The “end of history” thesis holds that liberal democracy is the final form of political organization — that once a country reaches it, the dynamics stop. The thesis is teleological: history is a process with a destination. Countries that haven't arrived yet are in transit; countries that have are stable. The prediction is that the basin of attraction around liberal democracy is deep, broad, and permanent.
Uhlig, Pirker-Diaz, Wilson, Metzler, and Wiesner (arXiv:2603.05994, March 2026) analyze the Varieties of Democracy dataset — quantitative regime scores for countries over decades — and find the dynamics are weakly non-ergodic. Countries do not converge to a fixed point. They wander in a two-dimensional regime space with sparse basins of stability that themselves shift over time.
The step sizes — how much a country's regime score changes from year to year — follow a heavy-tailed distribution near critical conditions. Most years, most countries change little. Occasionally, a country undergoes a large transition. The heavy tail means these large transitions are not outliers; they are characteristic of the process. The mean step size may not even converge — the distribution has enough probability mass in its tail that averaging over more data does not stabilize the estimate.
Sojourn times — how long a country stays near a given regime configuration — are also heavy-tailed. Some countries remain stable for decades; others flip rapidly. The distribution of waiting times does not have a characteristic scale. There is no typical duration of political stability.
A continuous-time random walk model reproduces the observed dynamics over three decades with high accuracy. The model does not need a destination. It needs only a landscape with shallow, mobile basins and a step-size distribution with a heavy tail. The walk is not directed toward democracy or any other attractor. The basins exist, countries fall into them, but the basins move — what was stable thirty years ago is not necessarily stable now.
The finding is not that democracy is fragile. It is that the concept of a permanent political equilibrium is inconsistent with the observed statistics. The dynamics are not converging. They are wandering, with the landscape itself evolving under the walkers' feet.
Uhlig, Pirker-Diaz, Wilson, Metzler, and Wiesner, "Critical dynamics govern the evolution of political regimes," arXiv:2603.05994 (March 2026).