The history of plantation slavery in the Americas has in recent years been enriched by a growing number of studies that focus on provision grounds and gardens– the small plots of land where, working in their “free” time, slaves grew food crops for their own use. Studies from the West Indies and the southern United States have shown that granting slaves access to land was a common practice in many areas of slave-based agriculture; indeed, slaves frequently gained by custom the “right” to gardens or provision grounds and to the time needed to work in them. Slaves in some regions not only obtained a large share of their daily diet from their provision grounds, but also managed to harvest sizable surpluses for sale in local markets. Even where planters relied chiefly on rations rather than on provision grounds to feed their bondsmen, slaves often succeeded in selling produce from their plots. Thus, in several plantation regions, provision grounds and gardens allowed slaves to develop an extensive and even impressive range of independent production and marketing activities.1

The possibility that similar independent slave production and marketing activities may have also emerged at times in Brazil stands at the center of an ongoing and sometimes heated debate among historians interested in Brazilian slavery.2Roças, as provision grounds and garden plots were known in Brazil, may seem a small matter for so much debate.3 But at issue here are broader questions about slavery and the relationships between slaves and masters. Should we, while recognizing the constraints of slavery and the overriding authority of slaveowners, search out ways in which slaves took an active part in shaping their own lives and those of their masters? Or should we see slaveowners as holding such absolute power over their bondsmen that slaves lacked any means to carve out a measure of autonomy in their daily lives?

These two conflicting perspectives on slavery lead, in turn, to sharply different interpretations of the role of provision grounds both in the lives of Brazilian slaves and in master-slave relationships. Thus, from the first perspective, provision grounds and garden plots might serve the interests of planters as a means of reducing their out-of-pocket expenses in sustaining their slaves. More than that, however, roças could provide slaves with an economy of their own. Over time and by custom, slaveowners might find themselves obliged to acknowledge their slaves’ “rights” to a roça, to the time needed to cultivate it, and to whatever produce it might yield. The importance of provision grounds and garden plots from this second perspective did not lie primarily in the amount of produce harvested from them; rather, it lay in the autonomy slaves enjoyed in cultivating them. Roças might be mere scraps of land, the hours to work in them few, and the harvests pitifully small, but here slaves, not their masters, made the decisions.

From the second perspective, when and where provision grounds existed, they benefited chiefly and even exclusively the planters, who freely manipulated the size of the plots and the time allotted to slaves to work in them according to their own interests. To the slaves, access to a plot of land for a garden may have represented not so much a right or a privilege as an additional burden. Authors adopting this perspective therefore not only question whether the practice of assigning provision grounds was at all common in Brazil; they also criticize as misguided and misdirected any interpretation that fails to stress that, even while cultivating provision grounds, slaves remained locked into an exploitative relationship entirely controlled by their masters. Thus, Jacob Gorender argues that the current interest in provision grounds and in independent activities by rural slaves gravely distorts our understanding of slave life, both by, supposedly, minimizing the absolute power of planters over their slaves and by falsely suggesting that Brazilian slaves may have found opportunities to shape, at least in part, their own lives. Scholarship on this topic, in Gorender’s view, reflects in turn a broader “neopatriarchalist” effort to rehabilitate slavery”–that is, a conservative attempt to revive and update older and now largely discredited arguments about the mildness of slavery in Brazil.4

Much of the debate has, until recently, been cast at a very broad and general level; indeed, discussions about provision grounds in Brazil have often concentrated on interpreting and reinterpreting a handful of well-known examples, such as the 1789 uprising at the Engenho Santana. Moving beyond those examples, this essay attempts to gauge the importance of provision grounds and independent slave marketing activities by focusing on sugar plantations and cane farms in the region known as the Bahian Recôncavo in Northeastern Brazil in the late eighteenth and early nineteenth centuries.5 It takes up three distinct but related questions: Was the practice of assigning provision grounds at all widespread on Bahian plantations and cane farms in the eight decades stretching from 1780 to 1860? If so, did plantation slaves draw a major share of their daily sustenance from the crops they cultivated in their roças? If they did have access to land for their own use, did Bahian slaves regularly sell or market surplus produce from their provision grounds?

The task of addressing those questions requires work with widely scattered and often sketchy sources to gather information about the daily lives of plantation slaves. It also involves weighing positive evidence against significant silences–that is, the lack of information in sources where, drawing on comparable research for other areas, one could reasonably expect to find references to independent production and marketing activities by slaves.6 For the Bahian Recôncavo, the balance of positive evidence and significant silences strongly suggests that although planters and cane farmers in the late eighteenth and early nineteenth centuries often granted their slaves access to roças, those slaves generally did not draw the bulk of their sustenance from those plots and only occasionally used them to market surplus produce. As a result, independent production and marketing activities by plantation slaves remained far less developed in the Recôncavo than in some of the main sugar-producing areas in the Caribbean. The explanation for that limited development lies not only within the plantations and cane farms of the Recôncavo, but also in the broader regional economy beyond their boundaries. Nevertheless, in the Recôncavo, just as in many other areas of slave-based agriculture, provision grounds allowed plantation slaves to create a restricted but still significant “economy” of their own.

Growth of the Sugar Trade and the Slave Labor Force

By the late eighteenth century, the captaincy of Bahia in Northeastern Brazil possessed a well-developed and mature colonial economy based firmly on slave labor. The heart of that economy lay in the Recôncavo, a fertile region of roughly ten thousand square kilometers surrounding the Bay of All Saints. Stretching back from the shores of the bay, amid still-extensive stands of forest, were hundreds of large estates and smaller farms where slaves planted sugarcane, tobacco, cassava, and various other crops. The Recôncavo also held a dozen or more villages, hamlets, and small towns, as well as the city of Salvador, Bahia’s capital and main port and one of Brazil’s largest urban centers. Located at the mouth of the bay, Salvador served as a focal point for both international and regional trade. Ships from Europe unloaded textiles, hardware, and other manufactures, while those from Africa set ashore hundreds, even thousands, of newly enslaved men and women every year. Into Salvador’s harbor also flowed the wide range of commodities produced in the Recôncavo and adjacent regions for sale in local markets and overseas shipment.7 The volume and variety of goods handled by Bahia’s internal trade could impress even an experienced English merchant like Thomas Lindley. In 1805 Lindley wrote, “the trade carried on within the immediate confines of the Bay, of which a great part is inland, is astonishing. There are a full eight hundred launches and sumacks [smacks] daily bringing their tribute of commerce to the capital ”: tobacco, sugar, cotton, and timber for export as well as maize, firewood, all manner of vegetables, whale oil, salt fish, earthenware pottery, and cassava flour for local consumption.8

Of all the products Lindley lists, sugar carried the greatest economic weight in Bahia. Sugar was Bahia’s oldest and most valuable agricultural export. The Portuguese colonists who first settled the Recôncavo in the mid-sixteenth century crossed the Atlantic with plans to cultivate sugarcane and to build mills for grinding cane and manufacturing sugar. They soon discovered that the heavy, clay-filled soils found on the northern side of the Bay of All Saints were ideally suited for cane. By the late 1580s, the settlers and their Brazilian-born descendants had established in the Recôncavo 36 engenhos (plantations with mills), two-thirds of them on the northern side of the bay. The mills on these engenhos manufactured sugar from cane cultivated both by plantation owners and by sharecropping cane farmers (lavradores de cana), who generally owned slaves and might even own land, but who did not possess their own milling equipment.

The engenhos and cane farms of the Recôncavo quickly transformed the region into one of the first centers of slave-based plantation agriculture in the Americas and a major supplier of sugar to the world market.9 Sugar would continue to dominate the landscape of the northern Recôncavo for the next three hundred years. By the 1790s, nine-tenths of the 221 engenhos in Bahia were located in the parishes and townships along or near the northern shore of the Bay of All Saints.10 The same districts contained an even larger number of cane farms.11

The presence of more than two hundred engenhos in the Recôncavo made Bahia one of the main sugar-producing regions in Portuguese America in the late eighteenth and early nineteenth centuries. Between 1796 and 1807, Bahia exported more sugar than any other captaincy in Brazil.12 This staple generally accounted for at least 45 percent of all revenues Bahia earned from exports in the period 1780-1860, and its share was often larger.13 As a result, the Bahian economy as whole could know no true prosperity when depressed conditions prevailed in the sugar trade.

Depression and stagnation did characterize the sugar trade for a hundred-year period starting in the 1680s. Competition from British and French colonies in the West Indies captured from Bahia and other Brazilian sugar-producing regions a progressively larger share of the world market.14 Then, at the end of the eighteenth century, wars in Europe and North America, the onset of the Industrial Revolution in Great Britain, and the Haitian Revolution raised prices and opened new and growing markets for Bahian sugar. The long years of stagnation and depression gave way, in the decades after 1770 and especially after 1790, to a new period of substantial, if not always steady, growth that would last through the 1850s. Whereas exports before the 1770s had generally stood at about 400,000 arrobas (slightly less than 5.9 million kilograms) of sugar a year, they reached an annual average of nearly 760,000 arrobas by the late 1790s. In the next three decades the volume of the Bahian sugar trade more than doubled. Although the 1830s were marked once again by stagnation and even decline, exports quickly recovered in the following decade, and by the early 1850s Bahia was exporting, on average, more than 3 million arrobas of sugar a year. Information on output shows the same long upward trend. Sugar production on Bahian engenhos underwent a fourfold increase between 1807 and the mid-1850s.15 Similarly, the number of engenhos in the Recôncavo rose from about two hundred in the 1790s to more than five hundred in 1842 and perhaps up to six hundred by 1860.16 It is likely that, during the same years, the number of cane farms in the Recôncavo also multiplied.

The sugar industry’s growth and expansion resulted in an increasing demand for slave labor, supplied in large part through the transatlantic slave trade. A recent set of estimates suggests that Bahia may have imported as many as 409,000 slaves from Africa between 1785 and 1852.17 Of course, not all of those slaves ended up on the sugar plantations and cane farms of the Recôncavo. Many were re-exported to other parts of Brazil. Of those who remained in Bahia, some were purchased by urban slaveholders in Salvador; others were no doubt sold to farmers who cultivated tobacco and cassava in the western and southern Recôncavo or sent to ranches and cotton farms in the backlands. Yet it is safe to assume that planters and cane farmers bought large numbers of African slaves in the late eighteenth and early nineteenth centuries.18

Slaves were needed not only to staff new engenhos but also to expand the work force on existing ones. At the Engenho Novo in the important sugar-producing parish of Santiago do Iguape, for example, the resident slave population rose from 137 in 1793 to just less than 200 by 1835. The number of slaves at nearby Engenho Maroim nearly doubled between 1791 and 1835. Perhaps the most spectacular increases in Iguape took place at the Engenhos da Cruz and Vitória. In the mid-1820s, Captain Tomé Pereira de Araújo had kept only 47 slaves at his new and still fairly small Engenho da Cruz. A decade later, his estate had a resident slave population of 167, nearly three-fourths of whom were African by birth. Some 12 kilometers southwest of Captain Tomé’s estate stood the Engenho Vitória, which in the mid-1780s was a smaller-than-average plantation. Around 1814 Vitória came into the hands of Pedro Rodrigues Bandeira, a wealthy merchant with connections to the slave trade. Bandeira quickly set about building up the engenho’s labor force; at his death in 1835, it numbered 241 slaves. That was nothing less than a fivefold increase over the fewer than 50 slaves who had worked there in 1786.19

Roças on Bahian Plantations and Cane Farms

Thus, between 1780 and 1860, Bahian planters and cane farmers confronted the problem of how to maintain and expand sugar production and, at the same time, feed a growing slave labor force. One solution would have been to rely on roças and, in effect, turn over responsibility for slave subsistence to the slaves themselves. Planters and cane farmers could provide their bondsmen with enough land and “free” time to grow the food they needed. In Bahia, though, assigning slaves provision grounds would not have eliminated entirely the need for issuing rations unless planters also introduced major changes in the slave diet.

Contemporary sources concur that jerked beef (carne seca or charque)–sometimes replaced by salt cod during Lent–and cassava flour, known in Brazil as farinha de mandioca or simply farinha, were the two main staples in that diet.20 Jerked beef came from far southern Brazil and from the distant backlands of the Northeast; salt cod was imported from Europe. But slaves could easily grow cassava to make farinha in a roça.21 On the same plots they could also plant maize, beans, and vegetables, while around their huts they might keep poultry and goats or pigs.

