Canadian Historic Sites: Occasional Papers in Archaeology and History No. 17

The Halifax Citadel, 1825-60: A Narrative and Structural History

by John Joseph Greenough

Colonel Calder Revises

I

The Citadel entered the second and final phase of its construction between 1840 and 1842. In these years the exterior of the fort, as definitively established by the revised estimate of 1836, was finally completed. There could be no fundamental alterations. In the second phase, the fleshing out of the granite and ironstone skeleton into a functional work of defence, a whole new set of problems arose. The difficulties encountered in the 1840s were in matters of detail — accommodation, waterproofing, interior partitions and so on. They required specific and detailed solutions which, of course, were quite beyond the general considerations provided for in the revised estimate and its supporting documents. Indeed, some of the problems were simply the result of the initial stages of building having taken so long. Many of the difficulties encountered with the cavalier, for example, arose from the fact that it was already more than 15 years old when the time came to make it fit for lodging troops, and it suffered from the maladies typical of any stone building left unoccupied for so long.

It was during this second phase that continuity in the building staff became important for efficiency. Colonel Jones had already been in Halifax for more than seven years and had, in effect, become the projector of the work. In the process, he had acquired enough experience with the Citadel to decide on matters of detail. He was also sufficiently well-established with the London authorities to be allowed a certain amount of latitude in his decisions. Any successor would have neither of these advantages.

It was probably for this reason that the Inspector General requested Jones to stay in Halifax until the work was completed. Then, less than a year later, London reversed itself. It is not known why. Possibly Jones himself requested it; he had been in Halifax eight years, longer than any other Commanding Royal Engineer. In any event, Jones was notified on 19 November 1841 that he was to be relieved.¹ His successor arrived on 8 March 1842.²

14 "Sketch of the North East and North Fronts of the Citadel," 1843. This plan shows the additional casemating proposed by Colonel Calder in his 1843 estimate. (Public Archives of Canada.)

15 "Plan and Elevation shewing the situation of Proposed lightning Conductors," 1846. The lightning conductors were installed (briefy) in this fashion and shortly thereafter failed. They were ultimately replaced by a different system. This is the earliest elevation of the magazine as built. (Public Archives of Canada.)

16 The north magazine and area wall, ca. 1880. The south magazine was identical.

II

Lieutenant Colonel Patrick Calder arrived just as the final season of work on the exterior walls was about to begin. The northern, western and southern fronts were virtually complete (except for a few problematical walls dating from Colonel Nicoll's time and a defective west ravelin), and the escarp and counterscarp on the eastern front were both more than half finished. The interior of the fort, however, had changed little since 1832, and indeed since 1828. The old powder magazine still stood, perched by that time on top of an island of earth in the centre of the parade square. The new magazines were not yet begun, nor were the bulk of the casemates; and the cavalier, which looked imposing enough, stood empty and incomplete.³ A newcomer walking into the place must have felt rather like a spectator blundering backstage at a theatre and seeing the sets from behind. Even an experienced engineer like Calder must have felt some discouragement at the amount of work still to be done.

The first summer passed quietly. The work done cost £12,742 — about the average amount spent in a working season.⁴ The only ominous event was the collapse of the area wall enclosing the stairs leading to the casemates of defence in the northwest bastion. It was the first such collapse since the early thirties and it immediately raised the question of the soundness of the other early walls. Calder's first progress report, dispatched on 30 June, contained an account of the collapse as well as of the other work in progress.

London's reply set the tone for Calder's relationship with his superiors for the next two or three years. The chief draughtsman of the Fortifications office, on examining the progress report, found it did not agree with his interpretation of the original 1836 estimate. His two complaints arose from an examination of the drawing accompanying the report. In one, the redan basement was shown without the area wall opposite; in the other, the main drain differed from that shown in a drawing in a previous report. Calder was instructed to account both for these discrepancies and for the failure of the area wall.⁵

The collapse of the area wall was easily explained; loading pressure and the poor quality of the mortar and masonry used in its construction were to blame.⁶ The chief draughtsman's complaints were another matter. Both were essentially trivial and were easy enough to correct; in the case of the redan basement, Calder's draughtsman had simply omitted to draw the area wall, since it was irrelevant to the matter at hand, and in the case of the main drain, an error had been made in copying the original drawing. But it was obvious from the nature of the complaints that Calder had not yet acquired the confidence of the Fortifications staff in London, and that Jones's estimate, detailed as it was, was still liable to differing interpretations on specific points. This last fact suggested to Calder that the overall plan was open to improvement. In his reply to the questions raised by his progress report, he made his first tentative suggestion for alterations. Could not two or three new casemates be provided in the rear of the basement area wall to provide storage space for the officers' quarters? Such casemates, "though eventually necessary," had apparently not been foreseen in the original plan.⁷

17 "Plan Elevation and Section of Retaining wall to two casemates of Defence," 1846. (Public Archives of Canada.)

