Stanford
NOVEMBER 26, 2009
Publications
Publications
About Us
About Us
Gallery
Gallery
Resources
Resources
Publications > Table of Contents > The Slave Market in Rio de Janeiro circa 1869
Spatial History Lab: Working paper; Submitted 28 August 2009; Currently in review
The Slave Market in Rio de Janeiro circa 1869

Movement, Context, and Social Experience
Zephyr Frank 1 & Whitney Berry 2
1. Stanford University, Spatial History Project, Associate Director
2. Stanford University, Spatial History Lab, GIS Research Assistant
The market for slaves in Rio de Janeiro underwent a series of major transformations over the course of the nineteenth century. Prior to 1831, slaves poured into the city during the period of the legal Atlantic slave trade. Tens of thousands more Africans landed in Brazil’s imperial capital before the illegal Atlantic trade was finally suppressed in 18501. Thereafter, buying and selling slaves shifted to a local market typified by individual sales2. The proportion of Brazilian-born slaves rose and the social experience of the market changed for all parties involved. Rather than a concentrated, large-scale process dominated by formal market spaces and professional slave traders, the mature slave market in Rio de Janeiro evolved after 1850 into a continuous (in the sense of both time and space) process that encompassed every neighborhood in the city. The absence of an individual slave would be felt in the neighborhood from which she originated, just as her new presence would be felt in the home and neighborhood of her new owner. Mapping the origins and destinations of slaves in this system, using detailed transaction data from the year 1869, highlights the ubiquity of slavery in Rio de Janeiro as well as the constant movement of slaves in and out of new environments. Connecting buyers and sellers to data regarding their wealth and occupations further highlights the significant changes experienced by all parties caught up in the system. Dynamic visualization techniques reveal meaningful patterns at multiple scales.

