Archaeology in Dominica: Everyday Ecologies and Economies at Morne Patate 2020017100, 2020017101, 9781683401605, 9781683401889

Citation preview

Archaeology in Dominica Florida Museum of Natural History: Ripley P. Bullen Series

H X

ARCHAEOLOGY IN DOMINICA Everyday Ecologies and Economies at Morne Patate

Edited by Mark W. Hauser and Diane Wallman

University of Florida Press Gainesville

Copyright 2020 by Mark W. Hauser and Diane Wallman All rights reserved Published in the United States of America 25 24 23 22 21 20

6 5 4 3 2 1

Library of Congress Cataloging-in-Publication Data Names: Hauser, Mark W., editor. | Wallman, Diane E., 1981– editor. Title: Archaeology in Dominica : everyday ecologies and economies at Morne Patate / edited by Mark W. Hauser, Diane Wallman. Other titles: Ripley P. Bullen series. Description: Gainesville : University of Florida Press, 2020. | Series: Florida Museum of Natural History: Ripley P. Bullen series | Includes bibliographical references and index. Identifiers: LCCN 2020017100 (print) | LCCN 2020017101 (ebook) | ISBN 9781683401605 (hardback) | ISBN 9781683401889 (pdf) Subjects: LCSH: Archaeology—Dominica. | Slavery—Dominica—History. | Dominica—History. | Dominica—Antiquities. Classification: LCC F2051 .A73 2020 (print) | LCC F2051 (ebook) | DDC 972.9841—dc23 LC record available at https://lccn.loc.gov/2020017100 LC ebook record available at https://lccn.loc.gov/2020017101 University of Florida Press 2046 NE Waldo Road Suite 2100 Gainesville, FL 32609 http://upress.ufl.edu

CONTENTS

List of Figures vii List of Tables ix Acknowledgments xi 1. Everyday Economies and Ecologies of Plantation Life 1 Mark W. Hauser 2. Dominica as an Evolving Landscape: Evidence of Changing Social, Political, and Economic Organization in the Eighteenth Century 31 Tessa Murphy and Mark W. Hauser 3. Tracing the Postemancipation Landscape of Dominica’s Lime Industry 48 Samantha Ellens 4. Building an Archaeological Chronology for Morne Patate 64 Lynsey A. Bates, Jillian E. Galle, and Fraser D. Neiman 5. Morne Patate House Yards, 1750–1900: An Overview 88 Khadene K. Harris 6. A Carbet among the Cabins: The Significance and Symbolism of a Possible Kalinago Household at Morne Patate 111 Lennox Honychurch, Diane Wallman, and Mark W. Hauser 7. Sourcing Coarse Earthenware at Morne Patate: The Impacts of French Colonialism and Local Exchange 129 Lindsay Bloch and Elizabeth Bollwerk 8. The Environmental Archaeology of Subsistence and the Socioecological Landscape at Morne Patate 153 Diane Wallman and Sarah Oas

9. Conclusion: Resilience and Capacity Building in the Age of Empires 168 William F. Keegan Bibliography 173 List of Contributors 193 Index 195

FIGURES

1.1. 1.2. 1.3. 1.4. 1.5. 2.1. 3.1. 4.1. 4.2. 4.3. 5.1. 5.2. 5.3. 5.4. 5.5. 6.1. 6.2. 6.3. 6.4. 6.5. 7.1. 7.2. 7.3. 7.4.

Timeline of Dominican history 5 Location of Dominica and Morne Patate 7 Excavations, 2015–2016 19 House areas, 2017 20 Chronology of house yards 21 Eighteenth-century settlements in Soufriere 34 Pulley for cableway in Soufriere 54 A scree plot of the amount of variation (inertia) accounted for by the successive CA dimensions 74 Number of ceramic sherds assigned to each phase 79 AI estimates for the five phases at Locus 1 and Locus 2 82 A Cudgelling Match between English and French Negroes in the Island of Dominica, 1779 95 Cottage and laborers on Bois Cotlette, Vaquero 96 Photograph of House Area A 102 House Areas A–D, F, and G 104 House Areas E and H 107 View on the River Roseau, Dominica, 1770/80 114 Villagers Merry-Making in the Island of St. Vincent, ca. 1775 115 A Negroes Dance in the Island of Dominica, 1779 116 A Leeward Islands Carib Family outside a Hut, ca. 1780 117 A Family of Charaibes, Drawn from the Life, in the Island of St. Vincent, 1801 118 Coarse earthenwares recovered from Morne Patate 138 PCA biplot of LA-ICP-MS results, log ppm values for 41 elements 140 Boxplots showing average elemental concentrations by compositional group 141 Scatterplot of calcium and hafnium showing the division of Huveaune compositional group into two possible subgroups 143

TABLES

1.1. 2.1. 3.1. 4.1. 5.1. 6.1. 6.2. 7.1. 7.2. 8.1. 8.2. 8.3.

Chronology of Morne Patate 8 Documented land use and slave population at Morne Patate 45 Value of lime juice exported from Dominica (£), 1870–1896 52 Chronological indicators for Morne Patate plantation-wide phases 77 List of domestic features according to occupation phase 103 Artifact types from House Area G 121 Approximate chronology of the site 121 Summary of sourcing results 144 Coarse earthenware types by site occupation phase 146 Local species noted by Atwood with corresponding binomial names 157 Identified fauna at Morne Patate 160 Ubiquity of major botanical remains at Morne Patate 163

ACKNOWLEDGMENTS

We would like to begin by stating how grateful we are to the authors who contributed to this volume and the research behind it. Their work provides a more holistic understanding of slavery at the edge of empire. The research for this manuscript was directly funded by a National Science Foundation award entitled Chronological Change in Domestic Economy and Provisioning Strategy (NSF Award ID 1419672) as well as supplemental funds provided by Northwestern University and the University of South Florida. The success of this proposal and the research that ensued was aided by a longer archaeological engagement funded through the WennerGren Foundation for Anthropological Research and an NSF High Risk Award entitled Archaeological Reconnaissance to Locate French and English Plantations in Dominica (Eastern Caribbean) 1700–1900 (NSF Award ID 0948578). The many anonymous referees who reviewed the manuscript provided valuable feedback and input into the research design that resulted in this work. The Thomas Jefferson Foundation through the Department of Archaeology at Monticello also financially contributed to this project in cataloging and making accessible research materials through the Digital Archaeological Archive of Comparative Slavery. This work would not have been possible without the enthusiastic support of property owners on whose land the archaeological research took place. George Blue, Christina Garner, and Simon and Wendy Walsh permitted us access and provided important material support for the crew. There were many from Dominica and beyond who worked on this project. Edward Thomas, Mitchell Laville, Dean Bellot, Dorival Bellot, Dora Bellot, Dan Wade, Keifia Stokes, Kirsha Reynolds, and Michael Samford were all valuable members of the research team who patiently worked with us and grew to be tireless advocates of the community’s archaeological heritage. Colleagues, both practicing and graduate students, from various institutions also contributed their efforts. Lindsay Bates and Isaac Shearn brought their

xii · Acknowledgments

valuable expertise to the field, which enriched the research at this small place. Alan Armstrong, Lauren Nareau, and Naphtalie Jeanty all contributed to this project in important ways. We would also like to thank those who volunteered on this project. William Keegan, Tom Hamilton, Michael Dion, Phil Springer, Sylvia Chappell, Barbara Cox, Amy Tyler, Kelly Delancy, and Andrew Walsh worked as volunteers. Finally, there were those that assisted in communicating research finds to a larger Dominican community. Paul Crask, through the auspices of Dominica magazine, visited the site frequently and wrote stories that were shared to residents and visitors of the community at large. CocoYea Cafe allowed the team to present talks about our research on a biweekly basis. We are grateful to the many people who provided valuable feedback on this manuscript, including Ken Kelly and Krysta Ryzewski. In addition, Kalyani Menon and Timothy Earle gave particularly valuable assistance. Volumes such as these are not the product of single authors but the work of many people, which extends beyond the confines of the table of contents or the works cited at the end of the book. There were many more people who provided valuable assistance who are not mentioned here. We would be remiss not to mention the community of Soufriere and Dominica more generally. On September 18, 2017, at 9:15 p.m., Hurricane Maria became the first category 5 hurricane to make landfall in Dominica. As has been widely reported in the media, the hurricane devastated the island nation’s infrastructure, housing stock, and economic base, and the government is currently seeking assistance from governmental, nongovernmental, and private organizations to assist in the long process of recovery. High winds, floods, landslides, and recovery efforts have exposed people to risks of food and water security. This provides a particularly poignant irony on the “Nature Island,” where water and nature’s abundance were often used as selling points for the island’s tourist industry. It should not go without saying that Soufriere and the surrounding towns were particularly hard hit. They have weathered the storm, but the economic and social impacts of such events will continue to play out, and our thoughts are with them as they continue to demonstrate their resilience in the face of the climate crisis.

1

Y Everyday Economies and Ecologies of Plantation Life Mark W. Hauser

A Small Place For they are caught in a collision and a clash that was inherent and in-built, and still is, between the plantation system, a system owned and dominated by external forces, and what we shall call the plot system, the indigenous and autochthonous system. (Wynter 1971:31)

Sitting on the crater’s edge, just above the small village of Soufriere in Dominica, there is a small place people call Morne Patate. The name refers to an area that is about 2 acres but once referenced a property that was approximately 300 acres. While unknown to most people outside of this island of 80,000 residents, the history of this place is crucial for not only Dominica but the region as a whole. As one of the newest of Britain’s Caribbean possessions, Dominica has long been overlooked in the histories of the Caribbean colonialism, especially as it relates to plantation slavery and its aftermath. The number of enslaved laborers, the wealth of its landowners, and the amount of produce exported never quite matched those of its older, wealthier, and more populated cousins, Jamaica and Barbados. Despite this, the development of places on this island, such as Morne Patate, is a story essential to understanding plantations and the modern world. In the quote that opens this chapter, Jamaican author and postcolonial scholar Sylvia Wynter references Jamaica Kincaid’s 1988 narrative nonfiction A Small Place, in which she describes the everyday predicaments of people when places are put to novel economic regimes. But Kincaid could have been easily describing the people negotiating the plantation agriculture during and after slavery. She describes colonies as contested landscapes caught between different priorities; one of the plantation system and the

2 · Mark W. Hauser

other the plot of everyday life. Reading the above quote through Donna Haraway’s observation (2015:162) that “slave gardens are an underexplored world,” we see that everyday ecologies and economies of enslaved people offer an important but different account of plantation landscapes. Such landscapes are implicated in matrices of power, global histories, and relations to markets and also have a residue that may be documented in the archaeological record. Wynter—and many of the scholars who extended her ideas—was not interested in discussing the material culture of everyday life as environmental practice. While much historical archaeological work employs material culture as evidence of people’s dispositions, activities, and beliefs, and while influential schools of thought in archaeology have recast the assumptions about interactions between humans and nonhumans, the material residues of everyday environmental practice has rarely been a focus within the archaeological treatment of plantation landscapes. These two impulses developed in tandem over the past several decades. For example, much of Eric Wolf ’s and Sidney Mintz’s theories of history and inequality were premised on the material world’s relationships to economies, environments, and power in everyday life (e.g., Mintz 1961; Wolf 1959). In the 1950s and 1960s attention to the environment was also evident in studies of how institutions, such as plantations, were understood to create foundations for social inequality. Since the 1970s, anthropological archaeological approaches have attended to cultural foundations and meanings as shaping and shaped by interaction, including the symbolic analysis of architecture and its communicative meaning. Historical archaeological inquiries sought to include interpretive variables, including ethnicity, style, and syncretism within an analysis of material culture (Armstrong 1990; Handler and Lang 1978). At the same time, studies foregrounding relationships between space and political economy directly took up materialist approaches wherein built landscapes shape access to social power and permit the accumulation of wealth. All of these approaches, however, largely construed economy as existing apart from ecology. Indeed, despite decades of work building on historical ecology (Watts 1966), only recently have scholars of colonial landscapes begun to consider the environmentality of features so commonly studied in historical archaeology—landscape, economies, and everyday life. A key question drove research for chapters in this edited volume: How did everyday economic and ecological relationships change with the introduction of sugar as an export commodity in Dominica? While there have been many studies considering such transformations from a macro

Everyday Economies and Ecologies of Plantation Life · 3

perspective (Delle 2014; Meniketti 2015, to name two recent examples), few have detailed the everyday activities of ordinary people that enable such activities (See Hauser 2017; Oas and Hauser 2018; Wallman et al. 2018). In answering this question, contributors to this volume identify three major themes that were important for this study: economies, ecologies, and everyday life. These themes not only lend significance to the results of an archaeological study of this place but also realize the potential for comparative studies of plantation landscapes in other times and other settings. Contributions document similar patterns that developed on other sugar islands but also emphasize important differences—from the continued presence and influence of indigenous peoples, to labor regimes seemingly less structured for the enslaved, to a material repertoire that marks distinct influences from Britain, neighboring French islands, and locally developed traditions. Material remains and cultural landscapes are an excellent tool through which to test the degree to which political boundaries shaped human, economic, environmental, and social interactions. Expanding this inquiry, chapters in this volume look at how ordinary people shaped and were shaped by changing economies and ecologies during the Morne Patate’s 200-year history. Defining the Plantation One way to analyze Morne Patate is as a plantation. The plantation is a frame of reference for each of the essays in this volume. Beginning in the 1950s, the plantation as a social institution became a central topic of anthropological inquiry, in part through the work of scholars working in the Caribbean and Latin America (Wolf and Mintz 1957; see also Rubin 1959). Much of the early work was spent defining the institution as a unit of analysis for social inquiry. For example, Eric Wolf and Sidney Mintz sought to describe and elaborate on the differences between haciendas and plantations. A hacienda, they argue, is “an agricultural estate, operated by dominant landowner and a dependent labor force, organized to supply a small-scale market by means of scarce capital, in which factors of production are employed not only for capital accumulation but also to support the status aspirations of the owner.” In contrast “plantations are an agricultural estate, operated by dominant owners (usually organized into a corporation) and a dependent labor force, organized to supply large-scale market by means of abundant capital, in which the factors of production are employed primarily to further capital accumulation without reference to the status needs of the owner” (Wolf

4 · Mark W. Hauser

and Mintz 1957:380). The differences between the two could therefore be characterized as the size of the market, the degree of specialization in production, the economic motivation of the owners, and the degree of control over the labor force. To be precise, the use of the term “plantation” in English predates European colonization of the Americas. According to the Oxford English Dictionary, a plantation in its earliest instance meant “something that has been founded, established, or implanted, as an institution, a religion or a belief, etc.” Four years later, the OED documents a different use, where it means the “act of planting seeds and plants in the ground.” By 1587 it came to mean “the settling of people, usually in a conquered or dominated country.” It was not until the seventeenth century (1626), that “plantation” was used to describe “an estate or large farm, especially in a former British colony, on which crops such as cotton, sugar, and tobacco are grown.” Scholars have extended these definitions to describe a broad condition of the past two centuries to include, “devastating transformation of diverse kinds of humantended farms, pastures, and forests into extractive and enclosed plantations, relying on slave labor and other forms of exploited, alienated, and usually spatially transported labor” (Haraway 2015:162). A long history of plantation archaeology exists in the Caribbean. While not the only kind of settlement found in the colonial Caribbean, nor the only context in which slaves labored, plantations have evolved as an important site of analysis for studying culture, economy, and power between the seventeenth and nineteenth centuries. By the mid-eighteenth century most people living in the anglophone and francophone Caribbean were living in a plantation settlement. Whether the estate grew cotton, cocoa, tobacco, coffee, or sugar, plantations contained dense villages and hamlets housing people of mostly African descent. These villages could be home for a handful of slaves on smaller farms growing tobacco or provisions (Armstrong 2003) or for as many as 400 slaves, as in the case of Habitation Loyola in French Guyana (Roux 2013). Plantations have become a useful unit of analysis to compare the economies, ecologies, and everyday life in colonial society. Importantly for those specifically interested in understanding the African diaspora in the Americas, plantations are sites where African cultural legacies in colonial Americas can be explored. On Dominica, plantation agriculture started comparatively late. Certainly, people were growing food for export as early as the seventeenth century (Boomert 2011a; Lenik 2012), but the racialized, intensive exploita-

Figure 1.1. Timeline of Dominican history.

6 · Mark W. Hauser

tion of the land that is synonymous with plantations did not begin until the second quarter of the eighteenth century. The history of agriculture on Dominica is very much tied to political contours of the region. Prior to French colonization, Dominica was a source of tobacco, hammocks, and arrowroot. When the French arrived, they began to grow provisions and coffee. Planters, British and French, took advantage of the new political order established by French colonization to intensify plantation agriculture through sugar, although coffee continued to be an important crop. As boom/bust cycles and blights continued, other crops—lime and cocoa—began to take on more important roles. As such, there were, in general, four types of plantations loosely framed around the crops grown on it. There were sugar estates that occupied large tracts of land and compelled between 120 and 180 enslaved people to work in gangs. There were coffee estates that occupied smaller tracts of land higher in elevation and compelled between 30 and 80 enslaved people to work through task-oriented labor. There were provision estates that compelled fewer than 20 workers on land not already used by sugar and coffee. Finally, there were plantations that grew all three. Morne Patate was one such plantation (Figure 1.2). Chapters in this volume focus on Morne Patate for several reasons. First, the size and production strategies of this settlement are similar to the different agricultural endeavors undertaken by planters and actualized by workers. Slavery was a major relationship of labor for many years of its operation. The plantations continued to operate well after slavery was over. As such, it is comparable with previously investigated sites including La Mahaudière in Guadeloupe (Gibson 2007, 2009; Kelly 2008), Crève Coeur in Martinique (Kelly and Wallman 2014), Trents in Barbados (Armstrong and Reilly 2014), Betty’s Hope in Antigua (Fox 2016; Wells et al. 2017), and Sugar Loaf and Bois Cotlette in Dominica (Hauser 2015). While the number of workers fluctuated, the population (ca. 120) made it an average yet sizable sugar estate. Second, its strength is the chronology. Unlike many plantations described above, the history of the settlement stretches back in time to before casual colonization by the French and formal colonization by the British. People continued to live there well into the twentieth century, making it an important study for tracking the changing economic and ecological relationships in everyday life (Table 1.1). Finally, the preservation of architecture, material culture, and dietary remains made it an ideal venue to study changes in those relationships in the everyday. Research conducted between 2015 and 2018 located both the planter’s house and village settlements, which have

Figure 1.2. Location of Dominica and loci of human interaction at Morne Patate (Loci 1–8).

La Soufriere Tenant

Soufriere and Undivided Estate; between the Morne Patate Elder (2/3) and the Younger (1/3) Morne Patate Soufriere sold to Samuel Chollet and Peregrine Bourdieu Morne Patate Undivided Estate;

Morne Patate Receivership, J. W. Foreman 1816, Jean Marie, LaFeuillie, Paquet, Mayne RaveriJohn Gordon 1817 ere (assumed) Morne Patate Undivided Estate; Mayne De Raveriere (part proprietor)

Morne Patate Undivided Estate;

Morne Patate Undivided Estate;

Morne Patate Undivided Estate

Morne Patate

Morne Patate

Morne Patate Ownership

Morne Patate Undivided Estate

1770s

1780s

1810s

1830s

1860s

1870s

1930s

1940s

1970s

1980s

1820s

1800s

1790s

La Soufriere Ownership

1750s

L. Rose and Co.

L. Tronchine

L. Tronchine

J. W. Foreman (Husband of Beamish) and others Heirs of J. W. Foreman

Marie Adele Beamish (part proprietor)

Jean Marie, Fleurie, Paquet, Mayne Raveriere

Nicholas Croquet de Belligny (The Elder and Junior) Nicholas Croquet de Belligny (The Junior)

Nicholas Croquet de Belligny (The Elder)

Nicholas Croquet de Belligny (The Elder)

Nicholas Croquet de Belligny (The Elder)

La Soufriere Ownership

1740s

Owner

Estate Name Legal Status

Decade

Table 1.1. Chronology of Morne Patate

n/a

n/a

130s

160s

120s

120s

120s

120s

100s

?

?

Enslaved Workers

x

x

x

x

x

x

x

x

x

x

x

x

x

x

x

x

x

x

x

x

x

x

x

x

x

x

x

Sugar Coffee Provisions Limes

Everyday Economies and Ecologies of Plantation Life · 9

clearly defined housing areas for laborers and a history of occupation for the period of interest. Furthermore, comparison of results from Morne Patate with other sites enabled us to situate the findings within Caribbean-wide trends. An Island Plantation on the Edge of Empire: The Environmental and Historical Setting One of the unexpected findings of this volume is that there is no simple comparison to be made before and after sugar’s introduction into the plantation economy. In framing the proposal to funding agencies, I had hoped, somewhat naively, that changes in imperial sovereignty of the island in 1763 would coincide with the environmental and economic priorities of its inhabitants. I anticipated an added benefit with a coincidental change in material repertoires documented through the remainder of the Atlantic world, where refined earthenware produced in Staffordshire became accessible table settings for people of many classes and cultural backgrounds. The research did not bear out these expectations. People were transforming the landscape well before Britain annexed the island, and they continued to transform it well after slavery was legally abolished. That being said, the relationships of commerce and agriculture that accompanied British colonization of the island acted as a discursive and material crescendo in relationships of commerce and agriculture. The island was a subject of plans and treatises written by English-speaking elites. The decades immediately following British colonization were the apogee of the slave trade on the island. Between 1763 and 1807 more than 100,000 captured Africans disembarked onto Dominica. While many were ultimately sold to other parts of the Caribbean, many stayed (Campbell 2007) and these decades witnessed some of the most intense construction of buildings and infrastructure prior to the twentieth century. In the years leading up to and after emancipation, new crops were grown, and some decisions about cropping were made by those who were once considered property. The vast majority of the land, however, remained the property of a limited few. Our research was informed by the physical and social geography of the island. Cultural anthropologist Michel-Rolph Trouillot (1988, 27–32) describes Dominica as a “patchwork of enclaves”—relatively independent regions constrained by mountainous topography and varied links to neighboring islands. He argues that, until recently, different communities in the

10 · Mark W. Hauser

northern, southern, and eastern parishes were relatively isolated from each other, each producing separate cultural trajectories. Take the example of Soufriere, the enclave where Morne Patate is located. Until recently the enclave could be described as an island unto itself. People who live in this enclave habitually speak Dominican creole. They maintain commercial ties with Dominica and Martinique. It was only in the late 1960s that a motor road connected the village with the remainder of the island. Until then most people going to town took a narrow path leading up the steep slopes of Morne Acouma. Conversely, they could take a canoe to Roseau. Compare this to the enclave on the northeast of the island around Calibishie. Here people habitually speak an Antiguan-based patois, Koi koi. People who lived in this enclave were more closely linked with Portsmouth and MarieGalante. Physical geography therefore plays an important role, but so do the economic networks and social histories of the people who lived there. Mountains and Enclaves: The Island Landscape In 1493 Christopher Columbus was on his second voyage when he sailed past the island. He named it Dominica after the day on which he saw it, Sunday, November 3. This was not, of course, the first name given to the island. The people who were living on it, the Kalinago, called it Wai’ti Kubili (Tall Is Her Body). They called the island this for two reasons. For people who lived on this island, “The earth was an indulgent mother who furnished them with all things necessary” (Rochefort 1666). This is hardly a surprise as the island was inhabited by humans, and Kalinago ancestors would have experienced tremors, earthquakes, and eruptions that would have reminded them of the island’s dynamic and sentient nature. The name also belies a fact about Dominica with which both past and present inhabitants have had to contend. It is a steep island, containing narrow river valleys separated by ridge lines and with limited coastal plains. This geography, one of the more extreme in the Caribbean, speaks to the youth of the island. The five volcanoes that compose the island’s spine are technically dormant, although their geothermal energy supplies heat and provides nutrients for sulfur springs that dot the landscape. These volcanoes have also shaped the island in essential ways for inhabitants. While no eruptions have been documented in Dominica since Europeans arrived in the sixteenth century, for those who lived in Dominica previously, these volcanoes would have been part of the living landscape. Soufriere is in some ways the

Everyday Economies and Ecologies of Plantation Life · 11

youngest part of the island. The most recent volcanic eruption took place there sometime between 1410 and 1590, and its ash composes the subsoil (Wadge 1985). The bay that Morne Patate overlooks, Soufriere, is a collapsed portion of a caldera. This simple geological fact made it possible for residents to capture both reef and pelagic fish not too far offshore. The steepness of the slope, the elevation of the land, and the composition of the soil vary significantly between enclaves. These differences are important because it helps us understand where people decided to settle, what crops could be grown where, and the consequences of settlement and agriculture on the land. The soil horizon on much of the island reflects its geological youth. While soil scientists have identified more than a dozen soil types on the island, five major classes are what most people encountered as they settled the land. These include kandoid soils, clays, alluvium, protosols, and young soils. Each of these soils has different properties important for cultivation, including how well it drains, the degree to which it leaches nutrients, and how well anchored it is to the subsoil. Because the island is so young, protosols, consisting of topsoil overlying pyroclastic ash, are a significant portion of the island’s soil horizon (22%). Land in Soufriere is 69% protosol. The remainder include “young soils,” “sulfur-rich soils,” and alluvium. Protosols drain well and retain nutrients. However, they are easily eroded if not anchored by root systems. In addition to soils, slope is important. As a general rule of thumb, people tend to build on slopes of 15 degrees or less, but on Dominica 44% of the land is on a slope between 20 and 90 degrees. In the enclave of Soufriere, the site of Morne Patate, 70% of the land rests on a slope within that range (Hauser 2020). The extreme slope means that limited land was available on which agricultural fields could be cleared and settlements constructed. Before Europeans arrived, settlements were located in the narrow patches of flat land and limited gentle slopes. When Europeans arrived, they moved into these previously occupied areas and also onto the more precarious slopes. They leveled platforms for houses on hillsides and piled up stones for agricultural terraces. Planters favored different landscape conditions for certain export crops. They felt the uplands were best suited for coffee (Girault 1985:70) and cocoa (Wood 1968:101). They also felt that sunny and well drained lowlands were best for sugar cultivation (Richardson 1992). Certain export crops, such as cotton, indigo, and tobacco, were more flexible and often were the first commodities planted in plantation colonies (Dunn 1972:53; see Watts 1990 for a

12 · Mark W. Hauser

thorough discussion of mid-eighteenth-century practices for ground preparation in the eastern Caribbean). Forested highlands were important for the hardwoods and the hunted game that they provided. In some enclaves, including Soufriere, these highlands also became the location where provision grounds were set aside for the enslaved workers. As a function of steepness, single enclaves contained land at different elevations. On Dominica 23% of the land could be characterized as lowlands, 27% as uplands, and 50% as highlands. In Soufriere the land is divided nearly evenly between these three zones. Soils also vary widely across the island. Due to the offshore fishery, the diverse elevation, and the soils, it is no surprise that Soufriere was a place of early human migration (Berard 2007) and early casual colonialization by Europeans (Hauser and Armstrong 2012). An Island in Between (ca. AD 800–1763) Dominica was annexed by Great Britain only in 1763, making it one of the last islands in the Caribbean to be officially colonized by a European power. Despite this young colonial history, it has a long human history. The earliest occupation of the island is in Delices, on the windward side of the island, and dates between 390 and 520 BCE (Shearn 2018:143). The societies of those who lived on the island were anything but small, static, or isolated. The peoples occupied nearly 61 archaeological sites that have been documented between 1950 and today. Here they adapted to new landscapes and, in so doing, modified the landscape (Shearn 2014). Since the third century BCE, islanders encountered new people from mainland South America and neighboring islands. In Soufriere previous archaeological investigations have identified sites in the low-lying alluvial plain from as early as AD 200 and as late as AD 1200 (Berard 2007). Morne Patate is not without its early component. We have no idea what these people called themselves or how they affiliated with each other culturally or politically. That being said, by the seventeenth century, Raymond Breton (1958: 1) said that island residents called themselves Kalinago, “of the islands,” to distinguish themselves from those who were from South America. While the Kalinago were neither politically nor culturally unified, they did prove to be a powerful military and economic force into the seventeenth and early eighteenth centuries. Europeans did settle here well before official colonization, and the island acted as a de facto dependency of Martinique (see Murphy and Hauser, this volume). Some of these settlers were seeking

Everyday Economies and Ecologies of Plantation Life · 13

advantage elsewhere; others were looking for “geographies of alternative possibility” (Hauser and Armstrong 2012). Jeannot Rolle, a gens de couleur from Martinique, is often cited as the first outsider to colonize the island, with the aim of establishing an economically viable colonial settlement in Grand Bay. These new settlers attempted to take advantage of the island’s ambiguous political status, profiting from a lack of government scrutiny and engagement in the broader world economy (Hauser and Armstrong 2012). These early settlements had significant implications for the economy, ecology, and everyday lives of people living here. What we know about Morne Patate from the documentary record is limited to a few enumerations in the years leading up to and after 1763. The estate was part of what was referred to as the Soufriere Quarter. According to a birth certificate found in Le Pecheur, the settlement then identified as Upper Soufriere Estate was established in the 1740s by Nicholas Croquet de Belligny. We have recovered no document that tells us how many enslaved people Belligny might have had laboring for him in these early years. We can surmise that it was not many. In 1743 Soufriere Quarter along with Point Michelle and Grand Bay had 210 slaves and 115 free men, women, and children. Rural settlements in these quarters grew cacao, coffee, manioc, yams, plantains, and coffee. By the time of British annexation in 1763, Belligny had at least 117 enslaved laborers working on two neighboring estates. A New West Indian Colony In 1761, as part of one campaign in the Seven Years’ War between France and Britain, Dominica was captured by Lord Rollo (1703–1765), commander of the British land forces, and Sir James Douglas (1703–1787), commander of ships. After the war Britain formally annexed the island, thus ending nearly 100 years of official neutrality. Sir William Young (1725–1788) was appointed the president of the Commission for the Sale of Lands in the Ceded Islands. He was responsible for overseeing and instituting a colonial agenda in Dominica, St. Vincent, Grenada, and Tobago. In Dominica, Young confronted three issues with the settlement of the island. First he had to secure the political and economic integrity of the island through the building of fortifications to monitor the French navy and contraband trade. He also had to encourage sugar production and settlement in order to generate wealth for the cash-strapped metropole. And he had to integrate French settlers and their enslaved laborers who occupied the land, grew coffee, and

14 · Mark W. Hauser

purportedly continued economic and social ties with the French colonies of Martinique and Guadeloupe. Evidence suggests that the commission’s efforts began to bear fruit. Before Britain annexed the island, Dominica was a de facto dependency of Martinique. Beginning in 1728 the Roseau-based commandant ordered enumerations of the island’s various quarters. These are generalized accounts stating the number of free and enslaved men, women, and children; the acreage of crops; and the number of animals on island. Before British annexation, crops include coffee, cocoa, yams, plantains, and cassava (Recensement général de l’île de la Dominique pour l’année 1753 [1753] no. 87, Dépôt des papiers publics des colonies [DPPC], Archives nationales d’outremer [ANOM], Paris). Documentary evidence reveals that Morne Patate grew rapidly. Two probates, one in 1777 and one in 1816, indicate that two estate houses were constructed, the second as a replacement for the first. It also indicates that by 1777 the estate had one boucan, a building devoted to drying coffee beans, a stable, and a warehouse. By 1816 either Belligny or his heirs had commissioned a boiling house, and many of the coffee lands had been converted to sugar. Between 1777 and 1827 the number of enslaved laborers living in the settlement fluctuated between 99 (in 1810) and 166 (in 1817). The majority of these enslaved lived in 35 to 36 “negro houses.” The quantity of sugar and coffee produced on the estate is not documented until 1827, when the estate is noted as having 121 slaves producing 21,000 pounds of sugar and 25,075 pounds of coffee. A Colony of Labor Redefined (ca. 1834–1950) In 1833 Parliament passed the Abolition Act, which legislated the end of slavery in the British Empire. Freedom would not be granted until after an eight-year apprenticeship. People were required to work for their previous masters weekly for thirty-five hours; any time worked beyond that was not required and needed to be compensated with wages. The apprenticeship system was abolished early in Dominica in 1838. In many cases, Soufriere included, enslaved laborers continued to live in the enclaves they once worked (Trouillot 1989). At the same time, Dominica also became a place of agricultural experimentation, where the boom/bust cycles of sugar, coffee, limes, and cocoa affected everyday life for workers.

Everyday Economies and Ecologies of Plantation Life · 15

As Khadene Harris points out, “the newly freed were acutely aware of opportunities that existed beyond the estate”(Harris 2016, 266). In some cases workers and laborers entered into a sharecropping relationship in which they continued to live and work the land. Property owners would retain a portion of the produce. New hamlets emerged on portions of land that sat at estate boundaries (Harris 2016). In other cases people fled the estates, moving to narrow strips of land that were intended for roadways and fortification. To make a living, these laborers were dependent on wages from working on estates and on cultivating land at the margins of plantations (Trouillot 1988). While acquiring property was theoretically possible for many of the former enslaved and their children, a combination of laws and market forces kept land out of reach for many. But a law requiring people to purchase no less than 40 acres kept “most people as landless labourers” (Honychurch 2001). These regulations, coupled with land taxes in 1886 and 1888, left wage labor one of the few viable options for people to make a living in the agricultural sector. At Morne Patate, people continued to live and work on the estate after slavery was abolished. All chapters in this volume, save one, address what came after slavery. The economic networks, the human–environmental engagements, and the construction of meaning in everyday life continued to inform the landscape. A sharp decline of coffee production began in the 1830s and was attributed to a blight (Cemiostosa coffelum), which reportedly destroyed most plantations. Some coffee planters, whose confidence was undermined by this blight and other labor concerns, uprooted their trees to grow cocoa, but Dominica remained known for its superior coffee well into the nineteenth century. Conversely, sugar estate owners doggedly continued to grow sugar regardless of inadequate production and profit. An 1843 almanac shows how Dominica’s economy diversified. That year 5.2 million pounds of sugar and 1.47 million pounds of coffee were exported. Goods most likely produced by the island’s new peasantry included arrowroot (4,000 pounds), firewood (853 cords), and charcoal (673 bushels). These were probably sold to neighboring islands. Poor returns from sugar meant that workers on the estate were asked to cultivate and harvest lime trees. In 1840 British agronomists suggested that the lime tree could be grown in fields once devoted to sugar. Lime concentrate would have been used in Rose’s Lime Cordial or further refined to make calcium citrate, an antacid.

16 · Mark W. Hauser

Archaeology at Morne Patate What happens to the everyday life of ordinary people when one crop replaces another in their agrarian economy? This seemingly simple question is made complex by the shifting legal and economic structures that framed crop replacement. For the many ordinary people considered here, institutions of slavery or its enduring effects structured their everyday lives. Regimes of labor structured the amount of time people had to grow their food for themselves, the amount of free time they had, and the amount of time devoted to caring for their household. The chapters of this volume focus on the socioecological changes that occurred on the southwestern tip of the Caribbean island of Dominica when the land and its occupants became subjected to the new and intensive practices of export sugar production on plantations. These chapters address the development of the plantation system on Dominica, a “marginal” colony of the British Empire, and how this development is reflected in material culture and the cultural landscape. The chapters use a multimethod, multiscalar approach to understand activities leading up to and in the wake of this economic transformation. Some of these activities were planned in distant drawing rooms of colonial planners. Others were improvised on the ground by people who were compelled to work by the institution of slavery. Many activities are reflected in evidence retrieved from archaeological and archival contexts. The different activities documented through the written and the archaeological records are what make Morne Patate such an interesting place to study. Archaeology of Everyday life at Plantations In archaeology, everyday life is a theme most closely linked with Maya archaeologist Cynthia Robin (2013). The everyday, she argues, concentrates archaeological work on “ordinary places and objects” and their extraordinary role in human society. Archaeologists concerned with “ordinary people” have shown how archaeological contexts such as household middens and kitchen areas contain remains made during the intimate and routine interactions between colonized and colonizing peoples (Deagan 2001; Robin 2013; Voss 2008). One of the ways that archaeologists have examined everyday life on plantations is to concentrate on the house yard (Armstrong 1990; Heath and Bennett 2000; Kelly 2019; Pulsipher and Goodwin 1999;

Everyday Economies and Ecologies of Plantation Life · 17

Wilkie and Farnsworth 2001). Using house yards “as a setting for daily activity” (Mintz 2010: 10), scholars have investigated the social, economic, and symbolic roles of these spaces under conditions of slavery and its aftermath. House yards often contained houses and attached kitchen gardens. As an archaeological deposit, they offer readily comparable information that allows us to discover substantive differences and similarities in material practices during the “sugar revolution” (see Harris, this volume). To understand everyday life, chapters in this volume take a decidedly multiscalar approach to the plantation. More than three decades ago, Bill Marquardt and Carole Crumley (1987) suggested that people experienced their world on multiple scales in space and time, and archaeologists should continuously move between different scales until the “effective” scale (where patterns begin to emerge) is achieved (Crumley 1979:166). Whether it is troubling property boundaries or political borders in understanding textural and archaeological sources (Murphy and Hauser; Bloch and Bollwerk, both this volume); analyzing household materials in the context of the Digital Archaeological Archive of Comparative Slavery (Bates, Galle, and Neiman, this volume); or considering the afterlives of slavery, both in the housing of people and the crops they tended (Harris; Ellens; Wallman and Oas, all this volume), the effective scale varies. By focusing on plantations as colonial settlements and those who inhabited those areas, one can begin to disentangle the varied ways in which colonial landscapes were constructed and contested (Voss 2008). A multiscalar archaeology is inherently a multimethod archaeology. Research conducted for this volume grew out of the Archaeological Survey of Colonial Dominica. This 10-year regional study of colonial period archaeology sought to understand the complex histories behind small Caribbean islands like Dominica. The documentation of plantation settlements, among others, provided baseline data by which to understand the development of plantation society there. All told, the survey identified eight plantations in the Soufriere region. As stated earlier, Morne Patate proved to be an important place to study the development of the plantation system and everyday life for three reasons: the long-term history of the place; the size and nature of the settlements people lived in during that history; and the integrity of the archaeological record. Moreover, there was evidence for the excellent preservation of botanical, faunal, and material evidence related to slave life. This is a rarity in the Caribbean that offered a singular opportunity to study human–environment interaction.

18 · Mark W. Hauser

There is one more reason that made Morne Patate an important place to study. Community members living in the nearby village of Galleon, including landowners George Blue, Christina Hendricks, Mitchell LaVille, and Simon and Wendy Walsh, were enthusiastic participants, providing intellectual and logistical support. While not initially designed as a communitybased project, their enthusiasm made it so. Through their help, we arranged biweekly community meetings and seminars. People would arrive from neighboring villages to hear our report on our findings but also to provide insight into the relevance, or lack thereof, of the questions we asked. Their participation dramatically changed how we understood the site and the methods we would use to go about and explore it. In order to explore the plantation as a whole, excavations focused on areas we believed to be house yards, including both domestic spaces in the village (Locus 2) and the estate house (Locus 1), agricultural lands, and provision grounds (Figure 1.3). Research at Morne Patate followed a multistage strategy of increasing intensity beginning first with a pedestrian survey in order to identify how different parts of the estate were used by past inhabitants. It was in these early surveys that Dominican archaeologist and crew chief Edward Thomas and I would rely most heavily on community members to understand different parts of the land and how they were used. We documented many places that were used in the past and the present. Some of these places were hidden-away hollows where people would make charcoal to sell in town. Others were gardens where people still grew food for the market. Still others were places that came to be associated with people in the past. The names people assigned to places were helpful. The estate house was located in what people called “le Cou,” the traditional creole name for the social space of the estate. All told, 15 separate loci of human interaction were identified on the estate. While each of these loci was examined, only three of them were intensively studied through standard archaeological methods: testing to determine the organization of the village, its size, and the layout of house areas; delineating and excavating to map how domestic areas were used; and recovering personal and portable possessions and retrieving faunal and botanical remains. The people responsible for much of this work were contributors to this volume, student volunteers, and Dominican assistants, many of whom lived in the community, including Michael “Togo” Samford; Dean, Dora, and Dorival Bellot; and Mitchell LaVille. They became skilled archaeologists and important advocates for the region’s heritage. They

Everyday Economies and Ecologies of Plantation Life · 19

Figure 1.3. Excavations at Morne Patate, 2015–2016.

helped organize school visits. Importantly, their contribution also extended into crucial insights about archaeological features. Ultimately, the way we studied the development of plantations and everyday life was through the house yard. We defined the house yard as the complex of spaces attached to a single house and excavated these areas in blocks. As an area defined by a relatively flat area on the steep slopes of Morne Patate, an excavation block could contain multiple house yards, tracking in space and over time, differences in organization, material assemblage, and dietary evidence. The team excavated eight such blocks, A–G, and the estate house. Some of these blocks, such as house area E, contained only one phase of occupation in which the area acted as a residence. Others, such as house area A and the estate house, appeared to have been used as a residential space throughout the site’s colonial history (see Figures 1.4 and 1.5). None of the house yards were excavated to their fullest extents. To do so would have been to sacrifice the sample size that was necessary to understand how different past actors engaged with the economic, social, and environmental surroundings. In each excavated area, architecture was identified through an alignment of postholes, stones, foundations, or floors.

Figure 1.4. House areas at Morne Patate, 2017.

Locus 3

Locus 2

Locus 1

1700

HA G

HA F

HA E

HA E

Maison1

1750

HA D

Farina House

Maison2

1800

.ca 1770 Sugar Introduced

Road

HA C

HA B

Factory

Maison3

Boucan/Glacee

Provision Ground/Garden

HA A

Stables

1850

Caretaker’s House

1900

1950

ca. 1840 Coffee 1884 ‘Entire or partial Blight abandonment of sugar’ ca. 1850 Lime Introduced KalinagoTerritory Established 1692 JeanottRolle Settles Grand Bay 1807 End of Slave Trade 1728 French Commandant 1831 Brown Privilege Bill 1763 British Annexation 1834 Emancipation Act 1778-1783 French Occupation 1838 End of Apprenticeship 1865 Made a Crown Colony

ca. 1730 Coffee Introduced

Figure 1.5. Chronology of house yards at Morne Patate.

22 · Mark W. Hauser

All identified structures had either intramural or extramural features that contained material or dietary evidence of everyday life in the form of storage pits and cooking hearths. These features are crucial as they gave us clues as to how space was organized within house yards. Some of the observations could be assigned functional explanations. For example, the cooking hearths revealed through excavation were located close to and upwind of the entrance to the houses they were associated with. This location would seem counterintuitive unless you relied on the smoke from the fire to deter the presence of mosquitoes. Other observations require inferences about how household membership was defined. Excavations revealed, for example, deep pits containing personal portable possessions, among other things, in contexts associated with the nineteenth century. These appear to be storage pits, and their presence indicates increasingly complex households in which some people might not have been related. These features could also be examined within their own terms to understand how everyday economies and ecologies changed over time. Contributors to this volume employed multiple techniques to analyze material culture, faunal remains, and macrobotanical specimens. Our study was a comparative endeavor relying on archaeological big data to see where Morne Patate sits (Bates, Galle, and Neiman, this volume), to gain a deep understanding of the varied taxa of species in the region and the taphonomic processes that effect their archaeological visibility (Wallman and Oas, this volume). By intensively examining one site and comparing this to a growing body of archaeological data recovered from field survey and other work across the eastern Caribbean, our excavations provided the data necessary to examine how broad historical processes are translated into the particular everyday activities. Contributors also engaged in a connective agenda. While the points of origin in many of the artifacts we recovered could safely be determined, the itineraries by which they got to the site and what those itineraries reflect are more vexing. Rather than looking at specific sources, we concentrated on measuring diversity. In what way was the diversity of commercial network reflected in material culture change over time? Here the work of archaeometry is basic. As archaeologists who document labor in the relatively recent past have demonstrated, published manuscripts, laws, legal agreements, census, and historical accounts help illuminate social and economic networks that are

Everyday Economies and Ecologies of Plantation Life · 23

difficult to establish solely through the archaeological record. Contributors to this volume consulted dispersed archives including the Records Office and the Dominica National Archives, Archives départementales de la Martinique in Fort de France, Archives nationales d’outre-mer in Aix-enProvence, the British Library, and the National Archives in the United Kingdom. Archival documents reveal something important about the history of labor in the eastern Caribbean. Whether it is the eighteenth or nineteenth century, Morne Patate was connected through capital, kinship, and commerce to estates in Martinique (Wallman and Oas; Murphy and Hauser, both this volume), Montserrat (Ellens, this volume), and beyond. Economies One of the questions addressed by this volume is how people living at Morne Patate were entangled in a broader world economy and how those entanglements transformed during the years leading up to and after the introduction of sugar. These changes included the introduction of new trade regimes including the Freeport Act, which opened up Dominica and Jamaica to intercolonial exchange. Dominica was connected to a wider world economy well before the French and English started to settle the island in the early eighteenth century. They also relied more heavily on slavery as a relationship of labor. With new colonial holdings, Britain acquired a ready market in addition to a source of raw and processed materials. The Freeport Act was intended to attract French sugar and make Dominica a marketplace for neighboring islands to purchase slaves and goods such as ceramics, textiles, and glass. The act also included the end of slavery. Plants also had a say in the matter. Sugar did not fare as well as many hoped, and a blight struck coffee in the 1830s. These different crops had different labor regimes that affected everyday life, including how many people were living in the settlement and how much free time they had to grow their own food (Dunn 1972; Higman 2005; Sheridan 1974). At the same time a system of informal and internal markets emerged as a form of everyday economic practice for the vast majority of the island’s population (Handler and Wallman 2014; Mintz and Hall 1960). These economies not only shaped everyday life under slavery but also continued well after its abolition. The archaeological record shows a dramatic change in the size and organization of the island village. It would have been nice to see if this change corresponded with the change in political sovereignty. Archaeological

24 · Mark W. Hauser

materials recovered from these superimposed household contexts in the village and the maison de maître indicate that the transformation of the estate did not happen all at once. During the transition, planters came to increasingly rely on the provision ground system in which the enslaved were compelled to grow their own food “in their free time.” Slaves were also responsible for their diet and the acquisition of household furnishings. After the agricultural transition, two trends are suggested: first, localized food production expanded; second, enslaved Africans began experimenting with different crops and garden locations. Evidence of animal resources was more limited than plant remains. Among the possible items that the slaves would have eaten are hunted animals (giant ditch frog, manicou, and agouti), some domestic animals (chicken and pig), fish, and produce from provision gardens in the nearby hills. To cook this food, slaves used local coarse earthenware (most likely made in St. Lucia or Martinique), earthen pottery made in southeastern France, and some cast iron cookery. They also relied on griddles, probably to make cassava bread from a New World crop used by local indigenous peoples. Artifact assemblages also indicate that well after Britain annexed the island, people continued to rely on commercial networks that connected Dominica to French Martinique and Guadeloupe. One of the ways that contributors to this volume address the theme of economies is through an examination of how the circuits of goods that people used changed over time (Bloch and Bollwerk, this volume; Bates, Galle, and Neiman, this volume). Diversity in economic networks represented in local, regional, and interregional goods highlights the degree to which political boundaries dictated human opportunities as represented by circulation of people and things (Smith 2005). With a denser settlement, we observe a more diverse economy with difference between houses. This difference is manifest in the elaboration of assemblages and the value that has been placed on them. The economic networks, the human–environmental engagements, and the construction of meaning in everyday life continue to inform the landscape after slavery was abolished. Changes for everyday life are documented in the house area excavations at Morne Patate. At Morne Patate, people continued to live and work on the estate. Nearly all excavated house areas contained discrete contexts, including features and floors, that were used in the years after 1838. Housing during this period differed in terms of material and construction but was similar in size to their earlier counterparts.

