Overview Documentary evidence Excavation history, procedure and methods Summary of research and analysis

Overview

	Map showing location of House 37 within the slave village.

House 37 was located on top of the rise overlooking the village and the great house, elevated about 40 feet above House 14. The site was on the southwestern margin of the zone of identifiable stone foundations, distant from the central pathway, and associated with the remains of stone enclosures and cut-stone graves. It was chosen for excavation partly because of its relatively isolated location and also because of its relatively complete state before excavation. When survey and excavation commenced in 1976, the house was easily identified by its substantial mound of rubble, with a tree growing through the middle, and intact sections of stone wall rising to 12 inches above the surface. Excavation continued into 1978, and the area of excavation was extended beyond the house foundations in some directions in order to search for enclosing walls and other features. House 37 was shown to have unusual features, shared in part with House 24. Occupation extended from the final decades of slavery and beyond abolition down to about 1850.

Documentary evidence
Detailed documentary data for the houses standing at New Montpelier in 1825, published in British Parliamentary Papers in 1832, includes information on 24 stone houses. These made up only 27 percent of the total village housing stock at that date, the other houses being wattled (wattle-and-daub) or Spanish-walled (timber frames infilled with stone and mortar). Although the houses of 1825 are described in some detail in the surviving documents, and associated with family household groups, it has not been possible to relate these specific descriptions to particular house foundations at the village site.

Only houses with stone foundations were visible by surface survey and only these stone foundations were excavated. Within the village a total of 42 complete stone foundations were identified and traces found of at least another ten. House 37 had a complete foundation but it is not certain that it is one of those described in the list of 1825. The unusual architectural features that it possessed may point to earlier or later construction. House 37 did not conform to the dimensions found for Houses 14 and 26, nor did it follow pattern of parallel lines associated with the layout of the houses surrounding Houses 14 and 26. Thus it is unlikely to have been one of the houses built in 1819 to shelter the enslaved people moved from Shettlewood. This is an important point, because it suggests that House 37 was less likely than Houses 14 and 26 to have been the product of design and construction closely controlled by the planter.

Generalized documentary evidence of objects that might survive in the archaeological record (such as iron cooking pots and thimbles) can be found for Montpelier, specifically Old Montpelier, in the early nineteenth century, but none of this evidence can be attributed directly to House 37 or any other house site. The list of houses from 1825 did enumerate the cattle, hogs and poultry belonging to each household but again it is impossible to link this evidence specifically to House 37. Descriptive accounts of plantation life at Montpelier exist from scattered points in the nineteenth and twentieth centuries, so that the broad context is well known.

Excavation history, procedure and methods
At the site of House 37, surface surveys and excavation commenced in 1976 and continued to 1978, with students from the University of the West Indies, staff from the Port Royal Archaeological Project, and volunteers, led by Barry Higman and Tony Aarons. Elevations were taken using the plane table and tied to a local datum, and a north-south grid established, for the entire village site. A separate survey was made of the mound covering the site of House 37, to take its elevations, before excavation was commenced.

Excavation was principally by levels of varied depth and excavation units or quadrats linked to the site grid, though construction trenches and other special features were treated separately. Most of the excavation was achieved using hand tools, generally trowels and brushes. Screening was used throughout most of House 37, using a 1/8 inch screen, but no flotation. Ceramics were mended to enable a Minimum Vessel Count (MVC), and the bores of clay smoking pipes measured using drills. Specialized analysis of the beads was carried out by Karlis Karklins. The faunal remains were studied by Elizabeth J. Reitz, assisted by Thomas Pluckhan and Philip Cannon, using the comparative skeletal collection at the Zooarchaeological Laboratory, Department of Anthropology, University of Georgia, Athens. In estimating the Minimum Number of Individuals (MNI), each house was regarded as a discrete analytical unit.

