When a scientific theory makes a prediction, it’s magical.
It’s like strolling into a casino, seeing the roulette wheel in motion, putting all your money on number 33, and watching the ball settle obediently into slot number 33.
In fact, when a scientific theory makes a prediction, it’s exactly like that. Newton’s laws of motion can predict precisely which slot the roulette ball will settle into. Some people have used these laws to make a lot of money in casinos... not to mention get into a lot of trouble with the owners of those casinos.
When a scientific theory makes a prediction that turns out to be right, it gives us a lot of confidence in that theory.
Maybe a little too much confidence.
And when a scientific theory makes no predictions, it gives us a lot of confidence in dismissing that theory.
Again, maybe a little too much confidence.
Prediction is one way scientists evaluate theories.
It’s not the only way.
It’s not the truest way.
Our knee-jerk reaction to any new theory – “Does it make predictions?” – is consistently leading us to dismiss bold, new ideas that might revolutionize our understanding of the universe.
Prediction? Who cares?
Empty predictions
Before Newton came up with his laws of motion and gravitation, we couldn’t predict the positions of the planets.
After Newton came up with his laws of motion and gravitation, we could predict the positions of the planets, with astonishing accuracy.
At least, that’s the way the story goes.
The only problem with that story is that it’s not actually true.
We could predict the positions of the planets, with astonishing accuracy, long before Newton.
Ask Kepler. Almost a century before Newton published Philosophiæ Naturalis Principia Mathematica, Johannes Kepler knew that the paths of the planets are elliptical. He could use this knowledge to predict the positions of the planets with astonishing accuracy.
Ask Tycho. Decades before Kepler realized that the paths of the planets were elliptical, Tycho Brahe observed the planets through instruments he made himself. He could use his data to predict the positions of the planets with astonishing accuracy.
Ask the Egyptians. Or the Babylonians. Or the Inca. Centuries or millennia before Tycho compiled his data, the priests of these civilizations, too, compiled calendars, which they, too, could use to predict the positions of the planets with astonishing accuracy.
There was nothing wrong with the predictions made by these priests, or by Tycho, or by Kepler.
The only thing they were missing was, well, any true understanding of what was really going on.
These predictions were accurate, but they were empty.
Explanation is more important than prediction
The Egyptian, Babylonian and Inca priests had no idea why the planets move the way they move.
Nor did Tycho.
With his ellipses, Kepler came one step closer, but he, too, had no idea why the planets move along elliptical paths.
Newton knew.
He understood that the same force of gravity that causes an apple to fall to the Earth also causes the planets to orbit the Sun.
He understood that the same laws of motion that cause the apple to fall at a speed that increases in proportion to the time it has been falling also cause the orbits of the planets to be elliptical.
Newton knew why the planets move the way they move.
His contribution to science was not to make better predictions.
His contribution to science was to give better explanations.
I mean, sure, Newton’s laws of motion and gravitation do make better predictions. We’ve since applied them to predict everything from the trajectories of Apollo spacecraft towards the moon to the fate of roulette balls on roulette wheels.
But the reason we think of Newton as one of the most important figures in the history of science is that he pulled together disparate phenomena – the falling of apples and the paths of the planets – into a single coherent theory.
Explanation is more important than prediction.
Pre-prediction theories
Some scientific theories don’t make any predictions at all.
Or, at least, not for a while.
Indeed, I’d go so far as to say that all scientific theories don’t make any predictions at all, for a while.
We were taught in school that the scientific method is some contrived cycle of hypothesis, prediction, experiment, conclusion.
This is nonsense.
As Paul Feyerabend argued in his groundbreaking work Against Method, there is no scientific method.
The most revolutionary scientists simply make stuff up.
When he formulated his theories of relativity, Einstein didn’t think: hmmm... let me come up with a hypothesis, let me make a prediction, let me perform an experiment, and then, only then, will I allow myself come to a conclusion.
No, Einstein simply made stuff up.
He imagined how the world would look if he were flying alongside a photon. He imagined. He used his imagination.
