by James Jinks
the size of the programme, physical and financial – the inter-connection of the ship and system, design and production, procurement and installation – the elements of bricks and mortar; assembly and stowage; selection and training, have all to be combined by January, 1968, to produce the first of a class of submarines which achieve consistently standards of reliability higher than we have ever demanded before.134
Past experience of procuring smaller and less advanced weapons systems had been ‘far from simple and subject to delays and difficulties over which we inevitably have no control’.135 The political directive to get the Polaris fleet to sea as quickly as possible meant that the programme could not afford to be delayed in any way by ‘teething problems’ common in other first-of-class ships.136
Although Macmillan had agreed to assign the Polaris fleet to NATO as a contribution to the Western Strategic Deterrent, the ‘supreme national interests’ clause in the Nassau Agreement meant that it could also be used independently. The level of deterrent strength for national purposes was therefore an important factor in determining the size of the Royal Navy’s Polaris fleet. In a paper for the Cabinet’s Defence Committee, Thorneycroft defined the requirement as ‘the continuous ability at any moment to achieve 50 per cent destruction of between 15 and 20 Russian cities including Moscow and Leningrad’, which could be achieved by having between 24 and 32 Polaris missiles on station at any one time.137
Before Nassau, the Admiralty envisaged using seven smaller, dual-purpose submarines, each armed with eight Polaris missiles, to meet this criterion.138 However, Carrington advised Thorneycroft that ‘the position may have to be reassessed’ after learning of McNamara’s concerns about the technical delays and vulnerability of the dual-purpose concept.139 A Polaris fleet consisting of four non-dual-purpose submarines, each carrying sixteen missiles, was now the favourite.140 One Polaris submarine, carrying sixteen missiles, constituted ‘a 7 or 8 city deterrent’, while two submarines carrying a combined total of thirty-two missiles constituted a twenty-city deterrent.141 A sixteen-missile fleet could be constructed and deployed two years earlier than the hybrid, in 1969 as opposed to 1971, a date that was now seen as ‘unacceptable and would open the door to major expenditure on gap filling’.142 A fleet of sixteen-missile submarines was also more cost effective. Although estimates indicated that each submarine would cost £38m compared to £32m for a hybrid, the combined cost of four sixteen-missile submarines in money and manpower was considerably lower per missile on station, £220m as opposed to £290m for a hybrid fleet.143
The size of the Polaris fleet was also linked to the second major decision that was taken: what to do about the Navy’s nuclear-submarine programme? The Admiralty had briefly considered converting HMS Valiant into a Polaris submarine, just as the Americans had done with the USS George Washington. However, this was discarded because Valiant was practically complete and Vickers would have to slow down work considerably, just at a time when they had built up a ‘good tempo’ on nuclear-submarine construction. It was also considered far better to overcome the problems of building an all-British submarine before ‘superimposing it on the difficult task of a Polaris programme’.144 Valiant was important because it was powered by the unproven British nuclear-propulsion plant that would also propel the Polaris submarines.145 Completing Valiant and the third nuclear submarine, HMS Warspite, would allow the Navy to ‘meet both building problems and early running problems on them rather than on the Polaris boats’.146 The Navy therefore decided to keep both nuclear submarines, ‘not least’, wrote Carrington, ‘because it seemed to me to offer an economical means of increasing the Navy’s nuclear hunter-killer capability at the same time as we built up the deterrent force’.147 Operationally, having ‘three SSNs available by the time the Polaris fleet was complete to sweep out exit routes’ (a process that submariners quickly dubbed ‘delousing’) when the Polaris submarines sailed from their operating base would also be of considerable advantage.148
When the Cabinet’s Oversea Policy and Defence Committee met on 23 January 1963, Thorneycroft argued for a fleet of five sixteen-missile-carrying submarines, whereas the Chancellor, Reggie Maudling, argued for either four sixteen-missile submarines or a fleet of six dual-purpose submarines. Thorneycroft countered that ‘the arguments in favour of the larger class of boat seemed overwhelming’ but in order to appease the Treasury agreed, along with the rest of the committee, to settle for four submarines, the minimum required ‘to maintain an adequate deterrent capability at all times’, though it also said it would revisit the question of a possible fifth submarine at the end of 1963.149 It accepted that construction of HMS Warspite should continue, but to avoid additional expenditure and straining the resources of the shipbuilding industry agreed that ‘no further Killer submarines should be taken in hand during the POLARIS submarine programme’.150 Although the Polaris Executive had only been in existence for a little over a month on 6 February 1963, the first ‘Longcast’ for the whole programme had been produced, showing all major critical dates, realistic targets for the construction, fitting out and trials of the submarines and base support facilities. The date for the operational deployment of the first submarine was given as June 1968, with the rest following at six-monthly intervals.
