Coko, Klodian
(2023)
Consilience and Hypothesis in the Nineteenth Century.
[Preprint]
Abstract
Perhaps the most important development in nineteenth century philosophical discussions on scientific
methodology was the dynamic re-emergence of the method of hypothesis. Simply put, the method of hypothesis
starts with the formulation of a hypothesis to explain observed facts. Then, the observable consequences of
the hypothesis are deduced. If the deduced consequences are verified by observation and experiment, the
hypothesis is validated. If the observed consequences of the hypothesis disagree with observation and
experiment, the hypothesis is modified or discarded. This development came as a reaction to traditional
scientific methodology which regarded scientific inferences mostly as inductive generalizations from empirical
facts. More specifically, for traditional methodology scientific investigation was a purely empirical process that
started with observation and experiment, and then cautiously, but surely, led to the formulation of general
statements that provided the true description and explanation of the observed facts. In this context, the
formulation of hypotheses about the object of investigation was considered unnecessary and potentially
harmful to scientific inquiry. During the nineteenth century, however, philosophers—especially the ones
sensitive to the complexities of scientific practice as revealed also by the study of the history of science—
started to realize that traditional inductive methodology could not accommodate new scientific developments,
particularly those related to the investigation of unobservable processes, entities, and phenomena. The
scientists investigating such unobservables, rather than generalizing from particular facts, constructed
hypotheses and deduced predictions from them which then they compared with the observable facts (Laudan
1981).
Amidst all the criteria for evaluating hypotheses, the ability of a hypothesis to successfully predict,
explain, and/or be supported by a variety of classes of empirical facts—especially facts that played no role in
the initial formulation of the hypothesis—was considered to be the highest criterion of truth. Support from
different classes of facts was thought to give rise to a no-coincidence argument for the truth of the hypothesis;
namely that it would be an improbable coincidence for a hypothesis (usually about unobservables) to be able
to accommodate a variety of (observable) facts and yet for it to be false. This criterion is found more
explicitly in William Whewell’s (1840b; 1847; 1858) notion of the consilience of inductions, but it can also be
encountered in the writings of many other nineteenth century philosophers such as John Herschel (1830),
William Stanley Jevons (1874), and Charles Sanders Peirce (1878, c. 1905). Throughout this chapter we use
the term ‘consilience’ to refer to the ability of the theoretical hypothesis to account for different classes of
facts (independently of whether the authors discussing this ability use this term to refer to it).
This chapter examines the method of verifying hypotheses in the thought of Whewell, J. S. Mill,
Herschel, Jevons, and Peirce. It focuses especially on expounding and evaluating the reasons these
philosophers gave for the epistemic force attributed to the criterion of consilience, i.e., their response to the
question: why the ability of a hypothesis to explain different classes of facts should be considered (or should
not, in Mill’s case) as an argument for its truth? The chapter is structured as follows. Section 2 examines the
structure, epistemic role, and epistemic import of Whewell’s consilience of inductions. Section 3 discusses
Mill’s and Herschel’s views on the method of the hypothesis and the criterion of consilience. Section 4
examines the same topics in the thought of Jevons and Peirce. Section 5 concludes by summarizing the main
points.
Monthly Views for the past 3 years
Monthly Downloads for the past 3 years
Plum Analytics
Actions (login required)
 |
View Item |
![[feed]](http://philsci-archive.pitt.edu/style/images/feed-icon-32x32.png){kind=icon}