"laws in the background, contingency in the details," and striking the best attainable balance between the two. They also include the vital role of consilience in minimizing heteronomy in the field (i.e., believing just what one pleases), and the equally vital role played in this respect by periodization.
Without a doubt, a major contributing factor to the lack-of-resolution problems we have been discussing is that there has tended to be only one research community organized as scientific research communities typically are -- the one constructed with such brilliant results as the Michigan survey-research group. Otherwise, we have only the cottage industry, coupled with the professional blinders that arise from the division of scholarly labor across more than one discipline. This has been a mighty brake on progress, but my ultimate sensation is not one of complaint but of steadily growing optimism. In recent years especially, there has been a considerable growth of opportunities for consilience to manifest itself more effectively as workers in previously partitioned research communities have come to be aware of each other. And there is a genuine desire for this: A few years ago, for instance, Terry Moe issued a call for people doing "positive theory of institutions" and scholars in the field of historically grounded, state-centered studies to move into interaction, to the benefit of both communities ( Moe, 1987:236-299). This call is clearly evoking a response, and it is easy to predict that it will grow in size and payoffs over time.
We seem to be at a threshold: An era of creative integration across these and other fields lies just ahead as people with convergent research problematics begin to find and react to each other. Much wheat will be winnowed along the way from a much larger supply of chaff. If, in this process, we learn enough to come to an agreement that critical-realignment theory should now be given a decent burial -- that would be progress, however unlikely I think such an outcome will be. In the meantime, this chapter, like Chapter 2, is an effort to point out why the supposed death of realignment has been prematurely announced. It is also an effort to suggest why and how generally better work with historical data in our field might be performed than has been the case hitherto.