Most visitors to Rock Hall Museum in Lawrence, NY, are quite surprised to learn that this stately manor was home not only to a well-to-do white family but also to one of the largest groups of enslaved Africans in a single household in what was then Queens County. Learning that the owners of the house were an Antiguan planter family known as the Martins, this discovery becomes less surprising. Still, the little known story of the African American majority in the household is currently based largely on conjecture and generalized patterns of and assumptions about master-slave relationships in the Northeast. In this paper we draw on the museum‚Äôs well-documented history of the household, complemented by a discussion of a fascinating set of fortuitous archaeological discoveries made in the west yard of the museum‚Äôs property, to bring out new details about the practices and beliefs of the slaves who lived at Rock Hall in the late-eighteenth and early-nineteenth centuries.
The Rock Hall Site
Owned and operated as a historic site by the Town of Hempstead, Rock Hall Museum is located in the Town of Lawrence, New York (Figures 1 and 2), in the southwest corner of what is now Nassau County. The museum property is the remnant of a 600-acre estate built in 1767 by Josiah Martin, a wealthy British sugar merchant from Antigua. While Martin lived on another estate in northwestern Long Island before 1767, at the age of 67 and in ill health, he sought to provide his family with a grander home. Fearful of the small pox epidemic in Manhattan, he chose to buy an existing farm farther from the city near the south shore of western Long Island.
Figure 1: The location of the Rock Hall site on Long Island. Drawing by Ross T. Rava.
Figure 2: Rock Hall Museum, 199 Broadway, Lawrence, New York. Courtesy Rock Hall Museum.
After this purchase Martin rebuilt the site to his elite Anglo-American tastes. Designed in the then-popular Georgian architectural style, the manor house is known for its high-quality construction with a central hall and four end-gable chimneys. According to historical records, the Martin family and their relatives lived at Rock Hall until 1818. After a series of misfortunes, the manor house and property went to auction and was purchased by the Hewlett family in 1824. The house was later deeded by the Hewletts to the Town of Hempstead in 1948 then restored and opened as a museum in 1953. Listed on the National Register of Historic Places, it has since become a popular destination for those interested in exploring the history of eighteenth-century Long Island.
Rock Hall was the manor house of a large farm that allowed the Martins to be almost self-sufficient. Corn, grain, and vegetable crops were grown on the property. Cattle, hogs, and sheep provided meat and dairy products. There were also plentiful wild fowl to hunt and fish to catch. Moreover, with a large slave labor force, Martin was well situated to oversee a productive plantation. Historic maps and surveys of the site in the early nineteenth century show that the house stood at the center of a complex of structures built to support the estate‚Äôs agricultural production. An 1817 survey by Morris Fosdick illustrates the estate structures (Figure 3) and supports a view of Rock Hall as modeled after a Southern plantation with a large formal manor house flanked on the east and west by a series of lower status outbuildings. It is possible that this plan reflects an influence of Josiah Martin‚Äôs family heritage as sugar planters in Antigua. Having spent much of his life on his family‚Äôs Antiguan plantations, Martin was certainly familiar and perhaps also comfortable with the typical spatial and social separation of his family from the domestic and farming activities performed by their slaves in Antigua. It would not be surprising, then, if he set up his estate in New York in a similar way.
Figure 3: Close up of Morris Fosdick‚Äôs 1817 survey of Rock Hall. Courtesy Rock Hall Museum.
One pattern indicated by the historic maps is a distinct use of the spaces on either side of the Martins‚Äô manor house. On the east side of the site (the left side of Figure 3), there were a number of utilitarian service structures likely used for farming storage, housing animals, and perhaps for household and farm production. Notably absent from these structures are any chimneys, which would have been necessary for human residences or a kitchen. Unfortunately, in the 1880s, a two-story addition to the manor house was built over the east yard, precluding any archaeological testing that could determine the accuracy of the survey. However, the museum still owns the land on the west side of the manor house (right side of Figure 3), and this area has not been visibly disturbed by later construction. Moreover, two of the service structures in the west yard (see Figure 3) are depicted with chimneys, suggestive of an ancillary dwelling structure and/or kitchen. As such, the west side is likely to have been where the slaves lived and worked at Rock Hall.
These details of the 1817 survey inspired the museum staff to consider undertaking archaeology at Rock Hall, though their interest was not initially in the slave housing that might be found. Like many historic house museums, Rock Hall offered public demonstrations of early American domestic cooking techniques because interpreting how food was prepared and served has long been seen as a powerful way to educate visitors, especially school children, about how our predecessors lived. Complicating this objective at Rock Hall was the fact that the manor house does not contain a historic kitchen. Despite its prominent chimneys and multiple hearths that served to heat rooms on the first and second stories, none show evidence of having been designed and used as a working kitchen hearth. Thus, it has long been assumed that at least one of the west yard structures depicted with a chimney on Fosdick‚Äôs 1817 survey must have been the original kitchen. In hopes of locating this missing kitchen so that it could be incorporated into the museum‚Äôs presentation of the site, archaeologists were invited to test the west yard to define its location.
Since the initial investigations in 1995, the west yard has been carefully studied and explored resulting in several archaeological reports. Drawing from these studies, we make the case in this paper that the structures depicted with chimneys in the west yard in the 1817 Fosdick survey included not only a kitchen but also a slave quarters, or more probably a structure that served both of these functions. As such, these reports provide a rare opportunity to recover data illustrating the material practices of enslaved African people on Long Island, and to put that material into the specific context of daily life at Rock Hall. Before considering this data directly, we want to first address the problems of visibility in the archaeology of slavery in the American Northeast, where few slaves are known to have occupied distinct dwellings or spaces apart from their owners. Lacking such clearly delineated spaces, archaeologists have long struggled to identify the remains of housing and associated separate spaces that could have been used by enslaved people.
