Indonesia has frequently been referred to in the literature as a good example of a core-periphery model, with Java the nuclear, congested heartland and the Outer Islands the less integrated, underdeveloped hinterlands. Yet while true in certain ways, in other significant respects this model is inaccurate. Factor and canonical correlation analysis of provincial per capita characteristics indicate that Jave does not exhibit many of the socio-cultural, interactional, and economically dynamic features expected on a national core: only Jakarta demonstrates overwhelming core characteristics. Contrasts among Outer Islands provinces are greater than those between the Java 'core' and the periphery. Through analysis of the factor scores, the spatial pattern of integration reveals that the most integrated and/or highly economically developed provinces on a per capita basis are separated both from each other and from Jakarta by less integrated and developed provinces. The implications of these findings for the continued integration of Indonesia and the direction of its economic development are discussed. National integration is a vast, multidimensional, dynamic, and highly complex concept, which is of vital importance in every nation-state. Integration, in the sense of a cohesive force which binds the various regions and diverse peoples of a country together into a functioning and mutually interdependent whole, is essential both for the continued existence of a state as a political entity, and for development. For without some degree of integration within a country, economic development in the sense of structural transformation of the system cannot take place.1 And the success of economic development depends to a considerable extent on the strength of those integrative, cohesive bonds, to ensure that the almost inevitably uneven spatial impact of development does not unduly exacerbate regional differences and tensions, and cause the breakup of the country. Hence an understanding of patterns of national integration is vital in economic development planning. Of the many historical, political, social, cultural, interactional, and economic elements that contribute to the idea of national integration, a substantial number can be studied only subjectively-aspects such as shared historical experiences, the role of national leaders, the value of symbols, and the perceptions and attitudes of the inhabitants toward the state.2 However, it is possible to quantify and analyse statistically three of the major dimensions of integration (the socio-cultural, interaction, and economic components), to consider their interrelationships, and to ascertain the degree to which each region or province is integrated into the national whole on each of these dimensions. This paper sets out to analyse the spatial patterns of national integration in Indonesia in the early 1970s. It is a topic of particular relevance because of the tremendous variety and uneven distribution of peoples, cultures, and levels of development in this archipelagic nation which extends over 3300 miles from west to east and consists of over 13 000 islands. Indeed Indonesia's need for greater integration is demonstrated in the twenty-one uprisings or regional rebellions it experienced in the first twenty years of its existence as an independent nation-state (from 1945-65).3 Trans. Inst. Br. Geogr. N.S. 6, 471-90 (1981) Printed in Great Britain This content downloaded from 157.55.39.161 on Mon, 23 May 2016 05:51:26 UTC All use subject to http://about.jstor.org/terms