Everyday Economies and Ecologies of Plantation Life · 25

One novel accommodation was the use of cut-stone flooring and foundations for housing. Yards continued to be symbolic and economic centers of everyday life (Harris, this volume). We observe an intensification of one spatial feature—storage pits. Materials recovered from these yards also show that workers and their children learned to read and write. They show broad diversification of economic networks upon which people relied (Bloch and Bollwerk, this volume). While the abolition of slavery and apprenticeship dissolved some of the predicaments that workers faced, new ones emerged. These new predicaments required different accommodations. With the inception of apprenticeship in 1834 and emancipation in 1838, the appearance of a new economic order arrived on the shores of Dominica, but plantation labor was still required for the crops grown on the island. As described above, the island as a whole underwent numerous changes in its agricultural base throughout the nineteenth century. Planters came to rely on cocoa and lime more and more as the coffee blight and fluctuations in the sugar market affected their returns. As Samantha Ellens (this volume) points out, the amount of lime juice increased significantly between 1870 and 1896, by 1,800%. The value of limes had increased so much on the global market that many of the estates were consolidated by Rose and Company at the turn of the twentieth century. At Morne Patate, however, owners continued to invest in sugar until the 1880s. It is only in the last quarter of the nineteenth century that the village appears to have been abandoned or, more probably, displaced. Ecologies The specialization of agricultural production resulted from global reorganization as early modern empires developed and included the cultivation of coffee, cocoa, and sugar in the eastern Caribbean. New production and exchange networks relied on the movement of free, indentured, and enslaved people to provide intellectual and physical labor in those regions and transformed local environments to more efficiently produce marketable commodities—at least in the short term. Little work, however, has focused explicitly on the political ecology of sugar colonies in the Caribbean. This is unfortunate; colonies and distant territories were likely characterized by “patterns of ecological power relations” (Grove 1997:183) such that we miss a rich source of evidence about how those transformations affected the everyday (Biersack and Greenberg 2006; Carrier and West 2009; Crumley 1994;

26 · Mark W. Hauser

Fisher and Thurston 1999; Wolf 1972:202). Environment should be used as a medium for agency, identity, and conflict (Kirch and Hunt 1997; Morrison 1994; Robin 2013; Stein 2002; West 2012). Tessa Murphy and Mark Hauser show that these concerns were part of the calculus that led to Nicholas de Belligny creating these agricultural enterprises at least two decades before Britain annexed the island. This enterprise is analogous to what Alfred Crosby (1986) has called “ecological imperialism.” Plantation settlements in northern colonies became botanical laboratories whereby “the plants on which Europeans historically have depended for food, fiber, and the animals on which they have depended for food, fiber, power, leather, bone, and manure, tend to prosper” (Crosby 1986: 6). Accounts detailed by Murphy and Hauser in this volume show that similar experimentation took place at Morne Patate. In its earliest incarnation, Belligny, the owner of Morne Patate, relied on crops that indigenous peoples had long been using. While there is no specific information on the estate, enumerations suggest cotton, tobacco, and cocoa were grown in the enclave with the express purpose of an export market. They also show that coffee, an Old World crop, was the crop grown in the greatest quantity. Coffee required experimentation with soil, water, and new arrangements of labor—chattel slavery being its most extreme form. The archaeological record shows a dramatic change in the size and organization of the village that coincided with a transition to sugar. According to one C14 date recovered from a feature that bore no cultural materials, Morne Patate was occupied at least as early as 771–903 AD CAL (80% probability) (Honychurch et al., this volume). The first consistent evidence of settlement were five post-and-beam houses, four of which were rectangular (Harris, this volume) and one was ovoid (Honychurch et al., this volume). Importantly, none of these houses appear to be grand in a manner that one might call a maison de maître in contemporary Martinique. In this loosely clustered set of houses, each home was surrounded by a large garden. While less structured, most of the people who lived on this estate were enslaved. There were important exceptions. In addition to property owners, the round house, according to Lennox Honychurch and colleagues (this volume), might have been inhabited by a Kalinago person or family working on the estate. After Britain annexed the island, sugar cultivation began as only one more crop in a long history of botanical experimentation on the estate. This experiment, as Murphy and Hauser describe (this volume), did have its consequences. According to one probate, soils were denuded and once-fertile

Everyday Economies and Ecologies of Plantation Life · 27

fields were considered abandoned or put to purposes for which the land was ill suited. In that probate, witnesses blame a hapless estate manager. To be clear, they did not question the decision to grow sugar, merely its implementation. Sometime between 1763 and 1917, Khadene Harris notes (this volume), a dramatic transition coincided with an increased density of people living in the regimented village. The regimented village with small houses (approximately 3 × 5 meters) had attached gardens (approximately 10 × 8 meters). Most likely these plank houses were placed directly on posts of wood and had walls made with dry stones. We also see a garden area distinct from the village during this period. Thick deposits of square-cut and wrought nails confirm plank wood construction techniques. To make this village, people cut into the loose protosols, creating a constant threat of landslides. The regimented village also meant that most of the food that people lived on was grown on provision grounds that were at some distance from the estate. As Donna Haraway notes, “Nurtured in even the harshest circumstances, slave gardens not only provided crucial human food, but also refuges for biodiverse plants, animals, fungi, and soils. Slave gardens are an underexplored world, especially compared to imperial botanical gardens, for the travels and propagations of myriad critters” (2015:162). Europeans were not the only ones planting new seeds. Diane Wallman and Sarah Oas (this volume) show that enslaved Africans began experimenting with different crops and garden locations and that these crops continued to be relevant after emancipation. Judith Carney and Nicholas Rosamoff (Carney and Rosamoff 2011) note that there is an essential botanical legacy of Africans that is often overlooked in the Caribbean. Enslaved Africans brought with them ideas on how to work the land, including intercropping. They also brought with them plants, including sorghum and millet. These key inputs are as vital to understanding the new ecologies brought about by plantation agriculture as the cane field or the coffee tree grove. These slave gardens were biotic, political, and social refuges. By examining socioecological systems, we can gauge the degree to which elites controlled, allocated, and limited resources (Biersack and Greenberg 2006; Crumley 1994; Hastorf 2003) and the degree to which nonelites exercised agency through everyday decisions in subsistence agriculture (Handler and Wallman 2014; Jamieson and Sayre 2010; Morehart and Helmke 2008; Morrison 2007; Oas and Hauser 2018). As Ellens (this volume) and Harris (this volume) point out, the later occupation of Morne Patate should not be treated as a single sequel to slavery.

28 · Mark W. Hauser

With emancipation and the end of apprenticeship, the relationship of people with the land changed significantly. Archaeological evidence suggests planters were no longer obliged to provide land for laborers to grow their food, although this practice continued in many places as a matter of custom. Add to that, people continued to rely on a combination of locally sourced foods and some imported domesticates (Wallman and Oas, this volume). The provision ground continued to be a location for people to grow their food, and housing continued to occupy yards that had been used for at least a century. Under the new colonial regime, people continued to be vulnerable, but in different ways. Upon emancipation, many people left their sites of bondage by seeking employment elsewhere and moving to the “Three Chains” set aside for roadways and forts (Honychurch 2001). Two archaeological observations document this vulnerability. Much of the reason why macrobotanical samples are so well preserved is that at least a part of the village was burned down in the late nineteenth century. While there was a late nineteenth-century occupation by workers at Morne Patate, few twentiethcentury goods were recovered (Bates et al., this volume). This observation corresponds to the migration of Morne Patate’s regimented village to Galleon. People became vulnerable to displacement. This period is inhabited by its predicaments, including new political arrangements and economic orientations where displacement was a genuine threat for people who made a living off the land they did not own. In Jamaica, for example, old slave houses on some estates continued to be occupied by tenant farmers throughout the nineteenth century until the boom/bust cycles of global markets found new uses for the land and those houses were knocked down (Armstrong and Hauser 2004). As Ellens (this volume) points out, the wages that people were paid were tiny—far smaller than what could have allowed them to stay. “Squatting” on Crown land came to be one of the only means through which people could build a home. Displacement was not just a physical removal from the land but a discursive act in which people became increasingly vulnerable. This precarity was not just economic but also physical. Increasingly intense weather associated with climate change has today become part of the predicaments residents have to negotiate on a daily basis. In September 2017 Dominica was ravaged by Hurricane Maria. While hurricanes are not a new threat to the region, the magnitude of the storm in the wake of Tropical Storm Erika in 2016, from which the island was still in recovery, resulted in historical devastation. As has been widely reported

Everyday Economies and Ecologies of Plantation Life · 29

in the media, the hurricane devastated the island nation’s infrastructure, housing stock, and economic base, and the government is currently seeking assistance from governmental, nongovernmental, and private organizations to assist in the long process of recovery. People displaced from the land and forced to live in these narrow villages were disproportionately affected. Concluding Thoughts There are many ways to organize a volume such as this. On first glance it might appear as if the volume is framed chronologically, and certainly there is a general arc followed by chapters hereafter. That being said, most of the chapters touch on Morne Patate’s history between the eighteenth and twentieth centuries. Rather, the editors (Wallman and Hauser) chose to follow a traditional structure. We begin with the geographic and historical context of the site (Chapters 1, 2, and 3); the archaeological record in space and time (Chapters 4, 5, and 6); analysis of economies and ecologies through objects of everyday life (Chapters 7 and 8); and implications of the research (Chapter 9). This structure could be loosely broken down into three narrower questions addressed in this volume: What makes a plantation a plantation? How are social and economic inequalities built into its landscape? How does the material record of enslaved workers speak about, with, or against the plantation as a concept and socioecological form? At Morne Patate the archaeological record forecloses the possibility of a history where the steepness of the slope and the vagaries of the weather and the soil do not play important roles. It also negates a history where people living on neighboring islands of the region also do not play equally important roles, whether those islands belong to the economic regime or not. Instead, the archaeological record emphasizes the diverse ways in which people interacted with the environment around them and the diverse economies they engaged in to make a living. This diversity is not only a story of a highly globalized twentieth century but is also deeply rooted in the years immediately before and after emancipation. Importantly, the predicaments created by and through plantation agriculture were part of the tools of governance in New West Indian colonies, such as Dominica. The following chapters reveal that life in the small place of Morne Patate was, to a large extent, a coproduction between newcomers and people who had made their home on Dominica for generations. People had different agendas or environmental priorities, some of which are better understood

30 · Mark W. Hauser

than others. Accounts, such as that of William Young (1764), state explicitly their agenda for colonizing the island and how that agenda fit into the project of empire. Other priorities can only be inferred from the material remains recovered by archaeologists. While there are many ways in which one might argue that Morne Patate is unique, the same argument could be made for any estate that had to deal with the vagaries of global markets, boom/bust cycles for agricultural commodities, and the effects that such intense uses wreak on the land. The chapters in this research show that power of place, in its very broadest meaning, matters.

2

Y Dominica as an Evolving Landscape Evidence of Changing Social, Political, and Economic Organization in the Eighteenth Century Tessa Murphy and Mark W. Hauser

In 1765 Louis de la Ferriere Constance created an inventory of Bois Cotlette, the plantation he owned in Soufriere, in southern Dominica. In addition to 51-year-old Louis, there were also 86 enslaved people on the plantation (Dominica National Archives [DNA] 1765 Deed Book [DB] B N.1:#40). These people included Mathurin, a 16-year-old from the Gambia; Collette, a 32-year-old Mina woman; and Francisco, a 27-year-old identified as “Sosso.” Some of these women and men may have only recently arrived in the British colony of Dominica; in the 12 months before Louis created this plantation inventory, five English slaving vessels collectively disembarked almost 800 individuals from the coast of West Africa onto the 290-square-mile island (Slave Voyages database, https://slavevoyages.org/ voyage/database#searchId=IeuQm2Fv). But others—like 16-year-old Paul, 46-year-old Martha, and 40-year-old “Mulatto wench” Lucrece—were, like their master, Louis, explicitly identified as “Creol,” meaning born in the Americas, if not necessarily in Dominica. The fact that people like Lucrece—an enslaved adult who was born in the Caribbean—lived on a plantation in Soufriere in 1765 is significant for a number of reasons. First, the presence of such individuals belies Dominica’s prior political status as “neutral.” Although European Crowns had mutually agreed that the island would remain outside of their respective realms, de la Ferriere Constance’s ownership of a well-established plantation just two years after Dominica was ceded to Great Britain in 1763 hints at Dominica’s entanglement in French colonial projects in the period prior to the Seven

32 · Tessa Murphy and Mark W. Hauser

Years’ War (1756–1763). Second, this inventory of Bois Cotlette hints at the particular characteristics of slavery at the margins of empire. With no legal access to the transatlantic slave trade, planters like Louis used other means to increase the enslaved population on their plantations. Whether engaging in an interisland slave trade or encouraging or relying on reproduction among their enslaved laborers, planters like Louis helped give rise to a Creolized enslaved population in Dominica. Third, the inventory sheds light on the place of Dominica in the wider Atlantic world in the period prior to 1763. Lacking formal access to the mercantilist networks intended to govern trade between metropole and colony, planters in Dominica favored subsistence and secondary export crops that could be illicitly trafficked to their family members and business associates in neighboring French colonies, rather than all the way to Europe. And finally, the inventory raises the question of what would happen to these Creoles—both the enslaved and their masters—when the frontier society of Dominica was incorporated into the existing British Empire after 1763. In the decades that followed, Soufriere experienced a number of significant changes that reflected the region’s shifting position in the wider Atlantic world. These included a dramatic increase in the enslaved population, a shift from mixed agriculture to sugar production, and significant investment in technologies to grow and process sugar. Nonetheless, planters in Soufriere—the majority of whom were descended from French Catholics who settled in the island prior to its cession to Great Britain—also retained several key features of the agricultural, slaveholding, and familial practices they had established in the first half of the eighteenth century. This chapter examines the link between administering land and administering colonial subjects. While scholars have explored imperial attempts to shape the environments of the West Indies (Grove 1995; Watts 1990) and to govern the people living in the islands (Weiss Muller 2017; Willis 2013), we argue that the two endeavors were inextricably linked. This work builds on that of Max Edelson, who makes the connection between mastering space and mastering empire. Instead of a close reading of maps, we use archival and archaeological evidence to explore how this administration was activated on the ground. Attention to evolving patterns of settlement and land use in Soufriere provides a microstudy of imperial attempts to master and manage the colonial landscape and testifies to the persistence of longstanding regional practices in the face of formal legislation.

Dominica as an Evolving Landscape: Evidence of Change in the Eighteenth Century · 33

Background to Colonial Settlement While there are few archival records for Dominica prior to 1763, French administrative documents provide some insight on labor and land use in the neighboring island during the period of its colonial dependency on Martinique. By 1729 the intensity of European settlement on Dominica prompted Martinique’s governor to appoint a commandant. Based in Roseau, Dominica, the commandant’s primary responsibility was to protect the island from English invasion and to oversee and protect Martinique’s commercial interests (Hauser 2015). The commandant relied on indirect monitoring via planters to enumerate the crops, populations (enslaved and free), and weapons at hand on the island. Enumerations taken in 1730, 1743, and 1753 document a dramatic increase in the number of enslaved laborers, first from 395 to 1,880, and ultimately to 3,530 individuals (Archives nationales d’outre mer [ANOM] Dépôt des papiers publics des colonies [DPPC] G1/498, Recensements de l’isle de la Dominique 1730, 1743, and 1753). These enumerations also describe an agricultural base of export commodities including cocoa (Theobroma cacao [L.]), coffee (Coffea spp.), cotton (Gossypium spp.), cassava (Manihot esculenta Crantz), choux caribe (Xanthosoma sagittifolium [L.] Schott), plantains (Musa spp.), sweet potatoes (Ipomoea batatas [L.] Lam.), and yams (Dioscorea spp.) These crops were most likely exported to neighboring French colonies, whether as foodstuffs for slaves on sugar estates or for supplying the large urban populations in Basse-Terre, Guadeloupe, and in St. Pierre, Martinique (Pérotin-Dumon 2000). By the 1730s small planters, mostly of French extraction, had also established plantations in Dominica to grow coffee and cocoa for the international market (Lenik 2010, 2012). Previous documentary and archaeological research established that Soufriere, in southern Dominica, contained a high density of such settlements including Morne Patate, Crabier, Bois Cotlette, Petit Coulibri, and Morne Rouge (Figure 2.1). Plantations such as these are important settlements for several reasons. As stated above, the presence of such plantations and the individuals who owned or labored on them belies Dominica’s prior political status as neutral. Dominica officially became a British colony at the end of the Seven Years’ War in 1763. Unlike Grenada or Quebec, which transferred from French to British rule as a result of the war, Dominica had not technically been a colony at all—instead, a series of treaties between the British, the French, and sometimes the indigenous inhabitants of the Lesser Antilles stipulated

Figure 2.1. Eighteenthcentury settlements in Soufriere.

Dominica as an Evolving Landscape: Evidence of Change in the Eighteenth Century · 35

that the island (along with St. Lucia, St. Vincent, and Tobago) should remain neutral, meaning outside the sphere of European imperial rule. This failed to deter the more than 1,500 settlers, along with their 5,000 slaves, who were already living in Dominica as of 1763. The existing population of Dominica in the wake of the 1763 Treaty of Paris was reported to be 1,718 free settlers, 5,872 slaves, and “from 50 to 60 Caraib familys” (The National Archives of the UK, Colonial Office, 101/1 N. 91). The existence of well-established plantations like Morne Rouge, Bois Cotlette, Crabier, and Morne Patate just two years after Dominica became a formal colony reveals that the island was neutral only in the sense that its connections to surrounding empires were not legally acknowledged (Hauser and Armstrong 2012). By the middle decades of the eighteenth century, Dominica was inextricably linked both economically and socially not to Great Britain but to the French colonies of Martinique and Guadeloupe—and by extension to the wider Atlantic world. Surviving Catholic parish records for Soufriere residents, including those who owned the Morne Patate plantation, offer some insight on when and why people migrated from Martinique to Dominica during the period in which the island was politically neutral. In November 1723 the Catholic priest for the parish of Precheur, in northern Martinique, baptized Marie Catherine Laverge LaFeuillée. Marie Catherine was the legitimate daughter of Sieur Adrien Louis Laverge LaFeuillée and his wife, Marie Catherine Plissonneau. Acting as godmother to baby Marie Catherine was her paternal aunt, Luce Laverge. This means that as early as 1723 both branches of young Marie Catherine’s family lived not in France but in the Caribbean, and it suggests that her parents were also Creole, or born in the colonies.1 In the ensuing years, Marie Catherine’s parents baptized several more children in the parish of Precheur—Francois Robert in 1724, Therese Louise in 1729, Suzanne in 1731, and Marguerite Rose in 1734 (ANOM DPPC, Etat Civil, Le Precheur, 1724, 1729, 1731, and 1734.). After 1734 the Laverge LaFeuillée family largely disappears from parish records in Precheur, Martinique. With the exception of Catherine’s older sister, Marie Luce, who married a man named Louis Adenet la Ronde in Precheur in 1736, Marie Catherine and her siblings do not appear in later records for the parish. This archival silence can be explained by the family’s migration outside the sphere of formal French rule. By the mid-1740s— and likely earlier—Marie Catherine LaVerge LaFeuillée was living not in the French colony of Martinique but in the neighboring neutral island of Dominica. An 1821 burial record for a man named Jean Francois Croquet

36 · Tessa Murphy and Mark W. Hauser

Deshauteurs indicates that he was born in Dominica in 1745 to Nicholas Croquet de Belligny, owner of the Morne Patate plantation, and his wife, Catherine LaVerge la Feuillée (ANOM DPPC, Etat Civil, Le Precheur, July 29, 1821). This means that at some point between her birth in 1723 and the birth of her son in 1745, Marie Catherine LaVerge la Feuillée left Martinique and established herself at Morne Patate, Dominica. The age of the older Creole enslaved people listed on a neighboring plantation’s inventory in 1765 suggests that the plantation’s owners undertook a similar migration around the same time as their neighbors on Morne Patate. While some of these Caribbean-born slaves may have been born elsewhere in the Americas, an 1817 slave registry distinguishes between Creoles born in Martinique and those born in Dominica. The oldest enslaved Dominica Creole in 1817 was listed as 70 years old, meaning he was born on the island around 1747—16 years before Dominica became a formal colony (DNA 1817 Register of Slaves:592). A wider scale exodus of small planters from Martinique in the middle decades of the eighteenth century lends further credence to this hypothesis of interisland migration in the same era. Three factors—an earthquake that resulted in the ruin of many cacao planters, the beginning of coffee production, and the introduction of a head tax on free people of color—prompted hundreds of people to abandon the French colony of Martinique and migrate to neighboring neutral islands beginning in the 1730s. As historian Léo Élisabeth notes, many of these migrants were people of African and Afro-European descent, who sought to avoid the increasingly restrictive legislation imposed on free people of color in France’s colonies. Between 1732 and 1733 (the first year the head tax began to be levied on free people of color), the number of free-colored individuals recorded in the census of Martinique declined by 476 (Élisabeth 2003:307). In justifying the collection of the tax, the Comte de Maurepas, French minister of the marine and colonies from 1723 until 1749, reasoned that free people of color “should consider themselves lucky to enjoy freedom, without aspiring to the same prerogatives as white creoles” [“doivent s’estimer heureux de jouir de la liberté, sans ambitionner les mêmes prérogatives des créoles blancs”] (Duval nee Mezin 1975:177). A desire to evade the increasingly discriminatory impositions of imperial rule was reflected in the societies formed by residents of Dominica. Although whites and free people of color were enumerated separately in population counts of Dominica created by the island’s militia captains in 1730 and 1731, the practice was subsequently abandoned. Instead the ability

Dominica as an Evolving Landscape: Evidence of Change in the Eighteenth Century · 37

to bear arms in defense of the sparsely settled island (as indicated by the census category of “hommes et garcons portant armes”—men and young men or bachelors bearing arms); to act as a legitimate marital partner (“filles nubiles”—nubile or marriageable girls); or to carry on one’s family name (“garcons au dessous de 12 ans”—boys under the age of 12) became more salient markers of belonging on this Caribbean frontier. Although these categories testify to the fact that early settlers in Dominica were able to shape the terms of belonging in the infant society, they also brought with them many established features of French colonial life. By migrating, free people of color and other small planters who settled in Dominica before it was incorporated into the British Empire participated in extending existing French practices of plantation agriculture, slaveholding, and trade beyond the boundaries of European colonial rule. The 1765 inventory of Bois Cotlette, an estate neighboring Morne Patate, provides a vivid illustration of this extension. Just two years after Dominica became a colony of Great Britain, Bois Cotlette already boasted 4,000 plantain trees, 3 quarrés (a French unit of measure equal to 3.2 acres, so 3 quarrés is roughly 10 acres) planted in cassava, and 2 quarrés, or about 6½ acres, planted in sweet potato (DNA DB B N.1:#40). While these subsistence crops would certainly have sustained the 86 enslaved individuals then living at Bois Cotlette, it is likely that the surplus was consumed elsewhere. One obvious market would be in northern Martinique, located just 20 miles across the channel from Soufriere and accessible in half a day’s journey by boat. As a key site for the production of export crops such as sugar and coffee, arable lands in the northern part of the French colony were largely devoted to export commodities. This left little room to grow the food necessary to feed enslaved laborers in northern Martinique. Planters in Soufriere could have secreted their crops by boat to Martinique, allowing them to sustain enslaved laborers in the French colony and, by extension, to support the French Atlantic economy in ways that have not been readily visible to historians. Some of Dominica’s contributions to the French Atlantic economy were more direct. When Dominica became a colony of Great Britain in 1763, export commodities produced in the island entered the mercantilist system that governed trade between England and her colonies. But before 1763 the produce of the neutral islands did not legally belong to any European crown. As of 1765 Bois Cotlette estate reported some 30,000 coffee “bushes” and 20,000 cacao “bushes.” We cannot know when these plants began bearing

38 · Tessa Murphy and Mark W. Hauser

fruit or how much, but if we date the planting of the coffee and cacao trees to around the time that Soufriere quarter began to be established, in the 1740s, we can infer that the mature plants would yield hundreds of pounds of cacao and tens of thousands of pounds of coffee each year. Because Dominica lacked legal access to European mercantilist networks, these yields would have been ferried, alongside manioc and sweet potato, to northern Martinique. Dominican coffee and cacao would then be folded into Martinican yields, allowing the commodities to be exported as the produce of the French colony. Dominica’s legal status outside the mercantilist systems that governed imperial commerce also extended to the transatlantic slave trade, and this in turn had significant effects on the demographic profile of the island’s enslaved population. For the period prior to 1763, the Slave Voyages database lists zero ships with a primary port of disembarkation in Dominica. (As a point of contrast, the database records more than 120,000 people disembarking as slaves in the French colony of Martinique during the same era [Slave Voyages database].) Of course, this does not mean that no slaving vessels ever made landfall in Dominica—as early as the 1720s officials in Martinique complained of illicit human traffic on the neighboring neutral island (ANOM Fonds Ministeriels C8A28 f. 15). But most enslaved people who were trafficked to Dominica in the period prior to 1763 came from neighboring islands rather than directly from Africa. Colonies such as Saint-Domingue (Haiti), Guadeloupe, Martinique, Barbados, Antigua, Jamaica, and Nevis were notorious for the high mortality rates of enslaved laborers, especially in the first few years of their residency in island colonies. For example, in 1688 it was estimated that Jamaica needed 10,000 slaves, the Leeward Islands needed 6,000, and Barbados needed 4,000 to maintain the existing labor force (Curtin 1969:55–60). John Ward suggests that by the second half of the eighteenth century, the mortality of slaves in the West Indies during their first three years after arrival from Africa was 15–20% (Ward 1988:127). Much of this mortality would have been in the first year (Bergrad 1990:69; Fraginals 1983:1207). Because planters in Dominica were rarely able to purchase enslaved people directly from transatlantic traders, they would not have experienced the same level of mortality among their enslaved population. Lack of access to the transatlantic slave trade may also have motivated planters in Dominica to encourage reproduction among their labor force. This practice echoes demographic findings for the early French Caribbean

Dominica as an Evolving Landscape: Evidence of Change in the Eighteenth Century · 39

by scholars like Gabriel Debien (1974) and Arlette Gautier (2000), suggesting that planters in Dominica may have drawn on existing precedents as they extended slaveholding beyond the legal boundaries of empire. The presence of a self-reproducing enslaved population on Dominica gave rise to a Creolized enslaved population decades earlier than in formal Caribbean plantation colonies such as Jamaica (Higman 1995). An examination of the list of enslaved children in the 1765 inventory of Bois Cotlette supports the theory of a Creolized enslaved population. Out of 86 slaves on the plantation, 35 people are listed as “under age” (i.e., approximately 13 years or younger). This means that children accounted for almost 41% of enslaved people at Bois Cotlette in 1765. This is a considerable percentage given that throughout the late seventeenth and eighteenth centuries, children are estimated to have accounted for 12% to 22% of all people forcibly transported from Africa (Eltis and Engerman 1993:310). Of the 35 children, 26—almost three-quarters (74%) of all enslaved children on the plantation—are listed as “creole.” Some, like three-month-old Eulalia and three-year-old Mary Clara, are listed as “yellow,” suggesting that the growing population of enslaved people owed in part to relations between white men and enslaved women in the island. The existence of plantations such as Morne Patate and Bois Cotlette raises the question of what would happen to these Creoles—both the enslaved and their masters—when the frontier society of Dominica was incorporated into the existing British Empire after 1763. The plantations also raise the question of how newly arrived Africans were accommodated into existing provisioning systems. Prior to 1763 slaves in Dominica performed different labor than many of their counterparts in established colonies, growing foodstuffs and secondary export crops but not sugar. They participated in a different market economy, traveling not on foot to Bridgetown or Kingston but by canoe to St. Pierre (traversing, in the process, both geographic and political boundaries). And they formed and maintained families, allowing for the early emergence of a Creole slave society. The development of settlements in Soufriere in the decades prior to 1763 testifies to the creation of a distinct francophone, Catholic, Creole culture—among the masters as well as the enslaved—and this culture would persist long after Dominica became a colony of Great Britain. After Dominica became a British colony in 1763, French planters were encouraged to remain and to continue producing coffee on leaseholds, while British investors were encouraged to purchase lands on which to

40 · Tessa Murphy and Mark W. Hauser

develop sugar estates. This resulted in many properties being repurposed from growing cocoa and coffee to sugar cultivation as new planters from Great Britain or from existing British colonies began to settle in Dominica (Atwood 1791). An enumeration taken during a brief French reoccupation of Dominica between 1778 and 1783 documents an intensification in sugar cultivation on the island in the first decades of British rule. This involved the importation of 14,308 enslaved people onto 281 plantations. By 1785, of the 45,128 acres carved up by these estates, 43% were in woodland, 18% cane, 12% coffee, 10% pasture, 8% were abandoned, and the remainder were fields devoted to victuals, cotton, and indigo. Of these 281 estates, 65 were devoted to sugar; between these sugar estates were 54 water mills, 18 cattle mills, and 6 windmills (ANOM DPPC, 1785). At least 4 of these new sugar estates were located in the parish of St. Mark, including the Morne Patate estate. Despite the considerable changes in land use, export production, and demography that accompanied Dominica’s transition from neutral island to British colony, certain elements persisted long after 1763. Records like the inventory of Bois Cotlette—which was created because Louis de la Ferriere Constance was deeding one-third of his plantation to family members in St. Pierre, Martinique—reveal the continuation of familial and economic ties between residents of Dominica who became British subjects and their friends and relations in the French Atlantic. These ties were reflected in some of the features that distinguished plantation society in Dominica from that found in many other parts of the British Caribbean. These include the persistence of mixed agriculture rather than a complete switch to sugar production; the division of property according to partible inheritance rather than primogeniture; and the importance of Catholicism among both the free and enslaved populations. Other records, such as a series of manumissions for enslaved mixed-race individuals aged about 21, further suggest that practices common in the early French Caribbean persisted in Dominica, a society located at the intersection of the French and British Atlantic worlds.2 Imaging a Sugar Plantation Although many features of francophone Catholic society persisted after Dominica became a colony of Great Britain, the Board of Trade also envisioned many changes for its new Southern Caribbee Islands (Edelson 2017).3 Agents of the Board of Trade therefore needed to balance many competing priorities: to encourage investment by British planters; to encourage

Dominica as an Evolving Landscape: Evidence of Change in the Eighteenth Century · 41

the continued settlement and investment in French holdings, such as Bois Cotlette, on the island to earn money from existing coffee plantations; and to alleviate the concerns of parliamentarians who were worried about the social instability that Caribbean colonies were perceived to introduce into the empire. This was no small feat, and agents of the Board of Trade spent nearly a year making preparations in London, amassing a coterie of bookkeepers, surveyors, artists, and attorneys to help administer the recently captured Southern Caribbee Islands. These efforts produced a series of texts that speak to the plans discussed above. They also do something else. These documents in some ways were about the colony, as imagined in the homeland, but they were also part of a discourse about the land and the people it could support. Take, for example, the pamphlet William Young published in 1764 entitled Considerations Which May Tend to Promote the Settlement of Our New West India Colonies by Encouraging Individuals to Embark in the Undertaking (Young 1764). The document prescribes a series of actions to reduce the social and political risks of acquiring a set of islands. As described above, there was an entrenched and, from Young’s perspective, troublesome French population that was seen to pose a security threat. There was also an existing class of British West Indian planters who had become politically powerful in London and in the Caribbean. They had island assemblies that often worked against the interests of the king’s governors on each of the islands. There was also a large number of French residents who had developed considerable knowledge about the islands but also had familial and social ties to neighboring French colonies (Niddrie 1966:21). As such, they could simultaneously facilitate the transformation of these lands into productive colonies and undermine their security in geopolitical struggles between Britain and its rival, France. Finally, there was an enslaved population, some of whom were born in the Caribbean but an increasing number of whom were born in Africa. Beginning in the 1750s there were many times more enslaved people than there were people who claimed ownership over them in Dominica. Young’s text also speaks to some of the priorities that framed Britain’s territorial acquisition. Young saw an opportunity in the amount of land each island contained. He enthusiastically described the potential of planting in the “fertile” ceded islands, as “there has been no such opportunity of improving private fortunes” (Young 1764:36). The number of potential acres to put to cash crops exceeded those of almost all the established British

42 · Tessa Murphy and Mark W. Hauser

colonies, including St. Kitts, Antigua, Barbados, and the Virgin Islands— only Jamaica offered more arable land. Wealthy merchants and poor settlers alike could take advantage of different types of land to grow different crops. Wealthy investors might buy fertile valleys to grow sugar while poor settlers could take advantage of hilly uplands to grow coffee. At the same time, the islands were not too large. Unlike Jamaica, where roads were a necessary infrastructure for developing the interior, Young wrote that “long land carriage of burdensome commodities is destructive to cattle” (Young 1764:32). Instead, on the ceded islands canoes and sloops could ferry much of the traffic between estates and transshipment ports. The potential for wealth, Young argued, was tied to the potential of the land. Old islands, being less mountainous, and almost entirely cleared of wood, [have] become extremely dry and unseasonable; at the same time the lands in them, by long and constant planting, have . . . lost their spring and spirit of vegetation, as to stand in need of more rains than they had before. (Young 1764:37) The soil, according to Young, was new and “will be rich, yield large and regular crops, ratoon longer, require less planting and be cultivated with moderate expense and fewer negroes” (Young 1764:39). His assessment was not based on mere speculation. Grenada, which had been more intensively settled by the French, had more than three dozen sugar estates by the 1760s. This island, in Young’s estimation, was comparable in geography and potential to St. Vincent and Dominica. Young argued that Grenada’s “soil produces a sugar of most excellent quality. It is well watered with rivulets and abounds with good provision grounds for the negroes, which save a considerable expense to the planters in their maintenance” (Young 1764:5). Dominica also possessed a well-forested and mountainous interior. This, Young argued, is on the whole an advantage, not a defect; for although there are some objections too great an inequality in their surface, and to large tracts of wood, yet as they contribute to insure rains and fertility, and to produce rivers, they are very beneficial in these climates. (Young 1764:31) In addition to affecting the physical geography of Dominica, British plans for the colony also transformed its demography, with important consequences for those who lived in the island. After British annexation of Dominica,

Dominica as an Evolving Landscape: Evidence of Change in the Eighteenth Century · 43

the number of enslaved people rose dramatically. In 1763 there were 5,872 slaves laboring in Dominica. By 1804 that number increased nearly fourfold, to 22,083. In 1791 Thomas Atwood published his History of the Island of Dominica. In it he describes many of the people who made up this increased population: “Many of them [English] brought negros who had only been in the capacities of domestics; some of those banished from other islands for their crimes; and others purchased negros just brought from Africa for the purpose of settling new estates.” Atwood goes on to vividly describe what life must have been like for them: They were immediately set to work, to cut down massy, hard wood trees, to lop and burn the branches, clear the ground of the roots, and to labor at difficult, though necessary business . . . [they] were not used to the climate, which, from the abundance of woods, was so unsettled, that it rained greatest part of the year; whilst they only had temporary huts covered with branches and leaves of trees to shelter them. (Atwood 1791:224–225) In response to the dramatic growth of the enslaved population, the Colonial Assembly—an elite body composed of wealthy British planters residing in Dominica—also sought to encourage white settlement and land cultivation. Throughout the colonial era, assemblies in British West Indian colonies passed what have been referred to as deficiency acts. In Dominica the act fined each owner 6 shillings per acre of land that was still in woodland. Yet these acts were not so much to rid private landowners of woodland but to promote white settlement. The act paid out of the public treasury 20 pounds per annum for every white person on each estate who was over the age of 14. Of course, as with any law, we should always be suspicious about the degree to which this act was enforced. However, we could read acts such as the woodland tax act as also promoting deforestation, the consequences of which we see today. At the time, absenteeism was high in Dominica, and an act such as this would encourage property owners to populate their estates with white families. The principled arguments documenting how land should and should not be used can be found in a number of other sources including acts of the privy council and the legislature. These also mirror the plans for the settlement of the island originally laid out by the Board of Trade, which suggested that each purchaser of land also be required to settle one white man and two white women for every 40 acres of land within three months of receiving his grant.

44 · Tessa Murphy and Mark W. Hauser

Archival Evidence for Changes in Settlement and Land Use at Morne Patate Beginning in 1763, land use and demography changed dramatically when Dominica was formally annexed by Great Britain. The Morne Patate estate provides a microlevel example of how these changes were experienced. The British Crown was explicitly interested in increasing the amount of sugar being exported from the West Indies, and officials viewed their newly acquired territories as perfect lands through which to accomplish this (Murdoch 1984). As such, the islands experienced what historians have sometimes referred to as a sugar revolution. Barry Higman has described the Caribbean sugar revolution as a combination of a land grab and a crop boom where land owners shifted their diverse agricultural base to monoculture on larger plantations employing enslaved laborers living in denser settlements composed of a population racialized as black and who produced higher capital output (Higman 2000). On Dominica the sugar revolution was short lived and by many accounts had failed by the 1850s (Honychurch 1995). Despite the limited success of sugar production on the island, the creation of sugar plantations on Dominica had significant repercussions for land policy, demography, and trade systems (Hauser et al. 2017). Morne Patate is situated on a dome in the center of the caldera, half of which forms Soufriere Bay. Initially called Upper Soufriere Estate, it was established by Nicholas Belligny (b. 1718, St. Pierre, Martinique) around 1745. While there are no specific documents related to land use on Belligny’s property between 1745 and 1763, the enumerations detailed above allow us to infer that slaves at Morne Patate labored to grow provisions, coffee, and cacao for neighboring Martinique. Several documents allow us to piece together changes in land use and labor at the estate after 1763 (Table 2.1). These include an indenture between the property owner and his heirs in 1777 (DNA 1777 DB T2, May 28), an 1816 probate taken in the wake of a lawsuit against one of the owners (DNA 1816 DB X4 July 16), the triennial register of slaves taken between 1817 and 1834 (DNA 1817–1831, Register of Slave Returns), and a register of estates compiled by the Dominican assembly in 1827 listing the number of slaves and produce for each estate (DNA 1827, Register of Registration of Estates). The 1777 indenture between Belligny and his heirs provides a description of the disposition of the land and buildings. It values the estate at 707,937 French livres or 35,396 British pounds. Importantly, Soufriere at that time

Dominica as an Evolving Landscape: Evidence of Change in the Eighteenth Century · 45

Table 2.1. Documented land use and slave population at Morne Patate Year

Properties

1777

Morne Patate & Soufriere Estate Morne Patate Estate Morne Patate Estate Morne Patate Estate Morne Patate Estate

1784 1816 1817 1827

Slave Acreage Population 288

117

88

—

—

120

—

166

—

121

Documented Land Use

Oxen/ Horses/ Cattle Mules

Coffee, Pasture, Provisions —

33

8

—

—

Coffee, Sugar, Pasture, Provisions —

15

12

—

—

Coffee, Sugar, Pasture

—

—

included both a 200-acre leasehold in pasture or under coffee cultivation (Lower Soufriere Estate) and a freehold of 88 acres (Upper Soufriere Estate). While the leasehold is ambiguous about the disposition of the land that would become Morne Patate (Upper Soufriere), the absence of a boiling house in the indenture implies that the land was either used for foodstuffs or for coffee. The indenture indicates that there was one “dwelling house built of stone, 60 feet long by 20 and galleries on three sides.” In addition, it indicates a stone grating house, a boucan, a house in the pasture, and two sheds. That year there were 117 enslaved people living in 35 “negro houses” who labored on both Upper and Lower Soufriere Estates (DNA 1777 DB T2 N.2). While Belligny had yet to adopt sugar, he had invested considerable capital at Soufriere. In 1784 Belligny divided up the property with his son, who received the part of the estate that would become Morne Patate (DNA 1784 DB Z N. 2:f. 70). The building stock of the estate had been modified considerably in the space of seven years. The indenture states that the estate house was 55 feet by 40 feet. It had a masonry foundation and a wooden frame. It contained six chambers, “one of them a store, a hall and two galleries.” This most likely means that the estate house was rebuilt. The probate also lists a pounding mill, a farine or flour house, a wooden building, a boucan “40 feet long by 30 feet wide,” a horse stable, a manager’s house, and a cistern “12 feet in diameter.” In addition, a sugar work was constructed in the big savanna consisting of a cattle mill and a boiling house. These records describe 120 enslaved people living in 36 houses on the estate (DNA 1816 DB X4, July 16).

46 · Tessa Murphy and Mark W. Hauser

While this document does not detail land use at the time, an 1816 probate documents several changes in land management following the inheritance of the land. The probate also identifies 15 parcels of land by name. Two parcels (38 acres) are described as being under cane cultivation, and five parcels (no acreage specified) were in coffee. The remaining lands were abandoned, and the document provides some clue as to why these lands might have been left in this state: Describing the canes—one remark has forcibly struck us that the cutting down of the Galba [Calophyllum antillanum Britton] Fences where canes are now planted, was highly injurious, in a situation so much exposed to the wind and must prove extremely injurious to the canes, at present not of sufficient growth to feel it as much as they will when further advanced. The canes lately planted put in without Dung of which however there is some collected on the ground. The advanced canes in which 2 have been much neglected. Geographer David Watts has synthesized many of the standard sugar cultivation practices in Guadeloupe, Martinique, Antigua, and Barbados by drawing on a broad range of published planters’ manuals from both the French and British (Watts 1990:384–391). These include the works of Richard Ligon (1657), Pere Labat (1722), William Belgrove and Henry Drax (1755), and John Ellis (for coffee) (Ellis 1774). While these manuals are too numerous to be summarized here, two practices of widespread importance were manuring and the creation of windbreaks between fields by planting trees (Watts 1990:399–401, 426; 1966:383–385). What is described in the 1816 probate is a set of agricultural practices that do not conform to published manuals. However, while both sugar and coffee crops seem to have suffered in 1816 at Morne Patate, the owners recovered quickly. By 1817 the number of slaves at Morne Patate had increased to 166 (DNA 1817–1831 Register of Slave Returns). In addition, records also indicate the building of a boiling house and the construction of a glacée, indicating that the owners continued to invest in both sugar and coffee production. In summary, following Dominica’s sugar revolution, documentary evidence indicates that Morne Patate was caught up in a land grab and crop boom between the 1770s and 1800s. The owners converted some of the fields to sugar production, increased the enslaved population through the transatlantic slave trade, and built factories to grow and process sugar. However, the estate was never fully converted to sugarcane and its by-products of

Dominica as an Evolving Landscape: Evidence of Change in the Eighteenth Century · 47

sugar, molasses, and rum and continued to grow and process coffee beans for export. Despite the poor shape of the property documented in 1816, by 1827 laborers had rehabilitated the land to the point where it could again be productive. In that year the Dominica Assembly compiled a register of estates with returns on produce (DNA 1827, Returns of Registration of Estates), which indicate that Morne Patate had 121 enslaved people producing 21,000 pounds of sugar and 25,075 pounds of coffee. In the parish, Morne Patate was second in sugar production only to (Lower) Soufriere Estate and first in coffee production. Notes 1. Marie Catherine’s mother, Marie Catherine Plissoneau, was almost certainly born in the Caribbean, as a number of Plissoneau’s sisters served as godmothers to subsequent children born to the family. Because it is unlikely that a group of adult sisters would have migrated to Martinique together, we speculate that the Plissoneau sisters were also born in the Caribbean. The burial record for Marie Catherine’s paternal grandfather, Louis Adrien Laverge, indicates that he was a militia officer in Marie-Galante, indicating a period of residence elsewhere in the Caribbean. 2. See, for example, DNA DB A N. 1, N. 20, Manumission, Jean Louis Bellot frees Charles Melor a creole mestif man aged 21 years, July 26 1788; DNA DB I N. 1, N. 214, Pierre Nicolas Croquet de Belligny manumits Pancrasse, a mulatto man of 22 years, June 11 1785; and DNA DB I N. 1, N. 56, Louis Adrien Crocquet de Beaubois manumits Germain a mulatto man, 25 years old, January 6 1785. These are only a few examples intended to indicate a pattern in which French planters manumit mixed-race slaves around the age of 20, a common practice in early colonial Martinique that I speculate may have persisted at the margins of the French Atlantic world. 3. Along with Dominica, Britain’s new colony of the “Southern Caribbee Islands” included St. Vincent, Tobago, and Grenada, where the colony’s government was initially headquartered.

3

Y Tracing the Postemancipation Landscape of Dominica’s Lime Industry Samantha Ellens

Competing alternative global methods of sugar production introduced during the eighteenth century, such as the European sugar beet, meant that the Leeward Islands were no longer competitive in the world sugar market (Sheridan 1974:181–183). On many islands, nineteenth-century postslavery economies emerged as populations adjusted to the new sociopolitical conditions accompanying emancipation. During this period failing or abandoned sugar estates provided an opportunity for entrepreneurs to establish alternative agroindustries and convert the old sugar plantation infrastructure to facilitate the production of new exports. This process also involved repurposing the laborers from the previous system of enslavement. On islands like Dominica, the citrus lime industry emerged as inhabitants reappropriated and restructured the post-sugar-era landscape to accommodate transitions to wage labor. During the nineteenth century wealthier estates upgraded to emerging steam-powered technologies to process their crops; however, by the midnineteenth century sugar production was in decline and was no longer a viable cash crop for most eastern Caribbean islands (Pulsipher and Goodwin 2001:167). Confronted with bankruptcy, many of the older, less efficient sugar estates continued production by means of the newly emancipated laborers. These laborers effectively subsidized the deteriorating system with their underpaid labor through métayage or sharecropping (Fergus 1994; Mintz 1985). On Dominica as well as other islands, the downsizing of sugar production after emancipation (ca. 1840) and the decline of sugar-related manufacture left intact agroindustrial island landscapes (see Meniketti 2006 for similar processes on Nevis). During this post-sugar-boom era, several

Tracing the Postemancipation Landscape of Dominica’s Lime Industry · 49

attempts at developing other export crops reappropriated and restructured island landscapes. Since the eighteenth century Dominica’s landscape has experienced a series of boom-and-bust cycles for export crops that have dominated its agricultural economy, including coffee, sugarcane, limes, and bananas (Nelson 2010:224). Extensive investigation of the peasant economy created on Dominica following the British abolition of slavery (see Trouillot 1988, 1989, 1992, 1993) describes Dominica as a “patchwork of enclaves” scattered across a steep mountainous topography (Trouillot 1988). The extreme topography and thick vegetation of the island played a functional role in shaping human occupation and demarcating settlement boundaries. In this framework, the landscape mutually constituted colonial social relations and embodied movements of people and objects, particularly during critical periods of transformation and reconfiguration of island socioeconomic structures. Historically, the struggle over restructuring emergent economic and social relations in the postslavery era was highly contested, including debates surrounding wages, access to land, and economic participation through terms negotiated rather than imposed (Heuman and Trotman 2005:xx). This chapter investigates aspects of the citrus lime industry (ca. 1852–1928) that promoted changes to the landscape of Dominica and the movements of goods and people through a transition to new dominant form of agricultural production. Using archival-based information from the National Documentation Centre (Roseau) relating to the lime industry in Dominica and the Lesser Antilles more broadly, the research engages with earlier literature concerning Dominica’s history for assessing the evidence for land use, resettlement, and the organization of labor within the network of lime circulation. Meant as a preliminary foundation for exploring the extent to which the agricultural shift in industry affected island laborers, many of whom constituted the previously enslaved population, this study is by no means an all-encompassing account of the reorganization of social and economic landscape of Dominica. Rise of Lime Even before its commercial success, the lime was a ubiquitous part of island life, with a variety of uses as a foodstuff and medicine. Among the other fruit trees, vegetables, herbs, and medicinal plants, lime trees were a common feature in West Indian yards or home gardens. Prior to emancipation

50 · Samantha Ellens

these kitchen gardens helped sustain the diets of enslaved laborers and laid a foundation for supplementary income leading into the future (Brierley 1985:1). In West Indian cuisine, the lime features as a flavorant for sauces, marinades, and beverages. It also held medicinal value in treating fevers and as a general tonic (Quinlan et al. 2016:443). Limes even held a reputation as an aphrodisiac, and the burning of its rind was capable of “bring[ing] back love lost” (Goucher 2013:17). Today the lime and house garden remain a part of everyday life in the West Indian agricultural landscape. During the second half of the nineteenth century commercial success of citrus lime cultivation on the island of Montserrat by Joseph Sturge and his lucrative enterprise, the Montserrat Company, inspired the industry to develop on other nearby Lesser Antillean islands like Dominica (Fergus 1982). As lime cultivation increased, the fruit was exported across the Atlantic from Caribbean British colonies, where limes were manufactured into various products in the food, beauty, medical, and chemical industries. Additionally, Caribbean limes were a vitally important provision on British navy ships as a means to stave off scurvy (Daily News 1910). At the beginning of the twentieth century Dominica would eventually come to surpass the fame of Montserrat for lime production, as Dominica became the world’s foremost exporter of the fruit, and several of its estates would substitute monocrop sugar production with monocrop lime production (Baker 1994:143). These two islands remained linked through their competing industries, interisland migration shifts, and cultural interactions, and each contributed to the shaping of the other’s landscapes of lime labor into the twentieth century (Honychurch 2003). The introduction of the lime industry to Dominica is largely attributed to Dr. John Imray, a Scottish physician who came to the island in 1832 and established lime groves on his estate in 1860 (Royal Botanic Gardens 1894a:113). Imray took notice of lime trees that grew unused fruit alongside the road on the way to his estate (Baker 1994:142). Seen as a versatile fruit, limes were commercially lucrative for the sake of their juice (and, later, other elements), which could be shipped to Europe and the United States to be manufactured into high-end food, beauty, medicinal, and chemical products (Nicholls and Holland 1929:155). Unlike the sugar, coffee, and cacao industries, the diversity in marketable products appeared to ensure a basic element of stability for the lime industry, as sales for the variety of products obtained from the fruit were not likely to rapidly decline all at once (Harrison 1935:72). Learning the many uses for lime, Imray studied the pro-

Tracing the Postemancipation Landscape of Dominica’s Lime Industry · 51

cess of extracting citric acid from Sicilian lemons and established his own factory. With Imray’s early success, he eventually entirely replaced the cultivation of sugar with limes on his estates (Baker 1994:142; see also Cracknell 1973:80). Other estate owners followed his example and turned to the cultivation of limes in lieu of sugar. The effect of Imray and other estate owners turning to cultivate this new crop was a marked increase in lime production during the early 1870s (Trouillot 1988:60). Between 1843 and 1873, exports of lime juice out of Dominica drastically increased from a meager 43 gallons to 7,317 gallons (Trouillot 1988:60–61). When the value of lime products increased and the industry gradually expanded, limes became the island’s leading export at the turn of the twentieth century (Honychurch 1995; Nelson 2010). As the establishment of limes as a major export gained momentum, English historian James Froude (1888:140) visited Dominica and wrote about the “ruins” of the island’s estate complexes and the sugarcane fields turned over to alternative uses. Through the British Encumbered Estates Act of 1854, which enabled the sale of properties laden with complicated debts, plantation holdings were consolidated and purchased by enterprising capitalists for record low prices (Green 1991). It was through this act that entrepreneurs established their agricultural enterprises, employing the freed population in citrus lime production. In the first years of the twentieth century, the L. Rose and Company had at least three estates in operation (acquiring more later) and became the main exporter of limes and lime products on Dominica; however, several local “colored” merchants also exported produce from their own estates (Green 1999:53). It should be noted that the planter class consisted primarily of white proprietors, merchants, attorneys, and, at times, even overseers and bookkeepers (Higman 2005). Although there existed some free people of color “with special privileges” who were able to somewhat integrate into the planter class, racial solidarity provided the undercurrent for class status within the colonial Caribbean in which whites shared common interests and perhaps an affinity to each other rather than to members of the working class, who were largely of African descent, whom they commonly degraded and oppressed (Delle 2014:110). Table 3.1 demonstrates the sharp increase in the value of Dominica’s lime export during the latter part of the nineteenth century. Besides whole fresh fruit, lime products were exported in several forms for a variety of purposes including concentrated and raw lime juice, essential oil, otto of limes

52 · Samantha Ellens

Table 3.1. Value of lime juice exported from Dominica (£), 1870–1896 Year

£

Year

£

Year

£

1870

83

1879

3,144

1888

7,392

1871

170

1880

No Record

1889

7,415

1872

691

1881

6,427

1890

9,175

1873

738

1882

5,102

1891

9,123

1874

1,600

1883

4,620

1892

11,189

1875

1,823

1884

3,315

1893

14,539

1876

1,834

1885

3,301

1894

10,986

1877

1,557

1886

5,076

1895

12,055

1878

2,267

1887

8,077

1896

14,933

Source: Dominica Colonial Reports—Miscellaneous 1903, 47.

