Canadian Historic Sites: Occasional Papers in Archaeology and History No. 17
by John Joseph Greenough
". . . and keep your powder dry!"
From Ballads of Ireland, Col. Oliver's
The attempts to staunch the casemates absorbed most of the energies of the engineering staff at Halifax during the last decade of the Citadel's construction. The problem involved was fundamentally the result of four different but related factors. The first was the necessity of completing the casemates in such a way as to allow them to perform their allotted functions effectively. This was vastly complicated by the second factor: pressure from the military authorities to use them as barracks. The third factor, in some ways the most frustrating, was the age of the work. A good many of the casemates had been standing empty for years before the construction finally reached a stage where they could be put to use, with the natural result that the process of staunching involved both building and repairing simultaneously. The fourth factor was the inadequacy of the original design. This was less because of incompetence on the part of Colonels Jones and Calder; the casemates were of comparable quality to those built elsewhere. But no one knew precisely what features could be used effectively in a permanent fortification in the damp Halifax climate.
These same four factors underlay the difficulties experienced with other parts of the work carried on at the same time as the staunching. The problem of waterproofing, moreover, ultimately affected almost all the other parts of the fortress. While the casemates remained unfinished, the ramparts, armement and parade ground could not be completed: the magazines could not be used except as storage depots for other works in the Halifax area, and the glacis could not be built. There was simply not enough labour to do all the work at once. This inevitably exacerbated the age factor, since the longer the remainder of the work was postponed, the more decrepit the existing buildings became. In the end the engineers found themselves caught in a kind of nightmarish race to get the fortress finished before its aging fabric went irretrievably rotten.
The last decade of construction was, therefore, characterized by interconnected routine work, with the dominant theme of casemate staunching played out against a counterpoint of increasing urgency. The period can be divided into three phases. In the first, lasting until about 1850, the momentum of building continued, all the while being gradually slowed and interrupted by the growing demands of the waterproofing problem. At this time the final provisions of the revised estimate were carried out and the last attempt was made to introduce new features into the original plan. By the end of this stage, it was obvious that the primary concern was not improving the work but preserving what had already been built. In the second phase, lasting from 1850 to about 1854, the waterproofing brought almost all other work to a complete standstill, while the decay of the older portions of the masonry was accelerated. In the third phase, from 1854 to 1856, all the problems, delays and faulty judgements of the previous quarter-century finally came home to roost, and the project came closer to foundering completely than it had at any point since the early 1830s.
The most characteristic activities of the first phase were the removal of earlier failed work and the abortive attempt to introduce prison casemates; of the second, the attempt to install the armament. The third phase was characterized by an almost frantic attempt to renew, restore on rebuild parts of almost all the major components of the fortress, including the cavalier and magazines. Even the casemates. after almost 10 years of continuous labour on the problem of waterproofing, remained a major source of worry and complaint. In the end, disaster was averted, but it had been (to use a Wellingtonian phrase) "a near run thing."
By the mid-1840s, one of the few remaining routine tasks which did not involve the casemates was correcting earlier mistakes and removing those features of the early design no longer felt to be necessary. The first casualty was the old (1812) magazine, which had been standing empty for 19 years and obviously impeded the completion of the parade square. In the spring of 1847, Calder got permission to remove it. As it was an almost embarrassingly solid piece of work, the Fortifications department did not want to spend the time and effort necessary to demolish it by conventional means. The only alternative was to blow it up, and even this took considerable time. Between 24 March and 6 April, working parties laboured with crowbars, picks and sledge hammers on the business of constructing galleries in the masonry walls for the gun powder charges. In all, 22 chambers were cut into the walls and were packed with charges of between 9 and 16 pounds of gunpowder each.
On 7 April, everything was ready. The officer in change of the demolition described the results.
Colonel Calder pronounced himself pleased with the operation. Indeed, he was so impressed with the speed and efficiency of the demolition that he proposed similar measures for one of the other failures earmarked for removal.
