On a whim, I decided to read C. Wright Mills' The Sociological Imagination, and am very glad I did. I had aquired - rightly or wrongly - an impression from textbooks that this was a pretty dry staking out of academic turf, bit it's actually quite a jolly table-thumping call-to-arms against bad academic practices and (for want of a better word) thinking. This exerpt is from a chapter taking the work of Talcott Parsons as its example, but it's not hard to think of many more examples one could point to since the book was published in 1959(!):

The basic cause of grand theory is the initial choice of a level of thinking so general that its practitioners cannot logically get down to observation. They never, as grand theorists, get down from the higher generalities to problems in their historical and structural contexts. This absence of a firm sense of genuine problems, in turn, makes for the unreality so noticeable in their pages. One resulting characteristic is a seemingly arbitrary and certainly endless elaboration of distinctions, which neither enlarge our understanding nor make our experience more sensible. This in turn is revealed as a partially organized abdication of the effort to describe and explain human conduct and society plainly.

When we consider what a word stands for, we are dealing with its semantic aspects; we we consider it in relation to other words, we are dealing with its syntactic features. I introduce these shorthand terms because they provide an economical and precise way to make this point: Grand theory is drunk on syntax, blind to semantics. Its practitioners do not truly understand that when we define a word we are merely inviting others to use it as we would like it to be used; that the purpose of the definition is to focus argument upon fact, and that the proper result of a good definition is to transform argument over terms into disagreements about fact, and thus open arguments to further inquiry.

The grand theorists are so preoccupied by syntactic meanings and so unimaginitive about semantic references, they are so rigidly confined to such high levels of abstraction that the 'typologies' they make up - and the work they do to make them up - seem more often an arid game of Concepts than an effort to define systematically - which is to say, in a clear and orderly way - the problems at hand, and to guide our efforts to solve them.

What he said.