Fire, Smoke, and Expertise in South Africa’s Grasslands

Fire, Smoke, and Expertise in South Africa’s Grasslands Abstract Fire as a management practice in South Africa’s grasslands inflamed heated debate throughout the twentieth century. Imported ecological ideas meshed with homegrown sectoral land management traditions to reinforce a powerful anti-burning narrative among experts. Farmers, however, developed their own theories on burning, and the history of fire research, policy, and management reveals a series of entanglements between ecological theory, management policies and recommendations, and practice that complicate narratives grounded in historiographical traditions focused on critiquing settler and colonial expertise. This article recommends three distinctions to make when thinking about the history of scientific expertise: first, between an individual’s abstract theorizing and his or her “thinking in the field”; second, between the influence of accepted scientific findings and the thinking guiding official policies of land management sectors; and third, between official policies and actual land management practices. This article provides overviews of long-term fire use in the country’s grasslands, the ecology of grasslands and fire in South Africa, of early debates over fire, and scientific fire research and management in the country. INTRODUCTION Cambridge-educated Scottish botanist John Bews introduced American ideas about vegetation, succession, and fire into South Africa in the 1910s, thereby providing a theoretical framework for an anti-burning narrative developed by local experts. Bews and several South African researchers including John Phillips, James (Hamish) Scott, and Winston Trollope tested and modified these ideas over the next half century in local grasslands. While Phillips in particular had concluded that fire was a natural and not necessarily destructive ecological agent in grasslands by the 1930s, foresters and agricultural officials—influenced by colonial traditions of forestry and botany—opposed burning. They linked it to vegetation degradation, soil erosion, and desiccation. The idea that anthropogenic burning had converted large areas of forest to grasslands or savanna, and that many of the country’s grasslands were human artifacts and did not constitute a so-called natural biome, persisted into the 1980s. From the 1920s, this anti-burning narrative was reinforced by concern over desiccation during extended dry periods, and later by fears that a natural disaster on the scale of the American dust bowl might occur in South Africa, a fear shared by British colonial experts in other parts of Africa. Farmers hotly debated the pros and cons of burning, and most continued to burn despite expert disapproval. Experimental studies began in the 1920s and included testing different approaches to burning as a grazing management practice. By the late 1930s, some of these showed that in some grasslands burning was an acceptable management practice, but it took another forty years for this to be officially adopted as policy rather than a sometimes “necessary evil.” A study of the evolution of grassland science, policy, and management in South Africa over the twentieth century reveals the complex interactions of scientific theories, experimental investigations, land management traditions, sectoral goals, and the interactions of land management bureaucracies, experts, and the farmers they were meant to be influencing. Influential experts worked in several contexts, as theoreticians, policy and management advisers, and researchers. Narrating this history requires engaging with the contradictions and accommodations these experts made as they moved between these operational modes and contexts and provides an opportunity to reflect on how historians write about experts and expertise in general. The history of the importation and misapplication of Western scientific ideas in colonial territories figures prominently in environmental historiography. Many of these narratives focus on foresters, commercial farmers, and agricultural experts and criticize their imposition of inappropriate land management techniques or their uncritical adoption in the Southern Hemisphere of ideas developed in the Northern Hemisphere. These studies frame science and expertise primarily in terms of relations of power between colonial authorities and the oppressed and have their roots in critiques of modernist and nationalist narratives emerging from African studies and South Asian subaltern studies in the 1970s and 1980s.1 As Jane Carruthers has shown, the environmental history of South Africa follows this pattern, with roots in a radical Marxist history that emerged in the 1970s, using class relations as a lens to reveal the subjugation and dispossession of Africans. As the country approached the transition to democracy in the early 1990s, the analytical emphasis shifted toward ecosocial themes, notably