The Invention of the 'Tropical Worker': Medical Research and the Quest for Central African Labor on the South African Gold Mines, 1903-36 Randall M. Packard The Journal of African History, Vol. 34, No. 2. (1993), pp. 271-292. Stable URL: http://links.jstor.org/sici?sici=0021-8537%281993%2934%3A2%3C271%3ATIOT%27W%3E2.0.CO%3B2-C The Journal of African History is currently published by Cambridge University Press.

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at http://www.jstor.org/about/terms.html. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use. Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at http://www.jstor.org/journals/cup.html. Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission.

The JSTOR Archive is a trusted digital repository providing for long-term preservation and access to leading academic journals and scholarly literature from around the world. The Archive is supported by libraries, scholarly societies, publishers, and foundations. It is an initiative of JSTOR, a not-for-profit organization with a mission to help the scholarly community take advantage of advances in technology. For more information regarding JSTOR, please contact [email protected].

http://www.jstor.org Sun Mar 9 06:45:16 2008

Journal of .4frican History, 34 (1993), pp. 271-292 Copyright 0 1993 Cambridge University Press

T H E INVENTION O F T H E 'TROPICAL WORKER':

MEDICAL RESEARCH AND T H E QCEST FOR CENTRAL

AFRICAN LABOR O N T H E S O U T H AFRICAN G O L D

M I N E S , 1903-36

BY R A N D A L L M. P A C K A R D

Emory Unizlersity A T the end of the South African War, in 1902, the Transvaal Chamber of Mines began to contemplate the possibility of recruiting African mine labor from territories beyond South Africa's northern border. Faced with local African resistance to low wages and unhealthy conditions, the mines hoped to meet their expanding labor requirements through the recruitment of men from the 'tropical' areas of Central Africa. By 1905, j,ooo Central African recruits were at work on the mines. T h i s number grew to 25,000 by 191I . F r o m its outset, the utilization of Central African workers was problematic for the industry that employed them, the Imperial government in London, government officials in the Transvaal and colonial administrators in the territories from which Central African recruits were being drawn. Most of all, it was problematic for the recruits themselves, for they died at an extraordinarily high rate. As Alan Jeeves has shown, pneumonia, meningitis, influenza and a range of other diseases took a much higher toll among Central African workers than among other African workers.' Death rates exceeding I O O per 1000 per annum were common during the first decade of Central African recruitment. Of all the ailments that plagued Central African workers, by far the worst was pneumonia, accounting for over half of Central African mortality. T h e high mortality rates of Central African workers before 191I resulted from a combination of factors. First, they were susceptible to certain strains of pathogens to which they had not previously been exposed. Second, the conditions of their recruitment were often very difficult. Central African workers often travelled long distances by foot. Moreover, recruiters provided them with few provisions for the journey to the mines. T h u s Central African workers often arrived on the mines in a seriously weakened condition. Finally, Central African mine workers were inexperienced with mine life and had little knowledge of what one needed to know to survive. I n most of these respects, the early disease experience of Central African workers does not appear to have been markedly different from the initial experience of earlier cohorts of workers coming from other areas of southern Africa during the 18gos, though health statistics for this period are admittedly scarce. Like those who had come before them, Central African workers were paying the high cost of transition to the unhealthy and dangerous conditions of mine employment. T h e y did not, therefore, represent a special category of Alan Jeeves, Migrant Labor in South Africa's Mining Economy 1890-1929 (Kingston, 19851, 323-33.

272

R A N D A L L M. P A C K A R D

workers, possessing a unique susceptibility to ill-health. Nonetheless, this was precisely how they came to be viewed within the mining industry. In 191I , the Minister of Native Affairs warned the mining industry that if the mines did not reduce what he termed 'the regular slaughter' of Central African workers, the government would be forced to ban their recruitment.' In response to this ultimatum, mine managers began to institute preventive workplace reforms designed specifically to reduce mortality rates among Central African workers. In the end, however, the mine owners' concerns about keeping down costs restricted measures of this hygienic type.3 I n place of meaningful and costly reforms of conditions in the work place, the mine owners invested in what appeared to be a quicker and cheaper solution to their problem, the development of a vaccine to protect Central African workers from pneumonia. In a stormy meeting with the Minister of Native Affairs, in which the minister threatened to close down Central African recruitment, the Chairman of the Executive Committee of the Transvaal Chamber of Mines used the promise of an effective vaccine to defend the mines' efforts to reduce Central African mortality rates. ' T h e absence of epidemics of pneumonia would no doubt cause an improvement, and it is for that reason that this year we have decided to establish a bacteriological institute to try to cope specially with this one disease, and to try and see if we cannot produce a serum which will immunize boys against the disease. '' In addition to establishing a bacteriology laboratory, the Central Mining group, which operated a large portion of the Rand gold mines, invited Sir Almroth Wright, a leading British bacteriologist who had developed a vaccine for enteric fever and was at the time doing research on a pneumonia vaccine in Britain, to move his research to Johannesburg. Sir Almroth accepted the invitation, apparently attracted by the prospect of testing his vaccines on a large population of highly susceptible African workers. By October of 191I , Sir Almroth had developed a vaccine and begun experimenting with it on Central African recruit^.^ Over the next 2 j years the mining industry supported pneumonia research and the quest for an anti-pneumonia vaccine, seeing them as keys to the Numerous communications from the Native Affairs Department to the Chamber of Mines stressed the need for the Chamber to improve working conditions, especially for tropical miners in the three years immediately prior to the closing of mine recruitment north of latitude 2 2 south. T h e following passage from a letter addressed to the Chairman of the Chamber of Mines from Henry Burton, Minister of Native Affairs, 1 2 June 191I (Chamber of Mines Archives [CMA] NIL+,Native Mortality, I ~ I I ) emphasizes , the seriousness of the situation: ' I have discussed this subject of the heavy mortality of tropical Natives with my colleagues, who are in agreement with me that unless a decided improvement can be effected at an early date the Government will have no alternative to the measure of entirely prohibiting the introduction of tropical natives'. See R. Packard, W h i t e Plague, Black Labor : Tuberculosis and the Political Economy of Health and Disease in S o u t h Africa (Los Angeles, 1989), 161-4, for a discussion of these efforts. ' C M A , Health Conditions Tropical Natives, 1911-12, 'Notes of Proceedings of a Meeting held on 13 July 1911 between the Minister of Native Affairs of the Union of South Africa, the Executive Committee of the Transvaal Chamber of Mines, and the Board of Management of the Witwatersrand Native Labour Association, Limited', 27. A. P. Cartwright, Doctors on the ~l/lines:A History of the ~l/linesMedical Association of South Africa (Cape T o w n , 1971), 26. -

"

