Parenthood, child-rearing and fertility in England, 1850-1914.

Fertility declines across Europe and the Anglo-world have been explained as the result of reversals of intergenerational flows of wealth. According to this theory, the child was transformed from an economically-useful household asset to an emotionally-valued parental burden. This article is based on a comparative study of changing understandings of parenthood in three provincial English localities between 1850 and 1914. It works from the premise that in order to make sense of reproductive behaviour, it is essential to examine the meanings that men and women attached to childlessness, child-rearing and parenthood. It is argued that there was not a universal shift that made children into burdens. New understandings of the duties of parenthood did develop, but these were founded on class-, gender- and place-specific interpretations. These encouraged a minority of fathers and mothers to believe that together they had the capacity to improve the lives of their sons and daughters in pioneering ways. Given that husbands and wives had distinct motives for avoiding rearing many children and that the discussion of reproduction was shrouded in silence, the dissemination and use of new ideals of family was crucial in enabling birth control to be thought about respectably within marriage.

Women who married in England in the 1860s bore an average of more than six children while their granddaughters who married in the 1910s bore fewer than three children, as the national birth-rate moved towards its nadir in 1933 (Anderson, 1990, pp. 28, 39). Some have explained this demographic change as the result of a transformative reversal of ‘intergenerational flows of wealth'. Birth control practices diffused down the class hierarchy as couples across Europe and the Anglo-world rationally adjusted to new socioeconomic circumstances (Becker, 1981; Caldwell, 1976). At the heart of this theory is the shift from the child as an economically-useful asset within the household economy to the economically-dependent ‘sacred’ child within the affective nuclear family (Gillis, 1996; Zelizer, 1985, p. 295). Yet there is a need for qualitative studies to assess the extent to which men and women did interpret their interaction with their daughters and sons in this way. This article argues that an understanding of the changing and diverse relationships between generations is essential in explaining why fertility declined as it did before 1914.

Most recent studies have emphasised the variety and complexity of reproductive behavioural change (Janssens, 2007a; Woods, 1987, pp. 309–311). Drawing principally on evidence from the 1911 census of married women's fertility, Szreter has argued that British couples’ decisions to reduce their family size were prompted by reassessments of the ‘perceived relative costs of child-rearing'. These decisions were based on married adults’ gendered roles within ‘communication communities', determined principally for the working class by local labour markets and for the middle class by national social status groupings (Szreter, 1996, pp. 546, 554). Quantitative analysis of fertility in 13 English districts similarly emphasised the strength of ‘community’ norms (Garrett, Reid, Schürer, & Szreter, 2001). Comparative studies in other national contexts have shown that religious denomination and ethnicity affected patterns of fertility through distinctive attitudes to family, sexuality and the planning of life-course events (Derosas & Van Poppel, 2006; Janssens, 2007b; McQuillan, 2004; Praz, 2005). It is not known whether there were distinctive understandings of raising children in England that resulted in diverse patterns of fertility or where boundaries between ‘communities’ lay. Nevertheless, it is widely accepted that it is necessary to move beyond ‘average’ national narratives, both quantitatively and qualitatively.

In quantitative terms, the use of the mean number of children born per woman in her fertile years has been called into question by the presence of divergent and extreme sub-groups. One such group was the increasingly large minority of couples with ‘highly restricted fertility', whose marriages remained childless or resulted in the birth of only one child (Anderson, 1998; Rothery, 2009, pp. 679–682). Micro-level longitudinal studies have suggested that some couples pioneered the adoption of new ‘reproductive strategies’ that reacted to the fall in child and infant mortality, so as to achieve a small ‘ideal (net) family size’ (Reher & Sanz-Gimeno, 2007; Van Poppel, Reher, Sanz-Gimeno, Sanchez-Dominguez, & Beekink, 2012, p. 325). In the Netherlands from the 1870s these active and planning couples were drawn from Liberal Protestant skilled and elite backgrounds, and thus of a similar class to those identified by Anderson with extremely low fertility (Van Bavel & Kok, 2010b). It has been suggested that Belgian restricted fertility was part of a ‘defensive strategy’ to prevent intergenerational downward social mobility, especially amongst the middle class (Van Bavel, Moreels, Van de Putte, & Matthijs, 2011, p. 338). In exploring these distributions, the attitudes of adults who did not rapidly follow what demographers class as the ‘pioneering’ trend towards investing in few ‘high quality’ children should be taken as seriously as the perceptions of those who did.

The tendency to focus on a smooth, national narrative of increasing reproductive control applies equally to qualitative research. Scholars have examined the scandals generated by the publication and reception of birth control advice (Banks & Banks, 1954; Caldwell, 1999; Cohen, 1993). Yet, more everyday, oral and implicit – and thus potentially more morally secure – settings could also be influential contexts in which family size was thought about in new ways (Watkins, 1990). Oral history testimonies have provided most qualitative evidence for these contexts in which knowledge of reproduction was communicated or silenced. Testimonies provide rich and apparently frank insights into attitudes to sex and sexuality, but unavoidably only offer historians the narratives of individuals who reached adulthood in the early or mid-twentieth century, decades after couples had started reducing their family size (Beier, 2008; Szreter & Fisher, 2010).

Since the 1980s the concept of the gender-less demographic actor has been challenged through studies that have emphasised the need to explore the different motivations of, and power relations between, men and women, which cannot be identified in quantitative data (Janssens, 2007b; Mackinnon, 1995; Watkins, 1993). Yet there is little historiographical consensus on the relative importance of open spousal conflict, unspoken tensions, strategic negotiation, or marital agreement on distinct gendered roles in effecting changes in reproductive cultures. The most frequently advanced argument suggests that rising public concern about maternal health and responsibilities drove the decline in family size (Ross, 1992; Seccombe, 1990). Explicit female agency in reducing family size has been suggested as a result of the economic power of highly-paid female textile workers (Gittins, 1982; Ittmann, 1995; Schwarzkopf, 2007) and amongst the increasingly educated and sometimes feminist women of the middle class (Mackinnon, 1997). This narrative of a fertility decline that followed female authority has been challenged by oral history testimonies which suggest that interwar young women sought to distance themselves from birth control, relying instead on masculine ‘skill’ and ‘considerate’ behaviour (Fisher, 2006; Szreter & Fisher, 2010, pp. 229–267). This indicates the need not only to examine the relative authority of husbands and wives in reproductive decision-making, but also to explore how fertility was positioned in relation to other aspects of gendered selfhood and familial intimacy.

Methods and contexts

This article draws on research that was conducted as part of a broader study of English parenthood between 1850 and 1914. A locally comparative and deeply contextualised approach was adopted, focusing on three large and contrasting provincial localities. The survival of archival sources determined the precise districts chosen. Table 1 provides details of employment in the studied areas.

Table 1.

Principal occupations (employing 10% or more) of fathers, mothers, boys and girls in the three study localities, drawn from sampled census enumerators’ books 1861 and 1901 (2.5% sample of all adults recorded as co-resident with an individual described as a son or daughter aged under 21 years; and separate 1% sample of all individuals aged under 21 years).

