How much data do we have on civilizations going back that far? I imagine that "progress" was not monotonically up, so there might be peaks of development above

How much data do we have on civilizations going back that far? I imagine that "progress" was not monotonically up, so there might be peaks of development above the trendline that we don't know about because of missing data