¹ On the census-based research and for references to many of the relevant works, see, e.g., Flávio Motta, José, “The historical demography of Brazil at the V centenary of its discovery,” Ciência e Cultura, 51:5–6 (1999), pp. 446–56;Google Scholar and the on-line bibliography maintained by the Núcleo de Estudos de História Demográfica at http://historia_demografica.tripod.com. Note that, by far, most of the currently available census-based literature deals with Southeastern Brazil (especially São Paulo and Minas Gérais); comparable research on the Northeast is relatively scant. Subsequent notes use the following abbreviations: APEB, Arquivo Público do Estado da Bahia; SH, Seção Histórica; SJ, Seção Judiciária; IT, Inventários e Testamentos; inv(s)., inventory(ies); ARC, Arquivo Regional da Cachoeira; RAE, papéis avulsos e encaixotados; BN-sm, Biblioteca Nacional (Rio de Janeiro), Seção de Manuscritos; CL, Bahia (Province), Collecção das leis e resoluções da Assemblèa Legislativa da Bahia …, 33 vols. (Bahia, 1862–89); gov., governor of Bahia; pres., president of the province of Bahia; jz, juiz; vig., vigário; S., São, Santo, or Santa; EE, Estudos econômicos; HAHR, Hispanic American Historical Review.

² The few studies dealing with these topics for Brazil generally focus on census categories for color or race. See n. 16 below. Elsewhere, by contrast, such topics have received more attention. See, e.g., Hakim, Catherine, “Census Reports as Documentary Evidence: The Census Commentaries, 1801–1951,” The Sociological Review, n.s., 28:3 (1980), pp. 551–80;CrossRefGoogle Scholar Deacon, Desley, “Political Arithmetic: The Nineteenth-Century Australian Census and the Construction of the Dependent Woman,” Signs 11 (1985), pp. 27–47;CrossRefGoogle Scholar Alonso, William and Starr, Paul, eds., The Politics of Numbers (New York, 1986);Google Scholar Conn, Bernard S., An Anthropologist Among the Historians and Other Essays (Delhi, 1987), pp. 224–54;Google Scholar Scott, Joan Wallach, Gender and the Politics of History (New York, 1988), pp. 112–138;Google Scholar Anderson, Benedict, Imagined Communities: Reflections on the Origin and Spread of Nationalism, 2d ed. (London, 1991), pp. 164–70;Google Scholar Anderson, Margo, “The History of Women and the History of Statistics,” Journal of Women's History 4:1 (1992), pp. 14–36;CrossRefGoogle Scholar and Celis, Leticia Mayer, Entre el infierno de una realidad y el cielo de un imaginario: estadística y comunidad científica en el México de la primera mitad del siglo XIX (Mexico City, 1999).Google Scholar

³ For source, see n. 15 below. In the rest of this essay, I have usually placed between quotation marks the categories and other classificatory terms used in the census to stress that their meaning may not always be self-evident. Although some of my findings may apply to other late eighteenth- and early nineteenth-century censuses in Brazil, the arguments presented here refer specifically to the 1835 censuses from Bahia and even more specifically to the census of Iguape. Rather than claiming that the Iguape census is typical of other Brazilian population counts, I try to use the Bahian censuses (including their peculiarities) to point out the possibilities of analyzing Brazilian household lists qualitatively and the need to question the categories and format of those lists.

⁴ In this regard, the 1835 Bahian censuses resemble in their general format U.S. censuses since 1850. See Anderson, , “The History,” pp. 19–20,Google Scholar which, despite differences in context, I have found useful here.

⁵ On liberalism in nineteenth-century Brazil, see, e.g., da Costa, Emilia Viotti, Da monarquia à república: momentos decisivos, 2d ed. (São Paulo, 1979), pp. 109–26;Google Scholar Graham, Richard, “Ciudadanía y jerarquía en el Brasil esclavista” in Ciudadanía política y formación de las naciones: perspectivas históricas de América Latina, coord. Sabato, Hilda (Mexico City, 1999), pp. 345–71;Google Scholar and the works cited in n. 70 below.

⁶ For this and the following paragraph, see, e.g., Bethell, Leslie and de Carvalho, José Murilo, “Brazil from Independence to the Middle of the Nineteenth Century,” in Cambridge History of Latin America, ed. Bethell, Leslie, 11 vols. (Cambridge, 1984–95),Google Scholar vol. 3, chap. 16; de Carvalho, José Murilo, Teatro de sombras: a política imperial (São Paulo and Rio de Janeiro, 1988);Google Scholar Barman, Roderick J., Brazil: The Forging of a Nation, 1798–1852 (Stanford, 1988);Google Scholar Graham, Richard, Patronage and Politics in Nineteenth-Century Brazil (Stanford, 1990);Google Scholar de Carvalho, Marcus J.M., “Hegemony and Rebellion in Fernambuco (Brazil), 1821–1835” (Ph.D. diss., Univ. of Illinois at Urbana-Champaign, 1989);Google Scholar Assunção, Matthias Röhrig, “Elite Politics and Popular Rebellion in the Construction of Post-Colonial Order: The Case of Maranhão, Brazil (1820–41),” Journal of Latin American Studies 31:1 (1999), pp. 1–38;CrossRefGoogle Scholar Mosher, Jeffrey C., “Political Mobilization, Party Ideology, and Lusophobia in Nineteenth-Century Brazil: Fernambuco, 1822–1850,” HAHR 80:4 (2000), pp. 882–912;Google Scholar and Ribeiro, Gladys Sabino, A liberdade em construção: identidade nacional e conflitos antilusitanos no Primeiro Reinado (Rio de Janeiro, 2002).Google Scholar Also see n. 11 below.

