Projects - Nature or Nurture? (VIDI) - Extended description

Research questions and theoretical framing

One of the major challenges for Western society today is to deal with the demographic balance between young and old. The increasingly aging population creates a number of problems which need to be dealt with urgently: how long can elderly work? And to what extent can they keep taking care of themselves? Can the current pension-model be sustained? To what extent can elderly rely on market and state for support? Are there other possible support models?

To arrive at a new balance, it is imperative to understand the underlying mechanisms that led to today’s situation. The proposed research looks at this ‘elderly puzzle’, and does so from a long-term perspective on the biological, institutional, and life-cycle factors that have led to the present-day outcome. These questions are rooted in a very long-term development towards higher life expectancy and high dependency on institutions for elderly care outside family circles – which is typical for Northwestern Europe (Reher 1998). At the root of the typical demographic regime of this area lays the emergence of a new marriage pattern during the late Middle Ages: the so-called (Northwestern) European Marriage Pattern (EMP), first described by Hajnal (1965). The EMP makes an historical distinction between areas such as the Netherlands, Belgium, and England from the rest of Europe, by their late marriage ages, small age differences between husband and wife, preference for newlyweds to establish their own households (neolocality), and large numbers of singles among both sexes.

The EMP is associated with issues that we are still familiar with today. First, low fertility leads to small numbers of offspring, and hence fewer children for the elderly to depend on during old age (Hajnal 1965). Second, the postponement of marriage causes a problem with respect to household life cycles, whereby parents may become dependent at a time when their children have small children of their own (Laslett 1988). Third, since the EMP involves neolocality, children leave the parental household at marriage, and elderly live on their own. The EMP was in the way of intergenerational support: elderly became dependent at the time when their children had children of their own. Two life-cycle squeezes (during early adulthood and retirement) coincided. In combination – according to historical demographers – these factors should have caused ‘nuclear hardship’ in Northwestern-Europe, whereby the elderly had little support, and should have suffered from a ‘retirement squeeze’ caused by decreasing income (Di Matteo 1998) combined with a lack of care (Laslett 1989).

In contrast, due to its earlier marriage age for women, relative high fertility, and lower outmigration, Southern Europeans could organize elderly care within families. It can be expected that such hardship would have led to a lower quality of life and life expectancy among Northwestern- European adults than in those regions where the European Marriage Pattern was not present. Data on early-modern life expectancy are very scarce, but those that are available, suggest – surprisingly – that such differences did not exist. When we look at life expectancy of adults aged 30, life expectancy in England is even slightly higher than in Italy and Spain (in this research we use life expectancy of adults as an indicator for living conditions of the elderly, rather than life expectancy at birth, which is to a large extent determined by child mortality (Riley 2001)).


Country; sex



Country; sex



England; all 1401-1425 24.1 England; men 1600-1649 29.8
England; all 1426-1450 25.0 Spain; men 1620-1680 23.2
Italy; all 1427 21.8 England; women 1600-1649 29.6
      Spain; Women 1620-1680 23.6

Table: Differences in life expectancy at age 30 (E30) in NW- and Southern-Europe
(Herlihy 1967; Acsádi 1970; Wrigley and Schoffield-1981; Soler Serratosa 1985)


Our project will first of all develop methodologies and databases to expand the data on early modern life-expectancy at adult age, as literature currently does not provide sufficient material to draw firm conclusions on intra-European differences. Furthermore, to explain how EMP-areas could produce equal – or perhaps even higher – life expectancies than non-EMP-areas, our project discusses two potential explanatory avenues: the biological and the institutional. A prominent theory in evolutionary biology, which is of particular interest in the context of this project, is the Disposable Soma Theory, which predicts that reduced fertility leads to increased longevity (first developed by Kirkwood (1977)). According to Kirkwood, individuals have a limited energy-budget, which can either be used for reproduction or maintenance of the body (soma) - resulting in a trade-off, whereby increased investment in reproduction results in reduced investment in maintenance of the body and subsequent reduced longevity. This links up to evolutionary theory which predicts that there should be a trade-off between reproduction and longevity, because animals must make use of the finite amount of energy that they can accumulate and utilize over their lifetimes. Animals may focus on reproduction or longevity to maximize their overall reproductive success and thereby representation of genes in future generations.

