And yet, it's an abiding ironies of seismicity that the largest earthquakes ever recorded in the continental United States took place not in California but near the small town of New Madrid, in the Missouri territory, between Dec. 16, 1811 and Feb. 7, 1812.
These quakes are fascinating both because of their size and frequency — three temblors, all between 7.0 and 8.1 in magnitude, in less than two months — and also because they are what University of Massachusetts professor Conevery Bolton Valencius describes in “The Lost History of the New Madrid Earthquakes” (
"Earthquakes can occur for many reasons besides crunching tectonic plates," Valencius writes: "small local tremors are frequently produced by deep drilling, for instance, and significant seismicity may be attributable to the 'glacial rebound' of continents relaxing back upward after being relieved of their heavy load of glacial ice. …Yet none of the current forms of explanation alone would seem to account for the scale of events that occurred in the Mississippi Valley in 1811-12."
It's not that intraplate quakes are unknown to us; two of history's worst tremors — "the 1556 earthquake in Shaanxi Province, [China], which buried at least 830,000, and the July 1976 magnitude 7.7 earthquake, which leveled the northeast China city of Tangshan, with an estimated death toll of 650,000" — were such events. But what they highlight is that earthquakes are strange, elusive, that they defy our expectations even as they conform to a pattern of some kind.
It's a seismological paradox, that even when we think we know what's coming, we never know what's coming. New Madrid is a case in point, lost to history, regarded as anomalous, if at all.
As Valencius points out, these quakes had far-ranging effects — on settlement patterns, among other things — even though they remain essentially lost to history. As to why that is, it's because they happened so early, and in so rural a region, that the most outrageous stories of the earthquakes (that they made the Mississippi River run backward briefly, for instance) have come to be seen as tall tales.
All of that raises what is perhaps the key problem of earthquake history: how to research past events when the physical landscape has changed. New Madrid was eventually denied altogether ("at an 1883 meeting of the American Association for the Advancement of Science, speaker James MacFarlane argued that stories of a midcontinent earthquake were mere legend"), as the railroad and timber industries fundamentally changed the terrain.
This is the risk in California also, where development has erased the traces of our seismic history, where we build too close to fault lines and often fail to adequately retrofit. In some sense, it has to do with what Norman Klein has called "our history of forgetting," but when it comes to earthquakes, forgetting is the one thing that we must not do.
"In many ways," Valencius writes, "this book has sought to remember. Not simply to remember these particular earthquakes, though they are interesting and may hold cautionary lessons for the future. Rather, this book has sought to remember how knowledge changes. … Sometimes even cracks running across the earth in front of us can go unseen because they fail to fit a current model. Across many sciences, events without apparent explanation can at times go ignored. We commit such unacknowledged forgetting at our peril."