¹ See, for example, Mendels, F., ‘Proto-industrialization: The First Phase of the Industrialization Process’, Journal of Economic History, vol. 32 (1972), pp. 241–61.CrossRefGoogle Scholar

² Clarkson, L. A., Proto-industrialisation: The First Phase of Industrialisation? (London, 1985), pp. 28–38CrossRefGoogle Scholar. Also Kisch, H., ‘The Textile Industries in Silesia and the Rhineland: A Comparative Study’, in Kriedte, P., Medick, H. & Schlumbohm, J., Industrialisation before Industrialisation (London/Paris, 1981), pp. 178–200Google Scholar, especially pp. 179–87.

³ Clarkson, , Proto-industrialisation, p. 52.Google Scholar See also: Berg, M., The Age of Manufactures: Industry, Innovation and Work in Britain, 1700–1820 (London, 1985), pp. 289, 313–14Google Scholar; Thomson, J. K. J., ‘Variations in Industrial Structure in Pre-Industrial Languedoc’, in Berg, M., Hudson, P. & Sonenscher, M., Manufacture in Town and Country before the Factory (Cambridge, 1983), pp. 61–91.CrossRefGoogle Scholar

⁴ Based on P. Kriedte, ‘The Origins, the Agrarian Context and the Conditions in the World Market’, in Kriedte, , Medick, & Schlumbohm, , Industrialisation before Industrialisation, pp. 12–37.Google Scholar

⁵ H. Medick, ‘The Demo-economic System of Proto-industrialisation, in Ibid., pp. 74–94, especially pp. 82–9. It should be noted, however, that recent studies have shown that the links between proto-industrialisation and increased demographic growth rates were neither automatic nor straightforward. Indeed, the demographic issues raised by the proto-industrial model are proving to be extremely complex, although there seems to be no reason to doubt that, in general, population increase did accompany proto-industrialisation. See, for example: Hill, B., ‘The Marriage Age of Women and the Demographers’, History Workshop, vol. 28 (Autumn 1989), pp. 129–47.CrossRefGoogle Scholar

⁶ H. Medick, ‘Household and Family in Agrarian Societies and in the Proto-industrial System: An Approach to the Problem’, in Kriedte, , Medick, & Schlumbohm, , Industrialisation before Industrialisation, pp. 38–73.Google Scholar

⁷ Clarkson, , Proto-industrialisation, p. 16.Google Scholar

⁸ Sodré, N. W., História da Burguesia Brasileira (Rio de Janeiro, 1976), pp. 25–35Google Scholar. Sodré speaks of the ‘feudal peripheries’ of the dominant slave system, but the implication is that those regions which exited from the system—including parts of Minas—retreated into feudal configurations. A recent return to the theme of feudalistic institutional forms imbedded in the Brazilian slave system is Hirano, S., Pré-capitalismo e Capitalismo (São Paulo, 1988).Google Scholar

⁹ See Novais, F. A., Portugal e Brasil na Crise do Antigo Sistema Colonial (1777–1808) (São Paulo, 1981), pp. 240–4, 264–5.Google Scholar

¹⁰ See, for example: Costa, E. V. da, Da Senzala a Colônia (São Paulo, 1982), pp. 42–6Google Scholar; Iglésias, F., A Economia Político do Governo Provincial Mineiro, 1835–1889 (Rio de Janeiro, 1958), pp. 130–1Google Scholar: Conrad, R. E., The Destruction of Brazilian Slavery, 1850–1888 (Berkeley, 1972), pp. 127–8.Google Scholar

¹¹ Furtado, C., The Economic Growth of Brazil (Berkeley, 1965), pp. 93–4.Google Scholar

¹² Filho, A. V. Martins and Martins, R. B., ‘Slavery in a Non-export Economy: Nineteenth-century Minas Gerais’, Hispanic American Historical Review, vol. 63, no 3 (Aug., 1983), pp. 537–68CrossRefGoogle Scholar; Martins, R. B., ‘Growing in Silence: The Slave Economy of Nineteenth-century Minas Gerais, Brazil’, unpubl. PhD diss., Vanderbilt University, 1980.Google Scholar