Calder waited almost a year for a reply to this proposal. When it finally became important to get the matter settled so that he could proceed with the rebuilding of the ravelin, he dispatched an informal query to London. "Col. Calder presents his compliments to the Inspector General of Fortifications and begs to acquaint him that the last paragraph of his letter No 193 . . . has not been replied to."3 The fact was that London had lost the original letter: one of the clerks had to annotate the margin of Calder's query, "I cannot put my hands upon the origl letter No 193."4 When it was finally found, the Inspector General responded by asking Calder why he wanted to proceed with the scheme. Calder restated his reasons.5 After another delay, Burgoyne decided to forbid the use of explosives in the demolition on the grounds that it might be possible to re-use some of the stone from the west ravelin in rebuilding.6
This ended the brief vogue for dramatic demolition of old mistakes. In fact, apart from the two cases mentioned above, a surprisingly small amount of the supposedly defective work of the early period was ever altered. Most of the work in question was, of course, in the escarp walls, and some of the basic rebuilding and repairs there had already been done by Nicolls and Boteler in 1831-32. Colonel Jones estimated in 1834-36 that only 574 feet of the remaining old walls would have to be rebuilt.7 This was only a portion of the original escarp and it was demolished by means less dramatic than explosives. In the end the engineers made do with the remaining old walls, partly because the masonry in question, though shoddily built, showed a complete disinclination to collapse. After the demolition of the west ravelin in 1848-50, the whole question of the old work was shunted aside and partly forgotten. It was not until 1855 and under rather different circumstances that it became again an issue.
As the last of the old work was being removed, Colonel Calder made the last attempt to introduce a new feature into the overall design of the Citadel. This was in response to a peculiar and specific sort of accommodation problem. The first soldiers to have the honour of inhabiting the Halifax Citadel had been the military convicts. As early as 1845, a strongroom and guardhouse had been fitted up for prisoners in two of the defence casemates (Nos. 54 and 55).8 This was apparently only a temporary arrangement to serve until cells designed for the purpose could be built. Such cells were included in the 1843 estimate for alterations and renewals and were to be located above the end casemates of the cavalier.9 But even after the cells were built there was still not enough room for the convicts. On 7 August 1847, Calder submitted a proposal for 12 more cells to be placed under the ramparts on the south side of the southeast salient.10 His design called for a complicated arrangement of two-storey arched compartments connected by a corridor at the rear. He estimated the total cost of the scheme at £2,410 19s. 7-1/2d.11
London not only approved the scheme but, in a rare burst of generosity, actually enlarged upon it. Calder shortly received a revised design which included two additional compartments for first-class prisoners and a more complicated system of heating and ventilation. The only objection which the Ordnance raised was to the proposed location of the new work. The south face of the southeast salient was considered inappropriate because of the lack of space available for the enlarged scheme, so it was suggested that the work should be put on the east side of the salient.12
Calder, doubtless amazed at this unexpected development, could only concur. He incorporated all the changes and re-submitted the design on 15 November.13 Even as he was doing so, however, London was having second thoughts about the whole project. The problem of accommodating prisoners was essentially an army matter, and the Ordnance had seen fit to submit the scheme to the Secretary at War for an opinion. The secretary, Mr. Fox Maule, disliked the idea and decided that it would be better policy to build a gaol large enough to hold all the garrison convicts somewhere outside the Citadel.14 The Board of Ordnance accepted the recommendation and instructed Burgoyne to inform Calder.15 In the end, the cells over the cavalier cookhouse remained the only military prison within the fortress.
It was not until 1846 that the Ordnance staff in Halifax addressed themselves to the task of composing an armament proposal for the Citadel. In that year, Lieutenant Colonels Calder and Jackson (the CRA) drew up a scheme which entailed 94 pieces of ordnance, including five 8-inch guns, thirty-one long 32-pounders, eighteen short 32-pounders, twenty 24-pounders, twelve mortars and eight howitzers (see Table 4).16 On 15 September 1846 the Director General of Artillery approved the plan and initiated the process of installation.17 Almost ten years elapsed before the bulk of the armament was installed.
*Adapted from a return in PAC, MG12, WO55, Vol. 880, p. 913.
The first stage of the process involved the manufacture of carriages for the guns, the acquisition of the guns themselves, and the construction of the stone platforms on which the greater part of them would be mounted. The first matter was the responsibility of the Royal Carriage Department; the second, of the Board of Ordnance, and the third, of the Engineer department in Halifax. Since a coordinated interdepartmental effort was involved, delay and complications were inevitable, and it was well over two years before all the orders were filled.
