October 28, 2026: Artifact Stories - Cowrie Shell

By Fraser Neiman and Corey Sattes

October 28, 2026

The Department of Archaeology receives a steady stream of inquiries from researchers and the public about our discoveries. We recently fielded a request for more information on a cowrie shell (Figure 1) found in the early-1980s excavations along Mulberry Row, the “Main Street” of houses and craft shops where many enslaved people lived and labored on the mountaintop. The request prompted us to have another look at this evocative artifact, its excavation context, and recent research into the ocean-spanning historical processes that brought it from its natural habitat in the tropical waters of the Indian Ocean to a subfloor pit in a slave dwelling at Monticello.

The Maldives to the Middle East to West Africa

Our story starts in the Indian Ocean, moves to West Africa via the Sahara, then to Europe and back to Africa, and finally to Virginia and Monticello. Money cowries (Monetaria moneta) and their close relative the ring cowrie (Monetaria annulus) are small sea snails that live in the shallow tropical waters of the Indian and Pacific Oceans (Heath 2016:20–21; Christie et al. 2019:480, 482). The money cowries that reached West Africa and are therefore relevant to our story came from the Maldives, a chain of islands southwest of India.

By the 12th century AD, the Maldives’ residents had developed and were exploiting a unique economic niche. They gathered cowries at low tide, buried them in the sand until the animals decayed, and then sold the cleaned shells by the basket to South Asian and Arab merchants in exchange for commodities like rice and cloth (Heath 2016:21; Haour and Moffett 2023:547).

South Asian merchants then funneled the shells into local economies where they were used as small change (Yang 2018:162). Arab merchants moved the shells to Red Sea and Persian Gulf ports and then overland to Cairo. From there, cowries were bulked with several other commodities—including salt, copper and brass, cloth, horses, beads, and paper—and transported south by caravan across the Sahara to West Africa (Yang 2018:163–164; Haour and Moffett 2023:548–549). In the 11th or 12th century, a caravan that had been headed south but was abandoned in the Mauritanian desert held more than 3,000 cowries packed alongside some 2,000 brass bars (Christie et al. 2019:480; Haour and Moffett 2023:549). In return, West African merchants sent gold and enslaved captives north. In the 1320s, Mansa Musa, the king of Mali, traveled to Cairo with a hundred camel-loads of gold and hundreds of slaves (Yang 2018:163). Once they were introduced into West Africa via the trans-Saharan trade, cowries were in common use as money in Mali and Gao by the 14th century (Yang 2018:165). It is likely that their early role as exotic, long-distance imports and later use within West Africa as a medium of exchange were ultimately responsible for their association with wealth and prestige and their roles in personal adornment and ritual in the region.

The European Connection

Starting in the early 16th century, European geographical expansion, fueled by advances in military and maritime technology, radically transformed West African trade mediated by cowries and linked it more closely with the trade in human beings. Once the Portuguese gained direct ocean access to Indian Ocean ports and South Asian goods, they packed cowries into their ships as ballast to help stabilize them on the trip back to Lisbon. In Lisbon they reshipped the shells, along with with European products, to West Africa, increasing the number of cowries imported into West Africa by orders of magnitude. On this leg, cowries served as ballast too. They took up no cargo space needed for valuable goods. However, unlike stones that that traditionally served this purpose, cowries were accepted as money on the African coast (Yang 2018:172–173). In 1522, a Portuguese vessel used cowries to buy captives at the mouth of the Niger (Haour and Moffett 2023:550).

In the 17th century, the Dutch and English East India Companies entered the trade. They purchased cowries in bulk in South Asian ports and shipped them to Amsterdam and London where the shells were unpacked, counted in hundreds, cleaned, and sorted by size, since smaller shells sold better in Africa. They were then advertised in newspapers and coffee houses and sold at auction to merchants outfitting African voyages (Yang 2018:174; Heath 2016:22). The markup at each stage was large. Cowries cost three to four times as much in Bengal as in the Maldives, and up to five times more again in West Africa (Yang 2018:174). In 1721, 12 pence bought a pound of cowries, about 400 shells, in Europe, and 30 pence at Whydah (Yang 2018:171). The French trader John Barbot, writing in the 1680s, traced the whole chain: from the Maldives to Indian ports, to Dutch and English ports, and then to Guinea for the purchase of slaves bound for the Americas (Yang 2018:177).