Study area	Auckland, County Durham (North-East England)		Burnley, Lancashire (North-West England)		Bromley, Kent (South-East England)
Census year	1861	1901	1861	1901	1861	1901
Principal paternal occupations	Mining: 45%Manufacture: 26%	Mining: 62%Manufacture: 16%	Textiles: 32%Manufacture: 17%Building: 14%Mining: 11%	Textiles: 33%Manufacture: 19%Mining: 12%	Agriculture: 31%Manufacture: 25%Professional: 10%Industrial service: 10%	Building: 17%Dealing: 17%Manufacture: 15%Domestic service: 13%Professional: 11%
Principal maternal occupations	No occupation listed: 97%	No occupation listed: 98%	No occupation listed: 77%Textiles: 18%	No occupation listed: 70%Textiles: 26%	No occupation listed: 85%	No occupation listed: 92%
Principal occupations of boys aged less than 21 years	Not in employment:63%Mining: 21%	Not in employment:71%Mining: 17%	Not in employment:61%Textiles: 21%	Not in employment:62%Textiles: 18%	Not in employment:79%	Not in employment: 78%
Principal occupations of girls aged less than 21 years	Not in employment: 92%	Not in employment: 90%	Not in employment:: 64%Textiles: 32%	Not in employment:65%Textiles: 30%	Not in employment:92%	Not in employment: 72%Domestic service: 18%
Total sample size	330 fathers;341 mothers;134 boys;117 girls	616 fathers;625 mothers;237 boys;218 girls	266 fathers;278 mothers;94 boys;111 girls	652 fathers;699 mothers;229 boys;230 girls	109 fathers;106 mothers;52 boys;37 girls	484 fathers;493 mothers;215 boys;199 girls

The first locality is Auckland, a coal-mining district in County Durham, centred upon the market town of Bishop Auckland. By 1901 62% of fathers worked in the mines. Another 16% worked in manufacturing, mainly in associated, heavy industries. Maternal employment was rare. About half of the population belonged to nonconformist Protestant denominations. Local government adopted a conservative, laissez-faire approach to social issues and the largely Anglican landed owners of the coal mines were influential. Sanitary conditions were poor in mining towns and villages and infant mortality was high.

The second area, the Lancashire town of Burnley, was also dominated by a nonconformist working-class population and, to a lesser extent, by a single industry. Cotton mills employed just under one-third of fathers, while just over one-tenth of fathers worked in small coal mines. Up to one-third of mothers recorded themselves as being in employment, primarily in weaving mills, and young female and male workers remained in demand. Local government contrasted politically with that in Auckland and many men and women prided themselves on their progressivism. Burnley was the northern stronghold of the Social Democratic Federation from the 1880s and of working-class suffragists from the 1900s. In spite of investment in public health infrastructure, infant mortality remained high.

Third, Bromley in Kent was a market town that, with the arrival of the railway from the 1850s, became only a 20 minute train ride from London. By 1901 almost one-quarter of fathers had middle- and upper-class occupations. These families lived in newly-built suburban villas where they enjoyed low mortality rates and a wide range of amenities. The southern part of the district offered low-paid and declining opportunities for agricultural work, so that working-class parents were increasingly employed in service provision. In contrast to the strength of Nonconformity in the industrial towns, the Church of England dominated and those involved in local government advocated conservative paternalism. Many elite women participated in a wide variety of voluntary philanthropic and increasingly political movements, including some for women's rights.

It is important to establish what we know about quantitative trends in average fertility rates in these localities. By 1911 the population of each district reached just over 100,000 people. There are three features to highlight from the fertility rates presented in Table 2. First, irrespective of locality, from the 1870s the average (mean) number of children that women bore declined. This was true of both legitimate and illegitimate fertility rates. Second, the average fertility rates for women in the three localities were sustained at contrasting levels, with the gap between high-fertility Auckland and low-fertility Bromley remaining fairly constant. Third, the precise timing and rate of decline differed in these three areas. Bromley matched the national pattern most closely, though the rate was below the national average with a more rapid decline from 1870. Auckland's fertility peaked in 1871 at an exceptionally high rate of 235 births per 1000 women aged 15–45 years; in spite of declining thereafter, by 1911 fertility rates were comparable to those in Burnley 60 years earlier. The birth rate in Burnley declined rapidly, especially in the 1880s. It was only Burnley that altered its relative position compared to the other localities and the national average, suggesting a local transformation in reproductive culture.

Table 2.

Number of births per 1000 females aged 15–45 years in the three study localities 1851–1911 (and % of 1851 rate), as recorded in the annual reports of the Registrars-General.

	1851	1861	1871	1881	1891	1901	1911
Auckland, County Durham	221.3	221.8	234.8	209.1	191.2	163.9	151.0
	(100%)	(100.2%)	(106.1%)	(94.5%)	(86.4%)	(74.1%)	(68.2%)
Burnley sub-Registration	156.0	151.6	167.6	167.1	123.9	105.8	84.6
District, Lancashire	(100%)	(97.2%)	(107.4%)	(107.1%)	(79.4%)	(67.8%)	(54.2%)
Bromley, Kent	129.0	129.9	131.1	117.8	94.1	80.8	59.1
	(100%)	(100.7%)	(101.6%)	(91.3%)	(72.9%)	(62.6%)	(45.8%)
England and Wales	145.4	147.6	152.2	147.5	132.7	114.5	98.0
	(100%)	(101.5%)	(104.7%)	(101.4%)	(91.2%)	(78.7%)	(67.4%)

Studies of diverse historical contexts have demonstrated that ‘natural’ fertility was far from unregulated before the fertility decline (Bengtsson & Dribe, 2006; Van Bavel & Kok, 2010a; Vann & Eversley, 1992, pp. 152–176). Likewise, there was no ‘contraceptive revolution’ to instigate the English fertility decline. The reduction in the birth rate was achieved primarily through the postponement or avoidance of marriage and within marriage through abstinence or withdrawal (Cook, 2004, pp. 40–164; Szreter & Fisher, 2010, pp. 229–267). This research does not focus on methods of fertility limitation, but evidence from the studied localities supports this conclusion. There were no references to the use of barrier methods of contraception before 1914; instead, abstinence, extended breast-feeding and abortifacients were most frequently noted as methods that might limit child-bearing. Given that this suggests that continuities in methods coincided with this transformation in rates between 1870 and 1914, it is important to consider what either encouraged people to use these practices more consistently or made a higher proportion of adults willing to countenance their use.

This study works from the premise that men and women had agency in sometimes thinking about and seeking to shape – often unsuccessfully or with unexpected consequences – their own lives, their capacity to generate future lives, and the lives of any sons and daughters. A wide range of qualitative archival and published primary sources were studied. Details are provided in Table 3, but these included: published literature, especially newspapers and advice manuals; local manuscript records of schooling, justice, government and philanthropy; census enumerators’ books; and all surviving personal documents such as letters, diaries, autobiographies and oral histories. Analysis was primarily qualitative, using a grounded theory methodology; sources were transcribed and coded in order to facilitate the identification of common and distinctive themes.

Table 3.

Principal primary sources.

