Skip to the content

Over the Rainbow

By Neil deGrasse Tyson

Natural History Magazine

The astrophysicist’s pot of gold lies along the entire spectrum.

Whenever cartoonists draw biologists, chemists, or engineers, the characters typically wear protective white lab coats that have assorted pens and pencils poking out of the breast pocket. Astrophysicists use plenty of pens and pencils, but we never wear lab coats unless we are building something to launch into space. Our primary laboratory is the cosmos, and unless you have bad luck and get hit by a meteorite, you are not at risk of getting your clothes singed or otherwise sullied by caustic liquids spilling from the sky. Therein lies the challenge. How do you study something that cannot possibly get your clothes dirty? How do astrophysicists know anything about either the universe or its contents if all the objects to be studied are light years away?

Fortunately, the light emanating from a star reveals much more to us than its position in the sky or how bright it is. The atoms of objects that glow lead busy lives. Their little electrons continually absorb and emit light. And if the environment is hot enough, energetic collisions between atoms can jar loose some or all of their electrons, allowing them to scatter light to and fro. All told, atoms leave their fingerprint on the light being studied, which uniquely implicates which chemical elements or molecules are responsible.

As early as 1666, Isaac Newton passed white light through a prism to produce the now-familiar spectrum of seven colors: red, orange, yellow, green, blue, indigo, and violet, which he personally named. (Feel free to call them Roy G. Biv.) Others had played with prisms before. What Newton did next, however, had no precedent. He passed the emergent spectrum of colors back through a second prism and recovered the pure white he started with, demonstrating a remarkable property of light that has no counterpart on the artist’s palette—these same colors of paint, when mixed, would leave you with a color resembling that of sludge. Newton also tried to disperse the colors themselves but found them to be pure. And in spite of the seven names spectral colors change smoothly and continuously from one to the next. The human eye has no capacity to do what prisms do—another window to the universe lay undiscovered before us.

A careful inspection of the Sun’s spectrum, using precision optics and techniques unavailable in Newton’s day, reveals not only Roy G. Biv, but narrow segments within the spectrum where the colors are absent. These “lines” through the light were discovered in 1802 by the English medical chemist William Hyde Wollaston, who naively (though sensibly) suggested that they were naturally occurring boundaries between the colors. A more complete discussion and interpretation followed with the efforts of the German physicist and optician Joseph von Fraunhofer (1787–1826), who devoted his professional career to the quantitative analysis of spectra and to the construction of optical devices that generate them. Fraunhofer is often referred to as the father of modern spectroscopy, but I might further make the claim that he was the father of astrophysics. Between 1814 and 1817, he discovered that when he passed the light of certain flames through a prism, the pattern of lines resembled what he found in the Sun’s spectrum, which further resembled lines found in the spectra of many stars, including Capella, one of the brightest in the nighttime sky. By the mid 1800s the chemists Gustav Kirchhoff and Robert Bunsen (of “Bunsen-burner” fame from your chemistry class) were making a cottage industry of passing the light of burning substances through a prism. They mapped the patterns made by known elements and discovered a host of new elements, including rubidium and caesium. Each element left its own pattern of lines—its own calling card—in the spectrum being studied. So fertile was this enterprise that the second most abundant element in the universe, helium, was discovered in the spectrum of the Sun before it was discovered on Earth. The element’s name bears this history with its prefix derived from “Helios” the Sun.

A detailed and accurate explanation of the role of atoms and their electrons in how spectral lines are formed would not emerge until the era of quantum mechanics a half-century later, but the conceptual leap already been made: Just as Newton’s equations of gravity connected the realm of laboratory physics to the solar system, Fraunhofer connected the realm of laboratory chemistry to the cosmos. The stage was set to identify, for the first time, what chemical elements filled the universe, and under what conditions of temperature and pressure their patterns revealed themselves to the spectroscopist.

Among the more bone-headed statements made by armchair philosophers, we find the following 1835 proclamation:

On the subject of stars, all investigations which are not ultimately reducible to simple visual observations are…necessarily denied to us… We shall never be able by any means to study their chemical composition…I regard any notion concerning the true mean temperature of the various stars as forever denied to us…

Auguste Compte, French Natural Philosopher

Quotes like that can make you afraid to say anything in print.

Just seven years later, in 1842, the Austrian physicist Christian Doppler proposed what became known as the Doppler effect, which is the change in frequency of a wave being emitted by an object in motion. One can think of the moving object as stretching the waves behind it (reducing their frequency) and compressing the waves in front of it (increasing their frequency). The faster the object moves, the more the light is both compressed in front of it and stretched behind it. This simple relationship between speed and frequency has profound implications. If you know what frequency was emitted, but you measure it to have a different value, the difference between the two is a direct indication of the object’s speed toward or away from you. In an 1842 paper, Doppler makes the prescient statement:

It is almost to be accepted with certainty that this [Doppler effect] will in the not too distant future offer astronomers a welcome means to determine the movements… of such stars which… until this moment hardly presented the hope of such measurements and determinations.

The idea works for sound waves, for light waves, and in fact, waves off any origin. (I’d bet Doppler would be surprised to learn that his discovery would one day be used in microwave-based “radar guns” wielded by police officers to extract money from people who drive automobiles above a speed limit set by law.) By 1845, Doppler was conducting experiments with musicians playing tunes on flatbed railway trains, while people with perfect pitch wrote down the changing notes they heard as the train approached and then receded.

