Fatal Crash Causes

Introduction, Understanding flying accidents, Pilot error, human factor, and bad luck, Risk-reducing tactics, Mid-air collisions, Low flying, Oxygen failure, Blood supply failure, Engine failure, Bad weather, 'Friendly' fire, Most KIA/KIFA crashes occurred in the UK, Good safety record of 320 Squadron, The art of riding the Flak, The high fatality episodes, High anxiety and discipline

1. Introduction

Wartime crashes of combat aircraft are obviously attributed to the enemy. Without War these crashes, and the fatalities resulting from them, would not have occurred. However, there is more to it. An analysis of all fatal crashes involving Dutch RAF aviators demonstrates that half of the fatalities came about without direct enemy action. Such crashes are commonly called flying accidents. The usual distinction is between operational and non-operational flying. An operational flight, or ops, is flown with the intention to attack the enemy, or to protect one's own bombers or vessels, or to survey enemy positions and movements. All other flying is considered non-operational. Aviators killed or missing on an operational flight are referred to as killed-in-action or KIA, or missing-in-action, MIA. Aviators lost on other flights are referred to as killed-in-flying-accident, KIFA. This author finds that schedule unsuffciently distinctive. All military flying in this episode came about as a result of War. Several flying accidents took place on ops, with no direct enemy involvement. Many men were lost over sea, as a result of flying accidents. That would generate the category MIFA.

Furthermore, the term KIA is a very general one. It does not describe the actual cause of death, that may or may not be related to a crash. As a way out of this potential confusion, author shall, in the tables below, distinguish only between fatalities resulting from direct enemy action, and those that came about without direct enemy action. But in the chapters about individual crashes, the circumstances of death shall be described as correctly as possible.

Causes for fatal accidents involving Dutch RAF aviators have been found to be as follows:

Table 7. Dutch RAF aviator loss causes

The conclusion is that about half of the number of fatalities ocurred with no enemy in sight. Flying combat aircraft in wartime was utterly dangerous, with or without an enemy in sight. The conclusion cannot be that the Dutch RAF aviators were somehow below average. The author did not compare these findings with crash analysis regarding other airforces, but has no indication that in those other forces matters were vastly different. In fact this study makes the point that accidents should be seen in the circumstances in which they occurred, rather than simply blaming the guy at the end of the line, the pilot. Full responsibility for whatever happens in the air, with no enemy in sight, can only be assumed if the pilot can be in full control, and has all relevant data. We shall see that in many accidents this was not the case at all.

Losses of Dutch RAF aviators caused by direct enemy action have been found to be of two types only: aircraft shot down by enemy fighter (1/3) or by enemy Flak (2/3). In these cases, the situation is clear. Loss in combat, directly caused by enemy action.

The percentage of losses with as yet unknown causes is high: 12%. These are cases in which direct enemy action is the likely cause. All of these occurred within reach of the enemy. The percentage of losses by presumed flying accidents is also high: 16%. These are cases, in which the cause of the loss is unknown, but the loss is most likely not due to direct enemy action. All of these occurred over the United Kingdom or the seas close to it, and usually close to base.

Adding the presumed flying accident fatalities and the presumed enemy action fatalities, we have a staggering 60 Dutch RAF aviator lost as a result of unknown causes. That's 26%. The statistics given above shall shift if and when those unknown causes become known. Data given has to be considered as preliminary, pending more and better data.

2. Understanding flying accidents

1. Pilot error, human factor, and bad luck

The causes for flying accidents given above, are not sufficiently descriptive. For instance, the cause 'collisions' describes what happened: two aircraft collided. But it does not describe how these aircraft came to collide. That leads to the second level of cause descriptions, aimed at the one who is responsible by definition, the pilot. And that shall also lead to questions about the definition that holds the pilot responsible for whatever happens when airborne, when there is no direct enemy attack.

The usual explanations for flying accidents are

- pilot's error, as in 'pilot did not recover the aircraft from a steep dive'. Pilot should have performed better.

- bad luck, as in 'aircraft crashed as a result of engine failure'. Pilot could not perform better.

Although these explanations may be perfectly valid and true, they do sound overly simplistic. Considering that reality is always highly complex, it could lead to a better understanding of historic events if we would probe deeper into the causes of these flying accidents. Careless use of the terms 'human error', 'pilot's error', 'human factor', and 'bad luck' can be confusing.

If 'pilot's error' is the verdict, then we shall ask what that error was, and how it came about. Below we shall differentiate between 'human factors' and 'pilot errors'. Human factors are usually involved, as humans were not designed to fly. Especially military aviation can take the human brain and body beyond their limits. Brain and body could not perform better in a given situation. On the other hand, pilot error implies that the pilot could have performed better, but somehow made the wrong decision. Below we shall see that the pilot-error-verdicht quite often is confused with the results of human factors, over which the pilot had limited or no control. When a pilot cannot be in full control, he cannot assume, nor should he be given in post-accident judgements, full responsibility for the outcome. Pilots did not 'fail to keep adequate lookout', if they could not see in clouds at night.

If 'human factor' is the verdict, then we shall ask what that factor was, and in which human it was to be found. Again, the reality of the time shall be found to be manifold and complicated. An aviator could get killed if his parachute failed to deploy properly. The human factor involved would not be located in the aviator, but in the ground crew. Many human factors can be distinguished. The most important ones have to do with lack of oxygen in the aviator's brain. These factors are mentioned in a separate chapter below. There are many reasons why an aviator could be in less than perfect condition, leading to an impairment of his ability to decide and act. Heavy alcohol consumption within 12 hours of the flight would be one. Deprivation of sleep would surely be another. Such data is usually not available for the analysis of the possible causes of a flying accident. There is an enormous gap between the last two factors. Heavy alcohol consumption is, or should be, under control of the individual. Sleep deprivation is, or should be, under control of those who are in command. Both statements are far too defined. They should be differentiated with factors that spring from the urgencies of War. There is little point in trying to design a structure that deals out responsibilities for whatever went wrong to all those concerned. The real point is that the aviator is at the end of the line. He is usually seen as the one who carrries all responsibility. Author hopes to have demonstrated that that's neither true nor justified in all cases. And then we do not have sufficient evidence in so many cases. Without evidence, there should be no verdict. And least of all a verdict over an aviator who cannot speak for himself any more.

