Critical rationalism is, first of all, the solution proposed by Karl Popper to the epistemological problem of the growth of knowledge. Second, it has come to be one description of the method by which science progresses. And, third, it has become an ideological position which both continues the project of the Enlightenment and celebrates the role of communication in the field of knowledge. The essence of critical rationalism is contained in four steps: (1) Through intuition, ignorance, or other circumstance, even blind luck, “we stumble over some problem” (Myth, p. 101). (2) We attempt a solution by boldly proposing a theory that entails material consequences that can be tested. (3) Our conjecture and tests are subjected to rigorous critical review and stringent testing by the scientific community. (4) The results of this process are used to refine our original understanding of the problem, and we return to step (1).
Critical rationalism, therefore, represents several differences from what Suppe (1989) calls the “received view” of science. First, it changes the starting place, which was and often still is considered to be careful observation. Second, it encourages dicey hypothesizing (Miller 1994). Third, it demands deductive rationalism as the basis of critique. Fourth, it requires both a review community and a community review. And last, it considers the process to be iterative and nonconclusive, although further refinement may be deemed currently unnecessary or too costly.
For Popper, the central epistemological problem was the growth of knowledge, which is best studied “by studying the growth of scientific knowledge” (Logic, p. 15). In crafting his argument for critical rationalism, Popper rejected inductionism, psychologism, probabilism, and verificationism, and advanced falsificationism and fallibilism.
Four Principles Rejected by Critical Rationalism
Inductionism supports a general principle of inductive logic by which we can move from statements of the particular (this swan is white) to statements of the universal (all swans are white). Popper rejected the possibility of inductive logic, because even a very large number of white swan sightings do not preclude the existence of a nonwhite swan. Accordingly, there is no way to move from experience (a series of singular events) to a justified hypothesis (a tentative universal statement) or from particular supporting tests of any number to a justified theoretical claim. The empirical basis of science, therefore, is not at the beginning of the process, but in the middle, within attempts to falsify a theory by demonstrating the failure of the material consequences deductively predicted from that theory.
Popper’s rejection of induction has implications for a description of science as moving in verifiable ways from observation to theory to hypotheses to confirmation, as proposed by Bacon (Cogitata et Visa, 1607; Novum Organum, 1620), and still used in standard textbooks. Popper’s starting point for science is not observation, but a problem. The goal of science is the continued refinement of that problem, which yields theory, hypotheses, as well as observations.
The role of psychology in the philosophy of science is a continuing source of debate (e.g., Fuller 1992). To provide an empirical basis for scientific claims, psychology has been introduced in the form of “the doctrine that the empirical sciences are reducible to senseperceptions and thus to our experiences . . . [This doctrine] is one which many accept as obvious beyond all questions” (Logic, p. 93). Popper’s rejection of the doctrine does not question that experience and its observations are central to science, but rather, he denies that experience could be the basis for accepting something as true (Logic, p. 98). As he points out, the ostensibly simple sensory statement, “this swan is white,” depends on the universal principles of swan-ness and whiteness. One’s unmarked experience of being certain of the swan in sight happens only once, and is unique to the individual. Neither the experience nor the conviction can be shared except through some transformation. That transformation depends on something that is not in the experience, and is variable according to the rules adopted to govern the transformation. In making this argument, Popper rejected the various forms of positivism, accepted that theory inhabits every observation, and went on to establish a new standard of objectivity.
Positivism is a philosophical system of thought that addresses the certainty of knowledge. It is attributed in Enlightenment times to Auguste Comte, but the principles reach back to the Greek empiricists (Aristotle, Epicurus, and the Sophists). Positivism and its modern expression in logical positivism assumed the independence of theoretical statements from statements of observation – a position that Popper explicitly rejected: “Every observation and, to an even higher degree, every observation statement, is itself already an interpretation in the light of our theories” (Realism, p. 222). Critics of positivism have used this position to declare that no observation is possible without theory and that, therefore, observations made under the aegis of one theory cannot be compared with observations made under another. This doctrine – the incommensurability of theories – is often used to defend one family of theories against attacks from another. Popper, however, rejected this extreme, as well (as he must in order to sustain falsification). He pointed out that all (known) sentient beings start out with the same a priori (the material world). Consequently, all theories can be compared, although the task may be more difficult across some theories. He offered the slogan, “I may be wrong and you may be right, and by an effort, we may get nearer to the truth” (Myth, p. xii).
Popper finds his standard of objectivity in this joint effort. The need for the new standard stems from his rejection of the presumed independence of observation and theory statements. Objective observations are replaced with hybrid statements that are an indivisible mixture of fact and interpretation. How, then, does science protect itself from this inherent bias? Popper’s answer is first to warn us against false expectations and then to note the work to be done: “Science has no certainty, no rational reliability, no validity, no authority. The best we can say about it is that although it consists of our own guesses, of our own conjectures, we are doing our very best to test them; that is to say, to criticize them and refute them” (Realism, p. 222). Popper does not expect the individual scientist to criticize his or her own theory, but does demand that the scientific community commit itself to the practice of critical review and refutation.