Cowries for Captives

By the 17th century, an enlarged “cowrie zone” had emerged that ran from the Volta River to the Bight of Benin and inland where cowries were everyday money. People exchanged them for food, work, fines, and taxes, as well as other people (Yang 2018:168–171, 179, 186; Heath 2016:22). Nearly all of West Africa’s major slave-exporting states lay within this zone (Yang 2018:171).

Demand for cowries from West Africans made cowries indispensable to slave traders. When the English slave ship Hannibal traded at Whydah in 1693–94, Captain Thomas Phillips reported that nearly half of any cargo had to be cowries or brass basins to make the other goods acceptable, and that about 100 pounds of cowries bought a man (Yang 2018:178–179). Over the 18th century, the English and Dutch alone shipped about 26 million pounds of cowries to West Africa, more than 10 billion shells (Yang 2018:175–176; Heath 2016:22). The price of a captive, measured in shells, rose from 40,000–50,000 cowries in the 1710s to 160,000–176,000 in the 1770s (Yang 2018:175–176). Cowries typically made up between a fifth and a third of the goods Europeans paid for slaves, and sometimes as much as half (Yang 2018:176–177).

Cowries in Virginia

Historians long assumed that the shells Europeans brought to Africa stayed there (Yang 2018:175). However, recent documentary research reveals that some unsold shells traveled with slave ships to the Americas. For example, Bristol shipping records reveal that cowries returned from Africa by way of Virginia and the West Indies between 1724 and 1734, including more than a ton on a single ship that had entered the York River in 1732. Archaeology demonstrates that some of those cowries made it ashore (Heath 2016:33–36). 

Barbara Heath recently surveyed the 353 cowries found on archaeological sites in Virginia. In her sample, 350 are from sites in the Coastal Plain and 252 are from in Yorktown, a key port that regularly received slave ships (Heath 2016:30–33). Excavations on the Yorktown property of Philip Lightfoot, a merchant heavily involved in importing captive Africans, yielded more than 150 cowries. Cowries also have been found on sites occupied by other Yorktown merchants who dealt in slaves, including Thomas Nelson and Cole Digges (Heath 2016:35). 

Archaeology offers clues to why some shells left West Africa. Heath argues they were rejected by African slave traders. Heath’s survey shows that 99% of the identified Yorktown cowries are M. annulus. She notes that documentary sources suggest West Africans preferred M. moneta until in the 19th century supply constraints forced them to accept M. annulus. A subsequent survey of cowries in West African archaeological assemblages from the 18th century and earlier strongly supports her hypothesis (Haour and Christie 2019). Nearly all the West African finds are M. moneta. It is not clear how the shells got from slave-ship hulls onto Yorktown merchants’ lots. However, the large number discarded there indicates that the shells had little or no monetary value for merchants on this side of the Atlantic. 

Data from Heath’s survey suggests that shells found on non-Yorktown sites, including sites in Williamsburg and rural sites across the Coastal Plain, also had their origins as rejects in the hulls of slave ships docked in Virginia ports. Over 90% of identified shells are M. annulus (Heath 2016: Table 1). How did these shells move out of Yorktown and similar ports? Heath suggests that enslaved people played a key role (Heath 2016:30). The numbers support her hypothesis. They reveal a roughly 9-fold increase in the odds of encountering a M. moneta shell on a site outside of Yorktown. While the numbers are small and 95% confidence intervals on the odds ratio are large (2.5 to 35), they point to a preference on the part of the people responsible for the movement of shells for M. moneta over M. annulus. The preference is a manifestation of the same tradition that led West African traders to favor M. moneta. 

More generally, cowries’ utility in economic transactions as small change and social transactions as signals among people familiar with the traditions of the “cowrie zone” are likely responsible for their spread out of port towns. Those traditions carry the centuries-long West African association of cowries with wealth and prestige, their consequent use in personal adornment and ritual, as well as their use as money. They were reinforced by the key role cowries played in the trans-Atlantic slave trade. In Heath’s Virginia sample, 26% of M. annulus and 48% of M. moneta cowries had their backs cut off to facilitate stringing on a cord or sewing to clothing, attesting to the importance of their signaling role. Although samples are too small to be statistically reliable, the figures also hint that people who came into possession of M. moneta were more likely to use it as a signal (odds ratio = 2.04, 95% confidence limits: .63, 6.66). 