If 'bad luck' is the verdict, then we shall try to reconstruct the circumstances that added up to the bad luck situation. Statistics are helpful in showing ways. Bad luck is a one-for-all explanation, that tells us nothing. We shall try to find out if there is more to be told.

The reader may question this questioning. Where did the author get the authority to raise these questions? As follows.

1. Authors take and have authority over what they write, and are personally responsible for what they wrote.

2. This author could not find a single publication in which the Dutch RAF episode is described with a sufficient level of detail and accuracy. Author did find an enormous lot of confusion over the subject. Even very basic data such as the number of Dutch RAF casualties, and the number of Dutchmen enlisted in the RAF in WW2, was not readily available anywhere. Everything had to be researched.

3. This author could not find a single publication in which the crash cause questions are raised.

4. As we are talking about half the number of all Dutch RAF fatalities, the issue is significant.

5. Author could not find self-reflective statements from the RAF that placed at least some of the wartime flying accidents in any sort of perspective. The 'pilot's error' and 'bad luck' verdicts are so simplistic that they cannot be automatically accepted as valid, complete, accurate and justified.

6. Author did find several cases in which the individual aviator was sentenced as guilty of his fatal accident, verdicts that can be demonstrated to be unjustified. As those individual aviators cannot speak for themselves, some-one has to step forward to defend the memory of these men.

In the analysis of automobile crash causes, the notions of 'driver's error' and 'bad luck' have been abandoned as rather useless. Most drivers are ignorant, and traffic designers just have to work with that. Luck cannot enter the designer's equasions. Crash causes have been reformulated as the results of inherently dangerous and/or unclear situations. Road design is ideally aimed at offering the driver crystal clear view and choices, whilst eliminating the known danger spots.

Unlike the usual estimation of skills of car drivers, the WW2 pilots were a sharp as they could be. If this did not come about by sufficient training, then certainly by their sense of imminent lethal danger.

In WW2 military flying, the airforce tacticians were the road designers. Obviously they had to work with a set of circumstances that differed greatly from those presented to peacetime road designers. Every improvement made by the airforce tacticians, had to be revised rather quickly as the enemy countered. The situation was never static, and the pressure was enormous. Inherently dangerous situations could not be avoided. In fact it is the essence of combat to seek these situations. A free and open debate about tactics was restricted to the highest levels of command, and results were classified. The enemy should not be able to read about new tactics in the newspapers flown daily by KLM/BOAC from RAF Whitchurch to Lissabon.

2. Risk-reducing tactics

The RAF designed tactics designed to lessen the risk of wartime flying. These tactics can be classified as follows:

1. Training instructions

2. Tactical instructions

3. Technical means

1. Training instructions, including P/O. Prune

1. Obviously flight safety was a main theme in aviator training. The RAF developed a sophisticated training schedule that, after the hectic days of the Battle of Britain, saw to it that pilots were prepared for their operational jobs.

2. The RAF gave it an effort to increase safety awareness in its aviators. Obviously via the various training schools, but also in a most original way. P/O. Percy Prune came to life in the RAF Tee Emms, the Training Manuals. P/O. Prune is a cartoon figure, that reached mythical status. He was the one who did everything wrong, thereby conveing the message how things should not be done. P/O. Prune's life and times were designed by Anthony Armstrong, who faced the task of transforming military aviation technicalities into messages that were instantly understood by all, via understatements and other humourous undertones.

Source: Tim Hamilton, 'The life and times of Pilot Officer Prune', London, 1991

P/O. Percy Prune & dog Binder, a page from a Training Manual illegally carried by a bomber pilot on operations as a good luck charm. It worked, the pilot survived the War.

Source: Tim Hamilton, 'The life and times of Pilot Officer Prune', London, 1991, p. 34

2. Tactical instructions

1. Aircraft crew were instructed about the places where Flak was massed. Crew also took the liberty to avoid such places on their own initiative.

Source: Jan P. Kloos, email 11/3/2005

2. Aircraft usually flew in formation. This applied to fighters as well as bombers, but their respective types of formation were different and, in the case of the fighters, evolved through the War as a function of experience and changes in mission type. Flying in formation massed aviator's eyes scanning for hostile aircraft, and it massed the defence of bombers with their machineguns. For bombers the parole was throughout the War: the tighter the formation, the better and the safer.

3. Flightpaths were selected to lead the enemy to believe that the attack was to take place somewhere else. With this, it was hoped that enemy defences would be less concentrated and organized. This tactic kept the enemy busy. That's a War goal in itself, especially at night.

4. After heavy losses during bomber raids in daylight, the RAF changed over to night flying. At night, aircraft were more difficult to detect by the enemy.

3. Technical means

World War 2 brought the birth of electronic warfare. A main example of risk-lessening measures was the use of 'Window', code name for tin foil strips, that were dropped over enemy territory in huge quantities. Whilst airborne, the clouds of tin reflector strips temporarily confused the German radar.

Technical means were wide and varied. Aircraft had radios, enabling aviators to give warning to the other of imminent danger. Aircrew had parachutes and dinghies. Bombers had escape hatches. Both fighters and bombers had armour plates for body protection of airmen. Fuel tanks were given self-sealing properties to protect against small caliber projectiles. Equipment was developed that enabled landings on friendly airfields under conditions of fog. This list is by no means exhaustive.

4. The drawbacks

As is happens, and especially in War, some of these risk-reducing measures had their drawbacks.

1. Tight formation flying brought about the risk of collisions. We have found formation flying to be the main cause for fatal flying accidents. During the bombing run, a tight formation was a massed and therefore easier target for enemy Flak. We shall describe fatal crashes that resulted from the explosion of a neighbour aircraft after it was hit by Flak. There has been more damage and non-fatal injury resulting from this cause, than could be described in these pages.

2. Devious flight paths to the target had the drawback of a longer time of exposure of the aircraft to enemy defences.

3. Night flying was very demanding on the aviators. Although navigation became easier as the War progressed, as a result of new electronic aids, aviators were highly dependent upon their own eyes. People are not designed to see a lot in the dark. Headlights, even navigation lights, the city lights of a car, could not be used because otherwise the aircraft would give its position away to the enemy. The only anti-collision system available to the aviators was their own eyesight. Vision was limited to a few meters or to zero in clouds, and very difficult in heavy rain and on dark nights.

