The Italian torpedo boats ”Castore” and ”Montanari” firing upon British MTBs and MLs at Tobruk harbour, part of Force A, Operation Agreement, 14 September 1942
Bombs exploding over Tobruk as part of the air attack
on 13th/14th September 1942. (IWM CM 2990)
‘Never in the whole history of human endeavour have so few
been buggered about by so many.’ So some unknown wag from Middle East Commando
quipped. And he wasn’t wrong. Special Forces are by no means popular with
regular commanders. Montgomery had little time for them, and whilst he wasn’t
involved in the planning of Operation Agreement and bears no responsibility for
its failure, he wasn’t slow to add his own stridency to the universal
recriminations.
Winston Churchill liked commando raids. He liked the
Henty-esque derring-do and the idea of clandestine assaults on the enemy where
he least expected to be attacked. The prime minister also liked the idea that,
despite the many reverses Britain had suffered and despite the ongoing crisis
in the Western Desert, the constant tale of woe from the Far East and the
parlous position of our almost-allies in Russia, we could somehow hit back.
Besides, the Germans didn’t really understand the whole commando concept:
amphibious operations and peripheral strategies were not really their style. Tobruk
was also dear to Churchill’s heart. The Great Siege rated as an Allied victory
when these were very scarce indeed. The final, humiliating fall was a massive
blow. He wanted the place back or at least convincingly denied to the Axis.
The idea of attacking the Axis at Tobruk was by no means a
novelty in the summer of 1942. Operation Agreement probably owed its genesis to
a scheme put forward in October 1940 at the very start of the Desert War. The
aim of the first proposal was much the same, to destroy fuel dumps and harbour
facilities. There was no shortage of suitable targets: four large oil tanks by
the harbour, each of 32,000 gallon capacity, another four Benzedrine vats
nearby, petrol stores north of the local settlement, and a dump seven miles
south of the town and the El Adem junction. The power station, magazines,
wireless stations and distillery were all ripe for destruction.
All of the main jetties projected from the north flank of
the harbour, together with a slipway. Water depth at the end of the piers was
between 14 and 20 feet. The coaling jetty and boom jetty provided anchor points
for a brace of booms cordoning the harbour. One of these stretched from Marsa
Agaisa to a bunker on the northern flank, whilst the other reached from the
southern side 400 yards west of the Marsa Sciarfa up to the coaling pier. The
entrances were mined and the fighting strength of the garrison around 17,000
strong; a formidable defence. The port was guarded by numerous coastal guns.
One vulnerable feature was the access to the port area from the lower of the
escarpments ringing the town. Two fairly narrow tracks came running downhill;
blocks placed across these could create a bottleneck and halt the flow of enemy
reinforcements.
There were two alternatives: a classic hit and run night
attack or a prolonged occupation that would breach the daylight hours. It was
recognized that, as ever, with raids, surprise was the key element. For that
reason it was deemed essential that the raiders be brought in by fast
destroyers. This was logical, but there weren’t enough of the sleek warships to
be found. Another essential was overhead fighter cover to facilitate the
withdrawal, tricky at best. Again, fighters were scarce.
#
Four groups of raiders were to be deployed. Group ‘A’ would
land to the west and hit the fuel dumps north of the aerodrome. The second,
Group ‘B’, would land simultaneously at the same spot, then attack those
installations east of the naval barracks and take out any planes that happened
to come within their sights. Group ‘C’ were to hit the coastal guns north of
the settlement and possibly the fuel stores in the same vicinity. The last
group would simply attack the town itself and generally ‘cry havoc’. This would
sufficiently distract the defenders and prevent them being a nuisance to the
other groups. This last formation, Group ‘D’, would comprise a single Special
Service Company, and they would target command and control centres during their
spree. The men would carry weapons and ammunition only.
To convey the raiders only a modest naval flotilla was
required: four destroyers, the same number of motor torpedo boats (MTBs) and a
single submarine. The approach, however, would require the involvement of the
entire Mediterranean Fleet as a grand diversion. The ships would sail as though
preparing to shepherd a Malta convoy, and the attack force would peel off
towards the Libyan coast. By the time they were steaming towards the hostile
shore, the lone submarine would have marked the landing zones and guided the
laden warships in. Meanwhile, the MTB flotilla, coming westwards along the
coast probably from either Alexandria or Crete, (still in Allied hands at that point),
would make smoke to cover the actual landings. As soon as the troops were
onshore, the landing boats would withdraw and be re-hoisted onto the
destroyers, which would then stand clear, safely out of range of the coastal
guns, until needed to take the raiders off.
