What is Popperian falsificationism?
Mar 12, 2019
Characterizing Popper’s Position
Karl Popper’s falsificationist position is characterized by his interlocuters as follows. A skeptic of induction, Popper believes that any argument whose premises seem to support its conclusion without entailing it is invalid. On the other hand, some precisely specified hypothesis can be disproven by a observation that it forbids. For example, the hypothesis that all coin flips yield heads is (questionably) falsified by one that yields tails.
However, as Popper acknowledges, observation reports, too, are liable to error and, therefore, are not to be treated as unassailable truths. Whether it is through miscalibration of instruments, failures of interpretation, or plain sensory lapses, one can be wrong in affirming the existence of a falsifying datapoint. As Geoffrey-Smith argues, the fallibility of observation reports mean that a falsifying observation must be accepted as a decisional matter rather than as a logical necessity in order to justify the deduction that the hypothesis in question is false. To make such a decision would require the non-deductive rationale that observation reports have proven sufficiently reliable such that they should be trusted. Because conclusions about hypotheses hinge on the non-deductive decision of whether or not to accept particular observation reports, it follows that the Popperian view of science cannot be purely deductive.
Non-Deductive Exclusions
Falsificationism excludes certain types of hypotheses from the realm of scientific inquiry, namely—and non-exhaustively—existential statements, probabilistic statements, and statements of the form “Some Fs are Gs”.
Existential statements cannot be falsified because it would require infallible and all-encompassing perception about what truly is in order to know what is not; there are an infinite number of potential existences that must all be surveyed to know that something is non-existent. Moreover, in Popper’s view, neither can existential statements be confirmed because all subjective perception is fallible, and, therefore, observation reports are no guarantee of truth. Any decision to believe in something’s existence can only be justified on inductive grounds: things that I have perceived before did exist (a claim which is also arguably unverifiable), and, therefore, it is likely that this object that I now perceive exists. It is unclear how verisimilitude can be achieved without the ability to confirm or falsify claims about existences.
Popper’s position also has strange implications for the scientific status of probabilistic hypotheses. The claim that a particular event is improbable is not falsifiable because, even if it is observed quite frequently, such an observation does not amount to an impossibility and, thus, does not guarantee the falsity of the hypothesis. Consequently, probabilistic claims are unassailable by deductive reasoning and are, as per Popper’s position, unscientific. In my view, this implication is quite worrisome, as many who profess to be and are generally regarded to be scientists are concerned with identifying and explaining probabilities in nature. For example, a cancer researcher might be interested in how and why a particular gene impacts the probability of developing certain kinds of tumors. According to the falsificationist, probabilistic conjecture is rooted in “baseless” induction and ought not be prioritized as the subject of experimentation at the expense of falsifiable hypotheses.
Semantic versus Epistemic Constraints
In assessing the promise of falsifiability as a scientific demarcation criterion, it is useful to compare Popper’s approach to the demarcation problem to that of his predecessors, as Larry Laudan does. In accordance with common intuition, many early Western philosophers such as Aristotle found it useful to focus on the distinction between episteme (knowledge) and doxa (opinion). What distinguishes the former from the latter, according to Aristotle, is the incorrigibility of the scientist’s fundamental assumptions. First principles are intuited using the senses, and all other scientific knowledge follows from these principles. A second Aristotelian stipulation was the difference between the know-how of a craftsman and the understanding of a scientist. One need not be able to provide a causal account in order to successfully build a ship, but scientists, unlike shipbuilders, are concerned with first principles and “reasoned fact”: what might explain the buoyancy of some type of wood versus another. Aristotle’s criteria provided a suitable basis for excluding from the domain of science those whose sole focus is correlation, such as mathematical astronomers during the time of Ptolemy. While Popper might dispute the knowability of Aristotelian first principles, the difference between the two positions is salient and informative. Unlike Aristotle, Popper deals with the demarcation problem by posing a semantic constraint. The feature of falsifiability pertains not to the epistemic status or belief-worthiness of a theory but rather to its logical form. As Laudan and Chalmers rightly point out, a solely semantic criterion of falsifiability grants scientific status to claims that are often regarded as outlandish and epistemically indefensible: horoscopes, the existence or non-existence of Big Foot, etc. Perhaps the falsificationist might respond that such hypotheses hold less scientific merit than other hypotheses by nature of their lessened degree of falsifiability. However, it is nearly impossible to quantify how falsifiable theories in each of these categories may be. For example, there are a theoretically infinite number of faith healing failures that could falsify a particular treatment’s purported curative applications. Thus, it becomes impossible to compare degrees of falsifiability and scientific character unless one theory entails another.