Friday, May 6, 2016

Shadow Pasts: Another Anti-Correllian 1940 With Meditations on the Debacle

"Posterity will hear of these Battles," said the most unfortunate Flight title ever.

The British Air Force in France, 10 May 1940*

Air ComponentThe 

Air Component Headquarters
HQ Air Component-
Air Vice-Marshall C.H.B. Blount

No 14 Group

HQ No. 14 Group-
Group Captain P.F. Fullard
60 (Fighter) Wing
HQ 60 Wing  Wing Commander J.A. Boret
85 SquadronHurricane Mk I  Lille-Seclin
87 SquadronHurricane Mk I  Senon
61 (Fighter) Wing
HQ 61 Wing  Wing Commander R.Y. Eccles
607 SquadronHurricane Mk I
615 SquadronHurricane Mk I
  A flight : Le Touquet
B flight : Abbeville
63 (Fighter) Wing - created on 10 May 1940
HQ 63 Wing  ?only mentioned in source (2)
3 SquadronHurricane Mk I  Merville
79 SquadronHurricane Mk I  Merville
70 (Bomber Reconnaissance) Wing
HQ 70 Wing  Wing Commander W.A. Opie
18 SquadronBlenheim Mk V  
57 SquadronBlenheim Mk V  
52 (Bomber) Wing
HQ 52 Wing  Wing Commander A.F. Hutton
53 SquadronBlenheim Mk IV  
55 (?) Squadron  source (2) says 59 Squadron
50 (Army Co-operation) Wing
HQ 50 Wing  Group Captain A.R. Churchman
4 SquadronLysander  
13 SquadronLysander  
16 SquadronLysander
51 (Army Co-operation) Wing
HQ 51 Wing  Wing Commander A.H. Flower
2 SquadronLysander  
26 SquadronLysander  
81 (C) SquadronDragonCommunications Squadron

Advanced Air Striking Force

Advanced Air Striking Force Headquarters
HQ Air Component-
Air Vice-Marshall P.H.L. Playfair
67 (Fighter) Wing
HQ 67 Wing  Wing Co. C. Walter
1 SquadronHurricane Mk I
73 SquadronHurricane Mk I
212 SquadronPhotographic Recce Squadron. Only mentioned in source (1)
501 SquadronHurricane Mk I
 BethenvilleArrived 10 May 1940. Only mentioned in source (2)
No. 71 (Bomber) Wing
HQ 71 Wing  Air Commodore R.M. Field
105 SquadronFairey Battle
114 SquadronBlenheim Mk IV
139 SquadronBlenheim Mk IV
150 SquadronFairey Battle
75 (Bomber) Wing
HQ 75 Wing  Group Captain A.H. Wann
88 SquadronFairey Battle
103 SquadronFairey Battle
208 SquadronFairey Battle
Auberivesource (2) gives 218 Squadron instead
76 (Bomber) Wing
HQ 76 Wing  Group Captain H.S. Kerby
12 SquadronFairey Battle
142 SquadronFairey Battle
226 SquadronFairey Battle
98 SquadronFairey Battle  Nantes Acted as a reserve for Fairey Battle squadrons. Only mentioned in source (2)


  • (1) The War in France and Flanders, L.F. ELLIS, London : Her Majesty's Stationery Office, 1953. SO Code : No. 63-111-2-2*
  • (2) La campagne de France (1e partie) : La bataille du NordBatailles A√©riennes no. 7 Jan-Feb-Mar 1999
If your immediate impression is that the British Air Force in France was rather larger than the impression given in the literature, you are probably not wrong about that: 28 squadrons is not large in some abstract sense, but, combined with 22 "heavy" and 10 light squadrons of Bomber Command still based in Britain, it is not a small commitment to the Battle of France, either. Too bad, as the French would say, that the 47 squadrons of Fighter Command and 19 squadrons of Coastal Command  could make so little contribution to the fight. (My source also indicates another 4 squadrons of Lysanders, presumably attached to the remains of the old Army Cooperation Command. For training?)  Another way of looking at it is that there were more combat squadrons of the RAF available for action on northwestern Europe than there were battalions and regiments of the British Army in France. 

