Chapter 1

The Sorcerer’s Apprentice: Bomber Command, 1939–42

On September 1, 1939, President Franklin D. Roosevelt sent an appeal to all the major European powers involved in the crisis over Poland to give a public undertaking that they would abstain from any air attacks against civilians or unfortified cities. The same day, Hitler told the American chargé d’affaires in Berlin that this had always been his preference and assured Roosevelt that German aircraft would attack only military objectives. The British prime minister, Neville Chamberlain, gave his guarantee the same day; a joint Anglo-French declaration followed on September 3, only reserving the right to act as they saw fit if the enemy failed to observe the same restrictions.¹ The Polish ambassador in Washington, whose country was already at war, agreed that Polish pilots would be told not to bomb open cities as long as the enemy did the same.² None of these expressions of goodwill were legally binding in international law.

The idea that bombing warfare could somehow be “humanized” had been explored by a British committee set up in July 1938 with the cumbersome title “Limitation of Armaments Committee, Sub-Committee on the Humanisation of Aerial Warfare.” The discussions of the committee, chaired by Sir William Malkin, went round in circles. The terms for a possible international agreement on limiting bombing to military objectives suffered not only from the realistic objection that such terms would be unenforceable in a real war, but also from the difficulty of defining what was meant by a military objective. In the end the committee proved more useful in giving the Air Ministry the opportunity to defend the idea that bombing arms factories and armaments workers was as legitimate as a naval blockade than it was in finding grounds for a diplomatic solution.³ In the absence of international agreement, the RAF was told to abide by the Hague Rules for air warfare, first drafted in 1923 but never ratified, but to do so only as long as the enemy did the same. The rules were spelled out in a cabinet decision and repeated regularly up to and beyond the outbreak of war: intentional bombing of civilians was illegal; only identifiable military objectives could be attacked from the air; and any such attack must be undertaken without negligent harm to civilians. In August 1939 the Air Ministry concluded that attacks on targets difficult to identify through cloud or at night would also be illegal, as would any operation in which the civil population, hospitals, cultural monuments, or historic sites were targeted. An interdepartmental committee, set up the same month to draft detailed instructions on rules of engagement for the British armed forces, stipulated that “it is clearly illegal to bombard a populated area in the hope of hitting a legitimate target.”⁴

These legal limitations reflected decisions taken in military staff talks between the British and French high commands in the spring of 1939 as they planned for a possible war. In April the two military staffs had agreed only to attack military targets in the narrowest sense of the term, at sea or at the fighting front. British bombers were to be used to help the battle on land, not to attack distant targets in Germany. The French were insistent that the RAF should not attack German cities while the balance of air resources so obviously favored Germany and French industry was not yet adequately protected.⁵ A few days before the outbreak of war the chief of the British air staff, Sir Cyril Newall, warned Air Chief Marshal Edgar Ludlow-Hewitt, commander in chief of Bomber Command, that his activities were bound to be restricted “for political reasons,” though Ludlow-Hewitt knew that Bomber Command’s small size and operational difficulties were enough on their own to inhibit offensive action.⁶ Newall worried that even if the RAF bombed legitimate objectives, the Germans would claim that they had killed civilians. There was strong political pressure to ensure that the democracies were not seen to violate the bombing proscription first. It was decided in October 1939 that only if German aircraft started to kill large numbers of civilians from the air—“promiscuous bombardment,” as it was called—would the RAF “take the gloves off.”⁷

Both sides at first stuck to their pledge not to attack targets in each other’s cities where civilians were at risk (though this did not prevent the German Air Force from killing noncombatants during its operations in Poland). Chamberlain had no interest in provoking German bombing of British towns, despite appeals from the Polish government in early September to begin bombing Germany as a gesture of assistance.⁸ It was nonetheless the case that for the first months of the war Bomber Command strained at the leash to be able to do what years of planning had prepared it for. A halfhearted agreement was reached with the French high command to initiate bombing of the German Ruhr-Rhineland industrial region—usually capitalized in British documents simply as the RUHR—if a sudden German attack threatened Belgium or menaced Franco-British forces decisively, but the French remained cautious about risking German retaliation, even during the invasion in May 1940.⁹ In the end it was the British who ended the international embargo agreed to in September. On the night of May 11–12, two days after the German invasion in western Europe, thirty-seven medium and light bombers attacked industrial and transport targets in the Rhineland city of München Gladbach (now Mönchengladbach), killing four people, including an Englishwoman who happened to live there. British raids were to continue throughout the war. The last was made on May 2–3, 1945, on the north German port of Kiel, just thirty-six hours before Allied troops occupied the city.¹⁰

“Taking the Gloves Off,” 1939–40

The political and legal restrictions imposed on Bomber Command were consistent with the widely held view in Britain that indiscriminate bombing was the hallmark of barbarism, whereas self-restraint was a feature of being civilized. Yet there were also powerful prudential arguments for not undertaking bombing from the start of the war, as the French realized. In the autumn of 1939, Bomber Command was not yet ready to launch any major offensive campaign. For all the talk in the 1930s of developing a “striking force” capable of taking the fight to the enemy heartland, progress in developing the technology of aircraft, bombs, bombsights, and navigation aids had been painfully slow. One of Chamberlain’s more desperate acts before the war was to ask Roosevelt on August 25, 1939, a week before the president’s appeal, to supply the American Norden bombsight for British use. Roosevelt declined, not because it would compromise his subsequent appeal to abstain from bombing, but because agreeing to this request would make it seem that the United States had taken sides in the conflict.¹¹ At the outbreak of war Ludlow-Hewitt was well aware of the deficiencies of his force, which had already been exposed when the Air Ministry in 1938 at last began to consider the practicalities of long-range bombing. The gap between ambition and reality was remarkably wide for a force committed to a bombing strategy, a reflection of the poor technical experience of much of the RAF leadership and the failure to define doctrine clearly. There were too few airfields capable of handling heavier aircraft, little experience in bombing training, a shortage of maps of northwestern Germany, and a total of only 488 bombers of all kinds, including light bombers destined to form part of the Advanced Air Striking Force when it was sent to France in late 1939. “Unrestricted air warfare,” ran an Air Ministry instruction, “is not in the interests of Great Britain.”¹²

RAF Bomber Command dated from the reorganization of the air arm in 1936. In contrast to the German Air Force, all bomber units were grouped together under a single commander in chief, based at Bomber Command headquarters near High Wycombe, northwest of London. This had important implications for the identity of the command, because its sole function was to bomb. As a force it was only offensive, and its main duty was to define what targets to bomb, to produce the technology to enable those targets to be destroyed, and to train the manpower to do it. This functional identity encouraged Bomber Command to construct a doctrine and a force independent of the army and navy, capable of striking a potential enemy at what was perceived to be its most vulnerable point. The command was disinclined to accept the role of auxiliary to the requirements of surface forces, and there was almost no planning for army cooperation to match German Air Force doctrine. Right up to the outbreak of war and beyond, the RAF argued that its contribution to the battlefield would be of little significance: attacks on railway communications were regarded as difficult and ineffective, while raids against marching columns of men were deemed to be a waste of bombing resources.¹³ Instead the Air Ministry drew up what were known as the Western Air Plans, a series of sixteen individual plans, some of which committed the bomber force to assist the Admiralty in the war at sea, but none of which committed the bombers to help the army in the field. Only two of the plans were seriously prepared for: W.A.4 involved long-range attacks against communications targets in Germany to slow a German advance; W.A.5 was for attacks on the German industrial economy, particularly the Ruhr area and the German oil industry.¹⁴ An Industrial Intelligence Centre, first established in 1931 under Desmond Morton (later Churchill’s intelligence adviser), drew up lists of vulnerable targets in German industry. Most of them could not be hit from British bases with existing aircraft, but they formed the basis for the RAF insistence that the most efficient use of bomber aircraft was against the enemy home front, not its armed forces.¹⁵ Yet vulnerable targets on the home front could not be hit as long as the government insisted on the letter of the law.

The bomber force was organized in September 1939 in five groups, spread out across east-central England and East Anglia, each group made up of between six and eight squadrons, a total of thirty-three in all. Although some of the aircraft were called heavy bombers at the time, the groups were actually equipped like the German Air Force with twin-engine light or medium bombers. The light bombers made up sixteen squadrons, ten consisting of the Fairey Battle, six of the Bristol Blenheim IV. The Battle emerged from a 1933 requirement for a monoplane light bomber but was obsolescent by 1939 and unsuited to daylight raiding; its operational function was unclear and its striking power negligible. They were sent to France as part of the air expeditionary force but were sitting ducks for German fighters. On May 14, 1940, as many as forty out of a force of seventy-one Battles were shot down in a single disastrous operation against the crossings at Sedan in the Ardennes, close to the Belgian border. The Blenheim was developed from 1935 as a fast medium bomber. It was weakly armed but could reach a maximum speed of 266 miles per hour. It could carry only 1,000 pounds of bombs around 700 miles. Its limited range and small bomb load made it unsuitable for a strategic role, and it played a larger part in the antishipping war along the German-occupied northern shores of Europe. Light bombers were sensibly phased out as the war progressed.¹⁶

Bomber Command had three principal long-range bombers in 1939: the Vickers Wellington, the Armstrong Whitworth Whitley, and the Handley Page Hampden. The Hampden came from a 1932 specification, the Whitley from 1934, making them both considerably older than their German counterparts. The Hampden had a top speed of 255 miles per hour and could carry up to 4,000 pounds of bombs around 600 miles; the Whitley Mk V with Rolls-Royce Merlin engines was the mainstay of Bomber Command in the first year, with a range of 800 miles with 3,000 pounds of bombs and a top speed of 222 mph. They were lightly armed and easy prey on unescorted sorties by day. The most successful of the medium bombers was the Vickers Wellington Mk IC (succeeded in 1941 by the Mk II and III), which made up more than half of Bomber Command’s strategic force by 1942. It was powered by Bristol Pegasus engines, had a top speed of 235 miles per hour, and could carry 4,500 pounds of bombs some 600 miles. Its geodetic construction—a lattice-like fuselage shell—made it exceptionally robust compared with the Hampden and Whitley, and later marks of the Wellington remained in service throughout the war.¹⁷ What the medium bombers lacked were effective navigational aids to match the German electronic systems (navigators and pilots were still taught astral navigation), powerful armament to be able to defend themselves against fighter assault, and bombs of sufficient destructive power. The standard British 250- and 500-pound bombs had a low charge-to-weight ratio (around one-quarter was explosive, against one-half in German bombs), but they also contained less destructive explosive content, without the addition of aluminum powder (standard in German bombs), and were prone to fail to detonate. The standard incendiary bomb was the four-pound Mk I magnesium bomb, which remained in production, with small modifications, throughout the war. Larger “firepot” bombs were developed in 1939 and 1940, designed to distribute a high number of incendiary devices, but they were plagued with technical difficulties. Only later in the war did Bomber Command acquire heavier oil-based incendiaries.¹⁸ The technical level of the force that went to war in 1939—aircraft, bombs, and equipment—can be described charitably as unsophisticated.

The early experience of the bomber force during the “Phoney War” confirmed the wisdom of not pressing for an immediate bombing offensive. The political restrictions confined the RAF to attacks on German naval vessels at sea and naval targets on the North Sea islands of Sylt and Helgoland, and on the German coast at Wilhelmshaven. Even these limited operations brought insupportable casualty rates: a small raid on the German coast on September 4 cost 23 percent of the bomber force; a raid by Hampdens against Helgoland on September 29 cost half the force. In October 1939 the “heavy bombers” were ordered to operate chiefly at night, as the Air Ministry had always expected.¹⁹ The only operations permitted in German airspace were propaganda runs dropping millions of leaflets. Some aircrews, struggling to cope with the excessive cold, chucked out the heavy bundles without cutting them first, making them a potentially more lethal weapon than had been intended. By March 1940 it was reported that the morale of Bomber Command crews was close to cracking after long and dangerous propaganda operations that seemed to contribute nothing to winning the war and exposed crews to excessive risk of accident.²⁰ The nighttime flights posed all kinds of difficulties. Interviews with operational crews confirmed that intense cold and long, risky flights over sea were compounded with the difficulty experienced over Germany itself, which they found to be “very black.” It proved almost impossible to find and hit a specific target in the midst of the blackout, even with leaflets, a fact that RAF planners had already realized some months before when drawing up a “Night Plan” to accommodate the shift from daylight to nighttime operations, in which it was admitted that hitting anything “will be largely a matter of chance.”²¹ In late March, shortly before the campaign in Norway, the chiefs of staff concluded that Bomber Command was too weak and unprepared to be able to do anything effective in the foreseeable future.²² In April, just prior to the German invasion of the Low Countries and France, the new head of Bomber Command, Air Marshal Charles Portal, who replaced Ludlow-Hewitt in March, told the air staff that he had only 260 serviceable bombers and 384 crews; he estimated that they would be capable of dropping 100 tons of bombs in the first week of bombing operations but only thirty tons by week three. Because of the high rate of loss anticipated, the force would be capable of only thirty-six sorties a day after two weeks of conflict.²³ This was a negligible effort on the eve of the German offensive.

It is all the more surprising under these circumstances that the RAF should have decided to take the gloves off in May 1940 when the German invasion began. The decision to permit British bombers to attack military-economic targets on German soil close to civilian populations was not an invitation to undertake the heavy city bombings of the later war years, if only because Bomber Command was manifestly incapable of inflicting them. But it was a threshold that had to be crossed consciously, given the heavy weight of political and ethical restrictions laid on the force from the start of the war. What was judged to be illegal in August 1939 had to be presented as legitimate when it was undertaken in the summer of 1940. Most explanations for the start of the British campaign have assumed that it was a response to the German bombing of Rotterdam on May 14, but the first raid, on Münchengladbach, had already taken place three days before, while Rotterdam was not mentioned in any of the cabinet discussions about initiating the bombing of German targets. The decision was taken because of the crisis in the Battle of France, not because of German air raids.²⁴ The actual circumstances surrounding the onset of bombing were more complex. By chance the German attack in the west on May 10, 1940, began on the same day that the Chamberlain government was replaced by a new one headed by Winston Churchill. Chamberlain had always opposed the use of bombers against urban targets, but Churchill had no conscientious or legal objections. As minister of munitions in 1917, he had been a prominent supporter of an independent air force and a campaign of long-range bombing against German industrial objectives. Later, as minister for air after the war, he had played a key role in securing the independent future of the RAF. He accepted the argument that bombing could be expected to produce important strategic results. A government headed by Churchill rather than Chamberlain was always more likely to endorse a bombing campaign.²⁵

One of the first issues discussed by Churchill’s new War Cabinet on May 12 was the virtue of initiating what was described as “unrestricted air warfare.” It was agreed that the RAF should no longer be bound by any moral or legal scruples to abstain from bombing; Germany’s wartime actions, Churchill claimed, had already given the Allies “ample justification” for retaliation. The chiefs of staff had since the outbreak of war approved the idea that if urgent military necessity made bombing imperative, it should not be limited by humanitarian considerations. On May 13 the cabinet considered again whether the crisis in the Battle of France was severe enough to justify bombing. Though there were arguments against running the risk of German retaliation, approval was given for a bombing attack against oil and rail targets in Germany on the night of May 14–15. On May 15 the cabinet finally took the decision to approve a full bombing strategy against German targets where civilians might be casualties, as long as they were “suitable military objectives.”²⁶ The driving force behind the decision was the deputy prime minister, Clement Attlee, who had been absent from the earlier cabinet discussions but who strongly favored raids on Germany, and continued to do so throughout the war. Churchill was anxious about the effect on American opinion if Britain began the bombing war, but the dire state of the ground campaign turned the tide. It was hoped that German bombers would be forced to strike back at Britain, while German fighters would be withdrawn from the land battle to defend Germany, though neither eventuality materialized. On the night of May 15–16, Bomber Command launched its first large-scale raid, with ninety-nine medium bombers on targets scattered across the Ruhr. To prevent further heavy losses of bombers in the land campaign in France, the air staff decided on May 19 to use the medium bombers only for attacks on Germany; on May 30, as the front in France collapsed, Bomber Command was ordered to stop using any bombers for direct support of the land battle and to concentrate on German industry, an admission of just how disastrously the British light bombers had performed in battlefront conditions against an air force that had advanced air technology and a sound doctrine for aircraft use.²⁷

The change in priorities necessitated a revision of the rules on the conduct of air warfare laid down in August 1939, which had made it illegal to attack targets in which civilians might be “negligently” killed. On June 4, 1940, the Air Ministry issued new guidelines for Bomber Command, canceling the earlier instructions. The intentional killing of civilians was still regarded as a violation of international law, but attacks could be made on military targets “in the widest sense” (factories, shipyards, communications, power supply, oil installations), in which civilian casualties would be unavoidable but should be proportional.²⁸ “Undue loss of civil life” was still to be avoided, which meant returning to base or jettisoning bombs safely if the target could not be properly identified.²⁹ Nevertheless, the ethical restraints imposed at the start of the war were eroded step by step as a result of the decision to initiate “unrestricted” bombing of targets in urban areas. In July, Bomber Command war orders were modified to allow pilots discretion in choosing any military or military-economic target (defined increasingly broadly) if the primary target was obscured or difficult to find, a policy almost certain at night to involve extensive damage to civilian life since, as Portal reminded the Air Ministry, a high percentage of bombs “inevitably miss the actual target.”³⁰ The final restraints were lifted in September and October 1940 after the heavy German attacks on London, which began on September 7 and continued almost without interruption for three months. The onset of the German “Blitz” is often regarded as the trigger for British attacks on German urban areas, but there had already been growing pressure from Bomber Command to be allowed to bomb less discriminately.³¹ In September the policy of bringing back unused bombs was suspended in favor of bombing anything that could be found worth bombing, though not merely at random. On October 30, Bomber Command was directed to focus on enemy morale by causing “heavy material destruction in large towns” to teach the German population what bombing could do.³² In setting aside the political and legal limitations that had operated during the Phoney War, this decision brought to an end the first stage, and it paved the way for the escalation of the RAF campaign during 1941 and 1942 into full-scale city bombing.