The most serious misunderstanding arose over the order for 24 siege gun platforms after Lieutenant Colonel Alderson's pattern. These were intended for mounting the mortars, howitzers and four of the 32-pounders.18 The Ordnance staff in Halifax included them in the order for traversing platforms and carriages sent in to the Carriage department in the spring of 1847.19 Two years later the Carriage department decided that the platforms might not be their responsibility. The gentleman in charge. Mr. Gordon, wrote to General Burgoyne,
After some discussion, the Board of Ordnance decided that the Carriage department ought to be relieved of the task of making the platforms, and instructed Burgoyne to ask the Commanding Royal Engineer in Halifax why they had not been included in the Engineer Demand of Stores in the first place.21
By the time this finally got back to Halifax, Colonel Savage had replaced Calder as Commanding Royal Engineer. Savage had no idea why his predecessor had requested the platforms from the Carriage department, and could only promise to include them in the Ordnance annual estimate as required.22 By then it was obvious that the armament could not be mounted at all until the problems of waterproofing the casemates were solved, and the whole question of equipment was temporarily sidetracked. Fortunately, the Artillery was in no hurry to mount the guns and, except for the occasional enquiry on technical matters, nothing more was heard about armament for two years.
By the spring of 1851, however, the Director General of Artillery was beginning to get impatient. The CRA was requested to report on "the condition of the fort with respect to its state of preparation for mounting the Ordnance."23 The CRA relayed the request to Colonel Savage,24 who answered that the Citadel would not be in any state to receive armament until the summer of 1853. Even this date proved optimistic. When the question was put to him again in January 1853,25 Savage was able to approve the mounting of only part of the armament for the following summer.
The remainder could not, he thought be mounted until the following year.
The Ordnance did its best to prevent Savage from carrying out his plans for 1854. The mounting of armament on the rest of the work depended on the completion of the staunching project and the construction of the ramparts and terreplein. While the former appeared to be going ahead successfully, London prevented the latter by refusing to allow Savage the sum provided for the service in the annual estimate for 1853-54. Three months after he had given his optimistic prediction to the CRA, Savage wrote to the Inspector General proposing that the funds allotted for completing the glacis be used instead for the terreplein and parade.27 London replied with surprising speed, granting permission to make the substitution.28 Since the work was not included in the annual estimate for the following year,29 it would seem that the ramparts were constructed in the summer of 1853, and, in all likelihood, most of the rest of the armament was mounted the following summer.
Whether it would stay mounted was another matter. By the fall of 1854, serious questions were being raised about the future of the cavalier, and after a brief period of optimism, it was becoming depressingly evident that the casemates were still displaying a pronounced tendency to leak.
The first indication that parts of the Citadel were falling to pieces came on 19 October 1852, when the Ordnance Storekeeper, Mr. Ince, discovered that the door of the north magazine would not open "in consequence of something having fallen against it."30 On examination, Colonel Savage discovered (probably to his horror) that "the floor, which was previously in a decayed state, had suddenly given way, from the weight of the powder and the decay of the joists."31 Savage had already provided for repairing the floor in the annual estimate for the following year, but the sudden collapse took him by surprise, and he could no longer wait for the estimate to be authorized. He therefore requested that the Respective Officers formally propose a special estimate. The Respective Officers replied three days later:
The next day, Savage formally requested permission to make immediate repairs, stating that the expense could be defrayed from the savings on various items of the annual estimate for the preceding year.33 London was quick to authorize the expenditure, and the repairs were carried out in the course of the winter.34
In spite of his experience with the north magazine, Savage was somewhat startled when, a few months later, he examined the floor of the south magazine while alterations to the powder bays were being made:
This discovery made it necessary to formulate yet another special estimate, but this time Savage decided to use a new method of repairing the floor. Acting on a suggestion from the Surveyor of the Ordnance, he proposed to use
He enclosed a special estimate and a demand for stores amounting to £158 5s. 0d.
Despite the fact that Savage's suggestion was made at the height of the asphalt mania, London decided that it would not be appropriate to use the material on the magazine floor. General Burgoyne recommended that the floor be repaired in the same way as the one in the north magazine (apparently with a new wooden floor) and the board approved his recommendation.37
The two magazine floors were repaired and the buildings restored to normal use by the summer of 1853. There followed a brief respite. It was to be a year and a half before the next serious problem arose.
By the fall of 1853, Colonel Savage thought that the end of the Citadel construction was in sight. The Ordnance annual estimate for the following year reflected this belief. There were only two items in it for the Citadel.38 One, amounting to £2,681 12s. 3d., was for the completion of the glacis and parade square, and this was believed to be the last major expenditure on the work. The Assistant Inspector General wrote, in forwarding the estimate to the board, "With the sum here proposed the Coming Rl Engineer expects to complete the Citadel in 1854-5."39
The second Citadel item for £1,256 2s. 11d. was for the renewal of the cavalier colonnade and was considered absolutely necessary for the occupation of the building by troops. This was an ominously large sum to be spent on repairs, but it could easily be explained. After all, the cavalier was almost 25 years old and repairs were a matter of routine in a building that age. At this point, no one seriously considered more drastic measures to be necessary.