The aviators were embedded in structures and systems, which which they had to make do. They had to follow orders, and the rest was classified. They could not perform beyond the capabilities of their machines. They could not perform beyond the capabilities of their own bodies. They had only limited control over the set of circumstances in which they were placed. Circumstances created by the enemy, by their own high brass, by technology, and by weather. Pilot's error as a verdict can be considered as always true in a flying accident. An aircraft of WW2 could not decide for itself; it had to be flown by the pilot through every cubic meter of airspace. But it remains to be seen whether this truth is also the full truth in all cases. The pilot's error verdict can only be valid if it is assumed that the pilots had all relevant data and were in full control. They most definitely had not and were not.

The surviving pilots could accept the 'pilot's error' and 'bad luck' explanations with ease. Flying an aircraft physically feels like being in control. Losing Squadron members was always painful and happened so frequently, that one's own survival prevented much interest in the details of those losses. Once the error verdict has been given, nobody likes to say much about the matter any more. Too painful.

At the time, the aviators were not given the full picture. And even if they had the overview, then still they had to act as ordered. The military cannot wage and win Wars, if the soldiers wish to question and debate every next move. Now, with the benefit of some overview, we can try to describe fatal crash circumstances, and we can try to make sense of it all. Some overview only, as relevant archives are still closed. The powers that be have decided that we, the public including the families of the lost men, should wait a hundred years before more comprehensive lessons can be learnt. This secrecy may be designed to protect those who made questionable decisions during WW2. After 100 years, none of those involved is alive any more.

Meanwhile, who is protecting the aviators, that are said to have made errors? Those who died in a Wartime flying accident, cannot speak for themselves. Those who were involved in a Wartime fatal flying accident, and survived, may have carried the burdon of the error verdict for the rest of their lives. To see if there is any justification in that, we shall try to study the multitude of circumstances that resulted in these crashes. There are things that need to be said, on behalf of those who cannot speak any more.

This shall be done in full awareness that we cannot claim to be complete and accurate. It has to be considered impossible to reconstruct all factors that acted in all crash situations. Therefore, any statement in this study has to be considered as subject to revision if new data would surface. Author felt that to be the better way, instead of ignoring it all and shrugging shoulders.

3. Mid-air collisions

The 47% fatalities through flying accidents were caused in again 47% of the cases by mid-air collisions. By far the most important factor in loss of life without direct enemy action. Below we shall try to understand how these collisions came about. Mid-air collisions by definition come about by aircraft that fail to keep distance. The tactics of the time prescribed aircraft to keep minimal distances, because it was believed that tighter was safer. Meaning safer against enemy attack. A distinction is made between bomber and fighter collisions, as these aircraft types operated under different circumstances.

1. Bombers

Bombers usually flew in formation. The main reason for formation flying was massing eyes and machineguns for the defence against fighter aircraft attack. When the allies became capable of sending in fleets of 1.000 or more bombers, formations were also needed to keep matters organized. Without such organisation, chances were higher that bombers would bomb or fly into each other, and bombing patterns would become more erratic. Flying very tight had the additional advantage, at least in theory, that air turbulence would effect the formation as if it was one aircraft only. A sudden drop of 100 meters as a result of air conditions, would lead to the same drop for all aircraft. Such phenomena were dealt to the pilot suddenly, and he could only react. A more pro-active way of controlling the aircraft would have been to keep inter-aircraft distances beyond the margins of what could happen in turbulence.

The discussion shall center on the type of formation flying done by 320 Squadron, Mitchell period.

The Squadron flew in so-called boxes of six aircraft. Three leading the others, flying at a wing's length distance (approx. 10 meters). The other three follow the leading Mitchells, 10 meters lower, noses almost under the tails of the leading aircraft. This type of formation flying requires a relative flying accuracy of a few meters only. Relative as in relation to the other aircraft. Pilots were quite capable of this extremely sharp type of flying, at airspeeds of 350 km/h and more. The margin for error was very small indeed, considering speed, inter-aircraft distance, and volatile atmospheric conditions. The pilot had to constantly watch the other aircraft. His eyes were not available for spotting enemy aircraft. That's why this type of very tight formation flying led to serious trouble for the 1940 British fighters in the Battle of Britain. Fighter pilots only had one set of eyes available. These were needed to spot the enemy, to constantly scan the entire airspace for hostile aircraft. That can hardly be done properly if the other guy is flying at your wingtip. These tactics were soon abandoned by Fighter Command and replaced with less dangerous types of formation flying.

This tight formation flying was done on the way in, to the target. Formations on the way out were less tight. The tight formation required that pilots could see the neighbour aircraft. If clouds were encountered, leading to an instantaneous reduction of visibility to zero, especially at night, the outer aircraft had to spread out, left aircraft to the left, right one to the right, rear group of three to greater vertical distance from the front group, all aircraft to maintain horizontal and vertical speed, and heading after the spreading correction.

This author believes that this type of tight formation flying could lead, and in fact has led, to collisions, especially if visiblity vanished suddenly, and under conditions of cross wind or turbulence. Spreading in clouds would have to be quite wide, to obtain a reasonable safety margin that would have prevented this type of accident. Crosswind could drift an aircraft to its invisible neighbour, whilst the compass showed the pilot the course he was ordered to fly. Whilst pilots were able to fly to a relative accuracy of a few meters under conditions of good visibility, clearly better that the accuracy of their flight instruments, their aircraft were not rock-steady platforms that could be flown to an absolute accuracy of say 50 meters in three dimensions in conditions of zero visibilty. Flight instruments were not accurate enough for that. It would be unjust to expect of the pilots to perform beyond the possibilities of their machines. If inter-aircraft distances would have been clearly greater to start with, then chances of colliding would have been lower. If cloud layers were thick, meaning low lower deck and high upper deck, and meaning several or even many minutes of flying without optical sight, than inter-aircraft distances of say 500 meters would still be hazardous. Furthermore, in zero visibility the pilot's eyes become glued to the navigational instruments. He can't see outside, and he has to constantly monitor the instruments to maintain course and horizontal as wel as vertical speed. Breaking out of the clouds could be quite sudden, pilot still looking at the instruments. If the other aircraft then appears, for whatever reason, to be very close indeed, there may be no time left to react in any useful way. Author believes that this may have happened in the case of the collision between the Mitchells flown by Maas and Manschot, on February 9^th, 1945, 15 minutes after take-off, directly after the aircraft came out of the upper deck of clouds over Tienen in Belgium. Of the total of nine crew, there were only two survivors.