In some abstract way, this is not surprising given how much more money the RAF had received than the army over the last three years or so. In another way, it is. Twenty-two squadrons of "heavy" bombers (a,b.c) is nothing compared with the late war armadas, but modern America itself could not afford 22 squadrons of B-2s. (Well, it probably actually could if it put its mind to it, but why bother when that would be enough B-2s to blow up the entire world. More. Sorry. Forget I said anything.) It's all in the technology. 

Yes, it is time for a counter-factual exercise! 

Image source.

Today's question: What if the Royal Air Force of 1940 looked like the Royal Air Force of 1945? Or something of a wish list of an RAF of 1945. Bomber Command, was, in 1940, a motley mix of Whitleys, Hampdens and Wellingtons in the heavy squadrons, and Blenheims with one squadron of Wellesleys in No. 2 Group, operationally under BAFF. Instead, I shall have Avro Lincolns in the heavy groups, because my blog, my rules. (In reality there were probably some left over Lancasters, still pretty amazing bombers by the standards of 1940!). No. 2 Group, with its Blenheims and single squadron of Wellesleys(!) could be replaced with any of the late marks of Mosquito bombers without much prejudicing the argument. 

The home-based squadrons of Fighter Command offer an embarrassment of riches, with Tempests, late model Spitfires, Spitefuls, and even the Gloster Meteor and de Havilland Vampire all available to replace the 38 squadrons of single-engined fighters, and Mosquito night fighters and Hornets for the long-ranged fighter role filled by Blenheim IVs and Vs, for lack of anything better in 1940.

BAFF is the real target rich environment, however. Naturally, you want to replace the Hurricanes with either Tempests or Spitfires --I'm indifferent as to which. But what about the Typhoon, you ask? It was still in service, you say. Patience, I answer.  The Blenheims of the BAFF were intended to function as "reconnaissance fighters," a role filled by Spitfires and Tempests in 1945. I would make them Mosquito fighter-bombers, except that is the role for the Battle replacements. 

From this, then--

General characteristics
Crew: 3
Length: 42 ft 4 in (12.91 m)
Wingspan: 54 ft 0 in (16.46 m)
Height: 15 ft 6 in (4.72 m)
Wing area: 422 ft² (39.2 m²)
Empty weight: 6,647 lb (3,015 kg)
Loaded weight: 10,792 lb (4,895 kg)
Powerplant: 1 × Rolls-Royce Merlin II liquid-cooled V12 engine, 1,030 hp (768 kW)

Maximum speed: 257 mph (223 kn, 413 km/h) at 15,000 ft (4,600 m)
Range: 1,000 mi (870 nmi, 1610 km)
Service ceiling: 25,000 ft (7,620 m)

Climb to 5,000 ft (1,520 m): 4 min 6 sec



.303 in (7.7 mm) Browning machine gun in starboard wing
1× .303 in (7.7 mm) Vickers K machine gun in rear cabin

4× 250 lb (110 kg) bombs internally
500 lb (230 kg) of bombs externally

To this:

DH.98 Mosquito B Mk XVI[edit]

(The definitive bomber version).

Data from Jane's Fighting Aircraft of World War II[171] and World War II Warbirds[172]

General characteristics
Crew: 2: pilot, bombardier/navigator
Length: 44 ft 6 in (13.57 m)
Wingspan: 54 ft 2 in (16.52 m)
Height: 17 ft 5 in (5.3 m)
Wing area: 454 ft2 (42.18 m2)
Empty weight: 14,300 lb (6,490 kg)
Loaded weight: 18,100 lb (8,210 kg)
Max. takeoff weight: 25,000 lb (11,000 kg)
Powerplant: 2 × Rolls-Royce Merlin 76/77 (left/right) liquid-cooled V12 engine, 1,710 hp (1,280 kW) each

Maximum speed: 361 kn (415 mph (668 km/h)) at 28,000 ft (8,500 m)
Range: 1,300 nmi (1,500 mi (2,400 km)) with full weapons load
Service ceiling: 37,000 ft (11,000 m)
Rate of climb: 2,850 ft/min (14.5 m/s)
Wing loading: 39.9 lb/ft2 (195 kg/m2)
Power/mass: 0.189 hp/lb (311 W/kg)


