Part 1

At 0400 on 23 March 1943, on the high plateau east of El Guettar in Tunisia, Friedrich Freiherr von Broich sat in his command car and listened to the sound of his tanks rolling into position. He had 50 serviceable panzers, about 6,000 men, a company of tank destroyers, and an assault-gun battery. The ground in front of him was open. The air before dawn was cold and still. There was every reason, as far as he could see, to believe that the next few hours would be a rout.

He had done this before. 5 weeks earlier, in February, German forces had torn through American positions in the battles that culminated at Kasserine Pass. They had destroyed 183 American tanks. They had captured more than 3,000 American soldiers. They had inflicted 6,000 casualties on a force that had seemed to run at almost every turn. Erwin Rommel had described the Americans as inexperienced amateurs. Albert Kesselring had expected the campaign to end with Tunis in German hands. Senior British officers, including Bernard Montgomery and Harold Alexander, had concluded privately that the Americans were hopelessly trained and badly led. British soldiers at the front had taken to calling them “our Italians.”

Now, at El Guettar, von Broich was looking across open ground at the same American formation he had beaten 5 weeks earlier: the 1st Infantry Division, the Big Red One, Terry Allen’s men. He expected the same result.

What he got was something else. By 1000, as the sun climbed over the Tunisian plateau, his tanks had run into a minefield that had not been there 5 weeks before. The panzers slowed to clear it and came under fire from carefully positioned antitank guns. American tank destroyers of the 601st Tank Destroyer Battalion engaged them from positions his reconnaissance had missed. American artillery fell on the armor with a concentration and coordination his scouts had not detected. Within roughly 1 hour, 30 of his 50 tanks were burning in the open. By midday, the 10th Panzer Division was retreating from a valley it had expected to own by lunch.

Von Broich did not understand what had happened. Captured documents confirmed that this was the same division, the same numbered regiments, many of the same officers. The Big Red One he had beaten at Kasserine and the Big Red One he faced at El Guettar wore the same patch, carried the same unit histories, and still had many of the same names on the rolls. Yet it was not the same army.

This would happen again in Sicily, in Italy, in Normandy, and in the Ardennes. The pattern became so common that, in postwar interrogations, German officers kept returning to a single word to describe it: incomprehensible. They had beaten an American unit one week and then faced what they swore was a different army the next. They could not understand where that speed of adaptation came from. They could not locate its source. They could not, despite every effort, reproduce it.

What von Broich was witnessing that morning was not luck. It was not an accident. It was the first visible result of a system that had been quietly built inside the United States Army over the previous 15 years, a system the Germans did not know existed, a system that, when it reached full speed in 1944, would turn American infantry divisions into the fastest-learning fighting organizations in modern military history. To understand why von Broich’s tanks burned at El Guettar, and why Fritz Bayerlein’s Panzer Lehr Division would be shattered in Normandy 18 months later, it is necessary to begin with something the Germans had that the Americans did not, and then to look at what the Americans built that the Germans could never copy.

For most of the 20th century, any serious student of war would have given the same answer to the question of which army in the world was best at tactics. Not the Americans. Not the British. The Germans.

This was not propaganda. It was professional consensus. Since the wars of Prussian unification in the 1860s, the German Army had refined a tradition of tactical excellence that every other modern military studied. It had a doctrine of Auftragstaktik, mission-type tactics, which gave subordinate commanders the authority to adapt their methods to achieve their commander’s intent without waiting for orders. It had the best noncommissioned officer corps in the world. Its field regulations were translated and studied at West Point, at Sandhurst, and at Saint-Cyr. Between 1939 and 1941, the Wehrmacht had defeated Poland in 5 weeks, France and the Low Countries in 6 weeks, and the Balkans in under a month. Entire armies had surrendered to German combined-arms operations that seemed to move at a speed their opponents could not process.

