intermediary representative strata (Lipow and Seyd 1996 ; Seyd 1999). Nevertheless, for all the manipulative potential of Mair's paradox of democratization as emasculation (of sub-leadership groups), it would be foolish to regard the mass party as some kind of paragon of democratic virtue; Michels pointed out as much early in the twentieth century, and few who witnessed the operation of the British Labour Party's system of delegatory democracy could have deluded themselves in this respect. In any case, if we are to take the notion of cognitive mobilization seriously (see Chapter 1 in this volume), we must give more credit to the potential and actual role played by the ranks of armchair members of modern parties. Many of them may be inactive when it comes to the mundane virtues of attending party meetings and running campaigns, yet they are better placed than most of their predecessors in terms of educational experience and access to independent political information. There are good reasons, therefore, to suppose that they possess the capacity to make informed and rational judgements about matters of candidate and leadership selection, and even party policy. Seen in this light, even the most strategically calculating electoralist leaders may find themselves operating in an authentically democratic context of sorts. At the very least, it would be surprising indeed if party elites generally managed to assume and maintain perfect control of their organizations in the top-down manner which the 'paradox of democratization' interpretation seems to imply. In short, there are still good reasons for supposing that political parties remain vital, if imperfect, mechanisms of linkage in modern democratic societies.