Ch. 4). Popper (2002) explains the value of falsification as follows: If the degree of falsifiability is increased, then introducing the hypothesis has actually strengthened the theory: the system now rules out more than it did previously: it prohibits more (62). Popper’s explanation of the value of the falsifiability solution to the problem of induction raises an issue. If falsifiability seeks to eliminate possibilities and probabilities, how can it help scientists or any theorists arrive at the truth? Surely, falsification only helps in terms of eliminating possible truths and as such does not miraculously identify solutions or outcomes. Popper (2002) explains this enigma as follows: The introduction of an auxiliary hypothesis should always be regarded as an attempt to construct a new system: and this new system should then always be judged on the issue of whether it would, if adopted, constitute a real advance in our knowledge of the world (62). In other words, Popper sees the falsifiability principle as a method for not only eliminating possibilities and possible truths but a method for directing further research and the introduction of new theories for further hypotheses testing. From Popper’s perspective, all theories are essentially suppositions as are any predictions they claim. Both are therefore capable of falsification. However, falsifiability is indiscriminate in that all theories and their predictions have different difficulty levels in terms of falsification. Scientists should target those suppositions that have a higher degree of difficulty because the more difficult or bolder prediction or supposition will lead toward are a greater degree of discovery once that supposition or prediction is not capable of falsification (Popper 2002, Ch. 4). In other words, Popper is expressing the view that falsification is as much about deduction as it is about discovery. In taking this view, Popper is rejecting the induction theory of scientific discovery. For Popper, rather than merely infer truths on the basis of observation and observances, scientists should strive toward a more exact deduction of the truth. This is accomplished by deduction or elimination of assumptions or pure conjecture and accepting as fact those things that cannot be falsified. However, this brings us back to the induction approach which is attributed to the positivist methods for scientific discovery. Accepting as true, that which cannot be falsified is no different from accepting as the induction schools of thought the fact that those things that are capable of observation are representative of the truth. The falsification theory immediately brings to mind the swan example used by Popper to discredit the induction theory. It sounds remarkably similar to Popper’s falsification principle. If only white swans are seen, then it can be inferred that all swans are white. Certainly this is no different from the falsifiability principle which assumes that anything that cannot be falsified is tantamount to a scientific discovery. Certainly the induction proponents are accepting that since it cannot be proved that there are swans of a different colour then certainly, one must accept that all swans are white. In other words, induction maintains that the supposition that all swans are white cannot be falsified. It therefore follows that induction is in reality no different from Popper’s falsification theory. Popper’s corroboration theory offers perhaps an explanation for this apparent correlation between falsification and induction. III. Corroboration According to Popper, “theories are not verifiable but they can be ‘corroborated’” (Popper 2002, 248). Clearly, Popper wants to set his approach to scientific discovery and research apart from conventional scientific research and discovery. Conventional or positivist approaches to scientific discovery typically look to empirical research studies or experience or observation to verify a theoretical assumption. As we have already seen, falsifiability principles put forth predictions or suppositions. When these predictions and suppositions do not pan out, then a negative theory is confirmed or essentially corroborated. Corroboration therefore functions to ascertain how a theory performs under testing in a particular space and time. It does not propose to predict how that theory will perform in a different space and at a different time. The assumption is always that the theory does not guarantee that it can stand up to another test (Popper 2002, Ch. 10). For Popper, the primary concern is explanation. This is true whether the outcome is corroboration or falsification. Mankind naturally gravitates toward a theory that is better corroborated because those theories have a tendency to provide a better explanation of some faction of realism. Moreover, a corroborated theory will typically not coincide with our own experiences and observations in those fields where the forbidden fact does not occur. In any event, when scientists or researchers in general are forced to choose from among conflicting theories, they will automatically select those theories that have a greater degree of corroboration unless those theories are not capable of being tested (Popper 2002, Ch. 10). According to Popper it is not possible for a statement purporting to support a probability relative to a hypothesis to amount to a claim that events are probable. In this regard, Popper puts forth a “positive theory of corroboration” (Popper 2002, 265). The concept of corroboration is a theory that stands up to rigorous tests that are designed to contradict it. The more frequently a theory is tested and the more rigorous a test, the greater the corroboration when the theory stands up to the testing. However, Popper acknowledges that it is not easy to ascertain the terms and conditions for testing or the extent to which a theory is corroborated. It is therefore impossible to anticipate a methodology that will provide a failsafe exercise (Popper 2002, Ch. 10). Ultimately, the sensible and logical approach to take in terms of ascertaining the degree of corroboration is to accept that all “knowledge” is bases on hypotheses (Popper 2002, 268). Therefore the question of not really knowing in advance