I have been working on this book for so long that I dislike to admit it, even to myself. Looking back I discover that I began lecturing on parties and party systems at the beginning of the 1960s. The original impulse was my unhappiness with Duverger's pioneering volume, which appeared in 1951. Duverger's everlasting merit was to seize a topic in want of a general theory. By 1963, however, David Apter correctly asserted, in surveying the theme, that [what is lacking is a theory of political parties.] In struggling with this manuscript I have since discovered that this lack has not been remedied – indeed, it has steadily grown.
A first draft was completed in 1967 at Yale and was circulated during that year. A number of friends quoted the 1967 draft believing, as I did at the time, that the book was about to appear. I owe them my humble apologies. The draft was more theoretically cumbersome than today's book, and its design was very much concerned with systems theory and with furthering structural-functional analysis. This is not to say that I have since dropped my theoretical ambitions. The difference is that in 1967 whatever I had to say was contained in one volume – whereas I am now publishing the first of two volumes. And my theoretical thread emerges as I go along, that is, especially in the second volume.
Since the rest is still to come, I ought perhaps to mention here where I stand with respect to systems theory. As the complaint correctly goes, whole systems analysis, on the one hand, and empirical research and findings, on the other, fall widely apart. Presumably, however, one way of bridging this gulf is to develop the in-between level, namely, subsystem or partial system analysis. I equally take it that the party system, and more precisely the party subsystem, is crucial to this end. Parties are the central intermediate and intermediary structure between society and government. Furthermore, insofar as they are a system, parties interact and such interactions can be viewed as mechanical propensities, as structures of rewards and opportunities that go a long way toward explaining the different performances of different types of party polities. Finally, I assume politics to be an independent variable, thus implying that parties and party systems mould (beyond the point at which they reflect) the political society. That is, before treating politics as a dependent variable, it behooves the political scientist to explore how much mileage is afforded by its autonomy.
These and other theoretical ambitions emerge in the second volume. But why