Slaves, Immigrants, and Freemen in the 19th-century
Transcrição
Slaves, Immigrants, and Freemen in the 19th-century
1 Ibmec MG Working Paper – WP14 Slaves, Immigrants, and Freemen in the 19th-century Brazilian Labour Market Sérgio de Oliveira Birchal (Ibmec MG) [email protected] (2004) 2 Slaves, Immigrants, and Freemen in the 19th-century Brazilian Labour Market Sérgio de Oliveira Birchal, Ibmec Business School I would like to thank Professors Colin M. Lewis, Mary Morgan, Janet Hunter, and Gareth Austin, from the Department of Economic History of the LSE, for reading and commenting previous versions of this paper. Address: Rua Curitiba, 2427/1201 – Lourdes – Belo Horizonte – Minas Gerais – Brazil – CEP 30170.121 Phone: (005531 33357721) Email: [email protected] 3 Slaves, Immigrants, and Freemen in the 19th-century Brazilian Labour Market Introduction Demographers have long acknowledged the importance of freemen in the composition of the population of 19th-century Brazil. At least half a century before the abolition of slavery, Brazil had the largest free coloured population of any slave society in the Americas. Coloured freemen were the largest single racial group within Brazil. This was even more true for 19th-century Minas Gerais1. However, economic historians have failed to spot the importance of free coloureds within the Brazilian working class during that period. They have largely been stranded in the juxtaposition between slave and immigrant labour, which has largely influenced the debate over the Brazilian labour market during the 19th century2. The general assumption has been that slave and, later, immigrant workers constituted the main sources of labour for coffee production and the emerging manufacturing industry. Non-slave Brazilian workers were considered unfit for disciplined and regular work and, therefore, were only marginally and sporadically employed. A large number of Brazilian economic historians have conceded that non-slave and poor Brazilians 1 See H.S. Klein and F.V. Luna, ‘Free Colored in a Slave Society: São Paulo and Minas Gerais in the Early Nineteenth Century’, The Hispanic American Historical Review, 2000, 80:4, pp.913-41; H.S. Klein and C.A. Paiva, ‘Freedmen in a Slave Economy: Minas Gerais in 1831’, Journal of Social History, Volume 29, Number 4, Summer 1996, pp.933-62; and H.S. Klein, ‘The Colored Freedmen in Brazilian Slave Society’, Journal of Social History, Volume 3, Number 1, Fall 1969, pp.30-52. 2 See, among others, E. Viotti da Costa, ‘1870-1889’, In: L. Bethell (ed.), Brazil: Empire and Republic, 1822-1930, Cambridge: Cambridge University Press, 1989, pp.161-213; W. Dean, ‘Economy’, In: L. Bethell (ed.), Brazil: Empire and Republic, 1822-1930, Cambridge: Cambridge University Press, 1989, pp. 217-56; B. Fausto, ‘Society and Politics’, In: L. Bethell (ed.), Brazil: Empire and Republic, 1822-1930, Cambridge: Cambridge University Press, 1989, pp.257-307; D.H. Graham, Population and Economic Development in Brazil, 1808 to the Present, Baltimore: Johns Hopkins University Press, 1979; R. Faoro, Os Donos do Poder, São Paulo: Globo, 1975; F. Fernandes, A Revolucao Burguesa no Brasil, Rio de Janeiro: Zahar Editores, 1975; S.B. Holanda, Raízes do Brasil, Rio de Janeiro: José Olympio, 21st ed. 1982,; M.S.C. Franco, Homens Livres na Ordem Escravocrata, São Paulo: USP, 1969; L. Kowarick, Trabalho e Vadiagem: A Origem do Trabalho Livre no Brasil, São Paulo: Brasiliense,1987; R. Graham, Britain and the Onset of Modernization in Brazil,1850-1914, Cambridge: Cambridge University Press, 1968; W. Dean, The Industrialization of São Paulo, 1880-1945, Austin: The University of Texas Press, 1969; S. Silva, Espansão Cafeeira e Origens da Indústria no Brasil, São Paulo: Alfa-Omega, 1976; T.H. 4 lived a vagrant life throughout the countryside and the cities, employed mainly in subsistence activities or in the most stagnated sectors of the economy. However, is that true? Given the size of free coloureds within the Brazilian population, is it possible that so many people have been so marginalized from the main economic activities in the country throughout the 19th century? More recently, this general assumption has been increasingly challenged3. Nevertheless, to date very few works have devoted themselves to the study of the employment of non-slave Brazilians in the most dynamic economic activities, especially in the emerging manufacturing industry in the South Eastern part of the country. This paper shows that in an important coffee growing province/state like Minas Gerais, which had the largest slave population throughout the 19th century, there is plenty of evidence that non-slave national labour was an alternative to slaves in several sectors. In the emerging industry of Minas Gerais non-slave Brazilians were widely employed and worked side-by-side with slaves. Immigrant workers were numerically insignificant and did not have the large participation that they had in the economies of Rio de Janeiro and São Paulo. Therefore, this work intends to investigate the labour market in 19th-century Brazil by describing and analysing the sources of labour available to mineiro4 entrepreneurs. The work is divided into three sections. The first briefly reviews the debate about the labour market in 19thcentury Brazil. The second section investigates the composition of the mineiro population in the Holloway, Immigrantson the Land: Coffee and Society in Sao Paulo, 1886-1934, Chapel Hill: University of North Carolina Press,1980; and E. Viotti da Costa, Da senzala a Colônia, São Paulo: Difusão Européia do Livro, 1966. 3 See, among others, M.L. Lamounier, ‘The “Labour Question” in Nineteenth Century Brazil: Railways, Export Agriculture and Labour Scarcity’, London: LSE/Dept. of Economic History, Working Paper No. 59/00, 2000; S.O. Birchal, Entrepreneurship in Nineteenth-Century Brazil: The Formation of a Business Environment, London: Macmillan, 1999; Klein and Luna, ‘Free Colored in a Slave Society’, pp.913-41; Klein and Paiva, ‘Freedmen in a Slave Economy’, pp.933-62. 4 Mineiro relates to things or people belonging to Minas Gerais 5 19th century. The third section describes and analyses the origin of the labour force employed in mineiro industry during the period 1810-1910. 1. A Brief Review of the Literature As mentioned above, the debate about the labour market in Brazil during the 19th century was basically restricted to counter posing slave and immigrant labour. The greater attention placed by historians upon these two sources of labour is due to their wide employment in the more dynamic sectors of the economy. Slaves constituted the principal source of labour for the main economic activity in the country – the production of coffee. Furthermore, as the abolition of slavery became inevitable, immigrant labour began to replace slave5. As historians devoted a disproportionate amount of attention to export activities, the composition of the labour force in manufacturing was largely ignored, and hardly considered as a source of employment for nonslave Brazilians. Historians inherited a prejudice against non-slave Brazilians, which was widely shared by the 19th century Brazilian elite. During the 19th century, rulers, policy-makers, and many contemporaries, tended to see non-slave Brazilians as vagabonds and unfit for regular, disciplined work. The so-called nonslave Brazilian – white, black, mulattos, and mestiços – was the least desired worker on coffee plantations6. The existence of slavery had also an important influence upon the work ethic in 19th-century Brazil. On the one hand, by degrading work to an extreme degree slavery developed the idea among poor free Brazilians that manual work was repulsive7. On the other hand, free labour had a particular meaning to the farmer: it was labour free from the burden of the trafficker’s 5 See, among others, Dean, The Industrialization of São Paulo, 1880-1945, pp.3-15; Graham, Britain and the Onset of Modernization in Brazil,1850-1914, pp.160-86; W. Baer, The Brazilian Economy: Growth and Development, Wesport: Praeger, 5th ed. 2001, p.17, and Kowarick, Trabalho e Vadiagem, pp. 35-41. 6 Kowarick, Trabalho e Vadiagem, pp.64, 68-9. 7 Holanda, Raízes do Brasil, pp.26-8. 6 charges8. Thus, employers in general (planters, officials, and entrepreneurs) shared these prejudices against the non-slave Brazilians, who they saw as indolent or culturally biased against regular and disciplined employment9. In 1856, for example, Mariano Procópio Ferreira Lage, a wealthy farmer10 and founder and chairman of the União e Indústria Turnpike, justified the recruitment of foreigners on the grounds that non-slave Brazilians were unreliable and lacked regular habits of work11. In 1869, this prejudice against non-slave Brazilians could also be depicted from the Brazilian Minister of Agriculture, Commerce and Public Works’ complaints about the lack of chances of employment for them in the industrial sector in Rio de Janeiro12. Three decades later, the owners of textile mills in Minas Gerais made the same kind of remarks about non-slave Brazilians13. In fact, textile industrialists were reported to treat their employees in the same way coffee farmers or sugar-mill owners treated their slaves14. As Carvalho has observed: “Although in the literature there is almost a consensus that there was an advance of modernity in Brazil after 1870, tradition was sufficiently strong to maintain the values of a rural, patriarchal and hierarchical society.”15 Even modernizers themselves took racist theories as the latest advances of science, which propagated disbelief in the black and brown population’s capacity to civilisation, and from this 8 J. Souza Martins, O Cativeiro da Terra, São Paulo: Livraria Editora Ciencias Humanas, 1979, pp.9-62. Lamounier, ‘The “Labour Question” in Nineteenth Century Brazil’, p.47. 