Keywords: slavery, Rio de Janeiro, slave market, slave mobility, property values

1
Historical Context
Rio de Janeiro, Brazil’s imperial capital, provides the setting for this study in spatial historical visualization. An emerging metropolis in the middle years of the nineteenth century, Rio de Janeiro’s population and economic status were unparalleled in Brazil. With 272,000 residents in the urban parishes according to the 1872 census, the city accounted for a disproportionate share of the country’s export trade (dominated by coffee) and was home to nearly all of the important institutions of the state. In this context, a population of nearly 50,000 slaves lived, worked, and were sometimes bought and sold3. Although slaves made up just under one-fifth of the population, they accounted for a much more significant proportion of the labor force, particularly in the manual trades and domestic service. The distribution of slaves was broad in Rio de Janeiro, with very few owners commanding more than ten bondspersons. In addition, given the urban environment and labor market, slaves were often rented out to third parties or expected to work outside of the supervision of their masters4. Thus, slaves, sometimes operating autonomously, were found in virtually every part of the city.
2
Argument
Contrary to life in rural areas, slaves in Rio de Janeiro rarely had the opportunity to marry and they tended to live alone or in small groups of unrelated individuals (the primary exception being slave mothers with young children)5. Their social ties were, therefore, more tenuous and dependent on institutions such as religious brotherhoods and, we hypothesize, neighborhood friends and acquaintances6. Therefore, the experience of being sold and forced to move to an entirely different neighborhood would likely have been much more significant than the case of a sale from one owner to another in the same street. What is more, buyers and sellers often came from quite distinctive social and economic backgrounds. This suggests that the characteristics of owners, as well as spatial distance, may have played a profound role in determining how a sale affected slave experience. There are no slave diaries or oral histories available to answer these questions in the context of sales within the city of Rio de Janeiro. Court records may provide some detail regarding individual cases, but these are problematic because they involve, by definition, unusual cases where conflicts erupted. Our approach is different. Taking over 1,107 slave sales from the year 1869, we mapped the locations of buyers and/or sellers in 351 cases7. Where possible, we connected these individuals to information about the neighborhood within which they resided by calculating neighborhood rental values according to data from the decima urbana (property tax roll) for the year 18708.
3
The spatial context for the analysis of these data is provided by a historical GIS for 19th century Rio de Janeiro, developed by our team at Stanford’s Spatial History Lab. We mapped the residential addresses of buyers and sellers within the space of the city using a historically-accurate street network.
4
Analysis
Our analysis centers on the notion of changing circumstances associated with slave sales. Slaves moved from sellers to buyers and, in the process, from one household and neighborhood to another. The greater the change in household and neighborhood, the more profoundly slaves experienced the effects of being sold in Rio de Janeiro’s complex urban setting. There are three important categories where we can observe the degree of change in our dataset. First, we examine the simple question of distance: how far did slaves move? Second, our data allows us to determine the gender of buyers and sellers, providing insight into this important dimension of change in circumstances. Third, using detailed information about property values, we can address the question of change in general neighborhood characteristics as well as in the specific value of the properties of sellers and purchasers. Our focus on these three categories is warranted for the following basic reasons:
5
1.
Since slaves were somewhat (or entirely) constrained in their movements, being sold over a great distance meant moving partly or wholly out of one context and into another. Maintaining connections to the old neighborhood would, we hypothesize, become increasingly difficult over greater distances. In addition, in cases where slaves were sold out of the city or purchased from sellers in other communities, the spatial dislocation would have been complete.
6
2.
The gender of a slave’s owner was consequential for at least two reasons. First, female slaves made up about 55 percent of the average woman slaveholder’s bondspersons; on the other hand, females made up about 36 percent of the average male slaveholder’s bondspersons9. Thus, on average, a slave sold from a male owner to a female owner would move from a world dominated by male slaves to one where female slaves made up a slight majority. Second, female owners, on average, were slightly more likely to manumit their bondspersons, and their propensity was to manumit female slaves10. Thus, a female slave sold from a male owner to a female owner might see a slightly higher chance of freedom in the long run. A male slave in the same scenario might actually see a slight decrease in the chances of manumission. In any event, the immediate effect of being sold would often be to place the prospect of liberty further over the horizon for a slave, as many owners who manumitted their bondspersons did so on account of their years of “good service,” a laurel slaves could only hope to garner after a significant period of time.
7
3.
The value of property occupied by the slave owner is a reasonable proxy for their economic standing. Likewise, the surrounding values in the neighborhood provide crucial information regarding the wider context of wealth or poverty within which the slave would be expected to live and work11. Note, of course, that some slaves lived outside of their owner’s residences, so this is a tendency, not a universal rule. A slave moving from a rich neighborhood to a poor one would, we hypothesize, experience quite different conditions of living and labor.
8
9
See patterns
Discover pathways
Analyze gender
According to our data, most slaves sold within the city moved significant distances. As the visualization shows, many slaves moved from the periphery of the city center and vice versa. Movement was, therefore, often from one kind of social world to another. Filtering by gender, the visualization and underlying data show that male slaves tended to move greater distances than their female counterparts as the result of a sale. For the portion of the data where we have two addresses located within the city, the average distance traversed by males was 1,408 and by females 1,198 meters. Women buyers and sellers, a relatively small minority in both cases, tended to transact across shorter distances, suggesting that gender influenced distance both with respect to the characteristics of the slave and the owners. Finally, in the limited number of cases where it was possible to calculate neighborhood rent statistics, on average slaves experienced a 41% absolute difference between the rent values associated with their previous home and their new one.
10
11
.CSV (146 KB)
Conclusions and Pathways for Further Exploration
The data and accompanying visualizations presented here reveal complex and meaningful patterns in slave experience. Rich data regarding buyers, sellers, and their slaves, illuminate a dynamic market system differentiated along lines of gender that sometimes thrust slaves into dramatically different living conditions. Because the data contains elements of both space and time, it moves beyond static portraits of slave life and begins to capture the dimensions of instability and change experienced by thousands of slaves in the city of Rio de Janeiro over the course of the nineteenth century. Connecting these events to detailed information regarding the wider context of slave’s points of departure and destinations yields further insights into how changes in ownership were significant for individual slaves and for categories of bondspersons.
12
There are significant limitations to our analysis, including a lack of full information with respect to most cases of slave sales. Furthermore, our data cover but part of one year, 1869, and only future archival work will tell whether these patterns were similar over longer stretches of time. Finally, our approach to understanding changes in slave experience is indirect and inferential. This probably cannot be helped, given the paucity of information available in the archives concerning what slaves themselves felt about their lives and owners. Nevertheless, here for the first time we see the outlines of the slave market in Rio de Janeiro at a particular point in time. We see slaves in motion and can begin to do the work of historical sleuthing and imagination necessary to give meaning to their paths through the city.
13
End Notes

1 Mary Karasch, Slave Life In Rio de Janeiro, 1808-1850 (Princeton: Princeton University Press, 1987), esp. p. A good summary history of the slave trade is found in Herbert Klein, The Atlantic Slave Trade (Cambridge: Cambridge University Press, 1999). For the African side of the story of the trade, see Joseph Miller, Way of Death : merchant capitalism and the Angolan slave trade, 1730-1830 (Madison: University of Wisconsin Press, 1988).

