Whewell on classification and consilience

https://doi.org/10.1016/j.shpsc.2017.06.007Get rights and content

Highlights

  • Whewell analyzed “natural affinity” as a particular methodology in nineteenth century systematics.

  • Scientists classified objects without knowledge about the causes of natural relationships.

  • Whewell's analysis of systematics led to his concept “consilience”.

Abstract

In this paper I sketch William Whewell's attempts to impose order on classificatory mineralogy, which was in Whewell's day (1794–1866) a confused science of uncertain prospects. Whewell argued that progress was impeded by the crude reductionist assumption that all macroproperties of crystals could be straightforwardly explained by reference to the crystals' chemical constituents. By comparison with biological classification, Whewell proposed methodological reforms that he claimed would lead to a natural classification of minerals, which in turn would support advances in causal understanding of the properties of minerals. Whewell's comparison to successful biological classification is particularly striking given that classificatory biologists did not share an understanding of the causal structure underlying the natural classification of life (the common descent with modification of all organisms).

Whewell's key proposed methodological reform is consideration of multiple, distinct principles of classification. The most powerful evidence in support of a natural classificatory claim is the consilience of claims arrived at through distinct lines of reasoning, rooted in distinct conceptual approaches to the target objects. Mineralogists must consider not only elemental composition and chemical affinities, but also symmetry and polarity. Geometrical properties are central to what makes an individual mineral the type of mineral that it is. In Whewell's view, function and organization jointly define life, and so are the keys to understanding what makes an organism the type of organism that it is. I explain the relationship between Whewell's teleological account of life and his natural theology. I conclude with brief comments about the importance of Whewell's classificatory theory for the further development of his philosophy of science and in particular his account of consilience.

Introduction

Snyder, 2006, Snyder, 2011 on William Whewell (1794–1866) includes a brief account of his views on classification. However, this aspect of Whewell's philosophy remains largely unexplored. Though his history and philosophy covered the breadth of natural science, he took pains to familiarize himself with the science of classification in particular (Whewell & Douglas, 1881, p. 122). He travelled to Germany to study with mineralogists and there encountered sophisticated biological classifiers as well (Rieppel, 2016, Whewell and Douglas, 1881, p. 98). Analysis of Whewell's thought on classification thus provides insight into the contemporary theories and practices of natural classification. This is particularly valuable given that the critical overthrow of accounts of pre-Darwinian biological classification as “unscientific” (Amundson, 1998, Wilkins, 2004, Winsor, 2003, Winsor, 2006) have cleared space for positive accounts of just what nineteenth century classification was about (Winsor, 2015). Amundson (2005) demonstrated that pre-Linnaean biologists (and commonfolk and theologians) believed that organisms could change species and that species could change in a variety of ways. Carolus Linnaeus established the belief that transformation of species, or of organisms from one species to another, is impossible (or rare enough that naturalists could ignore it in practice) (Amundson, 2005, p. 17; Locy, 1915, Osborn, 1894, Perrier, 1884, Thomson, 1899). This set up active research questions (Quinn, 2016a): how many forms of natural relationship are there between organisms and between species? How are taxa to be placed within the natural system? What constitutes evidence for relationship?

In this paper I sketch Whewell's attempts to impose order on classificatory mineralogy, which was in Whewell's day a confused science of uncertain prospects. Whewell argued that progress was impeded by the crude reductionist assumption that all macroproperties of crystals could be straightforwardly explained by reference to the crystals' chemical constituents. By comparison with biological classification, Whewell proposed methodological reforms that he claimed would lead to a natural classification of minerals, which in turn would support advances in causal understanding of the properties of minerals. Whewell's comparison to successful biological classification is particularly striking given that classificatory biologists did not share an understanding of the causal structure underlying the natural classification of life (the common descent with modification of all organisms).

Whewell's key proposed methodological reform is consideration of multiple, distinct principles of classification. The most powerful evidence in support of a natural classificatory claim is the consilience of claims arrived at through distinct lines of reasoning, rooted in distinct conceptual approaches to the target objects. Mineralogists must consider not only elemental composition and chemical affinities, but also symmetry and polarity. Geometrical properties are central to what makes an individual mineral the type of mineral that it is. In Whewell's view, function and organization jointly define life, and so are the keys to understanding what makes an organism the type of organism that it is. This theoretical framework enabled biologists' success in making natural classificatory claims. Whewell's account thus provides insight into pre-Darwinian systematists despite their ignorance or rejection of the evolutionary framework that is now central to the success of biological systematics.

First (section 1) I explain Whewell's diagnosis of the problem with his contemporary classificatory mineralogy. I present his analysis of classificatory science, modelled on botany, and show how he developed the method of natural affinity to solve the problem. In section 2 I discuss the justification for this principle in a non-evolutionary context, which Whewell ties to his views on natural theology (section 3). I conclude with brief comments about the importance of Whewell's classificatory theory for the further development of his philosophy of science and in particular his account of consilience.

