Primary and secondary qualities

First, a piece of vocabulary Locke insists on. A quality is not the same as an idea. An idea is the perception in the mind; a quality is the power in a body to produce that idea in us. The snowball has the qualities; it produces in me the ideas of white, cold, and round. Keep the two apart, and Locke’s whole theory becomes clear (Essay II.viii.8).

Primary qualities. Some qualities are utterly inseparable from body, no matter what you do to it. Divide a grain of wheat again and again — each piece still has solidity, extension, figure (shape), motion or rest, and number. You can never divide it into something that lacks these. These are the primary qualities: they are really in the body itself, in every part of it, at all times (Essay II.viii.9).

The crucial claim about primary qualities is about resemblance: our ideas of them resemble the qualities themselves. My idea of the ball’s roundness genuinely pictures a roundness that is out there in the ball. “The ideas of primary qualities of bodies are resemblances of them, and their patterns do really exist in the bodies themselves” (Essay II.viii.15).

Secondary qualities. Now the twist. Colours, sounds, tastes, smells, and warmth are not, Locke argues, anything resembling our ideas that sits in the object. They are merely powers to produce sensations in us by means of the primary qualities — the “bulk, figure, texture, and motion” of the object’s insensible corpuscles (its imperceptibly tiny parts). The tomato has no redness in it resembling my idea of red; it has a surface microstructure that reflects light so as to cause the idea of red in me (Essay II.viii.10, 13–14).

Locke backs this with arguments from variation. The porphyry stone looks red and white in the light — but put it in the dark and the colours vanish, though nothing in the stone’s real constitution has changed; only the light hitting your eye has. The same water feels warm to a cold hand and cool to a hot hand at the very same moment. The idea can flip while the object stays put — so the idea can’t be a straight copy of a quality resembling it in the thing (Essay II.viii.19–21).

Now the deep consequence — and it’s a problem. Locke’s theory is a form of representative (indirect) realism: the immediate objects of perception are our ideas, and through them we perceive external things. We never confront the tomato directly; we are directly aware only of our idea, which the tomato causes. The mind perceives its own ideas, and the world is inferred as their cause — perception is mediated by a “veil of ideas.”

That question is the sceptical risk. If all I directly perceive is my own ideas, how do I know any of them — even the primary-quality ones — truly resemble a mind-independent world I can never step outside my ideas to inspect? Locke trusts the resemblance for primary qualities; but he has opened a gap between idea and thing. His successors will exploit it ruthlessly: Berkeley will argue the whole primary/secondary distinction collapses (an idea can only resemble another idea), and Hume will turn the veil of ideas into thoroughgoing scepticism about the external world.