storehouse of Central African labor. It was a key, moreover, that could be purchased for much less than the cost of improving compound and working conditions. I n 1904, the Chamber had estimated that it would cost one and a quarter million pounds to increase the minimum space requirement in the 'native compounds' to 300 cubic feet per occupant, as had been recommended by the Transvaal's Medical Officer for Health.6 T h e bacteriological laboratory established by the mining industry had cost only ~ 4 0 , 0 0 0 Even . with the salaries of Sir Almroth and his staff, medical research was cheaper than environmental reform T h e mining industry's quest for an anti-pneumonia vaccine has been described by A. P. Cartwright, and it is not the intention of this paper to retell this ~ t o r y T . ~h e paper examines instead how the mining industry successfully employed the promise of an effective anti-pneumonia vaccine, which in fact never materialized, to defend its use of Central African labor and to offset calls for expensive reforms in living and working conditions on the mines.8 I t is about the discourse of medical research - the articulation of ideas concerning Central African worker mortality, its causes and control and how the mining industry appropriated and controlled this discourse. T h e appropriation of medical science and its subordination to the requirements of industrial capital accumulation have been discussed by numerous writer^.^ Within the mining industry Cartwright, Packard, Richardson and Burke1' and others have explored the ways in which the work of mine medical officers was made to serve the economic interests of the industry that employed them. T h e present paper, however, suggests that the relationship between medical science and the interests of the mining industry was both more complex and more subtle than the simple co-optation previously been suggested. T h e mining industry's effort to present an anti-pneumonia vaccine as a panacea for the problem of tropical worker mortality was challenged not only by critics outside the industry but also by its own medical community. Individual medical officers were much more independent in their opinions than previous studies have indicated. Many were, in fact, openly critical of the value of an anti-pneumonia vaccine for limiting the death of tropical workers. Some called for more serious reforms in the conditions of work. These criticisms, voiced publicly at meetings of the Transvaal Mine Medical Officers Association, ran contrary to the interests of the industry that employed them and had to be muted. T h e paper argues that the industry was able to overcome these internal criticisms by controlling the terrain of discourse on Central African worker T h e Transvaal, Report of the Coloured Labour Compound Commission (Pretoria, ~ g o j ) , Cartwright, Doctors on the Mines. I would like to thank D r s Cory Kratz, Alan Jeeves, Shula Marks, Barbara Rosenkratz and Joseph Miller for commenting on earlier drafts of this paper. Research support for the paper was provided by grants from the Social Science Research Council and Tufts University. V. Nal-arro, Medicine Under Capitalism (New York, 1976); D . Rosner and G. Markowitz, Deadly Dust, Silicosis and the Politics of Occupational Disease in Twentieth Century America (Princeton, I 99 I ) ; L. Doyal, The Political Economy of Health (London, 1979) ; M. Turshen, The Political Ecology of Disease in Tanzania (New Brunswick, 1986). l o Cartwright, Doctors on the wines; Packard, White Plague, Black Labor; G . Burke and P. Richardson, ' T h e profits of death: a comparative study of miner's phthisis in Cornwall and the Transvaal', J. Southern Afr. Studies, I V (1978), 147-272. xl-.

274 R A N D A L L M. P A C K A R D mortality, defining the problem as resulting from cultural and ultimately biological susceptibilities of 'tropical ' workers, rather than from the working and living conditions on the mines. Central to the mining industry's success in defining the discourse on Central African worker mortality was its professionalization and hence validation of medical research on the mines. Defining the problem of Central African mortality in terms of the inherent susceptibilities of Central African workers involved the social construction or invention of the Central African worker as a distinct social category with particular cultural and biological characteristics that differentiated him from other African workers on the mines. T h i s process of social construction resulted in the invention of the 'tropical worker' T h e term 'tropical worker' has become common usage among representatives of the mining industry as well as among scholars who have studied the mining industry. In this discourse it appears as a naturally occurring category. T h e paper argues instead that the term represents a rhetorical invention that emerged at a particular historical moment as part of the mining industry's efforts to define Central African mortality in ways that would limit its financial liabilities, while insuring continued access to this important source of labor. T h e invention of the 'tropical worker' within the South African gold mining industry was part of a broader history of western medical discourses on Africa and Africans. Drawing on Sander Gilman's work on the medical construction of difference in fin de siecle Europe,'' and in turn on Foucault's studies of the medicalization of power,'' recent studies by Comaroff, Vaughan, Packard and others13 have shown how the language and practice of colonial medical science served to define African identity.'Vackard and Epstein, moreover, have argued that images of the African body, disease and sexuality constructed during the colonial period have continued to define western medical discourse on Africa within the context of recent A I D S research.15 While these studies have made effective use of techniques of discourse analysis to uncover the constructive nature of western medical ideas concerning African clothing, sexuality, psychology, anatomy and disease, they have paid less attention to the historical matrices of social, political and economic relations within which such discursive processes took form. As a result, it is often difficult to understand why certain characteristics were ascribed to specific groups of Africans at particular moments in time. l 1 Sander Gilman, Pathology and Dzfference: Stereotypes of Sexuality, Race and Difference (Ithaca, 1985). l 2 Michel Foucault, T h e Birth of the Clinic (London, 1976); T h e History of Sexuality (London, I 979) ; ~l/ladnessand Civilization (London, 1989). l 3 R. Packard, ' T h e "healthy reserve " and the "dressed native " : discourses on black health and the language of legitimation in South Africa', American Ethnologist, XVI (1989), 77-93 ; Megan Vaughan, Curing their Ills : Colonial Poever and African Illness (Palo Alto, 1991); Jean Comaroff, ' T h e diseased heart: medicine, colonialism and the black body', in Margaret Lock and Shirley Lindenbaum (eds.), Analysis i n Medical Anthropology (Dordrecht, 1991). l4 They have also been influenced by the broader study of the cultural construction of difference found in such works as V. Y. Mudimbe's T h e Invention of Africa (Bloomington, 1988); Christopher Miller, Blank Darkness (Chicago, 1985); and more broadly, Edward Said's Orientalism (New York, 1979). R. Packard and P. Epstein, 'Epidemiologists, social scientists, and the structure of medical research on A I D S in Africa', Social Science and Medicine, X X X I I I (1991), 771-94.

T h e present paper suggests that invention of the 'tropical worker' as a distinct social and medical type reflected a general typologizing tendency within colonial medical culture. Moreover, it drew on a storehouse full of contemporary western cultural images of African 'otherness'. At the same time, however, the construction of the 'tropical native' on the Rand was strongly encouraged by the specific economic interests operating within the South African gold mining industry during the second and third decades of this century. Faced with an impending labor crisis and by fiscal constraints which limited their willingness and ability to improve working and living conditions on the mines, mine managers drew on these background images to define Central African workers in ways that deflected attention away from the mines' failure to effect environmental reforms. T h e nature of this definition, moreover, changed over time from cultural to biological as the interests of the mining industry shifted from making excuses for the high mortality among their workers to claiming the ability to prevent it. THE INVENTION OF THE TROPICAL WORKER

When the issue of high mortality rates of among African recruits from Central Africa first emerged in 1903, mine medical officers and Chamber of Mines officials referred to such workers as 'natives coming from tropical or sub-tropical districts' or more commonly as workers coming from British Central Africa. T h e terms 'tropical workers ' or simply ' tropicals ' did not appear in any of the documents from this early period. I t is impossible to identify the precise moment when these terms entered the discourse on the health of African mine labor. However, they did not become common usage within the mining industry prior to 191I , when the debate over the use of Central African mine labor intensified. I have found no use of the term before 1908, when it appeared in a report on 'Conditions Affecting the Health of Underground Workers on the Mines of the Witwatersrand'. T h e report was prepared by D r s L. G. Irvine and D . MacCauley, who had earlier given testimony on the health of 'native workers' before the Coloured Labor Compound Commission. I n a chart titled 'Mortality amongst Native Mine Workers: Comparative Territorial Death-Rates per 1000 from Disease', the terms 'tropical' natives and 'non-tropical' natives appeared in quotation marks.16 T h e use of quotation marks suggests that the terms were not common usage at the time. All other references to those workers in this report employed the same terminology used in 1903, i.e. 'natives from tropical areas, ' or refer to their specific territory of origln (e.g. Nyasaland) or ethnic origin (e.g. Nyassas). T h e Final Report of the Mine Regulations Commission's discussion of ' Health of Natives', published in I 9 10, employed the same restricted usage of the terms 'tropical' and 'non-tropical', again in quotations." A year later, however, both terms appeared frequently in published and unpublished sources. T h u s G. N. Maynard in his report to the Medical Officer of Health, Johannesburg, on 'Mortality amongst Natives Employed on Mines and l 6 CMA, Health of Underground Workers, 'Conditions affecting the health of underground workers on the mines of the Witwatersrand, Statement by Drs L. G. Irvine and D. Macaulay May, 1go8',Pretoria, 1909. l 7 Transvaal Government, Final Report of the Mining Regulations Commission (Pretoria, 1910).