Bombs: 4,000 pounds (1,800 kg)


GEE radio-navigation

Unfortunately, I'm lifting stat blocks from Wikipedia in a spirit of pure laziness, and the Wiki article doesn't have a stat block for the fighter-bomber variant. But you get the idea. We're going from a top speed of 264mph tto one of 415mph; and from a bomb load of 1500lbs to 4000. I believe the FB variant carried more like 2000lbs of bombs, but also had the standard air-to-air armament of 4x20mm cannons, and the inevitable racks for 12x60lbs of rockets. (Just because Grace Cook thinks that Flight magazine's obsession with aircraft rocket armaments is infantile does not mean that I deprecate the weapon. Although the operational research has established that they weren't very good at actually hitting things.) 

Finally, speaking of infantile, the interwar British army well-understood that the Air Council's emphasis on the strategic nature of air warfare, and the primary importance of the air superiority battle was just stupid fly boy stuff, and thus quite reasonably insisted that a large proportion of the aircraft that the RAF would fly over the land battle would consist of high-performance aircraft built with the most modern metallurgical advances, and equipped with a first-class fighter engine --and, since it was designed for "army cooperation," able to do nothing in the actual course of the battle of France except stooge around ducking from 109s. 

So let us hear it for the best and brightest minds of the anti-Air-Force-Mafia, and their glorious victory over the air marshals! Why, imagine if the Lysanders of 1940 had been the Hawker Typhoons which took over their role (for lack of anything better) in 1944? (Or, in an alternate alternate history, were just simply Hurricanes.) What a disaster the Air Force Mafia would have wrought!

Does that sound like sarcasm to you? Obviously it can't be anything of the sort. It would be implying criticism of a plane with a cannon that's bigger than a Volkswagen Beetle!

Renata cable ferry, sometime before 1965.
Oops, sorry. Wrong Beetle.

There we go. But since the Taylorcraft Auster AOPs were all in the Royal Artillery, all we've got left is the Typhoon. So, as ineffectual and anti-Army as it is to substitute dedicated army cooperation types with fighter-bombers during an all-out struggle for air superiority over the battlefield, that's what we're stuck with. Typhoons instead of Lysanders. 

So: It's 10 May 1940. The balloon goes up. The BAFF has 14 squadrons of Tempests and Spitfires, of which, say, 2 are photoreconnaissance types. It also has 4 squadrons of Typhoons, and 8 operational squadrons of Mosquito FB VIs. What do they do? Well, the battle starts off with a series of German coups de main, as long-planned offensives often do. The results were mixed --not as unfortunate as when Prince Eugene succeeded in capturing Louis XIV's worst general, but with some distinctly odd episodes. Bomber Command's first call was to react to one of the more unfortunate ones, the German parachute attack on Rotterdam, which not only failed to take the city, but left Dutch airports scattered with inoperative Ju52s, which the Blenheims of 2 Group took great pleasure in blowing up on the ground; something that Mosquitoes would do just about as well. 

That night, Bomber Command's heavies attacked airports some more, but also unleashed a staggering 9 Whitleys to attack Rhine bridges and German marching columns advancing on Fortress Holland through the Rhine bridgehead towns of Goch and Geldern. 

No results are known from these attacks.

The other, and by far the most successful German coup de  main of 10 May resulted in the fall of the brand-new Belgian fortress of Eben Emael, This allowed German Army Group B to quickly cross the Meuse/Maass water barrier and advance towards the Allied "Dyle Position," the intended stop line on which the advancing French and British would hold the Germans, allowing Belgium's mass conscript army to fight in position, as opposed to engaging in a war of marches, for which this 22 division army-raised-from-a-population-of-8.3-million was about as ready as you might expect from watching a representative sample of men aged 40+ perambulate.  

