David Teira

This article analyses how the French Academy of Sciences assessed Jaime Ferrán’s cholera vaccine submitted for the Prix Bréant in the 1880s. Ferrán, a Spanish independent physician, discovered the treatment in 1884 and tried it on thousands of patients during the cholera outbreak in Valencia the following year. His evaluation sparked a controversy in Spain and abroad on the vaccine’s efficacy. The Bréant jury did not see any evidence for it in Ferrán’s submission, a decision usually interpreted in terms of French scientific nationalism (or simple chauvinism): an outsider from the scientific periphery could not be awarded the Bréant. Drawing on the archival records of the award, we suggest that Ferrán failed instead to provide data that the Academy could consider unbiased, according to the contemporary standards for data presentation. We will illustrate these standards at work in the assessment of another submission from Spain, by Philipp Hauser, who received the Bréant for the thoroughness of his statistical endeavour

Jesús Zamora Bonilla, profesor de la Universidad Carlos III, es un autor bien conocido para el lector de Theoria por sus publicaciones en filosofía general de las ciencias y en filosofía de la economía. Sólo su distribuidora es culpable de que muchos lectores ignoren aún su primer libro, Mentiras a medias, un amplio estudio de la idea de verosimilitud que incluye una propuesta original, ya discutida en foros internacionales. El trabajo en epistemología general de las ciencias que aquí presentamos se articula, además, con sus aportaciones a la economía de la ciencia y a la propia reconstrucción racional de la metodología económica. No está de más, por tanto, el volver sobre los fundamentos de este proyecto, pues, como es sabido, no son pocas las dificultades que ofrece el concepto de verosimilitud.

A la vista del catálogo de nuestro Instituto Nacional de Estadística, quién dejará de sorprenderse al descubrir entre las colecciones auspiciadas por el INED francés una dedicada a los Clásicos de la economía y la población, con cuidadísimas ediciones de Condorcet, Süssmilch, Quesnay, Graunt... A esta colección se suma ahora, bajo la dirección de Eric Brian, otra serie de Estudios e investigaciones históricas, cuyo primer volumen comentamos aquí. Matemáticas y acción política, compilado por Thierry Martin, es, además, una excelente representación de los trabajos que en Francia se desarrollan en torno a la matemática social desde múltiples enfoques.

Breeds are classifications of domestic animals that share, to a certain degree, a set of conventional phenotypic traits. We are going to defend that, despite classifying biological entities, animal breeds are social kinds. We will adopt Godman’s view of social kinds, classifications with predictive power based on social learning processes. We will show that, although the folk concept of animal breed refers to a biological kind, there is no way to define it. The expert definitions of breeds are instead based on socially learnt conventions and skills (artificial selection), yielding groupings in which scientific predictions are possible. We will discuss in what sense breeds are social, but not human kinds and in what sense the concept of a breed is necessary to make them real.

Randomized controlled trials test new drugs using various debiasing devices to prevent participants from manipulating the trials. But participants often dislike controls, arguing that they impose a paternalist constraint on their legitimate preferences. The 21st Century Cures Act, passed by US Congress in 2016, encourages the Food and Drug Administration to use alternative testing methods, incorporating participants’ preferences, for regulatory purposes. We discuss, from a historical perspective, the trade-off between trial impartiality and participants’ freedom. We argue that the only way out is considering which methods improve upon the performance of conventional trials in keeping dangerous or inefficacious compounds out of pharmaceutical markets.

Experimental balance is usually understood as the control for the value of the conditions, other than the one under study, which are liable to affect the result of a test. We discuss three different approaches to balance. ‘Millean balance’ requires identifying and equalizing ex ante the value of these conditions in order to conduct solid causal inferences. ‘Fisherian balance’ measures ex post the influence of uncontrolled conditions through the analysis of variance. In ‘efficiency balance’ the value of the antecedent conditions is decided ex ante according to the efficiency they yield in the estimation of the treatment outcome. Against some old arguments by Worrall, we show that in both Fisherian and efficiency balance there are good reasons to randomize the allocation of treatments, in particular when there is no agreement among experimenters as to the antecedent conditions to be controlled for.

Statistical evidence is pervasive in medicine. In this chapter we will focus on the reliability of randomized clinical trials (RCTs) conducted to test the safety and efficacy of medical treatments. RCTs are scientific experiments and, as such, we expect them to be replicable: if we repeat the same experiment time and again, we should obtain the same outcome (Norton 2015). The statistical design of the test should guarantee that the observed outcome is not a random event, but rather a real effect of the treatments administered. However, for more than a decade now, we have been discussing a replicability crisis across different experimental disciplines including medicine: the outcomes of trials published in very prestigious journals often disappear when the experiment is repeated (see, e.g., Lehrer 2010, Begley and Ellis 2012, Horton 2015). There are different accounts of the reason for this replicability crisis, ranging from scientific fraud to lack of institutional incentives to double-check someone else’s results. In this chapter we will use the replicability crisis as a thread to introduce some central issues in the design of scientific experiments in medicine. First, in section 1, we will see how replicability and statistical significance are connected: we can only make sense of the p-value of a trial outcome within a series of replications of the test. But in order to conduct these replications properly, we need to agree on the proper design of the experiment we are going to repeat. In particular, we need to prevent the preferences of the experimenters from biasing the outcome of the experiment. If there is such a bias, when the experiment is replicated by a third party, the observed outcome will vanish. In section 2, we will argue that trialists need to agree on the debiasing procedures and the statistical quality controls that feature in the trial protocol, if they want the outcome to be replicable. In section 3, we will make two complementary points. On the one hand, replicability per se is not everything: we need trial outcomes that are not only statistically significant but also clinically relevant. On the other hand, trials are not everything: the experts analyzing the evidence can improve the reliability of statistical evidence, although they sometimes fail; we need to further study how they make their decisions. In section 4, we will use a controversy about the overprescription of statins to show how non-replicable effects are obtained in trials and how experts may fail at detecting such flaws, if the commercial interests are big enough.

Fact-checking agencies assess and score the truthfulness of politicians’ claims to foster their electoral accountability. Fact-checking is sometimes presented as a quasi-scientific activity, based on reproducible verification protocols that would guarantee an unbiased assessment. We will study these verification protocols and discuss under which conditions fact-checking could achieve effective reproducibility. Through an analysis of the methodological norms in verification protocols, we will argue that achieving reproducible fact-checking may not help much in rendering politicians accountable. Political fact-checkers do not deliver either reproducibility or accountability today, and there are reasons to think that traditional quality journalism may serve liberal democracies better.