The onset of the RAF bombing campaign in the summer of 1940 can certainly be explained by the change in government and the military necessity imposed by the German breakthrough and triumph in the west, but neither argument is entirely convincing. Unrestricted (though not yet unlimited) air warfare against Germany owed its genesis partly to the intense pressure applied by the RAF from the start of the war to be allowed to commence bombing operations over Germany regardless of the possible human cost. Military expediency also played a part, for RAF leaders had a force whose chief purpose was long-distance bombing; used in any other way, the strike force would no longer give the RAF the chance to demonstrate what strategic bombing could achieve. Bombing policy was predicated on offensive action and nourished by the idea, widespread among RAF leaders, that total war, if it came, would see the erosion of any distinction between the fighting man and the civilian war worker.³³ In the first few months of the war senior airmen argued repeatedly for attacks on the Ruhr as Germany’s Achilles’ heel. It was always recognized, by politicians as well as airmen, that such attacks would involve heavy civilian losses, as a War Cabinet paper in October 1939 made clear:

Germany’s weakest spot is the Rhur [sic], the heart of which is about the size of Greater London, and in which is concentrated approximately 60% of Germany’s vital industry. It contains, moreover, a population which might be expected to crack under intensive air attack. Such attacks would involve a heavy casualty roll among civilians, including women and children.³⁴

A planning document in October 1939 claimed that Bomber Command could, if allowed, bring industry in the Ruhr “practically to a standstill”; about a month later the Air Ministry produced a precise schedule of sorties needed to “K.O. the Ruhr” in a matter of weeks.³⁵ By the spring of 1940 there was a chorus of demands to allow the Ruhr plan to be put into action. Early in May, Viscount Trenchard, the doyen of the bombing lobby, expressed his regret to Portal that bombing had not yet been attempted “when I and others think it would probably have ended the war by now.” Portal himself, two days before the start of the German invasion, pressured Newall, the chief of the air staff, to reserve Bomber Command for attacks on German industry rather than fritter it away in direct support for the army.³⁶ All such views were represented by Newall to the chiefs of staff and the War Cabinet. The decision in May was clearly influenced by the widespread but unverifiable assertion that bombing would achieve something worthwhile.

The bombing lobby rested its case on a number of evident exaggerations. Both the accuracy and power of British bombing and its capacity to inflict decisive material and psychological damage on Germany were presented in terms quite incompatible with the reality of Bomber Command’s strength, range, and capability. The detailed study on the Ruhr bombing suggested that somewhere between 1,000 and 4,000 sorties were all that were required to knock it out. Calculations suggested that eight bomb hits would eradicate a power plant, sixty-four hits destroy a coking plant, and twelve hits destroy an aqueduct; average bombing error was given as seventy-five yards (sixty-nine meters) from low level and 300 yards (276 meters) from high altitude, figures that had never been verified under combat conditions (and proved entirely unattainable in practice).³⁷ It would take Germany months, so it was claimed, to recover from such an assault. There was no sense in this, as in other planning documents, of the exceptional operational and technical difficulties that would be encountered in carrying out such a program.

The effect that these operations would have had on the German war economy and German morale, even if they could have been executed, was subject to similar distortion. In the late 1930s intelligence assessments of German economic capacity for war almost all emphasized the fragility of an industrial economy regarded as taut and overburdened. Planners in the Air Ministry assumed that they faced an enemy whose strength was a façade, “politically rotten, weak in financial and economic resources,” and that the results of bombing were likely to be “decisive” for the outcome of the war.³⁸ This remained the prevailing view for much of the war despite all evidence to the contrary. It also colored the first overoptimistic reports of the impact of RAF raids in summer 1940 once the gloves were off. A Foreign Office intelligence report compiled from neutral eyewitnesses on May 30 suggested “terrible effects” in the Ruhr and a serious crisis of morale spreading through the German home front; a second report sent on to Churchill in early June by the foreign secretary, Lord Halifax, talked of the profound depression in Germany caused by the “violence and efficacity” of British bombing.³⁹ An Air Intelligence evaluation suggested “general dislocation of rail traffic” throughout Germany. The RAF planning staff considered that the first three weeks of bombing had produced “valuable results,” apparently justifying the decision to start it in the first place.⁴⁰

There was also the moral dimension. To abandon the principle that killing civilians from the air was wrong owed a good deal to the British perception of the German enemy. The legal issue involved was sidestepped by two arguments: first, that the Germans had begun unrestricted bombing and would do it again, given the chance; second, that Hitler’s Reich represented such a profound menace to Western civilization that the greater moral imperative was to use every means available to destroy it. The view that the Germans were responsible for bombing civilians first had a long pedigree, stretching back to the Zeppelin and Gotha raids of the First World War, which many RAF commanders had experienced as young officers just over twenty years before. During the 1930s, popular prejudices revived about German science and the German military conniving to produce lethal weapons of mass destruction to be unleashed from the air on an unsuspecting opponent.⁴¹ The bombing of the Basque town of Guernica in April 1937 by the German Condor Legion was popularly regarded in the West as evidence that the Germans had once again abandoned any pretense of civilized behavior. The campaign in Poland was scrutinized for evidence that German bombers were deployed for terrorizing and murdering civilians. Although the evidence was ambiguous, since it was understood that the German raids were directed at military targets as part of a combined-arms ground campaign, the RAF preferred to assume that the Germans had bombed indiscriminately. A report published by the RAF Tactical Committee in October 1939 of a speech made by a German air staff officer on the Polish campaign claimed that on the Germans’ own admission their operations went beyond the terms of the Hague Rules.⁴² That same month Newall told the commander of the Advanced Air Striking Force in France that because of German action in Poland, “we are no longer bound by restrictions under the instructions governing naval and air bombardment. . . . Our action is now governed entirely by expediency.”⁴³ Churchill himself later came to cite the German bombing of Warsaw and Rotterdam as moral justification for what was to be done to the German civilian population.

The assumption in all the discussions about restricting bombing was that it had force only so long as the enemy observed the same limitation, and in this sense Poland played an important part in paving the way for British action. Of course, German bombers had not yet bombed British cities, so the argument for attacking Germany came to be based on preemptive retaliation. Even before the war the RAF had taken for granted that the German Air Force would be bound by no scruples and would be “ruthless and indiscriminate” when the time came for a knockout blow.⁴⁴When a German raid on the Royal Navy base at Scapa Flow in March 1940 killed a nearby cottager (the first civilian casualty of the war), Churchill angrily berated the Air Ministry for not giving it maximum publicity as the likely start of “deliberate horror raids on civilians,” for which the Germans would carry the blame.⁴⁵ In April the propaganda department of the new Ministry of Economic Warfare recommended describing German reconnaissance missions as frustrated bombing raids—“driven off before they were able to drop their bombs”—so as to justify any British retaliation.⁴⁶ In May 1940 one of the arguments for bombing Germany was that sooner or later German leaders would do the same when it suited their strategic interests. “Do [the government] think,” wrote one bombing lobbyist, Marshal of the RAF Sir John Salmond, “that Hitler, who does not consider for a moment the slaughter of thousands on the ground and of devastating countries with which he has no quarrel, will shrink from killing civilians in this country?”⁴⁷ In the cabinet discussion on May 15, Sir Hugh Dowding, commander in chief of Fighter Command, argued in favor of anticipatory attacks because he felt convinced that sooner or later the German Air Force would start indiscriminate bombing.⁴⁸ Yet very soon the argument that Germany started the bombing became the standard version, both among the wider public and in the RAF, and it has remained firmly rooted in the British public mind ever since.

The view that German crimes, or potential crimes, made British bombing legitimate was legally dubious, since it amounted to claiming that two wrongs make a right, but it was morally rooted in the belief that German bombing was just one manifestation of the profound threat that Hitler and the National Socialist movement represented to the survival of the West. The terms of the contest in 1940 were easily presented as a struggle of light against dark, civilization against barbarism, to a public anxious not to see the war as a repeat of the pointless slaughter of 1914–18. The claim to the moral high ground gave an ethical purpose to British strategy that could be used to justify an air policy that in the 1930s would have been widely regarded as a moral lapse, like the Italian bombing of Ethiopians or the Japanese bombing in China. Churchill was among the foremost champions of the idea that every effort should be made to root out the source of Europe’s political poison. “The whole world is against Hitler and Hitlerism,” he announced in a radio broadcast in November 1939. “Men of every race and clime feel that this monstrous apparition stands between them and the forward movement that is their due, and for which the age is ripe.”⁴⁹ And although Churchill sometimes made the distinction between “Nazis” and “Germans,” he commonly used the pejorative “Huns” to describe the enemy, a term deliberately chosen in the First World War to contrast German barbarity with Western civility.

There was also widespread popular support for the argument that Hitler’s Germany was so wicked that any methods, even if they were morally questionable, should be used to destroy it. This was true across the political spectrum, even among those who had campaigned for peace in the 1930s. In an article on “Nazism and Civilisation,” Sir Charles Trevelyan, who had once advocated mass resistance to war, defined the war as a necessary crusade against a barbarous system that had discarded “all moral, international and public law” and had bombed women and children.⁵⁰Philip Noel-Baker, director of the International Peace Campaign in the late 1930s, observed that the bombing of Germany, of which he approved, was almost civilized “compared to the concentration camp and to the Himmler terror.” In a newspaper article just after the onset of the Blitz he pointed out that Hitler had smashed “every last remnant of the Laws of War” and that British hands were now free to take any measures to “bring his monstrous aggression to an end.”⁵¹ A leading member of the Women’s League for Peace and Freedom, which had publicly condemned bombing in the 1930s, abandoned her pacifism in 1940 on the grounds that “Nazism is a menace of evil corruption and lying and of all the forces of evil,” which could only be eradicated by the use of force. When pacifist clergy lobbied the archbishops of Canterbury and York in 1940 to condemn the British use of bombing, they received the following reply: “The moral issue involved in the victory of the allies is of greater importance than the harsh fact of fighting by methods that one deplores.”⁵² Many prominent churchmen, politicians, and intellectuals who might have condemned bombing under different circumstances supported it as a necessary evil. For people who were already predisposed to see the German threat in crude moral terms, it was a relatively simple step to the argument that the greater moral obligation was to secure the continued freedom of the West than to abstain from killing German civilians.

The British government and the RAF leadership were nevertheless aware that it was necessary to present the bombing as distinct from German practice to avoid the accusation that the British were no less barbarous than the Germans. In late April 1940 the minister for air, Sir Samuel Hoare, stated in a radio broadcast that the British would never imitate the enemy’s “dastardly conduct”: “We will not bomb open towns. We will not attempt to defeat the Germans by terrorizing their women and children.”⁵³ When the decision was finally taken on May 15 to start bombing, Churchill advised the Ministry of Information to publish a discreet press release on German killing of civilians from the air in France and the Low Countries, but to say nothing about British retaliation.⁵⁴ Pilots were at first instructed to ensure they could identify and hit specific targets before bombing them. Emphasis was put in every subsequent discussion on RAF bombing that it was directed only at military targets, even though the term was stretched almost to meaninglessness by the long list of economic and social objectives eventually classified as military, and the decision in October 1940 to permit the targeting of morale through attacks on city areas.⁵⁵ Throughout the war the public presentation of the bombing offensive in Parliament and in the press never deviated from the claim that the RAF bombed only military targets, unlike the enemy. When attacks against “industrial populations” were included in a draft directive in August 1942, the Air Ministry insisted that the term be altered to “industrial centres” to avoid the impression that civilians were deliberate targets, “which is contrary to the principles of international law—such as they are.”⁵⁶ When Air Marshal Harris tried to persuade the Air Ministry in the autumn of 1943 to be more honest in its publicity about bombing by showing that killing civilian workers was a stated aim of the bombing campaign, the ministry refused to change. “It is desirable,” ran the reply, “to present the bomber offensive in such a light as to provoke the minimum of public controversy.”⁵⁷ Discretion was always observed in describing British bombing as directed at military targets, even to the crews who could see the real results of the later raids that obliterated whole cities.

Much of the argument about whether to bomb or what to bomb was rendered void by the reality of Bomber Command’s offensive across the summer months of 1940. The optimistic intelligence reports were belied by the evidence of just how little Bomber Command could do with limited numbers and a small bomblift. Taking the gloves off revealed not a clenched fist but a limp hand. In late May priority was given to oil (Western Air Plan W.A.6) and communications targets, since their destruction was expected to affect the campaign in France directly. After the French defeat the priority given to oil remained in place, but a new directive on June 20 added the German aircraft industry to the list and proposed the firebombing of forests to induce a food crisis in Germany. It was argued, implausibly, that game driven from the forests by fire would be forced to eat the crops on the surrounding farmland. On July 30, Bomber Command was also directed to hit electric power stations where they could be located easily by night, while over the whole summer raids had to be carried out on targets in the ports from which an invasion might be launched.⁵⁸ Between June and August the modest bomb load of 2,806 tons was divided between these eight different objectives, spread across northern France, the Low Countries, and northern Germany, a little under one-quarter of the tonnage directed at enemy airfields.⁵⁹ The constant shifts in priority prevented the command from focusing on any one target system. Most operations had to be conducted in unfavorable conditions when the primary target was obscured by haze or cloud or shrouded by the blackout. Irregular, small-scale, dispersed, and difficult to assess, the early raids had the sole merit of forcing large numbers of Germans in northwestern Germany to seek shelter during the summer nights until ways were devised by German civil defense to minimize the time lost because of the alarms.

The large number of German targets and sudden shifts in allocation required a tactical approach that reduced even further any prospect of serious damage. In most cases only twenty to thirty aircraft were dispatched to each location, and only a fraction allocated to each specific target on the assumption that a handful of bombs dropped by five or six aircraft would be enough to do effective damage. Out of eighty-nine attacks made on Hamburg in 1940, fifty-eight were made with fewer than ten aircraft.⁶⁰ Most bombs carried were high explosive; incendiaries were spread in small quantities throughout the attacking force, which eliminated any prospect of a concentrated fire-raising operation. The number of bombers available for sorties declined over the summer months after the losses sustained in the Battle of France and because of the need to divert bomber aircraft to training units. The total number of sorties carried out between June and August was 8,681, but over 1,000 were against targets in France and the Low Countries, and around two-fifths were light bomber sorties. The tonnage of bombs dropped was just 0.9 percent of the tonnage dropped by Bomber Command in the same months in 1944.⁶¹

It soon became clear that the bombs were dropped with no particular accuracy. In the absence of electronic navigation or an effective bombsight, Portal’s claim that bombing would be inherently inaccurate proved entirely justified. A photographic survey of bombing raids on the German aircraft industry in July 1940 found that out of ten attacks only a couple of hangars had been destroyed; out of thirty-one oil installations bombed, only one appeared damaged.⁶² An American eyewitness account of an RAF attempt to bomb a Daimler-Benz factory in Stuttgart highlighted the exceptional efforts bomber pilots made to identify their target, circling the city for half an hour, but reported that not a single one of the bombs they subsequently dropped hit the plant.⁶³ One Bomber Command pilot later wrote that precise objectives in summer 1940 “were pointless” when crews could not even find the city they were supposed to target.⁶⁴ Most RAF bombs fell in the German countryside, but even those deliberately dropped on forests between June and August 1940, following repeated pressure from Churchill to try to burn down German woodlands, failed to ignite a fire. Experiments with setting fire to woods or crops continued for more than a year, until it was finally conceded that even under ideal climatic conditions most incendiaries burnt only the few inches around where they landed.⁶⁵

Little of this problem was due to German defenses. As the British were themselves to find when they tried to combat German night bombing in the opening weeks of the Blitz, little preparation had been made by the German Air Force for defense against nighttime operations. German antiaircraft fire could force enemy aircraft to fly higher at night, but could only hit them mainly by chance, and this despite a formidable array of antiaircraft weapons. By June 1940 there were 3,095 heavy guns, 9,815 light guns, and 4,035 searchlights organized into defensive zones around the vulnerable industrial areas. Guns were moved back from the frontier areas following the defeat of France, strengthening the home defense even more. Yet between January and June 1940 only two aircraft were claimed from antiaircraft fire, and in August and September not a single one. The principal effect was to force the attacking bombers to break formation, exacerbating the existing dispersion of the British effort.⁶⁶ Most of Bomber Command’s problems were self-inflicted. A postwar presentation of the early bombing effort by an official of the British Bombing Survey Unit in 1946 concluded that the forces were too small, the weapons incapable of a high degree of damage, targets could not be found, and too much effort was devoted to subsidiary operations: “great call on Air Force,” ran the lecture notes, “to attack and destroy targets beyond its power.”⁶⁷

Under these circumstances it is perhaps surprising that the offensive was continued at all. Its survival owed much to the confidence of Bomber Command’s new commander in chief, Charles Portal, that bombing would eventually deliver strategic dividends. Portal was a successful career airman who began flying in the First World War and was among the first to drop bombs on German soil. He played a key part in building up RAF strength in the late 1930s as director of organization. He was in charge of training when he was called to command the bombing war in April 1940. He was widely respected—“honest and unprejudiced,” according to one former colleague, “not much small talk,” according to another—and liked by Americans for his straightforward character and shrewd intelligence. A shy man who lunched alone day after day at the same London club, the private man was sealed off from his staff. He was convinced that bombing was a more effective way to wage modern war and remained a staunch defender of the bombing force when he was appointed chief of the air staff in October 1940.⁶⁸ He argued steadily through the summer months that Bomber Command should be allowed to do what it had been prepared for despite all the operational difficulties, and he shielded the force from criticism through the years that followed.

