I don’t know whether to laugh or cry when confronted by advocates of what one might term classical genealogy. That is, that nothing is proven until there are, say, three documented sources in support a given event. In Ireland of course, we are lucky to have even a single documented data source for an event, and often there are none. It follows, therefore, that if you are an advocate of a “classical” approach then rigorous genealogy cannot exist for most of us with Irish ancestry. And that is simply true.
So, what to do? Can we do anything?
I for one try to follow what might be loosely considered to be a Bayesian approach. That is, to utilize what is known, plus logical inferences therefrom, to determine a “most likely” reconstruction of a given family structure. In this approach, the resulting model for the historical family relationships is subject to revision as further data (if any) become available, leading to a revised “most likely” family structure.
In this approach a probability is assigned to each hypothesis, and we choose the most likely one. However, this assignment of probability can be subjective. That is, you may take the same known data and come to a different conclusion as to the most likely hypothesis. This is inherent in this approach.
Important points:
- The process has to start with known data as input! Not mere wishful thinking.
- Similarly, the assignment of probability to a given reconstruction has to be based on the known data, and the known data alone, and logical inferences therefrom.
- As more data are obtained, the “most likely” reconstruction will become more and more constrained. In the limit, the reconstruction will become fully determined, and will be identical to that obtained from the “classical” approach mentioned at the start.
Finally, I wish to repeat that I have described this methodology as Bayesian in a very loose way. Don’t take it as a literal statistical calculation.