Obliging slaves to grow their own food was a solution that apparently had its New World origin in the sugar-producing regions of Northeastern Brazil. On some islands in the West Indies, assigning slaves a plot of land for their subsistence was known as “the Brazil system” because Dutch colonists had introduced the practice after seeing it used in Pernambuco in the mid-seventeenth century.22 Recent research on other regions of Brazil has also uncovered evidence from the eighteenth and nineteenth centuries that slaves cultivated food crops for their own use on roças.23

Although provision grounds and gardens were found in many areas of plantation agriculture, their role in the slave diet varied greatly from one region to another. At one extreme, slaves in colonies such as Jamaica, French Guyana, and Grenada drew most of their sustenance from what they grew in their plots. At the other extreme, plantation slaves in Barbados, Antigua, and the southern United States depended chiefly on rations issued to them by their owners.24 This diversity makes it necessary to proceed cautiously in looking at provision grounds in Bahia. It warns against assuming hastily that what took place elsewhere in Brazil or in other sugar- producing regions also occurred in the Recôncavo. The need for caution is all the greater because Jacob Gorender has specifically argued that provision grounds were seldom found on Northeastern sugar plantations, where the agricultural calendar was long and where work on the main staple crop kept slaves occupied for nearly the entire year.25 Gorender’s argument, although logical, is based mainly on conjecture, because no study to date has investigated in any detail independent economic activities by plantation slaves in the main sugar-producing regions of Northeastern Brazil.

The first task, then, is simply to establish whether slaves on the engenhos and cane farms of the Recôncavo often cultivated provision grounds. On this question there is no lack of evidence. The length of the agricultural calendar did not by itself preclude the possibility of allowing slaves time and land to grow food crops in a roça. Bahian chroniclers from the late eighteenth century, such as José da Silva Lisboa, Luís Antônio de Oliveira Mendes, and Luís dos Santos Vilhena, all refer to the practice. In the 1830s, both Manoel Ferreira da Câmara, the owner of a large engenho in the Recôncavo, and Miguel Calmon, a representative of one of Bahia’s most important planter families, mentioned the custom of assigning slaves a plot of land for a roça.26

While Calmon and Câmara were writing about provision grounds, the administrator of the Fazenda da Capela de Santa Maria, a large cane farm in the parish of Iguape, paid slaves there for work on Sundays and holy days, time they could have spent in cultivating their roças.27 Miguel de Teive e Argolo, the owner of the Engenho Santo Estêvão in the township of São Francisco do Conde, did the same 20 years later. Argolo paid his slaves a total of Rs.58$000 (58 mil-réis) for four Sundays of work in January and April 1855. Although, unfortunately, he neglected to record in his accounts how many slaves worked on those Sundays, he paid them a considerable sum. A free blacksmith Argolo hired in April 1855 to repair equipment in Santo Estêvão’s millhouse earned only Rs.10$000 for ten days of work. That same year the plantation’s free overseer received a monthly salary of Rs.25$000.28

The slaves at Argolo’s estate in 1855 could have collectively used their Sunday earnings to purchase at least 38 alqueires (bushels) of cassava flour, enough to keep 49 adults fed for an entire month. Instead of flour, they might have bought 14 arrobas (about 212 kilograms) of fresh beef or up to 60 live chickens or even a full two tarefas (8,716 square meters) of sandy but productive land near the plantation.29 At about the same time that slaves at Santo Estêvão were being paid for Sunday work, two English Quakers visiting the Engenho Vitória in Iguape learned that adult slaves there cultivated “provision-grounds for their own benefit.”30

An 1854 survey of plantations in the parish of Nossa Senhora da Piedade de Matoim on the outskirts of Salvador, a few miles from Santo Estêvão, describes the Engenhos Novo Caboto and Matoim as “occupied by cane fields, pastures, and the slaves’ cassava plants (as mandiocas dos escraυos).”31 This survey, in mentioning the slaves’ roças, stands out from other descriptions of plantations in eighteenth- and nineteenth-century Bahia. Rarely do such descriptions refer to the location of roças, their size, the crops planted in them, or even their existence. This also holds true for sources, such as postmortem estate inventories, that might be expected to supply this information. Estate appraisers often recorded in great detail the plantation’s boundaries, the size and location of cane fields, the condition of fences, the quality of soils found on the property, and the tools– including broken hoes and axes–that plantation slaves used in their daily work. They sometimes even went so far as to count, describe, and appraise individual dendezeiros (oil palms), fruit trees, coconut palms, and touceiras de bananeiras (clumps of banana trees). Yet estate appraisers almost never took notice of the provision grounds cultivated by slaves. The descriptions of rural properties contained in the registros eclesiásticos de terras (ecclesiastical land registries) from the 1850s similarly fail to mention slave cultivation of roças (see  appendix).

The lack of explicit references obviously makes the historian’s work more difficult. As João Fragoso and Manolo Florentino argue, however, in not recording or evaluating provision grounds, estate appraisers implicitly acknowledged that the crops planted by slaves in roças did not belong to the property owner. In effect, both appraisers and heirs conceded that by custom and in practice, slaves had “rights to the crops they had cultivated for their own use.32

The rare references that do appear in Bahian inventories do not contradict Fragoso and Florentino’s argument. For example, in 1857, Antônio de Souza Ribeiro, a farmer in the southern Recôncavo who did not plant sugarcane, left his heirs, among other possessions, a coffee grove that had once belonged to a slave named Joaquim (“outra porção de cafezeiros que foram do escraυo Joaquim”) with an appraised value of Rs. 15$000. This coffee grove entered the inventory only because it no longer belonged to Joaquim. Joaquim was not even present among the slaves owned by Ribeiro at his death. Joaquim may have died or fled, or perhaps had been sold or freed; he had, in any case, surrendered or lost his claim to the grove.33

Not surprisingly, then, a European visitor has left the most detailed description of provision grounds on any Bahian engenho. The French diplomat Baron Forth-Rouen took advantage of a stopover in Bahia in 1847 to visit several engenhos in the Recôncavo. At a plantation belonging to the viscount of Pedra Branca, he found,

each [slave] possesses a lot (une portion de terre) that he chooses where he likes and cultivates when and how he pleases. They [the slaves] all have a horse; some have several horses that they hire out to their master. They also have cattle, sheep, etc.34

Forth-Rouen added that in this matter the viscount allowed his slaves exceptional freedom and that he constantly worried about their well-being; he required them to work only from nine o’clock in the morning until three in the afternoon.35 Unlike the viscount, most planters surely had a say in where slaves could grow food crops and when slaves could work in their provision grounds. It is also unlikely that many other slaves in the Recôncavo owned horses.36 What is not exceptional, however, is their freedom to plant their roças as they pleased.37 Thus, Bahia in the late eighteenth and early nineteenth centuries did not stand apart from other areas of slave-based agriculture in the Americas; like slaves on plantations in many parts of the West Indies and the southern United States, slaves on the engenhos and cane farms of the Recôncavo often cultivated provision grounds.

Subsistence and Surplus

The question now becomes whether provision grounds furnished those slaves with the bulk of their subsistence. Did slaves on many of the Recôncavo’s plantations and cane farms grow in their roças enough food, including cassava, to sustain themselves? According to José da Silva Lisboa, writing in 1781, they did. He explained that “the sustenance of slaves is not usually a burden to their masters because, by an almost universal custom, [the slaves] are granted the days of Saturday and Sunday to plant whatever they need and a plot of land.” 38 Yet Lisboa may have exaggerated. The church in colonial Bahia at times complained that planters allowed their slaves to work in their roças only on Sundays when they should rest, like other Christians.39 A decade after Lisboa wrote, moreover, Luís Antônio de Oliveira Mendes noted that planters often purchased basic provisions, such as farinha. Luís dos Santos Vilhena in the late 1790s likewise observed that only some planters refused to give their slaves rations and forced them to subsist solely on the produce of their garden plots. These planters, in Vilhena’s view, were the cruelest and most inhumane in the treatment of their bondsmen.40

Like Vilhena, Manoel Ferreira da Câmara in the 1830s also denounced as inhumane those slaveowners who failed to distribute rations. Because slaves could not survive on their provision grounds alone, he argued, this amounted to telling them, “Go out and steal to sustain yourselves.” Câmara offered the example of some of his neighbors who did not give out rations: their slaves had, he claimed, “stolen from me even leftover bits of yams, and not a single bunch of bananas escapes their grasp.”41 Miguel Calmon, writing at the same time, could find all sorts of advantages for the planter in the custom of assigning provision grounds. Among those advantages, significantly, he did not list the possibility of turning over to slaves full responsibility for their own subsistence. On the contrary, he followed Vilhena and Câmara in condemning those masters who gave their bondsmen a day instead of a ration.” Those slaveowners, according to Calmon, acted not so much out of “cruelty” as out of “stolid indifference” to the well-being of their slaves. Nevertheless, Calmon acknowledged that the result was hardly humane; he described the slaves belonging to those masters as the very picture of “misery, vice, and death.”42

All in all, these comments do not take us very far. They merely establish that on some plantations slaves obtained most of their subsistence from their roças. At no point do these authors even hint that this occurred on a majority of Bahian plantations. Moreover, in denouncing those planters who did not give out rations, all three authors do little more than resurrect moralistic arguments that reformist writers such as the Jesuits André João Antonii and Jorge Benci first put forth in the early eighteenth century.43 At most, we might very tentatively conclude that, with the revival of the sugar trade, the number of plantations where slaves did not receive rations diminished between the early 1780s, when Lisboa referred to the matter, and later decades, when Vilhena, Câmara, and Calmon took up the same question.

To go beyond the limited remarks of these authors, we need to approach the question from a different direction and with different sources. We can, for example, look for slaves selling produce from their roças. Almost everywhere slaves sold some of the produce they harvested, but the volume and frequency of those sales varied considerably from region to region. Especially in those regions of the Americas where provision grounds supplied most of the food eaten by slaves, they often yielded marketable surpluses as well. In Jamaica, St.-Domingue, Grenada, Martinique, and French Guyana, slaves not only grew much of their own food but also produced considerable amounts of vegetables and other foodstuffs for sale in local markets. As a result, around 1775, slaves in Jamaica held approximately one-fifth of all the circulating currency in that British colony. Nearly a third of Jamaica’s entire agricultural output in the early 1830s came from provision grounds cultivated by slaves in their free time. Slaves in Martinique used their plots in the early nineteenth century to become major suppliers of cassava flour in the island’s markets. Similarly, by the 1790s, slaves held a “virtual monopoly” over the supply of locally grown produce sold in Grenada and St. Vincent. Closer to Bahia, in the interior captaincy of Goiás, tithe records from the late eighteenth century indicate that slaves, acting on their own behalf and selling their own produce, played an important role in keeping that captaincy’s towns supplied with foodstuffs.44

Did Bahian slaves regularly market large surpluses of produce from their provision grounds? Stuart Schwartz argues that the resurgence of the sugar industry may have encouraged independent production and marketing activities by plantation slaves in Bahia and other sugar-producing areas of Brazil. The renewed growth in exports coincided with a general rise in the price of basic foodstuffs in urban markets. “If there was a time when slaves might profit from a marketing of their surplus,” Schwartz writes, “the late colonial era was that time.”45

The well-known example of the Engenho Santana near Ilhéus might suggest that plantation slaves in Bahia did frequently sell the produce of their garden plots in local markets.46 There, in 1789, a group of slaves rebelled, killed an overseer, and fled into the nearby forests. They took with them most of the engenho’s tools and thus brought work at Santana to a halt for nearly two years. Finally, in 1791, after enduring attacks from military expeditions, they negotiated a peace treaty with Santana’s owner, Manoel da Silva Ferreira, outlining the conditions under which they would agree to return to Santana: reductions in work quotas, elimination of disagreeable tasks, and guarantees that Ferreira would assign specific numbers of workers to certain jobs. In another clause, however, the slaves demanded more time to work in their roças and greater freedom to choose where they could clear ground to plant food crops.47 They also requested that Ferreira “make a large boat so that when it goes to Bahia [that is, Salvador] we can place our cargoes aboard and not pay freightage.” This clause clearly suggests that the slaves were accustomed to selling produce from their provision grounds; they now sought the right to ship that produce– free of charge–to Salvador’s market.

For Schwartz, who first brought this remarkable document to scholars’ attention, the peace treaty offers “a rare insight into slave wishes” and the slaves’ “perceptions of life on a large sugar plantation”–in short, “the slaves’ view of slavery.”48 Schwartz’s interpretation is convincing; without doubt, slaves on many Bahian plantations would have soundly endorsed the demands made by the runaways at Santana. The peace treaty also helps to establish that it was not impossible for slaves on Brazilian engenhos to market foodstuffs on their own behalf.