The second factor was the unstinting support of Churchill, whose interest in bombing waxed through the months after the fall of France. He asked to be informed about the operational performance of Bomber Command and intervened in every aspect of its activities, from the supply of bombers and bombs to the plan to burn down forests. His enthusiasm for bombing was a creature of the emergency facing Britain during the summer and autumn of 1940 when British forces had to demonstrate to the United States and to the peoples of occupied Europe that they still had some capacity for offensive action, however limited. It was also necessary to impress the British public that military action against Germany had not been abandoned after the evacuation from Dunkirk in late May and early June.⁶⁹ On July 8, Churchill wrote to the minister of aircraft production, Lord Beaverbrook, summing up his view of how bombing could help decisively in the overthrow of Hitler: “But there is one thing that will bring him back and bring him down, and that is an absolutely devastating, exterminating attack by very heavy bombers from this country upon the Nazi homeland.”⁷⁰ The choice of words was unfortunate, and Churchill’s comment has contributed to the recent debate about whether the bombing campaign from the outset had a genocidal purpose.⁷¹ This is almost certainly not what Churchill had in mind, since he wrote the letter months before the onset of the Blitz (which did prompt in him a rhetorical language of violent retribution). He was also content to reproduce the letter in his later history of the war, when he had had time to reflect on what to leave out, and indeed deliberately omitted much of the story of British bombing and his part in it.⁷² But it did express Churchill’s desperate hope that bombing Germany was perhaps, as RAF leaders had repeatedly asserted, one possible means to compel Hitler to abandon invasion plans, or even to dislocate the German war effort decisively. Harris kept a copy of the Beaverbrook letter with him after the war. “That was the RAF mandate,” he told the biographer Andrew Boyle in 1979, shortly before his death.⁷³

From the point of view of British strategy, approval of bombing was a decision that came at a high price. Bomber Command achieved negligible results against German targets and invited German retaliation. In early September, Hitler finally responded to British attacks by permitting a campaign against London and other cities that dwarfed anything that could be done in return. Between September 1940 and June 1941 more than 57,000 tons of high-explosive and incendiary bombs were dropped, principally on British port cities, which absorbed 85 percent of the tonnage. Around 43,000 people were killed in the ten-month campaign, more than ten times as many as were killed by RAF raids on Germany in the same period. It is possible that the Blitz would have been launched anyway, as British air leaders had expected, but it is also possible that without a British bombing campaign over the summer of 1940, British cities might have been spared the full horrors of the winter of 1940–41.

German Lessons, 1940–41

The onset of heavy German night bombing in September 1940 showed Bomber Command at last what a serious bombing offensive looked like. Attacks were made with hundreds of bomber aircraft concentrated against a single target, while diversionary or nuisance raids were made to confuse the defenses and create widespread disruption. Heavy use of incendiaries contrasted with the British preference for high-explosive bombs, and produced widespread area damage. At first the RAF thought the German campaign was flawed, because it assumed the attacks were designed to terrorize the population. “Notes on the Lessons to Be Learned from German Mistakes,” a survey produced two weeks after the first heavy raid on London, concluded that “the indiscriminate attack of cities is invariably uneconomical,” a view with which German air commanders would have concurred.⁷⁴ But it soon became clear that the pattern was to attack ports, food supplies, and the aircraft industry, and the evident ability of German bombers to attack at will and achieve a relatively high concentration of hits turned the RAF toward the idea of learning lessons from what the Germans got right. After nine months the Air Ministry arrived at the view that regular “Blitz” attacks on German city areas demonstrated the most profitable use to which Bomber Command could be put.⁷⁵ The shift in 1941 and 1942 to a policy of “area” bombing came about not as a result of the poor accuracy achieved in attacks on specific objectives, as is usually suggested, but by copying the Germans.

The German offensive was from this point of view a valuable learning tool, since it was difficult to evaluate clearly what Bomber Command was itself achieving over Germany. From June 1940 onward, Britain was cut off from Europe by German military successes. Until the autumn the RAF relied on hearsay and occasional news reports to form a picture of the effects of British bombing. The sustained German attacks could now be used to assess with greater scientific, technical, and statistical precision exactly what an air raid might achieve. The program of evaluation began almost immediately. In late September 1940 the Research and Experiments Department of the Ministry of Home Security supplied a detailed study of the effects of German bombs on different types of targets—oil storage, gasworks, power stations, aircraft factories, and so on—and arrived at a conclusion that was to have far-reaching consequences for the development of Bomber Command’s offensive. “It is axiomatic,” ran the report, “that fire will always be the optimum agent for the complete destruction of buildings, factories etc.” The department recommended using high-explosive bombs to create the “essential draught conditions” in damaged buildings, followed by heavy incendiary loads, and completed with more high explosive to hamper the enemy emergency services.⁷⁶ The evidence that concentrated use of incendiaries was the most effective form of air assault against large industrial centers gradually emerged as the key lesson to be learned from the experience of the Blitz. It was the seed from which area bombing slowly germinated during the year that followed.

The work of the Research and Experiments Department made an essential contribution to understanding what bombing could achieve. The department was set up in spring 1939 to help the Home Office evaluate the effects of bombing and was run by the former director of the Building Research Station, Dr. Reginald Stradling. He co-opted scientists and statisticians onto the staff, including the zoologist Solly Zuckerman and the physicist (and former pacifist) J. D. Bernal. In November 1941, Bernal established division RE8 to supply the Air Ministry directly with calculations on the effects of German bombs on British cities, the productive loss caused by bombing, and the likely impact of British bombing on German cities.⁷⁷ This work was supplemented by the Air Warfare Analysis Section, which looked at the weight and type of bombs to be dropped, and by the work of the Road Research Laboratory and Building Research Laboratory, both of which helped to estimate the nature of bomb damage and the vulnerability of German building structures.⁷⁸ Bombing research was in its infancy during the Blitz, but it benefited from the experience gained in the 1930s in recruiting senior scientists to work on particular aspects of air warfare.⁷⁹ Although the relationship between scientists and airmen was never formalized, the Air Ministry knew that it needed scientific input not only to provide effective technology (particularly radar and navigation aids) but also to make better sense of operations. Research questions were usually framed by the air staff and the subsequent expert reports circulated to those who needed them; sometimes it was the experts who took the initiative. This was particularly the case with the German-born Oxford physicist Frederick Lindemann (later Lord Cherwell), who was recruited by Churchill in 1939 to form a small Statistical Section in the Admiralty, and then followed Churchill to Downing Street in 1940. Lindemann took a special interest in bombing and was perhaps more responsible than anyone in keeping Churchill abreast of the many problems faced by Bomber Command in its early years.⁸⁰ His Statistical Section began at once in September 1940 to produce accurate figures on the damage inflicted by German bombs and to relate those figures to the density of urban population in different city zones. These figures were then applied to German cities to try to determine the areas where the highest damage could be done in terms of lives lost and houses destroyed.⁸¹ Why Lindemann was so committed to the idea of destroying the country of his birth has never been clear, but he played an important part in deriving conclusions about German bombing and projecting them onto potential German targets.

Of all the lessons drawn from what was thought to be German practice, the possibility of urban destruction was the most important. It was gradually understood that the German intention was to undermine British morale by inflicting concentrated attacks on ports and industrial cities, reducing the will to work by the destruction of services, amenities, and housing, and reducing food supplies. Planners at the Air Ministry described German bombing as the direct opposite of British practice: instead of attempting to hit precise targets, the German Air Force carried out attacks on particular industrial or commercial areas where multiple targets were clustered; German raids were concentrated but too inaccurate for any purpose, it was argued, other than “the ‘blitzkrieg’ of fairly extensive regions.”⁸² Studies of British cities also confirmed that the critical level of damage was inflicted by incendiary bombs dropped in large numbers. Particular attention was paid to the German bombing of London, Coventry, and Liverpool, but special studies were commissioned of the bombing of Hull and Birmingham with a view to understanding how fire combined with high explosive affected areas of different housing and population density. Damage was heaviest in the congested working-class districts, which suggested that these were optimum targets. A draft directive from the Directorate of Bomber Operations in the Air Ministry in June 1941 drew heavily from this research on the Blitz: “The output of the German heavy industry depends almost exclusively on the workers. Continuous and relentless bombing of these workers and their utility services, over a period of time, will inevitably lower their morale, kill a number of them, and thus appreciably reduce their industrial output.”⁸³

These arguments signaled an important change in the way “morale” was interpreted. The politicians, Churchill included, generally understood morale in political terms: heavy pressure from bombing would induce a social and political collapse, perhaps even a revolution. The German attacks on morale were more clearly economic in intent. In May 1941 the Ministry of Economic Warfare, which had been monitoring the ineffective impact of Bomber Command on precise economic objectives in Germany, sent a memorandum recommending that the RAF abandon military targets and focus instead on economic warfare against major industrial concentrations or “whole cities.” The idea stemmed from the effects of German bombing on the British workforce: “British experience leads us to believe that loss of output . . . through absenteeism and other dislocation consequent upon the destruction of workers’ dwellings and shopping centres is likely to be as great as, if not greater than the production loss which we can expect to inflict by heavy damage.”⁸⁴ Although the hope for a political dividend from bombing was not abandoned entirely, particularly by Churchill, the Air Ministry came to view morale as a barometer of productive performance rather than political outlook. The same term was used to cover both, but by the time morale became a specified objective in July 1941, it was used as a description of economic attrition—a form of “industrial blockade”—in which the working-class population was attacked as an abstract factor of production whose deaths, disablement, or absence would have economic consequences.⁸⁵

By the spring of 1941, the arguments in favor of imitating what was thought to be German practice had come to be widely broadcast and from a variety of different sources. The debt owed to German bombing was evident in the choice of the term “Blitz” to describe the nature of the new strategy. In April 1941 a review of bombing policy recommended “carefully planned, concentrated and continuous ‘BLITZ’ attacks delivered on the centre of the working-class area of the German cities and towns.” Notes produced in the Directorate of Bomber Operations in the Air Ministry a month later on the use of the bomber force also stressed “continuous blitz attacks on the densely populated workers [sic] and industrial areas.”⁸⁶ Later in 1941, when calculations were made of the ratio between weight of bombs and expected deaths among German workers, the measurements were given as “1 Coventry,” “2 Coventries,” and so on; an attack on the scale of “4 Coventries” was expected to yield 22,515 German deaths.⁸⁷ It is important to recognize that the emphasis on killing German workers and destroying their milieu was deliberate, not an unintended consequence of bombing factories. In November 1940 a memorandum on bombing policy, almost certainly penned by Harris, asked whether the time had not come to strike in full force “against the people themselves.” In May 1941 the director of Air Intelligence welcomed an attack on “the livelihood, the homes, the cooking, heating, lighting and family life of that section of the population which, in any country, is least mobile and most vulnerable to a general air attack—the working class.” The chiefs of staff in June finally endorsed morale attacks that induced “fear of death and mutilation.”⁸⁸The idea of collateral damage had been turned on its head: instead of the death of workers and the destruction of their housing being a side effect of hitting factories, damage to factories was a collateral effect of destroying working-class neighborhoods.

However, deliberate attacks on densely populated areas aimed at killing workers and disrupting civilian life once again raised awkward moral questions. A further memorandum drafted in May 1941 by the director of bomber operations pointed out that since October 1940, Churchill had freed Bomber Command from having to bomb with discrimination, so that “attacking the workers” was now permissible. “We do not mean by this,” he continued, “that we shall ever profess the German doctrine that terrorism constitutes an effective weapon of war.” Nevertheless, he recommended that no announcement of the policy should be made, and the details of plans for attacking the population should have very limited circulation; in the wrong hands, “all sort of false and misleading deductions might be made.”⁸⁹ In late November 1941, Sir Richard Peirse, the then commander in chief of Bomber Command, addressing a sympathetic audience of the Thirty Club, explained that for almost a year his force had been attacking “the people themselves” intentionally. “I mention this,” he continued, “because, for a long time, the Government for excellent reasons has preferred the world to think that we still held some scruples and attacked only what the humanitarians are pleased to call Military Targets. . . . I can assure you, Gentlemen, that we tolerate no scruples.”⁹⁰

There were also lessons to learn about how a Blitz attack should be carried out. An air staff memorandum on area attack pointed out how unwise it would be “if we fail to pick the brains of an enemy who has had so much experience in developing the required technique.”⁹¹ The method consisted of a high concentration of incendiary bombs dropped in a short period of time with the use of a target-marking force to indicate the urban area to be devastated. The proportion of incendiaries carried by German bombers was known to vary between 30 and 60 percent, concentrated in the first attack groups, while RAF bombers carried between 15 and 30 percent, diluted throughout the force. The critical problem was how to drop enough incendiaries to create fires that ran out of control, which meant smothering an area with firebombs. The attack on the City of London on December 29, 1940, was used as the model. The raid started twenty-eight conflagrations, fifty-one serious fires, 101 medium fires, and 1,286 small ones, and it was this level of assault that could be expected to overwhelm the emergency services.⁹² Detailed evidence from Britain’s other blitzed cities suggested that incendiary bombs had five times the destructive potential of heavy explosive per ton. They were best used, it was suggested by Air Intelligence, against cities with narrow streets and wooden structures. Germany’s old town centers were “ideal targets for large-scale incendiary attack” because German urban areas were denser and taller than their British equivalents. A salvo of 30,000 British four-pound incendiaries dropped in twenty minutes was regarded as a necessary minimum, though much larger quantities were found to be necessary later on. High explosive was needed to reduce the water supply and ventilate the buildings.⁹³ It was realized in the Air Ministry by the summer of 1941 that to maximize the effect of firebomb attack the equivalent of the German Kampfgruppe 100 was required, skilled in navigation so that it could carry out a fire-raising attack that the following bombers could use for their own navigation.⁹⁴ Portal used Kampfgruppe 100 as his example when he suggested in August 1941 to the government scientific adviser, Sir Henry Tizard, the need to move to a target-marking system as soon as possible.⁹⁵

These lessons were learned in the end both slowly and piecemeal. The structure for decision making in the RAF and the Air Ministry made it difficult to change quickly, while there remained honest differences of opinion about the most effective use for bombers. Major changes in policy over bombing required the approval of the chiefs of staff and Churchill’s Defence Committee. They had to be properly formulated and presented to the Air Council and the air staff in the ministry before they could be presented to higher authority. Much of the work on trying to understand German strategy and tactics was dispersed among different departments and usually written up in the first instance by junior staff. This situation improved when a Directorate of Bomber Operations was finally set up under Air Commodore John Baker in September 1940, at the prompting of the then deputy chief of the air staff, Air Marshal Harris. While strategy was decided at the highest level, the operational conduct of Bomber Command was left largely to the discretion of the commander in chief, who could modify or ignore instructions from the air staff with which he disagreed. Bombing policy was also subject to external civilian influence. The Ministry of Economic Warfare under the Labour minister Hugh Dalton and the ad hoc Lloyd Committee, set up in 1939 under Geoffrey Lloyd to deal specifically with the German oil situation, could make recommendations to the politicians that simply bypassed the RAF.⁹⁶ Air policy was not entirely haphazard, but it moved in the first years of war in fits and starts, trapped between exaggerated expectations and a beleaguered reality.

