Phase II and Phase III Archaeological Database and Inventory
18CH308 Two Friends
Diagnostic Artifact Data
Features
Summary
The Two Friends Site (18CH308) is an apparent domestic site and work yard dating to the middle of the 18th century. The site is located near the town of Benedict in Charles County, Maryland, along the ancient floodplain of the Patuxent River. The floodplain is quite wide near the site. It is bordered on the west by steep bluffs rising above the plain. The floodplain itself is nearly level and swampy drainages run through the southern portion. The soils are primarily Mattapex fine sandy loams. The area around the site is state and federally owned for wildlife conservation. It consists of a mix of leased agricultural fields and woodlands, with individual fields being separated by wide hedgerows.
The site was first documented in 1988 as part of a Phase I survey conducted throughout the aforementioned wildlife areas in an effort to document archeological sites for future conservation and/or study. The study was conducted mostly by surface walkover, even when surface visibility was poor. Only a few shovel tests were dug, and none were dug on sites that were visible on the surface. Specific information on the sites discovered is limited as the MHT library copy of the survey report is currently missing. However, according to MHT site files completed by the researcher conducting this project, Two Friends consisted of a moderately dense scatter of artifacts including 18th century glass, ceramics, oyster shell, and brick concentrated along the eastern edge of a hedgerow. Diagnostic artifacts included olive amber and olive green bottle glass, white salt-glazed stoneware, Westerwald, slip-decorated earthenware, and Buckleyware. Exact quantities are not known, so each type is only counted as a single occurrence in the tally of artifact tables above.
A second Phase I survey was conducted at the site in 2001 as part of a Maryland State Highway Administration construction project. About half of the site is located within a 10 acre field, where the Maryland SHA was preparing to construct a wetland to help mitigate the effects on other wetlands of construction of the MD 5 Hughesville bypass. Section 106 of the National Historic Preservation Act of 1966 required archeological study in the area to be impacted by the new wetland. The strategy chosen for the survey of the 10 acre field was plowing, followed by controlled surface collection based on a grid of 4.57 meter (15 ft) squares. After plowing, the site was immediately visible as a dense scatter of oyster shell on the field surface, and a noticeable darkening of the soil. A transit was used to grid the field into 45.7 meter (150 ft) squares, and the smaller divisions were measured out with tapes. The grid baseline was laid out very close to north-south in a line following the western field edge (hedgerow). The artifacts found in each 4.57 meter square were bagged separately, and labeled with the grid location of the southwest corner of the square in which they were found.
Surface visibility in most of the field was very good (greater than 80%) at the time of the survey. Visibility was reduced in some parts of the field by heaped vegetation, but this was generally in the lower-lying wetter areas, where sites were considered much less likely. Even in the worst of these wetter areas, visibility always exceeded 30%. Shovel tests were dug in some areas, but none were excavated within the boundaries of Site 18CH308. Recent objects found during the 2001 survey, such as plastic and aluminum pull tabs, were noted and discarded, as were coal, clinker, and similar waste products. Shell was retained, but once in the laboratory it was weighed without being counted or carefully cleaned. Thus, the artifact tallies above do not include full counts of these objects. About 30 eighteenth century artifacts were recovered during the surface survey within the boundaries of 18CH308. They consisted of olive wine bottle glass, a white clay pipestem, brick fragments, and sherds of tin-glazed earthenware, white salt-glazed stoneware, British brown stoneware, Staffordshire combed slipware, and coarse red earthenware. These items are included in the count of all Phase I and II artifacts given below and in the site-level tally in the tables above. These items are typical of an ordinary domestic occupation of the period and could have come from the home of a tenant, a small owner-farmer, or even from a slave quarter.
To help clear up questions regarding the occupation and to determine site significance, a Phase II examination was immediately undertaken. In order to place the archeological finds at Site 18CH308 in a more detailed context, documentary research was conducted on the history of the property and on some of its possible 18th century occupants. This archival research revealed that “Hilltop Farm”, the modern property that includes the site and about 400 acres surrounding it, descends from a property of some 670 acres assembled in the 1850s by John D. Bowling. In his will, Bowling states that he bought the tract from a Henry Sothoron. The Sothorons owned a number of large tracts in the area. Site 18CH308 is located at the eastern end of Bowling’s “Hilltop Farm” property, and the eastern-most Sothoron tract that was incorporated into “Hilltop Farm” was known as “Two Friends”. “Two Friends” was patented in 1671 by a Robert Gates and Justinian Dennis, and at that time, it was said to contain 525 acres. By 1696, the property was in the possession of Joshua Marshall, because all 525 acres were purchased from Marshall by John Sothoron.