Nevertheless, it would be unwise to conclude from this example that slaves on sugar plantations and cane farms in the Recôncavo regularly sold surplus produce from their roças. The grounds for caution have little to do with the fact that the Jesuit order owned and administered Santana until 1759, or that it was a larger-than-average estate with an unusually large proportion of creole slaves in its work force.49 Rather, generalizing from this example would be unwise mainly because the Engenho Santana, by its setting, was atypical of both sugar plantations in the Recôncavo and engenhos in the main sugar regions of Northeastern Brazil. Santana was located outside the Recôncavo, some two hundred miles south of Salvador at an isolated point in the sparsely populated township of Ilhéus. The whole township, with its extensive and heavily forested rural districts, had no more than two thousand inhabitants and only four or five engenhos at the end of the eighteenth century.50 More important, the town of Ilhéus apparently suffered a chronic shortage of foodstuffs. Townsfolk and travelers often found it difficult to purchase provisions, and as late as 1780, the township lacked a single butcher.51 Even the slaves at Santana recognized that the poorly developed local market in Ilhéus was no place to sell their produce; they asked in their peace proposal for easier access to Salvador’s market.52 Under precisely these conditions, a planter would be most likely to turn over to his slaves full responsibility for their own subsistence. But such conditions contrast sharply with those in the late eighteenth-century Recôncavo, with its more than two hundred engenhos, comparatively thick settlement, and well-stocked local markets where, if necessary, planters could easily purchase cassava flour and other provisions to feed their slaves.53

Therefore, before taking the incident at Santana as evidence that plantation slaves elsewhere in Bahia frequently engaged in independent marketing, the search must turn to sources that refer specifically to the Recôncavo, such as the municipal laws and ordinances (posturas) passed by town councils in the region. If rural slaves in the Recôncavo often sold the produce of their roças in local markets, we might reasonably expect municipal authorities, like their counterparts elsewhere, to take some notice of such activities. The governor of Martinique in the 1840s, for example, posted a police force of 30 to supervise a market where rural slaves regularly gathered to sell their produce. In the upland cotton districts of South Carolina and in several British colonies in the West Indies, special laws regulated marketing activities by slaves.54 Town councils in other regions of Brazil also approved legislation dealing with the independent sale of produce by slaves.55 Thus it would be truly surprising if Bahian authorities did not try to control or even prohibit a regular movement of slaves back and forth between plantations and markets. Town councils in the Recôncavo did approve numerous other ordinances regulating local markets and the sale of foodstuffs. They also imposed various restrictions on slaves in their jurisdictions. For example, slaves were prohibited from gathering in groups to dance, from gambling, and from purchasing poison. These laws reflected specific local concerns as well as a broader effort to control and limit any unsupervised time slaves spent away from their masters.56

The need for such legislation was particularly pressing in Bahia, where, between 1809 and 1835, both rural and urban slaves rose up in an unparalleled series of revolts and insurrections. A state of more or less constant fear, verging at times on hysteria, took hold of Bahia’s free population in those years. Even after 1835, the lasting memory of those revolts and unfounded rumors of new rebellions kept the fear alive and made clear the need for greater vigilance and stricter control over the slave population.57 Yet no town council in the Recôncavo in the nineteenth century ever passed a law aimed at rural slaves who frequented local markets.58

A set of emergency regulations proposed in 1816 reveals the same apparent lack of concern. In February of that year, rural slaves in the important sugar-producing townships of Santo Amaro and São Francisco do Conde rose up and burned several engenhos. They also attacked the town of Santo Amaro and killed a number of white residents. As soon as the four-day uprising had been crushed, a group of nearly 40 wealthy planters and local notables gathered in São Francisco do Conde and outlined a series of drastic measures to prevent future rebellions. They demanded that any suspicious-looking slave be summarily arrested and deported, and that slaves caught in acts of rebellion be hanged immediately, without a trial. The planters also demanded that the colonial government strictly control the movement of free blacks and mulattos. If the market fairs of the Recôncavo had served as regular meeting places for large numbers of rural slaves (as they did in some Caribbean sugar colonies), the planters of Santo Amaro and São Francisco do Conde surely would have targeted those markets in their proposed emergency regulations. But none of the measures they suggested refers to rural slaves who, on their own behalf, sold produce in local markets, or to dealings between such slaves and merchants.59

Thus, local authorities and planters in the Recôncavo–perhaps more so than anywhere else in Brazil–had sound reasons to fear any unsupervised activities by rural slaves, and equally sound reasons for trying to restrict or regulate situations, such as weekly markets, that might bring together hundreds of slaves from different plantations and place them in direct contact with free and freed blacks and mulattos.60 One justice of the peace from a sugar-producing parish put the matter clearly enough when, writing in 1830, he observed that “frequent gatherings of slaves [are] always dangerous.” Then, alluding to the rebellions and revolts that had already taken place, the same justice of the peace went on to note that “the crisis that currently threatens public security” had made those gatherings “all the more [dangerous].”61 Nevertheless, neither local authorities through municipal legislation nor even the planters of Santo Amaro and São Francisco do Conde who had just confronted a serious slave rebellion ever proposed restrictions or controls on rural slaves’ independent marketing activities. Their lack of any explicit concern with such activities not only stands in striking contrast with the official attention that colonial governments devoted to the matter in some parts of the West Indies, but would also seem to suggest that, on the whole, plantation slaves in the Recôncavo seldom sold surplus produce from their provision grounds in local markets.

Probate Records

Like municipal ordinances, postmortem estate inventories also represent a potentially useful source of information about sales of foodstuffs by rural slaves. For example, these probate records almost always contain declarations of unpaid debts, including whatever debts the deceased may have owed to slaves. Debts of this sort might easily arise where slaves sold their masters produce they had harvested from their provision grounds. Where, for whatever reason, the final settlement of an estate was delayed, the executors often submitted to the probate authorities running accounts of their expenditures in administering the estate. These accounts, which then became part of the inventory, document major expenditures, such as the acquisition of new equipment or the salaries paid to free overseers. In many instances they also record small sums spent in purchasing vegetables, poultry, maize, beans, farinha, and other foodstuffs–products that slaves could have furnished from their provision grounds. Researchers working with comparable probate records in other areas of Brazil have demonstrated that rural slaves in those areas did sell surplus produce from their roças and, in some cases, even livestock.62

Inventories of Bahian sugar planters and cane farmers also yield examples in which rural slaves sold produce on their own behalf and engaged in other independent economic activities. The examples go far to suggest that slaves in the sugar districts of Bahia could indeed achieve a restricted but still real measure of autonomy by cultivating roças and by exploiting in other ways their “free” time. Yet, numbering no more than a handful, the same examples fall short of demonstrating the widespread and regular sale of surplus produce by slaves on the engenhos and cane farms of the Recôncavo.

The first example comes from the debt for the sum of Rs.37$680, owed by José Manoel, a poor resident of Santiago do Iguape and sometime cane farmer, to Teodósio da Silva, a pardo slave.63 When José Manoel died in 1824, Teodósio received permission from his mistress, José Manoeľs daughter, to submit the following bill:

Account of how much Senhor Manoel owes to Teodósio da Silva the following items, to wit:

Teodósio’s bill, as transcribed by the local vicar, is not altogether clear. How, for example, had Florêncio, José Manoeľs son, come to owe Teodósio Rs. 13$000? Given the inconsistent use of pronouns, the first item is frustratingly ambiguous: who planted cane with Gabriel José Manoel or Teodósio? If it was José Manoel, had Teodósio then somehow financed the planting? If, as would seem more likely, it was Teodósio, then his bill shows that Bahian slaves could use their free time not only to cultivate garden plots but also to earn money as hired fieldhands. Not only did Teodósio work as a fieldhand, but he also raised livestock and was something of a part-time peddler. That meant that he had connections with local merchants who could furnish him with wares. All in all, Teodósio’s bill shows that he was involved in a remarkable and unusually wide range of activities. The amount José Manoel owed his daughter’s slave, moreover, was substantial. With Rs.37$680, Teodósio could have bought one or two horses or the freedom of an infant slave. He also could have kept himself, a wife, and three children supplied with all the farinha they needed for a full year.64

In the same parish of Iguape where, three-and-a-half decades later, Teodósio would submit his bill, Dona Ana Joaquina de São José, as executrix of her late husband’s estate, assumed full responsibility for administering the Engenho Maroim between May and December 1790. During those months, she hired free carpenters to carry out repairs on the plantation’s sugar mill. The accounts Dona Ana Joaquina prepared and then attached to her husband’s inventory record show, among other expenses, the wages she paid the carpenters and the food she purchased for their meals. On at least one occasion, she bought palm oil and beans from her slave Benta. Although unfortunately the accounts do not record the quantity she purchased, the sum she paid Benta, Rs.4$000, indicates a fair amount of each. Rs.4$000 would have bought a third of an alqueire, or approximately 11 kilograms, of beans in Salvador; and Benta, in all likelihood, did not receive the price then current in Salvador for her beans and oil.65

A third and final example comes from the inventory of José Francisco das Neves, owner of the Engenho Genipapo in the township of São Francisco do Conde. As administrator of the engenho, the executor made the following entry in his accounts:

1842. 8 February. A portion of cassava bought from the slave Bento Rs.6$000

Bento had sold a respectable amount of cassava. His “portion” may have occupied a full tarefa of land (4,356 square meters), planted with more than four thousand covas (pits or hollows) of cassava. That number of covas would have easily yielded him enough farinha to feed himself and four other slaves for a year.66

Clearly, plantation slaves in the Recôncavo did have roças, and sometimes they even harvested enough from them to sell a surplus. Yet the region had nothing to compare with the markets and fairs of Jamaica, St.¯ Domingue, Antigua, Martinique, Barbados, Grenada, and some of the other Caribbean sugar islands that were frequented by hundreds and even thousands of rural slaves, and that for this reason repeatedly drew the attention of both local officials and contemporary observers. Those markets were often held on Sundays. Sunday markets, of course, conflicted with Christian religious observances in both British and French colonies, and, on religious grounds, local authorities sometimes unsuccessfully tried to have them transferred to Saturdays. But Sunday was the one day of the week that slaves generally had “free.” Whatever their religious scruples, authorities had to give in to the fact that on no other day could large numbers of rural slaves afford to go into towns and sell produce from their provision grounds.67 Plantation slaves in Bahia also generally had their Sundays “free.” Yet the issue of a conflict between religious observance and slave marketeering never arose in the Recôncavo. In the small towns of the region, Saturday, a working day for slaves, and not Sunday, was and still is the day for the main weekly feira.68

The purchases of beans and palm oil at the Engenho Maroim and cassava at the Engenho Genipapo appear paltry compared not only with the quantity and variety of goods handled by slave markets in parts of the Caribbean, but even with the more modest examples found by João Fragoso in probate records from the coffee-growing township of Paraíba do Sul in the province of Rio de Janeiro. At the Fazenda Caxambu, the administrator acquired from that plantation’s slaves 39 percent of all the foodstuffs he purchased between 1880 and 1882. In one six-month period alone, the slaves supplied 391 alqueires of maize, almost 29 bushels of beans, and 60 sacks of rice. The slaves’ share at another coffee plantation in the 1850s reached 60 percent of all the foodstuffs purchased.69

Planters and cane farmers in Bahia frequently did buy basic foodstuffs, but not from their slaves. At local markets in Salvador and elsewhere in the Recôncavo, they bought for their slaves not only jerked beef but in many cases cassava flour and other essential provisions as well. Contemporary observers from the late eighteenth century onward noted that buying farinha to feed slaves was a well-established practice among planters and cane farmers in the Recôncavo.70 Likewise, the same probate records that seldom do more than hint at the existence of provision grounds often record large and regular purchases of farinha. Table 1 lists 30 inventories of sugar planters and cane farmers in which unpaid debts or accounts prepared by executors indicate purchases of farinha. The examples come from nearly all the major sugar-producing parishes of the northern Recôncavo; they also cover almost the entire period 1780-1860. The accounts attached to these inventories, moreover, record in several cases purchases of farinha large enough to provide every slave on the plantation or cane farm with a full ration of flour. Here, then, the silences in the sources give way to clear, positive evidence that, in the late eighteenth and early nineteenth centuries, plantation slaves in the Recôncavo generally did not draw the bulk of their sustenance from their roças, and that Bahian planters and cane farmers relied primarily on purchased rations to provision their estates.

Thus, in the years 1780–1860, provision grounds in the Recôncavo served mainly to supplement rations in the slave diet. Independent production and marketing activities by plantation slaves appear to have developed far less in the region than in the some of the main sugar colonies of the West Indies.

Roças, Plantation Work Routines, and the Regional Economy

The contrast between the Bahian Recôncavo and other regions of slave- based sugar production can be explained partly by differences in work routines and labor demands. The agricultural calendar on sugar estates in Northeastern Brazil contained, as Gorender has noted, almost no slack time. Planting in Bahia took place in two rounds, the first lasting from the start of June until early September and the second from February to May. The harvest season, during which work on engenhos and cane farms went ahead at an unrelenting and almost feverish pace, overlapping that of planting. Harvesting began in August and continued until early May, when torrential winter rains interrupted the exhausting work of cutting and grinding cane and making sugar from the cane juice. Lasting somewhere between 270 and 300 days, the harvest season was three to six months longer in Bahia than in the West Indies and fully three times longer than in Louisiana.71

The tasks of planting, weeding, harvesting, and milling therefore occupied Bahian slaves during all but a few weeks in May, leaving them with little time to devote to their roças. It made sense for planters and cane farmers to purchase farinha to feed their slaves rather than to count on the cassava the slaves might grow on their own. Moreover, expanding overseas markets for Brazilian sugar after 1780, sometimes accompanied by higher prices, may have led planters and cane farmers in the Recôncavo to intensify their labor demands and to curtail even further the amount of “free” time they allowed their slaves. The result would have been both an increase in sugar output and a decrease in harvests from the roças, making the purchase of farinha all the more necessary.