Portal’s successor as commander in chief of Bomber Command in October 1940 was Sir Richard Peirse, a senior career airman, who had been vice chief of staff before his appointment. He was a supporter of precision bombing, though more of a realist about what could be achieved, and not in principle opposed to area bombing. He stuck to the directives that called for attacks on oil, communications, and the aircraft industry, but his force remained small and divided between numerous targets. The number of sorties per month fell steadily, exacerbated by the slow supply of bombers and crew and the deteriorating weather, which grounded bombers on both sides through much of December and January. In September 1940 there were 3,597 sorties, by November only 2,039, and in January 1941 only 1,131. Even by the summer of 1941 the figures were little better than they had been a year before, reaching a peak of 3,989 in July.⁹⁷ The RAF also experienced, like the German Air Force, a high rate of wastage, due chiefly to accidents. The Bomber Command groups had 290 serviceable bombers at the beginning of October 1940, but only 212 at the end of November. Peirse told Portal that for every aircraft shot down by the enemy, he was losing six to accidents.⁹⁸ Part of the problem he attributed to the declining skill of bomber crews, too many of whom were rushed from the Operational Training Units (OTUs) to the frontline squadrons. The decision to accelerate promotion to allow sergeants to pilot aircraft produced evidence, so it was claimed, of “slackness and inefficiency.” A report from 7 Group in January 1941 highlighted poor discipline among recruits from the OTUs: “unpunctuality and absence without leave . . . some of them seem to think they can turn up when they like.”⁹⁹ For many crewmen in the winter of 1940–41 the reality was to fly long, dangerous operations over German territory in poor weather, with inadequate equipment, returning to bomber stations that were still improvised and poorly lit, and in the knowledge that most of the target photographs they took were of somewhere other than the place to which they had been directed.

The poor performance of Bomber Command owed a good deal to the priorities that had been given to home defense during the summer and autumn of 1940. But since this was one of Britain’s few offensive options, its deficiencies were very public. In early November, Churchill complained to Portal that bomb tonnage on Germany was “lamentably small,” given the amount of money and matériel devoted to it: “I wish I could persuade you to realize that there is a great failure of quantitative delivery.” In December 1940 he returned again to attack the “stagnant condition” of the command and its “deplorable” operational performance.¹⁰⁰ Peirse was sensitive to the accusations and continued to insist that his force would be a cardinal factor in reducing Germany’s war economy and will to war by the spring of 1941, while paradoxically explaining that weather and poor training were likely to inhibit anything his force could do.¹⁰¹ Churchill’s frustration contributed to pushing Bomber Command slowly toward a strategy that favored city bombing. He tried unsuccessfully to insist that Bomber Command attack Berlin in October. He told Peirse that he hoped his command would soon be bombing “every ‘Hun corner’ every night.”¹⁰²

In late November 1940 the War Cabinet endorsed a decision to retaliate for the attack on Coventry on November 14–15, when 503 tons were dropped on aero-engine plants, destroying much of the city center at the same time and killing 554 people. It was decided that a German city should be bombed indiscriminately in retaliation, although Coventry was evidently a military-economic target, scattered though the bombing had been. Portal supplied a list of four cities—Hannover, Mannheim, Cologne, and Düsseldorf—and told Peirse to mobilize every aircraft he could, even from the training units. Bombing was to continue all night, replicating the German practice of heavy incendiary attack, followed by high explosive, and then further incendiaries. For an operation code-named Abigail Rachel, Peirse chose Mannheim as his target and attacked it on December 16–17 “based on the experience we have gained from Coventry, Bristol etc.” But because of poor weather only 101 out of a planned 235 aircraft could be sent. Most claimed to have bombed the target, but in fact the advance group of Wellingtons failed to mark the center of the city, while other bombers scattered their loads widely over residential areas. There were thirty-four deaths and 476 houses were destroyed. When Peirse asked whether he could conduct a similar raid against Hannover, Churchill was noncommittal.¹⁰³

The city attacks were not repeated, though not from any moral qualms. The Blitz had finally eroded any serious concern about the morality of bombing the civilians of a state whose air force had killed almost 30,000 British civilians in four months. They were held in abeyance by the striking news given to the cabinet in mid-December that the small effort against German oil targets had probably reduced German supplies by 15 percent. The figure was a gross distortion of reality, as photoreconnaissance intelligence of plants bombed in December made clear, but Portal snatched at the news as a chance to redeem Bomber Command at one of the many critical moments in its survival over the early war years. The air staff worked out that there were enough aircraft to knock out seventeen oil plants and that the attacks could be repeated every four months to ensure that they remained inoperable.¹⁰⁴ The chiefs of staff approved the policy on January 7, 1941, and the War Cabinet a week later, with the rider that in adverse conditions area attacks might be made instead. The decision to focus on a single target made little sense in the light of what had already been learned about the pattern of German bombing, and the failure of the plan was evident within weeks. At the end of February 1941, Peirse had to confess to Portal that he had only been able to attack oil targets on three nights in the whole of January and February; towns had been attacked six times, but most of Bomber Command’s effort had in fact been devoted to German naval and port targets, which were easier to find and hit.¹⁰⁵ The oil plan was a peculiar fantasy given the current technical capability and evident inaccuracy of the bomber force. Its failure was masked by the sudden decision taken by Churchill in early March to focus the effort of Bomber Command entirely on the Battle of the Atlantic to try to break the blockade imposed by German sea and air attacks. Portal lacked Hermann Göring’s political muscle and could not resist the diversion. Naval priorities prevailed, and for four months Bomber Command began a largely fruitless campaign against German submarine pens and warships.

By the time Bomber Command was permitted to return to priorities in Germany, the weight of opinion inside and outside the RAF had consolidated in favor of morale attacks on working-class urban areas. Early in June 1941 the Air Ministry produced a new discussion document rejecting oil as a primary target. What was proposed was a compromise between what remained of the principle of precision and the desire to replicate German area attacks. Using material supplied by the Ministry of Economic Warfare, a concerted attack on railway transport in the Ruhr-Rhineland area was proposed. Precise targets were to be located in city areas so that “‘shorts and overs’ [which constituted most British bombs] will kill.” For most nights, however, it was proposed that the bomber force be used to attack the industrial workforce in the same Ruhr area.¹⁰⁶ Following German practice, it was also suggested that city targets on or near water would be more suitable to make sure that a sufficient proportion of the attacking force could find them. On July 9 the new proposals were issued as a directive to Bomber Command, after the phrase “the morals of the German people” had been altered to read “morale.” A list of suitable railway targets was appended with the caveat that for 75 percent of any month bombers would not be able to see their targets clearly enough for precision; for three-quarters of each month Bomber Command was expected to undertake “heavy, concentrated and continuous attacks of working class and industrial areas.”¹⁰⁷ There was no certainty how long this directive would remain in force. Peirse complained to Portal about the constant changes in priority: “I do not feel I am fully in touch with your ideas. I may be working with you or against you, I am not sure. But it is certainly difficult to work to any plan with this ever-changing background.”¹⁰⁸ Peirse came to accept that winning the bombing war required “attack of the German people themselves,” and this part of the directive remained in place, in one form or another, for the rest of the war.¹⁰⁹

The one lesson that the RAF and the government failed to learn from the German experience was, paradoxically, the reality of relative failure. German bombing did not dislocate the economy seriously, nor did it undermine civilian commitment to the war effort, as the Air Ministry could clearly see. Calculations were made which showed that potential output in the British economy was reduced by no more than 5 percent; even in cities heavily bombed, economic activity was restored to previous levels in between three and eight days.¹¹⁰ It was also difficult to argue that German “morale” would somehow crack if British morale had remained intact after nine months of remorseless assault. When the new directive was shown to the American chiefs of staff at the Argentia meeting between Churchill and Roosevelt in August 1941, they found it hard to reconcile the morale bombing of Germany with the “valorous experience” of the British people under German bombardment.¹¹¹ The RAF could not ignore this paradox. It was resolved by suggesting that the Germans lacked the qualities of endurance and pluck displayed by the British under fire. The general prejudice among senior airmen was that the German people, as one intelligence report put it, “will not stand a quarter of the bombing” dished out to Britain, though there were few sensible grounds for believing it.¹¹² A report in July 1941, following a meeting in Lisbon with American diplomatic personnel from Germany, suggested that the average German worker was a fit target because he displayed a “lack of moral fibre.”¹¹³ An air staff memorandum produced in September 1941 accepted that in the British case bombing tended to stiffen rather than weaken morale, but went on to argue that the Germans had based their campaign on judging the poor morale of their own people. Made of sterner stuff, British morale had not given way, but, the report concluded, “the wheel has gone full cycle, and it has become increasingly clear that one of the most (if not themost) serious chinks in the German armour is the morale of the civil population.”¹¹⁴ The paper concluded with an unambiguous statement of the purpose that now lay behind the British bombing offensive:

The ultimate aim of the attack of a town area is to break the morale of the population which occupies it. To achieve this we must achieve two things: first, we must make the town physically uninhabitable and, secondly, we must make the people conscious of constant personal danger. The immediate aim is, therefore, twofold, namely, to produce:

(i) Destruction, and (ii) The fear of death.

Here was a German lesson to be taught to the Germans.

Bomber Command in Crisis, 1941–42

By the time the July 1941 directive was issued, the war had suddenly changed its character. Heavy German bombing of British cities stopped in early June, and on June 22 as many as 4 million Axis soldiers poured across the Soviet border in the largest invasion in history. That same evening Churchill broadcast to the nation, pledging British support for the Soviet Union against “the bloodthirsty guttersnipe” who had now unleashed war against another suffering people. He announced that he had offered Stalin all technical and economic assistance, but the one military pledge he made was to promise “to bomb Germany by day as well as by night with ever-increasing measure” to give the German people a taste of their own medicine.¹¹⁵ On July 7, Churchill sent a telegram to Stalin explaining that the best Britain could offer as direct military assistance was bombing; this, Churchill continued, would force Germany to divert fighters to the west, and ease the pressure on the Soviet front.¹¹⁶ Churchill hoped privately that the new campaign would prompt Soviet bombers to attack Germany from the east: “A lot of German war industry should be vulnerable especially if we are bombing from the other side.” Stalin replied that he would prefer Britain to open a second front in northern France or Scandinavia.¹¹⁷

Churchill exaggerated what Bomber Command was capable of achieving and misunderstood the nature of Soviet air strategy, which favored ground support over long-range bombing. But Bomber Command used the German-Soviet war as a way to improve its low political stock. On July 21, 1941, Churchill and Attlee were invited to view a demonstration by the heavy bombers that were scheduled to come into large-scale service over the coming year. The party watched as five heavy four-engine bombers flew past at low altitude: a Short Stirling, an Avro Lancaster, a Handley Page Halifax, and two American bombers promised to Britain under the Lend-Lease scheme authorized in March 1941, the Boeing B-17 Flying Fortress and the Consolidated B-24 Liberator. The party was impressed in particular by the Lancaster, but there were reservations in the RAF about the American bombers with their more limited bomb loads.¹¹⁸ The political imperative of supporting the Soviet war effort suited Bomber Command in summer 1941 because it gave the command a prominence that its poor operational record scarcely warranted. Churchill needed bombing as something to trade with Stalin. Later in the war, Air Vice Marshal Richard Peck, in a speech surveying the course of British bombing, reminded his audience that in the summer of 1941 the air forces were given the task of supporting Russia by bombing Germany: “Not everyone has appreciated,” he continued, “the extent to which the bomber offensive was applied to aid the Russian armies.”¹¹⁹

The political imperative masked the operational reality. On June 22, the night of the German invasion, seventy medium bombers raided the north German port of Bremen; it was covered in haze and the bombing was scattered. The following night sixty-two bombers raided Cologne, where a few bombs fell on the city but there were no reported casualties; forty-one bombers raided Düsseldorf with no clear result; twenty-six aircraft attacked Kiel with little effect and one death.¹²⁰ These were no larger than the attacks still being mounted by the exiguous German force left in northern France after the end of the Blitz—raids on Birmingham with 94 and 88 aircraft, on Hull with 78, 64, and 114, on London with 60—and considerably less destructive.¹²¹ Most RAF attacks were still being made on targets on the French coast. On July 7, Churchill complained to Portal that he should stop bombing these Battle of the Atlantic targets and concentrate on “the devastation of the German cities” to take the weight off Russia. The war diary written up at Hitler’s supreme headquarters failed even to mention any of the British raids.¹²²

Summer 1941 was not the first time that bombing had been promoted for political reasons, but the fear that Germany might defeat the Soviet Union and turn back to Britain with all the resources of Eurasia at its disposal gave bombing an added urgency. It also made the operational inadequacy of Bomber Command more obtrusive. In early July, Churchill complained to Lindemann, Portal, and the air minister, Sir Archibald Sinclair, that the bomber force was little larger than it had been the year before, though it was supposed to be “indispensable for victory.”¹²³ High losses and the slow buildup of bomber production had indeed reduced the plans for expansion. There was worse to come. In July 1941, Lindemann asked Bomber Command whether he might investigate bombing accuracy by analyzing photographs taken during operations. This was a project that had only become possible since the early summer. When the war broke out, the RAF had day cameras but none suitable for night photography. Trials were carried out with the standard F.24 camera using a shutter mechanism and a large flash unit released manually through the bomber’s flare chute. When the flare was at maximum intensity, one of the bomber crew had to close the shutter. The result was a complex operation designed to be undertaken at the most dangerous point over the target. Research began on an automatic night camera, but it was not ready until 1942. The force had to make do in 1941 with a simplified camera with no shutter, which produced a poorer image but one regarded as adequate. In December 1940 there were still only thirteen cameras available; Peirse asked for five hundred so that most bombers could carry them. By March 1941 there were seventy-five, by September two hundred.¹²⁴ Taking an effective photograph was always difficult, with the ground obscured by smoke and the camera confused by flares, gunbursts, and searchlights. Pilots disliked the order to keep a level flight path while the picture was taken. Nevertheless, from June 1941 a growing stream of images became available for the first time, interpreted by a special unit set up at RAF Medmenham. Now that a fuller photographic record was possible, Portal willingly agreed to Lindemann’s request, perhaps not fully aware of what the results might show.¹²⁵

Lindemann instructed a young economist on his staff in the Statistical Section, David Bensusan-Butt, to examine 650 photographs taken from 100 raids between June 2 and July 25, 1941. The report was ready by August 18. The analysis showed that in general only one in five of all bomber aircraft sent on a mission reached within five miles of the assigned target; of those recorded as actually bombing, the proportion was one in four over Germany, one in ten over the Ruhr industrial area, and on moonless or hazy nights one in fifteen.¹²⁶ Churchill was alarmed by the revelations: “It is an awful thought,” he wrote to Portal, “that perhaps three-quarters of our bombs go astray.”¹²⁷ The RAF response was, not surprisingly, defensive. Portal pointed out that weather over Germany had been very poor in June and July; that the Butt Report covered only one-tenth of Bomber Command sorties; that inexperienced navigators probably took images too long after the release of the bombs (almost certainly the case, given the difficulty of operating the camera and seeing the bomb burst below); and, above all, that German raids tracked over Britain showed only 24 percent of German bombers reaching the target area. Even Lindemann admitted that conditions had not been ideal for photographic analysis in the summer months.¹²⁸ Portal was no doubt correct to argue that the Butt Report was subject to substantial methodological flaws, but the RAF’s own operational evidence gathered since 1940 had consistently shown a very wide margin of error between what the crews reported and what had actually been bombed. Given Bomber Command’s continued practice of sending raids to two or three cities on the same night, and in relatively small numbers, the aircraft likely to be hitting a particular aiming point in Germany on any one raid would be in single figures.

The Butt Report has generally been regarded as a turning point in the British bomber offensive, but its significance can easily be exaggerated. Peirse had asked the Air Ministry in December 1940 to speed up camera supply so that a proper survey of accuracy and bombing effort could be made.¹²⁹ Detailed examination of photographic evidence had already been carried out by Bomber Command in April 1941, and again in June, each time showing how overoptimistic were the reports of the crews and how wide the margin of error. Exaggerated reports were common to both sides in the bomber war, but the sober reality was well understood by the bomber crews. Robert Kee, a bomber pilot and future historian, later reflected on what his diary entries from late 1941 showed him:

It read pretty depressingly in terms of successful operations. . . . Here is an attempt to bomb Brunswick, hopelessly dark, bombed some incendiaries at what we hoped was Hanover. Düsseldorf also hopeless, bombed searchlight concentration. Kiel, this is three in succession. Kiel, hopeless again, very bad weather. . . . Mannheim, too much cloud, bombed searchlights.¹³⁰

In October the Operational Research Section of Bomber Command, established at Peirse’s request in September under the direction of Dr. Basil Dickens, reviewed accuracy for the three months following the Butt Report. It was found that the average performance was even worse than feared; only 15 percent of aircraft bombed within five miles of the target point.¹³¹

In truth, the Butt Report highlighted just one of the many problems facing the force in the late summer of 1941, important though it was. Losses began to increase substantially as the result of stronger German defenses, placing a heavier burden on a training program that turned out a growing number of crewmen with limited understanding of what was required of them. “The one failing of the whole training system,” recalled a rear gunner, “was that we weren’t told more of what to expect. We just learned it strictly from experience.” Peirse told the Air Ministry that up to 40 percent of the operational squadrons’ work consisted of essential additional training, which resulted in regular accidents. Most of the nonoperational flying was done during the day, which also prepared crews poorly for what to expect of nighttime conditions.¹³² In August, Bomber Command lost 525 aircraft destroyed or severely damaged (a wastage rate of 13 percent, many lost to accidents), but received only 106 replacements. In the following three months a further 578 aircraft were written off, many again on nonoperational flights. Raids carried out on Berlin for political effect had losses of 30 percent.¹³³ Between July and December 1941 the force showed a steady decline in its capability (see table 1.1):

Table 1.1: Bomber Command Statistics, July 1941–February 1942¹³⁴

*Aircraft missing are missing on operations. The number damaged or lost to accidents was higher; figures include all heavy, medium, and light bomber operations.

