Debates on marriage patterns and institutions for collective action

How is genealogical data useful for historical research?


The use of genealogies in historical research has been criticised by some, but we think the value and potential of this type of data is under-estimated.


The criticisms we often hear are that genealogies:

  • under-report childhood and female deaths;
  • tend to follow patrilineages (i.e. one male descendent to the next);
  • over-report larger, wealthier families and those marriages which produce children;
  • often contain incorrect information.


It is clear why such biases arise: for example, people wish to trace their ancestors along their blood-line, which tends to be more often possible along the paternal line and by definition excludes childless marriages.


However, it is not clear how people would conduct genealogical research if they did not reconstruct their blood-lines in this way, because it would be completely impractical for any individual to trace all of their indirect ancestry. If you go back 10 generations, you have 1,024 blood-line ancestors (1,024 great-great-great-great-great-great-great-great-grandparents) and those 1,024 ancestors undoubtedly have many millions of descendants that are your indirect ancestors.


The important point for us as historians and scientists is to understand what errors and biases we find in genealogical data. In this way, we can know what questions we are able to ask of the data and how to correct for errors or biases. It is true that information can simply be incorrect, perhaps because of inaccurate transcription or conflicting historical sources - see for example the debate on the origins of William Gifford, which are much debated. It is encouraging that there are lively debates among genealogists, because this process improves the quality of genealogies in the long term.


It should be recognised that there will be always be a degree of error in genealogical data, because not all of history is knowable and verifiable; nonetheless, we employ several methods to identify transcription and translation errors, such as checks on the plausibility of dates and ages and comparison of the same individuals and families between different genealogies. And importantly, we do not use those genealogies which have significant errors.


There are a number of advantages to using genealogical data:


  • There is a lot of genealogical research out there! This is a very important point, because larger samples of data allow us to get better estimates of demographic parameters, such as marriage ages or life expectancies.
  • The work of linking births, marriages and deaths from historical records and defining relationships between people has been done by human beings. In the future it may be possible to accurately link large numbers of records using computers, but these techniques are in their infancy, and besides many record sources are not yet digitised, so are not ready for computers to use. 
  • A genealogy is typically not limited to a single location, as with parish registers, for example; this means that we do not always lose information about marriages or deaths when people migrate away from their birth-place.


All historical sources have to be treated with caution and interpreted correctly. It is clear that genealogical data is a challenging resource to work with, but we recognise this in our research and are applying rigorous and scientific methods to address data quality issues, whilst recognising and utilising the full potential of genealogies to inform us about the past.