Probabilism is the practice of assigning truth-probability values to statements. Popper rejected all forms of induction, including those that depend on probability. In this form of induction, statements are arranged along a continuum anchored by ideal (but unattainable) end points of true and false, according to their likelihood of being true. For the swan example, the form might be, “all the swans I have seen are white, therefore it is most likely that all swans are white”; or, “all the swans I have seen are white, therefore, non-white swans are rare”; or, “all the swans I have seen are white, therefore, it is most likely that the next swan I see will be white.” For Popper, all these statements are inductions because a set of experiences is used to describe a universal set without any knowledge of the universal. In the absence of such knowledge, one cannot establish the probability of any possible subset. Our sense of the truth of the matter, thus, is a psychological condition, not an epistemologically valid conclusion (Logic, p. 255).
Popper spends a good deal of time on this pernicious problem in at least four different volumes (Logic, Realism, C&R, and Myth). To him, there is no way of developing the probability calculus because the degree of corroboration depends, not on the number of tests that support a hypothesis, but on the complexity of the hypothesis and the quality of the attempts to refute it. A thousand replications of one supporting protocol offers no decrease in the probability of dozens of other tests that will refute the hypothesis. Further, a hypothesis with a 90 percent chance of being true can still be false. Consequently, it remains merely a statement not yet falsified.
Finally, Popper holds that probabilism encourages scientists to test only what is known to work, to become uninspired and uninspiring: “Every interesting and powerful statement must have a low probability; and vice versa: a statement with a high probability will be scientifically uninteresting, because it says little and has no explanatory power. Although we seek theories with a high degree of corroboration, as scientists we do not seek highly probable theories but explanations; that is to say powerful and improbable theories. The opposite view – that science aims at high probability – is a characteristic development of verificationism: if you find that you cannot verify a theory, or make it certain by induction, you may turn to probability as a kind of Ersatz for certainty, in the hope that induction may yield at least that much” (C&R, p. 65).
Although Popper would not admit that a theory can have, for example, a 90 percent chance of being true, he argues that we can distinguish theories and hypotheses according to their degree of corroboration. This phrase may sound like a probability but, it is instead an evaluation of the quality of the refutation attempted and of the severity of the tests that have failed to falsify the theory and/or hypothesis in question. The mere accumulation of confirming instances adds no new information about or value to the theory/hypothesis. Consequently, he would reject a simple counting as in arguments beginning with “many studies have shown.” A theory advances on the basis of a complex analysis of the deductive consequences that a theory/hypothesis entails and of the quality of the tests that have demonstrated (or failed to demonstrate) their occurrence.
This argument appears to have implications for evidence from convergence. In convergence, a statement can be held to be true if the preponderance of evidence supports the truth of the matter. Popper would consider this an extension of the statement: “all the swans I’ve seen are white,” which itself depends on the methodology of looking at swans. He argued that the fact that a particular methodology generates consistent results simply reflects on the methodology, not on the validity of the finding as a universal statement.
Verificationism in science assumes that statements are meaningful only to the extent of their empirical content. Empirical content is material phenomena that can either be experienced directly or produce a necessary link to sensory experience. Verificationists use this principle as a demarcation between scientific and nonscientific statements, so that a statement with no empirical content is scientifically meaningless. This requirement would appear to rule out many variables that are of relevance to communication research, such as uncertainty, compliance, or apprehension. (The solution was operationalism – a contested, but generally accepted practice – by which a concept like uncertainty is defined in the methods of measurement adopted.) Verificationists also advance the argument that each hypothesis has a crucial test or series of tests that would justify its acceptance. Nearly all statistical studies in communication research can be seen to adopt this as a rhetorical, if not an epistemological stance.
Popper disagreed with verificationism on both points. He noted that atomic theory predated our ability to provide an empirical basis by several centuries. Atomic theory, nonetheless, was a worthy scientific problem that justified the effort to achieve an empirical basis. The immediate availability of empirical content, therefore, was not an acceptable demarcation criterion. His demarcation criterion was that the theory had to entail potentially falsifiable consequences.
Popper saw crucial tests as a return to inductionism – a point in his writings that has been lost in current notions of verification. No hypothesis or theory can ever be justified by an individual outcome or even the sum total of all supporting outcomes. No matter how many there are, outcomes can never deny disconfirming instances.
