commercial network in the 13th to the 16th centuries. The discussions that took place under these heads are as follows.1) Early State Formation, from Chiefdom to State.In Session 2, M. Yamagata reported on the development of Linyi polity in Vietnam, and R. Gurukkal on the Satavahanas in the Deccan and early polities in Tamil Nadu, South India. Gurukkal insisted on differentiating these South Indian polities from the state regarding the former as chiefdoms on the basis of no development class relation and sedentary agriculture. In Linyi though there was a change in the influencing agent, from China to India, probably in the 2nd century, her report on its early state formation seems to confirm ‘social nearness’ and ‘convergence’ in early state formation in South Asia and Southeast Asia pointed out by Kulke.The examination of the early state formation was further carried out by three scholars who studied the regional state formation of Bengal, Thailand and Sumatra. R. Furui stressed on the regional difference in development within Bengal itself before the emergence of a centralised state there. E. Nitta talked about the formation of Dvaravati state based on the network connecting port-cities and inland area. This point was extensively discussed in relation to a similar type of state formation in Sumatra. Though there remain many unclear aspects in the state formation of Srivijaya, the city-state formation based on the river system was suggested by Manguin. With respect to this, many participants pointed out the importance of recognizing the different ecologies in Southeast Asia itself (for example, Java and Sumatra). 2) The Role Played by Ideology and Caste (Varna) System in the Later State Formation.R. Champakalakshmi explained the part played by the ideology (Puranic religion and bhakti) in the process of later state formation taking up the Pallavas, Pandyas and Cholas for the period from the 5th to the 13th centuries. She emphasized the merging of the North Indian Brahmanical tradition with the local religious tradition in the Tamil country, which affected the religious policy of these states. She stressed upon the difference between the Pallavas/Pandyas and the Cholas with regard to the degree of importance given to the local Tamil tradition (vernacularisation), and though others shared her view, the time when the change occurred (since when in the Chola rule̶from the beginning or from the time of Rajaraja I) remained debatable.F. Matsuura explained change in the concept of kingship during the Angkor period by examining inscriptions and referring to studies by G. Coedes, I. W. Mabbett, H. Kulke and M. Vickery. The Cambodian concept of devara-ja invited discussion regarding divinity of kings not only in Southeast Asia but also in South Asia and Kulke denied divinity in both regions in his concluding comment. T. Aoyama studied state and social integration of Majapahit in Java. Through the analysis of an old Javanese narrative, Desawarnana, he emphasized that the Majapahit kings depended on the Indic (Hindu) ideology for the integration of state and society. Though many participants accepted his interpretation of the centralisation of Majapahit state based on sedentary agriculture, questions regarding how such integration was possible despite the non-functioning of caste system (varnasramadharma) in Java arose during the discussion. This aspect of his presentation awaits further studies.3) Mandala Theory and Samanta System.In their reports and in discussion Manguin and Aoyama expressed their dissatisfaction with the mandala state theory applied to Southeast Asian states by O. W. Wolters and others. Manguin argued that city-state formation depended on the river system in Sumatra and Aoyama favoured the interpretation of a centralised state for Java.052MODERN ASIAN STUDIES REVIEW Vol.5