The United States Army, by contrast, was barely an army at all. In September 1939, when Germany invaded Poland, it numbered fewer than 200,000 men. It ranked below Portugal. Its infantry still drilled with equipment from the First World War. Its tactical doctrine, set out in field manuals that had not been meaningfully revised in 2 decades, remained rooted in the trench warfare of 1918. Its armor doctrine assumed that tanks would never fight other tanks. Air-ground coordination was, in the words of one observer, essentially nonexistent.

When George C. Marshall became Chief of Staff in September 1939, he inherited a force that many European officers privately regarded as a joke. In 1941, the Army staged the Louisiana Maneuvers, the largest peacetime maneuvers in its history, involving nearly half a million men. The result was a mess. Combined-arms coordination failed at every level. Tank units lost contact with infantry. Artillery fell out of range. Radio discipline collapsed. Generals who looked strong on paper crumpled under the strain of moving large formations across difficult ground. The maneuvers did not prove that the United States Army was ready for war. They proved the opposite.

Yet something happened during those maneuvers that mattered more than anyone understood at the time. Marshall watched. So did his key subordinates. What they drew from Louisiana was not simply a set of tactical lessons. It was a conclusion about process. The United States Army, they decided, could not afford to train the way the old European armies trained. It did not possess a long institutional memory. It did not have a century-old tradition of tactical excellence behind it. What it could build, and what it would have to build, was a mechanism for learning faster than everyone else: a system that could take the lessons of one unit in one battle and feed them into every other unit before the next battle began.

In 1941, that remained only an idea. In November 1942, it faced its first real test when American troops landed in North Africa during Operation Torch. The early results were exactly what the Germans expected. By the time American forces reached Tunisia in early 1943, the weaknesses Marshall had seen in Louisiana were exposed in full. American armor was dispersed instead of concentrated. Antitank doctrine revolved around a weapon category, the tank destroyer, that assumed the enemy would attack at predictable points. Infantry and tanks rarely worked together properly. Air support was slow and poorly coordinated. At the top of the American command system in Tunisia sat Lloyd Fredendall, a man who spent more time in his underground bunker than at the front, issuing orders in a private slang that even his own subordinates sometimes struggled to decode.

On 14 February 1943, at Sidi Bou Zid, the Germans attacked. Combat Command A of the 1st Armored Division was committed piecemeal. German antitank guns and panzers destroyed it in sequence. The Americans lost more than 160 tanks in a single day. By 19 February, at Kasserine Pass itself, American positions had collapsed. By 22 February, they had lost more ground in a week than they would lose in any single week of the rest of the war, including the Battle of the Bulge. Rommel reviewed the performance and concluded, as he later wrote, that the Americans had not yet learned their lessons. The British commanders around him agreed. The German commanders agreed.

There was one man in a senior position who looked at Kasserine and drew a different conclusion. Dwight D. Eisenhower, writing privately to Marshall within days of the disaster, said, “All of our people, from the very highest to the very lowest, have learned that this is not a child’s game.” A few days later he wrote again: “We’re learning something every day, and in general, do not make the same mistake twice.”

The Germans heard none of this. They had no reason to believe an American general’s private confidence when their own eyes had shown them an American force in retreat. Their historical experience told them that green armies recovered slowly, if they recovered at all. The British had required nearly 3 years of defeats before they learned to beat the Germans. The Russians had spent 18 months paying catastrophic prices for their education. The Americans, they assumed, would pay the same price for the same lessons over the same span of time.

That was the German assumption on 23 February 1943. It was wrong by 23 March.

What happened in those 35 days was not magic. It was machinery, a machinery of learning that had been quietly assembled in the United States since the middle of 1942 and that was now running at a speed the Germans did not even know to measure.

