Life and death in Victorian numbers

Twenty years ago, Tony Wrigley and Roger Schofield resolved the conundrum of the cause of the growth of population in 18th-century England. The answer, it seemed, was a rise in the rate and a drop in the age of marriage. But in the 19th century, marriage ceased to be the major influence on population growth, and fertility was more likely to be controlled within marriage. At this point, a new conundrum appeared: why did couples decide to control their fertility, by what means, and who made the decision? And why did mortality start to drop and life expectancy to rise? Was it improvement in diet and real wages, investment in public health in cities, or autonomous changes in disease?

Robert Woods raises great expectations that these conundrums will be resolved, for he has written a stream of major articles and he has unrivalled technical expertise. Yet by the end of his impressively researched book, confusion is not so much dispelled as replaced by an informed, honest sense of complexity and uncertainty.

How is family limitation to be explained? Recent debate has focused on the importance of "stopping" versus "spacing". Couples might have children until they reached the desired size of family, and then stop so that the last child was born well before the end of the woman's fertile period. Or couples might space children at greater intervals over the fertile period. These different approaches had implications for the wife's experience of childcare and life chances; and they implied different methods of limiting births.

Recent work by Simon Szreter has stressed the role of spacing, which he explains by changing definitions of male sexuality. Spacing relied not on "artificial" or appliance methods but on male self-control or frigidity. And the emergence of smaller families is to be understood through the establishment of norms in local communities.

Woods agrees with this interpretation to some extent: he accepts that appliance methods were not crucial and admits the significance of spacing. He places the emphasis on a rapid change in public opinion and the emergence of a "new calculus of conscious choice".

Here his explanation differs: the sudden shift was caused by a revolution in mass education and the creation of a literate society, which narrowed the educational divide between men and women. This led to a form of domestic informal feminism aided by secularisation, which weakened the religious connotations of reproduction.

But how convincing is the rise of schooling as an explanation - what was the social mechanism by which more formal education led to fewer births? And how important were the changes in male attitudes to sexuality that have been considered in recent pioneering works by Michael Mason and John Tosh? The puzzle remains, as Woods admits in a section on "Why there are still no firm conclusions".

How is the fall in mortality and rise in life expectation to be explained? Life expectancy varied between districts, and especially between town and country, and remained variable throughout the period. In his earlier work, Woods argued that a movement of population from country to town would be enough in itself to slow the rise in mortality - but he assumed that mortality rates as a whole did continue to fall over the century, and that urban mortality did not itself get worse. More recently, Szreter and his collaborators have claimed that life expectancy at birth did deteriorate in towns in the second quarter of the 19th century and was reversed from the 1870s and 1880s only by large-scale investment in the urban infrastructure and public health.

Woods, not entirely convinced, leaves the reader confused. He comments that there is now more reason to think that urban mortality increased on the lines proposed by Szreter, but he then says that the deterioration was less than Szreter argues and was in any case due more to "particular circumstances" than to "general administrative deterioration". Unfortunately, he does not say why he thinks the rise in mortality was less significant and how much less significant, or why there was not a general problem of urban administration. Szreter's argument is that the improvement in mortality rates came with a removal of constraints on public investment for political reasons about 1870. Recent work by economic historians confirms the low level of investment at the start of the Victorian period and its rise at the end. Woods argues that although investment in public health might explain a decline in mortality from water-borne diseases, it cannot explain changes in mortality from scarlet fever or pulmonary tuberculosis, where he argues for independent changes in the disease. Not everyone will agree, and might point to the role of improved housing or control of infectious disease. Again, Woods's mass of data leaves us in a state of well-informed uncertainty.

Woods admits that demographers are more like hedgehogs than foxes: they know one thing well, and relate everything to that dominant thing. But his book suggests that foxes, who know a lot of things in a diffuse way, should now move into the territory. The great value of Woods's book is that the foxes know exactly where the conundrums remain, and where they should devote their energies.

Martin Daunton is professor of economic history, University of Cambridge.