GREEN'S METAPHYSICS OF KNOWLEDGE. 77 the sake of argument that however far we carry back the analysis of what constitutes an object, we still find ourselves dealing with relations ; are we not still compelled to believe that there cannot be relations unless something other than relations exists to be related, even though this " something " (apart from its relations) is "nothing for us as thinking beings " ? And if this be so, does the transcendental theory, in Green's hands, save us after all from the philosophic dualism of which he is so much afraid ? This objection Green attempts to meet, if I understand him rightly, by showing that simple sensations, unqualified by thought, are an unmeaning abstraction, have no place in the world of experience, and cannot therefore, since in truth they are nothiny, be the original cause or matter of anything. This may be true, but it does not really meet the point. In the first place, there is no need for us to suppose, as Green seems to think we must suppose, that this " not-thought " is simple sensation. In the second place, we are not obliged to imagine that by itself and apart from relations it is a possible object of experience. He does not perceive that it is possible to regard the "matter" of thought as necessary without also regarding it as independent. And therefore while he repeats with unwearied iteration that relations are necessary to turn sensations into facts, he seems to consider this a reason for ignoring the correlative and not less obvious truth that sensations, or, if not sensations, material of some kind, are necessary in order that relations may have some- thing to relate. The analogy of symmetry would therefore seem to suggest a dualistic addition to Green's theoiy. On that theory, as we have seen, the world of experience consists of related objects and of these alone, and as it assumes the existence of a single spiritual principle, which is the source of all relation, so it would seem but natural to suppose the existence of a non-spiritual principle which should be the source of that between which the relations are. If it be objected to this that the supposed non- spiritual principle is not an object of possible experience, I reply that neither, on Green's own theory, is the uni- versal spiritual principle an object of possible experience. If I be asked why we should thus suppose the existence of that which is not an object of possible experience, I reply, as Green replies (p. 54) in respect of the spiritual principle, that " its existence is implied in the existence of the world ". In short, if that which is related is, apart from its relations, an unthinkable and therefore non-existent abstraction, not less unthinkable and therefore equally non-