Submitted in accordance with the requirements for the degree of Doctor of Philosophy
The University of Leeds
School of History
The candidate confirms that the work submitted is his own and that appropriate credit has been given where reference has been made to the work of others.
This copy has been supplied on the understanding that it is copyright material and that no quotation from the thesis may be published without proper acknowledgement.
I am grateful to the researchers, archivists and librarians who have been generous with their time and advice during this study. Among them I include my fellow-students. It would be unfair to single some out for special mention, with the exception of my supervisors. Katrina Honeyman, despite serious illness in the last few months of the study, and John Chartres, have been dependable sources of insight, challenge and good-humoured support. My family, particularly Susan, David and Laura, also deserve my thanks for their support and encouragement.
Statistical methods have provided insight into the post-1860 fertility decline, but the deeper explanations lie in the choices of individuals, shaped by the cultures of local communities. This thesis combines the use of statistical and qualitative information, each form of enquiry guiding the other. It gains analytical strength by comparing three differing towns, Bradford, Leeds and Middlesbrough. Relationships between culture, employment, and fertility are examined by studying differences in local cultures and local labour markets, including female and child employment. The main argument of the thesis is for the impact of rising expectations about how a working-class family should live, underestimated by existing accounts. Working-class parents pursued higher living standards, not to emulate the better-off but, for example, to give their children better lives than they had experienced. To make these goals achievable, they chose to have fewer children, allowing more resources and attention for each family member. Such an explanation places a new stress on the nature of rising working-class consumption.
In all three towns, evidence is put forward for a strong growth in expectations about standards. This is demonstrated in relation to diet, housing, clothing, and leisure: the impacts of compulsory education and changing views of family life are also shown. The thesis shows that mothers and fathers came to see family limitation as a necessary part of pursuing these standards. Differences between the towns mainly reflect the varying frequency of female full-time work. Expectations of rising living standards placed the greatest pressures on working mothers, who had to pursue them by both wage labour and domestic labour. This made them the most susceptible to the appeal of family limitation. The study also shows, however, that fathers too had incentives to family limitation, which previous studies have underestimated.
For sources, qualitative evidence comes from cheap local newspapers with a substantial working-class readership. These new sources make evidence available which has never previously been used in research into the fertility decline. This is also true of the Bradford and Middlesbrough oral history collections used here. Materials in the Burnett Archive of Working-Class Autobiographies at Brunel University are also utilised. Use is made, too, of advice literature directed at the working class of the three towns, and reports of social conditions by, for example, doctors and social reformers. These materials are combined with quantitative evidence from the registration of births, Census Reports, Parliamentary Papers, and the Census Enumerators’ Books.
This thesis is significant because it is one of a surprisingly small number of studies which investigate the cultural causes of the fertility decline, a need identified by leading researchers. By placing attention, as it should, on the behaviour of the majority rather than better-researched elites, the thesis brings new sources such as cheap newspapers into the study of the fertility decline. Above all, its description of rising expectations about consumption and their impact sheds new light on the causes of the fertility decline, drawing attention to the adoption by working-class people of more demanding goals for the quality of family life, which they met with a shift to smaller families.
