NE of the persistent problems in contemporary democratic theory has been to specify the proper relationship(s) of democratic ideals to actual political practices. This problem is often expressed in the form of a dilemma between the demands of democratic ideals and empirical constraints on their realization. For example, Margolis argues that, "It may not be possible to develop a theory of liberal democracy which is compatible with both the empirically demonstrated capabilities of the citizenry and the theoretically demonstrated necessity of preserving liberal values" (1979: 93). There seems to be an inevitable trade-off between making democratic theory realistic or practical, on the one hand, and fulfilling democratic ideals, on the other. If we allow empirical research to constrain our ideals by leading us to underestimate what is possible, then we may undercut those ideals and fail to use them to guide us in improving the political system. If we insulate our ideals from empirical evidence and analysis, then we may be merely building elegant but wholly unrealistic and useless analytical systems. This problem is not unique to democratic theory. Any ideals or ideologies simultaneously express the way someone thinks the world does work, should work, and even might work if social and political behavior and institutions were changed. The problem arises when the world does not conform to these expectations. In the face of discrepancies between ideals and reality, one might cling to the former believing that the contrary evidence is merely an aberration or a misinterpretation. Or, more aggressively, it might be demanded that the world itself be changed to conform more closely to the ideal pattern. On the other hand, the collision of ideals and facts may prompt an adjustment in what one believes can and should be done. This latter strategy is often what is anticipated by the injunction to be "realistic" in political and moral affairs, i.e., to temper or modify our moral expectations on the basis of a clear view of empirical evidence and practical possibility. The appeal to realism is thus not unique to arguments in democratic theory but may be used for or against relatively narrow policies or the broadest world views.'