It began with paper. On 18 June 1942, the War Department’s Military Intelligence Service issued the first of a series of bulletins called Tactical and Technical Trends. A new issue went out every 2 weeks. They carried reports from American military attachés in every theater of the war, reports from observers embedded with British forces in North Africa and with Soviet forces on the Eastern Front, analyses of captured German equipment, translations of German and Japanese training manuals, and interviews with Allied soldiers who had fought German units. By the end of 1943, 40 bulletins totaling more than 2,000 pages had been pushed through the Army’s training establishments. No other army in the war ran anything quite like it on that scale. The Germans had intelligence summaries, but they were classified and tightly held. The British conducted their own analyses, but lacked the American printing and distribution infrastructure. The Soviets did something similar on a narrower front.

Then came the Combat Lessons pamphlets. The first appeared in early 1944 under the title Combat Lessons, No. 1: Rank and File in Combat, What They Are Doing, How They Do It. It was not a field manual. It was not signed by a general. It was a collection of short, specific, practical observations from American soldiers who had recently been in combat: a sergeant describing how his squad had dealt with a machine-gun nest in Sicily; a tank commander explaining why he had ordered a particular maneuver near Salerno; a forward observer setting out a method for adjusting artillery fire onto moving German armor. 9 issues were published between 1944 and 1945. They were printed in huge runs and distributed down to company level in every theater.

The Germans had nothing like this. Their doctrine assumed that insight belonged to commanders. The idea of asking a sergeant to explain his own tactical innovation, then printing it and sending it to every other sergeant in the army, would have struck a German staff officer as an inversion of how a professional military was supposed to function.

Then came the Army Ground Forces observer system, officers sent out under Leslie J. McNair to ride with combat units, record what worked and what did not, and then fly back to the United States to brief replacement training centers. Reports accumulated on defense against mortars, on Pathfinder teams in Operation Market Garden, on antitank weapons from the 2nd Infantry Division, and on hundreds of other specific problems encountered at the front. Each one fed back into the training pipeline at Fort Benning, Fort Sill, Fort Knox, and Camp Blanding.

Underneath all of this ran the quieter and more constant layer of the system: the after-action report. Every American unit, after every significant engagement, produced one. They were not simply filed and forgotten. They were collected, analyzed, excerpted, and redistributed. Lessons were extracted. Those lessons became the raw material of the next revision in the next training cycle.

The cycle time of this machinery was something no other army had achieved. A battle occurred. Within days, survivors produced a report. Within weeks, observers at higher headquarters had read it. Within a month, its observations might appear in a bulletin. Within 6 to 8 weeks, that bulletin could reach a replacement training center in Texas or Georgia. The curriculum changed. A few weeks later, the next shipment of replacements arrived at the front carrying the lessons of the unit they were replacing.

That was the feedback loop. That was what the Germans could not see and could not copy.

After Kasserine, the machinery moved at once. Within 48 hours of the battle’s end, Eisenhower acted. Fredendall was out. George S. Patton was in. Ernest N. Harmon had already been sent quietly to II Corps headquarters as what Eisenhower called a useful senior assistant, with secret orders to report on what was broken. Omar Bradley arrived to report directly on the command failures and stayed on as deputy corps commander. Laurence S. Kuter was brought forward to help systematize air-ground coordination under procedures that would become standard doctrine.

At the same time, the doctrinal lessons of Kasserine were moving through the American system at a speed the Germans would have found unbelievable. The dispersal of armor into small combat commands was identified as a central error. Antitank positioning doctrine was revised. Minefield employment was tightened. Artillery fire control, already being centralized through Fort Sill’s fire-direction-center system, was pushed closer to its full design potential. Most importantly, Terry Allen’s 1st Infantry Division spent the intervening weeks reviewing unit-level after-action reports and drilling the corrections.

The Germans did not know any of this was happening. They trusted their own intelligence picture, their own assumptions about how fast green armies could improve, and their own faith in the superiority of the German tactical system. Yet by 23 March 1943, when von Broich’s tanks rolled toward El Guettar, the 1st Infantry Division was not the division that had been beaten at Kasserine. Not in doctrine. Not in training. Not in the minds of the men serving the guns. It still wore the same patch and carried the same regimental numbers. It was still commanded by the same divisional commander. But it was already becoming a different army.

