Page:Popular Science Monthly Volume 28.djvu/262

252 effect; and these effects being invariably connected will, by ignorant people, be regarded as cause and effect, which they will not be. In fact, the reference of one phenomenon to another as its cause, in consequence of invariable sequence, may have the same essential error involved in it as had the classical example of Tenterden Steeple and the Goodwin Sands.

What is necessary in order that one thing shall be regarded as the effect of another, which may be called the cause, is not only that there shall be an invariable sequence, but also that it shall be possible to assert that the one could not take place without the other, or something equivalent. This invisible, impalpable chain between the one thing and the other must be postulated by the human mind: it constitutes the idea of cause; every child knows perfectly well what it is, and the profoundest philosopher does not go far, if at all, beyond the knowledge of the child.

Let me support what I have been saying by a quotation from Whewell's "Philosophy of the Inductive Sciences":

We see in the world around us a constant succession of causes and effects connected with each other. The laws of this connection we learn in a great measure from experience, by observation of the occurrences which present themselves to our notice, succeeding one another. But in doing this, and in attending to this succession of appearances, of which we are aware by means of our senses, we supply from our mind the idea of cause. This idea, as we have already shown with respect to other ideas, is not derived from experience, but has its origin in the mind itself; is introduced into our experience by the active not by the passive part of our nature.

And again Dr. Whewell writes:

That this idea of cause is not derived from experience, we prove (as in former cases) by this consideration: that we can make assertions, involving this idea, which are rigorously necessary and universal; whereas knowledge derived from experience can only be true as far as experience goes, and can never contain in itself any evidence whatever of its necessity. We assert that "every event must have a cause"; and this proposition we know to be true, not only probably and generally, and as far as we can see; but we can not suppose it to be false in any single instance. We are as certain of it as of the truths of arithmetic or geometry.

Here is a true postulate; and if to the postulate that every event must have a cause we add these postulates, (1) that causes in Nature are always of the same kind and always act in the same way, and (2) that no new causes come into existence, we should go a long way toward making the uniformity of Nature, if not axiomatic, at all events capable of tolerably simple and satisfactory demonstration.

But these latter postulates will perhaps scarcely be universally granted. I understand those disputants, who in the Metaphysical Society's discussion laid so much stress upon the duty of examining