Prof Przeworski, political scientist born in occupied Poland, scholar of democracy, is keeping a diary.

Over the years, I developed a theory of the conditions under which democracies process whatever conflicts that arise in society in liberty and peace. Indeed, my name is associated with one sentence I wrote some 35 years ago, namely that “democracy is when parties lose elections.” The conditions, I thought, required for elections to peacefully process conflicts are that elected governments do not make the electoral defeat too costly to temporary losers, so that they are “moderate,” and that they do not foreclose the possibility of being removed in elections, so that losing is temporary. Elections fail to maintain peace when they generate revolutionary transformations and, as the absence of precedents indicates, they never do. Unless the government use physical force, that is.

There is also statistical research which shows that democracies survive in countries with high per capita income and countries accustomed to peaceful alternation in office through elections. When I apply this statistical model to the US, with its income and its past 23 partisan alternations in the office of the president, I find that the probability that democracy would die in the US is 1 in 1.8 million country-years, zero.

Hence, neither my analytical nor statistical results equip me to understand the events that unravel hour-by-hour. I just cannot think of either some theoretical framework or of historical precedents that could serve to form expectations about what is about to happen. Is democracy dying in the United States?

The whole thing is a fascinating read — an engaged mind wrangling with what we are all seeing; a keen historical eye; deep personal experience.