Classification of Arts and Early Modern Sciences

The core problem for the sciences regarding classification during the period concerned was how to carve the world into kinds. For instance, while natural philosophers were engaged in debates over how to classify organisms, metaphysicians asked more foundational questions, such as whether there were natural kinds. Did nature come predivided into kinds? If so, then the task of science was merely to reveal these ultimate classes. And how might this task best be done? Was it even possible to ascertain nature's "joints"? Alternatively, if nature does not come already divided, what are the implications for the sciences? Independently of whether there are natural kinds, there remains the question as to whether there is an ideal system for sorting individuals. In the history of science are found the key foundational theories for the contemporary system of scientific nomenclature being developed in this period.

The problem of natural kinds remains in the early twenty-first century. The philosopher John Locke (1632–1704), an antirealist about species (he did not believe that the world came antecedently divided into distinct species/kinds), argued that in principle one can have no access to the "real essences" of things and as a result cannot ever hope to know how reality is "really" divided. Instead, the most for which one can hope is to develop an empirical system of classification based on nominal essences—the names or appearances of things: "the sorting of things is the workmanship of the understanding" (Essay, p. 415). A particular lump of matter is classified as gold because it appears to have the set of properties that have been assigned to the concept of the kind gold. This view was deeply unsatisfying to many, Wilhelm Leibniz (1646–1716) in particular. Leibniz argued that nature had to come prepackaged into kinds and furthermore that there existed some empirical (perhaps even a priori) evidence as to what those kinds in fact are. What is important about this debate is not its resolution—philosophers continue to argue whether there is one yet—but its impact on thinking about classification generally. This debate helped to liberate scientific thinking from the Aristotelian view of classification as definition. It was no longer deemed sufficient to classify the world by simply positing one or several definitions. How the world may be classified into scientific kinds has to obey certain empirical and analytical restrictions.

Much of the work came in response to the practical issue of how best to classify in the emerging sciences. A great deal of urgency was attached to developing coherent systems of classification, especially as human knowledge about the natural world and the variety therein continued to grow. Early modern scientific systems tended to be either artificial (classifying on the basis of convenience for identification) or natural (classifying according to natural kinds). Most of the classification systems in biology during the period were by the "habit" of the kind. So plants were categorized by whether they flower or whether they produce fruit. Animals were classified by whether they lay eggs or are nocturnal, and so on. The most important development, however, was the application of new rational systems of naming kinds. Carolus Linnaeus (1707–1778), a Swedish botanist, devised the precursor to the present system of nomenclature in the eighteenth century (although there were some, such as Jean Bauhin in the sixteenth century, who anticipated this system). His system of binomial nomenclature relied on the division between male and female as one of its fundamental kind distinctions (which is no longer used), but his basic methodology has been adopted as the standard for classification in the biological sciences.

Robert Boyle (1627–1691) is an exemplar of early modern thinkers who helped define "scientific" theories as rational and ordered methodologies. Boyle, now famous for his development of early chemical theories, argued passionately that chemical kinds had to be subject to empirical experimentation. The old chemical categories were deficient precisely because they were not subject to verifiable tests. Boyle further developed the distinction between primary and secondary qualities (though he coined the terms, the concepts can be traced back at least to Galileo), thus preparing the ground for additional scientific inquiry based on a classification of things in nature that were in principle subject to empirical testing. Thus in the debate over how to carve up nature into kinds, new meta-insights emerged that provided constraints on what sorts of classificatory schemes were acceptable. Even if one cannot know whether the particular details about the kinds one picks out in the world are accurate, there nonetheless emerges a theory of classification that indicates that how one classifies is not purely arbitrary.

It is worth noting an issue not addressed by the early moderns but that was fast approaching. All of the reasoning about classification in this period was pre-Darwinian. Phylogenetic systems of classification (those that classify according to evolutionary sequences) did not emerge until later and hence there was no pressure to suppose that there are deep connections between the kinds that are picked out in nature. Thus one of the constraints that would appear after the development of the theory of evolution (that species-kinds might be interrelated in definable ways) was not yet present. But one might speculate that the innovations in theory of classification in the previous century were part of what made evolutionary theory possible. That there are constraints on what could count as a good system of classification prepares one for additional deep connections in certain fields of inquiry.