This mood of optimism lasted for some time. In February, Lieutenant Parsons drew up his memorandum on the effectiveness of asphalt in the Citadel; while he admitted that it had not worked in the case of the cavalier, he did not speculate on the reasons.40 In forwarding Parsons' report to London, Savage noted that
Apart from this observation, which Savage appended almost as an afterthought to a long report, the whole question of the cavalier's suitability received little attention either in Halifax or in London.
When Lieutenant Colonel Richard Stotherd inherited Savage's command in June 1854, it seemed that he would have the good luck to be the first Commanding Royal Engineer in more than a quarter-century to avoid trouble with the Citadel. His first summer, in fact, passed quietly enough. The only matter concerning the Citadel which needed particular attention involved a special estimate (amounting to £22 12s. 10d.) which provided for altering the position of the stoves in the cavalier to keep the casemates warm in winter.42 This was approved by London in just over a month.43 Stotherd's first annual estimate, dispatched on 25 September, asked for only £1,902 for the Citadel, most of it for completing the glacis. Only £100 was for staunching the casemates and there were no items at all for repairing the cavalier.44
But during the winter of 1854-55, two events occurred which shattered the satisfaction of the Ordnance staff in Halifax, at least in regard to the Citadel, and Colonel Stotherd found himself faced with the worst crisis in the fortress's history since Colonel Nicolls's walls collapsed in 1830.
The first event was a systematic examination of the casemates in November 1854. This revealed that, despite all the measures undertaken in the preceding eight years, 21 of the casemates were to some degree damp. The extent of the problem varied from casemate to casemate. Some were only slightly wet: others were uninhabitable. The rampart casemates, however, were in relatively good condition compared to those in the cavalier. Except for the small end casemates and the rooms over them, the entire building was completely uninhabitable.
Stotherd reported all this to London in a rather gloomy letter. He was particularly dissatisfied with the cavalier.
As for the ramparts casemates,
He estimated that complete repairs would cost around £6,000, most of which would be needed to repair the cavalier, where he proposed to rebuild the entire top of the building from the springing of the arches up. He was less explicit about dealing with the leakage in the rampart casemates, but apparently he contemplated a continuation of the existing system of staunching.
Two weeks later, Stotherd dispatched a second letter requesting an immediate delivery of asphalt so that work on the casemates and cavalier could begin as soon as practicable in the spring.48 The response was surprising. After nearly 10 years of experimenting with asphalt in the Citadel, the Fortifications department was beginning to wonder whether it was, in fact, entirely suitable for waterproofing in the Halifax climate. The Assistant Inspector General, Colonel George Judd Handing, wrote back, enquiring whether "flat tiles laid in cement" would not be more suitable.49 One wonders whether Harding was aware that his suggestion had been tried before, with indifferent results, by Colonel Jones more than 10 years earlier.
Before Stotherd even got Harding's suggestion, the second, disastrous event occurred. On 8 February 1855 Halifax experienced one of its very rare earthquakes, and among the most vulnerable buildings in the entire city was the aged, decrepit and top-heavy cavalier. The report on the damage, submitted by the Clerk of the Works and two of the junior engineer officers, Captains Philip Barry and Henry Grain, was possibly the most pessimistic summary even produced in the entire course of the Citadel's construction.
They concluded by recommending that no attempt be made to staunch the arches while the walls were "in a condition apparently so irremidiable [sic]."
A second report, appended by Captains Barry and Grain, was if possible even more outspoken than the first:
The two officers then went on to suggest that the cavalier be demolished to make way for "a tower . . . to mount three or four heavy guns" which would both fulfill all the military functions of the cavalier and allow more space in the fort's interior.
This second report was not only outspoken, it was downright dangerous. In a mere half-page, two junior officers had managed to question the wisdom of the original designers of the Citadel, revive an idea which had been forgotten for nearly 30 years, and, worst of all, raise the whole question of the old remaining contract masonry which had been often condemned but never replaced. One can imagine Stotherd's reaction when he read it. It was beginning to look as if the major work in his command was about to disintegrate.