Here was the critical point. All amphibious operations may
be said to succeed or fail according to how efficiently and swiftly the men are
put ashore. The Allies did not, at this stage in the war, possess sufficient
specialized landing craft. They relied, as ever, upon innovation and making do
generally. Making do could take one of two forms. The normal ships’ boats could
be used. These were tried and sturdy, but clearly would be insufficient in
terms of available numbers. Additional boats could be found and more davits
fitted. Alternatively, some form of lightweight specialized craft could be built.
This seemed reasonable. The boats would have to be seaworthy and reasonably
easy to navigate, but making do like this rather depends upon the seas being
co-operative. The Mediterranean is rarely so obliging.
Though they opted for the second choice, the planners of
Operation Waylay, as the scheme was dubbed, were aware of the limitations. Calm
seas and light northerlies would be vital. It would take the raiders five hours
to accomplish their allotted tasks with a further period of three and a half
hours needed to complete the landings. Near-total darkness was clearly another
essential. What the Waylay team proposed was that the air raid should take
place after rather than prior to the raid. This and a naval bombardment would
provide the shield behind which the commandos would re-embark. We cannot say
exactly what influence the early idea exerted over those planning Agreement,
but in hindsight we can say it was probably a better plan, or at least less
flawed.
When Tobruk fell to Rommel, it wasn’t just a blow to
Churchill’s pride and Britain’s tottering esteem, but it netted the Axis a
significant haul of booty. The Allies had reaped a similar harvest when prising
the place away from the Italians. Hitler had written encouragingly to his ally
Mussolini, sufficiently so for Il Duce to plan his own triumphal entry into
Alexandria as the new Caesar, even if any such entry would largely be effected
on the back of German efforts. In fact, Rommel had failed, but the hot desert
summer of 1942 saw Allied fortunes at very low ebb. The Desert Fox had stumbled
at the final hurdle, but he was still unbeaten. With hindsight, it is possible
to see how the position had in fact shifted. The Allies, in Montgomery, would
finally have a general of equal worth, and the build-up of strength,
facilitated by Britain’s American allies, would finally tilt the balance
Monty’s way.
Admiral Andrew Cunningham had commanded the Mediterranean
Fleet with great élan and considerable success. Il Duce had been thrashed at
sea as comprehensively as on land. His successor, Admiral Sir Henry Harwood,
faced a difficult challenge. His ships had taken a fearful pounding trying to
succour Malta. He had neither battleships nor carriers and the RAF could
provide only very limited cover. The savaging meted out to Royal Navy ships
salvaging survivors from the mess on Crete had shown just what damage sustained
aerial attack could do. Churchill was still expecting the reduced fleet to
achieve prodigies, and, at the same time, to arrange blocking operations against
Tobruk and Benghazi.
This was a favourite obsession of the prime minister – using
blockships to bottle up enemy ports. It was highly difficult, dangerous, of
dubious long term value and bound to be expensive in terms of both ships and
men. Cunningham had vigorously resisted any such notions. In mid-April 1941, he
had been urged to commit one of his only battle-cruisers, HMS Barham, in an
attempt to block Tripoli. Cunningham considered the idea crazy and resisted.
Churchill personally intervened to press the scheme. The admiral ignored the
exhortation.
Once Tobruk had fallen to Rommel, such daft ideas were
resurrected. On 21 July, Harwood received a message from Whitehall (undoubtedly
inspired or even drafted by the prime minister) that he should send a destroyer
to attack shipping in the harbour: the signal did admit ‘this is a desperate
measure’. Hysteria was the order of the day, a mood ably caught by the
well-known commando Fitzroy Maclean: ‘In Cairo the staff at GHQ Middle East
were burning their files [‘Ash Wednesday’] and the Italian colony were getting
out their black shirts and fascist badges in preparation for Mussolini’s
triumphant entry’. It was in this heated and fearful context that the idea for
Operation Agreement took root and began to grow.
