The 1911 censuses of the British Isles included questions directed at currently married women, relating to the number of children they had borne in that marriage, the number of those children who were still alive and the number who had died. With the help of the demographic techniques of indirect estimation, the answers to such questions can be made to yield estimates of infant and child mortality over the fifteen or twenty years leading up to the census. Although the civil registration system was established and accurate, individual level records if birth, marriage and death are not available for research purposes, but the records of children born and died from the 1911 census can be used instead in analyses of fertility and early-age mortality together with independent variables measured by the census. Mobility in nineteenth-century cities, however, was very high: in Belfast fewer than 20 per cent of household heads in 1911 had been resident in the same house in 1901. Thus the independent variables measured at the time of the census may not have applied a few years previously when the children were at risk of death. This paper uses the 1911 and 1901 censuses of Belfast to explore the child mortality of movers and stayers, and examines the mobility histories of movers to assess the viability of using the characteristics measured in 1911 as proxies for those applicable when the children were at risk of death.

Email: amr1001@cam.ac.uk

Household formation, marriage and procreative careers in late nineteenth century Scotland
Eilidh Garrett, Alice Reid, University of Cambridge

Within the European Marriage Pattern, characterised by the importance of economic independence for newly-wed couples in general, marriage can be seen as a life-course process rather than an event, while entry into headship may be viewed as the key event affecting fertility. Changes in source material over the early modern and modern periods from court documents to registrations of events have created a spurious impression of increasing adherence to the solomnisation of marriage as the start of the union, but the implications of this for later periods when unions are measured exclusively by official registrations of marriage have rarely been explored. In particular the implications for measures of nuptiality, fertility and household formation over the demographic transition have been neglected. This paper will use registrations of marriage, maternal and paternal registrations of birth, subsequent marriage and living arrangements as measured by the census in late nineteenth century Scotland, to examine both procreative careers in the broadest sense, within two different socio-economic settings, and the consequences of such a view for measurements of fertility and household formation.

Email: amr1001@cam.ac.uk

Independence and interdependence: Householdformation patterns in eighteenth century Kythera, Greece
Violetta Hionidou, University of Newcastle

While earlier hypotheses concerning the household formation patterns of Greek populations were proved mistaken, further hypotheses have been proposed. Though the small number of existing studies prevents us from conclusively describe the household formation patterns in nineteenth century Greece, no studies exist referring to eighteenth century. This paper examines the household formation patterns on eighteenth century Kythera using nominal census and notarial sources. It demonstrates that Kythera did not belong either to the West or to the East, as these were described by Hajnal; neither did its household formation system conform to Laslett's Mediterranean tendency. The widespread prevalence of nuclear households in Kythera disguised the strong economic links between the paternal household and those of his sons. Thus, the residential independence, as demonstrated in the quantitative analysis of the census, contrasts the economic inter-dependence between the paternal and his sons' households, as depicted in the qualitative notarial sources.

Email: violetta.hionidou@newcastle.ac.uk

Long-term trends in cause-specific mortality in England and Wales
Mike Murphy & Mariachiara di Cesare, London School of EConomics

Cause-specific cohort mortality life tables are rarely available, in part due to lack of suitable data: cohort life tables ideally should include more than a century of data, and such long runs are rarely available in consistent form since classifications change. Although England and Wales has produced information on cause of death in national vital statistics since the middle of the 19th century by age and se, in earlier years, cause of death data were unavailable or coded inconsistently and/or without reference to specific pathologies. We use recently available annual data from 1911, coded using the ICD revision in force at the particular year (Office for National Statistics 2000; 2009). Rates for the period 1861 to 1961 were computed for decennial years. We use an elaborated version of the Preston, Keyfitz and Schoen (1972) classification taking into account emerging causes of death. We discuss how single year of age, cause-specific rates by sex for each year between 1861 and 2000 for the selected causes may be constructed and present estimates. Period and cohort life tables by cause of death were constructed for men and women in years 1861 to 2000. These values are used to derive summary indices such as numbers of years gained by deleting various causes of death. The values obtained from period and cohort life tables are compared, together with methods for projecting future cohort mortality by use of such models.

Email: M.Murphy@lse.ac.uk

Cohort patterns in historical tuberculosis mortality: a reanalysis
Romola Davenport, University of Cambridge

Background. The decline of tuberculosis mortality in the pre-antibiotic era is one of the most widely touted examples of a cohort effect on mortality trends. However the evidence for cohort effects rests almost entirely on two observations; the regularity of age patterns of mortality by birth cohort, and the progressive shift of the maximum mortality rate to older ages; and these are confined to males. Methods and Results. The datasets used in the original analyses were extended where possible to create continuous annual series of age-specific data, to allow analysis by discrete birth cohorts. Reanalysis indicated: 1. where the initial decline in tuberculosis mortality occurred within the period of observation, onset of decline occurred simultaneously in many age groups, in a pattern indicative of 'period' not cohort-dependent effects 2. there was no evidence of cohort patterns in tuberculosis mortality for any female population studied. Therefore any mechanisms proposed to underlie cohort patterns in male mortality must be sex-specific 3. sex ratios of tuberculosis mortality at older ages peaked in cohorts born around 1900, and resembled cohort sex ratios of lung cancer. Conclusions. These findings suggest that the decline in tuberculosis mortality before 1950 was not dependent on improvements in early life conditions, but was driven by period effects that reduced mortality at all ages. Additional factors, possibly smoking habits, impeded the decline of tuberculosis in adult males, and produced the sex-specific shifts in age distributions of mortality that were previously interpreted as evidence of cohort-dependent mortality decline.