studies of how colonial and apartheid-era interferences in, and regulations of, space and natural resource use resulted in inter- and intraracial social conflicts and disruption.2 Both historiographical strands, however, contributed to a focus on environmental justice, largely presenting experts and expertise as the instruments of oppression. William Beinart has since challenged the assumptions underlying such portrayals. In a 2009 article with Karen Brown and Daniel Gilfoyle, he sought to broaden debates over colonial (and postcolonial) science and expertise by moving beyond narratives of the short-sightedness of experts mobilized by authoritarian states, instead encouraging an interdisciplinary effort to understand the science and document the production of scientific knowledge and ideas.3 This is a much more useful approach for understanding historical interactions between scientific research, expertise, policy, and management and the many other social and ecological factors that shape how land is actually managed. Reappraisals of the complicated chronologies that have shaped present-day land use can provide a useful context for developing future environmental policies, even when they do not aim to extract “lessons learned.”4 Models for such reappraisals preceded Beinart’s call. Richard Grove already advocated in 1995 for a reconceptualization of environmental expertise, introducing the notion of “green imperialism” and questioning the centrality of the metropole in the development of global environmentalism and environmental science.5 Since 2003, Beinart has reappraised the careers of South African veterinarians, agricultural officials, and modernizing farmers, focusing on the positive contributions they made to land and livestock management, based on long careers spent developing expertise grounded in local knowledge.6 Later work by Bennett, Kruger, and Pooley has further explored the sophistication and diversity—and also limitations—of the intellectual engagements of scientific foresters, botanists, and agricultural experts in the region.7 Collectively, this research reveals the ways in which scientific theory, policy, and management were imbricated with social, cultural, environmental, and economic factors. It does so by focusing on the activities, influence, and circulation of individual experts in specific contexts and within polycentric networks. This more nuanced approach chimes with developments in the history of science characteristic of the “new imperial history.”8 One way to capture this complexity is to think about three distinctions when reconstructing histories of scientific expertise. First, we must not assume that an individual’s abstract theorizing (as expressed in academic journals) necessarily chimes with what might be termed their “thinking in the field.” Second, we should not confuse or overstate the influence of accepted or published contemporary scientific findings with the thinking that guided official policies of land management sectors. Third, we ought to distinguish between official policies and actual land management practices. This article explores these distinctions in the context of the history of grasslands fire management in South Africa in the twentieth century. It reconsiders the contributions of some key figures in the history of grassland research. Of these, only Phillips is now relatively well known in the historiography, figuring prominently in books by Anker, Bennett and Kruger, and Pooley. Phillips played an important role in the history of ecological thinking in South Africa and has been discussed in terms of both his commitments to ecological science and also to what Saul Dubow has called the South Africanization of science. While a history of ideas approach is valuable here, it also has its limitations when writing about researchers who got their boots dirty in the field and routinely considered both policy and practice in their work.9 This article is also a contribution to the historical literature on fire in Africa, which—Stephen Pyne’s summary comments on fire in Africa and the growing literature on fire in the southwestern Cape’s fynbos biome, notwithstanding—is woefully small considering the scale of burning that occurs across Africa’s grasslands and its rich history of fire management by indigenous people and settlers.10 GRASSLANDS AND FIRE Pollen data indicate that grasslands have been present in South Africa throughout the Holocene, occupying roughly the same region as they do today, and they were even more extensive during the Pleistocene.11 The grassland biome (figure 1) falls mainly east of 25°E and between 25°S and 33°S in South Africa. With the exception of the high Drakensberg Mountains, it is dominated by C4 grass species (more efficient at higher temperatures and light intensities than C3 species12), with the characteristic species being Themeda triandra, which is primarily adapted to defoliation by fire.13 The region’s