Using American Coarse Earthenware Types as a Tool for Site Interpretation and Comparison at Monticello

November 20, 2024

Paper presented at the Southeastern Archaeological Conference in Williamsburg, VA, November 13 - 16

A growing body of scholarship over the last twenty years has provided archaeologists with the analytical tools to identify different coarse earthenware types and study variation in site assemblages (Magid & Means, 2003; Bloch, 2015; Bloch & Heindl, 2015; Bloch & Bollwerk, 2024). My paper is a case study showing how standardized methods and terms developed by the DAACS (Digital Archaeological Archive of Comparative Slavery) project were used to analyze a distinctive group of coarse earthenware sherds from Site 30. The site is a Revolutionary-period quarter site for enslaved agricultural workers at Monticello. Certain physical characteristics distinguished them from other coarse earthenwares in the assemblage and led to the following research questions:

1.       Were the Site 30 sherds imported or American produced based on their physical attributes? Was there enough evidence to assign a regional production zone?

2.       Was the unusual surface appearance on some of the sherds evidence of an intentional applied decoration or unintended consequence of a glaze defect or uneven firing in the kiln?  

3.       Were these sherds unique to Site 30 or were similar examples found at other sites at Monticello or in the greater Chesapeake region? If they did occur at other sites, were they recovered from contexts with similar or different chronologies? The identification of similar sherds at other sites, especially sites with different chronologies, would suggest that they were not an anomaly but rather evidence of ongoing production by a particular potter or pottery workshop.

4.       How significant were coarse earthenwares in the overall ceramic assemblage of Site 30, and how does that compare with other assemblages at Monticello?  What can we learn about coarse earthenware use across the plantation landscape?

In summary, this paper explains the methods used to identify a group of unique coarse earthenware sherds from Site 30 and how this research led to a broader analysis comparing assemblages from other Monticello sites.

Background

Site 30 was discovered during the Monticello Department of Archaeology’s Plantation Archaeological Survey (Figure 1). The site is located 1/3 of a mile east of Thomas Jefferson’s house and is one of 40 undocumented sites discovered as a result of our survey work.  Artifacts collected from the survey date the site to the last quarter of the 18^th century. This date range coincides with what archaeologists call the “tobacco period” at Monticello when tobacco was the primary commercial crop that drove the plantation economy and economic and social relationships on the plantation. Jefferson’s decision in the 1790s to shift from tobacco to wheat cultivation in as part of a larger diversification effort marks the second phase in Monticello’s history which we refer to as the “wheat period.”  The occupation periods for most of the historic sites at Monticello tend to fall within these two chronological periods (Neiman 2010; Plantation Archaeological Survey; Crops at Monticello).

Figure 1. Results of the Plantation Survey project at the Monticello Home Farm, showing identified sites dating to Jefferson's ownership of the property

Archaeologists returned to Site 30 in 2022 to conduct a full-scale excavation of the site. Over the last three seasons, we have uncovered over 25,000 artifacts (Figure 2).  Of the 3,800 non-brick artifacts, 771 are ceramics. Since this paper focuses on American-made and imported coarse earthenware household vessels, native American, colonoware, and flowerpot fragments are not included in this analysis. The crew also has uncovered several features, including a subfloor pit and one large, unexcavated feature to the south of the sub-floor pit (Figure 3). Our current interpretation is that the site had one and possibly two more quarters for enslaved laborers who worked in nearby tobacco fields (Neiman, et al 2023).

Figure 2. Distribution Maps of recovered historic artifacts (left) and nails (right) from Site 30.

Figure 3. Identified features from Site 30. Left: location of features (light brown) across the site; Top-right: Overview of the subfloor pit; Bottom-right: Overview of the currently unidentified feature.

Methods

The ceramic assemblage from Site 30 contains a variety of different coarse earthenwares, including eleven sherds with several distinctive traits that caught our attention and prompted a deeper investigation into where they may have been produced and whether similar vessels had been found on other Monticello sites (see below for gallery of coarse earthenware images). These sherds and all coarse earthenwares from Monticello sites are cataloged using standardized methods and terminology first developed by the DAACS project in 2004 (www.daacs.org). DAACS records physical characteristics such as paste color, types of inclusions, inclusion density, decoration and surface treatment, vessel shape, and form for each individual sherd (Figure 4). DAACS developed a coarse earthenware typology using this attribute-based approach, and information from Dr. Lindsay Bloch’s ground-breaking elemental studies (Figures 5 and 6). Since 2015, Monticello analysts have used this typology to help record variation in coarse earthenware assemblages and learn more about production and consumption patterns at different sites (Galle 2018; Bloch 2015).