These factors may help to explain a planter preference for buying rations, but they cannot fully account for the very limited development of independent slave production and marketing activities in the Recôncavo. Planters in Barbados, for example, also relied chiefly on rations, both estate grown and imported, to feed their slaves. On that densely settled island, plantation slaves had access only to small “house spots” that measured at most 15 square yards; and, in contrast with slaves elsewhere in the West Indies, they were not allowed special time off to cultivate their plots. Independent slave marketing nevertheless flourished in Barbados.72 Even in those West Indian sugar colonies where planters relied on provision grounds for the bulk of their slaves’ subsistence, the “free” time they allowed slaves to work their crops was minimal, totalling only 19 working days a year on some islands.73 Wherever plantation slaves cultivated provision grounds, they had to compete with their masters for their own labor time. As markets expanded and prices rose in the late eighteenth century, that competition no doubt intensified. Yet in the Caribbean, independent slave production expanded during those years. The average size of provision grounds in the British sugar colonies nearly doubled between 1750 and 1800.74

A full explanation must go beyond labor demands and internal work routines on the sugar plantations and take into account the broader regional economy. Sugar was Bahia’s chief export and the Becôncavo one of the birthplaces of plantation agriculture in the New World, but unlike many British and French colonies in the West Indies, neither Bahia nor even the Recôncavo could be described as simply a vast plantation. Provision grounds expanded in parts of the Caribbean in the late eighteenth century because planters there could not count on any other reliable supply of foodstuffs. With the exception of provision grounds and estate-grown grain, production for export almost completely dominated agriculture on most islands.75 As a result, British islands that depended on imported provisions suffered widespread starvation when the War of Independence disrupted shipping between mainland North America and the West Indies. Out of necessity, planters on those islands either began using estate labor for food production or allowed their slaves additional time and land to grow their own provisions.76

In Bahia, by contrast, plantation agriculture coexisted with the commercial and even slave-based production of essential foodstuffs. Small and middle-sized farmers in the Recôncavo and elsewhere in coastal Bahia, often with the help of one or two slaves, produced large surpluses of cassava flour and other provisions for sale.77 Bahian planters therefore could obtain generally adequate supplies of flour to feed their slaves, either directly from neighboring farmers or in local markets.

The farmers who produced farinha for plantation slaves sometimes actually lived on the sugar estates. As in the West Indies, many plantations in the Recôncavo contained large tracts of hilly or less-fertile land that could not be profitably used for sugarcane. Whereas on many Caribbean plantations slaves used those lands to develop extensive provision grounds, small farmers, either as tenants, squatters, or under some other arrangement, often occupied comparable areas on Bahian engenhos. In the 1830s, more than two hundred such farmers lived in the wealthy sugar-producing parish of Santiago do Iguape, where they grew cassava and other food crops. In some cases, no doubt, these were poor subsistence farmers whose crops yielded only enough food to meet their household needs. But more than one-tenth of Iguape’s small farmers owned at least one slave, which strongly suggests that they regularly produced a marketable surplus of cassava flour.78

Even stronger evidence for the commercial production of farinha comes from the townships of Maragogipe and Jaguaripe, located across the Bay of All Saints from Iguape and the other traditional sugar-producing parishes of the northern Recôncavo. In 1780, the more than seven hundred roceiros (cassava farmers) living in these two southern townships had more than 5.6 million covas of cassava in their fields. By a very conservative estimate, the cassava they had planted would have yielded them at least 110,000 bushels of farinha, or more than twice as much as they needed to feed themselves, their families, and their slaves. A less conservative but more reasonable estimate would place total output at 168,000 bushels and the marketable surplus at more than 123,000 bushels.79 Half a century later, in 1825-26, the port of Jaguaripe alone sent nearly 250,000 bushels of flour to Salvador.80 This figure does not include shipments of flour ferried by smacks and launches directly from Jaguaripe to the main sugar-producing parishes on the other side of the bay. Nor does it include supplies of farinha that reached Salvador’s market from other cassava-growing districts in coastal Bahia to the north and south of the Recôncavo.

Bahian roceiros could produce substantial surpluses of flour in large part because they often employed slave labor on their farms. Three- quarters of the cassava farmers who appear in a 1781 survey of Jaguaripe owned slaves.81 Similar evidence for other dates and areas of Bahia indicates the widespread use of slave labor in cassava production.82 Of course, the one or two slaves that cassava farmers typically owned could not compare with the great gangs of slaves that staffed wealthy engenhos. Yet for roceiros and their families, ownership of even a single slave might in some years make the difference between merely meeting household needs and producing a surplus of flour for sale.

Whether they owned slaves or not, the roceiros of the southern Recôncavo supplied local markets with more than just farinha. As a list of the goods entering the market at Nazaré during one week in 1823 shows, they also produced a wide range of other foodstuffs for sale. The village of Nazaré, located in the township of Jaguaripe and the site of a regular weekly market, had, by the 1780s if not earlier, become the most important center for the farinha trade in the Recôncavo.83 In the week of January 12- 19, 1823, alone, Nazaré’s market handled more than 6, roo bushels of flour as well as 25 bushels of beans and another 405 bushels of maize. The pack horses, mules, and donkeys that descended on Nazaré during that week also brought fresh produce for sale: more than 6,000 limes, 2,000 cashew fruits, 191 cargas, or packloads, of bananas, 117 cargas of watermelons, 96 loads of pineapples, and, in smaller quantities, palm kernels, pumpkins, cucumbers, okra, jackfruits, gherkins (maxixes), coconuts, and cabbages– all in addition to 11 packloads of items classified generically as verduras (green vegetables). The list of goods that farmers from the surrounding countryside sent to Nazaré’s market in that same week goes on to include woven mats, palm oil, wooden troughs, tapioca, homespun cotton cloth, soap, loo dozen eggs, 120 head of cattle, and an even larger number of live chickens, ducks, and turkeys.84

Because they could provision their estates with flour purchased from neighboring roceiros and in local markets like Nazaré, Bahian planters and cane farmers could afford to restrict the amount of “free” time they granted their slaves for work on their roças. Perhaps more important, commercial production of farinha and other foodstuffs by small and middle-sized farmers in Bahia deprived plantation slaves of potential marketing opportunities. Slaves on the Recôncavo’s engenhos and cane farms did cultivate roças that could yield surpluses for sale; but when and if they succeeded in taking their surplus produce into the towns and villages of the region, they would have found the markets there already well stocked and occupied by free farmers, who often had slaves of their own.85

Roças and Master-Slave Relationships

The examples drawn from probate records, combined with the repeated references of late colonial chroniclers, nineteenth-century travelers, and authors like Miguel Calmon and Manoel Ferreira da Câmara indicate that, in the years 1780 to 1860, allowing slaves to cultivate provision grounds was indeed a common practice on sugar plantations and cane farms in the Recôncavo. But the evidence also shows that factors both inside and, more important, outside plantation boundaries restricted rural slaves’ opportunities to sell the produce of their roças. Roças, as a rule, did not supply slaves with the bulk of their sustenance; and, in contrast with some of the main Caribbean sugar-producing areas, plantation slaves in the Recôncavo were generally unable to use their provision grounds to establish widespread and extensive independent marketing activities.

If it is relatively easy to establish that roças in the sugar districts of Bahia were generally neither essential for slave subsistence nor crucial in provisioning local markets, it is far more difficult to determine what access to a roça may have meant to the slaves or how provision grounds may have affected master-slave relations. The sources seldom directly address such matters; certainly the advantages or disadvantages for either side were never the same or even clear-cut.

On the relatively few engenhos and cane farms in the Recôncavo where owners did not distribute rations, cultivating a provision ground may have appeared to slaves as anything but a privilege. Slaves on those estates could not afford to rest after long days of exhausting labor spent in the cane fields and at the mill house; instead, before sleeping, they would have had to work additional hours in tending their roças merely to avoid hunger. But where slaves did receive rations, as was more common in the region, a roça might provide them not only with additional provisions, but also with a healthier and more varied diet. Planters and cane farmers in the Recôncavo typically issued to their slaves only cassava flour and jerked beef. Kenneth Kiple has recently shown that a diet based primarily on those two staples can result in endemic beriberi and increased infant mortality.86 The vegetables, maize, and beans grown in garden plots and the small livestock or poultry raised around slave quarters could then substantially improve the slaves’ diet and give their newborn children a far better chance of survival.

Although Bahian planters and cane farmers were perhaps unaware of the possible health benefits, like slaveowners elsewhere, they did recognize other advantages in allowing their slaves access to a provision ground. They hoped that, by occupying slaves when they were not at work on the staple crop, independent production even on a small scale might serve to reduce slave disobedience and the likelihood of open rebellion. Calmon put the matter clearly enough in 1834:

[A] master should give his slave some property or make it easier for him to acquire some on his own and should encourage his industry. This is a powerful means of distracting him from the ideas that go hand in hand with his sad condition and of inspiring in him a desire to work and even of inviting him to form a family. The practice of encouraging him to plant a roça, especially with food crops, of allowing him to have some livestock or to exercise some trade–this without doubt will lead to his happiness since it can modify the unruly tendencies that slavery generates and fosters. ... 87

Thus, for slaveowners, provision grounds could fulfill an “ideological function,” binding slaves closer to their masters and diverting their energies into “safe” activities and away from threatening, subversive behavior.88 Nowhere perhaps was the need greater to “modify the unruly tendencies” fostered by slavery than in Bahia in the early nineteenth century, where, in a spectacular series of uprisings, both rural and urban slaves rebelled against their masters.

Yet if provision grounds were to serve an “ideological function” in the day-to-day struggle between masters and slaves, planters and cane farmers had to make real, if not always tangible, concessions. Those concessions began with the autonomy, however small and constrained, that plantation slaves gained in cultivating their own plots. In contrast with the other types of work they performed, slaves, not masters or overseers, decided what crops they would plant in their roças and even how much time and energy they would put into their crops.

There were other concessions as well. Calmon notes that where a slave had access to a roça, “he” would be more likely to form a “family.” Along the same lines, Francisco Peixoto de Lacerda Werneck, a coffee planter in the province of Rio de Janeiro, remarked in the 1840s that, with income earned by selling produce grown on his roça, a slave could buy not only tobacco for himself, but also “better clothing for his wife and children.”89

Implicitly both Calmon and Lacerda Werneck admit that, in the slave community, male slaves could, in practice, exercise patriarchal authority over their wives and children, and that access to a provision ground could reinforce that authority. Moreover, even to speak of families among slaves meant to recognize relationships that had only the most precarious legal status. Not until 1869 would Brazilian law prevent the forced separation of slave families.90 Yet nearly everywhere, the “economy” that slaves created for themselves by cultivating provision grounds or garden plots was a family economy.91 Work in a common “family” roça might even serve to strengthen kinship ties across generations. Just as in parts of the West Indies, slaves at the Engenho Vitória in Iguape were “allowed to name those to whom their property is to be given when they die.”92

Allowing slaves to transfer “property” among themselves represented a further concession, since no law in Brazil before 1871 guaranteed a slave’s right to a peculium; Brazilian slaves before that date could not legally own anything.93 Significantly, Calmon in his discussion of provision grounds speaks of slaves’ holding and acquiring “property.” He acknowledges, in other words, that slaves had won “rights,” if not by law then by custom, over the plots they cultivated. Bahian slaves transferred “property” in ways other than inheritance and not merely among themselves. On Sundays and holy days, slaves at the Fazenda da Capela de Santa Maria and the Engenho Santo Estêvão sold their labor for wages. Benta at the Engenho Maroim sold not her labor but the beans she had harvested in her roça and the oil she had made from palms planted there. Likewise, Bento received cash for his “portion” of cassava at the Engenho Genipapo, while Teodósio peddled blades and cloth goods and sold livestock.

As John Campbell has noted in his study of antebellum South Carolina, by holding, selling, and buying “property” of any type, slaves “challenged … one of the central premises of their enslavement”; namely, that they themselves were property to be held, bought, and sold.94 In Bahia, by challenging that central premise, slaves at times even succeeded in transcending their status as slaves. Four slaves at the Engenho Vitoria in the 1850s used what they earned from the sale of provision ground produce to buy their freedom.95 Examples such as these show that, like slaves elsewhere in the Americas, plantation slaves in the Recôncavo found in their “free” time and their roças a circumscribed but nonetheless significant “economy” of their own. Taking note of that economy does not imply any conservative effort to “rehabilitate slavery” or to restore outdated notions about the mildness of slavery in Brazil; on the contrary, it reveals one of the many “shifting terrains” where the day-to-day struggle between masters and slaves played itself out.96 Although factors inside and outside plantation boundaries restricted the development of the economy Bahian slaves built for themselves, they were slaves who, within the limits imposed by slavery, “negotiated.”97 They “negotiated” with their masters and even with slavery itself.