Von Broich saw the result. He did not see the cause. The question he could not answer was the question that would haunt every German commander who ran into this pattern afterward. Where, exactly, had that transformation happened? What was the mechanism? And why, despite every effort, could the German Army not build the same thing?

Part 2

The answer can be seen most clearly by looking closely at one division over one 5-week stretch: the 1st Infantry Division between 22 February and 23 March 1943.

By 22 February, when the German offensive at Kasserine finally ran out of steam, the Big Red One had been battered but not destroyed. Combat Team 16 and Combat Team 18 had taken serious casualties. Some of the division’s supporting artillery had been overrun. Confidence had been shaken. Yet the division remained in the field.

Its commander, Major General Terry Allen, was a veteran of the Meuse-Argonne, a polo player, a cavalry officer by training, and a man who swore like a private and drank like a cavalry officer from central casting. His assistant division commander, Brigadier General Theodore Roosevelt Jr., son of the president, had landed at Arzew on 8 November 1942 reciting Kipling from memory aboard the transport. Neither man was going to become someone else in Tunisia. But what they did in those 5 weeks changed decisively.

Within days of Kasserine, the division’s artillery began to reregister batteries under the centralized fire-direction-center doctrine refined at Fort Sill during the 1930s but not fully employed in combat at Kasserine. Gun positions were shifted so they could support one another. Forward observers were pushed down to platoon level, with universal map coordinates drilled until their use became automatic.

Antitank gun positioning was overhauled. At Kasserine, American antitank guns had often been placed too far forward, too visible, and too isolated. The new doctrine, drawn from British experience and from the first hard American lessons, called for guns positioned in depth, sited on likely avenues of approach, and integrated with mutually supporting fires. Minefields, which had not been a serious feature of the Kasserine defense, became a planning priority.

The 601st Tank Destroyer Battalion drilled under conditions intended to simulate the kind of German armored attack that had chewed through American positions at Sidi Bou Zid. They practiced firing from defilade, shifting positions after each engagement, and working under the direction of forward observers.

Air-ground coordination was tightened. At Kasserine, American aircraft had been largely absent from the most critical fighting. Kuter, newly installed on the air side of the command system, helped develop the direct-support procedures that would be used for the rest of the war in Europe. Liaison officers were embedded with ground units. Request procedures were simplified. Response times were cut.

Patton, meanwhile, did what Patton always did. He arrived in the II Corps area on 6 March and began firing men, touring units, finding soldiers without helmets and neckties, and imposing a new standard of order by force of personality. His relationship with Terry Allen was complicated and often contentious, but Patton largely left the 1st Infantry Division alone because he understood what Allen was doing.

At dawn on 23 March, the 10th Panzer Division attacked. German reconnaissance had reported that the American defense was thin. They expected to break through to El Guettar itself by midday. What they encountered instead was a defense built out of every lesson absorbed since 22 February.

They struck a minefield that stopped them cold. American forward observers called in artillery concentrations from gun lines that had already preregistered the likely approach corridors. When German armor tried to work around the minefield, it ran into American tank destroyers firing from defilade, manned by crews that had spent 3 weeks rehearsing precisely this problem.

Within about 1 hour of the opening engagement, 30 of the 10th Panzer Division’s 50 panzers were burning. The 1st Infantry Division lost 22 tanks of its own and took significant casualties, including heavy losses in the 601st Tank Destroyer Battalion. But the German attack broke. By noon, von Broich was pulling back.

Rommel, already reassigned and preparing to leave Africa for good, later reviewed the reports and wrote, “The Americans had fought brilliantly.” He added, with a generosity not always associated with him, that the Americans had made up for their lack of experience with far better and more plentiful equipment and with their tactically more flexible command.

