Home

Login

Contact us

Cette page existe uniquement en anglais

Home > Activities > Committees >Exploratory Mission Report

Report from the Exploratory Mission on Historical Demography

While historical demography has changed considerably from its origins fifty years ago, by many measures, the field is thriving. By some measures it is booming. Whereas early papers came exclusively from Europe and North America, recent work comes from virtually the entire globe. Whereas early research was largely from demography and history, with the occasional contribution from economics and geography, recent research has embraced such diverse fields as anthropology, cultural studies, education, epidemiology, genetics, medicine, political science, and public health. Whereas early work focused largely on the measurement of aggregate demographic indices, current work has shifted more to individual-level explanation of demographic behavior.

These successes are partly a consequence of sustained efforts by successive IUSSP historical demography committees during the last twenty years to develop the field and promote individual level methods not only throughout Europe and North America but also Latin America and East Asia¹. In a series of conferences, the IUSSP organized research agendas around areas of common intellectual concern, such as the mortality transition and the Malthusian Paradigm, as well as areas of shared geographic interest, such as Asia, to produce such recent volumes as the Decline of Mortality in Europe (1991), Old and New Methods in Historical Demography (1993), From Hunger to Modern Economic Growth (2000), Asian Population History (2001), and the IUSSP Population Atlas of the Second Millennium (in preparation) and World Population History of the Second Millennium (in preparation).

At the same time, the vitality of the field is also a result of the importance of longitudinal individual level demographic data for a variety of fields that have been outside the purview of previous IUSSP committees. Longitudinal panel data and historical individual level data have been central to research efforts in such fields as genetics, medicine, and PUBLIC health where they have developed largely independently from historical demography. These efforts using the utah genealogical data base have produced such recent spectacular successes as the identification of Breast Cancer One and Breast Cancer Two². Ongoing efforts to reconstruct the genealogical history of Iceland promise similar success in the near future as do other research efforts on, for example, the effects of early life conditions on later mortality. It is our hope that future iterations of the IUSSP Committee on Historical Demography will embrace these and other research agendas just as they promoted research on the mortality decline and the Malthusian Paradigm.

This is important for many reasons. Whereas early work shared a common intellectual and geographic agenda with common methods, current work is highly multi disciplinary. It is also depending on the field of research, more quantitative, more biological, more technical, more applied, and in such fields as anthropology more self-reflexive and self-consciously qualitative. On one hand, the development of multiple methods of multi-variate analysis in economics, political science, sociology, and the medical sciences have created a mathematical divide between the more scientifically literate and illiterate. On the other hand, the rise of an increasingly opaque post-modern critique have also created a similar divide among the culturally literate and illiterate. The ironic consequence is that in spite of the large numbers of papers and researchers using historical demographic data, research in historical demography is increasingly isolated and segmented as many 'historical demographers' no longer share either a common discipline nor a common problematic. As a result, the visibility of the field does not reflect the significant vitality of researchers using historical demographic data.

IUSSP would benefit the field of historical demography by creating a broad agenda and venue that would cross disciplines as well as political boundaries. We need intellectual integration as well as coordination that will allow researchers with shared agendas from different fields and different geographic areas to meet and discuss common problems and concerns. Such meetings should produce important methodological and intellectual benefits. At the moment, while there are several shared biological and demographic agendas, there are no commonly agreed upon methods and therefore no common measures. Moreover, much historical work has been confined within national boundaries. The result, as was apparent at the recent IUSSP Conference on the Population History of the Second Millennium, has been a dearth of truly comparative work and little broad synthesis of human demographic experience at the global level. As the only truly global academic organization in demography IUSSP is uniquely positioned to lead our field to a higher as well as broader level of intellectual concerns. This is especially true in the sub-field of our specific exploratory mission where only the union embraces all the academic disciplines that together comprise today's historical demography.

The creation of a common shared or at least understood agenda in historical demography should also facilitate the shared use of currently available data. There are already numerous data sets of historical individual level records, many of which are machine readable, which are under-used and often improperly organized, due to limitations in quantitative skills as well as data processing. At the same time, there is also a great increase in public use machine individual level censal data with the development of the large fixed one percent sample data sets available for the United States from IPUMS and increasingly for other countries as well. IUSSP could play a major role in facilitating the use of such data and the council should consider the creation of a subcommittee on historical demographic data to acquaint the field to these opportunities and to coordinate access to them.

