”It may be possible to excise all inductive ingredients from science, but if the operation were successful, the patient (science), deprived of all predictive import, would die” (Salmon 1981, 125).

In previous posts, I have suggested that we should approach the estimating of possible futures (of science) in a theory-driven way: We need to use theories to construct the possible structures of the future and attempt to determine the contents of these structures in order to understand the space of future possibilities.

But how should we determine, after we have mapped the possible structures of the future, on which structure we should act upon in the context of a practical decision? Here we have a version of the problem of induction: How can we infer from the past evidence for and against a theory whether the theory will hold in the future? How can we rationally choose one of the theories as our guide to the future?

Karl Popper suggested that we should avoid the problem of induction by rejecting induction altogether. According to Popper, evidence can only refute theories, not establish their truth. Surely, we have theories that have not been (seriously) refuted, but this does not guarantee their truth. Actually, many “theories” cannot be refuted precisely because they lack empirical contents (like some conspiracy theories) and, therefore, we should not equate “has not been refuted” with “is true”. However, there is an interesting class of theories: those that have high empirical contents and have not been refuted even though they have been seriously tested. According to Popper, such theories cannot be said to be true or verified but corroborated. Moreover – and this is important – from the corroboration of a theory we cannot infer that what it says about the future will be true. Otherwise we would reintroduce induction into science: From the past corroboration of a theory we would infer the future successes of the theory.

Should we follow Popper and choose, as the basis of estimating the future, those theories that have been corroborated to the highest degree? Perhaps, but Popperian picture cannot be the whole story due to the fact that corroboration does not allow us to infer that the theory will hold in the future. And we are very much interested in possible futures!

In an interesting paper from 1981, Wesley C. Salmon argues that Popper’s rejection of induction makes rational theory-based prediction impossible. I quote him at length because he was such a great writer:

We begin by asking how science can possibly do without induction. We are told that the aim of science is to arrive at the best explanatory theories we can find. When we ask how to tell whether one theory is better than another, we are told that it depends upon their comparative ability to stand up to severe testing and critical discussion. When we ask whether this mode of evaluation does not contain some inductive aspect, we are assured that the evaluation is made wholly in terms of their comparative success up to now; but since this evaluation is made entirely in terms of past performance, it escapes inductive contamination because it lacks predictive import. When we then ask how to select theories for purposes of rational prediction, we are told that we should prefer the theory which is ‘best tested’ and which ‘in the light of our critical discussion, appears to be the best so far’, even though we have been explicitly assured that testing and critical discussion have no predictive import. Popper tells us, ‘I do not know of anything more “rational” than a well-conducted critical discussion.’ I fail to see how it could be rational to judge theories for purposes of prediction in terms of a criterion which is emphatically claimed to be lacking in predictive import.

Basically, the problem is this: If understand theory-acceptance in terms of corroboration, then theory-acceptance does not tell us how to rationally estimate the future. Corroboration is, by definition, supposed to tell how to accept a theory without commitment to the success of the theory in the future.

The conclusion is that there has to be something more to a rational theory-choice than the measurement of the degree of corroboration. In order to use theories to estimate the future, we need to have some principles that connect the past success and corroboration of a theory to the future success of that theory.

Of course, Popper’s ideas of falsification and corroboration are outdated for reasons that go beyond the problem introduced above. Moreover, scientists make future predictions (about how pandemics will spread etc.) on the basis of their best theories and we should attempt to understand how this is done and whether it is done rationally (i.e. formulate the criteria of theory-choice that go beyond Popper’s falsificationism). A philosophical theory of science that defines integral parts of scientific activity as unscientific cannot be accept if we attempt to understand the actual science.

However, Popper’s problems are a useful reminder of the trickiness of estimating the future of a complex historical phenomena like science. Even if corroboration is not the whole story concerning theory-choice and even if we had some crazily good argument which connects the past corroboration of a theory to its future success, we would still have a problem: How to test and measure the corroboration of a theory that concerns a complex historical phenomenon? We are still far away from corroborated theories concerning historical phenomena like science. Actually, it is difficult to say how the testing of such theories is possible or even what “testing” in this connection means. I have written about the problems in earlier posts (here and here). So the formulation of criteria that connects past corroboration to future success is not even our main problem. Rather, our problem is to find criteria that tell us when a theory concerning historical phenomena is corroborated.

I hope that the reader will not find the depth of the problem demoralizing. I think that the explication of the depth of the problem is (ironically) a practically useful reminder of how difficult it is to choose a course of action on the basis of what we know. Moreover, I do not think we should be paralyzed by the problem; rather we should face it. We should attempt to find criteria (C) that tells us when it is rational to choose one theory concerning complex historical phenomenon over another. Using a variety of theories will make us aware of the variety of possible futures. However not every theory is equally good. We need to balance cautiousness with preference. And when it comes to a particular decision concerning the future, we need to acknowledge that we do not know for sure what will happen. Futures research cannot give direct answers. However, it can still give warranted answers if we find criteria (C). And we can become more aware of the possible futures by using a theory-driven approach.

It is better to live in uncertainty than in ignorance.

References:

Salmon, Wesley C. (1981). “Rational Prediction”. The British Journal for the Philosophy of Science, Vol. 32, No. 2.