Part I one of this article can be found here.
Another job that Smeed gave me was to invent ways to estimate the effectiveness of various countermeasures, using all the evidence from a heterogeneous collection of operations. The first countermeasure that I worked on was MONICA. MONICA was a tail-mounted warning radar that emitted a high-pitched squeal over the intercom when a bomber had another aircraft close behind it. The squeals came more rapidly as the distance measured by the radar became shorter. The crews disliked MONICA because it was too sensitive and raised many false alarms. They usually switched it off so that they could talk to each other without interruption. My job was to see from the results of many operations whether MONICA actually saved lives. I had to compare the loss rates of bombers with and without MONICA. This was difficult because MONICA was distributed unevenly among the squadrons. It was given preferentially to Halifaxes (one of the two main types of British heavy bomber), which usually had higher loss rates, and less often to Lancaster bombers, which usually had lower loss rates. In addition, Halifaxes were sent preferentially on less dangerous operations and Lancasters on more dangerous operations. To use all the evidence from Halifax and Lancaster losses on a variety of operations, I invented a method that was later reinvented by epidemiologists and given the name “meta-analysis.” Assembling the evidence from many operations to judge the effectiveness of MONICA was just like assembling the evidence from many clinical trials to judge the effectiveness of a drug.
My method of meta-analysis was the following: First, I subdivided the data by operation and by type of aircraft. For example, one subdivision would be Halifaxes on Bremen on March 5; another would be Lancasters on Berlin on December 2. In each subdivision I tabulated the number of aircraft with and without MONICA and the number lost with and without MONICA. I also tabulated the number of MONICA aircraft expected to be lost if the warning system had no effect, and the statistical variance of that number. So I had two quantities for each subdivision: observed-minus-expected losses of MONICA aircraft, and the variance of this difference. I assumed that the distributions of losses in the various subdivisions were uncorrelated. Thus, I could simply add up the two quantities, observed-minus-expected losses and variance, over all the subdivisions. The result was a total observed-minus-expected losses and variance for all the MONICA aircraft, unbiased by the different fractions of MONICA aircraft in the various subdivisions. This was a sensitive test of effectiveness, making use of all the available information. If the total of observed-minus-expected losses was significantly negative, it meant that MONICA was effective. But instead, the total was slightly positive and less than the square root of the total variance. MONICA was statistically worthless. The crews had been right when they decided to switch it off.
I later applied the same method of analysis to the question of whether experience helped crews to survive. Bomber Command told the crews that their chances of survival would increase with experience, and the crews believed it. They were told, After you have got through the first few operations, things will get better. This idea was important for morale at a time when the fraction of crews surviving to the end of a 30-operation tour was only about 25 percent. I subdivided the experienced and inexperienced crews on each operation and did the analysis, and again, the result was clear. Experience did not reduce loss rates. The cause of losses, whatever it was, killed novice and expert crews impartially. This result contradicted the official dogma, and the Command never accepted it. I blame the ORS, and I blame myself in particular, for not taking this result seriously enough. The evidence showed that the main cause of losses was an attack that gave experienced crews no chance either to escape or to defend themselves. If we had taken the evidence more seriously, we might have discovered Schräge Musik in time to respond with effective countermeasures.
Smeed and I agreed that Bomber Command could substantially reduce losses by ripping out two gun turrets, with all their associated hardware, from each bomber and reducing each crew from seven to five. The gun turrets were costly in aerodynamic drag as well as in weight. The turretless bombers would have flown 50 miles an hour faster and would have spent much less time over Germany. The evidence that experience did not reduce losses confirmed our opinion that the turrets were useless. The turrets did not save bombers, because the gunners rarely saw the fighters that killed them. But our proposal to rip out the turrets went against the official mythology of the gallant gunners defending their crewmates. Dickins never had the courage to push the issue seriously in his conversations with Harris. If he had, Harris might even have listened, and thousands of crewmen might have been saved.