Appendix: A Note on Sources

Probate Records

This study relies heavily on research with the collections of probate records, the Inventários e Testamentos, held at the Arquivo Público do Estado da Bahia in its Seção Judiciária and at the Arquivo Regional da Cachoeira, Bahia. At the two archives, the author systematically worked with all the postmortem inventories dated between 1780 and 1860 from the townships of Santo Amaro and São Francisco do Conde, and the parish of Santiago do Iguape in the township of Cachoeira, that were available for consultation during visits from 1986 to 1988.

These three locales encompassed the most important sugar-producing districts in the Recôncavo. The only well-established sugar areas not covered by the sample were the bayshore parishes in suburban Salvador.

I was further able to locate, among the probate records for the city of Salvador (housed in APEB under the heading Capital), a few inventories for sugar planters who owned properties in São Francisco do Conde, Santo Amaro, and Iguape, but who at the time of their death resided in Salvador.

After completing my survey of these collections, I eliminated inventories that were insufficient for the purpose of this study or that did not represent sugar producers–that is, planters or cane farmers. The final sample consisted of 162 postmortem inventories. I also examined another 322 inventories filed between 1780 and 1860 in Recôncavo parishes and townships where some plantations were located, even though sugarcane was not the main crop. Inventories from these districts reinforced conclusions based on the main sample.

Probate records not only list the property owned by planters and cane farmers, but also include information on unpaid debts and occasional sets of running accounts of expenditures and receipts prepared by executors. These sets of accounts vary greatly in the amount of detail they contain and the time span they cover. I located more than 60 sets of accounts attached to Bahian inventories (including accounts for rural estates not involved in sugar production). In Brazil, where plantation records are extremely rare, such accounts, together with inventories, are crucial sources of firsthand information about day-to-day operations on plantations and farms. In this regard, see Robert Sienes, “Escravos, cartórios, e desburocratização: o que Rui Barbosa não queimou será destruído agora?” Revista Brasileira de História 5:10 (Mar. 1985), 166-96, esp. 178-80.

Provincial and Municipal Records and Legislation

The study also draws on correspondence sent between 1780 and 1860 by various local officials in Salvador and nearby Recôncavo townships to the governors of the Captaincy of Bahia and, after 1822, to the provincial government. In APEB, Seção Histórica, I examined all the catalogued correspondence sent by the city council of Salvador and by the town councils of São Francisco do Conde, Santo Amaro, Cachoeira, Maragogipe, Nazaré, and Jaguaripe–the main townships in the Recôncavo. In addition, I consulted all the available correspondence from judges, farinha inspectors, and captains-major of local militia units in the same six townships, as well as the entire collections of correspondence from the administration of Salvador’s Celeiro Público (Public Granary), which, between 1785 and 1853, was the city’s only authorized market for cassava flour, beans, maize, and rice.

I was allowed to work with the entire set of materials from the late colonial period that, at the time, were unclassified and in the process of being recatalogued. At the same archive, I examined similar correspondence sent by local officials from other townships in Bahia and Sergipe that, although located outside the Recôncavo, regularly supplied foodstuffs for sale in Salvador’s market. Comprised in more than no maços (bundles) of manuscript materials, this documentation deals repeatedly with the supply and marketing of basic provisions and with local disturbances in both rural and urban areas, including episodes related to the marketing of basic foodstuffs or involving slaves.

The smaller collection of similar materials dealing with the supply and marketing of foodstuffs in Bahia that is held in the Seção de Manuscritos of the Biblioteca Nacional, Rio de Janeiro, was another source of data. At the Arquivo Regional da Cachoeira, I consulted the entire collection of papéis avulsos (loose, uncatalogued nineteenth-century materials), some of which dealt with various municipal concerns. (I did not, however, examine the documents that were simply heaped in piles in the archive’s back room, which included materials from the eighteenth, nineteenth, and twentieth centuries.)

At the Arquivo Municipal do Salvador, I examined the livros de posturas municipals (books of municipal ordinances) for 1650-1787 and 1829- 1859, which the archive has catalogued as 119.1 and 119.5, and the three surviving volumes dealing with trade in cassava flour in the city’s market: “Entradas de farinha, 1861-1864” (55.1), “Entradas de farinha, 1865” (55.2), and “Fiel do Celeiro Municipal e lançamento de farinha” (63.1). Other relevant posturas were published together with other provincial laws in Collecção das leis e resoluções da Assembléa Legislativa da Bahia … , 33 vols. (Bahia, 1862-89). Because posturas had to be approved by the provincial assembly, municipal councils sent proposals to the provincial president.

Apparently no municipal or provincial law regulating the sale of produce by rural slaves was passed after 1860. Further research with earlier colonial records might uncover ordinances of this sort for the Recôncavo. But here it should be noted that a number of town councils completely revised their posturas in the nineteenth century without including such laws. It is also possible that municipal authorities dealt with the matter through orders to the local police rather than through legislation.

Notarial Records

Notarial records of manumissions might also be expected to provide information on independent marketing by slaves. Research at the Arquivo Regional da Cachoeira examined 57 manumission letters recorded by the notary for the parish of Santiago do Iguape (Livro de notas de escrituras: Santiago do Iguape, 1831-1845). Unfortunately, none offers such information. Twenty of the letters explicitly require the slave to pay for freedom either immediately or at some future date, but none specifies where the slave would obtain the money for this purpose; the only contracts that refer to this matter are those in which a third party has contributed the funds.

Rural Land Records

The search for mention of slave roças led to the registros eclesiásticos de terras (ecclesiastical land registries) from the 1850s in APEB, SH, 4661, 4697, 4701, 4712, 4739, 4748, 4758, 4795, 4801, and 4815. These property registrations, filed by landowners to comply with the requirements of the 1850 Land Law, vary greatly in quality and in volume of information. The most elaborate of them describe in great detail a property’s boundaries and topographical features; some refer to the types of crops grown. But unfortunately, more than 1,800 registrations from 13 rural parishes (including the 6 traditional sugar-producing parishes of the northern Recôncavo) yielded no references to slave provision grounds. Future research might pursue the sesmarias, the official land grants that the crown distributed until 1821, such as that given to Pedro Rodrigues Bandeira.

Travelers’ Accounts

Finally, I consulted published works in English, French, German, and Dutch, as well as in Portuguese translations, by 38 European and North American travelers who visited Bahia in the nineteenth century. Mary C. Karasch has demonstrated that nineteenth-century travelers’ accounts can be an invaluable source of information about slaves and slavery in her Slave Life in Rio de Janeiro, 1808-1850 (Princeton: Princeton University Press, 1987). For Bahia, the nineteenth-century travel literature includes detailed descriptions of Salvador’s market and, less frequently, of market fairs in the towns of the Recôncavo. For a useful survey of the travel literature available for Bahia, see Moema Parente Augei, Visitantes estrangeiros na Bahia oitocentista (São Paulo: Editora Cultrix, 1980).

Urban Slaves

In focusing on slaves on sugar plantations and cane farms, this article sets aside independent production or marketing activities developed by Bahia’s urban slaves–quitandeiras, ganhadores, and ganhadeiras. The secondary literature contains scattered references to such activities by slaves in the city of Salvador. See Maria José de Souza Andrade, A mão de obra escrava em Salvador, 1811-1860 (São Paulo: Corrupio, 1988), 132-34; Maria Inês Côrtes de Oliveira, O liberto: O seu mundo e os outros, Salvador, 1790- 1860 (São Paulo: Corrupio, 1988), 18-20; João José Reis, “Slave Rebellion in Brazil: The African Muslim Rising in Bahia, 1835” (Ph.D. diss., Univ. of Minnesota, 1985), 211-25; Kátia M. de Queirós Mattoso, Bahia, século XIX: urna provincia no império (Rio de Janeiro: Editora Nova Fronteira, 1992), 436-37, 541-42. See also Jean Chrétien Baud, “Herinneringen,” in De Reis van Z. M. “De Vlieg,” Commandant Willem Kreekel naar Brazilië, 1807, ed. H. J. de Graaf, 2 vols. (The Hague: Martinus Nijhoff, 1975), 1:196-98; Edward Wilberforce, Brazil Viewed Through a Naval Glass: With Notes on Slavery and the Slave Trade (London: Longman, Brown, Green, 1856), 91-92; James Wetherell, Brazil. Stray Notes from Bahia: Being Extracts from Letters, & c., during a Residence of Fifteen Years, ed. William Hadfield (Liverpool: Webb and Hunt, 1860), 29-30; and Vilhena, A Bahia, 1:93, 126-27, 129-30.

On similar activities by slaves in Rio de Janeiro, see Karasch, Slave Life, 185-213; Leila Mezan Algranti, O feitor ausente: estudo sobre a escravidão urbana no Rio de Janeiro (Petrópolis: Editora Vozes, 1988), 65-72; and Marilene Rosa Nogueira da Silva, Negro na rua: a nova face da escravidão (São Paulo: Editora Hucitec, 1988). Likewise, except in passing, this article does not deal with the independent economic activities of rural slaves on tobacco and cassava farms or other non–sugar-producing agricultural establishments in the Recôncavo.

Research for this article was made possible in part by fellowships from the Fulbright Commission and the Social Sciences Research Council. The author wishes to thank Joseph Love, Nils Jacobsen, Ciro Cardoso, Laura Tabili, John Campbell, Suzanne Wilson, Judith Allen, Elizabeth Kuznesof, and Mark D. Szuchman for commenting on earlier versions of this article.

The research utilized the following archives: Arquivo Municipal do Salvador (AMS); Arquivo Público do Estado da Bahia, Salvador, Seção Judiciária, Inventários e Testamentos (APEB, SJ, IT), and Seção Histórica (SH); Arquivo Regional da Cachoeira, Bahia (ARC); Biblioteca Nacional, Rio de Janeiro, Seção de Manuscritos (BN, SM). The references also abbreviate the following: Santo Amaro (SA), São Francisco do Conde (SFC), and Cachoeira (Cach.).

1

See, e.g., Ira Berlin and Philip D. Morgan, eds., “The Slaves’ Economy: Independent Production by Slaves in the Americas,” special issue of Slavery and Abolition, 12:1 (May 1991), esp. the editors’ introduction, 1-27, which surveys recent literature. See also B. W. Higman, Slave Populations of the British Caribbean, 1807-1834 (Baltimore: Johns Hopkins Univ. Press, 1984), 209–12; Gabriel Debien, “La nourriture des esclaves sur les plantations des Antilles françaises aux XVIIe et XVIIIe siècles,” Caribbean Studies 4:2 (July 1965.3-27; Frantz Tardo-Dino, Le collier de servitude: la condition sanitaire des esclaves aux Antilles françaises du XVIIe au XIXe siècle (Paris: Editions Caribéenes, 1985), 134- 36; Dale W. Tomich, Slavery in the Circuit of Sugar: Martinique and the World Economy, 1830-1848 (Baltimore: Johns Hopkins Univ. Press, 1990), chap. 8; and Ciro Flamarion S. Cardoso, Escravo ou componês? o protocampesinato negro nas Américas (São Paulo: Editora Brasiliense, 1987), 54-87. Sidney W. Mintz and Douglas G. Hall were among the first scholars to investigate provision grounds and independent slave marketing in The Origins of the Jamaican Internal Marketing System, Yale University Publications in Anthropology 57 (New Haven: Yale Univ. Press, 1960). Mintz has continued to explore these topics. See, e.g, his “Caribbean Marketplaces and Caribbean History,” Nova Americana 1 (1980-81), 333–44, Caribbean Transformations (Baltimore: Johns Hopkins Univ. Press, 1974), chap. 7, and “Slavery and the Rise of Peasantries,” Historical Reflections/Réflexions historiques 6 (1979), 213-42. At present, the secondary literature discusses independent production and marketing activities by plantation slaves in various parts of the southern United States and nearly all the major British and French sugar-producing colonies in the West Indies. The literature on the Spanish Caribbean is less extensive, but see Rebecca Scott, Slave Emancipation in Cuba: The Transition to Free Labor, 1860-1899 (Princeton: Princeton Univ. Press, 1985), 14-16, 149-50, 259; and Laird W. Bergad, Cuban Rural Society in the Nineteenth Century: The Social and Economic History of Monoculture in Matanzas (Princeton: Princeton Univ. Press, 1990), 79-81.

