Critical rationalism

Last updated

Critical Rationalism is Karl Popper's answer to what he considered the most important problems of epistemology and philosophy of science: the problems of the growth of knowledge, notably by induction, and the demarcation of science. He adopted a fallibilist approach to these problems, especially that of induction, without falling into skepticism. His approach was to put in perspective the distinctive role of deductive logic in the development of knowledge, especially in science, in the context of a less rigorous methodology based on critical thinking. The central technical concept in the application of critical rationalism to science is falsifiabiity. Popper first mentioned the term "critical rationalism" in The Open Society and Its Enemies (1945), [1] and also later in Conjectures and Refutations (1963), [2] Unended Quest (1976), [3] and The Myth of the Framework (1994). [4]

Contents

Fallibilism, not skepticism

Popper admitted that the truth of statements cannot be obtained using only logical definitions and deductions, as this leads to an infinite regress. [5] For Popper, this does not prevent statements from being useful for solving problems, because they can be logically analyzed to draw logical consequences, possibly contradictions with observation statements linked to real tests. [5] Popper wrote that the bulk of scientific activities use deductive logic to evaluate theories. [6] [7]

Popper accepted Hume's argument and the consequences of Duhem's thesis and insisted that there is no logical method for accessing empirical truth, no inductive rule, not even to a small extent. However, he rejected skepticism, the idea that the search for truth is futile. He admitted that, although logic alone says nothing about empirical truth, statements can be related to reality through problem solving, scientific observations and experiments. [8] Popper has always insisted on this distinction between the logical aspect and the methodological aspect of science. [9] [10] In Realism and the Aim of Science, Popper speaks of a "preferred" theory rather than a "truer" theory when one theory is chosen over another following experimental results. [8]

Tarski's semantic theory of truth

Tarski inspired Popper with his semantic theory of truth. Alfred Tarski.jpeg
Tarski inspired Popper with his semantic theory of truth.

Popper was always aware that empirical truth eludes logic alone, and he was therefore reluctant to refer to the truth of scientific theories. This, he wrote, changed after reading Tarski's semantic theory of truth. He saw this theory as a way of talking about truth as a correspondence with facts. A key aspect of Tarski's theory, which Popper considered important, is the separation between the logical (formal) aspect of language as an object and its semantic interpretation. He saw this as a way of explaining the distinction between the logic of science and its methodology, or rather between logic and the metaphysical component to which methodology refers when it aims, for example, to test theories. The difference is that in Tarski's theory, "facts" are mathematical structures, not an external reality beyond the reach of logic and its language, and which we can only describe artificially in a meta-language as in the argument "Snow is white" (in the object language) is true because snow is white (in the meta-language). This use of Tarski's theory is accepted by some and sharply criticized by others. Popper used it, for example, in Realism and the Aim of Science, to explain the difference between the metaphysical versions of the problem of induction and its logical versions. He wrote that the metaphysical versions of the problem refer to the "meta-theory of physics" and compared this to what Tarski calls the "semantics" in its theory of truth. [11]

The role of methodology

For Popper, the bulk of activities in science use deductive logic on statements, [6] [7] but this logical part of science must be integrated within an adequate methodology. [12] The logical part is considered incapable of justifying empirical knowledge on its own. For example, Popper and the members of the Vienna Circle agreed that only statements can be used to justify statements, that is, the use of logic alone in science will not be linked to evidence. [13] [14] Logic uses accepted or provisionally accepted observation statements to determine whether a theory is logically refuted or not, but the "accepted" observation statement could be empirically false and that will not concern the logical part. [10]

Popper wrote that the bulk of scientific activity takes place in the logical part, using deductive logic to check the consistency of a theory, compare theories, check their empirical nature (i.e., falsifiability) and, most importantly, test a theory, which is possible only when it is falsifiable. He emphasized that, even when theories are tested against observations, deductive logic is largely used. [6] [7] Despite this intensive use of logic, Popper accepted, as do most philosophers and scientists, that logic alone does not connect by itself with evidence. Popper explained this dilemma by stating the existence of a natural separation (not a disconnection) between the logical and the methodological parts of science. [9] [10]

Lakatos described the aspects of scientific methodology leading to the rejection of research programs and their theories Professor Imre Lakatos, c1960s.jpg
Lakatos described the aspects of scientific methodology leading to the rejection of research programs and their theories