List of abbreviations 6
1. Introduction 8
2. Thinking about fertility and culture 13
3. Bradford, Leeds and Middlesbrough: fertility and society 54
Appendix 1: Description of the analytical work on the 1881 CEBs 107
Appendix 2: Preparation of the Cost of Living Series 115
4. Consumption aspirations and their impact on fertility 118
5. Perceptions of education and their impact on fertility 179
6. Changing identities, personal relationships and fertility 203
7. Conclusions 228
List of abbreviations AHRC Arts and Humanities Research Council
AWN Armley and Wortley News
BA Bradford Archives
BAWCA Burnett Archive of Working-Class Autobiographies
Between 1860 and 1920, fertility and family size in England declined in a momentous way. While families reaching completion in the 1860s typically had about five children (of whom an average of 3.5 survived long enough to start their own families), completed families in 1910 averaged 2.5 children, nearly all of whom survived.1 These changes reversed the historically high fertility rates of the late eighteenth and early nineteenth centuries and began a fall in birth rates which continued until the nineteen-thirties. They were a phenomenon with important social consequences. The proportion of the population under the age of fourteen fell from thirty-five per cent to twenty-five, so that there were fewer dependent children for each economically active adult, and more resources to go round.2 This appears to have been one of the main reasons why children’s health and welfare improved so notably in this period: adults also benefited from the reduced pressure on resources.3 Experiences of family life became more homogeneous as a much larger proportion of children grew up with only a small number of siblings, whereas in the 1870s, for example families with eleven or more children contained a quarter of all children born.4 In place of this mid-Victorian diversity, by 1920 most young adults were reaching maturity with similar expectations of what family life would, and should, be like.
Attention has been given to the reasons for this fertility decline ever since it was first observed by contemporary writers. Most modern scholars agree that there was no simple explanation, and the search for answers continues. This has not prevented some general histories of the later nineteenth century from taking a rather mechanistic approach to the fertility decline which relies on models already discarded by the demographic historians, or else simply neglecting its importance for social and cultural history. The present study came to the subject of fertility by a slightly roundabout route: beginning with an interest in working-class experiences of health care, it then focussed on the changing experience and social meanings of childbirth, which led to the oral history sources now used. At the same time as the absence of a compelling research question in this field was becoming problematic, two discoveries led to the focus on fertility and to the study’s main hypothesis. The first was De Vries’ well-argued but provocative claim that ‘breadwinner-homemaker households’ were best for family welfare in the late nineteenth century, with its focus on the meanings of consumption.5 The second was attendance at a conference of the Local Population Studies Society which showed the potential power of both Census and newspaper sources for investigating local cultures. Taken together, these experiences provided a new way to look at the fertility decline and a set of techniques for testing its validity.
The value and relevance of this study lie in four areas. First, it is an exploration of the changes in cultures which led growing numbers of women and men deliberately to limit their fertility within marriage. Previous research has shown that statistical aggregates can only tell so much: ultimately, demography happens at the level of individual decisions about sexual behaviour and fertility, and these are the results of individual perceptions which are shaped by prevailing cultures. Recent writers have emphasised the need to give centre stage to culture, rather than ever more sophisticated statistical analysis of demographic data, in order to examine the most interesting problems about the origins of the fertility decline. In the words of one, scholars must ‘think beyond the bounds set by disciplines back into the minds of their ... Victorian ancestors.’6 For example, demographers have begun to see changes in fertility as at least partially produced by changes in the socially constructed meanings of sexual behaviour.7 There have been surprisingly few studies which attempt such an interdisciplinary approach, however.
Second, the study focuses on the cultures of the working-class majority rather than on elites. It is important to get back to writing the history of the lives of the majority. Developments in historiography since about 1980 have had a tendency to move away from this, for example because the ‘linguistic turn’ led to concentration on the people who left the most written records available for textual analysis. If history is about ‘how we got here’, then it needs a focus on the lived experiences of the majority as well as on public affairs and elites. Most people in this period described themselves as ‘working-class’, so understanding whole population aggregates involves understanding trends in the working class. To write history in this way is also to affirm the agency – and historical importance – of ‘ordinary’ individuals, against the tendency to regard them as passive recipients of whatever came their way.
Third, this thesis presents significant new findings which call for some reinterpretation of the causes of the fertility decline. Alongside existing explanations such as the rising net cost of childrearing resulting from mass education and delayed entry into the labour market, it offers the new insight that the rising expectations about how a working-class family ought to live, largely fuelled by growing working-class self-confidence and assertiveness, were a major factor in the fertility decline. These expectations would have been unachievable with the large families of the early nineteenth century. The study shows how men and women responded to this increase in the perceived costs of childrearing by deciding to have smaller families in which each member could have more resources and attention.
