Settling the Harappan Chronology in Gujarat Using AMS Dates and Bayesian Statistics

Kalyan Sekhar Chakraborty1, 2, Prabodh Shirvalkar3, Yadubirsingh Rawat4, Sanjay Manjul5, Heather M.-L. Miller6

1Max Planck Institute for the Science of Human History, Germany

2Ashoka University, India

3Deccan College Postgraduate and Research Institute, India

4Gujarat State Archaeology (RTD), India

5Institute of Archaeology, India

6University of Toronto, Canada

Gujarat, the Western province of India, was extensively occupied during the Indus period. Explorations and excavations in this area by both Indian and international scholars have identified numerous settlements belonging to all three primary Indus phases. Interestingly, in Gujarat, both the Classical Indus tradition, with similarities to the core region, and the Sorath tradition, specific to Gujarat, coexisted during the Indus period. Numerous settlements from both these traditions have been excavated, and they have produced a significant number of radiocarbon and AMS dates. However, no efforts have been made to statistically categorise these vast numbers of dates in order to provide reliable date ranges with error margins for different Harappan phases, and to understand the interactions between the Sorath and Classical Harappan traditions. In this presentation, we will discuss the AMS dates from Kotada Bhadli, a Sorath settlement, and compare them with published dates from nearby settlements using Bayesian statistics. We will also discuss the use of animal skeletal remains as an alternative to charcoal in order to securely date archaeological settlements from India and to overcome limitations such as taxonomic identification, old wood effects, and the dispersion, bioturbation and unreliability of small charcoal.