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 Nicolls's Citadel

I

Although the genesis of the design for the present Citadel seems straightforward enough at first glance, the circumstances surrounding it are, in fact, rather obscure. A careful reading of the relevant documents reveals an essential uncertainty of purpose in the writings of the principals responsible for the design. Had the work been successfully completed without any major mishaps, the ambiguity surrounding its birth would be of no more than passing interest. As it happened, the adoption of the initial plan for the Citadel led directly to a decade of failure and confusion, and the origin of the trouble lay in the uncertainties evidenced in its inception and in the characters of the two men most directly responsible for it.

3 Colonel Gustavus Nicolls, RE. Portrait by his wife.

The first of these two was Sir James Carmichael Smyth. He and his fellow commissioners had the sometimes unenviable task of producing a coherent and reasonable general scheme in keeping with the framework laid down in the Duke of Wellington's 1819 memorandum and in his instructions to the commission. The major problem was that Wellington's instructions, though brief, were far too detailed. The duke was attempting to settle the defence of a country which he had never seen. Although his grasp of the overall strategic problems involved in the defence of British North America was sound enough, he faltered — sometimes badly — in his assessment of the value of specific locations. In fairness to Wellington, one ought to point out that he invariably phrased his suggestions in such a way as to give the commissioners the widest possible latitude in making their decisions. The problem was that Smyth and his fellow commissioners, in most cases, treated these suggestions with a reverence which their Victorian descendants usually reserved for Scripture. It was perhaps too much to expect that any engineer officer, no matter how competent, would have dared to contradict the duke himself but it would have been better if Smyth had displayed a little more independence in carrying out his commission.

This absolute devotion to Wellington's ideas was not, in itself, entirely bad. Smyth, however, combined it with an incurable optimism in estimating the amounts of money needed to construct the various works he recommended. It is difficult to be precise about the extent of his optimism, since so few of the works recommended were actually built, but it is worth nothing that in almost all cases the amounts estimated by the Commanding Royal Engineers (CREs) on the spot exceeded Smyth's figures (see Table 1). Those works which were finally constructed all cost more — some of them far more — than the figures proposed by the commissioners. Smyth was, by all accounts, a competent officer, so one is at a loss to account for his poor judgement. Perhaps he was merely ignorant of Canadian building conditions. Possibly the unrealistic estimates reflect Smyth's familiarity with political conditions in England and his awareness that excessive costs would deter Parliament from accepting his recommendations. In any event, the optimistic estimates contained in the final version of his report were to have serious consequences in the subsequent history of the Halifax Citadel.

4 "Outline map to illustrate a Report to His Grace the Duke of Wellington relative to His Majesty's North American Provinces" (1826). This map was inset in a large map of British North America drawn to illustrate the provisions of the Smyth report. It illustrates the relationship between the Citadel and the town and harbour batteries. The citadel was too far north to be of much use in defending the harbour and inconveniently situated for the landward defence of the town. The map shows clearly why a supporting fort on Port Needham Hill (north of the Citadel) was thought necessary. (Public Archives of Canada.)

Smyth's weaknesses were neatly complemented by those of the engineer officer most directly concerned with designing and constructing the Citadel, Colonel Gustavus Nicolls. Nicolls and Smyth had much in common. Both had enlisted in the Royal Artillery in 1794 and had transferred to the Royal Engineers in the following year. Both had risen through the regimental ranks in identical stages until 1813, when both were promoted lieutenant colonel. At that point their careers diverged dramatically. Most of Nicolls's career had been spent in colonial postings. He missed the opportunities afforded to officers who had had the good luck to serve in the peninsular campaigns and at Waterloo, with the result that he was still a colonel in the Royal Engineers — a mere major in the regular army — in 1825. Smyth, on the other hand, had attracted the patronage of the Duke of Wellington, married very well and, as we have seen, served with distinction in Europe and had been at Waterloo. By 1825 he was a major general in the army and a baronet.¹ Nicolls may well have resented his contemporary's striking success, but his resentment was either tempered or hidden by a well-developed sense of humility.

