qualities in no other way. They are found to be adjectives, somehow supervening on relations of the extended. The extended only is real. And the facts of what is called subjective sensation, under which we may include dream and delusion of all kinds, may be adduced in support. They go to show that, as we can have the sensation without the object, and the object without the sensation, the one cannot possibly be a quality of the other. The secondary qualities, therefore, are appearance, coming from the reality, which itself has no quality but extension.
This argument has two sides, a negative and a positive. The first denies that secondary qualities are the actual nature of things, the second goes on to make an affirmation about the primary. I will enquire first if the negative assertion is justified. I will not dispute the truth of the principle that, if a thing has a quality, it must have it; but I will ask whether on this basis some defence may not be made. And we may attempt it in this way. All the arguments, we may protest, do but show defect in, or interference with, the organ of perception. The fact that I cannot receive the secondary qualities except under certain conditions, fails to prove that they are not there and existing in the thing. And, supposing that they are there, still the argument proves their absence, and is hence unsound. And sheer delusion and dreams do not overthrow this defence. The qualities are constant in the things themselves; and, if they fail to impart themselves, or impart themselves wrongly, that is always due to something outside their nature. If we could perceive them, they are there.
But this way of defence seems hardly tenable. For, if the qualities impart themselves never except under conditions, how in the end are we to say what they are when unconditioned? Having once begun, and having been compelled, to take their