The pile of rubble contained many air spaces near its surface, showing that the stones belonged to the collapsed walls of the house. Removal of the superficial rubble exposed a more or less continuous stone foundation with external dimensions of roughly 22 by 19 feet, nearer to a square than a true rectangle, but not quite regular and with a protrusion to the north. The foundation varied in width from 12 inches in the western wall to 24 inches in the southern, with an even heavier section of 36 inches in the northern protrusion. Most of the stone was limestone, including massive blocks packed with smaller stones and cemented with lime mortar. Some of the blocks were worked or smoothed into regular shapes and a few of them had inscribed letters or numbers in their sides. Bricks were also common in some sections of the foundation. Most of these were imported, apparently from England, and probably were first used as firebricks in the sugar factory.

No evidence was found of a construction trench but the massive structure of the foundations suggests strongly that House 37 had stone walls all the way up to the eaves. The roof was probably shingled, as suggested by the large quantities of nails found in a fairly even scatter. Slate was excavated but some of it had mortar attached and it seems more likely to have been part of the walls than the roof.

	House 37 plan showing probable doorways, room divisions, plaster platforms, and floors.

One of the unusual features of House 37, shared with House 24, was its plastered floor. Whereas Houses 14 and 26 had floors of packed marl and earth, House 37 had a smooth plastered floor throughout, the plaster 1.25 to 1.50 inches thick and spread over a bed of packed marl. The plaster was a slaty gray colour unlike the mortar used in the walls.

Although House 37 was slightly smaller than Houses 14 and 26, it was more certainly divided into three rooms. The internal divisions in House 37 were of solid masonry rather than any lightweight timber alternative. These internal walls were narrower than the external walls, often less than 12 inches thick, and contained large proportions of cut-stone and brick. The plaster of the floor continued up all of the walls in a smooth curve. The internal walls were keyed into the external walls, demonstrating that they were built as part of the initial structure and part of a clear plan.

Another unusual feature of House 37, again shared with House 24, was the presence of raised masonry platforms in the ends of two of the rooms. These rose about 18 inches above the plaster floor and projected into the rooms near to 3 feet. The front of the platform was keyed and mortared into the adjacent walls, and contained some cut stones. The tops of the platforms were made of smoothed plaster just like that of the floor.

House 37 presented clear evidence of doorways. One entrance, about 2 feet wide, was located in the middle of the western side of the house. Another doorway, of the same width, was found near the centre of the eastern wall. There was also internal access from the southeastern to the northeastern room. A key and a hinge found in House 37 suggests the presence of at least one swinging door with a lock. It is unlikely that the house had windows of glass, at least during the period of slavery.

Summary of research and analysis
The location, orientation and dimensions of House 37 suggest that it was built independently, and that it did not belong to the series of houses laid out in parallel lines and constructed by the planters in 1819. The Mean Ceramic Date is 1818 and the Binford Pipe-stem Formula gives an average of 1775. The latter is average for the village but the Mean Ceramic Date is ten years earlier than the average, suggesting that House 37 preceded the planter-built houses rather than coming at the very end of the period of slavery. It may therefore be contended that House 37 was not only the product of construction work performed by the enslaved masons and carpenters of New Montpelier but was also a creation of the design of the enslaved. The raised platforms found in House 37 are strongly suggestive of the sleeping platforms found in the vernacular architecture of regions of West Africa from whence the people of Montpelier were taken.

Interpretation of the occupations of the persons occupying House 37 come most directly from the tool-related artifacts. During slavery, and beyond, annual distributions were made of various imported metal agricultural tools, notably cutlass or machete, bill and hoe, but other tools emerged from the excavations. House 37 yielded two hoe heads, immediately against the external walls of the house; these were all made of wrought iron and had shapes typical of the eighteenth century. Two bills (shorter, broader versions of the machete) and one cutlass blade were also found at House 37, more than at the other excavated sites. There was also an unusual sickle blade, a file, a broad iron spade blade, suggesting a family group largely employed in agricultural field labour.

Of the domestic life that was lived within the walls of House 37, little can be established, with items of furniture and methods of lighting hard to identify. If the interpretation is correct, beds were not needed in House 37 because they were replaced by the built-in sleeping platforms.

