Canadian Historic Sites: Occasional Papers in Archaeology and History No. 17
by John Joseph Greenough
Truth and Consequences
Lieutenant Colonel Richard Boteler assumed Nicolls's command on 29 October 1831.1 It must have been plain to him from the start that he had inherited a potentially dangerous and disturbing situation. We can, from his later letters, picture him in his first months on the station, picking his way around the rubble of the partly built Citadel, looking in dismay at the breaches in the newly built walls, at the new west ravelin, already twisted and misshapen,2 at the old magazine, tottering on its island of mud in the middle of the partly excavated parade square. Boteler asked questions of his subordinates; there were few answers. Colonel Nicolls could tell him more, but the colonel was already in Quebec City, thankful, no doubt, that the mess in Halifax had passed into other hands.
Finally, in January 1832, the Fortifications department dispatched copies of Nicolls's original estimates and later correspondence to the new Commanding Royal Engineer, and informed him in a brief note that, with respect to the revetments, the Inspector General could not "sanction work of an inferior or insufficient description, nor a substance of masonry less than was used by Vauban." The department also asked for Boteler's opinion.3 The controversy which was to swirl about Boteler and his successors had begun.
Boteler replied promptly, dispatching two letters and two expense statements to London on 14 February. The first of those letters, a summary of the state of affairs as he found them, was a long litany of woe and confusion.4 The very shape of the fort was in question. Was Nicolls's plan for a redan on the east front to be adopted? Where were Nicolls's plans for drains for the place? Was it intended to retain the old magazine? If it was, he begged to inform General Pilkington that it held only 1,344 barrels of powder and was "now standing on ground 10-1/2 feet above the level of the interior of the fort." Was there any intention to provide barrack accommodation beyond that in the three cavaliers? If not, he suggested that
As to the work already begun, he did not consider it advisable to continue with the west ravelin, since it was already twisted. He did not think that the gorge would bear being carried up to full height. He had similar reservations about the escarps; the one on the left face was already bulging. He noted that the sum included in the 1832 estimate for repairing the breach in the southwest bastion would only rebuild the right face, and there was no money for repairing the breach in the northwest bastion. In any event, he doubted the value of piecemeal repairs to the old work; as places were repaired, others might "not prove to be sufficiently good." He advised either waiting to see if the masonry would stand or tearing it all down and rebuilding.
With all those difficulties it was not an easy task to find work which could be undertaken. Boteler recommended continuing work on the counterscarp and gallery opposite the northwest bastion, despite the inconvenience of rubble spilling from the breach in the opposite escarp, since this was necessary in order to keep the masons busy.
Boteler enclosed a balance sheet detailing the amounts remaining unexpended of the parliamentary grants for the preceding three years.5 The balance showed that most of the money had been spent. Of the remainder, however, some could only be spent after the problems raised in his letter had been satisfactorily resolved. This list of problems, with Boteler's comments, is worth examining in detail.
There was £2,277 6s. 9-1 /2d. left from the 1829 estimate on the cavalier account. By Boteler's reckoning, all that remained to be done was to sod the roof, shingle the verandah and lay the lower floor. The cavalier was one of the few areas in which Boteler expected no problems. There was £188 0s. 3-3/4d. left from the 1829 estimate for four granite gun platforms. These belonged to the ramparts on the west front and could not be placed because of the condition of the walls. Another £145 11s. 0d. for the curbs at the salient angles could not be used for the same reason. The sum of £1,562 14s. 8-3/4d. left on the 1830 account for the casemates of reverse fire could be used, though Boteler doubted the wisdom of proceeding with the work. The £139 11s. 3d. for retaining walls, £40 0s. 8d. for curbs and £5 9s. 4-1/4d. for granite platforms, all for the west ravelin, could not be spent because of the danger of the ravelin collapsing. The remaining funds, mostly for excavation, could be used.
When Boteler's letter arrived in London, the engineer officers were astounded. Four cavaliers! Admission of the utter failure of previous work! An inadequate and improperly placed magazine! No plans for drainage! Whatever had happened? Who was to blame? Most important to all, what was all this going to do to the estimates? Would they have to go to Parliament again for money? The London staff had changed since 1828. Mann was dead; Wellington was loading the fight against the Reform Bill in the House of Lords. In their places were Sir Alexander Bryce and Sir James Kempt. It was Bryce who received the bad news first, and his immediate, instinctive reaction was to try to preserve economy.