Figure 4. Screenshots of the DAACS cataloging platform for ceramics and coarse earthenwares. Manuals can be viewed at daacs.org. 

Figure 5. Examples of imported coarse earthenware types. Images courtesy of Dr. Lindsay Bloch and daacs.org.

Figure 6. Examples of American coarse earthenware types. Images courtesy of Dr. Lindsay Bloch, daacs.org, and chipstone.org

The Site 30 sherds were analyzed using the DAACS methods and descriptions. The physical characteristics of the sherds were more typical of American coarse earthenware types rather than British or European-manufactured.  I attempted to take my analysis a step further by trying to determine which regional production zone the Site 30 sherds might have come from.  The red color and density of the paste, shown in Figure 7, and small number of hematite and quartz inclusions were traits more typical of the coarse earthenwares produced in Western Virginia and the greater Philadelphia, Alexandria, and Baltimore area rather than the coastal Viriginia (Bloch 2016:1-4, 12-14 ; Barka 2004; Comstock 1994:10-18, 49-50 ).  See Table 1 below for paste attributes associated with different redware types. The acronym PHAB is used in DAACS to refer to sherds produced in this Mid-Atlantic, urban region.

Table 1. Comparison of ceramic attributes between the sherds recovered from Site 30 and the three American redware types identified by Dr. Lindsay Bloch

Figure 7. Paste shots of Site 30 slipped redware

While I was confident that these sherds came from one of two regional zones, I did not think there was not enough evidence to choose one. The sherds had a translucent, lead glaze on the interior and exterior surfaces commonly found on both local ware types.

Figure 8. Profiles of two vessels of possibly slipped Redware from Site 30.

We were fortunate to have relatively large sherds from different parts of vessels, which in some cases could be mended together. This allowed us to identify at least two hollow vessels with similar dimensions (Figure 8). A reconstruction of a complete vessel using rim and base diameters as well as the curvature of body suggested these vessels were some type of bowl similar to the types of bowls produced by potters in the Philadelphia, Alexandria, and Baltimore areas (Figure 9; Magid et al 2003; Beaudry et al 1983). We cannot assume, however that this feature means that the bowls were produced in that region. We know that potters who initially worked in the PHAB area migrated south to the Shenandoah Valley of Virginia where they set up shops in the mid-18^th century. What we can say is that the Site 30 bowls were made in the PHAB tradition (Comstock 1994: 4-14, 69-76, 83-86, 197).

Figure 9. Examples of Slip-Decorated Coarse Earthenware Bowls Produced in Philadelphia. Images sourced from chipstone.org

The unique exterior and interior surface appearance of the sherds was unusual and not typical of the type of decoration and/or surface treatment we generally have seen on Western VA and PHAB vessels found at Monticello. Several rim sherds had a lighter “band” around the exterior rim (Figure 10). A more diffuse patterning of buff-colored, green, and orange colors could be seen primarily along the interior proximal rim.   This patterning was also seen on the interior base of one of the vessels, where it appeared thicker than on some of the body sherds. This raised the question of whether the multiple colors were evidence of some type of slip decoration or simply a defect in the glaze. These traits were different from the more typical trailed or marbleized slip decorations found on American coarse earthenwares at Monticello.

Figure 10. Examples of Site 30 Redware will bands of possible slip just below the interior and exterior rim.

Since the physical characteristics of the sherds did not provide enough evidence to support assigning them a specific ware type, I cataloged them under the more general term “American Redware.”  I also interpreted the lighter-colored areas on the sherds as evidence of an applied slip decoration.

The uniqueness of the sherds also prompted me to ask several of my colleagues to do an independent assessment of them.  They all agreed that the sherds were American redwares but that there was not enough evidence to assign them to a specific production area. There was not a consensus, however, on what caused the multi-colored patterning on the interior surfaces.  Several analysts agreed with me that the lighter colored areas were evidence of a thinly applied slip.  Dr. Lindsay Bloch, a leading authority on coarse earthenware technology and elemental analysis, disagreed with this assessment and offered several other possible explanations. She reasoned that the different color patterning on the sherds resulted from a problem with the glaze mixture or an event in the kiln referred to as “blooming.”  Blooming is a term used by potters to describe the uneven oxidation of the clay body and/or glaze in the kiln. She believed that the unusual surface appearance and the very dense paste of the sherds was not intentional but rather the result of a problem in the kiln. Bloch concluded that the sherds with this particular defect most likely came from a single lot of vessels fired at the same time (Personal communication with Lindsay Bloch, Feb. 1, 2024; Comstock 1994: 50-54, 60-64).