This was a bit of a crisis, and there wasn't much BAFF could do about it, even assuming that we repurpose Phantom Mission as RAF forward air controllers with 1945-era radios, scattered amongst the Allies. Ground attack aircraft can only attack targets which are called in for them in meaningful times. I'm going to add a codicil to my counterfactual, which is that the British Army has at least the radios needed for this work; but even if the RAF's 1940 Phantom Mission is repurposed as a number of RAF Regiment patrols of forward air controllers moving with Belgian and French forces, I do not see much opportunity for them to call in ground strikes. 

However, beginning on the morning of 11 May, Bomber Command could intervene, attacking the bridgeheads on the Meuse. Per Martin Middlebrook's definitive Bomber Command War Diaries (Now available on Kindle!), 23 Blenheims attacked the bridges, with 3 losses. (And 2 Bomber Command Blenheims abandoned reconnaissance missions due to weather. Assuming that the 1945 PR squadrons were a bit more pushy, the Allied Command would have had that much clearer a picture of what was going on.) On 12 May, 42 Blenheims hit the bridges, with a much heavier loss of 11 aircraft. After missing the 13th, on the 14th, 28 Blenheims raided, and 6 were lost. This was the last attack on the river bridges.

The losses tend to make this look like the Charge of the Light Brigade, albeit with much worse to come. Yet, suprisingly enough, Jeffrey Gunsberg notes that Erich Hoepner's 16th Corps' war diary characterised the Allied air harassment as heavy and effective. It is hard to guess just how much more effective 93 Mosquito sorties would have been over 93 Blenheims, but at the very least, we are talking about 4000lb payloads versus 1200. Losses would certainly  have been lighter, as a top speed of 400mph versus 266 (at 80% crusing speed: 320mph versus 214) means that much less time for AA to draw a bead, and a smaller interception window for enemy fighters. 

Night raids continued, but with little weight: 67 sorties against assorted communication targets over 3 nights, again with surprisingly little effect, although a 37 plane raid on the road and rail  nexus of Moenchengladbach handed the Germans a propaganda victory by killing an Englishwoman living there, along with 3 others. At least they found the town!

On the ground, on 12 May and continuing through the 15th, Army Group B came up on the advancing French and fought a meeting encounter battle, which, surprisingly enough, ended in a hard-fought French victory. (The British, meanwhile, were moving into place and sorting themselves out. Had the ground forces been equipped as they were in 1945, this would have been a more interesting battle, but that's for another day.) 

In this Battle of the Gembloux Gap, the Germans committed (per Wikipedia) 

. . . Luftflotte 2 supported Army Group B; its strength on 10 May included some 170 medium bombers and some 550 single-engine fighter aircraft and heavy fighters and although these numbers were not active during the first days of operation. The Oberkommando der Luftwaffe (OKL) reinforced Luftflotte 2 in the morning of 15 May with I. Fliegerkorps from Luftflotte 3 (which had some 300 medium bombers on 10 May). Above all, VIII. Fliegerkorps (which had some 300 Junkers Ju 87 Stukas on strength on 10 May and which specialised in ground-support operations) supported Hoepner at Gembloux.
This is the first all out struggle for air superiority over the battlefield. It is well within range of Fighter Command airfields in Britain, although the loiter time of even late-war fighters would not have been high, greater payloads having been largely eaten up by gigantic engines. On the other hand, they would have been available, as would all 16 squadrons of BAFF's single-engined fighter types, and presumably the greater part of Phantom's forward ground controllers.

A lot here depends on the effects of six years of aeronautical progress on air-to-air combat. Considering that RAF Tempests claimed a 7-1 kill ratio (It's in Wikipedia. It must be true!) I would say the table is being set here for a bit of a massacre of the innocents. Victory in air-to-air combat usually goes to the plane with the initial advantage, and that advantage depends on two factors: the window of interception opportunity, which in turn depends on aircraft speed and climb rate;  visibility, much higher with late war bubble canopies; and shooting, an area where gyroscopic gunsights and that four cannon broadside give the six-years-later planes a . . . considerable advantage. 