Falsificationism and Fallibilism
Though one cannot verify a hypothesis, Popper’s system does offer ways of evaluating a hypothesis or theory, proposing falsification both as a demarcation between science and nonscience, and as the appropriate means of assaying the possible truth content of a hypothesis. Falsification rests on deductive rather than inductive logic. A theory, to be scientific, must require certain empirical outcomes that can be deduced and tested to see if they actually occur. If the theory meets this requirement, it is falsifiable and meets the minimal requirements of science. The more outcomes it requires, and the more directly those outcomes can be tested, the more robust the theory. Any failure to achieve those outcomes falsifies the present theory and calls for its revision. Importantly, the occurrence of a predicted outcome does not verify the hypothesis, but simply declares it as not falsified. In the ordinary practice of inferential statistics, however, positive information is gained when the null hypothesis – a negation of a predicted outcome – is rejected. No information is considered gained when the null hypothesis cannot be rejected. This practice, on the face of it, is a verificationist, not a critical rationalist approach.
Popper proposed four classifications of statements about the world: statements that can be but have currently not been tested (not yet in the active domain of science); statements that have been tested and not falsified (scientific statements under continuing review; the location of current scientific activity); statements that have been tested and falsified (statements held to be scientifically false and presumably returned for revision); and statements tested, not falsified, and no longer being tested (statements acted upon by the scientific community as scientifically classified as true). The first type of statement is perhaps uninteresting or currently intractable to testing. The second is the bulk of research. The third is where the controversies lie, and the last type of statement is part of the background knowledge of science that will be used to pose new problems and their tests.
Popper’s original formulation of falsificationism has been the subject of much critique for its unfaithfulness to the actual practice of science (Lakatos 1970), its presumption of a common basis of comparison across theories (Feyerabend 1965), its inability to specify just when a theory has been falsified (Newton-Smith 1981; see also the Quine-Duhem thesis discussed in Klee 1997), or equally when it has been sufficiently corroborated (Salman 1978).
Fallibilism is the doctrine that all of science can be wrong both in the particular and in general. Popper believed that science had no special claim to authority or to certainty. While fallibilism has a long history, including the philosophy of pragmatism, Popper’s position should be understood within the cultural context in which he was writing (Germany in the 1930s, anticipating the rise of Hitler) and his personal distaste for dogmatism of any kind (Open). He avoided falling into unrelenting skepticism by holding that sufficient corroboration can justify classifying a claim as true, and acting on it as if it were true, without giving up the possibility that it may be false. Popper saw fallibilism as the opportunity for bold new conjectures about the world rather than as a limitation on human knowledge.
Critical Rationalism in Contemporary Communication Science
Critical rationalism has steadily advanced Karl Popper in the ranks of scholars rejecting positivism. Other key figures include Thomas Kuhn, Paul Feyerabend, Norwood Hanson, Imre Lakatos, Willard Van Orman Quine, Dudley Shapere, and Frederick Suppe. While most philosophers of science today would declare positivism dead, there remains a nostalgia for the clarity and simplicity that it provided. Positivist positions such as the unity of science, the observational basis of science, the corrective presence of an objective reality, and the mathematical framework of science still dominate discussions of communication as a science. Nonetheless, critical rationalism and its allies in the history and sociology of science have opened up spaces for practices in social science that were simply unavailable as few as 50 years ago, albeit still controversial today.
One particular appeal of critical rationalism is that it grounds science in communication practices. Science, in critical rationalism, progresses in the robust debate between contesting theories and their empirical evidence. In Popper’s figuration, it was, however, an enlightened, Habermasian debate with all members situated in a common epistemology. The continued fractionalization of the field suggests that communication researchers remain divided by epistemology as well as politics.
Communication is a very young science which has yet to develop a center. The conventionalism of critical rationalism is difficult to achieve because there is little agreement regarding background knowledge and frameworks of evidence. It is not clear that there is a common community of review, or an open, public critique in the pages of our journals. There is promise and possibility, but little of a unified practice.
- Note: Popper developed Critical Rationalism over a 60-year period. The major works referred to here are: The logic of scientific discovery, 1959 (1st pub. 1935) [Logic]; The open society and its enemies, 1945 [Open]; Realism and the aim of science, 1956 [Realism]; Conjectures and refutations, 1989 [C&R]; The myth of the framework: In defence of science and rationality, 1994 [Myth].
- Feyerabend, P. K. (1965). Problems of empiricism. In R. Dolodny (ed.), Beyond the edge of certainty. Englewood Cliffs, NJ: Prentice-Hall, pp. 145 –150.
- Fuller, S. (1992). Philosophy of science and its discontents. 2nd edn. New York: Guilford.
- Klee, R. (1997). Introduction to the philosophy of science. New York: Oxford University Press.
- Lakatos, I. (1970). Falsification and methodology of scientific research programmes. In I. Lakatos & A. Musgrave (eds.), Criticism and the growth of knowledge. Cambridge: Cambridge University Press, pp. 91–196.
- Miller, D. (1994). Critical rationalism. Chicago: Open Court.
- Newton-Smith, W. H. (1981). The rationality of science. London: Routledge.
- Salman, W. C. (1978). Unfinished business: The problem of induction. Philosophical Studies, 33, 1–19.
- Suppe, F. (1989). The semantic conception of theories and scientific realism. Urbana: University of Illinois Press.
Back to Communication Theory and Philosophy.