(a “superior” essence of lime), and pickled limes (Royal Botanic Gardens 1898:99). Concentrated lime juice was used for the preparation of citric acid used by calico printers in the textile trade (Anonymous 1894:328); and whereas the British government outfitted the Royal Navy with lime juice to combat vitamin deficiencies among its sailors, Americans introduced fresh fruits and extracts as an ingredient in alcoholic beverages (Trouillot 1988:61). As Dominica secured its position as the world’s largest producer of high-quality limes and lime products in the early decades of the twentieth century, multinational companies established four factories for manufacturing citrate of lime and another was built to produce citric acid crystals (Green 1999:53). L. Rose and Company In 1865 Lachlan Rose of Leith, Scotland, founded the L. Rose and Company to supply ships with various provisions including lime juice (Newton 2008). The Merchant Shipping Act was established in 1867, which required seafaring vessels to carry lime juice as a preventative against scurvy. This caused a significant increase in demand for the product. The L. Rose and Company saw increased sales of lime juice to ships, eventually becoming one of the most successful enterprises in Dominica’s agricultural history and contributing to the success of the citrus lime trade on Dominica (particularly during its height from 1903 to the mid-1920s). The company was in continuous

Tracing the Postemancipation Landscape of Dominica’s Lime Industry · 53

operation on the island up until its rapid decline in the 1970s (Honychurch 1995). The L. Rose and Company facility on Dominica’s southern coast provides a direct example of the transition from sugar to lime production on the island. In 1891 Lachlan Rose secured his company’s lime supply by purchasing two estates on the island and converting the old sugar factory in Soufriere for processing the fruit. Residents of the villages of Soufriere and Scott’s Head depended heavily on the company for employment as the entire coast surrounding Soufriere Bay was densely lined with lime trees and devoted to the industry (Honychurch, personal communication). The facility acted as the centralized point of manufacturing for several forms of lime products for which the surrounding estates in the Soufriere valley (e.g., Morne Patate and Morne Acouma) brought their lime crop to be processed. Given its well-preserved condition today, the facility offers potential as a case study for understanding the daily operations and transitions in lime processing on the island more broadly. The L. Rose and Company facility is centrally located in the village of Soufriere. An array of stone structures, including a watermill, boiling house, and warehouse remain largely intact and are highly visible from the paved roadway in town. Several existing components of the processing facility, such as an iron waterwheel, stone chimney tower, and furnace inscribed with the manufacturer’s mark “George Fletcher & Co Ltd., Derby” are related to the period of sugar production within the Soufriere estate complex and were later repurposed and employed for the processing of limes under the L. Rose and Company. The reuse of these elements speaks to the highly transferrable nature of sugar technology to the production of limes in the West Indies and denotes the ephemeral nature of the archaeological signatures relating to these functional changes (Naftel 1897:40). Nevertheless, some architectural modifications to the facility do act as material traces directly relating to the period of lime production under the L. Rose and Company. Evidence of technological development and increased transportation efficiency related to the production of limes are still apparent in Soufriere Valley, with many remnants of the upgrades to equipment and infrastructure that were necessary to export the increasing surplus of lime crop. Examples of technological improvements associated with the lime industry provide insight in the material ways in which the landscape was transformed throughout the trajectory of the lime industry with respect to how space

54 · Samantha Ellens

Figure 3.1. Pulley for cableway located on the north side of the L. Rose & Co. facility in Soufriere.

was subsequently repurposed and used. These include the construction of a jetty in Soufriere Bay in close proximity to the L. Rose and Company facility for exporting lime products via ship as well as an aerial cableway or “coulees,” which brought the limes from the surrounding hills to the processing facility to be crushed and boiled (Figure 3.1). Similar cableways have been used in other agricultural (e.g., bananas, sugarcane, tea leaves) and nonagricultural industries (e.g., mining, lumber) both to reduce costs and as a means for transporting cargo for long

Tracing the Postemancipation Landscape of Dominica’s Lime Industry · 55

distances over rough terrain. The development of freight ropeways prior to the twentieth century presented a new method of agricultural transportation (Booth 1965:8). Cable transportation has several advantages, including route flexibility (independent of road or bridge availability), high efficiency (follows the most direct route over varied terrains), and relatively lower operational costs than other methods of transport. In the “patchwork of enclaves” that make up Dominica’s rugged mountain terrain, this would be highly advantageous for enterprises like the L. Rose and Company because it would increase efficiency in both the time and effort required to transport products and materials between estates. As a fresh fruit, limes were more expensive and technically difficult to process than other products, such as dried cocoa, due to their greater bulk and fragility (Green 1999:52). The mechanization of transportation at Soufriere would have potentially decreased labor requirements as a means for more efficient transport across the valley, especially considering that the system previously would have relied upon laborers or pack animals transporting harvested fruit down from the orchards. Further transportation to ports was dependent upon the limited island road system, described in the 1860 Dominica Almanac as having a “peculiarly rugged and tortuous character” (Dominica Almanac 1860:26). The reorganization of the landscape through improved accessibility across enclaves, transportation networks, and infrastructure refinement demonstrates a localized shift in how the natural terrain contributed to shaping the design of the lime industry and its postemancipation relationships. Landscape of Limes With its massive changeover in agriculture from sugar to limes, Dominica offers an interesting case for exploring the ways in which the Caribbean landscape was constructed following emancipation. Understanding how aspects of the lime industry changed Dominica’s postemancipation landscape reveals how the struggle for landownership, by both planters and laborers, had political implications on the shifts in control over social and economic spaces and settlement patterns and social relations between classes. The transition from cultivating sugarcane fields to limes orchards, although lucrative for planter elites, was not a means of immediate improvement for the welfare of the laboring class on Dominica. The newly freed population desired land of their own and wanted an impartial judicial system governing

56 · Samantha Ellens

its tenure. Instead, they frequently encountered an oppressive system of labor control. Across the Caribbean, freed peoples were tackling similar grievances in the wake of emancipation as “indeed, some of the people believed that the boon of freedom was incomplete, if not insecure, without a small piece of land being attached to it” (Dominica Almanac 1864:55–56). In the case of the Soufriere Valley, much of the agriculturally viable land was appropriated for lime production, and labor opportunities outside of lime cultivation decreased. While sharecropping or métayage subsistence strategies existed as a means for Dominicans to support themselves, the colonial administration and local planters did everything possible to deter the growth of the free peasantry and disrupted the expansion of an autonomous peasant economy amid the cycles of active plantation monopoly throughout the nineteenth century (Green 1999:46). In her chapter of this volume, Khadene Harris discusses in detail the organization of house sites and the emergent labor networks after slavery in the nineteenth-century village occupation at the Morne Patate Estate. A transition away from subsistence farming into lime farming or wage labor on lime estates created an unequal power dynamic in which Dominican peasants struggled for capital and access to land. In the quest for social and economic freedom, free laborers desired access to land as well as opportunities for reasonably compensated work on the estates. On the contrary, local planters largely sought to limit wages on the estates and to restrict the movement of their laborers when off the estates (Heuman 1995:132). In the wake of British emancipation, the secretary of state for the colonies, Lord Glenelg, explained to the West Indies governors that in order for the economy to prosper, it was essential to “make it the immediate and apparent interest of the negro population to employ their labor in raising them” (Lockhart 1872, quoted in Berleant-Schiller 1995:60). This could be accomplished through hampering the laboring class’s ability to acquire land, consequently pressing them into wage labor on various estates. The fear regarding the loss of planters’ control over land is further expressed in a circular dispatch that same year, as the secretary of state called for information regarding the probable effects “should Crown Lands fall into the possession of individuals not possessing a proprietary title to them, or be even purchased by ‘persons without capital’ to cultivate them”(Lockhart 1872:55). Such activity was seen as a pestilence by the planters, who felt such relinquishing of land would diminish the supply of available field hands for wage labor on their estates (Berleant-Schiller 1995:61).

Tracing the Postemancipation Landscape of Dominica’s Lime Industry · 57

Since many of the emancipated laborers were raising their own food on the available vacant land, the laborers were not completely dependent on wages for their livelihood, which was seen as the key to plantocracy power (Delle 1999:140). In 1847 plummeting sugar prices, brought on by competing alternative global methods of production, forced large areas of cane field out of cultivation (Berleant-Schiller 1995:60). Left abandoned, these large, unprofitable tracts of land encouraged squatting, as individuals could remain undetected for a great deal of time and not be subject to tenancies. Purportedly, “fine virgin soil, woods totally free from any poisonous reptiles, filled with mountain cabbage trees and wild yams, and abounding with wild animals and birds suitable for food” were desirable resources left undeveloped, as “the interior of the Island, inaccessible for want of communication to the grower of the staple productions, presents to the poorer class of laborers temptations . . . which to persons of their habits would prove almost irresistible” (Lockhart 1872:56). Squatting was one of the most effective pathways for the Dominican peasantry to navigate the barriers for autonomous access to land in the political landscape of postemancipation Dominica. Occupation in the mountains had its appeal, as the necessary judiciary or military bodies were often underdeveloped or entirely absent in these areas. It was likely that individuals could remain undetected for a great deal of time as long as no estate owner sought immediate interest in that particular area; those residing along the coast occasionally worked on adjoining estates (Trouillot 1988:92). In the decades that followed, the choice of laborer plots eventually came to affect juridical and commercial territories, and the new configurations on the landscape would further provoke changes in labor, property, and distributive relations (Trouillot 1988:90). In July 1875 the introductory speech to the Third Legislative Assembly by Gov. Charles Monroe Eldridge noted the rise in the number of squatters, and later that year government botanist Henry Prestoe asserted that squatting was “one of the greatest evils the Colony now suffers from” (in Trouillot 1988:92). The failure to deter squatting was in large part beyond the attempts to identify the squatter due to the real impediment of the government’s ability to identify the boundaries of Crown land. Since the mid1800s, the governor’s unrestricted control of the Crown Lands Fund had used the fees that were collected from rentals or sales to ameliorate highprofile public officials, subsequently diminishing any chance at verification (Trouillot 1988:93). In 1894 Special Commissioner Robert Hamilton stated:

58 · Samantha Ellens

There is no proper and complete record of Crown lands nor private estate on the island. The Government have [sic] no means of knowing whether the lands on which many persons are living are their own private property, or whether they are the Property of the Crown. (in Trouillot 1988:93) With the lack of or imprecision of documentation, the chances of effectively detecting and expelling squatters on already abandoned or otherwise vacant land was largely ineffective. Squatters could work the lands without the imminent threat of eviction long enough to accrue a small savings from the sale of their supplementary provisions to formally legalize their possession of the plot in question or another elsewhere (Trouillot 1988:90). Squatting was consequently “an ephemeral configuration because of the limited security it provided for peasant labor process, a fragility reflected in the dislocation of the different relations of production” (Trouillot 1988:93). Nonetheless, there remained several regional and global dynamics within the British colonies that favored the planters’ desire to keep land out of the reach of most workers, including various taxation, forced labor, commodity politics, innovations in technology, and importations of labor (BerleantSchiller 1995:60). One of the most significant factors in supporting plantocracy resistance was British colonial policy, which encouraged limited access to property even where land was plentiful, like it was in Dominica (Moore 1987). Generally, the sale of Crown land was strategically priced in order to remain “out of reach of persons without capital” and always sold publicly to the highest bidder above a set minimum, with the goal not to force the cultivation of the present staples, by depriving the negroes of every other resource for subsistence, but merely to condense and keep together the population in such a manner that it may always contain a due proportion of laborers. (Great Britain 1846:3–4) A passage in the 1872 Almanac of Dominica identifies how the sale of Crown land was strategically priced to remain just out of reach of the laboring classes: To discourage the extension of cultivation, or to confine it to any particular field, is by no means desirable; but some security should, if possible, be taken, that all territory which is cultivated at all be cultivated well. The minimum price of land, therefore, shall be high enough to leave considerable portion of the population unable to buy it until

Tracing the Postemancipation Landscape of Dominica’s Lime Industry · 59

they have saved some capital out of the wages of their industry; and at the same time low enough to encourage such savings by making the possession of land a reasonable object to all. (Lockhart 1872:56) It would seem that land was priced steep enough to reinforce a landless working-class population but not too high as to completely discourage those industrious individuals from attempting to save enough from their wages in the hopes of purchasing their own plots (Berleant-Schiller 1995:60). Migration and Movement As the burgeoning commerce in lime cultivation expanded across Dominica, decreasing employment opportunities other than those associated with lime cultivation and difficulty in obtaining land impacted the resettlement circumstances associated with the shift in industry. During the early nineteenth century, most Dominican villages were located on the leeward coast, and surrounding lands were used for the cultivation of limes and cash crops. With a new economy demanding as much lime production as possible, land and the control over labor were at a premium as estates began to focus efforts toward increasing their outputs. Production facilities bought and processed produce from a variety of estates in the surrounding area and the labor itself was drawn from tenants and adjacent village populations. In 1844 the majority of the Dominican population resided on or near agricultural estates, or in detached shelters and villages along the coast (on Crown land known as the “King’s Three Chains”), with further settlement occurring on the abandoned coffee estates high up in the mountains (Dominica Almanac 1860, Honychurch 2001). Lime orchards, in particular, quickly became a dominating feature of the larger valleys (Hodge 1943:364). As land was reallocated to support increased lime cultivation, shrinking habitation areas caused the boundaries within village enclaves to shift as populations migrated to other more hopeful locales in search of alternatives (Trouillot 1988:193). The Soufriere Enclave on the island’s southwest coast contained a high density of agricultural properties, the largest of which included the Soufriere Estate and Morne Patate Estate. Like many other rural agricultural villages on Dominica, Galleon developed during the late nineteenth century within the Morne Patate hills as Afro-Caribbean descendants sought to establish independent communities away from the estates on which they

60 · Samantha Ellens

had formerly lived and worked (Honychurch 2001; Slocum 2017:425). A cistern was erected for the village in 1909 and remains a physical marker of the labor displacement and village formation. Overlooking the bay with scenic views of the fishing villages of Scott’s Head and Soufriere, Galleon has endured as a small rural community of Afro-Caribbean descendants. A substantial movement of labor also occurred between islands as laborers were recruited from other Leeward Islands to alleviate developing labor shortages on lime estates. Montserratians emigrated to Dominica en masse following emancipation to take advantage of the demands for estate labor (Honychurch 2003). Dominica had more mountainous and remote Crown land available for squatting than other Leeward Islands (e.g., Antigua and Montserrat) and the Dominican peasantry appeared unenthusiastic to continue work on the estates in lieu of independent subsistence (Berleant-Schiller 1995:54). Montserratian laborers were actively encouraged to undertake work in Dominica to replenish the shortfall in labor; as a result, the history of lime and cocoa production on Dominica closely tracks the peak years of Montserratian immigration to the island (Honychurch 2003). These types of labor migrations continued into the twentieth century as more than 2,000 laborers, primarily from Antigua and Montserrat, were recruited to work on the lands available in Dominica’s interior and the expanding lime industry (Green 1999:48). Cost of Cultivation The commercialized production of the newly introduced lime crop required a tremendous amount of not only monetary but also temporal commitment for planters to see any sizable returns on their investment. Lime seemed much less labor-intensive than sugarcane as orchards required far less planting than fields of cane (trees were to be planted three times as far apart as sugarcane stalks) and would benefit from as little human intervention as possible as opposed to the continuous cropping of sugarcane rattoons (Nicholls and Holland 1929:136, 155). Nevertheless, lime trees took up to ten years to reach full bearing (Royal Botanic Gardens 1894a:408). This meant that the enterprise required not only the startup costs to purchase land and establish orchards but also coverage of all yearly expenses associated with cultivation and processing for several years before receiving any profitability in the venture (Royal Botanic Gardens 1894b:116–117).

Tracing the Postemancipation Landscape of Dominica’s Lime Industry · 61

Unfortunately, these assets were largely out of reach for the majority of Dominica’s peasantry and were enough to discourage most small proprietors from attempting cultivation of the crop (Trouillot 1988:192). Despite being considered a hardier crop, limes were more expensive and technically difficult to process and transport than dried cocoa beans, which were preferred by the peasantry (Green 1999:52). If a prospective planter had enough capital to invest in lime production and the ability to wait the maturation period before seeking returns, this appeared to be a particularly sound longterm investment for those who could afford it. For the individuals laboring on the estates, agricultural reports for Dominica in 1898 show that the average wages for laboring on estates differed by gender and age, ranging from 9 pence to 1 shilling per day for men, 6 to 8 pence for women, and 3 to 6 pence for children (Naftel 1898). Work was paid for by the task, with weeding done by women and children. The laborers who lived on the estates were often tenants who paid rent for their gardens and worked for the landowner when required. On some estates, laborers were required to provide one day’s work a week without pay in exchange for dwellings and small gardens (Naftel 1898:20). To the southeast, on Pointe Mulatre Estate, tenants paid an annual rent of 30 pounds to the landowner. Economic control continued to be widely exercised by foreign elite as the benefits did not benefit the working class (Green 1999). If individuals did not own land, their prospects for economic mobility in what was predominantly an agricultural island remained improbable, as employment opportunities were scarce, aside from manual labor on estates. Comparatively, one appraisal of the monetary requirements for establishing a lime enterprise was given by a Mr. D. Morris in a paper entitled “On the Colony of the Leeward Islands,” read before the Royal Colonial Institute on April 14, 1891: It appears that it takes about £1,000 to purchase and establish 20 acres of good land with lime trees, to build a house for the manager, to erect a mill, with copper boilers to concentrate the juice, to pay for superintendence, and cover all expenses for seven years. At the end of this time the estate would yield at the rate of 40 hogsheads of concentrated lime juice, worth £12 per hogshead. This would amount to a gross income of £480. The yearly cost of cultivation and manufacture (including the cost of providing the hogsheads) would amount to £240. There

62 · Samantha Ellens

would, therefore, remain exactly £240.; and this sum would be the net income of a lime estate which had cost £1,000, spread over seven years. (Royal Botanic Gardens 1894b:116–117) From this account, lime cultivation required large monetary and temporal commitment via purchasing the land, necessary equipment, and covering expenses for the first seven years of lime maturation to see large returns on the initial investment. Unfortunately, the general prosperity related to lime cultivation did not last long into the mid-twentieth century as once again the island’s dependence on monocrop production proved devastating. The diversified market for limes and lime products could not protect the industry against losses obtained through the destruction of the trees themselves (Harrison 1935:73). A series of hurricanes beginning in 1916 and rampant lime tree disease in 1922 combined with several other factors to collapse the industry (Green 1999:49). In 1918 a U.S. embargo was placed on Dominican limes, followed by the end of demand for lime juice by the British navy; post–World War I, the island’s revenue had been reduced to the extent that government could not maintain its deteriorating roads (Baker 1994:51). Development of synthetic substitutes for citric acid and a drastic fall in cocoa prices also contributed to the breakdown of the island’s economic landscape during this period (Green 1999:49). Between 1921 and 1937 the acreage cultivated in limes had been reduced by two-thirds (Baker 1994:144; see also Cracknell 1973:87). Estates went bankrupt, and the rural Dominican peasantry was largely left to fend for itself, turning further into subsistence farming and residing in rural communities like Galleon in the Morne Patate hills. Conclusion The lime boom era intensified the marginalization of the Dominican peasantry since the fruit had fewer labor requirements than sugar (Baker 1994:145). Lacking the monetary and temporal investments required, the peasants could not afford resources necessary for establishing limes as a cash crop despite the seemingly lucrative commerce in limes during the nineteenth and twentieth centuries. Under these conditions, the livelihoods of these individuals were impacted through restrictive employment opportunities, constricting wages, mobility, and resettlement circumstances associated with the individual economic and social impacts aggravated by the

Tracing the Postemancipation Landscape of Dominica’s Lime Industry · 63

expansion of the lime trade. Accounting for the conditions (e.g., the material requirements and need for experienced labor) that allowed and encouraged movement (e.g., of profits, goods, people) during a critical period of transformation on Dominica highlights the dynamics and restructuring of the landscape that developed with the introduction of new postemancipation agroindustries. The landscape, in turn, contributed to the development of particular strategies, activities, and policies (e.g., the Encumbered Estates Act, the availability of land, settlement patterns, interisland labor flows) that shaped the trajectory of the lime industry on Dominica. This chapter has explored (aspects of) the transition to wage labor and the subsequent struggle for peasant autonomy that occurred during the changeover to a lime-based agricultural economy in the postemancipation context of Dominica and the British West Indies more broadly. Labor organization, landownership, and the socioeconomic strategies employed by individuals enmeshed within the lime industry speak to the significance and complexity of how people negotiate their environments through time and space. Research into the reorganization of space and economic ramifications of shifts to new agricultural industries, like the lime cultivation, are helpful for understanding not only the postemancipation dynamics of Caribbean lifeways in the postsugar era but also the ways that localized economic and political channels impacted the societal structure and legacy of such industries.

4

Y Building an Archaeological Chronology for Morne Patate Lynsey A. Bates, Jillian E. Galle, and Fraser D. Neiman

This chapter establishes an archaeological chronology for the Morne Patate Estate using a suite of statistical methods that Digital Archaeological Archive of Comparative Slavery (DAACS) researchers have used successfully to infer and evaluate similar chronologies on scores of archaeological sites from early-modern slave societies in North America and the Caribbean. While initially technical in its approach, this chapter is critical to understanding the evolution of the domestic core of Morne Patate. It provides a temporal framework for understanding how enslaved laborers, their descendants, and French and English owners and overseers created, reshaped, and used the landscape of Morne Patate between the mid-eighteenth and late nineteenth centuries. Our exposition begins with an outline of DAACS’s collaboration in the Morne Patate project. We offer a brief overview of the methods we employ, the models behind them, and the relationships among them: frequency seriation, mean ceramic dates, correspondence analysis, and termini post quem. We describe the resulting plantation-wide chronology for the estate (Locus 1) and village (Locus 2). We use the chronology to assign assemblages to temporal phases whose larger sample sizes facilitate the use of artifact abundance measures to highlight material trends across the plantation and through time. The Digital Archaeological Archive of Comparative Slavery and Morne Patate In 2015 the Digital Archaeological Archive of Comparative Slavery (DAACS; www.daacs.org) was asked to conduct the artifact identification

Building an Archaeological Chronology for Morne Patate · 65

and cataloging, context analysis and digitization, and map compilation and digitization for the Morne Patate project. Dr. Lynsey Bates visited excavations at Morne Patate in 2015 to help prepare artifacts for export and to advise on field recording methods. By 2016 all artifacts from the Morne Patate excavation seasons (2013–2016) had been exported to the DAACS Lab at Monticello, where each artifact was identified and cataloged into the DAACS database by DAACS staff using publicly and exhaustively documented DAACS protocols and standards.1 DAACS staff parsed and entered all Morne Patate field records into the DAACS database. Unit and feature plans were compiled and digitized by Bates. Complete archaeological data from Morne Patate, including artifactual, contextual, spatial, and image data, are now available to scholars and the public through www.daacs.org. Jillian Galle and colleagues (2019) provide details on the development and history of DAACS. Methods Building reliable chronologies is an essential first step in archaeological data analysis. In the case of the Morne Patate project, documentary and archaeological evidence pointed to occupations that span the British seizing control of Dominica from the French in 1763, the economic dislocations associated with the American Revolution, and ensuing decades of European wars (O’Shaughnessy 2000), the legal end of the transatlantic slave trade in 1807, and emancipation in 1834 (Ward 1988). These events, and the larger social and economic processes that underlay them, surely affected and were affected by the lives of Morne Patate’s residents, enslaved and free, and left traces in the archaeological record. Understanding this history requires placing Morne Patate’s archaeological assemblages in a chronological sequence and assigning calendar dates to segments of that sequence. The importance of a reliable chronology places a heavy epistemological burden on the methods and data we use for chronological inference. Over the long term, we can increase our own, our colleagues’, and our successors’ chances of success in establishing accurate chronological timelines by pursuing two complementary strategies. First, we want to harness analytical methods that expose flaws in potential interpretive claims that are based on them. To test chronological hypotheses, as one would other scientific hypotheses, we must account for flaws or errors, allowing us to evaluate if they are wrong (Mayo 2018). More on this below. Second, we need to

66 · Lynsey A. Bates, Jillian E. Galle, and Fraser D. Neiman

ensure that the methods, their implementation, the data, and the protocols used to capture the data on which our hypotheses are based are publicly shared and accountable (Marwick 2017; Marwick et al. 2017). To accomplish this, the data and protocols used below are available on the DAACS website (www.daacs.org). We have developed a reproducible workflow, described below. It is implemented in the R, a free, open-source statistical programming language. The code and data as well as an online supplement containing supplementary figures are available in an archival repository at the Open Science Foundation (see https://osf.io/52jfn/). We encourage readers to download the code, check our results, and adapt the code for their own research and data. How do methods facilitate the process of exposing flaws in claims based on them? We highlight two avenues. First, we need to be explicit about the models on which the methods are based. This makes it possible to check results for “goodness of fit” with a model’s assumptions and, critically, offers the opportunity to learn from a lack of fit. Second, we use two methods backed by different models so we can assess between-model agreement. Below we briefly review the models behind our methods and the relationships among them. Frequency Seriation Frequency seriation is the foundation for our approach to Morne Patate and other sites whose data are shared via DAACS. Frequency seriation is based on a simple model of how the relative frequency of types, usually of ceramics, varies across archaeological assemblages (Dunnell 1970; Lyman et al. 1997). The seriation model says that (1) there is temporal overlap among types—subsets of types do not appear and then disappear synchronously— and that (2) the type frequencies follow unimodal or battleship-shaped curves over time. The model implies that, given a data matrix with undated assemblages on the rows, types as columns, and entries the proportion of each type in a given assemblage, the row order in which the type proportions best fit the model is inferred to be a chronology. Culture historians realized that types needed to be defined so they are “historical”—that is, they had unimodal response curves against time. And the assemblages need to be large, similarly time averaged, more or less evenly sampled across a single temporal continuum, and derived from a single, spatially homogeneous “cultural tradition”(Dunnell 1970).

Building an Archaeological Chronology for Morne Patate · 67

Because the method is based on a model, it offers opportunities to expose flaws in hypotheses based on its application by assessing goodness of fit. The application can fail outright: it may be impossible to produce an ordering of rows in which the unimodal curves appear, suggesting the assemblages are effectively contemporary. Or it may be possible to achieve an order that more or less fits the model. In that case, identifying exactly where the data do not fit and investigating why offer learning opportunities. For example, amount of change in type proportions between adjacent rows may vary, revealing temporal gaps in the sequence or punctuated change linked to appearance in the sequence of assemblages generated by distinct cultural groups. Assessment of goodness of fit should advance our understanding of Morne Patate. If British control disrupted exchange networks and led to an influx of new social groups, we could expect to see a synchronous shift across all or most type frequencies in a seriated sequence. But checking the goodness of fit to the model is only one way to rule out flaws. A second essential avenue requires independent evidence that the seriation-derived order is in fact chronological. Types can demonstrate unimodal responses along nontemporal gradients, including geographical and social space (Kruskal 1971). So comparing an initial chronological hypothesis against one derived using different methods or data (e.g., stratigraphy, radiometric dating, a second seriation) is the key to detecting failures and to objective evaluation (on objectivity, see Kosso 2001). How can we realize the analytical potential of the seriation method at sites like Morne Patate, where we are confronted with tens of thousands of artifacts and scores of contexts? We need statistical methods backed by models that can then be related to the seriation chronology. We use two: mean ceramic dating and correspondence analysis. Mean Ceramic Dating When Stanley South proposed mean ceramic dating (MCD) nearly a half century ago, he recognized there was a connection to the unimodal response curves of the seriation model (South 1971). But the statistical details were only worked out by ecologists two decades later (ter Braak and Prentice 1988). MCDs are simply weighted averages of the historically documented manufacturing midpoints for each type found in an assemblage, where the weights are the type frequencies. An MCD offers a maximum-likelihood estimate of the mean date of an assemblage (the value that maximizes the

68 · Lynsey A. Bates, Jillian E. Galle, and Fraser D. Neiman

chance of getting the type frequencies in the assemblage) under a model in which type responses follow Gaussian functions, with means equal to manufacturing midpoints and identical variances. An obvious improvement to Stanley South’s MCD is to dispense with the equal variance assumption. To do that, we assume that type-manufacturing spans are 6 standard deviations long, so the standard deviation for each type is We then compute a BLUE (best linear unbiased estimate) MCD, weighting each midpoint (μi) not only by its counts but also inversely by its variance, so types with shorter spans get greater weight.2

The connection between the model behind MCDs and the seriation model is important because it immediately suggests a way to expose flaws in using MCDs to order assemblages into a proposed chronology. We check for goodness of fit to the model by sorting the assemblages on their MCD scores and plotting them in a seriation diagram. If the actual type proportions do not show unimodal response curves or an approximation to them, then time is not the primary factor in determining type frequencies. This simple check remedies a major defect in the MCD method: a data matrix of random numbers will return an MCD. Without checking goodness of fit to the underlying model, this flaw will go undetected. Other flaws in an MCD-based chronological hypothesis may be more subtle. As MCD-using archaeologists recognized almost immediately, the MCDs could be influenced by social or economic gradients as well as temporal ones (Turnbaugh and Turnbaugh 1977). For example, poor households might have fewer up-to-date ceramics than contemporary wealthy ones. This is one manifestation of a more general issue: the MCD model assumes that the Gaussian response functions of the types along calendar dates are the same in all times and places. Correspondence analysis can be seen as an attempt to address this issue by extracting estimates of the means of the types (analogous to the fixed manufacturing midpoints in MCD) from the data themselves so that they honor the actual trajectories of type responses in a set of assemblages from a particular time and place.

Building an Archaeological Chronology for Morne Patate · 69

Correspondence Analysis There are two complementary statistical motivations for correspondence analysis (CA).3 On the one hand, it is an ordination technique that offers archaeologists a way to visualize the pattern of similarity among a set of assemblages, based on the proportions of a large number of types that occur in them, by scoring them on a small number of underlying dimensions. Assemblages with similar type proportions will have similar scores and fall close together in a scatter plot of the dimension scores. The second motivation highlights its connection to the Gaussian response model behind MCDs and the battleship curves of seriation (Hill 1973; ter Braak 1985; ter Braak and Prentice 1988). Here the CA solution is derived from “reciprocal averaging.” We start by substituting random numbers between 0 and 1 for the type-manufacturing midpoints in the MCD equation. We then compute assemblage scores. Using the assemblage scores, we compute new type scores, as weighted averages of the scores of the assemblages in which the types occur. Then we compute new assemblage scores, and so on, until the scores stabilize. The result is a set of scores for the assemblages that place them on the gradient that we assume underlies their Gaussian responses as well as a “corresponding” set of type scores that, when scaled, estimate the positions of the types’ maximum popularity along the gradient. It is possible to compute additional underlying dimensions, up to one less than the number of types, using the same algorithm while constraining each set of scores to be uncorrelated with all the previous ones. Each successive dimension accounts for less variation, dubbed “inertia” in CA lingo, in the original data. A “scree plot” of the amount of inertia accounted for by each dimension against the dimension number often displays a steep decline on the first few dimensions, followed by a shallower one on the rest. The leading dimensions, to the left of the elbow, usually portray the substantively important variation among the assemblages. The price we pay for being able to use CA to accurately estimate assemblage locations on a single gradient is additional assumptions. The assemblage and type scores, which comprise the first dimension of CA, approximate maximum-likelihood estimates under a model in which the locations of the assemblages and the type maxima are uniformly distributed along the gradient and have Gaussian responses with equal variances (ter Braak 1985). A simple way to detect flaws in those assumptions is to assess goodness of

70 · Lynsey A. Bates, Jillian E. Galle, and Fraser D. Neiman

fit to the model: sort the assemblages on their dimension-1 scores and plot them on a seriation diagram. We should see unimodal responses if there is a single dominant gradient and if sampling error has not overwhelmed the unimodal type responses. In that case, and if the gradient is long enough relative to the variance of the responses, we also expect to see the “arch effect” in a scatter plot of the dimension-2 scores against the dimension-1 scores, where the former are a quadratic function of the latter. Unlike frequency seriation and MCDs, CA can help archaeologists identify cases in which a second gradient determines type responses independently of the first. For example, dimension-1 scores may register time while dimension-2 scores can simultaneously measure functional variation, say, food preparation versus consumption. Plotting CA scores against MCDs offers a way to identify which CA dimension captures a temporal gradient. It is usually the first, but we have encountered examples in which synchronic spatial gradients overwhelm a temporal signal that emerges on CA dimension 2. The CA dimension that is best correlated with MCDs is more likely to measure time for those assemblages than the MCDs themselves since the scores arise from the unique history of type trajectories in the analyzed assemblages. We check flaws in this expectation by comparing seriation diagrams with rows sorted using CA scores against those with rows sorted by MCDs. CA scatter plots allow us to see clustering among assemblages and gaps between them. These indicate discontinuities in type frequencies that may register depositional hiatus or sudden shifts in the identity of site’s occupants, their economic networks, or their activities. These patterns can offer useful clues to the historical processes that generated the assemblages. They also point to violations of the assumptions in the model behind CA. A critical assumption is that assemblages are uniformly distributed across the dominant (temporal) gradient. When they are not, variation in the gradient may actually be distributed across both CA dimensions 1 and 2. We will encounter an example at Morne Patate where dimension 1 separates early assemblages from later ones, while dimension 2 captures time within the later assemblages (for another example, see Smith and Neiman 2007). Terminus post quem The terminus post quem (TPQ) method is ubiquitous in historical archaeology. We use it as an additional check by plotting CA dimension scores for

Building an Archaeological Chronology for Morne Patate · 71

each assemblage against their TPQs. The two orders will be correlated if both measure time. However, we do not expect the correlation to be perfect. In theory, CA places assemblages on a continuous temporal gradient while TPQs can take only a few discrete values established by documented beginning type-manufacturing dates. Used alone, the TPQ method invites archaeologists to slip from the warranted inference that a deposit could date any time between the TPQ and the present to the unwarranted inference that it probably dates soon after the TPQ. The latter interpretation assumes that the absence of evidence for later types in an assemblage is evidence of their absence on the site when the deposit was sealed. This inference is only legitimate if we can show that the assemblage in question fits into a chronological sequence of assemblages in which the absent later types eventually do appear. This demonstration is the missing ingredient that the temporal sequence of assemblages from frequency seriation and related methods provides. Since TPQs are highly sensitive to stratigraphic excavation errors that may introduce one or two intrusive later sherds, we include robust estimates, TPQp95 and TPQp90, which are the 95th and 90th percentiles of the beginning manufacturing dates of all the sherds in an assemblage. Phases The individual assemblages that comprise the units for chronology building at Morne Patate have sample sizes that are too small to allow us to measure reliably temporal and spatial variation in the abundance of many artifact classes, particularly “small finds.” We need to use our chronology to construct counting units with larger samples. To do that, we use CA dimension-1 scores that our analysis indicates measure time reliably. To help identify clusters of assemblages with similar dates, we use weighted histogram of dimension-1 scores, where the weights are the sample sizes of the assemblages supplemented by kernel density estimates (Baxter et al. 1997). We assess the likely calendar dates of the phases using MCDs and TPQs. Analytical Workflow We have provided a brief outline of the methods we use to build archaeological chronologies along with the models on which they are based. Clarity on the models and their relationships enables us to use the methods to find flaws in the hypotheses we derive from them. We have assembled the

72 · Lynsey A. Bates, Jillian E. Galle, and Fraser D. Neiman

methods into a flexible, flaw-finding workflow composed of the following steps: 1. Compute the CA. Use a scatter plot of assemblage and type scores on the leading CA dimensions to identify outliers that may distort the results, eliminate them, and try again until results stabilize. 2. Once a stable CA solution is achieved, use the scree plot of inertia (variation) accounted for by each dimension to identify the dimensionality of the data. Focus on these dimensions. 3. Compute the MCDs, BLUE MCDs, and TPQs for the assemblages. 4. Use scatterplots of CA dimension-1 and -2 scores for ware types and for assemblages versus MCDs to identify which dimension(s) capture time. 5. Check fit of the CA solution to the frequency-seriation model by sorting the assemblages on the temporally sensitive dimension and producing a frequency-seriation diagram. 6. If both CA dimensions register time, identify the earliest cluster of assemblages on dimension-1 and assign this cluster to DAACS Phase 1. Compute weighted histograms and kernel density estimates to help identify the Phase 1 cluster. 7. Compute the CA without the Phase 1 cluster and circle back to steps 2–5. 8. Use weighted histograms and kernel density estimates of dimension-1 scores to help identify clusters of assemblages with similar positions along the temporal gradient. Assign the clusters to DAACS phases. 9. Compute MCDs, BLUE MCDs, and TPQs for the phases. Developing a Plantation-Wide Chronology We employed the foregoing workflow to develop a plantation-wide chronology for Morne Patate ceramic assemblages. The assemblages are derived from two areas. “Locus 1” is the presumed domestic complex that housed the estate’s owners and managers. It includes three excavation blocks: the Estate Block, the Stable Block, and Block E. “Locus 2” is the presumed site of the slave village and includes six excavation blocks: Blocks A, B, C, D, F, and G. The excavation blocks are composed of one or more two-meter quadrats. A third area, “Locus 3,” is thought to be the site of a provision ground. It was

Building an Archaeological Chronology for Morne Patate · 73

investigated using shovel test pits, which did not yield ceramic samples large enough to be included in this analysis. Implementing the methods described above requires counts of ceramic types in assemblages. We used the DAACS Ware Type field. DAACS Ware Types and their manufacturing date ranges are defined in the DAACS Ceramics Manual (https://www.daacs.org/about-the-database/daacs-cataloging-manual/). We know from decades of work that the DAACS Ware Types are likely to be “historical types” and therefore might be expected to yield respectable fits to the seriation model. We created assemblages using the archaeological contexts identified by the excavators in the field. Where we felt confident that several contexts in the same excavation quadrat or in adjacent quadrats belonged to a single lithological unit or layer, we assigned them to a DAACS stratigraphic group (SG) and used the SG to aggregate ceramic sherds into assemblages. To reduce the noise introduced by sampling error, only ceramic assemblages with more than five sherds and more than two ceramic types were included in our initial analysis. We excluded assemblages from topsoil, unit cleanup, and surface collections. Chronology building using CA is an iterative process. After running a first analysis, it was clear that several types were poor fits to the seriation model in these data: they occurred at low frequencies and in an unpredictable fashion across assemblages. These types could be identified because their dimension-1 and -2 scores made them outliers relative to the other types. And they made the assemblages in which they occurred outliers as well. We removed them from this analysis (American Stoneware; Astbury Type; Black Basalt; Delftware, Dutch/British; British Brown/Fulham Type; Nottingham; Refined Earthenware, modern; Refined Stoneware, unidentifiable; Saintonge). Three coarse earthenware types, Morne Patate Type 1, Morne Patate Type 1a, and Morne Patate Type 1b (see Bloch and Bollwerk, this volume), share similar attributes and were in close proximity along dimension-1 in this initial CA. We combined them into a single type that we called Morne Patate Type 1 Combined. After these adjustments, we achieved a CA solution that reveals fit and lack of fit to the seriation and CA models. Both are informative. A scree plot of the amount of variation (inertia) accounted for by the successive CA dimensions suggests that these data can be usefully summarized in a threedimensional space (Figure 4.1a). Patterns in the scores of types and assemblages across these three dimensions are revealing. The first dimension registers time: types that we know are early (White Salt Glaze, Tin-Enameled)

Figure 4.1. a, Scree plot of the proportion of inertia accounted for by the successive CA dimensions. b, Plot of ware-type scores on the first two CA dimensions; note the two types with high dimension-1 scores: “White Salt Glaze” and “Tin-Enameled, unidentified.” c, Plot of the assemblage scores on CA-dimensions 1 and 2; dark gray points denote Locus1 assemblages; light gray points denote Locus -1 assemblages. d, Plot of the assemblage scores on CA-dimensions 1 and 3. e, Plot of BLUE MCDs on dimension-1 assemblages scores; note the correlation in the early part of the sequence, but not in the late part. f, Plot of BLUE MCDs on dimension-1 assemblages scores; note the correlation in the late part of the sequence, not on the early part.

Building an Archaeological Chronology for Morne Patate · 75

have high scores while types we know are late have low scores (Whiteware) (Figure 4.1b). But dimension-2 also has a temporal component. The dimension-2 scores of Creamware, Pearlware, and Whiteware are negatively correlated with their known manufacturing midpoints. Plotting the assemblage scores reveals a U-shaped pattern—an arch— that is typical of CA solution for data arising from long temporal gradients along which type frequencies wax and wane in popularity (Figure 4.1c). Assemblages on the right side of the arch are early. The plot makes it clear that the Locus 1 assemblages from the estate are distributed across the arch. But the Locus 2 assemblages from the village area are concentrated on the left and later side. There are two assemblage clusters: a diffuse one composed of the early assemblages and a dense one composed of the later assemblages. The gap between them indicates the assemblages are not uniformly distributed along a single underlying gradient. The gap may register a temporal hiatus or, more likely, a shift in what cultural historians would have called the “cultural tradition” of the people generating the assemblages. The discontinuity in type-frequency trajectories is also registered in a seriation diagram of type frequencies in assemblages sorted by CA dimension-1 scores. It shows the early assemblages at the top of the diagram are distinctively dominated by White Salt Glaze and Tin-Enameled ceramics. The shift may register the British takeover of the island. What about dimension-3? Dimension-3 separates the early Locus 1 assemblages into two groups (Figure 4.1d). One (contexts 6182, 6224, 6225) is dominated by White Salt Glaze; the other, by Tin-Enameled (contexts 6210, 6208, 6155, Block E SG01, 6209) (online supplement, Figure 2). Both clusters contain contexts from the Estate Block and Block E. One Locus 2 context (Block A SG04) occurs in the latter cluster, raising the possibility that its contexts may be more closely related to the village-area occupation. On the other hand, there is no indication that there is anything but random variation between the Locus 1 and Locus 2 assemblages that fall into the later assemblages. The CA scores of the later assemblages from the two loci are indistinguishable, revealing that, when measured in terms of ware-type variation, they are part of the same evolving tradition. Our conclusion about the chronological significance of both dimensions-1 and -2 can be evaluated by plotting their scores against BLUE MCDs (Figure 4.1e and 4.1f). In the former case, we see a strong, negative, and nonlinear relationship. In the latter, we see an equally strong, negative

76 · Lynsey A. Bates, Jillian E. Galle, and Fraser D. Neiman

correlation for later assemblages, confirming our suspicion that dimension-2 captures time for them. Since variation on both dimensions-1 and -2 is chronologically significant, we cannot simply use dimension-1 scores to assign assemblages to chronological phases. A simple way forward is to identify those assemblages that are most important in determining the dimension-1 scores that separate the early from late assemblages. We can then remove them from the dataset and perform a second CA for the remaining assemblages, hoping that a single set of new dimension-1 scores will capture the chronological variation in the later part of the sequence. If it does, we can assign those assemblages to later phases based on their new dimension-1 scores. We created a histogram and a complementary kernel density estimate of dimension-1 scores, where the vertical axis measures ceramic assemblage size (online supplement, Figure 3). It shows that a majority of the assemblages and the ceramics they contain fall in a single cluster on the far left of the histogram. They have dimenion-1 scores less than 0.8. The highly skewed shape of the histogram is itself informative. There are at least two hypotheses that might explain it. First, the later phases of the occupation at Morne Patate witnessed a dramatic increase in either the number of (enslaved?) people living at the site, or in their per capita discard rate of ceramics, or both. The second possibility is that excavated quadrats happened not to intersect areas occupied by the bulk of Morne Patate’s residents in the early phase of the occupation. In other words, most of the site’s residents moved into the sampled areas in the later period. We explore change in occupation intensity in the next section of this chapter. To isolate and explore further the structure of the later assemblages, we assigned assemblages with scores greater than 0.8 to plantation-wide Phase 1 and computed a second CA solution without them. The scree plot of the new CA solution shows a dip in inertia values between dimensions-2 and -3, suggesting that, in this case, two dimensions summarize patterning in the data (online supplement, Figure 4). There is no indication of a significant third dimension. The plot of dimension-1 and -2 scores for types shows that dimension-1 captures time: earlier types have high scores (Creamware); later types have low ones (Whiteware) (online supplement, Figure 5). Dimension 2 may capture synchronic variation in assemblages whose significance remains opaque. Earlier assemblages vary more on dimension 2 than later ones (online supplement, Figure 6), with high dimension-2 scoring assemblages having more unidentified Caribbean Coarse Earthenware,

Building an Archaeological Chronology for Morne Patate · 77

Table 4.1. Chronological indicators for Morne Patate plantation-wide phases Phase

MCD

BLUE MCD

TPQ

TPQp95

TPQp90

Count

Unassigned

1823

1801

1840

1820

1820

180

P01

1768

1766

1830

1762

1762

320

P02

1797

1792

1840

1820

1775

1129

P03

1818

1800

1840

1820

1820

1858

P04

1843

1810

1840

1820

1820

3513

P05

1866

1824

1840

1820

1820

813

Morne Patate Type 1, Faience, and Redware, and low-scoring assemblages having more Albisola, Red Agate, and Frechen Brown. The overall similarity of the scores for Locus 1 and Locus 2 assemblages again suggests they are part of a single, evolving “tradition,” when variation is measured using ceramic ware types. We evaluated the inference that new dimension-1 scores represent time by plotting them against their BLUE MCDs (online supplement, Figure 7). As expected, the relationship is strong and negative. On the other hand, there is no relationship between dimension-2 scores and BLUE MCDs (online supplement, Figure 8), indicating that dimension-1 completely captures chronological variation in these later assemblages. We assigned each of the later assemblages to one of four plantation-wide phases, Phase 2 through Phase 5, based on the location of dips in the weighted histogram and kernel density estimation of dimension-1 scores (online supplement, Figure 9). We combined these four plantation-wide phases with the single early plantation phase (Phase 1) identified in the earlier CA, producing a total of five plantation-wide phases. These phases are groups of assemblages that have similar CA scores and similar MCDs and are therefore inferred to be broadly contemporary. We computed MCDs, BLUE MCDs, and the three TPQ measures for each plantation phase (Table 4.1). They suggest that Morne Patate was occupied from the third quarter of the eighteenth century into the third quarter of the nineteenth century. The significant amounts of time averaging displayed by both the phases and their assemblages make chronological precision difficult. However, plantation Phase 1 likely samples the period of French control while Phase 2 probably registers the arrival of the British and is followed by a massive increase in the number of people living in the

78 · Lynsey A. Bates, Jillian E. Galle, and Fraser D. Neiman

sampled areas of the site or in per capita discard rates or both. We suspect the plantation Phase 4 to 5 transition coincides with emancipation. The dispersal from Locus 2 of most of the plantation’s formerly enslaved people would explain the decline in the abundance of ceramics in Phase 5. Changes in Occupational Intensity across the Site With the plantation chronology in hand, we can now chart change over time in the number of ceramic sherds deposited within Locus 1 (Estate) and Locus 2 (Village) and among the different excavation zones or blocks that comprise them. The goal is to infer the pattern of increase and decrease in the size of the population using, breaking, and discarding ceramics in each of the blocks. A simple way to do this is to count and then plot the number of ceramic sherds found in the assemblages assigned to each phase for each excavation block. A key assumption here is that the different phases are characterized by similar amounts of time averaging. Figure 4.2 portrays the total number of ceramic sherds assigned to each phase within each block as a series of bar charts. Note that the y axis, which measures the number of sherds, is different for each of the graphs so that the patterns within each block are not obscured by variation in sample size among blocks driven by variation in excavation area. We have seen in earlier chapters that Locus 1 is composed of three excavation blocks, the Stable Block, the Estate Block, and Block E. All three have architectural evidence for multiple episodes of architectural construction. An early nineteenth-century foundation remains visible at the Estate Block, while excavators encountered postholes below overlying layers, which hint at earlier, eighteenth-century construction at the site. The phased bar chart for the Estate Block offers evidence for late eighteenth-century (Phases 1 and 2) occupation, an occupational or depositional hiatus in Phase 3, and then a resumption of deposition in Phases 4 and 5. The extant stable is thought to date to the early nineteenth century. Oral history suggests that the stable was turned into an overseer’s residence in the mid-nineteenth century. The phased bar chart (Figure 4.2) reveals that the stable was built on top of an earlier eighteenth-century occupation. The lack of assemblages from Phases 2–4 is striking. Occupation—by people— resumed in Phase 5, confirming the oral history about the arrival of an overseer in the postemancipation period.