Having settled on the minimum size of the Polaris fleet, the Navy turned to the Polaris missile. By 1963, the SPO had developed and deployed two versions of Polaris, the A1 with a range of 1200 nautical miles and the A2 with a range of 1500 nautical miles. It was also in the process of developing a third, the Polaris A3, with a range of 2500 nautical miles, capable of carrying more advanced re-entry systems and warheads. The Admiralty wanted to ‘merge into the American development programme and pick up those items which the Americans had developed, i.e. they would not go back and accept the A1 missile since they would then be two years astern of the US; nor would they insist on A3 if this was insufficiently developed’.151 The British were therefore initially attracted to the A2, which was available ‘off the shelf’. However, the SPO had already announced that it intended to phase out the A2 once it had completed the A3 development programme: if the Navy selected the A2, the SPO would have to keep its production lines open solely to meet British requirements. This was technically difficult and potentially very expensive. More importantly, the Royal Navy would have found itself in a position where it was the sole user of the A2 System and would have lost commonality with the US Navy.152
By January 1963, five out of six A3 test flights had failed, with the sixth described as a ‘partial’ success.153 Although Galantin told Zuckerman that it ‘would undoubtedly be successful and would be coming into service with the USN in August 1964’, relying on yet another undeveloped American missile so soon after the cancellation of Skybolt was difficult, and politically very dangerous.154 ‘If we go for the A3 missile and they do a Skybolt,’ warned one official, ‘we shall look foolish if we have paid over large amounts of dollars with nothing to show for them.’155 Carrington was ‘quite sure that the two cases are not comparable’.156 But he realized ‘that the great argument against it, especially after the Skybolt failure, is the element of risk in going for another weapon still under development’. But in reality the differences were really considerable. Skybolt had never been tested as a whole, nor its principles proven, whereas the American incentive to complete the A3 was much greater than Skybolt. Investment in the US Polaris programme was ‘immense’.157
There was one major problem with the Polaris A3. Should the British decide to purchase the A1 or A2, the question of development charges would not arise, since they had been incurred by the United States before the signing of the Nassau Agreement. If the British government purchased the A3 and any future missiles, the US wanted the British to make a contribution towards research and development costs. This was a reasonable request. McNamara told Ormsby-Gore that as the British were benefiting from ‘the 2.4 billion already spent on development and quite probably billions more in the future’ they we
re getting a ‘fantastic bargain’.158 But politically it was very embarrassing for the Prime Minister, who, in playing up the benefits of the Nassau Agreement, had told reporters that ‘in Polaris Britain was getting the benefit of £800m worth of expenditure to which the British taxpayer had contributed nothing’.159 Now the Americans were asking for an additional payment estimated at anything from £11m to £33m up until 1968.160 ‘They will tear me limb from limb in the House if this ever comes out’ was Thorneycroft’s initial reaction.161 Macmillan described it as ‘a great shock’.162 Fearing a ‘very important and perhaps decisive’ two-day Defence debate in the House of Commons on the Nassau Agreement on 30 and 31 January 1963, Macmillan was concerned that if he had to tell the House that the supply of Polaris was provisional, dependent ‘on a satisfactory settlement of the financial arrangement’, it ‘may result in the defeat of the Government with all the consequences which would follow’.163
Hard bargaining followed.164 Macmillan ‘refused to make an open-ended contribution to an unknown bill for Research and Development’.165 He felt that he had already paid for Polaris by accepting American demands over ‘assignment, allocation [to NATO] – call it what you will … a very difficult thing for me to defend here with the press and the Party’.166 To have to pay ‘a quite different form of commercial price to that which was included in the original Skybolt agreement – which has only broken down because of the failure of the American Government to go on with it’ was, in his view, unacceptable.167 He instructed Ormsby-Gore to go directly to Kennedy and offer just 5 per cent on top of the production cost of each missile.168 The President was ‘determined there should be no Anglo-American row’ over the issue and he instructed McNamara to accept the offer.169 ‘If,’ as a very satisfied Macmillan later recorded in his diary, ‘we bought fifty million pounds of missiles, we would pay fifty two and a half million pounds … Not a bad bargain.’170 The final settlement saw Britain spend just over £90m on missiles and spare equipment between 1963 and 1971/72.171