Email: rjd23@cam.ac.uk

The estimation of the migration bias on three different sources: civil registration of births and deaths and listing of inhabitants: the example of the city of Madrid arpund 1905
Diego Ramiro-Farinas, Barbara A. Revuelta-Eugercois, Spanish National Research Council

The urban penalty effect on mortality has been one of the key issues on the History of Population for a long period of time. The analysis until now has focused its attention on the living conditions of migrants and non-migrants in the city, and the effect in which social networks, housing and working conditions could affect their health in the urban setting. Simultaneously, there has been also an important debate about previous health conditions of migrants arriving to the city, and their effect on urban health when or whether their return to rural areas. But within this debate, there is an implicit assumption: registration of events is fully accurate. However, this idea is far from reality on a situation of strong migration movements towards big cities as in the process of the demographic transition. This was particularly important as deaths were not registered according to the person's place of origin but instead were registered according to their place of death. Accordingly, most of the temporary migrants were not registered in the listings of inhabitants and their deaths distorted urban mortality estimates. Therefore, in this paper we will firstly estimate the under-registration of births in the city by comparing birth and death registrations systems for children less than one year of age. Secondly, we will estimate the over-registration of urban mortality by linking birth and deaths registered in the city with the individuals present in the listing of inhabitants. Additionally, we will describe the composition by age, sex and cause of death of urban excess mortality. In order to do that, we will use information from the individual-level dataset of the deaths of the city of Madrid for the period 1880-1930. We will make use of births and deaths of the period 1905 and 1906 and the listing of inhabitants of 1905.

Email: diego.ramiro@cchs.csic.es

Revisiting a delayed and gradual fertility transition: new micro-level evidences for Alghero, Sardinia (1866-1936)
Marco Breschi, Massimo Esposito, Stanislao Mazzoni, Lucia Pozzi, University of Sassari

Unlike other regions, the demographic history of Sardinia is relatively unexplored. Nominative studies are almost absent, especially for the pre-demographic transition period. The most deep analysis on fertility decline is still that offered by Massimo Livi-Bacci within the framework of the Princeton Project, at the macro-level. According to the results of this project, Sardinian has been the Italian region with the slowest and most gradual fertility transition. In the 50s, fertility levels in the island were the highest in the country: TFR was slightly lower than 4 children per woman while, for the country on average, rates ranged around 2.3. In clear contrast with its previous reproductive history, today Sardinia is the Italian region with the lowest fertility level. The objective of this paper is to retrace, by using a large data-set, reconstructed at the individual and nominative level for the community of Alghero, the very original path towards fertility control followed by the Sardinian population. In particular, our analysis will cover also the decades following the 1st World Ward, a central period in the Italian demographic transition up to now never explored at the micro-level.

Email: lpozzi@uniss.it

The meaning of population growth for interwar Poland (1918-1939)
Malgorzata Radmoska, Max Planck Institute for Demographic Research

The paper focuses on discourse analysis of population growth in Poland 1918 – 1939, in the times of the reconstruction of the Polish state, after gaining its independence at the end of the First World War, and being geographically inconveniently situated between Germany and Russia, its previous invaders. It presents the changing meaning of the population growth and the role of the family, evaluating the population policy of the Polish state from the Neo-Malthusianism slogans in he 1920s till the late 1930s, where demographic growth is seen as an asset and serves as a political tool in bilateral and international relations. The Polish forerunners of sexual education and demographic concepts are evaluated, to which not enough attention has been paid in contemporary studies on Polish demographic thought. This qualitative study is based on Polish archival sources as well as demographic printed sources of that time.

Email: radomska@demogr.mpg.de

Population problems? Social policy and low fertility in post-war France and Great Britain
Emily Ann Marshall, Princeton University

This paper examines the historical context of current concerns about below-replacement European fertility and the role of social policy in addressing low fertility. Case studies of post-war Great Britain and France are used to examine state actors' perceptions about low fertility in the immediate post-war era, shedding light on the ways that public debates and policy regimes have evolved differently during the post-war twentieth century. As part of a larger project examining post-war differences in population policy in these two countries, this historical analysis relates post-war institutions and discourse to the often-neglected history of widespread concern over low fertility in Western Europe during the period between the World Wars. Using archival materials and mass-media coverage of the state-appointed British Royal Commission on Population and the French High Advisory Committee for Population and the Family in the 1940s, I examine how each organization understood the causes of low fertility, how each justified state involvement, and what kinds of measures each proposed as remedies. I argue that the commissions serve both as reflections of inter-war concerns about low fertility, as well as sites of transition into the post-war policy context. I find significant substantive disagreement over state limits on contraception, motivations for fertility behavior, and within-country differential fertility. Furthermore, the approaches of two organizations to the process of inquiry, as well as the types of evidence that were privileged in their deliberations, were quite dissimilar. These differences indicate possible sources of later differences in the social policies of the two countries.

Email: emarshal@princeton.edu