Fort Vancouver Historic Structures Report

History and location

No clear statement that a root house existed within the pickets of Fort Vancouver prior to 1853 has yet been found. Testifying under oath in 1865, Dr. H. A. Tuzo described the post buildings as they stood at the time of his arrival there in 1853. Among the other structures on the north side of the enclosure, he said, was "an excellent root house." [1]

Evidence of the continued existence of the root house after 1853 and until the spring of 1860 is conclusive. In 1866 Dugald Mactavish, who had served the Company for many years at Fort Vancouver as a clerk and commissioned officer, listed the buildings that he remembered standing within the stockade in 1858. Among them was a "Large Root House." [2]

During the spring of 1860 members of the British Boundary Commission party took several photographs of Fort Vancouver. One of these (Plate XXVIII, vol. I) shows a low structure, with a gabled roof rising from the ground level, situated in the northwest angle of the stockade, north of the sale shop building and extending almost to the north palisade line. This structure from its form clearly was the root house, though no map has yet been found that labels it as such.

The Root House is not listed in the inventory of Company buildings prepared by a board of army officers on June 15, 1860, the day after Chief Trader James A. Grahame turned the fort over to the military authorities, nor is it shown on the ground plan sketched by the same board (Plate XXX, vol. I). [3] Because this list included even "hovels," it is possible that the Company's officers ordered the building destroyed during the last few weeks of the firm's occupancy so that the cellar could be used as a dumping place for trash. Archeological findings tend to support such a hypothesis. [4] On the other hand, the board of officers simply may have considered the root house so far decayed as to be unworthy of notice.

Archeological excavations conducted in 1952 confirmed the location and structure of the root house as shown in the 1860 photograph. In the words of the supervising archeologist, Mr. Louis R. Caywood, the construction of the building, as revealed by the remains, left "no doubt" as to its function as a storage pit for root vegetables. [5]

There seems to be no serious difficulty, therefore, in identifying the building that the archeologists found in the northwest stockade angle and termed No. 26 on the "Summary Sheet, Archeological Excavations" (Plate I, vol. I) as the Root House and in being reasonably certain of its existence during the period 1853 to 1860. The trouble comes when one attempts to establish the existence of this structure prior to 1853.

A glance at the Emmons ground plan of 1841 (Plate III, vol. I) reveals that a "General Store House" (No. 17) stood in the same general vicinity as the later Root House. A closer examination quickly reveals that this earlier warehouse probably was not the Root House of the 1850s. Not only were the shapes and locations somewhat different, but the 1841 building abutted the west stockade wall, which, as has been seen, was probably at that time at line AD on the "Summary Sheet, Archeological Excavations" (Plate I, vol. I). The Root House, archeologists determined in 1952, did not abut wall AD but straddled it. "The construction of the root house," wrote Mr. Caywood in reporting these findings, "apparently obliterated all traces of a former stockade wall [AD] [under the new Structure]." [6]

It is plain, then, that the Root House was constructed after the west stockade wall at AD was moved westward at least to position HG sometime between 1841 and January-February 1845. The construction could also have occurred, of course, subsequent to the second moving of the west wall, in January-February 1845, to position IJ.

What may be a clue as to the construction date of the Root House seems to be the "Line of Fire" map of September 1844 (Plate V, vol. I). It shows a building in the northwest corner of the fort that in size, shape, and location seems almost identical with the Root House as outlined by the archeological remains. However, it is shown abutting a west stockade wall, which apparently the Root House never did. Therefore, the question is, was the building shown on the "Line of Fire" map the warehouse (No. 17) on the Emmons plan of 1841 abutting wall AD, or was it a new structure improperly shown as abutting wall HG instead of merely very near to HG as would have been the case if it were the Root House?

With the evidence at hand, it appears impossible to answer this question. As has already been developed, it seems to the writer that the west stockade in September 1844 probably was in the position HG, but it would be foolhardy to make a positive assertion to this effect.

This uncertainty is heightened by the failure of succeeding ground plans to support a case for the continued existence of the structure shown in the northwest corner on the "Line of Fire" map. The detailed Vavasour plan of 1845 does not show any structure whatsoever in the stockade angle where the Root House is known to have stood later (see Plates VI, VII, VIII, vol. I). The Covington map of 1846 places a small structure near the southeast corner of the bastion, but it seems to be square in outline and bears no resemblance to the shape of either the building on the "Line of Fire" map or the Root House as revealed by archeological evidence (see Plate XIII, vol. I).