¹³ Obviously, the Martins do not ignore the spread of coffee cultivation into a rather reduced area of the Mata Mineira region which, by the 1850s, had undoubtedly become the most dynamic pole of the provincial economy. They insist, however, and I fully agree, that in terms of labour absorption, the mineiro coffee fields were of very limited importance when compared to the basic food crops and ranching which characterised the rest of the vast provincial agricultural sector. As will be seen, I do take issue with their assumption that extractive and industrial activities were also of reduced numerical importance in the occupational structure of Minas.

¹⁴ Filho, Martins and Martins, , ‘Slavery in a Non-export Economy’, p. 549.Google Scholar

¹⁵ The Hispanic American Historical Review article was accompanied by comments which were quick to question how this anomaly could have occurred. See: ‘Comments on “Slavery in a Non-export Economy”’, Hispanic American Historical Review, vol. 63, no. 3 (Aug. 1983), pp. 569–90Google Scholar, with comments by R. B. Slenes (pp. 569–81), W. Dean (pp. 582–4) and S. L. Engerman and E. D. Genovese (pp. 585–90). The Martins reply came in: Filho, A. V. Martins and Martins, R. B., ‘Slavery in a Non-export Economy: A Reply’, Hispanic American Historical Review, vol. 64, no. 1 (02. 1984), pp. 135–46.Google Scholar

¹⁶ Slenes, R. W., ‘Múltiplos de Porcos e Diamantes: A Economia Escravista de Minas Gerais no Século XIX’, Cadernos IFCH/UNICAMP, no. 17 (06, 1985).Google Scholar

¹⁷ Júnior, C. Prado, Formacão do Brasil Contemporâneo. Colônia (São Paulo, 1976), pp. 197–203.Google Scholar The author notes that the first cattle drive from the South of Minas to Rio de Janeiro took place in 1765.

¹⁸ See Lenharo, A., As Tropas da Moderação: O Abastecimento da Corte na Formação do Brasil (São Paulo, 1979).Google Scholar

¹⁹ A disassociation of the plantation system and the commercial production of basic foodstuffs was not axiomatic, however, Cf. Irwin, J. R., ‘Escravidão e Trabalho em Sistema de Plantação’, Revista do Departamento de História, no. 6 (06 1988), pp. 5–14.Google Scholar

²⁰ Libby, D. C., Transformação e Trabalho em uma Economia Escravista: Minas Gerais no Século XIX (São Paulo, 1988), p. 97.Google Scholar

²¹ Furtado, , The Economic Growth of Brazil, pp. 55–8.Google Scholar

²² População da Provincia de Minas Geraes, Arquivo Público Mineiro, Belo Horizonte, Seção Provincial, 1833, Loose manuscripts in collection currently undergoing organisation.

²³ Luna, F. V. and Costa, I. del N. da, Minas Colonial: Economia e Sociedade (São Paulo, 1982), pp. 22–3.Google Scholar

²⁴ Furtado, , The Economic Growth of Brazil, p. 94.Google Scholar

²⁵ Maxwell, K., Conflicts and Conspiracies: Brazil & Portugal, 1750–1808 (Cambridge, 1973), pp. 128–9.CrossRefGoogle Scholar

²⁶ Luna, F. V. and Cano, W., ‘Economia Escravista em Minas Gerais,’ Cadernos IFCH/UNICAMP no. 10 (10. 1985), pp. 2–12.Google Scholar

²⁷ Klein, H. S., The Middle Passage: Comparative Studies in the Atlantic Slave Trade (Princeton, 1978)CrossRefGoogle Scholar, chaps. 3 and 4.