 Figure: Flow chart two potential explanations for higher life expectancy in EMP-areas


Although several studies have found supporting evidence for the disposable soma theory (Kirkwood and Rose 1991; Westendorp and Kirkwood 1998; Korpelainen 2000; Smith et al. 2009; Tabatabaie et al. 2011), others have failed to find any effect, or have found that increased reproductive effort actually prolongs lifespan (Doblhammer 2000; Müller et al. 2002; McArdle et al. 2006; Gögele et al. 2011). Moreover, a link between later age at first reproduction and increased longevity has been observed in more than one dataset (Pettay et al. 2005; Tabatabaie et al. 2011; Gögele et al. 2011), suggesting that not just the fertility but also the starting point of it is of importance, which points out the relevance of including marriage ages and other EMP-related aspects in this study.

In an interdisciplinary field of research where similar studies are producing conflicting results, there is also a strong argument for further research that focuses on variables that may have been overlooked – such as institutional differences between regions – because the absence of important explanatory variables can distort results. This project will therefore also focus on institutions of care and specific socio-economic variables as potentially important explanations, whilst also paying close attention to gender differences. There is a sound rationale for pursuing this avenue of research. It is recognized that conflicting results in previous studies could be due to inadequate control for socioeconomic factors (Le Bourg 2007; Gögele et al. 2011; Tabatabaie et al. 2011), whilst Lycett et al. (2000) observed differences in the effect of reproduction on longevity between different socioeconomic parts of society, with the lowest part being adversely affected by increased reproduction.

Furthermore, this study will focus on the number of offspring surviving to adulthood, as it is worth considering whether this may affect the longevity of parents, perhaps via the role of children as caretakers. The institutional explanation for greater longevity in the face of nuclear hardship, suggests that institutions formed a cushion to the negative side-effects of the EMP in NW-Europe. These include top-down provisions such as government poor relief (Kelly and Ó Gráda 2010), charity (van Leeuwen 1994), and forms of self-organized care –via e.g. the guild-system (van Leeuwen 2012). Apart from falling back on external arrangements, elderly could also opt to hire servants to 'replace' the children that had moved out or they never had, to provide support and care in old-age. In EMP-areas such servants were available: early adulthood service became a very important – and distinctive –  feature of the life-cycles of both men and women in Northwestern-Europe during the early modern period, with often 10 to 15 per cent of the population registered as servants. In the South of Europe this was hardly ever more than 5 per cent (Reher 1998).

This specific aspect of the labour market and of the life cycle of many youngsters in the Northwestern part of Europe may have been one of the 'solutions' for the nuclear hardship caused by the EMP. A first survey of differences between Northwestern- and Southern-Europe – which resulted from the ERC-project from Tine De Moor – has demonstrated that on the one hand there was a very large institutional variety in solutions for old age which may have created a more resilient system than that offered within the realms of the family (Bouman, Zuijderduijn, and De Moor 2012). This difference also became visible in the types of contracts used across Europe to deal with intergenerational transfers. Whereas in Northwestern-Europe the transfer of property was often a post mortem affair – disconnecting financial benefits from the duty to take care of the parents –, in Southern-Europe the right to inherit from parents was often contractually linked to the provision of care during their old age. As a result, elderly in the northwest had to look outside the family for old age provisions. On the other hand, the necessary financial institutions required to make savings during working life, and de-saving during retirement were in place, and were used within all layers of society (Zuijderduijn 2012). Although the above suggests that both demographic and socio-economic decision and activities seem to have affected longevity, there are no studies that address these from a life-cycle perspective, in particular in relation with the way couples were able to build up sufficient capital during the period of child-rearing.

One of the unknown elements in this is the age of dependency. Heavy manual labour and rudimentary medical practice meant that old age – in the past delineated by the deterioration of physical and mental capacities, which reduced labour capacity and left people unable to provide for themselves - came at an earlier age than today (Bourdieu and Kesztenbaum 2007). As a result, the population in demand of care included a large part of the people aged over 50 (Fogel 2004). The main question thus to be answered is whether the relatively high life expectancy in Northwestern-Europe was a consequence of reduced fertility or the diversity of institutions on which the elderly could rely to provide in their care. Was it the effect of nature or nurture? And how strong were both effects, and was there a genderdifference in these effects? When did one become more important than the other? And how were people in different household compositions capable of dealing with these life-events and income strategies throughout their entire life-cycle?

