8) Wesley Salmon attacks Karl Popper on the grounds that his theory of science does not allow for theory choice based upon past corroboration. Are Popper's responses adequate? Why/why not?
The problem of induction has been a pervasive one that has split philosophers and empiricists for years. Essentially, the issue lies in whether or not inductive reasoning can bring about true knowledge; whether anything can be truly known. The debate is split between whether or not a generalization based on a set number of observations can be true (e.g., since all elephants observed are gray, all elephants must, by virtue of fact, be gray), or that there are always possibilities for outcomes outside the scope of what has been previously observed or presupposed (Grattan-Guinness, 2004).
One person who sought to solve the problem of induction was the philosopher Karl Popper, in his book Logic of Scientific Discovery; "the theory to be developed in the following pages stands directly opposed to all attempts to operate with the ideas of inductive logic " (Popper, Ch. 1). In effect, Popper completely disavowed the idea of inductive logic; science cannot definitively answer a question definitively, as there is always the possibility for a different result in the future, and something that has not been looked at or anticipated as yet. To Popper, the goal of science is to take a look at existing theories and attempt to challenge them, adding further criticism to the table.
Popper believes that, in science, one should not attempt to justify theories, as it "begs for an authoritarian answer" (Conjectures and Refutations, p. 25). Instead, errors should be sought and corrected; when someone says that something is wrong, the scientific method would be used to fix said error. Since it is impossible to prove something right, as Popper would say, one can only prove things wrong.
Wesley Salmon, another philosopher, directly refuted and criticized Popper's perspective on inductive logic. According to him, he believes that predictions have to be made when conducting scientific studies, or else there would be no theory to test. As for the idea that incorrect theories have to be verified, Salmon claims that people who follow Popper's philosophy would have to pick from a number of theories that are out there, thus creating a dilemma of choice. According to Salmon, there is a dilemma for every Popperian scientist who wants to corroborate a theory: whether to state that a theory is correct because it has survived previous criticism, or to simply admit that there is no predictive power to Popper's preference to corroborated theories (Salmon 1967, p. 26).
In response to Salmon and other critiques of his philosophies, Popper had to defend his ideas of inductive reasoning. First, Popper claims that using corroborated theories is justified, since critical scrutiny has been applied to it. In essence, because people have examined it before; therefore, it is reasonable to assume that enough study has been done on it to place it in danger of being considered a true statement. As a result, it must be subjected to further scrutiny. Salmon feels there is a danger in preferring corroborated generalizations to uncorroborated ones, as those generalizations have not been proven false yet. However, Popper prefers to examine theories that have predictive content (i.e., it might have implications for experiences and studies in the future). To him, there is no way to sustain the philosophy that corroboration does not exert a practical influence on predictions made in the future, and therefore cannot be sustained.
Popper's attempt to defend his theories regarding inductive reasoning against Salmon's indications are inadequate. Because they end up relying on inductive reasoning in a very strong sense - due to the connection between corroboration and induction - it can be said that Popper disproves himself somewhat. When a theory has too many unfalsified hypotheses to debunk, Popperian scientists fall back on the theory that has the most pragmatic preference, so that whatever they are studying would be of the most use. Salmon successfully argues that Popper's theory is against the principles of deductivism, since (according to his debunking of inductive reasoning) there should be no reason to choose a more corroborated theory over a less corroborated one, even though he does. Popper's explanation of reliance on a more corroborated theory depends on subjective likelihood, and a relative sense that other data brings the truth closer to those theories.
In conclusion, Popper, in his defense of his theories against Salmon, relies eventually upon the idea that, in terms of what theory a Popperian will focus on debunking, they will choose a corroborated theory. This is done because there is no other way to truly determine what makes one theory better than another. However, when Popper implies that corroborated theories hold more credibility than uncorroborated ones, he implicitly accepts that induction occurs (since that is what induction is - inferring that a theory is more correct because study has been performed on it). With induction, people assume that one theory is more correct than others, because it has been studied more. There is a reasonable expectation of truth in that sense; the theory is thought to be correct, since the amount of information available supports that theory. By falling back to the idea that some theories, due to corroboration, are better than others, Popper's defense turns out to be the same thing he is decrying - induction. These theories are assumed to be more right than non-corroborated indications and theories; that is the essence of induction. With that said, Popper's responses to Salmon are inadequate to defend his idea of debunking inductive reasoning.
References
Grattan-Guinness, I. (2004). Karl Popper and the 'problem of induction': a fresh look at the logic
of testing scientific theories. Erkenntnis 60(1): 107-120.
Popper, K. (1959). The Logic of Scientific Discovery.
Popper, K. (1964). Conjectures and Refutations.
Salmon, W.C. (1967). The Foundations of Scientific Inference.
Salmon, W.C. (1981). Rational prediction. The British Journal for the Philosophy of Science.
32(2): 115-125.
Watkins, J.W.N. (1984). Science and Skepticism. Princeton University Press.
