Wesley Salmon objects to Popper’s theory of knowledge on the grounds that, contrary to its stated rejection of a principle of induction, in order to explain how one can rationally decide between competing unfalsified theories, it requires the adoption of a principle of induction. The advice to an applied scientist or engineer to act as if the best-tested theories are probably true and the untested theories are probably false, though no doubt excellent advice, does not have any claim to be dubbed ‘rational’ unless a pragmatic principle of induction is adopted.
If the applied scientist’s choice is guided by the best-tested scientific theories available to him, then it appears that he is assuming that what was successful in the past will remain successful in the future. This would be an assumption rejected by Popper, for it employs the principle of induction. However, if a scientist, following Popper’s theory of knowledge, renounces a principle of induction, then he is not allowed to say that ‘future unobserved events will resemble past observed events.’
Thus, the scientist cannot rationally choose between two types of unfalsified theories at any time: the currently best-tested theory and an untested theory. To prefer the currently best-tested theory over an untested theory, Salmon claims, the rational agent is compelled to be an inductive agent.
Q1: Why, should we rationally prefer (A) the currently best-tested theory over (B) an untested theory without making an inductive inference?
In order to make a rational prediction, it seems to me, one must make a *rational* choice of a premise for such an argument. … Thus, it seems to me, corroboration — the ground for theoretical preference — furnishes no rational basis for preference of one conjecture to another for purposes of practical prediction.” (Wesley Salmon, “Rational Prediction”, Philosophy of Science: The Central Issues, ed. Martin Curd, J.A. Cover (W.W. Norton & Company, New York/London), 437-8.)
If there aren’t grounds for regarding it as rational to rely on A over B that do not appeal to a principle of induction, then falsificationism runs aground on the pragmatic problem of induction, for falsificationism must implicitly assume something that which it explicitly denies in order to rationally practice theory-choice. However, if an answer for rationally preferring A over B without resorting to induction exists, then Salmon’s objection to falsification falls flat.
In other words, Salmon is right if and only if being rational requires justifying reasons for the future success of A over B. As Salmon says, “This, I take it, is the crucial thesis — that there is a rational basis for preferring one unrefuted generalization to another for use in a predictive argument. If that is correct, then Popper can legitimately claim to have solved the problem of rational prediction.” (ibid., 436)
If being ‘rational’ requires producing a reason for why the best-tested theory is more likely to be true than a theory that has not been tested, then no reasons can be found without relying on the assumption that future unobserved events will resemble past observed events. Due to the nature of corroboration, a currently untested theory may be true, or have more verisimilitude, while the best-tested theory may be false, or have a lesser degree of verisimilitude than the currently untested theory. The next question is then:
Q2: Is being rational producing a reason for believing in the future success of theories, or is it some other kind of behavior towards theories?
Popper provided an alternative form of ‘rational’ activity — the critical appraisal of how the scientist assesses his theories. Popper begins with an observation: “[S]ince 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.” Now, David Miller thinks Popper was mistaken.
I will attempt to sketch out an answer following Miller’s argument in “Putting Science to Work” to Q2 as follows:
Popper suggests that the target of our critical assessment is not the best-tested theory in our possession, but the best proposal for action: the proposal that has most resolutely survived criticism. The discussion should not focus on the hypotheses that are adopted by the scientist but on the proposals that they adopt and the decisions that they take. This meta-proposal is supposed to demonstrate that a scientist may prefer the presently best-tested theory over an untested theory. However, this is the starting-point for Miller in a different direction: technicians and engineers adopt proposals often without any knowledge of the scientific theories that limit the available proposals.
Question 1 can be rephrased, according to Miller, as follows:
Q3: Why, should the technician or engineer rationally prefer (A) the proposal that is limited by scientific theories that have presently survived intersubjective criticism over (B) a proposal that is not limited by scientific theories that have not presently survived intersubjective criticism?
Miller suggests that the rational thing to do is to follow the proposal that is limited by what scientific theories tell the technician. There are an infinite number of possible scientific theories that limit the technician in conflicting ways. We are lucky to have such a limited number that have survived criticism, or are even in principle coherent with our background assumptions — which happen to be these scientific theories that survived the most intense criticism that can be directed at them.
The answer to question Q3 is then as follows: there are rational decisions for the technician or engineer that do not appeal to some principle of induction.
Popper does not attempt the impossible — that is, to justify reasons for preferring a theory by nature of its possible future successes. Popper’s approach is then to attempt to achieve what is possible, which is the formation of a critical preference for one unfalsified theory rather than another.
While a scientist can choose to consider it rational to choose A, the unfalsified theory that has best survived all current testing over B, the unfalsified theory that has not been subject to any tests. Before B is even vocalized, much less imagined, it does not present itself as a viable option. Once it is put forward, it either is incompatible with our background assumptions or it is not. Most theories will necessarily be incompatible, while a few will survive (see here). Once this step occurs, we are not talking about criticized versus uncriticized theories, but of preferring two theories that have been criticized. Anyone who disagrees with this answer can give an example where she thinks that this advice would be bad advice.
However, it is no criticism of this proposal for rational behavior to say that its effectiveness hasn’t been authenticated. Mere doubt is not criticism. Furthermore, if induction is invalid, such authentication cannot be obtained, for it would require acquiring confirming instances that demonstrate that there are reasons why this proposal for rational behavior will continue to be successful in the future. Therefore, while this answer may not satisfy those that seek to justify their preferences, for such justification for preferences is a epistemological dead-end, there is no claim that the methodological proposal to tentatively adopt our best-tested theories, and to act in light of these theories, will be reliable, successful or true.
Miller, on the other hand, understands that the technician or engineer need not have any knowledge of the scientific theories that limit their proposals in order to make a rational choice. Polanyi’s concept of ‘tacit knowledge’ comes into play now: engineers have a long history of trial and error in the production of their craft. They know what doesn’t work. I, however, think a possible link between Popper and Miller can be found in an externalist theory of justification. If a technician or engineer makes a proposal that is limited in some way by relevant scientific theories that have most resolutely survived criticism, then I conjecture that they may have more successes in their proposals than proposals that are not limited in some way by relevant scientific theories. This solution is stronger than Salmon’s original requirement of solving the pragmatic problem of induction: technicians and engineers may have a justification for their proposal, for the technician or engineer assumes, unwittingly, to operate within the limits set down by current scientific theories, rather than outside these limits.
While this is no objective form of justification, it is a methodological or social ‘justification’: it is best not to attempt to produce technologies that contradict our current scientific knowledge. And yet, there is no prohibition against such an attempt — people may act irrationally and attempt to produce technologies that, for instance, violate the second law of thermodynamics. If they should be successful, however, then we are the better for it, but mere luck is possible in any activity, even the most irrational. Such a social ‘justification’ does not claim that uncriticized theories are in any way worse than currently best-criticized theories, or may have higher or lesser degrees of verisimilitude, only that one is to be preferred for practical action.