The paper in question is Peter Keen's "Information Systems and Organizational Change". This is considered a classic paper in software engineering, I'm studying it right now for a retrospective on a project that has been difficult. The paper is about why the implementation of MIS (management information science) systems can be difficult from a sociological and political science perspective.
The passage I'm confused by is as follows (bold emphasis mine):
4. Pluralism: The Need to Mobilize
Political science views organizations mainly as groups of actors, often with conflicting priorities, objectives, and values. (Allison ) The management literature generally assumes far more commonality of purpose. The Down- and-Out approach relies on this. Up-and-In evades the problem by limiting the scope of the project and hence the number of actors involved; it fails completely if consensus is not impossible. The more the organization is viewed as a set of loosely coupled units (Weick ) where joint action rests on negotiations (Strauss ), the more any strategy for implementation must emphasize the need to mobilize coalitions, to provide the necessary support for an innovative proposal. Obviously that process is based on political rather than economic rationality. The corollary of this argument is that lack of attention to the constraints on change imposed by pluralism in organizations will result in failure.
Everything makes sense except the bolded part. It seems, if we simplify the double-negative, that he means to say that if consensus is possible between the pluralistic groups, then the Up-and-In approach will fail. I do not see why this is so. It does not seem self-evident from my careful reading of the text why this would be true.
Here is a link to an online copy of the paper (I hope this is allowed): http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.83.3441&rep=rep1&type=pdf