what natural laws will persist in the future is substituted by the question of distinguishing between theories that are good and those that are bad. This is ultimately determined by reference to the positive theory of corroboration (Popper 2002, Ch. 10). Popper’s positive theory of corroboration aims at demonstrating the extent to which corroboration does not produce a probability but rather how it is inconsistent with probabilities. For Popper, the extent to which there is corroboration corresponds with the theory’s composition and the test outcomes. Corroboration is therefore a documentation of the veracity of previous performances. However, a theory might be able to predict events in the future, a documentation of previous performances is not a reliable source for prediction performances in the future. As Popper (2002) explains: The corroboration which a theory has received up to yesterday is logically not identical with the corroboration which a theory has received up today (273). Even so, it is not irrational to take action based on that specific theory which withstood testing to a greater extent than other theories (Popper 2002, Ch. 10). For Popper, these factors are important for embracing the idea “of content and of degree of corroboration” as it is the most logical tools” (Popper 2002, 395). Popper noted earlier on in The Logic of Scientific Discovery that: Once a hypothesis has been proposed and tested, and has proved its mettle, it may not be allowed to drop out without “good reason”. A “good reason” may be, for instance: replacement of the hypothesis by another which is better testable; or the falsification of one of the consequences of the hypothesis (51). In this statement the reader can appreciate the role of the greater degree of corroboration. Essentially, what this means is that no scientific discovery is written in stone. A different time and space may command different results with a greater degree of corroboration. In this regard, a corroborated theory today may be replaced by another and conflicting theory or hypothesis with a greater degree of corroboration at a later date and in a different place under different conditions. As Popper (2002) explains, a theorist may be compelled to look for an improved theory of “the experimental falsification of a theory, so far accepted and corroborated” (106) Popper (2002) puts the significance of corroboration in perspective as follows: We choose the theory which best holds its own in competition with other theories; the one which by natural selection, proves itself the fittest to survive. This will be the one which not only has hitherto stood up to the severest tests, but the one which is also testable in the most rigorous way (106). This approach to corroboration certainly finds currency in any field of research. In any research project where a systematic review of existing empirical research is conducted, the researcher will be expected to select the study with the more comprehensive and robust tests. A newer and more rigorously tested theory will be favoured over an older and less rigorously tested theory. However, the older theory may remain relevant if the newer theory later proves to be false. As Popper explains: A theory which has been well corroborated can only be superseded by one of a higher level of universality; that is, by a theory which is better testable and which, in addition, contains the old, well corroborated theory – or at least a good approximation to it (274). In other words, the newer theory must contain a good degree of correlation to the older theory in order to replace it. It must be based on quite similar conditions and circumstances. The newer theory must also be more reliable in terms of corroboration than the older theory. Otherwise it will not replace a good corroborated theory. Conclusion In Popper’s The Logic of Scientific Discovery, Popper took an approach that attempted to bypass as far as it was possible to, the notion of truth and untruths in terms of hypotheses or theories. For Popper the concept of truth was difficult and that scientific discoveries could be obtained without reliance on absolute or mathematical truths. Ultimately, Popper was arguing for an innovative concept of testing hypotheses and theoretical assumptions. In doing so, Popper set out to reject the induction principle. Specifically, Popper set out to assert that rather than canvas for the balance of probabilities relative to a theory or hypothesis, one ought to look instead to how a supposition stood up under testing. The reality is however, although Popper presents a different theoretical method for hypothesis testing, it is in principle no different from the induction approach. Popper’s deductionism approach invites the consideration of falsifiability while induction or positivists invite the consideration of observable facts. The induction theorists does take account of negative facts or falsification, such as the absence of non-white swans. Similarly, the deductionist approach takes account of positive facts in the sense that it looks for corroboration. When all is said and done, each theoretical approach to scientific discovery borrows one from the other so that each looks for outcomes in a logically systematic way. Essentially, both approaches to scientific discovery accept and acknowledge that there are no absolute truths and some form of analysis of existing facts is necessary. Whether this is in the form of induction or deduction is a matter of preference. Neither approach will produce an absolute theory, but both are based on logical reasoning in one form or another. Bibliography Garcia, C. Popper’s Theory of Science: An Apologia, Continuum International Publishing Group, 2006. Popper, K. The Logic of Scientific Discovery, Psychology Press, 2002. Shah, M. “The Logics of Discovery in Popper’s Evolutionary Epistemology,” Journal for General Philosophy of Science, (2008) 39(2): 303-319. Simon, H. “Does Scientific Discovery Have a Logic?” Philosophy of Science, (Dec. 1973) 40(4): 471-480. Turner, J. Handbook of Sociological Theory, Springer Publications, 2006. Willig, C. Introducing Qualitative Research in Psychology: Adventures in Theory and Method, Open University Press, 2002. Read More