A focus on regional variations may also show us to what extent biological effects can be moderated by the presence of institutional support outside of the family; likewise, variations in institutional provisions will also provide us insight on the quality effects of such varieties in care models.



In order to answer the main research question, several sub-questions need to be addressed. This will be done in three subprojects, each focusing on a specific aspect of the elderly care puzzle. All focus on the period 1500 to 1900, thus covering also the beginning of the formation of the welfare-state, and they will all cover developments in Northwestern-Europe (The Low Countries and England) and Southern-Europe (Italy and Spain). The subprojects rely on several databases that will be developed for the use of all project members, in collaboration with external partners.

Subproject A: Longitudinal effects of biological determinants on life expectancy and old age dependency

In Subproject-A we will examine the extent to which the physical demands of child-rearing (as indicated by parental age, family size, birth spacing, occupation, and presence of living grandparents) as opposed to expenditure of biological resources (e.g. pregnancy and childbirth) may have affected longevity. This might explain, for example, why a correlation between lifespan and reproduction has been observed in men, whereas men do not bear the physiological cost of reproduction (Le Bourg 2007); also why a socio-economic dimension has been observed in the link between longevity and reproduction (e.g. Lycett et al. 2000).

Subproject-B: Institutional responses to the changes in life expectancy

Subproject-B investigates how institutions other than the household allowed for old-age provision – family based support will be analyzed in Subproject-C. In particular, we will address the question of how societies responded to changes in old age dependency rates. How did societies in Northwestern-Europe respond to the challenges the EMP posed with respect to old-age provision? How long did it take them to come up with solutions, and which organizations were responsible for these? And why do we see great variation in the solutions societies in the Northwestern- and Southern-Europe came up with, both historically and today (Esping-Andersen 1990)? Addressing these questions will allow us to demonstrate which societies responded to ageing-related challenges, and which did not, and thus to enhance our understanding of processes of institutional change. We will focus on three institutional clusters: top-down care (provided by state or benefactors), bottom-up care (provided by craft guilds and other forms of collectivities), and commercial care (provided via markets).

Subproject C: Life cycle events, old-age provisions and life expectancy

Subproject-C studies the impact of life-cycle events and socio-economic behaviour on life quality and expectancy, by focusing on the one hand on household structures, and on the other on saving behavior. Household structures are determined by age at marriage, number of children, and relations between parents and offspring. All have a crucial impact on the family safety net of the elderly: parents can opt for a small number of children, who receive proper training, or a large number of less educated children. This quantity-quality trade-off (Klemp and Weisdorf 2012b) had an impact on the old-age safety net of the parents (was it better to have a few but trained adult children?). The main instrument used to reduce fertility, postponement of marriage, also determined the age-gap between parents and children, and hence the type of support the elderly could receive during old age (Bouman, Zuijderduijn, and De Moor 2012). In this respect modes of co-habitation were important as well: in Southern-Europe it was more usual for children to live with parents, in multi-generational units that provided for assistance for the elderly (Ruggles 2003). In the Northwest, neolocality was more usual: since the late Middle Ages households were generally small, consisting of couples and their infant children. What did this mean for possibilities for family assistance? Did weak ties between family members lead to a lack of support? If so, which of the old-age institutions discussed in Subproject-B did they then turn to?

Sources and databases


One of the major innovations in this project will be the combination of different databases which use different units of data collection and different historical sources in order to overcome biases in the analysis. Underneath an overview is given of the databases that will be constructed, their features, geographic delimitations, and (agreed) collaborators.




Data collection unit

Types of sources used

A Longitudinal non-location-specific family reconstructions Family trees / individuals Complete genealogical trees (based on a.o. parish registers)
B Location-specific family reconstructions Village / individuals Baptismal / (pre-) marriage / death / burial records
C Reconstruction life expectancy within religious communities (control group) Religious communities Life records
D Overview of types of provisions for elderly care and of intergenerational transfer Institutions  Regulations of institutions / Wills and testaments

Table: Overview of databases related to and / or resulting from this project





> Back to Projects