When the working season ended, Calder finally had the time to examine Jones's revised estimate in detail. He concluded that it could indeed be improved upon by a few judicious additions, and on 6 January 1843 he submitted his proposals for improvement for the consideration of the Inspector General.⁸ The tone of his letter was unprovocative and gentlemanly. He was not attempting to cast aspersions on Jones's ability, but merely recommending a series of minor improvements which either were too specific to have been considered within the broad scope of the revised estimate or had been made necessary by developments since 1836.

The changes included the provision of porches and shifting rooms for the new magazines, the cellars for the redan (already mentioned in his letter of 15 October), an alteration in the method of constructing the arches of the proposed cavalier additions, and the substitution of ramps for staircases leading to the west ramparts. All of these were minor changes which tended to increase the efficiency of the completed work at little additional cost.

Calder also wanted to add more casemates. His argument in favour of doing so was based on an absurd misinterpretation of Jones's intentions. The latter had proposed strengthening the interior retaining wall by building arches over the supporting buttresses to form small cells or recesses which could be used for a variety of purposes. Calder misread the wording of the estimate and believed that it had been Jones's intention to carry the arches all the way through to the escarp. He noted that this had not been done in the case of those parts of the retaining wall already built, and went on to argue that, even if it had been done, the resulting space would have been too narrow to be useful. He proposed instead the substitution of full casemates in most instances, two in the re-entering angles of the redan and an unspecified number on the other fronts.

The collapse of the area wall in the northwest bastion once again led to a reconsideration of the early work. Calder's opinion was that

The whole of the scarp of the north front (excepting 120 feet of the right face of the N. W. Bastion) as well as the adjoining face of the East front . . . [is] in such a state of dilapidation from the badness of the mortar used in the construction . . . [and] the inferior quality of the stone and the unworkmanlike manner in which it is built, as to render it advisable to take down & rebuild the whole from the level of the ditch.

This, he considered, would account for the bulk of the additional expense he proposed — £7,000.⁹ The other features he proposed could cost, in all, just over £5,000 for a grand total of £12,620

When Calder's letter was received in London, a copy was immediately dispatched to Colonel Jones for comment. He replied on 1 March.¹⁰ Apart from a mild rebuttal of Calder's misinterpretation of his design of the retaining wall, he was generally disposed to accept Calder's judgement. He did differ in certain points of detail. Jones had a curious theory about magazine construction; he disliked the idea of external porches and of north-end doors, both of which he considered unsuitable in the Halifax climate. Consequently, he suggested alterations to Calder's proposals for the magazines, while agreeing that porches and shifting rooms would improve the design. He raised a gentle objection to the proposed ramps:

[They] would give more ready access for guns &c to the Rampart but yet [they] seem objectionable from interfering with . . . the breadth of the Rampart at the Flanks.

The judicious wording of the objection is, however, typical of the tenor of Jones's letter.

A letter from Lieutenant Colonel Edward Matson (the Assistant Adjutant General of the corps) enclosing the Inspector General's comments was similar in tone.¹¹ General Mulcaster blamed Calder's misinterpretation of Jones's design on an incorrect transcript of the 1836 estimate, and enclosed a true copy so that the Halifax version might be altered to read correctly. The Inspector General directed that Jones's plan be followed with respect to the cavalier and referred Calder to Jones's objection to ramps, but these things aside, he was willing to consider the remaining items. Additional casemates could be brought forward as items in the estimates if it was found that "the casemated accommodation already contemplated [is] insufficient." The cellar and shifting rooms items were both accepted in principle. The matter of the magazine porches and doors was referred back to Calder with instructions to confer with the "Officer of Artillery and the Ordnance Storekeeper" on the subject. Matson concluded by requesting detailed drawings and estimates for the proposed changes.