10 See Birchal, Entrepreneurship in Nineteenth-Century Brazil, p.28; and W.L. Bastos, Mariano Procopio Ferreira Lage: Sua Vida, Sua Obra, Descendencia, Genealogia, Juiz de Fora: Paraibuna, 2nd ed. 1991, pp.15-270. 11 Companhia União e Indústria, Relatório da Assembléia Geral dos Acionistas, Rio de Janeiro, 1856, p.13. 12 Ministério da Agricultura, Commercio e Obras Públicas, Relatório da Repartição dos Negócios da Agricultura, Commercio e Obras Públicas, Rio de Janeiro, 1869, p. 23. 13 In 1884, Francisco de Paula Mascarenhas, one of the founders and owners of the Cachoeira mill and of the Companhia Cedro e Cachoeira complained to his elder brother about the difficulties to recruit people. See Companhia Cedro e Cachoeira, ‘Letter from Francisco de Paula Mascarenhas to Bernardo Mascarenhas, 15 September 1884, In ‘Caixa de Correpondências Recebidas No. 16’. 14 S.J. Stein, Origens e Evolução da Indústria Têxtil no Brasil, 1850-1950, Rio de Janeiro: Campus, 1979, p.63. 15 J.M. Carvalho, ‘Brazil 1870-1914. The Force of Tradition’, Journal of Latin American Studies, Vol. 24, Quincentenary Supplement: The Colonial and Post Colonial Experience. Five Countries of Spanish and Portuguese America (1992), p.145. 9 7 came the faith of many modernizers in the redeeming role of European immigration16. Therefore, modernizers’ rhetoric surely reinforced landowners’ complaints about labour scarcity as, among other things, a way to press government to subsidise the costs of labour reproduction as abolition of slavery had become inevitable. As mentioned before, many Brazilian economic historians shared this general assumption17. Thus, the mainstream literature has largely reproduced the view that non-slave Brazilians were restricted to subsistence activities, to occasional jobs or to begging in the countryside and in the cities. In the coffee growing areas, until the beginning of the large inflow of immigrants – which coincided with the abolition of slavery – poor and non-slave Brazilians were thought to take part in the productive process only on a casual basis18. Non-slave nationals were portrayed as moving constantly and occasionally rendering services to the large farms. As long as production continued to be based on slave labour, the large and increasing contingent of poor non-slave Brazilians would continue to be excluded from the productive system. In the coffee regions, especially in São Paulo, Brazilians only began to be employed after the abolition of slavery, even so on a casual basis. In the new dynamic zones of Eastern São Paulo, successive batches of immigrants would be employed, whereas non-slave Brazilians were employed in the stagnated areas where the immigrant did not go19. Furthermore, by looking the marginalization of non-white labourers in São Paulo’s post-abolition economy and society Andrews stresses how the legacy of slavery helped to shape the evolution of interactions among employers, workers 16 Carvalho, ‘Brazil 1870-1914. The Force of Tradition’, pp.156-7. In his paper commemorating the hundred years of the abolition of slavery in Brazil, Levine quotes several works that reinforce the idea that non-slave Brazilians were almost absent from the 19th-century Brazilian labour market. R.M. Levine, ‘”Turning on the Lights”: Brazilian Slavery Reconsidered One Hundred Years after Abolition’, Latin American Research Review, Volume XXIV, Number 2, 1989, pp.201-17. 18 See, for example, M.L. Lamounier, ‘Between Slavery and Free Labour: Experiments with Free Labour and Patterns of Slave Emancipation in Brazil and Cuba c. 1830-1888’, unpubl. PhD diss., University of London, 1993, p. 185. 17 8 (both white and non-white), and the state. Slavery produced, on the one hand, employers unaccustomed and unwilling to bargain with their former slaves and, on the other hand, a former slave population with very specific demands concerning the conditions under which they would work as free labourers. In the specific case of São Paulo, the flooding of the labour market with European immigrants, subsidized by the state, undercut labourers’ power of bargain 20. Thus, according to the mainstream literature, wherever there was a large inflow of immigrants they tended to fill the places in the productive activities and reduced substantially the employment of non-slave Brazilians. Even the emerging Brazilian industry hardly constituted an alternative for the mass of non-slave Brazilian workers21. In São Paulo, for example, immigrants predominated in the several sectors of the urban economy, particularly in the manufacturing activities, where the employment of Brazilians was small. As happened in the prosperous coffee regions, the Brazilian worker was only occasionally employed in the paulista22 industry. The number of immigrants was sufficient to supply paulista industry23. Thus, the emerging industry of São Paulo did not need to resort to the Brazilian labour force. Even authors looking specifically at the case of 19th-century Minas Gerais industry have reinforced this view24. Martins Filho and Martins, for example, in their seminal work on slavery in the 19th-century Minas Gerais, show that although Minas Gerais (unlikely São Paulo and Rio de Janeiro) was a non-export economy, it had the 19 Kowarick, Trabalho e Vadiagem, pp. 64, 68-9. G.R. Andrews, ‘Black and White Workers: Sao Paulo, Brazil, 1888-1928’, The Hispanic American Historical Review, Vol. 68, No. 3, August 1998, pp.491-524. 21 Kowarick, Trabalho e Vadiagem, pp. 65-8. 22 Paulista relates to the people or things belonging to São Paulo 23 Kowarick, Trabalho e Vadiagem, pp. 101-6. 24 See, among others, A. Martins Filho and R.B. Martins, ‘Slavery in a Nonexport Economy: Nineteenth-Century Minas Gerais Revisited’, The Hispanic American Historical Review, Volume 63, Number 3, August 1983, pp.53568; J..A. Paula, ‘Dois Ensaios sobre a Gênese da Industrialização em Minas Gerais: a Siderurgia e a Indústria Têxtil’, Anais do II Seminário sobre a Economia Mineira, Belo Horizonte, 1983, p.19-73; R.B. Martins, Growing in Silence: The Slave Economy of Nineteenth-Century Minas Gerais, unpubl. PhD diss., Vanderbilt University, 1980; D.C. Libby, Transformação e Trabalho em uma Economia Escravista: Minas Gerais no Século XIX, São Paulo: Brasiliense, 1988. 20 9 largest slave population in Brazil during the century. In trying to explain this apparent paradox – the sheer size of the mineiro slave population – they show that slaves were widely employed in a large number of non-agricultural sectors, such as cotton textile and iron manufacture25. Of course, where there was no inflow of immigrants, non-slave Brazilian workers were indeed more widely employed. In the Northeast, for example, after 1850, as the slave regime began to lose its hegemony in the sugar economy, with the migration of slaves, non-slave Brazilians were increasingly employed26. In several Northeastern provinces, the number of nonslave Brazilian workers employed in sugar production increased during the 19th century27. But it was not only in the Northeast and in the agricultural sectors that the wide employment of nonslave Brazilians can be observed. In Minas Gerais, the alternative to slave labour in nonagriculture sectors was mainly non-slave Brazilian workers. However, 19th-century Brazilian labour literature has been revised in several different directions. Martins Filho and Martins also argue that the scarcity of a voluntary supply of wage labour in 19th-century Minas Gerais was due to the existence of an open wide frontier, which offered free peasants plenty of land from which they could obtain an independent subsistence28. Lamounier, more recently, has argued that geographic mobility of non-slave labourers was the result of the seasonal, short-term nature of free worker employment in a slave-based rural society. Looking at the construction of paulista railways Lamounier observes that contractors were well supplied with labour. Free national labourers were locally available and could be 25 Martins Filho and Martins, ‘Slavery in a Nonexport Economy’, p.563. See, among others, B.J. Barickman, ‘Persistence and Decline: Slave Labour and Sugar Production in the Bahian Reconcavo, 1850-1888’, Journal of Latin American Studies, Vol. 28, No. 3, Brazil: History and Society (Oct. 1996), pp.581-633; J.H. Galloway, ‘The last Years of Slavery on the Sugar Plantation Of Northeastern Brazil’, The Hispanic American Historical Review, Vol. 51, No. 4 (Nov., 1971), pp.586-605; and P.L. Eisenberg, ‘Abolishing Slavery: The Process on Pernambuco’s Sugar Plantations’, The Hispanic American Historical Review, Vol. 52, No. 4 (Nov., 1972), pp.580-97. 27 Lamounier, Between Slavery and Free Labour, pp. 339-40. 26 10 rapidly trained to undertake semi-skilled tasks. Nevertheless, if the seasonable nature of agriculture employment suited the needs of railway contractors, it was a cause of concern for those requiring a substantial and constant labour force, because free national labourers usually resisted giving up links with their subsistence plots during planting and harvesting seasons29. Furthermore, Klein and Luna show that as soon as the early 19th century, there was a fair amount of economic mobility among free persons of colour in both Minas Gerais and São Paulo. Although they were not dominant in government positions and/or other highly socially regarded professions, free coloured Brazilians were an integrated mass of labourers who shared most of the characteristics of all free born and white population. In fact, they were an important, competitive, and integrated element within Brazilian society30. Thus, more recent works have began to portray a new picture of the free coloured. In the following section we will briefly investigate the composition of the mineiro population in order to show the importance of non-slave Brazilians within the mineiro labour force. 