ABSTRACT
However, there is no attempt to impose a ceiling on the size of population or area of islands to be considered here. King (1993) noted that there have been suggestions that small islands are those below 10,000 km2 and 500,000 people (Beller, 1986) or, alternatively, below 13,000 km2 and one million people (Dolman, 1985), but such distinctions are artificial. King found them ‘not particularly helpful’ for the Mediterranean (1993, p. 17); nor would they be anything but frustrating if imposed on small island studies elsewhere. Thus in the Indian Ocean, Beller’s threshold regarding population, if not area, would exclude Réunion and Mauritius from consideration, although both are framed by their insularity (though see Eriksen, 1993, for a discussion of this idea applied to Mauritius). Even more difficult would be Dolman’s definition. This would count Réunion as a small island, given its population of c. 624,000, but not its neighbour and direct comparator, Mauritius. The latter is slightly smaller at 1865 km2 (for the main island, 2040 km2 for the state) to 2510 km2
for Réunion, but has a larger population of 1.1 m, above Dolman’s threshold. The approach here is just to acknowledge that the effects of insularity tend to
be more pronounced the smaller the island is, but islands of any size will be brought into the discussion where appropriate. Chapter 8 will be a detailed consideration of the problems of scale.