Figure 2. Slipstreams behind a Mitchell.

One in the flightpath, caused by the entire aircraft, one reflected from the bottom side of the wings. Streams are approximately cone shaped. In them, the air is disturbed. An aircraft flying in these streams experiences bumps and shakes. This was in fact used as tactile feedback in zero visibility. The pilot could not see but feel where the leader was likely to be flying.

More about tight formation flying. When in clouds in daylight, a formation closes in so that aircraft can keep track of the other, on the premise that sight is usually about 15 meters.

Source: Marcel Daams, conversation 21/3/2005

Forming up to formation could be equally hazardous. Especially during sorties involving many aircraft, small groups were vectored to climb to their ordered altitudes over different locations, so as to avoid inter-group collisions. This enabled the aircraft to get through the layer of clouds, that was present so often over Southern England. Once above the clouds, the aircraft could see each other, if the night was not too dark. Once the pilots could see, they could form up for the way in. However, this risk-reducing measure had a flaw. The risk was not reduced for that small group vectored to circle and climb over a certain spot. The leading aircraft may have had to start its circling whilst still in the clouds. The next aircraft, that took off a minute later, could not see the first one, and had no way of knowing where exactly the other guy was. Author believes that this could explain the horrible collision between the Mitchells flown by Dobson and IJsselstein, on June 8th, 1944, some eight miles from their base at Dunsfold. All eight crew members were killed. The RAF sees as cause for this accident, that pilots failed to keep adequate lookout. It has yet to be demonstrated that, at the time of this crash, an adequate lookout was humanly possible.

Prevention of this type of flying accident, in the absence of anti-collision radar or similar means, would require vectoring all aircraft to different locations for their climbing circles, if the deck of clouds was thick and the night dark. With many aircraft up, that may have been impractical. If true, then the risk has been taken by high brass in a deliberate way. For lack of better ideas and equipment. Pilots should not be blamed for what is humanly impossible.

It has been argued that the 320 aviators had very little night flying experience. That is the full truth in the case of the Dobson/IJsselstein collision. 320 aviators flew no more than a single night flying exercise prior to operational night flying. The RAF threw these aviators into the dark. But the argument offers no explanation for this accident at all. No amount of experience could have prevented it, if pilots have to fly towards each other blind. Or perhaps experience was indeed gained as a result of this crash. Jan Kloos described how he choose his own climbing area under similar circumstances.

Source: Jan P. Kloos, email 11/3/2005

Such initiative would be called insubordination by some. The initiative at least gave a bit more aviator control over the set of circumstances in which he was thrown.

2. Fighters

In the later stage of the War, many Dutch RAF fighter pilots were detailed to fly in pairs, the smallest formation possible. Especially in the summer of 1944, when 322 (Dutch) and other Squadrons were engaged in Operation Diver, the fight against the V-1 flying bombs.

One would expect more collisions with the bombers, that usually flew in boxes of six aircraft, all the way to and from the target. The bombers were airborne for hours longer per sortie, flew in formation most of that time, and they flew in larger formations. However, we have found 21 Dutch RAF fighter pilots killed in flying accidents, of whom 7 died in collisions, which again is one third of all fatalities in flying accidents. One more Dutch RAF fighter pilot, H.F. Buiskool, was killed when practicing formation flying under conditions of limited visibility. He flew his Spitfire into a low hill.

These inter-fighter collisions occurred once when climbing in clouds, whilst we lost six Fleet Air Arm fighter pilots that flew their Hellcats into each other. These six losses have not been associated with adverse visibility or weather conditions. Lacking more detailed information, we have to assume that the tight flying parole was active here too. The most dangerous bit of FAA flying was considered to be landing on the deck of a carrier. Whilst the Dutch FAA aviators had their share of spectacular carrier landing accidents, none of these was fatal.

3. Conclusions

This author has been unable to find evidence that the RAF was aware of collisions occuring as the major cause by far for fatal flying accidents. Even so, it is the job of High Command to be aware of such matters. Jan Kloos does not remember any precise instructions given by the RAF regarding inter-aircraft distances in formation flying. Instructions were more general: tighter is safer and therefore better. These instructions were maintained for the bombers throughout the War. Meaning that High Command did not revise the general rule during a process that has taken the lives of 15% of all Dutch RAF aviators killed in the War in Europe.

Mentioning High Command responsibility is only part of the story. The aviators flew tight, and were proud that they could do that. There is psychology involved here too. The tight-means-safe rule was accepted instinctively, as in people nearing each other to find more safety in numbers when under attack. A tight formation felt stronger and looked good and impressive. Pilots were almost always quite sharp enough to fly so tight, provided that they could see the other guy. It was a yardstick by which their professionalism was measured. That's why tight formation flying was hardly ever questioned. This author argues as follows, being aware that psychology so very often wins over rational argument:

1. The defensive capability of a box of bombers would not have been less if aircraft would have maintained distances of say 50 rather than 10 meters. That would have added the extra milliseconds of reaction time, that could have made the difference between a very close call and utter disaster.

2. Defense was impossible in conditions of zero visibility, as an attacker could not be seen. Neither could an attacker see the bomber. Not until such times that airborne radar became common and accurate enough. Therefore, flying close together in zero visibility served no purpose. It shall be argued that a distance of say 500 meters could be too close, if you can only assume where the other aircraft are. Alternatively, formations could have flown much closer together in conditions of limited visibility, such as when climbing through clouds. This technique is practiced by formations of much smaller aircraft. With a distance of a few meters only, a part of the other aircraft may still be visible. No need to say that, although the technique may work for small aircraft, it remains an utterly dangerous way of flying.

4. Low flying

Eleven Dutch RAF aviators died in Europe in six aircraft as a result of extremely low flying. Amongst these, seven men remained missing-in-action as a result of flying too low over the North Sea. The other four fell in the UK during practice flights, and none over Europe's mainland. None of these cases have been associated with bad weather, or any other special condition such as formation flying. Therefore these accidents can, as far as data is available, truly be classified under pilot's error.