Bryce agreed with Boteler that it was unwise to undertake piecemeal repairs, and that it was necessary to wait and see how the work already completed would stand up over several winters. He suggested that casemating be substituted for cavalier construction. He had no firm opinions about Colonel Nicolls's proposed redan.
It was left to Kempt, in a pencilled marginal note on Bryce's letter, to assign blame for the situation and to speculate about the solution.
One wonders what Colonel Boteler thought as the winter of 1832 wore on. He had expressed his reservations about the Citadel project in strong language and had implicitly criticized his predecessor. What would London do? He got his answer in late May, and it was not reassuring. The Fortifications department, terrified by the prospect of asking Parliament for more money, demanded both results and economy demands which Boteler knew perfectly well were inherently incompatible. He was to "complete the work in an efficient manner, without increasing the amount of the original estimate or diminishing the projected casemate accommodation, and preserving if possible the Revetments of 1830, and 1831 which appear not yet to have proved defective."8 He was to report on Colonel Nicolls's proposed redan, which, Sir Alexander devoutly hoped, would "diminish the original Estimate of expense, and be a desirable alteration." The counterscarp gallery and mines on the east and south fronts were to be abandoned and the repair of the defective escarps was to be postponed until it was possible to find out whether they could be relied on. While Sir Alexander was "by no means disposed to sanction the hazard of a diminished revetment," he did wish, if possible, to "save those erected in 1830 and 1831," and Boteler was to do this, if necessary, by casemating. Finally the colonel was to report on the advisability of constructing "additional Magazine accommodation under the Ramparts in situations capable of thorough ventilation."
Fanshawe's private letter, which arrived with the same packet, was a little more explicit about some points. Sir Alexander, Fanshawe emphasized, was adamant about one thing; the revetments already built were to be preserved at all costs. Where it was impossible to relieve the pressure on the revetments by casemating, perhaps "additional buttresses, arches of discharge, or . . . dry walls in the rear" would serve as well. If it were absolutely necessary to rebuild failures, a special account of the sums expended was to be kept.9
The spirit of those two letters, with their enclosed comments from Bryce, was obvious, Boteler was being asked to work a miracle in order to preserve the department's honour. While we know that the Master General himself had agreed with Boteler's implicit criticism of Nicolls, no word of Kempt's approbation had seeped back to Halifax. Instead, the colonel got a curt injunction in Fanshawe's letter against making comments which might "excite controversial feelings." Boteler was to work wonders and he was not to rock the boat. After all, as Kempt's memorandum made clear, any criticism of Nicolls extended beyond him to the Inspector General's office itself, and Bryce had been Mann's deputy.
The Inspector General was sufficiently upset about Nicolls's performance to send him a copy of Boteler's letter of 14 February for comment. On 21 July Nicolls, writing from Quebec City, resolutely passed the buck back to the Ordnance.10 While it was true, he admitted, that he had never framed an estimate for the drains, he had shown them on his plan. Access to the ravelins through the ditch was considered sufficient at other posts Portsmouth, for example. While the barrack accommodation was insufficient for the garrison now proposed, it had been adequate for the number of men which Carmichael Smyth had originally required. As for the magazine, Nicolls wrote. "I believe there will be only a few spots outside Fort George from whence the ridge of the roof of this Magazine maybe seen; when the parapets [?] are complete; on this account no provision is made for another." This last was the weakest point in Nicolls's case (should the ridge of a magazine roof be visible from any point outside a fort?) but on the whole the colonel acquitted himself well. Nicolls, always the devious, ingratiating politician, succeeded in drawing attention to the fact that his original design had been faithful to the intentions of his superiors and had been approved by them. After he scored this point, all attempts to assign blame for the Citadel debacle temporarily ceased.