2. The Mineiro Population With one of the largest populations in Brazil during the 19th century31, Minas Gerais had also the largest slave population32. Furthermore, slaves constituted a large proportion of the 28 Martins Filho and Martins, ‘Slavery in a Nonexport Economy’, p.566. Lamounier, ‘The “Labour Question” in Nineteenth Century Brazil’, p.46. 30 Klein and Luna, ‘Free Colored in a Slave Society’, pp.913-41. 31 Although 19th-century population data may be unreliable (sometimes the several estimates are contradictory and often they were not rigorously collected) most estimates and the 1872 census point to the same conclusion: Minas Gerais had one the largest populations in Brazil during the 19th century. In the first decade of the 19th century mineiro population amounted to 350,000, making Minas the most populous province. In the 1820s, the population was estimated at around 600,000, increasing to more than 900,000 in the 1830s and to around 1,500,000 in the middle of the century. An estimate of 1870 calculated the population of the province in only 631,885 people, nearly a third of the estimate for 1867 (1,600,000) and 1869 (1,600,000). Nevertheless, even this low estimate confirmed the primacy of Minas as Brazil's most populated province. The census of 1872, found that MinasGerais was the most populous province with 2,039,735 people, a dominance confirmed by the censuses of 1890 and 1900. See Ministerio da Agricultura, Industria e Commercio, Directoria Geral de Estatistica, Recenseamento do Brazil realizado em 1 de 29 11 mineiro population33. If the size of the slave population in itself is an indication of the importance of slaves as a source of labour during the 19th century, this becomes even clearer when we take the relative participation of slaves in the labour force. During the period 1831-40 slaves represented nearly 37% of the mineiro labour force. In 1872, although the participation of slaves in the mineiro labour force was not as large as in Rio de Janeiro, where slaves constituted more than 46% of the total labour force, it was still very significant. Slaves still made up nearly 19% of the mineiro labour force. But non-slave Brazilians constituted the largest group of the mineiro population. Non-slave Brazilians represented more than 80% of the total population in 187234. Hence, non-slave Brazilians and nationals were by far the most important source of labour in 19th-century Minas Gerais35. Table 1 – Percentage of Foreigners in the Population of Minas Gerais and São Paulo, 1872, 1890, 1900, and 1900. 1872 1890 1900 1920 2.28% 1.49% 4.10% 1.48% Minas Gerais 3.67% 5.73% 20.22% 22.07% São Paulo Sources: Adapted from Ministerio da Agricultura, Industria e Commercio, Directoria Geral de Estatistica, Recenseamento do Brazil Realizado em 1 de Setembro de 1920, (Rio de Janeiro, 1924), Vol. IV, 1st part, p.LXIII and p.300-3. In contrast, foreigners constituted a small proportion of the mineiro population. As shown in Table 1, foreigners represented only 2.28% of the total population of the province in 1872, Setembro de 1920: Resumo Historico dos Inqueritos Censitarios Realizados no Brazil, Rio de Janeiro, 1922, I, pp.403-23. 32 See , A. Martins Filho and R.B. Martins, ‘Slavery in a Nonexport Economy: Nineteenth-Century Minas Gerais Revisited’, Hispanic American Historical Review, Volume 63, Number 3, August 1983, pp.535-68; H.S Klein and F.V. Luna, ‘Free Colored in a Slave Society: São Paulo and Minas Gerais in the Early Nineteenth Century’, The Hispanic American Historical Review, Vol. 80, No. 4, Special Issue: Colonial Brazil: Foundations, Crises, and Legacies, November 2000, pp.913-41; F. Iglésias, Política Econômica do Governo Provincial Mineiro: 1835-1889, Rio de Janeiro: Instituto Nacional do Livro, 1958, p.130; and Directoria Geral de Estatistica, Relatorio Annexo ao do Ministerio dos Negocios do Imperio de 1876, Rio de Janeiro, 1877, pp.5-8. 33 See Ministerio da Agricultura, Industria e Commercio, Directoria Geral de Estatistica, Recenseamento do Brazil realizado em 1 de Setembro de 1920: Resumo Historico dos Inqueritos Censitarios Realizados no Brazil, Rio de Janeiro, 1922, I, pp.403-23; and Libby, Transformação e Trabalho, p.47. 34 See Ministério da Agricultura, Indústria e Commercio, Recenseamento do Brazil Realizado em 1 de Setembro de 1920, Rio de Janeiro, 1924, IV, 2nd part, p.48 and Directoria Geral de Estatística, Relatorio Annexo ao do Ministerio dos Negocios do Imperio de 1876, Rio de Janeiro, 1877, p.15. 12 whereas in São Paulo foreigners represented almost 4% of the total population and was increasing rapidly in the following decades. In 1890, for example, the proportion of foreigners in Minas Gerais decreased to 1.49%, while in São Paulo the participation of foreigners increased to 5.73%. At the beginning of the 20th century, foreigners represented around 4% of the mineiro population. Nevertheless, at the same time, foreigners’ participation in the paulista population jumped to over 20%. Finally, in 1920 the participation of foreigners in Minas Gerais decreased to the same proportion of 1890, while in São Paulo foreigners continue to increase their participation in the paulista population, as shown in Table 1. Therefore, foreigners could hardly be considered an important source of labour in quantitative terms. The analysis of the mineiro population during the 19th century suggests that non-slave Brazilians and nationals constituted the main source of labour. The slave population – which was one of the largest in Brazil – constituted the second largest source of labour in the province. In contrast, foreigners were numerically the least important source of labour in Minas Gerais, although their importance as a source of skilled labour cannot be neglected, as is going to be shown in the following section. Furthermore, the analysis of the mineiro population also reinforces the main question posed by this paper: where was the large number of free coloured Brazilians employed? 3. The Labour Market in 19th-century Minas Gerais This section investigates the sources of labour from which entrepreneurs recruited their labour force. As demographic indicators examined above have shown, the majority of the mineiro population was made up of nationals and slaves, who constituted the bulk of the labour force in almost every industry examined in this work as both skilled and unskilled labour. 35 Libby, Transformação e Trabalho, p.54. 13 Nevertheless, some of these industries relied more than others on slave labour. This difference is due to several factors such as the local availability of slaves and alternative sources of labour, the timing of the emergence of each industry, the labour-intensity of each economic activity, and the nature of the technology involved in each specific industry. 3.1. The iron industry Slavery was of crucial importance for the mineiro iron industry for two main reasons. First, slaves were responsible for the diffusion of the first productive method. Second, the regular supply of slave labour, together with the physical isolation of the province created by natural difficulties of transport, represented the main competitive advantage that small foundries had against foreign competition. Until the 1880s, the mineiro iron industry expanded by relying heavily on slaves as its main source of labour36. Such a dependency on slavery can be observed in the first iron foundries established in the province. The São Miguel de Piracicaba foundry, established by Monlevade and Captain Luiz Soares de Gouveia in 182737, employed a large number of slaves. In 1840, there were 151 slaves. In 1864, as shown in Table 2, the São Miguel de Piracicaba foundry, is quoted as employing 103 workers, who most certainly were slaves. In the early 1870s, the number of slaves employed at the São Miguel de Piracicaba had not changed considerably and they were performing all types of work - skilled and unskilled, as Monlevade observed in response to a survey of the iron industry in Minas Gerais organized by the president of the province38. 36 Ibid., pp.135-6. Paula, ‘Dois Ensaios sobre a Gênese da Industrialização em Minas Gerais’, p.31. 38 F.A.M. Gomes, História da Siderurgia no Brasil, Belo Horizonte: Itatiaia, 1983, p.111. 37 14 One of the first successful foundries in Minas Gerais was the Patriótica, set-up by Eschwege, a German engineer, who came to Minas Gerais in 181139. At the beginning, Eschwege, did not buy slaves to work in his foundry. With his European anti-slavery background, he believed that the work could and should be carried out by non-slave Brazilians. Nevertheless, the difficulties in hiring a permanent and reliable labour force of non-slaves proved overwhelming. He also tried to hire slaves, but this did not work either. As soon as their masters judged that they were capable of doing the work they requested them back, clear evidence of the scarcity of skilled labour and the competition for it. In the end, Eschwege came to the conclusion that it was absolutely necessary to buy slaves, because he could rely on his slaves after training them as foundry masters and apprentices. From then on, his foundry functioned much better and more productively40. Table 2 – Labour Force of Selected Mineiro Foundries, 1811-1870 Foundries/Years 1811 1814 1821 1827 1831 1840 1864 1870 151 103 150 São Miguel de Piracicaba 23 23 56 Patriótica 34 155 Morro do Pilar 50 Girau Source: Adapted from D.C. Libby, Transformação e Trabalho em uma Economia Escravista: Minas Gerais no Século XIX, São Paulo: Brasiliense, 1988, pp.163-74; F.A.M. Gomes, História da Siderurgia no Brasil, Belo Horizonte: Itatiaia, 1983, pp.79-111; W.L. von Eschwege, Pluto Brasiliensis, Belo Horizonte: Itatiaia, 1979, Vol. II, p.209-48. During Eschwege's time - from 1811 to 1821 -, for example, thePatriótica foundry had a labour force made up of 20 slaves, two Brazilians, and occasionally a German foundry master. Slaves were mainly employed as woodcutters, charcoal burners, and as carriers of every type. Brazilians were employed in the smelting and reheating operations, which required more skill from the worker. These two Brazilians were probably the only two he was able to train and hold 39 40 Ibid., pp.79-85. W.L. von Eschwege, Pluto Brasiliensis, Belo Horizonte: Itatiaia, 1979, Vol. II, p.248. 