The proponents of the idea of a global climate crisis are fond of saying that when it comes to climate change, “the science is settled.” But in fact that is just a way of telling their opposition to shut up.

By definition, science can never be settled. Consensus does not constitute truth. Indeed every scientific theory, no matter how widely accepted or how often confirmed by experiment, always hangs by a thread. Only one prediction made by the theory that turns out not to be true is needed to negate it in part or in whole.

That’s how, in a nutshell, the scientific method works: first, gather data; second, use the data to form a hypothesis that can explain the data; third, test the hypothesis.

For instance, Europeans had long assumed that all species of swans are white, as in the northern hemisphere they all are. The Roman poet Juvenal coined the term “black swan” as an example of the ultimate rara avis. But when Europeans got to Australia, where the swans are in fact black, there went the hypothesis that all swans are white.

The night sky, one of nature’s greatest glories, now mostly lost in the glare of electric light, has fascinated humankind since time immemorial. While the vast majority of the lights in the sky do not move relative to each other, there are seven that do. The Greeks called them planets, from the Greek word planasthai, which means “to wander.” These seven astronomical misfits were the sun, the moon, and five bodies we call “planets” today: Mercury, Venus, Mars, Jupiter, and Saturn.

It was quickly noticed that these bodies did not actually wander at random. Instead, they moved in apparently complicated paths before returning to the same place in the sky over and over. What could explain this?

While a heliocentric model of the universe had been proposed as early as the third century B.C., by Aristarchus of Samos (ca. 310–ca. 230 B.C.), his predecessors Plato and Aristotle and the later astronomer and geographer Ptolemy (ca. A.D. 100–ca. 170) all favored a geocentric one. This was, perhaps, due to human ego, but also perhaps to our being hard-wired to perceive the earth and everything attached to it as unmoving. (After all, for hunters on the African savanna, there was a considerable evolutionary advantage to comprehending instantly that it was not the bush that was moving but the leopard behind it.)

Ptolemy’s treatise on astronomy, remembered as the Almagest, was one of the most influential scientific textbooks of all time and had canonical authority during the Middle Ages. The fact that it fit perfectly with the Church’s theology made it even more authoritative.

The mathematics was complicated and inelegant.

One problem for Ptolemy’s system is the so-called retrograde motion of the outer planets. Today, we know this is due to the fact that the earth, moving faster around a smaller orbit, laps the outer planets, causing them to appear to move backwards for a period against the fixed stars. Ptolemy got around this problem by proposing epicycles—essentially, a planet’s orbiting around a point that circulates on a larger orbit—and other complications to make the hypothesis fit the observed facts. Using Ptolemy’s system, it was possible to calculate with some precision the future locations of the planets in the sky. But the mathematics was complicated and inelegant, to put it mildly.

In 1543, Nicolaus Copernicus (1473–1543), a Polish polymath and clergyman, proposed a new heliocentric model of the universe. (He was apparently unaware of Aristarchus of Samos’s work.) It greatly reduced the amount of mathematical calculations needed to determine planetary orbits, although it did still rely on epicycles because Copernicus assumed that all bodies orbit in perfect circles.

The Copernican system was not without its critics. Where was the evidence that the earth rotated? And it was pointed out that if the earth circled the sun, then the nearer stars should appear to shift over the course of a year in relation to more distant ones, a phenomenon known as parallax. No such stellar parallax could be observed.

In fact, the parallax of stars was not measured until the 1830s, long after the heliocentric model had been universally accepted. The stars turned out to be much farther away than had been thought. No star is close enough to earth to show a parallax of even one second of arc. The proof of the earth’s daily rotation, accounting for the sun’s apparent motion, took even longer. It was only in 1851 that Léon Foucault used a free-swinging pendulum to demonstrate it.

Tycho Brahe (1546–1601), a Dane and the last great astronomer before the invention of the telescope in the early seventeenth century, relied on precise instruments to develop unprecedentedly accurate astronomical data. Johannes Kepler (1571–1630), Brahe’s assistant for a year at the end of Brahe’s life, used this data to show that astronomical bodies do not move in perfect circles but in ellipses, eliminating the need for epicycles.

Isaac Newton (1643–1727) then brought it all together in 1687 in perhaps the most famous scientific book ever written, the Principia Mathematica, in which he promulgated his laws of motion and of universal gravitation. Unlike Ptolemy’s clunky model of the universe, Newton’s was elegant in the extreme, passing the test of Occam’s Razor. No scientific theory has ever come as close to being “settled” as Newton’s theory of universal gravitation.

Newton’s hypothesis was that gravity is a force of attraction between objects that is proportional to the product of their masses and inversely proportional to the square of the distance between them. The predictive power of this hypothesis was so great, confirming the hypothesis over and over again, that Newton’s near-contemporary, the poet Alexander Pope, wrote that

Nature and Nature’s laws lay hid in night.