grasslands may be divided into semiarid regions, with 550 mm mean annual precipitation (MAP), mesic (735 mm MAP), and montane (1380 mm MAP). This biome is very rich in species diversity at 1000 m2 scale, but research has previously focused on grass species valuable for livestock production, rather than the plants that constitute this species diversity.14 Farmers and agriculturalists have customarily divided the grasslands into “sweet” (dry) and “sour” (moist) grasslands. As a rule, sweet grasslands occur in areas with MAP below 650 mm, often at lower elevations, where growing conditions are highly variable, linked to less predictable rainfall. The grasses are more palatable to livestock all year round because they have low fiber content and maintain aboveground nutrient content in winter. The offtake (by grazing, fire, or other means) removes a large percentage of production, except in years of exceptionally high rainfall. Changes in biomass are largely driven by rainfall rather than fire, which is relatively infrequent under natural conditions and has little effect on grass composition. Sweet grasslands are thus climate dependent. By contrast, sour grasslands (with a mean annual rainfall above 650 mm) are often found at higher elevations and along the Indian Ocean coastal belt where growing conditions are seasonally relatively fixed, making the grass unpalatable in the dry dormant season. The offtake removes only a small percentage of production, with a natural fire frequency of one to three years. Fire exclusion results in changes to grass composition and invasion by trees and shrubs. Sour grassveld is thus fire dependent.15 Figure 1. View largeDownload slide Map of South Africa’s grassland biome. Faint lines show country and provincial borders; the dark line outlines the grassland biome. The Dry Highveld grassland bioregion occurs on the interior central plateau of the country, where rainfall is below 600 mm per annum and is known as “sweet” grassland. Rainfall increases west to east across the Highveld, the vegetation transitioning from sweetveld into the sourveld of the Mesic Highveld grassland bioregion. The Drakensberg grassland bioregion occurs in the high Drakensberg Mountains, extending south through Lesotho as far as the Amathole Mountains in the eastern Cape, characterized by steep slopes and high rainfall. On the eastern foothills of the Drakensberg are the subescarpment grasslands, which are moist grasslands, in the mist belt. The letters refer to provinces as follows: A = Mpumalanga (formerly in Transvaal Province), B = Free State (formerly Orange Free State), C = KwaZulu-Natal (formerly Natal Province), D = Eastern Cape, E = Western Cape, and F = Kingdom of Lesotho. Credit: Simon Pooley, redrawn from Muscina and Rutherford, eds., The Vegetation of South Africa, Lesotho and Swaziland, Strelitzia 19 (Pretoria: South African National Biodiversity Institute, 2006), 44–45. Figure 1. View largeDownload slide Map of South Africa’s grassland biome. Faint lines show country and provincial borders; the dark line outlines the grassland biome. The Dry Highveld grassland bioregion occurs on the interior central plateau of the country, where rainfall is below 600 mm per annum and is known as “sweet” grassland. Rainfall increases west to east across the Highveld, the vegetation transitioning from sweetveld into the sourveld of the Mesic Highveld grassland bioregion. The Drakensberg grassland bioregion occurs in the high Drakensberg Mountains, extending south through Lesotho as far as the Amathole Mountains in the eastern Cape, characterized by steep slopes and high rainfall. On the eastern foothills of the Drakensberg are the subescarpment grasslands, which are moist grasslands, in the mist belt. The letters refer to provinces as follows: A = Mpumalanga (formerly in Transvaal Province), B = Free State (formerly Orange Free State), C = KwaZulu-Natal (formerly Natal Province), D = Eastern Cape, E = Western Cape, and F = Kingdom of Lesotho. Credit: Simon Pooley, redrawn from Muscina and Rutherford, eds., The Vegetation of South Africa, Lesotho and Swaziland, Strelitzia 19 (Pretoria: South African National Biodiversity Institute, 2006), 44–45. Hominins have been present in southern Africa for about 3 million years, but the available evidence suggests that although modern human hunter-gatherers used fire for hunting and fire-stick farming for at least 100,000 years, the practice had little impact on vegetation.16 Early Iron Age Bantu agropastoralists moved down the east coast from ca. 1700 BP, along the humid low-lying coastal plain (east of the 400-mm isohyet) into what are now KwaZulu-Natal Province and the Eastern Cape. They used fire for hunting, clearing land for agriculture, and iron smelting, and burned grassland and savanna to create fresh grazing for their livestock. The Highveld in the interior plateau of South Africa (above 1200 m above sea level) and its montane grasslands were settled