Figure 4.2. Total number of ceramic sherds assigned to each phase within each block.

80 · Lynsey A. Bates, Jillian E. Galle, and Fraser D. Neiman

Block E is adjacent to a glacée for coffee drying. Excavators encountered numerous postholes and small features, which Hauser has identified as architectural. The bar chart reveals that these almost certainly date to Phase 1. Locus 2 includes excavation Blocks A, B, C, D, F, and G. For this analysis, we combined the phased assemblages from Blocks F and G, which are near each other and have small samples, into a single Block FG. The phased bar charts reveal an order-of-magnitude increase in deposition in Blocks C, D, and FG during Phase 2. A similarly large increase occurs in Block A during Phase 3 and in Block B during Phase 4. During these same two phases, deposition at Blocks FG and D (respectively) drops off. All five blocks show precipitous drops in deposition in Phase 5. Only Blocks A and B remained occupied during the final phase. The foregoing patterns of change in the different excavation blocks support the hypothesis of two punctuations in settlement location, one during the early decades of British control and one at emancipation. Phase 2 witnessed a decline in the intensity of occupation and perhaps eventual abandonment of the estate area. This may have been ultimately linked to the economic and political uncertainty that plagued the Caribbean in the wake of the American Revolution. Simultaneously, there was a massive increase in deposition (and number of enslaved people responsible for it) into the village-area blocks, followed by continued increases in Blocks A, B, and C. This corresponds with the increase in enslaved laborers indicated by historical documents discussed by Hauser in the first chapter of this volume. This is followed by a precipitous decline at the Phase 4–5 transition, which we suspect is coincident with emancipation. However, some newly freed workers remained in Blocks A and B areas. Meanwhile, the estate area was reoccupied in Phase 4 and the occupation continued into the postemancipation period. Changing Domestic Economies through Time Our chronology also allows us to trace change over time in assemblage content within Locus 1 and Locus 2 while also measuring synchronic variation between them. We use an abundance index (AI), a flexible measure of variation in the frequency of one artifact class relative to another: AI = x/(x+y). In the equation, x is the assemblage count of the artifact class—the numerator class—whose abundance we hope to measure, while y is the count of

Building an Archaeological Chronology for Morne Patate · 81

a second artifact class—the denominator class—whose abundance we are using as a baseline. The analytical goal is to use AI estimates as proxies for discard rates of the numerator artifact class (x). Discard rates are in turn linked to breakage, use, and acquisition rates of the artifacts in question, all of which are linked to the strategies people employ to navigate their worlds and to the costs and benefits those strategies confer. We want to pick a denominator class whose discard rate (i.e., the number of sherds deposited per person per year) is more or less constant across the assemblages being analyzed. Or, if it is not constant, its discard varies much less than the discard rate of the numerator class. If the denominator class meets these assumptions, then variation in the AI will reflect variation in the discard rate of the numerator class. Galle has suggested that, for many eighteenth- and early nineteenthcentury sites in North America and the Caribbean, green glass from “wine bottles” is a promising candidate (Galle 2006, 2010, 2011, 2017). A second possibility is coarse earthenware. We use AI results for these two classes to determine if they do meet the assumptions required for a valid proxy. This offers an opportunity to catch and rule out flaws. To take advantage of it, we compute AI values for coarse earthenwares, with wine bottle glass as the denominator class. The result validates the use of the method with the Morne Patate assemblages and offers clues to changing domestic economies at the core of Morne Patate. Consider first the methodological implications. If discard rates for both wine bottle glass and coarse earthenwares are constant through time, then we expect a plot of AI estimates for coarse earthenwares, computed with wine bottle glass as the denominator class, to show no trend and no significant variation among assemblages. Figure 4.3a plots these AI estimates for the five phases at both Locus 1 and Locus 2, along with their 95% binomial confidence limits. The confidence limits offer an approximate and probably optimistic (in the sense that it is too low) measure of uncertainty in each estimate that is the inevitable consequence of sampling error. For Locus 2, the trend is flat and there is very little among-assemblage variation. This validates our reliance on wine bottle glass as the denominator class in what follows. The flat trend in the discard of utilitarian vessels suggests that Locus-2 residents—enslaved individuals living in household units, whether with kin or not—had relatively stable food processing, storage, and preparation routines through the occupation. The anomalous results for Locus 1 are a learning opportunity. AIs for

Figure 4.3. Abundance Index (AI) estimates for six artifact classes, all computed with wine bottle glass as the denominator class. Approximate 95% binomial confidence intervals computed using the Wilson score method (Agresti and Coull 1998).

Building an Archaeological Chronology for Morne Patate · 83

Locus 1 in Phases 1 and 2 are far higher than those for Locus 2. We suggest these higher rates of coarse earthenware discard indicate wealthier households with complex domestic economies that required more involved ceramic-mediated food processing, preparation, and storage routines than food processing routines within contemporary enslaved households at Locus 2. If this is right, it implies that the domestic-economic complexity of Locus 1’s new residents during Phase 4 declined to the level found in Locus 2, although it increased slightly in Phase 5. This pattern suggests that Locus 1 deposits in Phases 1 and 2 represent a resident-owner household, while their successors in Phases 4 and 5 were significantly less wealthy, perhaps hired overseers. The contrasting trajectories of change in AI values and discard rates at Locus 1 and Locus 2 in all coarse earthenwares persist when we break this large artifact class down into its two most abundant components: Caribbean coarse earthenwares and imported French coarse earthenwares (Figure 4.3b and 4.3c). However, the finer-grained classification reveals subtle differences late in the sequence. For Locus 2, we now see a significant uptick in the AI for Caribbean coarse earthenwares but not French ones in Phase 5. For Locus 1, the pattern is the opposite. This raises the possibility of a marginal increase in domestic-economic complexity at both loci, coupled with increased inequality between them, with Locus 1’s residents discarding slightly more costly imported coarse earthenware ceramics at higher rates. Subtle differences earlier in the sequence are also noteworthy. For Caribbean coarse earthenwares, an increase in the AI in Phase 2 at Locus 1 may point to import-substitution (Figure 4.3b). Perhaps discard of locally produced utility ceramics increased because they replaced imported ones, which were harder to get in the war-torn late eighteenth century. We would expect a complementary decrease in the French coarse earthenware AI in Phase 2. But the large confidence interval prevents evaluation of this idea. On the other hand, despite the statistical uncertainty, it is clear that French imports continued to be discarded at a relatively high rate in Phase 2 at Locus 1. This may point to the impact of the 1766 Free Port Act that enabled Dominican and Jamaican ports to trade freely with any ship that docked in their ports. Rampant smuggling, the lack of enforcement of British Navigation Acts, and the loosening of French trade laws in the late eighteenth century all contributed to the flow of goods throughout the Caribbean, especially in the Lesser Antilles, where trade between adjacent islands was easily accomplished. Dominica is sandwiched between Guadeloupe and

84 · Lynsey A. Bates, Jillian E. Galle, and Fraser D. Neiman

Martinique, French colonies that are still part of France’s overseas regions today. So it seems likely that French coarse earthenwares were both abundant and inexpensive on the island throughout the eighteenth and nineteenth centuries (Hauser and Kelly 2011; Kelly et al. 2008). Next we turn to AI estimates for more costly, imported refined wares (Figure 4.3d). Included here are fashionable porcelains, refined stonewares, tin-enameled wares, and refined earthenwares, all of which were key to strategies of conspicuous consumption related to food and drink (Galle 2010). The link to wealth offers an opportunity to use refined ware AIs to independently evaluate the ideas we offered to explain variation in the coarse earthenware AIs. The hypothesis of a drop in the wealth level of the residents of Locus 1 after Phase 2 is supported by a precipitous drop in the AI values for refined wares. In Phase 4, the new Locus 1 residents invested far less in status-driven displays than the earlier resident-owner occupiers. However, investment apparently increased in Phase 5, paralleling evidence for a modest increase in wealth based on coarse earthenware AIs. We also discovered significant, although more subtle, changes in refined earthenware AIs in Locus 2. A significant decrease in Phase 2 may point to a decline in consumption driven by war-related economic stress. The sustained increase across Phases 3 and 4 may point to marginal improvements in the motive and means to participate in consumption rituals related to food and drink, which in turn may be related to marginal improvements in living conditions made by slave owners anxious about the end of the transatlantic slave trade. The continued increase in the refined ceramics AI estimate in Phase 5 for Locus 2 may register expansion of economic opportunities for newly freed individuals within and outside of Morne Patate Estate. AI estimates for sherds from table-glass vessels (e.g., stemmed vessels, tumblers) serve as an independent check on the conclusions based on refined ceramics. Table-glass vessels were linked to conspicuous rituals of drink consumption. The pattern of change for table-glass discard in Locus-1 assemblages largely mirrors our findings for refined ceramics. We again see evidence for the decrease in discard rates after Phase 3, relative to Phase 1, although this is tempered by massive uncertainty about the AI value for Phase 2. There is also general agreement in the pattern of change for Locus 2. Discard rates for table glass drop from Phase 1 to 2. This is again followed by a modest increase after Phase 3. This contrasts with the much larger

Building an Archaeological Chronology for Morne Patate · 85

increase for Locus 1 in Phase 5, suggesting that social rituals diverged, with Locus 1 residents investing much more heavily in drinking. A unique feature of the Morne Patate ceramic assemblages is the abundance of gaming pieces for which there is evidence of on-site manufacture from ceramics, mostly refined, transfer-printed pearlware and whiteware ceramic plates. Discard rates for gaming pieces may be constrained by discard rates of the vessels from which they were manufactured. But the constraint is a loose one—there were plenty of discarded ceramics across the occupation—and additional factors must have been at work. Among the most important was variation in the amount of time individuals invested in social interactions involved in gaming, including the time they devoted to game-piece manufacture. This may in turn be linked to economic stress. For Locus 2, the pattern of change in AI values is a familiar one. We see a significant decline from Phase 1 to 2, followed by modest increases across Phases 3 and 4, and a much larger increase in Phase 5. On the other hand, small samples and the general rarity of gaming pieces generate so much uncertainty that we cannot make credible inferences about gaming-piece discard rates at Locus 1. Conclusion Establishing a robust chronology is critical to tracking changes in occupation intensity, and household complexity through time and space at Morne Patate. At Locus 1, the presumed location of the household of the estate’s owners or overseers, we have uncovered evidence for a steep decline in occupational intensity or perhaps a hiatus in the last decade of the eighteenth century (Phase 3), corresponding to the shift to sugar production on the island and associated transition on Morne Patate after British annexation. In Phases 1 and 2, the Locus 1 household has much higher wealth levels than the Locus 2 households. After Phase 3, Locus 1 wealth drops precipitously, leading to our inference that hired overseers replaced owners in the early nineteenth century (after Phase 3). In Locus 2, the presumed site of the slave village, we have detected a sharp increase in occupational intensity in Phase 2, followed by further increases in Phases 3 and 4, and then a sharp decline with emancipation in Phase 5. These trends likely reflect an increase in the size of the enslaved population from the start of British control in 1763 until emancipation. Emancipation was followed by dispersal of many enslaved

86 · Lynsey A. Bates, Jillian E. Galle, and Fraser D. Neiman

households. But significant numbers remained. There are subtle hints that enslaved people managed to achieve marginal increases in economic opportunities starting in the late eighteenth century and extending through the early nineteenth century (Phases 2, 3, and 4). This trend continued after emancipation. Finally, we have discovered evidence for deteriorating living conditions during Phase 2, with declining discard rates for material culture recovered from this period. These initial conclusions warrant further fine-grained exploration using a host of additional artifact classes, including a focus on coarse and refined ware vessel forms. We look forward to a spatially more fine-grained analysis of variation in trajectories of change among excavation blocks within both Locus 1 and Locus 2. However, smaller sample sizes for phases within excavation blocks will require more sophisticated statistical methods than we have employed here. Our goal has been to lay out a few major trends in each locus and thereby construct a foundation on which future analysis may build. The data are freely available to all scholars at www.daacs.org. Jump in! Acknowledgments We thank Diane Wallman and Mark Hauser for inviting us to contribute to this volume, and we appreciate their patience and good humor in working with us. We are grateful to Mark Hauser for selecting DAACS to conduct the analysis of the artifacts and field records from the Morne Patate excavations, which are freely and publicly accessible via the DAACS website. This work was made possible by the National Science Foundation and by the National Endowment for the Humanities and the Thomas Jefferson Foundation through their generous endowment of the Digital Archaeological Archive of Comparative Slavery. This article benefited from conversations with Khadene Harris, Elizabeth Bollwerk, and Leslie Cooper. Lynsey Bates, Leslie Cooper, and Elizabeth Bollwerk, senior archaeological analysts with DAACS, with help from Khadene Harris, Colleen Betti, Alan Armstrong, Lindsay Bloch, and Clive Grey, identified and cataloged all of the nonfaunal artifacts from the Morne Patate excavations.

Building an Archaeological Chronology for Morne Patate · 87

Notes 1. DAACS Cataloging Manuals, accessed January 13, 2020, https://www.daacs.org/ about-the-database/daacs-cataloging-manual/. 2. Where mi is the manufacturing midpoint for the i’th ceramic type, pi is its relative frequency, and si is its manufacturing span. The idea here is to weight the manufacturing midpoint not only by the frequency of each type but also inversely by the variance of the response function that describes the trajectory of change over time in the popularity or relative frequency of the type. We assume that, over time, the relative frequency of each type roughly follows a Gaussian response function, with the manufacturing start date three standard deviations below the manufacturing midpoint and the manufacturing end date three standard deviations above it. This implies that s[i]/6 is a reasonable estimate of the response function’s standard deviation, and its square is a reasonable estimate of the variance. 3. Textbook introductions for archaeologists include Shennan (1997) and Baxter (1994). Greenacre (2017) is the canonical statistical introduction. Useful archaeological case studies include Duff (1996) and Peeples and Schachner (2012), while de Leeuw (2007) offers an insightful overview of archaeological applications from a statistical point of view. Further details on and applications of the perspective we sketch below can be found in Neiman and Alcock (1995); Ramenofsky, Neiman, and Pierce (2009); and Smith and Neiman (2007). Baxter and Cool (2010) and Carlson (2017) provide basic guidance on using R for correspondence analysis.

5

Y Morne Patate House Yards, 1750–1900 An Overview Khadene K. Harris

Caribbean yards and their associated structures have long been of interest to archaeologists seeking to understand how the domestic spaces of enslaved laborers both embodied and reflected kinship ties, economic strategies, and broader sociopolitical processes. For scholars working in the region, yards are an elemental feature of Caribbean society. They are the spaces of repeated daily acts, where the mundane occurs—our peek into the subtlety of everyday life. As spaces of routinized activities, yards are excellent for tracking the slow pace of change brought on by larger political and economic shifts. Historical archaeologists have been able to do this by reconstructing how enslaved villages and their respective house yards were organized and reorganized over time and by exploring what may account for observable changes in the way one engages with domestic space (Armstrong 1990; Armstrong and Kelly 2000; Heath and Bennett 2000). Yards, then, have the potential to unveil networks and interconnections that are not immediately obvious but are just as vital to a better understanding of slave life and the years after its abolition. This chapter examines how house yards on the Morne Patate estate were organized between 1750 and 1900. In this more than one-hundred-year span, Dominica went from being a sparsely populated French outpost to a British slave colony, and finally to a predominantly Afro-Caribbean society made up of formerly enslaved laborers and their descendants. What follows is an examination of how the shift from French to British political rule, from enslavement to freedom, was manifested in the location, function, and layout of house yards on the Morne Patate estate. Scholarship on Caribbean house yards spans several disciplines and makes use of as many methodologies (Lightfoot 2015; Parsard 2018;

Morne Patate House Yards, 1750–1900: An Overview · 89

Pulsipher 1994). All this scholarly attention over several decades has left Caribbean scholars balancing or shifting between the yard as both a theoretical construct and a scale of inquiry. The earliest anthropological studies on yards were mainly interested in how and in what ways African cultural practices persisted within contemporaneous Caribbean populations (Beckwith 1929; Herskovits 1990). With data primarily from ethnographic observations and folklore studies, Caribbean yards and their associated cultural practices emerged as a key site for further examination. For Caribbean scholars, that one could access important cultural insights from Caribbean yards was an important development, but it would be decades before plantation archaeologists turned their attention to enslaved domestic spaces in an effort to expand scholarship on slavery. When plantation archaeologists began exploring slave villages and yards in earnest, the questions they posed varied, but generally they either explicitly or unintentionally sought evidence of an African presence in enslaved material culture (Emerson 1994; Fairbanks 1974; Ferguson 1980). The problems with searching for Africanisms have been discussed elsewhere (Palmié 1995; Singleton 1996) and are not within the scope of this chapter, but in the last four decades since the search for Africanisms was introduced and widely accepted, scholars have refined their collective understanding of ethnicity and the kinds of claims one can make about cultural knowledge based on data from house yards. Within the broader literature, the main focus has been on highlighting how enslaved and free people created a sense of community and undermined oppressive tactics within their domestic space despite the external constraints of enslavement and colonial rule. Ethnographic and archival studies of house yards tend to emphasize the centrality of the yard to kin relations and the ways in which familial roles intersect with the peasant labor process. Geographer Lydia Pulsipher’s work, for example, based on multiyear ethnographic research in Montserrat, demonstrates that house yards are the building blocks of modern West Indian society. She describes yards as embodying a way of life that is several hundred years old (Pulsipher 1993:50). Anthropologist Sidney Mintz argues similarly, maintaining that the house yard is the foundation for the Caribbean peasantry and the manner through which peoples of the past situated themselves and others (Mintz 2010:12). Both Mintz and Pulsipher are correct. Still, this chapter expands their observations by showing that Caribbean house yards have more analytic potential if there is more of a commitment to recording how and in what ways house yards change after dramatic shifts in political and

90 · Khadene K. Harris

economic practices. In this case, Dominica shifted to a British colony and ultimately became a society of free laborers. This chapter clarifies how and why house yards gained the prominence they have within Caribbean society. And, importantly, by adding temporal depth to our discussions of house yards, we are able to situate changing house-yard activities within broader regional–global shifts. For archaeologists working on sites of enslavement, house yards are discrete spaces within a larger plantation landscape; they index a particular scale of analysis—the level that offers the greatest detail about the lives of enslaved individuals (Armstrong 1990; Heath and Bennett 2000; Samford 1996). Archaeological examinations of house yards both during and after slavery can reveal how both enslaved and free laborers engaged with space to mediate coercive practices and how those practices change depending on the preoccupations of the colonial elite. Caribbean yards, despite their enduring quality, are especially sensitive to change. Natural disasters, trade wars, and labor uprisings, for example, can alter the spatial and material practices of enslaved and free people. Archaeologists Douglas Armstrong and Kenneth Kelly (2000) documented as much in a study of the Seville Plantation in Jamaica. Their research showed how, after a destructive hurricane, the enslaved population at Seville Plantation altered the prior linear layout of the village to a more communal design that reflected a degree of autonomy that was missing in other parts of their daily life (Armstrong and Kelly 2000:380). The patterns witnessed by Armstrong and Kelly at the Seville Plantation shape the expectations of this study and suggest that exploring the house yard as both material and conceptual holds great possibilities. This is illustrated using data from the Morne Patate estate in order to highlight how Dominican house yards take shape and are continuously transformed, making the yard important not only for cultural production but also as a way to contextualize local, regional, and global shifts. If, like Pulsipher and Mintz, we are to accept that house yards reflect a sort of cosmological metaphor for survival, then focusing on how spatial and material practices of enslaved people differed over time is key to understanding these spaces. In this way the explorations on Morne Patate give more context to a framework that considers the house yard foundational. By engaging with the material record, this chapter adds another layer to the current view of house yards, a view that encourages an exploration of the yard as both material and

Morne Patate House Yards, 1750–1900: An Overview · 91

cosmological and that positions its inquiry at the nexus of environmental, political, and cultural occurrences. I begin by discussing how historical archaeologists study yards and what sorts of methodological and interpretive concerns emerge from their inquiry. I then move on to a survey of Dominica’s broader political economic context and how such changes influenced the everyday practices and decisions of Morne Patate’s enslaved population. Following that, I discuss fieldwork results. I close by returning to the yard as conceptual tool for reconstructing daily life. I center the actions of Morne Patate’s enslaved and, later, free residents to encourage a more expansive look at what house yard are capable of revealing. The Archaeology of Yards House yards feature prominently in the study of enslaved life in the Caribbean and the greater Americas. Materially, they offer a counterpoint to what was designed to appear as an elite landscape. Most studies of enslaved domestic spaces define house yards as the domestic complex that includes a bounded area of land that surrounds a house (Heath and Bennett 2000:38). For the enslaved, the yard was the space of everyday activities, and the house was where individuals slept, stored valuables, and found protection from extreme weather (Mintz 2010:12). Most of the “living” was done outside in yards—a phenomenon that requires scholars to study both house and yard as relational. Still, yards pose their own unique challenge both for data collection and site interpretation because yards generally were routinely swept and kept clean. Archaeologically, this means that enslaved material culture can often appear as clusters or areas of high activity that do not always reveal patterns in spatial organization as they were lived by the house yards’ enslaved occupants. When archaeologist Theresa Singleton referred to the use of space as the most monumental aspect of material culture (Singleton 1992:x), it was both a reflection on the importance of spatial analyses in archaeology as well as reminder that social relations were mediated through the altering of space. It follows, then, that archaeological studies of Caribbean house yards have elevated three important research goals: First, to determine how enslaved villages were arranged over time and how their respective configurations reflect then-existing social relations and political economic conditions.

92 · Khadene K. Harris

Second, to identify specific activities undertaken in house yards and what that reveals about individuals’ roles within a domestic unit. And, third, to identify patterns within material remains and what such patterns reveal about enslaved laborers’ preferences and the networks that facilitate their acquisition. Previous studies of house yards suggest that the spatial layout of the house yard may vary in detail or the house yard can be relatively uniform in size and layout. For example, research on the Bois Cotlette estate enslaved village in the southern portion of Dominica showed that laborers carved their house yards into the hillside and reinforced their domestic space with dry-stone retaining walls that are often easy to identify from the surface. This meant that the house yards varied in shape and size. In contrast, the Sugar Loaf estate enslaved village in the northeast part of the island displayed the opposite pattern. Within the Sugar Loaf village, house yards were meticulously carved into the hillsides to create terraces that were relatively uniform (see Hauser 2015). Similar patterns have been observed in studies on neighboring islands. Habitation La Mahaudière, a sugar plantation located on Guadeloupe, was occupied from the mid-1700s until the early twentieth century (Gibson 2009; Kelly 2011). The estate’s occupation was briefly interrupted by the outcome of the French Revolution, which temporarily abolished slavery from 1792 until 1802. Because of the dramatic shift brought on by political and, undoubtedly, economic shifts, La Mahaudière offers a good comparison to Morne Patate, which experienced a similar political shift only three decades earlier. Results from La Mahaudière suggest that early in the estate’s occupation, housing was characterized by lightly built post-in-ground wattle houses, locally known as kaz en gaulettes. They are archaeologically visible through a set of randomly placed postholes dug into the underlying limestone bedrock (Kelly 2019). The French revolution prompted a brief disruption in the institution of slavery where many villagers left plantations to reunite with family members and live as free people. After Napoleon reinstituted slavery, the plantation was once again repopulated with enslaved people of African descent who were held captive in regimented villages that they were tasked with (re)organizing. The nineteenth-century village was located on the same site but is distinct in that the post-1802 dwellings were built using different practices and following entirely new logic. Replacing the post-in-ground structures, the plantation established a series of uniform houses with masonry half walls that were built in a grid-like disposition across the existing village site.

Morne Patate House Yards, 1750–1900: An Overview · 93

In contrast, Habitation Crève Coeur in Martinique, a sugar plantation similar in size, scale, and occupation, did not exhibit the high levels of regimentation that was observed at La Mahaudière. Archaeologist Ken Kelly and his team located and excavated house yards within an enslaved village located on a hill above the main house of the estate. Unlike La Mahaudière and Sugar Loaf, the village was not established on a grid. Rather, enslaved laborers took advantage of spaces that were more level than the surrounding slope (Wallman 2014). Excavations on these platforms revealed archaeological evidence of post-in-ground, wattled houses similar to those from prerevolution La Mahaudière. For the most part, however, archaeological excavation was frustrated by a lack of visibility, a condition Kelly attributed the necessity of elevating houses on steep slopes. The goals of the work conducted at Morne Patate follow in a similar vein. The project sought to better understand the evolution of the enslaved village at Morne Patate to gain a better comprehension of how enslaved laborers and, later, free people made a life for themselves in the wake of the transition from French to British rule and later in the shift from enslavement to freedom. A closer look at the house yards allows for a clearer interpretation of local lifeways that this project anticipates will reflect both cultural traditions and broader economic demands of the plantation economy. Methodologically, this means identifying the limits of each house yard, including the location of spaces where everyday activities were undertaken such as cooking and gardening, and determining how house yards were organized in the various occupational phases. How archaeologists decide to approach house yards depends on a number of factors: the present-day condition of the site, soil morphology, and what local resources were incorporated in house construction. House yards that are in heavily disturbed zones are unlikely to have intact domestic features. In fact, house yards of enslaved villages that were abandoned and then used for agricultural purposes are best approached by using site-wide survey methods that can reveal both the extent of the village and the location of high-activity areas. In instances where the house yard preservation is relatively better and conditions allow, a targeted excavation of multiple house yards is recommended.

94 · Khadene K. Harris

Dominican House Yards Like its Caribbean counterparts, Dominica’s enslaved population lived in small houses with attached yards that they likely constructed themselves from material sourced mainly from their immediate environment (imported nails and window glass that show up in assemblages are the main exception). Very little has been written on enslaved house yards in Dominica; much of the research conducted prior to this study was carried out by local archaeologist Lennox Honychurch, whose research incorporated mainly ethnographic observations and historical documents. Unlike islands such as Jamaica or Antigua, there are few firsthand accounts that describe the everyday living conditions of the enslaved in Dominica. Of the few that exist, Agostino Brunias’s paintings of Dominican life offer the most potential for aiding a reconstruction of enslaved domestic life. Brunias was brought to the island in the late 1760s by Gov. William Young. Brunias produced over the years a group of paintings and engravings of life in Dominica and St. Vincent that, for historians, raised “important questions about the racial, cultural and sexual dynamics of the islands” (Bindman 2017). Like the casta paintings of colonial Mexico, Brunias’s work should not be read as straightforward evidence of enslaved everyday life but as an invitation to consider the multiple possibilities for the organization of slave life. In fact, Brunias’s work has been interpreted as offering a benign look at slavery, lacking any signs of brutality and coercion (Bindman 2017). Still, the imagery is helpful because the paintings were produced right around the time that Dominica became an official British colony, and one of the goals of this chapter is to show the impact of these macrolevel political transitions on the enslaved. One of Brunias’s engravings, entitled A Cudgelling Match between English and French Negroes in the Island of Dominica (Figure 5.1), depicts enslaved laborers engaging in a leisure activity in a yard. Onlookers crowd the small yard, watching the performers from the cleared area that appears to have been swept. The structure that stands behind the performers appears to have a dirt floor, a thatched roof, and was made from wood panels. In the background is a banana or plantain tree. Despite the questions that arise from a closer look at Brunias’s work, the image is good for refining expectations of domestic dwellings at Morne Patate. It is likely that the structure depicted in the image housed a number of enslaved individuals who were a part of a broader community that constructed and arranged their house yards similarly.

Morne Patate House Yards, 1750–1900: An Overview · 95

Figure 5.1. Agostino Brunias. A Cudgelling Match between English and French Negroes in the Island of Dominica. 1779. Yale Center for British Art, Paul Mellon Collection. https://britishart.yale.edu/collections.

Another image that allows us to refine our expectations of how Dominican house yards were arranged is an image of formerly enslaved laborers on Morne Patate’s neighboring estate—Bois Cotlette (Figure 5.2). Captured by a traveler by the name of Vaquero and published at a point in the early twentieth century, the image is of a yard scene where a mix of adults and children stand in front of a structure that has a thatched roof and an assortment of wood panels. Vaquero describes the image as consisting of “coloured and black people on the estate, few of whom could speak any English” (1914:187). The structure in the image has a stone foundation and a dry-stone retaining wall surrounding the yard, presumably meant to signal its boundary. In the background is a banana or plantain tree. In the foreground, where the group stands for the picture, there is a cleared area that was likely routinely swept that formed the primary gathering space and most likely the front of the yard where one receives visitors. Despite 130 years separating the images, the similarities between Brunias’s painting and Vaquero’s image are striking. They further refine our expectations of how house yards are arranged and the kinds of materials involved in their construction. But what is not represented is also important. The images highlight the role of yards in leisure

96 · Khadene K. Harris

Figure 5.2. Cottage and Labourers on Bois Cotlette, Vaquero. 1914. Accessed from https://archive.org/details/cu31924103988915/page/n259.

activities and communal living. However, they don’t capture where in the yard occupants would cook, or wash, or maintain a garden. Nor do the images depict neighboring house yards. On their own, the images give the impression that house yards are detached spaces where layout and organization remain the same over time. We contend that house yards are anything but. These building standards were not stable; they changed as prevailing concerns or conditions changed. For example, by the early nineteenth century, housing in Dominica and neighboring islands increasingly employed stone and wood for foundations, floors, and walls. This was in part a response to increased availability of some building materials but also a response to a growing interest on the part of planters to appear as if they were ameliorating the conditions of their work force (Handler and Bergman 2009:7). From the texts and images there is evidence that houses and their yards varied significantly in time and space. These buildings as represented in the images above are important historical documents that can refine scholars’ interpretation of house yards. Archival and historical documents on the organization of enslaved domestic

Morne Patate House Yards, 1750–1900: An Overview · 97

spaces are few and are rarely ever written from the perspective of the enslaved. Thus, our use of historical documents should be strategic and should point us in ways that can deepen our understanding of the historical context that has shaped our understanding of the experiences of the enslaved and later free laborers. The primary claim of this chapter is that house yards are sensitive to broader socioeconomic and political changes, which are reflected in the size, layout, and use of space over time. The expectation is that house yards on the Morne Patate estate will bear similar organizational principles as the house yards represented in both Brunias’s and Vaquero’s depictions. However, this chapter pushes the analysis through an archaeological examination of multiple house yards and how they changed over time. By examining how multiple house yards were constructed, arranged, and reorganized over time, we hope to highlight the actions and decisions that enslaved and free laborers made in response to wider constraints brought on by both political change and market shifts. Morne Patate Located in the southernmost enclave on Dominica, Morne Patate and its neighboring plantations were established by French leaseholders in the early 1700s. The earliest study of the enclave suggests there were several Kalinago settlements in the area. There is archaeological evidence at Morne Patate of a prior indigenous occupation (See Honychurch, Wallman, and Hauser, this volume), although it is unclear whether they interacted with the earliest French settlers. These early European settlements started out as provisioning estates that grew food for larger plantations on Martinique to the south. Enslaved laborers grew cash crops such as coffee and cacao alongside everyday staples such as yams, cassava, taro, and plantains that were shipped by small boats and later sold to various plantations. This arrangement would remain in effect until 1763, when the British formally took control of the island as one of the conditions of the Treaty of Paris. Following this political shift, French leaseholders who remained on the island agreed to become British subjects in order to hold onto their leases. Such a move was a part of a larger effort to convert Dominica from a frontier colony to a profitable sugar economy. The result, after the dust cleared, was a transformed Soufriere whose small estates expanded to take advantage of Britain’s commitment to large-scale sugar production.

98 · Khadene K. Harris

This chapter maintains that these macroscale shifts altered enslaved domestic practices and is manifested in the way house yards were organized and the ways in which they changed over time. While scholars should be wary of periodizations and their tendency to flatten history’s complexities, there are measurable shifts in the way house yards are organized that map onto these larger economic shifts in ways that can expand our understanding of variation among Caribbean house yards. Pre-1770 French settlers began establishing small estates in Dominica as early as the 1720s (Honychurch 1995). During Britain and France’s decades-long war over territory in the New World, Dominica and other neutral islands were to remain unsettled. But, as Stephan Lenik notes, in spite of these agreements, there appears to have been de facto French control of Dominica by the first decades of the eighteenth century (2012:93). This was especially true of the Soufriere enclave, with its close proximity to Martinique and an “entrenched population of French planters and enslaved Africans distributed throughout” (Hauser 2015:607). As provisioning estates, Morne Patate and its neighbors were smaller and more modest than plantations on Martinique. Just before the introduction of sugar, Morne Patate had approximately a dozen enslaved individuals laboring on the estate. Much of the existing literature on the establishment and growth of sugar plantations rarely addresses the smallholdings that often existed before the introduction of sugar. Notable exceptions include the important work that has been conducted in Barbados that explores the social history of the early (mainly Irish) indentured workers who planted cotton before sugar assumed primacy (Reilly 2016). Slavery on smallholdings or farmsteads has been widely documented but only gained rigorous attention recently. Documenting racialized slavery on such a small scale is certainly challenging because the distinctions (both social and spatial) that plantation slavery invoked is often missing within these contexts (Phillippi 2016). One of plantation slavery’s definitive characteristics is the rigid separation of space meant to reinforce existing racial hierarches. On larger plantations this separation is marked by differences in the way domestic spaces were organized and the presence or absence of costly material goods. Before Britain’s formal political control, Morne Patate had a smaller labor force made up primarily of enslaved laborers who came from Martinique. As a

Morne Patate House Yards, 1750–1900: An Overview · 99

provisioning estate, the way land was used affected where, how, and with what materials enslaved laborers constructed their house yards. Houses likely were more diffuse, with attached yards and a sizable garden. Morne Patate at this point lacked the kind of infrastructure associated with largescale sugar or coffee production. After 1763, when William Young assumed the role of governor, he was charged with transforming Dominica from a colonial dependency of Martinique into a viable sugar colony that was able to replicate the successes seen on older British colonies like Barbados. Post-1770 In the early 1770s, at the encouragement of the British, French leaseholders began cultivating and producing sugar. The project was a part of a larger effort to increase Dominica’s presence in Britain’s market economy through sugar production. For the British, Dominica was the island where they could display their expertise in sugar cultivation. Among their first initiatives was a survey of all cultivable lands that was undertaken by a surveyor appointed by Young. John Byres produced a map in 1773 that not only mapped existing plantations but also carved up remaining land to sell to potential English buyers. The shift to sugar resulted in an expansion of infrastructure and a population increase. Britain’s model for transforming Dominica meant establishing order on the landscape; it meant fashioning discrete locales that aligned with the economic imperative and racial distinctions embedded in the plantation enterprise. In other words, it meant attempting to replicate relationships and patterns that seemed to work on its larger colonies. Following the introduction of sugar, there was an increase in the number of people and structures that Mark Hauser claims constitutes a horizon in its own right (2014a:144). For an estate like Morne Patate, this meant that the scale of production increased and, simply, there were more people on the estate to house and feed. When shifts like this take place, the estate and those who live and labor on it come to establish a new relationship with land as a resource. The same was true at Morne Patate, where house yards are reorganized to reflect both the plantation and its occupants’ current needs. At Morne Patate the village population increased and house yards were rearranged to accommodate this shift.

100 · Khadene K. Harris

Postemancipation, 1838–1900 Emancipation in 1838 had an adverse effect on the productivity of Caribbean plantations. Many of the formerly enslaved persons left plantations in search of better opportunities. Some went to work on neighboring plantations that either paid well or did not regulate their time. The actions of the formerly enslaved varied based on a number of factors that historical archaeologists are still attempting to contextualize. Anthropologists working in the region all agree, however, that land and the opportunities that it afforded were central to the actions of the formerly enslaved. Mintz in particular signals the emancipation period as the moment of shift toward a peasant society, where the formerly enslaved and their land-based labor were foundational to Caribbean society and cultural production. If the sugar boom in Dominica encouraged an expansion of both people and structures, emancipation in many ways led to a contraction of that very process. There is a wide body of literature that explores the impact of a free labor force on plantations that have shown that plantations do not disappear entirely—in fact, many were consolidated into larger corporations. On smaller islands like Dominica, newly agreed upon labor arrangements had direct influence on the way house yards were organized. Natasha Lightfoot’s research on postemancipation Antigua offers a useful comparison to Dominica. In her book Troubling Freedom (2015), Lightfoot details how house yards of those newly freed were shaped by the lack of job opportunities that plagued the island well into the twentieth century. Antigua planters decided to forego the four-year apprenticeship period and freed all enslaved laborers in 1834, as opposed to 1838 in Dominica and the rest of the British Caribbean. Additionally, Dominica, unlike Antigua, had a surplus of land available that the formerly enslaved managed to take advantage of to establish villages and houses independent of estates. This meant for Dominica’s laboring class more opportunities to negotiate over labor conditions, which allowed for variation in the way house yards were organized. In the years after emancipation, plantations, especially those close to hinterlands, experimented with two main labor strategies: wage labor and métayage. Wage laborers exchanged plantation work for the occasional use of the houses and the grounds. Métayers were granted a plot of land in exchange for up to half of the crops cultivated. Michel-Rolph Trouillot’s analysis of estates after slavery reveals that laborers had a preference for sharecropping or arrangements that allowed laborers ample time to work

Morne Patate House Yards, 1750–1900: An Overview · 101

on their own plots (1988:83). Wage labor was viewed as the least advantageous arrangement, with its unpopularity suggesting that the newly free saw fewer opportunities for independence from estate work in its undertaking. By 1842 métayage was common practice on many estates, with the efforts of métayers accounting for 20% of Dominica’s sugar crop (Chace 1989:103; Marshall 1996). A closer look at Dominica’s history and geography forces a reexamination of how the literature defines and describes Caribbean house yards and the conditions under which they assume their form over time. The realities of the postemancipation period meant that where one established a house yard had as much to do with economic stability as it did with maintaining social ties. The stable vision of yards cast by Mintz and Pulsipher was years in the making. With this research at Morne Patate, we offer a detailed look at this process. Archaeological Dimensions of the House Yard Archaeological excavations at Morne Patate (Figure 5.3) suggest that the earliest enslaved village was not regimented in ways that the later village was. Rather, the spatial dimensions of the house yard emerged from a dialectical relation between planned schemes, their adaptations to on the ground conditions, and forms of adaptation that can be known through intent and consequence, without one informing the other. Morne Patate was comparable in size and duration with both La Mahaudière and Crève Coeur. As with both of these estates, the Morne Patate enslaved village was located on a hillside above the state house and near the plantation’s road. Between 2011 and 2012 an archaeological survey and testing revealed evidence for a multiphase village consisting of some 32 house platforms on terraces carved into the mountainside. These terraces were located nearby but upslope from the estate house and its dependencies. Based on these observations, we identified and excavated eight platforms: A through H. Choice of platforms relied on a series of factors including ease of access, archaeological integrity, duration of occupation, and potential differences based on size and proximity to the estate house. In these excavations we identified numerous different features including, but not limited to, stone flooring, foundations, cooking hearths, trash middens, postholes, living surfaces, and storage pits. In many cases these features were intact and provided excellent opportunities to collect material

102 · Khadene K. Harris

Figure 5.3. Area photograph of House Area A; Courtesy of Simon Walsh.

and dietary evidence. It also enabled us to establish settlement pattern of houses and their organization over time (Table 5.1). From each context excavated, 10 liter flotation samples were collected, processed, and analyzed to capture changing food practices from recovered plant and animal remains (Wallman and Oas, Chapter 8 of this volume), and 69 excavation units were placed across seven areas identified as the 1780s estate house and stables as well as several house areas within the slave village—House Areas (HA) A, B, C, D, E, F, and G. Two additional units were placed in the area identified as the estate kitchen and midden, and 4 were placed within the estate house itself. These comprised HA H. In total, 15 domestic structures were identified, 8 of which underwent large area excavation using 2 × 2 m units. In some cases a particular house area enabled us to track the evolution of housing at Morne Patate during its entire duration. In house areas A, D, and H, there were three houses superimposed one on top of the other, corresponding, more or less, to the early, middle, and late phases of occupation of the site. In other house areas we recovered architectural evidence for one point in the estate’s history. While HAs B and C contain evidence of occupation during the middle phase, architectural evidence only exists for the

Morne Patate House Yards, 1750–1900: An Overview · 103

Table 5.1. List of domestic features according to occupation phase House Area

Phase 1

Phase 2

Phase 3

A

Surface Postholes UI feature

Surface Postholes

Surface Stone alignment

Surface Posthole Storage pit Storage pit Surface Postholes Retaining wall UI feature Cobble floor Foundation Roadway Postholes Surface Foundation Cobble floor

Postholes Kitchen hearth Storage pit Stone alignment UI feature

B

C D

E (+Stables)

Postholes Cooking hearth Retaining wall Postholes

F G H

Postholes Cooking hearth

Foundation

Foundation

site’s late phase. The reverse is true for HA F. While we have evidence for architecture in the sites earliest phase, there is little evidence for architecture in later occupation. HAs A and B (Figure 5.4) were the first within the village to be excavated and are located on what is today a small farm with a gentle slope to the south. Prior to excavations, the area was under preparation for crop replanting—which meant surface levels were heavily disturbed. Nonetheless, stone formations on the surface alluded to possible locations of enslaved housing. HA A, among all the enslaved house yards excavated, appears to have been occupied the longest, with the earliest levels having a mean ceramic date (MCD) of 1733, right around when the region itself was settled by the French. The size, orientation, and construction techniques of the structure that stood in HA A appears to have changed over time. In its earliest phase, the structure was likely constructed with wattle and daub and had a compact dirt floor. It appears that the house orientation shifted later, with the use of stone foundations and plank walls. The stone alignment uncovered in Phase 2 of HA A suggests that the structure was reoriented at least twice, as evidenced by the positioning of two cooking hearths associated with the

Figure 5.4. House Areas A–D, F, and G.

Morne Patate House Yards, 1750–1900: An Overview · 105

structure. Results from the flotation samples suggest that the occupants of HA A grew maize and millet in the years before the introduction of sugar to Morne Patate and included guava, passionfruit, okra, and coffee in the years after. Approximately 4 m to the south of HA A is HA B—a similarly sized yard space that was likely constructed using mainly plank walls and posts. HA B proved interesting because of the numerous ways in which it differed from HA A, despite their proximity. It is likely that HA B was constructed in the early nineteenth century and occupied well into the postemancipation period, when the estate was experimenting with different labor arrangements. It does appear that HA B was reoriented; it is unclear how, but two cooking pits and a small midden narrow the possibilities. The structure that stood in HA B likely used the same construction techniques that HA A did in its later phases—plank walls supported by postholes dug to bedrock. Flotation samples for HA B suggest that the occupants grew a number of cereals—millet, sorghum, and barley—in their garden alongside guava and passionfruit. HA C is located on what is today a neighboring property approximately 50 m northeast of HAs A and B. Excavating HA C posed less of a methodological problem as the area was already flat, and stone features were visible from the surface. The units uncovered an area that was likely occupied in the nineteenth century well into the postemancipation period. Botanical remains from HA C include maize, guava, and passionfruit, exhibiting less variety than house areas with longer occupations. HA D, excavated in the 2016 field season, posed similar methodological concerns that HAs A and B did. The area is located just off the side of the main road, parts of which were heavily disturbed and were also occasionally used to cultivate crops. The structure that stood in the yard was likely occupied prior to the introduction of sugar, evidenced by the presence of French-made tin enameled wares that were recovered from the earliest phase of occupation. The earliest phase of occupation has an MCD of 1783, and the latest phase has an MCD of 1823, well before slavery ended—unique among the house areas investigated. Galle, Neiman, and Bates’s discussion (Chapter 4, this volume) of the site’s chronology fleshes out the patterns in more detail, but it is likely that the occupants were not enslaved. More than the other house areas excavated, HA D had a higher proportion of French earthenwares—approximately 63%—and was located closest to the estate house. Results from flotation samples uncovered only maize, guava, and passion fruit.