The part of his job that Smeed enjoyed most was interviewing evaders. Evaders were crew members who had survived being shot down over German-occupied countries and made their way back to England. About 1 percent of all those shot down came back. Each week, Smeed would go to London and interview one or two of them. Sometimes he would take me along. We were not supposed to ask them questions about how they got back, but they would sometimes tell us amazing stories anyway. We were supposed to ask them questions about how they were shot down. But they had very little information to give us about that. Most of them said they never saw a fighter and had no warning of an attack. There was just a sudden burst of cannon fire, and the aircraft fell apart around them. Again, we missed an essential clue that might have led us to Schräge Musik.
On November 18, 1943, Sir Arthur Harris started the Battle of Berlin. This was his last chance to prove the proposition that strategic bombing could win wars. He announced that the Battle of Berlin would knock Germany out of the War. In November 1943, Harris’s bomber force was finally ready to do what it was designed to do: smash Hitler’s empire by demolishing Berlin. The Battle of Berlin started with a success, like the first attack on Hamburg on July 24. We attacked Berlin with 444 bombers, and only 9 were lost. Our losses were small, not because of WINDOW, but because of clever tactics. Two bomber forces were out that night, one going to Berlin and one to Mannheim. The German controllers were confused and sent most of the fighters to Mannheim.
After that first attempt on Berlin, Sir Arthur ordered 15 more heavy attacks, expecting to destroy that city as thoroughly as he had destroyed Hamburg. All through the winter of 1943 and ‘44, the bombers hammered away at Berlin. The weather that winter was worse than usual, covering the city with cloud for weeks on end. Our photoreconnaissance planes could bring back no pictures to show how poorly we were doing. As the attacks went on, the German defenses grew stronger, our losses heavier, and the “scatter” of the bombs worse. We never raised a firestorm in Berlin. On March 24, in the last of the 16 attacks, we lost 72 out of 791 bombers, a loss rate of 9 percent, and Sir Arthur admitted defeat. The battle cost us 492 bombers with more than 3,000 aircrew. For all that, industrial production in Berlin continued to increase, and the operations of government were never seriously disrupted.
There were two main reasons why Germany won the Battle of Berlin. First, the city is more modern and less dense than Hamburg, spread out over an area as large as London with only half of London’s population; so it did not burn well. Second, the repeated attacks along the same routes allowed the German fighters to find the bomber stream earlier and kill bombers more efficiently.
A week after the final attack on Berlin, we suffered an even more crushing defeat. We attacked Nuremberg with 795 bombers and lost 94, a loss rate of almost 12 percent. It was then clear to everybody that such losses were unsustainable. Sir Arthur reluctantly abandoned his dream of winning the War by himself. Bomber Command stopped flying so deep into Germany and spent the summer of 1944 giving tactical support to the Allied armies that were, by then, invading France.
The history of the 20th century has repeatedly shown that strategic bombing by itself does not win wars. If Britain had decided in 1936 to put its main effort into building ships instead of bombers, the invasion of France might have been possible in 1943 instead of 1944, and the war in Europe might have ended in 1944 instead of 1945. But in 1943, we had the bombers, and we did not have the ships, and the problem was to do the best we could with what we had.
One of our group of young students at the ORS was Sebastian Pease, known to his friends as Bas. He had joined the ORS only six months before I had, but by the time I got there, he already knew his way around and was at home in that alien world. He was the only one of us who was actually doing what we were all supposed to be doing: helping to win the War. The rest of us were sitting at Command Headquarters, depressed and miserable because our losses of aircraft and aircrew were tremendous and we were unable to do much to help. The Command did not like it when civilians wandered around operational squadrons collecting information, so we were mostly confined to our gloomy offices at the headquarters. But Bas succeeded in breaking out. He spent most of his time with the squadrons and came back to headquarters only occasionally. Fifty years later, when he was visiting Princeton (where I spent most of my life, working as a professor of physics), he told me what he had been doing.
Bas was able to escape from Command Headquarters because he was the expert in charge of a precise navigation system called G-H. Only a small number of bombers were fitted with G-H, because it required two-way communication with ground stations. These bombers belonged to two special squadrons, 218 Squadron being one of them. The G-H bombers were Stirlings, slow and ponderous machines that were due to be replaced by the smaller and more agile Lancasters. They did not take part in mass-bombing operations with the rest of the Command but did small, precise operations on their own with very low losses. Bas spent a lot of time at 218 Squadron and made sure that the G-H crews knew how to use their equipment to bomb accurately. He had “a good war,” as we used to say in those days. The rest of us were having a bad war.
