Dag Anckar, "Dominating Smallness: Big Parties in Lilliput Systems," Party Politics, 3 (April 1997), 243-263.

First Paragraph:
In their seminal work on 'Size and Democracy' Robert Dahl and Edward Tufte suggest that within systems of representative democracy 'the larger the political unit, the greater the extent to which public conflict is expressed and resolved through formal and impersonal organizations rather than through informal, face-to-face negotiations by the antagonists themselves' (Dahl and Tufte, 1973: 96). Since the political party is in representative democracies 'the most visible and in many ways the most important organization involved in conflict resolution' (p. 96), Dahl and Tufte expect that in larger systems political parties are more deeply involved in the everyday management of conflict. Furthermore, since the number who openly dissent are fewer in small systems in relation to the numbers holding the majority view, the likelihood of dissenters finding enough allies to pass the threshold for dissent thus being smaller (pp. 91-2), they expect that in larger systems parties are more evenly balanced in their struggles. In larger systems, as dissenters form a larger proportion of a total, the minority party or parties are less overshadowed by the majority party or coalition than would be the case in smaller systems.

Figures and Tables:
Table 1: The small island states of the world: number of parties (1993).
Table 2: Dominant parties in small island states: countryside configurations.
Table 3: Predominant party systems in the small island states.
Table 4: The occurrence of parties and predominant systems in miniature island states.

Last Paragraph:
Theories involving notions of size belong to a set of theories that are usually submitted to objections concerning determinism and the definition of, for instance, organizations as passive objects at the mercy of contingent variations of size (Panebianco, 1988: 183-4). The findings presented here certainly seem to suggest that there is ample room for active subjects and various intervening mechanisms and factors in the creation and shaping of party systems. Although the environments observed are, by and large, similar in size, the outcomes, in terms of parties and party systems, have been very different. Above all, the belief that small units, because of their smallness, produce party configurations characterized by a lack of parties or, alternatively, by predominance, has received only weak overall support. As a rule, smallness does not produce systems dominated by big parties; there are few giants in the lilliput systems. Various factors probably account for this, one of the most important being that traits like homogeneity, which are usually believed to appear in close association with diminutive size, do not in fact show a systematic association. On the other hand, however, notions of size are by no means totally meaningless. To the many existing observations of parties and party systems that emphasize the role of environment and environmental changes should be added the finding in this report that size apparently makes a difference if certain thresholds are passed. When one moves from small island systems to very small island systems, relevant differences appear to emerge, suggesting a link between miniaturism and certain party system characteristics. It may well be the case that research into the impact of size would in general benefit greatly from distinctions that advance further in the structuring of the independent variable than is usually the case. This point is fully argued in the volume by Dahl and Tufte (1973: 41-65, 94-109), which inspired this study, but the lesson does not always seem to have been learned in the relevant literature to a full extent.