In fact, the settlement represented a fantastic bargain. The ‘on a continuing basis’ clause in the Nassau Agreement meant that the 5 per cent formula applied to Polaris and ‘any later Marks of missile which we may want to buy’.172 The importance of this clause would become apparent almost twenty years later when Margaret Thatcher’s Government entered negotiations with the United States to purchase a replacement for Polaris known as Trident. The deal was so good that Ormsby-Gore had to warn Macmillan that it ‘was such a poor one from the American point of view that they may suggest that you do not spell out the effect of the agreement quite so precisely’.173 He urged the Prime Minister, when announcing the deal in public, ‘to explain that ours is a unique case’ as there was ‘likely to be strong criticism here of a deal which is so favourable to us’.174 McNamara was furious. He had told Zuckerman ‘that if ever a Polaris submarine was fired he hoped that the opening shot would be the then MOD Public Relations Officer, who had done a grand job in painting the picture of how the US had let us down over Skybolt; and no-one would have known that he, and not a missile, was in the tube!’175
There was only one shipyard in the United Kingdom capable of constructing a Polaris submarine within the very tight timetable – Vickers-Armstrongs at Barrow-in-Furness. As a paper to the Cabinet’s Economic Policy Committee put it, ‘they alone of British shipbuilders have experience of nuclear submarines, they already possess at Barrow much of the special equipment that is required for nuclear submarines [the government had already invested £1.5m in capital equipment for the construction of HMS Dreadnought] and they have very good working relations with the Electric Boat Company of America whose advice and technical assistance will be essential’.176 Vickers provided the Admiralty with assurances that it had the capacity to complete the submarines in the time required, without any detriment to the completion dates and cost of Valiant and Warspite.177 However, Vickers’ assessment was seen as ‘over optimistic’ and the Admiralty concluded that the shipyard had insufficient capacity to build more than two Polaris submarines at the same time. A second shipyard was required. Two shipyards with recent experience of submarine construction, Cammell Laird at Birkenhead and Scotts at Greenock, were considered. Cammell Laird had the strongest case. It was the second-largest shipbuilding firm in the UK and had recent experience constructing conventional submarines, as well as a modernized shipyard and excellent docking facilities.178 However, as they had never built a nuclear submarine before, Baker warned quite presciently that ‘we must not be surprised if Cammell Laird fail to honour their delivery promises’.179 Vickers would construct the first and third Polaris submarines while acting as ‘lead yard’, guiding Cammell Laird with the second and fourth.
An equally urgent question was where the Polaris submarines, their missiles and warheads would be based. From the outset, and largely derived from a recommendation in Palmer’s 1961 report, Devonport was ‘generally thought to be the best choice’.180 A number of documents prepared in early February 1963 all assumed that Devonport would be the operating base.181 But Mackenzie’s temporary replacement as FOSM, Commodore Ernest Turner, was ‘struck by the assumption that Devonport’ was ‘the best choice for a forward operating base’ and objected. He thought Devonport was in ‘danger of being accepted without a proper study of the requirement having been made’.182 A working party was established to examine, evaluate and make comparative costings of various locations across the UK.183 In total, the working party conducted eleven detailed site studies: Portland; Devonport; Falmouth; Milford Haven; Gareloch; Clyde Area (other than Gareloch); Fort William; Kyle of Lochalsh; Loch Ewe; Invergordon; and Rosyth.184 The three most important factors were operational, safety and cost. Safety was overriding. As one Admiralty brief concluded, ‘should the worst happen, Ministers must be able to stand up to searching fundamental questions about the reason for deciding on the acceptability of risks in a particular location in the first place and about safety measures taken on the spot. One question is bound to be “Was there no safer location? If so why was it not used?” ’185
The working party concluded that on safety grounds Devonport was not suitable due to the dockyard’s proximity to the population of Plymouth. On operational grounds the dockyard only had a limited number of berths for unrestricted regular access by nuclear submarines and it was too far away from sound-ranging facilities, which could not be provided locally. There was also only one possible site in the area for a new armament depot and, as the Admiralty did not own it, planning permission for development would be required. It was ‘unrealistic to ignore the probability of strong objections on the part of the present occupiers’ and the public when attempting to acquire the new land.186 ‘In view of its serious safety objections and operational limitations,’ concluded the working party’s final report, ‘we cannot recommend Devonport.’187