12

A F H 34

276

R A N D A L L M. P A C K A R D

Works in the Labour Area of the Transvaal' employed the terms 'tropical natives ', 'tropical boys ' and ' tropicals ' without quotation at several points.18 Similarly, in a letter to the chairman of the Transvaal Chamber of Mines dated 12 June 19I I , the Minister for Native Affairs referred initially to 'Natives from tropical areas'.lg However, by the end of the letter he began employing the term 'tropical natives' without quotations. All of the participants in the meeting between the Minister of Native Affairs, the Executive Committee of the Transvaal Chamber of Mines and the Board of Management of W N L A on I 3 July 19I I employed the terms tropical natives' and 'tropicals'. From this point on, the terms became common elements in the discourse on 'native labor'. I n short, the terms 'tropical workers' and 'tropicals' entered the discourse on Central African labor at precisely the moment that the mining industry's access to this source of workers came under attack from the Minister of Native Affairs in 191I . T h e introduction of the terms 'tropical natives' or 'tropicals' into the discourse on Central African labor may have resulted from a tendency to be economical in the use of words. After all, 'tropical native' or more simply ' tropicals ' captured the same meaning as 'natives from tropical areas ', but with fewer words. Yet the elision of human subject with geographical location, whatever its origins, had significant implications for the way in which workers coming from tropical areas came to be perceived. As a result of this usage, both the workers and their health problems became more strongly associated with their tropical ' o t h e r n e s ~ ' . ~ ~ T h i s rhetorical shift contributed to, and accompanied, an effort on the part of mining officials to construct Central African workers as culturally distinct from other African workers and as possessing a weaker physique. Through this process the health problems of Central African workers came to be viewed as both unique and inherently of their own making. T h e chairman of the Chamber of Mines clearly articulated this 'victim blaming' strategy in his defense of the mines treatment of 'tropical workers ' before the Minister of Native Affairs in I 9 I I :21 Chairman: When they [tropical workers] were first introduced it was not realized that they were a dzfferent class of boy to the East Coast boys south of latitude 22, amongst whom the death rate had been reduced to about 3 0 . [emphasis added]"

T h e Chairman went on to discuss how the mines issued clothing to the tropical workers to prevent sickness and how they discovered that chill l8 C h l A N 14, Native Mortality, I 9 I I , G. D. hlaynard, 'Reports re mortality amongst natives employed on mines and works in the labour area of the Transvaal', 5 Jan. 191I . CMA NIL+,N ative Mortality, 1911, Chairman, Transvaal Chamber of Mines to Minister of Native Affairs, 12 June 191I . 2 0 T h e difference in meaning is similar to that which occurs when one refers to people from the state of Georgia or Alabama in the United States as 'people from the South' as opposed to 'Southerners'. T h e latter term clearly carries with it a series of cultural associations that are not projected by the former phase. CMA, Health Conditions-Tropical Natives, 1911-12, 'Notes of Proceedings of a hleeting held on 1 3 July, 191I between the hlinister of Native Affairs of the Union of South Africa, the Executive Committee of the Transvaal Chamber of Mines, and the Board of hlanagement of the Witwatersrand Natives Labour Association, Limited'. 2 2 CMA, Health Conditions - Tropical Natives, I 9 I 1-1 2, ' Notes of Proceedings.. . 13 July I ~ I I ' ,2 0.

occurred nonetheless when the workers left the mine and returned to their barracks in perspiration-soaked clothing : ,4nd then it was that the system of change houses was introduced. The system of change houses we had great hopes for, but lately we have realized that greater attention must be paid to change house regulations being carried out. We have to deal with a class of boy who is not willing to clothe himself. .. [emphasis addedIz3 4 t another point in the discussions the Chairman commented further on the inherent irresponsibility of Central African workers in connection with the issue of clothing : There is the question of deducting from the boy's pay for clothing supplied by the mine. At present this clothing is supplied. He does not wear clothes in his own country, and the thing is to see that he uses his clothes and does not put them into his box or sell them for a small sum of money.z4 O n the question of excessive ventilation in the compounds and the resulting health problems it created, the Chairman again subtly shifted responsibility on to the 'tropical worker ' : M r Pritchard [former hlinister of Native Affairs] and many of the doctors will bear me out when I say that the ventilation in vogue is quite foreign to the native, especially the native who comes from tropical areas.. . [emphasis addedIz5 Following further efforts by the Chamber representatives to deny responsibility for ill health of Central African workers and to suggest that they had not realized the full dimensions of the problem until recently, the Minister provided a litany of mortality statistics and pointed out that there had been government concern about the health of Central African natives from the very beginning. Again the Chairman responded by shifting responsibility onto the tropical workers themselves: We had a report issued by Dr Turner a little time ago, which went to show that the actual condition of the native in his home was very bad, and that the coming of these boys here, where they m i x with a better class of native, is certainly improving their condition when they go back. [emphasis addedIt6 When the minister replied, ' T h o s e that survive', suggesting that many died and were unable to return home, the Chairman rebutted, ' T h o s e who die here might be of such low physique that they would have died in any case'. These passages reveal how mine officials constructed 'tropical workers ' as both culturally and physically susceptible to health p r ~ b l e m s . ~Not ' surprisingly, several of the images employed by Chamber officials reflected a CMA, Health Conditions- Tropical Natives, 1911-12, 'Notes of Proceedings ... I3 July 191I',21. 2 4 CMA, Health Conditions- Tropical Natives, 1911-12, 'Notes of Proceedings ... 1 3 July I ~ I I 29. ',

"j CMA, Health Conditions - Tropical Natives, I 9 I 1-1 2, 'Notes of Proceedings.. . 13 July I ~ I I 31. ',

26 CMA, Health Conditions- Tropical Natives, 1911-12, 'Notes of Proceedings ... 1 3 July I ~ I I 26. ',

'' This pattern occurred more broadly of course in the history of medical discourse on 'tropical diseases', a term which became a part of European medical terminology at the end of the nineteenth century. T h e term tended to define a wide range of health problems among colonized populations of the world in terms of climate and ecology, while directing attention away from social and economic factors which in many cases contributed to health problems in these peoples.

278 R A N D A L L M. P A C K A R D wider discourse on race and disease that was prevalent in South Africa at this time. T h e invention of the 'tropical' native incorporated the images of difference associated with the 'dressed native'. T h e 'dressed native' was an image of cultural maladaptation applied to newly urbanized Africans that attributed a wide range of African social and economic problems to their unsuccessful attempt to shift from 'barbarism to civilization'. T h i s failure was symbolized in the African's inappropriate use of clothing.28Yet these images of inappropriate or aberrant behavior were applied to 'tropical workers' in a way that distanced them not only from the European worker but also from other African workers. It should also be pointed out that the tendency to ascribe particular susceptibilities to disease as well as other social and behavioral characteristics to specific 'tribal groups' was common by this period. Megan Vaughan in her recent study on colonial power and African illness points out that 'biomedical discourse in Africa operated through the specification of the features of groups, rather than the minutiae specification of the features of individuals T h e construction of the 'tropical worker' was unique, however, in the particular set of cultural and physical characteristics it ascribed to a population of Africans it defined not on the basis of ethnic or even cultural identification but rather as a result of their habitation of an imagined 'tropical ' world. Central African workers came from a very wide range of social, cultural, geographical and climatological environments and often had little in common other than their status as colonial subjects and mine recruits, a nondistinguishing status they shared with all African mine workers. Nonetheless, within the mining industry's discourse on African health, 'tropical workers ' took on a homogenous and exotic cultural identity associated with their 'tropical' origins. Whatever distinctions might have been made at an earlier point in time between ' h'yassas ', ' Ndebeles ' or 'Tongas ' were submerged within the larger homogenizing category of ' t r ~ p i c a l ' . ~ ' I n conjunction with efforts to blame the victim, the Chamber defended their access to tropical labor by promoting Sir Almroth Wright's work on an anti-pneumococcal vaccine. A drop in pneumonia mortality following the introduction of inoculation in the last two months of I Q I I and first two months of I Q I led ~ the chief medical officer for the Witwatersrand Native Labour Assdciation, the W N L A , which was primarily responsible for recruiting Central African labor, to claim that the inoculations were a success.31D r S. F. Lister, who carried out a study comparing the pneumonia '"ackard, ' T h e "healthy reserve" and the "dressed native"', 80-83. " Vaughan quotes the following statement by the medical superintendent of the Robben Island Asylum in Cape Town in 1890: ' T h e pure native races, like the Zulu and Kaffirs, are seldom affected with leprosy; but among the Korennes and cross-breeds between native women and nomadic Boers of the coast districts are to be found a large number of cases ... '. She goes on to cite similar comparisons made by European doctors working in Northern Rhodesia in 1898, between the 'promiscuous' and 'insanitary' Bisa and the 'sanitary' Bemba. Finally she shows how western psychiatry ascribed psychiatric characteristics to specific ethnic categories; Vaughan, Curing their Ills, 8 1 . 30 By the 1920s, the only characteristic that joined the various peoples who fell under the category of 'tropical' was the mining industry's inability legally to recruit them. 31 CMA, Health c o n d i t i o n s Tropical Natives, 191 1-12, G. A. Turner to C . W. Dix, Sec. WNLA, 5 Feb. 1912.

