N ANY COMMUNITY or period of time, two sorts of change are presumably always going on at once: real change, which is happening but may not be perceived, and perceived change, which may or may not be taking place. History, in practical terms, is largely the story of perceived change. Change in an objective sense may well be taking place, but the materials out of which books on history are made almost never reflect it directly. Instead, they tell the story of various, partial recordings of change which, however the fact is disguised, really add up to interpretations.' Even quantitative history, the latest and most ambitious of objectifying techniques, does not provide a satisfactory account of transformations in the external world. I do not wish to argue that esse equals percipi, or that pure objectivity, like subjectivity, is unattainable--these are problems for the philosopher-merely that, for the historian, in the everyday practice of his craft, objectivity and subjectivity exist in a rough continuum. Drawing a hard-and-fast line between them is impossible. Of course, the historian tries to use the evidence in such a way that his own bias or that of the original participants is balanced by countervening factors, to which he often attempts to give an air of theoretical respectability. But in the last analysis his approach boils down to common sense. If one looks back over historical writing since the nineteenth century, there is no simple way of summing up the various trends. But there are two repeatedly stated goals. One is a closer and closer accounting for change in the external world. The other is a gradual awakening to patterns of inner development. Rightly or wrongly, the difference between the two has often been looked upon as the opposition of objectivity to subjectivity. About a century ago, economic history emerged from legal and political history as the most faithful adherent to the external-world hypothesis; in the twentieth century, especially since World War I, social history has become its sister discipline, incorporating economic history into a larger and more flexible framework-the social system of which economy is a part-while retaining much of the vigor of method. The link between the two is provided by new methods of counting, which also furnish the theoretical justification for retaining the empirical framework. No one now doubts that quantification is a good thing. But its success is relative, not absolute. Those who entertain the wildest illusions about statistical methods