Popper wrote that any criterion, including his famous falsifiability criterion, that applies solely on the logical structure could not alone define science. In The Logic of Scientific Discovery , he wrote "it is impossible to decide, by analysing its logical form [as do the falsifiability criterion], whether a system of statements is a conventional system of irrefutable implicit definitions, or whether it is a system which is empirical in my sense; that is, a refutable system." [15] [16] Popper insisted that falsifiability is a logical criterion, which must be understood in the context of a proper methodology. [9] [10] The methodology can hardly be made precise. [17] It is a set of informal implicit conventions that guide all the decisions that surround the logical work, which experiments to conduct, which apparatus to build, which domain will be financially supported, etc., aspects that were raised by Lakatos in The Methodology of Scientific Research Programmes. [18]

Popper's philosophy was criticized as if the logical part existed alone. For example, Putnam attributed to Popper “the fantasy of doing science using only deductive logic”. [19] Putnam further criticized Popper's description of the logical part of science by referring to methodological problems. For example, he wrote "I claim: in a great many impor­tant cases, scientific theories do not imply predictions at all." [20] Because Popper does not believe in inductive logic, Wesley Salmon wrote that, for Popper, "there is no ampliative form of scientific argument, and consequently, science provides no information whatever about the future." [21] Regarding the methodological part, Feyerabend wrote that there is no method in science. He considered and rejected methodological rules, but they were those of a naive falsificationist. [22]

In contrast, Popper emphasized both parts of science and spoke of methodology as a means of correctly using falsifiability and the usual logical work in science to make it useful in a method of conjectures and refutations to be used in usual critical discussions. Falsififiability says hypotheses should be consistent and they should logically lead to predictions, which confrontation with observations should be considered in critical thinking. [23]

Marxism, politic and psychoanalysis

The death of friends in a demonstration instigated by the communists when he was about seventeen, strongly contributed to Popper's position regarding the search for contradictions or criticisms and the attitude of taking them into account. He blamed Marxism which thesis, Popper recalls, "is that although the revolution may claim some victims, capitalism is claiming more victims than the whole socialist revolution." He asked himself "whether such a calculation could ever be supported by 'science'." He then decided that criticism was important in science. [24] This, Popper wrote, made him "a fallibilist", and impressed on him "the value of intellectual modesty." It made him "most conscious of the differences between dogmatic and critical thinking." [25]

The failure of democratic parties to prevent fascism from taking over Austrian politics in the 1920s and 1930s traumatised Popper. He suffered from the direct consequences of this failure since events after the Anschluss (the annexation of Austria by the German Reich in 1938) forced him into permanent exile. His most important works in the field of social science— The Poverty of Historicism (1944) and The Open Society and Its Enemies (1945)—were inspired by his reflection on the events of his time and represented, in a sense, a reaction to the prevalent totalitarian ideologies that then dominated Central European politics. His books defended democratic liberalism as a social and political philosophy. They also represented extensive critiques of the philosophical presuppositions underpinning all forms of totalitarianism. [26]

Sigmund Freud developed the theories of "Psychoanalysis." Popper considered these theories to be unscientific because they were not falsifiable. Sigmund Freud, by Max Halberstadt (cropped2).jpg
Sigmund Freud developed the theories of “Psychoanalysis.” Popper considered these theories to be unscientific because they were not falsifiable.

Popper saw a contrast between the theories of Sigmund Freud and Alfred Adler, which he considered unscientific, and Albert Einstein's theory of relativity which sparked the revolution in physics in the early 20th century. Popper believed that Einstein's theory, as a theory properly grounded in scientific thought and method, was highly "risky," in the sense that it was possible to deduce consequences from it that differed considerably from those of the then-dominant Newtonian physics. [27] One such prediction, that gravity could deflect light, was verified by Eddington's experiments in 1919. [28] When he tackled the problem of demarcation in the philosophy of science, he realized that "what made a theory, or a statement, scientific was its power to rule out, or exclude, the occurrence of some possible events—to prohibit, or forbid, the occurrence of these events." [29] He thought that, in contrast, nothing could, even in principle, falsify psychoanalytic theories. This led him to posit that "only attempted refutations which did not succeed qua refutations should count as 'verifications'." [30]

A little later, Popper realized that theories can be "immunized" against falsification using auxiliary hypotheses. In Logik der Forschung, he introduced the notion of "(degrees of) content". He proposed that only modifications that increase the empirical content of a theory should be considered. [31]

In a series of articles beginning in 1979, Adolf Grünbaum argued, with examples, that Freudian psychoanalytic theory is in fact falsifiable. He criticized Popper's analysis of Freud's psychoanalytic theories and, on this basis, questioned the applicability of the demarcation criterion in general. [32]

