Methods of Investigation

There are three ways of treating a subject: --

In the first mode, you begin with a definition, and that definition is necessarily assumed as the truth. As the argument proceeds, the conclusion from the first proposition becomes the base of the second, and so on. Now, it is quite impossible that you can be sure that you have included all the necessary, and none but the necessary, terms in your definition; as, therefore, you proceed, the original speck of error is multiplied at every remove; the same infirmity of knowledge besetting each successive definition. Hence you may set out, like Spinosa, with all but the truth, and end with a conclusion which is altogether monstrous; and yet the mere deduction shall be irrefragable. Warburton's "Divine Legation" is also a splendid instance of this mode of discussion, and of its inability to lead to the truth: in fact, it is an attempt to adopt the mathematical series of proof, in forgetfulness that the mathematician is sure of the truth of his definition at each remove, because he _creates _it, as he can do, in pure figure and number. But you cannot _make _any thing true which results from, or is connected with, real externals; you can only _find _it out. The chief use of this first mode of discussion is to sharpen the wit, for which purpose it is the best exercitation.