course to take. There are no grounds for supposing that they experienced the rising costs of prosperity, in terms of home comforts and luxuries, any earlier or more insistently than any other middle-class groups with similar income levels. It is very possible, on the other hand, that the professional class were more affected, and affected earlier, than other groups by the rising cost of educating their children. Not only were professional men likely to set a high value on a good education, but also if they were keen on their sons following in their footsteps, as so many in fact did, they were excellently placed to know what kind of education or training was currently appropriate for entering their own profession. The decline of ‘Old Corruption’ and its final dismemberment in the 1830s and 1840s removed the traditional route of patronage and favour as the way forward in the professions – more obviously in some, like the Church, the law, and perhaps medicine, than in others such as the army or the civil service – and more formal and more prolonged education began to be substituted as the avenue of entry. In some cases that meant staying longer at more expensive schools, or going on to university, in others it meant taking articles which had to be paid for and which yielded little or no income for the trainee; in either case the cost of launching a son into an acceptable career had increased. It is, moreover, apparent that businessmen who contemplated having their sons follow them in the family firm did not favour such prolonged or expensive schooling; they did not become patrons of the public schools much before the 1890s, and their sons generally went into business at the age of sixteen or seventeen to learn on the job. The nine ancient public schools, those investigated by the Clarendon Commission in 1864, probably remained very much the preserve of the aristocracy and gentry and were not much penetrated by the sons of professional men. But the great crop of new public schools founded or radically reconstructed from grammar school origins between 1840 and 1870, for example Cheltenham, Clifton, Marlborough, or Radley, of which there were at least thirty altogether, catered primarily for the sons of the professional class. The total cost of educating a son and fitting him to enter a profession was of the order of £1500 to £2000, spread over perhaps ten years; with an outlay of that size in prospect it would be most understandable if parents developed a desire to restrict the number of their children to that which they could afford to educate adequately.
It is true that secondary education for girls, a truly revolutionary departure, made its appearance only a little later, with the North London Collegiate (day school) and Cheltenham Ladies’ College (boarding school) in the early 1850s, both of which had the daughters of professional families much in mind. The growth of girls’ public schools thereafter, however, was gradual rather than spectacular; Woodard, for instance, began his scheme for studding the country with strict Anglican public schools for boys with Lancing in the late 1840s, but did not move into public schools for girls until the 1860s. It was not until the 1880s and 1890s that there was anything like a rush of foundations of girls’ high schools and boarding schools, and until then it is a fair presumption that expensive secondary education for their daughters remained a decidedly secondary consideration with even upper-middle-class parents. Education-based family limitation, therefore, at least in its initial forty to fifty years, was most likely aimed at limiting the number of sons whose future prospects it was desired to maximize, and one would expect professional-class parents to go on having children until a desired number of sons had appeared, a strategy which could well have led to wide variations in the size of families between couples who shared the same family-limitation objectives. This could also help to explain why family sizes, which still remained at over four children for the pioneering professional groups until the 1880s, plunged down to a little over two by 1911, as more equal treatment for the two sexes became more widespread, educationally.
There is little in all this to suggest that the upper middle class was following, after a time lag, the aristocratic example. It is true that some of these groups, especially the army, Church, and law, lived on the fringes of aristocratic society and were prone to seek to emulate aristocratic lifestyles in such matters as servant-keeping and domestic apparatus, as closely as possible. But since the expense of such things does not appear to have been the dominant motive in professional-class family limitation, it cannot have acted as the link in any social diffusion chain. Moreover the educational factor was distinctive to the professional classes, and was not strongly felt by the aristocracy who had long been accustomed to spending rather heavily on the education of their sons and to a lesser degree of their daughters, at least since the early eighteenth century and probably since the sixteenth. As to aristocratic behaviour rendering family limitation respectable, so that the upper middle class then felt free to adopt it themselves without risk of censure, not only is there no evidence that the subject was referred to or commented on by respectable people, but also a section at least of the upper middle class would not have accepted aristocratic behaviour as any guide to respectability. It is more sensible to conclude that the upper middle class had their own good reasons for embarking on restriction of family size, were responding independently to their own circumstances and needs, and were not emulating the class above them.
The rest of the middle classes started to restrict their family sizes from the 1880s, and it is not altogether clear whether they were copying the earlier examples of their social equals, responding to the general movement in educated opinion, or adapting to the arrival in their particular groups of circumstances similar to those which had begun to affect groups in the professions thirty or more years before. Possibly a mixture of the three, with most influence coming from the third factor. Manufacturers and industrialists, as noted earlier, were impossible to identify separately in the census returns; but a family survey carried out privately by Charles Ansell in 1874, a postal questionnaire answered by over 25,000 clergy, lawyers, doctors, merchants, bankers, manufacturers, and peers, showed that the merchants, bankers, and manufacturers had larger completed families than the other groups. Insofar as the merchants and bankers had fallen into line with the professional families by the 1890s, it is not unlikely that the manufacturers had done so as well. It is perhaps excessively determinist to suggest that civil servants, whose families declined sharply in size with marriages made after 1871, responded instantly to the opening of the civil service to competitive examinations in 1870, since that would assume that civil servants hoped to breed future civil servants and began at once to tailor the number of their children to the more elaborate and advanced education which had implicitly been made necessary. Nevertheless, education-triggered family limitation applies as well to them as it does to manufacturers, whose ambitions for gentility for their offspring and more prolonged and modern education were supplying more public schoolboys by the 1890s. The very rapidly growing core of the lower middle class, the commercial clerks, were already restricting their families in all post-1861 marriages and from 1871 onwards their families were only fractionally larger than the upper-middle-class mean; yet these were miles away from being customers of the public schools. The clerks formed one of the fastest growing occupational groups of the second half of the nineteenth century, male clerks quadrupling in number between 1861 and 1891, from 91,000 to 370,000, and female clerks making an appearance from 1881 onwards. This expansion was sustained only by massive recruitment from other, largely manual working-class, groups, and the essential qualifications for entry were literacy, good handwriting, accuracy, and respectability. Recruitment was, therefore, likely to attract the working-class children from the top of the class who had performed best and benefited most from their schooling, and were likely to put a high value on getting as good a schooling as possible for their own children in turn. Moreover, clerks distanced themselves from those they regarded as their social inferiors less by any income differences, since they earned scarcely as much as many skilled workers, than by careful cultivation of status differences expressive of moral superiority, which in their own estimation included shunning public elementary or Board schools for their children and sending them to local private, fee-paying, schools. Here, at a lower level of incomes and standards, was the same category of motives for having fewer children.
The clerks could well have added that it was important to their social self-esteem to make it clear that they were different from the working classes who bred like rabbits, and that far from there being any question of emulating their superiors they were engaged in underlining contrasts with their inferiors. The working classes, however, were far from homogeneous in their family habits; they were not all big breeders; and it is more than doubtful if they were in the business of emulating the middle classes themselves. Although unreflecting and vulgar Malthusians had spoken as though the lower orders were an