Falsifiability, probability statement and metaphysics

Popper identified scientific statements with falsifiable statements and distinguished them from metaphysical statements. But, he considered metaphysical statements useful in science. In particular, probabilistic statements are non falsifiable and thus metaphysical in Popper's terminology. [33] There are many other kinds of metaphysical statements that are useful in Popper's view. For examples, "all men are mortal" is metaphysical, because it does not matter how much old is a man, perhaps he will die next year. Popper considered it useful, because it suggests other hypotheses that are more precise and more useful such as all men die under 150 years. [34] Similarly, probabilistic hypotheses suggest other hypotheses that are falsifiable such as the acceptation criterion for the null hypothesis in a statistical test. [35] A statistical hypothesis such a Chi-Square test is not a universal statement. It concerns a given study, but it is falsifiable and useful in critical discussions. A probability statement, such as "the probability of both head and tail are 1/2" is not falsifiable in the sense that it is compatible with any sequences of tails and heads. The problem of strengthening probability statements to make them falsifiable (incompatible with some sequences) and thus not metaphysical was named by Popper the problem of decidability of probability statements. [36]

Critical thinking, not support

Popper distinguished between trusting a theory because it is true and preferring a theory because it has been more severely tested. [37] Some have argued that, indirectly, Popper was adopting an inductive principle when he proposed to "prefer" a more severely tested theory. [38] [39] [40] For Popper, the term "induction" refers to a logical method of justification, and he emphasized that this preference does not result from such a logical process whose premises would be the results of rigorous tests. For Popper, results of rigorous tests are rather used in critical discussions. He wrote: [37]

[T]here is no 'absolute reliance'; but since we have to choose, it will be 'rational' to choose the best-tested theory. This will be 'rational' in the most obvious sense of the word known to me: the best-tested theory is the one which, in the light of our critical discussion, appears to be the best so far, and I do not know of anything more 'rational' than a well-conducted critical discussion.

Moreover, critical discussions must consider how much the theory prohibits and thus is unlikely to survive the tests, as well as whether the theory supersedes previous theories by generalizing them as when speaking of all heavenly bodies instead of only planets. [41] [42] Popper regularly emphasized that criticism in critical discussions requires the use of background knowledge, but rejected the view there will always be a set of assumptions beyond rational assessment. [43] In particular, Popper brought this point in the context of the empirical basis of science, which he compared to a swamp. [44]

The critical rationalism approach to evaluating scientific theories can be generalized to non-scientific domains. [45] Critical rationalists hold that any claims to knowledge can and should be rationally criticized, and, if they have empirical content, can and should be subjected to tests which may falsify them. They are either falsifiable and thus empirical (in a very broad sense), or not falsifiable and thus non-empirical. The general principle of critical rationalism is the same in both cases: we critically analyze the hypotheses using our "background knowledge". [46] In the case of scientific hypotheses, background knowledge is used while observation statements are discussed or analysed. [46]

Bayésianisme vs conjecture and refutation

Use of probability in a verificationist approach, "similar in some ways to that of modern pragmatists and positivists", has been traced back to Carneades. [47] In the first half of the 20th century, Reichenbach and Carnap argued that "the only criterion of theory-confirmation ought to be agreement with observed facts; the theory would thus be the 'most probable' one ... within a formal theory of inductive probability." [48] Carnap studies have been related to Bayesianism. [49] [50] Theories are assigned a probability, outcomes also have a probability and, given an outcome, Bayes' theorem can be applied to revise the a priori probability of each theory. [51] Bayes' theorem is useful when we have the background knowledge needed to establish the a priori probability of characteristic parameters of the application domain and the probability of the observed data depends on these parameters: the parameters that fit the data and therefore the domain get revised with a higher probability. [52]

Andrew Gelman 2012.jpg
Cosma Shalizi 2.jpg
For Gelman (left) and Shalizi (right), the application of Bayes' theorem uses the hypothetico-deductive approach to revise the models.

This view on the growth of knowledge has been criticized. Andrew Gelman and Cosma Shalizi, for example, wrote that the use of Bayes's theorem in practice is closer to the hypothetico-deductive approach, as proposed by Popper and others, than to the approach according to which the revision of probabilities is the sole consequence of the observed data.. In their work, they "examine the actual role played by prior distributions in Bayesian models, and the crucial aspects of model checking and model revision, which fall outside the scope of Bayesian confirmation theory. [53]

Critical rationalism is againt the use of probability to assess theories. Popper explained that the greater the informative content of a theory the lower will be its probability. [54] [55] He wrote that in "many cases, the more improbable (improbable in the sense of the calculus of probability) hypothesis is preferable. [55] He also wrote that "it happen quite often that I cannot prefer the logically 'better' and more improbable hypothesis, because somebody succeeded in refuting it experimentally." [56]

Justified true belief

Critical rationalism rejects the classical position that knowledge is justified true belief. David Miller noted that, for Popper, knowledge is neither justified nor believed, and that, generally, scientific knowledge is not true (in any logical sense). [57] [58] Musgrave wrote that "Popper’s theory of science, and his cure for relativism, rest upon his rejection of the traditional theory of knowledge as justified true belief." [59]

Variations

Mario Bunge's scientific realism draws on Popper's critical rationalism. MarioBungesmall.jpg
Mario Bunge's scientific realism draws on Popper's critical rationalism.