That phrase mattered. Rommel had noticed, without fully being able to name it, that the Americans at El Guettar were now operating at a tempo they had not reached at Kasserine. Decisions were being made faster. Units were coordinating more smoothly. The gap between a report arriving at a command post and a response going out to the field had narrowed. He could see the effect. He did not understand the machinery producing it.

What the Germans could not grasp was that the 1st Infantry Division at El Guettar was not better because it had received better men or substantially better equipment. Its rosters were largely unchanged. On paper, its tanks were not superior to the panzers facing them. What had changed was the volume and speed of information flowing through its system. The after-action reports from Sidi Bou Zid had been absorbed. The doctrinal lessons from Kasserine had been drilled. Observations from British officers who had spent 2 years fighting Rommel had been passed down and practiced. At El Guettar, for the first time in the war, the American learning machinery worked in combat at full visible effect.

It would not be the last time.

Hans von Luck, who had fought in France, North Africa, and Russia, later wrote that the Americans adapted immediately to a changed situation and fought with great doggedness. Friedrich von Mellenthin, chief of staff of the 48th Panzer Corps in Russia and later a division commander in the West, described it in harsher language after the war. Speaking of the Americans, and not as a compliment, he said, “They are masses and we are individuals.”

He was describing something real, but he drew the wrong conclusion from it. The masses he spoke of were learning collectively through a system that pumped lessons from each individual soldier into every other individual soldier. The individuals he admired were increasingly forced to learn alone, because German doctrine and German political culture no longer allowed honest reporting to move freely from the bottom upward.

By the time Mellenthin made that remark, the American system had already bloodied the Wehrmacht’s best in North Africa. In Normandy, it would do something even more startling. What had taken 35 days between Kasserine and El Guettar would, in the hedgerows of France, happen in 11.

In June 1944, the largest amphibious operation in history put almost 9 divisions of American infantry and armor ashore on the Normandy coast during the first week of Operation Overlord. Almost immediately, those divisions collided with a terrain problem that nobody had seriously planned for: the bocage.

For centuries, Norman farmers had enclosed their fields with earthen banks topped by dense hedgerows. Some banks rose 10 feet high. Some hedges formed walls of thorn and root so thick that they were nearly impenetrable. The fields between them were often only a few hundred yards across. The Germans had moved into this country with a defensive doctrine that exploited every yard of it. Every hedgerow could conceal a machine-gun position. Every corner could hide an antitank ambush. Every field could become a killing ground.

American tanks could not get through. When they tried to ride over the hedges, their hulls tilted upward, exposing their thinly armored bellies to German Panzerfausts and antitank guns. Their own main guns, instead of covering the enemy, pointed uselessly at the sky. Infantry casualties, as Americans crossed field after field under observed fire, rose to levels that alarmed the high command.

Panzer Lehr Division, under Fritz Bayerlein, was the best-equipped armored division the Germans possessed in the West. It stood in position to block the American breakout in the Saint-Lô sector. Bayerlein, who had served as Rommel’s chief of staff in Africa, knew the terrain and believed he knew his enemy. He expected the Americans to grind slowly through the bocage for weeks, taking heavy casualties.

The answer that came did not come from a general or from a staff officer. It came from a conversation in a bivouac of the 102nd Cavalry Reconnaissance Squadron, where a group of tankers were discussing the hedgerow problem. According to the account later given by Max Hastings, one of the men, an enlisted soldier identified in the surviving record as a Tennessee hillbilly named Roberts, offered a suggestion. Why not put saw teeth on the front of the tank and cut straight through the hedges?

Most of the men laughed. One of them did not.

Sergeant Curtis G. Culin III, 29 years old, from Cranford, New Jersey, serving with the 102nd Cavalry Reconnaissance Squadron of the 2nd Armored Division, went looking for material. He found scrap steel from German and Czech hedgehog beach obstacles scattered across Normandy to stop Allied landings. He welded 4 tusk-like prongs onto the front of a light tank and tested it against a hedgerow.

