From 1707, the year he was made a fellow of Trinity College Dublin at twenty-two, Berkeley filled notebooks arguing on the page with Descartes, Locke, Malebranche, Newton and Hobbes. The account Trinity had been teaching when he matriculated at fifteen, a boy up from Kilkenny, was the one Locke published in 1690: what a mind meets is never a thing but an idea of a thing. Colour, sound and warmth belong to us; size, shape and motion belong to matter. And matter itself, the stuff doing all the causing, stands permanently behind the curtain, inferred and never once encountered. That picture left two questions nobody could answer: how unthinking stuff produces a thought, and how anyone could ever check.
His solution was to abolish the curtain: nothing stands behind our experience holding it up, because there is nothing for such a thing to be, and for a sensible object to exist is for it to be perceived. He took this to be a rescue of common sense — the philosophers had put an unknowable substance between people and the apple they were eating, and he was taking it away and handing the apple back.
The rest came fast: An Essay Towards a New Theory of Vision in 1709, the Treatise Concerning the Principles of Human Knowledge in 1710, the Three Dialogues in 1713. He was twenty-eight and essentially finished with metaphysics.
The lever was Locke's own distinction: if heat and colour exist only in a perceiver — and Locke granted that — then extension and motion go the same way, since nobody can form an idea of a shape without some colour or other filling it in. Once the two halves fall together, matter has no work left to do. What keeps the world orderly and continuous when no one is looking is not substance but God, in whose perception it stands.
Almost nobody followed him. His first readers largely met the argument with incomprehension and filed its author among the dealers in sceptical paradox — the precise reverse of his intention. He spent the next two decades elsewhere: on De Motu in 1720, on a deanery, and on a college for Bermuda for which the Commons asked the Crown to grant twenty thousand pounds that never came, stranding him for nearly three years in Rhode Island. He divided his library between Harvard and Yale, left a ninety-six-acre farm to Yale, and sailed home.
Then, in 1734, the year he was consecrated Bishop of Cloyne, he turned the same instrument on mathematics. The Analyst asked what Newton's vanishing increments actually were, given that they were treated as real when convenient and as nothing a line later. The mathematicians were furious and Berkeley was right: calculus did not get a rigorous foundation until the limit was properly defined, roughly a century after he raised the objection.
The metaphysics had a longer fuse. Hume took the attack on abstract ideas and pushed it past anything Berkeley wanted, and Kant built his defence of the external world in explicit resistance to it, taking up the question Berkeley had pressed: what a world could be, in itself, apart from any experience of it.