Nicolls's letters to his superior officers make interesting reading. He never contradicted. He greeted suggestions with praise and gratitude. He was deferential and complimentary. He never ventured to criticize. He was quite capable of calling the attention of "His grace the Master General" (Wellington) to the fact that the neck of the Halifax isthmus bore "so strong a resemblance to the lines of Torres Vedras (that so effectively put a stop to the success of the French in Portugal ...)" that he could not "refrain from noticing it."² Occasionally this weakness completely usurped his better judgement. In 1830, Lord Beresford (the Master General of the Ordnance at the time) differed with Nicolls's strategic assessment of a local prominence known as Cape Hill near Annapolis Royal. Beresford based his objections on a vague memory of the geography of the place; he had served there as an ensign forty-odd years earlier.³ Nicolls, whose acquaintance with local conditions was of a decidedly more recent vintage, did not venture to disagree. Instead he drew up plans for a work for the hill which he took the liberty of "naming Fort Beresford . . . it having emanated from His Lordship's recollections from having quarters at Annapolis."⁴

Gustavus Nicolls, therefore, was the last man either to resist the suggestions or to contradict the financial judgement of Sir James Carmichael Smyth, especially since the latter had the backing of so formidable a figure as the Duke of Wellington and good relations with virtually every senior officer in the engineer corps, from the aged Gother Mann (the Inspector General of Fortifications) on down. Picture the two men touring the defences of Halifax in the late summer of 1825, Smyth suggesting, Nicolls agreeing and enlarging on the suggestions. Between them, they fathered the present Citadel. They were also largely responsible for the disasters which befell their inadequate and slightly peculiar offspring.

II

In the case of the Citadel, Wellington presented the commissioners with the most ambiguous of his suggestions:

It would be most desirable to look at the ground upon which Fort George at Halifax, now stands, with a view to either its reform or the construction of a work of larger capacity upon that ground by way of keep to the works destined for the defence of the harbour, which might be garrisoned by two or three hundred men.⁵

This contradictory passage reveals the duke's fundamental uncertainty about the strategic value of the hill in the overall framework of the Halifax defences. It appears to suggest that the Citadel was less important than the harbour defences. On the other hand, it does not reject outright the possibility of a major building on the site. But it does indicate that Wellington had in mind a modest work, and it does not explicitly mention the possibility of permanent construction.

When Nicolls and Smyth came to consider the duke's recommendation, they decided that a "work of larger capacity" was clearly called for. To make a case for such a work, a variety of reasons was given. The commissioners argued that a work on the hill would

[protect the town] . . . support . . . the sea batteries. . . . give confidence to the troops and militia advancing to meet an advancing enemy, and . . . enable the General Officer in command to move to any other part of Nova Scotia with his disposable force . . . without exposing his stores . . . to be taken and destroyed.⁶

Smyth himself added the argument that expenditure on a permanent work would, in the long run, be cheaper than piecemeal expenditure on temporary fortifications.⁷ He also elaborated on what, in his opinion, was the nature of the threat to the town.

In Canada and Halifax the enemy is at our door. If our minister in Washington is deceived, if our generals are indolent or supine, a war may be declared and an invasion take place before the ministry in England are aware that hostilities are even contemplated. The construction of the fortress as proposed becomes consequently more urgent and in dispensible.⁸

Nicolls's contribution to the debate was phrased in his usual mannor:

Sir James C. Smyth has assigned several good reasons for the construction of a work on Citadel Hill, — I will take the liberty of adding one more, — viz, the good effect it would have on the Morale of the natives, as well as the contrary on that of their neighbours the Americans, who when on their frequent visits to this harbour, see its shores bristling with cannon on every side, and the British flag flying on the Citadel, on a fort respectable and strong for this side of the Atlantic, are thoroughly deterred from making an attack on Halifax.⁹