Nicolls's counter-attack was not forwarded to Halifax until September,11 but long before this Boteler had taken steps to protect himself in the event of the failure of his direct assault of 14 February. His position was, after all, unenviable. If he could not convince London that the situation was indeed serious and that expensive changes were necessary to complete the fortress properly, he would fail. His professional reputation was at stake. Shortly after he launched his direct assault, he changed to a different tack. The station records were sketchy; twelve plans, seven of them from before 1826,12 and a few dozen letters. If London could not be made to see the gravity of the situation by direct means, perhaps a persistent series of inquiries on points of detail would serve. By the end of the year, the Ordnance had received more letters from Boteler on the subject of the Halifax Citadel than it had received from Nicolls in the preceding four years, and the flood showed no signs of crossing. In the end, Boteler achieved his purpose, but the deluge was to involve the Fortifications department in the intimate details of the Citadel's construction, and began a long series of transatlantic exchanges which was to hinder and occasionally paralyze Boteler's successors.
The first such consultation involved the counterscarp and gallery opposite the northwest bastion. As this was one of the few areas where Boteler felt that work could proceed, he wished to be able to start construction as soon as the weather allowed. There was, however, a problem. Nicolls's plans were vague. While the ditch deepened at the salient, the gallery behind the counterscarp was apparently intended to remain in the same plane throughout the entire length of the wall, with the result that the loopholes were 6 feet 3 inches above the ditch near the west ravelin and 9 feet 3 inches above it at the salient. Should he build the gallery in this fashion, or should he incline it so that the loopholes were all at the same height above the ditch?13 A month later, Boteler reminded London of the problem, this time enclosing a copy of Nicolls's plan of the gallery and stating that the wall would be built according to plan if he did not receive instructions to the contrary.14 London finally replied on 25 May.
The Inspector General also suggested changes in the construction of the loopholes, and enclosed a sketch of the new arrangement.16
Fanshawe's letter did not arrive until the working season was well under way, and was therefore too late for its suggestions to be of practical value. Boteler, therefore, politely acknowledged its receipt and went on to say that he was proceeding along the lines indicated in his two earlier letters17 proof, if any were needed, that the whole object of the correspondence was not so much to elicit suggestions from London as to make his superiors aware of his difficulties. In fact, a new problem had arisen since his last letter. The salient of the counterscarp fell on "made ground" ground which had been filled up to form the glacis and in places the foundation of the gallery had to be "carried to a considerable depth, in one part 12'6" below the bottom of the ditch." Boteler had met this difficulty by "building up the foundation . . . as far as the level of the bottom of the ditch." and proposed to erect the gallery, following the official plan, on top of this. The Fortifications department, apparently satisfied with Boteler's judgement, did not reply to his letter.
A month after his questions about the counterscarp and gallery, Boteler dispatched a long list of statements and questions about the north and south ravelins.18 He noted that there was not enough money to complete the gorge of the north ravelin and that he had insufficient information to commence construction of the guardhouse and ditch in either. Was it Colonel Nicolls's intention to provide caponiers for these ravelins? Would it be possible to lower the escarps of both ravelins by two feet? London's reply took the form of four statements by the Inspector General in the margin of Boteler's letter.19 The first three dealt with matters of detail. The escarps could be lowered, if this did not expose the revetments of the body of the fort to distant cannonade; the caponiers were superfluous and cost money; a sunken area was to be provided around the ravelin guardhouses. The fourth statement contained an important concession:
No longer was Boteler explicitly enjoined to preserve economy at all costs. The tide was beginning to turn in his favour.
As the summer of 1832 wore on, the results of Boteler's tactics began to be evident in the financial balance sheets. Ironically, the problem was not that Boteler was spending too much but that he was spending too little. As we have already seen, when Boteler took over his command there was over £3,000 unexpended on the Citadel accounts, some of it money which had been voted as early as 1829. London's response to this fact was an injunction to spend the money; as long as the total expenditure during 1832 did not exceed the cumulative grants up to that time (reckoned at about £71,000) both the Ordnance and the Treasury would be happy.20 The Inspector General, earlier in 1832, had cut the annual grant by £3,409 17s. 2d. to £17,656 14s. 5-1 /2d., but saw no need for any further reduction.21 This gave Boteler a total of about £20,000 to spend. By the end of the working season, £3,000 remained unused.22 The failures of the preceding four years had taken their toll. Too much of the work could not proceed without some sort of guidance on basic matters such as the shape of the fort and the means of remedying the failures, as well as specific information on lesser topics such as the height of the escarp and the arrangement of the loopholes. A coherent policy could be formed only in the light of detailed information which, it had become apparent, neither Boteler nor London possessed. A few plans Nicolls's brief and insufficiently detailed estimates and a few dozen letters were all either side possessed, and these were not enough. The work was in a state which bordered on paralysis.
