II

Appendix Y

The success of a seaborne assault depends on the ability of the attacker to sustain a more rapid rate of reinforcement by sea than the defender is able to do by land. The Germans, well aware of this, had organised their coastal defences on the assumption that the attackers would be dependent on the early capture of deep-water ports in order to maintain their ascendancy in the bridgehead. An examination of this problem by the staff of GHQ Home Forces, made during the previous year, had revealed the notable fact that no group of French ports could, on any reasonable calculation of the time that it would take to put them into working order, provide the capacity to support a single operation on the required scale. This meant that maintenance over open beaches, perhaps for months on end, would have to supplement the flow of men and supplies through the ports. If beach maintenance could be developed on a scale that would make us independent of existing ports for an extended period, we should have gone far towards circumventing the enemy’s entire system of coastal defence, whose first aim was to deny us the use of these harbours. The beaches at Caen provided facilities for such development and at the same time satisfied the other basic requirements of a seaborne assault. It was this that determined the choice of Normandy as the objective for OVERLORD and which gave rise to the demand for the MULBERRY harbours.

The decision to invade west of the Seine placed in the hands of the deceivers a gift of priceless value: a cover objective which possessed all the more obvious attractions for an assault landing. So long as one overlooked the vital and determining factor of port capacities, the Pas de Calais had most things in its favour: a short sea passage, excellent air cover and possibilities of strategic development.

As originally planned1 OVERLORD was to have employed some thirty operational divisions in the United Kingdom, with a subsequent build-up of from three to five divisions a month from the United States. The assault was to have been made by three divisions, two follow-up divisions being landed on the second tide. The first phase of the operation was to have been completed with the capture of Cherbourg fourteen days after the initial landing, troops on the right flank driving west to cut off the Cherbourg Peninsula and then advancing northwards to the port itself. In the second phase the Allied forces were to have advanced south-west to cut off the Brest Peninsula with the object of capturing the Brittany ports and turning them into American bases.2 Subsequently, as the battle moved eastwards towards Germany, it was anticipated that the British would take over newly captured ports, relinquishing the more westerly ones to the Americans, thus securing for the latter the shortest available sea route for direct shipment from the United States. This arrangement would bring the Americans on the right flank and the British on the left, which in turn would demand that the American base installations and ports of embarkation should be in the West of England and those of the British in the East.

Of the three assault divisions due to take part in OVERLORD, the 3rd British Infantry Division was to train throughout the winter with its naval assault force in the Moray Firth. The 3rd Canadian Infantry Division was to be similarly employed in the Solent and surrounding waters, and the American infantry division, not yet specified, in the Plymouth area. Of the two follow-up divisions, the American formation would be located in the Bristol Channel during the spring of 1944, and the British in the Thames Estuary and Humber. On about 1st April, 1944, a month that is to say before the projected date of the invasion, the 3rd British Infantry Division was to concentrate in the South of England, its assault force sailing down the east coast, through the Straits of Dover, and coming to anchor at Portsmouth. This journey would take about a fortnight. At the same time, the two follow-up forces would move south. Fourteen days before the invasion the position would be as follows: three assault forces would be located respectively at Portsmouth, Poole and Plymouth, while the follow-up forces would be found on the south coast of Cornwall and in the Thames Estuary.

The high regard in which we held the German Intelligence Service led naturally to the conclusion that the invasion preparations which were already being set in motion could not be hidden from them for long. ‘After January 1944 it would be impossible to conceal from the enemy that a large-scale operation was pending. All that the cover plan could do in the cross-Channel area was to lead the enemy to believe that the assault was elsewhere than was in fact planned.’3 It would be the task of deception to move the apparent ‘centre of gravity’ of the Allied invasion preparations from the South-West to the South-East.

Although planning proceeded on the assumption that all our activities would be more or less accurately observed, it was agreed that the most significant pointer to our intentions would be provided by the landing craft associated with the assault divisions.4 Whatever the Germans might think before the assault force sailed from the Moray Firth, once it had been observed to pass through the Straits of Dover – and any hope of concealing that part of its journey was thought to be out of the question – it seemed inevitable that they should draw the conclusion that we intended to attack somewhere to the west of the Straits where all our assault craft would then be concentrated. It was against this background that the formulation of a cover plan was undertaken.