²⁸ Schwartz, S., Sugar Plantations in the Formation of Brazilian Society: Bahia, 1550–1825 (Cambridge, 1985), p. 349Google Scholar. ‘Throughout the period 1600–1825, Brazilian-born blacks never seem to have made up more than one-third of the slave force.’

²⁹ For an analysis using large comarca/município aggregates see Paiva, C. A., Libby, D. C. and Grimaldi, M., ‘Crescimento da População Escrava: Uma Questão em Aberto’, Anais do IV Seminário sobre a Economia Mineira, CEDEPLAR/FACE/UFMG, (08. 1988), pp. 11–32Google Scholar. A second article deals more with growth patterns in a few selected districts for which chronological series of data are available. Libby, D. C. and Grimaldi, M., ‘Equilibrio e Estabilidade: Economia e Comportamento Demográfico num Regime Escravista’, Papeis Avulsos, no. 7 (12. 1988), pp. 26–43.Google Scholar

³⁰ Elben, J. E., ‘On the Natural Increase of Slave Populations: The Example of the Cuban Black Population’, in Engerman, S. L. and Genovese, E. D., Slavery in the Western Hemisphere: Quantitative Studies (Princeton, 1974), pp. 211–46.Google Scholar

³¹ Higman, B. W., Slave Population and Economy in Jamaica, 1807–1834 (Cambridge, 1976), pp. 134–8.Google Scholar

³² In fact, the census for Minas was undertaken a year after the originally scheduled date. The original Recenseamento publication is rife with tabulation errors which have recently been corrected by a study group at the Centro de Desenvolvimento e Planejamento Regional of the Faculdade de Ciências Econômicas, Universidade Federal de Minas Gerais. The figures used in the present article are corrected and were kindly provided by Professor Clotilde Paiva. For a general examination of the problems which the Recenseamento presents to researchers see Paiva, C. A. and Martins, M. do C. S., ‘Notas sobre o Censo Brasileiro de 1872’, Anais do II Seminário sobre a Economia Mineira, CEDEPLAR/FACE/UFMG, (09. 1983), pp. 149–63.Google Scholar

³³ It should be noted here that the data for 1873 were organised by parishes which did not necessarily coincide with the administrative districts, the former being somewhat larger when differences did occur.

³⁴ See: Costa, I. del N. da, Populações Mineiras (São Paulo, 1981)Google Scholar. In his study of nominal lists dating from 1804, the author designates jornaleiros, i.e. day labourers, as belonging to the service sector when, in fact, such individuals represent an embryonic proletariat which was almost certainly only able to find employment in the primary and secondary sectors. Admittedly, the number of jornaleiros registered in the 1804 lists is very small, probably because they were seldom heads of household. Using the same classification when sources indicate large groups of jornaleiros, however, would be most misleading.

³⁵ The Recenseamento's subcategories were: seamstresses, mine workers, metalworkers, wood workers, textile workers, masons, leather workers (saddle makers and tannery workers), dyers, tailors, hat makers and cobblers.

³⁶ That region was composed of the Jequitinhonha–Mucuri–Doce river valleys where, as will be seen, the cottage textile industry was extremely active. The regional sample is somewhat biased due to the inclusion of the Peçanha district which, in 1831, was a frontier environment where women frequently helped in laying out the new fields. It could be said that the district did not fit into the general ambience of settledness which marked most of the province at that time.

³⁷ Libby, Transformação e Trabalho, chap. 3.