2

In the debate over provision grounds, what I call the first perspective is perhaps best represented by the writings of Ciro Flamarion S. Cardoso; e.g., Agricultura, escravidão, e capitalismo (Peŧrópolis: Editora Vozes, 1979), chap. 4, Escravo ou camponês, and “The Peasant Breach in the Slave System: New Developments in Brazil,” Luso-Brazilian Review 25:1 (Summer 1988), 49-57. Jacob Gorender has most forcefully advocated the second perspective in O escravismo colonial, 3d ed. (São Paulo: Editora Atica, 1980), 258-67, “Questionamentos sobre a teoría econômica do escravismo colonial,” Estudos econômicos 13:1 (Jan.-Apr. 1983), 7-39, esp. 17-35, and A escravidão reabilitada (São Paulo: Editora Atica, 1990), 70- 81. Other works in this debate include Stuart B. Schwartz, “Resistance and Accommodation in Eighteenth-Century Brazil: The Slaves’ View of Slavery,” HAHR 57:1 (Feb. 1977), 69– 81; idem, Slaves, Peasants, and Rebels: Reconsidering Brazilian Slavery (Urbana: Univ. of Illinois Press, 1992), 49-55, 83-84; Antônio Barros de Castro, “A economía política, o capitalismo e a escravidão,” in Modos de produção e realidade brasileira, ed. José Roberto do Amaral Lapa (Petrópolis: Editora Vozes, 1980), 67-107, esp. 94-101; Mário José Maestri Filho, A servidão negra (Porto Alegre: Mercado Aberto, 1988), 78-80; João José Reis and Eduardo Silva, Negociação e confliŧo: a resistência negra no Brasil escravista (São Paulo: Companhia das Letras, 1989), chaps. 1-2; Silvia Hunold Lara, “Trabalhadores escravos,” Trabalhadores ([June?] 1979), 4–20. For studies based on original research that deal with the issues in this debate, see note 23.

3

This essay uses the words provision ground, garden, and garden plot as equivalents for the Portuguese term roça; for the sake of clarity, roça is used only in reference to slave provision grounds. Although in some English-speaking parts of the Caribbean contemporaries distinguished between gardens, yards, and provision grounds, no comparable distinction appears in Brazilian sources. The word roça also meant a cleared field that had been planted with cassava or other food crops; by extension, it was commonly used to refer to farms that produced cassava flour and other basic foodstuffs for sale in local markets. See, e.g., “Rellação dos moradores do Rio Cahippe, Estiva e roças respectivas que usão da planta de mandiocas …”, 1781, BN, SM, I-31,30,52; João da Costa Carneiro de Oliveira, “Breve compendio … sobre a villa de Jaguaripe e estado actual de plantação de mandioca nas Rossas de Nazaré … ,” 1799, Anais da Biblioteca Nacional 16 (1914), 181-82; and Friedr[ich] Asschenfeldt, Memoiren aus meinem Tagenbuche, geführt während meiner Reisen und meines Aufenthaltes in Brasilien in den Jahren 1847 bis 1847 (Oldenburg: C. Fränkel, 1848), 56. Likewise, Bahians in the late eighteenth and early nineteenth centuries also used roça to refer to suburban properties, which in other areas of Brazil might have been called chácaras. See, e.g., Lista das roças numeradas, e seus senhorios, que existem na 2a divizão da Freguezia de Santo Antonio Além do Carmo desta Cide da Ba neste prez6 anno de 1804 … ,” BN, SM, I-31,20,39. One of the first references to chácaras in Bahia that I located is W[illia]m Gore Ousely, Descriptions of Views in South America, From Original Drawings Made in Brazil, the River Plate, the Parana, &·c., & c. with Notes (London: Thomas McLean, 1852), 13. But even in the mid- and late 1850s, local documentation continued to refer to such properties as roças. See, e.g., Anna Amélia Vieira Nascimento, Dez freguesias da cidade do Salvador (Salvador: Fundação Cultural do Estado da Bahia, 1986), 39. Thus it is unclear whether, in the mid-nineteenth century, Bahians commonly used the term chácara, which is of Andean origin.

4

Gorender, A escravidão reabilitada, 14-17, 70-81. He uses the term neopatriarchalist because he regards much of the recent literature on slavery in Brazil as representing a “return” to interpretations that, beginning with Gilberto Freyre’s work in the 1930s, stressed the patriarchal and comparatively benign nature of Brazilian slavery. See, e.g., Freyre, Casagrande e senzala: formação da família brasileira sob o regime patriarcal, 22d ed. (Rio de Janeiro: Livraria José Olympio, 1983).

5

The terms rural slaves and plantation slaves in the text should be understood as referring to slaves on sugar plantations and cane farms. For sources on urban slaves, see  appendix.

6

For examples of how significant silences may be effectively used as evidence, see M. I. Finley, Economy and Society in Ancient Greece, ed. Brent D. Shaw and Richard P. Sailer (Harmondsworth: Penguin, 1983), 206, 214–15, 225-30.

7

Kátia M. de Ģueirós Mattoso, Bahia. a cidade do Salvador e sen mercado no século XIX (São Paulo: Editora Hucitec, 1978), 29-59; Stuart B. Schwartz, Sugar Plantations in the Formation of Brazilian Society: Bahia, 1550-1835 (Cambridge: Cambridge Univ. Press, 1985)> chap. 4; B. J. Barickman, “The Slave Economy of Nineteenth-Century Bahia: Export Agriculture and Local Market in the Recôncavo, 1780-1860” (Ph.D. diss., Univ. of Illinois, 1991), 21-41; Milton Santos, A rêde urbana do Recôncavo (Salvador: Imprensa Oficial do Estado, 1959), 3, 13-15; Dauril Alden, “Late Colonial Brazil, 1750-1808,” in The Cambridge History of Latin America, ed. Leslie Bethell, 8 vols. (Cambridge: Cambridge Univ. Press, 1984-91), 2:606.

8

Thomas Lindley, Narrative of a Voyage to Brazil; … With General Sketches of the Country, Its Natural Productions, Colonial Inhabitants, & c. and a Description of the City and Provinces of St. Salvadore and Porto Seguro . . . (London: J. Johnson, 1805), 103-5, 261-62.

9

Schwartz, Sugar Plantations, chaps. 1-3; Vera Lúcia Amaral Ferlini, Terra, trabalho, e poder: o mundo dos engenhos no Nordeste colonial (São Paulo: Editora Brasiliense, 1988), 161.

10.

“Quadro dos engenhos das vizinhanças da Bahia, com os nomes dos possuidores” [ca. 1790–95], BN, SM, 7,3,27. See also José da Silva Lisboa, “Descripção da cultura da Capitania da Bahia,” 1799, Anais da Biblioteca Nacional 36 (1914), 122-23; and Luís dos Santos Vilhena, A Bahia no século XVIII, 2d ed., 3 vols. (Salvador: Editora Itapuã, 1969), 2:479-81, 483-84.

11

Information on the number of cane farmers is scant. Working with an 1816-17 survey of slaveholders for the townships of Santo Amaro and São Francisco do Conde in the northern Recôncavo, Schwartz found an average of three to four slaveowning cane farmers perengenho. Sugar Plantations, 451.

12

For exports of white and muscavado sugar, see José Jobson de A. Arruda, O Brasil no comercio colonial (São Paulo: Editora Atica, 1980), 360-61, 375-76.

13

Barickman, “Slave Economy,” 52-57.

14

Schwartz, Sugar Plantations, 177-201.

15

Barickman, Slave Economy, 85–98 (which also includes a discussion of sugar prices).

16

“Matricula dos engenhos da Capitania da Bahia pelos dizimos reais administrados pela Junta da Real Fazenda,” 1807–1874, APEB, SH, 632. After 1842, planters were no longer legally required to register new engenhos. An official survey from 1873 lists 635 engenhos in the Recôncavo. “Trabalho da Commissão [da Bahia],” table B, in Informações sobre o estado da lavoura (Rio de Janeiro: Typografia Nacional, 1874), n.p. Because relatively few new plantations apparently were established after 1860, the survey may be taken as a fairly accurate guide to the number of engenhos operating in the mid-nineteenth century.

17

David Eltis, Economic Growth and the Ending of the Transatlantic Slave Trade (New York: Oxford Univ. Press, 1987), 243-44· Eltis’ estimates suggest that of this total, three-fourths arrived in the 4 ½ decades between 1785 and 1831, and slightly more than one hundred thousand between 1831 and 1852. The fluctuations in volume shown in Eltis estimates generally correspond closely to trends in the Bahian export economy, and particularly to trends in the production and export of Bahian sugar. On those trends, see Barickman, “Slave Economy,” 42-116. .

18

On the high proportion of African-born slaves on Bahian engenhos and cane tarms and the large size of Becôncavo sugar estates, see Barickman, “Slave Economy,” chap. 5; Schwartz, Sugar Plantations, chaps. 13 and 16; João José Reis, População e rebelião: noţas sobre a população escrava na Bahia na primeira metade do século XIX, Revista de Ciências Humanas 1:1 (1980), 143-54.

19

Inventories of Ana Rita Evaristo Duque de Estrada e Menezes, Iguape, 1793, ARC., IT; Félix Alves de Andrade, Cach., 1791, APEB, SJ, IT; Manoel Francisco Barreto, Cach., 1785, ARC, IT; Alistamento das pessoas que habitão dés do Engenho da Crûz athê o Engenho Novo [1825 or 1826], ARC, Papéis Avulsos; and “Relação do numero de fogos, e moradores do Districto da Freguezia de Sant-Iago Maior do Iguape, da Comarca da Villa da Cachoeira da Provincia da Bahia,” 1835, APEB, SH, 6175-1. On Bandeira, see Catherine Lugar, “The Merchant Community of Salvador, Bahia, 1780-1830” (Ph.D. diss., State Univ. of New York at Stony Brook, 1980), 235-37; and John Norman Kennedy, “Bahian Elites, 1750–1822,” HAHR 53:3 (Aug. 1973), 420-21. In 1814, Bandeira received from the crown a sesmaria (official land grant) confirming his ownership of Vitória and other nearby properties. Brazil, Ministério da Justiça, Arquivo Nacional, Repertório das sesmarias da Rahia (Rio de Janeiro; Seção dos Ministérios, 1968), 19.

20

See Barickman, “Slave Economy,” 123-32.

21

On making farinha, see Barickman, “Slave Economy,” 459-62; Carlos Borges Schmidt, L·avoura caiçara, Documentário da vida rural, 14 (Rio de Janeiro; Ministério de Agricultura, 1958), chap. 4; and [Manuel] Pinto de Aguiar, Mandioca–pão do Brasil (Rio de Janeiro: Civilização Brasileira, 1982), 133-36. For instances in other areas where plantation slaves grew cassava and made flour for their own use or for sale, see Berlin and Morgan, Introduction, 9; Woodville K. Marshall, “Provision Ground and Plantation Labour in Four Windward Islands: Competition for Resources During Slavery,” Slavery and Abolition 12:1 (May 1991), 55; Cardoso, Escraυo ou camponês, 79; Higman, Slave Populations, 212; Tomich, Slavery in the Circuit of Sugar, 271, 274.

22

Manuel Diégues Júnior, População e açúcar no Nordeste (Rio de Janeiro: Comissão de Alimentação, 1954), 70; Tomich, Slavery in the Circuit of Sugar, 262.

23

Maria Regina Mendonça Furtado Mattos, “Vila do Príncipe, 1850-1890: Sertão do Seridó. Um estudo de caso da pobreza” (Masters thesis, Univ. Federal Fluminense, 1985), 124-27, 139–41, João Luís Ribeiro Fragoso, “Sistemas agrários em Paraíba do Sul (1850- 1920): um estudo das relações não capitalistas de produção” (Master’s thesis, Univ. Federal do Rio de Janeiro, 1983), 67-69; idem, and Manolo Garcia Florentino, “Marcelino, fllho de Inocencia Crioula, neto de Joana Cabinda: um estudo sobre as famílias escravas em Paraíba do Sul (1835-1872),” Estados Econômicos 17:2 (May-Aug. 1987), 166-71; Silvia Hunold Lara, Campos da violência: escraυos e senhores na capitania do Rio de Janeiro, 1750–1808 (Rio de Janeiro: Paz e Terra, 1988), 208-20; Diana Soares de Galliza, O declínio da escravidão na Paraíba, 1850-1888 (João Pessoa: Editora Universitária/UFPb, 1979), 148-50; Eurípedes Antônio Funes, “Goiás 1800-1850: um período de transição da mineração à agropecuária” (Master’s thesis, Univ. Federal Fluminense, 1983), 122-25. See also Maria Helena P. T. Machado, Crime e escravidão: trabalho, luta, e resistência nas lavouras paulistas, 1830-1888 (São Paulo: Editora Brasiliense, 1987), 117-23.

24

Berlin and Morgan, Introduction, 5-6; Higman, Slave Populations, 204-14; Cardoso, Escravo ou camponês, 79; Kenneth F. Kiple, The Caribbean Slave: A Biological History (Cambridge: Cambridge Univ. Press, 1984), 67.

25

Gorender, O escravismo colonial, 259; idem, A escravidão reahilitada, 72-74.

