Popper and phylogenetics, a misguided rendezvous

Abstract

Abstract. Popper's falsificationism is frequently referred to as a general normative reference system in phylogenetics. Referring to falsificationism, phylogeneticists have made four central claims, including that frequency probabilities (1) cannot be used for inferring degrees of corroboration and (2) cannot be used in phylogenetics because phylogeny is a unique process, (3) likelihood methods represent verificationist approaches, and (4) the congruence test is a Popperian test. However, these claims are inconsistent with Popper's theory. Moreover, phylogeneticists have proposed four strategies for dealing with the unfalsifiability of cladograms, including (1) interpreting re-interpretations of putative synapomorphy as homoplasy as Popperian ad hoc manoeuvres, (2) decoupling corroboration from falsification, (3) interpreting the tree with the highest likelihood as the most corroborated tree, and (4) interpreting tree hypotheses as Popperian probabilistic hypotheses that do not have to be falsifiable. These strategies are also inconsistent with Popper's theory. Four fundamental problems and a problem with Popper's formula for measuring degree of corroboration demonstrate that Popper's theory does not live up to its own claims. Moreover, neither historical nor experimental sciences can be conducted in a way that is consistent with the principles of falsificationism. Therefore, phylogeneticists should stop referring to falsificationism when defending a specific methodological position.

A sort of revolution: Systematics and physical anthropology in the 20th century

“…These questions have largely subsided from the journals of biological systematics, but they continue to crop up in the writings of philosophers and historians of biology. Interested readers can refer to the useful summaries and analyses of this literature by Hull () and Vogt ().…”

“…A cladogram forbids nothing to happen, because biological theory gives us no grounds for expecting that the next character we look at will fit the pattern inferred from its predecessors, or that synapomorphies will ultimately outnumber or outweigh other sorts of resemblances in our data set. As the German systematist Lars Vogt () writes, “a given tree, in combination with descent with modification as background knowledge, does not prohibit any specific character‐state distribution … As a consequence … cladograms are not falsifiable in principle and thus not testable in a Popperian sense.”…”

Parsimony analysis of unaligned sequence data: some clarifications

“… Popper's philosophy of science has been discussed extensively within systematics (see, for example, Gattei, ; Rieppel, ,b; Vogt, , ; Kluge, ; Crother and Murray, and references therein) and diverse scientific disciplines (e.g., Hansson, ; Persson, ). We are aware of some of the alleged shortcomings of Popper's position that, according to some authors, render Popper's falsificationism fatally flawed (e.g., Vogt, ).…”

“… Popper's philosophy of science has been discussed extensively within systematics (see, for example, Gattei, ; Rieppel, ,b; Vogt, , ; Kluge, ; Crother and Murray, and references therein) and diverse scientific disciplines (e.g., Hansson, ; Persson, ). We are aware of some of the alleged shortcomings of Popper's position that, according to some authors, render Popper's falsificationism fatally flawed (e.g., Vogt, ). Nevertheless, we are also aware that several theoretical components tightly related to Popper's philosophy of science, such as simplicity, unification and maximization of explanatory power, have been repeatedly recognized as epistemological virtues of scientific explanations (Farris, ; Koertge, ; Norton, ; Baker, ), and as such, we consider that Popper's () characterization of scientific criticism, which allowed him to criticize the inductive logic and establish his demarcation criterion, is an extremely useful tool to consider in the epistemological debate. …”