The major obstacle to the formation of a coherent policy was money. Boteler seems to have realized from the start that the deficiencies could not be made good and the work completed for the £116,000 allowed in the original grant. The problem was to convince London of this fundamental fact. Boteler's chance arose over Colonel Nicolls's proposed redan. On his arrival in Halifax, he had found a letter from the Inspector General asking for detailed information on the project.23 Boteler provided it. Estimate, plan and covering letter were dispatched on 13 April 1832.24 Having taken pains in his covering letter to state that he based his calculations on Colonel Nicolls's original estimate of 1825, Boteler reckoned that the additional expenditure for the alteration would be between £2,152 4s. 8-1/4d, and £3,254 11s. 2-1/2d. He emphasized that the greater figure was for the construction of Nicolls's proposal in all respects. Even this sum only allowed for a 30-foot escarp at the redan salient, making it substantially lower than the salients of the two adjacent bastions.
London was quite properly shocked. "Sir Alexr Bryce was not prepared from Col. Nicolls's letter . . . to expect any excess beyond the original estimate, even were his propositions to the full extent sanctioned."25 Once again, the Inspector General demanded the impossible; Boteler was to remedy the low escarp at the salient, adopt the full extent of Nicolls's proposal, and stay within the original estimate.
The Inspector General's letter contained one significant change in tone. Earlier answers to Boteler's letters had called for reports on specific problems, but this letter was sufficiently vaguely worded to be taken as a request for a general report. In addition, Bryce's marginal annotation of Boteler's enquiries about the ravelin was delivered in the same packet. The two allowed Boteler the freedom to offer suggestions based on his knowledge of local conditions. London had finally given Boteler a loophole, and, in the autumn of 1832, he prepared to step neatly through it.
There is no evidence in the surviving correspondence that London ever requested anything so formal as a detailed estimate for the completion of the Citadel, but that was exactly what Boteler set about drawing up. In fact, he produced three of them, and, not content with transatlantic letters, decided to go to London to argue his case in person. He set out on the Calypso in late January 1833. He never reached London. The ship foundered and took Richard Boteler with it.26
Of all the engineers who supervised the building of the Halifax Citadel, Boteler had the most difficult task. It fell to him to retrieve Nicolls's mistakes and to force London to recognize the necessity of a thorough reassessment of the work. Had he lived, the transition from Nicolls's inadequate planning to the more detailed work which was necessary for the completion of the fort might possibly have gone smoothly. His death, coming when it did, was an unmitigated disaster. In the confusion which followed, the Board of Ordnance found itself saddled with no fewer than eight different detailed estimates for the completion of the Citadel, and an administrative stalemate set in which lasted for more than three years. In the end, the matter was settled as Boteler had intended, but by then the project had fallen hopelessly behind schedule, and limped on for another 22 years before finally being declared finished.
Finding a successor to Boteler proved to be no easy task. The new Inspector General, Major General Robert Pilkington, recommended Sir George Hoste.27 Hoste, who had been a member of the Smyth commission, prudently declined.28 The next candidate was Lieutenant Colonel Rice Jones, the Commanding Royal Engineer at Chatham, who accepted. By this time the Fortifications department was keenly aware of the disadvantages of sending out a new CRE without extensive prior consultations on the course to be followed once the CRE arrived at the station. But upon what could such consultations be based? The Inspector General's office had not yet seen Boteler's detailed plans; they had gone down with the Calypso. A request was dispatched to Halifax for copies, and Jones was instructed to remain in England until they arrived.29
When Boteler left for England, his command had temporarily passed into the hands of Captain Loyalty Peake. Peake had had no part in the formation of Boteler's estimates, but he was well enough acquainted with the situation to realize that Boteler's revised estimates exceeded the amount originally provided for the construction of the Citadel, and that London would probably not be pleased with them. After Boteler's death, Peake saw a golden opportunity arising. Rarely had a junior officer been in charge of so important a project. If he could suggest an economical solution to the problem, the Inspector General would be certain to notice him favourably. In any case, he had little to lose. The difficulties in finding a successor for Boteler and the decision to keep Jones in England until more information could be gotten from the colony gave Peake the time he needed, and he used it to draw up four estimates of his own. Between September 1832 and June 1833, therefore, no fewer than seven supplementary estimates for the completion of the Citadel wore formulated.