Despite its language, Nicolls's explanation of the reasons behind the building of the present Citadel is the only one which makes much sense. None of the explanations dealt at any length with the strategic value of such a work, and indeed the meager explanations which were offered were contradictory. In an era when the largest gun in common use in the British army had a maximum range of just over 3,000 yards,¹⁰ the Citadel could not effectively support the sea batteries. A gun mounted on the extreme southern end of the hill could only mask Georges Island and the middle reaches of the harbour — neither of which was an important factor in the event of a sea-borne attack. Nor was the hill in itself particularly well situated to defend the town against a land attack. Nicolls himself admitted that the first line of defence against such an attack would be the neck of the Halifax isthmus, which was out of sight of the Citadel.¹¹ The commissioners conceded that the hill could be properly defended only if it were supported by temporary works on adjoining high ground (notably Fort Massey Hill) and a permanent work on Needham Hill to the north.¹²

The best that could be said was that the Citadel, supported by the works described above and by a field army, could assist in the defence of the town against a land attack, and in this sense was intended as a keep. However, "keep" can mean any work, from a blockhouse on upward, and one wonders if perhaps a less elaborate work (like Captain Fenwick's towers) would not have served the purpose equally well.

No one connected with the project, with the possible and ironic exception of Nicolls, ever seems fully to have understood the fallacy in the strategic reasoning behind it. There is no evidence, at least in North American documents, that any questions were raised about the scheme, except in terms of purely technical aspects of the final design. Wellington's tentative and ambiguous assessment of the value of a work on the hill was accepted, and the commission recommended, without reservation, the present work on Citadel Hill.

III

The actual design was Colonel Nicolls's work. It is impossible to determine how much of it was contributed by Smyth and his fellow commissioners; their report is not sufficiently specific. They pronounced themselves in perfect agreement with Nicolls on the principles upon which he proposed to base his design, and enjoined him to submit plans and estimates at his "early convenience."¹³

The commissioners did, however, impose two restrictions, both of which were to have serious consequences. The first involved the question of the labour force for the new work.

[Colonel Nicolls] states that in turning the arches and other important parts of the construction of a fortress, which require great attention and superior work, he would prefer not employing contractors. . . . We . . . agree with [him] that it will be desirable to employ a company of Sappers in Nova Scotia, but we still recommend that whatever can be done by contract should be agreed under proper securities and subject to a vigilant superintendance.¹⁴

This decision led directly to the employment of contract labour in the building of the escarp walls, which was to have dire consequences a few years later.

The second restriction imposed by the commissioners was concerned with the estimated cost of the work. The commission decided, with its usual optimism, that the fortress would cost about £160,000.¹⁵ Most of the other engineers involved in the design of works recommended by the Commission blithely disregarded the commissioners' estimates, but Colonel Nicolls was of a different nature. He adhered to the estimates so scrupulously that he found himself forced to compromise in fundamental matters of design in order to keep the costs down. The exact nature of his compromises will he discussed later in this chapter.

5 "Plan No 1" (1825). This was Nicolls's original plan for the Citadel. In the course of construction, the eastern front was redesigned, the north cavalier and the caponier were abandoned, the magazine was demolished and new magazines and additional casemates were added. Despite these changes, the west, north and south fronts as finally constructed are virtually identical with the original design. (Public Record Office, London.) (click on image for a PDF version)

IV

Nicolls drew up his plans and estimates, which were duly dispatched on 20 December 1825. "You will easily perceive," he wrote to Mann, "that the trace has been formed more to answer the extent and nature of the ground than according to any regular system of fortification."¹⁶ It had indeed; compared to textbook plates, the trace was peculiar. It resembled a stubby arrow feathered at both ends. For this oddity Nicolls proposed to spend a total of £115,999 16s. 3 3/4d.¹⁷ Despite its peculiarities, General Mann could easily have discerned in Nicolls's plan echoes of earlier proposals and suggestions for fortifying the hill, including at least one of his own.