William Warren Bartley developed a variation of critical rationalism that he called pancritical rationalism. [60] [61] [62]

The Argentine-Canadian philosopher of science Mario Bunge criticized Popper's critical rationalism, [63] [64] while drawing on it to formulate an account of scientific realism. [65] [66] [67]

See also

People

Citations

  1. ↑ Popper 2013, pp. 435–437.
  2. ↑ Popper 2014, Intro., sec. XV.
  3. ↑ Popper 2005, p. 132.
  4. ↑ Popper 2014b, p. xii.
  5. 1 2 Popper 2005, ch. 7.
  6. 1 2 3 Popper 2002, ch. 1, sec. 3.
  7. 1 2 3 Popper 1972, ch. 8, sec. 4.
  8. 1 2 Popper 1983, ch. 1 sec. 2.
  9. 1 2 3 Thornton 2018, sec. 3.
  10. 1 2 3 4 Popper 1983, Intro. 1982.
  11. ↑ Popper 1983, ch. 1, sec. VI-5.
  12. ↑ Gattei 2009, p. 36.
  13. ↑ Popper 2002, sec. 7, 25.
  14. ↑ Shearmur & Stokes 2016, ch. 5 sec. 6.5.4.
  15. ↑ O'Hear 1982, ch. VI, sec. 2.
  16. ↑ Shearmur (2006), p. 275.
  17. ↑ Popper 2002, chap. 4, sec. 23.
  18. ↑ Lakatos 1999.
  19. ↑ Agassi 2008, ch. 10, app. 5.
  20. ↑ Putnam 1974, p. 224.
  21. ↑ Salmon 1978.
  22. ↑ García 2006, ch. 3, sec. 3.3.
  23. ↑ Thornton 2018, sec. 4.
  24. ↑ Popper 2005, ch. 8.
  25. ↑ Popper 2005, chap. 8.
  26. ↑ Thornton 2015.
  27. ↑ Popper 2005, p. 37, chap. 8.
  28. ↑ Shapiro & Shapiro 2010.
  29. ↑ Popper 2005, p. 42, chap. 8.
  30. ↑ Popper 2005, p. 43, chap. 8.
  31. ↑ Popper 2005, pp. 43–45, chap. 8.
  32. ↑ Grünbaum 2008.
  33. ↑ Shearmur 2006, p. 271.
  34. ↑ Popper 1974, sec. 17.
  35. ↑ Gillies 1995, sec. 3.
  36. ↑ Popper 2002, chap. 8.
  37. 1 2 Garcia 2006, p. 33.
  38. ↑ Garcia 2006, sec. 4.5.
  39. ↑ Drieschner 2005.
  40. ↑ Afisi 2013, sec. 3.1.
  41. ↑ Garcia 2006, p. 91.
  42. ↑ Popper 2002, sec. 36.
  43. ↑ Thornton 2018, sec. 5.
  44. ↑ Watkins 2014, sec. 7.4.
  45. ↑ Bartley 1982, sec. XXVI.
  46. 1 2 Popper 2014, ch. 10, sec. 4.
  47. ↑ Popkin 2015.
  48. ↑ McCullin 1976.
  49. ↑ Gower 1997, chap. 11.
  50. ↑ Kreuzman 2000.
  51. ↑ Lin 2024, sec. 1.
  52. ↑ van de Schoot et al. 2014.
  53. ↑ Gelman & Shalizi 2013.
  54. ↑ Corvi 2005, p. 45.
  55. 1 2 Popper 1972, sec. 1.8.
  56. ↑ Popper 1971.
  57. ↑ Miller 2011.
  58. ↑ Miller 1994, sec. 3.1.
  59. ↑ Musgrave 1974, p. 562.
  60. ↑ Rowbottom & Bueno 2009.
  61. ↑ Yoshida 2019.
  62. ↑ Bartley III 1999.
  63. ↑ Bunge 1983a, pp. 323–376.
  64. ↑ Bunge 1983b, pp. 59–113 (70).
  65. ↑ Agassi & Bar-Am 2019.
  66. ↑ Quintanilla 1982.
  67. ↑ Pickel 2004.

References

Further reading

This page is based on this Wikipedia article
Text is available under the CC BY-SA 4.0 license; additional terms may apply.
Images, videos and audio are available under their respective licenses.