historical sociology

The term commonly applied to sociological analysis based on historical data sources-either primary (such as original documents in archives) or secondary (the written history produced by historians themselves).

There has been widespread and sometimes acrimonious methodological debate among historians and sociologists as to the boundaries and relationship between the two disciplines. In the early 1960s E. H. Carr (What is History?) argued that ‘The more sociological history becomes and the more historical sociology becomes, the better for both. Let the frontier between them be kept open for two-way traffic.’ However, Carr's views can be contrasted with those of Charles Wilson (Professor of Modern History at the University of Cambridge), who used the occasion of his inaugural lecture (in 1964) to observe that ‘the economic historian and historical sociologist have, it seems, never had things so much their own way … There is, I think, no great harm done if they go on talking their latest professional jargon to each other in private; if economic historians prattle of liquidity, variables, or backward sloping curves, or sociologists of motivation, elites and social roles. Provided always that we return to some common plain language which recovers this specialized shorthand for purposes of civilized intercourse and ultimate history’. The less controversial view of most sociologists and historians is probably that aired by Philip Abrams (Historical Sociology, 1980), who argued that ‘history and sociology are and always have been the same thing’, so that any dispute about their relationship to each other was merely a matter of prevailing institutional arrangements rather than one of intellectual substance. History tends to be individualizing ( ideographic ) and to describe singular or unique phenomena, whereas sociology is generalizing (nomothetic) in formulating theories that apply to categories of phenomena, but this is a matter of emphasis rather than a hard and-fast principle of method, since (to quote, Central Problems in Social Theory, 1979) ‘there simply are no logical or even methodological distinctions between the social sciences and history-appropriately conceived’. The common objective of practitioners in both disciplines is a causal analysis of the meaningful behaviour of individuals and groups, with a proper appreciation of process, context, and change .

Notwithstanding these epistemological matters, the major issue confronting historical sociologists relates to the practical difficulties of using primary historical materials as evidence, since (as E. P. Thompson so eloquently put it) we cannot interview tombstones. Such materials include public and private written documents such as official reports, surveys, parish registers, records of organizations, letters, and diaries. Among othe things, the researcher needs to establish the authenticity of documents, identifying their authorship and degree of completeness; the credibility of documents, given likely sources of error and distortion, the possible motives of their authors, and the different conditions under which each testament was produced; and the representativeness of the various surviving documentary materials. In other words issues of reliability are especially acute. The question of validity is also to the fore, since few surviving sources were constructed for precisely the purposes which the modem researcher has in mind, so that he or she is usually attempting to establish the meaning of data by reading ‘against the grain’ of the purposes for which they were originally compiled.

It is often the case, therefore, that extant historical data cannot directly answer the questions which are asked of them by the sociologist. In particular, it has proved very difficult for the historical sociologist to explore the subjective aspects of behaviour, especially given the traditional reliance on written documents as sources of evidence. People's motives for action, the meaning and feeling they might attach to (for example) their relationships with kin, their attitudes , values , and beliefs, and the salience of all of these things in their daily lives, have proved among the most elusive of quarries. Such subtleties cannot easily be recovered from the official statistics and semi-official records which are by far the most common source of information regarding individual and collective behaviour. While oral histories , diaries, letters, novels, and so forth can give us an impression of the ‘temper’ of a particular time, they are themselves often partial and have methodological problems of their own which restrict the extent to which the historical sociologist is able to draw inferences from these materials.

Such scepticism should apply, of course, to secondary as well as primary historical materials. There is an unfortunate tendency among some historical sociologists to treat the written accounts produced by historians as if these were somehow unproblematic collations of the historical facts about particular episodes. Such accounts are then raided to provide illustrations with which to fill the empty boxes provided by preconceived sociological models and theories. Historical sociology on this grand scale-in which one ignores the specific historicity (or context) of events, ignores any disputes surrounding the interpretation of the original data offered by the historian, and merely collects his or her inferences as facts which can be arranged in attractive patterns according to the requirements of one's sociological model-is said by its critics to offer practitioners ‘a delightful freedom to play “pick-and-mix” in history's sweetshop’. One such critic has specifically cautioned sociologists against turning too readily to historical sources, where appropriate sociological data can be generated, since to do so is to abandon a position of comparative advantage whereby the sociologist can generate his or her own evidence (via appropriate surveys, interviews, or observations) whereas the historian is necessarily condemned to work with such evidence as can be culled from among the stock of documentary ‘relics’ that happen to have survived (see, ‘The Uses of History in Sociology’, British Journal of Sociology. 1991).

Despite this methodological controversy-or perhaps one should say fuelling it-some of the most prominent works in recent years have been within the domain of historical sociology. These include Barrington Moore's The Social Origins of Dictatorship and Democracy (1966), Immanuel Wallerstein's The Modern World System (1974,1980), Michael Mann's The Sources of Social Power (1986), and Theda Skocpol's States and Social Revolutions (1979), all of which have provoked an enormous secondary literature and no little controversy.

Moore's text exemplifies this sort of work and was significant as the first of a new wave of historical sociology in the late 1960s. The central argument of the book is that it was variations in three sets of social relations which determined whether particular nation-states took a democratic or dictatorial route to modernity. On the basis of sustained reflection on the transition from feudalism to capitalism in a range of societies which includes England, France, Germany, Japan, Russia, China, and India, Moore suggests that the critical sets of relations are the following: the relations between the landed aristocracy and the monarchy, which remained evenly balanced where there was a democratic outcome; the relations between the aristocracy and the peasantry, which involved the granting of considerable autonomy to the latter, where there was a democratic outcome; the relations between the aristocracy and the urban bourgeoisie, which were mutually supportive vis-à-vis the monarchy, where there was a democratic outcome. Ther are numerous critical treatments of this work. The best of these is probably Dennis Smith's Barrington Moore: Violence, Morality and Political Change (1983).