Back in the 90’s, Steve Jobs gave an interview about his early success stories, his childhood, his science accomplishments. At some point, he picked up a heavy book and held it for a handful of seconds in the air, just to drop it with a thud to the floor. Then he asked
- “Why did the book fall?”
Of course, shish, everyone knows that Earth’s gravity pulled the book down. However, he goes to digress, gravity is just a theory that allows us to calculate the acceleration of mass particles subject to the presence of other particles. He goes on:
- “We can compute very well the position, speed and acceleration of the particles in the book but nobody ever knew why the book falls.”
Which is an astonishing proposition. Yes, we can compute the orbits of planets with ridiculous precision, we can even find new planets using this theory. But nobody really knows why.
So it begs to question: is it relevant to ask “why” things happen in first place? Or we should be content knowing how to precisely predict them?
Are you doubting science?
There’s a chicken running down the road. (The chicken will not cross, I promise) My uncle Fred tells me the good news:
- “I computed the chicken’s speed and I can tell you that the chicken is running faster and faster. In fact, it fits an exponential with incredible precision to the point I can compute the exact point where the chicken has started running.”
- “Well, and before that?”
- “We believe that the chicken just appeared there. In fact, the calculations are so precise that we gave it a name: it’s the ‘Chickenity Law’. We just published a paper and won a Nobel prize for it. “
- “And why is the chicken running in first place?”
- “Because of the Chickenity Law, of course.”
- “But Chickenity Law is just a formula you found…”
- “You religious bigot, are you doubting science?”
Was Einstein a Monkey?
The laws of physics we have today were the result of centuries of experimentation and fitting the resulting data, and creating new models when they do not fit the data.
In econometrics there is an interesting name for this: overfitting. Overfitting is when you select models that fits your data perfectly but fails to describe reality. One example of overfitting? Newton’s laws.
We all know Newton’s laws are wrong, wrong in the sense that they fail to describe perfectly the universe. Newton’s laws are at best an approximation of the universe. How do we know it? Mercury’s orbit. Einstein’s relativity also converges to Newton’s equations when speed is much lower than the speed of light.
Well is Einstein’s equations the final word then? Hardly. Michio Kaku has released a documentary describing how Physics has been struggling for the past twenty years to create models that don’t fail fitting the data we have. String theory, multiverse? These are not science.
The Infinite Monkey Theorem states that
a monkey hitting keys at random on a typewriter keyboard for an infinite amount of time will almost surely type a given text, such as the complete works of William Shakespeare.
Is it far fetched to think of Einstein’s works as the result of numerous failures to fit ? Is Einstein really right or in fact it just so happens that his theories just fits the data? One might claim to know but in reality – nobody knows.
Science as Religion
What is my concern is that the science I love has left the building. In her place it’s her ugly cousin. It has ceased to be a method to collect and analyze data. It has become a religion. Worse, it has become a sect, full of dogmas and radicals ready to strike back at the sign of questioning.