We have identified five broad topics in historical demography that would benefit particularly from IUSSP developmental efforts. We outline those topics that fall within the purview of the expertise of the committee below. Several of the topics, like the ones on longevity and kinship, overlap with the work of other committees, which shows the relevance of historical demography for other areas of our discipline and may require collaboration between the historical demography committee and other committees to organize conferences. All topics are multi-disciplinary and benefit from a global approach, which only IUSSP could provide.

Proposals for Exploratory Conferences and Workshops in Historical Demography:

1. Kinship: Longitudinal Perspectives and Prospects, 2002-2003.
2. Longevity: Childhood Conditions, Social Mobility and Other Factors that Influence Survival to Old Age, 2003-2004.
3. Temporality: Integration of Short- and Long-term Economic-Demographic Models, 2004-2005.
4. Re-Categorization: Deconstruction-Reconstruction of Intellectual and Technical Tools of the Historical Demography, 2005.
5. Spatial and Longitudinal Comparison, 2006.

Assuming that the IUSSP Council is interested in this intellectual agenda, we are willing to facilitate such activities and suggest two other scholars, who supplement our expertise, and who could make important contributions to the organization and funding of these conferences and workshops. We detail the agenda below.

1. Kinship: Longitudinal Perspectives and Prospects

The historical demographic study of kinship stands at the confluence of two longstanding streams of academic research and one new current with the potential to overtake them both in terms of immediate intellectual promise. On the one hand, anthropologists have had a long interest in delineating kinship systems and kinship ties (Feinberg and Ottenheimer 2001, Schneider 1984). On the other hand, historical demographers, inspired by anthropologists, have devoted considerable resources to delineating kinship networks in historical societies where kinship is thought to have been a major organizing principle of social groups (Levi 1985, Plakans 1984, Sabean 1984). More recently, research pioneered by Mark Skolnik and others has shown that when such longitudinal genealogical data are extended to the present-day and matched to current medical records, they can produce family histories of disease with important implications about the heritability of morbidity, mortality, and for that matter other forms of social behavior (Davies and White 1995).

The biological sciences, in other words, as in the study of longevity have shown important new potentials for historical demographic research on kinship. Moreover, such research can be accomplished without contemporary links and without morbidity data. Alan Bittles (1991) and other studies) and his colleagues have shown, for example, that historical information on consanguinity over just two generations - common grandfather or great grandfather - can have significant associations with the pattern of infant mortality and perhaps even fertility. Other work by Geraldine Mineau, Joanna Mountain, Gabriele Doblhammer and many others have similarly shown other potentials for biological analysis of longitudinal historical demographic individual level data.

Analysis of the longitudinal reconstruction of kinship over far more generations will allow us to refine such patterns in some detail. In theory, there is already a mass of identified historical data, much of which has been processed, that lends itself easily to such analysis. Family reconstitution from parish register data stretching over generations makes it possible to identify kinship ties, measure their characteristics, and examine how they influence individual outcomes (Macfarlane 1986). So does the longitudinal study of individual level data from household registers, which is now common in historical demography (Bengtsson and Campbell, forthcoming). The essential problem has been with the constraints of traditional disciplinary perspectives. While some studies have made use of sources such as genealogies and court records to study kinship networks in the past, the emphasis has often been on describing how networks formed and then persisted over generations rather than on measuring their actual importance (Emler, Kassler, Levi, Mitterauer, Sabean and others). Recent work, however, have shown an interest in transcending such boundaries and building links to the social sciences and biological sciences (Cameron Campbell and James Lee, Catherine Capron and Michel Oris, Theo Engelen and Jan Kok, Andrejs Plakans and Charles Wetherall, Paul-Andrè Rosental, Noriko Tsuya, and Giles Postel-Vinay).

A conference and workshop organized by the IUSSP would allow these historical scholars whose disciplinary affiliations span anthropology, biology, economics, genetics, history, medicine, public health, and sociology and whose research range from Belgium, Canada, China, France, Germany, Italy, Japan, Switzerland, and the United States to discuss common problems, to develop common methods, and to determine common problematics. At least one-third of the invitees should be from the biological sciences, another quarter from the more traditional historical social sciences, and the remaining half from our core constituency, the demographic historical social sciences.