Of Boteler's three estimates, the most elaborate incorporated all the changes proposed in the correspondence of the previous summer.30 The new features incorporated in the estimate included the redan, two new magazines (each consisting of a pair of linked casemates in the western bastions) and 16 now casemates, the bulk of them in the north, west and south fronts. The southern and eastern counterscarps were to be built without galleries or mines. Granite was to be substituted for ironstone in the wall facings as "granite is very abundant in the neighbourhood of Halifax and of the very best quality."31 The remaining items of the estimate were for the completion of other parts of the fort according to the original plan. The total expenditure was estimated at £92,378 5s. 8-1/2d.
Boteler's first estimate was, therefore, his assessment of the probable cost of implementing the suggestions made by London. Those did not necessarily accord with his own views, He thought that "it would be better not to place [?] casemates under the ramparts of the north, south and west fronts," and he disliked the idea of abandoning the southern and eastern portions of the gallery and countermines and the south cavalier.32 He therefore drew up a second estimate,33 intended to supersede those items in the first estimate which dealt with the casemates and counterscarp, and to show the comparative costs of the two schemes. In the place of the casemates, this estimate proposed a "substantial retaining wall"34 to take some of the loading weight off the escarps. The estimated cost was £79,014 2s. 10-1/2d., plus another £10,000 for the south cavalier.35
Boteler's third estimate36 was intended to supplement either of the others. The bulk of it was concerned with the probable costs of making good earlier building, should it be necessary to do so. The amount of the estimate was £15,975 14s. 1d.
Peake's four estimates were arranged in a similar fashion; the first three presented alternative schemes for completing the fort while the fourth dealt with the cost of replacing earlier work. Peake's approach to the problem was, however, only superficially like Boteler's. Boteler had begun with the assumption that additional spending would be necessary in order to complete the work and drew up his estimates accordingly. He was not an innovator; indeed, as we have seen, he personally wished to retain the essential features of Nicolls's scheme and produced his second estimate to show that this could be done at a reasonable cost. Peake began with the opposite assumption; the Citadel could be completed for the amount specified in the original estimate if drastic alterations were made in the physical shape of the fortress. In proposing such alterations, he altered Nicolls's original concepts beyond recognition.
Peake was merely continuing a process which had begun with Nicolls himself. In Nicolls's original idea, the four fronts of the Citadel were reduced to a regular order by duplication on opposite fronts and by the uniform provision of auxiliary features like the counterscarp gallery and mines. Insofar as this arrangement was based on the idea of four fronts of more or less equal strength, it was a triumph of geometry over common sense. Nicolls's proposal to substitute a redan on the eastern front was a recognition of the fact that that front differed, both in its relationship to the adjacent ground and in its accessibility to any enemy from the other three. Peake carried this reasoning to its logical extreme. Each of the four fronts, he argued, was unique; each presented different problems to an attacking enemy and each had special advantages or disadvantages for the defenders. With this belief as his starting point, Peake produced a scheme in which no two fronts were at all alike.
He left the west front exactly as Nicolls had designed it. Most of the work had been done, if inadequately, and it would have been too expensive to make any radical changes. On the eastern front he accepted the idea of a redan, but considered the counterscarp and gallery unnecessary, suggesting the substitution of "a palisaded covert way" instead.37 His argument for this proposal was that the nature of the ground and the close proximity of the town rendered it unnecessary to make this front as strong as the others. The north front he considered the most vulnerable because of
To remedy these faults, Peake proposed that "A Caponnier . . . be added, and the Counterscarp with gallery and mines . . . be continued from the Salient (N.W.) until it meets the proposed covert way at the N.E. Salient." The south front was not, he thought, such a serious problem.