	Auckland, County Durham	Burnley, Lancashire	Bromley, Kent
School, church associational records	79 sets of elementary school logbooks3 sets of church, Sunday school or charity records	46 sets of elementary schoollogbooks10 sets of church, Sunday school, private school or charity records	33 sets of elementary school logbooks14 sets of church, Sunday school, private school or charity records
Local state records	Poor Law and public health minutes5 sets of school board minutes	Petty session court records from 18844 sets of school board minutes	Poor Law and public healthminutesPetty session court records from18905 sets of school board minutes
Personal sources	21 sets of personal papers	20 sets of personal papers	43 sets of personal papers15 oral history testimonies
Local newspapers	Every edition of principal local newspapers was read for the three study localities 1859–61, 1869–71, 1879–81, 1889–91, 1899–1901, 1909–11. Additional editions of these newspapers were sampled around selected dates
National literature	The content of every edition of the 13 child-rearing manuals that were published in at least five editions or 10,000copies 1800–1914 was read and compared word-by-wordA sample of national newspapers was also searched for specific themes and events
Census enumerators’ books 1861 and 1901	2.5% sample of all men and women resident in the same household as at least one individual aged less than 21years and described as son or daughter1% sample of all children aged less than 21 yearsFurther contextual details were accessed from all censuses 1841–1911 for named individuals

First, this article examines critically the conceptualization of children as economic and social burdens by considering the extent to which elite and then non-elite parents understood the ‘cost’ of their children to be increasing. Second, the emergence and communication of new cultural ideals of family are explored, by focusing on three models that legitimised particular family sizes. In each section, any distinctive patterns relating to class, place, gender or ethnicity are identified. It is argued that those fathers and mothers who reduced their fertility earliest both perceived their duties towards their sons and daughters to be rising to new heights and drew on familial ideals that could be used to construct birth control as a moral practice that could legitimately be explored within marriage.

The cost of children

In seeking to analyse demographic responses through the ‘strains’ placed on adults, it is necessary to chart the changing demands that were made on the time and resources of parents by the law, the economy and societal expectations (Friedlander, 1983, pp. 266–269). Yet it is also essential to gauge the extent to which men and women knew about these pressures and how far they interpreted them as ‘burdens’ that were sufficiently weighty to influence their behaviour. It will be suggested that parents understood their roles in two class-specific ways. Elite fathers worried about the sustained responsibilities that their relationships with teenaged and adult sons and increasingly daughters placed upon them. Less wealthy parents, especially mothers, focused on the immediate and increasing possibilities for providing care for younger children. Such contrasting perceptions of parenting influenced understandings of the points in the life-course when it was especially beneficial to try to avoid bearing more offspring.

Elite perceptions of sons and daughters

Men from substantial property-owning families where multiple servants were employed were most likely to use a language of ‘costs’ in discussing relationships with children. The correspondence between Charles Darwin, who was rearing his ‘large & growing family’ (at that point of seven children) and other gentlemen was typical in being littered with such fears, to such an extent that he remarked that T never meet any one who is not perplexed what to do with their children’ (Darwin, C. letter to J. Higgins, 1850, December 21; Darwin, C. letter to S. Covington, 1857, February 22). Research into elite masculinities has catalogued the increasing cult, and resulting expense, of boys’ public schools in the second half of the nineteenth century (Banks, 1954; French & Rothery, 2012, pp. 39–136; Tosh, 1999/2007, pp. 102–122). Gentlemen expressed concerns in all of the studied private letters and diaries about the future challenges of providing suitably for sons. Men's unease related to three long-lasting duties: the expense of providing their sons with an increasingly lengthy education; the social and organisational responsibility of selecting judicious career paths for their sons as competition multiplied; and unease about the provision of a fairly-divided inheritance. The advertised prices for tuition at boys’ private school ranged widely, but an increasingly high proportion of schools that appealed to parents around London offered the most expensive tuition and boarding.[1] For fathers who aspired to give their sons a university education, the annual cost of supporting a son at an Oxbridge college was more than triple the cost of a boarding school (Marsham-Townshend, H., accounts, 1897).

Early nineteenth-century paternal worries had related overwhelmingly to the demands of rearing young men, such that sons were perceived to be three-times as troublesome as daughters (Darwin, C. letter to W.D. Fox, 1852, March 7). From the mid-nineteenth century, fathers expressed growing concerns about the challenges of educating and supporting daughters until a difficult to engineer, late or non-existent marriage. By the 1880s in southeast England there was the option of sending girls to private schools that were as expensive as those of their brothers. Additionally, young women increasingly expected a lengthy period of carefreeness to continue after their formal education ended. One gentlewoman recorded her friend's unease when she received a proposal of marriage in her early 20s in 1855: ‘The poor child said amid her tears “I thought I was a child and that every body would think me only a child, but now I find I am a woman'” (Hall, E. M. diary, 1855, December 27). This perception that girlhood continued until marriage made it pressing for fathers to protect their daughters’ vulnerability and to provide for dependence that potentially lasted for many decades. From the 1850s, men increasingly understood their paternal duties to have doubled because they expanded with the birth of each daughter as well as each son.

In the mid-nineteenth century the primary means by which men prudentially avoided these responsibilities was by postponing marriage. Fathers urged this strategy on young men. One instance – recorded in unusual detail – occurred in Lancashire in the late 1850s when Philip Hamerton, a 24-year-old gentleman, was warned against early marriage by his prospective father-in-law, a banker and father of four. He ‘expressed the regret that in our class, a family of children should be a cause of weakness instead of strength', claiming that sons ‘only weaken the father by draining away his income'. He warned that daughters had rejected their customary ‘middle-class’ roles as substitute servants and ‘do nothing of the least use and require to be first expensively educated, and afterwards expensively amused’ (Hamerton, 1897, pp. 213, 224–226). The postponement of marriage was not solely a response to perceptions of parenthood, but these formed part of a set of beliefs that publicly emphasised the disabling ties of early marriage. In this case, the banker's daughter rejected the proposal and never married; Hamerton married a Parisian woman the same year, but had only three children. Importantly, while men drew on this rhetoric of grudgingly accepted liabilities, fathers also privately expressed intense feelings of loneliness, grief and sometimes hostility when their sons or daughters left home (Quennell, 1976, p. 12). Active fatherhood was integral to male identities throughout this period, but it was principally through a language of burdensome duty that the significance of these relationships could be discussed with male peers and communicated intergenerationally.

Given these prudential warnings, it is not surprising that age at marriage rose most for elite men. Smaller increases in age at marriage for professional and gentry women reduced the number of years over which they were at risk of pregnancy and excluded their most fertile years (Rothery, 2009, pp. 680–682; Szreter & Garrett, 2000, pp. 64–67). Before the late 1870s wealthy women seldom bore their final child before they reached their mid-40s and this ‘starting’ effect was the principal means by which elite fertility was limited.