³⁸ I wish to thank Professor Robert W. Slenes for calling my attention to this method of tabulation. See: de Estatística, Directoria Geral, Relatorio e Trabalhos Estatísticos (Rio de Janeiro, 1874), pp. 49–51.Google Scholar

³⁹ I have discussed the tedious problems of the two series at considerable length elsewhere. See Libby, , Transformação e Trabalho, pp. 28–45, 89–92.Google Scholar

⁴⁰ The mineiro iron industry may have been in decline by the 1870s, but the evidence is inconclusive. Evidence from the 1860s confirms the existence of at least 140 foundries. Most of these were small-scale operations based on a hybrid of African and eighteenth-century European technology which had developed during the first three decades of the nineteenth century. The average foundry employed from eight to twelve workers, although a few had several forges and utilised as many as eighty workmen. The only Catalonian forge in the province appears to have engaged over a hundred trained slaves. Dependence on slave labour was characteristic of the entire industry right up to emancipation. The iron produced by the foundries fuelled what seems to have been a large number of toolmaking shops about which, unfortunately, very little is known. I would conclude that in the 1870s between foundries and tool shops, this particular mineiro proto-industry could have employed as many as 4,000 individuals. Cf. Libby, Transformação e Trabalho, chap. 3.

⁴¹ Schwartz, , Sugar Plantations in the Formation of Brazilian Society, p. 446.Google Scholar In his analysis of an 1816–17 Bahian census of slave holdings, a canvass which was concentrated in the heavily plantation-dominated Recôncavo region, Schwartz finds that fully 84% of all slave owners whose occupations were listed were engaged in agricultural pursuits. Craft workers and small manufacturers accounted for only three per cent of all slave holders, while the corresponding figure for the service sector was three per cent. A similar analysis for Minas clearly demonstrates the striking difference between regions dedicated to mono-cultural export activities and those that were not. Mineiro artisans and manufacturers comprised 24 per cent of all slave holders, those engaged in agriculture 60 per cent and members of the composite service sector 16 per cent. Cf. Libby, , Transformação e Trabalho, pp. 110–18.Google Scholar

⁴² Marcílio, M. L., ‘Población y fuerza de trabajo en una economía agraria en proceso de transformación. La provincia de São Paulo a fines de la época colonial’, in Sánchez-Albornoz, N., Población y Mano de Obra en América Latina (Madrid, 1985), pp. 115–26.Google Scholar

⁴³ This relatively large marginal segment contrasts with findings for Minas. The number of vagabundos, mendicantes, esmoleiros, etc. observed in the 1831–40 manuscript nominal lists is statistically insignificant. It is tempting to suggest that the relative absence of marginal segments in the mineiro population may very well reflect the option which proto-industrialisation represented in terms of survival strategies for the poor.

⁴⁴ A study of the cotton export boom in São Paulo, provoked by the United States Civil War, affirms that, prior to the 1860s, cotton cultivation had been abandoned for many decades. Thus it would seem even more unlikely that a paulista cottage textile industry took hold during the nineteenth century. See Canabrava, A. P., O Algodão em São Paulo: 1861–1875 (São Paulo, 1984), p. 21.Google Scholar

⁴⁵ Beauclair, G., ‘A Pré-indústria Fluminense, 1808–1850’, unpubl. tese de Doutoramento, Universidade de São Paulo, 1988.Google Scholar

⁴⁶ Furtado, , The Economic Growth of Brazil, pp. 55–8.Google Scholar

⁴⁷ Kriedte, , ‘The Origins, the Agrarian Context, and the Conditions in the World Market’, in Kriedte, , Medick, and Schlumbohm, , Industrialisation before Industrialisation, p. 16.Google Scholar The author does mention that the less fertile mountainous regions of Europe were usually the first to proto-industrialise because feudal relations were generally weaker there or broke down earlier. If one really wanted to stretch comparisons, it could be said that the post-gold rush mountains of Minas were also ripe for proto-industrialisation because colonial relations of production were weaker or breaking down more rapidly than in other regions of Brazil.

⁴⁸ Ibid., p. 33.