experience of inoculated and non-inoculated Central African workers at the Premier Mines, supported this conclusion. Accordingly, the chairman of the Chamber of Mines, in a letter to the Minister for Native Affairs, expressed the utmost confidence in the vaccine. H e noted that, while the experiments were still in progress, 'there is every reason to think that they will be attended by a very large amount of success'.32 Others within the mine medical community were more sceptical of the value of inoculations and pointed to serious errors in Lister's evaluation methods. Moreover, despite the continued use of inoculations, pneumonia mortality rose in I 9 I 3. In the same year, George Maynard, medical advisor to the W N L A , conducted a statistical analysis of mortality records, comparing inoculated and non-inoculated workers on several mines. H e concluded that ' I n these records there is no evidence that prophylactic inoculation has had any influence in case mortality, except possibly for a very short time after i n o ~ u l a t i o n ' . ~ ~ Neither the Chamber's efforts to blame the victim nor their attempt to sell the wonders of the anti-pneumococcal vaccine succeeded in distracting the government from the continued high levels of Central African mortality occurring on the mines. Accordingly, the Government banned recruitment north of 22" S latitude in the middle of I 9 I 3. Yet the Chamber's efforts were not a complete failure. They had begun to stigmatize Central African workers and to shift the focus of attention away from the mines and onto the susceptibility of this class of workers. This shift would eventually lead to a refocusing of the problem along biological lines and to the legitimation of efforts to solve the problem of tropical mortality through inoculation. Central to the completion of this process, as we shall see, was the creation of the South African Institute for Medical Research and the professionalization of medical research on the mines. REFORMS VERSUS MEDICAL T E C H N O L O G Y : THE CREATION O F THE SOUTH AFRICAN INSTITUTE FOR MEDICAL RESEARCH

T h e withdrawal of Central African labor had a sobering effect on many of the Rand mining houses. For some it meant that the era of total worker neglect was over. T h e costs of continuing to consume labor like a raw material were becoming too high. As Sam Evans, chairman of the Crown Mines Company, 3 V M L 4 ,Health conditions-Tropical Natives, 1911-12, President of the Transvaal Chamber of Mines to the Minister for Native 'qffairs, 13 March 1912. 3 3 Quoted in Cartwright, Doctors on the Mines, 27. T h e Chamber's efforts to present Sir Almroth's vaccine as a panacea for the problem of 'tropical' labor mortality would seem to suggest that they viewed the cause of this mortality as primarily biological, i.e. it was due to susceptibility to a particular pathogen. T h i s would seem to contradict the arguments made before the Minister of Native Affairs in July in which they ascribed tropical worker mortality to cultural and behavioral differences. T h i s is not necessarily the case. Instead Chamber officials appear to have viewed Sir Almroth's vaccine as a solution to a disease which took a higher toll among tropical workers because tropical workers Lvere culturally different from other workers. Chamber officials would later come to view the cause of tropical susceptibility in purely biological terms. T h i s shift, as we will see, reflected the changing economic interests of the mining industry and the professionalization of medical research on the mines.

280

R A N D A L L M. P A C K A R D

noted in a memorandum to Sir Lionel Phillips, chairman of Rand Mines Limited, ' If we let things slide we shall lose the Portuguese natives as well as the tropical^'.^^ In addition, the loss of shift time due to accidents or disease was proving very costly. In a survey of conditions on the mines in April of I 9 I 3, the Native Labour Bureau found that 10 per cent of the men were incapacitated every day. This amounted to 191,343 lost shifts, or nearly one shift per W ~ r k e r . ~ ' I n an effort to generate reforms, Evans invited Colonel William Gorgas to visit the Rand in 19I 3 and to make recommendations for lowering mortality rates. Gorgas' success in lowering pneumonia mortality rates among 'negro' workers on the Panama Canal had greatly impressed Evans, who had visited the Canal in 1012. Colonel Gorgas, along with two of his medical officers, visited the Rand in December 1913. They carried out an inspection of the mines over the next three months and presented their report to the Chamber on 25 February. Gorgas' recommendations, while hardly startling by today's standards, were radical by those of the mining industry in I 9 I 4 in identifying the need for reducing overcrowding, improving the diet and expanding and centralizing medical services. Gorgas criticized the industry for not providing adequate accommodation for its work force and recommended a minimum of fifty square feet per man in every room. H e went on to propose an even more radical solution to the housing problem, the creation of family housing. He also criticized the mine diet and proposed that the ideal system for feeding workers was in a family setting. H e recommended that the workers be paid cash to buy their own food that their wives could prepare in their Gorgas' emphasis on working and living conditions was not well received by the mining houses. Although the mines made improvements over the next twenty years, the changes came slowly and, as before the ban, were limited in scope.35 One of the prime movers in these reform efforts was D r A. J . Orenstein, an assistant to Gorgas, who was hired by the largest mining house, Rand MinesICentral Mining group, to head up its sanitation department. Orenstein, like Gorgas, was a great advocate of disease prevention through sanitation measures. Over the next two decades he became a major voice for environmental reform within the mining industry. Though he was widely respected within the industry, his recommendations were not readily accepted. For example, in 1916 he recommended that dividers be placed between the bunks of workers to reduce opportunities for disease transmission. T h e adoption of this practice was still incomplete by 193I .38 31

Cartwright, Doctors on the Mines, 30.

" Jeeves, Migrant Labour

in S o u t h Africa's Gold Mining Economy, 232-4.

William Gorgas, Recommendations as to Sanitation Concerning Employees on the

Mines on the R a n d (Johannesburg, 1914). 3' See Packard, W h i t e Plague, Black Labor, ch. 6 , for a discussion of the achievements and limits of mining reforms between the wars. 38 Transvaal Archives Depot [TAD], Government Native Labour Bureau [GNLB] 386 33/44, D r Culver, 'Investigations into Health Conditions on the Mines, 193 I '. Discussions within the Chamber of Mines during the 1920s suggest that Orenstein's views were respected and that he was viewed as a major figure within the mine medical community. Yet one also has the impression that the Chamber officials respected him more for the image of reform that he projected to the wider South '4frican and international community than for the specific reforms that he proposed. T o the latter their was consistent resistance. 36