Two caveats should be placed on this explanation for middle-class fertility decline (Banks, 1954). First, these wealthy men who discussed new paternal duties were spatially concentrated in London and the surrounding commuter counties, as well as, to a lesser degree, the suburbs of the large provincial cities (Rubinstein, 1988). Yet in suburban London men with lower-middle-class jobs seldom expressed these concerns about long-lasting paternal duties. The sons of clerks or shopkeepers attended small private schools, but entered paid work early in their teenaged years, often acquiring residential, social and financial independence (Baxter, ca. 1920s, pp. 7–10; Wells, 1934, pp. 115–154). Other sons and daughters provided essential labour in their parents’ businesses. In 1885 George Allen, a publisher living in Kent with eight children aged 11–27, described proudly how ‘I have six [children] at home all more or less in the business.’ He compared his family with that of his Australian brother-in-law:

We were much interested in the account of your children also we had a laugh at what you said about the reverse of things in Australia to what they are here – I mean as to large families of course. In my case a large family is a boon but in most cases it is the other way on I think especially nowadays… (Allen, G. letter to J.T. Hobbes, 1885, November 29).

It is revealing that Allen was not only aware that his expanding middle-class social circle viewed ‘large families’ increasingly unfavourably, but also that he was confident that this judgment did not fit his own experiences. In fact, for men of his generation who married before the late 1870s, this divergence – between an awareness of elite male concerns that many older children became burdensome and an acceptance that once married they personally expected to raise a large family – was common.

Second, in none of the studied sources did fathers identify the task of providing immediate care and education for children aged under 12 years as an increasing liability. Further, no mothers expressed the view that their duties to their younger or older children were newly demanding. Throughout this period many expressed fears about child-birth and weariness at the constraints of breast-feeding. In 1878 41-year-old Sibella Bonham Carter wrote to her mother-in-law about nine months after the birth of her eleventh child, voicing these concerns:

This afternoon I shall again drive out late, for I must call upon some friends whom I have hardly seen this year. It is shocking how little one can do when there is a tiny Baby to be attended to. Now that I have weaned mine I have much more leisure (Bonham Carter, S. letter to ‘Granny', 1878, June 27).

What changed amongst elite women from the late 1870s was that – partly due to the norm of a longer, idealised phase of relative freedom in girlhood – these women aspired to a wider variety of roles after child-bearing. Most prominent in diaries and letters were the attractions of practical and emotional closeness to husbands, of philanthropic and public roles, and of entertainment and travel. Elite women who married after the late 1870s – whether conservative, feminist or neither – rarely bore children beyond their mid-30s. This created an extra decade of less maternity-intense married life. For those women who had married in their early- or mid-20s, this meant that they bore families of around five children, while the majority who had remained single for longer, bore three or fewer. None of the women revealed how they avoided child-bearing in their later fertile years. Nevertheless, for these elite women, it was understandings of novel non-maternal roles from the final quarter of the nineteenth century, rather than perceptions of newly weighty maternal responsibilities, that were foremost in their minds.

The much-cited anxiety about the rising burden of providing for children was not a universal and immediate concern of all middle- and upper-class parents. Instead it formed a distinctively sustained part of a culture of elite prudential masculinity. It was only once these concerns about paternal duties coincided with wives’ expressed and realisable aspirations that stretched beyond the maternal that elite fertility was systematically controlled not just through delayed ‘starting', but also through ‘stopping’ within marriage.

Working-class and lower-middle-class perceptions of sons and daughters

The above explanation applied to no more than one-tenth of the English population (McKibbin, 1998, pp. 44–45; Trainor, 2000, p. 681). The majority of parents, including those from the lower middle class, did not consider their relationships with teenaged and adult children to be changing. Instead, it was the provision of suitable care for the youngest children that parents thought to be newly within their power. Although caring roles were taken on more frequently by mothers nationwide, this interpretation was often also expressed by fathers, especially in districts dominated by the textile industry.

The most universal change in understandings of non-elite parents’ roles related to new expectations of care for children aged under 12 and for ill children. Published commercial and didactic literature expressed rising concerns about the ‘suffering’ of children, especially in the first two years of life. Representations of both fathers and mothers as the potential saviours of their tragically ‘delicate’ children became emotive in widely-published advertisements for children's patent medicines, coinciding with the decreased frequency with which parents experienced the death of a child beyond infancy (Woods, 2000, pp. 247–309). These caring duties were extended for mothers by the increasingly high standards of domesticity for which ‘respectable’ mothers laboured, especially in the early twentieth century (Bourke, 1994; Davin, 1978; Ross, 1993, pp. 195–221). Concerns were intensified by school attendance. Teachers expected high standards of cleanliness, but parents also considered schools to be unhealthy spaces. This prompted fathers and especially mothers to take on new roles in overseeing schooling by intervening to protect their children in these threatening non-domestic environments (Pooley, 2010, pp. 537–541). The 1911 census has suggested that husbands who worked as shopkeepers and clerks were amongst the earliest to rear extremely small families from the 1870s (Anderson, 1998, pp. 182, 192–193). These fathers and mothers seldom expressed the belief that a norm of investing in children was diffusing from the state or the elite, but instead focused on their expanding and progressive parental capacity to improve their children's lives voluntarily through providing a better quality of care and education than they had experienced.

This is not to suggest that the financial cost of rearing children remained unchanged, but non-elite fathers and mothers seldom claimed that the rearing of children was newly economically draining. Living standards rose from the 1850s, especially through increasing real wages for industrial workers (Boyer, 2004, pp. 282–290). The highest rates of employment early in childhood occurred before 1870, so that a sudden loss of children's income did not coincide with the period in which working-class parents reduced their family size (Humphries, 2010, pp. 172–209). In the studied localities, teenaged and adult children were increasingly valuable contributors to the household economy of their ageing parents, especially through the rising demand for daughters as live-in servants who ‘tipped up’ wages (Census, 1861 and 1901).

The introduction of compulsory elementary schooling in 1880, which was most financially burdensome until it was made free from 1891, did coincide with increased non-elite fertility control. However, it should be remembered that most mid-nineteenth-century parents chose to give their sons and daughters sufficient schooling to make them literate before education became compulsory, usually paying fees that were higher than those charged by board schools (Vincent, 1989, pp. 22–29, 66–72). Further, as a result of the failure of many school boards to enforce the legislation, fees were not a cost that suddenly increased to unprecedented levels. School attendance authorities in both the large coal-mining district of Auckland and rural Bromley systematically failed to enforce attendance until at least 1902. Further, all of the studied Auckland schools and some Bromley schools capped parents’ costs, so that if more than two or three children attended from the same family, the youngest children were schooled for free. Parents with a large number of closely spaced children were thus recognised as deserving of special assistance.

The exception to this was in the Lancashire town of Burnley, where the duties of parents were widely emphasised and rigorously enforced. This region experienced an early and rapid decline in working-class family size from the 1880s. Men and women who worked as cotton weavers were paid identical wages (though opportunities for promotion were reserved for men) and fathers were routinely responsible for domestic chores and childcare (Gittins, 1982, pp. 95–124; Schwarzkopf, 2004, pp. 29–46). By 1901 28% of parents recorded themselves as part of a dual-income couple, so that a sizable minority of both fathers and mothers faced the double demands of paid work and intensifying expectations of childcare (Census, 1901). While recorded maternal employment became rarer and more concentrated only in the poorest households in Bromley and Auckland, in Burnley the proportion of mothers recorded in work increased. Although this view was not expressed by parents in the studied sources, the opportunity cost of leaving work to care exclusively for many children was thus unusually high.

