Can scientific models be falsifiable?
Let’s read two stories of changes in scientific models and try to find out the answer. If you are bored to read, skip directly to the Conclusion at the end… You will get the meaning.
Story 1: Modelling sun spots…
A group of researchers believes that contrary to a common model sunspots form near the Sun’s surface.
The SPOTSIM project, which studies the formation of sunspots, was awarded competed time on the Mare Nostrum supercomputer in Spain. The awarded resources amount to 20 million CPU hours – a computing time equal to approximately 500 years on a normal laptop. The group of researchers including scientists from Aalto University and the Max Planck Institute for Solar System Research (MPS) in Germany hopes to pinpoint the important role played by turbulent plasma flows within the Sun.
‘The simulations are exceptionally large; one data cube in one simulation is approximately 700 gigabytes in size. One simulation produces tens of data cubes, but these cannot all be stored. We would like to have a total of around 15 terabytes of research data in long-term storage,’ researcher Petri Käpylä from the Aalto University in Finland and the Leibniz-Institut for Astrophysics in Germany describes the project. (1)
Story 2: Modelling universe expansion
UBC physicists may have solved one of nature’s great puzzles: what causes the accelerating expansion of our universe? A group of
PhD student Qingdi Wang has tackled this question in a new study that tries to resolve a major incompatibility issue between two of the most successful theories that explain how our universe works: quantum mechanics and Einstein’s theory of general relativity.
The study suggests that if we zoomed in-way in-on the universe, we would realize it’s made up of constantly fluctuating space and time. Their calculations provide a completely different physical picture of the universe. In this new picture, the space we live in is fluctuating wildly. At each point, it oscillates between expansion and contraction. As it swings back and forth, the two almost cancel each other but a very small net effect drives the universe to expand slowly at an accelerating rate.
But if space and time are fluctuating, why can’t we feel it? “This happens at very tiny scales, billions and billions times smaller even than an electron,” said Wang. (1)
With all these changes in scientific models, the question to be asked is a very simple and yet difficult one: Can a scientific model ever be wrong?
Sure, one would say. If it does not “fit the data”.
Can a scientific theory ever be disproved?
Sure, one would say. If it does not “fit the data”.
But what data? And if it does not fit would that mean that it is wrong or that it needs “corrections”? What scientists usually do is adjust the model to fit the data in a better way, not discard the model as a whole. Creating new models is hard. Adjusting is easier. And helps the funders sleep better at night.
All models can be correct. With the proper adjustments…
This makes science more of a tool rather than a path to the “truth”.
Someone might argue that in this way we degrade science into something much less than what it really is. And for most science is the ultimate tool to discover the “truth” and “reality” (another big word). But throughout the history of science the creation of a scientific model was never the main event of scientific discovery. The latter has more to do with short bright flashes of insight which light up a shady area of the cosmos, without necessarily formulating a specific model to represent the data and make predictions.
The distinction is a subtle one and I am not sure if even I understand it well. I can only give a hint of what I see with the corner of my mind when contemplating on the subject…
Newton is great not because of his equations based on the planetary movements data. These have anyway been amended and corrected many times since he first came up with them. (the above-mentioned “adjustment”) And despite the fact that newer theories have come up to replace the theory of gravity (e.g. the bending of space-time by Einstein) Newton’s equations can still be used to get the same results as the newer theories! Newton was not great because of the scientific model he created. Newton is one of the greatest scientists of all times because he tried to read the mind of the Creator and dared to imagine an all-reaching invisible force which affects everything in the universe instantly.
True science is about great ideas.
Modern science is about equations.
True science is about poetic imagination.
Modern science is about reading a computer screen.
But equations are mere tautologies as Wittgenstein so eloquently explained.
See science into the falsifiable. Seek science into the ideas which could vanish into thin air. It does not matter whether these ideas are right or wrong – at the end they just form the context of a potential modelling of the cosmos. But they are not the modelling itself.
Newton’s model of gravity based on the idea of gravity can never be wrong – equations can always be adjusted to match the data.
But Newton’s idea of that invisible all-catching instant force may be wrong.
And this is what makes it so precious…
Any flash of light is bright only because of the darkness that follows it…