106 · Khadene K. Harris

HA E (Figure 5.5), also excavated in 2016, is located in the plaza area close to the estate house, which suggests that the structure that stood in that space predates the drying area. HA E is a combination of two excavation blocks. Excavations uncovered a series of postholes that were relatively well defined, giving a good sense of the structure’s orientation. Just 6 m south of HA E is the estate’s stable—a dry-stone structure that is currently in ruins. Excavations revealed an additional structure that, like HA E, predates the stable and was likely a wattle-and-daub structure held together by wooden posts. It was likely a laborer’s house that was constructed during the site’s earliest occupation when the estate grew provisions for Martinican plantations. The site map (Figure 5.5) illustrates the orientation of both structures. Underneath that is fill ranging from 15 cm to about 1.2 m. In each block there was evidence of some architecture including nine postholes in one block and five in another. The arrangement of these postholes is telling. Located less than 20 m from each other, they appear to be postholes for pilons sunk into the underlying subsoil. Prior to building the glacée, the surface here was steep and uneven. HA H was composed of excavation blocks at the planters’ residence or the maison de maître and the kitchen. Although not an enslaved dwelling, the house shows evidence of being rebuilt and reoriented over time. The block had evidence of two masonry structures superimposed on top of each other. An 1816 probate indenture states that the estate house was 55 feet by 40 feet. It had a masonry foundation and a wooden frame. It contained six chambers, “one of them a store, a hall and two galleries” (Dominica National Archives, Book of Deeds X4, July 16, 1816). The foundation that is visible today conforms to these dimensions. Finally, HAs F and G were outliers. Test units revealed a cobble plantation road. Associated material culture places the road’s construction around the 1780s. The roadway aligns with other patches of cobble roadway documented through the study area. It suggests that road through which one enters the site today was not the one that would have been used in the years immediately following 1763. Both HAs F and G were excavated in 2017 and are located in the area between the enslaved village and the estate house. Preliminary examination of the layout of the yard and the material culture retrieved suggests that indigenous Dominicans were the likely occupants of these domestic structures. Analyses are ongoing and will not feature in my subsequent discussion but pose an interesting line of inquiry on native Dominican interactions with early French settlers.

Figure 5.5. House Areas E and H.

108 · Khadene K. Harris

Discussion Following our research goals, excavations were designed with the intent to recover as much information as possible about the location, size, orientation, and use of associated yard space. My subsequent discussion centers the shifting practices of HAs A, B, C, D, and E. The question is not whether enslaved spatial practices shifted alongside larger political economic shifts but how they shifted. Based on observations of patterns among the pre-1770 house areas that we located and excavated, house yards in Morne Patate’s early history were larger in size and were not initially arranged as a centralized village, as was custom on larger estates. There are two likely reasons for this pattern. The first is that Morne Patate originally started out as a provisioning estate that grew food for larger estates in Martinique. Morne Patate’s operation was on a smaller scale and likely involved agricultural practices that the enslaved were familiar with. Further, a small provisioning estate likely required less differentiation based on labor. On sugar plantations, in particular, laborers were differentiated based on the type of job they had in the sugar-producing process. These practices shaped hierarchies among the enslaved that were subtle but still very material. At Morne Patate, the lack of differentiation among the enslaved did not mean that relations were equal. The landscape was organized in ways that reinforced racial difference—the estate house was larger and had more expensive goods than HA E and the structure that predated the stable. Nonetheless, the fact that the enslaved housing areas were spread widely apart with attached gardens adds to a body of literature that seeks to highlight the variation in estate organization and how a particular labor process can shape domestic life among the enslaved. After large-scale sugar production was introduced to Morne Patate, the number of house yards in the village increased to accommodate the expanded labor force. This meant that the scattered arrangement of space gave way to a consolidated village that remained in use until the late 1880s—some 40 years after slavery ended. During the decades of Dominica’s sugar boom, house yards in some ways contracted. HA A best illustrates how that process unfolded. The earliest structure was constructed using wooden posts. Later on, as the enslaved village expanded and familiarity with the estate’s immediate resources grew, houses shifted to using stone foundations from dry-stone ravines located near the estate. Even further, the house was reoriented at least once. These successive changes in house yard layout map onto

Morne Patate House Yards, 1750–1900: An Overview · 109

the plantation’s wider chronology where we see pre-British spatial practices reflecting those of a small farmstead with little differentiation in house yard activities. After the agricultural transition to sugar, there are increased differences in the size, design, and construction of houses and a greater reliance on provision grounds that were located in the hills above Morne Patate. When slavery ended in 1838, the plantation experienced yet another drastic shift; however, its material signature is more subtle than what is observed in the shift to sugar. Nonetheless, despite the complicated realities of emancipation, Morne Patate’s occupants were able to alter their domestic space in ways that reflected their expanding notions of freedom. Conclusion This look at Morne Patate house yards has engaged with current understandings of house and yard patterns in the Caribbean. This work emphasizes three major themes. First, it maintains that house yards are spatial entities that are defined by everyday actions that change over time. This means that to identify the layout and extent of the house yard is to define the ways in which the enslaved engaged with and manipulated their domestic space. In other words, space is dynamic and cannot be separated from the experiences, movements, and interactions of the wider society. In this chapter, house yards are defined as an enclosed area that immediately surrounds a domestic structure and is considered an extension of that dwelling. As bounded spaces, house yards require a fine-grained examination in order to grasp the subtle changes they exhibit over time. This allows one to consider how both the coercive strategies of colonial officials and the economic impulse of the plantation economy were manifested in the use and alteration of domestic space among the laboring population. Second, the house yard is essential to the formation and maintenance of social relationships among the enslaved. Several visitor accounts from the late eighteenth century recognized and wrote of the importance of the house and yard as a focus of social interaction. Accepting this notion does not presume much about specific kinship patterns (Mintz [1985], in his post-slavery ethnographic study, discusses household composition), but here I want to highlight the house yard as a means by which the enslaved sought to foster social ties and to reinforce relationships. In this way, the work that we do is in direct conversation with broader scholarship on practices of (re)connection that are constitutive of life among the enslaved.

110 · Khadene K. Harris

These works—working in the wake of Orlando Patterson’s notion of social death (2018)—tend to discuss the social lives of the enslaved in terms of “hopeful stories of heroic subalterns versus antinomies of doom” (Brown 2009:1235). With this project we hope to transcend these commitments by exploring linkages and interconnections (and their absence) that might expand our view of the world(s) that the enslaved operated in—using their material culture and an interpretation of their domestic spatial practices. And finally, house yards are symbolic, acting as a repository of tradition and central to the occurrence of psychic phenomena. Mintz (2010) and others have argued that the house and yard in its representation can reveal how affective and ceremonial life is organized. Accordingly, the house and yard may be read as reflecting locally defined conceptions of life and death (and other foundational principles as well, like hierarchy and kinship) as mediated through the use of material objects and the manipulation of space. Ethnographic accounts that detail birth and death rituals among peasant populations provide compelling context. Mintz writes of the planting of umbilicus or “navel-string” beneath a tree that remains within the yard. He states: “the links between the dead ancestors and the newborn children are represented in the yard” (2010:20). While these three key themes have undoubtedly pushed our analysis of house yards within the Caribbean, we need to pay more attention to the variations within this common pattern—questioning, for example, the underlying reasons for the location of provision grounds. Or perhaps, on sites where possible, what informs architectural decisions and when? How much do these actions rely on access or shifts within local/regional/global networks? This archaeological study of Morne Patate offers some answers to these questions.

6

Y A Carbet among the Cabins The Significance and Symbolism of a Possible Kalinago Household at Morne Patate Lennox Honychurch, Diane Wallman, and Mark W. Hauser

This research explores a time period when indigenous peoples alternatively identified as Carib or Kalinago (the term they currently prefer) remain unaccounted for in the historical record of British Dominica. On the eve of the British conquest of Dominica, Caribs were key players in an extensive and vigorous regional trade. By the 1800s they all but disappeared from the documentary record. Only after a century did American Frederick Ober publish Camps in the Caribbes (Ober 1880), estimating that the Kalinago were “less than 300 altogether” (Ober 1907). What happened in between? Despite their absence from the documentary record, the Kalinago did not disappear but became a part of the colonial social structure on Dominica in varying ways. Based on the recovery of Kalinago material culture and architecture at the site of Morne Patate, we examine how indigenous peoples were integrated into colonial networks during this little-known period, and we explore the significance of a possible Kalinago household within the plantation. Archaeological testing in the uplands of Soufriere revealed evidence for a household (House Area G) that contained diagnostic Amerindian materials normally attributed to the fourteenth to sixteenth centuries alongside architectural evidence and late-eighteenth-century European material goods. The size and relatively intact nature of the pottery sherds recovered indicate that materials were not subject to taphonomic processes such as plow zones or landslides. Located in relatively close proximity to a slave village and the estate house at the Morne Patate, House Area G provides enticing evidence

112 · Lennox Honychurch, Diane Wallman, and Mark W. Hauser

for the immersion of this cultural group into eighteenth-century plantation society. The archaeological investigation of a late-eighteenth and earlynineteenth-century household asks three questions: What was the history and shape of a Kalinago household located in southwest Dominica? What role did its occupants play in the plantation in which it was found? And how did their material lives differ from neighboring slave households whose occupants the documents describe as African? The answers to these questions contribute a better understanding of the development of colonial society and postcolonial communities in the Caribbean. Documenting architectural differences among plantation settlements through time provides a means of tracing the transformation of agricultural communities following the annexation by the British. Such discussions have tended to focus on the variation in the size and organization of enslaved houses and yards to describe the potential agency of its occupants in the context of highly regimented labor regimes and of social control. This chapter presents an analysis of one household identified at Morne Patate and defines the occupants’ social role within the larger settlement. We also investigate the relationship between the built landscape and social organization and consider how architecture might serve to communicate information. In the uplands of St. Mark Parish many of the platforms that made up the regimented slave villages remain visible, although their functions might have changed throughout time. The organization of these platforms as well as the structures they contained communicate information about the occupants. In the case of House Area G, we argue that the occupants emphasized their identity as distinct from the surrounding enslaved Africans. This study examines architectural variation in the Morne Patate slave village during a moment when increased regularity in domestic architecture was emphasized. It begins by using contemporary pictorial and documentary evidence to describe how these houses articulated with social hierarchy in colonial Dominica. We then document variation in architecture found in house platforms in the uplands of St. Mark—that is, the way houses were built, their shape, and their organization. This research considers whether the design of the buildings had symbolic significance and how they might have expressed claims to racialized status. Finally, it examines the change in settlement organization to understand how British rule affected the variation of domestic space.

The Significance and Symbolism of a Possible Kalinago Household at Morne Patate · 113

Background House design and the built environment on plantations represent a confluence of cultural expression, racialized social organization, and physical or environmental limitations. Vernacular architecture and domestic spaces in these contexts are often considered reflections of agency or identity and are symbolic of the powered structure of plantation systems (Glassie 2000; Vlach 1995). Here we explore the meaning behind a single distinctive archaeological signature of a house at Morne Patate contextualized by portrayals of Kalinago houses and material culture in the historical record. Visual culture has been an important body of material through which historians have constructed the social milieu of colonial society. Works of art have been even more important in assisting archaeologists in identifying the function and context of use of material culture retrieved from archaeological excavations. Despite these uses, the limitations of art as a source of information about slave life are legion. This is especially the case with islands, such as Dominica, where limited archival sources have led to an even more limited historiography. In this chapter we compare two bodies of work—one from a trained artist catering to an overseas audience; the other from the writings or recollections of contemporary visitors or residents of Dominica, including the private unpublished journal of a middling physician who visited Dominica in the last quarter of the eighteenth century, a freedman involved in a 1791 slave revolt, and a government official. We focus our attention on depictions of architecture and material culture within these sources to critically evaluate the composition and role of a household at Morne Patate. Between the 1760s and 1770s Italian painter Agostino Brunias visited the ceded islands as one of the many specialists who were part of Lt. Gov. William Young’s retinue. Brunias was trained in Rome and, according to Lennox Honychurch (2004:106), “would do in the Caribbean, what he had done for the English tourists in Rome; providing in paint, snapshots of their experiences and souvenirs of the places they had visited, in those days before photography.” Art historians have described him as belonging to the verité ethnographique tradition. Art historian Kay Dian Kriz (2008:23) argues that paintings of urban and rural scenes in Dominica were a kind of advertising in which “the island’s appeal as a potentially civilised space, as well as a profitable one, had to be emphasized.” Beth Fowkes Tobin (1999) suggests that Brunias followed larger conventions from natural history. In so doing, these

114 · Lennox Honychurch, Diane Wallman, and Mark W. Hauser

Figure 6.1. View on the River Roseau, Dominica, Agostino Brunias, 1770/80, Art Institute Chicago, https://www.artic.edu/artworks/79037/view-on-the-river-roseaudominica.

paintings not only cataloged everyday life in the abstract but also naturalized racial classifications between black, white, Carib, and mulatto through stance, clothing, and associated personal portable possessions. Such works were painted and circulated in the context of an Atlantic world where the increasing visibility of slavery and its violence were part of a charged set of politics in Great Britain. Works such as Brunias’s endorsed new colonial regimes because of the stable plantation society they depict. Here we focus particularly on the representation of housing in Dominica and neighboring St. Vincent in Brunias’s paintings. In these artworks, houses provide a backdrop to the scenes of everyday life that promote a vision of Dominica as a stable plantation society and encode to the reader cultural characteristics of the subjects in the foreground. We can break the groups depicted by Brunias into three general categories: the planters, the enslaved and free black class, and the Kalinago. Some paintings focus explicitly on specific groups while others may incorporate an amalgamation of the different groups in a scene in an attempt to normalize and soften the social relations of plantation society. Very few plantation houses are depicted by Brunias. Most European housing in his work is restricted to townhouses in the background of market scenes and distantly depicted maison de maîtres in the hills above Roseau (Figure 6.1). These structures, to the last, are modest endeavors. They are generally no more than to two to three rooms that lacked the elaboration of the new order of estate houses (French or English) built during Dominica’s sugar revolution.

The Significance and Symbolism of a Possible Kalinago Household at Morne Patate · 115

Figure 6.2. Villagers Merry-Making in the Island of St. Vincent, with Dancers and Musicians, a Landscape with Huts on a Hill, Agostino Brunias, ca. 1775, National Library of Jamaica Digital Collection. Accessed at http://slaveryimages.org/s/slaveryimages/ item/2432.

In scenes depicting ordinary life of slaves, their housing is surprisingly uniform, if not rudimentary (Figures 6.2 and 6.3). The most common slave house represented is referred to in modern Creole as the ti Kai (petite caz fr., “small hut”), a one-room structure built on a substantial foundation, often with plank flooring. It is associated with enslaved laborers and has features recognizable to a European eye, such as a door and sometimes windows. Indeed, two distinguishing features mark it as different from other more substantial buildings in Brunias’s paintings. First, slave houses generally have a thatch roof but are sometimes also depicted with plank roofs. Second, they are much smaller in size than townhouses or the maison de maître, typically represented as a single story. According to Brunias’s depictions, these houses were constructed with both wooden planks for walls and what appears to be wattle and daub walls. The third subject of Brunias’s scenes of everyday life are the Kalinago community spaces. In these paintings, Brunias depicts thatch huts that are rectangular in shape. Carbet is a French term referring to the main meeting

116 · Lennox Honychurch, Diane Wallman, and Mark W. Hauser

Figure 6.3. A Negroes Dance in the Island of Dominica, Agostino Brunias, 1779, Yale Center for British Art, Paul Mellon Collection. https://collections.britishart.yale.edu/ vufind/Record/3622709.

house and overall Carib settlement. The original Island Carib word for this structure is “táboui.” The French resided with the Tupi-Guarani tribe of Amerindians in Brazil and brought some of these individuals to work in Martinique, Dominica, and Guadeloupe. The Tupi-Guarani who came to Dominica used words from their language to describe similar things on the island. The word “karbay” was used to describe these houses and transferred to common use in the Creole language developing in Dominica. Karbay eventually became the accepted word for this house by all groups and is considered today to be a Kalinago-derived word. Carbet were part of a complex of structures that constituted the area were Kalinago men and women slept and worked. The main carbet was a large building, in most cases about 60 feet long and 30 feet wide. It was made of tall round wood posts and was either a rectangular or an oval shape with a tall, steep roof. The thatch which supported the roof was also used to tie hammocks for sleeping. The roof was thatched with palm leaves or the leaves of “roseaux” reeds (Gynerium saggitatum). Brunias depicts such housing in two paintings, A Leeward Islands Carib Family outside a Hut (Figure 6.4) and Family of Charaibes, Drawn from the Life in the Island of

The Significance and Symbolism of a Possible Kalinago Household at Morne Patate · 117

Figure 6.4. A Leeward Islands Carib Family outside a Hut, Agostino Brunias, ca. 1780, Yale Center for British Art, Paul Mellon Collection, https://collections. britishart.yale.edu/vufind/Record/1670141.

St. Vincent (Figure 6.5). While these paintings depict indigenous people in St. Vincent, they can also be used to understand some of the architectural variation on Dominica. These paintings depict post-and-beam structures with thatch roofs and walls. Additionally, Brunias depicts the smaller, openended structure with a hammock hung inside. According to his paintings, such structures were rectangular, appearing similar to laborer houses on plantations, and were portrayed as less permanent and less recognizable to the European eye.

Figure 6.5. A Family of Charaibes, Drawn from the Life, in the Island of St. Vincent, Agostino Brunias, 1801. New York Public Library Digital Collections, http://digitalcollections.nypl.org/ items/88b18789-9a1a-45ce-e040-e00a18067148.

The Significance and Symbolism of a Possible Kalinago Household at Morne Patate · 119

While many scholars have argued that Brunias’s paintings reflect an idea of a colony rather than the colony itself and that objects and features in the paintings are interchangeable props used to make a point, these paintings have interpretive value nonetheless (Tobin 2011). As long as we approach the images critically, we can use these depictions to better interpret the archaeological record, especially considering the dearth of primary sources from this period in Dominica. As we will see in the following pages, structures represented in Brunias have their analogs at Morne Patate. The Built Landscape Studies of slavery’s built landscape have recognized a number of factors that shape the design and organization of housing, including material, cultural, economic, and social considerations (Delle 2000; Farnsworth 2001; Hauser and Hicks 2007). Explanations for variation in housing throughout the Caribbean stress these variables to different degrees. Research on domination and resistance have tended to focus on the role of social space and the degree to which the relatively powerful could scrutinize the activities that took place within it (e.g., Lenik 2012). Others have focused on the built environment as an expression of underlying understandings of space and variation as a reflection of different worldviews (e.g., Armstrong and Kelly 2000). This approach draws attention to the built landscape as a culturally ordered setting used for a range of human activities. A third strand of this research, more often employed in the analysis of monumental architecture like great houses, has emphasized what buildings can communicate within the matrices of power (e.g., Delle 2014). We argue that houses at Morne Patate are representative of the systems of power and cultural worldviews of the different residents of the site, but we highlight the communicative power of the house, as it reflects one’s place and identity within the racialized plantation structure. In Caribbean plantation contexts, regularities in enslaved house and house yard design evolved out of an interplay among differing cultural perspectives on the house structure, power dynamics, economic demands placed on the household, and space available to resolve these issues. Documentary evidence suggests that such houses in Dominica were built cooperatively among the slaves during their free time and on land that was assigned by plantation owners, with materials they had at hand. This evidence also suggests that the size of slave houses in this area could range from

120 · Lennox Honychurch, Diane Wallman, and Mark W. Hauser

single-room structures to two-room structures. The archaeological record shows that most houses (House Areas A, B, C, D, E, and S) occupied immediately after the “sugar revolution” at Morne Patate were situated on parallel terraces oriented along a more or less cardinal direction. Houses within these platforms were rectangular and built on posts using planks as floors and walls. The houses did not vary much in size, measuring approximately 3 m to 4 m along one dimension and 5 m to 7 m along the other. Cooking hearths associated with these houses were located upwind of the doorway, allowing smoke to act as a deterrent for mosquitoes. House Area G, however, proved to be distinct along many dimensions from the rest of the house yards. The platform itself is like the remainder of the house areas. It is the smallest of the house platforms examined in this research, but only marginally so, covering approximately 60 square meters. The house sat on a terrace immediately upwind and upslope from the estate house but was in alignment with the remainder of the house yards. The architecture and organization of this house, however, displays a highly distinctive pattern from the rest of the slave village houses. Forty square meters were excavated from House Area G, revealing four distinct and stratified deposits. The interfaces of these deposits were determined by a combination of soil color and texture. The strata appear to coordinate with the occupation history of the site. The topmost layer is twentieth-century agricultural soil. The bottommost layer is a deposit of pyroclastic flow that, according to one geologist, dates sometime between 1410 and 1590, when the dome that makes up Morne Patate erupted (Wadge 1985). Evidence indicates that structures in Levels 2 and 3 and their associated deposits were abandoned in the following centuries. Domestic refuse in the form of late eighteenth-century trade goods (dipmolded glass bottles, creamware, faience, and salt-glazed stoneware) and coarse earthenware were recovered from all levels (Table 6.1). Imported ceramics were particularly useful in establishing the age of different occupations (Table 6.2). While the mean ceramic dates cluster to the period leading up to the transformation of Morne Patate’s landscape to produce sugar, the terminus post quem for Levels 1 and 2 highlight occupations as recent as the mid-nineteenth century. Given the presence of and density of French faience and creamware, materials deposited at Level 3 most likely represent a household occupied in the years leading up to Morne Patate’s transformation into a sugar estate. Given the density of pearlware and whiteware, Level 2 represents an occupation in the years prior to emancipation.

The Significance and Symbolism of a Possible Kalinago Household at Morne Patate · 121

Table 6.1. Artifact types recovered from House Area G Artifact Types

Level 1

Small Finds Button Gaming piece Tobacco pipes Slate Glass Bottle Glassware Clay Ceramics Local Coarse Earthenware Metal UI Nails Totals

Level 2

Level 3

Level 4

Totals

1 1 5 34

1 3

1 2 10 34

65 3

189 2

20 2

274 7

50 5

146 86

46 53

9

242 153

6 131

35 41 540

6 4 135

9

41 51 815

2

Table 6.2. Approximate chronology of the site MCD

TPQ

n

Level 1

1793.277778

1840 (Ironstone)

55

Level 2

1780.068182

1820 (Whiteware)

160

Level 3

1764.25

1775 (Pearlware)

47

The preservation does not permit as detailed an examination of architecture as that of the estate house, but changes are nevertheless visible in the structure and organization of the houses. Seven postholes were identified at the interface between Levels 2 and 3, and 13 postholes were identified at the interface between Levels 3 and 4. All of the postholes were 5–8 cm in diameter and 11–15 cm deep with small rocks placed on their perimeter, presumably to support the poles. Post-1770 The arrangement of postholes in Level 2 creates a rectangle approximately 3.5 m × 4.5 m. This is well within the range of house sizes documented at the remainder of Morne Patate. During the years immediately following

122 · Lennox Honychurch, Diane Wallman, and Mark W. Hauser

sugar revolution, housing in the village was rectangular and built from plank boards—usually white cedar. Local stones were sometimes placed underneath the floorboards to help support the structure or even out on the ground surface. These stones were usually laid without modification, in rough courses. Dwelling walls consisted of white cedar planks nailed into the posts with wrought nails that were reused from previous structures, borrowed from neighboring buildings, or provided by the estate owner. The arrangement of the postholes also seems to indicate that doorways for the structures were located on the upslope side, facing away from the maison de maître. In 1823 an anonymous author published an account of his travels throughout the eastern Caribbean. He spent a majority of his time in Dominica, where he had the chance to visit sugar estates and coffee estates in the vicinity of Roseau, Point Michelle, Soufriere, and Grand Bay. A French estate described by the author had a village extending behind the maison de maître. The negro houses extended in two rows, at a short distance from the mansion-house. All the married negroes had a house to each family, and the men who had no families had a large house, properly fitted for their accommodation, like a barrack. . . . On extensive estates these different buildings form small towns of two to four or five hundred people. (Resident 1828:60) The author’s description is an important one for what it says and what it doesn’t. First, it describes that slave villages were composed of a variety of living arrangements where kinship might be one idiom through which residence is defined. Some houses were reserved for families; some houses were reserved for bachelors. Importantly, each household is defined as incorporating a community of people. Contemporary accounts can help define what that community looked like. The anonymous author would go on to describe slave housing as “cottages, neatly thatched with palm or plantain leaves. Some have floors of wood and are well furnished with a bed, cooking utensils, & etc.; but this depends on the station and industry of the occupier” (Resident 1828:71). His account indicates that the location of the village was the decision of the planter or his manager. How slaves used land is a different story. While the author’s descriptions of estates are vague, they do provide some clue about the land set aside for slaves:

The Significance and Symbolism of a Possible Kalinago Household at Morne Patate · 123

The plantation negroes are provided with good houses, each containing two, some of them four apartments. Their cottages are thatched with leaves of the palmetto tree, or dyed Guinea grass. They have poultry-yards, and gardens railed in; and the latter produce all sorts of tropical fruits and vegetables. (Resident 1828:239) It is clear from this account that houses were not uniform in the design or contents but that the variation in housing was very much tied to the status of its occupants. These rectangular structures were often by themselves in a yard complex with an attached garden. The boundary of this yard was usually delimitated by living fences such as wild pineapple. Concentrations of charcoal in a matrix of soil might signpost a cooking hearth or a charcoal pit. The size and proximity to contemporary housing indicate for whom and when the meal was intended. For example, cooking hearths located next to the house might have been intended for members of the household during the morning or evening meals. A larger cooking hearth found in or near an agricultural field might have been a temporary mess set up to feed slaves during the harvest. Domestic fire use was not only concentrated in cooking. Dr. James Clark, a Scottish planter and physician on Dominica who wrote a treatise on diseases in Dominica, complained about the practice of slaves who “are so fond of fire, that they often lighted it up by stealth” after he “gave orders that no fires should be allowed in negro-houses” (Clark 1797:137–138). He feared the smoke would cause “jaw drop” tetanus. What Clark might not have recognized was that slaves were most likely employing their own preventive measures. Even according to Clark, some contemporary opinions suggested fire was considered a “purifier of the air . . . especially in ships and close quarters” (Clark 1797:67). In his own practice he suggested placing smoldering logs leeward of slave hospitals. While unbeknownst to Clark, smoke kept the mosquitoes away, reducing the chance for people to contract yellow fever or other mosquito-borne diseases. At Morne Patate, it appears that slaves consistently placed their cooking fires upwind of their doorways. The post-1740 structure and organization of House Area G follows the typical plan of slave housing in Dominica. Further, it aligns with the architectural conventions identified for other house areas on the site. Importantly, there would have been little to distinguish the occupants of this yard from other yards at Morne Patate based solely on the architecture.

124 · Lennox Honychurch, Diane Wallman, and Mark W. Hauser

Pre-1770 As noted above, excavation uncovered 13 postholes at the interface between Levels 3 and 4 at House Area G. The most obvious feature of this pre-1770 house area is the arrangement of postholes in relationship to the rest of the site. On the southwest side there appears to be an entrance. To the northwest of the structure there was a small shallow feature with some charcoal present. A dramatic difference is visible in the relatively large size of the structure compared to the others on the plantation. Postholes documented at the interface between Levels 3 and 4 were in the shape of an ellipsoid roughly 5 m × 4 m. In comparison to the remainder of the village, no ferrous construction materials were recovered from Level 3. The lack of nails indicates that the structure was built using methods distinct from the remainder of the village. Based on these architectural and archaeological signatures, we suggest the structure occupying House Area G appears to be a carbet. The closest analog to a structure such as the one revealed at House Area G are structures excavated in Argyle, St. Vincent, and La Poterie, Grenada. According to Raymond Breton, a windward-looking location with commanding coastal views is typical for early colonial indigenous settlements (Breton 1666). According to sixteenth- and seventeenth-century missionaries, villages typically were located on top of a hill, overlooking the water, and near a fresh stream (Borde 1674; Breton 1666; Dutertre 1667; Rochefort 1658). Extensive survey has seemed to confirm these observations (Hofman et al. 2019). Settlements consisted of large central men’s houses (táboui/carbet) surrounded by smaller huts for women and children (mánna). Gardens in which Kalinago would plant cassava and maize were scattered around the village in the forested interior. According to those same sources, the village could be abandoned upon the death of the chief. Excavations by Leiden University in 2009–2010 at Argyle in St. Vincent have revealed that 300 posthole features were attributed to two táboui and nine mánna. Dating between cal. AD 1500 and 1620 (Hofman 2012), these 11 houses were in two phases of settlement at Argyle (Hofman 2012). In addition to these 11 houses, features were also assigned to dependency structures including drying racks, cooking sheds, and barbacoas (Hofman and Hoogland 2015). The material culture repertoire at Argyle consists of Cayo pottery and European trade wares found in the eroded cliff to the seaside, which

The Significance and Symbolism of a Possible Kalinago Household at Morne Patate · 125

probably represent the garbage areas behind the houses. Cayo pottery has been documented archaeologically in Guadeloupe, Dominica, St. Vincent, and Grenada and in isolated contexts on other Windward and Leeward Islands (Boomert 2011b; Bright 2011; Hofman 2012; Hofman et al. 2019). Typical Cayo vessels have a variety of restricted and unrestricted shapes with simple and composite contours, some reminiscent of the Late Ceramic Age styles in the Caribbean (Meillacoid and Chicoid series; see Rouse 1992) of the Greater Antilles and some more related to the pottery styles (Koriabo complex, see Boomert 2011b) of the Guianas and northern Brazil (Hofman 2012). Cayo is a mix of locally produced earthenware with an amalgamation of shapes and decorative designs with zoomorphic modeled appliques and punctations originating from adjacent areas. The European tradewares include a fragment of an olive jar that, according to Goggin’s typology (1960), is the middle period (1500–1700), glass (some worked pieces), beads (chevron and seed beads), and European faience or majolica. Notable is a rim fragment of an indigenous or Afro-Caribbean pot inlaid with white European seed beads, affirming the indigenous-European-African intercultural dynamics at the onset of the encounters in this region. The implications for the similarity of the structure recovered at Morne Patate with the smaller structures documented at Argyle are twofold. First, evidence suggests that there is a degree of material continuity on this mideighteenth-century site with earlier practices. Material culture associated with this structure included dip-molded wine bottle glass, French faience, tobacco pipes, and broadly written local coarse earthenware. The local coarse earthenware included coil-built ceramics, mostly of a generalized form, and wheel-thrown storage jars, most likely made for processing sugar in Martinique. Levels 3 and 4 also contained diagnostic indigenous pottery possibly related to the Late Ceramic Age or early colonial Cayo complex (ca. AD 1400–1640) (Boomert 1986, 2011a; Hofman and Hoogland 2015; Hofman et al. 2019; Keegan and Hofman 2016). One of the diagnostic characteristics of this type is appliqué fillets with impressed decorations on or near the rim of the vessel, such as were documented on vessels at Morne Patate. Undecorated griddles were also found at House Area G as well as later deposits, discussed earlier. The griddles at the rest of the site are important because they represent pottery used by Africans in the early-nineteenthcentury Caribbean that is either made by indigenous peoples or influenced by their designs.

126 · Lennox Honychurch, Diane Wallman, and Mark W. Hauser

Second, the people who lived in this structure considered themselves different enough to build a distinct style of house. There is not much in the way of documentary evidence that supports the claim that Kalinago were living on a functioning estate during the sugar revolution. But what evidence does exist is compelling. Take the account of Jean Louis Polinaire, a “freeman” of Martinique who moved to Dominica. Taken at the completion of his trial for his principal role in the 1791 New Year’s Day revolt, Polinaire (“The Examination of Polinaire taken the seventh day of February 1791”; National Archives Unit of the Commonwealth of Dominica; Roseau, Dominica; CO 71/20) provides important details about a day when the island’s entire enslaved population in the southeast rose up in revolt to create an independent state on the windward side of the island. His account is brief and is mostly concerned with the conspiracy that led to the revolt. Polinaire described one of the ringleaders, Pharcelle, coming down from the mountain to a spring to gather water where he was spotted by “a Caraibe named Bigaire who lives on Mr. LaRonde’s Estate” (Polinaire 1791:2). Implied in this document is that Kalinago did live on plantations and that they also had some rights to property there. It is also implied that the Kalinago worked in concert with at least some of the planters. This was not isolated. In 1771 Joseph Senhouse became the collector of Roseau in Dominica and was later appointed comptroller of customs. He remained there until 1777 when he traveled to Barbados. During his stay in Dominica he purchased a sugar estate, Lowther Hall. Importantly, he was a firsthand witness to many of the administrative challenges of Dominica as a colony and the ways that those subject to its new regime negotiated its challenges. Lowther Hall was a fairly large estate outside of present-day Castle Bruce. In his diary he states that the area had 10 or more families, one of which “cultivate[s] a piece of ground for their own benefit” on his estate. It is not clear from the diary what the terms of the permission were or what Senhouse received in return. That being said, his account of contemporary material culture would appear to be at least firsthand. He describes the houses as “very small . . . resembling the letter ‘A’ or the roof of a house set upon the ground, covered with the branches of the coconut palm, etc. which are generally open at one end.” He goes on to say, “Under this covering they swing in hammocks of their own manufacture” (Senhouse 1985, 1778:283). Thomas Atwood, Dominica’s eighteenth-century historian, commented on these relationships in general: “they [the Kalinago] are very serviceable to

The Significance and Symbolism of a Possible Kalinago Household at Morne Patate · 127

planters near their settlements who they chiefly supply with fish and game” (1791:221). While it is uncertain who lived at House Area G, the material culture and design of the architecture raise the possibility that it was someone who might have identified with Kalinago families in the northeast. The presence of the Cayo pottery and of the circular structure with no ferrous material indicates a break in the pattern of house construction on Morne Patate. Additionally, the historical accounts seem to indicate that Kalinago who lived on plantations might have held different economic and social positions on estates than enslaved peoples. Discussion Historical archaeology has long been interested in what housing means. Building on the work of Vlach (1995), Glassie (2000), and others, this school has sought to understand the underlying grammar that inflected how people made use of their space. Past and present types of housing as well as their dimension and orientation, these authors would argue, carry enormous ideological baggage about family and its definition. Choices regarding housing materials, elements, shape, and organization reflect cultural ideas and social structures. By examining changing styles, archaeologists gain insight into cultural traditions and transformations. Continued studies of such forms have shown considerable variation in shape, organization, and construction. Matthew Johnson (2003) notes that over the fifteenth and sixteenth centuries in England, there was a process of “closure” involving new taxonomies of domestic space that coincided with a complex of cultural and social changes. Flexible interior spaces, typified by the medieval open hall, were increasingly segmented into rooms associated with particular charges. Ceilings were introduced to conceal timber supporting roofs, closing off that space even further from an increasingly important taxa—the outside. As such, “closure” in spaces within the home also involved a growing representational distinction between inside and outside space. Inside space was domestic and connotated notions of “comfort.” Outside space was considered wild, functional, and labor-filled. As an aesthetic predicated upon the enclosure of inside spaces, the house became a material boundary from the outside, “natural” space. Here culture is a not a thing or shared set of

128 · Lennox Honychurch, Diane Wallman, and Mark W. Hauser

values—although that is certainly at play. Rather, it is an assemblage of ideas materialized through the work of people encountering predicaments emerging from new economic and social orders. We argue for a similar approach here. While, indeed, difference in architecture might reflect an underlying and different cultural order among inhabitants, it might also reflect a kind of communication. Painting in the mid- to late eighteenth century, Brunias depicts clear social distinctions in the construction of houses while obscuring divisions of status, power, and race. If we consider the limited but informative primary accounts of Kalinago residents on plantations, these households assumed different positions within the plantation structure. By building and occupying a structure distinct from the others in a regimented slave village, the occupants might have been signaling this difference in status—slave versus non-slave.

7

Y Sourcing Coarse Earthenware at Morne Patate The Impacts of French Colonialism and Local Exchange Lindsay Bloch and Elizabeth Bollwerk

Coarse earthenware vessels are of particular interest to archaeologists because the low-firing technology employed enabled a variety of producers— from potters operating in households or small workshops to international companies producing wares in industrial factories—to participate in their manufacture. The resulting range of wares, which varied in price and function, can provide insights into enslaved and emancipated consumers’ ability and choice to participate in local and global economies and exchange networks. In this chapter we explore the coarse earthenwares recovered from household quarters of enslaved and later emancipated residents at Morne Patate Plantation, located on the island of Dominica, during the eighteenth and nineteenth centuries (ca. 1760–1880). The coarse earthenwares from Morne Patate present a prime opportunity for analyses because during much of the period that the plantation was occupied, Dominica was actively claimed by two European powers, France and Britain. Each governing body attempted to assert domain via settlement, military presence, and legal orders. The material patterning of goods like coarse earthenwares enables us to understand how these displays of colonial dominance affected the lives of Dominican residents by promoting or restricting access to certain kinds of goods and, conversely, how colonial dominance could be subverted through participation in local markets. The Caribbean coarse earthenwares at Morne Patate exhibit significant variation in surface treatment and paste attributes, indicating several

130 · Lindsay Bloch and Elizabeth Bollwerk

distinct recipes or sources. The majority of sherds were wheel-thrown, but hand-built or molded vessels were also present. While there is limited historical evidence of pottery production on Dominica during the colonial period (Hauser 2011), the ceramic assemblages at Morne Patate contained numerous coarse earthenware vessels with attributes consistently associated with pottery produced in the adjacent French colonies of Martinique and Guadeloupe (England 1994; Kelly et al. 2008). This suggests the possible interisland trade of locally produced earthenwares for domestic and industrial uses. Household contexts at Morne Patate also contained an abundance of lead-glazed, European-style wheel-thrown vessels. The majority of these are consistent with published descriptions of coarse earthenwares from French production centers including Vallauris, Huveaune, and Saintonge as well as the Italian type Albisola, but there has been significant ambiguity in these type descriptions. In this study, we sought to establish distinct coarse earthenware types for the sherds recovered from Morne Patate through patterned elemental variation (via laser ablation inductively coupled plasma-mass spectrometry, or LA-ICP-MS) as well as macroscopic attribute analysis. This combined strategy made it possible to assess whether attributes such as surface treatment or production method (e.g., wheel-throwing versus hand-building) were markers of distinct production origins or were within the range of internal production variation from the same source material or location. The resulting identification into European—mainly French—production groups speaks to the global trade networks within which the households of enslaved and emancipated residents of Morne Patate participated. The varying types of locally made coarse earthenwares also serve as material evidence of the social and economic networks that tied residents of Dominican sugar plantations to other islands in the Caribbean. Background The island of Dominica has a unique colonial history in the Caribbean, remaining unoccupied by Europeans throughout the sixteenth century. This was due to its status as a neutral territory for displaced native Kalinago inhabitants and its rocky coastline that discouraged exploration and settlement. It came to serve as a site of contention for British and French powers in the Lesser Antilles. Both recognized its strategic location. For France, control of Dominica cemented the country’s dominance in this part of the

Sourcing Coarse Earthenware at Morne Patate · 131

Lesser Antilles; conversely, Great Britain was keen to break up France’s Caribbean holdings (Baker 1994:63). The mountainous terrain made planting difficult, and territorial uncertainties stifled the development of large-scale plantations until the mid-eighteenth century. Agriculture initially focused on coffee and a variety of subsistence crops, many of which went to provision Martinique. While nominally a British holding in the aftermath of the Seven Years’ War in 1763, French planters carried on operating their plantations across the island as legal leaseholders. The continued presence of French planters, however, did not prevent the British from attempting to initiate a large-scale agricultural shift from coffee to sugar as the primary cash crop on the island during the last quarter of the eighteenth century (Hauser 2011). Critically, the ports of Dominica remained free. This encouraged trade with the neighboring French colonial islands of Guadeloupe to the north and Martinique to the south. The island’s status as entrepôt drew in trade from throughout the Caribbean, which also led it to become known as a locus for smuggling, especially of slaves. Despite its active role in the interisland trading networks of the eighteenth century, merchants on Dominica complained that supply ships were bypassing the island, resulting in shortages of goods. At times residents keenly suffered from a lack of even basic resources, especially in the wake of natural disasters, impacting subsistence for slaves on the plantations. This insecurity encouraged self-sufficiency and likely led to the enslaved inhabitants pursuing informal trade relationships within and beyond the island (Hauser 2014b). The plantation of Morne Patate is located in the southwestern corner of Dominica, near the capital of Roseau. It operated as a plantation during the eighteenth and nineteenth centuries, spanning several significant political and economic transitions of Dominica: from French control to British, from diversified agriculture to large-scale sugar production, and from slavery to emancipation, which occurred on Dominica in 1838. Morne Patate has been investigated archaeologically since 2010. Excavations have uncovered evidence for plantation industrial zones, slave housing, and gardens (Harris, this volume; Hauser, this volume). The artifacts discussed here are associated with the village that housed enslaved laborers on the plantation. Due to the economic history of Dominica, it was expected that during the period in question at Morne Patate, residents would have relied upon a mixture of imported French and English goods along with materials produced on the island itself and surrounding locales.

132 · Lindsay Bloch and Elizabeth Bollwerk

Earthenware Production in the Caribbean Earthenware production in the region can be loosely organized into two important and intermingling streams of influences. On the one hand, there is a long-standing tradition of ceramic manufacture by indigenous peoples who continued to produce pottery during the colonial period. On the other hand, Europeans and enslaved Africans brought their own manufacturing methods and ways of using pottery with them. These methods were then adapted by potters in the colonies. It is not the purpose of this chapter to disentangle the various cultural streams informing ceramic manufacture. Instead, we work from the premise that potters following each of these traditions influenced each other in myriad ways. Some of those influences were caught up in the matrices of power that were endemic in slavery. While most eighteenth-century Caribbean colonies relied to some extent on European imports of coarse earthenwares for domestic and industrial uses, local economies of coarse earthenware production also developed. Previous research on these industries has focused on ceramics produced to meet two general needs: domestic earthenware cooking pots and storage jars, and industrial earthenware for sugar production or architecture such as tiles. In the British colony of Jamaica, production of Caribbean coarse earthenware was undertaken at the household level, produced using handbuilding methods and non-European technologies (Armstrong 1990:153; Hauser 2008, 2011; Reeves 2011). Bowls and pots were sold through an informal, island-wide market system. Archaeological investigations at plantations in Jamaica have found these coarse earthenwares to be abundant within plantation household contexts (Armstrong 1990; Bates 2015; Galle 2017; Reeves 1997). In Jamaica, industrial earthenwares and those manufactured with European methods such as wheel-throwing and kiln-firing were not produced. Instead, the market for those goods continued to be driven by importation from Great Britain and obtained through the consignment system. In the eastern Caribbean are a number of locations where both traditions of pottery were produced, including Barbados (Handler 1963), Antigua (Hauser and Handler 2009), St. Lucia (Hofman and Bright 2004), Martinique, and several islands in the Guadeloupe archipelago (Kelly et al. 2008). In Barbados, the production of earthenware was more formalized, with English-style factories developing early to accommodate the needs of sugar production and household activities. Planters themselves brought

Sourcing Coarse Earthenware at Morne Patate · 133

over English potters to establish workshops and train indentured and enslaved workers to manufacture the necessary vessels (Bloch 2019; Handler 1963; Scheid 2015). There is very little evidence for colonial-era production of hand-built Caribbean coarse earthenware in Barbados. Hand-built wares that are present at sites such as Trents Plantation may reflect earlier preColumbian habitation on the island (Bloch 2019). On the islands of the French Caribbean, pottery production was more diversified and included wheel-thrown as well as hand-built vessels. Kenneth Kelly and colleagues (2008) surveyed Caribbean coarse earthenware produced on kiln sites in Guadeloupe and Martinique (see also England 1994; Gibson 2009). They identified at least three separate compositional groups among these islands. In contrast, Dominica’s ceramic chronology is poorly studied. Less is known about pottery found on Dominican sites than that from neighboring islands. It is also challenging to draw clear distinctions between African and indigenous influences on material culture. This is due in part to the fact that the colonial settlement of Dominica was more gradual and complex, coupled with the persistence of indigenous residents, the Kalinago, on the island. This has made it difficult to parse the arrival, interactions, and disappearance of different groups archaeologically. In Dominica, there are very few historical accounts related to pottery manufacturing. The Royal Commission for the Colonial and Indian Exhibition noted in 1886: “Coarse pottery is manufactured at the north end of the island and exported to Guadeloupe” (West India Royal Commission 1886:367), but further details on this production center were not provided. Later documents attribute pottery made during the 1890s to “Island Carib” communities in Dominica, assuming the wares to represent a precolonization tradition (Ober 1913:361; Thomas 1953). Ethnographers of the 1930s recognized that local Dominican potters were of African descent (Delawarde 1937; Taylor 1938:140) but were implementing practices they defined as “Carib.” These accounts may be referencing pottery made in the style of Cayo ware, a long-standing indigenous tradition of pottery manufacture in the Caribbean, as late as the nineteenth century. Cayo pottery includes vessels in a variety of forms and often with applied zoomorphic and other decorative designs. It has been hypothesized that the origins of the ware are in South America, given similarities to pottery styles of the Guianas such as the Koriabo complex (Boomert 2011b; Evans and Meggers 1960), and in

134 · Lindsay Bloch and Elizabeth Bollwerk

northern Brazil (Boomert 1986, 2011a; Hofman et al. 2019). The ware also resembles the Antillean Late Ceramic Age styles of Meillacoid and Chicoid described by Irving Rouse (1992). Two sherds with zoomorphic applique decoration were recovered from Morne Patate. Early European visitors to Dominica described some of the ceramic forms they encountered, and the functions of these vessels have been synthesized by Corinne Hofman and Alistair Bright (2004). The most common vessels were the canari; the boutéicha, a water jar; the toürae, a kettle, pot, or marmite to cook stews; and the boutalli, a griddle. The function of a canari vessel is ambiguous. While some argue that it was used to ferment cassava beer (de La Borde 1674), others argue that smaller versions of it were used as a pot with which to cook stews (Barbotin 1974). Even today, the term “conaree” is used on some islands to describe cast-iron cooking pots. It is most likely that there was originally a specific function attributed with the term “canari,” but over time it became a general class of domestic pottery that could serve multiple functions (Boomert 1986). While the vast majority of locally made earthenwares tended to be hollow forms, some coarse earthenwares recovered from Morne Patate were likely griddles. Raymond Breton’s description of locally made griddles notes that the shape of a griddle on which the women from that time dry their starch [paste] and of which it retains the name, it is of clay, placed on three stone rocks elevated half a foot or more, heated by the fire, the manioc flour is poured over to be baked on it, such is the oven of the savages. (1892:93–94) The starch Breton refers to has traditionally been believed to be cassava flour, but it could have also been maize flour (Pagán-Jiménez et al. 2015). Flat coarse earthenware sherds were recovered from both the estate and the village at Morne Patate. These are likely griddle fragments. They were associated with one context in the area of the estate and three different contexts located in Block B in the village. The other ceramics recovered from these contexts were primarily imported refined earthenwares, tin-enameled wares, and French coarse earthenwares. The location of the griddles in spaces occupied by enslaved individuals and their association with refined ceramics suggests they likely date to the historic occupation at Morne Patate.