⁷ On trade statistics, see, e.g., Luz, Nicia Villela, “Ensaio de interpretação,” in Latin America: A Guide to Economic History, ed. Conde, Roberto Cortés and Stein, Stanley J. (Berkeley, 1977), p. 170.Google Scholar

⁸ On census-taking before 1872, see de Souza e Silva, Joaquim Norberto, Investigções sobre os recenseamentos da população geral do Impèrio e de cada província …, 1870;Google Scholar facs. reprint, publ, together with Resumo histórico …, [by Francisco José de Oliveira Vianna], (São Paulo, 1986); Alden, Dauril, “The Population of Brazil in the Late Eighteenth Century: A Preliminary Study,” HAHR 43:2 (1963), pp. 173–205;Google Scholar and Marcílio, Maria Luiza, “Levantamentos censitários da fase proto-estatística do Brasil,” Anais de História (Assis), 9 (1977), pp. 63–75.Google Scholar

⁹ See, e.g., Santos Vilhena, Luís dos, A Bahia no sécula XVIII [c. 1799]Google Scholar, annot. Amaral, Braz do [1st ed. publ, as Recopilação de noticias soteropolitanas …, 1921], 3 vols. (Salvador, 1969), vol. 2, pp. 459–60;Google Scholar Wetherell, James, Brazil. Stray Notes from Bahia …, ed. Hadfield, William (Liverpool, 1860), p. 95;Google Scholar and Mattos Monteiro, Hamilton de, Nordeste insurgente (1850–1890) (São Paulo, 1981), pp. 33–45.Google Scholar

¹⁰ The first post-independence attempt to carry out a provincial census in Bahia seems to have been an almost complete failure. See pres. to the capitão-mor de Caetité, and to the ouvidor da Comarca de Jacobina (both, 22/11/1824), APEB, SH, 1621, fols. 342–3; jz de fora (Maragogipe) to the pres. (4/7/1826), APEB, SH, 2470; jz de fora? (S. Amaro) to the pres. (24/5/1826), APEB, SH, 2580. To date, from the attempted count, only a few dozen nominal lists (none, it would seem, covering an entire parish) and a few aggregated tables, all from what was then the township of Cachoeira, have come to light. ARC, PAE.

¹¹ Reis, João José, “A elite baiana face aos movimentos sociais, Bahia: 1824–1840,” Revista de História 54:108 (1976), pp. 341–84;CrossRefGoogle Scholar idem, Slave Rebellion in Brazil: The Muslim Uprising of 1835 in Bahia, trans. Brakel, Arthur (Baltimore, 1993);Google Scholar Jancsó, István, Na Bahia contra o Impèrio: História do ensaio de sedição de 1798 (São Paulo, 1995);Google Scholar Kraay, Hendrik, ‘“As Terrifying as Unexpected’: The Bahian Sabinada, 1837–1838,” HAHR 72:4 (1992), pp. 501–27;Google Scholar idem, , Race, State, and Armed Forces in Independence-Era Brazil: Bahia, 1790s–1840s (Stanford, 2001);Google Scholar Barickman, B.J., “‘Tame Indians,’ ‘Wild Heathens,’ and Settlers in Southern Bahia in the Late Eighteenth and Early Nineteenth Centuries,” TAm 51:3 (1995), pp. 325–68.Google Scholar

¹² Reis, , Slave Rebellion, chaps. 11–12;Google Scholar da Cunha, Manuela Carneiro, Negros, estrangeiros: os escravos libertos e sua volta à África (São Paulo, 1985), pp. 72–81, 96–9;Google Scholar CL, vol. 1, pp. 1–2, 22–6, 3844. Freed Africans would remain the target of discriminatory laws for many years after 1835. See CL, vol. 4, pp. 257, 261, 281, 323, 415, 530; vol. 5, pp. 260, 530; vol. 6, pp. 61, 175–76.

¹³ Silva, , Investigções, p. 79 Google Scholar (for quotations from Martins’ address); CL, vol. 1, pp. 19–21. Although falas (addresses) and relatónos (reports) by provincial presidents were routinely published, I was unable to locate a published copy of Martins’ 1835 address at the Arquivo Público do Estado da Bahia or the Biblioteca Nacional; nor does it appear in the list of published reports in Graham, Ann Hartness, Subject Guide to Statistics in the Presidential Reports of the Brazilian Provinces, 1830–1889 (Austin, Tex., 1977), pp. 18–27.Google Scholar Therefore, I have had to rely on secondary sources for quotations from Martins’ address.

¹⁴ Pres. to the vig. da freguesia de Santana desta Cidade and to the jz de paz do 1° distrito do Curato da Sé (both, 13/4/1835), APEB, SH, 1653, fols. 28–9. The model forms were not transcribed in the outgoing correspondence cited here, but their general appearance can be surmised from the letter sent to vicars and from the overall similarity in the format of the surviving censuses. For vicars’ excuses, see APEB, SH, 5212. Note that, given the lack of separation of Church and State in imperial Brazil, vicars were in effect public servants.