So the Battle of the Gembloux Gap is still a German defeat, but now decorated with German planes falling from the sky here and there. I can only imagine the average Landser's response if a few tanked-up Meteors or Vampires were able to make a quick battlefield sweep. Probably something about Perfidious Albion luring poor  Germany into a trap, though.

Of course, it's all actually a German trap, as on 12 May, the Germans complete their crossing of the Forest of Ardennes and arrive on the banks of the French Meuse. 

I haven't stolen nearly enough stuff from Wikipedia already today. The Meuse at Sedan. Source: By Donar Reiskoffer, CC BY-SA 3.0,
 This is going to get ugly. For the French, mostly, but once the Germans are across, those bridges are going to exercise a malevolent fascination for Allied air commanders. Bomber Command's war diaries testify to heavy Blenheim losses in attacks on the bridges, but those war diaries do not include the records of the BAFF, lost in the evacuation. We do, however, know that on 14 May, in a desperate attempt to stop the German penetration, 63 Battles and 8 Blenheims were committed, and 40(!) were lost. Since the official list includes 35 Battles, the Blenheim detachment actually suffered heavier losses proportionately. Apart from some unfortunate episodes involving Stukas and "flying gas tank" variants of the Wellington and, later, japanese bombers, which occasionallly achieved 100% losses, these are pretty much the heaviest losses of any air attack in WWII. 

I don't think that it is in doubt that Mosquitoes would have done better, although this does not look like very favourable circumstances for any air attack. Fortunately, the RAF, like a reasonably grown-up air force, has reserves that will allow it replace these losses, at least to a point. 

The breakthrough at Sedan led the War Cabinet to finally authorise Bomber Command to strike Germany beyond the Rhine and carry out really heavy attacks. On the night of 15/16 May 1940, 99 British heavy bombers attacked "16 different targets in the Ruhr area." Three separate factories, for example, were attacked by 9 aircraft each, while 12 other aircraft hit "communication targets."

You have to commend the Air Council for its optimism, although a casualty list of 2 people killed and 6 wounded (2 in Muenster, not on the target list or reported as attacked) suggests that this optimism was a bit misplaced.

Now I'm reminded of something.

Specifically, this aerial photograph of the marshalling yards at Aubenoye, France, part of a series of attacks made from the night of 7/8 March 1944, and continuing through D-Day. IN these nigh operations, Bomber Command intended to cut German communications with the Normandy battlefields. It is generally deemed that they were successful in doing so, and you certainly cannot argue that the attacks were not heavy and well-concentrated on vital targets. Only one of these raids, the attack against Hamm, Germany, involved the use of more than 22 squadrons of bombers. Generally speaking, 99 Lincolns, each carrying 14,000lbs of bombs, would have accomplished this much. In fact, given a penetration speed of 260mph at 20,000ft, it is likely that they would have been able to bomb in daylight, given some Hornets to drive off any Bf109s that could reach altitude in time. In a particularly crazy speculation, I suggest that Bomber Command could even have used its greater speeds to carry out day and night raids on the same day, at least at this crisis of the air land battle.

Meanwhile, and in fairness to the Bomber Command of real 1940, this sarcasm is directed at a particularly weak effort. On 17/18 May, 48 Hampdens visited Hamburg, setting 36 fires, gutting a fertiliser factory, damaging 150 buildings, and killing 36 people. Hamburg was always a vulnerable target due to the shoreline providing a convenient navigational marker, while the deep, fog-, and smog-shrouded valleys of the Ruhr were virtually impossible targets until they became sitting ducks with the introduction of hyperbolic radio navigation aids.

That'd be your Gee/Oboe/Decca/Loran. So good luck to the Krupp-Essen of Alternative 1940, because it's going to need it.

What immediate effect much earlier, more effective bombing would have had is another question: but bear in mind that a release from Cabinet restrictions would have been an opportunity for Bomber Command to carry out all of its long-planned stunt attacks. Presumably, the morning of 16 May would have seen the German nation waking up to blown dams in the Ruhr, a breached Bielefeld Viaduct and Dortmund-Ems Canal, and the Kiel Canal closed by combined pressure/acoustic/magnetic mines. 