15 before deciding to buy slaves, as mentioned above. In 1831, when Eschwege had already left the foundry, the Patriótica foundry continued to rely basically on slaves. As shown in Table 2, at this time, the foundry employed a total of 56 workers, 55 of them slaves and one non-slave manager. This implies that the skilled tasks were shifted from the hands of non-slave Brazilians to the hands of slaves and further confirms the dependence of one of the first iron foundries established in Minas Gerais on slaves as the main source of both skilled and unskilled labour. The Morro do Pilar foundry, which was financed by the Imperial government and organized by Manuel Ferreira da Câmara, seems to have employed both slaves and non-slaves. According to Schoenewolf – a German foundry master who worked at both the Morro do Pilar and the Patriótica foundries –, there were 34 people working at the foundry in 1814, as shown in Table 2. Of the 34 people employed in the foundry, 15 were employed as smelters and ironfounders, eight as blacksmiths, six as carpenters, and two negroes, two youths, and a overseer. Schoenewolf gives no information on the number of slaves and non-slaves within the labour force. However, based on evidence given by the German foundry master on how Câmara beat one of the blacksmiths, it is reasonable to believe that part of the labour force was slave. In 1821, Schoenewolf described the labour force of the foundry as made up of 155 people, as shown in Table 2: one manager, two masters, six overseers, one blacksmith, two master-carpenters, 28 workers in the furnaces and the hammermill, 17 apprentices, and 70 slaves for the charcoalbox41. Apart from the 70 slaves, the German foundry master gave no further information about the condition of the rest of the labour force. However, the number of overseers suggests that slaves were being employed in tasks other than the preparation of charcoal. The dependence of 41 Ibid., pp.209-13. 16 the Morro do Pilar on slave labour is indisputable. However, the employment of a large number of non-slave Brazilians is also beyond doubt. Another important foundry in Minas Gerais in the first half of the 19th century was the Girau, considered one of the largest foundries in the period 1831-40. Founded in 1813, by 1840 it employed a total of 50 workers, as shown in Table 2, 49 slaves and one manager. Apart from the manager, all the skilled and unskilled workers were slaves, including two slave foremen, three smelters and iron-founders, and several blacksmiths42. Smaller factories also employed slaves, though they employed a considerable number of non-slave Brazilians. In Itabira do Mato Dentro, for example, there were three small foundries in 1840 employing, in addition to the owners, a total of 15 slaves and two non-slaves. In Itabira do Campo in 1831, there were another three foundries employing 19 slaves and 12 non-slaves. From 1831 to 1840, there were 24 foundries in the Metalúrgica-Mantiqueira region of Minas Gerais43. There is no information available about the labour force of two of the 24 foundries. As a whole, the 22 remaining foundries employed 168 slaves and 70 non-slaves, giving indisputable evidence of the importance of slaves as a source of labour for the smaller foundries. However, it becomes clear that the iron industry also provided enough for non-slave Brazilians even in the first half of the 19th century when slaves were abundant and relatively cheap. There is very little information available for the second half of the century. Nevertheless, in 1853 there were 84 ironworks, besides the numerous small shops, in Central Minas Gerais. The 84 ironworks employed at that time about 2,000 workers, slave and free44. In 1864, in Santa 42 Libby, Transformação e Trabalho, p.163. There is evidence suggesting that the iron industry was heavily concentrated in the Metalúrgica region in 1821. During the 1850s, 80% of the mineiro foundries were located in the Metalúrgica-Mantiqueira region and the same trend is observed for the period 1863-66, although the information is incomplete and scattered. Ibid., pp.152-60. 44 Martins Filho and Martins, ‘Slavery in a Nonexport Economy’, p.564. 43 17 Bárbara (a district within the Metalúrgica-Mantiqueira zone), there were 21 foundries. One of these is the São Miguel de Piracicaba foundry that employed 103 workers, as mentioned above. The remaining 20 foundries employed a total of 178 workers, each employing between four to 16 people, but there is no information about the balance between slaves and non-slaves. However, on the evidence provided by the São Miguel de Piracicaba foundry, it is reasonable to believe that a large proportion of the 178 workers employed by the other 20 foundries were slaves. In 1883, Gorceix – a researcher from the Mining School of Ouro Preto – observed that most of the foundries established in the four main producer districts45 relied heavily on slave labour46. Thus, slaves represented the main source of labour for the mineiro iron industry until the abolition of slavery at the end of the 1880s. Their employment was widespread among both small and large foundries during this period. Nevertheless, as indicated above, there is evidence suggesting that non-slave Brazilians were also widely employed, although to a lesser extent. Foreigners were an important source of skilled labour for the mineiro iron industry. Foreign ironworkers played a critical role in the industry at the beginning of the 19th century setting-up iron foundries, assembling machinery, training the native labour force, and supervising production. The largest mineiro foundries established during this period relied upon foreign technicians. In the operation of the Patriótica foundry Eschwege depended on the help of a fellow German foundry master47. At the Morro do Pilar, Schoenewolf, the German foundry master who had worked for Eschwege, was hired as supervisor of the production in 1814. He proved to be absolutely vital to the successful production of the foundry. Before his arrival all attempts of production had failed. Moreover, in an attempt to improve production at the foundry, English 45 The districts are: Itabira do Mato Dentro, Santa Bárbara, Conceição do Mato Dentro, and Caeté. Libby, Transformação e Trabalho em uma Economia Escravista, p.174. 47 Gomes, História da Siderurgia no Brasil, p.83. 46 18 ironworkers were requested from the Imperial government48. After Schoenewolf had returned to Germany in 1821, seven German technicians were recruited to work at Morro do Pilar. Nevertheless, the foundry was finally shut down ten years later. One of those German technicians continued in Minas Gerais and was found by Gardner, an Englishman, in 1840 operating a small foundry close to the village of Conceição49. Thus, foreigners represented an important source of skilled labour. Furthermore, the fact that foreigners continued to own the main foundries of this period is a testimony of their qualitative – rather than quantitative – importance. Finally, the São Miguel de Piracicaba foundry provides undisputable evidence of the employment of women and children in the mineiro iron foundries. As mentioned above, among the 151 slaves employed by the foundry in 1840, 42 were women and 14 children under ten years of age50. There is also evidence of the employment of two youths at the Morro do Pilar foundry, as mentioned above. Although there is no further evidence, it is reasonable to believe that the employment of women and children was widespread and not only restricted to textile mills, as will be shown further on. 3.2. The Cia. União e Indústria Turnpike As in the case of the railways51, legislation prevented the Companhia União e Indústria (CUI) to hire slaves in the construction or the operation of the turnpike52. However, as was also in the case of the railways53, during the years of the construction of the União e Indústria a large number of slaves were hired, managed, and paid by local builders contracted by the company to 48 von Eschwege, Pluto Brasiliensis, pp.208-11. Libby, Transformação e Trabalho em uma Economia Escravista, p.163. 50 Ibid., p.165. 51 Lamounier, ‘The “Labour Question” in Nineteenth Century Brazil’, p.4; O.N. Mattos, ‘ Vias de Comunicacao’, In: S.B. Holanda, Historia Geral da Civilização Brasileira, São Paulo: Difel, 1971, pp.42-59; R.J. Katinsky, ‘Ferrovias Nacionais’, In: S. Motoyama (ed.), Tecnologia e Industrialização no Brasil: Uma Perspectica Histórica, São Paulo: UNESP, 1994, pp.37-65. 52 Companhia União e Indústria, Relatório da Assembléia Geral dos Acionistas, Rio de Janeiro, 1856. 