If Dr. Bloch’s hypothesis was correct, this would suggest that the Site 30 sherds were an anomaly resulting from a single production episode and would not likely to show up in other site assemblages. One way to test this theory was to see if sherds with similar attributes had been found at other Monticello sites.  

Figure 11. Examples of American redware from other sites at Monticello. a) Home Farm Quarter Site 8; b) West Kitchen Yard/Dry Well/ MRS 1; c) Building o; d) Home Farm Quarter Site 6; e) Building D/j (Blacksmith/Nailery); f) Building o

I ran a query in the DAACS database to see if I could find coarse earthenwares with the same attributes as the Site 30 sherds and I found seventeen similar sherds at five other Monticello sites (Figure 11). Interestingly, many of the sherds with the distinctive patterning had been cataloged as being slipped. Two of the sites, Site 6 and Site 8, are quarter sites located not far from Site 30 on the south slope of Monticello Mountain (Figure 12; Bon-Harper 2006; Neiman 2018).  The three other sites are located along an area on top of the mountain that Jefferson referred to as Mulberry Row (Figure 13).  Mulberry Row refers to both the road and a series of workshops and dwellings for enslaved and white workers. These sites are called Building o, Building D/j or Blacksmith/Nailery, and West Kitchen Yard/Dry Well/MRS 1 (Hill 2003; Neiman 2010; McVey 2011; Bates et al 2013).  The names refer to the historic structures that Jefferson had built on those sites. 

Figure 12. Location of Home Farm Quarter sites containing similar American redware to Site 30

Figure 13. Location of Mulberry Row sites containing similar American redware to Site 30.

The presence of the unique coarse earthenware sherds on other sites at Monticello was the first piece of evidence challenging Dr. Bloch’s hypothesis (Figure 14).  I felt that this argument would be strengthened if I could show that the sherds found at the other sites also came from contexts or projects with different chronologies than Site 30. Variation in chronology would be further evidence that these wares were produced at different times.  To test this theory I used both context and project MCDs to identify the chronological period when the sherds were deposited. Although MCDs are not calendar dates, they do provide a midpoint date for a ceramic assemblage and can be used to determine whether a context or project falls within the two broad historic periods at Monticello: the tobacco period (1770-early 1790s) and the wheat period (late 1790s-1820).

Figure 15. Ware type frequencies across various sites at Monticello

My choice of which time-averaged date to assign to use depended on the site formation process for the various sites. Sites 6, 8, and 30 were plowzone sites with multiple cabins which makes context MCDs less reliable since the deposits do not reflect discrete, undisturbed strata. The contexts from these sites also tended to have small ceramic counts, which can skew a context MCD. Instead of relying on context MCDs for the quarter sites, I chose to use the project MCDs. The mountaintop sites, on the other hand, had stratified contexts with robust ceramic sample sizes which made their context MCDs a more reliable way of dating the various deposits and creating a site chronology. Both the project MCDs and Context MCDs provided a general date that could be placed within the broader plantation chronology.

Temporal evidence indicated that the unique Site 30 wares were found at sites with different occupation periods (Table 2).  For example, the project MCDs for Site 30 and Site 8 fell within the tobacco period of occupation while Site 6 dates to the wheat period. The relatively broad time span suggests that sherds found at these sites were not necessarily produced at the same time.  Context MCDs from the mountaintop sites also show variation in chronology. Context MCDs for Building o and Building D/j date the deposits to the early tobacco period at Monticello. The context MCDs from the West Kitchen Yard, however, span a broad period of time, offering further proof that the vessels likely came from different production events.

Table 2. Sherd count and chronology data from Monticello sites with American redware similar to Site 30.

Analysis of coarse earthenwares from the six sites proved beneficial in researching the Site 30 sherds and testing whether they were an anomaly. It also caused me to look more closely at the ceramic assemblages from all the sites to better understand the relative importance of coarse earthenwares within each of the assemblages.