At least, that's how it goes in the fantasy land of alternate history scenarios. I suspect that the "real" version of this fantasy would extend the destruction of these civil engineering mega-structures through ten days or more. Either way, it's not exactly good news for either the German economy or the offensive in the West. I'm also excluding the more pedestrian attacks on assorted bridges carried out by fighter-bombers and Mosquitoes.  

On 16 May, the success of the German breakthrough to the south led the BEF to pull back to the line of the river Escaut/Scheldt. If the German breakthrough were not successful, then it is pretty much game over for Germany; assuming that the heavier weight of RAF air attacks have not so far changed the course of (alternate) history, it is now time for the RAF to wage a more conventional battle for air superiority and ground support over the Forward Edge of the Battle Area. At this point, the history of the land campaign becomes almost impenetrably obscure due to British unwillingness to admit that they pulled their troops out of the line to save their line of retreat to the coast, making Allied defeat in the Battle of France inevitable. The official history is almost as weaselly as an old-time Economist editorial, but finally gets around to admitting that the British Commander-in-Chief demanded to be permitted to withdraw from the position, unhinging the Belgians and forcing them into a fighting retreat which their army could not possibly prosecute.

So it's all the Belgians' fault, in other words. 

This is the last occasion at which the Allied position is tenable. It will require a successful counterattack from the north across the German penetration along the line of the Somme, something that will only be possible with time and preparation, but for which there are intimations of potential success, given that the French DLMs (the French armour divisions equipped for assault and exploitation) are north of the breakthrough.

Oh, and if the leading mobile divisions of Army Group A are not reinforced in a timely way by the slower-moving infantry divisions which follow, by, say, air interdiction. 

It is as well to note that by this point, British single-engined fighters available to be committed to the battle in this scenario exceed 50 squadrons. While not all of Fighter Command will be available by a long shot, British air superiority assets are by now comparable in numerical strength to German, and possess a vast performance superiority that might very well be achieving that 7-1 kill ratio superiority of which I was skeptical above. 

In short:

i) The Allies will by now have air superiority, if not air supremacy (It is very possible that by now the Germans owuld have withdrawn their air assets from the day battle by now in the face of the casualties likely resulting from these lopsided air fights);
ii) A powerful arm of close air support;
iii) A crushing weight of strategic air interdiction in the rear of the battle;
iv) The incidental capability to carry out very heavy and damaging attacks on German industry. 

For these reasons, it seems very likely to me that Fall Gelb would have reached its high mark on the 21. Given that a British armoured division and 3 infantry divisions are awaiting immediate deplooyment in Britain to further increase the overall numerical balance against the german, it seems to me that, sometime before the end of May, a "Miracle of the Canal du Nord" would have bundled them back to the Meuse.

All of this is, of course, absurd. The useful point of the exercise lies in understanding why it is absurd. For that, I turn to a 1946 article in Aircraft Production, of which its editor was so proud that he reran it in Aircraft Engineering. 

A now all-but-forgotten but authoritative aviation monthly, Lockwood Marsh's paper is good for some pretty interesting insights into wartime industry. This article, recounting a visit the Kingston-on-Thames works of Hawkers Aircraft, of the Hawker-Siddeley Group, is one of those. Kingston was then building the achingly beautiful Sea Fury for the Royal Navy. Generations of aviation enthusiasts who have watched surviving examples swap world piston-engine speed records with the Grumman Bearcat.

It's just an amazing plane, with none of the tubbiness typical of older radial-engined fighters. The 18 cylinder, 54L, 2480hp Bristol Centaurus engine drags this 9,420lb airframe (14,000lb auw, but that looks to me like a catapult limit, and not the actual aerodynamic limiting weight) around the sky at a maximum speed of 460mph at 18000ft. That's more than 100mph faster than a Messerschmitt Bf109E of 1940, and at a greater altitude, too, which is quite an achievement for a carrier fighter. 

This comparison, across, after all, just five years of war, should give anyone pause. What is going on here?