49 19 the construction of portions of the road. Nevertheless, there is plenty of evidence that a considerable number of non-slave Brazilians were also employed, as both skilled and unskilled labour. Furthermore, a considerable number of foreigners were employed, mainly as skilled labour. As shown in Table 3, in 1855 between 515 and 828 slaves were employed in the construction of the turnpike. Slaves most probably represented a large proportion of the total labour force during this period, as the figures for the following years suggest. In 1856, 1,102 people were employed; among them, 900 slaves who constituted nearly 82% of the total labour force. Of the 900 slaves, 48 were elderly or youths, some of whom were employed in the making of charcoal; 96 were employed as bricklayers, cooks, etc., and as their assistants and apprentices, in the several workshops and sections of the turnpike; and the remaining 756 worked in the construction and maintenance of the turnpike. This demonstrates that in 1856 slaves were employed as both skilled and unskilled labour. The remainder 202 were non-slaves, most of whom were probably Brazilians. In 1857, there were 804 slaves working in the construction of the turnpike, but there is no information about the total number of workers employed, as indicated in Table 3. However, on the basis of the number of workers employed in 1856 (1,102) – and as there is no information of any important alteration in the labour force –, it is realistic to suggest that the total number employed in 1857 may have been around 1,000 and that slave labour continued to account for a large proportion of the labour force. It is also reasonable to conclude that non-slaves must have constituted approximately 20% of the labour force, of whom a large proportion was most probably made up of Brazilians. In 1858, 2,636 workers were employed. Among them, 1,136 were employed on the section of the turnpike between Juiz de 53 Lamounier, ‘The “Labour Question” in Nineteenth Century Brazil’, p.16. 20 Fora and Paraíba. Of this number, 800 were slaves. In other words, 70% of the 1,136 during that year. The other 1,500 were employed on the section between Petrópolis and Paraíba do Sul. Nevertheless there is no information about the status of these workers54. On the basis of the number of slaves working on the first section, it is reasonable to believe that slave labour represented a large proportion of the remaining 1,500 workers. Furthermore, if the proportion of slaves and non-slaves employed on the first section was about the same on the second section, it is reasonable to conclude that approximately 500 non-slaves were employed in 1858. In 1859, with the construction of the section between Pedro do Rio and the Paraibuna River, a total of 3,500 workers were employed, as shown in Table 3. However, there is no indication of the balance between slaves and non-slaves, nor between skilled and unskilled workers, nor about those employed directly by the company and those employed by contractors hired to build parts of the turnpike. Table 3 – Slave and Non-Slave participation in the Construction of the União e Indústria Turnpike, 1855-65. Years TOTAL Slaves % Non-Slaves % 1855 515 - 818 1856 1,102 900 82 202 18 1857 804 st 1 Section 1,136 800 70 336 30 1858 2nd Section 1,500 *1,050 *70 *450 *30 Total 2,636 *1,850 *70 *786 *30 1859 3,500 *2,450 *70 *1,050 *30 1865 344 Sources: Companhia União e Indústria, Relatório da Assembléia Geral dos Acionistas, (1856), pp.13-5; (1857), pp.21-3; (1860), p.7; (1866), Annexo 12.; A.O. Esteves, "Mariano Procópio", Revista do Instituto Histórico e Geográfico Brasileiro, Vol.230, Jan-Mar, 1956, p.138; D.A. Giroletti, "A Companhia e a Rodovia União e Indústria e o Desenvolvimento de Juiz de Fora, 1850 a 1900", Universidade Federal de Minas Gerais, Mimeo., Belo Horizonte, 1980, p.30. * These figures were calculated based on the 1st Section percentages. 54 D.A. Giroletti, ‘A Companhia e a Rodovia União e Indústria e o Desenvolvimento de Juiz de Fora, 1850 a 1900’, unpubl., Universidade Federal de Minas Gerais, 1980, p.30. 21 With the completion of the construction of the turnpike, the number of workers employed by the company fell steadily. In 1865, for example, the company employed a total of 344 people, but there is no information concerning the employment of slaves or non-slaves. However, from the 30 names listed as forming the staff of the company, 22 are without any doubt names of a Portuguese origin55, which is strong evidence of the employment of non-slave Brazilians as the company could not hire slaves itself according to the terms of the concession for the construction and the operation of the turnpike. For the period 1866 onwards, there is no information either about the total number of workers employed by the company or about the employment of slaves. Slaves employed by the CUI were recruited basically from three sources: from other companies, directly from private masters, and from the company's own shareholders. On 7 June 1855, for example, the CUI signed a contract with the Companhia de Cocaes to hire 305 slaves. Furthermore, in the company report of 1856, slave owners – to whom the company had advanced payment for the hire of their slaves – appeared as the company's sole sundry debtors. In the same report, the company informed that at the end of 1855 it had increased by more than 25% the price it had previously been offering to slave owners for the hire of slaves. Several reasons accounted for this change in the company's recruitment policy: the increase in the cost of labour caused by competition from other employers, the small supply of workers, and the increasing amount of work caused by the development of the construction of the turnpike União e Indústria56. In the report of 1857, the chairman thanked a shareholder for the services he had rendered to the company. Among other things, he hired out to the company more than a hundred slaves from his own stock that were working on the construction of the turnpike57. 55 Companhia União e Indústria, Relatório da Assembléia Geral dos Acionistas, Rio de Janeiro, 1866, Annexe 12. Companhia União e Indústria, Relatório da Assembléia Geral dos Acionistas, Rio de Janeiro, 1856, pp.14-32. 57 Companhia União e Indústria, Relatório da Assembléia Geral dos Acionistas, Rio de Janeiro, 1857, p.38. 56 22 Non-slave Brazilians were a source of skilled and unskilled labour for the CUI, as pointed out above. In 1856, for example, the company employed a total of 80 non-slave Brazilians for the gravelling of the road58. The report of 1857 informed that a Brazilian engineer was hired to take the place of one of the French engineers whose contract had expired. It also stated that the number of Brazilians recruited had increased and that the workshops established in Juiz de Fora employed both Germans and Brazilians59. Finally, foreigners were also an important source of labour for the CUI. The company employed several foreigners in the construction and operation of the turnpike. In 1856, for 60 example, 20 German workers were recruited in Hamburg to work in the company's workshops . From 1853 to 1856, the company employed two French engineers to supervise the construction of the road61. According to the report of 1857, the French engineers – whose contracts had expired – had been replaced by a German and a Brazilian engineer, as mentioned above. During the same year, the company also hired the two sons of the German engineer to work as his assistants62. Apart from the engineers already mentioned, the company also employed two foreign architects, and a foreign surveyor63. In 1860, the company employed more than 50 free Africans in the construction of a bridge over the river Kagado and of a branch of the turnpike to Mar de Hespanha. These Africans workers were paid by local farmers interested in the construction of the branch64. Further evidence of the employment of foreigners in the operation of the turnpike is found in the list of the staff of the company in the report of 1866: William 58 A.O. Esteves, ‘Mariano Procópio’, Revista do Instituto Histórico e Geográfico Brasileiro, Vol.230, JanuaryMarch 1956, p.138. 59 Companhia União e Indústria, Relatório da Assembléia Geral dos Acionistas, Rio de Janeiro, 1857, pp.21-3. 60 Companhia União e Indústria, Relatório da Assembléia Geral dos Acionistas, Rio de Janeiro, 1856, p.13. 61 Esteves, ‘Mariano Procópio’, pp.149-52. 62 Companhia União e Indústria, Relatório da Assembléia Geral dos Acionistas, Rio de Janeiro, 1857, p.21. 63 Giroletti, ‘A Rodovia e a Companhia União e Indústria’, p.27. 64 Companhia União e Indústria, Relatório da Assembléia Geral dos Acionistas, Rio de Janeiro, 1861, p.7. 23 Morrit, inspector; Theodor Krauss, treasurer; León Boullié, Julien Audemars, César Ansaldi, and Ulysse Dauphin, drivers65. Another important source of foreign labour was the colony of immigrants "D. Pedro II" set-up by the CUI in Juiz de Fora. According to the decree that granted the concession for the construction of the turnpike, the company was obliged to establish a colony of immigrants and settle 2,000 people there. By 1858, a total of 950 had arrived in the colony. Two years later, 1,144 lived there. Among the 667 adults living in the colony in 1860, approximately 196 worked for the company, most of whom as skilled labour66. In 1867, there were 1,082 Germans and 117 Brazilians living in the colony67. Thus, from evidence presented above, it is reasonable to conclude that slaves constituted an important source of both skilled and unskilled labour for the CUI, at least during the years of the construction of the turnpike, and that they made up a large proportion of the total labour force employed by the company during the period. Slaves were mainly recruited from slaves owned by other businesses, directly from private masters, and from the company's own shareholders. There is also evidence that non-slaves, of whom Brazilians were probably the majority, constituted around 20% to 30% of the labour force during the years of the construction of the turnpike. It seems that after the completion of the turnpike the participation of non-slaves increased and that a large part of the staff of the company was probably made up of Brazilians. Finally, foreigners were employed as both skilled and unskilled labour. Foreigners were recruited directly in their country of origin or from the immigrants of the Colony "D. Pedro II" established by the company. 3.3. The Textile Industry 65 Companhia União e Indústria, Relatório da Assembléia Geral dos Acionistas, Rio de Janeiro, 1866, Annexe No.12. 66 Giroletti, ‘A Rodovia e a Companhia União e Indústria’, pp.31-8. 67 Ministerio da Agricultura, Commercio e Obras Publicas, Relatorio da Repartição dos Negocios da Agricultura, Commercio e Obras Publicas, Rio de Janeiro, 1869, Annexe L, p.130. 