2. Longevity: Childhood Conditions, Social Mobility and Other Factors that Influence Survival to Old Age

The importance of conditions in early life for health and mortality in later life is well-known from the existing literature on developing countries (Schrimshaw 1995) but also from studies in historical demography. Epidemiologists and demographers studying the mortality decline in the 1920s and 1930s were well aware of the connection between early-life conditions and mortality later in life (e.g. Derrick 1927; Kermack et al. 1934). Analyzing aggregated data they noticed that mortality for infants and children went down much earlier than for adults. Each generation seems to experience the same relative mortality from childhood to old age (Kermack et al. 1934:699). The importance of cohort factors for the analysis of the mortality decline has also been emphasized by Preston and van de Walle (1978) for urban France and by Fridlizius (1989) for Sweden. Kannisto and Wilmoth who recently have re-evaluated the evidence are, however, questioning earlier conclusion. This implies that those factors that vary over the life course, such as income and improvements in public health and medicine, etc. are more important than conditions in childhood. Most historical studies of cohort mortality are done at a high level of aggregation, like the ones discussed above. Today, however, several historical cohort studies are done at individual level and the area is expanded rapidly.

Within medical research the work by Barker has been of major importance. In his 1994 book Mothers, Babies and Disease in Later Life, he summarizes the medical evidence regarding the connection between nutrition during infancy as well as during the fetal stage and adult health outcomes. It is not only chronic malnutrition during early life that can be of importance for health later in life. Temporary disturbances in nutrition, in particular during some periods in the fetal stage, may also have long-term effects on health, since later improvements cannot always compensate for prior loss (Barker 1994). The relevance of Barker's findings has, however, been much debated within medicine during the last years as this field of research has turned into a big industry.

Fogel (1993) has stressed the importance of early-life conditions on adult mortality in historical research. He has been using final heights as a measure of net nutrition and health during infancy and early childhood³. If individuals are well nourished and healthy during childhood their cells and organs develop better and they reach higher heights and live longer. It is net nutrition (intake minus claims), rather than gross nutrition that is of importance. Thus, improvement in health and height may either be the result of better nutrition (better diet) or less claims due to e.g. lower prevalence of diseases. The problem is that, looking only at heights, it is difficult to separate the nutrition effect from the disease effect. It is therefore important to include indicators of conditions early in life into the analysis, not only heights and longevity. Bengtsson and Lindström (2000) have this way been able to show how the importance of the disease load in the first year of life for mortality in old age using historical records on individuals.

Besides the factors discussed so far, genetic factors most likely have an impact on human longevity, even if other variables are more important in determining the final outcome. Genetic factors, for example, set limits to the final height achieved by an individual, which in turn has been seen as an important factor behind adult mortality risks (e.g. Elo and Preston 1992:194). Other factors that could be important are social mobility (changes in access to individual resources over the life course) and societal interventions, like vaccination campaigns, etc.

The factors influencing adult- and old age mortality could be summarized under four broad headings: genetic factors, early-life conditions, life course transitions (e.g. social mobility) and prevailing living conditions. We believe that only by taking all these variables into consideration simultaneously, can one correctly assess their relative importance on health and mortality. This in turn has strong implications not only for research concerning the importance of early-life conditions on mortality later in life, but also for the long-term mortality development in general. The research questions posed are clearly multi-disciplinary, linking social sciences and history with epidemiology and social medicine. Hence, the workshop organized by the IUSSP will include researchers from demography, economic history, economics, biology, medicine and social medicine working on historical as well as contemporary populations.

3. Temporality: Integration of Short- and Long-Term Economic-Demographic Models

From a large number of aggregated studies on pre-industrial populations, we know that short-term changes in food prices have a strong impact on demography, particularly on fertility but also on mortality and migration⁴. However, the links between economy and demography are not clarified. How different household and family characteristics, such as social status, household composition, and previous fertility history moderates the effects of economic stress are largely unknown. Aggregated studies also fail to explain the causal links, i.e. whether the fertility response to short-term economic stress is due to malnutrition and bad health of parents, to splitting up of couples due to temporary migration, or to birth planning. Similarly, they cannot explain whether the link between food prices and mortality is due to malnutrition or spread of decease as a result of temporary migration. Thus even though the results at aggregated level are very consistent, the links between economic and demographic variables remains unclear.

This has necessitated the construction of a variety of large, individual level longitudinal data sets of specific historical communities with information on individual occupation, household structure, and individual or household status, and parallel serial information on either grain prices or farm wages. It has also necessitated the development of a methodological approach combining individual life histories with community wide economic indicators within a life event history framework (Bengtsson 1993, Lee, R 1993)⁵. This way one can consequently study how economic conditions such as occupation, food prices, and wages and social conditions such as household and kin composition and relationships affect a person's ability to marry, to have children, to migrate, even to live.