Slave Housing in the Northeast
The study of slavery in the Northeast, and in New York in particular, has advanced significantly in recent years. Prompted in part by the discovery in 1990 of the human remains of enslaved persons at the African Burial Ground site in Lower Manhattan, this research has shed new light and brought greater public awareness to the existence of slavery and the presence of African American communities in colonial and early American New England and New York. This recent work has explored a diverse array of important topics such as demographic patterns, the dynamics of the slave trade, family life during slavery, work routines, health, religion, and cultural expression as well as overt and hidden forms of resistance.
The question of slave housing in the Northeast, however, continues to be a topic of particular interest and frustration to historical archaeologists. The difficulties become apparent when we compare slave housing in the North with that of the South and the Caribbean where plantations were characterized by distinct slave quarters, if not entire slave villages, constructed to house the dozens or sometimes hundreds of slaves who lived and labored there. This domestic segregation between masters and slaves allows archaeologists to collect material culture from slave quarter sites with a high level of confidence that the objects are rightfully associated with the enslaved community. Since John Otto‚Äôs influential analysis of Cannon‚Äôs Point Plantation in coastal Georgia, for example, historical archaeology has opened a fruitful line of inquiry into comparative studies of slave, master, and overseer assemblages in the South. Similarly, Leland Ferguson and others have drawn on archaeological evidence to argue that slaves produced most of the ceramics known as colonware, which tended to be found more in slave quarter sites than anywhere else on South Carolina plantations.
In the Northeast, by contrast, enslaved Africans were more thinly distributed among the slaveholding population. It was not uncommon for a Northern family to own a single slave and therefore require no additional housing. Distinct housing built for slaves was a rare sight. Without a comparable artifactual pattern of social and spatial segregation to that of the South, the archaeology of slave life in the Northeast faces significant challenges. To help overcome this limitation, we review here some historical literature, which offers precedents for investigating the sleeping and working arrangements of Northeastern slave households. Historians typically describe the pattern of slave housing in the Northeast as follows: ‚ÄúMost slaves ‚Ä¶ lived on small to medium sized farms with perhaps one other slave. Scattered throughout Long Island and often isolated from one another, Long Island slaves usually lived in close proximity to their masters.‚ÄĚ Richard Shannon Moss similarly concludes that ‚Äúmost enslaved Africans and slaveholders shared the same living space ‚Ä¶‚ÄĚ Historians largely agree that¬†since most¬†slaves¬†in the Northeast¬†were owned individually or in small groups on separate farms, they likely lived more as subordinate members of their owners‚Äô families than as members of a larger enslaved community.
Historians James and Lois Horton have called this arrangement ‚Äúfamily slavery‚ÄĚ such that ‚Äúmasters and their family ate, slept, and socialized in a common residence with their few, often no more than one or two, slaves. [For example,] Quasho, Pompey and his wife Phyllis and Prince, the household cook shared two rooms above the kitchen in the LaBaron home in Plymouth, Massachusetts.‚ÄĚ Historian Graham Russell Hodges prefers the term ‚Äúkitchen family‚ÄĚ to emphasize the spatial distinction between where masters and slaves would most likely be found. Hodges writes: ‚Äúmasters considered bondspeople part of their kitchen family, people with whom they worked and lived comfortably ‚Ä¶ Servants lived in the dark and airless upper stories or, as oral traditional has it, in barns and outbuildings.‚ÄĚ After researching the record of slavery in Rhode Island, archaeologist James Garman agrees and elaborates:
On farms with enslaved African Americans, the kitchen ell was the domain of the enslaved. From a more enlightened twentieth-century perspective, this can be read as an indictment of EuroAmerican racism, which relegated African Americans to the service portion of the house. However, it can also be read as a ceding of space on the part of EuroAmericans or, more strongly, as a wresting of a measure of semiprivate space by the enslaved.
By the early eighteenth century, it had become the common practice in the Northeast for slaves to occupy separate, usually ancillary spaces within their master‚Äôs household. The problem for archaeologists is that these spaces often were within the standing frame of the house itself (such as attics, cellars, kitchens, or back rooms). As few houses from the time of slavery still stand and those that do have mostly undergone significant alternations, the likelihood of recovering material culture from such spaces is low. Notwithstanding this limitation, Long Island nevertheless boasts one very fortunate discovery of such an attic space that would have been used by slaves and that remained largely unaltered since slaves lived at the site. This is at the Henrik I. Lott House site in the Marine Park section of Brooklyn. When archaeologists from Brooklyn College excavated the site, they discovered a trapdoor in a closet ceiling that granted access to a passage between two small, windowless rooms under the eaves. Archaeologists discovered candle drippings, a cloth pouch, oyster shell and corncobs under the floor boards in these rooms. Despite this great find, most archaeologists are not so fortunate to find spaces unaltered since they were last used by enslaved African people.
Further research on slave housing in the North does offer some hope if we look to draw distinctions between different types of slave owners. In her foundational 1943 paper, ‚ÄúSlavery on Long Island,‚ÄĚ Anne Hartell wrote that ‚ÄúThe slaves usually slept in the attics on the smaller farms. A more pretentious household might have had quarters for the slaves among its outbuildings.‚ÄĚ In his 1941 article about slave life in New York, Edwin Olson came to a similar conclusion about slave housing diversity:
There was no uniform method of housing slaves in New York. With but a pair of Negroes to provide for, the average slaveowner usually found sufficient room for them in the attic or basement of his home. Infrequently an addition was built to the house for the purpose. Sometimes the attic was used as a quarters for a large number of slaves, as in the case of the Phillipse Negroes in Yonkers. In Albany County it was quite common for slaveowners to provide their slaves with separate establishments, cabin-like structures, located to the rear of the family house.