24 In relation to the mineiro textile industry, slaves were a less important source of labour. This is due to the fact that the first mills were established in the mid 1870s, a time when slaves were scarce, expensive, and mostly concentrated in the coffee-growing areas, Furthermore, the analysis of the evolution of the stock of slaves within the various sub-regions of Minas Gerais shows that slaves were becoming even scarcer where the textile mills were situated68. However, it is important to note that there is no agreement about this issue in the literature69. Yet, slaves were employed in small numbers, as in other mills all over Brazil70. They were usually to be found in ancillary activities, such as in construction work, in the carpentry and blacksmith workshops and the sawmill, in transport, in the cleaning of mills, and in several activities related to the breeding of animals and farming, but there were occasions in which a few slaves were employed as operatives in the mills71. Textile mills usually resorted to the hiring of slaves, a practice very common among industrialists in Brazil during the 19th century72. From 1873 to 1877, the Companhia Mascarenhas Irmãos, which owned the Cedro mill, consistently employed slaves. In 1873, for example, the mill hired two slaves, paying their owner 140$000 Milreis and 70$000 Milreis respectively for an annual contract73. The company also hired slaves from shareholders74. But the Cedro mill also possessed its own slaves. In 1873, the company bought a slave for 1:500$000 Conto75. There is evidence that the Companhia Mascarenhas & 68 R. Conrad, The Destruction of Brazilian Slavery:1850-1888, Berkeley: University of California Press, 1972, pp.285-293. 69 See Martins Filho and Martins, ‘Slavery in a Nonexport Economy’. 70 Although the total employment of slaves tended to decrease after the 1850s, they were still being employed in the Brazilian textile mills, as mentioned by Stein, Origens e Evolução da Indústria Têxtil no Brasil, p.64. 71 D. Giroletti, Fábrica Convento Disciplina, Belo Horizonte, Brasília: Editora UNB, 2nd ed., pp.61-2. 72 L.C. Soares, ‘Urban Slavery in Nineteenth-Century Rio de Janeiro’, unpubl. PhD diss., University of London, 1988, p.72. 73 Companhia Cedro e Cachoeira, ‘Caixa de Correspondências Recebidas No.5’, Contract for the hiring of slaves. 74 Companhia Cedro e Cachoeira, ‘Copiador de Cartas, 1872-1879 - Mascarenhas & Irmãos’, Letter from Bernardo Mascarenhas to Caetano Mascarenhas, 27 October 1879, p.108. 75 Companhia Cedro e Cachoeira, ‘Caixa de Correspondência Recebida No.5’, Escritura de Compra e Venda de Manoel Cabra. 25 Barbosa, owner of the Cahoeira mill, also hired slaves. In 1878, the manager of the mill – Francisco de Paula Mascarenhas – wrote to Francisco de Assis Vianna saying that: "We believe that you made a mistake when you counted the number of days that your slave worked here (...)."76 In 1883, the Companhia Cedro e Cachoeira (CCC) bought a slave77 and there is evidence that the employment of slaves at CCC continued until 188778. Taking into account the fact that the Cedro and the Cachoeira mills were owned by members of the same family, Mascarenhas, that the mills were close to each other, and that the CCC – the company resulting from the merger of the two mills – employed and owned slaves, it is reasonable to conclude that the employment of slaves at the Cachoeira mill was not sporadic. The only textile mill to be operated basically by slaves was the São Sebastião mill, set-up in 188479. Apart from a few non-slave operatives, the large majority of the 75 workers80 employed at the mill was made up of slaves. The employment of slaves by the mineiro textile mills continued until the last years of slavery. The Cassu mill, for example, employed a total of 61 workers in 1886, of whom only three were slaves. During the same year, the Bom Jesus D'AguaFria mill employed ten slaves81. Thus, although slaves did not represent a source of labour for the mineiro textile industry, they were consistently employed in small numbers. 76 Companhia Cedro e Cachoeira, ‘Copiador de Cartas, 1878-1880 - Mascarenhas & Barbosa’, Letter from Francisco de Paula Mascarenhas to Francisco de Assis Vianna, 10 October 1878, p.118. 77 Companhia Cedro e Cachoeira, ‘Caixa de Correspondências No.14’, Receipt of Rs.1:350$000 signed by Antônio Diniz Mascarenhas, 27 July 1883. 78 See Companhia Cedro e Cachoeira, ‘Caixa de Correspondências Recebidas No.18’, Letter from Francisco de Paula Mascarenhas to Bernardo Mascarenhas, 14 August 1885; Companhia Cedro e Cachoeira, ‘Quarto Relatorio apresentado á Assemblea Geral dos Accionistas da Companhia Cedro & Cachoeira, em 15 de março de 1887’; and Companhia Cedro e Cachoeira, ‘Caixa de Correspondências No.21’, Letter from Francisco de Paula Mascarenhas to Bernardo Mascarenhas, 1 February 1887. 79 Libby, Transformação e Trabalho, p.229. 80 P. Tamm, Uma Dinastia de Tecelões, Belo Horizonte, 2nd.ed. 1960, p.110. 81 Libby, Transformação e Trabalho, pp.234-5. 26 Nevertheless, non-slave Brazilians were the most important source of labour for the mineiro textile industry. As mentioned above, to the exception of the São Sebastiao mill, which was operated basically by slaves, the larger part of the labour force employed by the mineiro textile mills was made up of non-slave Brazilians. As mentioned above, of the 61 people employed by the Cassu mill 57 were non-slave Brazilians. In 1883, the Biribiry mill employed 130 people: apart from the foreigner machinist, all were non-slave Brazilians82. Although there is very little information available about the composition – in terms of nationality – of the labour force of other textile mills for any specific year, there is evidence of the wide employment of non-salve Brazilians, especially of women and children. The large employment of women and children is not peculiar to the mineiro textile mills since in many different countries the bulk of the labour force employed in textile mills was made up of women and children.83. In the case of Minas Gerais, as most mineiro mills were established in rural areas, and as the majority of adult men were employed on the land, the most accessible source of labour for the mills were women and children, especially orphans84. This is further evidence of the seasonal nature of agriculture employment85 and points to the problems textile entrepreneurs faced with worker-life-cycle’ employment in the mills. 82 Ibid., pp.234, 236. For a more detailed debate on the employment of women and children in the British textile industry see E.P. Thompson, The Making of the Working Class, London: Gollancz, 1980 and P. Mathias, The First Industrial Nation: The Economic History of Britain, 1700-1914, London: Meuthuen, 2nd ed. 1983. For a debate on the employment of women and children in the Japanese textile industry see, among others, K. Odaka, ‘Redundancy Utilized: The Domestic Economics of Female Domestic Servants in Pre-War Japan’, In: J. Hunter (ed.), Japanese Women Working, London: Routledge, 1993 and J. Hunt, ‘Gendering the Labour Market: Female Textile Workers in Interwar Japan’, In: C.M. Lewis (ed.), ‘Workers and ‘Subalterns’: A Comparative Study of Labour in Africa, Asia and Latin America’, London: LSE/Dept. of Economic History, Working Paper No. 73/03, 2003. 84 A.M. Vaz, Cia. Cedro e Cachoeira: História de uma Empresa Familiar, 1883-1987, Belo Horizonte: Cia Cedro e Cachoeira, 1990, pp.193-7. 85 See the discussion in part 1, p.8. 83 27 Orphans were usually concentrated in orphanages and were thus more easily recruited. The CCC employed a large proportion of women, and children. The Cedro mill, for example, employed a total of 130 workers in 1882, 60 of who were children86. In 1883, only 20% of the labour force of the CCC was made up of men. The rest was made up of women and children. The composition of the labour force did not change in the following year, as women, boys, and girls represented 45%, 23%, and 12% of the total labour force respectively. For the years 1885 and 1886, there is no information about the percentage of men, women, and children employed by the company. Nevertheless, there is evidence of the continuing employment of children at the company in 188587. In 1887, the Cedro mill employed a total of 233 workers, out of whom 21% were men, 45% women, 23% boys, and 11% girls. In the following year, the mill employed a total of 253 workers: 45% were women, 23% were boys, and 9% were girls. In 1889, the figures for boys and girls are included in the figures for men and women respectively. Nevertheless, the group of women made up the larger proportion of the total labour force representing 65%88. There is further evidence that the company continued to employ women and children in large proportions89. The same phenomenon can be observed in other mills in Minas Gerais. The first records of the labour force of the Companhia Cachoeira de Macacos (CCM) appeared only in the 1920s. Nevertheless, from the records of those workers who had started to work at mill at the turn of the century, it is possible to observe that they were all Brazilians and locals. Furthermore, most of 86 Companhia Cedro e Cachoeira, ‘Copiador de Cartas de 1881 - Mascarenhas & Irmãos’, Inquiry of the Sete Lagoas District Council on 13 March 1882, pp.488-92. 87 Companhia Cedro e Cachoeira, ‘Caixa de Correspondências Recebidas No.18’, Letter from Francisco de Paula Mascarenhas to Bernardo Mascarenhas, 9 October 1885. 88 Vaz, Cia. Cedro e Cachoeira, p.196. 