It often happens in philosophy that the half-truth takes its place beside the whole truth and assumes on its own account the position of something permanent. But the fact is that the half-truth, instead of being a fixed or self-subsistent principle, is a mere element absolved and included in the whole. The metaphysic of understanding is dogmatic, because it maintains half-truths in their isolation: whereas the idealism of speculative philosophy carries out the principle of totality and shows that it can reach beyond the inadequate formularies of abstract thought. Thus idealism would say:--The soul is neither finite only, nor infinite only; it is really the one just as much as the other, and in that way neither the one nor the other. In other words; such formularies in their isolation are inadmissible, and only come into account as formative elements in a larger notion. Such idealism we see even in the ordinary phases of consciousness. Thus we say of sensible things, that they are changeable: that is, they _are,_ but it is equally true that they are _not._ We show more obstinacy in dealing with the categories of the understanding. These are terms which we believe to be somewhat firmer--or even absolutely firm and fast. We look upon them as separated from each other by an infinite chasm, so that opposite categories can never get at each other. The battle of reason is the struggle to break up the rigidity to which the understanding has reduced everything. 33.] The _first_ part of this metaphysic in its systematic form is Ontology, or the doctrine of the abstract characteristics of Being. The multitude of these characteristics, and the limits set to their applicability, are not founded upon any principle. They have in consequence to be enumerated as experience and circumstances direct, and the import ascribed to them is founded only upon common sensualised conceptions, upon assertions that particular words are used in a particular sense, and even perhaps upon etymology. If experience pronounces the list to be complete, and if the usage of language, by its agreement, shows the analysis to be correct, the metaphysician is satisfied; and the intrinsic and independent truth and necessity of such characteristics is never made a matter of investigation at all.