> Well, you can't predict what an individual molecule would do, motion wise, but you can statistically predict the behavior of a collection of lots of molecules

> Well, you can't predict what an individual molecule would do, motion wise, but you can statistically predict the behavior of a collection of lots of molecules and find it has certain characteristics, the same holds true for whole populations. Saying you can predict the behaviour of lots of molecules is only true for very small and simple systems. There's a hard reason why weather predictions become random noise seven days out. Similarly, society has a lot of feedback loops that cannot be gauged. I can recommend "Private Truth, Public Lies" by Timur Kuran as a good take on the subject.