This exchange — gentlemanly, tactful and blandly reasonable — was in vivid contrast to the acrimonious exchanges which had greeted Boteler's first letters on the subject of alterations ten years earlier. Even as recently as 1840, Calder's proposals would probably have provoked a row, but in the intervening three years, attitudes had mellowed. The ensuing history of Calder's proposals, though almost as complicated as that of the 1836 estimate, was relatively harmonious. The era of bitter controversy was at last over.

18 "Plan of Upper floor of Redan," 1844. (Public Archives of Canada.)

19 "Sections of cavalier showin [sic] the mode proposed for rendering the arches secure against leakage," 1849. This plan illustrates the drainage provisions of Colonel Savage's staunching estimate of 30 April 1849. It is impossible to tell to what extent it was carried out. (Public Archives of Canada.)

III

The Inspector General's invitation to justify the increase in casemate accommodation prompted Calder to do something which no one had thought of doing before. In late April, he canvassed the other department heads to find out how much space they would need in the Citadel, both in peacetime and for a siege of two months.¹² Since he wanted an argument for additional casemates, he encouraged his colleagues to submit the largest possible claims for space. The Deputy Commissary General replied that he would need three casemates for a summer siege and at least three more for a winter one. (No commissariat stores were kept in the Citadel in peacetime.¹³) The Barrack Master needed two casemates under any conditions;¹⁴ the Commander, Royal Artillery (CRA), needed at least three;¹⁵ the Ordnance Storekeeper, four. This gave Calder a maximum figure of 16 casemates beyond the ones he needed for the normal garrison of one regiment. He considered this sufficient justification for bringing forward 16 additional casemates in his new estimate.¹⁶

The estimate was completed on 22 May 1843.¹⁷ It provided for all of the features mentioned in Calder's letter, excepting the ramps for the western ramparts. It also contained provision for fitting up the rooms over the end casemates of the cavalier and reconstructing the roofs of the magazines and ravelin guardhouses. In all, it amounted to £12,879 19s. 7d.

In his explanatory letter, Calder said little which was new. He had consulted both the CRA and the Ordnance Storekeeper on the arrangement of the magazines, and they had both accepted his proposals. As for Jones's objections to doors facing north, he noted that "all the magazines in Halifax stand north south and that each of them have [sic] doors in both ends." The two new aspects of the scheme were scrupulously accounted for. The cells over the cavalier end casemates were in response to a suggestion from the Major General Commanding. The substitution of rafters for cement on the dos d'anes of the magazines and guardhouses was the result of "the latter having shown itself unfit to resist the effects of this climate in the trials that have been made on the last mentioned Buildings."

The most interesting features of the covering letters were the three statements of accommodation appended to them. These were intended to support Calder's argument for more casemates, and they detailed the number of men intended for the Citadel's garrison. In all, the fort was designed for two field officers, 17 officers, 609 NCOs and privates and 39 women (the proportion of soldiers' wives allowed under regulations). In addition, provision was made for a 35-bed hospital in the cavalier and a school room, as well as for the usual assortment of storerooms. The average number of privates per casemate in time of peace was 22.¹⁸

London acted very quickly. The Inspector General dispatched the estimate to the Master General and board on 1 July 1843.¹⁹ In his accompanying letter, Mulcaster briefly reviewed the background of the proposals and recommended their acceptance.

The Estimate amounts to £12,879 . . 19 . . 7 and although its details have not yet been investigated by the Surveyor and some internal Fitments are omitted, it may I apprehend be taken as an Estimate sufficiently approximative to enable the Master General and Board to determine upon the additional Bomb proof accommodation and the omissions and renewals . . . which had not been originally provided for or have become necessary.

He admitted that the renewals were "discreditable to the department," but could see no way of avoiding the expenditure. He concluded,

Should the Master General and Board sanction the view I have taken, after a careful consideration of the above named Reports and circumstances, I propose making the necessary communication without delay for the Commanding Royal Engineer's guidance, in proceeding with the construction, and that the detail of the several additions be examined annually as they may be provided for in the Estimates for Parliament.