It requires the conception of a formal analytic framework to understand population responses to exogenous stress as well as the construction of a number of specific detailed and dynamic models of demographic behavior, a process, which has taken years. One has to refine such well-known sociological concepts as 'the family' to make them usable for longitudinal analysis and to redefine such economic concepts as 'living standards' to fit them into a conceptual framework as one analyze mortality, nuptiality, fertility, and migration behavior. This approach has been developed within the EurAsia Project on Population and Family History but has also been applied lately among other scholars using data sets arrived from family reconstitutions and other ones.

The purpose a workshop organized by IUSSP is to expand in this direction but also to go one step further and integrate the type of micro/macro models for analyzing effects of short-term economic stress on demographic events with models for long-term relationship between economy and population. A first attempt have been done at a workshop in Sweden in 2000, which focused on the relationship between the long-term standard of living, using traditional macro indicators such as average real wages, GDP per capita, heights, etc, and changes in the short-term response to economic stress over time. The approach and the type of models proposed are by no means limited for use on historical populations only but could and have been used also on contemporary ones.

4. Re-Categorization: Deconstruction-Reconstruction of Intellectual and Technical Tools of the Historical Demography

During the fifties and sixties, the pioneers of historical demography sought to measure quantitatively the changes in the levels and characteristics of nuptiality, mortality and fertility. How many babies did a couple have in a "natural fertility regime"? What were the dimensions and structures of household? At what age did people marry? The definition of the notions used was less important than the biases in the measures' methods. The positive approach was a priority and criticisms of sources were restricted to the reliability and completeness of the reported facts.

Probably because of the influence of the development of the new cultural history of sciences, of 'Critical Theory', of the dialogue with anthropologists and philosophers, but also of the political and the ideological stakes of their field, historians of population have begun to focus more critical self-reflexive attention to their field. Some scholars, for example, have focused attention on the political links between the emergence of fertility indices and social Darwinism (Le Bras, 1981), on the consequences for demographic approaches of political and cultural history (Teitelbaum and Winter, 1985), and on the design of censal categories and social demographic survey questionnaires (Tanur 1992).

Generally speaking, several studies have emphasized the fact that the categories of analysis and the categories of data presentation are not 'natural'; that they have been created in a special context (historical, ideological, scientific…), with specific means and for precise uses. These questions have been addressed on the example of the notion of 'demographic aging' by Bourdelais (1997).

Today, it seems very important for the historical demographers to be able to present a critical view of their works, of the notions they deal with, of the categories used and their results. It is necessary from a scientific point of view for the dynamic of the discipline, but also to be able to answer to the post-modernistic objections to our field.

Several directions could be selected:
The analysis of the emergence of notions used in our field, or of demographic indexes: legitimacy, definition of the family, ethnicity, but also demographic aging, demographic dynamism, shapes of age-pyramids naming, uses of the age groups and of the age, measures of migrations. The problems of administrative categories: what are the cases for which it is a problem to use it and why. For example, occupational statutes, nationality, maternal language, and causes of death. Are we able to propose some solution? How to propose re-categorization or new ways of measuring? What change in our results because of this re-categorization? What are the benefits for our knowledge and our comprehension of the demographic evolution? What are the theoretical basis behind our new proposals and techniques? Choices of aggregation level, of multivariate analysis, network analysis, etc.

5. Spatial and Longitudinal Comparison

One of the most important developments in historical demographic information is the completion of 1 percent stratified samples of all the historical censuses of the United States and the initiation of similar projects in Canada and the united Kingdom and elsewhere. These data offer an unprecedented opportunity for detailed reconstruction of spatial demographic patterns in these countries in the past and the comparison of such patterns over time. The main task is to develop models to make and simultaneous use of changes over space and time and to avoid traditional regional boundaries. Certain epidemiological models as well as chaos models might constitute a useful starting point. It might even be worth while to review Torsten Hägerstrand's ideas about time-space models, which were very promising when they came but that never meant the breakthrough many anticipated.

Tommy Bengtsson, Patrice Bourdelais and James Lee

References:

Alter, G. (1988) Family and the Female Life Course. Madison, Wisconsin and London: The university of Wisconsin Press.

Barker, D. J. P. (1994) Mothers, Babies and Disease in Later Life. London: BMJ Publishing Group.

Bengtsson, T. (1993) "Combined Time Series and Life Event Analysis", in Reher, D. & Schofield, R. (eds.), Old and New Methods in Historical Demography. Oxford: university Press Oxford.

Bengtsson, T. and Campbell, C. (eds., forthcoming) Families, the economy, and death in past times: Relationships between economic conditions, household context, and mortality in Europe and Asia before the twentieth century.