T h e mines did not, however, ignore the tropical health problem. Instead, they reconstructed the issue in terms that favored themselves. In the same year that Gorgas was invited to review health conditions on the Rand, the Chamber funded the creation of the South African Institute for Medical Research. T h e Institute represented a major step in the professionalization of medical research on the mines. It was staffed with professional researchers who were freed from the day-to-day responsibilities of caring for the sick and wounded. Over the years the Institute gained an international reputation for its work on silicosis, tuberculosis and malaria. Yet its first task was to develop an anti-pneumonia vaccine to combat 'tropical' mortality on the mines. In attempting to do so, the Institute contributed to the medicalization of the problem of tropical mortality and legitimized a narrowly biological response to health problems on the mines that largely ignored the circumstances in which miners lived. By the time the Institute took up the problem of pneumonia, 'tropical' workers had been defined as the primary cause of 'tropical' mortality. Given this presumption, there was little need to consider wider environmental factors. One could focus in on the worker himself. This was precisely what researchers at the Institute chose to do. They emphasized the physical characteristics of the tropical worker and specifically his susceptibility to particular strains of the pneumococcal b a ~ t e r i u m . ~Within ' the Institute, pneumonia, and tropical health in general, became defined as a problem of bacteriology and immunology rather than occupational health and safety. T h e Institute directed its resources toward identifying the various strains of pneumococci affecting mine workers and developing vaccines to provide protection against them. It made no effort to examine or reform the working and living conditions that were contributing to high levels of infection within the mining industry. T h e Institute's biological definition of the problem differed significantly from the broader sanitation perspective of physicians like Irvine, MacCauley and Maynard, who had analysed the problem of Central African worker mortality earlier in the century in terms of air quality, living conditions and length of employment, as well as other employment condition^.'^ T h e medicalization of the problem of Central African worker mortality gave medical legitimacy to the industry's efforts to avoid responsibility for these deaths by blaming the victim. It also enabled the mines to claim that 39 It is also significant that this shift in emphasis reflected, and perhaps contributed to, a broader transformation in the discourse on race and disease in South Africa at this time. As I have noted elsewhere, explanations for African susceptibility to tuberculosis shifted from behavioral (or cultural) explanations to biological ones during the teens and twenties. R. Packard, 'Tuberculosis and the development of industrial health policies on the Witwatersrand, 1902-193z', J . Southern Afr. Studies, X I I I (1987), 187-209. T h e redefinition of pneumonia along narrowly medical lines resembles in many respects the shift that occurred in the definition of silicosis within public health circles in the United States between the end of the nineteenth century and the 1920s. David Rosner and Gerald Markowitz assert that the new breed of industrial hygienists that emerged in the U S during this period were 'primarily physicians whose training led them to see industrial disease in much narrower terms. T h i s group emphasized personal hygiene, the laboratory and the identification of specific toxins or germs in their attempt to improve workers' health ... ' (Rosner and hlarkowitz, Deadly Dust, 45-6) T h i s description accurately assesses the new cadre of professional medical researchers who came to work at the South African Institute of LIedical Research.

282

R A N D A L L M. P A C K A R D

they were doing something about tropical mortality while limiting their expenditures on more costly environmental reforms. Finally, as we will see, the narrowly medical definition of the problem of Central African mortality better served the industry's changing economic interests following the ban of tropical workers in 1913. Whether mining officials foresaw these benefits when they established the Institute for Medical Research is unclear. Their primary motivation was simply finding a medical solution to the problem of Central African worker mortality. What is clear is that these officials repeatedly employed the Institute's legitimizing function to push their case for the use of Central African labor. It is also clear that they did this despite the Institute's lack of success in developing an effective pneumoccocal vaccine.

POST-WAR L A B O R CRISES A N D THE ' S E L L I N G ' OF LISTER'S POLYVALENT PNEUMOCOCCAL VACCINE

Over the next twenty years, the mining industry, with the assistance of its medical experts at the South African Institute for Medical Research successfully constructed a set of arguments for the use of Central African labor that centered on the tropical workers' susceptibility to pneumonia and the mines' possession of a silver bullet solution to this problem in the form of an effective pneumococcal vaccine. These arguments were rolled out each time the industry faced a labor shortage. In fact the industry's interest in a vaccine appears to have ebbed and flowed in line with these shifts in labor demand. T h e first of these shortages occurred at the end of the First World War. In a letter addressed to the prime minister, the chairman of the Chamber of Mines, M r E. A. Wallers, wrote, 'As compared to the pattern eighteen months ago, when the mines were not short of labour to an appreciable extent, there are employed today on "mines and works" practically the same number of natives, but the number employed on the gold mines shows a decrease of 20,000, although their requirements have increased by over ~ o , o o onatives during that time'.41 T h e mines' labor shortage was caused by four factors. First, the demand for labor within the industry had increased as a result of an expansion of operations designed to meet the rising war-time demand for gold. I n addition, the mines had increased their use of African labor to offset shortages of equipment. Secondly, the demand for African labor in other industries was also increasing with the reopening of the diamond mines and the expansion of operations in the collieries. Thirdly, the recruitment of African labor for the war effort continued to pull labor away from the mines. Finally, the outbreak of Spanish influenza temporarily reduced the mines' African work force by 62 per cent in the second half of 1918.~' Within the context of this emerging labor shortage, the mine owners looked for ways to overcome the government's ban on Central African C M A N8, Native Labour Contingent, 1916-18, E. A. Wallers to Prime Xlinister Louis Botha, 7 June 1917. 12 H. Phillips, Black October : T h e Impact of the Spanish Influenza Epidemic of 1918 on S o u t h Africa (Pretoria, ~ g g o ) 3, .

recruitment, in effect since I 9 I 3. TO accomplish this, Chamber officials employed two sets of arguments. First they emphasized the costs attached to not providing the mines with an adequate labor supply. T h e y pointed particularly to the possibility that low-grade mines would have to shut down. I n doing so, the mine owners employed their perennial trump card, the threat of structural unemployment for white workers. I n a letter to the Minister for Mines in November of 1917, Wallers wrote ' I feel convinced that there would be a very large number - and here I refer particularly to the white workmen -thrown out of e m p l ~ y r n e n t ' . ~ ~ Secondly, Chamber officials tried to convince the government that Central African labor would be safe on the mines. T h i s was a difficult argument to make, given the minimal changes in working conditions that had occurred since the ban. T h e Chamber, nonetheless, pushed their case by claiming that the industry now possessed an effective anti-pneumococcal vaccine. During the war D r S . F. Lister had been appointed to the newly created South African Institute for Medical Research. H e had devoted the majority of his time to developing a more effective vaccine and to testing the vaccine on thousands of African workers. Lister, who was later knighted for his work in identifying various types of pneumococci, maintained that the vaccine was effective. H e and others pointed to the marked decrease that had occurred in pneumonia mortality following the introduction of the vaccine. T h e Chamber lost no time singing the praises of the vaccine in their letters to the government. Writing to the Prime Minister in June of 1917 about the mines' labor crisis, Wallers noted Under these circumstances, I venture to think that the time has come when the v e r y successful results which have attended Dr Lister's investigations at the South

-African Institute in the preparation of a prophylactic for pneumonia be tested on tropical natives on a practical scale. It will be remembered that the heavy mortality amongst tropical natives was mainly due to pneumonia, which accounted for more than half the total number of deaths. The disease is, however, now almost nonexistent on the mines where Dr Lister's prophylactic has been used. [emphasis addedI4' I n a second letter to the prime minister, dated 16 November 1917, the Chamber's president reiterated his claims about the wonders of Lister's vaccine and asserted that ' i t seems practically certain that, had D r Lister's vaccine been known at the time of the employment of the Tropicals on the gold mines, the very high death rate which was responsible for their removal would not have prevailed '." Wallers concluded the letter by suggesting that Parliament be asked to authorize the importation of 2,000 tropical workers for the express purpose of testing D r Lister's vaccine. These passages reveal the extent to which the problem of Central African worker mortality had been medicalized. T h e problem was tropical worker susceptibility to pneumonia. T h e solution was Lister's vaccine. Gone was 13 CM4 N8, Native Labour Contingent, 1916--18,E. 4 . Wallers to Minister of Mines, 16 NOV.1917. CM4 N8, Native Labour Contingent, 1916-18, E. A. SVallers to Prime Minister Louis Botha, 7 June 1917. '''CMA N8, Native Labour Contingent, 1916-18, E. A. Wallers to Prime Minister Louis Botha, 16 Nov. 1917.