Unfortunately, the resulting article might as well be in Martian for anyone but an industrial engineer. Nevertheless, there's an interesting story to tell here. Marsh phrases it in terms of evolution:

Although the Hawker Sea Fury is an individual design, its structure has been developed step-by-step from its forebears. The Hurricane had a tubular fuselage and a metal-covered wing: on the Typhoon, the rear fueselagte becacme a monocoque; while in the Tempest the semi-elliptical wing appeared. The wing of the Fury is generally similar in construction to that of the Tempest; but differs in the important poitn that it is continuous benath the fuselage instead of being bolted to the sides. With the introduction of folding wings, when development was centred on the Navay Sea Fiury in 1945, further modifications were made that resulted int he present wing design.

This gradual growth, type from type, over a period of years, has, in many ways, assisted the design engineers, so that a fairly closely integrated production control exists at Hawkers . .

Sure. Evolution. I get it. Marsh can't really get at all the evolution that's been going on, especially as what he wants to tell us readers about is this "integrated production control." I have a different perspective. I want to stop and gape in wonder. 

The Hurricane's "tubular" construction was the famous "Hawker Construction Method," introduced with the Hawker Hart (actually the Hoopooe, IIRC?) The Air Ministry wanted an all-metal aircraft, and Hawker gave them one.

It looked like this, and the metal construction material consisted of assembling it out of steel tubes, with jointing clamps riveted closed. So, basically, the same way a bicycle is made --if it's a bicycle made without once bending or welding the tubes. Hawker adopted this glorified Meccano construction method for the fairly obvious reason that it was still basically a woodworks shop, and was run like one. It could go to some suppliers, order a few pallets of standard parts, and then assemble them to the structure called for by the drawing office. 

No heat treatment that might ruin temper was required, and just as well, considering that the plant didn't have furnaces to start with. The thought was that the tubes might eventually be reinforced by running box girders down their middle, but no-one had done the math on this, and no-one was exactly eager to start. Fortunately, the rest of the industry worked at the same level, and, after seeing off a challenge from Boulton Paul's similar patent aircraft construction method, the Hawker technique swept the industry, as Depression austerity led the Air Ministry to place orders for approximately a million Harts, and Hind follow-ons, in order to reduce costs by extending production runs. 

The first Hart was delivered to the Aircraft and Armaments Experimental Establishment in June of 1928. Hurricanes were still being built in a slightly more complex version of the Hawker Method in 1939. The first Fury flew on 21 February 1945, less than eighteen years later.

I don't know about you, but I was working at my current job eighteen years ago, and I find this amazing. Why?

In the last six years, Kinston-on-Thames has become a Wonderland of specialised machine tools. Much of this work has the potential to heat metal and undo the heat treatments used to improve the mechanical performance of aluminum alloy parts. Fortunately, Kingston now has furnaces, and refrigerators, and spraying rooms: all the complicated paraphenalia needed to make sure that the Fury's structure can withstand the loadings imposed by this engine, and by its payload. 

Nor is this a case of tools improving on, or replacing, only hand labour.

This is how the Hawker Hart and even the Hurricane were designed. (Actually, it is Lockheed's Skunk Works on its first project, the P-80, but you get the idea.)

And this is the next step, at least at Consolidated. (I don't think that a design as small as the Fury would need to be lofted, but you get the idea.)

And this is one of any number of devices for replacing "brain work," and making the Fury possible.

It's not the best example I can think of, but it is the best I could come up with before I gave up on searching my hard drive. Sorry: some day I'll be patient enough to dig up that weird Boeing automatic drawing table again.

No, wait. As soon as I say that, I find it again.

I know that it doesn't look like "artificial intelligence," but considering that 21st Century artificial intelligence researchers gave up on AI that looks like AI ("Open the pod bay doors," etc), I'm not apologising. Much.

Machines, people, skills. Wrapping our minds around these communities of practice requires some kind of geography of the mind: a place to put them. Fortunately in this case, they are also geographies of place!

This picture was taken by Ant Basterfield, from the old dispersal pads for the Bristol Aeroplane shadow factory, near Hutton, North Somerset. Some thousand BEaufighters were built here during the war, and many Westland helicopters, after. Source. Nowadays it looks like it would be a good site for a grow-op or something.