Source: TNA, AIR 9/150, DBOps to DCAS, September 11, 1941; AIR 22/203, War Room Manual of Bomber Command Operations, 1939–1945, 20–21; AIR 41/41, RAF Narrative, “The RAF in the Bombing Offensive Against Germany: Vol. 3,” App. C, E1.

There are many explanations for the crisis now confronting Bomber Command. The new strategy of attacks on city areas was only possible with better equipment, and the directives failed to take sufficient account of what technology was currently available. The most pressing need was for larger aircraft capable of carrying a much greater tonnage, dropped with greater accuracy. This was an obvious solution, and the pressure to accelerate output and improve navigation came from all sides. Yet the heavy bombers that Churchill had been shown in July were still only available in very small numbers because of persistent problems with technical development, while improved navigation was still at the experimental stage despite more than eighteen months of war. The Stirling and the Halifax made their first sorties in February and March 1941. The Short Stirling Mk I was the only one of Britain’s wartime bombers designed from the start to have four engines. It was first commissioned in 1936, the prototype made its first successful flight in December 1939, and the first aircraft came into service late in 1940. Powered by four Bristol Hercules engines, the Stirling had a top speed of 270 miles per hour and a range of 590 miles with a full bomb load of 14,000 pounds. It had a limited flying ceiling but good defensive capabilities with three powered turrets, yet it was plagued with technical problems that had to be ironed out in 1941 and early 1942. The same was true for the Handley Page Halifax, which also stemmed from a 1936 specification and was originally designed as a twin-engine medium/heavy bomber. In 1937 it was converted to four Rolls-Royce Merlin engines, and the prototype flew in October 1939. The Halifax Mk I was developed rapidly and was in service by November 1940. Its first operation, against the French port of Le Havre, was made in March 1941. It had a low operational ceiling of 18,000 feet, a top speed of 265 miles per hour, and a range of 1,260 miles with a maximum bomb load of 13,000 pounds. The aircraft exhibited persistent development problems, had slow handling characteristics, and took high losses. Output continued because it was difficult to disrupt production schedules already laid down, but it was an unpopular aircraft with bomber crews.¹³⁵

The third heavy bomber, the Avro Lancaster, grew out of another twin-engine development, the Manchester, also first specified in 1936. The Manchester was designed around twin Rolls-Royce Vulture engines, but these proved to be a constant source of technical delays. The prototype flew in July 1939, and the first service aircraft were delivered in November 1940. The first raid was against the French port of Brest in February 1941, but repeated engine failure led to the cancellation of further production and only 209 were built. In late 1940 a Manchester Mk III was produced with four Rolls-Royce Merlin engines. It was renamed the Lancaster, and for Bomber Command it was an unexpected godsend. The Lancaster had a much better performance: a top speed of 287 miles per hour, a ceiling of almost 25,000 feet, and a range of over 1,000 miles even with its heaviest load of 22,000 pounds. The usual load was somewhere between 14,000 and 18,000 pounds and the range correspondingly farther. Its carrying capacity was larger than that of any other bomber used in the European theater, and four or five times that of the standard German medium bombers. Some 6,750 Lancasters were produced during the war, the mainstay of the later force. Unlike the Halifax, the Lancaster had a more modest loss rate (3.92 percent compared with 5.75 percent), absorbed less production effort, carried an average of almost twice as much bomb weight, and was easier to service.¹³⁶ But it only began operations in modest numbers in 1942. In 1941 the small total of heavy bombers dropped only 4,000 tons, against the 31,500 tons dropped by the Wellington medium bombers.¹³⁷

The advent of heavier aircraft would mean that the RAF could take advantage of both a new generation of heavier bombs under development and the rapid expansion of bomb production. The prospect of increasing the aggregate payload was regarded as the critical factor in the offensive, but it had to be postponed until the heavy bomber force became available. The 250- and 500-pound General Purpose bombs were still extensively in use in 1941; larger 1,000-, 2,000-, and 4,000-pound bombs, more suitable for the larger bomber models, were developed during the Blitz and brought into use in small numbers. These Medium Capacity (MC) and High Capacity (HC) bombs had a higher charge-to-weight ratio, a thinner metal shell, and a much greater blast effect. However, they still lacked aluminized explosive, which would have increased that effect more than threefold; only later in the war was Lindemann finally able to persuade the RAF to adopt it. The 4-pound incendiary bomb remained standard equipment but was supplemented by the larger 30-pound firebomb with a blend of phosphorus, rubber, and benzol gel, 400,000 of which were ordered in June 1941 and 3 million used by the end of the war.¹³⁸ All these bombs became available in quantities too large for the existing bomber force to use. In April 1941, 12 million incendiaries were ordered for the rest of the year and 36 million for 1942; because of magnesium shortages, however, output was only 2.2 million in the nine months of 1941 and 11.8 million in 1942, but these figures were more than enough for a force not yet converted fully to mass incendiary bombing. By the end of 1941 there was a surplus of more than 2 million bombs, and monthly production was double monthly expenditure.¹³⁹ By the summer of 1941 around 11,000 tons of high-explosive bombs were being produced and filled every month, though Bomber Command had dropped an average of just 948 tons a month between January and April and averaged only 1,884 tons on Germany between July and December. In October 1941 there were unused stocks of 121,000 tons of bombs.¹⁴⁰ This was the reverse of the German problem during the Blitz, when there had been the airplanes but insufficient filled bombs; the RAF had the bombs but not enough heavy and medium bombers to use them.

Both the new aircraft and the new bombs were slow to join Bomber Command in any significant numbers. Only 41 heavy bombers were produced in 1940 and 498 in 1941, compared with an output of 4,703 medium bombers.¹⁴¹ These were modest figures against the plans drawn up in the spring of 1941 to create a force of 4,000 heavy bombers by the spring of 1943. Bomber production had taken a backseat during the summer and autumn of 1940 when priority went to fighter aircraft for defense against German raids. The minister of aircraft production, Lord Beaverbrook, was later blamed by the Air Ministry for trying to kill off “the Big Bomber programme,” but this ignored the serious technical problems encountered in trying to develop and get into service large and complex aircraft in a matter of months.¹⁴² In May 1941, Portal informed Beaverbrook’s successor, John Moore-Brabazon, that he did not want any further heavy bombers developed during the war because of the long lead time between designing a bomber and seeing it into service.¹⁴³ The “4,000 Plan” was always unrealistic. It called for production of at least 1,000 bombers a month over a two-year period, more than twice the number produced during 1941 and 1942. It was already evident by the summer of 1941 that bomber production had hit a serious bottleneck. The RAF pinned its hopes on being able to persuade the United States to make up the shortfall.

The efforts to get America to solve Britain’s bomber crisis went back to the early spring of 1941, when Lend-Lease was finally approved. The RAF delegation in Washington had the challenging task of persuading the American service chiefs to accept the transfer of substantial quantities of modern aircraft, and in particular heavy bombers, from their own rearmament program. Air Vice Marshal Slessor negotiated the aircraft requirements with the Army Air Corps, commanded by General Henry “Hap” (for “Happy”) Arnold. The American offers were enshrined in what became known as the Slessor-Arnold Agreement, a generous commitment, subject to circumstances, to supply Britain on a 50–50 basis from all American aircraft production.¹⁴⁴ The agreement failed, however, to address the problem of the heavy bomber, where Britain’s deficiency was most marked and American output still in its infancy. Arnold visited Britain in April 1941 and was told that the British aircraft industry could not produce more than 500 of the 1,000 heavy bombers needed each month. The United States was asked to fill the gap. Arnold agreed that up to four-fifths of American heavy bomber output could go to Bomber Command by the summer of 1942, but this would consist of fewer than 800 aircraft. By then it was evident that the American air force would renege on the original agreement as relations with Japan deteriorated. At the staff discussions at the Churchill-Roosevelt Argentia summit at Placentia Bay in August 1941, Arnold refused to confirm the American offer. In his diary he noted, “What the British want—my God what a list and what things—no promises.”¹⁴⁵ During September the full extent of American withdrawal from the initial Slessor Agreement became clear. The bombers destined for Britain had been fitted with the Norden bombsight, which was still embargoed for British use, and ensured that the bombers could not be released to the RAF. Instead of the 800 bombers expected, the British were granted 238 with no promise of any further deliveries beyond July 1942. It marked the end of the Slessor Agreement and the end of any prospect of developing a force of 4,000 bombers.

The most urgent problem facing Bomber Command was the search for some form of electronic aid for navigation, without which even larger numbers of bombers would still have restricted striking power. In the summer of 1941 the problem was not simply the failure to hit a precise industrial or railway target, but the inability, under conditions of night, poor weather, and German defenses, to find an entire city. Given that these failures almost nullified what Bomber Command was trying to do, the long period that elapsed in trying to find appropriate tactics or technology is difficult to explain. The technology itself was not exotic, and the capacity to interfere with German electronic navigation in the winter of 1940–41 made evident that British science was capable of replicating German practice. The Telecommunications Research Establishment had begun work on a system using radio pulse transmitters in 1938, known as G (for Grid), but usually described as Gee. The system worked by sending pulses from three ground stations that could be measured on a cathode-ray tube carried in the receiving aircraft; where the coordinates intersected it was possible to estimate to between a mile and six miles the aircraft’s position. Like the German system, it had limited range and was less accurate the farther away from the ground stations the aircraft was. It worked generally no farther than western Germany. The system was shown to Bomber Command in October 1940, and service trials began in May 1941. The first experimental operation using Gee was conducted by two Wellington bombers on August 11, 1941, but one crashed on German soil. The delay in introducing Gee was partly a result of delays in the production of one of its vacuum tubes, but the main problem was the argument between those who favored putting Gee in a small number of target-finding aircraft, which would lead in the rest of the force, and those who argued that it was something that should be made widely available for the benefit of all. This was to become a central conflict in deciding the best tactics for attacking German cities, and it undermined efforts to develop a more appropriate operational system more rapidly. The use of Gee was postponed until enough sets were available to supply much of the command; its first operational use was not until March 8, 1942.¹⁴⁶

The arguments over the introduction of Gee also involved the best tactics to adopt to achieve Bomber Command’s new objectives and to counter the threat posed by German antiaircraft defenses. Since the summer of 1940, when the German Air Force had relied principally on antiaircraft fire, a more sophisticated defensive system had been constructed combining antiaircraft fire, night fighters, searchlights, and radar. The original air defense system, like that of the British, had been based on the assumption that attacks would come by day. The German defenders soon realized that the pattern of British bombing was difficult to predict. A few daylight raids were made, but most raids were small night attacks defined because of their modest scale as nuisance raids (Störangriffe), whose object, it was assumed, was to intimidate the population and disturb the rhythm of industrial labor. Then came heavier raids in the spring of 1941, again scattered and unpredictable but deliberately directed, so the German authorities believed, against “open cities and residential areas” as simple terror attacks.¹⁴⁷ Night attacks meant that antiaircraft fire, without radar assistance, was effectively blind. The decentralized pattern of British raiding made it difficult to know what to protect. German air observation posts were set up around fifteen to twenty kilometers from predicted target areas, but nighttime conditions reduced the prospect of accurate information. The numerous sound detectors used in conjunction with searchlights were found to be vulnerable to the British tactic of throttling back the engines to dampen the noise as aircraft approached a potential danger zone. (British crews also believed that throwing empty milk bottles or beer bottles out of their aircraft confused enemy equipment. The “whistling bottle” was said to interfere with sound location and trigger the searchlights to switch off.)¹⁴⁸ For the German side a concerted defense was difficult to mount, because RAF bombers failed to damage essential war-economic targets, which were guarded by “air defense strongpoints,” and instead scattered their loads over an extended area with few evident objectives. The Butt Report could essentially have been written by the Germans months earlier.¹⁴⁹

On March 3, 1941, the German Air Force established a new command system to cope with the British offensive. General Hubert Weise was appointed Luftwaffenbefehlshaber Mitte (air force commander, center) with the task of constructing an effective air defense wall around northern Germany. He centralized air defense by taking over the defensive functions of the Luftgaue (air regions) in northern, western, and central Germany. On May 1, 1941, he set up the first dedicated night-fighter organization under General Josef Kammhuber as Jagdführer Mitte (fighter leader, center) and integrated it with the searchlight and antiaircraft artillery batteries deployed in northern Germany and the Low Countries. A “Kammhuber Line”—generally known in German as Himmelbett (heavenly bed)—was constructed from the Swiss border through the Belgian city of Liège to the German-Danish border, consisting of a series of map “boxes” in each of which a small number of fighters were controlled by a new and improved radar, code-named Würzburg.¹⁵⁰ Only one fighter could be controlled at a time, but once a bomber had been identified, it became with practice easier to direct a fighter to combat position. The night fighters were not yet fitted with AI (Airborne Interception) radar, like British night fighters. But the German version, code-named Lichtenstein, was in the process of development and was finally installed in 1942, though it was not popular with pilots, who assumed the large external antennae would reduce performance. The searchlights were numerous and powerful, but it was found they were wrongly positioned. From mid-1941 they were spaced out at least three kilometers apart to ensure a better prospect of trapping a bomber overhead. The antiaircraft batteries were gradually supplied with the new Würzburg radar, which, like the British experience with antiaircraft radar, proved difficult to operate with poorly trained personnel and was prone to technical problems. As radar-guided fire improved, the batteries found the supply of radar too slow. By the spring of 1942 only one-third of antiaircraft guns had the new apparatus.¹⁵¹

The fighters worked in two distinct ways. The first echelon engaged in what was called “dark” night fighting, using radar-equipped ground controllers to guide them to their target; behind the night-fighter boxes was a line of searchlights, soon to have their own radar guidance system, which was used by a second echelon of night fighters for “light” fighting against bombers trapped in a searchlight cone. No dedicated night fighter had been developed before the start of British bombing, but the Junkers Ju88, the Messerschmitt Me110, and the Dornier Do17 (later Do217) were converted to the role in 1940 and formed the mainstay of the force thereafter. The night-fighter force had expanded by the start of 1942 to four groups totaling 265 aircraft, a modest fraction of the total German Air Force establishment. The British tactic of allowing bomber crews to work out their own route to the target meant that the raiding group became spread out in area and time, making it easier for each German night fighter to locate and destroy them in their individual boxes. By September 1941, night fighters assisted by searchlights had claimed 325 enemy aircraft, while “dark” night fighting added a further 50.¹⁵² Antiaircraft fire claimed 439 aircraft shot down between January and September 1941, though many of these, if true, were from daylight operations mounted by other RAF commands.

The steady increase in losses might well have pushed Bomber Command to adopt new tactics. The decision to focus on incendiary bombing of urban areas ought to have encouraged a tactical shift to larger and more concentrated raids. The advantages were obvious: the concentration of the bomber stream would mean that the individual fighter boxes in the Himmelbett line and the searchlight wall behind them would be swamped; most bombers would be through the line and to relative safety until they reached one of the inland gun belts. Above all, tight formation and a bomber stream would allow a raiding group to drop all its bombs in a short period of time, maximizing their impact and reducing bomber casualties.¹⁵³ Opinion in the Air Ministry and the air staff nevertheless remained divided. Peirse favored retaining the loose, decentralized formations and encouraging the crews to find the best way to their target and back. A tighter formation, it was claimed, would place a heavier burden on pilots, while it would become easy prey for the “dark” night fighters waiting in the Kammhuber Line. Bomber Command had reached an impasse, exaggerating the threat from the German defenses, yet incapable of responding creatively to the new strategic imperatives.