z84 R A N D A L L M. P A C K A R D the language of cultural and behavioral maladaptation that marked earlier Chamber discussions of tropical labor mortality. Central African workers were now presented as flawed only in their susceptibility to pneumonia. T h e fact that tropical workers also fell victim to such other ailments as dysentery, malaria and bilharzia was either not discussed or dismissed because each of these conditions could be controlled medically.46 This new language of medical technology and capability not only differed from that used by Chamber officials in earlier arguments, it also reflected the changing economic and political interests of these officials. Prior to the War the mines had been trying to avoid costly reforms by placing responsibility for the death of the tropical workers they were currently recruiting on the workers themselves. Once access to these workers had been lost, the problem facing the mines changed. They needed to justify the reintroduction of Central African workers. T o argue, as they had before, that tropicals were culturally or physically unsuitable would undermine justifications that these workers should be reintroduced. Instead, arguments for the reintroduction of tropical workers presented these workers as particularly well suited for mine work. Their only problem was their susceptibility to a disease that the mines' medical program could control through vaccination. In this way the image of the tropical workers evolved in tandem with the mining industry's changing situation vis-a-vis their Central African labor supplies. T h e mining industry's re-definition of the problem of Central African worker mortality in terms of pneumonia and vaccine development was based on highly questionable evidence, despite its tone of scientific authority. Included in \Valler's letter to the prime minister was a copy of a memorandum entitled 'Brief Statistical Abstract of Results Following the Use of D r F. S. Lister's Pneumococcal Vaccine'. T h e abstract, submitted by the director of the South African Institute for Medical Research, compared the mortality rates for workers on three mines before the introduction of the pneumococcal vaccine in I 9 I 6 with figures for workers on these mines after inoculation was introduced. T h e figures in all three cases showed a marked decrease in pneumonia mortality rates. T h e figures, taken at face value, were impressive and clearly supported the Chamber's case that Lister's vaccine was an effective form of protection against pneumonia. Yet the figures and the arguments they appeared to support ignored one critical factor. T h e populations being contrasted were not comparable, due to the removal of nearly 20,000 Central African workers from the mine's work force between I 9 I 3 and 19I 6. T h e pre-vaccine control populations contained a higher percentage of Central African workers than the post-vaccination population. Their departure would have led to a dramatic decrease in pneumonia deaths even if the vaccine were totally ineffective." T h e impact of this factor on the resulting statistics was evident in the figures presented. Figures for the pre-vaccination population for two of the mines contained rates only from before the ban took effect in 1913.~' T h e pre-vaccination figures for the third mine, Crown hlines, however, ' 6 ClLlA 5 8 , Sative Labour Contingent, 1916-18, Employment of Tropical Workers, Percival \Yatkins to A. I . Girwood, 6 July 1917. " A. J . Orenstein, 'Vaccine prophl-laxis in pneumonia', Journal of the Medical A s sociation of S o u t h Africa, v (193 I ) , 339-46. Orenstein, 'Vaccine prophylaxis in pneumonia ', 345.

covered the period from I 9 10 to I 9 I 5 and thus included two post-ban years. T h e figures for the Crown mine accordingly showed the lowest pre-vaccine rates and the smallest decline in mortality. In short, available statistics were of little value for assessing the effectiveness of the vaccine. It was perhaps for this reason that the South African Institute for Medical Research suggested that a new trial be organized using 2,000 Central African workers. In fact, there appears to have been some doubt about the ability of the vaccine to protect Central African workers at all. Writing to D r A. I . Girwood, chief medical advisor to W N L A in July of 1917, D r A. P. Watkins of Village Deep mine suggested that it would be safe to reintroduce tropical workers due to the availability of Lister's vaccine. Yet he noted, ' I n this connection it might be advisable to discover if the strains of pneumococcus causing pneumonia, found in the natives of these [tropical] regions zcere the same as, or different f r o m , any of those at present in the Polyvalent Vaccine now in use [emphasis added] '.4"he statistical abstract sent to the Prime Minister sounded a similar note of caution: ' I t is confidently anticipated that a further and considerable reduction in the incidence and death rate will be achieved when other groups of the Pneumococcus, now in our possession, are included in the vaccine. 'jO Both of these notes of caution reflected an awareness among the mine medical professionals of what was to be the major obstacle to the development of an effective anti-pneumococcal vaccine. S o single pneumococcal agent Lvas responsible for the pneumonia cases occurring on the mines. There were in fact several such bacteria. I n constructing a vaccine Lister had tried to include a range of such agents producing a polyvalent vaccine. Yet there was doubt whether the ones he had included were those that really affected tropical workers. Moreover, as it would be discovered later, antigenic variation occurred in the bacteria at a rate that made it impossible to produce a vaccine that would be effective for more than a short time. In other words, the characteristics of the pneumococcal bacterium varied over time and space, making the creation of a vaccine extremely difficult. For some medical personnel doubts about the effectiveness of the vaccine took the form of open scepticism. D r A. J. Orenstein, pointed out that the pneumonia mortality rates for the Crown mines, where inoculations had been widely used, were no different from those on the Durban Deep mine, where no workers had been i n ~ c u l a t e d . ~ ~ There thus appears to have been a disparity between the cautious view of the medical community, with the South African Institute for Medical research being fairly optimistic and Orenstein being openly dubious, and the unbridled optimism of Wallers and the Chamber officials. T h e Chamber in its quest for Central African labor chose to ignore the warning signals from the professionals and push ahead full speed under the banner of what they presented as Lister's medical miracle. Were the Chamber officials simply being self-serving and duplicitous in their representation of medical facts about the effectiveness of Lister's

" C M A 5 8 , Sative Labour Contingent, 1916-18, Employment of Tropical Natives, Percival TVatkins to A. I. Girwood, 6 July 1917. " O CSIA S 8 , Sative Labour Contingent, 1916-18, E. A. TVallers to Prime Minister Louis Botha, 16 Nov. 1917. '' Annual Report, Sanitation Department, Rand Mines Ltd, 1918.

286

R A N D A L L M. P A C K A R D

vaccine? That conclusion is hard to avoid. Yet it must also be recognized that the people they had placed in charge of the development of the vaccine were nearly as optimistic. Lister himself was clearly convinced that he had developed a successful vaccine. Moreover, Watkins-Pitchford, the director of the Institute, had reason to accept Lister's conclusion, or at least not to question it too rigorously. T h e Institute, after all, was totally dependent on Chamber funding. In addition, its first and most important assignment had been to develop a usable anti-pneumococcal vaccine. He may well have seen, perhaps correctly, that continued support for the Institute, or at least for his directorship of it, was dependent on making, or appearing to make, progress in this area. One must also acknowledge the widespread success of other vaccines in combatting infectious diseases at this time. T h e western bio-medical community both within and outside of South Africa were increasingly confident the n other hand, in the ability of autogenous vaccines to combat d i ~ e a s e . ~ W debates over the ability of vaccines to combat plague at the turn of the century and influenza during the great influenza pandemic of I 9 I 8 suggest that within South Africa vaccines were not universally seen as a viable response to disease.53 Despite their representations, the Chamber was unsuccessful in convincing the government to permit the requested experimental entry of 2,000 Central African workers. This refusal was partly due to difficulties between the British government and the Portuguese over the recruitment of workers from northern Mozambique.j4 It was also because the Chamber failed to convince the Minister of Native Affairs and the British government of either the need for or the viability of tropical recruitment. In the period following World War I , the Chamber continued its quest for tropical labor, making it a central issue in its representations to the Low Grade Ore Commission of I 9 I 9-20. Yet the intensity of its efforts diminished during most of the 1920s as the mining industry's ability to attract African labor from other sources increased. T h e negative impact of the I 9 I 3 Land Act on the ability of African farmers to resist the calls of mine recruiters, especially in the eastern Cape, combined with the Chamber's success in coordinating and controlling recruiting activities after 1920, provided the mines with a more reliable, if not always abundant supply of African labor. Predictably, interest in the development of a pneumonia vaccine waned between 1920 and 1927. Lister continued to work on his vaccine and was promoted to the position of director of the Institute for Medical Research. At the same time, a number of mine medical officers ran tests to evaluate the vaccine's actual effectiveness. T h e results of these tests were discussed from time to time within the Transvaal Mine Medical Officers Association, which was established in 1921. A review of these discussions indicates that there was little if any consensus among association members concerning the effectiveness of the vaccine. Nor does one get a sense that there was a great deal of urgency attached to resolving the problem.55 j2 H. J . Parish, Victory with Vaccines - T h e Story of Immunizations (Edinburgh and j3 Phillips, Black October, I 13-18. London, 1968). j4 C M A N8, Native Labour Contingent, 1916-18, Secretary for Native Affairs to Secretary, Transvaal Chamber of Mines, 8 April 1918. j5 Proceedings of the Transvaal M i n e Medical Oficers Association, I 92 1-7.

