The invasion of Normandy did not begin on a beach. It began sixteen minutes after midnight, in a field, in silence.
Six wooden gliders, towed across the Channel by bombers and then cast off to come down without engines, landed beside two bridges over a canal and a river at the eastern end of the invasion area.
The lead glider stopped forty-seven yards from its bridge. The pilot, Jim Wallwork, had flown the whole approach on a stopwatch and a compass, at night, in a machine made of plywood, and he went through the cockpit windscreen on landing and was the first Allied soldier to touch Normandy, unconscious.
The bridges were taken in about ten minutes. Holding them mattered because they were the only crossings on that flank, and whoever held them controlled whether German armour could reach the beaches from the east.
It is genuinely difficult to convey the scale, so here are some numbers.
Nearly seven thousand vessels crossed the Channel that night. Around twelve thousand aircraft flew in support. Some hundred and fifty-six thousand men landed on the first day, by sea and by air.
Behind that, in England, were the depots that had been filling for two years. There were so many American troops in the south of England that a joke went round that the only thing keeping the island from sinking was the barrage balloons.
Five beaches were used, along about fifty miles of coast. From west to east they were Utah and Omaha, which were American; Gold, which was British; Juno, which was Canadian; and Sword, which was British.
Every man going ashore that morning had been issued, among other things, a paper bag for seasickness, a French phrasebook, and about two hundred francs of specially printed invasion currency.
And behind the beaches, from about half past one in the morning, three airborne divisions were coming down in the dark to hold the ends of the line: two American in the west, one British in the east.
The airborne landings went badly wrong, and then that turned out to help.
The transport aircraft ran into heavy cloud and anti-aircraft fire, formations broke up, and paratroopers were dropped over an area many times larger than intended. Men landed miles from their objectives, alone or in twos, in fields and hedgerows and flooded meadows, in the dark, with no idea where they were.
Some drowned in ground the Germans had deliberately flooded. Others spent the night walking, using a small tin cricket clicker they had been issued as a recognition signal: one click, answered by two.
What they did, in almost every case, was gather up whoever they found and attack whatever was nearby. Roads were cut, telephone lines pulled down, crossroads seized by improvised groups of men from three different regiments who had never met.
And the German command, receiving reports of paratroopers from everywhere at once, across a hundred miles, could not work out where the actual attack was. A drop that had gone wrong looked, from the other side, like a drop of enormous size and unreadable purpose.
The seaborne landings began at half past six in the morning in the west and about an hour later in the east, because the tide runs along that coast and low water arrives at different times.
On four of the five beaches it went, by the standards of what had been feared, well. Not easily and not without loss, but the defences were suppressed, the specialised armour got ashore, and by the afternoon troops were moving inland.
On Omaha it very nearly failed.
Omaha is a curved beach overlooked along its whole length by bluffs a hundred feet high, and the defences there were stronger and better manned than intelligence had indicated. The preparatory bombing had gone long and landed inland. Most of the amphibious tanks meant to swim ashore were launched too far out in a heavy sea and sank.
So the first infantry arrived on an open beach, against undamaged positions, without armour. For a couple of hours the situation was bad enough that the American commander offshore considered evacuating it.
What changed it was not a plan. It was small groups of men, junior officers and sergeants and privates, working along the sea wall, finding gaps in the wire, and going up the bluffs where there was no path, in ones and twos, until the positions above were taken from behind.
Meanwhile, the German response was slower than it should have been, and the reasons are worth listing because none of them were accidents.
Rommel was in Germany. Several senior commanders were away at a map exercise. The weather had convinced everybody that nothing was likely for a fortnight.
The armoured reserve, which was the one thing that could have thrown the landings back into the sea, could not be moved without an order from Hitler's headquarters, and Hitler was asleep, and his staff would not wake him. The order releasing two panzer divisions came in the middle of the afternoon.
And the deception held. German intelligence continued to believe that Normandy was a feint and that the main landing would come at the Pas de Calais, and formations that could have moved west were kept in place for weeks, waiting for an invasion by an army that did not exist.
Reconnaissance also failed almost completely. Only two German fighters made a strafing run along the beaches all day, out of an air force that had once controlled the sky over France.
By the evening of the sixth of June, all five beaches were held, though none of them had reached the objectives set for the first day, and two of them were not yet joined up.
Allied casualties for the day are usually put at around ten thousand, of whom something over four thousand were killed. The figure was far lower than the planners had feared, and it is a large number of people.
The town of Caen, which Montgomery's plan had optimistically listed as a first-day objective, was not taken for another six weeks.
Nobody standing on those beaches that evening knew whether it had worked. A beachhead can be pushed back into the sea, and that had happened to Allied landings before. What made it permanent was not the sixth of June but the following fortnight, when the build-up over the beaches proved faster than the German build-up by road, because the roads were being attacked from the air.
There is one more thing that belongs in the day. The Norwegian, Polish, Free French, Dutch, Greek and Czechoslovak crews and units in that fleet are usually folded into the word Allied and lost there. Some of the French troops who came ashore at Sword had last seen France in nineteen forty, on the way out through Dunkirk.
The campaign in Normandy took until the end of August and was far bloodier, week for week, than most of what came before it.
The D in D-Day does not stand for anything interesting. In military planning, D-Day means simply the day an operation begins, and H-Hour the hour, so that everything else can be scheduled relative to them before the date is known. There were D-Days for many operations. This one took the name over.
The proper name for the landings was Operation Neptune, which was the assault phase of Operation Overlord, the wider campaign.
Jim Wallwork, the glider pilot, survived the war and lived until two thousand and thirteen. Asked repeatedly about the accuracy of that landing, he said that the second glider was closer.
Here is something to think about tonight. On Omaha, the plan collapsed within twenty minutes and the beach was taken anyway, by people improvising with what was in front of them. What is the point of making a detailed plan if the plan is going to stop being true almost immediately?