Canadian Historic Sites: Occasional Papers in Archaeology and History No. 5

Excavations at Lower Fort Garry, 1965-1967; A General Description of Excavations and Preliminary Discussions

by James V. Chism

Introducion

Historical Sketch

Construction of this major Hudson's Bay Company establishment was begun in the fall of 1830 to replace the Company post in what is now the city of Winnipeg. Although it soon became apparent that it was an unsuccessful location for the trade, the establishment went through several periods of expansion associated with the arrival of British troops in 1845; the establishment of a large Company farm and industrial operations in the mid-1860s; and the arrival of several Canadian military and paramilitary groups in the early 1870s. In the 1880s, activity at the site was considerably reduced and most structures were taken down and removed. The fort housed several miscellaneous activities during its final decline in the 1880s, including the first mental hospital in the new province of Manitoba. The Company closed down its store in 1911 after which the fort served as a country club until 1963, when it was opened to the public as a National Historic Park.

Geographic Setting

The geography of Lower Fort Garry may be tentatively described with reference to Soils Report No. 5 of the Manitoba Soil Survey (Ehrlich et al. 1953). One may expect that exploitation of forested land, modern alteration of drainage patterns and modern agricultural practices will have changed soil patterns from their 19th-century condition. These will be noted when possible.

Lower Fort Garry is located on a 35 ft. high bank near the foot of the former St. Andrew's Rapids of the Central Lowland Landscape Area. The river bank bordering the site on the east is classified (Ehrlich et al. 1963) by the Manitoba Soils Survey as a soil associate of the neighbouring Selkirk-Beausejour Subarea; however, steepness of the bank and lack of any floodplain makes this classification seem questionable. The site is on a low river levee and is characterized by black soils overlying fine-textured Red River Clay. The clay is a 35-ft. to 40-ft. thick calcareous lacustrine deposit containing scattered granite and dolomitic limestone boulders. The clay and boulders overlie the dolomitic limestone Red River Formation of the Ordovician.

The fort enclosure and site land south of the enclosure lie on well-drained soils. A deeply cut creek south of the enclosure probably has traces of Riverdale Silty Clay, a local alluvial soil with a weakly developed profile. The deeply cut nature of this creek has probably contributed considerably to draining the slightly lower-lying land to the west of the low levee bordering the river.

The site area north of the enclosure is characterized by a low gravel ridge overlying what appears to be a somewhat less well-developed and somewhat sandier black earth. This ridge has probably contributed to the poorly drained nature of land immediately west of this point, although a shallowly cut creek has developed at the north edge of the site.

The area to the west of the site is composed of well to intermediately drained associates of Red River Clays with spotty occurrences of poorly drained associates of Osborne Clay. Poorly drained areas are often alkalinized and degraded, and where uncultivated, today they may be represented by marshy meadows such as those which are immediately west of the gravel ridge mentioned above.

The pattern of Company land usage reflects these land forms. As nearly as can be determined from Ingram's data, by inspection of the grounds and archaeological investigation, nothing was built on the land side of the levee with the possible exception of the as yet unlocated pig sties. An examination of historical and archaeological maps will also reveal to the reader that all buildings for livestock were placed north of the fort enclosure on the well-drained gravels.

One may well wonder why the main establishment was not built nearer the deeply cut creek where it would presumably have had easier access to the river and a wider strip of well-drained land. Less tangible values appear to have operated in the selection of the somewhat more imposing site chosen with its sweeping view both up and down the river. Practicality took precedence again when the deeply cut creek, which presumably drained a much larger land area than the shallow creek to the north, was chosen as the site of the later brewing and distilling complex.

It has been noted by Ingram that in selecting this particular stretch of the river, the Company governor felt that the area had agricultural potential with particular emphasis on sheep ranching. An examination of soil maps for the Red River from the present international boundary to its mouth at Lake Winnipeg shows few places with as wide a belt of well to intermediately drained Red River Clays and with only a spotty occurrence of poorly drained associates of Osborne Clay.

It has been proposed by Morton (1951) and Warkentin (1967) that the location of 19th-century activity was along the river because of the dry levees and accessibility to water for consumption and transportation. Accepting this thesis for the subject at hand, it appears that Lord Selkirk should have located his Red River colony here instead of the low marshy area now occupied by the city of Winnipeg.

A review of the Red River census from the 1840s indicated to the writer that land under cultivation by settlers near and in Winnipeg was generally less than five acres. A small group cultivated about 15 acres and one man was cultivating between 40 and 60 acres, but this may have been an earlier Company farm.

In 1857, on an area equivalent to approximately eight settlers' lots, the Hudson's Bay Company had 100 acres under cultivation without drainage ditches.

We might assume that the governor had a sharper eye for land than the earlier colonizers.

