Why ceasefires fail — three patterns from the data

The Uppsala Conflict Data Program has tracked the outcome of every ceasefire in armed conflicts since 1989. The summary statistic is striking: roughly two-thirds collapse within 18 months. The interesting question is why. Three patterns recur.

Pattern 1: No verification mechanism

Ceasefires without independent monitoring (UN, OSCE, regional body, or trusted civil society) almost always break down. Each side accuses the other of violations, and without a neutral arbiter, accusation becomes pretext. The 1993 ceasefire in Angola, the early-stage Yemen ceasefires, the 2014 Minsk I agreement: all suffered from weak or absent verification.

Pattern 2: No political track in parallel

A ceasefire is only a pause in fighting. Without a parallel political negotiation tackling root grievances, both sides use the pause to re-arm and re-position. Strong cases — the Colombian peace process, the Belfast Good Friday agreement — combined ceasefire with substantive political talks from day one.

Pattern 3: Excluded spoilers

If significant armed groups are not part of the agreement, they have a direct incentive to sabotage it (otherwise they become irrelevant). Hard-line factions in Israel-Palestine, dissident republican groups in Northern Ireland, splinter FARC dissidents in Colombia: each became a long-running risk because they were not at the table.

What this means practically

Designers of future ceasefires can borrow a checklist: independent verification, parallel political process, and inclusion of all significant armed groups (or a credible plan to neutralize spoilers). It does not guarantee success, but it shifts the odds significantly.