He didn’t perform the experiment. He couldn’t. As Einstein himself reasoned, it’s impossible to fly alongside a photon, since it’s impossible to accelerate to the speed of light.
Einstein, when he first formulated his theories of relativity, wasn’t making any predictions. He was just playing with ideas.
And that’s true for all scientific theories. At the earliest stages, when scientists are just playing with ideas, their theories don’t make any predictions.
Eventually, of course, they do. Einstein’s general theory of relativity, for example, predicted that light from distant stars is deflected as it passes close to the sun. This prediction was tested by astronomers, who observed such a deflection during a solar eclipse, and confirmed that it matched Einstein’s calculations.
Famously, Einstein’s initial calculations were incorrect. If astronomers had made their observations during earlier eclipses, they would have measured a deflection that failed to match his calculations. Fortunately for Einstein, clouds blocked the sun during these earlier eclipses, giving him time to correct his calculations before the seminal observations were made.
Here’s the point.
All scientific theories – even Einstein’s theories of relativity – make no predictions when they’re first formulated.
All scientific theories – even Einstein’s theories of relativity – are prone to making incorrect predictions while they’re being worked out.
This is not a problem.
What is a problem is dismissing a new theory solely because it doesn’t make any predictions.
If we did that, we’d dismiss all new theories, because – I’ll say it again – all scientific theories make no predictions when they’re first formulated.
Same predictions, different day
These days, there’s a deeper problem with demanding that new theories make predictions.
The trouble is, our existing theories already make pretty precise predictions.
Newton had this problem. Tycho’s data and Kepler’s ellipses could already predict the positions of the planets pretty precisely. If all you cared about were predictions, why would you have cared about Newton’s new theory, which, for all its insights into motion and gravity, couldn’t, at first, predict the positions of the planets any more precisely than Tycho or Kepler could almost a century previously?
Today’s physicists have this problem in spades. General relativity and quantum mechanics can already predict the positions of planets and particles extraordinarily precisely. If all you care about are predictions, why would you care about any new theory of physics, which won’t, at first, predict the positions of planets or particles any more precisely than general relativity and quantum mechanics already do?
So why don’t today’s physicists just give up and go home?
Well, imagine this:
Imagine that today’s physicists come up with a new theory that interprets the same observations of the universe in a fundamentally different way, just as Newton once came up with a new theory that interpreted the same observations of the universe in a fundamentally different way.
Imagine that today’s physicists’ new theory made the same predictions about the precise positions of planets and particles as general relativity and quantum mechanics already do, just as Newton’s new theory made the same predictions about the precise positions of planets as Tycho’s data and Kepler’s ellipses already did.
Now imagine that today’s physicists’ new theory resolved deep inconsistencies between general relativity and quantum mechanics and yielded deep insights into the true nature of the universe, just as Newton’s new theory once yielded deep insights into the true nature of the universe.
Wouldn’t you be interested in that new theory?
It’s entirely possible for fundamentally different theories to make the same predictions.
And it’s entirely unreasonable to assume that the earliest of those theories is the one that’s right, just because it was proposed a few years or decades or centuries sooner.
Does Wolfram Physics make predictions?
In my exploration of Wolfram Physics, I’ve come across one objection more than any other.
Over and over again, people have told me that the Wolfram model must be rejected because it makes no predictions.
It’s true that Wolfram Physics is pre-prediction.
I mean, it does make predictions. It predicts Einstein’s equations. It predicts Schrödinger’s equation.
But it doesn’t make any predictions that differ from those of general relativity and quantum mechanics.
Yet.
It’s not that the Wolfram model can’t make predictions.
It might predict, for example, that the dimensionality of the universe fluctuates from three dimensions.
It might predict that emissions from the collapse of a star into a black hole are shaped by the discreteness of the hypergraph.
It’s just not there yet.
Wolfram Physics does, however, interpret the same observations of the universe in a fundamentally different way.
It promises to resolve deep inconsistencies between general relativity and quantum mechanics and yield deep insights into the true nature of the universe: the nature of space, the nature of time, the nature of matter, the nature of consciousness.
I’m seriously interested in this new theory.
Aren’t you?