Probability is probably the most powerful conceptual tool that we have yet developed to understand and manage uncertainty. Scientists, demographers, economists and other researchers who are engaged in forecasting the future regularly use probabilistic techniques to manage the massive uncertainties they face. Rather than trying to estimate the future value of the variables they are interested in exactly, they tend to think in terms of ‘probability distributions’ over ranges of possible values. In effect, they draw on the power of modern computing to model a very large number of possible futures in order to see how frequently different outcomes occur, rather than trying to model a single ‘best estimate’ scenario. The probabilistic estimation of uncertain variables was pioneered in the field of nuclear safety in the 1970s and has since become a workhorse of the natural and social sciences. Decades of experience have refined both theory and method. Probabilistic models are not infallible (as the financial crisis has revealed), but they capitalise on the power of computing and offer a balance of pragmatism and rigour that makes them far superior to the deterministic ‘best estimate’ models they displaced.
Historians have been slow to see the potential to approach to uncertainty about past in the same way. It is easy for ancient historians in particular to assume that we have a near monopoly on massive uncertainty, but there are many other fields in which estimates have to be based on subjective assessments of what is likely, rather than hard data. Though the scientists who forecast climate change and the demographers who forecast population growth have some data about past trends, projecting those trends into the future always involves subjective judgement. The key difference between those fields and our own is that they are more sophisticated in their understanding of uncertainty and their methods for managing it.
The project aims to demonstrate that probabilistic modes of analysis that have been developed to manage uncertainty about the future can be redeployed as powerful tools of historical analysis.