Opticks

Overview

The Opticks is four hundred pages, published in 1704, and it is nothing like the Principia. That book is in Latin, proceeds by geometry, and is impenetrable; Newton said himself he had made it difficult on purpose to avoid being pestered by people who did not understand mathematics. This one is in English, describes experiments in enough detail to be repeated, and can be followed by anyone patient.

The delay in publishing is part of the story. Newton had done the crucial prism work in 1665 and 1666, during the plague years, and published a paper on it in 1672 that provoked a long and bad-tempered dispute, particularly with Robert Hooke. He withdrew, sat on the material for three decades, and brought out the book the year after Hooke died.

The structure is Euclidean. Newton opens with definitions and axioms, then proceeds through propositions, each supported by numbered experiments rather than by constructions. He is explicit at the start that he will not explain light by hypotheses but prove its properties by reason and experiment, which is the methodological claim the whole Enlightenment took from him.

Then, at the end, he does the opposite. The Queries are a series of speculative questions, thirty-one of them by the final edition, in which Newton floats everything he suspects and cannot show. Keeping them separate, and phrasing them as questions, is how he squares speculation with his own rule.

The argument

Book One: white light is a mixture. The central experiment is simple enough to have been done in schoolrooms ever since. Newton admits a thin beam of sunlight through a hole in a shutter into a darkened room and passes it through a prism. The received view held that white light is pure and that a prism modifies it, adding colour. If that were so, the patch on the far wall should be round, like the beam. It is not. It is an elongated spectrum about five times longer than it is wide.

Newton’s explanation is that white light is not simple at all but a mixture of rays that are bent by different amounts, so the prism does not change the light, it separates what was already there. To prove it he devises what he calls the experimentum crucis, the crossroads experiment. He lets the spectrum fall on a board with a small hole, so that one colour alone passes through, and sends that single ray into a second prism. If the prism manufactured colour, the second prism should change it again. It does not. The ray is bent further, by the amount characteristic of its colour, and comes out the same colour it went in. Each ray has a fixed refrangibility and a fixed colour, and neither can be altered by refraction, reflection or anything else he can do to it.

He then completes the case by recombining. Passing the spectrum through a lens or a second, inverted prism reconstitutes white light, which shows that white is the compound rather than the element.

From this follows a practical consequence Newton cared about: a lens will always disperse colours as it refracts, so a refracting telescope will always show coloured fringes. He concluded the defect was incurable and built a reflecting telescope instead, using a mirror. He was wrong that it was incurable, since achromatic lenses combining different glasses were made half a century later, but the reflector is the design every large telescope uses.

He also divides the spectrum into seven named colours, adding indigo between blue and violet, explicitly so that the number should match the seven intervals of the musical octave. It is the one piece of inherited harmony in a book otherwise built on measurement, and it is why schoolchildren still learn a colour most people cannot distinguish.

Book Two: thin films and periodicity. The most technically impressive part. Newton studies the colours produced by thin transparent films, soap bubbles, the air gap between a lens and a flat plate, and measures with great precision the rings that appear, which are still called Newton’s rings.

What he finds is that the colours depend on the thickness of the film, and repeat periodically as the thickness increases. This is straightforwardly a wave phenomenon, and it is the strongest evidence for a wave theory available at the time. Newton, committed to light as a stream of particles, explains it with what he calls fits of easy reflection and easy transmission, a periodic disposition that a ray falls into along its path. It is an awkward hybrid, and it is honest: he had measurements that his own theory could not naturally accommodate, and he published them.

Book Three: diffraction and the Queries. A shorter book on what Newton calls inflexion, the bending of light at the edges of obstacles, which he investigates and does not resolve.

Then the Queries, which are the most-read part of the book today. They range enormously. Whether bodies act on light at a distance and bend its rays. Whether light acts on bodies to produce heat by setting their parts into vibration. Whether the rays are small bodies emitted from shining substances. What ether might be, and whether it explains gravity. Whether animals and plants are produced by chemical forces. And, in the late Queries added in 1717, a series of speculations on the forces between the small parts of bodies that effectively founded eighteenth-century chemistry’s research programme. Query 31 ends with remarks on method and on natural philosophy’s bearing on moral philosophy.

How to read it

Read the opening definitions and axioms, then Book One Part One entire. That is the prism work and the argument, and it is perhaps eighty pages.

Do the experiment if you can. A glass prism is cheap, and the difference between seeing a spectrum and reading about one is considerable.

Skim Book Two unless the measurements interest you, but read Newton’s account of the fits. Watching a great scientist build an ungainly mechanism to accommodate data that does not fit his theory is more instructive than any account of the scientific method.

Read the Queries. They are the best writing in the book and the clearest window into what Newton actually believed as opposed to what he was prepared to claim.

Keep the methodological slogan in proportion. Newton says he does not feign hypotheses, and the Queries are thirty-one pages of hypotheses. The distinction he is drawing is between what the experiments establish and what he suspects, and the care with which he separates them is the point rather than the abstinence.

Questions it raises

• Newton says he proves properties by experiment rather than explaining by hypotheses. Are the Queries consistent with that, or a confession that the rule cannot be kept?
• The crossroads experiment is meant to be decisive. What alternative explanation could have survived it, and did any?
• Newton had periodic data that fitted waves and held to particles. What was he weighing, and would you have decided differently?
• Colour is said to belong to the observer and refrangibility to the ray. Does that distinction hold, and what does it imply about other sensible qualities?
• Why insert indigo to make seven? What does that tell you about the persistence of harmonic thinking in a quantitative science?
• The eighteenth century learned its idea of scientific method more from this book than from the Principia. Was that the right lesson to draw?