Regarding Long Island, Lynda Day writes that ‚Äúslaves occasionally occupied their own small dwellings on the property of their owners, but most slept over kitchens or in the northwest corner bedroom of the main house.‚ÄĚ Day includes a 1944 Historic American Buildings Survey photograph of the ‚ÄúServant‚Äôs Quarters of Caleb Smith II in Commack‚ÄĚ as an illustration of her first point. Grania Marcus also mentions this example and includes this 1962 description: ‚ÄúThe rustic ‚ÄėOld Slave House‚Äô itself probably dates back to the late 1700s. It stood southwest of the Caleb Smith domicile along the edge of the highway. It thereby was almost a hundred feet closer to Jericho Turnpike than was the home of their slaveowner.‚ÄĚ Caleb Smith owned four slaves in 1790 and may have owned more in earlier years.
The pattern seems to be that wealthier slave-owning families were more often known to have constructed separate houses for their slaves. To date, these separate constructions amount in most cases to a single dwelling house for slaves per farm or household. These instances are not common, but they seem to follow the pattern of slaveholding found in the Northeast: while the majority of slaveowners in the rural Northeast owned fewer than four slaves, there was typically also a stratum of very wealthy, or according to Hartell, ‚Äúpretentious,‚ÄĚ whites who owned ten or more. These were the sorts of families that would have built slave quarters, if for no other reason than to provide housing for what was most likely the majority of their household population.
Graham Russell Hodges expands on this line of reasoning to also suggest that there was a change in the pattern of slave housing through time associated with the wealth of the slaveholder:
During the late colonial period, any closeness in living conditions between owners and masters disappeared. Bergen County, New Jersey offers examples of three ranks. Accounts of the architecture of wealthier households reveal physical separation of master and slave. Blacks were exiled either to the wing or, more often, a separate building in the rear of the farm. Real estate advertisements mention ‚ÄėNegro-Houses.‚Äô Separate ‚ÄėNegro Kitchens‚Äô combined housing and cooking area ‚Ä¶ Among the middling sorts, separation occurred within the home. In the house of Abraham Demarre of Closter in Bergen County, for example, two doors led from the main house in the slaves‚Äô wing. Cut into each door was a small square spy-light through which the master could survey his slaves. In poorer Bergen homes, older parts of the house were used for a combination kitchen and slave quarters. Slaves lived over ovens, in separate huts and barns, in crude tents, and even in caves.
We see here a clear pattern of difference in slave housing created by the wealth and status of the slave owner.
This observation is especially useful in this paper as the Martin family at Rock Hall can be firmly situated in the upper echelon of the status hierarchy of slave owners on Long Island. It is also notable that the prominent examples of separate structures for slaves on Long Island sites are solely associated with wealthy white owners of larger numbers of slaves. For example, in addition to the wealthy Caleb Smith II family, the homes of two generations of the Lloyd family of Lloyd Neck near Huntington are depicted in an 1811 map with separate small structures with chimneys. An archaeological excavation of the likely slave quarter at Joseph Lloyd Manor recovered a dense concentration of household artifacts. While still under analysis, this project will certainly add useful information to what we currently know about slave housing and slave life on Long Island. In addition, the estate of William Floyd near Mastic is known to have had at least one separate structure to house their slaves. The Sylvester family, which built a provisioning plantation on Shelter Island, owned as many as twenty-four slaves in the seventeenth century. Based on a concentration of brick debris in one area, archaeological research ‚Äúhas suggested that there may have been a separate living area in close proximity to the main house, but this has not yet been confirmed.‚ÄĚ Finally, a photograph of the Van Brunt ‚Äúslave quarter‚ÄĚ structure is depicted in Grania Marcus‚Äô Discovering the African American Experience in Suffolk County volume. Jacob Van Brunt, a wealthy farmer in Setauket, is listed as owning three slaves in the 1790 census and likely owned more in earlier years.
A particularly useful comparison for the study of Rock Hall is the Isaac Royall house in Medford, Massachusetts, which was well researched by Alexandra Chan. Like the Martins of Long Island, the Royall family came to the Northeast from Antigua. Born in Maine, Isaac Royall travelled to Antigua as a young adult to make a living in his family‚Äôs slave trade business. He eventually settled on the island of Antigua and became the owner of a sugar plantation in 1712. Over the subsequent years, Royall purchased provisioning farms in Massachusetts and Rhode Island. Following a slave revolt conspiracy on Antigua in 1736, he decided to move permanently to his farm in Medford, Massachusetts. In addition to his family he brought with him a number of enslaved Africans. Records also show that he already sent other slaves ahead to work on the provisioning farms. A 1754 census lists twelve slaves at the Isaac Royall household. Moreover, 1739 and 1752 probate inventories affiliate Isaac Royall with as many as thirty-nine slaves who would have lived and labored on his multiple farms in New England.
Given Royall’s large number of slaves, it is not surprising to discover that his ¬†Medord estate consisted of the master‚Äôs family home, a separate slave quarter, and several other outbuildings. The slave quarter is described as having been built in three phases: ‚Äúa very small one-story brick kitchen, replaced with a two-story brick kitchen around 1739, and, several years later, a wood frame addition.‚ÄĚ The final form of the slave quarter structure contained several rooms and likely served as an ‚Äúout kitchen‚ÄĚ as well as slave housing. While records are not overtly clear that an ‚Äúout kitchen‚ÄĚ was ever a designated slave quarter as well, archaeological excavations adjacent to the structure revealed a distinct pattern of artifacts suggesting that the area was a site not only of everyday labor but of some distinctive recreational activities and religious practices.
One telling piece of evidence was the concentration of various artifacts on the side of the slave quarter structure farthest from the master‚Äôs house. This location would have allowed the slave quarter structure to serve as a physical barrier preventing direct surveillance of its occupants’ activities. Chan writes: ‚Äúthe trash-strewn work yards emphasized the separation of master and slave, work and leisure, clean and unclean. To the enslaved on the estate ‚Ä¶ work yards might have connoted family ties and time away from the master: long smokes on the stoop out back, and impromptu games of checkers or marbles, some quiet time to sew, make beads, or tell stories of Africa, and freedom to the children who never knew it.‚ÄĚ Notably, these suggested activities derived from direct archaeological evidence excavated from the work yard, which included handmade and marked tobacco pipes, marbles and altered ceramic gaming pieces, handmade beads, sewing equipment, a repaired ceramic plate, a black teacup with heavy use-wear consistent with making medicines, and even a worked stone ‚Äúarrowhead amulet‚ÄĚ that may reflect an African practice of wearing symbolic objects associated with the ancestors.
