With documentary sources, as with one's friends, it is never easy to recall exactly when mere acquaintance passed into familiarity. I am somewhat shaken to see from its footnotes that when I wrote my chapter for the Chapman symposium in 1969–70 that I made no use of fire insurance policies for the dating of house-building, relying mainly on conveyances, estate-plans, rate books, press advertisements and family papers; the same range of sources employed over a far wider geographical area in Christopher Chalklin's recent study. My love affair with insurance policies is therefore relatively recent, and that of an ageing researcher, not to be compared with the amatory techniques of others who were using them for industrial history well before 1971. I have been an old man in a hurry, a scanner rather than a scholar; and these practical notes from a scanner will be superseded if the S.S.R.C. project for an index to the main registers is ever realized.

The developing techniques of historical nominal record linkage can make substantial contributions to the questions raised by our present understanding of the urban middle class in the first half of the nineteenth century. Cheap printing, institutionalization, and increasing political and state action provided a growing variety of information about individuals – directories, poll books, lists of shareholders, pewholders, stallholders, members of committees and societies, signatures to petitions and requisitions, wills, insurance policies, and by the 1840s marriage and census data. By their nature most of these listings were concerned with the politically and socially active, and with those above a minimal level of social status and economic power. In practice, this meant predominantly but not exclusively members of a potential middle class. Under certain constraints the lists may be merged to provide surrogate answers to some ghostly questionnaire regarding patterns of association, behaviour and social status, major social divisions and networks, and the characteristics of those who took part in institutions and activities traditionally identified with the middle class.

Professor Marion Bowley, writing of the housing boom of the 1930s which was uniquely based upon sales to owner-occupiers, remarked that ‘we have no means of knowing statistically the extent of owner-occupation before the Great War…Actually, it is more important to know why the change occurred [in the inter-war period] than to know its actual size, and there is no need to be led astray in a futile search for non-existent statistics’. The same ban would seem to extend to analysis of investment ownership of housing before 1914. What I want to suggest is that it is indeed possible to obtain statistics relating not only to owner occupation but also to the overall structure of the urban housing market; and that far from leading us astray, it is in fact important to acquire precise figures in order to understand with more certainty the structure of the nineteenth-century housing market – something of which we have remarkably little knowledge.

This article surveys some of the more interesting writings on Dutch urban history over the last decade against the background of the older literature of the subject. It also deals briefly with some current research projects. The choice of what to include has perforce been a subjective one since the field itself is so vast, but an effort has been made to indicate the main stream of historical research now going on. How much more widely that field might have been drawn is perhaps indicated by the subjects that have been relatively neglected: education, town planning, housing and rents, wage movements, patterns of consumption, leisure pursuits, retail distribution.