This chapter presents the analysis of necessary conditions, and shows that strong claims about finding necessary conditions in PB research have been premature. They are likely to have been an artefact of small-N research strategies where deviant cases that reduce evidence for necessity are not as obvious as they become when comparing the full range of evidence across research studies. There is quite a lot of evidence to show meaningful citizen control occurs without autonomous civil society demand and organizing, bureaucratic support for programmes, and strong financial resources respectively. That is not to say that case researchers were not on the right track with their claims, only that these claims need some further disambiguation to better identify the scope or context in which a condition becomes necessary in combination with other conditions. This is a crucial finding because it suggests that there is no context that is universally and unambiguously unable to produce some good PB outcomes, and therefore there is no place where PB will always fail (at least in the short term). Nevertheless, there is strong evidence that political leaders’ commitment to participation is almost always necessary for success.

A researcher's task in presenting research is to distil excessive complexity and explain social phenomena with appropriate levels of parsimony. To this end researchers often construct typologies, and make claims about necessary and sufficient conditions based on their observations. Brian Wampler applies a familiar controlled approach in his distinguished comparison of eight cases of PB in urban Brazil (2007a). This is a useful contribution for scholars who wish to explain the emergence of emancipatory participatory governance through democratic innovation. It has been commended not only for its attentive and detailed narrative of processes within cases, but also its comparative design that moved beyond best-case examples to include variations in outcome and explanatory conditions to explain causation more systematically. In common with other examples in the emerging field of research on participatory democratic innovations, comparative work up to now has relied on more traditional small-N comparative designs. Therefore, in this final part of the book I often introduce stages of the research process by presenting a version of what a QCA replication of that study might look like. I then present wider analysis that accumulates more cases across studies.