⁴⁹ See, for example: Martins, R. B., ‘A Indústria Têxtil Doméstica de Minas Gerais no Século XIX’, Anais do II Seminário sobre a Economia Mineira, CEDEPLAR/FACE/UFMG (09. 1983), pp. 85–8.Google Scholar

⁵⁰ Alden, D., ‘The Population of Brazil in the Late Eighteenth Century: A Preliminary Survey’, Hispanic American Historical Review, vol. 43 (05, 1963), pp. 173–201.CrossRefGoogle Scholar Analysing census material dating from 1772 to 1782, the author finds that Minas accounted for slightly more than a fifth of the total colonial population. The mineiro population is estimated to have been nearly 320,000, that of Bahia 289,000, Pernambuco's 240,000 and that of Rio de Janeiro 216,000. Of all the other captaincies, only São Paulo had a population above 100,000 (116,975).

⁵¹ Banner, J. C., Cotton in the Empire of Brazil: The Antiquity, Methods and Extension of Its Cultivation, Together with Statistics of Exportation and Home Consumption (Washington, 1885), pp. 10–13.Google Scholar

⁵² Maxwell, , Conflicts and Conspiracies, p. 63.Google Scholar

⁵³ Carvalho, D. de, Noticia Histórica sobre o Algodão em Minas, (Rio de Janeiro, 1916). p. 9.Google Scholar

⁵⁴ Ibid., p. 10.

⁵⁵ Novais, , Portugal e Brasil na Crise do Antigo Regime Colonial, pp. 272–3.Google Scholar

⁵⁶ Inventario de Teares de Minas Geraes em 1786. Arquivo Público Mineiro, Seção Colonial, 1786, manuscript, microfilm. Unfortunately, the Inventario has not been made fully available to the public as yet. It may well represent the only source which will afford reliable data as to rates of productivity in the mineiro cottage textile industry. At this point, one can only speculate that such rates were quite low.

⁵⁷ Martins, , ‘A Indústria Têxtil Doméstica’, pp. 78–9.Google Scholar

⁵⁸ Saint-Hilaire, A. de, Viagem pelas Províncias do Rio de Janeiro e Minas Gerais (São Paulo/Belo Horizonte, 1975), p. 119.Google Scholar

⁵⁹ Ibid., pp. 119, 171–2, 228, 284, 289 (Quotation p. 284).

⁶⁰ Saint-Hilaire, A. de, Viagem pelo Distrito dos Diamantes e Litoral do Brasil (São Paulo/Belo Horizonte, 1974), p. 47.Google Scholar Saint-Hilaire specifically mentioned that cotton was shipped into Tapera from northern production areas such as Peçanha and Minas Novas.

⁶¹ Ibid., pp. 102, 111.

⁶² Saint-Hilaire, A. de, Viagem as Nascentes do Rio São Francisco (São Paulo/Belo Horizonte, 1974), pp. 91, 96, 121, 136.Google Scholar

⁶³ Saint-Hilaire, A. de, Segunda Viagem do Rio de Janeiro a Minas Gerais e a São Paulo (São Paulo/Belo Horizonte, 1975), pp. 38, 49, 52Google Scholar (Quotation p. 52).

⁶⁴ Sturz, J. J., A Review, Financial, Statistical & Commercial, of the Empire of Brazil and its Resources: Together with a Suggestion of the Expediency and Mode of Admitting Brazilian and Other Foreign Sugars into Great Britain for Refining and Exportation (London, 1837), pp. 111–12.Google Scholar

⁶⁵ Ibid., 110–12.

⁶⁶ Even those nominal lists that tended to contain information about the occupation only of heads of household registered numerous spinners and weavers, usually widows or single women who were running their respective domiciles. Thus, absolutely none of the dozens and dozens of nominal lists from 1831 to 1840 failed to register the existence of the cottage textile industry in Minas.

⁶⁷ Saint-Hilaire provided one example of water power being applied to ginning operations. Since, as was already mentioned, his direct contacts with the workings of domestic industry were very rare, it would be difficult to generalise on the basis of this observation. Saint-Hilaire, , Viagem pelas Provincias…, p. 172.Google Scholar

⁶⁸ Berg, , The Age of Manufactures, p. 237.Google Scholar ‘Where in 1715 seven carders and twenty-five weavers kept two hundred and fifty worsted spinners employed, hand jennies reduced the weaving spinning ratio to one weaver to four spinners.’