Considering contemporaneous archaeological evidence, it seems likely that Rock Hall would follow some of these class-correlated patterns of separate slave housing in the Northeast. The Martins were a very wealthy family with a long history of association with Atlantic slavery, and the construction of Rock Hall in 1768 coincides with Hodges‚Äô assessment of the late colonial period as a time when class distinctions among slave owners became more pronounced. We also know that Rock Hall was home to a large number of slaves during the Martins’ tenure. Moreover, having been raised in a planter family in Antigua, Josiah Martin was directly familiar with Caribbean plantation layouts and modes of labor management that emphasized segregation (personal and architectural) between masters and slaves. Based on this information, we suggest that the Rock Hall estate likely included a separate structure to house the Martins’ slaves. Like at the Royall estate, this house may have also served additional functions. In particular, the slave quarter at Rock Hall was probably also the principal kitchen for the estate. As we discuss below, we have located what we believe to be some remains of this structure. Following the lead of James Garman, we also present further evidence that enslaved Africans used this structure to practice activities drawn from African and West Indian traditions, thus illustrating their retention of a distinct African-derived culture.
Archaeological Methodology and Initial Discoveries
Initial field-testing for evidence of a kitchen outbuilding in Rock Hall‚Äôs west yard involved a shovel-test survey for archaeological deposits over the whole west yard site (Figure 4). This technique assesses the distribution of artifacts and deposits across a large area and helps to locate the presence of significant buried architectural features, such as foundations and chimney bases. After several promising locations were identified, three archaeological units were excavated, revealing two possible building support piers that showed signs of at least two repair episodes. In addition, a suite of eighteenth- and nineteenth-century artifacts was also recovered (see Figure 5) including ceramic, bottle, and cut bone fragments, which indicate that the west yard was a site of typical domestic activities associated with food preparation and household service. While the materials seemed to suggest that the west yard was the site of the historic kitchen, there was no evidence of a substantial hearth structure to be expected in the house of an elite family of this period. The location of the kitchen outbuilding thus remained a mystery.
Figure 5: Photo and profile view of Unit 1 showing a concentration of stone and brick that may have served as a pier support for the kitchen outbuilding at Rock Hall. Photograph from Silver and McLean 1995/2005, Drawing by Ross T. Rava.
West Cellar Passage
Theorizing that the kitchen outbuilding most likely had been in the west yard, Rock Hall researchers concluded that a cellar entrance of the same side of the manor house probably served as a passageway between the two buildings. After cooking the food in the kitchen outbuilding, the Martins‚Äô slaves would have brought prepared food into the house through the west cellar entrance and into the basement, where there was likely a warming kitchen used for final plating and service. This cellar entrance was abandoned around 1830 and sealed around 1880 by the Hewletts (Figure 6). Given that slavery in New York State was abolished in 1827, closing such a ‚Äúslave entrance‚ÄĚ to the manor house may have been an attempt by the new owners to erase signs of slavery at their new home. Furthermore, according to an 1836 Fosdick survey of the site, the Hewletts built a hyphen to connect a large building in the west yard to the manor house, perhaps to provide their servants, now wageworkers, with more direct access to the manor house dining room (Figure 7).
Prompted by the Museum‚Äôs plan to reopen the sealed cellar door as a means of entry to the basement, an archaeological study of the adjacent ground was conducted in 2003. This project involved the excavation of four units (see units 4-7 in Figure 12). This area was largely protected from disturbance since it lay beneath a two-story porch constructed by the Hewletts in the 1880s (see Figure 8), which was removed in the 1950s during the restoration of the manor house by the Town of Hempstead Parks and Recreation Department. Unfortunately, in 1982 a five-foot deep ‚ÄúFrench drain‚ÄĚ was installed around the entire perimeter of the manor house to control rain seepage into the basement, destroying most of the archaeological evidence of the west cellar stairwell. Archaeological excavation beyond the reach of the French drain revealed a partial demolition deposit in the site of the original cellar staircase (Figure 9). This excavation showed that the cellar entrance ceased to be used after 1830, when the hyphen structure was placed above it, and that it was sealed up with sandstone blocks from the stairwell walls.
Figure 8: Circa 1930 view of west side of Rock Hall manor house showing west porch built by the Hewletts in the 1880s. Courtesy Rock Hall Museum.
Figure 9: South wall profile view of Unit 7- Rock Hall manor house west cellar entrance mitigation. The image shows the demolition debris found in the site of the cellar stairway. Undisturbed archaeological deposits are illustrated on the right below the ground surface. Drawing by Ross T. Rava.
In this undisturbed area, the archaeology showed that a thick layer of seashell fill was used to level the ground surface for the hyphen and that the intact eighteenth-century pebble ground surface, with small fragmented domestic artifacts, was untouched (Figure 10). In addition to ceramic, bottle glass, cut bone, charcoal, sea shell, and clay pipe stem fragments, archaeologists also recovered from this buried surface an unusual quantity of eighteenth-century artifacts, including brass round-head straight pins, bent wrought iron nails, and lead bird shot adjacent to the remnants of the original cellar stairway (Figure 11). Initially, the museum curators and archaeologists considered these seemingly random constellations of artifacts to be accidental‚ÄĒstraight pins spilled from a sewing basket as someone exited the cellar; scattered lead shot dropped by a hunter walking through the yard or into the cellar; bent iron nails, which dated from the late-eighteenth century, perhaps castoffs from the remodeling of the manor house roof around 1800. As discussed below, however, other archaeological findings from the west yard have since prompted us to reconsider the significance of these artifacts and their relation to the Rock Hall slaves.
