Newton: Optics

When Isaac Newton was a young man, he bought a glass prism at a fair. He darkened his room, made a small hole in the window shutter, and let a thin beam of sunlight pass through the prism. On the far wall appeared a long band of colors: red, orange, yellow, green, blue and violet.

People in Newton's day thought the prism added the colors to the light, the way paint adds color to a wall. Newton thought the opposite: the colors were already hiding inside the white sunlight, and the prism only sorted them, bending each color by a different amount.

To prove it, he used a second prism. He let only one color, say red, pass through a little hole and into the second prism. If prisms added colors, more colors should come out. Instead, red went in and red came out, just bent a bit more. And when he put all the colors back together with a lens, he got white light again!

Who, when and the problem

Newton did his first prism experiments in 1665–1666, while still a young Cambridge graduate, and lectured on optics as professor from 1670. In 1672 he sent the Royal Society a short paper, "A New Theory about Light and Colours," and a small reflecting telescope he had built. The paper started years of arguments with Robert Hooke, Christiaan Huygens and others, and Newton disliked the quarrels so much that he waited to publish a full book until after Hooke's death. Opticks appeared in English in 1704; a Latin edition followed in 1706, and Newton's last corrections went into the fourth edition of 1730.

The problem: what is color? The common view, going back to Aristotle and held in new forms by Descartes and Hooke, was that white light is pure and simple and that colors are made by changing it, for example by mixing it with darkness at the edge of a prism. Newton set out to settle the question by experiment instead of by argument.

The main ideas

• White light is a mixture. Sunlight is made of rays that bend by different amounts in glass ("differently refrangible"), and each kind of ray has its own color.
• A prism sorts, it does not create. A single color separated out by one prism is not changed by a second one: it bends again by its own amount and keeps its color.
• Colors can be mixed back into white, and Newton arranged the colors around a circle to predict the results of mixing them.
• The colors of objects come from which rays they reflect most strongly. A red apple reflects red rays more than others.
• Thin films and rings. In Book II Newton measured the colored rings formed in the thin layer of air between a lens and a flat glass, now called Newton's rings, and found they repeat at regular thicknesses.
• The Queries. Book III ends with a series of questions about light, heat, gravity, chemistry and the nature of matter, which became a research program for the next century.

Colors are in us, as well as in the light

Newton is careful here. Red is not a paint on the ray. A ray has a certain power to make us see red. That distinction between the physics of light and the sensation of color is still how color science thinks today.

Why seven colors?

Newton named seven colors: red, orange, yellow, green, blue, indigo and violet. The spectrum itself has no sharp edges, and in early work he named five. In Opticks he divided it in the same proportions as the notes of a musical scale, which has seven steps before the octave. So the "seven colors of the rainbow" are partly a choice, made by a man who also loved the idea that nature's patterns rhyme.

Setting the record straight

• Newton was not the first to make a spectrum. People had seen prism colors for centuries, and Descartes had explained the rainbow's angle in 1637. What was new was Newton's proof that white light is a mixture, and his measurements.
• Newton was wrong about telescope lenses. He concluded that the colored fringes made by lenses could never be removed, so he turned to mirrors. In 1733 Chester Moor Hall, and in 1758 John Dollond, made "achromatic" lenses from two kinds of glass that cancel most of the color. The mistake came from assuming all glass and water spread colors in the same proportion.
• Particles or waves? Newton leaned toward light being made of tiny bodies, but he put this as a question (Query 29), not a claim, and he also used waves of some kind to explain his rings.

What later science kept, and what it changed

• Kept: white light as a mixture, dispersion, colors from selective reflection, and the color circle, which led to the modern color wheel and to the idea that three primary lights can make most colors.
• Changed: Thomas Young (1801–1803) used Newton's own ring measurements to work out the wavelengths of light, showing that each color is a wave of a different length: about 700 billionths of a meter for red, 400 for violet. In the 1860s James Clerk Maxwell showed light is an electromagnetic wave.
• Then changed again: in 1905 Einstein proposed that light also comes in packets of energy, photons. Today light is described by quantum theory, which has room for both pictures. Newton's guess about particles was not simply wrong, nor was Huygens's wave picture.

A different style of science. The Principia is written like geometry. Opticks is written in plain English, as a series of experiments that anyone with a dark room and a prism could repeat. It became a model for the experimental science of the 18th century, as the Principia was for the mathematical. Benjamin Franklin and many others learned their science from it.

The Queries. In the later editions the book ends with 31 Queries. They ask whether heat is a vibration, whether bodies and light turn into each other, whether chemical changes come from forces between small particles, and whether an ether fills space. Newton framed them as questions on purpose: these were things he could not yet prove. Lavoisier's chemistry and Faraday's fields, later on this ladder, took up several of them.

Color and the poets. Some writers felt Newton had "unwoven the rainbow." Goethe, in his Theory of Colours (1810), rejected Newton's theory and studied color as an experience instead. Physicists sided with Newton on the physics, but Goethe's careful descriptions of how we see colors are still read by artists and psychologists. Is something lost when we explain a rainbow? Or gained?

Compare: Huygens's Treatise on Light, the next rung, explains reflection and refraction by waves, but says nothing about color. Newton explains color, but struggles with why light bends. Each had the piece the other lacked.