The precise chain of title for the Two Friends tract is difficult to reconstruct between the period of John Sothoron’s ownership and the tract’s purchase by Bowling, but it seems likely that it remained in the control of the Sothoron family for the entire duration. The inventory of John Sothoron, taken around the time of his death in 1707, gives some idea of the kind of farming undertaken on this land and the style of life to which the Sothorons aspired. John’s inventory lists only 3 slaves, but it seems likely that he owned others who had been distributed among his heirs through his will. Certainly, the archival record reveals that future generations of Sothorons owned more. John’s inventory lists 55 head of cattle, 26 sheep, 30 hogs, and 10 horses, representing the diverse livestock typical of Maryland plantations in this period. No plow or other gear for grain farming is listed, so John Sothoron probably grew mainly tobacco and Indian corn. The household goods include 66 ounces of silver, 4 feather beds, linens, and other comforts. Particularly interesting are possible signs of industrial activity. Five-hundred and seventeen pounds of “hammered iron and cast”, and brass, may represent the work of a smith, and “26 Tand hides” and a “parcel of leather and tools” suggest that tanning and possibly shoemaking or harness making were being done on the plantation as well.
John Sothoron’s will, dated 1707, left to his wife, Mary Sothoron, part of his “dwelling plantation”, “that tract of land known as Two Friends”. After her death, this tract was to fall to his youngest son, Benjamin Sothoron. The will of this Benjamin Sothoron, dated 1745, gave to his brother, Samuel Sothoron, 125 acres at the southeast end of “my land called Two Friends”. Based on the remains encountered at the site (see below), it is unlikely that 18CH308 represents the remains of the main Two Friends plantation, which would date earlier and would presumably be a much larger site. It seems more likely that the Two Friends Site was established during the ownership of Benjamin Sothoron, before 1745. Benjamin owned several slaves (eight are mentioned in his will), so it is possible that the site represents a slave quarter. The archival record suggests that the family had a propensity to establish houses for younger relatives on parts of their land, and the site could also represent the home of one of these junior Sothorons. It could also be a tenancy.
The excavation of nineteen 91.44 X 91.44 cm (3 X 3 ft) test units was the primary field method used. The test units were dug according to natural strata, which in every case consisted of a plowzone and subplowzone layers. In most cases the subplowzone soils were sterile subsoil, and only one approximately 10 cm level of the subsoil was excavated, but in other cases features were found. Features were left intact or were tested, according to the judgment of the researchers. A metal detector was also used to search the area around the site for concentrations of nails that might represent structures. Targets were flagged, and shovel tests were dug at each flag. Shovel test pits (STPs) were approximately 45.72 cm (18 in) in diameter, and were excavated into the sterile subsoil. All soil was screened through hardware cloth. A brief inspection was also made of the field to the west of the hedgerow (outside the area of impact for the SHA project) to better define the boundaries of the site. The test units were concentrated in the core of the site, where most of the artifacts had been found during the Phase I surface collection. One test unit was dug in the southeastern part of the site, well away from the others.
About 300 artifacts were recovered from the 19 Phase II test units, as well as 39 pieces of animal bone and close to a thousand fragments of oyster shell. An artifact-rich core area was evident in the north-central portion of the site. Higher artifact counts were observed in this area, as well as darkened soil and large amounts of oyster shell and ash. The core was interpreted to be the probable remains of an 18th century trash midden. In Test Unit 1, within the core, a piece of this midden was found intact underneath the plowzone. The unplowed midden consisted of a shallow, irregular pit, containing dark soil, varying in depth from 6 to 24 cm below the plowzone. The artifacts appeared essentially identical to those in the plowzone. A large cow bone, however, was present in a deeper midden that was not excavated. This midden was designated Feature 1. Feature 1 was also identified in a test unit 3.05 meters to the east and in two adjacent units approximately 7 meters to the southwest. A second feature (Feature 2) was identified in these same adjacent units. This was a small posthole and mold, thought to be part of a fence. However, about 25 handwrought nails were found in these two units, so it was possible that a building stood nearby.
Near the end of the Phase II fieldwork, a metal detector was used to check the fields surrounding Site 18CH308 for metal artifacts. The main purpose of this survey was to search for concentrations of nails that might indicate the locations of outbuildings. This survey was made along regular transects, using the same grid that had been used for the Phase I surface collection, and provided nearly 100% coverage of the southern half of the wetland project area. The only concentration of targets identified was near the test units in the core of Site 18CH308, northeast of Test Unit 1. Three nails were found in this area. A few bits of metal were also found in the southern part of the site.