Sometime early in 1944, 218 Squadron stopped bombing and started training for a highly secret operation called GLIMMER, which Bas helped to plan, and whose purpose was to divert German attention from the invasion fleet that was to invade France in June. The operation was carried out on the night of June 5-6. The G-H bombers flew low, in tight circles, dropping WINDOW as they moved slowly out over the English Channel. In conjunction with boats below them that carried specially designed radar transponders, they appeared to the German radars to be a fleet of ships. While the real invasion fleet was moving out toward Normandy, the fake invasion fleet of G-H bombers was moving out toward the Pas de Calais, 200 miles to the east. The ruse was successful, and the strong German forces in the Pas de Calais did not move to Normandy in time to stop the invasion. While Bas was training the crews, he said nothing about it to his friends at the ORS. We knew only that he was out at the squadrons doing something useful. Even when GLIMMER was over and the invasion had succeeded, Bas never spoke about it. My boss, Reuben Smeed, was a man of considerable wisdom. One day at Bomber Command, he said, “In this business, you have a choice. Either you get something done or you get the credit for it, but not both.” Bas’s work was a fine example of Smeed’s dictum. He made his choice, and he got something done. In later life he became a famous plasma physicist and ran the Joint European Torus, the main fusion program of the European Union.
The one time that I did something practically useful for Bomber Command was in spring 1944, when Smeed sent me to make accurate measurements of the brightness of the night sky as a function of time, angle, and altitude. The measurements would be used by our route planners to minimize the exposure of bombers to the long summer twilight over Germany. I went to an airfield at the village of Shawbury in Shropshire and flew for several nights in an old Hudson aircraft, unheated and unpressurized. The pilot flew back and forth on a prescribed course at various altitudes, while I took readings of sky brightness through an open window with an antiquated photometer, starting soon after sunset and ending when the sun was 18 degrees below the horizon. I was surprised to find that I could function quite well without oxygen at 20,000 feet. I shared this job with J. F. Cox, a Belgian professor who was caught in England when Hitler overran Belgium in 1940. Cox and I took turns doing the measurements. My flights were uneventful, but on the last of Cox’s flights, both of the Hudson’s engines failed, and the pilot decided to bail out. Cox also bailed out and came to earth still carrying the photometer. He broke an ankle but saved the device. In later years, he became rector of the Free University in Brussels.
After the War, Smeed worked for the British government on road traffic problems and then taught at University College London, where he was the first professor of traffic studies. He applied the methods of operational research to traffic problems all over the world and designed intelligent traffic-light control systems to optimize the flow of traffic through cities. Smeed had a fatalistic view of traffic flow. He said that the average speed of traffic in central London would always be nine miles per hour, because that is the minimum speed that people will tolerate. Intelligent use of traffic lights might increase the number of cars on the roads but would not increase their speed. As soon as the traffic flowed faster, more drivers would come to slow it down.
Smeed also had a fatalistic view of traffic accidents. He collected statistics on traffic deaths from many countries, all the way back to the invention of the automobile. He found that under an enormous range of conditions, the number of deaths in a country per year is given by a simple formula: number of deaths equals .0003 times the two-thirds power of the number of people times the one-third power of the number of cars. This formula is known as Smeed’s Law. He published it in 1949, and it is still valid 57 years later. It is, of course, not exact, but it holds within a factor of two for almost all countries at almost all times. It is remarkable that the number of deaths does not depend strongly on the size of the country, the quality of the roads, the rules and regulations governing traffic, or the safety equipment installed in cars. Smeed interpreted his law as a law of human nature. The number of deaths is determined mainly by psychological factors that are independent of material circumstances. People will drive recklessly until the number of deaths reaches the maximum they can tolerate. When the number exceeds that limit, they drive more carefully. Smeed’s Law merely defines the number of deaths that we find psychologically tolerable.
