These days we're hearing and talking about a future that is suddenly very difficult to predict. Uncertainty is frequently invoked. I know what it's supposed to mean, but the scientific context is very different. It's one of those words - theory, hypothesis, work, energy, power, model, significant, natural, ... - with a very specific scientific meaning at odds with their popular meaning. (science communication is hard)
Who knows what's going to happen? frequency comes to mind, while uncertainty is well defined in the context of science. It's the quantification of how tightly we can constrain what we don't know. That's a very powerful concerpt. Being the simplest of the sciences, physics has the most rigid requirements. In order for a discovery to be accepted by particle physicists uncertainty that it could be something else has to be lower than one part in three point five million - the five standard deviation level. People living in this world spend much of their time thinking about what could go wrong and devising other measurements to disprove their original hypothesis. We speak in terms of confidence levels.
Henry Cavendish was one of the first to work at understanding and minimizing uncertainty when he weighed the Earth to determine the gravitational constant. Even today the measurement is difficult as gravity is incredibly weak as forces go. Without going into details, Cavendish built an exquisitely sensitive torsion balance that measured the attraction between small lead spheres and a much larger lead sphere. His early measurements made sense, but was it just gravity? Perhaps there was a bit of a magnetic effect .. he rebuilt the experiment with a large magnetic ball replacing the lead ball. This wasn't to rule out magnetism as much as to set a quantitative limit on how large the effect was. Then he turned to thermal effects heating and cooling different parts of the experiment. He was left with his best measurement and the level of confidence that he had removed all other controllable effects (including human error).
The game becomes trying to make a more precise measurement and perhaps seeing something new in the process - some measurable effect that can't be explained conventionally. A good deal of science advances in this way.
Here's a cute introduction to uncertainty that I almost really like - the bear in the moonlight - almost as it doesn't touch on fundamental uncertainty.
There's another deep form of uncertainty - you've probably heard of Heisenberg's uncertainty principle and Schrödinger's poor cat. Broadly stated Heisenberg showed that it was impossible to perfectly know both the position and momentum of anything. He called it ungenauigkeit (inexactness) or unbestimmtheit (undeterminedness). Niels Bohr called it unsicherheit (unsureness). Even today German physicists (like their Prime Minister, Angela Merkel) will call it unschärfe (fuzziness). All of these terms are more descriptive of the physics than uncertainty. The use of uncertainty suggests to some there is an exact value and we just can't measure it. Instead there is just a probability distribution. It is indeed fuzzy - at a microscopic level all of the atoms that make us up are fuzzy - it's just our world is mostly too large to notice.
In some parts of my life I try to avoid what is usually thought of as uncertainty, but when you want to really understand something diving in and reducing your uncertainty and increasing your level of confidence is a powerful tool.
Comments