The working party concluded that Rosyth and Faslane on Gareloch were the most suitable locations. On safety grounds there was very little between them, but on operational grounds Faslane was judged to have a number of advantages. The Captain of the Third Submarine Squadron, Ian McGeoch, had transferred the Squadron and its Depot ship, HMS Adamant, to the area from Rothesay in 1957.188 The location was close to the Royal Navy’s northern exercise areas, a sound range in Loch Fyne and the Clyde degaussing range, a means of erasing the magnetic field from a submarine to camouflage it against magnetic detection. This was important because of the rigid requirements of the operating cycles. In order to ensure that two submarines were kept on station as much as possible it was important to use the lay-off periods as productively as possible rather than wasting time moving the submarines to exercise areas (Rosyth was over 500 miles from the nearest sheltered sound range).189 Faslane was also ideal because the large amount of shipping and numbers of submarines moving in and out of the Clyde area would help disrupt enemy surveillance of the Polaris fleet. The approaches to Gareloch offered very unfavourable conditions for hostile anti-submarine surveillance,
whereas the approaches at Rosyth were much more open. But Faslane did have some disadvantages. It was estimated that constructing the base would cost £20m–£25m, whereas Rosyth would cost around £2.5m less. The proposed site was also very close to a Ministry of Power emergency oil storage installation which might have to be moved as the whole base would become a target for a Soviet nuclear strike. In addition, Metal Industries Limited had just secured a lease for jetty space after years of protracted negotiations with the Ministry of Transport. The prospect of renegotiating the contract, at considerable time and expense, was not attractive to the Admiralty.
Faslane’s proximity to two suitable sites for a Royal Naval Armament Depot to assemble, test, store, load and unload the Polaris missiles and warheads overcame these disadvantages.190 A site at Glen Douglas was ruled out almost immediately because to load and unload the missiles from the submarines, it would have been necessary to moor within a few yards of the busy Glasgow to Oban road. The Glen Douglas site was also adjacent to a NATO depot and the working party concluded that there would be strong political objections to putting the national strategic nuclear deterrent in such close proximity to a NATO facility. The second site at Coulport was far more suitable. Although configuration of the land meant that development was likely to be expensive, the fact that the Admiralty already owned an existing Hydro Ballistic Research Establishment there was of considerable advantage, as was its deep-water jetty and proximity to Faslane.191 From start to finish it took the working party a fortnight to recommend Faslane.192 The Admiralty Board agreed that ‘the operational advantages and other conveniences of Faslane made it the proper selection, notwithstanding the additional cost’.193
The Admiralty desperately needed more information about Polaris from the Americans. But in order to access it a new agreement was required that would turn what had been agreed at Nassau into a practical working arrangement. In mid-February 1963 a technical and political team was dispatched to Washington to conduct a series of negotiations that would lead to one of the most important and enduring US/UK post-war agreements, the 1963 Polaris Sales Agreement (PSA). Once in Washington the technical team split into various sub-committees, each responsible for studying a particular area such as the re-entry body, warhead, shipyard services, security, patents, warranties, indemnities and financial matters.194 Both sides agreed early on that it was impossible to produce an agreement that included the full specifications of items and equipment required to meet British needs. The aim rather was to produce a ‘permissive and not binding’ agreement that ensured the British had ‘the option to buy, if we wished, any equipment which the Americans produced as part of the Polaris weapons system’.195 They also agreed to set out the detail at a later date in a series of technical arrangements, basic ground rules to implement the programme at the working level. Article II of the agreement authorized each country to enter into ‘such technical arrangements consistent with this Agreement, as may be necessary’. By 1970, once all four British Polaris submarines had put to sea, a total of forty-eight Technical Arrangements were in operation.196