Beginning in 1927, however, the mines were faced with a new labor crisis, as the Portuguese government reduced the quotas it imposed on the number of Mozambican workers going to the mines. Between 1927 and 1929 the annual number of Mozambique workers on the mines decreased by roughly I 5,000 men. While the mines were able to make up some of this deficit from other sources, their overall African work force dropped by nearly ~ o , o o o workers.56 This prompted the Chamber again to step up its efforts to gain access to Central African labor. W N L A sent a senior representative to tour various Central African locations and report back on the availability of labor. H e returned with glowing reports of large reservoirs of available healthy labor to the north. T h e Chamber, accordingly, began pressing for the reopening of Central African recruitment." T h e Chamber's renewed interest in tropical recruitment also energized efforts to promote Lister's anti-pneumococcal vaccine. In September 1927 the Executive of the Association issued a questionnaire on pneumonia and efforts to control it to all gold and coal mines. T h e Association also formed a sub-committee to evaluate the responses to this questionnaire and submit a report to the Gold Producers Committee of the Chamber of Mines.58T h e timing of the formation of a committee to evaluate in a systematic way the pneumonia position on the mines and the submission of a report to the Gold Producers Committee was not coincidental. It represented, instead, an example of how increased labor requirements of the mining industry intensified the industry's interest in pneumonia research. T h e Transvaal Mine Medical Association Report to the Gold Producing Committee, submitted in April of 1928, hardly gave that Committee the ammunition it needed to make their case for the reopening of Central African recruitment. T h e report was inconclusive concerning the effectiveness of vaccinations. \;lrhile some medical officers attested to the vaccine's value, statistics from the Rand Mines Ltd, which had the most developed medical services and kept the best health records within the industry, indicated that despite the use of inoculation pneumonia mortality increased between I 920 and 1927. T h e report recommended that more research be carried out. Despite this weak medical case, the Chamber once again presented its argument for the reintroduction of Central African labor but were again unsuc~essful.~~ THE DEPRESSION AND THE REOPENING OF 'TROPICAL' RECRUITMENT

By the time that the Low Grade Ore Commission met in 1931, the mines were no longer experiencing a labor shortage. T h e Depression had produced a glut of African labor for the mines and ended the temporary shortage produced by the Portuguese government quotas. Nonetheless, the Chamber chose to push for the reintroduction of Central African workers. There were several reasons for this decision. First, the Chamber officials recognized that j6 David Yudelman and Alan Jeeves, ' N e w labour frontiers for old: black migrants to the South African gold mines, 1920-1985', J. Southern A f r . Studies, X I I I (1986), 124. 5' Yudelman and Jeeves, ' N e w labour frontiers for old', 108. ' V a r t w r i g h t , Doctors on the Mines, I I I . j9 Cartwright, Doctors on the Mines, I 1 2 .

288

R A N D A L L M. P A C K A R D

the depressed economic conditions would not last and that their future needs would require expansion of their recruiting activities to the north. Secondly, they realized that the Depression had created widespread unemployment that made the government more susceptible to requests made by the mine owners. T h e threatened closure of low grade mines and the loss of white jobs within an economy in which white unemployment was already high clearly carried more weight than it had in the past. Moreover, the importance of the mining industry to the economy, given its counter-cyclical nature, could not be overlooked. Finally, the Depression, by allowing the mines to be more selective in their recruitment of African labor, had greatly reduced the incidence of a number of diseases and particularly tuberculosis. Whether or not the mines were aware of the demographic reasons for this improvement, they could and did use the improved health of their work force to push their case for the reintroduction of 'tropical' labor on a trial basis.60In short, the Chamber officials saw the early thirties as providing a window of opportunity for regaining access to Central African labor and accordingly pushed their case. I n doing so, they once again employed the promise of an effective pneumonia vaccine, supported by medical evidence that was at best no more than questionable. D r Lister, now Sir Spencer Lister, gave testimony to the Commission and used the opportunity to stress the importance of his own research and its value for the health of 'tropical' workers on the mines. Lister, moreover, minimized the role of improved hygiene and thus the arguments of those, like Orenstein, who advocated environmental reform as a solution to the pneumonia problem : General hygiene methods will play their part towards this desiratum [lowering tropical mortality] but to a limited extent, both as regards time and degree. T h e only specific way of raising resistance to the pneumococcus is by employing the principle of active or passive immunization and active immunization is here the method of choice and in effect means inoculation with a suitable vaccine, a procedure I strongly advise should be carried out on all Mine Native Labourers. T h e validity of this procedure is founded upon - ( I ) laboratory experimentation, ( 2 ) special animal experimentation, (3) practical results 'in the field' statistically. T h e mass of evidence on this subject is too great to be discussed in this ~ t a t e m e n t . ~ '

While Lister acknowledged the problem created by the existence of a variety of existing pneumococcal bacteria, he assured the committee that this obstacle could be overcome. Lister's testimony, in effect, not only supported the mines' use of Central African labor but did so by defining the problem of 'tropical' mortality in terms of immunization, vaccines and laboratory experimentation. This was the authoritative voice of medical science spoken by a distinguished knight of the realm. T h e medical basis for Lister's arguments was, however, questionable. T h e mass of evidence that Lister could not provide to the committee, if anything, contradicted his conclusion. Writing in the 1930 Annual Sanitation Report of the Central Mining/Rand Mines group, D r Orenstein observed that during 1930 African mine workers on the Crown Mines, New Modder and

" "

See Packard, White Plague, Black Labor, 185-93, for a more detailed discussion of the effects of the Depression on labor supplies and workers' health on the gold mines. CCXIA, LOTVGrade Ore Commission 1930-1, Exhibits 112-86, Statement by Sir Spencer Lister, Director of the South African Institute for Medical Research, Dec. 1930.

Kew Modder B were not inoculated, inoculations having stopped on I August 1929. H e went on to state, ' I t will be observed that the group of mines in which inoculation was stopped for 1930 gives better results in respect to incidence rate than the group of inoculated mines'. H e concluded, 'anti-pneumonia inoculation plays no significant role in diminishing the incidence of pneumonia, or mortality consequent upon attack'." Although Orenstein, as noted above, was a long-time opponent to the use of silver bullet approaches to health problems and an advocate of environmental reform, he was not alone in his opposition to inoculations. Several other medical officers reported similar findings. At a March 1931 meeting of the Transvaal Mine Medical Officers' Association, Drs Dauberton and Donaldson reported an actual decrease in pneumonia mortality following the cessation of inoculation^.^^ A statistical analysis of pneumonia cases occurring on several mines between 1928 and 1937 would later confirm the impression that the vaccine had little effect. T h e analysis showed that between 1928 and 1930, where inoculations were in vogue, the incidence trend was upward. By contrast between 1930 and 1934, when no inoculations were given, the trend was down.64 Despite these facts, the Chamber, with its august medical expert Sir Spencer, continued to argue that vaccinations would insure the safety of 'tropical' labor on the mines.65As Lister noted in his statement to the Low Grade Ore Commission, ' I am of the opinion that making use of all means now at our disposal, that the employment of tropical natives as hline Labourers on the Rand will not lead to a death rate greater than that which has been estimated to occur in their own country'.66 By the end of the ~ g z o sit, was becoming increasingly evident that the only medical professionals within the mining community who believed in the efficacy of Lister's vaccine were Lister and the staff of the Institute. Even Lister recognized that there were problems due to the existence of a wide range of infectious agents. I n this context, it is important to recognize that mine medical officers were becoming increasingly professionalized during the 1920s. FVith the creation of the Transvaal Mine Medical Officers' Association, physicians working on the mines began to see themselves as physicians first and mine employees second. T h u s , while they remained cognizant of their financial dependence on the Mines, they became increasingly vocal in their professional concerns about worker health and safety.