The Monticello cowrie

The Monticello cowrie was excavated in 1983 at the Building t Site, on Mulberry Row, the street of slave houses and workshops adjacent to Jefferson’s mansion (Figure 2). Building “t” names the structure Jefferson described on his 1796 Mutual Assurance Declaration. It was one of three buildings constructed along Mulberry Row in the early 1790s to serve as living quarters for enslaved people. However, the cowrie came from an earlier structure that the three houses replaced, a building that Jefferson called “the Negro Quarter” in his plans for Mulberry Row dating to the 1770s. This was a barracks-style dwelling. The cowrie was excavated from one of four subfloor pits, small personal storage closets, that its enslaved resident had dug into the dirt floor of the Negro Quarter. Figure 2 is an archaeological plan of the Building t Site excavation. The four subfloor pits are labeled F01, F02, F03, and F04. The context within the subfloor pit (F02) from which the cowrie shell was recovered is highlighted in yellow.

The other artifacts that archaeologists recovered from Feature 2 help us date the period over which they and the cowrie were discarded, before they were entrained in the dirt with which the subfloor pit was filled. Table 1 provides known manufacturing dates. The small ceramic sample yields a mean ceramic date of 1776. This estimate, the absence of later ware types (e.g., pearlware), the dominance of hand-wrought nails, and the clay pipe stems support a 1760–1780 date range. The one anomaly is two cut nails, which date to the early 19th century. A careful look at the site plan reveals that the subfloor pit fill was intruded by a later posthole (F09), dug to accommodate a parking lot fence in the 1930s. Our best guess is that the excavators did not completely remove the fill of the posthole before they dug the subfloor pit fill, inadvertently mixing the two later nails with the older assemblage.

Table 1: Counts and calculated dates for Feature 02, Subfloor Pit. 

A mean ceramic date (MCD) is a weighted average based on the known manufacturing dates of ceramic ware types and the frequency of each type, giving greater weight to types that are more represented in the assemblage. A BLUE MCD gives an additional weighting to ware types with shorter manufacturing ranges (Bates et al. 2020:66). The terminus post quem (TPQ) of an artifact assemblage – literally the latest date after which it was deposited. The traditional estimate of the TPQ is the latest (or maximum) starting manufacturing date among all the artifacts comprising that artifact assemblage. Example: a coin dated 1932 has a TPQ of 1932 because it could not have been created or deposited prior to that year (Renfrew and Bahn 2019:140). The 90th percentile TPQ removes the top 10% of outliers.

Summary of Feature 02 Artifacts Relevant to Dating

• Creamware, 1762-1820, n=2 (Noël Hume 1969)
• Delft, 1600-1800, n=1 (Noël Hume 1969)
• Staffordshire Slipware, 1670-1795, n=1 (Noël Hume 1969)
• Whieldon-type Ware, 1740-1775, n=1 (Noël Hume 1969)
• Wrought nails, c. pre-1820s, n=31 (Nelson 1968; Wells 1998)
• Machine cut nails, c. 1820s-1890s, n=2 (Nelson 1968:10)
• Pipe stem, 4/64” bore diameter, 1750-1800, n=5 (Noël Hume 1969)
• Pipe stem, 5/64” bore diameter, 1710-1750, n=2 (Noël Hume 1969)

The comparative perspective afforded by the research we summarize above illuminates the Monticello cowrie. It is M. monetaria, the species more highly prized in West Africa and apparently by enslaved people in Virginia. Its back has been sliced off, a treatment more common for this species elsewhere in Virginia. Two grooves are visible on the shell surface made by the abrasion thread passing through the hole to attach the shell to a cloth surface. It was used as an adornment, not in financial transactions. While the precise intent behind its owner’s decision to display the shell is unknowable, it is likely that they thought there were observers in the immediate environment familiar enough with West African traditions to be able to decipher the signal being sent. 

It is also impossible to know exactly how the cowrie arrived at Monticello. The most likely route is the one mapped out by Heath: from shell stocks rejected by sellers in West Africa to the properties of slave traders in Virginia, from which they were dispersed in networks of African-descended people for whom they had economic and social utility. However, we cannot rule out its transport across the Atlantic in the possession of a newly enslaved African. 

While recent research affords a better understanding of the ocean-spanning processes operating over centuries that landed the cowrie at Monticello, it also raises questions. First, the Monticello cowrie is one of only three found in the Piedmont (Heath: Table 1). It is not clear why the network links across which cowries moved rarely extended into this region. Second, the three sites from which those three cowries were recovered are all in Charlottesville! The two others are Pavilion VI at UVA and the Catherine Foster Site, located in a free Black neighborhood just south of the UVA Lawn. This Charlottesville cluster is to some extent an artifact of intensive archaeological research at Monticello and UVA. However, equally intensive research programs at Poplar Forest and Montpelier have failed to produce cowries. So there is a real historical phenomenon to be explained here. Why did Charlottesville have denser or stronger links to the cowrie network?