Like many of the shadow factories of WWII, Bristol Hutton does not sound like it had a terribly impressive war. The Beaufighter has had intermittent good press, but it is still the plane that the RAF was stuck with until the Mosquito came along, and who really cares about the boring postwar British industrial economy of Andy Capp and everyone on the dole, etc, etc. Financial services! That's where it's at!

Britain built 26 "shadow factories" to serve the air war proper. Since my copy of  Roger David's Shadow Factories is apparently walking to Vancouver, and I am certainly not off to the library to dig through Postan on British War Production, I will settle here for Wikipedia here, and its [incomplete] list.Twenty-six factories, twenty-six communities of practice. It would be easy to tell a melancholy story of decline and despair framed around abandoned shadow factories, but it would also be a misleading one. Castle Bromwich, which actually got as far as making Lancasters during the war, is still doing stampings, final assemblies, and detailing for Tata Motor's Jaguar marquee, although the plant seems to be on borrowed time. On the other hand, as you will notice if you follow up on the Clifton link, and find out about the Magnesium Elektron corporation (long since incorporated into a larger Group), which makes hafnium for the British nuclear submarine fleet. 

This might be telling, in that it was another intervention by the military-industrial complex which redirected Clifton. Who knows if it would not be another housing estate or distribution centre were it not for that contract. Or underperfoming, like Castle Bromwich.

If this is getting to sound like another encomium precisely to defence needs as leading, pulling and directing industry towards technological frontiers of practice, well, that's because it is. It doesn't need to be an endorsement of it, as long as there's another way of going about giving industry a helpful shove in the right direction. It's just that there doesn't seem to be. 

A big push, administered by Neville Chamberlain, of all people, in 1937:

An impetus, begun in terms of Blenheims, completed in terms of Land Rovers, just about the time that these young men would have retired. Now exhausted? I don't know. I do know that closing down communities of practice means opening them up again before they can exist once more. Once you close a factory, in other words, something has to happen before it opens up again. Maybe the invisible hand of the market will accomplish that.

I can, however, tell you that relying on the invisible hand to work the ice cream when you can't get a clerk in to do it has not been successful for us laely.

But something something comparative advantage.

*(Every credit is due to France 1940, but unfortunately this is another of those disembodied voice websites. Thank you, Some Guy!) Also, the actual source is the official history, Elles, War in France and Flanders.)

1). Carbonitriding. Obviously written by an engineer rather than a physicist, and one would prefer a description in terms of quantum chemistry, as this would make it clearer that "resistance to impact loading," as they say in the automotive industry, is actually an issue of thermal conductivity and resistance to dislocation migration. This would make it less counterintuitive that carbonitriding is also used to protect surfaces against heat and chemical attack, for example (probably) in the barrel of the 17 pounder.  I should also admit down here that the actual purpose of the Banner Lane plant was to produce parts for Bristol Hercules engines, and not tank transmissions. I am still going to make the claim, though, as wider use of carbonitriding plant would have brought down its costs and made longer production runs practical. 


  1. Fun fact: the Mossie's power to weight ratio after an engine failure on take-off with full load was only 33 per cent less than the Battle's in normal operations, a tribute to just what a terrible aircraft the Battle was and the vastly design- and craft-intensive process of continuous improvement that put the Merlin through 76 major versions between 1030 and 1710hp.

  2. After praising shadow factories, it's a bit embarrasssing to admit that the Battle's lifespan was extended by its selection for manufacture at Austin's shadow plant of Longford, most recently the home of the MG6. "A convenient plane for a new company to build," I think the official history has it. Although I suspect that those would have been the planes transferred to the Commonwealth Air Training Plan, and that the ones massacred over France would have been from the initial tranche, which baptised Fairey's plant at Heaton Chapel, Stockport, previously National Aircraft Factory Number 2, operated for the Ministry of Munitions by Crossley Motors. (The things you learn.)

    Later the home of the Fulmar, Barracuda, Halifax, Beaufighter and Gannet, Stockport is, uhm...

    Well, it's next door to McVities, so at least this proves that business success isn't contagious.