With the War Cabinet’s approval for a campaign “focussed on the morale of the enemy population,” Harris proceeded to put into practice his conviction that the only way to destroy anything in a city was to destroy everything in it, by setting huge conflagrations with tons of incendiary bombs. Harris was disappointed that two subsequent 1,000-plane raids he mounted following the attack on Cologne—the Ruhr cities of Bremen and Essen were hit in June 1942—were not as successful. Even the Cologne raid, he felt, ought to have done more destruction. But then the average bomb load of his planes was only a ton and a half; most of the force available to Bomber Command was still two-engine medium bombers. Harris was confident that once the new heavy bombers began to arrive in numbers, his program for “the elimination of German industrial cities,” as he put it, would swiftly be accomplished.26
WITH BLACKETT’S DEPARTURE to the Admiralty in December 1941, E. J. Williams had taken over as head of the Coastal Command Operational Research Section, and within a few months he believed he had figured out a way to dramatically increase the effectiveness of the antisubmarine air forces even without prising a single new airplane away from Bomber Command. Based on some preliminary theoretical calculations, Williams sent Air Marshal Joubert a memorandum proposing an urgent study of maintenance and flying schedules of the aircraft in the command, and in June 1942 assigned a single member of his now twelve-person ORS staff, Cecil Gordon, to the task.27
Gordon, leaving behind his drosophila in Aberdeen, had just joined the unit that summer. He was pudgy, Jewish, abrasive, and communist. “Definitely not officer material,” noted his Admiralty file. He had a high-pitched, grating voice, an awkward walk, a nervous fidget, and an unkempt appearance that was a source of wonder even to his left-wing scientist friends, not exactly known for fastidious dressing themselves. Once Gordon tried to donate some of his old clothes to a secondhand shop: the shop rejected them.
Gordon had literally been a card-carrying communist, joining the party in the 1930s and resigning apparently only just before he was hired by Coastal Command. His Marxist convictions nonetheless remained deep and indelible. Growing up in an impoverished Russian immigrant family in South Africa—his father was a fabulously unsuccessful peddler of geegaws in the native areas—Gordon had been powerfully influenced at the University of Cape Town by the antiestablishment views of Lancelot Hogben, the head of the zoology department where Gordon studied and later taught as a lecturer. Hogben was a socialist, militant atheist, ardent feminist, and outspoken critic of eugenics and the theories of scientific racism. In 1931 Gordon followed Hogben to the London School of Economics, where his mentor had just been appointed professor of social biology; he then completed his Ph.D. in genetics at University College London under another famous left-wing biologist, J. B. S. Haldane, before rejoining Hogben at Aberdeen. Gordon was thoroughly steeped in left-wing politics during his time in London and Aberdeen. He met his future wife at a socialist student dance; he chaired a British-Soviet Unity Campaign; a story made the rounds that he had been seen outside the Finchley Road tube station, dressed in dinner jacket, peddling copies of the Daily Worker one evening while on his way to dine with his wife’s wealthy parents.
He was also impatient, vociferous, opinionated, and domineering in conversations. His mind was always “temporarily one-track,” a colleague remembered, whatever obsession of the moment taking over his thoughts and conversation “to the exclusion of everything else for the time being.” He could be brilliant and fascinating, relating some novel idea he had pulled from his enormous range of reading. When the topic of his momentary obsession was a problem of work, however, he could be “monotonous,” and when it was one of his many grievances, “a bore.” Even his enemies and detractors, though, acknowledged that he had a fierce intellectual and moral honesty, which he applied as ruthlessly to himself as to others.28
Within three weeks of being assigned to the maintenance problem Gordon had worked out the basic mathematical concepts and had thrown himself into an experimental field test of the ideas he had devised. The calculations did not require more than algebra but were lengthy and incorporated vast amounts of data that Coastal Command had collected about the work of the maintenance shops: the number of man-hours spent on various tasks, the intervals between inspections, the average time of failure of various components.
Gordon’s basic conclusions, though, were simple and stunning. First was that there were strangling bottlenecks in the maintenance chain: the availability of skilled labor was the limiting factor at several choke points, which meant that while some repair departments worked constantly others were left idled with nothing to do. The analysis suggested that maintenance crews were accomplishing only three quarters of the work they could have in the total man-hours they had available if the work were more rationally organized. It was the four-washtub problem writ large.29
But a far greater obstacle to increasing the number of flying hours per month, Gordon realized, had to do with the RAF’s “serviceability” policy that set a goal of having 75 percent of the aircraft of every squadron ready for operational duty at all times. The counterintuitive mathematical result Gordon discovered was that by accepting a lower serviceability rate, the total number of monthly flying hours would increase. The best policy would in fact be to ignore serviceability altogether; it was the wrong quantity to be measuring or even paying attention to. A serviceability standard made sense for fighter defense squadrons, which had to be ready at any time to put as large a force in the air as possible with minimal warning. But for the squadrons of Coastal Command, which had a steady day-to-day mission of patrolling against a constant enemy presence, the insistence on keeping a large portion of aircraft ready at any time could only be met by holding back a significant number that could otherwise be flying useful missions.
