Times as Misty Window
In many areas of life, both within academia and without, people are interested in discovering and understanding what occurred in the past. The very nature of subjects such as archaeology and history are to discover about the past. It is central to our understanding of the universe - determining how it came into being. It is a key strand of biology, to determine the evolutionary history of life on earth and discover the relationships between species. Outside of academia, accurate reconstruction of what happened in the past is, for example, a key feature of police work, and is of great interested to millions of individuals researching their family history and genealogy.
The common feature of all these situations is that information that was readily available at the time that the events originally occurred is lost as time passes. Time becomes not a clear window by which we can see all that has occurred, but a misty window through which only some information is transmitted. In computer science and information theory such a situation is modelled as a noisy channel along which information is transmitted. As information passes along the channel, some of it may be lost and some of it may be corrupted. Modelling how information is transmitted can lead to accurate analyses of how much of the past we can ever hope to reconstruct accurately, or how we could combine different data sources to build a more accurate reconstruction of the past.
The use of techniques originally developed to reconstruct the evolutionary history of species, has been seen in many different fields in recent years. These 'phylogenetic methods' have been applied to analyse the evolution of languages, the evolution of culture (from weaving patterns to folklore), the copying history of ancient manuscripts and the evolution of industry. Indeed any situation in which there is a process of repeated imperfect copying of information, whether by biochemical processes, by teaching a younger generation or by scribes, could be suitable for analysis with phylogenetic or information theoretic techniques.
There has been similar progress in use of techniques developed to study aspects of ecology where there is missing data, such as capture-recapture models. Some of these models have been extended to political history and prosopography, but these and other statistical methods have not been widely applied.
The question to be explored in this activity is: can these techniques be applied, directly, by analogy, or with appropriate adaptation, to a wider range of fields facing the same underlying question - what exactly happened in the past?
During 2011-12 a number of seminars were held to allow the University to establish capacity within this area. Members from this group will convene a series of seminars on this theme during 2012/13. Speakers include Dr Magnus Bordewich, Prof Rus Hoelzel and Dr Jamie Tehrani.
For full details of all the events please visit the Events Listing