Isolated stands of scrub oak, poplar and willow can still be seen west of the present park where modern farming has not yet removed them. Stunted oak, poplar, willow, elm and occasionally white spruce occur on the bank along the river both north and south of the park land. The less common elm is taller than the oak, which suffers from the calcareous nature of the soil. In the creek bottom south of the fork, relatively larger elm and oak grow, no doubt due to richer soils provided by flood deposits. One may assume from the surrounding wooded lands that much of the park site was also wooded prior to the 19th century.

The river bed is cut deeply enough into the dolomitic limestone so it can be quarried with relative ease on the site to provide both stone and lime for construction.

The poplar stands would have served well for above-ground building material but it is difficult to believe that oak in the immediate area would have been large enough to have been used for below-ground and heavy stress members in anything but very small structures. There are large oaks to be found in very small numbers in more recent alluvial situations as discussed above; however, within a few miles down the river these stands become larger. The banks up the river nearer to Winnipeg support stands of very large, old oaks, adequate for use in framing larger structures.

General Condition of the Site

The walls, bastions and six original buildings of Lower Fort Garry were extant when archaeological investigations began at the site in 1965. The buildings and three of the bastions had been variously altered in the 19th and 20th centuries. Five of the buildings were in the fort enclosure; an additional building had been "reconstructed" as a museum, and the grounds within the fort enclosure had been disturbed in many areas by flower beds, greenhouses, water and sewage lines, power and telephone cables, sidewalks, widening of the original road within the fort, the construction of the museum and of a kitchen annex and a shelter for a York boat, the installation of a flagpole, outdoor display signs, horseshoe pitching pits and the operation of heavy machinery within the fort.

A portion of the site north of the enclosure had been under cultivation for gardens. A parking lot was built on part of this garden area. In recent years heavy equipment had filled and levelled other areas north of the enclosure during construction of the museum and parking lots. A portion of the gravel ridge had been disturbed by quarrying at some unspecified time and the entire area had obviously served as a refuse dump in the 20th century.

The fort land south of the enclosure was utilized as a golf course and had been relatively undisturbed except for the installation of greens and tees and natural erosion along the deeply cut creek flowing through that portion of the site. One original building was standing which probably incorporated its formerly detached summer kitchen as an annex.

The river bank bounding the site on the east appeared relatively undisturbed except for sewage and water supply installations. That portion located to the north was covered by 20th-century refuse.

West of the enclosure the construction of a four-lane highway had replaced an original 19th-century road, so one assumed that damage to structures had been minimized. Indications of a probable burial ground were still to be seen west of the Canadian National Railway line, although it was located in a wood lot and root disturbance was a clear possibility.

The most serious damage to the archaeological subsoil had been through activity associated with the construction of the new museum and parking lot and by plowing in the north field. Most other disturbance merely constituted a nuisance for the excavation undertaken in the three-year period reported here, although there have been isolated exceptions and also future excavations may find the situation more serious.

The survival of the majority of the site was due to the fort's use as a country club. Had all land to the north and south of the fort been broken for farming, the majority of the building sites would have been destroyed beyond archaeological recovery.

Many of the buildings at the fort had been taken down and moved or sold by the Hudson's Bay Company. Superstructure was removed, leaving only foundations, some flooring, cellars, the rotted sills of log buildings and an occasional remnant of wall missed in destruction and protected by slope wash. The remaining foundations and sills together with contemporary documentary material on this fort and others built by the Hudson's Bay Company made it possible to predict with fair accuracy the missing details of superstructure. Comparative material was also close at hand in the remaining buildings at the fort and buildings of the period throughout the Red River country.

Excavation Procedure

Investigations were carried out to determine the archaeological potential of the site, particularly those portions owned by the province of Manitoba, whose land began 200 ft. beyond the north and south walls of the fort enclosure. An initial inspection of the site was made using the original Watson map of 1928 (Fig. 2) and the David Lee (1965) superposition of Watson's map over the Dominion survey map of 1874 (Fig. 3). Lee included an overlay of certain structures from the Watson and Dominion survey maps which demonstrated a general disagreement as to location and dimensions of buildings. This comparison was not particularly helpful since the survey map was a rough map and was drawn on a very large scale.

