Adult mortality in Catalonia in the 16th and 17th centuries

Francisco Villavicencio, Max Planck Institute for Demographic Research
Joana-Maria Pujadas-Mora, Universitat Autònoma de Barcelona
Fernando Colchero, Max Planck Odense Center and University of Southern Denmark
Anna Cabré, Universitat Autònoma de Barcelona

Due to the lack or incompleteness of census data and death records, little is known about mortality patterns of Catalonia in the 16th and 17th centuries. The purpose of this paper is to present a methodology to estimate adult life expectancy with unknown ages-at-death, using data extracted from historical marriage licenses records of the Diocese of Barcelona. The analysis method to be used is the Bayesian Survival Trajectory Analysis (BaSTA), a software package for estimating age-specific survival from capture-recapture/recovery data under a Bayesian framework that was originally designed to study the survival in wild animals with unknown ages and unknown ages-at-death. Marriage licenses records from the Diocese of Barcelona between 1451 and 1905 are conserved in set of 291 books known as “Llibres d’Esposalles”. The use of BaSTA is justified by the type of information registered in each marriage record and by the fact that the age of the grooms remains unknown. Over the 31-year period from 1598 to 1629, additional information concerning the spouses’ parents is available, including a notation indicating if they were alive or not at the moment of their children’s marriage. This enables the realization of a nominal record linkage among marriage records to reconstruct individual’s lifespans: there will be several observations for each individual (the own marriage, and the marriages of the offspring) knowing if he or she was alive at each observation. Our model uses parametric laws of mortality as Gompertz and Makeham, and also Model Life Tables that are introduced to select the best mortality pattern. Several simulations have been carried out, obtaining similar values of life expectancy at age 15 (around 30 years for both sexes), which are plausible according to previous literature.

  See paper

Presented in Session 115: Multiple aspects of mortality and health

´