The artifacts recovered during both Phase I and Phase II work at 18CH308 included 82 architectural objects, 1 possible furniture item (a brass finial), 1,173 kitchen-related artifacts, 7 tobacco-related artifacts (white clay pipestems), 1 arms artifact (a gunflint), and 16 miscellaneous objects (11 unidentified pieces of glass, and 5 pieces of metal). The architectural artifacts include 55 pieces of brick, 9 handwrought nails, and 18 unidentified nails. The kitchen-related artifacts were 59 ceramic sherds (6 Staffordshire comb slipware, 7 tin-glazed Delftware, 1 creamware, 1 Chinese porcelain, 1 red-bodied slipware, 15 coarse redware, 6 British brown stoneware, 17 white-salt-glazed stoneware, and 5 gray stoneware), 133 container glass fragments, at least 942 oyster shell pieces, and 39 animal bones.
The site as a whole was determined to be roughly 3.5 acres in size, but most of the artifacts were found in a core area measuring about .65 acres. This core area was found to include intact subplowzone features that contained artifacts. Thus, the site was considered capable of answering research questions related to rural life in 18th century Maryland. Site 18CH308 was determined eligible for listing on the NRHP. Although only part of Site 18CH308 was within the project area, construction of the wetland would have significant impacts on the site and about 40% of the core area would be destroyed by construction. Since the site could not be avoided by redesign, Phase III mitigation by data recovery was undertaken prior to the wetland construction.
Researchers returned to the site in 2002 to conduct data recovery. Work began with additional sampling of the plowzone. Forty-nine additional 91.44 X 91.44 cm units were dug within the threatened portion of the site, roughly one test unit per every 9 m2. The 68 units excavated during Phase II and Phase III work therefore composed a sample of about 9% of the plowzone. No additional features were identified during this stage of Phase III sampling, but midden identified during Phase II was noted in several units. Artifacts from the plowzone sampling include at least 449 architectural artifacts, a clothing item (brass button), at least 928 kitchen-related artifacts, at least 4 arms artifacts (3 gunflints and fragments, 1 gunflint core), at least 46 tobacco-related artifacts (white clay pipe fragments), and at least 68 miscellaneous objects (45 unidentified glass fragments, and 23 metal pieces). The architectural items include 303 pieces of brick, 1 fragment of mortar/plaster, 56 handwrought nails, 88 unidentified nails, and a handwrought spike. The kitchen assemblage included 324 ceramic sherds (3 Midlands-mottled, 38 tin-glazed Delftware, 11 creamware, 23 yellow-bodied slipwares, 1 buff-bodied slipware, 143 coarse redwares, 8 other earthenwares, 3 Chinese porcelain, 26 brown salt-glazed stoneware, 48 white salt-glazed stoneware, and 20 gray salt-glazed stoneware), 398 container glass fragments, 3 tumbler fragments, a glass handle from a piece of unidentified tableware, a knife, and 201 animal bones.
The artifacts encountered in the plowzone suggested an occupation period for the site of perhaps 1740 to 1780. Very few “small finds” were encountered in the plowzone (i.e. coins, pins, thimbles, etc.), which are regularly dropped wherever people live, or fall through the floorboards of their houses. Their absence from the Two Friends site collection suggested (although hardly proved) that the excavated part of the site might not have been a domestic area. One particularly interesting feature of the plowzone collection was the complete absence of recognizable window glass. By 1750, owner occupied farmhouses generally had glass windows, and window glass has been found on many tenant sites as well. While the excavations were generally limited to the portion of the site that was actually threatened by wetland construction, three units were dug outside the project area within the western hedgerow line. Artifact counts in these units were much higher than the rest of the site. The hedgerow units also yielded much higher counts of architectural artifacts (nails and brick) than the others, suggesting that a structure may have stood in the hedgerow area.
After the plowzone sampling, a backhoe with a smooth bucket was used to remove the bulk of the plowzone from the threatened portion of the site. Less than 10 cm of plowzone was left in place in this area, which had to be removed with a shovel later. The stripped area measured 18.29 X 33.53 m (60 X 110 ft) and primarily uncovered the extent of the midden in the site core. No fence lines were observed, nor was any sign of outbuildings seen. A few postholes and molds were noted in the midden area. After the plowzone had been entirely removed, field technicians began to see boundaries within the broad smear of midden. Ultimately, 18 individual pit features could be defined and constituted the “midden”. Since some of these features actually consisted of more than one overlapping pit, it appeared that the Colonial residents dug at least 22 pits in this excavation area. In addition, 14 other disturbances were assigned feature numbers, but were ultimately determined to be natural in origin. Features were recorded in plan view and profiled where possible, and flotation samples were taken from the feature fill. Details of the ethnobotanical findings are provided in the linked Ethnobotanical Profiles document (see below).