Bengtsson, T. and Lindström, M. (2000) "Childhood misery and disease in later life: The effects on mortality in old age by hazards experienced in early life, southern Sweden, 1760-1894", Population Studies, 54, 263-277.

Bengtsson, T. and Reher, D. (1998) "Short and Medium Term Relations Between Population and Economy" in Núñez, C-E, ed, Debates and Controversies in Economic History. Madrid.

Bittles, A. H. (1991) "Consanguinity: a major variable in studies on North African reproductive behavior, morbidity and mortality?" in Demographic and Health Surveys World Conference, August 5-7, 1991, Washington, D.C.: proceedings. Volume 1. 1991. 321-41 pp. Institute for Resource Development/Macro International, Demographic and Health Surveys [DHS]: Columbia, Maryland.

Bourdelais, P. (1997) L'age de la vieillesse. Histoire du vieillissement de la population. Paris: Odile Jacob. Nouvelle version. Paperback edition.

Davies and White (1995). Breakthrough: the Race to find the Breast Cancer Gene. New York: John Wiley and Sons.

Derrick, V. P. A. (1927) "Observation on (1) Error on age on the population statistics of England and Wales and (2) the changes of mortality indicated by the national records", Journal of the Institute of Actuaries, vol. 58.

Elo, I. T. and Preston, S. H. (1992) "Effects of Early-life Conditions on Adult Mortality: A Review." Population Index, Vol. 58:2, pp. 186-212.

Feinberg, R and Ottenheimer, M (eds, 2001) The Cultural Analysis of Kinship: The Legacy of David M. Schneider. university of Illinois Press.

Fogel, R. W. (1993) "New Sources and New Techniques for the Study of Secular Trends in Nutritional Status, Health, Mortality, and the Process of Aging." Historical Methods, 26:1, 5-43.

Fridlizius, G. (1989) "The Deformation of Cohorts. Nineteenth Century Mortality Decline in a Generational Perspective." Scandinavian Economic History Review, Vol. 37:3, pp. 3-17.

Galloway, P. R (1988) "Basic patterns in annual variations in fertility, nuptiality, mortality and prices in pre-industrial Europe", Population studies, 42, 2.

Kermack, W. G., McKendrick. A. G. and McKinlay, P. L. (1934) "Death rates in Great Britain and Sweden: Some regularities and their Significance", The Lancet, March.

Komlos, J. (1993) "The secular trend in the biological standard of living in the unitied Kingdom", Economic History Review, 46, 115-144.

Le Bras, H. (1981) "L'histoire secrète de la fécondité", Le Débat, 8, 76-100.

Lee, R. (1993) "Inverse projection and demographic fluctuations: A critical assessment of new methods" in Reher, D. And Schofield, R. (eds) Old and New Methods in Historical Demography. Oxford: Clarendon Press.

Macfarlane, A. (1986) Marriage and love in England: Modes of reproduction, 1300-1840. Oxford:Basil Blackwell.

Plakans, A. (1984) Kinship in the past an anthropology of European family life, 1500-1900. Oxford: Basil Blackwell.

Preston, S. H. and Van de Walle, E. (1978) "urban French mortality in the nineteenth century", Population Studies, 32, 2.

Sabean, D. W. (1984) Power in the blood popular culture and village discourse in early modern Germany. Cambridge: Cambridge university Press.

Schrimshaw, N. S. (ed.) (1995) Community-Based Longitudinal Nutrition and Health Studies. Classical Examples from Guatemala, Haiti and Mexico. INFDC. Boston.

Schneider, D. M. (1984) A Criticism of Studies of Kinship. Ann Arbor: university of Michigan Press.

Steckel, R. (1995) "Stature and the standard of living", Journal of Economic Literature, 33, 1903-1940.

Tanur, J. M. (ed.) (1992) Questions About Questions: Inquiries into the Cognitive Bases of Surveys. Russel Sage Foundation.

Teitelbaum, M. and Winter, J.M. (1985) The Fear of Population Decline, 1879-1984, Orlando, Florida: Academic Press.

Footnotes:

¹ These efforts were begun under the chairmanship of Roger Schofield who established a Latin American sub-committee and continued under David Reher who established an Asian sub-committee.

²For an account of this process of discovery, see Davis and White (1995). The umeå Demographic Database has similarly been used to detect certain genetic eye diseases.

³Other important work on health and heights are done by by Steckel (1995) and Komlos (1993).

⁴For an overview see Galloway (1988), Lee, R (1993) and Bengtsson and Reher (1998).

⁵Alter (1988) and others has previously explored the use of life event analysis for analyzing event over the individual and family life course.