An exceptionally active range of mutual, philanthropic and local government agencies sought to ensure that Burnley parents fulfilled their responsibilities. School fees were on average twice those in the other studied localities and a higher proportion of children stayed in education for the increasingly sought-after higher grades. In 1871 it was found that 81 % of pupils paid between two- and four-pence per week, but more than one-tenth of pupils were paying four-pence or more (‘Meeting of the School Board’, 1871). These fees rose in all of the schools over the following 20 years. None of the studied Burnley schools recorded a policy of capping the fees paid by families, so that the cost of schooling large families was unusually expensive. Although most people justified these costs by citing the perceived high quality of education and facilities, by 1891 a radical School Board member complained that ‘in nearly all the board schools of the country the fees were less than were now charged in Burnley’ (‘Burnley School Board’, 1891). The Board was also rigorous in seeking to ensure that children attended school for 10 half-days per week; it was, unusually, willing to warn and fine ‘respectable’ parents who failed to comply with the attendance by-laws (Pooley, 2009, pp. 80–87, 174–178).

One typical case will be used to indicate this parental pressure. In 1901 an inquest was reported in The Burnley Gazette, following the death of a five-year-old girl from burns she sustained while in the care of her eight-year-old sister. The coroner warned her mother and especially father, both of whom were weavers, as follows:

Do you think it is a proper thing for you men to take upon yourselves the responsibilities of parents, to go and get married, and have children, and then leave them alone in the house? … The jury had taken the most lenient view, but if people would have children, and then, out of their own selfishness and greed for getting money, without making proper provision for looking after them, went and left children in the house, juries would have to take notice of it, and would bring in, instead of a verdict of accidental death, a verdict of manslaughter. (‘The care of operatives' children’, 1901)

The words of the coroner formed a powerful conclusion to a case in which the parents had described their efforts to provide non-parental child-minders. The coroner ended by stating that ‘he hoped the reporters would insert his remarks in all the papers, so that it would be known that all well thinking men would not allow that sort of thing to go on.’ The ‘responsibilities of parents’ were painfully prominent and frequently displayed to readers of these local newspapers, explicitly challenging parents’ most intimate decisions about when it was appropriate to marry and how to provide suitable childcare. Caring duties were thus made publicly integral to both fatherhood and motherhood.

From out of this unusually gendered regional economy and the, linked, exceptional culture of rigorously promoted parental responsibilities emerged England's most popular affordable child-rearing manual. The wife's handbook was published by Henry Allbutt, a doctor in the nearby woollen manufacturing town of Leeds, in 1886 at the price of sixpence. By its forty-fifth edition in 1913, 430,000 copies had been sold. It has been presumed that this popularity resulted from its explicit instructions on methods of birth control (Fryer, 1965, pp. 169–171). Yet Allbutt also provided unusually practicable instructions on childcare and offered a pioneering conceptualisation of the relationship between health, marriage and parenthood. Unlike other manuals that treated female and child health separately, the volume was structured around the mother's life-course, so that child-rearing was integrated into a longer time-span of ‘sufferings of women and children'. The book opened with the warning that ‘From the first marriage-night no woman under forty-five years of age can consider herself SAFE'. Allbutt drew on a new rhetoric of idealised parental omnipotence by which the child's constitutional health could be assured from birth. If parents ignored the laws of ‘Nature', they would continue ‘bringing into the world feeble and puny children, born but to suffer and die'. Yet if they followed the ‘laws', fathers and mothers gained the power ‘to rescue from death and disease children who may be born’ (Allbutt, 1887, pp. 3–5). In linking behaviour in the marital bed to the duties of parenthood, Allbutt emphasised the responsibilities of knowledgeable and considerate husbands. The regulation of fertility was conceptualised as part of a culture of maternal and especially paternal selflessness engendered by their perceived capacity to ease children's suffering. We know too little about the readership of advice literature. Nevertheless, The wife's handbook was advertised nationwide, so it is likely that it spread this culture beyond the occupational context from which it originated, thus encouraging a more sustained nationwide shift in perceptions of parental duty.

Thus, we can explain the early fall in non-elite fertility through the awareness that ‘respectable’ parents expressed from the 1880s of the increasing duties that they believed they bore towards their youngest and most vulnerable children. Non-elite fertility declined most in those areas where these responsibilities were thought most significant and where fathers were most involved in childcare. It is plausible to suggest that these expectations for new, demanding forms of care for younger children – both created by working-class parents and advocated by outside institutions – encouraged parents to seek to increase the gap between each pregnancy.

Ideals of family

Studies of sexuality have been unanimous in establishing the dominance of a culture of silence throughout these decades. Young women were shielded from reproductive knowledge and couples struggled to discuss birth control. Cook has examined Edwardian mothers’ efforts to pass on the revulsion that they felt about their own bodies and sexual experiences to their innocent daughters (Cook, 2012). Contexts in which this silence could be respectably broken were therefore important. This section focuses on three cultural ideals of family that were used to explore family size in new ways: a disciplined familial model; an ideal of the domesticated small family; and an aspiration for unbounded fertility. The comparative potency of these ideals in the studied localities is traced. It will be suggested that even when commentators sought to condemn birth control, the act of speaking of reproduction through positive familial ideals created alternative routes through which men and women could think about and potentially negotiate the adoption of birth control.

The disciplined family ideal

The first context in which a link between familial ideals and family size was newly publicised was neo-Malthusian. Fertility control was conceptualised as one aspect of the disciplined and health-endowing regimes that had been advocated since Chartism as a means for fathers to strengthen their own character and improve their families’ lives (Clark, 1992; Gleadle, 2003). The trial of birth control campaigners, Charles Bradlaugh and Annie Besant, in 1877 was exceptional in giving these views national publicity and is often interpreted as a trigger for fertility decline (Caldwell, 1999; Mason, 1994, pp. 68–73, 195–213).

Yet, for at least a decade before the Bradlaugh-Besant case, the principles that underpinned this neo-Malthusian culture were widely and mundanely disseminated in the working-class dominated textiles towns of East Lancashire and West Yorkshire. Not only – as established above – were duties towards young children understood to be heavy, but in this geographically-defined context, women and men were publicly urged to transform civic morality through reformed familial relationships. Importantly, these debates took place in public spaces that were the epitome of Christian morality. One typical example of this was a Wesleyan missionary mothers’ meeting in 1870, attended by more than 200 women. The nonconformist temperance lecturer and father of five, William Bell from the neighbouring textiles town of Bradford, declared that having five children was better than having a dozen since it enabled the family's bread to last longer (Census, 1871). Although not a surprising conclusion, it is revealing that it was a comparison that was accepted as uncontroversial in such a pious setting. The speaker expanded on his understanding of the enormous influence of ‘woman as a companion, a wife, a mother, and a worker', using the example of a ‘well-educated young woman’ he claimed to have met in Preston, a nearby textiles town. The woman had declared that she did not want a ‘man', but a ‘husband, one that would look after the home. A househusband; one who would keep the house together. One who wouldn't sit at home smoking while his wife was working in the mill.’ (‘A mother's meeting’, 1870) Irrespective of whether this was an apocryphal tale, similar models of authoritative wage-earning femininity and familial masculinity were commonly promoted from at least the 1870s. Efforts to limit family size were understood as a realistic mechanism for familial and then civic reform.