The title page of Nicolls's estimate reads: "General Estimate of expense of reconstructing in masonry, altering and adding to Fort George" (emphasis mine).¹⁸ This insistence on the relationship between Nicolls's design and the third Citadel (Straton's) is particularly appropriate. The two had much in common. Both contained four bastions and were alike in outline;¹⁹ both made use of cavaliers. Nicolls's ramparts were at least as high as those of his predecessor, and were occasionally higher,²⁰ despite the fact that in his excavations of the fort's interior, Nicolls had cut down the crest of the hill by as much as 20 feet. There were divergences, most of them resulting from one factor: Nicolls's use of permanent building materials. He was, therefore, able to make use of elaborate fortification techniques which had been denied Straton.

The greatest difference between Nicolls's and Straton's traces of the fort, however, was in their respective conceptions of the difficult northern and southern fronts. Nicolls considered Straton's trace unacceptable; the fronts were "so short as not to admit regular flanks."²¹ Both Fenwick and Arnold had proposed solutions for this defect, but Nicolls discarded both men's ideas and selected a method which Arnold had previously rejected, that of flanking from reverse fire casemates in the counterscarp.

The individual elements of fortification which Nicolls used fell into two classes: those which his predecessors had proposed and which had never been built, and those which (so far as we know) Nicolls originated himself. The casemates and caponier come under the former heading; the counterscarp gallery, countermines and ravelins come under the latter.

Casemates had found their way into both Fenwick's and Arnold's plans in one way or another, but in neither plan had they been put to such a variety of uses as in Nicolls's design.

In so small a work without casemated cover, troops may be shell'd out immediately.

The smallness of the work also admits of but a weak diverging fire being brought on the ground around it. By Casemated Cavaliers this fire is greatly increased and the Troops have at all times a Barrack secure from shells. — And for this reason as being the most exposed, I have also placed a Casemated Defensible Guardhouse on each of the . . . Ravelins, there not being a Covert Way.

The ditches of the Ravelins have been flanked by Casemates in the Body of the place, — the fire from the interior outwards, when it is to be procured, being preferable to that from the exterior outward.²²

In all, Nicolls proposed a total of 34 casemates including 16 single storey casemates in pairs under the ramparts, 7 two-storey casemates in each cavalier, and a casemated guardhouse in each ravelin. Of the total, 20 casemates (those in the ravelins and under the ramparts) were intended primarily for defence; the remainder were to be bombproof barracks.

The caponier was to serve two purposes; it was to be a flank defence for the west ditch and a communication with the west ravelin. The idea of using the caponier to defend the west ditch had first appeared in Arnold's design for the northern and southern fronts, outlined in his letter of November 1825. (See . . . we have nothing on Citadel Hill but a heap of ruins ... above.)

V

Nicolls may have planned a counterscarp gallery and counter-mines because it was impossible to form a covert way as a first line defence. In any event, he seemed to consider them to be a logical outgrowth of the four reverse-fire casemates.

[The north and south fronts] have . . . been flanked by casemates of reverse fire from the Counterscarp which also serve as Galleries for Mines, and I have included in the Estimate a Counterscarp Gallery around the direct Galleries to run out 20 feet beyond them allowing for Mines being exploded at that distance without injuring to [sic] the Counterscarp, or that low Galleries may be made to branch out at leisure.²³

The counterscarp gallery was a relatively unusual feature. Ravelins, on the other hand, were common in bastion fortifications, but none of Nicolls's predecessors had proposed their use. Straton lacked the wherewithal to build them properly, and ravelins on the northern and southern fronts as he designed them would have made the fronts look ludicrous. The spirit of Fenwick's design was such that ravelins would have been entirely irrelevant. According to Arnold's plan, there would have been insufficient room for them on the eastern and western fronts. Considering the size of Nicolls's ravelins on those sides. Arnold may very well have been right.

Arnold recommended, as we have seen, the occupation of a good deal of ground on the northern and southern fronts, beyond the limits of Straton's trace, to provide adequate flank defence and to take advantage of the commanding nature of the ground. This second reason presumably justifies Nicolls's occupation of much of the same ground with ravelins.

Three of the ravelins, those on the north, west and south fronts, were basically alike. In each of them, the guardhouse was placed in the centre of the gorge and was surrounded by a shallow ditch which took up most of the area beneath the ramparts in the ravelins' interior. The only important differences among the three were, first, the size of each (the northern and southern ravelins were identical and larger) and second, the means of access. The north and south ravelins were to be "entered from the ditch by wooden stops to be drawn up into the Guardhouse"²⁴ while on the western front there was to be a casemated two-storey guardhouse, the lower storey of which was to connect directly with the caponier.

