The war that wasn’t there

When collecting data, some observations will always be hard to confidently classify in accordance with stated definitions of war, civil conflict, or political violence. This research note draws on the experiences of the Uppsala Conflict Data Program in the last decade in managing such unclear cases. After explaining the difference between unclear and non-cases, I describe the data generating process and how this uncertainty is distributed over time in the data. This exercise reveals that the 1980s may have been more conflict-filled than the 1990s, challenging arguments about the stability of the bipolar global order as well as the sudden ‘rise’ of warfare in the immediate post-Cold War era. The final section suggests different ways that researchers may use existing information regarding unclear cases as a way to conceptualize the nature of civil strife without having to engage in additional data collection.