Flying ultralow served tactical purposes: to escape from enemy radar sight, and to make it more difficult for an enemy fighter to pursue and attack. That meant that low flying had to be exercized. Six out of 120 aircraft with fatal Dutch RAF casualties is 5%, which can be considered as the usual outer edge of phenomena under the Gaussian curve. Pilots could have flown a bit higher and be safe, but somehow did not. There is little to analyse here.

For the men lost whilst flying very low over the sea, the analysis should differ. The 320 Sqn Coastal Command patrols were often flown very low, to escape from enemy radar sight, and this type of flying was done for many hours during an operation. Hours with only the sea in sight. A slight hiccup, in the local conditions or in the control of the aircraft, or an engine failure, could be fatal in a second. A seagull smashing into to cockpit windscreen could bring a very low flying aircraft down. Author cannot classify such accidents as merely pilot's errors. These pilots were required to fly at the edge for hours at a time. The slightest distance beyond that edge meant death.

Source: Tim Hamilton, 'The life and times of Pilot Officer Prune', London, 1991, p. 19

5. Oxygen failure

The six fatalities due to mechanical failure include four due to engine failure, and two due to presumed oxygen failure. We have two types. Failure of the oxygen supply system, and failure of the human system. The first one is a technical problem, the last one comes under the heading 'human factors'.

1. Failure of technical system

After a crash, it seems unlikely that the oxygen system could still be investigated for proper functioning. Oxygen failure is presumed as the cause for the crashes of J.W. van Hamel in Spitfire Mk. XIV and E.H. den Hollander in Hellcat Mk. I, both fighter pilots. Aircraft dived into the ground from great height, no enemy action involved. Pilots are presumed to have been unconscious. Oxygen failure is presumed to be the cause for unconsciousness. Pilots could also lose consciousness as a result of high G-forces during a steep dive. But the oxygen supply regulator is known to have frozen up on occasion.

Source: Jan P. Kloos, email 14/3/2005

It happened to René Wittert van Hoogland in 1935, when breaking the altitude record with a Fokker D-XVI. At an altitude of 11.000 meters, and a temperature of minus 54 degrees Celcius in the open cockpit, the oxygen system froze up. The pilot regained consciousness at about 1.000 meters, and lived to tell what had happened.

Source: R. Wittert, 'Het Vergeten Squadron', Amsterdam, 1983 (1976), p. 11

At 30.000 ft, temperatures are in the minus 50 degrees Celsius area. Unlike bombers, fighters did not have a cockpit heating system. The regulator was mounted at the side of the cockpit, a location colder than the dashboard, that had the aircraft's engine directly in front of it. The oxygen was stored under pressure. The regulator decreased pressure. Expanding gases cool down. The result of cooling gases and very low area temperatures may have been a frozen oxygen regulating valve. The technology was available to prevent this. Already in 1940, the British fighters had a wing gun air heating system, to prevent failure at high and cold altitudes. An extra hot air hose to the oxygen regulator could have saved lives.

WW2 Allied oxygen regulator, officially called 'Regulator, oxygen,diluter demand', Bendix, type 2872-B2.

Source: author's collection

2. Failure of human system

Apart from a problem with the oxygen supply system, several conditions could endanger a pilot's oxygen level in the blood, leading to loss of consciousness. These conditions are summarized under the term hypoxia, lack of oxygen. Aircraft have changed since WW2, the human body has not. Therefore these oxygen-related problems still exist. Also in civil aviation, but the occurrence is usually much more violent in military aviation. Hypoxia comes without warning. The aviator is unaware of the condition, as the warning system in his brain is one of the first centers in the nervous system to be affected. Pilots may even experience a sense of well being, before vision becomes reduced, confusion sets in, and shortly after that consciousness is lost.

The general rule is that the human body needs extra oxygen in the thinner air at altitudes above 10.000 feet, and at night above 5.000 feet. Going on oxygen at lower altitudes at night, was found to improve night vision.

Even with sufficient oxygen supply, the oxygen supply to the brain can be disturbed by high G-forces, such as exist when pulling out of a steep dive. The heart has difficulty in pumping enough blood to the brain, and 'the pilot was unable to pull out of a steep dive'. Whilst this certainly is a human factor causing flying accidents, it cannot be seen under the heading of 'pilot's error', as the pilot was ordered to make that steep dive. Either as an exercize, or as forced by the urgencies of combat.

Another type of hypoxia is carbon monoxide (CO) poisoning. CO is an odorless gas, produced as a result of fuel combustion in the aircraft's engine. The gas attaches itself with great ease to the hemoglobin molecules in the blood, impairing the vital ability of these molecules to transport oxygen. This type of hypoxia could occur if exhaust systems were leaking into the cockpit. Symptoms of CO poisoning are headache and dizziness. In a later stage vomiting, coma and death. CO is a slow killer, if not by itself, than by impairing the pilot's ability to fly in combat as a result of a killer agent that he could not see or smell. Again a human factor, this time caused by defective equipment and not by pilot's error.

Smoking cigarettes is a self-induced mild form of CO poisoning. Smoking was totally acceptable at the time, and we have to assume that many aviators smoked on board, especially bomber crews on long hauls to the target. Combat aircraft of WW2 would be well ventilated. If not by design, than by hasty construction and the air streams that entered via machinegun ports. Therefore, cockpits and gun turrets were unlikely to become smoke holes poisoning all occupants, smokers as well as any non-smokers present. Whilst smoking would decrease the oxygen supply to the brain, it can be argued that the addicts would have been in a worse shape if they had to refrain from smoking for a few to many hours.

Smoking introduced another hazard of the human factor type. The smokers had to lose ash and sigarette butts somewhere. They were not issued with nice and designed-for-the-purpose ash trays with a screw lid. They had to improvise. Meaning that many work places messed up. This could play havoc with their eyes, if the aircraft had to fly aerobatics to avoid fighters or Flak. In combination with poor maintenance it could be a factor in escape hatches that would not open when needed. See the crash of Wils & crew, 29/12/1944. Here the enemy Flak has taken one life of a Dutch RAF aviator, that would not have been lost if this human factor would not have been present. And this human factor was not located in the humans of the crew.