Artifacts encountered within the various features excavated as a part of the Phase III study include 95 activity items, 581 architectural objects, 1 clothing artifact (a brass button), 2 possible furniture objects (a brass knob and a small iron strap hinge), 3,550 kitchen-related artifacts, 157 tobacco-related objects, 7 arms items, and 115 miscellaneous objects. The activity items were a possible shackle, an iron sickle blade, and 93 horse bones. Architectural remains consisted of 412 pieces of brick, 120 handwrought nails, 47 unidentified nails, and 2 handwrought spikes. The kitchen assemblage includes 371 ceramic sherds, 1,073 glass objects (10 tumbler pieces, 20 pieces of table glass, & 1,043 bottle/container glass shards), a piece of iron kettle, a knife blade, and 2,104 faunal remains (bones and shell). The ceramic sherds include 113 tin-glazed Delftware, 22 yellow-bodied slipware, 169 redware, 1 pearlware, 1 Chinese export porcelain, 1 other porcelain, a coarse earthenware fragment, 16 white salt-glazed stoneware, 1 Westerwald stoneware, 4 brown salt-glazed stoneware, 7 gray salt-glazed stoneware, 25 white stoneware, 4 gray stoneware, and 6 brown stoneware sherds. The faunal remains are quite varied and include several species not typically considered food remains. Here, horse bones have been considered “activity items” and feline bones have been counted as “miscellaneous objects”, while all other faunal remains have been considered kitchen remains. The actual use of the latter bone grouping is ambiguous because these species can be food remains, but the context suggests they may be remnants of specialized activities taking place at 18CH308. The “kitchen” fauna include 593 cow bones, 307 pig bones, 21 goat bones, 59 sheep bones, 2 rabbit bones, 420 large mammal bones, 373 medium mammal bones, 4 small mammal bones, 6 unidentified mammal bones, 9 goose bones, 1 duck bone, 2 chicken bones, 3 unidentified small bird bones, 12 unidentified bird bones, 100 snapping turtle bones, 72 box turtle bones, 29 other turtle bones, 1 frog bone, 30 longnosed gar bones/scales, 8 unidentified fish bones, 14 unidentified animal bones, and at least 38 pieces of oyster shell. Tobacco-related artifacts were 72 clay pipe fragments and 85 snuff bottle fragments. Arms objects include 4 gunflint cores and 3 gunflints. Miscellaneous objects were 47 pieces of unidentified glass, 22 unidentified pieces of iron, 22 other pieces of metal, and 24 cat bones.
The midden deposits at the Two Friends Site represent a specialized refuse area from the period 1740 to 1780 rather than an area used for disposal of the full range of trash from a domestic site. Comparison of the midden to the excavation units dug in the western portion of the site shows that the midden contained much less ceramic material and more glass. The animal bone encountered included horn cores and other bones usually thrown away at a butchering site rather than discarded with the kitchen trash. Alternatively, the presence of the horn cores could suggest that tanning took place nearby (something that is attested to archivally). The remains of gunflint manufacture were also present, so the area may have simply served as a generalized workplace or work yard. On the other hand, the presence of Delftware bowls, oyster shell, and certain types of butchered bone shows that some of what ended up in the pits was household trash. The midden area probably represents a work yard near a residence (probably just to the west) where some household trash was disposed of along with the waste generated on the spot.
If the site was a residence, who lived there? The site is obviously not the big house of the Two Friends plantation, which would have been a much larger and grander site with a longer occupation period. The evidence of the bones suggests that slaughtering for the plantation house residents may have been done at the site, which implies that the residents were not independent tenant farmers. The collection of faunal material, however, does not strongly resemble those from other slave occupied sites, particularly in the preponderance of beef over pork and the small amounts of wild game. Perhaps the head of household was a favored servant, such as an overseer. Faunal assemblages at known overseer’s house sites do resemble assemblages from the “big house” in some respects and assemblages from slave cabins in others. More work would be required to make this connection, however, and understanding of the site could benefit from further examination of the western part of the site which was not impacted by the wetland construction.
Only part of the Two Friends site has been excavated. Over half of the site should have survived construction of the wetland by SHA and is preserved on state-owned land managed for wildlife conservation. Thus, the site should be considered a significant resource for better understanding rural lifeways in Charles County during the mid-late 18th century. Indeed, the results of study thus-far point to more questions than answers and additional work is warranted should an opportunity present itself.
|