Though an introductory passage emphasises, in general terms, the need for employing every artifice to draw the enemy’s attention to the Pas de Calais as a major condition affecting the success of the operation, it is curious to note that the only specific demand for deceptive action made in the OVERLORD outline plan is for two ‘major diversions’, one against the Pas de Calais and the other against the Mediterranean coast of France, both to begin about D minus fourteen.5 This perhaps surprising limitation of deception to a relatively minor role may be partly explained by the fact that when OVERLORD was being written in the summer of 1943, minds were much preoccupied by the growing strength of the German Fighter Air Force. Not only was this diversion against the Pas de Calais ‘to contain the maximum forces both ground and air in that area, but it was also to ‘form part of the general air plan for the reduction of the German Fighter Force’.6 When, two months later, STARKEY failed to tempt a single German aircraft off the ground, the hopes of achieving decisive results in this way were abandoned and the disadvantages of pursuing such a course were recognised. ‘The Germans would very soon realise that the landing of such a force without a follow-up was only a diversion; the threat to the Pas de Calais could not then be maintained and the German air and ground forces would be available to reinforce the main assault areas. As this is clearly to our disadvantage, all idea of making a diversionary landing with the object of bringing on air battles has been omitted.’7 The objects of this cover plan thereafter became and subsequently remained:

‘(a) To induce the German Command to believe that the main assault and follow-up will be in or east of the Pas de Calais area, thereby encouraging the enemy to maintain or increase the strength of his air and ground forces and his fortifications there at the expense of other areas, particularly of the Caen area.

‘(b) To keep the enemy in doubt as to the date and time of the actual assault.

‘(c) During and after the main assault to contain the largest possible German land and air forces in or east of the Pas de Calais for at least fourteen days.’8

The plan was divided into three phases: the preliminary phase, from the time that the plan was issued until the sailing of the naval assault force from the Moray Firth; the preparatory phase, from the departure of that force for the south coast until the touch-down of the assault; and the post-assault phase, from the invasion date until our forces were firmly established in France. In order to convey the desired impression in the vital matter of landing craft, the proposal was to make the Germans believe that we had enough assault craft for one more assault and one more follow-up division than we really possessed; four assault divisions instead of three; three follow-up divisions instead of two.9 This was to be achieved by the creation of specially trained assault wireless deception units and by the construction of dummy landing craft. The imaginary assault force was to train somewhere on the Scottish coast like the force operating with the 3rd Infantry Division. The imaginary follow-up division was to be located in the neighbourhood of Rosyth. On 1st April, when ‘S’ Force, the naval assault force training with the 3rd Infantry Division, moved south, the two imaginary forces were also to be shown coming down the east coast, but instead of accompanying ‘S’ Force through the Straits of Dover, they were to put in to the Thames Estuary, dummy craft being floated out so as to denote their arrival to German reconnaissance aircraft. Thus, a fortnight before the invasion took place, it was hoped that the enemy would have the following picture: four assault divisions, one in the Thames Estuary, two in the Isle of Wight area and one in the Plymouth area; and three follow-up divisions, two in the Thames Estuary and one on the south Cornish coast. This result is obtained by adding the real to the imaginary. Portsmouth and Southampton, so it was argued, were as conveniently placed for an attack on the Pas de Calais as against the Seine bay. The presence of one assault and two follow-up divisions east of the Straits would give a bias in favour of the former.

