40

Logic II 40

{Section title: Natural Classification does not proceed by Definition.}

items of his thought. For the living thought, itself, in its
entirety, the reader has to dig into his own soul. I
think I have done my part, as well as I can. I am
sorry to have left the reader an irksome chore before
him. But he will find it worth the doing.

So then, a natural class being a family whose
members are the sole offspring and vehicles of one
idea, from whch they derive their peculiar faculty, to classify
by abstract definitions is simple a sure means of
avoiding a natural classification. I am not decrying
definitions. I have a lively sense of their great value in
science. I only say that it should not be by manes of
definitions that one should seek to find natural classes.
When the classes have been found, then it is proper to try
to define them; and one may even, with great caution
and reserve, allow the definitions to lead us to turn
back and see whether our classes should be ought not to have