Fourth and last, the study makes use of new sources which have hardly ever been previously exploited for research, in particular a group of cheap local newspapers, much less well-known to historians than the Leeds Mercury or the Bradford Observer, which were read by, and targeted at, the working class or at least a substantial element of it. These new sources make evidence available which has never previously been used in research into the fertility decline. The same point can be made of the study’s use of Bradford and Middlesbrough collections of oral history testimony made in the 1980s.
The study is a comparison of three northern towns, Bradford, Leeds and Middlesbrough.8 These were selected because they were large centres of population, meaning that their characteristics would have a significant impact on aggregate fertility at a higher, even national, level. (Leeds and Bradford were both among the ten largest towns in England and Wales at this period.) While it is not possible to claim that they were representative of the whole country, they stand for one important segment of its population. Secondly, the three towns were quite diverse in their economic, social and cultural characteristics, and in their fertility experience, as chapter three will show. Their selection therefore allowed useful comparisons to be drawn, helping to identify factors which were major influences on fertility. The diversity of the three towns also meant that findings which were common to all three were more likely to be generalisable to other English towns.
The study begins in 1860 because the 1860s were the last decade before local aggregate birth rates started to decline, before legislation on universal education, and before the key economic changes began which John Clapham described as the ‘gigantic hinge’, shifting the distribution of employment and wealth in England from northern industry to southern international commerce and banking.9 The 1860s are therefore the last decade before important changes set in. The arguments for closing in 1920 are less strong, since the fertility trend resumed its downward trajectory after this. 1920 has been selected to exclude the period of the postwar collapse of staple industries, which brought new social changes.
The study represents a dialogue between the quantitative study of information about birth rates, employment, and real wages, and the qualitiative study of cultural phenomena such as diets, clothing, and attitudes to childrearing. Carus and Ogilvie have recently described this approach as the micro-exemplary method.10 The approach is dialectical, in the sense that examination of one type of evidence raises questions which need to be answered by examination of the other type, in an iterative process. In this study, quantitative evidence came principally from national records of the registration of births, from the Census Reports of 1861 to 1921, from Parliamentary Papers, particularly of labour statistics including wages, and from the (digitised) local Census Enumerators’ Books (CEBs) for 1881. Qualitative evidence came from the newspaper and oral history sources already mentioned, working-class autobiography, advice literature directed at the working class of the three towns and reports of social conditions by, for example, doctors and social reformers. The merits of each kind of source and the use made of it are discussed in chapters three (for the quantitative material) and four (for the qualitative).
The thesis is structured as follows. Chapter two introduces the concepts necessary in this study. It reviews the literature on the fertility decline, and sets out some new interpretations of its causes, introducing the central idea of rising expectations about living standards, and discussing the literature on the history of consumption which contributes to this insight. Chapter three highlights relevant features of the local economy, society and culture in each place, before moving into the detailed exposition of quantitative material which sets out the contours of the fertility decline in the three towns selected for study, reporting on the calculation of local birth rates, and then on investigations into male, female and child employment, wages and the discovery of local trends in real wages. Chapter four presents qualitative evidence for rising expectations about how a working-class family ought to live, in the fields of material living standards and leisure. Chapters five and six develop this theme further with special reference to the perceived needs of children. Chapter five concentrates on the impact of the extension of education, and also discusses the effects of the waning of child labour. Chapter six widens the discussion to take in the impact on fertility of changing ideas about the nature of childhood and family life, which made bringing up a family more demanding. Chapter seven draws together the findings of this study and discusses their implications, considering in turn how different the towns’ fertility experiences were, the relative importance of economic and other underlying causes, chronology, including the time-lags with which different processes produced a decline in fertility, methodological lessons learned from this study, and directions for further research.
