I really enjoyed reading Collier and Adcock because their article is very structured, well-argued and most of all it finally enters into the realm of what we are concretely going to do with a definition of democracy. Not that I don’t like theoretical debates but , as underlined by the authors, “how scholars understand and operationalize a concept can and should depend in part on what they are going to do with it”( 537). Therefore I really liked their idea of seeing the definition of democracy, not only as a pleasurable philosophical struggle between ideas and academic’s egos, but also as a tool to test causal relationships.
In their article, Collier and Adcock assess the definitions of different authors to see the underlying justifications for their choice between dichotomous and gradated definitions. Then, they criticize those justifications by clearly balancing advantages and disadvantages, arriving to the conclusion that it is often arguable to use both dichotomous and gradated definitions depending on the aim of the research. Moreover, they emphasize the importance of limiting the study of democracy in a certain time and space to avoid incomparable comparisons. Justifications must be as specific as possible and built into the framing of the research question.
Near the end of the article they develop the idea of bounded wholes, which I found very interesting. Starting from Sartori’s definition of a bounded whole : ” a system constituted by multiple attributes, all of which must presumably be present for a case to be classified as an instance of the concept” (543), they focus on the conceptual interaction among the defining attributes of democracy.
It is very interesting to see that for example, the absence of civil liberties cancels the interpretation of the other attributes as being democratic, whereas decree powers do not. Therefore, defining attribute do not all have the same importance in a definition. Some are crucial and some are less crucial. This conclusion however, is not a real one, as it brings us back to the starting point : what are the minimal fundamental components of democracy?!
Finally, the only critic that I could easily see arising against their pragmatic approach is the threat of being to relativist because nothing is ever wrong depending on what we are looking for and how we justify it. For example, when they discuss a solution to compare transitions to democracy with a dichotomous classification, their advice is to “compare regimes according to whether they have achieved full democratization in relation to the norms of the relevant time period” (552). What will the result of this be? Democratic under certain conditions of space and time? Is this not a gradation of democracy at the end?
However, I could use their own argument to take their defense. Finding a coherent and theoretical definition of democracy was not the aim of their article. Pragmatism was and I think they achieved it by giving us the tools to choose between dichotomous and gradated definitions. The truth of a definition for them, can be evaluated after the measure not before. It’s true if it works.