“But it is all based on assumptions.”

“That may be; but granting the assumptions, it deduces from them necessary consequences.  And all true science is of that type.  A law of Nature is not a mere description of a routine; it’s a statement that, given such and such conditions, such and such results follow of necessity.”

“Still, you admit that the conditions have to be given!  Everything is based ultimately on certain successions and coincidences of which all that can be said is simply that they exist, without any possibility of getting behind them.”

“I don’t know about that,” he said, “but at any rate it would be the ideal of Knowledge to establish necessary connections throughout; so that, given any one phenomenon of the universe, all the rest would inevitably follow.  And it is only in so far as it progresses towards this consummation that Knowledge is Knowledge at all.  A routine simply given without internal coherence is to my mind a contradiction in terms; either the routine is necessary, or it’s not a routine at all, but at best a mere appearance of a routine.”

“I think,” I interposed, “we must leave you and Wilson to fight this out in private.  At present, let us assume that your conception of Knowledge is the true one, as we did with his, and examine it from the point of view of the Good.  Your conception, then, to begin with, seems to me to be involved in the same defect we have already noted—­namely, that it may be knowledge of Bad just as much as knowledge of Good.  And I suppose you would hardly maintain, any more than Wilson did, that the Good may consist in knowledge of Bad?”

“But,” he objected, “I protest altogether against this notion that there is Knowledge on the one hand and something of which there is knowledge on the other.  True Knowledge, if ever we could attain to it, would be a unique kind of activity, in which there would be no distinction, or at least no antagonism, between thinking on the one hand and the thing thought on the other.”

“I don’t know,” I said, “that I quite understand.  Have we in fact any knowledge of that kind, that might serve as a kind of type of what you mean?”

“Yes,” he replied, “I think we have.  For example, if we are dealing with pure number, as in arithmetic, we have an object which is somehow native to our thought, commensurate with it, or however you like to put it; and it is the same with other abstract notions, such as substance and causation.”

“I see,” I said.  “And on the other hand, the element which is alien to thought, and which is the cause of the impurity of most of what we call knowledge, is the element of sense—­the something given, which thought cannot, as it were, digest, though it may dress and serve it up in its own sauce?”

“Yes,” he said, “that is my idea.”

“So that knowledge, to be perfect, must not be of sense, but only of pure thought, as Plato suggested long ago?”