One challenge in using CBDB data in prosopographical research is seeing the patterns in large amounts of data. To do this we can use statistical software and we can use geographic information system (GIS) software. For China’s history the China Historical GIS (CHGIS) project makes datasets of the administrative system between 221 BC and 1911 AD and major non-administrative towns for 1820 and 1911 freely available. Using GIS software such as ArcGIS or MapInfo (or even GoogleEarth) we can combine CBDB output with CHGIS datasets.
This paper illustrates some ways in which CBDB data can be linked and mapped in a GIS.
When we send a geospatial query to the CBDB database we can ask that the report be output for use in a GIS. The resulting text (.txt) file will include the x/y coordinates for all places mentioned in the query as well as totals for places. For example in this query the user asked for listing of all those in CBDB who gained a jinshi degree between 1127 and 1199. He added this to the GIS (adding the .txt file as an “x/y event”). Once it appeared on his screen he used the GIS program to display the results by quantity and place.
The result shows concentrations in the Liangzhe East and West, Fujian, Jiangnan East and West, and Chengdu circuits. His next step was to see whether this pattern changed during the last 75 years of the Southern Song.
This suggests some further narrowing of geographic representation in government. Given that CBDB skews towards those who rose higher in government he then wanted to see if this pattern was also true generally. There are two complete lists of degree recipients for the Song, from 1148 and 1256. The 1148 list is complete in CBDB. Comparing that against the locations and quantities of all Song jinshi degrees in CBDB gives the following.
We see a similar spatial distribution with some expected and unexpected differences (unexpected are the greater representation of the eastern Sichuan basin and the few successful candidates from the northeast of modern Zhejiang; expected is the number of candidates who still have a formal household registration in the lost capital of Kaifeng).
Other kinds of queries sent to CBDB can also be output in forms that can be put into a GIS. The following are examples of different kinds of social networks. The first is a listing of those who were listed (and proscribed) as members of the Yuanyou and Yuanfu reign period coalitions of officials who tried to get rid of Wang Anshi’s New Policies. The goal was to see if the two groups came from the same places.
In a GIS all the data underlying the visual display can be queried. In this case, having seen where the coalition member come from, we want to find the names of Yuanfu coalition members from Shaowu in Fujian. In all the examples used here we are able to look at the tables from which the data come and ask for more information about particular cases on the map.
The second example maps marriage networks, a social connection that CBDB can output for any given individual or family line. In this case we compare the spatial extent of the marriage networks of the Northern Song statesman Sima Guang and the Southern Song statesman Shi Hao.
The third example is the network of non-kin social connections of Lü Zuqian. In this case we want to distinguish between first, second, and third order connections.
The final example brings together CBDB data with information from other CHGIS-based data sources, using a GIS program to combine them all. In this case the question was whether there were obvious correlations between commercial activity as represented by the commercial tax quotas as of 1077 and the success of localities in the civil service examinations during Northern Song.
