On Measuring and Explaining Union Democracy: A Reply to Dr. Martin's Critique

Abstract

The Martin communication generates more heat than is warranted by the extent and nature of the disagreement over indicators of democracy, and is mistaken in emphasizing contextual variables at the expense of formal organization. The measurement of democracy. Since democracy is multidimensional conceptually, only a number of measures of various aspects of union democracy over a period of time could do justice to the situation in any particular union. At this exploratory stage, there is no reason to exclude any measure which adequately represents a relevant concept, even if there is disagreement as to its importance. The aspect of democracy we chose to emphasize was effectiveness of opposition, which was measured primarily on the basis of electoral results. As we pointed out : 'There would be little point in singling this out for empirical studies on trade unions if there were little problem in this respect. The fact is, however, that it is precisely effective electoral opposition or competition which is alleged to be absent in the overwhelming majority of unions' (1970: 149). Martin has stated: 'Union democracy exists when union Executives are unable to prevent opposition factions distributing propaganda and mobilizing electoral support' (1968: 207). In those American unions in which there was no electoral opposition for top or second-ranking during 1949-1966, and in some cases for much longer, it is a reasonable assumption that such opposition was 'prevented*. (All were required to hold elections periodically, at intervals no longer than five years.) In those British and American unions which had fairly close elections or defeats of top officers, we know that organized opposition of a factional nature was involved in most cases, if not in all. Thus there is probably a moderate correlation between Martin's survival of faction, or tolerance of organized opposition, as it is variously stated, and our mean closeness of elections over a period of time. The fact that there is ordinarily ťno union equivalent to the Parliamentary electoral pendulum' (1968: 207) does not invalidate the usefulness of electoral data.