¹⁵ “Relação do Numero de Fogos, e moradores do Districto da Freguezia de Sant-Iago Maior do Iguape, … da Villa da Cachoeira” (1835), APEB, SH, 6175–1. The other surviving censuses are of S. Gonçalo dos Campos and S. José das Itapororocas, both mainly rural parishes, and S. Pedro Velho, an urban parish in Salvador. APEB, SH, 5683, 5684, and 5685. Surprisingly, the census of S. Pedro Velho was, it seems, prepared less carefully than the other three; e.g., it apparently excludes children under 7 and does not record naturalization or the ages of African-born slaves. The first scholar to call attention to the 1835 censuses was Azevedo, Thales de, Povoamento da Cidade do Salvador, [3d ed.] (1949; Salvador, 1969), p. 233.Google Scholar On Iguape, see Barickman, B.J., A Bahian Counterpoint: Sugar, Tobacco, Cassava, and Slavery in the Recõncavo, 1780–1860 (Stanford, 1998), chaps. 5, 6.Google Scholar

¹⁶ Like the other categories in the 1835 censuses, those used to classify individuals by color should also be questioned. I have dealt with the matter in “As cores do escravismo: escravistas ‘pretos,’ ‘pardos’ e ‘cabras’ no Recôncavo baiano, 1835,” População e família 2 (1999), pp. 7–62. On census categories for race or color, also see Nazzari, Muriel, “Vanishing Indians: The Social Construction of Race in Colonial São Paulo,” TAm 57:4 (2001), pp. 497–524;Google ScholarPubMed and Nobles, Melissa, Shades of Citizenship: Race and the Census in Modern Politics (Stanford, 2000)Google Scholar (which deals with published censuses from 1872 onward).

¹⁷ The census-takers in Iguape and S. Gonçalo dos Campos used the column for “observations” mainly to distinguish freeborn individuals (regardless of color) from freed former slaves; in S. José das Itapororocas, they recorded the distinction under “occupation.”

¹⁸ Anderson, , Imagined Communities, p. 106.Google Scholar

¹⁹ Quoted in Silva, , Investigções, p. 18.Google Scholar But note that tables with aggregated information are attached to the censuses of Iguape, S. Gonçalo dos Campos, and S. Pedro Velho. If a similar table was prepared for S. José das Itapororocas, it is no longer attached to the census.

²⁰ By contrast, the surviving 1825–26 censuses from Cachoeira often locate “households” by citing landmarks, local place names, rural estates, etc.; e.g., “no Dendê,” “na Varge [sic] Grande,” “Fazenda da Boavista,” and “no Tanque do Engenho do Rosário.” Although no doubt largely incomprehensible to provincial authorities in Salvador, such references would have been highly meaningful and precise for local officials. Also note that, in the urban parish of S. Pedro Velho, where the census groups “fogos” by streets, lanes, squares, etc., “household” numbers are apparently street addresses.

²¹ See the manuscript censuses, dating mainly from the 1780s, in APEB, SH, 596; the 1775 census of S. Pedro Velho (Salvador), transen in da Costa, Avelino Jesus, “População da Cidade da Baía em 1775,” in Actas do V Colóquio Internacional de Estudos Luso-Brasileiros (Coimbra, 1964), pp. 195–238;Google Scholar the 1818 census of Santarém (Bahia), APEB, SH, 246; “Mapas estatísticos da Comarca de Porto Seguro” (1812–20), BN-sm, I-31,19,15; and the 1825-26 censuses from Cachoeira, in ARC, PAE. At most, these censuses indirectly register birthplace when, in recording color, they classify free and enslaved non-whites as “crioulos” (Brazilian-born “blacks”), “pardos,” or “Africans” or, less often, by referring to the supposed ethnic origins of Africans. On birthplace in censuses elsewhere in Brazil, see Marcílio, , “Lev-antamentos,” p. 71;Google Scholar Nero da Costa, Iraci del, Populações mineiras: sobre a estrutura populacional de alguns núcleos mineiros no alvorecer do século XIX (São Paulo, 1981), pp. 218–24;Google Scholar and Paiva, Clotilde Andrade, “Minas Gérais no século XIX: aspectos demográficos de alguns núcleos populacionais,” in Brasil: História econômica e demográfica, ed. Nero da Costa, Iraci del (São Paulo, 1986), pp. 186–87Google Scholar, n. 15.