Finally, the oxygen supply to the brain can be disturbed by hyperventilation, an unconciously increased rate of breathing. This may come about as a result of stress, anxiety and pain. It leads to a deficiency of carbon dioxyde in the blood, leading to dizziness, hot and cold sensations, and finally unconsciousness. The remedy is to take it easy, which is a remedy that could hardly be given to combat aviators. A human factor certainly, but not a pilot's error. War is the error here.

6. Blood supply failure

Intended is blood, and with that oxygen, supplied to the brain, needed to keep consciousness. It can be considered as a form of hypoxia, but then with a much faster development curve. The blood supply could be disturbed by G-forces, as in violent turns and steep dives. The transition from consciousness to unconsciousness could happen in a second, giving the aviator no chance to redirect the fast moving mass of the aircraft. Several of the Dutch RAF aviators, mostly fighter pilots, 'failed to recover from a steep dive'. They may have started the dive recovery manoeuvre too late, but it is entirely possible that they lost consciousness as a result of that dive. Pilots differed in the amount of G-forces they could take. Those who discovered that tightening the leg muscles helped, could take more. But they were on their own, in working out these matters. Eventually, and during Wold War 2, this led to the discovery and development of the pressure suit. The suit would wrap the body, especially the lower section, in a layer of water, and later air. This was instrumental in keeping the blood supply in the upper section intact. The pressure suit enabled the pilots to withstand the results of steeper dives and tighter turns than would be possible without the suit. But many pilots would die, before this human factor was countered to a fair degree with technological means.

During and after the War, new techniques would be developed that enabled a pilot to sustain higher levels of exposure that would previously have led to hypoxia. War had become a great accelerator of research and development. Research data was quite often obtained without prior planning, as a result of the contingencies of War, and quite often involving the loss of life. High Command cannot be held accountable for failure to protect its aviators better, if the knowledge and technology to do so did not yet exist. But blaming pilots for making errors that led to fatal flying accidents, would be equally misguided if human factors beyond the pilot's capability of decision making were involved.

7. Engine failure

The number of fatal accidents that resulted from engine failure is quite low. Engine failure understood as occurring spontaneously, not as a result of enemy action. Engine failure could be fatal if it occurred during the take-off phase, or when flying very low. That's how Sitters and Van Hulzen lost their lives in Héricourt, France, in May 1940. We can't prove the point, but believe that the loss of two Fokker T-VIIIw's, who suddenly dived in from very low altitude over sea, may have been caused by a sudden failure of one of the two engines too.

However, in many cases engine failure left the pilot or the crew with the options to either bail out, or to glide to an emergency landing. But not always. Engine failure could be sudden, violent, and fatal. This happened to two Spitfires Mk. XIV's of 322 (Dutch) Squadron. The aircraft came equipped with US Packard built V-1650 Merlin engines, that sometimes had a troublesome supercharger. This turbo-like supercharger could break down, resulting in what appeared to be an exploding engine, and even a disintegration of the aircraft in mid air. As the supercharger was mounted at the rear of the engine, close to the bulkhead that separated the cockpit from the engine bay, such a malfunction could become fatal not only to the engine and the aircraft, but also directly to the pilot.

Merlin engine showing the large supercharger at the rear of the engine. RAF Museum Hendon 060201 Merlin

8. Bad weather

Bad weather could play havoc with military aviation in World War 2. However, the Dutch RAF/FAA aviators were mostly spared from this potential accident cause. The exception is the loss of Swordfish Mk. II Nr. LS244 of FAA 860 (Dutch) Squadron on 11/1/1944 in the Atlantic Ocean North of Ireland. The aircraft was flying escort duties from the merchant aircraft carrier 'Acavus'. The aircraft had to be emergency-landed at sea; a fuel shortage is presumed. The three man crew remained missing: Off Vl 3kl J. Slakhorst, Pilot, Off Wnr 3kl G.A.Q. Krijnen, Navigator, and Korp Vltg Telegr M. Boelhouwer. If winds would turn out very different from the predicted situation, either as a result of errors or of quickly changing atmospheric circumstances, then FAA aviators could get into serious trouble. Needless to say that in the vast expanses of the Atlantic Ocean, chances of survival were slim.

9. 'Friendly' fire

Officially, no Dutch RAF aviator was killed by 'friendly fire'. We have made a strong case that Rijklof 'Charlie' van Goens was accidentally shot by Allied anti-aircraft artillery off the coast of Dover. With a detailed explanation of the complicated technological environment in which this accident took place. We have demonstrated that there was a wide area of lethal danger beyond the boundaries of the Gun Belt, where the pilots were allowed to fly. In fact had to fly, as their bases were located within the Gun Belt. Meanwhile, Rijk got the blame. Our fingerpointing in this case is towards high brass, who demonstrated to be not in control of the circumstances where they could have been, or who did not care enough to do so. Not a pilot's error, but a high brass error situation. High brass escaped accountability, by covering up the loss, and by keeping the archives closed for a hundred years.

Regarding the loss of Hilbrand Holtrop, we can assume friendly fire to have been the cause, but we cannot produce sufficient evidence to prove it.

Source: "Steen", by this author, Soest, May 2005.

These are the only cases found so far, one most likely, one possible. Meaning only one or two out of 120 Dutch RAF fatal aircraft crashes have been caused by friendly fire. The USAF had to experience a higher percentage of friendly fire incidents, mainly due to a more widespread use of proximity fuzes on AA-shells in the Pacific theatre.