He therefore proposed to complete the south front without a ravelin, but with a wide ditch, caponier, gallery and mines and a covert way; the last was to be an extension of the one proposed on the eastern front.
The core of Peake's scheme, therefore, was the use of caponiers. He listed six advantages to be gained from building them;
Above all, the caponiers had the advantage of being cheap. They provided the means by which some of the more expensive features of the original plans could be dispensed with, and "the several Fronts completed at a moderate expense and their capabilities of defence nearly equalized."
Peake estimated the additional money needed to complete his basic scheme at £53,997 12s. 10-1/4d.38 He produced, in addition, two variations on it, the first dispensing with the north and south caponiers and reinstating the south ravelin,39 the second encompassing both ravelin and caponiers.40 The cost of the first variant was put at £55,770 9s. 1/4d., and that of the second at £61,510 10s. 11-1/4d. Peake's fourth estimate, for tearing down and rebuilding escarps in the southwest and northwest bastions, amounted to £7,242 8s. 9-3/4d.41
We now come to the difficult problem of trying to ascertain the amounts by which the various schemes of Peake and Boteler would have exceeded the original estimate. If any contemporary calculation was done, no trace of it has been found, and the contemporary material which survives concerning Citadel expenditure before 1836 is frequently contradictory. The overall cost was to be computed by adding the estimated total of the new project to the amount of money already spent under the original grant. The problem lies in determining the latter figure. According to Peake, £55,718 had been expended as of 30 April 1833.42 The surviving Citadel account book, however, states that no less than £86,570 had been granted by the end of 1833.43 How does one account for the discrepancy? Had the unexpended balance on the Citadel account increased from £13,000 to £30,000 in less than a year?
Were the figures in the account book which was only begun after 1836 mwildly inaccurate? Or did Captain Peake manipulate his calculations to produce the lowest possible figure? Given the information presently available, it is impossible to tell which explanation is correct, but the last one is the most likely. The date Peake chose for his calculations 30 April was significant, since it fell before the beginning of the 1833 working season. By the time he wrote his letter of 12 June, several thousand pounds more would have been spent. The calculations which follow are, therefore, based on the minimum cumulative expenditure under the 1825 estimate; the total amount needed in excess of the estimate may have been anywhere up to £30,000 more.
The accompanying table (Table 3) details the calculation of the excess or saving produced by both Boteler's and Peake's schemes. In the case of each of the five basic schemes, the total amount of the new estimate is added to the £55,718 which, according to Peake, had been spent on the Citadel to 30 April 1833; the sum of these two figures is the estimated total cost for each scheme. This total is then compared to the original estimated cost (£116,000, in round figures) and the excess or saving calculated. To this is added the amount estimated for rebuilding old work; the total of the two is the total excess. The difference between the largest and smallest total excesses is more than £54,000. The least expensive is Peake's basic scheme (Peake's estimate No. 1) which represents a saving of £6,285 over the 1825 estimate. The most expensive is Boteler's first scheme, coupled with his estimate for rebuilding, which represents an excess of £48,071. On paper, at least, the range of alternatives was comprehensive.
*Boteler No. 3.
On 12 June 1833 Captain Peake bundled up the whole lot seven estimates, a covering letter, two explanatory letters, reports by Captains Wentworth and Rivers, and a list of plans and sent the entire collection off to London.44 Altogether, it amounted to more than 400 folio pages. One can almost hear the gasps of alarm when this monstrous collection was trundled into the Fortifications department. Pages and pages of figures, enough to keep the clerks busy for a month; the very complexity of Peake's report was its downfall. Colonel Jones was presently to be sent out to the station. He could read all these documents, of course, but only as a means of increasing his knowledge of the situation. He must produce his own report simple, coherent and (subject to London's approval) final. As for the fruits of Peake's and Boteler's labours, they were put aside and forgotten until further alterations were proposed ten years later.