Hardly less significant than the assembly of landing craft, though perhaps less easy for the enemy to observe, would be the location of short-range fighter aircraft. ‘The progressive build-up of short-range aircraft in the South-West and South of England after the New Year will be an increasing indication of impending operations threatening the Cherbourg-Caen-Havre area.’10 This danger was to be overcome by the display of dummy aircraft in the South-East and by wireless deception.11

Finally, there was the general disposition of forces in Southern England to reckon with. It will be necessary at the same time to ensure that the apparent size of the military and air forces, and preparations in East and South-East England are enlarged by discreet display and other deceptive methods on a scale commensurate with the display of craft in that (i.e. the Southern) area.’12 Without waiting for the completion and acceptance of the plan, COSSAC had already in September 1943 requested GHQ Home Forces, to issue a new camouflage policy to Commands which, it was hoped, would tend to move the balance of our visible preparations from West to East. For Southern and Western Commands a policy of total concealment was inaugurated, while Eastern and South-Eastern Commands were to pursue what was described as a policy of ‘discreet display. This studied neglect was to be achieved, not by relaxation of camouflage discipline, but by the omission of technical camouflage advice.

As we were then assuming the presence of German spies and counting on a sustained programme of air reconnaissance, it was accepted that the army could only make its contribution to this great deceptive display by employing real troops, in a false role, and by actually carrying out in the Eastern and South-Eastern counties all those visible preparations which are normally associated with the mounting of a seaborne assault. During the preliminary phase ‘the apparent capacity of the sector exclusive Thames Estuary-inclusive Yarmouth should be increased to one and a half divisions per day’,13 and in the preparatory phase, that is to say, when the real forces were moving to concentration, ‘large-scale movements will be carried out with formations not immediately required overseas into the East and South-East sectors, to full capacity of these sectors, to simulate the movement of these forces overseas. These movements will be synchronised with the concentration and movement of the expeditionary force in other sectors.’14

Accepting the view that all these activities, real and false, were under constant observation, it would indeed have been difficult to deceive the enemy about the date of the invasion. It is, therefore, not surprising to find that the plan devotes little space to this aspect of the problem. Admittedly, as a consideration affecting timing, it says that, ‘in order to achieve surprise, cover activities should normally lead the enemy to believe that the main operation is not due to be launched until about D plus twenty’, but it then enumerates several good reasons why this would be difficult to achieve. For example, ‘the arrival of Naval forces in the south coast from Rosyth about D minus fourteen will make the imminence of the operation difficult to conceal’, and ‘it will be difficult, if not impossible, to conceal the final loading of the assault forces, as this must be done by day as well as by night’.15 Generally speaking, it seemed best to aim simply at the fulfilment of the second object of the plan, ‘to keep the enemy in doubt as to the date and time of the actual assault’.

Let us now turn to the post-assault phase. That this was to be regarded as a very subsidiary affair may be judged both by its anticipated duration and by the size of the force engaged. ‘For about fourteen days after D Day, a force of not less than one assault, one follow-up and four build-up divisions will be represented in Eastern and South-Eastern Commands.’16 The passage quoted goes on to say that the force would be made up of formations not required overseas during the first fourteen days, including two United States divisions carrying out normal full training in Eastern Command, and formations not forming part of the expeditionary force. Failing real formations, the force was to be made up by deceptive methods, including wireless and the use of dummy equipment. Thus, real divisions were to be used if available, sham ones faute de mieux.

The first draft of the plan appeared on 4th September. After undergoing seven revisions, it was submitted in its final form to the British Chiefs of Staff on 20th November, 1943. The latter preferred not to give their approval until the main deception plan for the winter of 1943-4 had been agreed. A request was therefore made for permission to issue it on a provisional basis only, so that it might be studied by those concerned with the preparation of the detailed plans for OVERLORD, with which it would have to be closely co-ordinated. The Chiefs of Staff were also asked to approve certain items of the plan, preparations for which, if they were to be ready in time, would have to be put in hand forthwith. These were three in number: the introduction of wireless silence periods,17 the adoption of specific measures to lead German aircraft and ground observers into the belief that the Eastern counties had a greater capacity for the accommodation of invasion troops than was in fact the case, and the adoption of the general camouflage policy of concealment in the West and display in the East. The last-named was in fact a request for confirmation of a policy which had already been adopted. Approval to these requests was given on 30th November, with the proviso that the visible preparations in the East of England should be delayed for as long as possible.