These ideals were promoted as shared by all moral citizens, not as class-specific concerns. One case of these cross-class relationships was described autobiographically by Mary Brown, the wife of a doctor. She was the mother of three children born before her thirtieth birthday, who – like the elite women discussed above – later took on roles that made her a leading figure in the public life of the town. From her arrival in Burnley in 1879, when her youngest child was aged two, Brown sought to carry out ‘social work’ on behalf of the working men and ‘factory girls’ through Sunday school classes. She recounted how:

Some years later when my influence was making itself felt, two young men came to see me, and this was their reason. “There are girls,” they said, “that work in our mills, that we call Mrs. Doctor Brown's disciples. They are different and it is all because of you. We want you to come and speak to us young men that we may be better company for them young girls.” (Brown, 1937, pp. 37–38).

Of course, only the stories of people affected figured in her account. Yet her efforts were just one of many reformist missions that made family life a concern of civil society and latterly of local government, with the remaking of masculinity as a particular priority. Importantly, these initiatives were successful as a result of both the high proportion of elite residents involved and their mutual character, which meant that they were sustained through active working-class initiative.

It was because of this that birth control publications were welcomed in this region. Bradlaugh was an admired visiting radical speaker before the Fruits of Philosophy trial. The reporting of the campaigns of Bradlaugh and Besant made them two of the most prominently featured national figures in the local newspapers. One typical occasion that cemented this celebrity status was a two-night Mechanics’ Institute philosophical debate in 1879 between Bradlaugh and a Congregational minister, chaired by a Baptist minister, which attracted ‘crowded audiences’ ('Has or is Man', 1879). One attendee was Philip Snowden, the son of Wesleyan weavers, who later attributed the campaigners’ appeal to the attraction of their ‘individualistic radicalism'. He also recalled their power as orators, recording T can see him [Bradlaugh] now as he stood on the platform. He was a massive figure, with a fine head and a powerful voice, and in declamation he was a tremendous force', while Annie Besant ‘even on such a subject as birth control… showed herself a superb orator’ (Snowden, 1934, pp. 43–44). In these accounts it is less what was said that was notable than the opportunity to hear these radical matters spoken of by such eloquent celebrities and under the auspices of organisations that gave these topics legitimacy. On Bradlaugh's death, The Burnley Gazette, published a feature on his life, including information on the pair's profession of ‘what are called Malthusian opinions on the population question’ and describing how:

The news of his death has produced a profound impression upon the minds of his many friends and admirers in this town, and it is the general feeling that his life has been a generous and noble sacrifice to high principles, sincere love of truth and unselfish devotion to the duties of citizenship and humanity. (‘Echoes’, 1891).

Not all the town's residents agreed with the reformers’ agenda. However, it would be surprising if their widely-documented lives and the moral contexts in which they were accepted as speakers did not encourage men and women to think about family size as something that could be placed respectably under personal control.

The contrast with the relationship between family life and public concerns in Auckland in County Durham is clear. In this coal-mining district, which was similarly dominated by an industrial working-class and nonconformist population, the local newspaper reported on the secularism of Bradlaugh and Besant, but in the sampled newspapers there is no indication that they were invited to speak in the district, no mention was made of their attitudes to birth control, praise was not offered of their principles and there was no coverage of Bradlaugh's death. The characteristics that gave them celebrity status in Burnley as defenders of the people and as reformers of the family were presumably considered of no public interest, morally unsuitable or both.

Nevertheless, it must be noted that even in these districts where public discussion of family life was rare, some individuals were vociferous promoters of birth control. One neo-Malthusian in Auckland was a coal-miner, Robert Robinson, whose wife variously worked as a mantua-maker and grocer. When their daughter was born in 1877, they gave her a middle name of Bradlaugh (Census, 1911). In 1880 Robinson invited a Parisian to speak at a ‘Fruit of Philosophy meeting’ and the reaction of the public and of local journalists highlights the contrast with the textile towns. The audience was reported as follows:

with the exception of three women – two old and one young – there were only about thirty males present, the public showing their appreciation of the event by keeping on the sunny side of the door. Those three women were evidently in search of information under difficulties …. There's no accounting for taste. ('Scrutator', 1880).

As a result of the small audience, the meeting was cut short. The ridicule expressed towards those who showed an interest in the discussion of fertility, especially women, suggests the infrequency with which the topic was discussed. Robinson justified the meeting a fortnight later by attacking the columnist ‘Scrutator', as one of the ‘would-be smart and “touch me not” moral sages’ and a ‘high-class moralist'. He contended that ‘the population question is open to discussion', objecting to ‘working men and their wives being insulted by vile imputations and suggestive motives, which he doubts whether “Scrutator” would dare to suggest against the eclat of society’ (Robinson, 1880). It is revealing that it was with accusations of class prejudice, rather than alternative familial principles, that Robinson defended his meeting.

While the public in the coal-mining district responded with indifference or ridicule to this attempt to make intimate life of public import, these issues are absent from local newspapers in suburban Bromley in Kent. It is not surprising that the middle-class journalists did not celebrate the radical democratic campaigners as their less elite Burnley colleagues did. Of course, both the ‘population question’ and the actions of Bradlaugh and Besant featured, usually negatively, in the national press of which elite Bromley parents were readers (Banks & Banks, 1954). However, it is significant to note the subject's exclusion from the medium that was most accessible to working-class readers and the perceived immorality of these discussions in this geographical context. The cultural construction of disciplined and moral family life as a means of civic reform was thus powerfully normative in one locality, but was only patchily and controversially communicated elsewhere in England.

The leisured family ideal

A second familial ideal – of the leisured, domesticated small family – became prominent only from the late 1890s and especially the 1900s. By this point, as discussed in the opening section, many elite fathers and mothers had significantly limited their fertility for decades. There were two new important pathways through which the ideal of rearing only a small number of children was communicated through national published literature to many lower-middle-class and some working-class adults. In the districts surrounding London this textual model was given particular social authority by the presence of neighbouring wealthy families whose comfortable lives contained only a couple of children.

New idealised visual images of families were disseminated nationally from the late 1890s. Commercial advertisers had previously depicted whole family groups rarely, instead imagining children alone or occasionally in the care of a mother or sibling. Out of a sample of more than 500 editions of local newspapers published between 1859 and 1901 and of popular child-rearing manuals published before 1895, there were no advertisements that depicted a father, mother and children together. Yet while these images of individuals and dyads remained in use until 1914, an alternative image of family emerged from the late 1890s. In the national press, advice manuals and some local newspapers (mostly in Bromley in Kent), nuclear family groups – containing two parents and only two children – featured increasingly frequently. These images idealised modern family life that was domesticated, comfortable and leisured. The family group was shown spending exclusive time together, usually sat around a dining table. Cadbury's was one of the first companies to disseminate this ideal of the modern private family who enjoyed healthy mutual bliss through drinking cocoa (Chavasse, 1913, p.xxi).