The east ravelin connected to the body of the work by a bridge which entered at the mid-point of the gorge. Another bridge, approached through a passage under the ramparts on the right face, led to the exterior. In the eastern ravelin, the guardhouse was shaped irregularly and had no ditch. It was located on the left side of the gorge, immediately adjacent to the ramparts.

The shape of the fort made its interior cramped: the distance from curtain rampart to curtain rampart was less than 150 feet. It would seem that the four bastions were intended to be hollow, although contemporary plans vary on this point. The ramparts on the west side were somewhat thicker than those on the east;²⁵ this allowed more space in the northern and southern ends.

What interior space there was in the northern end of the fort was almost entirely taken up with the two identical cavaliers, one on a north-south axis between the curtains, and the other on an east-west axis fitting rather snugly between the bastions. Each consisted of seven two-storey casemates surmounted by a masonry and earth parapet, a terreplein, possibly of wood or earth (neither the plans nor the contemporary documents are explicit on this point) and curbs and racers for seven guns on traversing platforms. Both cavaliers were intended as quarters; the northern one was to be "a convenient Barrack for 320 men" and the eastern one "Officers Quarters for 4 Captains and eight Subalterns."²⁶

Certain peculiarities in the design of these buildings deserve comment. For one thing, the only provision made for access from the lower to the upper storeys of the casemates was by means of staircases in a wooden verandah which was to run along the interior side of each cavalier. As it was intended to remove the verandahs (to keep them from being set on fire) during an attack, it is interesting to speculate how Nicolls intended, in such a situation, to get men and ammunition to the guns on the roof. Another odd detail was the arrangement of the chimneys for the fireplaces in the casemates. The chimneys were to run through the exterior wall and emerge flush with the masonry parapet on the roof. Obviously Nicolls intended never to light fires during a siege.²⁷

Nicolls provided no detailed account of the armament proposed for the work. It is likely that he had no more than an approximate idea of the type and calibre of the ordnance to be mounted as he drafted his plans. He did make allowances in his estimates for platforms and embrasures in the appropriate places, as well as for traversing platforms in each of the north and south ravelins — two in each face — as well as four traversing platforms in the west ravelin and three in the east. He planned one embrasure at each of the bastion and ravelin salients, and seven on each of the cavalier roofs. The plan also shows two mortar platforms in each of the western bastions.²⁸ The 16 rampart casemates were intended to mount guns. The total number of gun positions would have been 63, a number which may be taken as an approximation of the number of guns intended for the work.

The fort was provided with seven sally ports. One of them provided access to the caponier. There were two in each curtain, and one in the re-entrant angle of both northern and southern fronts, all leading to the ditch. The two in the western curtain emerged opposite the rudimentary place d'armes flanking the west ravelin; they therefore provided access to the only defensive position proposed for the top of the glacis.

6 "Plan No 3" 1825. This, Colonel Nicoll's original design for the cavalier, was much altered in the course of construction. (Public Record Office, London.)

VI

Nicolls did not give a detailed account of the dimensions of the works in his proposed fort at any point in either his estimate or his covering letter. The estimate, in fact, gave only a cursory account of the cost of each individual work, without detailed calculations of materials, labour and workmanship involved. The only entries which come close to accounting for the extra services necessary for construction on the scale Nicolls proposed are as follows: a recommendation for the purchase of 12 horses "for the service of the work;"²⁹ one entry for £385 for "scaffolding, wheeling, planks, etc.," and another entry for £585 for "Repairs to tools, etc."³⁰ Similarly, there were few references to building materials. The estimate called for "granite quoins at the Salient angles of the shoulder [sic] of the bastions," but did not specify the type or quality of stone to be used in the remaining 99 per cent of the escarp wall.³¹ The whole question of labour was dealt with in a single paragraph.