Whilst the notion of the earlier concept, Operation Waylay,
may have formed a viable precedent, both Stirling and Haselden had put forward
ideas for limited attacks against Tobruk and Benghazi. These were plans for
clean, surgical strikes involving land forces only, SAS, LRDG and other
raiders. What would become the plan for Force B was Haselden’s idea, a group of
commandos, sneaking through Tobruk’s defended perimeter, attacking fuel
installations, then withdrawing swiftly across the desert. This would be something
on the scale of the actual LRDG attack on Barce, which did achieve some gains,
though these came at a high price.
When suggesting an attack on Benghazi, Stirling had,
unwittingly, opened a Pandora’s Box by proposing to add a naval element including
a blockship. Fitzroy Maclean, always a beau sabreur of the cut likely to appeal
to the prime minister, found himself dining with Churchill in Cairo. He
recalled:
The plans for a raid on Benghazi had been greeted with
enthusiasm at GHQ. With such enthusiasm that by the time they came back to us
they were practically unrecognizable. The latest scheme envisaged a major
operation against Benghazi, to be carried out in conjunction with similar large
scale operations elsewhere.
Another commando element was added by the Special Boat
Service (SBS). This unit was the brainchild of Lieutenant Roger Courtney, who
had joined the commandos in mid-1940. He had the idea of specialist raiders who
would approach from the seas using folding kayaks. Initially nobody seemed
interested, so he adopted the bold and unorthodox tactic of launching his own
private raid against HMS Glengyle moored in the Clyde. He got aboard undetected
and wrote his initials on the door of the captain’s cabin, seizing some booty
as further proof. He flung his gains at the astonished feet of his superiors
then dining in the Inverary Hotel. Message received, he was promoted and given
a dozen men to nurture his new unit.
The Folboat was some 16 feet in length, a rubberized canvas
surface stretched over a timber frame with front and rear buoyancy bags. These
handy, folding kayaks took two men and their kit. The Folbot troop became No 1
Special Boat section early in 1941 and the team was deployed to the
Mediterranean as part of Layforce. Courtney’s raiders successfully carried out
a series of missions and returned to the UK in December to recruit a second
formation. The original bunch were grafted onto Stirling’s SAS as the Folboat
Section and carried out a further series of raids during the early summer of
1942. It was felt by the Directorate of Combined Operations that up to half a
dozen teams, taken in by three subs, could paddle into Tobruk Harbour and fix
limpet mines to Axis vessels there. Once their charges were planted, they would
slide out of the harbour and get back by moving only at night and hugging the
coast till picked up by MTBs. The idea foundered, however, as there weren’t
enough submarines available.
Rommel’s Achilles heel was his supply route. If both Tobruk
and Benghazi were hit and successfully put out of action, even temporarily,
then his logistical troubles would multiply, forcing him to extend his supply
lines even further. Such a deprivation of resources at a time when both sides
were girding their loins for what would be the decisive clash could reap a huge
dividend for the Allies, though by no means all of the planners were convinced.
On 3 August the Joint Planning Staff (JPS) set out their
initial considerations in Paper 106. The prime objective was to destroy harbour
facilities and installations at both ports, as it was thought that this may
well lead to the rapid defeat of Rommel by land forces. This was wildly
optimistic at best, wishful thinking embodied in tactical planning. Operation
Agreement began with a wish that was built up into a plan; pious hopes, fed by
frustration and desperation, led from the start. The planners went on to detail
the units that could be employed to find sufficient forces without pinching
from 8th Army. This was fine by Monty, as the operation wasn’t his and the
resources would not be his. If it went well, he could look to grab some of the
credit; if it failed, he could simply stand clear.
Ideally, both places should be attacked simultaneously, if
the forces available were sufficiently strong. If not, then Tobruk would remain
the prime target. In each case the tactics would be similar. The landward party
would be responsible for rushing the coastal guns, taking these under new
management and turning them against their previous owners. Amphibious raiders would
be responsible for most of the demolitions, which would be blown in daylight.