The board took less than two weeks to decide in favour of the new estimate,²⁰ and authorization was dispatched to Halifax on 18 July.²¹

IV

The method proposed by the Inspector General of approving funds for the new estimate signalled the beginnings of a change in the Ordnance accounting system. At some point between 1844 and 1847, the authorization of each item of expenditure as it arose in the annual estimates became standard procedure (in contrast to the old system of approving a general estimate and making annual grants against it). The new system had obvious advantages. It eliminated the embarrassment of over-running the original grant, as the Citadel account did at some time between 1847 and 1849 (the accounts for these years have not been located). It also, however, had one disadvantage. Like all changes, it produced a certain amount of confusion during its transitional stage. Not all the people involved understood the significance of the change, and one who did not was Patrick Calder who, in 1846, submitted yet another supplementary estimate for the completion of the Citadel.

The origins of this document are obscure. On the title page, it was credited as being in response to the Inspector General's letter of 18 July 1843²² authorizing the earlier estimate for alterations. But the surviving copies of the Inspector General's letter of that date contain no indication that such an estimate was requested or even contemplated. Possibly Calder genuinely misread the letter; possibly the title page was wrong and the new estimate was in response to a later communication from London, since lost. Unless new evidence comes to light, it is unlikely that we will ever know the truth of the matter.

In its format, the new estimate reflected the new accounting system. The items were divided into six classes:

(1) Works first detailed in Calder's first estimate for renewals, and subsequently authorized in the annual estimates for 1844-45 and 1845-46.

(2) Works from the same source, brought forward in the current annual estimate and not yet approved.

(3) Additional services found to be necessary since the 1843 estimate.

(4) Services in the 1843 estimate "ordered to be brought forward as excess."

(5) Works necessary because of failures.

(6) Services necessary for the installation of the armament.

Of the 17 items, 14 were new since 1843. These included water tanks, a well on the glacis, flagging for the areas, lightning conductors for the magazines, water pipes and gargoyles for surface drainage, flagging for the cavalier dos d'anes, fitments for the casemates, and a picket fence around the glacis to keep out trespassers. In addition to the new features, provision was made for rebuilding works which had been considered adequate three years earlier. These included the west ravelin in its entirety and six casemates of defence (four in the curtain and two in the northwest bastion) which had been part of the initial construction. Calder had intended to provide for curbs and platforms for the guns, but since no decision had ever been formally made on the armament of the work, he was unable to estimate the overall cost of the service. The entire estimate amounted to £26,563 3s. 1-3/4d.

Calder's covering letter was brief. It repeated the time-honoured phrase used by successive engineers in submitting revised estimates: "I have reason to think the amount of this estimate . . . will complete the work."²³ He went on to say that he had considered returning to the use of caponiers, but had discovered that they had been removed from Jones's first estimate for reasons of economy. Apart from this, and a few comments on the lack of information about armament, he let the estimate (which was the most detailed one yet drawn up for the project) speak for itself. His arguments for each individual feature were contained in the preamble of each item. Thus the rebuilding of the west ravelin was necessary because "the gorge [had] fallen down carrying with it part of the guardhouse"; besides this, the escarp faces had "cracked from the foundations upwards in several spots."

One feature of the estimate was Calder's emphasis on securing an adequate water supply. He considered the two wells insufficient for a garrison in the event of a siege, and proposed two complementary methods of supplementing them. The first method involved the construction of two water tanks under the casemate next to the guardroom "to be supplied with rain water collected from the ramparts of the work by the surface drains" (item 4). The second involved the provision of protected access to a well on the glacis near the northeast salient (item 5). The means of access proposed was a tunnel, like a countermine, from the counterscarp gallery. These two, in conjunction with the two existing wells would, Calder considered, be enough to supply the fort.

The new Inspector General's assessment of the estimate was favourable but cautious²⁴ (John Fox Burgoyne had been appointed to the post in July 1845.) He was disposed to accept most of the new features as "desireable" with the exception of the picket fence, which was, he thought, extravagant. But Burgoyne withheld final decision until he had better information. He therefore ordered that the document be returned to Calder for revision, that the CRA in Halifax be consulted on the subject of armament and that a scheme be submitted to the local commander of the forces for approval.