²² See, e.g, the works cited in nn. 6 and 11 (esp. Barman, , Brasil, pp. 17–30;Google Scholar and Ribeiro, A liberdade, chap. 1); as well as Cunha, , Negros, pp. 81–86.Google Scholar

²³ Rodrigues, José Honório, A Assembléia Constituinte de 1823 (Petrópolis, 1974), pp. 56–60,Google Scholar 127–30; Carneiro da Cunha, Paulo Octávio, “A fundação de um império liberal: discussão de princípios,” in História geral da civilização brasileira, ed. de Holanda, Sérgio Buarque and Fausto, Boris, 11 vols (São Paulo, 1960–84), tomo II, vol. 1, p. 254;Google Scholar Júnior, Caio Prado, Evolução política do Brasil: Colônia e Império, 15th ed. (São Paulo, 1986), pp. 54–55.Google Scholar

²⁴ “Constituição política do Império do Brasil,” in Constituições do Brasil, comp. Campanile, Adriano and Lobo Campanhole, Hilton, 10th ed. (São Paulo, 1989), pp. 749–50,Google Scholar 767–70 (Tít. 1°, art. 1; Tít. 2°, art. 6, §§ I-IV; Tít. 8°, art. 179). Also see Pimenta Bueno, José Antonio (de São Vicente, Marquês), Direito público brasileiro e análise da constituição do Império (1857; Brasília, 1978), pp. 442–47.Google Scholar Here I set aside the distinction between “active” and “non-active” citizens.

²⁵ See the sources cited in nn. 6 and 11. Also note that Brazilians at the time often used “pátria” (homeland, fatherland, nation) to refer to their native province rather than to “Brazil.” See, e.g., Barman, , Brasil, pp. 26–27.Google Scholar

²⁶ “Lista das pessoas que se achão assistentes na Freguezia de N. Sr^a de Nazaré” (1779), APEB, SH, 596. Elizabeth Kuznesof reports that some late colonial censuses in São Paulo had a similar format. Personal communication. The same was true for U.S. censuses before 1850. Anderson, , “The History,” pp. 17–18.Google Scholar Also note that, although not present in Iguape, free Brazilians born in provinces other than Bahia do appear in the other 1835 censuses.

²⁷ Cunha, , Negros, pp. 63–86,Google Scholar esp. pp. 74–81; Reis, , Slave Rebellion, p. 192;Google Scholar [Agostinho Marques] Malheiro, Perdigão, A escravidão no Brasil …, 3d ed., 2 vols. (1866–67; Petrópolis, 1976), vol. 1, pp. 140–43,Google Scholar who also noted that no law prevented the naturalization of freed Africans. But, from the research by Cunha and Reis (cited above) and by Côrtes de Oliveira, Maria Inês, O liberto: o seu mundo e os outros, Salvador, 1790/1860 (São Paulo, 1988), p. 30,Google Scholar n. 24; it would seem that few freed Africans in Bahia underwent naturalization.

²⁸ Jz de fora (Maragogipe) to the gov. (20/3/1814), BN-sm, II-33,24,22; jz de fora (Cachoeira) to the pres. (25/3/1827 and 24/9/1828), APEB, SH, 2270; jz de paz (Iguape) to the pres. (24/9/1828), APEB, SH, 2394; Reis, , Slave Rebellion, pp. 49, 59–64.Google Scholar Note that the aggregated tables attached to the censuses of Iguape, S. Gonçalo dos Campos, and S. Pedro Velho all specify the number of freed Africans living in those parishes.

²⁹ Reis, , Slave Rebellion, esp. chaps. 4–12,Google Scholar p. 141, and (for quotation), p. 121.

³⁰ CL, vol. 1, pp. 22–26; Reis, , Slave Rebellion, pp. 221–30;Google Scholar Cunha, , Negros, pp. 75–81.Google Scholar In the mid-18305, 10 mil-réis would have purchased 9.3 alqueires (337.3 liters) of cassava flour (the chief breadstuff in the local diet), enough to provide an adult with a full year's standard ration of flour. Barickman, Bahian Counterpoint, chaps. 3, 4.

It is impossible to know how many freed Africans in Iguape were aware of the deportation law and therefore hid from the census-takers. But a serious undercount seems unlikely given the special exemptions in rural areas and given that the generalized fear of freed Africans would have encouraged special care in enumerating them. Furthermore, the number of Africans (slave and freed) in the censuses of Iguape and S. Gonçalo dos Campos is broadly consistent with the other available information about the population of the two parishes. Also note that, although the proposed “colony” was never established, a significant number of freed Africans did return, often more or less willingly, to West Africa. Verger, Pierre, Fluxo e refluxo do tráfico de escravos entre o Golfo de Benim e a Bahia de Todos os Santos dos séculos XVII a XIX, trans. Gadzanis, Tasso (São Paulo, 1987), pp. 360–68,Google Scholar 529–30, 599–635; Cunha, , Negros, pp. 101–216.Google Scholar

³¹ Reis, , Slave Rebellion, p. 192 Google Scholar (including police chief's statement); idem, “A elite baiana,” p. 382 (for the letter to the minister of justice); and Verger, , Fluxo, p. 359 Google Scholar (for quotation from Martins’ speech).

³² In this regard, the 1835 censuses had a clear precedent in the 1816-17 surveys of Bahian slaveholders, ordered by the colonial government in response to slave revolts and conspiracies, especially a large 1816 uprising. See Schwartz, Stuart B., Sugar Plantations in the Formation of Brazilian Society: Bahia, 1550–1835 (Cambridge, 1985),Google Scholar chap. 16. But, unlike the 1835 censuses, the 1816–17 surveys merely listed the number of slaves owned by each slaveholder, without providing individual-level information about the slaves, and did not collect any information on freed former slaves as such.