The whole lot would re-embark at dusk. This draft was then reviewed by the
Director of Military Operations General HQ Middle East Forces (DMO GHQ MEF). He
broadly concurred and 8th Army would not be placed to start attacking before 30
September. Rommel was getting new tanks through Benghazi plus some 1,200 to
2,000 tons of supplies per day; rather less was coming in through Tobruk.
If the raids could be carried out by the middle of August,
then Rommel might be seriously embarrassed. The initial drafts highlighted the
risks and advised that losses might be heavy. The final draft acknowledged
casualties could be as a high as one hundred percent of those taking part in
the landings. It wouldn’t be feasible to launch the raids until the first week
in September and the JPS were never more than lukewarm. The commander-in-chief,
however, seized upon the idea and his reaction to Paper 106 was galvanic; ‘I am
in NO rpt NO doubt that it is essential rpt essential that these operations
take place in August and that probable losses must rpt must be accepted’. It
doesn’t get much plainer than that. Operation Agreement was now pretty much
assured, and there wasn’t going to be room for doubters.
If this wasn’t emphatic enough, the commander-in-chief went
on to stress how important the psychological aspect would be, uplifting for the
Allied troops and depressing for the Axis. If any lingering qualms persisted,
the JPS were exhorted ‘to adopt a more vigorous and offensive habit of
thought’. The die was cast, and the operation would be vigorous and aggressive.
It is probably not entirely coincidental that Churchill was in Cairo at this
time. The tone of the communication does rather suggest his style. GHQ got the
message and agreed their planners were falling short of the bulldog
temperament. If the PM was so adamant, who were they to object?
To deliver before September was problematic, even with the
most snappish of bulldogs barking. It couldn’t be done during the dark moon
period in August, though Admiral Harwood did not apparently consider that full
darkness was necessarily vital. Overland elements would have to approach via
Kufra, an immense distance to cover, and some of those among the raiding
parties might not be fully trained up. As Peter Smith points out, Operation
Agreement had become Topsy; it just kept on getting bigger. A further strand,
Operation Hoopoe, an attempt to recapture Siwa Oasis, was now bolted on. An all-arms
force there could create merry hell with enemy transport and oblige the DAK to
detach substantial forces to remove the threat. The bigger the threat, the
bigger the response, so the idea was to beef up the attacking force, providing
AA (anti-aircraft) cover and giving the enemy something really massive to worry
about.
It was known that the Italian garrison at Siwa was quite
small, at best a weak battalion, with no armour and only a quartet of 37mm
Breda AA guns. It could be attacked by LRDG/SAS, with some armour to add a heavier
punch, and then the main force could move up and take ownership. This would be
a hefty contingent including a regiment of Honeys, Bren carriers, transport,
signals, guns, engineers, medics, RASC and RAF detachments. Happily, the
commander-in-chief wisely decided against such a commitment and Hoopoe went in
the basket. Popski, most irregular of irregulars, summed up the comic opera of
GHQ in unflattering terms:
Friends joined in with suggestions picked from boyish
books that they had pored over in earnest only a few years before, Drake and
Sir Walter Raleigh, Morgan and the Buccaneers were outbidden; new stratagems
poured out in a stream of inventiveness.
This is probably not such a gross exaggeration.
Lieutenant-Colonel Calthorpe on the planning staff undertook a review of the
planning process as it stood in the latter part of August. He stressed that the
prime objective at both ports was to take and hold the enemy’s defensive ring
or those elements that could fire on the demolition parties. As with Operation
Waylay, the prime factor was surprise. Where Calthorpe differed was in the
timing of the air raid. He wanted this before and not after, to cover the
approach rather than screen the withdrawal. The reasoning behind this is
understandable, but what price surprise? The enemy would not require high
levels of tactical insight to twig that they were being softened up prior to an
attack. There was also the matter of timing; it had to be either mid-August or
from 8 to 13 September when moonlight was minimal.
Friends joined in with suggestions picked from boyish books
that they had pored over in earnest only a few years before, Drake and Sir
Walter Raleigh, Morgan and the Buccaneers were outbidden; new stratagems poured
out in a stream of inventiveness.
This is probably not such a gross exaggeration.