26

José da Silva Lisboa, “Carta muito interessante … para o Dr. Domingos Van- delli … era que lhe dá noticia desenvolvida sobre a capitania da Bahia,” 1781, Anais da Biblioteca Nacional 32 (1910), 501; [Luís Antônio de Oliveira Mendes], “Discurso preliminar, historico, introductivo com natureza de descripção económica da Comarca e Cidade da Bahia …” [ca. 1790], Anais da Biblioteca Nacional 27 (1905), 321; Vilhena, A Bahia, 1; 185- 86; Manoel Ferreira da Câmara Bittencourt e Sá, Memoria ojferecída á Sociedade de Agricultura, Commercio, e Industria da Bahia (“Memoria sobre a possibilidade de plantarem os lavradores de canna todo, ou grande parte do mantimento de que precisão, sem quebra no producto de canna; e meis [sic; i.e., meios] de conseguir.”) (Bahia: Typographia do V. B. Moreira, 1834), 6-7; Miguel Calmon du Pin e Almeida, Ensato sobre o fabrico do assucar, offerecido á Sociedade d’Agricultura, Commercio, e Industria da Provincia da Bahia (Bahia: Typographia do Diario, 1834), 60. Vilhena wrote: “… tais [senhores de engenho] há que … Ihes facultam somente o trabalharem no domingo, ou dia santo, em um pedacinho de terra a que chamam roça . . .” (p. 185, emphasis in original).

27

Inv. of Manoel Estanislau de Almeida, Iguape, 1838, ARC, IT. The Fazenda da Capela de Santa Maria was not, despite its name, owned or administered by the church.

28

Inv. of Maria de Assunção Freire de Carvalho, SFC, 1848, APEB, SJ, IT. Argolo had the accounts for his engenho attached to this inventory to help settle a dispute among Carvalho’s heirs.

29

For prices for farinha, fresh beef, and chickens in 1855, see Kátia M. de Queirós Mattoso, “Au nouveau monde, une province d’un nouvel empire: Bahia au XIXe siècle” (Thèse de doctorat ďétat, Univ. de Paris-Sorbonne, 1986), annexes, 445-61. For land prices, see the survey of engenhos in the parish of Nossa Senhora do Socorro (where Santo Estêvão was located), 1854, APEB, SH, 4597.

30

John Candler and Wilson Burgess, Narrative of a Recent Visit to Brazil, to Present an Address on the Slave-Trade and Slavery Issued by the Religious Society of Friends (London: Edward Marsh, 1853), 57.

31

“Survey” of the engenhos in the parish of Nossa Senhora da Piedade de Matoim, 1854, APEB, SH, 4597.

32

Fragoso and Florentino, “Marcelino, filho de Inocencia Crioula, 171; Manuela Carneiro da Cunha, Sobre os silêncios da lei. Lei costumeira e positiva nas alforrias de escraυos no Brasil do sécalo XIX, Cadernos IFCH Unicamp, 4 (Campinas: Univ. de Campinas, l983), 5 and passim (on positive law that denied slaves the right to own anything before late 1871 and on widespread customary practices that, by contrast, recognized a slave’s right to a peculium). Fragoso and Florentino do not imply that the absence of evidence in probate records proves the existence of provision grounds. Rather, they argue that the lack of references indicates that neither appraisers nor heirs insisted on asserting the dead slaveowner’s legal rights to the crops from roças. Along the same lines, an inventory of the Engenho Vitória from early 1871 (i.e., before the passage of the “Law of the Free Womb”) records no roças. On that estate, according to Candler and Burgess, the practice was to allow slaves to cultivate provision grounds. See Inv. of Egas Munis Barreto de Aragão, Iguape, 1871, ARC, IT; and Candler and Burgess, Narrative. (Egas Munis already administered Vitória on his father’s behalf at the time of Candler and Burgess’ visit.) The same pattern holds for the examples of Benta at the Engenho Maroim and Bento at the Engenho Genipapo mentioned in the text. For general information on customary rights to land in nineteenth-century Brazil, see Nancy Naro, “Customary Rightholders and Legal Claimants to Land in Rio de Janeiro, Brazil, 1870-1890,” The Americas 48:4 (Apr. 1992), 485-517.

33

Inv. of Antônio de Souza Ribeiro, Maragogipe, 1857, APEB, SJ, IT.

34

Sophie-Élie-Alexandre, Baron Forth-Rouen, “Bahia en 1847,” in Mélanges américains, by Henri Cordier (Paris: Jean Maisonneuve et Fils, 1913), 121.

35

Ibid., 120-21.

36

In Pernambuco in the early nineteenth century, Henry Koster observed, “many … [plantation slaves] rear pigs and poultry, and occasionally a horse is kept, from the hire of which money may be obtained.” Koster, Travels in Brazil (London: Longman, Hurst, Rees, et al., 1816), 433 (emphasis added).

37

Cardoso, Escravo ou camponês, 95 and passim; Berlin and Morgan, Introduction, 15; Mintz, Caribbean Transformations, 189. There is no reason to believe that provision grounds were less common on the tobacco farms of the Recôncavo than on the sugar plantations and cane farms, but evidence is still scarce; but see Inv. of Vicente José de Cerqueira Pinto, Cach., 1829, APEB, SJ, IT.

38

Lisboa, “Carta,” 501.

39

Ronaldo Vainfas, Ideologia e escravidão: os letrados e a sociedade escravista no Brasil colonial (Petrópolis: Editora Vozes, 1986), 109; Schwartz, Sugar Plantations, 1.37-38·

40

[Mendes], “Discurso,” 89–90; Vilhena, A Bahia, 1:185–86.

41

Câmara, Memoria, 6-7. Câmara was also condemning those slaveowners who, to save money, cut back on rations.

42

Calmon, Ensaio, 59-60 (emphasis in the original).

43

André João Antonii [João Antônio Andreoni], Cultura e opulencia do Brasil por suas drogas e minas, texte de l’ édition de 1711, trans. Andrée Mansuy (Paris: Institut des Hautes Etudes de 1 Amérique Latine, 1968), 126–28; Jorge Benci, Economía cristã dos Senhores no gaverno dos escravos: livro brasileiro de 1700 (1705), with preliminary study by Pedro Alcântara Figueira e Claudinei M. M. Mendes (Rio de Janeiro: Editorial Grijalbo, 1977), 57-64. On these reformist writers, see Vainfas, Ideología e escravidão, 107-10, 149-59; and Cardoso, Escravo ou componês, 94-95, 121.

44

Mintz, Caribbean Transformations, 194-206; idem, Slavery and the Rise of Peasantries,” 226-27, 230-31; A. J. G. Knox, “Opportunities and Opposition; The Rise of Jamaicas Black Peasantry and the Nature of the Planter Resistance,” Canadian Review of Sociology and Anthropology/La Revue Canadienne de Sociologie et ďAnthropologie 14:4 (Nov. 1977), 386; Higman, Slave Populations, 213-14; Cardoso, Escravo ou compones, 77-79, 88, 108; Tomich, Slavery in the Circuit of Sugar, 274; Marshall, “Provision Ground, 56; Funes, “Goiás,” 122-25.

45

Schwartz, Slaves, Peasants, and Rebels, 83.

46

Schwartz brought to light the incident at Engenho Santana and published the relevant documents in “Resistance and Accommodation. The summary here closely follows that which Schwartz provides in his Sugar Plantations, 159. Most of the authors engaged in this debate (see note 2) discuss this incident and the extent to which it allows for generalization.

47

In still another clause, the slaves demanded that Ferreira give them “casting nets and canoes ... to enable [them] ... to live.” This clause reminds us that, for slaves living near the sea or rivers, the right to fish could be just as important as the right to farm.

48

Schwartz, Sugar Plantations, 159, and “Resistance and Accommodation.”

49

João José Reis, “Resistencia escrava em Ilhéus: um documento inédito,” Anais do Arquivo Público da Bahia 44 (1979), 286-88; Schwartz, Slaves, Peasants, and Rebels, 51– 54. Gorender has insisted that the practice at Santana of allowing slaves to cultivate a roça and permitting them to sell produce should be attributed to the paternalism of the previous Jesuit administration. He therefore argues that the peace treaty cannot be used to reach any broader conclusions about independent production and marketing activities by slaves on Brazilian sugar plantations. Gorender, “Questionamentos,” 26-35, and A escravidão reabilitada, 74. As Schwartz notes, however, Gorender’s argument is not convincing. The Jesuits were not at all paternalistic in their administration of Santana; moreover, the estate was expropriated from the Jesuits some 30 years before the 1789 rebellion took place. See Schwartz, Sugar Plantations, 530, n. 103, and Slaves, Peasants, and Rebels, 52, 58-59, n. 42.

50

On Santana’s isolated setting, see Reis, “Resistência escrava,” 288–89. For the area’s population, see Baltasar da Silva Lisboa, Ouvidor e Juiz Conservador, Ilhéus, “Memoria sobre o corte das madeiras na Comarca dos Ilhéos,” ca. 1800, BN, SM, II-34,3,6, fol. 7. For the number of engenhos, see José da Silva Lisboa, “Descripção,” 123.

51

Gonçalo Francisco Monteiro, Ouvidor Interino, Ilhéus, to Governor of Bahia, Feb. 27, 1796, APEB, SH, 184; Baltasar da Silva Lisboa, “Memoria sobre a comarca dos Ilhéos,” 1802, Anais da Biblioteca Nacional 37 (1915), 9; idem, “Memoria sobre o corte das madeiras na Comarca dos Ilhéos,” fol. 4; Maximiliano, Príncipe de Wied-Neuwied [Maximilian zu Wied-Neuwied], Viagem ao Brasil, trans. Edgar Süssekind de Mendonça and Flávio Poppe de Figueiredo, 2d ed. (São Paulo: Companhia Editoral Nacional, 1958), 339; Joh[ann] Bapt[ist] von Spix and Carl Friedr[ich] Phil[ipp] von Martius, Reise in Brasilien auf Befehl Sr. Majestät Maximilian Joseph I. Königs von Baiern in den Jahren 1817 bis 1820, 3 vols. (Munich: I. J. Lentner, 1823-31), 2:677. On the scarcity of butchers, see “Officio do ex- Governador da Bahia Manuel da Cunha Menezes para Martinho de Mello e Castro, sobre a Capitania dos Ilhéos,” 1780, Anais da Biblioteca Nacional 37 (1910), 473.

52

Slaves at Santana may have preferred to sell their produce in Salvador because prices were higher than those paid on the engenho itself. In the 1750s, Santana’s administration purchased farinha from the estate’s slaves for less than the current market price. Schwartz, Slaves, Peasants, and Rebels, 32. But it is not clear that prices were much higher in Salvador than in the town of Ilhéus. For another example in which a slave from southern Bahia sent produce to Salvador’s market, see “Rol das pessoas q. embarcarão farinhas, arrozes, efeijao, q. estão por pagar, no Barco do Thene Jozê Maciel Frra … , " Nov. 22, 1785, APEB, SH, 201-32.

53

Cardoso, Escraυo ou componês, 58-59, 87-88; Berlin and Morgan, Introduction, 5-6. Berlin and Morgan, surveying the literature, point out that even in regions where land was plentiful, planters were less likely to rely chiefly on provision grounds to feed their slaves if they had ready access to off-plantation supplies, and more likely to do so in areas where they could not easily acquire foodstuffs from off-plantation sources. Indeed, Santana seems to have relied largely on internal supplies of cassava flour: inventories from the 1750s indicate that the engenho maintained a casa de farinha to prepare flour and used estate labor to grow enough cassava to meet perhaps half its needs. At the same time, the casa de farinha at Santana was occupied two days a week in processing cassava the plantation’s slaves had grown on their own. The slaves, in turn, sold surplus flour to the engenho. Schwartz, Slaves, Peasants, and Rebels, 52-53. On later food production at Santana, see Reis, Resistência, 290. In the sample inventories, by contrast, only about one-third of all cane farms and one-fifth of all sugar engenhos possessed equipment to make flour. References to cassava fields are even less common in those inventories. See also Barickman, Slave Economy, 154–62.

54

Tomich, Slavery in the Circuit of Sugar, 276; John Campbell, “As ‘A Kind of Freeman ?: Slaves Market-related Activities in the South Carolina Upcountry, 1800–1860." Slavery and Abolition 12:1 (May 1991), 143; Mintz, Caribbean Transformations, 197-98; Mintz and Hall, Origins of the Jamaican Internal Marketing System, 15, 19-20; Hilary McD. Beckles, Natural Rebels: A Social History of Enslaved Black Women in Barbados (New Brunswick: Rutgers Univ. Press, 1989), 74-77; Marshall, “Provision Ground,” 56-58; David Barry Gaspar, “Slavery, Amelioration, and Sunday Markets in Antigua, 1823-1831,” Slavery and Abolition 9:1 (May 1988), 1–28, esp. 5, 13, mentioning similar legislation on Montserrat in the early eighteenth century and on St. Kitts and Nevis in the early nineteenth century. See also Bergad, Cuban Rural Society, 80.