In forwarding the reports on the earthquake damage to London, Stotherd adopted a cautious, almost contradictory stand on the suggestions contained in them. He began by confessing that, since it was his first winter in Nova Scotia, he was far from being an expert on the effects of the local climate. He then went on to state that, in its present condition, he could not recommend the staunching of the upper parts of the cavalier. But he was uncertain about the best course to adopt.
On the other hand, he noted that the cavalier had once been
Such a roof would, he estimated, cost around £600.
The Ordnance was not disposed to accept any radical suggestions. In fact, the whole apparatus of the Ordnance department was under tremendous strain because of the Crimean Wan, and the department was to undergo a major revolution in the near future. The officials in London, uncertain about their own futures, were not about to make major decisions. Their only response to Stotherd's letter and the gloomy reports it enclosed was a brief note asking whether it was necessary to restore or replace the building at all. No mention was made of the possibility of tearing the cavalier down, and Stotherd was requested to report on the "extent of the repairs required" so provision could be made for them in the annual estimate for the following year.53
This was virtually the last instance of the Board of Ordnance handing down a decision on matters relating to the Citadel. Appropriately enough the board ended its superintendance of the work on a note of administrative equivocation. Stotherd was enjoined to await events. He did not have to wait long; events were quick to catch up with him. He was soon facing both a political challenge from forces which had never before had any effective control over Ordnance works, and the pressures of providing necessary services within the Citadel. The first of these, which was to be the most difficult to manage, will be discussed later. The second was to shape the concluding stages of the construction of the work.
London was wrong in assuming that the cavalier was of little importance to the Halifax garrison. It was true that the station was well below strength in the winter of 1854-55 because most of the British army was in the Crimea. Even the small remaining garrison, however, needed more barrack space. On 21 June Stotherd submitted an estimate amounting to £944 0s. 7d. for the restoration of the cavalier.54
The scheme put forward in the estimate was essentially an elaboration of the roofing proposal which Stotherd had made at the end of his February letter. Besides installing a timber roof, it proposed to alter and enlarge the chimneys, to point the defective masonry joints and to whitewash the rooms. This implied the abandonment of the cavalier as a defensive work. Although the guns were left in place, the enlargement of the chimneys and the installation of the roof would make it difficult to get the gun positions cleaned for action in time of war and impossible to fire them in peacetime.55
Authority to proceed with the scheme was quickly forthcoming.56 By August Stotherd was able to report that he expected to be finished with the work within two months.57 By this time, Stotherd had found solutions to most of the remaining problems of the Citadel. He no longer thought in terms of major alterations, but only of minor repairs which, he hoped, would be sufficient to silence criticism of the work and to keep it in a tolerably good state of repair. It is difficult to escape the conclusion that most of the items proposed were at least partly cosmetic in nature, but they did at least manage to keep everyone satisfied. In this rather undignified way, the Citadel project limped into its ultimate stage.
The nature of Stotherd's work is demonstrated by the type of item he inserted in the annual estimate for 1856-57. Of the £2,900 estimated for the Citadel, over two-thirds (£1,795) was for minor repairs of one sort or another, including £959 for repairing the asphalt over the arches, £38 for pointing the arches in the redan, and £529 for pointing masonry in the escarps, counterscarps and magazines.58 This list covers two of the three major sources of complaint (the old escarps and the waterproofing) in the cheapest way possible.
In a report on the defence of the Nova Scotia command, submitted at the same time as the annual estimate, Stotherd defended his policy, especially in regard to the pointing.
The effectiveness of Stotherd's measures was varied. His assessment of the strength of the old walls was borne out by subsequent experience with them (see "The Very Model of a Modern Major General"). The experiment with the roof of the cavalier proved equally successful. A tabular statement of the condition and usage of the casemates drawn up in June 1856 reported that there was only a slight appearance of damp on the west wall and this could be easily corrected by additional pointing of the masonry.60 The same statement revealed, however, that Stotherd had been less successful with the other casemates. A surprising number of them still leaked or showed evidence of damp on one or another of their internal walls. The report treated each case individually; there was no longer any attempt to assign blanket causes for the problem. One was damp because of faulty drainage; another because of decaying masonry; a third because the terreplein had not had time to settle properly and so on, down a whole list of similar minor faults. In other words, the problem had reached the stage where it could be treated as a minor housekeeping difficulty, and no further large sums of money were needed to correct it.
As for the other features of the fort, most required only minor alterations. Most of the armament had been installed.61 After a bad start, marred by the complete undrinkability of the water, the water tanks were in the course of being repaired.62 It was not a particularly heroic ending but, with the exception of the glacis, the Citadel was virtually finished.