Lieutenant-Colonel Calthorpe on the planning staff undertook a review of the
planning process as it stood in the latter part of August. He stressed that the
prime objective at both ports was to take and hold the enemy’s defensive ring
or those elements that could fire on the demolition parties. As with Operation
Waylay, the prime factor was surprise. Where Calthorpe differed was in the
timing of the air raid. He wanted this before and not after, to cover the approach
rather than screen the withdrawal. The reasoning behind this is understandable,
but what price surprise? The enemy would not require high levels of tactical
insight to twig that they were being softened up prior to an attack. There was
also the matter of timing; it had to be either mid-August or from 8 to 13
September when moonlight was minimal.
As planning moved beyond feasibility into detail, it was
recognized that both attacks should ideally go in on the same date. Surprise
only comes around once. As the assault on Benghazi could not be staged before
the end of August, it made compelling sense to deliver both the following
month. The delay would allow for sorely needed training and preparation.
Destruction of enemy supplies would be as damaging in September as it would
have been in August.
Broadly then, the plan for Tobruk was that a land-based
element would attack the coastal guns at the south-east end of the harbour
before moving westwards to seize the additional guns on the south side. This would
have to be accomplished in darkness, so detailed local knowledge of the tricky
and broken ground east of the port was clearly essential. Assuming this part of
the operation succeeded and the requisite signal was given by 0200 hours, a
flotilla of MTBs with around 200 reinforcements would slide into the cove at
Mersa Umm Es Sciausc, previously secured as a beach head. The full complement
would then advance westwards to silence the southern battery and destroy the
various dumps and facilities that lay in their path.
To the west at Mersa Mreira, a strong party of marines would
come ashore and sweep along the northern flank, dealing with the guns and
facilities there, gathering in a shoal of lighters. Once the flanks were secure
the MTBs would pull out of the cove and accelerate into the harbour itself,
where they’d torpedo any targets of opportunity, hiding themselves amongst the
debris. With all of the enemy guns in British hands, the MTBs would cut out or
sink lighters. They’d be joined by the two Tribal Class destroyers, which would
take precautions to disguise themselves as a ruse against air attack while the
shore party manned captured AA defences. The whole force would ship out at
dusk.
This was the plan for Tobruk, bastard child of Operation
Waylay. It marked a very significant leap from the modest spoiling raids
proposed by Stirling and Haselden. It was bold, certainly, but reliant upon a
whole series of disparate groups being able to coalesce on time and in the
dark. It counted upon a weak enemy garrison, stunned by the ferocity of the air
raid and yet not on alert. The seaborne elements were dependent on the right
weather conditions. Good communications between the interlocking units was
essential, but British radios didn’t always work.
It was very complex. At Benghazi, rather surprisingly, it
was decided that Stirling’s ‘L’ Detachment could manage the job without
amphibious support. He would have a small naval party along, but ensuring the
raid was more hit and run obviated the need to take and hold coastal guns. It
scaled down the complexities a very considerable extent and minimized the
potential loss of men and ships. As Peter Smith points out, this reasoning
could as easily have been applied to the attack on Tobruk. That process did not
occur.
#
The final orders for Operation Agreement were set out in the
commander-in-chief’s Combined Operation Instruction No. 1, dated 21 August and
issued to Captain Micklethwait, lieutenant colonels E. M. H. ‘Mit’ Unwin (Force
A) and Haselden (Force B). The overall tactical aims were to destroy petrol and
oil installations, to sink enemy shipping, degrade harbour and dock
installations and to bring away port lighters. Those that could not be ‘cut
out’ were to be shot up and sunk.
Exact orders of battle (ORBAT) are as shown in the appendix,
but Unwin’s seaborne invaders would comprise his own unit, 11th Battalion RM
(Royal Marines), with attached AA and Coastal Defence gunners, sections of
engineers, signals and medics. The naval elements would be led by Captain
Micklethwait, commanding two Tribal Class destroyers, HMS Sikh and Zulu, which
would deliver the marines and their cumbersome, improvised landing craft.