- 2 -
Thinking about fertility and culture
This chapter introduces the concepts necessary in this study. The study bridges the gap between two fields, demographic history and cultural history. In this chapter, the literature of each is examined in turn for its relevance to the late nineteenth-century fertility decline, and more particularly its manifestation in urban northern England. This survey demonstrates what the current literature can explain, some of its limitations, and the opportunity which these create: an opportunity to add to the understanding of the fertility decline by exploring its cultural origins. This chapter is more than a literature review in the strict sense: it both reviews the literature relevant to cultural explanations of the fertility decline, and proposes some new interpretations of its causes for subsequent testing. Specifically, the chapter introduces the role of rising working-class expectations about living standards, which play a central part in the study.
The demographic literature demonstrates how the ever greater refinement of statistical methods always falls short of its goal of explaining the fertility decline. A good example is Minoru Yasumoto’s demographic study of Leeds in the earlier nineteenth century, which proceeds entirely by the statistical analysis of vital events and address data, and can offer only the most limited explanations for the trends which it finds.11 Meanwhile, the cultural history literature has almost completely overlooked the connection between rising working-class expectations about living standards and family limitation. The need is for studies which combine the demographic and cultural approaches in a dialectic fashion.12 It is true that there have been many of these since Peter Laslett and John Harrison wrote about population and family in Clayworth and Cogenhoe in 1963.13 However, as the survey in this chapter will show, there is no study of the later nineteenth-century fertility decline which adequately addresses the impact of rising expectations. The present work steps into this gap.
There is a large literature about the demographic history of England in the nineteenth and early twentieth centuries: the secular decline in fertility was so striking that it attracted attention even while it was still taking place.14 Those with a mainly statistical approach have been eager to establish exactly what the trends in births, deaths and marriages were (‘vital events in the demographers’ terminology); historians more interested in interpretation have discussed both how couples could have reduced their marital fertility and why they did so when they did. The second question is proving particularly difficult to resolve. As Bob Woods observed in 1992, ‘hypotheses abound, but the evidence remains tantalising in its vagueness and insecurity.’15 In 1971, John Habakkuk could write that we knew little of the chronology of the ‘small planned family...the methods by which it was effected, and about its fundamental causes.’16 The survey of demographic history sources with which this chapter begins will adopt his threefold division, starting with the chronology. It will then look at fundamental causes because these are logically prior to methods. Some of the historiography has unhelpfully concentrated on methods and fallen into an assumption that women and men must always have wanted smaller families. Nevertheless, for an underlying motive, once established, to turn into actual family limitation, an effective method must be available, so the discussion of demographic sources will conclude by confirming what is known about the availability and acceptability of means of family limitation.
Substantial progress is being made in the reconstruction of a chronology of fertility, and this helps the sifting of different explanations by winnowing out those producing different fertility patterns from the ones now reconstructed. Before the modern rise of computing in historical research, the use of record linking to reconstitute the vital events of past families (‘family reconstitution’) was almost prohibitively time-consuming, and in 1984 Michael Teitelbaum was pessimistic about the prospect of being able to study vital events at a level below the aggregated tables produced for counties by the Registrars General.17 Today, however, researchers can have considerable confidence about the chronology of the English fertility decline, if not about the answers to Habakkuk’s two deeper questions. The use of computers has permitted the linking of birth registrations to particulars of the mother and father held on other documents, at a scale which allows powerful analytical methods including family reconstitution to be employed on a scale which offers insight into entire communities of substantial size, rather than, as in the 1960s, Laslett’s small settlements of Clayworth and Cogenhoe. The leading role in this kind of study has been played by the ESRC Cambridge Group for the History of Population and Social Structure.18
This research indicates that fertility in England and Wales peaked around 1815-1820, declined sharply in the 1820s and 1830s before slowing up and then rising gently until the 1860s or 1870s. After this it fell to levels never before seen, reaching a minimum in the 1930s. Figure one illustrates this long-term trend by showing the gross reproduction rate (the number of daughters born per woman). Thus the aggregate fertility decline can be dated, according to preference, from about 1820 or about 1875, but both are genuine turning points in the long-term trend. As will be discussed below, however, the England and Wales aggregate masks a fractured and complex picture once it is broken down either geographically or occupationally.