The last year of the War was quiet at ORS Bomber Command. We knew that the War was coming to an end and that nothing we could do would make much difference. With or without our help, Bomber Command was doing better. In the fall of 1944, when the Germans were driven out of France, it finally became possible for our bombers to make accurate and devastating night attacks on German oil refineries and synthetic-oil-production plants. We had long known these targets to be crucial to Germany’s war economy, but we had never been able to attack them effectively. That changed for two reasons. First, the loss of France made the German fighter defenses much less effective. Second, a new method of organizing attacks was invented by 5 Group, the most independent of the Bomber Command groups. The method originated with 617 Squadron, one of the 5 Group squadrons, which carried out the famous attack on the Ruhr dams in March 1943. The good idea, as usually happens in large organizations, percolated up from the bottom rather than trickling down from the top. The approach called for a “master bomber” who would fly a Mosquito at low altitude over a target, directing the attack by radio in plain language. The master bomber would first mark the target accurately with target indicator flares and then tell the heavy bombers overhead precisely where to aim. A deputy master bomber in another Mosquito was ready to take over in case the first one was shot down. Five Group carried out many such precision attacks with great success and low losses, while the other groups flew to other places and distracted the fighter defenses. In the last winter of the War, the German army and air force finally began to run out of oil. Bomber Command could justly claim to have helped the Allied armies who were fighting their way into Germany from east and west.
While the attacks on oil plants were helping to win the War, Sir Arthur continued to order major attacks on cities, including the attack on Dresden on the night of February 13, 1945. The Dresden attack became famous because it caused a firestorm and killed a large number of civilians, many of them refugees fleeing from the Russian armies that were overrunning Pomerania and Silesia. It caused some people in Britain to question the morality of continuing the wholesale slaughter of civilian populations when the War was almost over. Some of us were sickened by Sir Arthur’s unrelenting ferocity. But our feelings of revulsion after the Dresden attack were not widely shared. The British public at that time still had bitter memories of World War I, when German armies brought untold misery and destruction to other people’s countries, but German civilians never suffered the horrors of war in their own homes. The British mostly supported Sir Arthur’s ruthless bombing of cities, not because they believed that it was militarily necessary, but because they felt it was teaching German civilians a good lesson. This time, the German civilians were finally feeling the pain of war on their own skins.
I remember arguing about the morality of city bombing with the wife of a senior air force officer, after we heard the results of the Dresden attack. She was a well-educated and intelligent woman who worked part-time for the ORS. I asked her whether she really believed that it was right to kill German women and babies in large numbers at that late stage of the War. She answered, “Oh yes. It is good to kill the babies especially. I am not thinking of this war but of the next one, 20 years from now. The next time the Germans start a war and we have to fight them, those babies will be the soldiers.” After fighting Germans for ten years, four in the first war and six in the second, we had become almost as bloody-minded as Sir Arthur.
At last, at the end of April 1945, the order went out to the squadrons to stop offensive operations. Then the order went out to fill the bomb bays of our bombers with food packages to be delivered to the starving population of the Netherlands. I happened to be at one of the 3 Group bases at the time and watched the crews happily taking off on their last mission of the War, not to kill people but to feed them.
Freeman Dyson was for many years professor of physics at the Institute for Advanced Study in Princeton. He is famous for his contributions to mathematical physics, particularly for his work on quantum electrodynamics. He was awarded the Lorentz Medal in 1966 and the Max Planck Medal in 1969, both for his contributions to modern physics. In 2000, he was awarded the Templeton Prize for Progress in Religion.
This is the real story of the Afghan biometric databases abandoned to the Taliban
By capturing 40 pieces of data per person—from iris scans and family links to their favorite fruit—a system meant to cut fraud in the Afghan security forces may actually aid the Taliban.
The covid tech that is intimately tied to China’s surveillance state
Heat-sensing cameras and face recognition systems may help fight covid-19—but they also make us complicit in the high-tech oppression of Uyghurs.
How Amazon Ring uses domestic violence to market doorbell cameras
Partnerships with law enforcement give smart cameras to the survivors of domestic violence. But who does it really help?
Why you should be more concerned about internet shutdowns
Governments are turning off the internet to silence dissenters at an ‘exponential’ rate—and threatening civil society, says the chief operating officer of Google’s Jigsaw project.
Get the latest updates from
MIT Technology Review
Discover special offers, top stories, upcoming events, and more.