God said, “Let Newton be!” and all was light.

Newton’s majestic theory (which is what a hypothesis becomes after it has been rigorously and repeatedly confirmed) allowed astronomers to predict the orbits of astronomical bodies with unprecedented precision. This came in handy when the planet Uranus was found by William Herschel in 1781. Uranus had been catalogued previously as a star, perhaps as early as ancient times, for it is visible to the naked eye under excellent seeing conditions. But Herschel noted that it was moving relative to the background stars and therefore could not be a star itself.

Herschel at first thought he had found a comet. But when Anders Johan Lexell, a Swede working at the Russian Academy of Sciences, used Newton’s equations to determine its orbit, it was found to be nearly circular, like planetary orbits, not highly eccentric as with comets. Lexell calculated that Uranus would take almost exactly eighty-four years to revolve around the sun. In 1783, the great French polymath Pierre-Simon Laplace (sometimes known as “the French Newton”) worked out all six orbital elements of Uranus.

But as Uranus advanced around the sun, astronomers noticed that the seventh planet was not exactly where Newton’s theory predicted it would be. It wasn’t off by much, but the gap was measurable. As telescopes improved rapidly in the early nineteenth century, the discrepancy became ever more obvious. There were only two possibilities: either Newton was wrong, or, as the British mathematician John Couch Adams suggested in 1841, another body was perturbing Uranus’s orbit.

François Arago, the director of the Paris Observatory, encouraged the mathematician Urbain Le Verrier, who specialized in celestial mechanics, to calculate the position of a planet that could cause the observed orbital anomaly of Uranus. On August 31, 1846, Le Verrier announced to the Académie Française that his calculations were complete. On September 18, he sent them to the astronomer Johann Galle of the Berlin Observatory, where the letter arrived on September 23.

That very night, Galle found the new planet, now called Neptune, within one degree of Le Verrier’s postulated position, making himself and Le Verrier world-famous almost overnight. It was considered the greatest triumph yet of Newton’s theory of universal gravitation.

Then, in 1859, Le Verrier reported that the observed orbit of the planet Mercury was at slight variance with Newtonian theory. Mercury, the nearest planet to the sun, has the most eccentric orbit of all the planets. During its eighty-eight-day orbit, Mercury gets as close to the sun as 29 million miles—this point is called the perihelion—and as distant as 43 million miles. More, its perihelion precesses around the sun at a rate calculated by Le Verrier as 1.5556 degrees per century.

But Newton’s theory predicted that it would precess at 1.5436 degrees per century. In other words, Mercury’s perihelion was precessing 360 degrees every 23,142 years, while Newton said it should be taking 23,322 years.

Le Verrier suggested that, as with Uranus, there must be an undiscovered planet, this time closer to the sun, perturbing Mercury’s orbit. Given Le Verrier’s triumph in predicting Neptune, astronomers all over the world vied for the honor of discovering the next new planet. They even gave it a name, Vulcan, after the Roman god of fire.

But while Neptune had been quickly found in the utter blackness of the outer solar system, Vulcan, it was reckoned, would be hiding in the glare of the sun. That would make it not only difficult to find, but even dangerous to look for. Except at dawn and sunset, you cannot look directly at the sun with the naked eye for more than a fraction of a second because of its dazzling brilliance. Accidentally imaging the sun with a telescope, without a suitable filter, would blind you.

For the next several decades, astronomers looked in vain for the elusive Vulcan. But they kept trying, for Newton’s celestial mechanics said it must be there, and the science was, well, settled.

Then, in 1915, Albert Einstein, the greatest physicist since Newton, published his General Theory of Relativity. Einstein had a radically different idea regarding the nature of space than Newton, who had written in the Principia that he assumed that space was everywhere and always the same. Einstein said that space-time would be warped by the presence of very massive objects, such as the sun.

Using Einstein’s equations instead of Newton’s, the perturbation of the orbit of Mercury disappears, as, therefore, does the need for planet Vulcan, a very neat confirmation of Einstein’s theory.

This did not diminish Newton’s scientific reputation in the least; he is still widely regarded as the greatest scientist who has ever lived. It only showed that his theory of universal gravitation was not quite universal.

Indeed, classical Newtonian physics is still used every day for such things as plotting the courses of space probes like the New Horizons spacecraft that flew past Pluto in 2015. For it is only in the vicinity of huge objects such as the sun (which is 333,000 times as massive as the earth) that Einstein’s relativistic effects become large enough to matter.

But if the science of the mighty Newton can be unsettled by a single data point, to say that the science of something as complex—and poorly understood—as the earth’s climate is settled, is ludicrous.