The reductio ad absurdum of this policy, Blackett later observed, would be never to fly at all: that would achieve a serviceability rate of 100 percent. (Blackett drew an analogy to an incident he recalled a friend relating in the years between the wars: arriving at a rural train station one night, he approached the lone taxi waiting out front and asked to be driven to his destination. The taxi driver refused—explaining that a local police ordinance required one taxi to be waiting at the station at all times.) Gordon’s numbers suggested that if all serviceable aircraft were flown every day when weather permitted, the serviceability rate would sink to 30 percent, but monthly flying hours would substantially increase. The other way to think about it was that to get the most out of a squadron’s aircraft, the best plan was to fly enough to ensure that the maintenance shops were fully employed at all times. To get more flying hours, in other words, you had to increase the breakdown rate. That would mean more aircraft needed repair at any given time, but the total throughput of the maintenance shops would increase.
Implementing Gordon’s scheme for what would come to be called “Planned Maintenance, Planned Flying” was a prime case where supreme tact was required; in its bare outline it sounded like nothing so much as a criticism of the efficiency of a unit and a slap at workers for sitting around and doing nothing. Gordon was hardly tactful. But his sheer force of intellect seemed to make up for it. It also helped that Churchill was right on top of his work. On July 5 the prime minister asked the Air Staff for a copy of Gordon’s initial report and quickly requested the first lord of the Admiralty and the secretary of state for air to sign off on the proposal to put it to a test: Churchill especially warned that until Coastal Command fully implemented Gordon’s ideas to increase the number of sorties per squadron, “there can be no case for transferring additional squadrons from Bomber Coastal Command.” Churchill also, significantly, grasped at once the essential point that “it is true the standard of serviceability in Coastal Command will fall off if the aircraft make more frequent sorties,” but that this was an acceptable trade-off.30
Coastal Command agreed to place one unit, No. 502 Squadron of Whitleys, under Gordon’s mathematical directi
on as a test. For five months Gordon tracked the status of every aircraft in the squadron as the changes were implemented. The crux of the matter was the percentage of time each plane spent in each of four different states: flying; serviceable but not flying; being serviced; or awaiting service. As expected, the percentage of time a plane spent waiting to be serviced increased. But the increase in flying hours more than made up for it: flying hours nearly doubled.
In the process, the scientists poked into everything from inspection schedules to the time spent waiting for spare parts to arrive to the rate at which various components of an aircraft fail. One of their interesting discoveries was that for a large class of components, routine inspections were literally a waste of time: a part like a fuse or a spark plug lead was “just as likely to go wrong at one moment as the next” and inspections revealed little or nothing. In fact, when they started plotting the failure rates of various parts, they found that inspections in many cases increased breakdowns, apparently the result of disturbing components that had been working fine. The results produced by the rationalization of maintenance procedures and flying schedules in 502 Squadron were another of those operational research contributions that spoke for themselves. Joubert ordered the policy instituted throughout the command—and asked the ORS scientists to take charge of implementing it.31
The mathematician T. E. Easterfield, who joined the group later that year and was promptly grabbed by Gordon to help with the work, found that the general attitude among the aircrews was that the scientists “might be mad, but they got results.” Easterfield thought the scientists were viewed as “licensed jesters” whose odd behavior was tolerated because they served a purpose. Gordon’s one-track mind and tenaciousness were madder than most; but so, too, his results were commensurately greater than most. “Marvelously efficient chap!” remarked one RAF officer to Easterfield about Gordon—before adding, “But he was no gentleman!” To the RAF aircrews, Gordon was known as Joad, a nickname they bestowed on him after C. E. M. Joad, the eccentric and discussion-monopolizing resident philosopher on the popular BBC radio talk show The Brains Trust.32
WILLIAMS AND BLACKETT WERE meanwhile working on another approach to tighten the noose around the U-boats. Since 1941 air patrols had been regularly flying over the Bay of Biscay, the roughly 300-mile-by-300-mile bottleneck that all of the U-boats had to transit between the French ports and their stations on the mid-Atlantic convoy lanes. Williams calculated that in the second half of 1941 every U-boat that put to sea on a patrol had a 30 to 35 percent chance of being attacked by Allied aircraft during either its outward or inward passage of the bay.33 In response, the Germans had taken to transiting the bay submerged during daylight hours, surfacing at night to recharge their batteries.
