We hear them continually on TV: hence they occur first when it is our turn to talk. In this regard, talk may be said to be the enemy of writing. If y… - Jacques Barzun

Convince yourself that you are working in clay, not marble, on paper not eternal bronze: Let that first sentence be as stupid as it wishes.

Now, analysis, the breaking of the wholes into parts, is fundamental to science, but for judging works of art, the procedure is more uncertain: what are the natural parts of a story, a sonnet, a painting? The maker's aim is to project his vision by creating not a machine made up of parts but the impression of seamless unity that belongs to a living thing. Looking at an early example of systematic criticism by analysis — say, Dante's comments on his sonnet sequence La Vita Nuova — one sees that the best he can do is to tell again in prose what the first two lines mean, then the next three, and so on in little chunks through the entire work. We may understand somewhat better his intention here and there, but at the same time we vaguely feel that the exercise was superfluous and inappropriate. Reflection tells us why: those notations taken together do not add up to the meaning of the several poems. In three words: analysis is reductive. Since its patent success in natural sciences, analysis has become a universal mode of dealing not merely with what is unknown or difficult, but also with all interesting things as if they were difficult. Accordingly, analysis is a theme. Depending on the particulars of its effect, it can also be designated reductivism.

History, like a vast river, propels logs, vegetation, rafts, and debris; it is full of live and dead things, some destined for resurrection; it mingles many waters and holds in solution invisible substances stolen from distant soils. Anything may become part of it; that is why it can be an image of the continuity of mankind. And it is also why some of its freight turns up again in the social sciences: they were constructed out of the contents of history in the same way as houses in medieval Rome were made out of stones taken from the Coliseum. But the special sciences based on sorted facts cannot be mistaken for rivers flowing in time and full of persons and events. They are systems fashioned with concepts, numbers, and abstract relations. For history, the reward of eluding method is to escape abstraction.