Historians have long searched for answers about today's world from the past, such as "Why do civilizations collapse?". Although historians look for answers from language, today mathematicians like Peter Turchin, a professor at the University of Connecticut, are using math to gain further insight. Turchin is the driving force behind a field called “cliodynamics,” where
scientists and mathematicians analyze history in the hopes of finding
patterns they can then use to predict the future. And unless something changes, according to Turchin, the U.S should expect a large amount of violence (terrorist activity, uprising) in the year 2020. A summary of these 100 year "waves of violence" in the U.S can be seen below: It is interesting to note that these spikes in violence occur in 50 year cycles in the U.S, and that these "secular cycles" occurred in all past agrarian states in which records were available (i.e Ancient Rome, Medieval England, Dynastic China, Russia).
Although Turchin is not able to apply many big data techniques in his analysis due to the lack of historic data sets, he admits that creating models on these historical data sets were not even possible until recent history when old documents started to become digital.
http://www.wired.com/wiredenterprise/2013/04/cliodynamics-peter-turchin/
This is a pretty thought provoking presentation. As you mentioned before, there is a "missing spike" in 1820. Given this, does the remaining three spikes give enough predictive evidence to say that another "violence" spike will happen in 2020? Turchin's reasoning for the "missing spike" is the "relative prosperity" that was present in 1820. If this is the case, it would be interesting to note if other civilizations that Turchin mentioned having missing spikes were more prosperous around those times as well.
ReplyDeleteFurthermore, the parent article stated that only recently have historic data sets been digitalized to the point that are now available for use. These visualizations can change drastically depending on future availability of more historic data sets. In essence, I don't know how accurate these patterns are given the lack of data that there seems to be.