During the late 1800s, with the widespread use of spectrographs in astronomy, coupled with the new science of photography, the field of astronomy was reborn as the discipline of astrophysics. One of the preeminent research publications in my field, the Astrophysical Journal, was founded in 1895, and, until 1962, bore the subtitle: An International Review of Spectroscopy and Astronomical Physics. Even today, nearly every paper reporting observations of the universe gives either an analysis of spectra or is heavily influenced by spectroscopic data obtained by others. Since it takes much more light to generate a spectrum of an object than it does to take a snapshot of it, the biggest telescopes in the world, such as the 10-meter Keck telescopes in Hawaii, are tasked primarily with getting spectra. In short, were it not for science’s ability to analyze light by breaking it up into its component colors, we would know next-to-nothing about the universe.

Astrophysics educators surely face a pedagogical challenge of the highest rank. We deduce nearly all of our scientific information about the structure, formation and evolution of things in the universe from the study of spectra, but the analysis of spectra is removed by several levels of inference from the things being studied. Analogies and metaphors help, by linking a complex, somewhat abstract idea to a simpler, more tangible one. The biologist might describe the shape of the DNA molecule as two coils, connected to each other the way rungs on a ladder connect its sides. I can picture a coil. I can picture two coils. I can picture rungs on a ladder. I can therefore picture the molecule’s shape. Each part of the description sits only one level of inference removed from the molecule itself. And they came together nicely to make a tangible image in the mind. However easy or hard it is, one can now talk about the science of the molecule. To explain how we know the speed of a receding star from us requires five nested levels of abstraction:

  • Level 0: Star.
  • Level 1: Picture of a star.
  • Level 2: Light from the picture of a star.
  • Level 3: Spectrum from the light from the picture of a star.
  • Level 4: Patterns of lines lacing the spectrum from the light from the picture of a star.
  • Level 5: Shifts in the patterns of lines in the spectrum from the light from the picture of the star.

Going from Level 0 to Level 1 is a trivial step that we take every day when we get our pictures back from the photo processor. But by the time your explanation reaches level five, your audience is befuddled or just fast asleep. Therein is why the public hardly ever hears about the role of spectra in cosmic discovery—it’s just too far removed from the objects themselves to explain efficiently or with ease. In the design of exhibits for a natural history museum, or for any museum where real things matter, what you typically seek are artifacts for display cases—rocks, bones, tools, fossils, memorabilia, and so forth. All these “Level 0” specimens require little or no cognitive investment before you give the explanation of what the object is. For astrophysics displays, however, we cannot place stars or quasars on display because they would vaporize the Museum.

Most astrophysics exhibits are therefore conceived in Level 1, leading principally to displays of pictures—some quite striking and beautiful. The most famous telescope in modern times, the Hubble Space Telescope, is known to the public primarily through the beautiful, full-color, high-resolution images it has acquired of objects in the universe. The problem here is that after you view such exhibits, you leave waxing poetic about the beauty of the universe yet you are no closer than before to understanding how it all works. To really know the universe requires forays into Levels 3, 4, and 5. While much good science has come from the Hubble Telescope (including the most reliable measure to-date for the expansion rate of the universe) you would never know it from media accounts that the foundation of our cosmic knowledge continues to flow primarily from the analysis of spectra and not from looking at pretty pictures. I want people to wax poetic not only from exposure to Levels 0 and 1, but also from exposure to Level 5, which admittedly requires a greater intellectual investment on the part of the student, but also (and perhaps especially) on the part of the educator.

It’s one thing to see a beautiful color picture, taken in visible light, of a nebula in our own Milky Way galaxy. But it’s another thing to know from its radio wave spectrum that it also harbors newly formed stars of very high mass within its cloud layers. This gas cloud is a stellar nursery, regenerating the light of the universe.

It’s one thing to know that every now and again, high-mass stars explode. Photographs can show you this. But x-ray and visible light spectra of these dying stars reveal cache of heavy elements that enrich the galaxy and are directly traceable to the constituent elements of life on earth. Not only do we live among the stars, the stars live within us.

It’s one thing to look at a poster of a pretty spiral galaxy. But it’s another thing to know from Doppler shifts in its spectral features that the galaxy is rotating at 200 kilometers per second, from which we infer the presence of 100 billion stars using Newton’s laws of gravity. And by the way, the galaxy is receding from us a 110 the speed of light as part of the expansion of the universe.

It’s one thing to look at nearby stars that resemble the Sun in luminosity and temperature. But it’s another thing to use hyper-sensitive Doppler measurements of the star’s motion to infer the existence of planets in orbit around them. At press time, our catalog is rising though a hundred such planets outside the familiar ones in our own solar system.

Fortunately, for the magnetohydrodynamicists among us, atomic structure changes slightly under the influence of a magnetic field. This change can be seen in the slightly altered spectral pattern caused by these magnetically afflicted atoms.

It’s one thing to observe the light from a quasar at the edge of the universe. But its another thing entirely to analyze the quasar’s spectrum and deduce the structure of the invisible universe, laid along the quasar’s path of light as gas clouds and other obstructions take their bite our of the quasar spectra.

And armed with Einstein’s relativistic version of the Doppler formula, we deduce the expansion rate of the entire universe from the spectra of countless galaxies near and far, and thus infer the age and fate of the universe.

One could make a compelling argument that we know more about the universe than marine biologist knows about the bottom of the ocean or the geologist knows about the center of the Earth. Far from an existence as powerless stargazers, modern astrophysicists are armed to the teeth with the tools and techniques of spectroscopy, enabling us all to stay firmly planted on Earth, yet finally touch the stars without burning our fingers, and claim to know them as never before.