"

Barlo~vRand Archives, Annual Sanitation Report of the Central hIining/Rand hIines G r o u p , 1930, 3 . Cartwright, Doctors on the Ll/lines, I 1 3 . " Cart~vright,Doctors on the Jllines, 121-2. " Lister's status as a Knight of the Realm certainly strengthened his ability to influence discussions of tropical mortality. One ~ v o u l dlike to know more about the role of Chamber officials and mine owners in Lister's acquisition of this honor. In this context, it should be noted that the Chamber employed Sir Lyle Cummins to serve as technical advisor to the Tuberculosis Research Committee in 1926. Cummins' opinions supporting the idea that African Tvorkers were physiologically susceptible to T B coincided with the economic interests of the mining industry. His presence insured that the Committee's report reflected those interests. See Packard, White Plague, Black Labor, 206-7. CCMA, Low Grade Ore Commission 1930-1, Exhibits I 12-86, Statement by Sir Spencer Lister, Director of the South African Institute for LlIedical Research, Dec. 1930.

"

"

290 R A N D A L L M. P A C K A R D Yet the professionalization of medical care on the mines was a two-edged sword. For while mine doctors viewed themselves as professional physicians, they also saw themselves as men of science. Their objections and concerns concerning Central African mortality were increasingly articulated in the same language of medical research and investigation that had been fostered by the South African Institute for Medical Research. Discussions concerning the health of tropical labor focused narrowly on the question of vaccine efficacy. Although Orenstein stressed the role of food and sanitation at a 1922 meeting of the Mine Medical Officers' Association and D r Daubarton again pointed to the need for better sanitation and separate bunks at a meeting in 1927, mine medical officers were for the most part drawn into the realm of vaccine research. Experiments using different vaccines were conducted on individual mines and the results presented and debated at Association meetings. By contrast, the effects of living and working conditions on the mines seldom entered their discussions. Thus while mine medical officers and the professional researchers of the Institute often disagreed about the efficacy of inoculations, they increasingly debated within a common idiom of medical research. This common usage reinforced the tendency to view the problem of tropical labor mortality in narrow bio-medical terms. T h e Chamber's reading of the political environment operating within the national government favoring access to Central African laborers in the early 1930s was accurate. T h e government put up only token resistance to the Chamber's request to reopen recruiting in the north. T h e minister of health tried to get the mines to commit to the creation of a centralized health service in return for allowing the reintroduction of Central African labor. T h e Chamber had long been opposed to this idea, because it threatened to restrict the ability of individual mine managers to manage their labor costs. They accordingly rejected this quid pro quo.67 Nonetheless, the government eventually gave in to the mines' request, granting permission in 1933 for the introduction of 2,000 Central African workers on a trial basis.68

" T h e case for a centralized medical system along the lines of that established by D r Orenstein within the Rand Mines Tvas put forth by D r E. N. Thornton, Acting Secretary for Public Health, in ~vrittentestimony to the Low Grade Ore Commission in 1931. (CRIA, 22/1931, L O WGrade Ores/Tropical Natives [Central Health Administration], E. N. Thornton to T h e Chairman, Low Grade Ores Commission, 19 Jan. 193 I .) Chamber officials pointed to higher disease rates on Rand mines as an indication of the failure of the system Orenstein had constructed. Thornton argued that the higher rates reflected better record keeping and case detection. T h e willingness of the Chamber and the Institute for Medical Research to risk the lives of Central African ~vorkersto prove the effectiveness of a vaccine that was of questionable value needs to be examined more closely. I t is easy to explain the mine owners' support for this action in terms of their overall economic interests and their desire to believe in the efficacy of Lister's vaccine. These factors may also explain the attitude of the Institute's medical researchers. Yet the willingness of medical researchers to experiment on Central African workers also encouraged the dehumanizing practices that Tvere an every-day part of mine medical culture. T h e rapid examination of long lines of naked men, the use of finger prints and numbered metal discs instead of names to identify individual African ~vorkersand the recurrent association of physical and medical traits with particular 'tribal' groups all worked to efface a ~vorker'sindividual identity and humanity and encouraged the use of Africans as research subjects.

"

T h e experiment with 'tropical' labor was a relative success.69While the mortality rates for Central African workers were considerably higher than those for other workers prior to 1939,they were much lower than before the ban had been instituted. There were two reasons for this decline. First working and living conditions had improved to some degree on the mines since 1913.Secondly, and more important, the Chamber insured that mine managers treated the experimental Central African workers with kid gloves. Once the experiment demonstrated lower rates of mortality the government permitted the mines to increase the number of Central African recruits steadily over the next decade. During this period, the precautions taken during the experimental phase were no longer enforced. In addition, the mines preferred to employ 'tropical workers' in the hottest and most unhealthy areas of the mines underground on the assumption that they were better able to withstand the effects of heat, a practice that reflected their own construction of the 'tropical worker '. Predictably, the incidence of pneumonia among Central African workers rose, increasing from 5o/rooo to 12511000,or by I 50 per cent, between 1939 and 1946.This increase was not reflected, however, in an increase in pneumonia mortality, thanks to the introduction of sulfonamide therapy in 1939. T h e use of sulfonamide, developed outside South Africa, received its first large-scale human test among African mine workers on the Rand. Its success in curing workers with pneumonia meant that disease ceased to be a killer of Central African labor and a liability for the mining industry. T h e introduction of this new and effective medical silver bullet brought to an end the mines' search for a solution to the main medical obstacle to the recruitment of Central African labor. With their access to Central African labor assured, at least for the foreseeable future, pneumonia ceased to be a subject for serious research on the mines, despite its continued existence as a major source of morbidity and suffering. There were only three meetings of the Mine h/Iedical Officers' Association dedicated to discussions of pneumonia between 1939 and 1966.T h e health costs of having invented the physically vulnerable 'tropical worker', rather than investing in improvements in working and living conditions, had for the moment been suppressed by the availability of increasingly effective medical t e ~ h n o l o ~ y . ' ~

I n 1903 the S o u t h African mining industry began recruiting African labor from Central Africa in order to shore u p their labor supplies. F r o m the outset, Central African recruitment was problematic, for Central African mine workers died at very high rates. T h e primary source of Central African mortality was pneumonia. I n response to this high mortality the Union government threatened to close down Central African recruitment, a threat which they carried out in 1913. F r o m 191 I to I 933, the mining industry fought to maintain, and then after I 9 I 3 to regain access to Central African labor. Of central importance in this struggle were efforts

"

See Packard, W h i t e Plague, Black Labor, 230-1, for a more detailed description of this episode. ' O These costs would re-emerge, however, in the 1970s and 1g8os, following the withdra~valof tropical labor and the move toward labor stabilization. These costs were revealed in the sharp rise in T B rates on the mines. See Packard, W h i t e Plague, Black Labor, 309-17.

292

R A N D A L L M. P A C K A R D

to develop a vaccine against pneumonia. While the mine medical community failed to produce an effective vaccine against pneumonia, the Chamber of Mines successfully employed the promise of a vaccine eventually to regain access to Central African Labor in 1934. T h e mines achieved this goal by controlling the terrain of discourse on the health of Central African workers, directing attention away from the unhealthy conditions of mine labor and toward the imagined cultural and biological peculiarities of these workers. In doing so the mines constructed a new social category, 'tropical workers ' or ' tropicals '. T h e paper explores the political, economic and intellectual environment within which this cultural construction was created and employed.