Figure 10: Close up of Rock Hall Manor‚Äôs eighteenth-century pebble yard surface exposed during the excavation adjacent to the west side cellar entrance. Note the thick layer of shells in the far wall of the excavation unit that was laid down over this surface, fortunately protecting it from any subsequent disturbances. Photograph by Ross T. Rava.
Figure 11: Artifacts found during mitigation of the 18th century yard surface adjacent to the west cellar stairway. Clockwise: lead bird shot; cut and hand-wrought iron nails; brass straight pins; small red sandstone chips. Photographs and Image prepared Ross T. Rava.
With the cellar entrance mitigation completed, the attention of the museum staff returned to locating the possible site of a kitchen outbuilding in the west yard. Having re-opened the cellar passage, their long-term goal was to lead visitors from the basement, up the re-built cellar staircase, and into a re-constructed colonial kitchen. However, they needed more evidence to justify this interpretation. So, in 2005, archaeologists were asked to return to the site to see if they could locate the kitchen outbuilding foundation or the remnants of a large cooking hearth in the west yard. This time a 19 ¬Ĺ‚ÄĚ long steel rod was used to probe through the ground along north to south and east to west transects in order to locate possible features (Figure 12). One area in particular was identified for further investigation. Subsequent excavation of this area in 2005 revealed a feature composed of a dense deposit of burned ash and sea shells mixed with crushed shell and brick fragments that took the form of a small fireplace base (Figure 13).
This surprising find called for additional research. Could a hearth be made of burnt shells and ash? Based on standard models of large colonial hearths, which were typically massive brick structures, the feature was initially rejected as a fireplace base. Moreover, at just over 5 feet long, the feature base was too small to have been the cooking hearth for a colonial kitchen outbuilding serving the large Martin household. Nonetheless, the project‚Äôs principal investigator, Dr. Annette Silver of TAS Archaeological Services, conferred with Dr. A. Reginald Murphy, Director of Heritage and Archaeology, Antigua and Barbuda National Parks Authority, about the find. Dr. Murphy noted that the deposit resembled the remains of colonial-era tabby fireplace bases found throughout Antigua. He also reported that tabby was a common construction material in the West Indies. We have also learned that tabby was used in house construction in coastal Georgia and Florida.
Tabby is a concrete made by a very labor-intensive process of collecting, cleaning, crushing and burning sea shells to extract lime. It appears to derive from a North African ‚ÄúMoorish‚ÄĚ technology imported to the Americas by Spanish settlers (who probably called it tapia, which means ‚Äėmud wall‚Äô) and later adopted by English colonists. Large shell middens left behind by Native Americans were the most likely source of the shells. It is notable that, while shell middens are widely distributed in North America, tabby concrete was only used in areas associated with large numbers of enslaved Africans.
Given the likelihood that at least some of the slaves at Rock Hall came with the Martins from Antigua, this may mean that the tabby construction method discovered at Rock Hall would have been familiar to the Martins and their slaves. In light of this information, the tabby structure is now accepted as a likely fireplace presenting material evidence of the artisanal influence that the black slave community had on the architecture at Rock Hall. From this starting point we have begun to build a revised interpretation of ¬†Rock Hall that considers in more detail the presence of the enslaved African laborers who lived there. In particular, we highlight three factors that we think support the idea that the tabby fireplace was part of the kitchen outbuilding that served the Martin household and that likely did double duty as slave housing.
First, the location of this fireplace establishes that Fosdick‚Äôs 1817 survey (Figure 3) which had been used as the basis for the kitchen search, was not an accurate representation of the Martin‚Äôs Rock Hall estate. At the suggestion of the archaeologists, attention shifted to consider Fosdick‚Äôs 1836 survey (Figure 7). This survey was previously dismissed by most Rock Hall historians as inaccurate in its depiction of the sizes and locations of the west yard outbuildings. However, upon careful examination and measurement, the location of the tabby fireplace feature base was found to coincide with the middle of the north wall of the large west yard building shown on the 1836 survey, exactly where a fireplace traditionally would have been constructed. In addition, subsequent archaeological studies verified the accuracy of Fosdick‚Äôs 1836 survey as it correctly predicted the location of another outbuilding in the west yard.
Second, Josiah Martin‚Äôs property deed dated September 21, 1767, states that at the time of the sale another farmhouse with many outbuildings stood on the 600 acres that he bought. The large structure shown on the survey in the west yard may indicate that earlier farmhouse and may predate the construction of Rock Hall‚Äôs manor house. We think that at least some of the cooking for the manor house from 1767 until 1880 (when the east-wing kitchen was built) took place in this structure rather than in a separate kitchen outbuilding specially built by the Martins. It is very likely that building piers discovered in the 1995 archaeological testing were among the supports of this large structure, which would have stood at ground level and thus left no substantial foundation to be discovered by archaeologists in later years. Moreover, any remnants of a larger, more permanent cooking hearth that may have been located on the south wall of this structure would have been destroyed by the construction of a brick pathway in the twentieth century (see Figure 14).
Figure 14: Morris Fosdick‚Äôs 1836 survey overlaid on Rock Hall Museum‚Äôs current survey. Note the projected footprint in the west yard of the large kitchen outbuilding that perhaps predates the manor house. The inset photographs show the actual location of the other building features, including the tabby fireplace base as well as features discovered in the far west yard, that align with the 1836 survey. Prepared by Ross T. Rava for the Hofstra University Center for Public Archaeology.
Third, documentary evidence of the Rock Hall slaves, coupled with historical understandings of other wealthy Northeastern planter households, further suggests that the archaeological remains in the west yard once served as slave quarters. Presuming that the large structure in the west yard was a pre-existing farmhouse, it is very possible that Josiah Martin would have used it to shelter his extensive household during the construction of the new manor house. Yet, after completing his manor house, why would he have kept a large old building so near his new beautiful home? We believe the answer relates to the actual size of Josiah Martin‚Äôs household and their West Indian plantation heritage. In 1790 his household included nine family members and, as mentioned above, seventeen enslaved Africans. We think therefore that at least some of the enslaved community continued to reside in the large farmhouse after the Martins moved into the manor house. We also propose that this farmhouse was converted into a kitchen as well as slave quarters and that the slaves, themselves, constructed and perhaps even designed the tabby fireplace feature that would have stood along the north wall. The small size of the hearth may reflect a subdivision of the larger structure into separate spaces, perhaps to create a living quarters for the domestic staff apart from the kitchen area. The tabby fireplace would have been a handmade renovation using imported West Indian technologies by and, perhaps for the benefit of, the slaves themselves.
