Sunday, 17 January 2010
Woolworths and Leylines
The Guardian's always excellent Ben Goldacre strays into the world of archaeology with a nice piece on the latest claims about the sacred geometery of the prehistoric world - also worth reading for the comments below. The work in question claimed that prehistoric monuments were so arranged as to form a network of triangulated points that were used by past societies to navigate around the country. It also reports a counter analysis that showed that similar patterns could be found in the spatial distribution of Woolworths
The key point, of course, is not whether prehistoric societies ritualised their landscapes through monument construction (something accepted by all mainstream and 'alternative' archaeologists), but how data is used and analysed. Like any study which involving recognising patterns in huge amounts of data, it never really confronts the fact that if we have enough points of data (whether these are inscriptios, Bronze Age mounds etc etc) and subject them to enough analyses seemingly meaningful patterns will be found. However, the trick is proving whether these apparent patterns are a function of meaningful action by a past society or just a freak of statistics. Another example of this is the work by Charles Thomas drawing on the approach developed by David Howlett on Biblical Latin Style on the early medieval inscriptions of Wales and Western Britain. Thomas's analyses of these inscriptions seem to show messages (and even images) hidden within these simple inscriptions (this is best laid out in his his book Celts: Messages and Images. Stroud: Tempus, 1998). He argues that these messages can be made visible by certain mechanisms such as letter counting and the ascription of numeric values to letters. However, his critics point out that it is possible to identify hidden messages using such techniques in almost any text if you analyse it in enough different ways - is it a case of the 'wisdom of ancients' or simply infinite monkeys producing, if not Shakespeare, then at least Biblical Latin?