Comparison of the six ceramic assemblages reveals some common patterns as well as interesting differences (Figure 15). Refined earthenwares composed the highest proportion of ware types on all six sites regardless of their location or their chronology.  The site assemblages begin to vary considerably, however, when comparing proportions of other ware types. At Site 8 and Site 30 assemblages, coarse earthenwares  are the next most common ware type followed by stonewares.  At Buildings D/j, o and the West Kitchen Yard, porcelain was found in greater proportion to either coarse earthenwares or stonewares.

There are several possible explanations why refined earthenwares are the dominant ware type at all the sites. One hypothesis is that enslaved laborers at Monticello had access to this ware type and preferred it to other ceramics. Another explanation is that these counts are somewhat inflated as a result of higher breakage and discard rates compared to the larger and more solid stonewares and coarse earthenwares. Furthermore, since most of refined earthenwares we find at Monticello are tablewares, it is more likely that a broken or chipped tableware would be considered unusable and discarded. Many utilitarian vessels were made from coarse earthenware or stoneware. A coarse earthenware or stoneware utilitarian vessel missing one or two pieces could still be functional and thus would not necessarily be discarded.

Coarse earthenwares along with stonewares composed important portions of the assemblages from Site 8 and Site 30 but appear in significantly less proportions on the other 4 sites. This pattern does not seem be temporally related since Building o and Building D/j also date to the earlier period. It may reflect social and economic differences within the enslaved community or unequal access to certain markets. Another trend that supports this theory is the higher percentage of porcelain in the Mulberry Row assemblages. Enslaved men and women who lived and worked along Mulberry Row and in the South Dependency encountered greater opportunities to perform tasks for Jefferson and members of his family in exchange for small cash payments which may account for the significant amount of porcelain, a relatively high-end ceramic, in those assemblages.

Conclusions and Future Research

The goal of my paper was to explain the methods I used to analyze distinctive coarse earthenware sherds from Site 30 and how this research led to a broader study of coarse earthenware assemblages from other Monticello sites. Based on this analysis, I concluded that the Site 30 sherds were American-made and had traits associated with the types of pottery made in both Western Virginia and the PHAB area.   By identifying similar sherds in other site assemblages, I was able to determine that the unique paste and surface appearance of the Site 30 sheds was not a one-time phenomenon or anomaly but more likely evidence that these distinctive wares came from a particular potter or pottery workshop. At the same time, I was unable to determine with any degree of certainty what caused the different colored patterning on some sherds. Further analysis of this trait might include microscopic analysis of the surface or LA-IC-MS studies.

The identification of similar vessels across the plantation landscape lent itself to a broader comparative analysis of the site assemblages. This research showed that refined earthenwares were the dominant ware type on all sites regardless of their chronology. A closer look at the ceramic assemblage data (e.g. sherd size) needs to be done before we can begin interpreting this information. Coarse earthenwares were the next largest group of ware types on two of the quarter site assemblages, while Chinese porcelain composed the second largest group of ceramics found in the MR assemblages. This data suggests a possible difference in household wealth.  Intersite comparison of coarse earthenware did not provide clear evidence of type variation since the data from certain assemblages was not analyzed using current coarse earthenware types and descriptions.  One possible solution to this problem would be to revisit legacy assemblages cataloged prior to 2015 and reanalyze the coarse earthenwares using current cataloging protocols. Documentary research exploring the various ways in which enslaved and free workers at Monticello acquired goods outside of the plantation would help interpret broader trends in the material record and further our understanding of the overall plantation economy.

Acknowledgements

Special thanks to all who assisted in this project: Corey Sattes, Elizabeth Bollwerk, Iris Puryear, Lindsay Bloch, Fraser Neiman, Derek Wheeler, Crystal O’Connor, Charlotte Devine, John Dukes, Susan Palazzo, Liz McCague, and Sean Devlin.

Bibliography

Barka, Norm. 2004. “Archaeology of a Colonial Pottery Factory: The Kilns of Ceramics of the “Poor Potter” of Yorktown.” In Ceramics in America. ed. Robert Hunter, pp.15-47.

Bates, Lynsey, Elizabeth Sawyer (née Clites), and Fraser Neiman. 2013. West Kitchen Yard/Dry Well/MRS1. Background. Digital Archaeological Archive of Comparative Slavery. https://www.daacs.org/sites/west-kitchen-yard-dry-well-mrs-1/#background  Accessed 11/13/2024.

