But the same disinclination to repeat myself word for word, or to say the same thing a second time in other and worse words, after I have deprived myself of the better, has occasioned another defect in the first book of this work. For I have omitted all that is said in the first chapter of my essay “On Sight and Colour,” which would otherwise have found its place here, word for word. Therefore the knowledge of this short, earlier work is also presupposed. Finally, the third demand I have to make on the reader might indeed be tacitly assumed, for it is nothing but an acquaintance with the most important phenomenon that has appeared in philosophy for two thousand years, and that lies so near us: I mean the principal writings of Kant. It seems to me, in fact, as indeed has already been said by others, that the effect these writings produce in the mind to which they truly speak is very like that of the operation for cataract on a blind man: and if we wish to pursue the simile further, the aim of my own work may be described by saying that I have sought to put into the hands of those upon whom that operation has been successfully performed a pair of spectacles suitable to eyes that have recovered their sight—spectacles of whose use that operation is the absolutely necessary condition. Starting then, as I do to a large extent, from what has been accomplished by the great Kant, I have yet been enabled, just on account of my earnest study of his writings, to discover important errors in them. These I have been obliged to separate from the rest and prove to be false, in order that I might be able to presuppose and apply what is true and excellent in his doctrine, pure and freed from error. But not to interrupt and complicate my own exposition by a constant polemic against Kant, I have relegated this to a special appendix. It follows then, from what has been said, that my work presupposes a knowledge of this appendix just as much as it presupposes a knowledge of the philosophy of Kant; and in this respect it would therefore be advisable to read the appendix first, all the more as its content is specially related to the first book of the present work. On the other hand, it could not be avoided, from the nature of the case, that here and there the appendix also should refer to the text of the work; and the only result of this is, that the appendix, as well as the principal part of the work, must be read twice.