89 See Companhia Cedro e Cachoeira, ‘Caixa de Correspondências Recebidas No.35’, Letter from Francisco de Paula Mascarenhas to Theóphilo Marques Ferreira, 8 January 1893; and Companhia Cedro e Cachoeira, ‘Caixa de 28 them were children and teenagers when they were first admitted90. Among the 32 workers employed by CCM whose records were found some were brothers and sisters91. This seems to indicate that family relations were an important source of recruitment. The analysis of the records of the labour force of the Companhia de Tecidos Santanense (CTS) leads to the same conclusions: most of them indicate the employment of locals, notably children, teenagers, or women92. The Cassu mill employed in 1886 15 women over 14 years old, 10 girls between the ages of eight and 14 years old, and 20 boys between the ages of eight and 14 years old. During the same year, the textile mill of Montes Claros employed a total 81 people; 70 workers were orphans or abandoned youths. Among the 66 people employed by the União Itabirana mill, 40 were boys and girls. Women made up the majority of the labour force of the Bom Jesus d'Àgua Fria mill: of the 140 workers employed at the mill, 80 were women. The same can be observed at the Biribiry mill in 1883. Of the 130 workers employed at the mill during this year, 110 were female aged between 10 and 30 years old93. As the analysis of the records of the labour force of both the CCM and the CTS indicates, non-slave Brazilians were frequently recruited locally. All of the workers listed by the records of both companies were born in the Metalúrgica zone, where both mills were located94. The CCC also mainly recruited locals95. Furthermore, the CCC often recruited children in orphanages from Correspondências Caetano Mascarenhas, 1883-1912 - No.149’, Letter from Caetano Mascarenhas to Francisco de Paula Mascarenhas, 16 July 1894. 90 Companhia Cachoeira dos Macacos, ‘Livro de Registro de Empregados, 1926-1931’, pp.1-56 and ‘Livro de Registro de Empregados, 1935’, pp.3-203. 91 Companhia Cachoeira dos Macacos, ‘Livro de Registro de Empregados, 1926-1931’, pp.1-56 and ‘Livro de Registro de Empregados, 1935’, pp.3-203. 92 Companhia de Tecidos Santanense, ‘Registro de Empregados’. 93 Libby, Transformação e Trabalho, pp.234-6. 94 Companhia Cachoeira dos Macacos, ‘Livro de Registro de Empregados, 1926-1931’, pp.1-56; Companhia Tecidos Santanense, ‘Livro de Registro de Empregados, 1935’, pp.3-203. 95 Companhia Cedro e Cachoeira, ‘Caixa de Correspondências Recebidas No. 18’, Letter from Francisco de Paula Mascarenhas to Bernardo Mascarenhas, 30 October 1885; and Companhia Cedro e Cachoeira, ‘Caixa de 29 the neighbouring towns96. Non-slave Brazilians were also recruited through relatives97, suppliers, agents, and clients98. Although foreigners were not quantitatively representative in the mineiro population, they represented an important source of skilled labour for the mineiro textile industry. Until the last decades of the 19th century neither slaves nor Brazilian wage earners had the required mechanical skills. Mill owners needed competent technicians and were thus willing to hire foreigners99. First in England and later in the USA and in continental Europe, foreigners were recruited to set-up mills, to operate and maintain the machinery, to train the labour force, and to supervise production. At least in the first years, every master in the Cedro mill, apart from the blacksmith and carpenter masters, was recruited abroad. They were recruited by agents, preferably from the same country where the machinery had been bought100. The purchase contract for the machinery of the Cedro mill, for example, included the provision of a technician to assemble and operate it101. Before merging with the Cedro mill, the Cachoeira mill had also hired foreign technicians to supervise work. William Hutchinson, an Englishman, was hired in 1876 for two years to Correspondências Caetano Mascarenhas, 1883-1912 - No.149’, Letter from Caetano Mascarenhas to Francisco de Paula Mascarenhas, 24 October 1893. 96 Vaz, Cia. Cedro e Cachoeira, p.193. 97 Letter from Bernardo Mascarenhas to Antônio Pinto Mascarenhas in 14 March 1884, reproduced in Giroletti, Fábrica Convento Disciplina, pp.64-5; Companhia Cedro e Cachoeira, ‘Caixa de Correspondências Recebidas No.16’, Letter from Francisco de Paula Mascarenhas, 11 August 1884; Companhia Cedro e Cachoeira, ‘Caixa de Correspondências Caetano Mascarenhas, 1883-1912 - No.149’, Letter from Caetano Mascarenhas to Francisco de Paula Mascarenhas, 16 July 1894. 98 Letter from Francisco de Paula Mascarenhas to Robert L. Kerr, 1 June 1878, reproduced Giroletti, Fábrica Convento Disciplina, p.67; Companhia Cedro e Cachoeira, ‘Caixa de Correspondências Recebidas No.29’, Letter from Theóphilo Marques Ferreira to Gontijo, Mascarenhas & Cia., 1 April 1891; Letters from Antônio Hygino M. do Rego to Francisco de Paula Mascarenhas, 17 April 1896, and from Manoel Pimenta Figueiredo to Aristides José Mascarenhas, 15 July 1898, reproduced in Giroletti, Fábrica Convento Disciplina, p.67. 99 For the employment of foreigner technicians in the country as a whole during the 19th century see Stein, Origens e Evolução da Indústria Têxtil no Brasil, pp.64-5. 100 Vaz, Cia. Cedro e Cachoeira, p.53. 30 supervise the setting-up and operation of the mill, and to train the labour force. Three years later, two foreign machinists, John and William Lomas, and a foreign weaver, Andrew White, replaced William Hutchinson, who returned to England. In 1882, all three technicians left the company. John Lomas did not return from a visit to England, and William Lomas and Andrew White ran away during the night. In the following year another Englishman, James Winders, was hired102. Such a high turnover of foreign workers indicates that their employment was erratic, either because the conditions of employment were far from ideal or because foreign workers were not as reliable as many in Brazil thought, or even both. This shows that the perception of non-slave Brazilians as non-reliable had not a solid basis as foreigners proved not to be that reliable as well. The employment of foreign technicians continued after the merger of the Cedro and the Cachoeira mills into the CCC until the last years of the century. James Winders, for example, continued to work for the new company until 1889. In 1883, the company tried to hire William Hutchinson once again, but he decided not to go back to Brazil103. In 1884, the company looked for an experienced and skilful dyer in England104. In 1886, William Hutchinson agreed to consider the matter of rejoining the company, as the CCC approached him once more105. At the end of this year, the company tried to recruit a mechanic in England who understood about carding, spinning and weaving, and was also capable of erecting machinery and keeping it in 101 Companhia Cedro e Cachoeira, ‘Caixa de Correspondências Recebidas No.2’, Contract of purchase of machinery signed by Mascarenhas & Irmãos and Gme. Van Vlick Lidgerwood, 27 September 1870. 102 Vaz, Cia. Cedro e Cachoeira, pp.86-8. 103 Companhia Cedro e Cachoeira, ‘Caixa de Correspondências Recebidas No.14’, Letter from William Hutchinson to Robert L. Kerr, 7 January 1883. 104 Companhia Cedro e Cachoeira, ‘Caixa de Correspondências Recebidas No.16’, Letter from Robert L. Kerr to Bernardo Mascarenhas, 8 July 1884. 105 Companhia Cedro e Cachoeira, ‘Caixa de Correspondências Recebidas No.20’, Letter from Robert L. Kerr to Bernardo Mascarenhas, 2 August 1886. 31 good order106. This shows clearly how limited the mineiro labour market was. It seems that it was easier to find a general technician abroad – which was not an easy task per se – than to find a more specialized local skilled worker. During the same year the company was told that Hutchinson was reluctant to accept the company's offer unless he was offered greater inducements107. It seems that the CCC did not have a better alternative and decided to pay what Hutchinson was requesting. In 1889, Hutchinson rejoined the company, but two years later he returned to England leaving his son, Hebert, and a relative, William, in his place. They had come out with him two years previously. Both of them worked for the company for another three years, before returning to England for good. William Hutchinson returned to the company in 1892 to assemble machinery of the São Vicente mill and stayed until 1894108. In the following year, as William Hutchinson again declined to return to the company109, the company tried to recruit another foreign machinist110. The company finally decided to recruit John Lomas, who had worked at the Cachoeira and the São Sebastião mills. John Lomas left the company months later and until the beginning of the 20th century the company did not hire any other foreign technician111. Other mills also had to resort to foreign skilled labour. In 1875, the Brazil Industrial textile mill recruited five English technicians to assemble the machinery112. The CCM also hired an English technician to install the machinery of the company, but he soon returned to 106 Companhia Cedro e Cachoeira, ‘Caixa de Correspondências Recebida No.20’, Bernardo Mascarenhas, 11 December 1886. 107 Companhia Cedro e Cachoeira, ‘Caixa de Correspondências Recebida No.20’, Bernardo Mascarenhas, 11 December 1886. 108 Giroletti, Fábrica Convento Disciplina, pp.94-5. 109 Companhia Cedro e Cachoeira, ‘Caixa de Correspondências Recebidas No.41’, Francisco de Paula Mascarenhas, 23 May 1895. 110 Companhia Cedro e Cachoeira, ‘Caixa de Correspondências Recebidas No.18’, Francisco de Paula Mascarenhas, 8 August 1895. 111 Giroletti, Fábrica Convento Disciplina, p.96. Letter from Robert L. Kerr to Letter from Robert L. Kerr to Letter from Robert L. Kerr to Letter from Robert L. Kerr to 32 England113. The Cassu mill also depended on foreign technicians114, and in 1893, the CTS also requested the help of a foreigner technician from the CCC to assemble and put into operation its machinery115. The above mentioned decision of the CCC not to hire any other foreign technician from by the end of the 19th century points to a trend that can be observed in other mills too. In 1883, the manager of the União Itabirana mill stated that: "The machinist who is supervising production for 18 months is a Brazilian, who is a local and learned the work at the mill; (...)