Beaudry, Mary C., Henry Miller, Janet Long, and Fraser Neiman. 1983. “A Vessel Typology for Early Chesapeake Ceramics: The Potomac Typological System.”  Historical Archaeology, 17 (1): 18-43.

Bloch, Lindsay. 2015. “Made in America? Ceramics, Credit, and Exchange on Chesapeake Plantations.”  Unpub. Ph.D Dissertation, University of North Carolina.

Bloch, Lindsay. 2016. “Reanalysis of Coarse Earthenwares from Williamsburg-Area DAACS Sites.” Report submitted to the Digital Archaeological Archive of Comparative Slavery.

Bloch, Lindsay and Elizabath Bollwerk. 2024. “Coarse Earthenware at Flowerdew Hundred: Tools of Colonization in the 17^th Century Chesapeake.”  Paper presented at the Southeastern Archaeological Conference, Williamsburg, Virginia.

Bloch, Lindsay and Brenda Hornsby Heindl. 2015. “In One Kiln: Technologically Revising Research on Earthenware and Stoneware Production in Virginia.” Paper presented at the Council for Northeast Historical Archaeology Meeting, Fredericksburg, Virginia.

Bon-Harper, Sara. 2006. Site 8. Background. Digital Archaeological Archive of Comparative Slavery. https://www.daacs.org/sites/site-8/#background  Accessed 11/13/2024.

Comstock, H.E. 1994. The Pottery of the Shenandoah Valley Region. The Museum of Southern Decorative Arts.

Crops at Monticello. n.d. Monticello website. Thomas Jefferson Foundation, Inc. https://www.monticello.org/research-education/thomas-jefferson-encyclopedia/crops  Accessed 11/13/2024.

Diagnostic Artifacts in Maryland: Colonial Earthenware Ceramics. n.d. Jefferson Patterson Park and Museum https://apps.jefpat.maryland.gov/diagnostic/ColonialCeramics/colonial-earthenware.html  Accessed 11/13/2024.

Diagnostic Artifacts in Maryland: Colonial Earthenware Ceramics. 2015. “Colonial Earthenware Ceramics.” Diagnostic Artifacts of Colonial Maryland. https://apps.jefpat.maryland.gov/diagnostic/ColonialCeramics/colonial-earthenware.html  Accessed 11/13/2024.

Galle, Jillian et al. 2018. “DAACS Cataloging Manual: Ceramics.” chrome-extension://efaidnbmnnnibpcajpcglclefindmkaj/https://daacs.wpenginepowered.com/wp-content/uploads/2018/10/DAACSCeramicManual.pdf.

Hill, Martha. 2003. Building o. Background. Digital Archaeological Archive of Comparative Slavery. https://www.daacs.org/sites/building-o/#background  Accessed 11/13/2024.

McCartney, Martha and Beverly Straube. 2016. “George Thorpe’s Inventory of 1624: Virginia’s Earliest Known Appraisal, with photo essay Ceramics in Early Virginia.”  In Ceramics in America. ed. Robert Hunter, pp.2-32.

McVey, Shannon. 2011. Building D/j. Background. Digital Archaeological Archive of Comparative Slavery. https://www.daacs.org/sites/building-o/#background  Accessed 11/13/2024

Magid, Barbara H. and Bernard K. Means. 2003. “In the Philadelphia Style: The Pottery of Henry Piercy. In Ceramics in America. ed. Robert Hunter, pp.47-86.

Neiman, Fraser. 2010. Monticello. Background. The Digital Archaeological Archive of Comparative Slavery. https://www.daacs.org/plantations/monticello/  Accessed 11/13/2024

Neiman, Fraser. 2018. “The Mysteries at Site 6: Archaeological Clues about Slavery at Monticello.” Monticello Magazine (Winter), pp. 18-19.

Neiman, Fraser, Crystal O’Connor, Derek Wheeler, and Corey Sattes. 2023.“Archaeology at Monticello’s Site 30: Some questions, tentative answers, and more questions.” Monticello Livestream. Thomas Jefferson Foundation, Inc.   https://bit.ly/3TLvxY2 Accessed 11/13/2024.

Plantation Archaeological Survey. n.d. Monticello website. Thomas Jefferson Foundation, Inc. https://www.monticello.org/research-education/for-scholars/archaeology-daacs/current-research/plantation-survey   Accessed 11/13/2024.