Nog omwerken:

Het sneuvelen van twee NL soldaten in Afganistan op 17/01/2008, door friendly fire, maakte de ogenschijnlijk verrassende stelling los in de krant, dat toegenomen technologie een toename van friendly fire-incidenten brengt. FF casualties in WO2 2%, in de Golfoorlog niet minder dan 17%. Inzake de NL RAF vliegers: ik ken minder dan een dozijn berichten over NL RAF vliegers die FF ontvingen, en slechts één geval van naar alle waarschijnlijkheid FF, met fatale afloop. Op 235 NL RAF fatalities in Europa is dat minder dan een half procent. En dat geval werd in de doofpot gestopt. Ik meen te hebben aangetoond dat dit geval werd veroorzaakt door samenkomst van verschillende nieuwe technologieën: IFF, proximity fuzes voor de AAA shells, en AA gun laying radar. IFF had friendly fire moeten voorkomen, maar was niet gekoppeld aan de radar. Dat was technisch mogelijk, maar het kwam er niet van door minder dan optimale samenwerking tussen de Britten met hun IFF systeem, en de Amerikanen met hun radar en VT-fuzes. De IFF operators moesten de waarnemingen van hun IFF ontvanger, met handbediende antenne, letterlijk doorschreeuwen naar de AAA vuurleiding. Terwijl de vier AA guns op enige tientallen meters afstand elk hun 8 schoten per minuut stonden af te geven. Deze technologie was superieur, maar nog niet helemaal goed genoeg. Aan het einde van de projektielbaan betaalde de vlieger daarvoor de hoogste prijs. Ik heb me ertegen verzet dat de beschikbaarheid van de hypermoderne spullen werd gemaskeerd door de vlieger de schuld te geven. Destijds was dat wellicht begrijpelijk vanuit security overwegingen, maar zo is het in de boeken blijven staan, en dat is niet terecht. Voor het overige lijken de mensen lui te zijn. Het eenvoudige maar onjuiste oordeel over pilot error bleef overeind. De pilot zweeg, want die was dood. Het komplete technische verhaal is ingewikkeld, en dus kennelijk slechter verteerbaar. Bovendien mankeren dan nog in de incidentverklaring de psychologische elementen, die volgens mij ook een stevige rol spelen. Een rol die niet essentieel minder wordt door toegenomen technologie. Dat zou je graag willen geloven, maar ik zie dat als denkfout. En wel een die in de komende tijd wellicht meer publiek wordt, in oproepen aan de overheid om haar soldaten beter te beschermen.

Als hierin een les zit, dan is het waarschijnlijk dat het mankeren van voldoende 'oortjes' hooguit een element is in de verklaring van het vallen van die recente FF slachtoffers. Bovendien, verbale kommunikatie onder vuur, in het donker, in vreemd terrein, is geen garantie tegen FF incidenten. Oorlogswerkelijkheid is daarvoor te ingewikkeld.

3. Most KIA/KIFA crashes occurred in the UK

Those who were killed in action fell in the following countries, analysed in number of men and in number of downed aircraft:

Table 8. Losses of men and aircraft versus countries in Europe

So 27% of the crashed aircraft went down over England or Northern Ireland, usually close to base. These UK crashes are exclusively caused by flying accidents, killing 58 Dutch RAF aviators, 40% of all Dutch RAF aviators killed as a result of flying accidents in Europe. The division by country excludes presumed crashes in the case of men missing-in-action. 82 of the 90 Dutch RAF aviators remained missing-in-action over the seas around Europe and the UK.

4. Good safety record of 320 Squadron

A division between aircraft type (fighter/bomber/trainer/other) versus cause of fatal accident reads as follows:

Table 9. Aircraft type versus cause of fatal accident

Het percentage verliezen door flying accidents, gemeten in aantallen vliegtuigen, is bij de bombers (33%) duidelijk het laagst, en bij de trainers, zoals te verwachten, het hoogst (82%). 'RAF Bombers' en 'Nederlanders' betekent in hoofdzaak 320 Sqn. Daarmee is het beeld in het geheugen van André Hissink, dat er bij 320 relatief weinig ongevallen waren, bevestigd, en in maat en getal uitgedrukt. Dat moest maar eens te boek gesteld worden, want ik meen dat dat nog nooit is gedaan.

Het aantal fatale ongevallen waarover we nog te weinig weten is met 16% nog vrij groot. Naarmate er meer bekend wordt, als dat tenminste lukt, zullen deze cijfers dus nog een beetje opschuiven. Vermoedelijk in de richting van méér verliezen door toedoen van de vijand, en dus relatief minder door vliegongevallen. Maar ik verwacht niet dat het beeld, hierboven gegeven, daardoor drastisch anders zal worden. 320 Sqn heeft 62% van de verliezen van NL RAF vliegers gedragen. Intussen was 320 Sqn wèl het veiligste Squadron, als er geen vijand in de buurt was. En dat dus in weerwil van het nauwe, in mijn ogen gevaarlijke, formatievliegen, het door de RAF in het diepe gooien van de 320 vliegers toen de nachtvluchten begonnen, en het feit dat bomber ops bijna altijd langer tot veel langer duurden dan fighter ops.

5. The art of riding the Flak

Two thirds of the Dutch RAF bombers that were lost as a result of direct enemy action, were shot down by Flak.

For the bomber crews, survival in the air depended on knowledge about enemy Flak positions, skill and even talent of the Commanding Officer in the air to select flight paths and lead the attack, and also on luck. Absence of bad luck would be the more appropriate expression.

Bomber crews were faced with the following dilemma. If the bomb load was to hit the target, then the bombing run had to be flown in a predictable way, meaning on exactly the right course and altitude, and then straight and level. But flying straight and level offered the enemy anti-aircraft gunners on the ground the best opportunity to zero in on the attacking aircraft. The acuteness of this dilemma was augmented vastly by the fact that enemy anti-aircraft defences would be concentrated in potential target areas, and along the obvious attack flight paths.

The attackers had the following options:

1. Attack from great height. High altitude meant that the attackers were more difficult to hit by Flak. But bombing accuracy decreased with increasing altitude. The attack altitudes chosen were a trade-off between these two factors.

2. Enemy air defence positions, as had become known through experience and/or photo reconnaissance, had to be avoided as far as possible. This was part of the pre-flight mission plan.

3. Besides pre-ops planning, airborne officers leading the attack had the freedom to amend attack flight paths as they saw fit in reaction to circumstances encountered there and then.

This translates into two decision-making aspects:

A. Evasive action in reaction to Flak encountered at the spot.

B. Evasive action in reaction to Flak anticipated at spots.

A. is purely reactive. Evasive action is flown as Flak shells are seen to come up and explode.

B. is instinctive. Evasive action is flown as a result from a gut feeling how enemy defences may be deployed.

A. can be learned, B. requires talent, a sixth sense. And it is here that Commanding Officers differed greatly. Some had that talent, some lacked the experience needed to develop a skill in this complicated area.