³³ The figure of 1,914 excludes 5 freed children of slave mothers (identified as such) and 105 individuals listed (under “occupation”) as family members and as also holding conventional occupations. Likewise, in S. Gonçalo dos Campos, just under 90% of all free residents (excluding “household” heads) had a family relationship of one sort or another as their only recorded “occupation.” I am grateful to Victoria Weinberg for suggesting in this context the terms “professional family member” and “professional kin.”

³⁴ “Agregados” are also found in the 1835 censuses (including 3 individuals so classified in Iguape). Use of “agregado” and “doméstico” varies within and across the censuses; that variation would seem to reflect little more than individual census-taker preferences. The “domésticos” and “agregados” listed within “fogos” in the 1835 censuses should not be confused with individuals who maintained their own households on someone else's land in exchange for odd services and occasionally a nominal rent and who often were also known as agregados.

³⁵ The census lists 241 “domésticos” with no other recorded “occupation” as well as 2 “domésticos” who were also “sons” and 31 who also held conventional occupations, for a total of 274 (11.1% of all free residents excluding “household” heads). The next-most-common “occupations” within the free population were seamstress and farming (lavoura), 179 and 169 individuals, respectively, including heads as well as 44 persons also classified as family members or as “domésticos.”

³⁶ The censuses of S. Gonçalo dos Campos and S. José das Itapororocas also record relationship to the head under “occupation”; in the census of S. Pedro Velho, that information comes in the column for “names,” under subheadings for “wife,” “children,” etc.

³⁷ The census lists a conventional occupation for only one married woman living with her husband: a freed “black” ganhadeira (day laborer), married to a freed “black.” In S. José das Itapororocas and S. Gonçalo dos Campos, wives were also generally listed simply as “wives” under “occupation.” Exceptionally, conventional occupations (always in addition to “wife”) were recorded for about one-third of all wives living with husbands in the Capela de S. Luzia, one of 5 subdistricts in S. Gonçalo dos Campos. Along the same lines, according to Elizabeth Kuznesof, in a 1765 count, census-takers in S. Paulo “often seemed to treat the category of ‘widow’ as if it were an occupation.” “How Does Development Begin? Wealth and Family Strategies in São Paulo, 1765 to 1836,” paper presented at the RMCLAS conference, Portland, Oregon, April 2002, p. 13 (cited by permission).

³⁸ Cf. Anderson, Margo, “(Only) White Men Have Class: Reflections on Early 19th-century Occupational Classification Schemes,” Work and Occupations 21:1 (1994), pp. 5–32;CrossRefGoogle Scholar idem, “The History,” p. 18; Hakim, , “Census Reports,” pp. 554–55;Google Scholar Deacon, “Political Arithmetic.”

³⁹ Moraes e Silva, Antonio de, Diccionario da lingua portugueza …, 3d ed. (1789; Lisbon, 1823),Google Scholar s.v. “família”; “Constituição política,” p. 758 (Tit. 4°, cap. VI, art. 92); Graham, , Patronage, p. 7;Google Scholar Bueno, Pimenta, Direito Público, p. 191.Google Scholar Sons-at-home over 25 could, however, vote if they were public employees.

⁴⁰ Pimenta Bueno, p. 191 ; Pinto Paiva, Maria Arair, Direito político do sufrágio no Brasil (1822–1982) (Brasília, 1985), p. 108.Google Scholar Also Anderson, cf., “(Only) White Men,” esp. pp. 8, 26–27.Google Scholar

⁴¹ On the categories of illegitimacy in Brazilian law, see Lewin, Linda, “Natural and Spurious Children in Brazilian Inheritance Law from Colony to Empire: A Methodological Essay,” TAm 43:3 (1992), pp. 352–96.Google Scholar

⁴² See, e.g., Kusnezof, Elizabeth, “Sexual Politics, Race, and Bastard-bearing in Nineteenth-Century Brazil: A Question of Power or Culture?,” Journal of Family History 16:3 (1991), pp. 241–60;Google Scholar de Almeida Figueiredo, Luciano Raposo, Barrocas familias: vida familiar em Minas Gerais no século XVIII (São Paulo, 1997).Google Scholar

⁴³ Likewise, less than 1% of all slaves were listed as having any “occupation” (e.g., carpenter, driver, seamstress, etc.) other than “slave.” Again it would seem that, when offered information about a slave's occupation (as a specific activity), the census-takers in Iguape—and also in S. Gonçalo dos Campos and S. José das Itapororocas—recorded it, but made no systematic effort to obtain that information. The census of S. Pedro Velho leaves the column for “occupations” blank for slaves, using instead a subheading in the column for “names” to classify slaves as slaves. U.S. censuses also failed to collect information on slave occupations. Anderson, , “The History,” p. 19.Google Scholar By contrast, Bahian post-mortem inventories regularly record conventional occupations for slaves since occupations influenced the slaves’ appraised value.

⁴⁴ All “whites” in the census are classified as “livres” (free; i.e., freeborn). But, insofar as the census designation “white” indicates an absence of known (“black”) African ancestry, “whiteness” does not serve as a clear divider between the free and the slave populations or between slaveholders and slaves since only 17.2% of the free population was “white” and nearly 47% of all slave-owners in Iguape were non-“whites.” See Barickman, “As cores.”

⁴⁵ See, e.g., Sienes, Robert, Na senzala, urna flor: esperanças e recordações da família escrava— Brasil Sudeste, século XIX (Rio de Janeiro, 1999),Google Scholar which also cites many of the other studies on slave families.