Peirse’s lackluster command finally produced a growing chorus of criticism. The Directorate of Bomber Operations insisted that Bomber Command begin serious operational preparations for large-scale incendiary attacks on enemy cities. Assessments were produced by Air Intelligence of the degree of necessary concentration based on German practice. The Air Warfare Analysis Section tested the possible effects of heavy salvos of incendiaries on a large-scale map of the City of Westminster to see what damage might be done. Around 100,000 incendiary bombs were now considered a suitable load to begin a major conflagration. Zone maps of German cities were drawn up showing the most densely populated residential areas (Zones 1 and 2A), the suburban areas (Zones 2b and 3), and the outer industrial areas (Zone 4), with recommendations to deliver the maximum bomb load on the two central zones where large numbers of workers were packed together and to leave the industrial areas alone. In October, Peirse was sent detailed instructions on carrying out an experimental incendiary raid on a German city. The subsequent raid on Nuremberg on the night of October 14–15 proved an inauspicious start: most aircraft bombed a small town outside Nuremberg and only one Stirling hit targets in the city, injuring six people. No major fires were started.¹⁵⁴

The most dangerous criticism came from the top. In response to a paper from Portal in late September 1941 spelling out the long-term plan for 4,000 bombers, Churchill replied, “It is very disputable whether bombing by itself will be a decisive factor in the present war. On the contrary, all that we have learnt since the war began shows that its effects, both physical and moral, are greatly exaggerated.”¹⁵⁵ Portal objected that he saw no reason to regard the bomber “as a weapon of declining importance,” but went on to ask Churchill whether the RAF should now be looking for a new strategic concept. Churchill’s reply in early October was equivocal. On the one hand he assured Portal that bombing was still a strategic priority, but on the other he played down the likelihood of a satisfactory strategic outcome:

I deprecate, however, placing unbounded confidence in this means of attack, and still more expressing that confidence in terms of arithmetic. . . . Even if all the towns of Germany were rendered uninhabitable, it does not follow that the military control would be weakened or even that war industry could not be carried on. The Air Staff would make a mistake to put their claim too high.¹⁵⁶

This was the start of Churchill’s growing disillusionment with what bombing could deliver. His initial enthusiasm had been based on a very limited understanding of what bombers were currently capable of achieving. As a politician he was interested in the prospect that air attack might provoke a political reaction in Germany, but the erratic intelligence available suggested that bombing had done very little to undermine German war willingness, while the clearer evidence nearer home showed that the British political system and social structure had survived intact. Morale was now viewed by the RAF less as a means of political pressure, more as a war of economic and social attrition, or, as Portal put it, “interference with all that goes to make up the general activity of a community.” But to Churchill, who had imagined a more immediate and politically significant effect from bombing, the idea of long-term and unpredictable attrition was an unexciting prospect.

Peirse made one final effort to redeem his reputation and that of his force. On the night of November 7–8, 1941, he marshaled the largest force yet sent out on operations over Germany, some 392 aircraft, including 43 heavy bombers. The weather forecast was poor but he persisted with the operation. The chief target was Berlin, but of the 169 bombers sent there, only 73 reached the capital, where they distributed their bomb loads with very limited effect. Only fourteen houses were destroyed, nine people killed, and thirty-two injured. Other bombers attacked Cologne, which suffered five deaths and two houses destroyed, and Mannheim, where no bombs fell at all. During the night 37 bombers were lost, more than 9 percent of the force; for the task force sent to Berlin the loss rate was 12.4 percent. One squadron recorded in its diary that the mission was “practically abortive.”¹⁵⁷ Berlin was not bombed again until January 1943. Peirse was summoned to see Churchill on the following day and told to suspend the offensive over the winter to conserve his shrinking force. Small raids were carried out when possible, but the assault on morale ordered in the summer of 1941 effectively came to an end with little achieved. The air staff investigated the Berlin raid and concluded that Peirse had been negligent in sending out his force in the knowledge that high winds, storms, and icing would be met by the crews. The decision was taken in December to replace him, and he was finally removed in early January 1942 after Churchill had been shown the documents on the disastrous Berlin raid. On January 8, Peirse was appointed to command Allied air forces in Asia, facing the Japanese. Air Vice Marshal John Baldwin, commander of 3 Group, became his temporary replacement until a new commander in chief was in post.¹⁵⁸

Bomber Command found itself in a state of limbo in the last months of 1941 and the first two months of 1942. The crews were only too aware of the crisis surrounding their commander in chief and the failures of the force. Over 3,000 had been casualties during 1941. In December the Directorate of Bomber Operations investigated the views of the group commanders about the state of the force and found evidence of a feeling of “hopelessness and ineffectiveness” among the operational units, largely on account of the difficulties in navigating and target marking. When they found a target, the report continued, “they stumbled on it more by luck than judgement.”¹⁵⁹ The overwhelming evidence that British raids were still dispersed and ineffective exposed Bomber Command to close scrutiny by the chiefs of staff. The talk in the interregnum imposed by Peirse’s redeployment was about the possibility of winding up the offensive. In a note on “Use of the Bomber Force” drafted early in 1942, the government scientist Patrick Blackett speculated that with a few more reverses the navy and army might insist on the “dismemberment of the Air Force as a unit.”¹⁶⁰ Sir Stafford Cripps, Lord Privy Seal, told the House of Commons late in February 1942, winding up a debate on the current strategic situation, that bombing strategy was among the things under consideration: “The Government are fully aware of the other uses to which our resources could be put.”¹⁶¹ The day before this speech the new deputy director of bomber operations, Group Captain Sydney Bufton, fresh from command of a bomber squadron and a champion of concentration and target marking, warned his superior of the situation now faced by the command:

At the present time there is a great deal of criticism of our strategic bombing offensive. This is being voiced not only in Army and Navy circles and in Parliament, but also more generally by members of the public. The criticism cannot be countered by promises of results which we expect to obtain in the future, and rightly cannot be met by evidence of any decisive results which our bomber force has achieved in the past. These results so far have been nebulous, inconsistent and indecisive.¹⁶²

One week before this a new commander in chief had been appointed to Bomber Command—Air Marshal Sir Arthur Harris.

Harris and the Americans

Harris was in Washington on the morning the Japanese navy bombed Pearl Harbor, December 7, 1941. He had been sent in July as a member of the delegation negotiating for American aircraft deliveries to the RAF. His telegrams back to London said much about his personality. In September he dismissed the prospect of American belligerency—“these people are not going to fight . . . they have nothing to fight with”—and thought they engaged in “plain double cross” in reducing aircraft allocations to Britain.¹⁶³ Harris complained to Air Chief Marshal Wilfrid Freeman, vice chief of the air staff, about how hard it was to carry out missionary work “with a people so arrogant as to their own ability and infallibility as to be comparable only to the Jews and the Roman Catholics.” The problem, Harris continued, was the American conviction “of their own superiority and super efficiency—and of our mental, physical and moral decrepitude.”¹⁶⁴ During the morning of December 8 he was summoned to see Henry Stimson, Roosevelt’s secretary of war, and Robert Lovett, assistant secretary of war for the Army Air Forces, to discuss supplies for Britain in the wake of the Japanese attack. “They were dazed,” Harris wrote to Portal, “and Stimson himself hardly able to speak.” The American politicians asked Harris to give back at once 250 aircraft already supplied to the RAF so that they could defend Hawaii. Harris telegrammed Portal for urgent instructions about what to “save from the wreck if wreck is unavoidable.”¹⁶⁵ Two weeks later Portal arrived in Washington to attend the first major wartime conference between Roosevelt, Churchill, and the Allied military chiefs. During the first week of January 1942, Portal told Harris that he wanted him to replace Peirse; Harris agreed and Churchill approved the appointment, which was made official from February 22, 1942, after Harris had sailed back to England.¹⁶⁶

The Japanese attack promised to transform the bombing war more certainly than the German invasion of the Soviet Union, because it brought into the conflict a power capable of colossal military output and an air force already committed to the concept of long-range strategic bombing. Yet the outcome of the Arcadia Conference in Washington between December 22, 1941, and January 14, 1942, left the bomber offensive as one small part of the wider strategic objectives agreed to between the two leaders. On the way to the conference Churchill cabled to Roosevelt a long memorandum on Allied strategy, which included a short passage on bombing asking the United States to send at least twenty bomber squadrons to help boost Britain’s offensive. “Our own bomber programme,” he added, “has fallen short of our hopes.”¹⁶⁷ During the twelve meetings between the British and American teams, however, bombing was discussed only once, when the U.S. side insisted that their bombers would only be manned by American crews, confirming that Britain would get no further heavy bombers from American production.¹⁶⁸ In the list of strategic priorities, bombing was included as a contribution to item “(d)”: “wearing down and undermining German resistance by air bombardment, blockade, subversive activities and propaganda.”¹⁶⁹ On January 7, Churchill, briefly in Florida for his health, summed up what the two men had agreed. Bombing hardly featured except for Churchill’s fears that the Blitz might be renewed. He assumed that most American airpower, including the bombers, would have to focus on the Pacific War for the coming year.¹⁷⁰ The role of bombing in Allied strategy for the foreseeable future was regarded as modest and peripheral.

The Americans were not unprepared for involvement in the European bombing war. Indeed, as early as 1935 American airmen had begun thinking about building bomber aircraft that could fly across the Atlantic to project long-distance airpower against a hostile state. Writing in 1939, General Arnold, chief of the Army Air Corps, addressed the question “Can We Be Bombed?” and concluded that the answer was yes: “We are vulnerable to bombing. Such bombing is feasible.”¹⁷¹ On the day the German army invaded in the west, May 10, 1940, Arnold proposed the development of a new bomber with a 4,000-mile radius of action capable of attacking European ports to disrupt “the launching of expeditionary forces against the Western hemisphere.”¹⁷² During 1940 and 1941 the U.S. Army Air Forces had been instructed by Arnold to collect detailed intelligence information on German industrial and economic targets, much of it supplied by the British Air Ministry. Consistent with air force thinking, this material was designed to support the idea that attacks against the vulnerable industrial web would unravel the enemy’s capacity to make war. When Roosevelt instructed the American armed forces to draw up a “Victory Program” in the summer of 1941, the air force was asked to prepare a plan of the resources needed to fulfill a strategic air campaign against Germany. In six stifling days in Washington in August 1941, a team assembled by Lieutenant Colonel Harold George worked day and night to produce a detailed plan for a putative offensive. The result was AWPD-1, a detailed survey of 154 German targets in three key target areas: electric power, fuel oils, and communications. Production of 11,800 heavy bombers, to be employed on precision bombing in daylight, was considered sufficient for the task, though the air force currently had only a few hundred. Unlike the RAF, which had never embraced a serious counterforce strategy, the American planners—like the German Air Force in 1940—assumed that enemy airpower would be an essential intermediate target, whose destruction would make the obliteration of the primary objectives possible.¹⁷³ Morale was not considered a useful target and was not included on the list. Again unlike the RAF, the American planners did not argue about the legality of bombing urban targets or hitting civilians.¹⁷⁴ The German economic web, with its vital centers, was treated as an abstraction; the metaphor of the “social body” created a language that distanced those planning the bombing from the reality of civilian deaths.

Roosevelt was pleased with the plan. He had supported American air rearmament steadily since 1938 and in spring 1941 authorized a schedule of production that included 500 heavy four-engine bombers a month.¹⁷⁵ Despite his appeal in September 1939 to avoid bombing civilians, he shared Churchill’s uncritical view that bombing was a possible war winner in the face of German aggression. He had a long-standing personal hostility to Germany and the Germans, and an abhorrence of Hitlerism. American reports sent back to Washington at the start of the war in Poland highlighted the ruthless destruction of Polish towns from the air and underlined how shallow had been Hitler’s positive response to Roosevelt’s plea.¹⁷⁶ Roosevelt, like Churchill, proved susceptible to the extravagant fears of German airpower and scientific ingenuity painted by unreliable intelligence. Since the Munich crisis, when the president had advocated to his cabinet the idea that European states should bomb Germany in concert to halt Hitler’s aggression, Roosevelt had retained extravagant notions of what airpower might achieve. His special adviser Harry Hopkins noted in August 1941 Roosevelt’s conviction that bombing was “the only means of gaining a victory.”¹⁷⁷ In the United States as in Britain, the air forces became the unexpected beneficiaries of political support at the highest level, without which the complaints and blandishments of the other services would have been more difficult to resist.

The sudden coming of war with Japan, Germany, and Italy in December 1941 nevertheless exposed how flimsy were the American preparations so far. The United States possessed no strategic bomber force and had to build one from scratch. Most of the small number of B-17 Flying Fortress bombers were stationed in Hawaii and the Philippines to protect against possible Japanese aggression. A real fear was the possibility of air attack either on the eastern seaboard from German bases or on the Pacific coast from Japanese carrier aircraft. Civil defense preparations were already in place, organized by the Office of Civilian Defense set up in May 1941, and were activated at once in vulnerable areas on the outbreak of war. War-essential factories in coastal areas were ordered to begin a program of camouflage and to black out windows with black paper and layers of opaque paint. All aircraft plants, even in areas not obviously exposed to risk, had to prepare concealment and obscurement plans, while the American Chemical Warfare Service developed units to distribute a five-mile smoke screen around vulnerable targets.¹⁷⁸ Air-raid wardens patrolled Washington streets to enforce the blackout drills, and in June 1942 it was decided that coastal cities should operate a permanent “dim-out” against the threat of air raids, with veiled vehicle lighting and low-visibility streetlamps.¹⁷⁹

Strict civil defense instructions were issued for the control of traffic during air-raid alerts, and in August 1942 the Federal Works Agency produced a 173-page air-raid protection code, covering every subject from behavior in air-raid shelters to compulsory fire watching. As in Britain, dispersed sheltering was favored, with no more than fifty people in any one shelter, but unlike the European experience, basements and cellars were regarded as hazardous. In tall buildings with a reinforced skeleton it was recommended that shelters should be constructed on the upper floors, though not on the top floor; the exact position could be calculated by working out the load-bearing properties of the ceiling once debris had collapsed onto it. The structure of the air-raid precautions system resembled that of the British, with volunteer auxiliary firemen, fire-watching units, first-aid volunteers, decontamination, and rescue battalions.¹⁸⁰

The Office of Civilian Defense, run in 1942 by James Landis, a Harvard law professor, was responsible for organizing the volunteer and full-time personnel. Thousands of Americans spent much of the war period engaged in drills and practices that made increasingly less sense as the war went on, though continued speculation about the possibility of German bombing kept the civil defense force in being. In May 1943 there were fears after the German defeat in Tunisia that Hitler would seek a propaganda coup by launching bombing aircraft from German submarines against East Coast cities. The gas threat was also an ever-present anxiety. In June 1943, Roosevelt announced that any use of gas by the Axis powers would provoke immediate retaliation “throughout the whole extent of the territory” of the enemy state.¹⁸¹ As in Europe, civil defense was also designed to get the American public to identify with the war effort as democratic participants; since American bombing was predicated on attacking the social and economic web of the enemy, the American people could now be viewed as an active part of the war. The Civilian Defense journal was deliberately titled Civilian Front to reflect war in the modern age. This rationale was explained by Landis in an editorial in 1943:

Civilian Defense is more than insurance for ourselves. It is a military duty. Modern war is not confined to battle lines. It is all the arms, resources and production of one people against all the arms, resources and production of another. A food warehouse or a machine tool plant 3,000 miles from the spot where the land forces are locked in combat is as legitimate a military objective as a pillbox on the battle line. . . . That is our assignment and it is a military assignment as definite as that given an armed task force ordered to take and hold an enemy position.¹⁸²

Imagination rather than reality shaped these views as they had in prewar Europe, but in the eyes of the American public they helped to legitimize American bombing of German urban targets when this began early in 1943.

The American bombing campaign took a long time to evolve. The Eighth Air Force was activated on January 28, 1942, in Savannah, Georgia, under the initial command of Colonel Asa Duncan. Because of the commitment made at the Arcadia Conference to mount an invasion of Europe, or possibly North Africa, during the coming year, the Eighth Air Force was expected to play an air support role as well as prepare for strategic operations from airfields in England. Arnold sent Colonel Ira Eaker to Britain to establish contact with Bomber Command and to learn about its operations. Eaker met Harris in Washington before they both left in late February, and an immediate rapport was established between the two men, despite the differences in their personality: Eaker was diffident and earnest, Harris opinionated and brusque. Eaker arrived in London on February 21, a day before Harris assumed command at High Wycombe. After a period staying with Harris, Eaker in April set up an American headquarters in the nearby Wycombe Abbey School for Girls, after the pupils had been forced to leave. Code-named Pinetree, the site became the wartime command center for the Eighth Bomber Command, with Eaker (now brigadier general) as its commander, but as yet with no aircraft or personnel.¹⁸³ It was made clear from the start that the American force was not under RAF command, though it was expected to learn a great deal from British experience. Eaker wrote to Harris later in the summer that he regarded him as “the senior member in our firm—the older brother in our bomber team.”¹⁸⁴ Arnold appointed Major General Carl Spaatz, one of the most senior American airmen, as overall commander of the Eighth Air Force, including its fighter, reconnaissance, and service branches, but Spaatz remained in the United States for five months while the air force organization was established, the training programs initiated, and the service and procurement system organized. He finally took over from Duncan on May 10, 1942.¹⁸⁵ Both Eaker and Spaatz were selected by Arnold because they had shared with him the struggle to establish American air forces during the years of isolation, and both supported his view of the strategic importance of independent airpower. Spaatz had visited Britain in July and August 1940 and had been unimpressed by what appeared to be indiscriminate German night bombing, but impressed by the possibility that daylight bombing in close formation could afford sufficient protection against fighter penetration and achieve greater bombing precision.¹⁸⁶ These were lessons that governed the operational and tactical development of the American bomber force in 1942 and 1943.