Sourcing Coarse Earthenware at Morne Patate · 135

European Imports In addition to locally produced wares, imported vessels were found in domestic contexts at Morne Patate. The majority of the French earthenware exported to the Caribbean colonies in the eighteenth and nineteenth centuries came from the production centers of Biot and Vallauris and from the Huveaune Valley. Biot vessels have a buff-colored paste with a clear or slightly yellow-tinged lead glaze. The forms were predominantly very large shipping or storage jars, sometimes repurposed for water storage in domestic contexts of the Caribbean (Hauser 2017; Losier 2012:177). Vallauris pottery has a pale paste with abundant inclusions and typically has a transparent lead glaze on the interior of vessels. The primary forms of Vallauris pottery exported to the Caribbean were marmites (cooking pots) and storage vessels, although bowls and some tableware forms were also produced (Amouric and Vallauri 2007). While the majority of sherds found at Morne Patate could not be identified to specific forms, many were from hollow, utilitarian vessels. Huveaune pottery is typically pink to orange in color with very few inclusions and a soft texture. Some vessels exhibit slip decoration. The forms typically found on Caribbean sites include milk pans, chamber pots, and basins. Examples of milk pans were recovered from Morne Patate as well as some plates. Finally, a few fragments of Saintonge were also recovered from Morne Patate. Produced in the southwest of France, the interiors and exteriors of these vessels can exhibit a clear lead glaze that ranges from green to yellow in color depending on the surface paste color and firing atmosphere. Common vessel types include cooking pots, skillets, mugs, colanders, storage jars, and jugs. Saintonge is most common on seventeenth-century French colonial sites such as New France (Métreau 2016; Monette 2010). Only three sherds in the Morne Patate assemblage were identified as Saintonge, and none could be identified as specific forms. Methods Ceramic Sourcing Sourcing ceramics via elemental analysis relies upon measurable differences in elemental concentrations from one production location to another as a

136 · Lindsay Bloch and Elizabeth Bollwerk

function of the specific geological origins of the clay deposits used to produce the pottery (Weigand et al. 1977). When embarking upon a sourcing project, one must understand the underlying geological variation within the sources or regions under investigation in order to assess the likelihood of chemical variation and minerals or suites of elements that may drive patterned variation among the samples. Here we follow a broad definition of source that relies upon distinct geological provinces. It was expected that European and Caribbean earthenwares would have characteristic variation. Within these two broad regions, we anticipated subsetting into more refined groups based upon the unique depositional histories of the potential production sources. Lacking reference material from known pottery production sites, we relied upon additional lines of evidence such as previous studies and known regional geology to infer the geographic origins of the compositional groups, as described below. We focused on local Caribbean and French sources as previous research has shown these to be the most abundant wares found in the French Caribbean (Gibson 2009). Geology As with most of the Lesser Antilles, the island of Dominica is volcanic in origin. The majority of the landmass is composed of intermediate volcanic products in different stages of weathering (Lindsay et al. 2005:257). Areas on the west side of the island have thinner and rockier soil layers, while the older deposits on the east consist of well-weathered clayey soils (Shearn 2014:283). The terrain is very rugged, and the island is the most volcanically active in the region, especially on the southern tip near the capital of Roseau and the location of Morne Patate. While the geology of neighboring islands of Guadeloupe and Martinique is also predominantly volcanic, caps of sedimentary carbonate rocks overlay the islands of Grande-Terre, Marie-Galante, and Désirade in Guadeloupe. This variation results in two broad soil types within this part of the archipelago. The first originates from intermediate volcanic rocks such as andesite and the other from sedimentary calcareous deposits such as limestone. These geological signatures may be used to interpret the sources of Caribbean coarse earthenwares found at Morne Patate. Volcanic soils may be quite distinctive even within small geographic ranges as the deposits from each eruption have different elemental signatures (e.g., Carr et al. 2003). These Caribbean geological signatures contrast with potential earthen-

Sourcing Coarse Earthenware at Morne Patate · 137

ware sources in Europe, which nominally include France and England. As noted above, the majority of the French earthenware exported to the Caribbean colonies in the eighteenth and nineteenth centuries came from the Mediterranean basin, especially the production centers of Biot and Vallauris, near Cannes, and the Huveaune Valley, located inland from the busy port of Marseille. The landscape of Alpes-Maritime around Cannes is composed of iron-rich kaolinitic clays weathered from crystalline rocks such as feldspar (Petrucci 1999:26). Clay deposits contain iron nodules, gypsum, and quartz grains. Vallauris and Biot pottery has been made from these clays from the sixteenth century onward, beginning with the arrival of Genoese immigrant potters. In the southeastern region of Provence containing the Huveaune Valley and Marseille, the bedrock is limestone, with very fine-grained calcareous clays developing atop it. It was expected that pottery from these two locales—Cannes and the Huveaune Valley—would have distinct elemental signatures. We also anticipated that several sherds of pottery identified as Saintonge would form their own compositional group. This pottery type, produced in southwest France in the province now known as CharenteMaritime, is composed of a kaolin-rich, pale, fine clay. Although these French products have been described in some colonial contexts (e.g., Amouric and Vallauri 2007; Losier 2012; Métreau 2016; Mock 2006) and in limited French archaeological studies (Abel 1987; Abel and Amouric 1991; Petrucci 1999), they have not been explicitly defined through compositional analysis. In part, this is due to a lack of archaeological investigations into the pottery workshops themselves. Consequently, the results of this study provide groupings that can be further tested by future studies that focus on assemblages recovered from workshop or other production areas. Samples Seventy-seven sherds from the site of Morne Patate on Dominica were analyzed via LA-ICP-MS to obtain their elemental composition (Figure 7.1). They were purposely chosen from among the 3,091 coarse earthenwares recovered from excavations at Morne Patate in order to represent visually distinctive groups based on paste and surface characteristics, such as glaze or burnishing, and evidence of manufacture (e.g., wheel-thrown versus hand-built). Given the existence of European-style pottery production during the colonial period in the Caribbean, such as in Barbados (Scheid 2015)

138 · Lindsay Bloch and Elizabeth Bollwerk

Figure 7.1. Coarse earthenwares recovered from Morne Patate.

and neighboring islands (Kelly et al. 2008), and a lack of elemental data on French types, we sampled across all coarse earthenwares recovered from Morne Patate. The study included 40 samples that were preliminarily identified as European coarse earthenwares, including Vallauris, Huveaune, and Saintonge French coarse earthenwares as well as unknown “European-like” samples. The remaining 37 coarse earthenwares were presumed to be Caribbean in origin but exhibited a range of variation in surface treatment, color, and paste inclusions. These sherds had been preliminarily sorted into three

Sourcing Coarse Earthenware at Morne Patate · 139

main types based on visual differences. It was anticipated that there would be clear differences in the elemental signatures of European and Caribbean coarse earthenwares, given their distinctive geology, as described above. Instrumentation Lindsay Bloch conducted the LA-ICP-MS analysis in 2016 at UNC–Chapel Hill in the Mass Spectrometry Lab. The Excite 193 ultrashort pulse excimer laser and ablation system (Teledyne/Photon Machines, Bozeman, MT) was coupled to an Element XR double-focusing magnetic sector field inductively coupled plasma-mass spectrometer (Thermo Fisher Scientific, Bremen, GER). Laser ablation was chosen for this study because the small beam size made it possible to avoid large inclusions within the ceramic paste that might attenuate the chemical fingerprint. Furthermore, the results could be directly compared with previously analyzed historic coarse earthenwares from Great Britain, mainland North America, Jamaica, and Barbados (Bloch 2016, 2019). A small piece of each sherd was removed and mounted on a microscope slide. The freshly broken edge was placed face up and smoothed to facilitate even sampling with the laser. The laser was targeted to avoid inclusions and voids larger than 30 μm, instead focusing on the fine clay matrix. This reduced the dilution effects of abundant inclusions such as quartz sand (SiO2) or ferruginous nodules. Ablation lines were 600 μm long and 110 μm wide. Three lines were analyzed on each sample. Instrument settings followed Bloch (2016). Data were collected on 44 isotopes. Elemental values were calculated following Bernard Gratuze (1999), averaging the results of the three ablation lines to obtain a representative sample of the clay matrix. Reference standards included NIST SRM 679 (Brick Clay) and NIST SRM 610 and 612 (Trace Elements in Glass) and were analyzed alongside samples. The resulting elemental concentrations were log-transformed and analyzed in R (version 3.4.1). Elements potassium and gold were below limits of detection for some samples, so they were removed from analysis. Lead was also removed to avoid potential contamination from glazed surfaces of sherds.

140 · Lindsay Bloch and Elizabeth Bollwerk

Results Clustering Principal components analysis (PCA) was used as an exploratory technique to understand the patterns of elemental variation within these samples. Five main groups were present within these data. Figure 7.2 shows the overall sample assemblage along the first two principal components. Axis 1 separated samples depleted in trace elements from those enriched in trace elements. In general, this distinction serves to separate Caribbean (depleted) from European (enriched) samples. Types 1 and 2 were characterized by depletion in most elements. They separated from one another according to concentrations of calcium and cesium, among other elements (Figure 7.3). Although sharing elemental similarities, Types 1 and 2 could be distinguished further on the basis of visual attributes. Type 1 exhibited a range of

Figure 7.2. PCA biplot of LA-ICP-MS results, log ppm values for 41 elements.

Figure 7.3. Boxplots showing average elemental concentrations by compositional group.

142 · Lindsay Bloch and Elizabeth Bollwerk

variation in a number of attributes, and the paste showed uneven firing colors (Figure 7.1, sherds A and B). There were two distinct variants of Type 1 based on size and density of inclusions. The paste of Type 2 sherds was finer, with few inclusions compared to Type 1. Vessels of Type 2 also tended to be very well oxidized (Figure 7.1, sherds C and D). It is likely that both Type 1 and Type 2 are products locally produced in the Lesser Antilles, prepared from similar volcanic clay sources. Sherds identified as Type 3 were elementally the most distinctive of the Caribbean coarse earthenwares (Figure 7.1, sherds E and F). Sherds of this group were enriched in transition metals such as chromium and nickel. Type 3 was very coarse, and all examples were hand-built. This type may possibly represent Cayo or another indigenous pottery production tradition in Dominica or the adjacent islands. The European sherds readily separated into two main groups. The larger group contained samples visually consistent with the French production group Huveaune (Figure 7.1, sherds G and H). This group was characterized by high calcium and accessory elements such as strontium and barium, which is expected for pottery produced from calcareous clays. Several sherds that did not share the formal characteristics typical of Huveaune also fell into this group, some forming a distinct subgroup (Figure 7.4). The group outliers include two of the three samples that had been preliminarily identified as Saintonge products (Figure 7.1, sherd I). One of the pale, greenglazed sherds of possible Saintonge exhibited significant elemental variation from the others, but more specimens would be needed to determine whether this was a normal range of variation for the ware type. If these are true Saintonge, it is likely that several of the additional sherds in this cluster are also Saintonge. To date, there has been no elemental analysis conducted to define these French products. As Catherine Losier (2012:177) notes, there are a number of potential French production regions that produced wares with the same visual characteristics at this time. It is likely that this group, while defined primarily by the Huveaune products, also contained French coarse earthenwares from other production regions that share similar geological formations. Given the small sample sizes, we chose not to separate this group further. Sherds in the final compositional group were visually consistent with published descriptions of Vallauris, another French coarse earthenware (Figure 7.1, sherds J and K). These samples were characterized by depleted calcium and iron but enrichment in alkali metals such as cesium.

Sourcing Coarse Earthenware at Morne Patate · 143

Figure 7.4. Elemental scatterplot of calcium and hafnium showing the division of Huveaune compositional group into two possible subgroups.

Following definition of the five compositional groups, we validated group assignment using Mahalanobis distance probabilities (Table 7.1). Eight sherds were unable to be assigned to a group. These sherds may represent outliers but true members of one of the existing groups or may be individual specimens of as-yet undefined compositional groups. Six were most likely Caribbean in origin, and two were likely European. Given that Dominica was a British colony for much of the time in question, we also considered whether some sherds might represent products from British production zones such as London or Liverpool. When merged with a previous LA-ICPMS dataset of known British wasters (Bloch 2016), none of the sherds within this sample had a predicted assignment to a British source, again verified with Mahalanobis distance probabilities. This indicates that the unidentified European sherds most likely represent French or continental European

144 · Lindsay Bloch and Elizabeth Bollwerk

Table 7.1. Summary of sourcing results Compositional Group

Hand-Built, unidentified

WheelThrown

Total

Caribbean Earthenware Type 1

4

11

15

Caribbean Earthenware Type 2

2

8

10

Caribbean Earthenware Type 3

6

—

6

Huveaune

—

22

22

Vallauris

—

16

16

Unassigned Total

3

5

8

15

62

77

sources. The presence of British coarse earthenwares at Morne Patate is discussed in greater detail below. Comparison with Previous Sourcing Studies The three Caribbean types, which formed coherent compositional groups through both elemental and physical attribute variation, were then considered alongside the previous historic pottery sourcing study in the region (Kelly et al. 2008). This study involved neutron activation analysis (NAA), which is a bulk elemental technique, and petrography of earthenwares from Guadeloupe, Martinique, and St. Martin. The elemental results could not be directly compared because the incorporation of inclusions in the NAA analyses affected the elemental ratios. However, based on their published descriptions, our Type 1 is consistent with Group 1 defined in their study, which represents earthenwares produced in Guadeloupe. This group was defined through petrographic analysis by density and type of inclusions, which included feldspars and volcanic glass, also seen in the Morne Patate Type 1 samples under 10× magnification. Kelly and colleagues (2008) also found fine and coarse variants of Group 1, as we did. Elementally, NAA results for Group 1 showed enrichment in manganese, rubidium, and rare earth elements, which was not consistent with the LA-ICP-MS results, but this discrepancy could be explained by the presence of those elements in the inclusions rather than the clay fraction of this paste. The Morne Patate Type 2 established via LA-ICP-MS is consistent with Kelly and colleagues’ Group 2. This group, which was composed predominately of waster sherds from Martinique, was characterized by enrichment

Sourcing Coarse Earthenware at Morne Patate · 145

in transition metals such as chromium and iron. We found similar patterns in the LA-ICP-MS results. The description of these wares, having oxidized surfaces and reduced cores, and inclusions of abundant quartz and feldspars, is consistent with the Type 2 wares recovered from Morne Patate. Both wheel-thrown and hand-built vessels were present. The third Caribbean earthenware compositional group in this study consisted entirely of hand-built specimens. Inclusions were very large and predominantly quartz. It is possible that this type is also equivalent to Kelly and colleagues’ Group 3, but additional lines of evidence are required to assess their similarity. Assemblage-Level Results Based on the sourcing results and consequent type descriptions, all 3,091 sherds of coarse earthenware recovered from Morne Patate were cataloged, incorporating the new types. The results are summarized in Table 7.2. It is clear that of the European imports, France is the largest contributor, with at least 55% overall. However, it is notable that when the Morne Patate assemblage is divided into phases (Bates et al., this volume), there is a steady decrease in the proportion of European earthenwares over time. In the earliest phase (ca. 1760), French coarse earthenwares make up nearly two-thirds of the coarse earthenware assemblage, with most of those coming from Huveaune. By Phase 5 (ca. 1840), less than half of all coarse earthenwares are European, and only 41% can be concretely identified as French. Very few British coarse earthenwares were recovered from Morne Patate, but they are present in every occupation phase. The phasing of domestic contexts at Morne Patate also demonstrates temporal variation in the Caribbean coarse earthenwares. The expansive Type 1, which has several variants, is present throughout the assemblages. However, it is most common early in the plantation occupation. Types 2 and 3 are more common during later phases of occupation. This may signal a shift from direct trade with Guadeloupe to an intensification of trade with Martinique around the turn of the nineteenth century. However, if the Type 3 sherds are pre-Columbian in origin, this suggests some of the contexts may represent depositional fill or mixed deposits.

1768

1797

1818

1843

1866

1823

P01

P02

P03

P04

P05

Unphased

b

a

1801

1824

1810

1800

1792

1766

blue MCDa

564

33

33

177

116

176

29

Type 1

413

10

46

139

154

58

6

Type 2

78

—

21

47

3

7

—

117

8

4

30

22

32

21

421

12

35

130

117

82

45

3

—

—

3

—

—

—

1,281

23

76

563

262

323

34

10

4

2

4

—

24

—

—

4

8

10

2

180

11

6

43

39

55

26

European Caribbean Type 3 Unidentified Huveaune Saintonge Vallauris Albisola Britishb Unidentified

Best Linear Unbiased Estimator MCD (Galle 2010) Includes Red Agate, Coarse, Slipware, Staffordshire/N. Midlands, and Coal Measures types.

Total

MCD

Morne Patate Phase

Table 7.2. Coarse earthenware types by site occupation phase

3,091

97

221

1,140

723

747

163

Total

Sourcing Coarse Earthenware at Morne Patate · 147

Discussion The results of these analyses—the fact that French coarse earthenwares far outweigh British wares—confirm that enslaved and emancipated inhabitants of Morne Patate continued to procure and use imported French coarse earthenwares during and long after the French had ceded their unofficial political control of Dominica. Despite the British governance, Dominican households, including those at Morne Patate, mostly relied upon French or locally made coarse earthenwares from neighboring French colonies, not British goods. Types 1 and 2 provide tangible evidence of the trade of everyday objects with Guadeloupe and Martinique, respectively. Although we are accustomed to conceive of seas as boundaries, in the colonial Caribbean, the sea was a road that tied together colonies into extended communities sharing mutual needs. Both Guadeloupe and Dominica relied upon Martinique as the primary French entrepôt (Kelly 2009). Under French rule, Dominica became established as a provisioning ground for Martinique, creating trade networks among small landholders and large-scale planters that persisted over time. Despite the shift to British governance, Dominican households, including those at Morne Patate, continued to rely upon French or locally made coarse earthenwares from neighboring French colonies, not British goods. French coarse earthenwares far outweighed British wares. Yet, during the course of occupation at Morne Patate, assemblages of coarse earthenwares shifted from being dominated by French marmites and storage jars to locally made utilitarian vessels. A similar pattern is found in the British mid-Atlantic, with American coarse earthenwares nearly fully supplanting British imports over the course of the eighteenth century (Bloch 2016). Could this be evidence of plantation inhabitants choosing to buy local? Through her analysis of Guadeloupe probate inventories, Myriam Arcangeli (2015) found that recorders were keenly aware of the origins of their coarse earthenwares, so the transition was more than simple substitution. Moreover, many other classes of goods were not replaced by local products. For example, refined earthenwares continued to be procured from European sources throughout the eighteenth and nineteenth centuries. The ability of local coarse earthenwares to supplant imported wares suggests that they were perceived to be at least as good as European wares, they were

148 · Lindsay Bloch and Elizabeth Bollwerk

more readily accessible or cheaper, or they possessed some other quality that we have yet to ascertain. Conclusion In this study we have demonstrated that residents of Morne Patate had a variety of earthenware products at hand to meet their daily needs. What is particularly striking is the persistence of French and locally produced coarse earthenwares despite the takeover of the island by the British Empire in 1763. The reasons behind the continuation of use are not immediately clear. Given that the French were allowed to maintain their landholdings in Dominica even after the British took control, it could have taken time for the primary trading networks to shift away from those focused on France and French landholdings in the Caribbean to those circulating British goods. Alternatively, Kelly and colleagues (2008) suggest that, in at least some cases, the driving factors may have been what the vessels contained—that local earthenware may have reached Dominica as a carrier for other commodities—rather than serving as the primary commodity itself. More research is needed to tease apart how these and other factors may have impacted the strategies enslaved individuals used to participate in local and global markets. Nevertheless, we show that the increased use of ceramic compositional analysis will help unravel these questions by providing a useful method of tracing the movements of ceramics over time and space. Type Descriptions The following type descriptions for coarse earthenwares recovered at Morne Patate were developed for use in cataloging by the Digital Archaeological Archive of Comparative Slavery (DAACS), following the DAACS Cataloging Manual (DAACS 2018). Vallauris General Description Vallauris is a medium- to thick-bodied coarse earthenware whose paste color ranges fall into the buff, pink, and orange categories in the DAACS Paste Color Range. The core is often pink with whiter, oxidized sections near the exterior. It contains abundant inclusions of sand-sized particles of

Sourcing Coarse Earthenware at Morne Patate · 149

iron oxides, white rock (often gypsum), and quartz. The interior is nearly always lead-glazed with a clear glaze, which results in a glazed interior color ranging from light orange to dark reddish brown, depending on firing conditions. The exterior is typically unglazed. Burned or heavily reduced Vallauris may resemble Caribbean coarse earthenwares (i.e., “locally made CEW”). Transparent yellow glaze may be an indicator that the heavily burned sherds are indeed Vallauris. Residue/sooting/fire clouding on the exterior is also common on Vallauris sherds. Forms According to Heather Gibson (2007:169–170) forms include marmites (cooking pots) and poêlons (deep skillets with long handles), both of which often have charred exteriors, showing evidence of their use for cooking on open flame. Marmites recovered have several variations: large straight-sided marmites measuring 22–25+ cm in diameter and small straight-sided marmites measuring less than 20 cm in diameter. References Arcangeli 2015; Gibson 2007; Losier 2012; Métreau 2016; Petrucci 1999 Huveaune General Description Huveaune is a thin-bodied coarse earthenware. The vessels feel lightweight for their size, and the paste is rather chalky in texture. The paste color is orange to red, and paste contains very few visible inclusions, generally redder and more refined than Vallauris. Both the interior and exterior surfaces are usually lead-glazed. Glaze colors are predominantly transparent, producing a caramel or ginger color. Some interiors have a thick, white slip, resulting in creamy white- or yellow-glazed surfaces. Green-glazed vessels are rare in Caribbean contexts, as are vessels with slip-trailed decoration. Forms According to Gibson (2007:171–172) shallow milk pans (milk pan–style vessel with a flat base, short inverted, truncated cone shape, and a thick folded rim) are typical, as are chamber pot–style vessels, which are taller and more narrow than milk pans with straight sides.

150 · Lindsay Bloch and Elizabeth Bollwerk

References Abel 1987; Abel and Amouric 1991; Gibson 2007; Losier 2012; Métreau 2016 Saintonge General Description Saintonge is relatively thin with buff to pink paste that tends to have large chunks of white rock. Otherwise, Saintonge sherds are relatively smooth. Clear lead glazes can range in color from pale yellow to olive green. Green lead glaze on the interior can sometimes occur over a white slip. Forms Saintonge takes a variety of tableware and utilitarian forms, including jugs, storage jars, and milk pans. References Losier 2012; Métreau 2016; Monette 2010 Biot General Description Biot is a very thick-bodied, coarse earthenware with buff paste that is made gritty by abundant quartz, calcareous, and ferruginous inclusions. The interior surface is nearly always covered with a white slip covered in a clear lead glaze. The majority of the exterior surface is typically unglazed with the exception a small strip that runs from the neck to the upper body or shoulder incidental to interior glazing. These vessels were hand-built. Forms The most common forms are large, barrel-shaped storage jars that were used for both wet and dry goods. These jars are quite large, often between three and four feet tall. References Arcangeli 2015; Losier 2012; Métreau 2016

Sourcing Coarse Earthenware at Morne Patate · 151

French Coarse Earthenware General Description The ware type “French Coarse Earthenware” is used to generally describe coarse earthenware sherds that do not easily fall into the Saintonge, Huveaune, Vallauris, or Biot ware categories. However, they display a constellation of characteristics seen in the identifiable French types, with familiar French CEW attributes in a variety of combinations and degrees. These sherds are clearly European in manufacture. Sherds whose paste color matched one of the Redware color chips and did not exhibit any similarities to French ware types (e.g., paste inclusions and density of paste) were cataloged as Redware. Morne Patate Type 1 General Description Type 1 exhibits a range of visible variation, which was also reflected in the compositional results. While elementally somewhat similar to Type 2, Type 1 sherds have a paler paste than Type 2 and more inclusions. The group may be separated into Type 1a and Type 1b on the basis of inclusion size and density. Type 1a inclusions are finer and are usually “Greater than 7.5%,” while Type 1b inclusions are larger and usually “Greater than 15%.” Type 1b inclusions are also often visible on the exterior and interior surfaces. Visible surface inclusions are not a hallmark of Type 1a. Type 1a is also more thinly potted than Type 1b, which is chunky and thicker. This may indicate a functional difference between Types 1a and 1b, though fragmentation has hampered the identification of vessel types at Morne Patate. Type 1a has very smooth exterior surfaces, likely a result of production on the pottery wheel. A self-slip is often discernible on surfaces of 1a and 1b sherds. Sooting/residue is occasionally present on 1a and 1b sherds. Morne Patate Type 2 General Description Morne Patate Type 2 is generally a finely potted coarse earthenware whose body is thinner than Huveaune, Vallauris, and MP Type 1. Type 2 paste

152 · Lindsay Bloch and Elizabeth Bollwerk

color is deep red with very few inclusions. The paste is highly fired, which produces sharp-edged breaks. The vessels are generally wheel-thrown although occasionally a sherd displays evidence of molding. Both interior and exterior are unglazed/bisque. The exterior surfaces are often smoothed or burnished, sometimes to the extent that the sand-sized quartz and volcanic inclusions cannot be seen or felt. They are often a little rougher on the interior, which does not appear to have been smoothed. Nearly all Type 2s are fully oxidized, with no firing core. Morne Patate Type 3 General Description Morne Patate Type 3 is composed of a thin, fine red paste with chunky, highly visible quartz inclusions. The paste is fired at a lower temperature than other coarse earthenwares at Morne Patate. The edges readily erode. Paste color is a deep red, with uneven oxidation through the body. The paste also has volcanic rock, some of which appears to be micaceous. The exterior is occasionally burnished in addition to a self-slip. The exterior and interior surfaces are unglazed. This ware type is almost always hand-built.

8

Y The Environmental Archaeology of Subsistence and the Socioecological Landscape at Morne Patate Diane Wallman and Sarah Oas

From the seventeenth through twentieth centuries, the Caribbean region experienced unprecedented demographic and environmental change, with the rise and fall of sugar monoculture and the institution of chattel slavery. These transformations were a result of power imbalances at many scales, and the economic, ecological, and social consequences of the migrations and interactions were significant and long-lasting. During the colonial period, enslaved communities developed socioecological practices to survive and adapt within the oppressive plantation structure through the establishment of diverse domestic economies, internal marketing systems, and creative subsistence strategies. Environmental data, specifically animal and plant remains from archaeological sites, provide insight into these localized histories and offer important perspectives on the human ecodynamics of plantation landscapes. This chapter presents the results of paleoethnobotanical and zooarchaeological analysis from Morne Patate. Through an investigation of the subsistence-based practices developed by the enslaved laborers and habitants of the plantation, Morne Patate provides a unique case study to explore the socioecological dimension of colonial Dominica. A volcanic island dominated by steep mountainous topography, Dominica hosts a rich diversity of ecosystems and biota. This diversity highlights the amount of land and variability in the type of soils to support export and subsistence agriculture. The island is one of the most biologically diverse in the Caribbean and is home to 155 plant families and over 1,200 species of vascular plant and tree species (FAOSTAT, http://www.fao.org/statistics/ en/; Nicolson et al. 1991). Dry scrub woodland plant communities abound

154 · Diane Wallman and Sarah Oas

on the leeward side of Dominica where Morne Patate is situated, although much of this woodland has been heavily modified by human subsistence activities. Endemic fauna on the island are also diverse, with 18 species of terrestrial mammals (mostly bats), 19 species of reptiles, nearly 500 recorded species of fish, and an incredibly high diversity of avifauna. This ecological setting provided abundant natural resources for colonial settlers on Dominica, and in combination with the importation of plant and animal species from Europe, Asia, and West Africa, colonial residents developed unique foodways and culinary traditions. Examining household food production and consumption practices within the Caribbean plantation system allows for insight into these localized political ecologies and how individuals, households, and communities managed their resources in this dynamic socioecological context. Historical Background As discussed in Chapters 1 and 2, until 1763 Dominica was ostensibly controlled by the French and remained one of few refuges of the Kalinago people as it was declared a “neutral” territory, along with several other islands in the Lesser Antilles. Before this time, Dominica was scattered with small mixed-crop agriculture estates of coffee (Coffea sp.), cacao (Theobroma cacao L.), cotton (Gossypium sp.), plantain (Musa sp.), manioc (Manihot esculenta Crantz), and taro (Colocasia esculenta [L.] Schott) and contained approximately 200 coffee estates run primarily by French-Caribbean settlers and missionaries (Atwood 1791; Hauser 2015; Honychurch 1995; Lenik 2014). After 1763, under the authority of Britain, sugar production developed on the island, much later than many of the islands in the Lesser Antilles. As a colony, Dominica was characterized by difficult terrain, absentee landowners, and relative autonomy from direct control by the British crown (Hauser 2015; Hauser and Armstrong 2012; Honychurch 1995). With the adoption of sugar production and the associated increase in enslaved labor, the social and ecological relations on the island transformed. From early on in the plantation system in Dominica, enslaved laborers were relatively self-reliant regarding subsistence, cultivating crops on provision grounds and yard gardens, raising livestock, and harvesting the island’s bountiful resources (Hauser 2015; Trouillot 1988). These practices that began during slavery produced what Michel-Rolph Trouillot (1988:73) calls “a peasant breach in the slave mode of production” in Dominica, with

The Environmental Archaeology of Subsistence and the Socioecological Landscape · 155

enslaved individuals participating in small-scale subsistence agricultural practices and marketing surplus products. Sidney Mintz (1961) refers to this process as the development of a “proto-peasantry” during slavery in similar ways throughout the Caribbean. Thomas Atwood (1791:256), a government official of Dominica in the eighteenth century, notes in his history of the island that some planters in Dominica did provide some weekly allowances to the enslaved community, including “biscuit, indian corn, beans, salt fish, mackerel or herrings.” On plantations in Dominica, however, similar to other British and French colonies in the Lesser Antilles, he observes that the enslaved laborers mostly self-provisioned and were allotted one day a week, typically Saturday, in addition to Sunday to work their gardens, tend to livestock, fish, and market their goods. In 1834 Britain emancipated enslaved individuals in Dominica, with an “apprenticeship” period to transition to full freedom in 1848. Upon full emancipation, great numbers of formerly enslaved laborers left plantations, establishing settlements on the interior and coast, continuing traditions of subsistence gardening, and growing cash crops for market, fishing, and other small-scale economic activities (Honychurch 1995; Trouillot 1984). After former enslaved laborers gained their freedom, there began a “continuous struggle” between British colonists/officials and the small-scale cultivars (ex-slaves) in Dominica to “establish a form of control over the labour process” (Trouillot 1989:705). These processes and struggles, along with Dominica’s geography and its liminal space between empires during the colonial period, had significant influence on the island’s social, economic, and ecological trajectories. Several accounts from the eighteenth and nineteenth centuries provide insight into the flora and fauna that composed the diets of the inhabitants of Dominica, including that of Thomas Atwood, already mentioned in 1793, and Joseph Sturge and Thomas Harvey, two British visitors to the island in 1837, on the eve of emancipation. Atwood (1791:41) notes that enslaved households bred hogs, horses, horned cattle, sheep, goats, rabbits, guinea pigs, and agouti and also procured numerous species of sea and river fish, including “groupers, cavallies, snappers, silks, baracutas, kingfish, mackerel, jacks, sprats.” He also observed the common consumption of several species of land crabs, lobster, crayfish, lizards, iguana, and frogs (crapaux) and of imported and wild birds such as geese, ducks, guinea fowls and others. In terms of the staple plants, he identifies the common provisions on the

156 · Diane Wallman and Sarah Oas

island as plantain, banana, manioc or cassava, yam (Dioscorea sp.), sweet potato (Ipomoea batatas [L.] Lam.), cushcush (Dioscorea trifida L.f.), tania (Xanthosoma sp.), eddoe (Colocasia esculenta [L.] Schott), wild yam (Rajania cordata L.), Guinea corn (Sorghum bicolor [L.] Moench), Indian corn (Zea mays L.), and rice (Oryza sp.) (Atwood 1791:84). Table 8.1 includes a complete list of the species mentioned in this text. According to Joseph Sturge and Thomas Harvey (1838:101), during the transitional period to full emancipation, apprentices received “no allowances at all except of clothing and presents of pork, flour and fish at Christmas. They support themselves by cultivating their grounds on the steep sides of the mountains, and by catching sea and river fish.” Sturge and Harvey (1838:101) observed “many wild hogs in the woods, and a small species of boa constrictor, the iguana is not uncommon; and there is a large edible frog, which is caught in great numbers and esteemed a delicacy.” The crapaux, or mountain chicken (Leptodactylus fallax), is a frog that is still considered a national delicacy in Dominica. As described by Atwood (1791:57), this species was harvested at night under torchlight and cooked in fricassees or soup. Atwood also observed the frequent consumption of the grugru worm, or grubworm, by inhabitants of Dominica, which was roasted with sauce of lime, salt, and pepper. Morne Patate As discussed elsewhere in this book, with its precipitous topography, settlements in Dominica are situated primarily in valleys of mountain watersheds and are referred to as enclaves (Hauser 2015; Hauser and Armstrong 2012; Lenik 2011; Trouillot 1988), a settlement pattern that has largely persisted from the colonial period until today. As noted by mid-nineteenth-century visitors to Soufriere Bay, “little of Dominica, except the river levels and the fertile sides of the ravines, has been brought to cultivation” (Sturge and Harvey 1838:98). The site of Morne Patate is in the enclave of Soufriere on the southwestern coast of the island. The site lies in a valley approximately 213 m up and approximately ½ km (geodesic distance) from the Soufriere Bay, which is formed by two projecting reefs at the southwestern tip of the island. Investigations focused on several loci, including occupation areas associated with enslaved laborers, and on contexts associated with the estate house and the main kitchen, with organic materials recovered in varying quantities throughout the site.

Table 8.1. Local species noted by Atwood with corresponding binomial names Staple provisions (Atwood 1791:84) Plantains Bananas Manioc/cassava Yams Sweet potatoes Cushcushes Tania Eddoe Wild yams Guinea corn Indian corn Rice

Musa sp. Musa sp. Manihot esculenta Dioscorea sp. Ipomoea batatas Dioscorea trifida Xanthosoma sp. Colocasia esculenta Rajania cordata Sorghum bicolor Zea mays Oryza sp.

“Island plants” (Atwood 1791:100) Oranges Lemons Limes Citrons Shaddocks Water lemons Granadillas Sapodillas Pomegranates Alligator pears Mountain pears Pineapples Rose apples Star apples Sugar apples Custard apples Mamma apples Guavas Sea-side grapes Cocoa nuts Conk nuts Soursops Papaws Cashew apples Tamarinds

Citrus sp. Citrus limon Citrus sp. Citrus medica Citrus maxima Passiflora laurifolia Passiflora ligularis Manilkara zapota Punica granatum Persea americana Opuntia stricta Ananas comosus Syzygium jambos Chrysophylum cainito Annona squamosa Annona reticulata Mammea americana Psidium guajava Coccoloba uvifera Cocos nucifera Passiflora maliformis Annona muricata Carica papaya Anacardium occidentale Tamarindus indica

Imported Botanicals (Atwood 1791:101) Strawberries Raspberries Muscadine grapes

Fragaria sp. Rubus sp. Vitis rotundifolia (continued)

Table 8.1—Continued Figs Musk melons Watermelons Cucumbers Gourds Pompions English, American, and West Indian beans and peas Cabbages Carrots Turnips Parsnips Lettuces Radishes Horseradish Asparagus Artichokes Spinach Celery Onions Shallots Thyme Sage Mint Rue Balm Parsley

Ficus sp. Cucumis melo Citrullus lanatus Cucumis sativus Cucurbitaceae Cucurbita sp. Fabaceae Brassica oleracea Daucus carota subsp. sativus Brassica rapa subsp. rapa Pastinaca sativa Lactuca sativa Raphanus sativus Armoracia rusticana Asparagus officinalis Cynara cardunculus Spinacia oleracea Apium graveolens Allium cepa Allium cepa var. aggregatum Thymus sp. Salvia sp. Mentha sp. Ruta sp. Melissa officinalis Petroselinum sp.

Fish and shellfish (Atwood 1791:41) Groupers Cavallies Snappers Silks Baracutas Kingfish Mackerel Jacks Sprats Lobsters Conk Wilks Crabs

Serranidae Caranx hippos Lutjanidae Lutjanus campechanu Sphyraenidae Scomberomorus cavalla Scombridae Carangidae Clupeidae Panulirus argus Strombidae Cittarium pica Gecarcinidae

Note: Categories of plants/animals are as distinguished by Atwood as are local names of species. Authors made best estimates for corresponding scientific classification.

The Environmental Archaeology of Subsistence and the Socioecological Landscape · 159

The faunal and botanical samples were collected from a variety of contexts, including house and yard deposits, pit features, hearths, and the provision grounds from the plantation’s occupation. Botanical and faunal materials were recovered from the 58 flotation samples, of approximately 10 L each (see Oas and Hauser 2018), that were collected during excavations. Faunal materials were also recovered from dry screening. Due to differences in taphonomic processes and preservation, the recovery of faunal and botanical remains varies among the contexts. Those features or areas that allowed for good preservation of botanical remains did not necessarily contain well-preserved faunal materials and vice versa. Combined, however, the environmental data offer a representative sample of foodways throughout the plantation’s operation, from both enslaved laborer occupations as well as the estate house. Zooarchaeological Analysis Faunal samples were recovered primarily from flotation samples and dry screening of strata, and features and were analyzed using standard zooarchaeological methods. The remains came from household and yard deposits, pit features and hearths associated with slave cabin, and a midden associated with the kitchen of the estate house. Regarding the samples from the estate house midden, it is difficult to ascertain the origin of the deposits or exactly who on the plantation consumed the resources identified in the midden. The food was very likely prepared by enslaved cooks and could represent the consumption patterns of the planters, the enslaved, or both. For this analysis, the loci are divided into two general contexts, those from the pre-sugar period, and those after the introduction of sugar (post-1770). Only specimens identified to the level of Family are included here. Overall, the faunal remains indicate a reliance on domestic animals throughout the plantation’s occupation, including cattle (Bos taurus), sheep/ goat (Ovis/Capra), chicken (Gallus gallus), and primarily pig (Sus scrofa) (Table 8.2). The imported herbivore domesticates (cattle, sheep/goat, and pig) are represented largely by teeth, and the postcranial specimens from medium and large mammals are highly fragmented, moderately weathered, and mostly unidentifiable. As preservation is not ideal at the site, this bias could represent density-mediated attrition, but the presence of primarily cranial elements suggest that these animals were likely raised on site. Analysis identified dog and Old World rodents (Rattus sp. and Mus sp.). Further,

Table 8.2. Identified fauna at Morne Patate (only includes taxon identified to Family or below) MNI Phase 1

Taxon

Common Name

Mus sp.

Old World Mouse Old World Rat

1

Agouti

1

Rattus sp. Dasyprocta leporina Canis domesticus

NISP Phase 1

MNI Phase 2

NISP Phase 2

1

3

1

3

22

1

0

0

1

25

Bos taurus

Domesticated Dog Domestic Cattle

1

3

4

32

Ovis/Capra

Sheep/Goat

1

1

3

20

Sus scrofa

Domestic Pig

1

2

4

35

Gallus gallus

1

1

1

4

Boa constrictor

Domestic Chicken Boa constrictor

2

115

Leptodactylus fallax Albulidae

Mountain Chicken Bonefish

1

1

1

1

Clupeidae

Herrings/shads

2

2

13

13

Belonidae

Needlefish

3

3

29

29

Hemiramphus sp.

Halfbeak

14

14

25

25

Exocoetidae

Flying Fish

1

1

1

1

Holocentridae

Squirrelfish

1

1

Serranidae

Grouper

1

1

15

15

Rachycentridae

Cobia

1

1

Carangidae

Jacks

6

6

Lutjanidae

Snapper

2

2

1

1

Haemulidae

Grunts

2

2

7

7

Coryphaenidae

Mahi Mahi

1

1

Labridae

Wrasses

2

2

Acanthuridae

Surgeonfishes

1

1

Balistidae

Triggerfish

4

4

127

364

Total

32

35

The Environmental Archaeology of Subsistence and the Socioecological Landscape · 161

the inhabitants exploited the native snakes (Boa constrictor), agouti (Dasyprocta leporina), and there is evidence of consumption of the mountain chicken (Leptodactylus fallax) in an enslaved laborer house area. Overall, the results of the analysis indicate that the inhabitants of Morne Patate consumed a diverse variety of fish, with 15 distinct species identified. These are all taxa that could be found in the Soufriere Bay or in the surrounding waters. The Soufriere Bay is an extinct volcanic crater, with a complex of reefs along the Scott’s Head (or Cashacrou Peninsula) to the south and deep plunging waters directly off the west edge. This pattern of both reefs, plunging pelagic waters, and the interface of the Atlantic and Caribbean waters create habitats for both pelagic and reef fish. The fish species in the assemblage from Morne Patate reflect this mix of habitats. The households on the estate could have procured these species without having to leave the bay. While sample size is likely somewhat of a factor, species richness increases during the second phase of the plantation’s occupation. In particular, the number of fish species procured increases, and there is evidence of the consumption of crapaux and potentially boa. The results suggest that household-level fishing economies developed as early as the mid-eighteenth century as important domestic livelihoods. The fish recovered at Morne Patate were available locally in the bays nearby the plantation, and initial observations of their size suggest they were procured using nets and traps or pots. Historic accounts corroborate these procurement methods. As described by Sturge and Harvey (1838:94–95), fisherfolk in Dominica with “three or four canoes, loaded with stones” would take a large net about ten feet deep, and from sixty to one hundred yards in length, to some distance from the shore, which they let down; the lower edge being weighted with lead, the upper supported by pieces of cork. The stones in the canoes are then thrown with great violence into the sea in such a direction as to frighten the fish towards the shore, when a canoe at each extremity drags the net rapidly to the beach, and is secured. The species identified are many still fished today by local fisherman in Soufriere Bay. Modern studies reveal that the fishing techniques and catch composition remain consistent with those suggested by archaeological evidence. Contemporary research indicates that the primary gear used by modern fisherman is the seine net, followed by two types of traditional fish pots for

162 · Diane Wallman and Sarah Oas

catching a wide variety of reef fish species and predatory species such as dolphinfish (Coryphaena hippurus), groupers (Serranidae), morays (Muraenidae), and squirrelfishes (Holocentridae) (Theophille and Hutchinson 2012). With regard to the boa, two individuals were identified in the midden behind the kitchen of the great house. According to Atwood (1791:53–54), the fat of these snakes was esteemed as an excellent remedy for rheumatism or sprains, and it was mixed with a strong rum and rubbed on the skin. The flesh was also eaten by the French, and the skin made into sheaths. There are also some foodstuffs that elude archaeological investigation, such as the grubworm. Also noted in the ethnohistorical literature in Dominica is the consumption of “trez trez” or tri tri, which are juvenile gobi. Atwood (1791:40) describes enslaved people acquiring these fish, which were caught in large volumes in baskets lined with a tablecloth or sheet held down with stones. This is a practice that continues today on Dominica but will not leave an archaeological signature. Paleoethnobotanical Analysis Pre-1770 The remains recovered from pre-1770 contexts included mostly fruits and vegetables with fewer cereal grains. One exception to this was a hearth feature associated with a pre-sugar enslaved household containing maize, an African cereal, Guinea corn or sorghum, and large quantities of charred root and shoot material (Table 8.3). Excluding the cereal found in the hearth feature, the only other cereals recovered were a few maize cupules from the lowest level of House Area A. This suggests that both maize and potentially sorghum were early and enduring additions to household foodways at Morne Patate. One additional domesticated species, eggplant (Solanum melongena L.), was also recovered from two enslaved households. Eggplant is a domesticate from Southeast Asia that spread into the Mediterranean and eventually Africa, and African varieties of eggplant were introduced and commonly grown by slaves throughout the Caribbean (Carney 2016; Higman 2008:173–74; Hodge and Taylor 1957:604–605; Oas and Hauser 2018). Local plants served as sources of flavoring, nutrition, and medicine for enslaved households at Morne Patate. Excavations recovered several species of seeds and fruits endemic to Dominica, pre-dating 1770, that are available

6

8

6

3

4

31

6

17

5

3

House Area A

House Area D

House Areas E and G

Stables

Estatea

Post-1770

House Area A

House Area B

House Area C

Estate Midden

X

X

X

32%

X

X

15%

Maize

X

3%

Barley

Enslaved household context covered by later estate house.

27

Pre-1770

a

No. Samples

Context

X

6%

Millet

Cereals

Table 8.3. Ubiquity of major botanical remains at Morne Patate

X

6%

4%

Sorghum

X

X

7%

Eggplant

X

X

X

42%

X

X

7%

Guava

X

X

X

10%

X

4%

Passion fruit

Fruits and Vegetables

X

3%

Okra

X

3%

Coffee

X

10%

Fennel

Beverages and Seasonings

164 · Diane Wallman and Sarah Oas

today in dryland forests adjacent to the plantation. These fruits include guava, hackberry (Celtis sp.), and wild lime (Zanthoxylum sp.) (Table 8.3). Guava is a historically and currently popular fruit indigenous to the Caribbean (Parry 1955). This fruit may be eaten fresh or stewed or otherwise processed into purees, pastes, jellies, jams, or cheeses (Higman 2008:198). Hackberry and wild lime are commonly found today in the dry scrub woodland plant communities on the leeward side of Dominica that might reflect either dietary or medicinal uses (Ayensu 1981:176; Nicolson et al. 1991). Post-1770 The recovery rate of seeds, particularly domesticated cereals, and other plant materials more than doubled in household samples after the introduction of sugar at Morne Patate (Table 8.1). The diversity of domesticated cereals also increased as, in addition to maize and sorghum, both millet (Pennisetum glaucum [L.] R.Br.) and barley (Hordeum vulgare L.) were identified. Among the cereals, maize clearly dominates, recovered as kernels, cupules, and cob fragments in nearly a third of the post-1770 samples. The recovery of cobs suggests that at least some of the maize was likely grown and cooked (likely roasted) locally. Given fertile ground, maize is a highly productive crop that likely made it a desirable addition to gardens and nearby provision grounds (Higman 2008:223). As discussed earlier, Atwood (1791:84) suggests that maize (Indian corn) was grown on the island for subsistence. Additionally, maize continues to be grown on Dominica today (FAOSTAT, http://www.fao.org/statistics/en/). In House Area A, we identified various grains and African cereal crops (Oas and Hauser 2018). Excavation from floor and feature contexts recovered several grains of pearl millet and probable pearl millet (cf. P. glaucum), which is an important arid-tolerant staple of sub-Saharan Africa. Additional African cereals identified include several partially fused grains of sorghum and some accompanying chaff from House Area B. Sorghum is a valuable and hearty grain with relatively high yields and less sensitivity than maize to hot and dry conditions (Berleant-Schiller 1983; Muchow 1989). Throughout the Caribbean and the United States Southeast, sorghum was a common addition to provision and plantation agriculture due to its high yields and tolerance of aridity and was grown as fallow for sugar fields (Carney 2016; Carney and Rosomoff 2011; Higman 2008:229–32).

The Environmental Archaeology of Subsistence and the Socioecological Landscape · 165

In House Area B, one hearth feature contained several charred barley seeds and barley chaff. The presence of this crop is surprising as wheat and barley are temperate-adapted plants, and historical accounts suggest that the cultivation of these plants had little to no success throughout the Caribbean (Carney and Rosomoff 2011:105; Higman 2008:36, 235). Other domesticated plants identified in House Area B include a combination of Old and New World species, including okra (Abelmoschus esculentus Moench.), coffee, and fennel (Foeniculum vulgare Mill). Okra was an important African introduction that was a universal staple of provision grounds in the Americas (Carney 2016; Carney and Rosomoff 2011:135, 238; Higman 2008:174–175). Okra was used in a multitude of ways: it was incorporated into soups and stews; the leaves were prepared like spinach; and the buds could also be cut, processed, dried, or boiled and served in variety of dishes. The presence of coffee cherries in a houseyard floor context in House Area A corroborates the pre-1830s context of the household before the subsequent coffee blight (Gurney 1840:63). The coffee cherries also indicate a degree of local coffee production for personal consumption or sale or perhaps some illicit harvesting of coffee from plantation production. Fennel, a plant of Mediterranean origin with edible aromatic leaves and seeds, was identified only in the estate kitchen midden. The presence of this plant in the estate context offers a point of contrast between the foods and flavorings used in enslaved and estate household contexts. Fennel was, and still is, a common plant used in both French and English cuisine. A number of local fruits are also represented in post-1770 samples. These include cocoplum (Chrysobalanus icaco [L.] L.), guava, hackberry, passion fruit (Passiflora sp.), and wild lime. Cocoplum is a small tree found in coastal areas throughout the Caribbean, and its fruits are commonly eaten and used for jams. Passion fruit are often eaten fresh but are also commonly made or mixed into drinks. The climbing vines of passion fruit species often thrive in the living fence materials of fruit trees and shrubs around fields and gardens and provide an additional source of food and flavoring (Higman 2008:204). The expansion of agricultural fields, provision grounds, and household gardens post-1770 would all have provided new areas for passion fruit vines to colonize. In general, there was a lower recovery rate of plant foods in the pre-1770 contexts, which could have occurred due to several reasons. This difference may be explained by taphonomic factors or by the fact that fewer people resided on the plantation before the transition to sugar production. The

166 · Diane Wallman and Sarah Oas

pattern may also indicate that enslaved laborers were not growing as many of their own provisions in the earlier phases of the plantation. Throughout the sequence, however, we recognized a mixture of local and nonlocal cereals, wild plants, and imports. Similar to issues of preservation bias with faunal remains, certain plant foods will leave no macrobotanical trace in the archaeological record but may be identified through additional analyses, such as pollen or starch grain. At Morne Patate, excavations recovered various cassava griddles and cassava processing tools (see Chapters 4 and 6). These objects strongly suggest the status of cassava as a staple crop, which corroborates the significance of this tuber in the diet of the enslaved. Conclusion This research contributes to an understanding of the role of enslaved individuals and communities and their descendants in the development of island traditions and landscapes. The millions of Africans taken captive who survived the Middle Passage encountered entirely new social and natural environments in the Americas under extreme duress. Despite these pressures, enslaved Africans and their descendants negotiated this new reality in various ways. The management of, and adaptation to, local landscapes for subsistence and livelihoods was fundamental to coping with the challenging economic, environmental, and power structures that defined plantation agriculture. Under colonial rule, disenfranchised and enslaved communities developed informal ways to resist or respond to the dominant system, such as small-scale fishing, pastoralism, horticulture, and agriculture. Enslaved peoples incorporated local wild and domestic species with plants and animals from Europe, West Africa, and Asia, developing a creolized system of foodways, exploiting and managing their local landscape to support survival and community well-being in spite of their enslavement. At Morne Patate, the faunal and botanical evidence offer a detailed picture of the socioecological transformations that took place on Caribbean plantations. Over time the enslaved laborers and planters at the estate developed increasingly complex and localized subsistence practices as the plantation shifted to a focus on sugar production. Enslaved laborers at Morne Patate raised Old World domesticates, hunted local fauna, fished in the local waters, and cultivated European, African, Asian, and local plants for subsistence and medicinal uses. While these practices did not have the same

The Environmental Archaeology of Subsistence and the Socioecological Landscape · 167

level of impact as large-scale monoculture on the landscape, they became an important part of the socioecological landscape in colonial Dominica. These practices endure on Dominica, with small-scale agriculture and fishing playing a very important socioeconomic role as a source of income and food security for many islanders.