⁴⁶ The census lists only 5.2% of all slaves in Iguape (7.2% of those 18 or older) as married or widowed. The census-takers in Velho, S. Pedro took the failure to record slave kinship a step further by not even registering the marital status of slaves. In “Família escrava em Lorena (1801),” EE 17:2 (1987), p. 247,Google Scholar Iraci del Nero da Costa, Robert W. Sienes, and Stuart B. Schwartz note that, “as a rule,” Brazilian household lists do not provide clear information about slave families. Yet there are exceptions. See, e.g., ibid., pp. 248–95; Motta, José Flávio, Corpos escravos, vontades livres: posse de escravos e família escrava em Bananal (1801–1829) (São Paulo, 1999);Google Scholar and the census of S. José das Itapororocas, which, unlike the other three 1835 Bahian censuses, does more or less regularly record slave family ties. The exceptions suggest that the census-takers in Iguape could have registered family ties amongst slaves and that their failure to do so should be seen as a deliberate oversight.

⁴⁷ Malheiro, Perdigão, A escravidão, vol. 1, pp. 58–61 Google Scholar (p. 61 for quotation). On kinlessness as a fundamental component in the definition of slavery, see Finley, M.I., Ancient Slavery and Modern Ideology (Harmondsworth, Middx, 1983), pp. 75–77;Google Scholar Patterson, Orlando, Slavery and Social Death: A Comparative Study (Cambridge, Mass., 1982), pp. 7–9;Google Scholar and Meillassoux, Claude, The Anthropology of Slavery: The Womb of Iron and Gold, trans. Dasnois, Alide (Chicago, 1991), pp. 9–140.Google Scholar Also note that the U.S. Congress rejected proposals to list slaves in the 1850 census by name (rather than number) and to identify the mothers of slave children. Anderson, , “The History,” 19.Google Scholar For the 1869 law, see Brazil, Collecção das leis do Imperio do Brasil, tomo 29, pt. 1 (Actos do Poder Legislativo) (Rio de Janeiro, 1869), p. 129 (Decreto n. 1695).

⁴⁸ Moraes e Silva, Diccionario, s.vv. “fogo,” “família,” “pai.” Also see Bluteau, Raphael, Vocabulario portuguez, e latino, …, 8 vols. (Coimbra, 1712–21), s.vv. “familia,” “fogo”;Google Scholar and Graham, , Patronage, 17.Google Scholar The percentage cited in the text excludes the 7 “fogos” with absent heads.

⁴⁹ By contrast, widowed mothers were not listed as “household” heads in a 1798 census of Ubatuba (S. Paulo) once their sons had reached maturity, according to Metcalf, Alida C., “Recursos e estruturas familiares no século XVII, em Ubatuba, Brasil,” EE 13:no. esp. (1983), p. 783.Google Scholar

⁵⁰ Consider, e.g., the already mentioned cane-farming “fogo” headed by dona Ana Joaquina do Amor Divino; while she would have owned one-half of the “household's” property (which included 24 slaves), the share belonging to José Bento, her 26-year-old son, would have amounted to only one-fourteenth of that property. Conversely, lack of property would seem to explain, for the most part, the 11 generally poorer “fogos” in which adult sons appear as heads and their mothers were classified as “mothers.” On inheritance laws, see Lewin, “Natural and Spurious Children.”

⁵¹ The colonel's 1839 post-mortem inventory includes an appraisal for an African-born slave named Jerônimo, described as being “de boa idade” (of good age; i.e., middle-aged), but not yet “velho” (old). Inv. of Domingos Américo da Silva (1839), APEB, SJ, IT. It thus seems safe to assume that Jerônimo belonged to the colonel and was allowed to live separately with his freed wife. Between the Engenhos S. Catarina and S. Domingos, the census lists 3 other “fogos”: nos. 397 and 398, headed by free workers employed at the plantations; and no. 400, left blank, suggesting that it was an unoccupied dwelling.

⁵² Samara, Eni de Mesquita, As mulheres, o poder e a família: São Paulo, século XIX (São Paulo, 1989), p. 25 Google Scholar n. 41 ; Laslett, Peter, Introduction to Household and Family in Past Time, ed. Laslett, Peter, assist. Wall, Richard (Cambridge, 1972), pp. 23–25.CrossRefGoogle Scholar Among the few authors to offer a substantially different definition are Eisenberg, Peter, Os homens esquecidos: escravos e trabalhadores livres no Brasil, séculos XVIII e XIX (Campinas, 1989), p. 333;Google Scholar and Slenes, , Na senzala, pp. 71,Google Scholar 117. Also cf. de Castro Faria, Sheila, A Colônia em movimento: fortuna e família no cotidiano colonial (Rio de Janeiro, 1998), pp. 355–90.Google Scholar

⁵³ In like fashion, many artisan-headed “fogos” in the parish no doubt united dwelling and work spaces.

⁵⁴ The percentage cited in the text, which also takes into account “fogos” headed by cane farmers, is based on the 62 individuals with the following “occupations”: engenho administrator (administrador de engenho), nurse (ama, enfermeiro, enfermeira, ocupada do serviço das crias), crater (caixeiro), engenho employee (servente de engenho, serventuário), overseer and head overseer (feitor, feitor-mor), kettleman, sugar master, assistant sugar-master (banqueiro), molasses-hauler (condutor de mel), and still-master (alambiqueiro).