later, from around 700 years BP, possibly as a result of increasing domestic stock numbers and resource depletion on the moister coastal lowlands. The key impacts during the early Iron Age in the eastern region were clearing and cultivation in the rich soils of valley bottoms, with lasting impacts on vegetation structure and composition, notably patchiness in savanna regions. By the late Iron Age, settlements shifted to the hilltops and the interior treeless grasslands, and regional economies developed where political power and wealth were linked with cattle. This resulted in overgrazing in some (notably northern) regions of what is now South Africa.17 Under colonial rule, livestock farming with sheep, cattle, and goats became the main land use practice in regions too dry for agriculture, resulting in land degradation in some areas. In grassland regions suitable for agriculture, extensive areas were converted to maize and cereal crops, particularly in the former Orange Free State and Transvaal provinces to supply the burgeoning urban centers developing around diamond and gold mines. From the mid-1800s, sugarcane was planted in the coastal lowlands of KwaZulu-Natal on the east coast. Plantations of introduced trees (mostly pines and eucalypts) were planted to supplement the country’s meager timber resources, first by state foresters in the Cape Colony in the 1860s, and then, following the creation of a national forestry department in 1910, in the moist upland grassland regions of what are now KwaZulu-Natal, Mpumalanga, and Northern Provinces.18 Private wattle (Acacia) plantations were established in Natal from the 1860s.19 FIRE, FARMERS, AND EXPERTISE For European settlers in the region, the cultural legacies of the anthropogenic transformations of the grassland and savanna biomes included both extensive use of burning by farmers and colonial experts’ hostility to burning practices, in particular botanists, foresters, and soil conservationists, all of whom regarded burning as a primitive and destructive practice. These experts believed that most bush fires were anthropogenic and resulted in the desiccation of the land, destruction of the indigenous flora in particular forests, favored less palatable grasses and scrub, and resulted in soil erosion.20 Analysis of the agricultural journals published in the Cape, Natal, and Transvaal colonies and territories (1898–1910) of what became the Union of South Africa in 1910, up to 1926, reveals the major debates about veld burning before prolonged scientific studies were set up. There were many rhetorical appeals to the authority of science, and concepts borrowed from the emerging science of ecology were used to justify a variety of opinions. Only after World War I, however, did South African scientists begin to publish some preliminary findings on fire and pasture management derived from experimental research.21 Controversy was fierce among farmers over the costs and benefits of using fire to manage veld, with most farmer correspondents to the journals in favor of controlled burning.22 Department of Agriculture officials and editors of the journals advocated new and improved farming methods to replace the need for veld burning, but they acknowledged that most of these were simply too expensive for the average farmer.23 South African farming mechanized late with tractors widely adopted by white farmers only after World War II.24 A 1932 report on the country’s grasslands acknowledged that research into “artificial pastures” and “grass silage” were in their infancy.25 In the meantime, judicious veld burning was a cheap and effective, if imperfect, method of veld management requiring no special equipment and no sustained use of labor. The main reason given by farmers in the agricultural journals for burning, especially in sourveld, was to get rid of old dry grass, which was useless as fodder, and force up new grass, which was edible and nutritious for livestock.26 They also argued that dry grass was a hazard when wildfires swept through their land. In contrast, the anti-burning camp (mostly experts, with some farmers, often from sweetveld regions) asserted that veld burning led to the “coarsening of veld grasses,” resulting in reduced stock-carrying capacity.27 Furthermore, they argued that destruction of the vegetation and humus layer by fire denuded and hardened the soil, thereby reducing its capacity to both absorb and hold moisture and leaving it prone to erosion.28 They maintained that these evil effects of burning contributed to the general desiccation of the country, negatively impacting climate, and rainfall in particular.29 Expert opposition to fire informed by the rhetoric of science began in the mid-1800s with concern over its desiccating effects. The key figures were the Cape Colony’s official botanists Ludwig Pappe and John Croumbie Brown, both active during periods of prolonged drought in the colony. Key Afrikaner agricultural bureaucrats