2 Plan of Lower Fort Garry published by Robert Watson in 1928 (Hudson's Bay Company, Winnipeg) (cick on image for a PDF version). INSIDE THE FORT 1 Present entrance to Lower Fort Garry. 2 Originally men's house, soldiers' canteen in 1870. Later women's asylum. 3 Northwest bastion — Company's bakehouse. 4 North gateway. 5 Wooden house, used by Dr. Young, the prison doctor. 6 Prisoners' yard. 7 Penitentiary and asylum. 8 Remains of the oven built by prisoners. 9 Small gateway between bastion and stockade. 10 Northeast bastion. This has always been used as a powder magazine. 11 Old storehouse (frame building). This building lodged soldiers of the Wolseley Expedition of 1870. 12 Guard room and sergeants' mess, built 1870 for soldiers of Wolseley's Expedition. 13 and 14 East gate pillars. 15 The store deep in the ground in the centre of this gateway was placed there in 1883. 16 Southeast bastion — ice house. 17 Position of old tethering posts. 18 Part of south wall where Riel, Lepine and O'Donohue, with their followers, clambered over at midnight in 1870. 19 Hudson's Bay Company's second retail store and fur loft. 20 Foundations of another store. 21 Old site of the Lower Fort Garry bell. 22 Foundations of meat warehouse. Removed 1882. 23 Flagstaff. 24 Southwest bastion, washhouse and cookhouse. 25 Foundations of an old stable. 26 The residence. For a time the home of Sir George Simpson, Governor of Rupert's Land. 27 and 28 Old cannon. 29 The bell. 30 and 31 Sundial. 32 Former position of the Company's flagpole. 33 Old gateway of inner enclosure. OUTSIDE THE FORT 101 Lime house (frame). 102 Vegetable garden of penitentiary. 103 Hay yard for cattle. 104 Cattle yard. 105 Ox stable (stone). 106 Horse stable (stone). 107 Cattle stable. 108 Cow stable. 109 Prisoners' root house. 110 Old pathway to Selkirk. 111 Pathway to the river. 112 Stableman's house. 113 Path to cattle yard. 114 Penitentiary stockade. 115 Entrance gate to prisoners' yard. 116 Prisoners' ice house. 117 Old fashioned turnstile. 118 Stone marking the boundary line between the parishes of St. Andrew's (south) and St. Clement's (north). 119 Seat by the Red river. 120 Steps leading down to the river landing. 121 Pathway to York boats' landing beach. 122 Men's house and canteen. 123 Blacksmith's shop (log). 124 Farm Manager's house (log). 125 Engineer Abell's cottage (stone). 126 Grain flailing building (log). 127 Root house (log). 128 Beer cellar (log). 129 Store (log). 130 Malt kiln (stone). 131 Grist mill. 132 Saw mill (log). 133 Dwelling and brewery (log). 134 The creek. 135 The spring well. 136 Landing place for the Company's steamers, Polly, Colvile, Chief Commissioner, etc. 137 Lime kilns (store). 138 Miller's dwelling house (log). 139 Indian camping ground. 140 Supposed position of burial ground of soldiers of Wolseley Expedition, 1870. 141 Hudson's Bay Company farm lands. 142 Scene of First Indian Treaty, 1870. 143 Western post marking dividing line between the parishes of St. Andrew's and St. Clement's. 144 Hudson's Bay Company farm lands. 145 Indian camping ground. Burial grounds: The burial grounds for men at Lower Fort Garry were at St. Andrew's church (Church of England) four miles south, and at Little Britain (Presbyterian) about a mile south, also at St. Clement's (Church of England) about two miles north (Watson 1928).

3 Dominion Survey of 1874 (Public Archives of Canada).

A small part of the project's effort was directed toward support of immediate restoration planned for the principal residence, the fur loft-retail store and the southwest bastion. An even smaller amount of time was required for testing areas prior to their disturbance for purposes of construction or maintenance.

The major portion of the project's energy was oriented toward investigating as wide a range of activity and time as was possible in three field seasons, with emphasis on features outside the fort enclosure. Archaeological examination of major structures was also emphasized. These structures represented very nearly the full time span for occupation of the fort, as well as many functions. Such an approach can be justly criticized (Noël Hume 1969) and the painstaking examination of secondary structures, fence lines, and so on, must be done to add proper detail for interpretation of the site.

Horizontal control in excavations at Lower Fort Garry was maintained by the utilization of a base line surveyed through the long axis of the site. This line, which lay along the river side of the fort enclosure at 40deg; 50' 20"N., was designated grid north, and was extended through the site in 100-ft. intervals marked with steel pins driven flush with the surface. The grid system was oriented parallel and perpendicular to the base line, and each square of the grid was described by the north-south and east-west distance in feet at its southwest corner from the base line point zero near the northeast bastion. After the 1965 season, the project found it convenient to rely less on the grid system and more on maintaining control through triangulation of points in the excavation from points on the base line. This was particularly convenient when clearly architectural features were under excavation.

Nomenclature of excavation units followed that required by the National Historic Sites Service (Rick: 1965). The largest unit of excavation was the "Operation." Such a unit might have been a building, a portion of a building or a general area. The operation in turn was subdivided into "Sub-operations" which were either culturally significant, operationally convenient, or both. This might have been a room within a building, a building within an area, a series of five foot squares or simply a measured portion of a larger area. Vertical control was maintained by using distinct soil strata. It was then presumed that such an archaeological "Lot" could be defined as having a meaningful cultural context. A lot could also represent the precise location of a specific feature, artifact or grouping of artifacts. In the event of unclearly defined or thick deposits, it was sometimes useful to utilize mechanical lots of 0.5 ft. It was required that measurements be recorded in tenths of feet, making it necessary to convert to inches when such units are needed.