⁵⁵ Inv. of Domingos Américo da Silva (1839), APEB, SJ, IT. Given the concentrated pattern of land-holding in Iguape, most free engenho employees (even those living in their own “fogos”) and the bulk of the free population would have lived on land owned by planters. On landholding in the parish, see Barickman, , Bahian Counterpoint, pp. 113–18.Google Scholar Also note that determining whether census “fogos” were both residential and production units requires a prior definition of what constituted a production unit.

⁵⁶ The same holds trae for the other three 1835 censuses and also, apparently, for other Brazilian censuses. But, because of differences in economic setting, large slaveholding “fogos” were less common in the other three Bahian parishes than in Iguape. Yet, even the census of urban S. Pedro Velho includes “fogos” such as rua de João Pereira, no. 78, with 22 members, 19 of whom were slaves.

⁵⁷ Invs. of Domingos Américo da Silva (1839), APEB, SJ, IT; of Manoel Estanislau de Almeida (1838), ARC, IT; and of Clara Maria do Sacramento and Joaquim da Costa e Melo (1832), ARC, IT. Postmortem inventories of planters and cane farmers in the Recôncavo suggest that 2–5 slaves typically inhabited a senzala. Also see the sketch of a slave hut (Negerhütte) at a Bahian engenho, by Naeher, Julius, Land und Leute in der brasilianischen Provinz Bahia (Leipzig, 1881), p. 99.Google Scholar Hut lists with information about domestic groups among rural slaves are rare in Brazil, but see Graham, Richard, “Slave Families on a Rural Estate in Colonial Brazil,” Journal of Social History 9:3 (1976), pp. 82–402;CrossRefGoogle Scholar and Schwartz, , Sugar Plantations, pp. 394–406.Google Scholar

⁵⁸ In fact, in Family life and illicit love in earlier generations: Essays in historical sociology (Cambridge, 1977), chap. 7, Laslett refers to slave “domestic groups” on U.S. plantations as “households,” applying to them the same methods he uses in analyzing households in European tax lists, censuses, etc. Yet his discussion relies on private plantation records. U.S. censuses did not list slave “households.” Anderson, , “The History,” pp. 17–19.Google Scholar In 6 “fogos” in Iguape, the first name listed belongs to a slave, but they were engenhos or cane farms with absentee owners or tenants or recently deceased owners whose estates were still in probate and are identified as such.

⁵⁹ See the description of the huts (choupanas) occupied by the rural poor in Vilhena, , A Bahia, vol. 3, p. 927;Google Scholar the photos (c. 1890) of thatched huts in Ferrez, Gilberto, Bahia: velhas fotografías (Rio de Janeiro, 1986), pp. 164–67,Google Scholar 169–70; and Barickman, B.J., ‘“A bit of land, which they call roça’: Slave Provision Grounds on Sugar Plantations and Cane Farms in the Bahian Recôncavo, 1780–1860,” HAHR 74:4 (1994), pp. 649–87 (on provision grounds); and Sienes, chap. 3 (more generally, on senzalas as co-residential domestic groups).Google Scholar

⁶⁰ Here it might be tempting to invoke Laslett's term “houseful,” which, he argues, is to be used where several “households” in a census occupy the same “premises”; i.e., a single building divided into discrete residential units or a “number of conjoined or contiguous buildings” such as “farmhouse with a yard surrounded by outhouses … suitable for occupation.” Introduction, p. 36. In some contexts, Laslett's distinctions between “household,” “houseful,” and “premises” would no doubt be useful, but not in dealing with the larger slave-owning “fogos” in the 1835 Bahian censuses. The censuses do not in any way distinguish within those “fogos” different co-residential groups of slaves, domestic slaves (who as such may have routinely slept in their master's house) from field hands, or even slave quarters from the owner's dwelling space. Classifying slave “family” units, where they can be detected, as “households” would also entail using a definition of “fogo” not employed by the census-takers and would violate Laslett's own dictum that clearly divided blocks of names in censuses “must” be regarded as “households” or “families” (p. 24). Also see, e.g., Berkner, Lutz K., “The Use and Misuse of Census Data for the Historical Analysis of Family Structure,” Journal of Interdisciplinary History 5:4 (1975), pp. 721–38;CrossRefGoogle Scholar and Rebel, Hermann, “Dispossession in the Communal Memory: An Alternative Narrative About Austria's Descent into the Holocaust,” Focaal 26–27 (1996), p. 170.Google Scholar

⁶¹ Perdigão Malheiro, vol. 1, pp. 35–36, 58 (emphasis in original). As often noted, the 1824 constitution excluded slaves from “the political community” by not even mentioning the word “slave” in any of its 179 articles.