Most of the Coastal Command aircraft by this point were equipped with 1.5-meter-wavelength ASV Mark II radar sets, but the results had been disappointing. The aerials for the radars were an ungainly sight, mounted on a row of four pylons perched along the spine of the planes’ fuselages. The radar’s detection range was not much better than what the naked eye could achieve by day. At night, the problem turned out to be not how far the radars could see ahead, but rather their minimum range. The powerful radio pulse sent out by the transmitter would instantly fry the circuitry of the sensitive receiver located right next to it, so the receiver was automatically switched off for a fraction of a second as each pulse went out. But this meant the system was unable to detect echoes bouncing off very close objects, which would arrive back at the receiver during the hundred thousandth of a second or so while it was still switched off. In practice, anything closer than three quarters of a mile was invisible. If the sea was choppy, waves on the surface would add a clutter of echoes that extended out even farther. The ASV planes could thus pick up on their radar screens surfaced U-boats at night at a range of as much as ten to twenty miles, but as they approached they would lose the radar contact at a point where the boat was still well beyond the range the eye could see at night; there was no way to follow through with the actual attack.34
One way to bridge the gap between the minimum radar range and the point of attack was to equip the patrol planes with powerful searchlights that could be switched on for the final attack run. The first of these needed their own Ford V-8 engine and 35-kilowatt generator, or a bank of batteries completely filling the bomb bay, to supply the required electricity. An inventive RAF officer assigned to Coastal Command headquarters, Squadron Leader Humphrey de Verd Leigh, came up with the idea for a more efficient carbon arc lamp that would need only seven 12-volt batteries, which could be trickle-charged from a small generator driven by one of the aircraft’s engines. The whole package weighed less than 600 pounds.35
In a February 1942 paper, Williams had calculated that a balanced day and night force with 50 Leigh Light–equipped aircraft and an additional 100 long-range Catalinas or Liberators could increase the chances of intercepting each U-boat on its inward or outward passage to well over 50 percent.36 He was also able to provide a reassuring calculation proving that the Germans had not yet taken the step of countering the Allied radar by equipping the U-boats with warning receivers that could pick up a radar signal, which would give them time to dive before the would-be attacker came within striking distance. The poor success of No. 502 Squadron that same month when it tried to carry out night radar attacks using dropped flares to illuminate the target area made some Coastal Command officers worry that the Germans already had deployed such a threat-warning receiver. But Williams showed that the number of sightings by day per flying hour for ASV and non-ASV equipped aircraft were still roughly the same.37
By June 1942 the first Leigh Light–equipped planes were operating and the number of nighttime attacks shot upward.38 Again, the operational researchers were able to produce convincing calculations showing that the sighting rate was very close to the theoretical values expected, assuming the Germans still had not deployed a radar warning device.
The impact on German morale was instantaneous. The security of the night had been shattered: Dönitz ordered his boats to reverse procedure and return to surfacing by day, the idea being it would be a more even fight in daylight when the U-boat crews could visually spot an approaching airplane and possibly have time to submerge. By July the average number of U-boats at sea reached seventy and the return to the convoy attacks was keeping the tonnage total at a steady half million or so a month.39 But it was a much tougher and grimmer fight than ever. On July 27, Dönitz made a radio broadcast to the German people unlike anything they had ever heard from a high official of the Third Reich. He told his listeners that the “harsh realities of submarine warfare” meant Germany must expect sacrifice and losses. Dönitz later explained that he was worried by the exaggerated hopes of easy victory that had been raised by a torrent of official propaganda and felt that a dose of caution was needed to prepare the public that, as he said in his broadcast, “even more difficult times lie ahead of us.”40 It was, thought the British Admiralty, a clear indication that Dönitz was planning to pour even more of his force into the convoy battle.
In September he issued new orders emphasizing that the war was now a fight for survival. The immediate impetus for the orders was a chaotic incident in which the British troopship Laconia, carrying 1,800 Italian prisoners of war from North Africa along with some British soldiers and women and children, was torpedoed 500 miles north of Ascension Island by U-156. The German submarine assisted the survivors, taking nearly 200 on board and towing four lifeboats while radioing a message that it would not attack any Allied vessel that came to assist. Dönitz approved, and requested Vichy French ships from West Africa to help in the rescue of the British passengers and Italian prisoners. Two days later an American B-24 from Ascension Island spotted the surfaced U-boat, and attacked. The German captain immediately ordered his passengers onto the lifeboats, cut them loose, and escaped serious damage by diving, but one of the lifeboats filled with Italian POWs was sunk. A few hours later the Vich
y French ships arrived and picked up the remaining survivors.
It was a confused incident and it was unclear how much of Dönitz’s humanitarian concerns were prompted by the consideration that one of his captains had torpedoed a ship full of Axis prisoners. But it became the pretext for his order of September 17 that was the final culmination of the progressive brutalization of the war at sea:
To all Commanders:
1. All attempts to rescue members of ships sunk, including attempts to pick up persons swimming and placing them in lifeboats, righting capsized lifeboats, or supplying provisions and water, must cease. Rescuing survivors contradicts the most primitive demand of war, to annihilate enemy ships and crews.
2. The order [previously issued] for bringing back commanding officers and chief engineers remains in force.
3. Survivors are to be picked up only in cases where their interrogation would be of value to the U-boat.
4. Be severe. Remember the enemy in his bombing attacks on German cities has no regard for women and children.41
At some point that fall Dönitz issued another order, which was formally reiterated the following year in a war diary entry that referred to this earlier directive, calling the attention of his U-boat commanders to the usefulness of torpedoing the rescue ships attached to convoys. “In view of the desired annihilation of ships’ crews,” Dönitz stated, “their sinking is of great value.”42
Blackett's War Page 28