Finally, the discard date of the Monticello cowrie (1770–1780; Heath 2016: Table 1) falls at the tail end of cowrie deposition in Virginia. Of the shells Heath could date, 78% come from contexts predating 1775 (Heath 2016:29–30). The likely cause is the shrinking supply of rejected cowries coming to Virginia ports, as the importation of African captives into Virginia dwindled after mid-century and effectively ended during the American Revolution (Heath 2016:30, 37–38). The puzzle is that the other two Charlottesville cowries, from the Foster Site and Pavilion VI at the University of Virginia, come from deposits dated to the mid- to late 19th century and mid-20th century, respectively (Heath 2016: Table 1). The cowries might be redeposited from earlier deposits. On the other hand, they might have been curated across generations. Closer study of the two assemblages with cowries, along the lines of our analysis of the Negro Quarter subfloor pit assemblage above, would allow us to determine if they contain late-18th century artifacts along with the later ones that date their final deposition. Stay tuned!

References

Bates, Lynsey A., Jillian E. Galle, Fraser D. Neiman. 2020. Building an Archaeological Chronology for Morne Patate. In Archaeology in Dominica: Everyday Ecologies and Economies at Morne Patate, edited by Mark Hauser and Diane Wallman, pp. 64-87. University Press of Florida, Gainesville.

Christie, Annalisa C., Alastair Grant, and Anne Haour. 2019. Cataloging Cowries: A Standardized Strategy to Record Six Key Species of Cowrie Shell from the West African Archaeological Record. African Archaeological Review 36(4):479–504. https://doi.org/10.1007/s10437-019-09351-z. 

DAACS. 2026. “Glossary” Digital Archaeological Archive of Comparative Slavery.” https://www.daacs.org/about-the-database/glossary/. Accessed October 15, 2026. 

Haour, Anne, and Annalisa Christie. 2019. Cowries in the Archaeology of West Africa: The Present Picture. Azania: Archaeological Research in Africa 54(3):287–321. https://doi.org/10.1080/0067270X.2019.1648726

Haour, Anne, and Abigail Moffett. 2023. “Global connections and connected communities in the African past: stories from cowrie shells. African Archaeological Review 40(3): 545-553.

Heath, Barbara. 2016. “Cowrie Shells, Global Trade, and Local Exchange: Piecing Together the Evidence for Colonial Virginia.” Historical Archaeology 50(2):17-46.

Jefferson, Thomas. 1776-1778. Monticello: outbuildings (notes), circa 1776-1778. N88 [electronic edition]. Thomas Jefferson Papers: An Electronic Archive. Boston, Mass.: Massachusetts Historical Society, 2003. https://www.masshist.org/thomasjeffersonpapers/doc?id=arch_N88&mode=lgImg. Accessed October 15, 2026.

Jefferson, Thomas. 1796. Monticello: building insurance, recto. N133; K136 [electronic edition]. Thomas Jefferson Papers: An Electronic Archive. Boston, Mass.: Massachusetts Historical Society, 2003. http://www.thomasjeffersonpapers.org/. Accessed October 15, 2026.

Monticello: building insurance, recto, 1796, by Thomas Jefferson. N133; K136 [electronic edition]. Thomas Jefferson Papers: An Electronic Archive. Boston, Mass. : Massachusetts Historical Society, 2003. http://www.thomasjeffersonpapers.org/

Nelson, Lee H. 1968. Nail Chronology as an Aid to Dating Old Contexts. National Park Service.

Noël Hume, Ivor. 1969. A Guide to Artifacts of Colonial America. Knopf, New York.

Renfrew, Colin and Paul Bahn. 2019. Archaeology: Theories, Methods, and Practice. Eight Edition. New York: Thames and Hudson.

Wells, Tom. 1998. Nail Chronology: The Use of Technologically Derived Features. Historical Archaeology 32(2): 79-99. 

Wheeler, Derek. 2004. PDF of composite excavator’s plan compiled by DAACS from original field drawings. Digital Archaeological Archive of Comparative Slavery, Building t site page. https://www.daacs.org/sites/building-t/#images. Accessed October 15, 2026.

Yang, Bin. 2018. Cowrie shells and cowrie money: A global history. Routledge.