⁶² Thus, in part, the 1835 census bears out the classic arguments about the patriarchal family as the fundamental social unit in colonial and nineteenth-century Brazil put forth (with different emphases) by Gilberto Freyre, Sérgio Buarque de Holanda, and Antônio Candido (de Mello e Souza). I explore the matter as well as some of the implications for census-based research on family and other social-demographic issues in late colonial and early nineteenth-century Brazil in “Households of the rich (but not necessarily famous): Plantation and wealthy cane-farming households in early nineteenth-century Bahia,” paper presented at the RMCLAS conference, Portland, Ore., April 2002. Also see Kuznesof, Elizabeth Anne, Household Economy and Urban Development: São Paulo, 1765 to 1836 (Boulder, Col., 1986), p. 5, n. 11;Google Scholar Vainfas, Ronaldo, Tropico dos pecados: moral, sexualidade e inquisição no Brasil (Rio de Janeiro, 1989), pp. 107–113;Google Scholar and Faria, , A Colônia, pp. 47–52;Google Scholar and, for recent historical studies of the family in the Northeast, Lewin, Linda, Politics and Parentela in Paraíba: A Case Study of Family-Based Oligarchy in Brazil (Princeton, 1987);CrossRefGoogle Scholar Queirós Mattoso, Katia de, Família e sociedade na Bahia do século XIX (Sào Paulo, 1988);Google Scholar and Borges, Dain, The Family in Bahia, Brazil, 1870–1945 (Stanford, 1992).Google Scholar

⁶³ Graham, , Patronage, pp. 17–23 Google Scholar (17 for quotation). Although used in Brazil since the 1821 election of deputies to draft a constitution for what was then the United Kingdom of Portugal, the Algarves, and Brazil, fogo for electoral purposes lacked a legal definition until an 1842 decree defined it as “a house, or part thereof, inhabited independently by a person or family so that a single building may contain two or more fogos.” Porto, Walter Costa, Dicionário do voto (São Paulo, 1995), pp. 173–74.Google Scholar Before and perhaps also after 1842, defining and counting electoral fogos could generate local political disputes. On one such dispute in a parish near Iguape, see vig. (Nazaré) to the pres. (11/9/1836), APEB, SH, 5212.

⁶⁴ de Mattos, limar Rohloff, O tempo saquarema (São Paulo and Brasília, 1987), chap. 2, esp. pp. 119–20;Google Scholar Moraes e Silva, Diccionario, s.vv. “familia,” “fogo,” “casa”; Bluteau, Vocabulario, s.vv. “familia,” “fogo.”

⁶⁵ See, e.g., M[anoel] F[erreira] da C[amara Bittencourt e Sá], “Carta II,” in de Brito e outros, João Rodrigues, Cartas economico-politicas sobre a agricultura e commercio da Bahia, dadas á luz por I.A.F. Benevides (Lisbon, 1821), p. 84;Google Scholar de Holanda, Sérgio Buarque, preface to Davatz, Thomas, Memórias de um colono no Brasil (1850), trans, and annot. de Holanda, Sérgio Buarque (São Paulo, 1941), pp. 26–27;Google Scholar Graham, , Patronage, p. 20.Google Scholar On Latin usage, see Glare, P.G.W., ed., Oxford Latin Dictionary (Oxford, 1972),Google Scholar s.v. “familia”; Herlihy, David, Medieval Households (Cambridge, Mass., 1985), pp. 2–3.Google Scholar

⁶⁶ See Mattos, , O tempo, pp. 119–20.Google Scholar Also see Ortner's, Sherry B. general remarks on the family as an administrative unit in “The Virgin and the State,” Feminist Studies 4 (1978), p. 29.CrossRefGoogle Scholar

⁶⁷ de Almeida e Arnizáu, José Joaquim, “Memoria topographica, histórica, commercial e politica da Villa da Cachoeira…” (1825), Revista do Instituto Histórico e Geográfico Brasileiro 25 (1862), pp. 131, 134.Google Scholar

⁶⁸ Ibid., pp. 134, 141–42. Also see Vilhena, , vol. 3, pp. 925–27.Google Scholar

⁶⁹ Schwarz, Roberto, Ao vencedor as batatas: forma literária e processo social nos inícios do romance brasileiro (São Paulo, 1981),Google Scholar chap. 1. But Carvalho Franco, cf. Maria Sylvia de, “As idéias estão no lugar,” Cadernos de debate, 1 : História do Brasil, 2d ed. (São Paulo, 1976), pp. 61–64.Google Scholar

⁷⁰ See, e.g., Kraay, , Race, esp. pp. 5–6;Google Scholar Bosi, Alfredo, Dialética da colonização (São Paulo, 1992)Google Scholar, chap. 7; da Cunha, Manuela Carneiro, Antropologia do Brasil: mito, história, etnicidade (São Paulo, 1986), pp. 139–44;Google Scholar Graham, , Patronage, pp. 72–79;Google Scholar Costa, , Da monarquia, pp. 109–126.Google Scholar Also cf. Malerba, Jurandir, Os broncos da lei: liberalismo, escravidão e mentalidade patriarcal no Império do Brasil (Maringá, 1994).Google Scholar

⁷¹ Anderson, , “The History,” pp. 17–20;Google Scholar Barker, Diana L., “Regulation of Marriage,” in Power and the State, ed. Littlejohn, Gary et al. (London, 1978), p. 256.Google Scholar Also see Smith, Hilda L., All Men and Both Sexes: Gender, Politics, and the False Universal in England, 1604–1832 (University Park, PA, 2002);Google Scholar and Pateman, Carole, The Disorder of Women: Democracy, Feminism and Political Theory (Cambridge and Oxford, 1984).Google Scholar