Evasive action could be flown in three ways:

A. Changing flight paths in the horizontal plane. This would make it difficult to the enemy Flak gunners to zero in on the attacking aircraft. However, deviations in the horizontal plane could not be made in the final stages of the bombing run.

B. Changing flight paths in the vertical plane. Enemy AA-shells had to be set to explode at a certain altitude; the enemy did not have proximity fuzes that could deal with this artillery issue in an automated way. Changing the flight path in a vertical plane would lead to enemy shells exploding too early or too late to do damage.

C. Combination of A. and B. Aerial warfare is a highly three-dimensional matter, and using the evasive potential of both the horizontal and the vertical plane at the same time was the best defensive technique. It boiled down to flying the most unpredictable pattern on the way in, keeping the actual bombing run as short as possible in time and distance, deliver the eggs, and get out in a pattern as unpredictable as possible. All of this done in a tight box formation of six aircraft. As was the case in the 2TAF attacks. The >1.000 bomber raids of the heavies resulted in less room for individual survival decisions and actions. Flying unpredictable patterns in these raids would easily have led to inter-aircraft collisions, or to bombs hitting the aircraft below.

It should be obvious that this combination of pre-ops planning based on reconnaissance data, and of on on-the-spot decisions by the Commanding Officer up there, as controlled by experience, gut feeling, and talent, was extremely difficult for many of the CO's, whilst some seemed to have a natural talent for it. A talent attributed by Jan Kloos to Wing Commander Alan Lynn DSO DFC, who guided the 2TAF Mitchells through many lethal situations.

The above applies to the loss of the Bakker crew, shot down over Lanvéoc-Poulmic airfield on 25/10/1943, see Chapter 45. It was the considered opinion of the eyewitnesses that Commander Bakker, who did not get a chance to develop experience, let alone a sixth sense, made a bombing run that was too long and with too little evasive action.

Riding the Flak could not be an exact science. We mentioned the need for a sixth sense. But even with that, there would be deadly surprises. The German Flak guns had wheels, enabling quick relocations, that could make the most meticulous Allied reconnaissance data obsolete within a day. This accounted for surprize Flak hits. The German guns and gunners were good. Without the help of gun laying radar and proximity fuzes, the gunners occasionally managed to achieve a direct hit with the first rounds. That could be translated as 'lucky' hits. There is no defence against Flak in an unknown and invisible position, manned by gunners who manage to hit with the first shot fired from a distance of several kilometers.

We feel that it is fair and justified to mention these intricacies with which the men at the time had to deal. As usual, the reality of the time was manifold and complicated. Failing to understand that, means a programmed failure to make sense of it all.

This chapter was written with great help from Jan Kloos

6. The high fatality episodes

If we plot the number of Dutch RAF aviators lost against the 60 months of WW2 described here, we can distinguish 5 episodes in which the loss of life exceeded ten men per month:

Graph 4. Dutch RAF monthly losses, expressed in number of men and aircraft

1. August 1941

During this month 19 lives are lost in the crashes of five 320C Hudsons, of which four are shot down by enemy fighters off the coast of Norway. Clearly the Hudsons were no match for the enemy fighters. Most of the casualties remain missing-in-action.

2. May 1942

Three Hudsons are lost over the North Sea, during attacks on shipping. All 12 crew perish. Two become victims of enemy fighters, one falls to AA-fire.

It is believed that these two danger episodes had to do with the very low altitude attacks of the time, making the attackers vulnerable to enemy shipborn Flak. Although it is probably true that these attacks were rather unsuccesful due to bombs that were designed for delivery from higher altitudes, the fact is that six out of seven of the Hudsons were downed by enemy fighters, not Flak. After these tragic experiences, 320C Sqn reverted to higher altitude bombing of shipping. Losses were reduced, but we may wonder whether this resulted from the modified tactics, or from the enemy's reduced ability to engage with fighter aircraft, as the War progressed.

3. October 1943

This month saw the loss of two aircraft in flying accidents, killing two Dutch RAF aviators, and most of all, the losses of three 320B Sqn Mitchells to enemy Flak, killing nine Dutch aviators. All lost over the Brest and Cherbourg hotspots. More aircraft were shot up, and aviators injured, sometimes even very seriously, but that is beyond the scope of this study.

4. June to September 1944

This is the invasion episode, that brought the highest loss of life by far, of 25 aircraft and 53 men in a period of three months. As far as known, most fell to enemy Flak. Most casualties fell in 320B Sqn, but this episode saw its complement of flying accidents too. Most noteworthy is the collision of two 320 Sqn Mitchells over Horsham, Sussex, whilst forming at night up for the attack flight. As these losses have a substantial impact on statistics, they are dealt with in some length below.

5. February 1945

During this episode, three fighter aircraft went down over Holland for different reasons, killing the pilots. Two Dutchmen lost their lives in four-engined RAF bombers lost during attacks on Germany. About half of the losses resulted from the exeedingly tragic collision of two 320 Sqn Mitchells over Tienen in Belgium. At the end the next Chapter, there shall be a substantial subchapter about this accident.

7. High anxiety and discipline

If we look at the aviator pictures in this study, we usually see young men smiling. That's what people usually do when looking into a camera. However, we often see faces that look at least ten years older than the men really are. A few Dutch RAF aviators were affected by their flying experiences to such a degree, that they became inoperative. This came about as a result of physical injury sustained during operations, but also as a result of psychological injury. It should be noted in the clearest possible way that the RAF allowed the men involved to become non-operational. This has to be considered as one of the strong points of RAF command, demonstrating a broader view than can be seen in other forces. Author believes, with the RAF of the time, that there is no point in pushing men beyond their limits, by waving the book of discipline in front of their faces. Such men may, unwillingly, become a hazard to the other crew. Author believes that it shows strength in RAF High Command decision making, when it allowed men to go on rest rather than to push them beyond their personal limits. That is, after the Battle of Britain had been won. This study carries quite a bit of implicit or explicit criticism of RAF command decisions, that were ill informed and/or ill researched, if at all, but this point stands out, and should be mentioned. RAF High Command may have blundered in several ways, leading to loss of life of which we can now, with the benefit of hindsight, raise serious questions about, but some aspects were dealt with by the RAF in a most commendable way, that has received very little attention in literature.

Polish memorial at Newark, England