The Martin Family‚Äôs Relationships with Enslaved Africans
Another factor may be at work here as well. As a child, Josiah Martin lost his father, Major Samuel Martin, during a bloody slave uprising on their family‚Äôs Antigua sugar plantation in 1701. In addition, Josiah and his family barely escaped being killed in Antigua in 1736 during a failed slave plot to massacre all of the island‚Äôs white inhabitants. Later, Josiah‚Äôs brother, Samuel Martin, became well-known in the British Atlantic world for his oft-reprinted 1754 treatise, entitled An Essay Upon Plantership, which called for improving the living and working conditions of slaves as a way to reduce the likelihood of revolt and thus to ensure the survival of slavery as an acceptable and profitable practice.
Given Josiah Martin‚Äôs family history, it is reasonable to believe that he agreed with his brother and that in his move to New York, he brought only his most trusted, useful, and seasoned slaves to set up a new home in a distant place. While we do not know of any records that can confirm this statement, it is also not out of the question that some of these slaves may have been working for the Martins for generations. As far we can tell, they faithfully cared for his family and performed the domestic and farming chores without resistance, even if this fidelity was manufactured by a master concerned with cultivating an image of peaceful household relations. We can read aspects of these relations on the landscape as they are embodied by the use of a separate quarter for the slaves, which afforded a modicum of privacy, personal space, and opportunities to build for themselves a tabby fireplace within a standing structure. Whether this separation was symbolic of a benign master-slave relationship or if Martin saw this separation as advisable from a security standpoint in the face of the potential for slave rebellion, we cannot say.
Moving to the following generation, however, some additional evidence provides insight into the relations that these spaces may have reflected. After Josiah Martin‚Äôs death in 1778, his son, Dr. Samuel Martin, returned to the manor house to become head of the household. Dr. Martin never married, however, and when he died in 1802 he provided in his will that five children ‚Äúfrom my late mulatto woman Molly‚ÄĚ be freed and their education paid for. It seems quite likely that Dr. Martin had a paternal interest in these children. Again, we cannot say with any specificity what the relationship between Dr. Martin and Molly was like, but a degree of intimacy associated with fatherhood seems evident. We have to also question who Molly‚Äôs parents were, as a biracial woman. Could it be that the Martins, like many slave owners, were in constant regular negotiations with their slaves over the dynamics of control and autonomy that defined their personal, physical, and intimate relationships? Returning now to reconsider the findings from previous archaeological work in the west yard at Rock Hall we think provides one way to develop an answer.
Enslaved African Spaces at Rock Hall
Given the everyday household intimacy of master and slave at Rock Hall in the late-eighteenth century, we presume that the archaeological record associated with the west yard, where the lives of master and slave regularly intersected, will necessarily contain evidence of how these relations were experienced. The tabby fireplace and its indication of a separate residence for captive laborers is one example. Further evidence comes from a re-analysis of the material collected from the eighteenth-century yard surface adjacent to the west cellar entrance. This was the location where the straight pins, bent iron nails and lead shot were found. Upon further examination, we discovered that these materials were in fact contained within an area marked by two half buried small red sandstones that resemble stone ax heads (Figures 11 and 15). We propose that these stones created the boundaries of a symbolically charged liminal spot at the top of the staircase and that the pins, nails, and shot reflect not mere accidents or spills but spiritual practices undertaken to protect this entrance. We believe that the staircase was marked because of its great significance to those whose lives were in large part defined by the passage from kitchen to basement at Rock Hall. Placing the artifacts in this spot strongly suggests the Martins‚Äô captive laborers were using their own religious practices to protect their entrance into the manor house.
Figure 15: Rock Hall Manor west cellar entrance mitigation site showing position of red sandstone chips that contained the assemblage of bird shot, straight pins, and iron nails. Photograph by Ross T. Rava.
Supporting evidence of enslaved people marking and protecting liminal spaces in such ways derives from a series of recent archaeological discoveries across the United States illustrating the retention of African spiritual practices. At many slave quarter and post-emancipation tenant sites, as well as within the homes of slave owners, caches or bundles of materials such as pins, nails, beads, quartz crystals, ceramic shards, pierced coins, buttons, animal bones, sea shells, crustacean shells, and smoothed stones have been found. These caches are often located by doorways, chimneys, and windows, which seems to reflect a process of marking and protecting openings. Drawing especially from evidence collected in the WPA slave narratives, researchers found many references to the use of caches of items in the practice of Voudoun (Hoodoo), which is based on beliefs in powerful ancestral spirits and that some persons are blessed with a power to put these spirits to work for harm or protection.
The Archaeology in Annapolis project team of the University of Maryland recently discovered a similar set of artifacts as those found at Rock Hall bundled together in the gutter in front of a colonial-era house. The find is described as: ‚Äúabout the size of a football, the compacted clay and sand bundle originally sat in clear public view stationed in front of a house. X-rays show the object served as a container holding hundreds of pieces of lead shot, pins and nails. A prehistoric stone axe extends upward from the top of the bundle.‚ÄĚ These are essentially the same materials used in the cache that we argue was purposefully placed at Rock Hall. Notably, the Annapolis object has been interpreted by African cultural experts, with one stating that ‚ÄúIf Yoruba in origin, the bundle would likely represent the image of Eshu Elegba, the god of chance, confusion and unpredictability, the god of the crossroads. The axe blade could replace the comb in other representations of the Eshu, and it is also indicative of the power of Shango, the god of thunder and the lightning bolt.‚ÄĚ It is certainly fascinating to think that the materials recovered at Rock Hall at the top of staircase may be associated with a god of the crossroads.