The food history of the household can be reconstructed, at least in some of its aspects. The documentary record for Montpelier describes the annual distribution, in the 1820s, of imported iron cooking pots and knives. These are present among the excavated artifacts, though House 37 had the lowest ratio of iron pot sherds in its artifact assemblage. It also yielded one metal spoon and an eating fork. Wear marks on stones may indicate that they had been used to grind corn or other grain. It is known, however, that vessels and utensils crafted from organic materials, such as wood, calabash and bamboo, were in common use, and these items were not recovered archaeologically.

Most of the many ceramics found at House 37 related to the preparation and consumption of food, and most of these items were imported (British-made) goods, purchased in the local markets or obtained from the plantation’s resources. It is surprising that so little locally made pottery was found, in light of the vigorous African tradition that survived in the island. Once again, it is difficult to assess the extent to which the ceramics, whether local or imported, were balanced by objects of organic materials.

Of the excavated ceramics, House 37 had the largest total number of sherds (1,103) but not the largest number of unique vessels, suggesting an above average rate of breakage. The MVC showed House 37 to have 76 unique vessels, fewer than House 26 (84 vessels) but substantially more than House 14 (50 vessels). House 37 had the largest representation of locally made ceramics or yabbas but they still made up only 2.6 percent of the total. In terms of shape and type, the vessels of House 37 included large proportions of bowls (36.8 percent) and plates (36.8 percent), and it also had above average proportions of jugs, and a teapot. House 37 had a relatively high ratio of plates to bowls, matching exactly the ratio at House 14. Compared to the other house sites, House 37 was characterized by relatively high proportions of pearlware (39.5 percent) and, to a lesser extent, creamware (17.1 percent). Both the creamware and the pearlware of House 37 were however more often plain or shell-edged than at House 26. House 37 also yielded a teapot, with a molded handle and a knobbed lid.

Direct evidence of the food consumed at House 37 is confined to animal sources and the data obtained from the vertebrate faunal analysis carried out by Elizabeth J. Reitz. House 37 yielded an MNI or 9, about average for the village overall. The most common species were unidentified rodents (2) and pig (2), with individual representatives of cow, chicken and sheep/goat. Contemporary descriptive accounts refer to the use of all these as potential food but whether the individuals found in House 37 were actually eaten is less certain as is the question which parts were used as human food. House 37 was the only excavated site in which there was no evidence of sawing of the bones, though there were indications of cutting. Fish, distributed in the largest quantities in pickled and salted form to the enslaved people of New Montpelier, left little trace. House 37 was also unusual in yielding a freshwater turtle. These animals could be caught in the nearby Great River.

Clothing and costume is known archaeologically principally through its technologies and accessories, whereas the documentary record specific to Montpelier emphasizes the kinds and amounts of cloth distributed by the planters during slavery along with caps and hats, and general descriptive accounts focus on style and fashion. The point of overlap is found in the needles, thimbles and scissors, that occur in both contexts. House 37 yielded three thimbles, a complete pair of scissors and three separated blades, but no needles. Outside the documentary record specific to Montpelier, House 37 yielded 13 buttons, a relatively small number, along with three belt buckles. Surprisingly, House 37 yielded just one bead. Analysis by Karlis Karklins argues that the evidence from the beads suggests overall that Houses 14, 24 and 26 were occupied at about the same time, and that House 37 does not conflict with this interpretation since its single bead had a counterpart in House 26.

This interpretation fits closely the conclusions drawn from the other datable artifacts as well as documentary record for New Montpelier, that the houses were occupied in the later decades of slavery and for perhaps two decades beyond abolition. The people who lived in House 37 experienced the brutality and hardships of slavery, the rebellion of 1831/32, and the apprenticeship, but remained on the estate after 1838 (or, less probably, were replaced by like individuals) to endure the immediate post-slavery period at least down to the abandonment of sugar production. As to the particular status of the occupants of House 37 within the community of the village, the fact that they lived in a house built of stone did not necessarily point to their being relatively well off, and it seems most likely that they were field labourers. The existence of the sleeping platforms does however suggest the possibility that the people who built and lived in House 37 were able to recreate a design brought with them from Africa.

Barry Higman

Australian National University

October 2006