⁶⁹ Saint-Hilaire, , Viagem pelas Provincial, p. 111.Google Scholar The author observed that in São João d'El Rei merchants ginned the cotton which was bought in outlying areas, however, he made no mention as to whether the prepared fibres were then shipped to Rio de Janeiro or sold locally. It may very well have been that in the towns and in those rural areas where cotton did not grow well producers purchased their cotton already ginned.

⁷⁰ Martins, , ‘A Indústria Têxtil Doméstica’, p. 88.Google Scholar

⁷¹ Clarkson, , Proto-industrialisation, pp. 53–4.Google Scholar

⁷² Oddly enough, in the 1873 Recenseamento sample a fair number of males–8.4% of the textile labour force–was engaged in the domestic textile industry. There is no way of knowing whether or not they were concentrated in spinning or weaving, since the data is aggregated for the whole industry. In the Alto Paranaíba region, where the cottage industry appears to have declined less than in the rest of the province, fully 14.3% of the textile labour force was male. Perhaps this demonstrates that, despite the general tendency toward de-industrialisation, domestic thread and cloth production could represent a more viable survival strategy than pure subsistence farming. On the other hand, the already-discussed unreliability of the Recenseamento data must always be born in mind.

⁷³ This idea is developed in Cardoso, F. H., Capitalismo e Escravidão no Brasil Meridional (Rio de Janeiro, 1977), pp. 208–12.Google Scholar

⁷⁴ See Berg, The Age of Manufactures, chap. 6.

⁷⁵ Ibid., p. 153. The quote is from Godelier, M., ‘Work and its Representations’, History Workshop, vol. 10 (1980).CrossRefGoogle Scholar

⁷⁶ Matos, R. J. da Cunha, Corografia Histórica da Provincia de Minas Gerais (1837) (Belo Horizonte, 1981), vol. 1, pp. 101–2.Google Scholar

⁷⁷ Sturz, , A Review…, p. 112.Google Scholar The author calculated the costs of transportation to the coastal markets as amounting to from 50 to 80% of the final price of mineiro goods.

⁷⁸ Burton, R. F., Exploring the Highlands of Brazil (New York, 1969), p. 134.Google Scholar In 1867 Burton observed that the cotton goods produced in São João d'El Rei were ‘strong and outlast many lengths of machine woven stuffs…’

⁷⁹ See Canabrava, O Algodão em São Paulo, for the effects of this crisis in São Paulo.

⁸⁰ See Libby, , Transformação e Trabalho, pp. 214–56.Google Scholar

⁸¹ See P. Kriedte, ‘Proto-industrialisation between Industrialisation and De-industrialisation’, in Kriedte, , Medick, and Schlumbohm, , Industrialisation before Industrialisation, p. 139.Google Scholar

⁸² See Libby, , Transformação e Trabalho, pp. 384–8Google Scholar. The annual earnings of early textile workers in Minas did not even equal the annual costs of renting and maintaining an adult male slave in the 1870s and 1880s.

⁸³ Arquivo Público Mineiro, Seção Provincial, Códice 570, 1855, manuscript.

⁸⁴ Schlumbohm, , ‘Relations of Production–Productive Forces Crisis in Proto-industrialisation’, in Kriedte, , Medick, and Schlumbohm, , Industrialisation before Industrialisation, pp. 110–11.Google Scholar

⁸⁵ Ibid., pp. 98–101.

⁸⁶ Cf. Franco, A. A. de M., ‘Páginas do Passado Brasileiro: Tropas e Tropeiros’, Cultura Político, vol. 2, nos. 12 and 13 (1942), pp. 233–8, 262–5.Google Scholar

⁸⁷ Thomson, , ‘Variations in Industrial Structure in Pre-industrial Languedoc’, p. 91.Google Scholar