Recent research in the northeast, where the separation of slaves and masters was less pronounced, has also begun to identify comparable patterns of cached or bundled materials at sites associated with enslaved and free African Americans. A prominent example is a collection of broken glass vessels on the surface of a burial ground at Parting Ways, located in Plymouth, Massachusetts, a small community of formerly enslaved African Americans. Closer to Rock Hall, a cache of beads, pottery fragments, bent nails, pins, brass buttons, and chipped prehistoric stone tools was discovered under the floor boards at top of the basement stairs at the Lattings Hundred site in Huntington Village on Long Island. In this instance, family documents record the presence there of a well-regarded slave woman who was likely also a spirit practitioner and who may have used this cache to protect the intersection of the worlds of the slaves (basement) and master (upper floors) at this house in a way that would have been familiar to the African slaves living at Rock Hall.
The purpose of known caches documented by anthropologists and other scholars was to invoke ancestral spirits to either protect or harm people entering or exiting a doorway. Such talismans and the rituals associated with them ensured that, at vital cultural intersections, proper respect was paid to the spirits whose power gave direction and meaning to everyday activities and acts of creation (such as cooking and domestic work) that also defined master-slave relations. The presence of these materials at Rock Hall may reflect one important result of the Martins‚Äô strategy for the preservation of slavery: to allow a social and physical space within the world of the master for the slaves to invoke their own spiritual beliefs and thus to create a space for themselves within, rather than against, slavery. Whether or not the Martins actually pursued such a strategy, the separate slave quarters in the west yard certainly opened possibilities for the enslaved to maintain their spiritual beliefs and subtly challenge their subordination.
So far the search for the site of a kitchen outbuilding at Rock Hall, if it ever existed, has not been successful. However, this search has yielded fresh insights into the lives of the slaves who worked and lived at Rock Hall in the eighteenth century. On-going research and re-analysis of previous findings at Rock Hall has identified another possible West African religious cache of buried items just north of the cellar entrance, and possibly another in a shovel test pit located opposite the cellar entrance and inside of the proposed location of the west wall of the large structure shown on the 1836 survey. It now appears possible that these religious caches and the tabby fireplace base may form the north, east, and west axis of a West African religious spiritual cosmogram, such as those documented at sites elsewhere in the African diaspora. This evidence suggests that, despite living in close proximity to their master and his family, the enslaved Africans at Rock Hall found ways of carving out their own physical and spiritual spaces. Further research and investigation is needed, however, to tell us even more about the complex social relationships at this site and to confirm our propositions about the conciliatory relationship between the Martins and their slaves.
 We are grateful to the Friends of Rock Hall for their consistent support of this fieldwork and analysis over the last decade. We are also very grateful to Jennifer Anderson for her careful editing and many useful suggestions that have improved this paper. Of course, we are entirely responsible for its content.
 Rock Hall is now located in Nassau County, a new county created from eastern Queens after the consolidation of western sections of Queens with New York City in 1898.
 Hibbard, Shirley G. Rock Hall: A Narrative History (Dover Publications Garden City, NY, 1997), p.3.
 Hibbard 1997, 73; Wilke, Douglas A., The Historic Structure Survey of Rock Hall (1980), on file at Rock Hall Museum.
 Hibbard 1997, p. 13.
 Wilke, 1980.
 LaRoche, Cheryl and Michael Blakey, ‚ÄúSeizing Intellectual Power: The Dialogue At The New York African Burial Ground,‚ÄĚ Historical Archaeology 31(3):84-106, 1997; Perry, Warren, Jean Howson, and Barbara A. Bianco, eds., New York African Burial Ground Archaeology Final Report, Volume 1. Prepared by Howard University, Washington, D.C. for The United States General Services Administration Northeastern and Caribbean Region, 2006.
 e.g., Berlin, Ira and Leslie M. Harris, Slavery in New York (New York Historical Society/The New Press, 2005); Lepore, Jill, New York Burning: Liberty, Slavery, and Conspiracy in Eighteenth-Century Manhattan (Vintage, 2006); White, Shane, Stories of Freedom in Black New York (Harvard, 2002); Hodges, Graham Russell, Root and Branch: African Americans in New York and East Jersey, 1613-1863 (North Carolina, 1999); Hodges, Graham Russell, Slavery and Freedom in the Rural North: African Americans in Monmouth County, New Jersey, 1665-1865 (Madison House/Rowman and Littlefield, 1997); Piersen, William D. Black Yankees: The Development of an Afro-American Subculture in New England (Massachusetts, 1988); Horton, James O. and Lois E. Horton, In Hope of Liberty: Culture, Community and Protest Among Northern Free Blacks, 1700-1860 (Oxford, 1997); Gerzinga, Gretchen¬† H. Mr. and Mrs. Prince: How an Extraordinary Eighteenth-Century Family Moved Out of Slavery and into Legend (Amistad/Harper Collins, 2008); Fitts, Robert K. ‚ÄúThe Landscapes of¬† Northern Bondage,‚ÄĚ Historical Archaeology 30(2): 54-73, 1996; and Garman, James C. ‚ÄúRethinking ‚ÄėResistant Accommodation‚Äô: Toward and Archaeology of African-American Lives in Southern New England,‚ÄĚ International Journal of Historical Archaeology 2(2): 133-160.
 Otto, John S., Canon‚Äôs Point Plantation, 1794-1860: Living Conditions and Status Patterns in the Old South (Academic, 1984).
 Ferguson, Leland, Uncommon Ground: Archaeology and Early African America, 1650-1800 (Smithsonian, 1992).
 Marcus, Grania, Discovering the African American Experience in Suffolk County, 1620-1860 (Society for the Preservation of Long Island Antiquities/Amerion House, 1995), p. 72.