2 For a discussion of professional slave traders in Rio de Janeiro, before and after 1850, see Luís Carlos Soares, O "povo de cam" na capital do Brasil: a escravidão urbana no Rio de Janeiro do século XIX (Rio de Janeiro: 7Letras, 2007), esp. pp. 51-53, 56. Soares does not analyze the small-scale local slave sales that form the basis of the present analysis, focusing his attention, instead, on larger slave traders. Relying on published newspaper data for the buying and selling of slaves, Soares comes to the conclusion that there were far fewer slave sales in the city after 1850.

3 Brazil, Recenseamento Geral . . . 1872 (Rio de Janeiro: Typ. Leuzinger), p. 61. The census enumerated 48,939 slaves in 1872: 24,886 men and 24,053 women.

4 Karasch, Slave Life.

5 One out of 964 slaves in the slave sale database with civil status noted was listed as married; the remaining 963 appeared as unwed. This is not, it should be noted, an artifact of the category—slaves for sale. Independent records yield rates of marriage almost as low. For instance, in São José parish, the baptism records of slaves in 1850 and 1868-71 show that married mothers made up just 11 of 448 cases. For the study of slave marriage in rural plantation zones, see Robert Slenes, Na senzala, uma flor esperanças e recordações na formação da família escrava Brasil Sudeste (Rio de Janeiro: Nova Fronteira, 1999). A detailed case study is also offered in Sandra Lauderdale Graham’s Caetana Says No: women's stories from a Brazilian slave society (Cambridge: Cambridge University Press, 2003).

6 For the importance of lay religious brotherhoods, see Mariza de Carvalho Soares, Devotos da cor: identidade étnica, religiosidade e escravidão no Rio de Janeiro, século XVIII (Rio de Janeiro: Civilização Brasileira, 2000). For a revealing portrait of slaves interacting among themselves in Rio, see Sidney Chalhoub, Visões da liberdade: uma história das últimas décadas da escravidão na Corte (São Paulo: Companhia das Letras, 1990). A discussion of neighborhoods and slave experience can be found in Sandra Lauderdale Graham’s House and Street: The Domestic World of Masters and Servants in Nineteenth-Century Rio de Janeiro (Cambridge: Cambridge University Press, 1988).

7 Meia-siza, Recebedoria do Rio de Janeiro, 1869, AGCRJ.

8 Decima urbana, 1870, AGCRJ.

9 This calculation is based on 181 estate inventories (70 women, 111 men) drawn randomly from the Arquivo Nacional for the period 1855-1860. ANRJ, inventarios. Nearly identical percentages (57 percent female slaves among women owners, 36 percent among men) were also calculated from a large corpus (N = 1,096) of vaccination records involving adult slaves during the years 1851-1854. AGCRJ, vaccinations.

10 Analysis of 85 manumission records from the periods 1854-55 and 1868-70 suggests that women made up between one-fifth and one-third of manumitters, and that they overwhelmingly manumitted female slaves (about 85 percent). Using these same records, we see that female slaves accounted for about 63 percent of all manumissions. ARQUIVO NACIONAL, CARTAS DE LIBERDADES, livro n.53 ano 1854/55 do 1.oficio de notas do RJ. Periodo 24/03/54 a 21/06/1855 and ARQUIVO NACIONAL, CARTAS DE LIBERDADES, livro n.74 ano 1868/70 do 1.oficio de notas do RJ. Periodo 10/09/68 a 04/05/1870.

11 For a similar argument based on the study of the movement of artisans within the space of the city, see Zephyr Frank, “Layers, Intersections, and Flows,” Journal of Social History 41:2 (winter 2007).

Acknowledgments We'd like to thank the following people for their contributions: Mithu Datta (GIS Specialist, SHL), Erik Steiner (Lab Director, SHL), Ryan Delaney (RA, SHL), David Sabeti (RA, SHL), Hannah Gilula (RA, SHL), Lucas Manfield (RA, SHL), Luciana Barbeiro (Cecult, UNICAMP), Chester Harvey (RA, SHL), Meredith Williams (Stanford)and Tereza Cristina Alves.

Author Information Correspondence and requests for materials should be addressed to Zephyr Frank zfrank@stanford.edu or Whitney Berry wberry@stanford.edu

Rights and Permissions Copyright ©2009 Stanford University. All rights reserved. This work may be copied for non-profit educational uses if proper credit is given. Click here for additional permissions information.