55

See the municipal laws from Diamantina and Itajubá, Minas Gerais, and Cabo, Pernambuco, and the provincial law from Pará that refer specifically to slaves as sellers, all of which date from the nineteenth century, published in Children of God’s Fire: A Documentary History of Black Slavery in Brazil, ed. Robert Edgar Conrad (Princeton: Princeton Univ. Press, 1983), 261-62, 264, 266. For similar legislation from townships in the province of São Paulo, see Ademir Gebara, O mercado de trabalho livre no Brasil (1871-1888) (São Paulo: Editora Brasiliense, 1986), 103-5.

56

Cf. Gebara, O mercado, 103-5.

57

Reis, “Slave Rebellion,” esp. 94-112, 309-16.

58

On sources for municipal and provincial legislation, see  appendix. With one ambiguous exception, provincial legislation also failed to deal explicitly with independent marketing acitvities by rural slaves. In 1822, Bahia’s provisional governing council issued an emergency wartime decree ordering police and militia escorts to arrest any slave wandering about the countryside without a note from his master, but excepted “slaves who conduct wagons, beasts of burden, or who are carriers of provisions to fairs and markets. The decree does not refer to slaves as sellers, and the provisions they carried may well have belonged to their masters. Receipts and bills of sale attached to inventories reveal that planters and farmers frequently sent their slaves into town to deal with merchants on their behalf. For the text of the decree, see Conrad, Children, 255-56.

59

Conde dos Arcos, Governor of Bahia, to the Câmara de Maragogipe, Aug. 12, 1816, APEB, SH, 5473, which includes a copy of the revised proposed regulations that were endorsed by the colonial government; Eduardo de Caldas Britto, Levantes de pretos na Bahia, Revista do Instituto Geográfico e Histórico da Bahia 10:29 (1903). 88-94. On the uprising, see Reis, “Slave Rebellion,” 99.

60

On legislation discriminating against freed slaves (especially freed African-born slaves), see Manuela Carneiro da Cunha, Negros, estrangeiros: os escravos libertos e sua volta à Africa (São Paulo: Editora Brasiliense, 1985), 68—86; and Reis, Slave Rebellion, 309-16. Townships in the Recôncavo also passed ordinances discriminating against freed slaves.

61

Manoel Ferrás da Mota Pereira, Juiz de Paz, Iguape, to President of Bahia, Oct. 4 1830, APEB, SH, 2394.

62

Mattos, “Vila do Príncipe,” 124-27, 139-41; Fragoso, “Sistemas agrários,” 67-69; Galliza, O declínio, 148-51. Also more generally, on the usefulness of these probate records for research on provision grounds and independent marketing activities developed by rural slaves, see Cardoso, Escravo ou camponês, 121, and “Peasant Breach,” 52.

63

Inv. of José Manoel, ARC, IT (Iguape, 1825). The fact that Teodósio, as a slave, had a surname is surprising since slaves in Brazil seldom possessed surnames. Teodósio’s use of a surname is even more surprising because José Manoel is listed in his own inventory as lacking any surname. Nothing in the documentation explains how or when Teodosio acquired or adopted “da Silva” (perhaps the most commonplace of all family names in Brazil) as his surname. The explanation may, however, be in some way related to the fact that Teodósio was a pardo and could therefore claim some degree of kinship with a white person.

64

Estimates of the purchasing power of Rs.37$680 are based on judicial evaluations in other inventories carried out in the years 1823-1826 and on the price of farinha as recorded in Mattoso, “Au nouveau monde,” annexes, 445-61.

65

Inv. of Félix Alves de Andrade, Cach., 1791, APEB, SJ, IT. Price of beans from Mattoso, “Au nouveau monde,” annexes, 445-61. Mattoso did not collect information on the price of palm oil. The entry in the accounts refers simply to “oil” (azeite), but this was surely palm oil and not olive oil (azeite doce). Olive oil was imported from Europe, but oil palms were grown in Bahia. Furthermore, in current expressions such as comida de azeite, the term azeite is used in Bahia to refer to palm oil.

66

Inv. of José Francisco das Neves, SFC, 1837, APEB, SJ, IT. Area under cultivation was estimated from contemporary judicial appraisals of cassava fields in other inventories. The number of plants and their yield were estimated with ratios discussed in Barickman, “Slave Economy,” chaps. 6 and 7. I have assumed here a conservative yield of 20 alqueires per thousand covas. The only other example I was able to locate in accounts attached to inventories was the sale in 1843 of a capado (gelded goat or pig or possibly ram) by a slave named Alvaro at the Engenho Buranhém. Inv. of Joaquim Ferreira Bandeira, SFC, 1842, APEB, SJ, IT.

67

Mintz, Caribbean Transformations, 204-6; Beckles, Natural Rebels, 80; Marshall, “Provision Ground,” 57-58; Tomich, Slavery in the Circuit of Sugar, 275; Gaspar, “Slavery, Amelioration, and Sunday Markets. See also Roderick A. McDonald, Independent Economic Production by Slaves on Antebellum Louisiana Sugar Plantations, Slavery and Abolition 12:1 (May 1991), 194-95·

68

For early references to Saturday as market day, see Lisboa, Carta, 503; Antônio Augusto da Silva, Juiz de Fora, Maragogipe, to Governor of Bahia, Mar. 16, 1817, APEB, SH, 234. By the late 1820s, if not before, market fairs were held in Nazaré on Mondays and Fridays as well as on Saturdays. Tuesdays were by then the main market day in Conceição da Feira, near Cachoeira, and in Santana dos Olhos d’Agua (modern Feira de Santana), located just west of the Recôncavo. The markets at both Conceição da Feira and Santana dos Olhos d’Agua, however, not only handled locally grown produce but also specialized in cattle brought from the backlands of Bahia. Domingos José Antonio Rebello, Corographia, ou abreviada historia geographica do imperio do Brasil (Bahia, 1829), reprinted in Revista do Instituto Geográfico e Histórico da Bahia 55 (1929), 173, 175, 180; José Joaquim de Arnizáu, “Memoria topographica, historica, commercial e politica da Villa da Cachoeira da Provincia da Bahia” [1825], Revista do Instituto Histórico e Geográfico Brasileiro 25 (1862), 134. Field observations in the late 1980s and 1992 found market fairs held only on Wednesdays, Fridays, and Saturdays in Nazaré, Maragogipe, São Gonçalo dos Campos, and Cachoeira. In all four towns, the main feira was held on Saturdays. Saturday was, at the time, the only market day in the village of Santiago do Iguape. On Saturday as being currently the main market day in the rural Recôncavo, also see Sylvia Maria dos Reis Maia, “Dependency and Survival of Sapeaçu Small Farmers, Bahia, Brazil” (Ph.D. diss., Boston Univ., 1985), 70, 138.

69

Fragoso, “Sistemas agrários,” 67-69; Fragoso and Florentino, "Marcelino,” 170.

70

Barickman, “Slave Economy,” 182—89.

71

Schwartz, Sugar Plantations, 100-5. For the harvest season in Louisiana, see McDonald, “Independent Economic Production,” 184.

72

Beckles, Natural Rebels, chap. 4. Gaspar documents extensive independent marketing activities by plantation slaves in Antigua, where conditions closely resembled those in Barbados. “Slavery, Amelioration, and Sunday Markets,” 5-8.

73

Marshall, “Provision Grounds,” 53-54.

74

Ibid., 62-63; Berlin and Morgan, Introduction, 8-9; Cardoso, Escraυo ou camponês, 59.

75

See, e.g., Richard B. Sheridan, Sugar and Slavery: An Economic History of the British West Indies, 1623-1775 (Baltimore; Johns Hopkins Univ. Press, 1973), chaps. 7-10; and Higman, Slave Populations, 46—71.

76

Richard B. Sheridan, Doctors and Slaves: A Medical and Demographic History of Slavery in the British West Indies, 1680-1834 (Cambridge: Cambridge Univ. Press, 1985), 154-62; Berlin and Morgan, Introduction, 8.

77

On market-oriented cultivation of cassava in Bahia in the period 1780-1860, see Barickman, “Slave Economy,” esp. 206-8 and 450-55. On the earlier development of commercial production of foodstuffs in Bahia, see Francisco Carlos Teixeira da Silva, “A morfologia da escassez: crises de subsistencia e política econômica no Brasil colônia (Salvador e Rio de Janeiro, 1680-1790)” (Ph.D. diss., Univ. Federal Fluminense, 1990), esp. chap. 5. Da Silva’s research also failed to locate indications that plantation slaves in the Recôncavo regularly marketed surplus produce. Personal communication, Sept. 1986.

78

“Relação do numero de fogos,” 1835. On the small farmers of Iguape and patterns of land tenure in the parish, see Barickman, “Slave Economy,” 321-31; on small farmers in other sugar-producing parishes, ibid., 35. In the cassava-growing townships of the southern and western Recôncavo, small and middle-sized farmers often owned land. See ibid., 305−11.

79

Lista dos lavradores de mandioca de Jaguaripe, 1780, APEB, SH, 199; “Lista das mandiocas que se achão nos lavradores do districto da Villa de Maragogipe . . . ,” 1780, APEB, 187. For the method used in estimating output and marketable surplus, see Barickman, “Slave Economy,” 451-55.

80

Monthly “Rellação das farinhas que forão exportados para o Celleiro Publico da Bahia . . . deste termo de Jaguaripe,” covering the 12-month period Oct. 1, 1825-Oct. 31, 1826 (excluding June 1826, for which information is lacking), APEB, SH, 1609, 2440, 2470.

81

‘Ofício do Capitão-mor da Vila de Jaguaripe ao Governo da Bahia que remete as relações dos lavradores ... de mandioca’’ (1781), BN, SM, 1-31,30,52.

82

Barickman, “Slave Economy,” 358-59.

83

Lisboa, “Carta,” 503; Oliveira, “Breve compendio,” 182; Maximiliano, Viagem, 462-63; Rebello, Corographia, 180-81.

84

“Mappa especulativo dos efeitos entrado [sic] pelas estradas de Nazaré Termo de Jaguaripe na semana e feira de 12 de Janr0 de 1823, BN, SM, 11-34,8,29. See also Rebello, Corographia, 180—81.

85

The evidence from the Bahian Recôncavo for the late eighteenth and early nineteenth centuries thus corroborates Berlin and Morgan’s argument that the presence of nonplanter farmers in a region limited the development of the slaves’ economy”. Introduction, 6. Having mainly the North American experience in mind, however, Berlin and Morgan do not consider the possibility that the nonplanter farmers who provided foodstuffs to feed plantation slaves may have also been slaveholders.

86

Kenneth F. Kiple, “The Nutritional Link with Slave Infant and Child Mortality in Brazil,” HAHR 69:4 (Nov. 1989), 677—90.

87

Calmon, Ensato, 60. Coffee planters in the province of Rio de Janeiro in the mid-nineteenth century, cotton planters in South Carolina, and sugar planters in the West Indies all saw similar advantages in assigning provision grounds to their slaves. Eduardo Silva, “O Barão de Pati do Alferes e a fazenda de café da velha província,” introduction to Memória sobre afundação de uma fazenda (edição original de 1847 e edição modificada e acrescida de 1878), by Francisco Peixoto de Lacerda Werneck (Brasília: Senado Federal, 1985), 35-37; Berlin and Morgan, Introduction, 19-20; Campbell, “As ‘A Kind of Freeman, ” 152-53.

88

Eduardo Silva, “A função ideológica da brecha camponesa,” in Reis and Silva, Negociação e conflito, 22-31.

89

Werneck, Memória, 63.

90

Emília Viotti da Costa, Da senzala à colonia, 2d ed. (São Paulo: Livraria Editora Ciencias Humanas, 1982), 377—78. On kinlessness (or, following Orlando Patterson, “natal alienation”) as a fundamental component in the definition of slavery, see M. I. Finley, Ancient Slavery and Modern Ideology (Harmondsworth: Penguin, 1983), 75-77; and Orlando Patterson, Slavery and Social Death: A Comparative Study (Cambridge: Harvard Univ. Press, 1982), 7-9.

91

Berlin and Morgan, Introduction, 15.

92

Candler and Burgess, Narrative, 57.

93

See Da Cunha, Sobre os silêncios, 5 and passim.

94

John Campbell, “In the Marketplace: The Slave Economy in the South Carolina Cotton Country, 1795-1860” (Unpublished ms., Univ. of Arizona, 1991), 1.

95

Candler and Burgess, Narrative, 57.

96

Berlin and Morgan, Introduction, 21.

97

The expression negotiate is taken from Eduardo Silva, “Entre Zumbi e Pai-João, oescravo que negocia,” Jornal do Brasil (Rio de Janeiro), Caderno especial, Aug. 18, 1985, 3, reprinted in Reis and Silva, Negociação e conflito, chap. 1.