From the mid-1900s, this domesticity began to be imagined in less class- or place-specific ways. Whereas domestic servants had featured in earlier images of domesticity, they were seldom shown in familial scenes from c.1905 onwards. Depictions of furnishings and dress became simple and cosy rather than luxurious. This standard of domesticity was becoming affordable for skilled working-class families in these decades, if only in their front rooms and on Sundays. It would be surprising if the sudden emergence of these images in a wide range of working-class and middle-class publications did not have an impact on men's and women's perceptions of whom a respectable English family contained and how it behaved.

This model of the private family that depended on the company of only a couple of children was also publicised unintentionally as modern and advantageous through the reporting of discussions of ‘national degeneration’ from the late 1890s onwards. Commentators warned that while the degenerate poor continued to bear many children, wealthier parents restricted their family size as they became ‘selfishly’ seduced by amusements. One journalist summarised the causes of fertility decline in the mid-priced Illustrated London News as follows:

if… people give their entertainments at restaurants instead of at home, and if much of life's leisure is spent in frivolities which make the home a desolate and deserted place, we cannot feel surprised that the family circle of old has come to represent an almost extinct idea. (Wilson, 1904)

Amongst newspaper commentators there was little consensus as to whether male or female choices were most responsible for this perceived transformation through fertility of familial lifestyle. Whereas male choices were typically described in approving terms as those of the ‘longheaded, far-sighted rising man’ (Our declining birth-rate, 1910), women who did not prioritise maternity were more pejoratively imagined as ‘too advanced for that, too educated, too cultured, too fond of talking about the equality of the sexes’ (Harris, 1899).

Importantly, clergymen from all denominations were the leading voices in expounding these warnings that the national press publicised further. These pious men, with the authority to speak on matters of private conscience, legitimised the discussion of the link between reproduction and family life, by making couples’ reproductive behaviour a matter of national significance. Although ministers spoke to warn against pursuing this path of ‘selfishness', their speeches simultaneously presented the small family as modern, educated, elite and beneficial to one's personal success.[2] All of the identifiable married eminent clergymen who spoke in this way had large families. The positive tone of their condemnation was perhaps due to the average family size of ministers being amongst the smallest by 1911, presumably led by their junior colleagues (Szreter, 1996, pp. 312–313). In making intimate life the subject of unprecedented public discussion, these arbiters of morality raised popular awareness – and arguably acceptance – of the principle of restricted parental fertility.

It was in the London-based press and by newspaper correspondents who identified themselves as dwelling in the capital's wealthier suburbs that ‘race suicide’ was discussed most vigorously in the decade before 1914. ‘Degeneration’ was only haphazardly reported in the provincial local newspapers, which were read by a higher proportion of working-class households (McKibbin, 1998, pp. 503–506). This was especially true in coal-mining Auckland where one tiny re-published article mentioned these debates in the sampled newspapers. Journalists presumably thought these heated ‘national’ discussions were of little interest in a district where large families continued to be normal.

However, in wealthy southern districts the belief that family size was inversely related to social status influenced the language and aspirations of at least some non-elite adults. One unusually explicit example of this is a letter sent by ‘A mother of three’ to a newspaper in suburban Kent in 1909, complaining that:

I have lately applied to five different landlords for a cottage, and each have refused me on account of my children. Will either of them tell me what to do with them, as the law will not let me kill them, and the Union [workhouse] is full…. Because I've three children people say the birth rate is dying. As far as I am concerned it might never have lived, since small fingers tear off the wall paper that is never put on, and so prevent us getting a cottage. ('A mother of three', 1909)

The polite national discussion of the threat of the ‘dying’ birth rate had publicised a new vocabulary on which poor parents, as well as middle-class fathers such as the publisher discussed above, drew. The mother sought accommodation for a rent of eight shillings out of her meagre weekly wage of 18 shillings. The working class in this prosperous locality were vulnerable to the moral judgments and actions of their wealthy neighbours who wielded authority as landlords, employers and philanthropists, to an extent that was unthinkable in the working-class-dominated industrial districts. The mother's family size of three children was below average for rural labourers, but it is unclear whether she agreed with those who categorised her fertility as the result of deliberate control. The model of the compacted domesticated family increasingly resonated with men and women from across the socio-economic hierarchy. However, fertility limitation within marriage to meet these norms was imposed through external moral and material pressure as well as selected through aspirational choice.

The unbounded fertile family ideal

In examining patterns of fertility change, it is also essential to explain why many men and women did not reduce their fertility. Alongside the emergence of these new models of disciplined civic and of domesticated private family life, other parents proudly celebrated their capacity to father or bear many children up to 1914. These ideals of unrestrained fertility were notably potent amongst fathers living in districts whose labour market was dominated by highly-paid male labour and amongst ethnically Irish Catholic mothers.

In coal-mining districts maternal paid work and paternal domestic labour were negligible, gendered roles were sharply defined, and men and women formed themselves into distinct social worlds (Church, 1986, pp. 611–637). In Auckland in County Durham women's fertility rates were about 50% higher than average for England and Wales. Although women married slightly later in the late nineteenth century, marital fertility increased, reaching a new recorded high in 1908 ('Bishop Auckland Health', 1909). Virility was greatly valued by fathers, but women often expressed less pride in their large families. Harold Heslop, born in County Durham in 1898, later described women's attempts to increase the number of months between pregnancies by breast-feeding until their children reached three years. In spite of his mother's efforts, Harold recalled that ‘our family was increased to five boys. Grandmother Whitfield was not highly excited about the fecundity of her daughter, but, being a wise woman, she did not expostulate in the presence of my father’ (Heslop, 1994, p. 82). Isabella Heslop eventually bore eight children in 18 years of marriage to William Heslop, a Primitive Methodist pit foreman. She died in childbirth aged 39. Although only their son's account exists, the unchallengeable patriarchal rights of men in these coal-mining districts gave women few effective means of resisting normative expectations that men should father large families. Sustained gendered interests were expressed through unspoken and accepted campaigns in opposite directions, rather than through explicit conflict over the frequency of child-birth.

While it was more common for married women to focus on mothers’ struggles, some expressed pride in bearing and rearing many children. These attitudes were understood to be particularly central to Irish Catholic identities. Such attitudes to family size can be glimpsed from insults used by ethnically Irish women in disputes in both industrial localities. In 1861 more than 10% of fathers and mothers in coal-mining Auckland had been born in Ireland, compared to almost 6% in the textiles town of Burnley and under 3% in suburban Bromley.