Section snippets

Natural affinity

Whewell was appointed Chair of Mineralogy at Trinity College in 1828, and published an Essay (Whewell, 1828) on mineralogical classification that same year. In that work and his later History (1837a) and Philosophy of Inductive Sciences (1840b), he contrasted the disorganized state of mineralogy to the success achieved in zoology and especially in botany. He was Chair of Mineralogy until 1832 and was then appointed in moral philosophy. Though he researched a wide variety of philosophical and

Justifying natural affinity

In developing his concept of natural affinity, Whewell attempted to ground this comparative method by specifying the sense of “most important.” Some justification was needed to identify which characters are most important with respect to a function, and which functions are most important with respect to an organism. The requisite sense of importance is not the same as the importance for survival. It is unclear why importance for the organism's needs, preferences, or survival would be relevant

Whewell's natural theology

Historically and philosophically, teleo-functionalist accounts of organisms and natural theological final cause ascriptions are separable. For example, Owen's anatomical research program utilized means-ends reasoning to describe and explain organic forms and functions without reference to Design (Griffiths, 2007; see Lennox, 1992 for discussion and examples, descended from Aristotle). Each part can only be understood in terms of its relation to the system, which implies a teleological cause. A

Conclusion

Consilience is more than simply the agreement of more and more data. The concept of consilience appeals substantively to explanatory virtues (Lipton, 2004). Addition of data, including data of different types, need not in itself lead to the conclusion that a hypothesis has explanatory merits beyond its probable truth.

Several distinct versions of consilience can be drawn from Whewell's work. Whewell describes consilience as “one of the most decisive characteristics of a true theory” (Whewell,

Acknowledgements

I thank Jim Lennox for comments, and Jim Mead, Neal Woodman, and participants at the American Institute of Physics' Third Biennial Early-Career Conference for discussion.

Parts of this research were supported by a grant from the University of Pittsburgh Provost's Development Fund, a Smithsonian Predoctoral Fellowship, and a Postdoctoral Fellowship at California Institute of Technology.

References (74)

  • F.A. Bather

    The anniversary address of the president

    Proceedings of the Geological Society

    (1927)
  • J. Beatty

    Darwin's cyclopean architect

  • Biological Individuality: Integrating scientific, philosophical, and historical perspectives. (2017). (S. Lidgard, & L....
  • A.J. Cain et al.

    Phyletic weighting

    Proceedings of the Zoological Society of London

    (1960)
  • A.P. de Candolle

    Théorie Élémentaire de la Botanique: Ou, Exposition des Principes de la Classification Naturelle et de l'Art de Décrire et d'Étudier les Végétaux

    (1813)
  • A.P. de Candolle et al.

    Elements of the philosophy of Plants: Containing the principles of scientific botany… with a history of the science, and practical illustrations

    (1821)
  • F. Franklin et al.

    Mill's natural kinds

    Mind

    (1888)
  • C. Gegenbaur
  • P.E. Griffiths

    The phenomena of homology

    Biology & Philosophy

    (2007)
  • M.H. Haber

    Species problems

    Metascience

    (2013)
  • P.M. Henson

    Evolution and Taxonomy: John Henry Comstock's Research School in evolutionary entomology at Cornell University 1874-1930

    (1990)
  • P.M. Henson

    The Comstock research school in evolutionary entomology

    Osiris

    (1993)
  • J.F.W. Herschel

    Whewell on inductive sciences

    Quarterly Review

    (1841)
  • J. Jaszczak et al.

    Merelaniite, Mo4Pb4VSbS15, a new molybdenum-essential member of the cylindrite group, from the merelani tanzanite deposit, lelatema mountains, manyara region, Tanzania

    Minerals

    (2016)
  • L. Laudan

    William Whewell on the consilience of inductions

    The Monist

    (1971)
  • J.G. Lennox

    Teleology

  • P. Lipton

    Inference to the best explanation

    (2004)
  • W.A. Locy

    Biology and its makers

    (1915)
  • P.D. Magnus

    Heuristics and biases in evolutionary biology

    Biology and Philosophy

    (1996)
  • P.D. Magnus

    No grist for Mill on natural kinds

    Journal for the History of Analytic Philosophy

    (2014)
  • P.D. Magnus

    John Stuart Mill on taxonomy and natural kinds

    HOPOS: The Journal of the International Society for the History of Philosophy of Science

    (2015)
  • E. Mayr

    The growth of biological Thought: Diversity, evolution, and inheritance

    (1982)
  • E. McMullin

    The inference that makes science

    (1992)
  • J.S. Mill
    (1843)
  • L.A. Necker

    Le Règne Minéral Ramené aux Méthodes de l'Histoire Naturelle

    (1835)
  • A. Novick

    On the origins of the quinarian system of classification

    Journal of the History of Biology

    (2015)
  • L.K. Nyhart et al.

    Individuals at the Center of Biology: Rudolf Leuckart's polymorphismus der individuen and the ongoing narrative of parts and wholes. with an annotated translation

    Journal of the History of Biology

    (2011)
  • Cited by (6)

    View full text