."116 The substitution of Brazilians for foreigners in the more technical jobs at the end of the 19th century could be the result of several factors: the emergence of a pool of local skilled labour; the result of the various problems faced by mineiro entrepreneurs with foreigners previously employed; a reluctance of foreign workers to return to – or to work – in Brazil, because conditions of employment were not as promised or expected; or/and the inability of mineiro entrepreneurs to pay sufficiently high wages to attract foreign workers. But it also shows that much on the job training was being carried out. To sum up, slaves were consistently employed in small numbers by the mineiro textile industry, but only occasionally as operatives in the mills. Mills owned slaves, but they also hired slaves from shareholders and from private masters. Foreigners were an important source of skilled labour, setting-up mills, operating and maintaining the machinery, training the labour force, and supervising production. They were mainly recruited through agents and suppliers, and 112 Stein, Origens e Evolução da Indústria Têxtil no Brasil, p.52. N.A.M. Freitas, ‘Cia. Têxtil Cachoeira dos Macacos: Empresa que deu Origem a uma Cidade’, unpubl., Fundação Mineira de Arte Aleijadinho/Escola Superior de Artes Plásticas, 1990, p.27. 114 Reproduced in Libby, Transformação e Trabalho, p.234. 115 Companhia Cedro e Cachoeira, ‘Caixa de Correspondências Recebidas No.35’, Letter from Manoel José de Souza Moreira to Aristides Mascarenhas, 24 April 1893. 116 Reproduced in Libby, Transformação e Trabalho, p.238. 113 33 towards the end of the 19th century Brazilians were replacing them. In quantitative terms they were not very representative within the mineiro textile labour force and many foreigners proved not to be as reliable as contemporaries thought. Finally, non-slave Brazilians were the most important source of labour. Mills employed large numbers of locals, mainly women and children, who were recruited in orphanages, by relatives, agents, clients, and suppliers. 3.4. The Electricity Industry Finally, regarding the electricity generating industry, there is very little information concerning the labour force employed by both the Companhia Mineira de Eletricidade (CME) and the Companhia Força e Luz Cataguazes-Leopoldina (CFLCL). Nevertheless, it is possible to draw some conclusions. First of all, as the CME – the first electricity generating company to be organized in Minas Gerais – was established at about the same time of the abolition of slavery in Brazil (the company was established in January 1888, four months before the abolition of slavery) it seems very unlike that the company had employed slaves in its labour force. Thus, the mineiro electricity generating industry could only resort to two sources of labour: non-slave Brazilians and foreigners. Foreigners were regularly employed at the CME. In 1889, for example, during the installation of the plant, the company hired two US technicians, who had arrived together with the equipment ordered from the Westinghouse Electric Company. Moreover, in 1891, Bernardo Mascarenhas sent his electrician, Wan Wagenen, who was certainly a foreigner, to the USA to discuss the project of the new plant he was planning to build with the engineers of the Westinghouse 117 . Thus, it seems that foreigners were employed mainly in those more technical positions. Furthermore, in 1893 the company informed its customers that the increase of the price 34 of the domestic lighting service was due, among other things, to the increase in the salaries of its foreign and Brazilian employees118. However, there is no information concerning the number of Brazilians and the composition – in terms of gender, age, occupation, etc. – of the labour force employed by the company. Nevertheless, based on the information about the employment of foreigners, it is reasonable to believe that, at least at the beginning, Brazilians were mainly employed in the less skilled jobs. The CFLCL did not employ any slaves at all, simply because the company was organized in 1905, 17 years after the abolition of slavery in Brazil. It also seems that the company employed only Brazilians, since there is no reference to the employment of any foreigner even for the most technical jobs. This shows the existence of a pool of skilled Brazilians available already in the early 20th-century mineiro and/or Brazilian labour market. The first manager of the company, hired to supervise the construction of the plant and the installation of the equipment, was Elpidio de Lacerda Werneck, an electrical engineer who lived in Leopoldina and was most probably a Brazilian119. The construction of the power station and the installation of its equipment involved a number of engineers whose names suggest that they were all Brazilians120. There is very little information about the nationality of the labour force employed by the CFLCL after the inauguration of the power station. Nevertheless, there is evidence that the company continued to employ basically Brazilians121. Such a large employment of Brazilian skilled workers can only be explained by the emergence of a local pool of skilled labour, which was possible most probably 117 P. Oliveira, Companhia Mineira de Eletricidade: Pioneira da Iluminação Hidrelétrica na América do Sul, Juiz de Fora: Tip. Lar Catolico, 1969, pp.33-9. 118 Companhia Mineira de Eletricidade, ‘Declaração da Companhia Mineira de Eletricidade’. Letter issued by the Companhia Mineira de Eletricidade informing its customers the increase in the price of the domestic lighting service, 10 June 1893. 119 ‘Companhia Força e Luz’, in Jornal Cataguazes, 12 March 1906. 120 ‘Companhia Força e Luz’, in Gazeta de Leopoldina, 24 March 1907. 121 Companhia Força e Luz Cataguazes-Leopoldina, Relatorio da Diretoria, (1911), p.5. 35 due to the previous establishment of other electricity generating companies in the country such as the CME (1888), the São Paulo Tramway, Light and Power Company (1899), the Companhia Ituana de Força e Luz (1903), and the Empresa Luz e Força de Jundiaí S/A (1904)122. Thus, for obvious reasons, both the CME and the CFLCL did not employ slaves. Whereas the former company employed Brazilians and foreigners, who performed the more technical work, it seems the latter one employed only Brazilians. Conclusion This work revises the widely accepted assumption that free coloured Brazilians in the 19th century were marginalized from the main economic activities due to the fact that they were unfit for disciplined and regular work, a view shared by contemporary rulers, policy makers, and economic historians alike. According to this general assumption, as slavery was abolished, European immigrants were employed in large numbers, thus providing the necessary labour force for the main Brazilian economic activities. By looking at the case of 19th-century Minas Gerais, one of the main coffee growing province/state and with the largest slave population throughout the century, this paper shows that free coloured Brazilians were widely employed in several industrial sectors during this period. Although the mineiro iron industry, established in the early 19th century, employed a large number of slaves, non-slaves, which women and children, were also widely employed. The Cia. União e Indústria, which began to build a road linking the southern part of Minas Gerais to Petrópolis in the mid-1850s, employed a large number of slaves during the construction works. However, non-slave Brazilians represented a important percentage of its employees during this stage. As the construction of the road finished most of the personnel employed by the company was made up of non-slave Brazilians. In the mineiro textile industry 122 F.A.M. Gomes, ‘A Eletrificação no Brasil’, Caderno História & Energia, São Paulo, No.2, October, 1986, pp.5- 36 slaves constituted a very small proportion of the labour force, although a few ones were employed as operatives. The bulk of the textile labour force was made up of non-slave Brazilians, most of whom were women and children. Due to the time when they were established, mineiro electricity firms never employed slaves. Finally, foreigners were employed in all the industries investigated. Nevertheless, they were numerically very small, although very important in the more skilled jobs. They were vital not only in the establishment of foundries and mills, but also in the operation and maintenance of machinery, in the supervision of production, and in the training of the native labour force. Finally, the work gives plenty of evidence that mainstream literature has for a long time failed to acknowledge the importance of free nationals within the Brazilian labour force even in the most dynamic economic activities such as the emerging industry. The case of Minas Gerais shows clearly that they were neither employed only occasionally at the local industry nor unfit for regular and disciplined work. The seasonal nature of agricultural employment, prejudice from employers, work ethic based on slavery, among other factors explains part of the feeling shared by contemporaries. Furthermore, the work shows the extent to which mainstream economic historians have simply bought and inherited prejudices from 19th-century rulers and employers in general. Furthermore, assertions about the marginalization of free coloured Brazilians from the main economic activities in the 19th century seems to be a myth fuelled by a biased literature that has given a disproportionate attention to export activities that formed a relative small (though a very important) part of the Brazilian economy. 12.