The Nineteenth Century, a Period of Change 119
be preceded by definition. However, he equivocates when discussing essential prop-
erties, and says that “to the genus plant belong all those entities which are endowed
with the property of possessing leaves, stalks, roots, &c.” Even so, he realizes that
classification, which he calls methodization, is either physical or logical^13 and that
division is of several kinds, analytic when performed on individuals, logical when
performed on collective entities, and possibly also physical. There is a difference
between dividing, say, Vertebrata into mammals, birds, fishes, lizards, and serpents
in terms of obvious features if you want a “slight and general idea” of vertebrates,
but “it cannot suffice for the naturalist, who must always be assured of the all-
comprehensiveness of his classes and subclasses; he must always be enabled to ascertain
precisely to which of them he should refer any individual animal that comes under
his observation.”^14 So, he applies a binary dichotomy of lungs/no lungs (fish), with
mammaries/not with mammaries (mammals), winged (birds) not winged (reptiles).^15
In the rest of his work, he seems to have reinvented the Universal Language
Project. His uncle, Jeremy Bentham, in his 1817 Chrestomathia, revives, explicitly,^16
Porphyry’s dichotomous (“bifurcating”) mode of classification, counterpointing
it to D’Alembert’s classification of knowledge in the Encyclopedia. His division
of knowledge^17 begins with “Eudaemonics, or Ontology,” which he divides into
“Coenoscopic” and “Idioscopic” ontologies (general and particular properties), and
so on, but lower down he starts introducing privative categories, such as “no-work-
producing,” “not-state-producing,” and so on. Moreover, some of his dichotomies,
such as Nature/Man, information/passion, seem as arbitrary as Plato’s original. Still,
Jeremy Bentham’s recasting of the Arbor Porphyriana as a bifurcating tree diagram
is significant.^18
By the middle of the nineteenth century, despite the arguments naturalists were
now having over the meaning of the term species, the “genera plus differentia” defi-
nition remained widely accepted by logicians until, under the weight of the new set
theory and the biological pre-eminence of the use of the term, the older logic was
relegated to specialists in metaphysics and medievalists. Here, for example, is the
definition of a widely used dictionary of science and the arts in 1852:
SPECIES. (Lat.) In Logic, a predicable which is considered as expressing the whole
essence of the individuals of which it is affirmed. The essence of an individual is said
to consist of two parts: 1. The material part, or genus; 2. The formal or distinctive part,
or difference. The genus and difference together make up, in logical language, the spe-
cies: e.g. a “biped” is compounded of the genus “animal,” and the difference “having
two legs.” It is obvious that the names species and genus are merely relative; and that(^13) Bentham, op. cit., 98.
(^14) Bentham, op. cit., 112.
(^15) Bentham, op. cit., 114.
(^16) Bentham 1983, Table IV.
(^17) Bentham 1983, Table V.
(^18) For a more comprehensive overview, but flawed in several cases I believe, of the Benthams’ logical
enterprise, see McOuat 2003. The flaws relate to identifying fixism with essentialism (McOuat agrees
in communication), and to a lesser extent not recognizing the much older tradition of the debate over
binary privative logic versus multiple species within genera. However, the paper has a much wider
agenda, and these are minor problems. Thanks to Charissa Varma for sending me this paper.