 Moss, Richard Shannon, Slavery on Long Island: A Study in Local Institutional and Early African-American Communal Life (Garland, 1993), p. 83.
 Pierssen 1988.
 Horton and Horton 1997, p. 16.
 Hodges 1997, p. 53.
 Garman 1998, p. 152.
 Hartell, Anne, ‚ÄúSlavery on Long Island,‚ÄĚ Nassau County Historical Journal 6(2): 55-71, 1943, p. 59.
 Olson, Edwin, ‚ÄúSocial Aspects of Slave Life in New York,‚ÄĚ The Journal of Negro History 26(1): 66-77, 1941, p. 68.
 Day, Lynda R. Making a Way to Freedom: A History of African Americans on Long Island (Empire State, 1997), p. 31.
 Marcus 1995, p. 111.
 Hodges 1999, p. 112-114.
 Matthews, Christopher N. and Jenna W. Coplin, ‚ÄúThe Archaeology of Captivity and Freedom at Joseph Lloyd Manor,‚ÄĚ African Diaspora Archaeology Newsletter, December 2007. http://www.diaspora.uiuc.edu/news1207/news1207.html#3
 McGovern, Allison Manfra, Personal Communication, February 2012.
 Mrozowski, Stephen, Personal Communication, February 2012. He continued to say: ‚ÄúBased on what we know at this point it seems just as likely that early slave residents were in out-buildings associated with the main core of the plantation.‚ÄĚ See also Stephen A. Mrozowski, The Historical Archaeology of Sylvester Manor: Special Issue, Northeast Historical Archaeology, 36(1), 2007; Katherine Hayes. Before Race: Europeans, Africans, and Indians at Long Island’s Sylvester Manor Plantation, 1651-1884 (New York University Press, New York, 2013).
 Marcus 1995, p. 72
 Chan, Alexandra A. Slavery in the Age of Reason: Archaeology at a New England Farm (Tennessee¬† 2007), pp. 47-57.
 Chan 2007, p. 82.
 Chan 2007, p. 41, Fig 19.
 Chan 2007, p. 42.
 Chan 2007, p. 226.
 Chan 2007, p. 138-189.
 Silver, Annette and McLean, Jo-Ann. Stage 1B Archaeological Survey, West Yard, Rock Hall, Lawrence, NY , 1995/2005.
 Hibbard 1997, p. 9.
 TAS Archaeological Services. Stage 2 Archaeological Mitigation West Yard Cellar Entrance Site, Rock Hall, Lawrence, NY, 2004.
 TAS Archaeological Services 2004.
 TAS Archaeological Services. Kitchen Site Probe Survey, West Yard, Rock Hall, Lawrence, NY, 2005/2006. TAS Archaeological Services. Stage 2 West Yard Archaeological Survey, Rock Hall, Lawrence, NY, 2005.
 Adams, Dannis, Tabby: The Oyster Shell Concrete of the Lowcountry (Beaufort County Library, 2007).
http://www.beaufortcountylibrary.org/htdocs-sirsi/tabby.htm#Tabby (last accessed July 6, 2012).
 This project overlaid the 1836 survey onto a current survey of the Rock Hall museum property. Subsequent testing and excavations in 2008, 2010, and 2011 of outbuilding sites in the current west yard area confirmed the near perfect accuracy of the 1836 survey. See Rava, Ross T. Stage 2 Archaeological Subsurface Investigation, Possible Outbuilding Site, The Far West Yard, Rock Hall, Lawrence, NY, 2008; Center for Public Archaeology, Hofstra University; Continuation of the Archaeological Stage 2 Investigation, the Far West Yard, Rock Hall, Lawrence, NY, 2010; and Center for Public Archaeology, Hofstra University, Continuation of the Archaeological Investigation, the Far West Yard, Rock Hall, Lawrence, NY, 2011.
 Hibbard 1997, p. 5.
 Hibbard 1997, p. 3; Rebovich, Samantha Landscape, Labor, and Practice: Slavery and Freedom at Greencastle Estate, Antigua (Ph.D. Dissertation in Anthropology, Syracuse University, 2010).
 Will of Dr. Samuel Martin, Liber of Wills B, p 324, Queens County Record Office. Hibbard 1997, p. 14.
 Ferguson 1992, Brown, Kenneth L. ‚ÄúMaterial Culture and Community Structure: The Slave and Tenant Community at Levi Jordan‚Äôs Plantation, 1848-1892‚ÄĚ in Working Toward Freedom: Slave Society and Domestic Economy in the American South, Larry E. Hudson, Jr., ed., (University of Rochester Press, 1994), pp. 95-118; Fennel, Christopher. Crossroads and Cosmologies: Diasporas and Ethnogenesis in the New World (University Press of Florida, 2007); Leone, Mark P., An Archaeology of Liberty in an American Capital: Excavations in Annapolis (University of California Press, 2005); Matthews, Christopher N., ¬†An Archaeology of American Capitalism (University Press of Florida, 2010); Russell, Aaron.‚ÄúMaterial Culture and African American Spirituality at the Hermitage.‚ÄĚ Historical Archaeology 31(2):63-80, 1997; Samford, Patricia M., Subfloor Pits and the Archaeology of Slavery in Colonial Virginia. (University of Alabama Press, 2007); Woodruff, Janet, Gerald F. Sawyer, and Warren R. Perry, ‚ÄúHow Archaeology Exposes the Nature of African Captivity and Freedom in Eighteenth and Nineteenth Century Connecticut,‚ÄĚ Connecticut History 46(2): 155-183, 2007.
Leone, Mark P., ‚ÄúA Unique, Early Artifact of African Worship Uncovered in Annapolis,‚ÄĚ African Diaspora Archaeology Newsletter, December 2008. http://www.diaspora.uiuc.edu/news1208/news1208-1.pdf
 Deetz, James F., In Small Things Forgotten: Archaeology and Early American Life (Anchor Books, 1996); ¬†Metcalf, Reginald, Personal Communication, July 2008.