A 200 year old lesson: Scientific predictions are worthless unless tested

Your theory predicts something novel? How nice. But no one will pay you any mind unless you test it.

“He who loves practice without theory is like the sailor who boards ship without a rudder and compass and never knows where he may cast.”
–Leonardo Da Vinci

Our best scientific theories of the day — of any day, really — are only as useful as the phenomena they predict. We have confidence that the Sun will rise and set not merely because it’s always done so, but because the laws of physics, proven and validated throughout the centuries, dictate that this behavior must continue under these laws. But sometimes, the predictions that a theory makes are patently absurd. Does that mean the theory is wrong? Sometimes, but not always. You see, sometimes it’s our intuition that’s wrong. Sometimes, nature truly is absurd. The only way to decide? To do the experiment, and test the theory, for ourselves.

Newton’s sketch of Halley’s Comet, as he drew it and published it in his Principia Mathematica. Image credit: Isaac Newton / Principia Mathematica.

Imagine yourself back in history, a hundred years after Isaac Newton. His treatises on a variety of topics — mathematics, astronomy, gravitation, mechanics and optics — had been verified better than any other scientific disciplines in history up to that point.

Many of these fields had been further developed as well, and it was discovered that Newton’s theories not only served as a solid foundation for each of these fields, but that they often provided deep insights into the fundamental workings of the Universe when applied to new phenomena. This was true for practically all of the aforementioned areas, but with one exception: the behavior of light.

The behavior of white light as it passes through a prism demonstrates an apparent ray-like nature, in concordance with Newton’s description. Image credit: University of Iowa.

Newton was insistent that light behaved like a ray, refracting, diffracting and reflecting according to the laws he laid out in his important book: Opticks. Through this work, he was able to account for a whole host of phenomena, including the behavior of colors, all verifiable through experiment. Indeed, the first sentence of his book opened as such:

My Design in this Book is not to explain the Properties of Light by Hypotheses, but to propose and prove them by Reason and Experiments.

But 100 years after Newton, an experiment was performed that simply couldn’t be accounted for by Newton’s conception.

The classical expectation of sending particles through either a single slit (L) or a double slit (R). Image credit: Wikimedia Commons user inductiveload.

If you passed a beam of light through a single, narrow slit, you’d expect it to arrive on the other side, perhaps more intense towards the very center than at either end as you moved away. If you passed a beam of light through two slits, you’d expect two central peaks, each one fading away as you moved away from it. At least, that would be true if light were made of corpuscles, or particles.

But when the experiment was performed with these slits closely spaced together, you didn’t wind up seeing two peaks at all, but rather a great number of peaks, with dark spaces in between them.

The bright and dark fringes that appear on the far side of a two-slit experiment performed with light can only be explained by a wave-like, rather than ray-like nature. Image credit: Wikimedia Commons user inductiveload.

This sort of phenomenon could not be accounted for with any ray-based (or corpuscular-based) theory of light, but rather required that light fundamentally behaved as a wave. When Thomas Young performed his double slit experiment in 1799, he recognized that this type of phenomenon could only result if — as others such as Huygens had theorized before — light was fundamentally behaving as a wave. This same pattern of interference, with constructive peaks and destructive minima, was familiar to anyone that had performed the analogous experiment with water waves.

The wave-like nature of light passed through two slits, as illustrated by Thomas Young’s original work, dating from 1803. Image credit: Wikimedia Commons user Quatar.

But light also appeared to have corpuscular (or particle-like) properties as well. Newton’s treatise on Opticks, after all, was able to explain how light reflected and refracted perfectly, without treating light as a wave. The new revelation — and the new experimental results — didn’t invalidate the older ones at all. Quite to the contrary, if light really were a wave, it should show up in all instances that a wave-like behavior ought to evidence itself.

Light, whether passed through two thick slits (top), two thin slits (middle) or one thick slit (bottom), displays evidence of interference, pointing to a wave-like nature. Image credit: Benjamin Crowell.

So the top theorists of the time, many of whom were enamored with the infallibility of Newton, set out to see if the idea that light was a wave led to any predictions that were absurd. And in 1818, that’s exactly what the famed French mathematician and physicist Simeon Poisson set out to do.

He imagined what would happen if he had a source of light that emitted a single wavelength — assuming it was a wave, of course — and that it spread out as it left the source until it encountered a spherical object. The light that hit the sphere would either be absorbed or reflected away, and what you’d be left with was a ring of light showing up on the screen behind it.

Shining coherent (e.g., laser) light around a spherical, opaque object is one of the clearest ways to test the wave-like versus the particle-like nature of light. Image credit: Auburn University.

But if light were truly a wave, you would get some very bizarre phenomena, some that you might expect and some that’s completely counterintuitive. You might expect that you’d get a series of light-and-dark fringes outside the sphere, similar to the interference pattern observed in the double slit. But what no one expected was that Poisson’s calculations showed that at the very center of the shadow on the screen, there should be a single bright point, where the wave nature of light all constructively interfered at the most unlikely of places.

A theoretical prediction of what the wave-like pattern of light would look like around a spherical, opaque object. The bright spot in the middle was the absurdity that led many to discount the wave theory. Image credit: Robert Vanderbei.

How absurd! And so, Poisson elegantly reasoned that the wave nature of light was a ridiculous notion, and must be wrong. But Poisson committed the cardinal sin of theoretical hubris: he drew a conclusion without performing the crucial experiment at all!

The circumstances of this were particularly maddening: this was at a competition sponsored by the French Academy of Sciences to explain the nature of light, and the entrant who proposed the wave theory — Fresnel — was basically laughed out of the room by Poisson, who was one of the judges. But the head of the committee stood up for the entrant instead, and decided to do what a scientist must do in good conscience. François Arago, who later became much more famous as a politician, abolitionist, and even prime minister of France, performed the deciding experiment himself, fashioning a spherical obstacle and shining monochromatic light around it. The result?

The results of an experiment, showcased using laser light around a spherical object, with the actual optical data. Image credit: Thomas Bauer at Wellesley.

The spot is real!

I myself have referred to this — like many others — as the Poisson spot in the past, but I do so no longer. From this point on, in honor of the scientist who actually put the science to the experimental test, it shall be known as the Spot of Arago!

A model of the experiment, with the bright spot actually tested and found by Arago. Image credit: Thomas Reisinger, cc-by-sa 3.0, E. Siegel.

What’s perhaps most amazing about this is that if you craft a perfectly circular obstacle, the intensity of light at the very center is actually equal to the completely unobstructed intensity, with small circular fringes around the spot itself. Slight imperfections in the sphere only enhance the additional “wiggles” seen in the shadowed portion.

Imperfections in the smoothness of the sphere lead to additional interference perturbations, but the central spot always dominates. Image credit: Thomas Reisinger, created using GNUPlot, under cca-sa-3.0.

So the next time you run across what appears to be a theoretical absurdity, either because you believe such a thing must be so or cannot be so, don’t forget the vital importance of putting it to the experimental test! It’s the only Universe we have, and no matter how solid the footing of our theoretical predictions, they must always be subject to the scrutiny of unrelenting and continuous tests. After all, you never know what secrets the Universe will reveal about itself until you look!

This post first appeared at Forbes, and is brought to you ad-free by our Patreon supporters. Comment on our forum, & buy our first book: Beyond The Galaxy!