It worked.

On 14 July 1944, Major General Leonard T. Gerow, commanding V Corps, came to watch. Omar Bradley arrived soon afterward. Bradley later described what he saw. A tank fitted with what looked like a crossbar carrying 4 tusk-like prongs backed away, accelerated to about 10 miles per hour, and crashed through a hedgerow. A Sherman modified the same way repeated the performance. Bradley wrote that this was “a soldier story, so absurdly simple that it had baffled an army for more than 5 weeks.”

The device had been made by Curtis G. Culin III. Within hours, Lieutenant Colonel J. B. Medaris, the corps ordnance officer, was on the radio to every ordnance unit in First Army. Production was ordered around the clock. Scrap metal was collected from wrecked beach obstacles along the invasion coast. Welding crews worked through the night.

The clock began on 14 July. By 25 July, roughly 500 of the devices had been built and installed. 3 out of every 5 American tanks in First Army now carried what the troops were calling rhinos. In 11 days, an idea voiced in camp had become equipment across a field army.

Operation Cobra began on the morning of 25 July 1944. Nearly 1,800 American heavy bombers dropped an immense carpet of explosives on Panzer Lehr’s positions. Then American armor went forward with Culin’s devices cutting openings through hedgerows that Bayerlein had believed would channel the attackers into German kill zones.

Within hours, Panzer Lehr had lost most of its combat effectiveness. Within 2 days, the German defense in that sector had ruptured completely. Bayerlein’s own postwar description of the bombardment was simply that it was hell. He wrote of a front line that looked like the surface of the moon. He estimated his personnel losses at roughly 70 percent dead, wounded, crazed, or numb. Then he added the sentence that touched the exact nerve of the problem. He had gone into the fight certain that the Americans could not break the bocage defense quickly. He had expected the terrain to hold them for weeks. Instead, they had arrived at his line with a solution he had never seen before, and they had arrived with it fitted to 3 out of every 5 tanks.

Bayerlein had been with Rommel in Africa. He had seen how long it took the British to learn how to beat the Germans. Now, in 11 days, he had watched an American army identify a terrain problem, invent a solution at the enlisted level, validate it, scale it across a field army, and hurl it against him. He did not understand how that was possible.

For the second time in this story, a senior German commander reached for some version of the same word. At El Guettar it had seemed incomprehensible. At Saint-Lô it seemed no less so. Between those 2 moments, the American learning cycle had not slowed. It had accelerated.

Part 3

By the summer of 1944, the final question was no longer whether the Americans were learning faster. The real question was why the Germans, who could see the outlines of the system by then, still could not build one of their own.

The system was not secret. Captured copies of the Combat Lessons pamphlets had turned up in German intelligence files by the summer of 1944. Tactical and Technical Trends was known to exist. German liaison officers on the Eastern Front had seen rough Soviet equivalents in combat reporting. By 1943, and certainly by 1944, German staff officers could describe in broad terms what the Americans were doing. They still could not reproduce it.

The reason had little to do with paper, printing, or distribution. It rested on 3 things deeper than any manual.

The first was the question of whose observation counted. In the German tradition, tactical insight belonged to commanders. The word they used for what a good officer possessed was Fingerspitzengefühl, a feel in the fingertips for terrain, timing, and enemy intention. It was treated as a virtue of the man, not a product of the system. When men like Mellenthin and Hermann Balck spoke after the war, they described combat in the language of individual genius. A good commander saw what subordinates did not. A good sergeant executed intent. The notion that a 29-year-old sergeant from New Jersey might produce one of the most important tactical innovations of the Normandy campaign, and that his name would stand beside generals in the official record, did not fit comfortably inside that tradition. It was not a question of whether such men existed. It was a question of who was permitted to have ideas.

