Edge.org, a highfallutin sort of scientific blog - don't check it right now because it's unavailable - that people say you should be reading but you're not, probably for a good reasons, asks the all caps question above.
For sure it is the idea that there exist hard and soft sciences, or nonsciences. One wouldn't call it a scientific idea except that it is held by scientists and non-scientists alike, and forms the core of the classification of areas of study for the purpose of university administration. The separations is founded on the fact that science progressed by the analysis of some particularly simple phenomena - simple in the sense that they can be simplified by approximation to involve a small number of factors and therefore can be described directly by consideration of fundamental physical laws. Originally we can mention planetary and stellar motion, and electric currents and magnetic attraction.
Now if you really believe in "exact sciences" go compute the fuel consumption of an automobile using Newton's laws and see how exact your answer will turn out. Or if you believe in non-falsifiable sciences, check out the 5 year old experiment performed in the rich economies - provoking a recession by purely financial methods, following textbook recipes, just to show how right Keynes was.
The fictional division between hard and soft sciences is convenient for academics on both sides, who derive satisfaction from putting down their counterparts because, respectively, they cannot make precise statements, or can arrive at truth from mere experiments. It also serves to track young minds into academic tribes from whose confines few are able to escape. The distinction has no ground on reality and should be retired.