Appendix to Power and Size:

Urban and Polity Size Swings and changes in the distribution of power among states in interstate systems since the bronze age

v. 8-11-17

Mari (modern Tell Hariri, Syria) 2900 BCE - 1759 BCE (middle chronology)

Christopher Chase-Dunn, Hiroko Inoue and Levin Welch

IROWS Working Paper # 108 available at

Appendix url:

Table of Contents

Wilkinsonís Power Concentration Codings


Our Estimates of Largest City and Polity Sizes


Comparison between 50 year and 100 year estimation time points for the Central PMN


Excel File Combining Wilkinsonís Power Concentration with City and Polity Size Estimates


Short paper discussin the population size of Hangzhou and its connection with changes in the geopolitical structure of East Asia


List of resources for studying the Aegean PMN 1600 bce Ė 600bce

Warfare Data

Cioffi and Lai, Ancient China 2697 bce-729 bce

Data set:


Figure A1: Component of Early East Asian warfare intensity scores (1900 bce- 700 bce (50 year intervals)

Figure A2: Warfare, Power Concentration, City Sizes and State Sizes in Early China (1900 bce-700 bce) 50-year intervals

The correlations based on the numbers in Figure A2are interesting.  Warfare intensity is correlated with year Pearsonís r= .45 (more wars in more recent years). This could be because the number and size of wars increased over time or because our knowledge of the existence of wars improved. The Xia epoch of Chinese history that constituted the early part of these estimates (before 1797 bce) is not well-documented, and our knowledge of the Shang epoch 1797 to 1120 bce is less well-known that the Western Zhou epic (1120 to 729 bce). Improvements in the completeness of documentation may explain the correlation between conflict and year.

It is hard to see in Figure A2 5 because of scaling, but there is a negative correlation between war intensity and power concentration (-.49), and this may not be due to a secular trend because usually power concentration is not correlated with year. Geopolitical theory generally predicts that more power concentration will produce less war because the preponderance of a great power dissuades challengers. The correlation between city size and war intensity is .38.  This is probably mostly due to the fact that both increase with year (trend).  We will calculate the partial correlation controlling for year but have not done that yet.

††††††††††† The correlation between the territorial size of the largest polity and war intensity is -.07. probably zero.  Better temporal resolution may find that warfare decreases the size of the largest city, but we do not see it with the fifty-year intervals.


Brecke and Kang, East Asia 1400 cd- 1999 ce

Breck Conflict Catalogue 1500 ce- 1854 ce