9

Y Conclusion Resilience and Capacity Building in the Age of Empires William F. Keegan

Mark Hauser begins this collection with a simple question: What happens when you replace one crop with another? To find an answer, the contributors draw attention to the agricultural ecology, nonmarket economics, and political economy. They address three episodes of stability punctuated by what some must have viewed as catastrophic events—British rule and emancipation. The contributions highlight the different objectives and outcomes of a dual economy composed of the hegemonic plantation focused on maximizing profit in the external market and the local subsistence economies imposed on enslaved labor. While the former experienced repeated episodes of “boom” and “bust” (Chapter 1), the self-reliance of the slave economy achieved resilience through adjustments to local farming and fishing conditions (Chapter 8), through the development of local and interisland networks of production and exchange (Chapter 7), and through local institutions based on filiation and affiliation (Chapter 2; see Keegan 2018). Interrogating Slavery Morne Patate is located in the southwestern corner of Dominica, near the capital of Roseau. It operated as a plantation during the eighteenth and nineteenth centuries, spanning several significant political and economic transitions on the island. These included the shift from French control to British legal authority, from diversified agriculture to large-scale sugar production, and from slavery to emancipation. The island was ceded to Britain in 1763, and emancipation occurred on Dominica in 1838.

Conclusion: Resilience and Capacity Building in the Age of Empires · 169

Archaeological investigations under the direction of Mark Hauser were initiated in 2010. Virtually every aspect of Morne Patate has been explored by his teams, especially the plantation industrial zones, slave housing, and gardens (Chapter 1). A detailed chronology for the entire site was established using frequency seriation, mean ceramic dates, correspondence analysis, and termini post quem based on the DAACS database (Chapter 4). These methods yielded temporal controls for archaeological deposits from activity-specific contexts. There are advantages when competing empires don’t really want you. Dominica was not officially claimed by either Britain or France during the early colonial period. The island was “neutral” territory, even if it was French planters who first settled here. A significant reason is that the nine volcanic peaks create an essentially vertical landscape that was not conducive to the forms of plantation agriculture that developed on neighboring islands. The terrain also limited the location of core support centers (towns or cities) that could serve the needs of dispersed plantations in the surrounding periphery (Chapter 1). Finally, communication between these towns was further restricted to overwater networks, which resulted in what MichelRolph Trouillot (1988) has called “enclaves.” Surviving Slavery The chapters in this volume highlight three mechanisms that contributed to the emergence of a resilient Dominican peasantry. First, by encouraging marriage and concentrating laborers in tightly packed villages, the planters unwittingly promoted the emergence of local institutions (Chapter 5). The result was a distinct slave society anticipating emancipation (Chapter 3). Organizational structures self-organized to facilitate independence from the plantation (see Ensor 2013). Second, Dominican planters favored subsistence and secondary export crops from the beginning (pre-1763); their markets were neighboring French colonies versus the French metropole (Chapter 2). The enslaved who labored on these estates cultivated provisions for themselves and the estate. Thomas Atwood provides an accounting of provisions on Dominica in the eighteenth century (Chapter 8:Table 8.1). Diane Wallman and Sarah Oas analyze the faunal and botanical samples collected from different contexts on the Morne Patate estate to further document the diversity of plant and animal use (Chapter 8:Table 8.2). In addition to the typical suite of cow,

170 · William F. Keegan

sheep, chicken, and pig, inhabitants consumed a wide diversity of meats (e.g., rabbits, guinea pigs, agoutis, land crabs, iguanas, frogs, sea and river fishes, and birds). They cultivated plantains, bananas, cassava, yams, sweet potatoes, sorghum, millet, maize, eggplant, and several others and even experimented with temperate crops such as barley, which are not known to grow well in tropical climates. In sum, the subsistence and local export economies during the 100 years prior to emancipation were diversified and refined in ways that encouraged independence. Third, lacking formal access to the mercantilist networks intended to govern trade between metropole and colony, early eighteenth-century French planters in Dominica favored crops that could be illicitly trafficked to family members and business associates in neighboring French colonies rather than exported all the way to Europe. The shift to sugar production failed to change this peripheral status. Dominica lacked the production capacity and large harbors needed to attract external merchants and cargo ships. The study of local earthenware vessels clearly demonstrates the relative isolation in which the Dominican economy operated (Chapter 7). The island relied heavily on products produced in neighboring islands—and did so through smaller-scale networks of exchange that involved local vessels (see Hauser, Chapters 1 and 2). Such local networks must have involved the participation of enslaved labor to crew the vessels and porter their loads. Imagining Slavery Living on the margin presents a “refracted reality” of the dominant image. Through refraction, a truer image of “reality” is observed (Keegan 2007). The use of artwork by the Italian Painter Agostino Brunais (1760s and 1770s) to interpret slave houses in Dominica is in keeping with this theme (Chapter 6). Brunias’s paintings depict a sanitized landscape in which “the island’s appeal as a potentially civilised space, as well as a profitable one, had to be emphasized” (Kriz 2008). With housing as the backdrop, the foreground scenes of “everyday life” depict the planters, the enslaved and free black class, and the native Kalinago. The discovery of an unusual oval structure within the highly regimented slave village, diagnostic Amerindian materials, and the absence of ferrous construction materials (e.g., nails) appear to indicate the presence of a Kalinago household. The occurrence, especially of indigenous Cayo-style pottery, which normally is attributed to the fourteenth to sixteenth centuries, in

Conclusion: Resilience and Capacity Building in the Age of Empires · 171

association with eighteenth-century European material goods counters the narrative of indigenous extinction. Continuing interactions between Kalinagos and enslaved Africans facilitated the emergence of postemancipation communities through the transfer of local knowledge assembled over centuries of island life. Exploiting the visual in contemporary artworks leads me to the emotional content of modern Caribbean literature. In the novel Salt, Earl Lovelace (1966:7) captures the trauma of emancipation: Four hundred years it take them to find out that you can’t keep people in captivity. Four hundred years! And it didn’t happen just so. People had to revolt. People had to poison people. Port-of-Spain had to burn down. A hurricane had to hit the island. Haiti had to defeat Napoleon. People had to run away up to the mountains. People had to fight. And then they agree, yes. We can’t hold people in captivity here. But now they had another problem: it was not how to keep people in captivity. It was how to set people at liberty. The British solution to emancipation was to protect both labor and land in their service (Chapter 3). Few opportunities were offered to newly freed laborers beyond underpaid métayage or sharecropping. Crown land was strategically priced in order to remain out of reach of persons without capital. The newly capitalized economy sought continued dependence on the estate that initially was instituted as a period of “apprenticeship.” Moreover, despite a failing sugar-based economy, labor shortages were mitigated by recruitment from other Leeward Islands. As Samantha Ellens (Chapter 3) writes, “Dominican peasantry appeared unenthusiastic to continue work on the estates.” Furthermore, the en masse emigration of Montserratians following emancipation (Honychurch 1995) suggests that Dominicans fled the estate in equal proportions. It is hard to imagine a place where geological, ecological, economic, and social histories have so conspired to facilitate “squatting.” Although the British made every effort to deny access to land, the inaccessible interior provided safe haven for those who fled plantation life. So what happens when one crop replaces another? If you are dependent on the external markets of empires, then you are sucked into a repetitive cycle of boom and bust (Chapter 1). However, there are alternatives. The chapters in this volume revolve around life on the estates. Their significant contribution highlights the many ways that the enslaved communities at

172 · William F. Keegan

Morne Patate repurposed practices imposed by the planters to claim their freedom. They reproduced self-reliance in subsistence, family and social groups, and engagement in the local and broader economy prior to emancipation. They survived the sugar boom and the sugar bust, the lime boom and bust, and even the collapse of the British Empire.

BIBLIOGRAPHY

Abel, Veronique 1987 La céramique commune à Marseille au XVIIe siècle: L’exemple de dépotoirs domestiques du site de la Charité. Archéologie du Midi médiéval 5:153–165. Abel, Veronique, and Henri Amouric 1991 Les ateliers de l’Huveaune à l’époque moderne. Ve Colloque international de la céramique médiévale en Méditerranée occidentale:84–94. Amouric, Henri, and Lucy Vallauri 2007 Céramiques mediterannéenes et du midi français dans les colonies d’Amérique: Fin XVIIe–XVIIIe s. relecture et nouveaux apports. In French Colonial Pottery: An International Conference, edited by George Avery, pp. 199–257. Northwestern State University Press, Natchitoches, Louisiana. Anonymous 1894 American Druggist and Pharmaceutical Record. American Druggist Publishing Company. Arcangeli, Myriam 2015 Why Quality Mattered: A Case Study Using Ceramics from the French West Indian Colony of Guadeloupe. Historical Archaeology 49(4):87–109. DOI:10.1007/ BF03376982. Armstrong, Douglas 1990 The Old Village and the Great House: An Archaeological and Historical Examination of Drax Hall Plantation, St. Ann’s Bay, Jamaica. University of Illinois Press, Champaign. 2003 Creole Transformation from Slavery to Freedom: Historical Archaeology of the East End Community, St. John, Virgin Islands. University Press of Florida, Gainesville. Armstrong, Douglas V., and Kenneth G. Kelly. 2000 Settlement Patterns and the Origins of African Jamaican Society: Seville Plantation, St. Ann’s Bay, Jamaica. Ethnohistory 47(2):369–397. Armstrong, Douglas, and Mark W. Hauser 2004 An East Indian Laborer’s Household in 19th Century Jamaica: A Case for Spatial, Chronological, and Material Analysis in Determining Ethnicity. Historical Archaeology 38(2):9–21. Armstrong, Douglas V., and Matthew C. Reilly 2014 The Archaeology of Settler Farms and Early Plantation Life in Seventeenth-Century Barbados. Slavery & Abolition. 35(3):399–417. Atwood, Thomas 1791 The History of the Island of Dominica: Containing a Description of Its Situation Extent, Climate, Mountains, Rivers, Natural Productions, &c. &c. London: J. Johnson.

174 · Bibliography

Ayensu, Edward S. 1981 Medicinal Plants of the West Indies. Reference Publications, Inc., Algonac, Michigan. Baker, Patrick L. 1994 Centring the Periphery: Chaos, Order, and the Ethnohistory of Dominica. McGillQueen’s University Press, Montreal. Barbotin, Maurice 1974 Archéologie caraïbe et chroniqueur: Extrait des Bulletins de la Sociétés d’Histoire de La Guadeloupe. Imprimerie J. Owen, Nérac. Bates, Lynsey A. 2015 Surplus and Access: Provisioning and Market Participation by Enslaved Laborers on Jamaican Sugar Estates. PhD dissertation, University of Pennsylvania, Philadelphia. Baxter, Michael J. 1994 Exploratory Multivariate Analysis in Archaeology. Edinburgh University Press, Edinburgh. Baxter, Michael J., Christian C. Beardah, and Richard V. S. Wright 1997 Some Archaeological Applications of Kernel Density Estimates. Journal of Archaeological Science 24(4):347–354. Baxter, Michael J., and Hilary E. M. Cool 2010 Correspondence Analysis in R for Archaeologists: An Educational Account. Archeologia e Calcolatori 21:211–228. Beckwith, Martha 1929 Black Roadways: A Study of Jamaican Folk Life. University of North Carolina Press, Chapel Hill. Belgrove, William, and Henry Drax 1755 A Treatise upon Husbandry or Planting. D. Fowle in Ann-Street, Near the Town-Dock, Boston. Berard, Benoit 2007 The “South-Dominica” Archaeological Mission: The Soufrière Site. Paper presented during the XXIIth Conference of the International Association of Caribbean Archaeology, July 23–29, 2007, Kingston, Jamaica. Bergad, Laird W. 1990 Cuban Rural Society in the Nineteenth Century: The Social and Economic History of Monoculture in Matanzas. Princeton University Press, Princeton, New Jersey. Berleant-Schiller, Riva 1983 Grazing and Gardens in Barbuda. In The Keeping of Animals: Adaptation and Social Relations in Livestock Producing Communities, edited by Riva Berleant-Shiller and E. Shanklin, pp. 73–91. Allenheld, Osmun, Totowa, New Jersey. 1995 From Labour to Peasantry in Montserrat after the End of Slavery. In Small Islands, Large Questions: Society, Culture and Resistance in the Post-Emancipation Caribbean, edited by Karen Fog Olwig, pp. 53–72. Frank Cass, Portland. Biersack, A., and J. B. Greenberg 2006 Reimagining Political Ecology. Duke University Press, Durham, North Carolina. Bindman, David 2017 Representing Race in the Eighteenth-Century Caribbean: Brunias in Dominica and St Vincent. Eighteenth-Century Studies 51(1):1–21. Bloch, Lindsay 2016 An Elemental Approach to the Distribution of Lead-Glazed Coarse Earthen-

Bibliography · 175

ware in the Eighteenth-Century Chesapeake. American Antiquity 81(2):231–252. DOI:10.7183/0002-7316.81.2.231. 2019 Sourcing Domestic and Industrial Ceramics from Trents Plantation, Barbados Using LA-ICP-MS (Laser Ablation-Inductively Coupled Plasma-Mass Spectrometry). In Pre-Colonial and Post-Contact Archaeology in Barbados: Past, Present, and Future Research Directions, vol. 5 of the Taboui series, edited by Kevin Farmer, Douglas Armstrong, Maaike de Waal, Niall Finneran, and Matthew C Reilly. Sidestone Press, Leiden. Boomert, Arie 1986 The Cayo Complex of St. Vincent: Ethnohistorical and Archaeological Aspects of the Island Carib Problem. Antropológica 66:3–68. 2011a Searching for Cayo in Dominica. In The Proceedings of the XXIII Congress of the International Association for Caribbean Archaeology, June 29–July 3, 2009, Antigua, edited by Samantha A. Rebovich, pp. 655–677. Dockyard Museum, English Harbour, Antigua. 2011b From Cayo to Kalinago: Aspect of Island Carib Archaeology. In Communities in Contact: Essays in Archaeology, Ethnohistory and Ethnography of the Amerindian CircumCaribbean, pp. 291–306. Sidestone Press, Leiden. Booth, Walter G. 1965 The Design and Application of Aerial Ropeways. Master’s thesis. McMaster University, Hamilton, Ontario, Canada. Borde, Sieur de la. 1674 Relations des moers, coustumes, religion, gueeres et voyages des Caraïbes suavages des Isles Antilles de l’Amerique. Paris: Louis Billaine. Breton, Raymond 1666 Dictionaire francois-caraibe. Auxerre: Par Gilles Bouquet impr. ordinaire du Roy. 1892 Dictionnaire caraïbe-françois. Teubman, Leipzig. 1958 Father Raymond Breton’s Observations of the Island Carib: A Compilation of Ethnographic Notes Taken from Breton’s Carib-French Dictionary Published in 1665. Translated by Marshall McKusick, and Pierre Vérin. Human Relations Area Files, New Haven, Connecticut. Brierley, John S. 1985 West Indian Kitchen Gardens: A Historical Perspective with Current Insights from Grenada. Food and Nutrition Bulletin 7(3):1–10. Bright, Alistair J. 2011 Blood Is Thicker Than Water: Amerindian Intra- and Inter-Insular Relationships and Social Organization in the Pre-Colonial Windward Islands. Sidestone. Leiden. Brown, Vincent 2009 Social Death and Political Life in the Study of Slavery. American Historical Review 114(5):1231–1249. Campbell, Susan 2007 Africans to Dominica: 100,000 Middle Passages from “Guinea” to the Eastern Caribbean, 1764–1808. Dominica Academy of Arts and Sciences 25. https://web.archive.org/ web/20130418033700/http://da-academy.org/AfricanstoDominica.pdf. Carlson, David L. 2017 Quantitative Methods in Archaeology Using R. Cambridge University Press, Cambridge, U.K.

176 · Bibliography

Carney, Judith A. 2016 Fields of Survival, Foods of Memory. In Geographies of Race and Food: Fields, Bodies, Markets, edited by Arun Saldanha and Rachel Slocum, pp. 61–78. Ashgate, Farnham, Surrey, U.K. Carney, Judith, and Richard Nicholas Rosomoff 2011 In the Shadow of Slavery: Africa’s Botanical Legacy in the Atlantic World. University of California Press, Berkeley. Carr, M. J., M. D. Feigenson, L. C. Patino, and J. A. Walker 2003 Volcanism and Geochemistry in Central America: Progress and Problems. In Geophysical Monograph Series, edited by John Eiler, pp. 153–174. American Geophysical Union, Washington, D.C. Carrier, James G., and Paige West 2009 Introduction: Surroundings, Selves and Others: The Political Economy of Environment and Identity. Landscape Research 34(2):157–170. Chace, Russell 1989 Protest in Post-Emancipation Dominica: The “Guerre Negre” of 1844. Journal of Caribbean History 23(2):118–141. Clark, James 1797 A treatise on the yellow fever, as it appeared in the island of Dominica, in the years 1793– 4–5–6: to which are added, Observations on the bilious remittent fever, on intermittents, dysentery, and some other west India diseases; also, The Chemical Analysis and Medical Properties of the hot mineral waters in the same Island. Printed for J. Murray and S. Highley, No 32, Fleet Street, London. Cracknell, Basil Edward 1973 Dominica. Stackpole Books, Harrisburg, Pennsylvania. Crosby, Alfred W. 1986 Ecological Imperialism: The Biological Expansion of Europe, 900–1900. Studies in environment and history. Cambridge University Press, Cambridge [Cambridgeshire]; New York. Crumley, Carole L. 1979 Three Locational Models: An Epistemological Assessment for Anthropology and Archaeology. Advances in Archaeological Method and Theory 2:141–173. 1994 Historical Ecology: A Multidimensional Ecological Orientation. Historical Ecology: Cultural Knowledge and Changing Landscapes:1–16. Curtin, Philip 1969 The Atlantic Slave Trade: A Census. University of Wisconsin Press, Madison. Daily News 1910 Montserrat. Daily News, December 21, 2. Accessed November 1, 2016. http://nla.gov. au/nla.news-article76829673. de La Borde, Sieur 1674 Relation des moeurs, coustumes, religion, guerres et voyages des Caraïbes sauvages des Isles Antilles de l’Amérique. Louis Billaine, Paris. Deagan, Kathleen 2001 Dynamics of Imperial Adjustment in Spanish America. Ideology and Social Integration. In Perspectives from Archaeology and History, edited by Susan E. Alcock, H. D. John, Terence N. D’Altroy, Kathleen D. Morrison, and Carla M. Sinopoli, pp. 179–194. Cambridge, Cambridge University Press.

Bibliography · 177

Debien, Gabriel 1974 Les esclaves aux Antilles francaises XVIIe-XVIIIe siècles. Société d’Histoire de la Guadeloupe, Basse Terre. Delawarde, Javier 1937 La vie paysanne à la Martinique: Essai de géographie humaine. Imprimerie Officielle, Fort-de-France, Martinique. de Leeuw, Jan 2007 Correspondence Analysis of Archeological Abundance Matrices. https://escholarship.org/uc/item/9vd6p5zk. Department of Statistics Papers, University of California Los Angeles. Delle, James A. 1999 The Landscapes of Class Negotiation on Coffee Plantations in the Blue Mountains of Jamaica: 1790–1850. Historical Archaeology 33(1):136–158. 2000 The Material and Cognitive Dimensions of Creolization in Nineteenth-Century Jamaica. Historical Archaeology 34(3):56–72. 2014 The Colonial Caribbean: Landscapes of Power in Jamaica’s Plantation System. Cambridge University Press, Cambridge. Digital Archaeological Archive of Comparative Slavery 2018 DAACS Cataloging Manual: Ceramics. Electronic document, last update October 10, 2018. https://492nzz341b7zv7n2p3rfrebt-wpengine.netdna-ssl.com/wp-content/ uploads/2018/10/DAACSCeramicManual.pdf, accessed November 13, 2018. Dominica Almanac 1860 Dominica Almanack. Augustus Theodore Righton (Dominican Office), Roseau. 1864 Dominica Almanack. Augustus Theodore Righton (Dominican Office), Roseau. Dominica Colonial Reports 1903 Roads and Land Settlement. Report on the Expenditure of the Parliamentary Grant in Aid. Vol. 23. London: Darling & Son, Ltd. Duff, Andrew I. 1996 Ceramic Micro-Seriation: Types or Attributes? American Antiquity 61(1):89–101. Dunn, Richard S. 1972 Sugar and Slaves: The Rise of the Planter Class in the English West Indies, 1624–1713. University of North Carolina Press, Chapel Hill. Dunnell, Robert C. 1970 Seriation Method and Its Evaluation. American Antiquity. 35(3):305–319. Dutertre, Jean Baptiste 1667 Histoire générale des Antilles habite’es par les françois. Thomas Iolly, Paris. Duval née Mezin, Christiane 1975 La condition juridique des hommes de couleur libres à la Martinique au temps de l’esclavage. History, Université de droit d’économie et de sciences sociales de Paris, Paris II. Edelson, S. Max 2017 The New Map of Empire: How Britain Imagined America before Independence. Harvard University Press, Cambridge, Massachusetts. Élisabeth, Léo 2003 La société martiniquaise aux XVIIe et XVIIIe siècles: 1664–1789. Éditions Karthala, Paris. Ellis, John 1774 An Historical Account of Coffee: With an Engraving and Botanical Description of the

178 · Bibliography

Thee: To Which Are Added Sundry Papers Relative to Its Culture and Use, as an Article of Diet and of Commerce. Dilly, London. Eltis, David, and Stanley L. Engerman 1993 Fluctuations in Sex and Age Ratios in the Transatlantic Slave Trade, 1663–1864. Economic History Review 46(2):308–323. Emerson, M. C. 1994 Decorated Clay Tobacco Pipes from the Chesapeake: an African Connection. In Historical Archaeology of the Chesapeake, edited by Paul Shackel and B. J. Little, pp. 35–49. Washington, D.C.: Smithsonian Institute Press. England, Suzannah 1994 Acculturation in the Creole Context: A Case Study of La Poterie, Martinique. PhD dissertation, Cambridge University, Cambridge. Ensor, Bradley E. 2013 The Archaeology of Kinship: Advancing Interpretation and Contributions to Theory. University of Arizona Press, Tempe. Evans, Clifford, and Betty J. Meggers 1960 Archeological Investigations in British Guiana. Vol. 177. Bureau of American Ethnology Bulletin. U.S. Government Printing Office, Washington, D.C. Fairbanks, Charles H. 1974 The Kingsley Slave Cabins in Duval County, Florida, 1968. Conference on Historic Site Archaeology Papers 7:62–93. Farnsworth, Paul 2001 “Negroe Houses Built of Stone Besides Others Watl’d + Plaistered”: The Creation of a Bahamian Tradition. In Island Lives: Historical Archaeology of the Caribbean, edited by Paul Farnsworth, pp. 234–271. University of Alabama Press, Tuscaloosa. Fergus, Howard A. 1982 Montserrat’s Days of Lime and Cotton. Caribbean Quarterly 28(3):10–18. 1994 Montserrat: History of a Caribbean Colony. MacMillan Caribbean, London. Ferguson, Leland 1980 Looking for the “Afro” in Colono-Indian Pottery. In Archaeological Perspectives on Ethnicity in America, edited by Robert L. Schuyler, pp. 14–28. Baywood, Farmingdale, New York. Fisher, Christopher T., and Tina L. Thurston 1999 Dynamic Landscapes and Socio-Political Process: The Topography of Anthropogenic Environments in Global Perspective. Antiquity 73(281):630–631. Fox, Georgia L. 2016 Transformations, Economics, and Bitter Outcomes: Archaeological Investigations at Betty’s Hope Plantation—a Case Study Reflecting 300 Years of Caribbean Sugar Production. Economic Anthropology 3(2):228–239. Fraginals, Manuel Moreno, and Stanley L. Engerman 1983 Plantations in the Mid-Nineteenth Century: Some Comparative Perspectives. American Historical Review 88(5):1201–1218. Froude, James Anthony 1888 The English in the West Indies: Or, The Bow of Ulysses. Longmans, Green, London. Galle, Jillian E. 2006 Strategic Consumption: Archaeological Evidence for Costly Signaling among Enslaved Men and Women in the Eighteenth-Century Chesapeake. PhD dissertation, Department of Anthropology, University of Virginia, Charlottesville.

Bibliography · 179

2010

Costly Signaling and Gendered Social Strategies among Slaves in the 18th-Century Chesapeake. American Antiquity, 75(1):19–43. 2011 Assessing the Impacts of Time, Agricultural Cycles and Demography on the Consumer Activities of Enslaved Men and Women in Eighteenth-Century Jamaica and Virginia. In Out of Many, One People: The Historical Archaeology of Colonial Jamaica, edited by James Delle, Mark Hauser, and Douglas Armstrong, pp. 211–242. University of Alabama Press, Tuscaloosa. 2017 The Abundance Index: Measuring Variation in Consumer Behavior in the Early Modern Atlantic World. In Material Worlds: Archaeology, Consumption, and the Road to Modernity, edited by Barbara J. Heath, Eleanor E. Breen, and Lori A. Lee, pp. 162–191. Routledge, London. Galle, Jillian, Elizabeth Bollwerk, and Fraser Neiman 2019 The Digital Archaeological Archive of Comparative Slavery: A Case Study in Open Data and Collaboration in the Field of Archaeology. In New Life for Archaeological Collections, edited by Rebecca Allen and Benjamin Ford. University of Nebraska Press, Lincoln. Gautier, Arlette 2000 Les familles esclaves aux Antilles francaise, 1635–1848. Population 6:975–1001. Gibson, Heather R. 2007 Daily Practice and Domestic Economies in Guadeloupe: An Archaeological and Historical Study. PhD dissertation, Syracuse University, Syracuse, New York. 2009 Domestic Economy and Daily Practice in Guadeloupe: Historical Archaeology at La Mahaudière Plantation. International Journal of Historical Archaeology 13(1):27–44. Girault, Christian 1985 Recherches de géographie économique aux Antilles: le cas haïtien. Espace géographique 14(4):285–288. Glassie, Henry 2000 Vernacular architecture. Indiana University Press, Indianapolis. Goggin, John 1960 Spanish Olive Jar: An Introductory Study. Publications in Anthropology. 1. Yale University, New Haven. Goucher, Candice 2013 Congotay! Congotay! A Global History of Caribbean Food. M. E. Sharpe, Armonk, New York. Gratuze, Bernard 1999 Obsidian Characterization by LA-ICP-MS and Its Application to Prehistoric Trade in the Mediterranean and the Near East: Sources and Distribution of Obsidian within the Aegean and Anatolia. Journal of Archaeological Science 26:869–881. Great Britain 1846 Regulations or Instructions under Which Crown Lands in the West India Colonies of British Guiana Are Permitted to Be Put Up for Sale. Parliamentary Papers 30(514). Green, Cecilia 1999 A Recalcitrant Plantation Colony: Dominica, 1880–1946. NWIG: New West Indian Guide/Nieuwe West-Indische Gids 73(3/4):43–71. Green, William A. 1991 British Slave Emancipation: The Sugar Colonies and the Great Experiment, 1830–1865. Oxford University Press, New York.

180 · Bibliography

Greenacre, Michael 2017 Correspondence Analysis in Practice. Chapman and Hall/CRC, Boca Raton, Florida. Grove, Richard H. 1995 Green Imperialism: Colonial Expansion, Tropical Edens, and the Origins of Environmentalism, 1600–1860. Cambridge University Press, New York. 1997 Ecology, Climate and Empire: Colonial and Global Environmental History, 1400–1940. White Horse Press, London. Gurney, Joseph John 1840 Familiar Letters to Henry Clay of Kentucky: Describing a Winter in the West Indies. Press of M. Day, New York. Handler, Jerome S. 1963 Pottery Making in Rural Barbados. Southwestern Journal of Anthropology 19(3):314– 334. Handler, Jerome S., and Stephanie Bergman 2009 Vernacular Houses and Domestic Material Culture on Barbadian Sugar Plantations, 1640–1838. Journal of Caribbean History 43(1):1–36. Handler, Jerome, and Fredrick Lange 1978 Plantation Slavery in Barbados: An Archaeological and Historical Investigation. Harvard University Press, Cambridge, Massachusetts. Handler, Jerome, and Diane Wallman 2014 Production Activities in the Household Economies of Plantation Slaves: Barbados and Martinique, Mid-1600s to Mid-1800s. International Journal of Historical Archaeology 18:441–466. Haraway, Donna 2015 Anthropocene, Capitalocene, Plantationocene, Chthulucene: Making Kin. Environmental Humanities 6(1):159–165. Harris, Khadene 2016 Postemancipation Shifts: Land Labor and Freedom in Dominica after 1838. In Archaeologies of Slavery and Freedom in the Caribbean: Exploring the Spaces in Between, edited by Lynsey. A. Bates, John M. Chenoweth and James Delle, pp. 264–278. University Press of Florida, Gainesville. Harrison, Lucia Carolyn J. 1935 Dominica: A Wet Tropical Human Habitat. Economic Geography 11(1):62–76. Hastorf, Christine A. 2003 Andean Luxury Foods: Special Food for the Ancestors, Deities and the Élite. Antiquity 77(297):545–554. Hauser, Mark W. 2008 The Archaeology of Black Markets: Local Ceramics and Economies in Eighteenth-Century Jamaica. University Press of Florida, Gainesville. 2011 Routes and Roots of Empire: Pots, Power, and Slavery in the 18th-Century British Caribbean. American Anthropologist 113(3):431–447. DOI:10.1111/j.15481433.2011.01352.x. 2014a Blind Spots in Empire: Plantation Landscapes in Early Colonial Dominica (1763– 1807). In The Archaeology of Slavery: A Comparative Approach to Captivity and Coercion, edited by Lydia Wilson Marshall, pp, 143–165. Southern Illinois University Press, Carbondale. 2014b Land, Labor, and Things: Surplus in a New West Indian Colony (1763–1807). Economic Anthropology 1(1):49–65.

Bibliography · 181

2015

The Infrastructure of Nature’s Island: Settlements, Networks and Economy of Two Plantations in Colonial Dominica. International Journal of Historical Archaeology 19(3):601–622. 2017 A Political Ecology of Water and Enslavement: Water Ways in Eighteenth-Century Caribbean Plantations. Current Anthropology 58(2):227–256. 2020 Mapping Water on the Nature Island: Enslavement and Environment in the Caribbean. University of Washington Press, Seattle. Hauser, Mark W., and Douglas Armstrong 2012 The Archaeology of Not Being Governed: A Counterpoint to a History of Settlement of Two Colonies in the Eastern Caribbean. Journal of Social Archaeology 12(3):310– 333. Hauser, Mark W., and Jerome Handler 2009 Change in Small Scale Pottery Manufacture in Antigua, West Indies. African Diaspora Archaeology Newsletter 12(4). Hauser, Mark W., and Dan Hicks 2007 Colonialism and Landscape: Power, Materiality and Scales of Analysis in Caribbean Historical Archaeology. In Envisioning Landscape: Situations and Standpoints in Archaeology and Heritage, edited by Dan Hicks, Laura McAtackney, Graham Fairclough, and Graham J. Fairclough, pp. 251–274. Left Coast Press, Walnut Creek, California. Hauser, Mark W., and Kenneth G. Kelly 2011 Colonies without Frontiers: Inter-Island Trade in the Eighteenth and Nineteenth Century Caribbean. In Islands at the Crossroads: Migration, Seafaring, and Interaction in the Caribbean, edited by L. A. Curet and M. W. Hauser, 41–56. University of Alabama Press, Tuscaloosa. Hauser, Mark William, Anna S. Agbe-Davies, Mary C. Beaudry, Charles R. Cobb, Karen B. Graubart, Krysta Ryzewski, Alice V. M. Samson, and Barbara L. Voss 2017 A Political Ecology of Water and Enslavement: Water Ways in Eighteenth-Century Caribbean Plantations. Current Anthropology 58(2):227–256. Heath, Barbara J., and Amber Bennett 2000 “The little spots allow’d them”: The Archaeological Study of African-American Yards. Historical Archaeology 34(2):38–55. Herskovits, Melville J. 1990 The Myth of the Negro Past. Beacon Press, Boston. Heuman, Gad 1995 Post-Emancipation Resistance in the Caribbean: An Overview. In Small Islands, Large Questions: Society, Culture, and Resistance in the Post-Emancipation Caribbean, edited by Karen Fog Olwig, pp. 123–134. Frank Cass, Portland, Oregon. Heuman, Gad, and David Vincent Trotman 2005 Contesting Freedom: Control and Resistance in the Post-Emancipation Caribbean. MacMillan, London. Higman, B. W. 1995 Slave Population and Economy in Jamaica, 1807–1834. University of West Indies Press, Kingston, Jamaica. 2000 The Sugar Revolution. The Economic History Review 53(2):213–236. 2005 Plantation Jamaica 1750–1850: Capital and Control in a Colonial Economy. University of the West Indies Press, Kingston, Jamaica.

182 · Bibliography

Jamaican Food: History, Biology, Culture. University of West Indies Press, Kingston, Jamaica. Hill, Mark O. 1973 Reciprocal Averaging: An Eigenvector Method of Ordination. Journal of Ecology 6(1):237–249. Hodge, Walter Henricks 1943 The Vegetation of Dominica. Geographical Review 33(3):349–375. Hodge, Walter Henricks, and Douglas Taylor 1957 The Ethnobotany of the Island Caribs of Dominica, Webbia 12:513–644. Hofman, C. L. 2012 Caribbean Encounters: Rescue Excavations at the Early Colonial Island Carib Site of Argyle, St. Vincent. Analecta Praehistorica Leidensia 43/44:63–76. Hofman, Corinne L., and Alistair J. Bright 2004 From Suazoid to Folk Pottery: Pottery Manufacturing Traditions in a Changing Social and Cultural Environment on St. Lucia. New West Indian Guide / Nieuwe WestIndische Gids 78(1–2):73–104. Hofmann, Corinne, and Menno Hoogland 2015 Beautiful Tropical Islands in the Caribbean Sea: Human Responses to Floods and Droughts and the Indigenous Archaeological Heritage of the Caribbean. In Water & Heritage Material, Conceptual and Spiritual Connections, edited by W. Willems and H.P.J. van Schaik, Sidestone, The Hague. Hofman, Corinne L., Menno Hoogland, Arie Boomert, and John Angus Martin 2019 Colonial Encounters in the Southern Lesser Antilles: Indigenous Resistance, Material Transformations, and Diversity in an Ever-Globalizing World. In Material Encounters and Indigenous Transformations in the Early Colonial Americas: Archaeological Case Studies, edited by Corinne L. Hofman and Floris Keehnen, pp. 359–384. Brill, Leiden. Honychurch, Lennox 1995 The Dominica Story: A History of the Island. 3rd ed. MacMillan Education, London. 2001 Slave Valleys, Peasant Ridges: Topography, Colour and Land Settlement on Dominica. Paper presented at the Roseau Country Conference, Dominica. January 2001. 2003 Inter-Island Migration and Cultural Change: The Impact of Montserratians on Dominica. Paper presented at the Montserrat Country Conference, November 13–14, 2002. 2004 Chatoyer’s Artist: Agostino Brunias and the Depiction of St Vincent. Journal of the Barbados Museum and Historical Society 50(1):104–128. Jamieson, Ross W., and Meridith Beck Sayre 2010 Barley and Identity in the Spanish Colonial Audiencia of Quito: Archaeobotany of the 18th Century San Blas Neighborhood in Riobamba. Journal of Anthropological Archaeology 29(2):208–218. Johnson, Matthew H. 2003 Housing Culture: Traditional Architecture in an English Landscape. Routledge, New York. Keegan, W. F. 2007 Taíno Indian Myth and Practice: The Arrival of the Stranger King. University Press of Florida, Gainesville. 2018 Caribbean Kinship as Instituted Process. The Journal of Island and Coastal Archaeology 14:1–19. Keegan, William F., and Corinne L. Hofman 2016 The Caribbean before Columbus. Oxford University Press, Oxford. 2008

Bibliography · 183

Kelly, Kenneth G. 2008 Creole Cultures of the Caribbean: Historical Archaeology in the French West Indies. International Journal of Historical Archaeology 12(4):388–402. 2009 Where Is the Caribbean? French Colonial Archaeology in the English Lake. International Journal of Historical Archaeology 13:80–93. 2011 “La Vie Quotidienne”: Historical Archaeological Approaches to Plantation Slavery in Guadeloupe, French West Indies. In French Colonial Archaeology in the Southeast and Caribbean, edited by Kenneth G. Kelly and Meredith Hardy, pp. 189–205. University Press of Florida, Gainesville. 2019 Building a Better Village? Transformations in French West Indian Slave Village Architecture from the Ancien Régime to Emancipation. In Archaeology of Domestic Landscapes of the Enslaved in the Caribbean, edited by James Delle and Elizabeth Clay, pp. 66–86. University Press of Florida, Gainesville. Kelly, Kenneth G., Mark W. Hauser, Christophe Descantes, and Michael D. Glascock 2008 Compositional Analysis of French Colonial Ceramics: Implications for Understanding Trade and Exchange. Journal of Caribbean Archaeology Special Publication 2:85– 108. Kelly, Kenneth G., and Diane Wallman 2014 Foodways of Enslaved Laborers on French West Indian Plantations (18th–19th Century). Afriques: Débats, méthodes et terrains d’histoire 5. DOI:10.4000/afriques.1608. Kincaid, Jamaica 1988 A Small Place. Farrar, Straus & Giroux, New York. Kirch, Patrick Vinton, and Terry L Hunt 1997 Historical Ecology in the Pacific Islands: Prehistoric Environmental and Landscape Change. Yale University Press, New Haven, Connecticut. Kosso, Peter 2001 Knowing the Past: Philosophical Issues of History and Archaeology. Humanity Books, Toronto. Kriz, K. D. 2008 Slavery, Sugar, and the Culture of Refinement: Picturing the British West Indies, 1700– 1840. Yale University Press, New Haven, Connecticut. Kruskal, Joseph B. 1971 Multidimensional Scaling in Archaeology: Time Is Not the Only Dimension. In Mathematics in the Archaeological and Historical Sciences, edited by F. R. Hodson, D. G. Kendall, and P. Tautu, pp. 119–132. Columbia University Press, New York. Labat, Jean-Baptiste 1722 Nouveau voyage aux isles de l’Amérique: Contenant l’histoire naturelle de ce pays, l’origine, les moeurs, la religion & le gouvernement des habitans anciens & modernes, les guerres & les événemens singuliers qui y sont arrivez pendant le séjour que l’auteur y a fait. Vol. 5. Chez P. Husson, Paris. Lenik, Stephan 2010 Frontier Landscapes, Missions, and Power: A French Jesuit Plantation and Church at Grand Bay, Dominica (1747–1763). PhD dissertation, Anthropology, Syracuse University, Syracuse, New York. 2011 A Jesuit Plantation and Church in the Caribbean Frontier: Grand Bay, Dominica (1748–1763). In Proceedings of the XXIII Congress of the International Association for Caribbean Archaeology, June 29–July 3, 2009, Antigua, edited by Samantha A. Rebovich, pp. 147–159. Dockyard Museum, English Harbor, Antigua.

184 · Bibliography

2012

Mission Plantations, Space, and Social Control: Jesuits as Planters in French Caribbean Colonies and Frontiers. Journal of Social Archaeology 12(1):51–71. 2014 Plantation Labourer Rebellions, Material Culture and Events: Historical Archaeology at Geneva Estate, Grand Bay, Commonwealth of Dominica. Slavery & Abolition 35:508–526. Li, Tania Murray 2005 Beyond “the State” and Failed Schemes. American Anthropologist 107(3):383–394. Lightfoot, Natasha 2015 Troubling Freedom: Antigua and the Aftermath of British Emancipation. Duke University Press, Durham, North Carolina. Ligon, Richard 1657 A True & Exact History of the Island of Barbados. Printed for Humphrey Moseley, London. Lindsay, J. M., R. B. Trumbull, and W. Siebel 2005 Geochemistry and Petrogenesis of Late Pleistocene to Recent Volcanism in Southern Dominica, Lesser Antilles. Journal of Volcanology and Geothermal Research 148(3):253–294. DOI:10.1016/j.jvolgeores.2005.04.018. Lockhart, Alexander Rumsey 1872 The Dominica Almanac. Official Gazette Office, Roseau. Losier, Catherine 2012 Approvisionner Cayenne au cours de l’Ancien Régime: Étude Archéologique et historique de l’économie et du Réseau Commercial d’une Colonie Marginale, La Guyane (XVII et XVIII Siècles). PhD dissertation, l’Université Laval, Quebec. Lovelace, Earl 1996 Salt: A Novel. Persea, New York. Lyman, R. Lee, Michael J. O’Brien, and Robert C. Dunnell 1997 The Rise and Fall of Culture History. Plenum, New York. Marquardt, William H., and Carole L Crumley 1987 Theoretical Issues in the Analysis of Spatial Patterning. In Regional Dynamics: Burgundian Landscapes in Historical Perspective, edited by William H. Marquardt and Carole L. Crumley, pp. 1–18. Elsevier Science, Amsterdam. Marshall, Woodville 1996 “We Be Wise to Many More Tings”: Blacks’ Hopes and Expectations of Emancipation. In Caribbean Freedom: Society and Economy from Emancipation to the Present, edited by Hilary Beckles and Verene Shepherd, pp. 12–20. University of Texas Press, Austin. Marwick, Ben 2017 Computational Reproducibility in Archaeological Research: Basic Principles and a Case Study of Their Implementation. Journal of Archaeological Method and Theory 24(2):424–450. Marwick, Ben, Jade d’Alpoim Guedes, C. Michael Barton, et al. 2017 Open Science in Archaeology. SAA Archaeological Record 17(4):8–14. Mayo, Deborah G. 2018 Statistical Inference as Severe Testing: How to Get beyond the Statistics Wars. Cambridge University Press, Cambridge. Meniketti, Marco G. 2006 Sugar Mills, Technology, and Environmental Change: A Case Study of Colonial AgroIndustrial Development in the Caribbean. IA. The Journal of the Society for Industrial Archeology 32(1):53–80.

Bibliography · 185

Sugar Cane Capitalism and Environmental Transformation: An Archaeology of Colonial Nevis, West Indies. University of Alabama Press, Tuscaloosa. Métreau, Laetitia 2016 Identifier la céramique au Québec. Cahiers d’Archéologie. CÉLAT, Montréal. Mintz, Sidney W. 1961 The Question of Caribbean Peasantries: A Comment. Caribbean Studies 1:31–34. 1985 From Plantations to Peasantries in the Caribbean. Wilson Center, Latin American Program, Washington, D.C. 1989 Caribbean Transformations. Columbia University Press, New York. 2010 Houses and Yards among Caribbean Peasantries. In Perspectives on the Caribbean: A Reader in Culture, History and Representation, edited by Philip W. Scher, pp. 9–24. John Wiley and Sons, Hoboken, New Jersey. Mintz, Sidney W., and Douglas Hall 1960 The Origins of the Jamaican Internal Marketing System. Yale Publications in Anthropology. Yale University Press, New Haven, Connecticut. Mock, Kevin 2006 An Analysis of the Morphological Variability between French Ceramics from Seventeenth-century Archaeological Sites in New France. MA thesis, Department of History, University of Maine, Orono. Monette, Yves 2010 Rapport d’analyse sur les céramiques vernissées vertes de France et de Saintonge découvertes sur le site de l’Îlot de palais à Québec. Rapport de recherche. Centre Eau Terre Environnement, Institut national de la recherche scientifique, Université du Québec, Québec City. Moore, Brian L. 1987 Race, Power, and Social Segmentation in Colonial Society: Guyana after Slavery, 1838– 1891. Gordon and Breach, New York. Morehart, Christopher T., and Christophe G. B. Helmke 2008 Situating Power and Locating Knowledge: A Paleoethnobotanical Perspective on Late Classic Maya Gender and Social Relations. Archeological Papers of the American Anthropological Association 18(1):60–75. Morrison, Kathleen D. 1994 The Intensification of Production: Archaeological Approaches. Journal of Archaeological Method and Theory. 1(2):111–159. 2007 Rethinking Intensification: Power Relations and Scales of Analysis in Precolonial South India. Seeking a Richer Harvest, edited by Tina L. Thurston and Christopher T. Fisher, pp. 235–247. Springer, Boston. Muchow, R. C. 1989 Comparative Productivity of Maize, Sorghum and Pearl Millet in a Semi-Arid Tropical Environment II: Effect of Water Deficits. Field Crops Research 20:207–219. Murdoch, D. H. 1984 Land Policy in the Eighteenth-Century British Empire: The Sale of Crown Lands in the Ceded Islands, 1763–1783. Historical Journal 27(3):549–574. Naftel, C. O. 1897 Report on the Forest Land and Estates of Dominica and on the Agricultural Capabilities of the Island. In Colonial Reports. Roseau: Guardian Office. 1898 Dominica: Report on the Agricultural Capabilities of Dominica, Colonial Reports: 2015

186 · Bibliography

Miscellaneous, Great Britain Colonial Office Command Papers, no. 9. H. M. Stationery Office, Great Britain. Neiman, Fraser D., and Nat W. Alcock 1995 Archaeological Seriation by Correspondence Analysis: An Application to Historical Documents. History and Computing 7(1):1–21. Nelson, Velvet 2010 Dominica’s Lime Industry: Agriculture and Identity in the Early Twentieth Century. National Identities 12(3):219–236. Newton, D. 2008 Trademarked: From Aertex to Wright’s Coal Tar. History Press, Stroud, U.K. Nicholls, Henry Alfred Alford, and John Henry Holland 1929 A Text-Book of Tropical Agriculture. Macmillan and Co., London. Nicolson, Dan Henry, Robert A. DeFilipps, and A. C. Nicolson 1991 Flora of Dominica, Part 2: Dicotyledoneae. Smithsonian Contributions to Botany, no. 77. Smithsonian Institution Press, Washington, D.C. Niddrie, David L. 1966 Eighteenth-Century Settlement in the British Caribbean. Transactions of the Institute of British Geographers, no. 40 (December): 67–80. Oas, Sarah E., and Mark W. Hauser 2018 The Political Ecology of Plantations from the Ground Up. Environmental Archaeology 23(1):4–12. Ober, Frederick Albion 1880 Camps in the Caribbees: The Adventures of a Naturalist in the Lesser Antilles: D. Douglas, Edinburgh. 1907 Our West Indian Neighbors: The Islands of the Caribbean Sea, “America’s Mediterranean”: Their Picturesque Features, Fascinating History, and Attractions for the Traveler, Nature-Lover, Settler and Pleasure-Seeker. J. Pott & Co., New York. 1913 A Guide to the West Indies, Bermuda and Panama. Dodd Mead, New York. O’Shaughnessy, Andrew Jackson 2000 An Empire Divided: The American Revolution and the British Caribbean. University of Pennsylvania Press, Philadelphia. Pagán-Jiménez, Jaime, Reniel Rodríguez Ramos, Basil A. Reid, Martijn van den Bel, and Corinne Hofman 2015 Early Dispersals of Maize and Other Food Plants into the Southern Caribbean and Northeastern South America. Quaternary Science Reviews 123:231–246. Palmié, Stephan 1995 Slave Cultures and the Cultures of Slavery. University of Tennessee Press, Knoxville. Parry, J. H. 1955 Plantation and Provision Ground: An Historical Sketch of the Introduction of Food Crops into Jamaica. Revista de Historia de América, 39:1–20. Parsard, Kaneesha Cherelle 2018 Barrack Yard Politics from CLR James’s “The Case for West-Indian Self Government to Minty Alley.” Small Axe: A Caribbean Journal of Criticism 22.3(57):13–27. Patterson, Orlando 2018 Slavery and Social Death: A Comparative Study, With a New Preface. Harvard University Press, Cambridge.