The first echelon of American air force personnel arrived on May 11, a second one a week later, but the first 180 aircraft only arrived in mid-July, and just 40 were heavy bombers. American planning, unlike British, had to be based from the start on the assumption that an invasion might take place somewhere in 1942, so that most of the initial aircraft deliveries were of light or medium bombers for army support roles at the expense of a strategic bombing capability.¹⁸⁷ Until the decision taken by Churchill and Roosevelt in July 1942, against strong American objection, to undertake a limited invasion in North Africa (code-named Torch), American air planning had to be based on the assumption that a landing in France would be undertaken before October. The result would have been to divert American aircraft almost entirely to a role in support of surface forces, and this possibility compromised the early efforts to turn the Eighth Air Force into a principally strategic force. The prospect of an invasion of Europe (code-named Sledgehammer) also prompted the head of the American military mission (Special Observer Group) appointed in the spring of 1941, Major General James Chaney, to insist that Eaker and Spaatz integrate with his organization rather than set up a new independent command. The jurisdictional battle was resolved only because Eaker refused to be based in London under Chaney’s close supervision. The arguments over invasion also affected relations with the British, who tried to insist for the sake of operational efficiency that American fighter aircraft be absorbed into RAF Fighter Command and that at least 400 American heavy bombers be given in the first instance to Bomber Command, which could utilize them immediately. Arnold visited London in late May 1942 and succeeded in reducing this demand to a tentative 54 but could not promise that American-flown bombers would be in action much before the autumn.¹⁸⁸ He found London very different from his last visit during the Blitz: “Men, women and children have lost that expression of dreaded expectancy,” he wrote in his diary, “they have a cheerful look on their faces. . . . Pianos are playing, men are whistling. London is changed.”¹⁸⁹ He returned to Washington with enough achieved to prevent the further emasculation of the Eighth Air Force’s still nonexistent capability.

Harris arrived in England shortly before Eaker and moved at once to High Wycombe to take up the command left in abeyance by the sacking of Peirse. He remained the longest-serving bomber commander of the war. He began his air career in the First World War when he left Rhodesia, where he had emigrated as an adventurous teenager in 1910, to join the Royal Flying Corps. He became a major and ended a dramatic operational career in 1918 as a training officer. He remained in the fledgling RAF and saw active service in the Middle East, where he helped to define “air policing” methods by using light bombers to intimidate recalcitrant populations in Iraq and Palestine. He held high office in the Air Ministry in the 1930s, and played a key part in planning what was known as the “Ideal Bomber” (the Lancaster was a distant descendant). In 1939 he became commander of 5 Group, Bomber Command, before becoming Portal’s deputy when he was appointed chief of staff in October 1940. From June 1941, Harris was in Washington, absent from the ongoing arguments about air tactics and the diminishing impact of the command, though not unaware of the problems.¹⁹⁰

On most accounts Harris was judged an effective officer, and he impressed many of those who met him with a shrewd intelligence and a mordant wit. He established a working relationship—though not always frictionless—with Churchill and the American air leaders. He gave the impression of a straightforward, no-nonsense personality, who spoke his mind and changed it little. He had scant sympathy with those of his colleagues or his men who displayed any weakness. The crews who followed behind the target markers he termed “rabbits”; the crewmen who expressed doubts about bombing civilians were “weaker sisters.” The civilian critics of bombing were “Fifth Columnists,” his junior critics at the ministry simply “impertinent.”¹⁹¹ His blunt talk became a hallmark of his relations with anyone who crossed him, however senior. In April 1942, Wilfrid Freeman, then vice chief of the air staff, told Harris after a typically robust exchange that he had spent years getting used to his “truculent style, loose expression and flamboyant hyperbole,” but could still be surprised by the level of verbal injury Harris was willing to inflict. So fearsome was Harris’s reputation that when in early 1947 the Air Ministry proposed a conference on the wartime bombing campaign based on the critical report of the British Bombing Survey Unit, Claude Pelly wrote to his coauthor Solly Zuckerman that they needed adequate warning if Harris decided to come from retirement in South Africa so that they could “make the best of a couple of Continents’ start. Iceland or Southern Pacific?”¹⁹²

Harris had two important prejudices that colored his entire period as commander in chief. He held an exceptional hostility to the Germans, which made it possible for him not only to run a campaign of city bombing with high civilian casualties in mind, but also to relish, in his own choice of words, “this lethal campaign.” Harris was known to see the First World War as unfinished business, and he had an instinctive hostility to totalitarian systems, right or left. But neither perhaps explains sufficiently why he regarded the death of ordinary Germans as something to be sought in its own right. “We have got to kill a lot of Boche,” he famously wrote in April 1942, “before we win this war.”¹⁹³ During 1943 and 1944 he wanted the Air Ministry to state unequivocally that killing the German people was what his command was for. In later life he never wavered from his conviction that there was nothing ethically objectionable to killing the enemy civilian in total war, which was a view widely shared at the time, but his complete indifference to the fate of the Germans he bombed, even in Dresden, is more difficult to understand. When the biographer Andrew Boyle asked Harris in 1979 about his “aggressive philosophy where Germans were concerned,” Harris did not respond.¹⁹⁴

His second conviction was his unyielding belief that the heavy bombing of urban areas was the best use to which the current bombing technology could be put. He contested, often bitterly, any attempt to divert the forces under his command to other purposes, and when compelled to do so, he fought to have his bombers returned to what he saw as their only rational function as soon as possible. The destruction of cities, Harris insisted to the end of the conflict, would “shorten the war and so preserve the lives of Allied soldiers,” though it cost the lives of half his operational crews.¹⁹⁵This stubborn refusal to accept that any other strategy might yield more strategically useful and less damaging results made him into the Haig of the Second World War. Harris’s reputation, like Douglas Haig’s before him, has been a historical bone of contention ever since.

Though Harris’s appointment no doubt marked a turning point in the bombing war, he was not, as is so often suggested, the originator of the area-bombing campaign. He arrived at his command after a brief interregnum in which the officers in the Air Ministry in favor of large-scale incendiary attacks on residential areas had been able to exploit the absence of a field commander to put in place an unambiguous commitment to the strategy they preferred. A new directive was sent to Baldwin as acting commander in chief on February 14, 1942, modifying the directive of July 1941 by removing communications as a primary target and focusing the force entirely on “the morale of the enemy civil population and in particular of the industrial workers.” A list of cities was appended to the directive, with the vulnerable central zones highlighted and the bomb tonnage necessary to destroy them recommended.¹⁹⁶ In February 1942 the Directorate of Bomber Operations, which had prepared the directive, explored the vulnerability of particular cities to large-scale conflagration and chose Hamburg (rated “outstanding”), followed by Hannover, Cologne, Düsseldorf, Bremen, Dortmund, and Essen.¹⁹⁷ The zoning system developed in 1941 was now applied to these cities to show the value of hitting the “closely built-up city centre” (Zone 1) and the “completely built-up residential area” (Zone 2A). Attacks on these central zones were estimated to be up to twenty times more effective than attacks on the outer industrial and suburban zones. The damage done to a large working-class area was expected to affect the output of numerous factories through absenteeism or death, where an attack on a single factory target would affect only that one.¹⁹⁸ This was the background to the famous minute sent to Churchill by Lord Cherwell on March 30, 1942, in which he calculated that 10,000 RAF bombers would by mid-1943 be able to drop enough bombs to dehouse one-third of Germany’s urban population. “Investigation,” ran the minute, “seems to show that having one’s house demolished is most damaging to morale.” Churchill was so impressed that he insisted on circulating the minute to the War Cabinet. It generated at the time a great deal of argument from other scientists who criticized the arithmetic (Patrick Blackett thought it exaggerated by a factor of 600 percent), and it has attracted much discussion from historians, but in effect it simply advertised a shift in bombing priorities that had already been agreed upon and was now in place.¹⁹⁹

Harris did make a difference when he took over Bomber Command, because he was an aggressive and single-minded defender of his force against all efforts to divert it to other purposes or to compromise the directive he had been given. He also argued forcefully against the widespread criticism of the command—“ignorant and uninstructed chatter,” he called it—because of the damaging effect on bomber crew morale to be regularly reminded that their efforts were “futile.”²⁰⁰ But Harris did realize how limited bombing still was without a substantial increase in the size of the bomber force and an end to the dispersion of bomber aircraft to other theaters. When he arrived at the command he had at his disposal only a few hundred bombers, of which a large part were still medium Wellingtons. He understood that this force was incapable of achieving what the new city-bombing directive suggested. He complained to Norman Bottomley, deputy chief of staff, that what he needed was a force of at least 2,000 bombers; such a force, he claimed, would not only destroy his list of twenty cities but “knock Germany out of the war.”²⁰¹

Harris nevertheless set out to demonstrate what his limited numbers could achieve. On March 8–9, 1942, 211 aircraft (including 37 heavy bombers) armed with Gee navigation attacked Essen and the Krupps complex. Dense industrial smoke obscured the city; no bombs hit Krupps, a handful of houses were destroyed, and ten people were killed. A second raid on Essen two days later killed only five people; the bombs were scattered over sixty-one different villages and towns.²⁰² A raid on Cologne on March 13–14 proved more effective thanks to better target marking despite a gloomy night. The most successful attack was made against the Baltic Sea port of Lübeck on the night of March 28–29. Although the city was beyond the range of Gee, there was a full moon and good visibility. The 234 bombers attacked in three waves, carrying two-thirds incendiaries against the lightly defended and densely constructed “old town” area. Around 60 percent of buildings in the city were damaged and 312 people killed, the heaviest casualties in Germany so far. A series of four raids were then made against the northern port of Rostock between April 23–24 and 26–27, again aiming for the main city area, 60 percent of which was damaged or destroyed, though thanks to effective civil defense only 216 inhabitants died. These were the first raids where incendiary damage could be inflicted on the central areas of a combustible target along the lines planned in 1941, and they inflicted high levels of urban destruction. They were also the first raids that the German authorities took seriously; following the Rostock raid a special category of “great catastrophe” was introduced to define larger and more destructive attacks.²⁰³

The reaction to the first Gee raids at the Air Ministry was nevertheless unenthusiastic. The director of bomber operations, John Baker, accused Harris of misunderstanding the nature of the incendiary attacks he had recommended, by carrying too much high explosive. Harris was sent a memorandum summing up the opinion of British fire chiefs about the relative value of high explosive and incendiary, which showed that in almost all cases more than 90 percent of the damage had been caused by fire. Baker suggested carrying at least 200,000 four-pound incendiary bombs to maximize the damage.²⁰⁴ On May 8, following the Rostock raids, Baker’s deputy, Sydney Bufton, also wrote to Harris with the evidence from plotted photographs that his attacks on Essen in March and again in April showed that 90 percent of bombs had fallen from between 5 and 100 miles from the Essen aiming point. Plots of twelve raids on Essen between March and June 1942 showed that in seven of them fewer than 5 percent of aircraft got within 3 miles. The raids on Rostock, which was easier to locate, being near the coast, showed that 78 percent of the photographs taken were not of the town.²⁰⁵ A few weeks before this, on April 14, the chiefs of staff had asked Churchill to authorize a second study of bombing results by Justice Sir John Singleton, to see what might be expected from bombing over the following eighteen months. The decision was prompted by Cherwell’s minute on “de-housing,” which suggested very significant consequences by the end of that period with more bombers and greater accuracy.²⁰⁶ Singleton’s report was produced by May 20 using material supplied by Baker and Bufton, though without the statistical foundation used in the Butt Report from the previous August. Singleton concluded that the use of Gee had had mixed results, but that in general, efforts to improve the level of accuracy and concentration had been a failure. He did not believe that over the following six months “great results can be hoped for.”²⁰⁷Cherwell wrote to Churchill a week later that Singleton had been disappointed, “as any layman would be, by the inaccuracy of our bombing.”²⁰⁸

On the question of greater accuracy Harris was generally unhelpful. The arguments over developing a target-finding force equivalent to the German Kampfgruppe 100 had begun in 1941 but were still unresolved when Harris took over. He was opposed to the idea of using the introduction of Gee as an opportunity to develop specialized units to find, identify, and illuminate a target city. Together with other senior commanders, he thought the creation of an elite corps would leave poorer-quality crews to follow behind and would sap the morale of the rest of the force. He favored keeping “lead crews” in each bomber group to find and mark the target, and was impervious to the evidence that this practice failed to produce a concentration of bombing effort. At a meeting with group commanders and the Directorate of Bomber Operations in mid-March, Harris made it clear that he entirely rejected the idea of a target force and was supported by all five group commanders.²⁰⁹ The argument highlighted the extent to which the individual commanders in chief and their subordinate commanders enjoyed independence from the air staff at the ministry in the way they chose to run their campaigns. It was nevertheless difficult for Harris to ignore all the evidence of continued inaccuracy and the political and service pressure to improve it. Failure to do so might, as an air staff memorandum pointed out in May, make it increasingly difficult “truthfully and logically” to resist pressure to divert bombers to other uses.²¹⁰ In March, Bufton sent out a questionnaire to squadron and station commanders in Bomber Command asking them whether they approved the creation of a target-finding force. The replies were unanimously in favor. A squadron commander based at Oakington, near Cambridge, told Bufton that the senior officers’ First World War experience was valueless in the new conflict: “The crocks . . . must be swept from the board.”²¹¹

Bufton sent the results of the survey to Harris, but it made little difference. Harris found five squadron commanders who were prepared to argue the opposite case. The most he would concede was the idea of raid leaders for each group, which built on existing practice. The crisis point came in June when Wilfrid Freeman, acting on Portal’s behalf as vice chief of staff, finally seized the initiative after weeks of fruitless argument with Harris over tactical issues. He told Harris that he would have to accept the logic of a specialized force. Harris met Portal and despite a trenchant rearguard action finally agreed to the establishment of what he insisted on calling the Pathfinder Force to distinguish it from the air staff title of target finding. Even then Harris found ways to obstruct the proper functioning of the new force, which remained short of the most effective aircraft and highly trained crews. An Australian pilot, Group Captain Donald Bennett, was appointed on July 5, 1942, to command the new units; the Pathfinder Force was activated on August 15 and undertook its first operation three days later against the north German coast port of Flensburg. It proved an awkward baptism. Strong winds drove both the Pathfinders and the main force off course, and instead of on the German city, the bombs fell on two Danish towns and injured four Danes.²¹² An Air Ministry minute in early August noted that despite the agreement to form a target-finding force, “a lack of enthusiasm and sense of urgency in high quarters permeates the whole command, and will inevitably result in a complete failure of the T.F.F. [target-finding force] at its inception.”²¹³

Harris found himself, like Peirse before him, fighting against a chorus of criticism both inside and outside the RAF. During May he began to plan a sensational air raid to try to still public criticism and stamp his mark on his new command. He won approval from Portal and Churchill for the plan to send 1,000 bombers against a single German city. It was a risky promise, because it depended on the cooperation of Coastal Command in releasing their bomber aircraft for the raid and the use of aircraft from the training units. Bomber Command itself had just over 400 frontline aircraft. The city chosen was Hamburg, which like Lübeck and Rostock was easily identifiable as near the coast. The object, Harris wrote, was to wipe it out in one night, or at most two. The target was large, near, and “suitably combustible.” The aim was to carry every single incendiary possible and to create an “unextinguishable conflagration” by bombing in a continuous stream and in a short period of time, a gesture toward the tactical recommendations made by the Air Ministry.²¹⁴ The code name Operation Millennium, like later code names, betrayed its apocalyptic purpose. By May 23 plans were prepared with details of German defenses and three routes to the target. Coastal Command agreed to release 250 aircraft, only to find that the Admiralty countermanded the offer. Harris had at the last moment to recruit training personnel and trainee pilots to raise his force to a total of just over 1,000. The weather worsened over the week that followed, and by May 26, Cologne was chosen as a possible alternative. Hamburg was finally abandoned as the primary target and waited another year for its firestorm.