However, this lack of delineation on maps cannot be taken as absolute proof that the Root House was not in existence in 1845 and 1846. Both of these ground plans either omitted or incorrectly represented several minor structures. [7] And maps drawn after 1853, when the Root House almost surely was in existence, also do not show a structure that can be clearly identified as the Root House. The carefully executed Bonneville survey of 1854 (Plate XIX, vol. I) pictures no building at all on the Root House site. A map drawn by Brvt. Capt. T. R. McConnell in 1854 from the Bonneville survey shows a very small structure in the northwest corner of the fort. [8] And even the very detailed survey prepared under the direction of Capt. George Thom in 1859 depicts a very small building directly north of the trade shop that in no way resembles the large Root House (see Plate XXIV, vol. I).

But perhaps the most telling evidence in favor of a hypothesis that the Root House was not constructed until after 1846 is found in the Company's 1846-47 inventory of its buildings at Fort Vancouver. Three "Root Houses 60 x 20" are listed in that census, but all are carried under the heading "Out Buildings." Seemingly, then, there were no root houses inside the fort when that inventory was taken. [9]

Yet the possibility remains that the structure shown in the north west stockade angle on the "Line of Fire" map of September 1844 was indeed the Root House. In that case the west fort wall shown on that map would have been line HG, and the Root House would have been erected across the line of the older wall AD when that wall was demolished between 1841 and 1844. But to assign a date as early as 1844--indeed any date before 1853--to the Root House must be on a highly tentative and speculative basis only. Proof is lacking.

The location of the Root House is now designated as Building No. 26 on the site plan of Fort Vancouver National Historic Site.

Construction details

Between the remains of the building discovered during archeological excavations and the glimpse of the structure given in the 1860 photograph, the dimensions, materials, and construction of the Root House are known in considerable detail. Undoubtedly a reexcavation of the site in the presence of an historical architect would provide additional facts.

Rotted wooden sills, situated forty inches below the present ground surface (the depth of the historic ground level at this point was not mentioned by Mr. Caywood), outlined a structure with a north-south length of fifty-five feet and a width of twenty-one feet. Charred remains of a wooden floor were found. No evidence of wood or stone walls seems to have been uncovered at the portions excavated, although the existence of wooden walls is not necessarily eliminated by this fact. Along the earthen east wall there was evidence of two posts projecting at an upward angle. Mr. Caywood believed they "may have been part of the low gable roof which came to the ground line." [10]

The 1860 photograph shows that the roof was formed of vertically laid planks, probably lapped "board and batten" fashion. The ridge boards were also heavy planks (see Plate XXVIII, vol. I).

The location and style of the entrance are not known, but further careful archeological excavations at the site might reveal traces of a stairway or other evidence that would throw light on this matter. Comparative studies of European and American root houses of the same or an earlier time undoubtedly would provide other clues as to typical construction details for this type of structure.

Recommendations

a. Careful archeological excavations should be conducted in the northwest stockade angle in an attempt to determine the sequence of structures in that area. Establishment of such a sequence might throw light on the construction date of the Root House.

b. The site of the Root House should be completely excavated and all architectural details should be carefully examined and mapped.

c. If the refuse that archeologists discovered in the Root House cellar in 1952 is still there, it should be carefully screened for materials, such as the "interlocking sheets of iron," that would throw light on the architecture of other fort structures.

d. Upon the basis of the archeological findings and the 1860 photograph, architectural drawings of the Root House should be prepared by historical architects versed in Hudson's Bay Company construction methods and in the design of root cellars in general. A special study of nineteenth-century British and Canadian root cellar design would be desirable.

e. In view of the uncertainty concerning the date of erection of the Root House, it is recommended that a decision concerning the reconstruction of this building be delayed until it is seen if renewed archeological excavations produce information as to the time of building. If it can be shown that the Root House was not erected until after 1846, the building should not be included in the reconstruction project, interesting though it is. If no further information concerning the date of erection can be found, it is recommended that reconstruction be authorized but with a very low priority.

1. Br. & Am. Joint Comm., Papers, [2:]176-77. The 1846-47 inventory of Company improvements at Fort Vancouver included "3 Root Houses, 60 x 20," but they were listed under the heading "Out Buildings" and thus were not within the stockade. Ibid., pp. 118-19.

2. Ibid., pp. 202-3.

3. Ibid., [9:]75—77.

4. Caywood, Final Report, p. 20. The evidence as to whether the building was burned is inconclusive. The wooden floor was "badly charred," as if the structure had suffered a fire, but the sills showed no signs of burning.

5. Ibid. In actuality, the discovery of the Root House sills, revealing the dimensions of the structure, seems to have been the first step in the identification of this building. Only about half of the Root House site was excavated in 1952.

6. Ibid.

7. For example, the Vavasour plan which does not show the privies, were a conspicuous part of the fort scene; the Covington map omits the magazine and one or more structures in the Beef Store-Wheat Store vicinity.

8. See Hussey, History of Fort Vancouver, Plate XVI.

9. H.B.C.A., B.223/z/5, MS, fols. 265d-266.

10. Caywood, Final Report, pp. 19-20, and Excavation Drawings, sheets 4 and 7.