The second was the question of what could be reported honestly. American after-action reports could state plainly that a particular attack had failed because of a specific tactical mistake and that the mistake should be corrected in training. The report would be filed, read, and used. Writing it did not end a career. In the later years of the Third Reich, German reporting operated under a political pressure that made such honesty dangerous. Pessimism could be read as defeatism, and defeatism might be treated as a criminal offense. By 1944, Hitler had dismissed, humiliated, or destroyed enough senior commanders for reporting unwelcome truths that honest reporting had largely dried up at the top. Down at battalion level, German officers still recorded what they saw, and many of those records survived. But the vertical circulation of bad news into the training and doctrinal system had broken. Honest reporting was punished. Honest reporting was exactly what fed the American system.

The third was speed. American industrial capacity could print Combat Lessons in enormous numbers and distribute it down to every company in every theater. American transportation could take an observer from a foxhole in Belgium to a classroom at Fort Benning in a week. American replacement training centers could rework their curriculum in days. By 1944, when the German Army was increasingly unable to move paper and men across its own internal lines, the American feedback loop was still accelerating.

Put those 3 things together and the answer that von Broich could not find at El Guettar, and that Bayerlein could not find at Saint-Lô, becomes visible. It was not that the Americans were inherently smarter. It was not that their soldiers were inherently braver. It was certainly not that their tanks were better than the panzers and Tigers they faced. It was that the American Army had built a system in which the observation of a rifleman in a hedgerow in July could reach a training instructor in Georgia by September, shape a revised curriculum by October, and return to the front in the person of a replacement arriving in November. The same loop ran again and again through 3 years of war. By the time Panzer Lehr faced it in Normandy, American infantry divisions were learning roughly 10 times faster than their opponents.

Mellenthin’s postwar insult became, by accident, one of the clearest descriptions of what he had seen. They were masses, and his own soldiers were individuals. The masses had learned together. The individuals had learned alone.

That is why the scene on the morning of 23 March 1943 matters so much. On the plateau east of El Guettar, as von Broich sat in his command car and watched his tanks begin to burn in the open, the decision of the war was already present in a form he could not see. It did not lie only in the panzers dying in front of him, or in the Americans dug in across the valley. It lay in the invisible mechanism behind them: a printing press in Washington, a training depot in Texas, a room at Fort Sill where firing tables were being revised, a recruiting office in New Jersey where a future sergeant would soon put on a uniform, and a feedback loop running every day in every theater, pulling lessons from the dead and wounded and pouring them back into the living.

Von Broich did not understand it. Bayerlein would not understand it either. None of them ever fully did. For the rest of the war they fought an army that was, in the most literal sense, a different army every week.

The final verdict is not that German soldiers ceased to be formidable. They did not. Unit for unit, man for man, German infantry and armor remained dangerous into the last weeks of the fighting. Nor is the verdict that the Germans lost because their equipment had become uniformly inferior. That too would be false. What they lost, against the Western Allies, was the race of adaptation.

They had built their army around the belief that excellence came from individual genius cultivated over a career. The Americans built theirs around the belief that excellence could come from a system that turned every engagement, every failure, and every sergeant’s bright idea into training material for everyone else. Over 3 years, the system beat the genius. In Normandy, over little more than 10 days, it beat the genius faster.

Curtis Culin received the Legion of Merit. Later he lost his left foot to a mine in the Hürtgen Forest. He died in Greenwich Village in 1963, largely unknown outside his hometown of Cranford, New Jersey. Roberts, the Tennessee sergeant whose joking remark helped spark the idea, disappeared from the record. Neither man stands prominently in most histories. They are not on many monuments. The generals have their statues. The privates and sergeants mostly do not.

Yet war was a system, and the men who fought it were never only numbers. They had names. Curtis Culin had a name. Roberts of Tennessee had a name. Fritz von Broich watched his tanks burn at El Guettar because somewhere behind American lines, a sergeant’s observation from a week earlier had already reached the gunner on an antitank weapon. That was the sentence the German Army never learned to write. By the time it understood the meaning of it, the war was already over.