After first being approved, then canceled, then reinstated on May 30, the raid against Cologne was authorized by Harris for that night. A total of 1,047 bombers were sent off, but only 868 claimed to have attacked the main target, dropping 1,455 tons of bombs, two-thirds of them incendiaries, though only 800 tons fell on the city itself. The concentrated stream allowed the bombers to complete the raid in just an hour and a half, which may explain why the first reports from the city suggested that only between 50 and 100 bombers had been overhead. A later report from the local National Socialist regional leader confirmed the actual scale: it was, he wrote, “the most successful concentrated enemy air attack to date.”²¹⁵ Some 3,330 buildings were destroyed and 7,908 damaged; 486 people were killed and over 5,000 injured; 59,100 were rendered temporarily homeless. This represented a loss of 5.2 percent of Cologne’s buildings. Heavy though the raid was, it was impossible to wipe a city out, as Harris had hoped.²¹⁶ He planned to continue the large raids as long as he had the force of bombers acting together. On June 1–2 another “1,000” raid was made on Essen, with far less success: only eleven houses were destroyed and 15 people killed. The last large “1,000” raid, Millennium II, was against the port of Bremen on the night of June 25–26. Out of a force of 960 aircraft, 696 claimed to have hit the city, but destroyed only 572 buildings and killed 85 people, suggesting that many of the bombs missed the target area altogether.²¹⁷ This was the end of the “1,000” plan. Despite the effort to overwhelm the Kammhuber Line by using a concentrated bomber stream, losses were the highest of the war, 123 bombers from the three raids. This threatened to eat into Bomber Command’s training system, and the large raids were discontinued. Some of the OTUs were close to mutiny at the loss of training staff and the demands placed on novice crews sometimes forced to fly obsolescent aircraft to make up the numbers on each raid.²¹⁸

The gesture did something to reinstate Bomber Command’s reputation, particularly with a British public impatient for more rapid progress, but the situation faced by both Bomber Command and the Eighth Air Force in the summer and early autumn of 1942 was more dangerous to the future of the bombing campaign than the crisis in 1941. The summer of 1942 represented a low point in Allied fortunes. The Pacific and southern Asia were dominated by a rampant Japan, held at bay by the victory at Midway in early June, but a formidable obstacle for sustained counterattack. In North Africa the British Commonwealth forces abandoned most of Libya, lost Tobruk, and retreated into Egypt. Field Marshal Erwin Rommel seemed poised to seize the Suez Canal. The Battle of the Atlantic had reached a critical point, and on the Eastern Front, German forces poured toward the oilfields of the Caucasus and the Volga city of Stalingrad. The many areas of crisis left Allied strategy in confusion, and the bombing offensive was the unwitting victim of efforts to plug the many strategic gaps that were opening up with Axis success. Field Marshal Jan Smuts, recruited to Churchill’s War Cabinet, urged the prime minister to send Bomber Command to North Africa where he thought it would do more good.²¹⁹ To try to allay these pressures, Harris wrote directly to Churchill to persuade him that Bomber Command was still the potentially war-winning instrument it had hoped to be two years before:

We ourselves are now at the crossroads. We are free, if we will, to employ our rapidly increasing air strength in the proper manner. In such a manner as would avail to knock Germany out of the War in a matter of months, if we decide upon the right course. If we decide upon the wrong course, then our Air power will now, and increasingly in the future, become inextricably implicated as a subsidiary weapon in the prosecution of vastly protracted and avoidable land and sea campaigns.²²⁰

Harris appended a document to show that his force had at present just thirty-six squadrons with 584 aircraft, or exactly 11 percent of the entire RAF and Fleet Air Arm, and added that of this percentage half the operational effort went to help the Royal Navy. A few weeks later Harris calculated that his force had dwindled to twenty-two effective squadrons available for bombing Germany.²²¹

In the last months of 1942, Bomber Command waited to see what the strategic outcome would be. Harris knew that the command would benefit from a number of technical and tactical innovations that were in the pipeline. As predicted, Gee had had a very short life. It was first jammed by German countermeasures on August 4, and a wide network of stations was set up to interfere regularly with its transmissions. Two new systems had been in development at the Telecommunications Research Establishment. The first was known as Oboe (the transmission noise resembled the sound of the instrument). Two ground radar transmitters, one at Dover and one at Cromer in Norfolk, emitted pulses that were received by an aircraft transmitter and relayed back to the master station, allowing an exact fix of the plane’s position. When the aircraft was over the aiming point, the second station sent out a bomb-release signal. The system was accurate but had a range of only 270 miles, covering the Ruhr but little else, and could only be used by one aircraft at a time. The second system was a more radical innovation. Taking advantage of the British discovery of the cavity magnetron, which permitted much narrower radar wavelengths, an airborne radar device, H2S, was devised that gave a map of the ground area by recording stronger echoes from built-up areas. This could be used over longer distances and could not be jammed as beams could be. Both instruments were available for operational use in early 1943. Their potential effectiveness was magnified by the fortuitous development of a new fast twin-engine bomber, the de Havilland Mosquito. Begun initially as a private venture in October 1938, the aircraft was uniquely made of wood, and powered by two Rolls-Royce Merlin engines. It was designed as a light bomber and relied on its high speed to avoid enemy fighter interception. The Air Ministry showed little interest until Air Marshal Freeman, in charge of research and development, saw the aircraft late in 1939 and ordered work on a prototype. It first flew on November 25, 1940, and saw operational service a year later, when it was used extensively for daylight bombing. It could fly at almost 400 miles per hour (faster than Battle of Britain fighters) and had a service ceiling of at least 28,000 feet. It was so difficult to intercept that it had lower losses than any other Bomber Command aircraft. Its special operational characteristics made it a natural choice for the new Pathfinder units, but in January 1943 there were still only sixteen Mosquitos available.

Bomber Command was nevertheless unable to demonstrate after the Cologne raid of late May 1942 that it merited the kind of strategic profile that Harris had argued for in June. The hope that American entry into the war might soon lead to a strengthened bombing effort was undermined by the slow establishment of the Eighth Air Force, which until July did not know whether there would be time to mount any raids against German targets at all before starting direct preparations to aid a cross-Channel invasion. Even more than Bomber Command, the American bomber force lived in the future. The slow buildup of aircraft and personnel postponed any serious possibility of action against Germany into 1943. Table 1.2 shows the buildup of the Eighth Air Force during 1942, but none of the operations it describes took place over Germany.

Table 1.2: Eighth Bomber Command Operational Statistics, August–December 1942

Source: AFHRA, Maxwell, AL, Eighth Air Force collection, 520.056-188, Statistical Summary of Eighth Air Force Operations, August 17, 1942–May 8, 1945; Richard G. Davis, Carl A. Spaatz and the Air War in Eurape (Washington, DC: Center for Air Force History, 1993), App. 17.

As with the German Air Force and Bomber Command, an operational learning curve had to be followed before crews with no operational experience could be released against improving German defenses. Pressure from Washington insisted that Spaatz and Eaker organize a demonstration to satisfy American and British opinion, and Arnold named Independence Day, July 4, as the day to carry it out. Spaatz had no aircraft of his own, so he recruited six Douglas A-20 light bombers serving with the RAF to make a suicidal attack against four German airfields on the Dutch coast. The RAF colors were painted over and the six aircraft sent off on the morning of July 4. By the end, one-third of the force was lost, seven aircrew were dead and one a prisoner. Three weeks later a surviving crewman committed suicide. The press on both sides of the Atlantic made the most of the raid, but it was a futile gesture.²²² Arnold pressed his commanders to speed up the organization of real operations. The U.S. bomber offensive was launched on August 17 with an attack by twelve Boeing B-17 Flying Fortress bombers on the railway sheds at Rouen in northern France. Eaker flew with the mission, which was protected by RAF fighters. All aircraft returned safely after striking the target. Ten days later, after three more missions over occupied Europe, Eaker reported to Spaatz the current assessments of accuracy. The bomb plots seemed to show that 90 percent of the bombs dropped fell within a one-mile radius of the aiming point, almost half within 500 yards. He concluded from this that daytime bombing with the Norden bombsight was ten times more accurate than RAF night bombing. Limited though this experience was, Eaker, like Harris, thought that Allied bombing by day and night would be adequate “completely to dislocate German industry and communications.”²²³ But unlike the British and German learning curve, the early raids convinced the American side that daylight raids were possible.

While Harris waited for a response from Churchill on the future of the offensive, the prime minister flew to Moscow on August 12 for urgent talks with Stalin. The object was to explain to Stalin why the Western powers had decided in July to abandon the idea of a cross-Channel invasion in 1942. The meeting was famously combative: Stalin argued against every explanation provided by Churchill in insulting terms until the point when Churchill explained the plans for an Anglo-American bomber offensive. Roosevelt’s representative, W. Averell Harriman, wired back the result to his leader: “Stalin took over the argument himself and said that homes as well as factories should be destroyed. . . . Between the two of them they soon destroyed most of the important industrial cities in Germany.”²²⁴ Harris and Spaatz were both fortunate that bombing was still required in the summer of 1942 as a means to placate the Soviet Union over the failure to open a second front. Although Churchill knew about the poor progress of the offensive, it could not easily be abandoned now that there was to be no cross-Channel operation. On August 17, Churchill asked Portal and Sinclair to lay on an operation against Berlin to show Stalin that he had been in earnest, but he was told that Harris regarded the operation as too costly with only 300 serviceable bombers and a great many inexperienced crews. Though Churchill argued angrily in favor of an attack, Harris told Portal that it would seriously damage the expansion of the command. “As I have frequently pointed out to you,” he wrote in late August, “Bomber Command is now quite definitely too small for the tasks it is expected to carry out.”²²⁵

As a result, when Harris asked Churchill for a “firm and final decision” on September 4 about the future of the bomber offensive, he received a guarded response. Churchill remained committed to bombing Germany, since he could not easily terminate such a conspicuous element of Britain’s war effort, but he thought it would have no decisive results in 1943 nor bring the war to an end; “better than doing nothing,” he concluded.²²⁶ This was a view widely shared in military and political circles by the autumn of 1942, for which the chief priority was breaking the submarine blockade, using bombers if necessary, and supporting American participation in the ground offensives planned for North Africa and Europe. Leo Amery, one of Churchill’s cabinet colleagues, found the Harris memorandum “entirely unconvincing” and thought bombers should be used for “tactical co-operation with the army and navy.”²²⁷ One of the scientists at the Air Warfare Analysis Section warned the Air Ministry that Bomber Command could not hit enough of German industry to do any decisive damage. “I am aware that this view of night bombing,” he continued, “is shared by a very large number of thoughtful people.”²²⁸ When the chiefs of staff considered the future of the bombing campaign in November 1942, Portal was subjected to a hostile cross-examination by his colleagues. General Alan Brooke, chief of the imperial general staff, thought the air force lacked a clear plan of campaign, underestimated the German defenses, exaggerated the possible bomblift, and overstated the damage likely with blind bombing. The one slim advantage, he concluded, was its political value, bringing “the horrors of war home to the German people.”²²⁹

Harris took out his own frustration on others. He deplored the decision by the Canadian bomber squadrons, which composed a growing fraction of the command, “to huddle into a corner by themselves,” even more the prospect of supplying them with Lancaster bombers at the expense of British crews.²³⁰ He was scathing about the American Eighth Air Force, despite the public image of friendly collaboration, for taking airfields in East Anglia away from British squadrons, forcing them to fly dangerous return routes to bases farther west and north, and without contributing “the smallest assistance” to the bombing campaign against Germany. He asked the Air Ministry to challenge American leaders to state categorically “whether it is their intention to proceed with the air bombardment of Germany,” and, if so, when it would start. If no adequate answer was forthcoming, Harris recommended taking some of the airfields back, to which the ministry gave qualified agreement.²³¹ To Portal he sent a bitterly sarcastic denunciation of the efforts to divert his force to what he called panacea targets: “In sum,” he concluded, “they spell the end of our effective Bomber offensive against Germany.” He spent the rest of the war grimly contesting every attempt by what he called “Panacea Target mongers” and “Diversionists” to prevent him from bombing city areas.²³²

There was little that Bomber Command could do over the autumn months to still the chorus of complaints. Evidence from attacks on Germany showed that despite the advent of the Pathfinders, levels of accuracy were still strikingly low. In December 1942 the government scientist Henry Tizard asked the command for details of its performance in recent weeks against Ruhr targets and was told that in good weather around one-third of bombs were landing within three miles of the aiming point, but in most raids the figure was still 15 percent, and sometimes zero.²³³ Surveys of raids on Mainz and Munich showed a wide spread of bombs, with most incendiaries destined for Munich falling in open country. “There is at the present time,” wrote Bufton in response to these findings, “a lack of grasp throughout the Command of a common tactical doctrine.”²³⁴ There was also no effective way of measuring what impact the bombing was having on the German economy, military machine, and popular morale. During 1942 the command dropped 37,192 tons of bombs on German soil, compared with 22,996 in 1941, but most of these bombs failed to hit the target area, and the raids cost some 2,716 bombers lost on operations or through accident.²³⁵ The first scientific analysis of a major raid was supplied in November 1942 by division RE8 of the Research and Experiments Department, which used British models to calculate the likely degree of homelessness, lost man-months, and financial cost of the 1,000-bomber raid on Cologne six months before. The first statistical assessments of acreage destroyed and of the ratio between high-explosive and incendiary damage were only ready in January 1943.²³⁶ Until then, claims that cities had been wiped out or obliterated were mere guesswork. In fact, during 1942 the damage to the German economy and society remained limited. A small number of spectacular raids in the late spring had not been sustained, and the German civil defense and repair organization coped with the consequences with little pressure. The German economy cushioned the bombing and expanded weapons output by more than 50 percent during the year. Postwar calculations in the United States Bombing Survey suggested a loss of potential overall production of 2.5 percent due to British bombing, or roughly half the impact of the German Blitz on Britain. During the course of 1942, 4,900 Germans had been killed, two for every bomber lost.²³⁷ The one solid achievement was to compel the German enemy to divert aircraft, guns, and ammunition to defense against bombing, when they could have been used for the fighting fronts in North Africa and Russia.

At the chiefs of staff meeting on November 18, 1942, Churchill opened the discussion on bombing with the remark that at the moment it had “petered out.” He continued that the answer was not megalomania—a none too oblique reference to Harris—but a more modest and achievable program.²³⁸ There are a number of familiar explanations given for the failure to produce an operationally effective and sustained bomber offensive in the first three years of war—economic restraints on aircraft production, the demands of other theaters, the long program of training and preparation—but none of them is sufficient to understand why evident tactical, operational, and technical changes were not made sooner and consistently or a clear and convincing plan devised (or indeed why the whole strategy was not abandoned in favor of using the resources more productively). By the autumn of 1942 neither the British nor American air forces had a bombing plan beyond destroying working-class districts and attacking a limited number of industrial objectives in western Germany, and no effective effort had been made to evaluate what even this modest program might achieve strategically. The British War Cabinet finally asked the Joint Planning Staff to draw up a bombing plan in late August 1942, but nothing was approved before the end of the year.²³⁹ Roosevelt the same month ordered Arnold to produce a comprehensive plan for the future air war, and the result, AWPD-42, was the clearest outline yet produced of how a bombing offensive should be organized and with what object, though it was still not a definite operational directive. Arnold complained to Harry Hopkins a few weeks later that what was still missing was “a simple, direct plan, tied to a definite date.”²⁴⁰ American frustration at the lack of strategic direction and the slow buildup of the Eighth Air Force made Arnold decide to send Spaatz to join Dwight D. Eisenhower in North Africa with a view to eventually making him overall commander of all American air forces in Europe. Spaatz was reluctant to lose operational control of the Eighth Air Force, which was now taken over by Eaker, while the start of operations in Africa, as had been feared, diverted the bombing effort to the Battle of the Atlantic and postponed even longer the start of American bombing over Germany.²⁴¹

The most remarkable failure in the British offensive was the slow development of target finding and marking, the dilatory development of effective electronic aids, marker bombs, and bombsights, and the inability to relate means and ends more rationally to maximize effectiveness and cope with enemy defenses. The lengthy learning curve cost Bomber Command 14,000 dead from September 1939 to September 1942. A central explanation is the poorly defined relationship between the Air Ministry, the air staff, and the commanding officers. A great deal of responsibility was delegated to the commander in chief, which in turn was delegated to the group commanders in the field. This created a wide gap between the essential scientific and tactical evaluation available from the staff in the ministry and the officers whose task it was to organize operations. The Ministry of Economic Warfare in a letter to the air minister observed that this gap reduced the prospect of learning from experience and of collectively evaluating the best use to be made of the bomber force. The ministry wanted a greater say in bombing operations, and was thus a scarcely neutral observer, but the predicament was a real one, made worse by Harris’s strident defense of his independence.²⁴² At the same time, directives were worked out by ministry officials for the air staff with too little reference to the commanders in the field on questions of technical requirements and operational feasibility, and with no clearly articulated strategy behind them, since this was not the officials’ job. The result, as a memorandum produced in May 1942 suggested, was “considerable criticism and loss of faith on both sides.”²⁴³ The crews were caught between these two poles, asked to perform impossible tasks, taking high casualties, and receiving little explanation for the wider purpose of their missions. Bufton, himself a former squadron commander, summed up this sense of frustration: “They feel that they can do more than they are doing; they grope somewhat blindly in an effort to find where the failure lies.”²⁴⁴