Positivists and analysts alike believe that the words is and ought belong to different worlds, so that sentences which are constructed with is usuall… - Jacob Bronowski

Of course, everyone is free to prefer his favourite article of faith to the scientific, that is the empirical method. But do not let us imagine that his faith is then anything except a piece of comfortable and customary superstition. To try to make a nice distinction between what science can predict and what is somehow supernaturally determined is a piece of elegant but really quite shameless self-deception. Science is a practical study of what can be observed, and the prediction from that of what will be observed. To say that causes are somehow getting under this observable world, when anything under it is essentially unobservable, is neither helpful nor meaningful; it is just a piece of faithful comfort. We might as well say that the electrons are really pushed about by blue fairies with red noses who know exactly what they are doing, only it happens that every time we look in their direction these fairies instantly hide. If they are essentially unobservable, beyond all hope of future unravelling, then it simply does not make sense to bring them into any system, logical, metaphysical, or even religious.

Life is not an examination; we do not get marks for the steps; what matters is getting the right answer. So it is perfectly possible to base a system of prediction on no principle except trying to get the right answer. This is exactly what all plants and animals do. The bat avoids obstacles by shouting at them that shrill cry just beyond my hearing, and then listens for the echo. Whatever system it has for translating the echo into a prediction it has found by evolution, and evolution has found it by trial and error.

The [eighteenth] century settled down in two camps of Philistines: the literary Philistines and, largely to spite them, the scientific Philistines. It was the beginning of the mistaken opposition between them from which we still suffer.