Bibliography · 187

Peeples, Matthew A., and Gregson Schachner 2012 Refining Correspondence Analysis-Based Ceramic Seriation of Regional Data Sets. Journal of Archaeological Science 39(8):2818–2827. Pérotin-Dumon, Anne 2000 La ville aux Iles, la ville dans l’île: Basse-Terre et Pointe-à-Pitre, Guadeloupe, 1650– 1820. Karthala, Paris. Petrucci, Jean Ferdinand 1999 Les poteries et les potiers de Vallauris 1501–1945. PhD dissertation, L’EHESS, Marseilles. Phillippi, Bradley D. 2016 From Coercion to Compensation: Labor Systems and Spatial Practice on a Plural Farmstead, Long Island. PhD dissertation, Department of Anthropology, Northwestern University, Evanston, Illinois. Pulsipher, Lydia 1993 Changing Roles in the Life Cycles of Women in Traditional West Indian Houseyards. In Women and Change in the Caribbean: A Pre-Caribbean Perspective, edited by Janet Momsen, pp. 50–64. James Currey, London; and Ian Randle, Kingston, Jamaica. 1994 The Landscapes and Ideational Roles of Caribbean Slave Gardens. In The Archaeology of Garden and Field, edited by Naomi F. Miller, Kathryn L. Gleason, pp. 202–222. University of Pennsylvania Press, Philadelphia. Pulsipher, Lydia M., and Conrad M. Goodwin 1999 Here Where the Old Time People Be: Reconstructing the Landscapes of the Slavery and Post-Slavery Era in Montserrat, West Indies. In African Sites Archaeology in the Caribbean, edited by Jay B. Haviser, pp. 9–37. Markus Weiner, Princeton, New Jersey; and Ian Randle, Kingston, Jamaica. 2001 “Getting the Essence of It”: Galways Plantation, Montserrat, West Indies. In Island Lives: Historical Archaeologies of the Caribbean, edited by Paul Farnsworth, pp. 165– 203. University of Alabama Press, Tuscaloosa. Ramenofsky, Ann F., Fraser D. Neiman, and Christopher D. Pierce 2009 Measuring Time, Population, and Residential Mobility from the Surface at San Marcos Pueblo, North Central New Mexico. American Antiquity 74(3):505–530. Quinlan, Marsha B., Robert J. Quinlan, Sarah K. Council, and Jennifer W. Roulette 2016 Children’s Acquisition of Ethnobotanical Knowledge in a Caribbean Horticultural Village. Journal of Ethnobiology 36(2):433–457. Reeves, Matthew 1997 “By Their Own Labor”: Enslaved African’s Survival Strategies on Two Jamaican Plantations. PhD dissertation, Department of Anthropology, Syracuse University, Syracuse, New York. 2011 Household Market Activities among Early Nineteenth-Century Jamaican Slaves: An Archaeological Case Study from Two Slave Settlements. In Out of Many, One People: The Historical Archaeology of Colonial Jamaica, edited by James A. Delle, Mark W. Hauser, and Douglas Armstrong, pp. 183–210. University of Alabama Press, Tuscaloosa. Reilly, Matthew C. 2016 “Poor White” Recollections and Artifact Reuse in Barbados: Considerations for Archaeologies of Poverty. International Journal of Historical Archaeology 20(2):318–340. Resident, A. 1828 Sketches and Recollections of the West Indies. London.

188 · Bibliography

Richardson, Bonham C. 1992 The Caribbean in the Wider World, 1492–1992: A Regional Geography. Cambridge University Press, Cambridge. Robin, Cynthia 2013 Everyday Life Matters: Maya Farmers at Chan. University Press of Florida, Gainesville:. Rochefort, César de 1658 Histoire naturelle et morale des iles Antilles de l’Amerique: Enrichie de plusieurs belles figures des raretez les plus considérables qui y sont d’écrites. A. Leers, Roterdam. 1666 The History of the Caribby Islands, viz., Barbados, S’Christophers, S’Vincents, Martinico, Dominico, Barbouthos, Monserrat, Mevis, Antego, &c. in all XXVIII. In two books. The first containing the natural; the second, the moral history of those islands. Illustrated with several pieces of sculpture, representing the most considerable rarities therein described. With a Caribbian-vocabulary, translated by John Davies. Printed by J.M. for T. Dring and J. Starkey, London. Rouse, Irving 1992 The Tainos: Rise and Decline of the People Who Greeted Columbus. Yale University Press, New Haven, Connecticut. Roux, Yannick Le 2013 Loyola, l’habitation des jésuites de Rémire en Guyane française. In Situ. Revue des patrimoines (20). Royal Botanic Gardens, Kew 1894a Cultural Industries in Dominica. Bulletin of Miscellaneous Information (Royal Botanic Gardens, Kew) 1894(96):405–410. doi:10.2307/4115496. 1894b West Indian Lime. (Citrus Medica, L., var. acida, Brandis.). Bulletin of Miscellaneous Information (Royal Botanic Gardens, Kew) 1894(88):113–116. doi:10.2307/4113168. 1898 Bulletin of Miscellaneous Information: Additional series: H. M. Stationery Office. Rubin, Vera (editor) 1959 Plantation Systems of the New World: Papers and Discussion Summaries of the Seminar Held in San Juan, Puerto Rico. Research Institute for the Study of Man, Pan American Union, Washington, D.C. Samford, Patricia 1996 The Archaeology of African-American Slavery and Material Culture. William and Mary Quarterly 53(1):87–114. Scheid, Dwayne 2015 The Political Economy of Ceramic Production in Barbados: From Plantation Industry to Craft Production. PhD dissertation, Department of Anthropology, Syracuse University, Syracuse, New York. Senhouse, Joseph 1985. The Diary of Joseph Senhouse [1777–1778]. Journal of the Barbados Museum and Historical Society 37(3):276–295. Shearn, Isaac 2014 Pre-Columbian Regional Community Integration in Dominica, West Indies. PhD dissertation, Department of Archaeology, University of Florida, Gainesville. 2018 Pre-Columbian Settlement Trajectories in Eastern Dominica: Report on Initial Radiocarbon Age Estimates. Journal of Island and Coastal Archaeology 13(1):132–146. Shennan, Stephen 1997 Quantifying Archaeology. University of Iowa Press, Iowa City.

Bibliography · 189

Sheridan, Richard B 1974 Sugar and Slavery: An Economic History of the British West Indies, 1623–1775. University of West Indies Press, Jamaica. Singleton, T. A. 1992 Preface. In African-American Gardens and Yards in the Rural South, by Richard Westmacott. University of Tennessee Press, Knoxville. 1996 The Archaeology of Slave Life. In Images of the Recent Past: Readings in Historical Archaeology, edited by Charles Orser, pp. 141–165. AltaMira, Walnut Creek, California. Slocum, Karla 2017 Caribbean Free Villages: Toward an Anthropology of Blackness, Place, and Freedom. American Ethnologist 44(3):425–434. Smith, Karen Y., and Fraser D. Neiman 2007 Frequency Seriation, Correspondence Analysis, and Woodland Period Ceramic Assemblage Variation in the Deep South. Southeastern Archaeology 26(1):47–72. Smith, Michael L. 2005 Networks, Territories, and the Cartography of Ancient States. Annals of the Association of American Geographers 95(4):832–849. South, Stanley 1971 Evolution and Horizon as Revealed in Ceramic Analysis in Historical Archeology. Research Manuscript Series Book 15. http://scholarcommons.sc.edu/archanth_ books/15. Stein, Gil J. 2002 From Passive Periphery to Active Agents: Emerging Perspectives in the Archaeology of Interregional Interaction. American Anthropologist 104(3):903–916. Sturge, Joseph, and Thomas Harvey 1838 The West Indies in 1837. Reprint, Dawsons of Pall Mall, London. Taylor, Douglas 1938 The Caribs of Dominica. U.S. Government Printing Office, Washington, D.C. ter Braak, Cajo J. F. 1985 Correspondence Analysis of Incidence and Abundance Data: Properties in Terms of a Unimodal Response Model. Biometrics 41(4):859–873. ter Braak, Cajo J. F., and I. Colin Prentice 1988 A Theory of Gradient Analysis. Advances in Ecological Research 18:271–317. Theophille, Brent, and Sharon D. Hutchinson 2012 Economic Analysis of the Targeting Strategy of Artisanal Fishers in Dominica. Marine Resource Economics 27(4):323–342. Thomas, Leon 1953 La Dominique et les derniers Caraibes insulaires. Les Cahiers d’Outre-Mer 6(21):37– 60. Tobin, Beth Fowkes 1999 Picturing Imperial Power: Colonial Subjects in Eighteenth-Century British Painting. Duke University Press, Durham, North Carolina. 2011 Colonizing Nature: The Tropics in British Arts and Letters, 1760–1820. University of Pennsylvania Press, Philadelphia. Trouillot, Michel-Rolph 1984 Labour and Emancipation in Dominica: Contribution to a Debate. Caribbean Quarterly 30:73–84.

190 · Bibliography

Peasants and Capital: Dominica in the World Economy. Johns Hopkins University Press, Baltimore. 1989 Discourses of Rule and the Acknowledgment of the Peasantry in Dominica, WI, 1838–1928. American Ethnologist 16(4):704–718. 1992 The Caribbean Region: an Open Frontier in Anthropological Theory. Annual Review of Anthropology 21(1):19–42 1993 Coffee Planters and Coffee Slaves in the Antilles: The Impact of a Secondary Crop. In Cultivation and Culture: Labor and the Shaping of Slave Life in the Americas, edited by Ira Berlin and Philip D. Morgan, pp. 124–137. University of Virginia Press, Charlottesville. Turnbaugh, William, and Sarah Peabody Turnbaugh 1977 Alternative Applications of the Mean Ceramic Date Concept for Interpreting Human Behavior. Historical Archaeology 11(1):90–104. Vaquero, V., and M.H.F. Saville 1914 Life and Adventure in the West Indies; A Sequel to Adventures in Search of a Living in Spanish-America. J. Bale, Sons & Danielson, Ltd., London. Vlach, John M. 1995 “Snug Li’l House with Flue and Oven”: Nineteenth-Century Reforms in Plantation Slave Housing. Perspectives in Vernacular Architecture 5:118–129. Voss, Barbara 2008 Poor People in Silk Shirts’ Dress and Ethnogenesis in Spanish-Colonial San Francisco. Journal of Social Archaeology 8(3):404–432. Wadge, G. 1985 Morne Patates Volcano, Southern Dominica, Lesser Antilles. Geological Magazine 122(3):253–260. Wallman, Diane 2014 Slave Community Food Ways on a French Colonial Plantation. Bitasion: Archéologie des habitations-plantations des petites Antilles-Lesser Antilles Plantation Archaeology:45–68. Wallman, Diane, E. Christian Wells, and Isabel C. Rivera-Collazo. 2018 The Environmental Legacies of Colonialism in the Northern Neotropics: Introduction to the Special Issue. Environmental Archaeology 23(1):1–3. Ward, John Robert 1988 British West Indian Slavery, 1750–1834: The Process of Amelioration. Clarendon Press, Oxford. Watts, David 1966 Man’s Influence on the Vegetation of Barbados, 1627–1800. Hull University Press, Hull, U.K. 1990 The West Indies: Patterns of Development, Culture and Environmental Change since 1492. Cambridge University Press, New York. First published 1987. Weigand, Phil C., Garman Harbottle, and Edward V. Sayre 1977 Turquoise Sources and Source Analysis: Mesoamerica and the Southwestern U.S.A. In Exchange Systems in Prehistory, edited by Timothy K. Earle and Jonathon E. Ericson, pp. 15–34. Academic Press, New York. Weiss Muller, Hannah 2017 Subjects and Sovereign: Bonds of Belonging in the Eighteenth-Century British Empire. Oxford University Press, New York. 1988

Bibliography · 191

Wells, E. Christian, Christopher K. Waters, Anthony R. Tricarico, and Georgia L. Fox 2017 Agroindustrial Soilscapes in the Caribbean: A Geochemical Perspective from Betty’s Hope, Antigua. Environmental Archaeology 22(4):381–393. West, Paige 2012 From Modern Production to Imagined Primitive: The Social World of Coffee from Papua New Guinea. Duke University Press, Durham, North Carolina. West India Royal Commission 1886 Her Majesty’s Colonies. A Series of Original Papers Issued under the Authority of the Royal Commission. William Clowes and Sons, London. Wilkie, Laurie A., and Paul Farnsworth 2001 Sampling Many Pots: An Archaeology of Memory and Tradition at a Bahamian Plantation. University Press of Florida, Gainesville. Willis, Aaron 2013 The Standing of New Subjects: Grenada and the Protestant Constitution after the Treaty of Paris (1763). Journal of Imperial and Commonwealth History 42(1):1–21. Wolf, Eric R. 1959 Specific Aspects of Plantation Systems in the New World: Community Sub-Cultures and Social Classes. In Plantation Systems of the New World, edited by V. Rubin, pp. 136–212. Research Institute for the Study of Man, New York. 1972 Ownership and Political Ecology. Anthropological Quarterly 45(3):201–205. Wolf, Eric R., and Sidney W. Mintz 1957 Haciendas and Plantations in Middle America and the Antilles. Social and Economic Studies 6(3):380–412. Wood, D. 1968 Trinidad in Transition. Oxford University Press, Oxford, U.K. Wynter, Sylvia 1971 Novel and History, Plot and Plantation. Savacou 5:95–102. Young, William 1764 Considerations Which May Tend to Promote the Settlement of Our New West India Colonies by Encouraging Individuals to Embark in the Undertaking. Printed for J. Robson, London.

CONTRIBUTORS

Lynsey A. Bates is contract archaeologist for the Digital Archaeological Archive of Comparative Slavery. She is coeditor of Archaeologies of Slavery and Freedom in the Caribbean: Exploring the Spaces in Between. Lindsay Bloch is collections manager of the Ceramic Technology Laboratory and Florida Archaeology at the Florida Museum of Natural History. Elizabeth Bollwerk is senior archaeological analyst for the Digital Archaeological Archive of Comparative Slavery. She is coeditor of Perspectives on the Archaeology of Pipes, Tobacco, and Other Smoking Plants in the Ancient Americas. Samantha Ellens is a doctoral candidate in anthropology at Wayne State University. Jillian E. Galle is project director of the Digital Archaeological Archive of Comparative Slavery at Monticello. She is the editor of Engendering African American Archaeology. Khadene K. Harris is assistant professor of anthropology at Kenyon College. Mark W. Hauser is associate professor of anthropology at Northwestern University. Lennox Honychurch is the most prominent expert on Dominican history and heritage. His most recent book is In the Forests of Freedom: The Fighting Maroons of Dominica.

194 · Contributors

William F. Keegan is curator of Caribbean archaeology at the Florida Museum of Natural History and professor of anthropology at the University of Florida, Gainesville. He is author (with Corinne Hofman) of The Caribbean before Columbus and editor (with Corinne Hofman and Reniel Rodríguez Ramos) of The Oxford Handbook of Caribbean Archaeology. Tessa Murphy is assistant professor of early American history at Syracuse University. Fraser D. Neiman is director of archaeology at Monticello and a lecturer in the departments of anthropology and architectural history at the University of Virginia. Sarah Oas is postdoctoral research associate in the School of Human Evolution and Social Change at Arizona State University. Diane Wallman is assistant professor of anthropology at the University of South Florida.

INDEX

Page numbers followed by the letters f and t indicate figures and tables. Abundance indexes (AI), 80–85 Agriculture: Bois Cotlette plantation, 37–38; boom-bust cycles, 49; and everyday life, 16, 23; history of, 6, 14–15, 23–25, 154; landscape considerations, 11–12; and landscapes, 11–12; neutral period, 33, 37–38, 154. See also specific crops Antigua, 38, 60, 100, 132 Apprenticeship period, 14, 25, 28, 100, 155–56, 171 Arcangeli, Myriam, 147 Archaeological Survey of Colonial Dominica, 17 Armstrong, Douglas, 90 Atwood, Thomas, 43, 155, 162 Barbados, 38–39, 132–33 Barley, 105, 163t, 164–65, 170 Bates, Lynsey, 65 Belligny, Nicholas Croquet de, 13–14, 26, 44 Bois Cotlette plantation: agriculture, 37–38; enslaved people, 6, 31, 39; house yards, 92; inventory of 1765, 31–32, 37, 40; and Martinique, 37; photographs of, 95–96 Breton, Raymond, 12, 124, 134 Bright, Alistair, 134 British colonial period, Dominica: autonomy during, 154; beginning of, 12–13, 33, 35; deficiency act, 43; difficulties with, 13–14; emancipation strategies, 171; enslaved people, 42–43; and enslavement, 9, 40; French planters, 39, 41, 97; goals for, 40–43; house yards, 99; lime industry, 50; manumission, 40, 47n2; Morne Patate growth during, 14, 99; neutral period continuities,

40; Soufriere, 32, 97; and sugar, 40, 44, 99, 154; transition to, 9, 39–40; white settlement encouragement, 43 British Empire: Caribbean colony concerns, 41; deficiency acts, 43; Encumbered Estates Act, 51, 63; land access restrictions, 58; Southern Caribbee Islands colony, 40–42, 44, 47n3; sugar expansion interests, 44 Brunias, Agostino, 94, 113 Brunias, Agostino, work of: overview of, 94, 113; A Cudgelling Match, 95f; A Family of Charaibes, 118f; houses in, 94–95, 114–17, 118f; interpretive value of, 119; A Leeward Islands Carib Family, 117f; natural history conventions, 113–14; A Negroes Dance, 116f; as sanitized, 170; social distinctions in, 128; social groups depicted, 114; View on the River Roseau, 114f; Villagers Merry-Making in the Island of St. Vincent, 115f Byres, John, 99 Cableways, 54–55 Carib people. See Kalinago people Cassava, 124, 134, 156, 166 Cayo pottery, 125, 133–34 Cereals, 162, 164–65 Clark, James, 123 Cocoa: and coffee blight, 15, 25; cultivation landscapes, 11; and migration, 60; peasant preference for, 61; price drops, 62; smallholders, 33; sugar transition, 40 Coffee: blights, 15, 23, 25, 165; cultivation of, 11, 26, 42; and enslavement, 26, 97; estates and plantations, 6, 33, 59, 154; exports of, 15, 33, 131; glacées, 80; House Area A findings, 105, 165; importance of, 6; labor regimes, 23; land use, 11, 40; local production, 165; and migration, 36; Morne Patate production, 8t,

196 · Index Coffee—continued 14, 26, 44–47, 97; Morne Patate remains, 78, 105, 163t; Soufriere production, 45; sugar transition, 14, 40, 131 Colonies as contested landscapes, 1–2 Correspondence analysis (CA): overview of, 69; assumptions, 69; and frequency seriation, 70; and MCD, 68–70; Morne Patate chronology development, 68–76; scatter plots, 70; and terminus post quem, 70–71 Crève Coeur, 6, 93, 101 Crosby, Alfred, 26 Crown lands, 57–59, 171 Crumley, Carole, 17 De la Ferriere Constance, Louis de, 31 Digital Archaeological Archive of Comparative Slavery (DAACS), 64–65, 72–73, 76 Discard rates, 81 Dominica: British-French contention over, 130–31; in Caribbean histories, 1; ceramic chronology, 133; ceramic forms, 134; cultural divergence, 9–10; Delices, 12; ecological diversity, 153–54; economic connections, 23–24, 35, 37; economic problems, 62; economies, informal, 23; economies, local, 32; enclave settlement pattern, 9–10, 49, 55, 156, 169; geology of, 11–12, 136; goods shortages, 131; history timeline, 5f; isolation of, 170; location, 7f; names for, 10; plantations on, 5–6, 25; ports, free, 131; settlement history, 12–13, 59; soil, 11; South American migration, 12; surveys of, 99; as trade hub, 131; villages, leeward coast, 59 —geography of: and cableways, 55; slopes, 11, 153, 169; and social relations, 49; volcanic, 10–11, 136, 153, 161, 169 Earthenware, coarse: archaeological interest in, 129; Caribbean versus French, 83, 84; Cayo, 133–34; chronology development use, 73, 76, 81, 83–85; conaree, 134; in Dominica, 133–34; European imports, 132, 135, 147–48; everyday life insights, 129; griddles, 24, 125, 134, 166; at House Area G, 120, 121t, 125; in Jamaica, 132; production, Caribbean, 132– 34; sources of, 130; trade in, 83, 130, 144–45, 147; uses of, 24, 132; variation in, 129–30

Earthenware, French: Biot, 135, 137, 150; versus Caribbean, 83, 84; Caribbean exports, 135; Dominican presence, 147–48; elemental data lack, 137–38, 142; Housing Area D, 105; Huveaune, 135, 137–38, 141f, 142, 143f, 149; Saintonge, 135, 137–38, 142, 150; source regions, 135, 137; Vallauris, 135, 137–38, 141f, 143, 148–49 Earthenware, Morne Patate types: overview of, 130; assemblage-level results, 145, 146t; Biot, 135, 137, 150; British, 143–47; Caribbean, 140–42, 144–45, 151–52; ceramic sourcing, 135–36; descriptions of, 148–52; French coarse, 142, 151; and geology, 136–37; group assignment, 140–44; Huveaune, 135, 137–38, 141f, 142, 143f, 149; LA-ICP-MS analyses, 130, 139; local trade evidence, 144–45, 147; methods overview, 130; previous study comparisons, 144–45; principle component analyses, 140; Saintonge, 135, 137–38, 142, 150; sample selection, 137–39; trace element presence, 140; Type 1, 140–42, 144–45, 146t, 151; Type 2, 140–42, 144–45, 146t, 151–52; Type 3, 142, 144t, 145, 146t, 152; Vallauris, 135, 137–38, 141f, 143, 148–49 Ecological imperialism, 26 Ecology, 2 Edelson, Max, 32 Eggplants, 162, 163t Eldridge, Charles Monroe, 57 Élisabeth, Léo, 36 Emancipation: Antigua, 100; apprenticeship period, 14, 25, 28, 100, 155–56, 171; British solution to, 171; in Dominica, 14–15, 49, 100, 168; and house yards, Morne Patate, 109; and lime industry, 48; Morne Patate and, 15, 24–24; Morne Patate chronology, 77–78, 80, 85; as peasant society shift, 100; and Soufriere, 14–15; trauma of, 171 —effects of: on house yards, 100–101; on labor, 100–101; labor control struggles, 56–57, 155; on landscapes, 55–59; land struggles, 56; migration, 59–60, 171; on productivity, 100; sugar industry, 15, 48, 57 Enclave settlement pattern, 9–10, 49, 55, 156, 169 Enslaved people: coping strategies, 166; Creoles, 32, 36; everyday life of, 2, 94–96, 113;

Index · 197 food, 154–55, 162, 166, 169–70; gardens, 49–50, 155; houses, depictions of, 94–95, 115; revolts, 126; social relationships, 109– 10; subsistence of, 154–56, 166, 169–70 —Morne Patate: eighteenth-century population, 45, 98; food preparation, 24; gardens kept by, 27; nineteenth-century population, 46–47; provision ground system, 24; tasks of, 44. See also House yards Enslavement: and British Empire transition, 39; built landscapes of, 119; creolized population, 38–39; end of, 15, 23; and plantation landscapes, 1–2; reliance on, 23; on smallholdings, 98; spatial separation enforcing, 98–99; trade, 31–32, 38–39. See also Emancipation Environmental data and local history, 153 Environmental practices, 2 Estate house: overview, 45, 106; construction of, 14, 45; excavation of, 18–19, 102; food evidence, 159, 165; goods recovered from, 108; location of, 18. See also Morne Patate chronology European houses, 114 Everyday life: overview of, 16; agriculture’s impact on, 16, 23; depictions of, 94, 114–15, 170; as environmental practice, 2; evidence of, 22, 24; and landscapes, 15; and plantation archaeology, 4, 16–17, 19, 22. See also House yards Fennel, 163t, 165 Fishing, 161–62 Food production and consumption: barley, 105, 163t, 164–65, 170; botanical remains, 163t; cassava, 124, 134, 156, 166; cereals, 162, 164–65; domestic animal reliance, 159; eggplants, 162, 163t; enslaved people’s, 154–55, 162, 166, 169–70; fennel, 163t, 165; fish, 161–62; fruits, 162–65; historical accounts of, 155–56; maize, 162–64; okra, 165; paleoethnobotanical analysis, 162–66; post-sugar introduction, 164; sample collection, 159; snakes, 161–62; sorghum, 105, 162–64; staples, 155–56, 157–58t; zooarchaeological analysis, 159–62 France, geology of, 137

Free Port Act, 23, 83, 131 French planters, 39, 41, 97, 131, 170 Frequency seriation, 66–67, 70, 71–72, 75 Froude, James, 51 Fruits, 162–65 Galleon, 18, 28, 59–60, 62 Gibson, Heather, 149 Grenada, 42, 125 Griddles, 24, 125, 134, 166 Guadeloupe, 92, 125, 131, 133, 136 Haciendas versus plantations, 3–4 Haraway, Donna, 2, 4, 27 Harris, Khadene, 15, 27 Harvey, Thomas, 155–56 Hauser, Mark, 26, 99, 169 Head taxes, 36 Higman, Barry, 44 Historical archaeology, 2, 70, 88, 127 Hofman, Corinne, 134 Honychurch, Lennox, 26, 94, 113 House Area A: versus B, 105; botanical remains, 105, 163t, 164; house superimpositions, 102; occupation of, 103; photographs of, 102f; reorientations, 103, 108–9 House Area B, 102–3, 105, 163t, 165 House Area C, 102–3, 105, 163t House Area D, 102, 105, 163t House Area E, 106, 163t House Area F, 103, 106 House Area G: Argyle structure similarity, 124–25; artifact types recovered, 121t, 125; botanical remains, 163t; carbet evidence, 124; chronology, 121t; coarse earthenware, 120, 121t, 125; cobble road, 106; excavation of, 106; identity signaling, 119, 126, 128; Kalinago occupation evidence, 111–12, 126–27, 170; occupation of, 120; versus other houses, 120, 123, 127; postholes identified, 121, 124; size of, 121, 124; stratified deposits, 120 House Area H, 102, 106 Houses: carbet, 115–17, 124; communicative aspects, 119, 128; culture’s influence on, 127–28; depictions of, 94–95, 114–17, 118f; Kalinago communal, 115–17; meanings of, 113, 119, 127; photographs of, 96f; research on, 119; versus yards, 91

198 · Index Houses, enslaved people’s: building materials, 92, 94, 96; construction materials, 122; construction of, 119; cooking hearths, 120; descriptions, primary source, 122–23; factors influencing, 119; fire use, 123; living arrangements, 122; sizes of, 119–20; wattled, 93 House yards: archaeological interest in, 88; archaeology of, 91–93; Bois Cotlette estate, 92; boundary markers, 123; change, sensitivity to, 90, 97; cooking hearths, 123; Crève Coeur, 93; and cultural practices, 89; definitions of, 91, 109; Dominican, 94–97; emancipation’s impact on, 100–101; in engravings, 94–95; and enslavement, 89–90; and everyday life, 88; on Guadeloupe, 92; and houses, 91; La Mahaudière, 92; and material culture, 91; in photographs, 95–96; research methods, 93; scholarship on, 88–90; and social relationships, 109; spatial aspects, 109; spatial layouts, 92–93; Sugar Loaf estate, 92; as survival metaphor, 90; symbolism of, 110 House yards, Morne Patate: arrangement of, 108; British colonial period, 99; cooking hearths, 22, 123; domestic features, 103t; and emancipation, 109; excavations, 101–3, 105–6; features of, 101–2; flotation samples, 102; French colonial period, 99, 108–9; House Area site maps, 104f, 107f; literacy evidence, 25; maps of, 20f; research methods, 93; settlement patterns, 102; spatial aspects of, 109; spatial dimensions of, 101; storage pits, 22, 25; and sugar transition, 99, 108–9; timeline of, 21f. See also individual House Areas Hurricanes, 28–29, 62, 90 Imray, John, 50–51 Indigenous extinction narratives, 171 Indigenous peoples. See Kalinago people Indigenous settlements, 124 Jamaica: coarse earthenware production, 132; Seville Plantation, 90 Johnson, Matthew, 127 Kalinago people: colonial society integration, 111–12; community spaces, 115–17; Dominica, name for, 10; House Area G, 111–12, 126–27, 170; houses of, 115–17, 118f; name

meaning, 12; plantation residency, 126–28; and post-emancipation society, 171; settlement characteristics, 125 Kelly, Kenneth, 90, 93, 133, 148 Kirz, Kay Dian, 113 La Mahaudière plantation, 6, 92–93, 101 Land, 32, 57–59, 171 Landscapes: and agriculture, 11–12; contested, 1–2; and enslavement, 1–2, 119; and everyday life, 15; and lime industry, 53–56, 63; postemancipation changes in, 55–59; sugar transition, 99 Lenik, Stephan, 98 Lesser Antilles: Antigua, 38, 60, 100, 132; Barbados, 38–39, 132–33; colonial power conflicts, 131–30; earthenware production, 142; enslavement, 155; geology of, 136; Grenada, 42, 125; Guadeloupe, 92, 125, 131, 133, 136; lime industry, 50; Montserrat, 50, 60, 171; Saint Vincent, 124–25; sugar transition, 154; trade among, 83. See also Dominica; Martinique Lightfoot, Natasha, 100 Lime industry: citrate manufacturing, 15, 52; collapse of, 62; costs of cultivation, 60–62; and emancipation, 48; exports, 50–52; and geography, 54–55; growth of, 25, 50–51; John Imray and, 50–51; labor, 56, 59–61; and landscapes, 53–56, 63; L. Rose and Company, 51–54; monocropping, 50; Montserrat, 50; peasantry marginalization, 62–63; plantation consolidation, 51; processing difficulties, 55; Soufriere facility, 53; sugar plantation connections, 15, 48, 53; and transportation infrastructure, 54–55 Limes, uses for, 15, 49–52 Losier, Catherine, 142 Lovelace, Earl, 171 Lowther Hall sugar estate, 126 L. Rose and Company, 51–54 Maize, 162–64 Marquardt, Bill, 17 Martinique: Crève Coeur, 6, 93, 101; Dominica, trade with, 37–38, 144–45, 147; Dominica’s dependency, 12, 14, 33, 147; earthenware production, 133; and enslavement, 38; export crop emphasis, 37; French Creoles,

Index · 199 35, 47n1; geology of, 136; migration from, 35–36; and Morne Patate, 10; trade with, 131 Mean ceramic dating (MCD): overview of, 67–68, 87n2; and correspondence analysis, 68–70; Morne Patate chronology, 67–68, 71–72, 75–78, 87n2; problems with, 68 Métayage, 48, 56, 100–101, 171 Migration: and cocoa, 60; and coffee, 36; and lime industry, 59–60; from Martinique, 35–36; from Montserrat, 60, 171; to Morne Patate, 36; neutral period, 35–37; postemancipation, 59–60, 171; from South America, 12 Mintz, Sidney: on Dominican slavery, 155; on emancipation, 100; haciendas versus plantations, 3–4; history and inequality theories, 2; on house yards, 89, 110 Montserrat, 50, 60, 171 Morne Patate: overviews of, 1, 44, 131, 156, 168; African agricultural practices, 27; animal consumption, 24; archaeological investigations of, 169; botanical experimentation, 26–27; chronology of, 6, 8t; coffee production, 8t, 14, 26, 44–47, 97; coffee remains, 80, 105, 163t; crop replacement, 16; displacement from, 28; enslaved people’s village, 112; establishment of, 13, 44–45, 97; fire, nineteenth-century, 28; geography, 11; growth of, 14, 99; houses, 99; human interaction loci, 7f; indigenous occupation evidence, 97; land use, documented, 45–47; layout of, 108; migration to, 36; organizational changes, 23–24; versus other sites, 6, 9; as plantation, 3, 6; provision grounds, 27–28; settlement history, 12, 26–27; size, 98; size changes, 23–24; sugar production, 25–26, 46–47, 166–67; value of, 44 —research at: archaeometry, 22; architectural identifications, 19, 22; archival, 22–23; community support, 18–19; comparative, 22; excavations, 18–19; human interaction loci, 18; site value, 6, 9, 17–18 —stable: botanical remains, 163t; chronology of, 72, 78; documentary evidence for, 14, 45; domestic features of, 103t; excavation, 102, 106. See also Estate house; House Areas; House yards Morne Patate chronology: Block E, 78, 80; British takeover, 75, 77–78, 80, 85; coffee

glacée, 80; economic changes, 80–86; emancipation, 77–78, 80, 86; estate area, 78, 80; Free Port Act, 83; French control, 77–78; occupational intensity changes, 76–80, 85; occupation hiatus, 78; phase definitions, 71, 77–78, 79t; stables, 78; sugar transition, 85; village area, 78, 80; wealth declines, 83–84 —Estate (Locus 1): overview, 72; correspondence analysis, 73–76; discard rates, 80–81, 83–84; earthenware, 83–84; economic changes, 84–86; excavation blocks, 72, 78; gaming pieces, 85; occupational intensity, 85; resident changes, 81, 83; sherd deposits, 77–78, 79t; table-glass vessels, 84 —Village (Locus 2): overview, 72; correspondence analysis, 73–76; discard rates, 80–81, 83–84; earthenware, 83–84; economic changes, 84–86; excavation blocks, 72, 78; gaming pieces, 85; occupational intensity, 85; residents, 81, 83; sherd deposits, 77–80; table-glass vessels, 84 Morne Patate chronology development: abundance indexes, 80–85; analytical workflow, 71–72; correspondence analysis, 68–76; and DAACS, 64–65, 72–73, 76; data analysis, 72–77; frequency seriation, 66–67, 72, 75; histogram, 76; importance of, 65, 85; mean ceramic dating, 67–68, 71–72, 75–77, 87n2; methodology overview, 64–66; modeling, 65–66; phase dating, 71; sampling error correction, 73; selection of, 72–73; sherd assignments, 78, 79f; summary of, 65; terminus post quem, 70–72, 77 Morris, D., 61–62 Multiscalar archaeology, 17 Murphy, Tessa, 26 Neutral period: overview, 33, 35; agricultural base, 33; agriculture and trade, 37–38, 154; British control continuities, 40; colonial entanglements, 31, 35; commandant, 33; early, 98; end of, 13; enslavement, 31–33, 38–39, 41; French control during, 98; French Creoles, 35–36, 47n1; migration, 35–37; reasons for, 169; societies, 36–37 Oas, Sarah, 27 Ober, Frederick, 111 Okra, 165

200 · Index Patterson, Orlando, 110 Peasant laborers: cocoa beans versus limes, 61; emergence of, 169–70; land, desire for, 56–57; and lime industry collapse, 62; marginalization of, 62–63; métayage, 48, 56, 100–101, 171; planters, conflicts with, 56; slavery connections, 155; squatters, 57–58, 60, 171; subsistence of, 156; wage systems, 56–57, 100–101 Plantations: anthropology of, 3; archaeology of, 2, 4; definitions of, 3–4; on Dominica, 5–6, 25; versus haciendas, 3–4; house yards, 16–17 Polinaire, Jean Louis, 126 Prestoe, Henry, 57 Pulsipher, Lydia, 89 Robin, Cynthia, 16 Rolle, Jeannot, 13 Rose, Lachlan, 52 Saint Mark Parish, 112 Saint Vincent, 124–25 Scott’s Head, 53 Senhouse, Joseph, 126–27 Sharecropping, 56 Singleton, Theresa, 91 Slaves. See Enslaved people Snakes, 161–62 Social death, 110 Sorghum, 105, 162–64 Soufriere: British incorporation, 32, 97; Catholic parish records, 35–36; as cultural enclave, 10; enslaved people in, 13; geography of, 10–11; highlands, 12; L. Rose and Company in, 53; and Martinique, 37; postemancipation, 14–15; settlement history, 12; settlements, 33, 34f; slope, 11–12; soil, 11 Soufriere Bay, 161

Soufriere Estate, 59 South, Stanley, 67–68 Southern Caribbee Islands colony, 40–42, 44, 47n3 Squatters, 57–58, 60, 171 Staple provisions, 157–58t Sturge, Joseph, 50, 155–56 Sugar: and the British, 40, 44, 99, 154; cultivation processes, 46; as export commodity, 2–3, 9; labor regimes, 23 Sugar industry: and emancipation, 15, 48, 57; labor classifications, 108; lime industry connections, 15, 48, 53; Morne Patate, 25–26, 46–47, 166–67; political ecology of, 25–26 Sugar Loaf estate, 6, 92 Sugar transition: and British takeover, 99, 131; and cocoa, 40; and coffee, 14, 40, 131; landscape transformations, 99; Lesser Antilles, 154; and Morne Patate, 85, 99, 108–9 Terminus post quem (TPQ), 70–72, 77 Tobin, Beth Fowkes, 113 Treaty of Paris, 35, 97 Trouillot, Michel-Rolph, 9–10, 100–101, 154 Tupi-Guarani people, 116 Visual culture, 113 Wage labor, 100–101 Wallman, Diane, 27 Ward, John, 38 Watts, David, 46 Wolf, Eric, 2–4 Wynter, Sylvia, 1–2 Young, William: and Agostino Brunias, 94, 113; on Caribbean colonization, 30, 41–42; Dominica position, 13–14; surveys commissioned, 99

Ripley P. Bullen Series Florida Museum of Natural History Tacachale: Essays on the Indians of Florida and Southeastern Georgia during the Historic Period, edited by Jerald T. Milanich and Samuel Proctor (1978) Aboriginal Subsistence Technology on the Southeastern Coastal Plain during the Late Prehistoric Period, by Lewis H. Larson (1980) Cemochechobee: Archaeology of a Mississippian Ceremonial Center on the Chattahoochee River, by Frank T. Schnell, Vernon J. Knight Jr., and Gail S. Schnell (1981) Fort Center: An Archaeological Site in the Lake Okeechobee Basin, by William H. Sears, with contributions by Elsie O’R. Sears and Karl T. Steinen (1982) Perspectives on Gulf Coast Prehistory, edited by Dave D. Davis (1984) Archaeology of Aboriginal Culture Change in the Interior Southeast: Depopulation during the Early Historic Period, by Marvin T. Smith (1987) Apalachee: The Land between the Rivers, by John H. Hann (1988) Key Marco’s Buried Treasure: Archaeology and Adventure in the Nineteenth Century, by Marion Spjut Gilliland (1989) First Encounters: Spanish Explorations in the Caribbean and the United States, 1492–1570, edited by Jerald T. Milanich and Susan Milbrath (1989) Missions to the Calusa, edited and translated by John H. Hann, with an introduction by William H. Marquardt (1991) Excavations on the Franciscan Frontier: Archaeology at the Fig Springs Mission, by Brent Richards Weisman (1992) The People Who Discovered Columbus: The Prehistory of the Bahamas, by William F. Keegan (1992) Hernando de Soto and the Indians of Florida, by Jerald T. Milanich and Charles Hudson (1992) Foraging and Farming in the Eastern Woodlands, edited by C. Margaret Scarry (1993) Puerto Real: The Archaeology of a Sixteenth-Century Spanish Town in Hispaniola, edited by Kathleen Deagan (1995) Political Structure and Change in the Prehistoric Southeastern United States, edited by John F. Scarry (1996) Bioarchaeology of Native American Adaptation in the Spanish Borderlands, edited by Brenda J. Baker and Lisa Kealhofer (1996) A History of the Timucua Indians and Missions, by John H. Hann (1996) Archaeology of the Mid-Holocene Southeast, edited by Kenneth E. Sassaman and David G. Anderson (1996) The Indigenous People of the Caribbean, edited by Samuel M. Wilson (1997; first paperback edition, 1999) Hernando de Soto among the Apalachee: The Archaeology of the First Winter Encampment, by Charles R. Ewen and John H. Hann (1998) The Timucuan Chiefdoms of Spanish Florida, by John E. Worth: vol. 1, Assimilation; vol. 2, Resistance and Destruction (1998) Ancient Earthen Enclosures of the Eastern Woodlands, edited by Robert C. Mainfort Jr. and Lynne P. Sullivan (1998) An Environmental History of Northeast Florida, by James J. Miller (1998) Precolumbian Architecture in Eastern North America, by William N. Morgan (1999) Archaeology of Colonial Pensacola, edited by Judith A. Bense (1999) Grit-Tempered: Early Women Archaeologists in the Southeastern United States, edited by Nancy Marie White, Lynne P. Sullivan, and Rochelle A. Marrinan (1999; first paperback edition, 2001) Coosa: The Rise and Fall of a Southeastern Mississippian Chiefdom, by Marvin T. Smith (2000) Religion, Power, and Politics in Colonial St. Augustine, by Robert L. Kapitzke (2001) Bioarchaeology of Spanish Florida: The Impact of Colonialism, edited by Clark Spencer Larsen (2001)

Archaeological Studies of Gender in the Southeastern United States, edited by Jane M. Eastman and Christopher B. Rodning (2001) The Archaeology of Traditions: Agency and History Before and After Columbus, edited by Timothy R. Pauketat (2001) Foraging, Farming, and Coastal Biocultural Adaptation in Late Prehistoric North Carolina, by Dale L. Hutchinson (2002) Windover: Multidisciplinary Investigations of an Early Archaic Florida Cemetery, edited by Glen H. Doran (2002) Archaeology of the Everglades, by John W. Griffin (2002; first paperback edition, 2017) Pioneer in Space and Time: John Mann Goggin and the Development of Florida Archaeology, by Brent Richards Weisman (2002) Indians of Central and South Florida, 1513–1763, by John H. Hann (2003) Presidio Santa María de Galve: A Struggle for Survival in Colonial Spanish Pensacola, edited by Judith A. Bense (2003) Bioarchaeology of the Florida Gulf Coast: Adaptation, Conflict, and Change, by Dale L. Hutchinson (2004; first paperback edition, 2020) The Myth of Syphilis: The Natural History of Treponematosis in North America, edited by Mary Lucas Powell and Della Collins Cook (2005) The Florida Journals of Frank Hamilton Cushing, edited by Phyllis E. Kolianos and Brent R. Weisman (2005) The Lost Florida Manuscript of Frank Hamilton Cushing, edited by Phyllis E. Kolianos and Brent R. Weisman (2005) The Native American World Beyond Apalachee: West Florida and the Chattahoochee Valley, by John H. Hann (2006) Tatham Mound and the Bioarchaeology of European Contact: Disease and Depopulation in Central Gulf Coast Florida, by Dale L. Hutchinson (2007) Taíno Indian Myth and Practice: The Arrival of the Stranger King, by William F. Keegan (2007) An Archaeology of Black Markets: Local Ceramics and Economies in Eighteenth-Century Jamaica, by Mark W. Hauser (2008; first paperback edition, 2013) Mississippian Mortuary Practices: Beyond Hierarchy and the Representationist Perspective, edited by Lynne P. Sullivan and Robert C. Mainfort Jr. (2010; first paperback edition, 2012) Bioarchaeology of Ethnogenesis in the Colonial Southeast, by Christopher M. Stojanowski (2010; first paperback edition, 2013) French Colonial Archaeology in the Southeast and Caribbean, edited by Kenneth G. Kelly and Meredith D. Hardy (2011; first paperback edition, 2015) Late Prehistoric Florida: Archaeology at the Edge of the Mississippian World, edited by Keith Ashley and Nancy Marie White (2012; first paperback edition, 2015) Early and Middle Woodland Landscapes of the Southeast, edited by Alice P. Wright and Edward R. Henry (2013; first paperback edition, 2019) Trends and Traditions in Southeastern Zooarchaeology, edited by Tanya M. Peres (2014) New Histories of Pre-Columbian Florida, edited by Neill J. Wallis and Asa R. Randall (2014; first paperback edition, 2016) Discovering Florida: First-Contact Narratives from Spanish Expeditions along the Lower Gulf Coast, edited and translated by John E. Worth (2014; first paperback edition, 2016) Constructing Histories: Archaic Freshwater Shell Mounds and Social Landscapes of the St. Johns River, Florida, by Asa R. Randall (2015) Archaeology of Early Colonial Interaction at El Chorro de Maíta, Cuba, by Roberto Valcárcel Rojas (2016) Fort San Juan and the Limits of Empire: Colonialism and Household Practice at the Berry Site, edited by Robin A. Beck, Christopher B. Rodning, and David G. Moore (2016) Rethinking Moundville and Its Hinterland, edited by Vincas P. Steponaitis and C. Margaret Scarry (2016; first paperback edition, 2019)

Gathering at Silver Glen: Community and History in Late Archaic Florida, by Zackary I. Gilmore (2016) Paleoindian Societies of the Coastal Southeast, by James S. Dunbar (2016; first paperback edition, 2019) Cuban Archaeology in the Caribbean, edited by Ivan Roksandic (2016) Handbook of Ceramic Animal Symbols in the Ancient Lesser Antilles, by Lawrence Waldron (2016) Archaeologies of Slavery and Freedom in the Caribbean: Exploring the Spaces in Between, edited by Lynsey A. Bates, John M. Chenoweth, and James A. Delle (2016; first paperback edition, 2018) Setting the Table: Ceramics, Dining, and Cultural Exchange in Andalucía and La Florida, by Kathryn L. Ness (2017) Simplicity, Equality, and Slavery: An Archaeology of Quakerism in the British Virgin Islands, 1740–1780, by John M. Chenoweth (2017) Fit for War: Sustenance and Order in the Mid-Eighteenth-Century Catawba Nation, by Mary Elizabeth Fitts (2017) Water from Stone: Archaeology and Conservation at Florida’s Springs, by Jason O’Donoughue (2017) Mississippian Beginnings, edited by Gregory D. Wilson (2017; first paperback edition, 2019) Harney Flats: A Florida Paleoindian Site, by I. Randolph Daniel Jr. and Michael Wisenbaker (2017) Honoring Ancestors in Sacred Space: The Archaeology of an Eighteenth-Century African-Bahamian Cemetery, by Grace Turner (2017) Investigating the Ordinary: Everyday Matters in Southeast Archaeology, edited by Sarah E. Price and Philip J. Carr (2018) New Histories of Village Life at Crystal River, by Thomas J. Pluckhahn and Victor D. Thompson (2018) Early Human Life on the Southeastern Coastal Plain, edited by Albert C. Goodyear and Christopher R. Moore (2018) The Archaeology of Villages in Eastern North America, edited by Jennifer Birch and Victor D. Thompson (2018) The Cumberland River Archaic of Middle Tennessee, edited by Tanya M. Peres and Aaron DeterWolf (2019) Pre-Columbian Art of the Caribbean, by Lawrence Waldron (2019) Iconography and Wetsite Archaeology of Florida’s Watery Realms, edited by Ryan Wheeler and Joanna Ostapkowicz (2019) New Directions in the Search for the First Floridians, edited by David K. Thulman and Ervan G. Garrison (2019) Archaeology of Domestic Landscapes of the Enslaved in the Caribbean, edited by James A. Delle and Elizabeth C. Clay (2019) Cahokia in Context: Hegemony and Diaspora, edited by Charles H. McNutt and Ryan M. Parish (2020) Bears: Archaeological and Ethnohistorical Perspectives in Native Eastern North America, edited by Heather A. Lapham and Gregory A. Waselkov (2020) Contact, Colonialism, and Native Communities in the Southeastern United States, edited by Edmond A. Boudreaux III, Maureen Meyers, and Jay K. Johnson (2020) An Archaeology and History of a Caribbean Sugar Plantation on Antigua, edited by Georgia L. Fox (2020) Modeling Entradas: Sixteenth-Century Assemblages in North America, edited by Clay Mathers (2020) Archaeology in Dominica: Everyday Ecologies and Economies at Morne Patate, edited by Mark W. Hauser and Diane Wallman (2020)