One typical incident occurred in Burnley in 1910 when a fight broke out between 37-year-old Annie Westwell and 32-year-old Ada Naughton, allegedly watched by ‘500–600 people'. Naughton was reported to have insulted Westwell, a waste-sorter who had been married to a coal-hewer for 13 years, by saying ‘“I have nine kids … and they are all Irish fox terriers. She has no kids”.’ (‘Exciting street scene watched by 600 people’, 1910). Both Ada Naughton and her husband, a labourer, were born in Lancashire, but she appears to have understood her large family as an essential part of an ethnically Irish identity, such that her tenacious children were a point of pride with which she could insult her older childless adversary. Interestingly, in the census of the following year, Ada initially recorded that she had only borne five children during 13 years of marriage, all of whom she claimed remained alive. This was subsequently corrected, apparently by the enumerator, to a record of only four living and five dead children out of the nine children born (Census, 1911). Comparison with civil registration records suggests that one baby had recently died, but in the heated street-scene in 1910 Ada enumerated at least four already-dead children amongst the ‘nine kids’ about whose births she expressed angry pride. High female fertility – understood in terms of the number of children born, not the number raised to adulthood – was drawn upon by mothers as part of a proud ideal of culturally Irish Catholic family life. Contingent parental experiences, such as bereavement, natural infertility, and gendered marital authority shaped the relish with which such ideas were used.

It is likely that at the same time as new ideals of both disciplined and domestic small families became more prominent, those who bore large families felt the need to defend more vociferously the children with which they were surrounded. Far from acknowledging or adopting the labels of moral weakness or class deterioration, these adults constructed their family size as a marker of a proud sectarian identity. While published and official agencies communicated models of family life more rapidly and nationally, the oral communication of alternative norms carried a potency and immediacy that these could lack. Models of family were relational and in explaining why some men and women reduced their family size, it is essential to also understand the ideals of unbounded fertility from which they sought to differentiate themselves.

Conclusions

Men's and women's understandings of child-bearing, child-rearing and childlessness help us to explain why some adults reduced their fertility in late nineteenth and early twentieth-century England. This evidence for shifting meanings is inevitably patchy and riddled with silences. Individual-level longitudinal data would be invaluable in enabling the correlation between experiences, ideas and births to be examined across fathers’ and mothers’ reproductive careers.

This study does not support the theory that adults adjusted their fertility in response to a universal transformation of children from economically-valued assets to emotionally-valued burdens. First, emotional, social, organisational and financial bonds between parents and children were inseparable. The imposition of a language of rational calculation does not further understandings of reproductive behaviour. Second, the extent to which parents perceived their children to be increasingly burdensome has been overestimated. Some groups felt their responsibilities were rising – most notably elite men in relation to their older children from the 1850s and ‘respectable’ non-elite parents, especially mothers, in relation to their younger children from the 1880s. Other parents showed no signs of believing that children were increasingly impoverishing or exhausting. This continuity in perceptions over time and between generations helps to explain why two out of the three studied localities and most occupations did not alter their position in a ranking of average fertility rates (Woods, 2000, pp. 119–122). Third, these understandings of duty were neither simply diffused from elite to non-elite parents, nor legislatively imposed by the national state. Instead, men and women drew on a variety of often local and contingent experiences in concluding that they had an increased, shared capacity to improve the lives of their children. For example, while middle-class couples might appear superficially to have adopted a single ‘pioneering’ norm of controlling fertility, wealthy professional parents and lower-middle-class shopkeepers were motivated by contrasting perceptions of how their responsibilities towards their children were changing. The motivations to alter reproductive behaviour were locally and patchily expressed, so it is not surprising that fertility declines occurred with different timings and rates in the three studied localities.

Men's and women's motivations for controlling fertility as a result of their experiences of parenthood were fundamentally different. The size of a family was most likely to be limited when husbands as well as wives considered not bearing a child every two years during their fertile lives to be advantageous. From the studied evidence, some sort of marital agreement emerged in three contexts, which suggest interesting parallels with demographic evidence from other national contexts. First, men's and women's roles were experienced most similarly in areas with high rates of valued female employment and male unpaid domestic work, such as in the northern textiles districts. Notions of equal civic citizenship were also strong, further strengthening the ideal of mutually investing in children as a means of improving society. Second, in suburban London, ‘respectable’ lower-middle-class men and women agreed upon complementary gendered roles. Diligent male breadwinners supported increasingly feminised and intensive childcare, especially for ‘delicate’ children, alongside higher standards of domesticity and schooling. This was understood to be the pathway through which parents offered their children explicitly ‘modern’ advantages. Third, upper-middle-class and upper-class fathers expressed concerns about their long-lasting duties to support older sons and daughters from the 1850s. Yet it was only from the 1870s, once wives expressed aspirations to take on roles that were less maternity-focused, that couples significantly reduced their fertility after marriage. Each of these strategic and often unspoken areas of agreement between husbands and wives depended on women expressing their aspirations and men valuing their wives’ contributions. Relative equality in class and gendered terms made it more likely that men and women successfully restricted their family size. Marital conflict was seldom mentioned in the surviving sources, but it must not be assumed that it did not form part of these relationships.

These perceptions that the tasks of parenthood were newly significant did not automatically lead to changes in reproductive behaviour. As Woods and Caldwell hypothesised, the most rapid and profound transformations in fertility occurred when cultural understandings of family life emerged that enabled couples to think – for the first time – about the limitation of births in contexts that were understood to be moral (Caldwell, 1999; Woods, 1987). These malleable models gained particular power in specific social and geographical contexts. It is suggested that the use of the ideal of the disciplined, pure family in textile towns of northern England from the 1870s and of the domestic leisured family in suburban districts surrounding London from the late 1890s contributed to the rapid local falls in average fertility in the following decades. Moral spaces and incontrovertibly respectable people – clergymen, philanthropic women, civic meetings, local government initiatives, and their reporting in the press – were essential in publicising ideals of small families. Oral history testimonies have demonstrated that birth control was understood to be a masculine responsibility, defined ‘as a sexual matter rather than a domestic one’ (Fisher, 2006, p. 211). Yet while methods for limiting fertility were conceptually detached from women's child-rearing responsibilities, it was through the rhetoric of family that both men and women began to respectably express the appeal of fewer children. Prior to 1914, it seems likely that these moral contexts were essential in making birth control methods, which were understood to be shamefully obscene, sufficiently thinkable to be widely practised by men and women who thought of themselves as ‘respectable'.

Children had always made demands on their parents. Attention to adults’ perceptions has revealed that some fathers and mothers understood their parental duties to be increasingly interconnected and significant. They believed the ways in which their sons and daughters were raised had far-reaching positive implications for their own lives, their children's life-chances or societal well-being. Simultaneously, parents from other local, ethnic and socio-economic backgrounds, especially in the studied coal-mining district, did not perceive their duties to be newly challenging or actively celebrated a model of highly fertile parenthood. Chronological specificity is essential in explaining motivations. Utterly different expectations of parenthood were powerful in the 1850s, when only elite men sought to limit their duties by postponing marriage, compared to the 1910s when reproductive control within marriage was widely legitimised through moral familial ideals. Parents who reduced their family size later did not simply follow the same process with the same motivations as their predecessors. The shifting interpretations that fathers and mothers placed on their relationships with their children thus help to make sense of the partial, locally-diverse and class-specific pathways by which some women and men were motivated to transform their reproductive behaviour prior to 1914.