A System of Logic: Ratiocinative and Inductive, 7th Edition, Vol. II — Key Ideas to Explore

Empirical Laws and the Limits of Experience

Mill defines an empirical law as a uniformity of sequence that has not been resolved into any ultimate law. In Chapter XVI, he argues that such laws commonly depend on collocations—the particular arrangements of permanent causes—which themselves are not subject to law. This means that generalizations resting solely on the Method of Agreement can only be received as empirical laws, and they cannot be trusted beyond the bounds of actual experience. Mill's phrasing is precise: he states that derivative laws are 'almost always contingent on collocations' (Chapter XIX), and that the collocations of permanent causes are 'not reducible to any law' (Chapter XVI). This directly challenges any simplistic view of induction as a straightforward accumulation of facts.

Chance, Causation, and the Calculation of Probabilities

Mill devotes two chapters to chance and its elimination, grounding the doctrine of chances ultimately in causation. He defines chance as the absence of law, but insists that the calculation of chances depends on the assumption of causation. In Chapter XVIII, he writes that the doctrine's 'ultimate dependence on causation' is its true foundation, and he provides a theorem relating to the cause of a given event. This is not a mathematical treatise but a philosophical justification: Mill is concerned with how we can eliminate chance to discover residual phenomena. His discussion of the 'elimination of chance' (Chapter XVII) shows a careful attempt to reconcile probabilistic reasoning with deterministic causation.

Classification, Types, and Nomenclature in Science

In the later chapters, Mill turns to classification and nomenclature, using concrete examples from chemistry and natural history. He introduces the concept of a 'type-species' for inductive inquiry—the one that exhibits the phenomenon under investigation 'in the most eminent degree.' This is not a mere taxonomic convenience but a methodological necessity. Mill also praises the nomenclature of elementary chemistry, where the name of a compound expresses its composition, calling it 'an ideally perfect Nomenclature.' He notes that such perfection is rare because most kinds require many characters for distinction. These passages reveal Mill's attention to the practical tools of science, not just abstract principles.

Readers approaching Volume II should note that Mill's argument is cumulative: earlier chapters on empirical laws and chance prepare the ground for his later treatment of classification. The excerpts show a philosopher deeply engaged with the limitations of inductive inference, yet committed to refining its methods. Pay attention to how Mill distinguishes between ultimate laws, derivative laws, and empirical generalizations—this hierarchy is central to his system. The work rewards careful reading of its examples, which often carry as much weight as the theoretical claims.

There’s a particular ache in finishing Mill’s careful stripping away of chance—how he shows that what we call certain is only experience holding its breath. It reminded me, strangely, of Freud’s quieter admissions in Reflections on War and Death — Text and Context, where even our most reasoned frameworks buckle under grief. Some collocations we never fully classify; we just learn to live beside them.