Haselden’s Force B was to come out of the desert, the most
daring part of the plan. He would have a squadron of Major Campbell’s 1st
Special Service Regiment, Y Patrol of the LRDG, led by Captain Lloyd Owen, a
squad of Buck’s SIGs commanded by Buck himself, plus Lieutenant Russell with
further detachments of AA and CD artillerymen (Lieutenant Poynton), engineers
signals (Captain Trollope) and RAMC (Royal Army Medical Corps) (Captain
Gibson). Force C, which was to land in support of Haselden east of Tobruk,
would be formed by a company of the Argylls under Captain Macfie, a machine-gun
platoon of the Northumberland Fusiliers, two sub-sections of engineers, AA
gunners and medics. The force, coming in from Alexandria, would be conveyed in
15 to 20 MTBs.
Additional naval support would be provided by Forces D and
E, which comprised the light AA cruiser HMS Coventry, several Hunt Class
destroyers from No. 5 Destroyer Flotilla and a single submarine HM Taku,
responsible for delivering the pathfinder or ‘Folbot’ party. The entire show
would be preceded by a massive RAF raid, bombing the northern shore of the
harbour from 2130 hours on Saturday 13 September until 0340 hours the next
morning. This opening deluge would be Haselden’s cue to begin his attack inside
the perimeter, targeting the AA and CD batteries at Mersa Umm Es Sciausc, a
cove which lies toward the south-eastern extremity of the main harbour. It was
here that Force C would disembark once Haselden signalled the beach head was
secure.
Force C had to be in position by 0200 hours at the latest.
Besides bringing in the assault troops, they were to beat up enemy shipping
beyond and inside the harbour. An hour later the destroyers transporting Force
A were to land their marines 1½ miles north of the town at Mersa Mreira. The
invaders would then deal with the gun emplacements guarding that flank, fight
their way into the port and generally enjoy themselves blowing things up. The
warships too would enter the harbour to add the weight of their guns.
Of the three main elements, the role of Force B was most
critical. If Haselden’s commandos failed to secure the batteries and beach
head, then the whole plan would have to be aborted. Force B would have to send
the signal for success before 0200 hours on the 14th. If this wasn’t picked up,
both Forces A and B would withdraw. The original RAF element was to involve
additional air raids on 12, 13 and 14 September on selected targets not only
along the North African coast but also on Crete. Low flying aircraft would come
in close on the night of the 13th to distract and confuse enemy radar and
lookouts.
It was whilst the whole of Force B was concentrated at Kufra
on 1 September that Haselden issued his operational orders. The force would
march out from the oasis on 6 September piled into in eight 3-tonners (in
addition to Y Patrol’s vehicles) and motor across the desert to a forming up
area in the vicinity of Sidi Rezegh of evil memory by 1200 hours on D1. Moving
out at dusk on the 12th and less the LRDG contingent the commandos would sneak
into the Tobruk perimeter via the eastern approach. Here they’d come through
camouflaged as depressed and scruffy Allied POWs, guarded by DAK abteilung who
would in fact be Buck and his SIGs posing as Germans.
Assuming the ruse worked and they passed through without
being rumbled, they’d approach the cove at Mersa Umm Es Sciausc through the
maze of wadis cutting through towards the shore. A track was known to run down
past the aerodrome at El Gubi. Lieutenant T. B. Langton from the Irish Guards
and ‘borrowed’ from the SAS would be the pathfinder. Force C had to be safely
ashore by 0230 hours. Once the bombs began to fall (and it was hoped the roar
of engines and explosions would drown the MTB approach), Force B would split
into two assault groups.
One contingent with the artillerymen and engineers would
descend upon enemy gun positions on both flanks of the cove. They’d deal with
any opposition and turn the captured ordnance on any ships trying to get clear
of the harbour. Three sections were to take the eastern positions and the
remainder those to the west. Taking these guns was deemed vital. If they could
not be silenced, then the whole operation should be called off. Both German and
Italian passwords were known and the attackers would use the name ‘George
Robey’.
David Lloyd Owen and his piratical crew would not be left
idle. Their task was, two hours after the commandos had got through the wire,
to follow on and attack a radar station. The place was to be thoroughly
destroyed before midnight, and at dawn LRDG would fall upon the Axis landing
fields at El Gubi and wreak their customary havoc. Having had their fun, they’d
then set up a block astride the Bardia road to deal with any enemy
reinforcements coming up from that direction.