By the middle of July, Calder and Colonel Jackson (the CRA) had drawn up the armament proposal (see "... and keep your powder dry!" below).²⁵ Calder then proceeded to revise his estimate. Most of the revisions were minor. Asphalt was substituted for flagging in the magazine areas and an entry (item 3-1/2) was inserted for providing area walls in all three ravelins. Calder still did not estimate for the number of curbs and platforms needed for the proposed armament, although he did provide for 19 curbs for dwarf platforms, 12 wooden ground platforms and 12 wooden mortar platforms. The bulk of the revision consisted of alterations in the calculation of expense. The overall cost of the works proposed in the new estimate was put at £27,977 10s. 2-1/4d. excluding armament.²⁶

Calder's explanatory letter was, as usual, brief. He enclosed a list of replies to the specific points raised by the Inspector General, and the armament proposal, signed by Colonels Calder and Jackson, and endorsed, as Burgoyne had instructed, by Major General Dickson, the General Officer Commanding in Nova Scotia. The replies, for the most part, revealed that Calder agreed with the Inspector General's opinions except in the matter of the picket fence. Calder maintained that Burgoyne had misinterpreted his original suggestion.

The enclosure proposed for the Glacis is a common picket fence and not palisading. It is very near as cheap as an ordinary post and rail fence, and affords greater protection against trespass of every description in a Country where whatever belongs to the crown is almost considered a common good, more especially where land has for a length of time lain unenclosed and been dailly [sic] overrun by cattle, goats, geese, &c.

Calder requested that the lightning conductor estimate be revised in London according to the most approved opinion, this being a subject "where such diversity of opinion" existed. He debated the virtues of enclosing the ravel in guardhouses with an area:

[It] would be an improvement as a work of defence was the interior space sufficiently large, and it would render the building more wholesome in some situations, but in this Climate where a deep narrow ditch is liable to be filled with snow . . . it is apprehended that the walls might receive more injury and the building be less fit for occupation than at present.²⁷

He concluded that proper drainage would meet at least some of the objections.

London answered on 15 September. Calder was instructed to bring forward the items providing for the water tanks, the well, the magazine areas, the lightning conductors, the water pipes and the cavalier roof in ensuing annual estimates. The Inspector General stood firm on the subject of the glacis enclosure and instructed Calder to substitute a post and rail fence for his proposed pickets. Calder's objections to the ravelin areas were also dismissed:

Your objection to the ditch or area to the guardhouse would apply to all ditches and Buildings in that climate if care be not uniformly and constantly taken day by day to keep the footings of all buildings . . . clear of snow.

He was, therefore, enjoined to bring forward estimates for the areas when "the guardhouses in these outworks require reconstruction." As for the artillery plan, it was at present being considered by the Director General of Artillery and Calder would be notified when it was finally approved.²⁸

The same day that this response was sent, the Director General wrote to Burgoyne, pronouncing himself satisfied with the artillery proposals.²⁹ The proposals were then submitted to the Board of Ordnance, which communicated its approval on 10 October.³⁰ A week later notice of the decision was dispatched to Calder.³¹

In his letter instructing Calder about the disposition of his proposals, the Assistant Inspector General, Colonel Edward Fanshawe, reminded him to adhere in future to the new system of annual accounts and to submit proposals for new works in the appropriate annual estimate. This spelled the end of the tradition of all-inclusive Citadel estimates. Calder's revision of his second supplementary estimate was the thirteenth³² and last of a long and frequently confusing line. The change was symbolically appropriate. Despite all the disasters and crises of the preceding decades, the Citadel was visibly nearing completion, and major estimates were no longer appropriate to the situation.

It is a striking fact that all five engineers who held the post of Commanding Royal Engineer between 1828 and 1846 felt it incumbent on them to draw up large-scale estimates for the Citadel. Quite apart from the fact that the majority of these estimates were in response to genuine needs, we can, I think, discern in this pattern an attempt by each of the engineer officers to impose his own ideas on the work, to leave a monument to himself. To a greater or lesser extent, all five of them succeeded. But after Colonel Calder, no engineer had this opportunity. Calder's successors did not even have the chance, unlike Boteler, Peake and Jones, to gain some satisfaction from correcting, or trying to correct, someone else's disastrous mistakes. Calder's predecessors (excepting Nicolls) may well have looked on the work with a certain amount of satisfaction. To his successors, it was nothing more than an embarrassment.

Already in 1846 one future source of trouble was beginning to develop — hardly a disastrous problem, merely an irritating one which seemed to have no easy or permanent solution. It was becoming evident that the majority of the new casemates had a disconcerting tendency to leak.