There is nothing peaceable about the Atlantic, as my father, who went back and forth across it in wartime convoys, could attest. Its inexhaustible palette of moods includes all manner of awfulness. There is, though, something happily pacific, and distinctly un-Pacific, about the shores on to which its tempests blow. The edges of the Pacific are endlessly subject to earthquakes, volcanoes and tsunamis. The edges of the Atlantic, by contrast, are startlingly stable.
Yes, there was the great Lisbon earthquake of 1755, which killed tens of thousands. But that is remembered precisely because it was so exceptional, and had profound effects. The levelling of Lisbon changed not just Portugal’s history but Europe’s ways of thinking about good and evil. In Japan you can expect such things in every lifetime. And until about 50 years ago, this global asymmetry was both inexplicable and more or less unquestioned.
In the 1960s, a few geologists from both sides of the Atlantic worked out how the ocean that separated them – and all the other ocean basins – had come to be. Their plate-tectonic revolution, like all scientific revolutions, did more than just answer outstanding questions. It realigned them, turning apparently separate questions into aspects of the same problem.
At the heart of plate tectonics was the idea that the Earth’s crust is forever being created and destroyed. It is created at the ridges that run through all of the oceans. It is destroyed in subduction zones at the edges of oceans, where it plunges back down into the depths. Both ridges and subduction zones, the revolutionaries told the world, were boundaries between things no one had previously imagined called tectonic plates, the former being where plates pull away from one another, the latter where they converge.
Plate tectonics thus explained a range of things which had seemed to have nothing in common. In the 1950s measurements of the magnetic fields in rocks had shown that the continents had once been at different latitudes; this was explained by the continental drift brought about by the plates. Half a century of seismology had shown that small earthquakes clustered along ocean ridges and big ones at some ocean margins; this was explained by the stresses the crust underwent as it was stretched and compressed.
The most ancient of the puzzles solved was the way the east coast of the Americas echoes the west coast of Europe and Africa. It had been the subject of speculation in the 18th century by the natural historian Alexander Humboldt, who saw the Atlantic as something like a winding river valley, and it had even been noted in the 16th by the philosopher Francis Bacon. Plate tectonics showed it to be the result of the sundering of an ancient supercontinent.
In 1964 the Cambridge geophysicist Teddy Bullard put these observations on a firmer footing. Using computers and spherical geometry, he showed exactly how the zigs of the east and the zags of the west could nestle against each other if the ocean were removed from the scene and the continents snuggled up together.
Since then the history of the break-up has been mapped with some precision. About 200m years ago, what is now Mauritania and what is now the eastern seaboard of the United States began to pull apart; as they did so, new crust formed at the nascent mid-Atlantic ridge between them. The process has continued, albeit with fits and starts, ever since, widening the Atlantic a few centimetres a year – roughly the pace at which fingernails grow. And because crust is being made at the ridge, not destroyed at the edges, the basin has little earthquakes at its heart but none on its rim.
This is not a sustainable arrangement. Even at fingernail speed, the Atlantic cannot grow for ever on a finite planet. There are, though, two different ways it could end. Either the Atlantic keeps on widening, the Pacific shrinking, until America pushes up against Asia, or the Atlantic starts to develop subduction zones of its own, which eat up old ocean floors faster than the central ridge can produce new ones; the ocean narrows and eventually collapses. The process may already have begun: the fault that produced the Lisbon quake could be the start of a subduction zone.
Like all the best scientific revolutions, plate tectonics did not only answer old questions, but posed new ones. It does not promise to resolve them – I doubt anyone will ever be able to say with certainty whether the Atlantic will keep opening or start closing. But whichever course it takes matters not a whit in human terms. Just as there are some questions that no one bothers to answer, so there are others which no one needs to answer.
Some questions fascinate because they can be answered, others because they can’t be. Knowing how the Atlantic arose falls into the first category; an origin that can, in principle, be unearthed or explained is exciting, one that is for some reason necessarily obscure or unobtainable is by and large not. Questions about where things are headed tend to fall into the second. Clear answers about future events may have practical value, but they have little else to recommend them. The undecidable future, though? That never palls, any more than a seascape can exhaust itself.