1. Small societies and small states Interest in small states as a separate field of study is still at an early stage of development, although the first studies were carried out in the 1950s and 1960s (Fox 1959, Robinson 1960, Benedict 1966). Researchers disagree on how best to distinguish between large and small states. Population size is usually taken as the main criterion, although common alternative or supplementary indicators are surface area and the size of the economy. The cut-off has often been set between 1 and 2 million people, although some scholars (e.g. Bray and Packer 1993) suggest that it would be more appropriate to examine issues along a continuum of size. 49 states out of 188 member states of the United Nations had populations below 2 million in 1999. Countries with population less than 100,000 are often called microstates, which can be distinguished by more specific characteristics. Within Europe, states with the population between 100,000 and 2 million include Cyprus, Estonia, Iceland, Luxembourg and Malta, with Latvia and Slovenia reaching only slightly over 2 million. It is important to distinguish between the size of a state and the size of a society in the development of small state theories. The majority of contemporary studies of small states refer to Benedict (1966), who has noted that the main criteria of size for 'territories' ('states') are area and population, whereas the criteria of size for 'societies' are the number and quality of role-relationships. He pointed out that small societies do not exist only in small states, for they may also exist in large states that have high degrees of segmentation (including, for example, minority groups, islands and other relatively isolated and/or closed labor markets or communities within larger states). Benedict (ibid.) claims further that, just as it is possible to have a small-scale society in a large state, it is also possible to have part of a large-scale society in a small state. He presents Luxembourg and Monaco as two examples of states, which are closely related to neighboring states and thus are not considered as small societies by him. Consequently, although the theories of small states refer to certain 'states', the same characteristics may apply to small 'societies' within large states, and may not apply to small states which are parts of larger societies. For the same reason, small states may be defined differently in the study of economics and political science (going up to 20 million of population), whereas the study of public administration (and also sociology) has more implications from the notion of small societies, where the cut-off line between small and large states is substantially lower. In this article, the term of 'small states' refers to 'small societies' which have less than 2 million inhabitants. Previous studies suggest that small countries are not simply smaller versions of large countries. Differences between large and small states are not merely quantitative--essential qualitative differences can also be found. Benedict (1966) has shown in his study of the social anthropology of small societies, that people in small societies grow up within an interdependent network, where each person plays several roles; thus nearly every social relationship serves many interests. In a small society individuals interact with each other over and over again in a wide range of social situations. In such conditions the decisions and choices of individuals are influenced by their relationships with other individuals in many contexts. Consequently, relationships in small societies seldom concentrate on a single act or specific function, but tend instead to be functionally diffuse and to last for a long time, though their specific content changes over the course of life. Talcott Parsons (1939, 1951) has characterized such role-relationships as 'particularistic', where role-relationships extend over a considerable time-span and the roles involved are usually ascriptive. …