The Estimate has been formed on the Principal of Workmanship being performed 3/4 by Civil Artificers 1/4 military. . . . — But this will vary materially according to the circumstances, as well as in regard to the Military assistance to be had as what part of the workmanship may be performed by contract; which I may offer my opinion, as to works of Fortifications I consider not likely to be more economical or the works to be equally well performed as by military Artificers, supposing the principal part to belong to the Corps of Royal Sappers and Miners; — as to stone, the principal part of the material, I much doubt the Department obtaining it by contract as cheap as by quarrying.³²

This last sentence is the only reference to the manner of supplying the raw materials, except for a recommendation that the necessary bricks be sent from England as ballast, "as the Bricks here are of very inferior quality."³³

Nicolls's estimate was, therefore, somewhat less precisely worded than one might expect. This made it easier for the colonel to conceal the compromises he had made in formulating the design. There were two major ones: the retention of the old powder magazine and the unusual thinness of the escarps.

Nicolls retained the powder magazine he himself had built in 1812 for use in the new Citadel. The magazine was a stone, bombproof building with a capacity of 1,344 barrels of powder,³⁴ located in the new fort at the southern end of the eastern curtain. In his covering letter, Nicolls mentioned it only once, to note that it could be advantageously used in the new work.³⁵ Nicolls's own section drawings clearly showed that the floor of the old magazine was 10 feet higher than the proposed level of the parade square of the new fort. Moreover, the magazine roof was somewhat higher than the adjacent ramparts.³⁶ Nicolls mentioned neither fact in either his covering letter or his estimate, and this omission seems to have gone unremarked in London.

Nicolls's escarp sections were another, less obvious problem. It is difficult to ascertain the dimensions of the escarps. In this, the modern researcher is a good deal better off than the gentlemen in the Fortifications department were at the time, since he, at least, has access to the contract specifications of 1828-1829 and 1830. The Fortifications department had no information whatsoever in Nicolls's estimate and covering letter; their only guides were his section drawings. These were contrived in such a way that, in almost all cases, they showed the escarp where it was broken either by a sally port or by the gate.³⁷ This circumstance, obviously, made accurate measurement of the escarp almost impossible. It also obscured the fact the Nicolls's escarp sections were rather less substantial than the fortifications textbooks permitted. A comparison between Nicolls's escarps and Vauban's recommendations (see Table 2) shows that Nicolls's escarps were, on the average, two feet thinner than they should have been. The same comparison also reveals that Nicolls's buttresses were up to three feet shorter than Vauban recommends, and did not in all cases run up the whole height of the wall.³⁸

*Columns 1 and 2 are derived from John Muller, A Treatise containing the Elementary Part of Fortification . . . (Ottawa: Museum Restoration Service reprint, 1968), p. 50; column 4 is derived from PAC, MG12, WO55, Vol. 1558, part 7, p. 50; Columns 3 and 5 are derived from NHPSB Plan 02-1825-12-2. These last figures are less accurate than the others.
†Measured at right angles to escarp wall.
§Greater figure is width next to wall.

VII

It is difficult to assess Colonel Nicolls's design for the Citadel. On the one hand, it is a competent piece of work, more sophisticated than previous plans and better adapted to the site than any of them, with the possible exception of Arnold's. On the other hand, Captain Fenwick's towers would have been cheaper and strategically more suitable for the hill. Nicolls's fort is admirable enough in itself, but its utility can be questioned. It is doubtful whether there was any purpose for the fort other than the one Nicolls himself suggested; to show the flag.

The suitability of the work, however, is not as important to its subsequent history as the adequacy of the specifications for its components set forth in Nicolls's estimate. Those were demonstrably insufficient to meet the demands of the local climate and soil conditions. The work had barely gotten under way when their insufficiency became embarrassingly obvious. Within four years of the beginning of construction it was apparent that major alterations (and more money) were necessary if the work was to be properly finished. By a misguided but entirely characteristic attempt to please his superiors, Nicolls not only put his own competence as an engineer seriously in question but also delayed the completion of the Citadel by almost a quarter of a century.