trained in the United States following the South African (or Boer) War (1899–1902) became influential after union in 1910. They drew on ideas about vegetation and pasture science developed in the prairies of the American Midwest. The period from 1918 throughout the 1920s was very dry in South Africa, and a Drought Investigation Commission was convened, reporting in 1923. They concluded that the ongoing desiccation of the country was the result of bad farming practices, including veld burning, rather than declining rainfall.30 The commission drew on current ideas about successions to climax vegetation, influenced primarily by climate, to argue that “[g]rass-veld which grows so vigorously that it cannot be grazed down by full stocking is naturally tree-veld rather than grass-veld,” and that farms that could only be rendered fit for stock farming by burning should possibly be given over to timber production.31 By transposing arguments about (climate-driven) sweetveld onto (fire-driven) sourveld areas, they obscured the natural dynamics of fire-regulated sourveld areas. While it is true that fire exclusion would result in a shift to scrub and trees, fire exclusion was not “natural” in most sourveld grassland areas. As noted earlier, grasslands have been present in South Africa since the Pleistocene and thus fires must have been present before humans arrived on the scene, and further, colonial records indicate that settler farmers have been burning grass since the seventeenth century and learned the practice from indigenous herders.32 Between 1900 and 1913, farmers and officials frequently speculated that impoverishment of the soils and thus the veld by burning caused animal diseases including lamsiekte (which was actually botulism in phosphorus-deficient cattle from ingesting carrion and bones infected with clostridial toxins) and gallsickness (in reality a tick-borne disease).33 In the Transvaal, it was believed that burning could eradicate growths of “moulds and bacteria … one of which, Bacillus botulinus, may cause deaths amongst cattle.”34 Albert Moggs of the Division of Veterinary Services concluded that dunziekte (emaciation, caused by ingesting plants of the genus Senecio) and stijfziekte (an unidentified viral disease) were associated with burning and overstocking.35 Many African and white farmers saw burning as a valuable way of purifying or sanitizing the veld through controlling “vermin, snakes, ticks, etc., by removing the cover.”36 In 1899 Magistrate F. Y. Gibson of Nongoma District in Zululand advocated not burning grass routinely because grass burning could then be deployed against locusts.37 Fire was used to destroy cichlid moth larvae, believed to poison cattle.38 By the 1930s, most scientific authorities had discounted veld burning as an effective means of eradicating ticks and instead favored dipping. This is more expensive, however, and African farmers in the Maluti/Drakensberg area and in eastern Pondoland still list pest control (especially ticks) as a reason for controlled burning.39 Some correspondents to the agricultural journals argued that burning encouraged weeds and invasive plants, to the detriment of nutritious grasses.40 Several stressed the destructive effects of veld fires on trees, offering suggestions on prevention and firefighting.41 Agricultural experts acknowledged the problem of dealing with weeds with inadequate labor on terrain unsuitable for mowing. It was no use trying to change “the cultural custom of four or five generations” without offering “a sufficient substitute for the [condemned] practices,” wrote Charles Eustace Pillans of the Cape Department of Agriculture. He proposed paddocking combined with systematic weed eradication and cultivation of useful stock-food plants.42 H. J. Choles, an editor for Agricultural Journal of the Union of South Africa, recommended slowly phasing out burning, replacing it with heavy stocking and grazing as a means of veld management.43 Fire historian Stephen J. Pyne has critiqued fire research in general for focusing on the technical study of fire behavior (led by foresters) and neglecting the social dimension.44 There is much truth in this. However, agricultural researchers in South Africa in the early 1900s were well aware of the socioeconomic constraints on good farming practice, and therefore why farmers resorted to veld burning. In 1913 Choles observed, The firing of the veld is a time-honoured custom in South Africa, and dates from the first days of great, poorly stocked farms when, as is still sometimes the case, holdings were beyond the capacity of the limited number of stock they carried to keep the grass short; and as the grass shot up in the late summer months and dried out into coarse, unpalatable herbage, it was necessary either to mow it or burn it off to provide for succulent young shoots for the impoverished stock in the spring. Mowing was quite out of the question over the extensive areas that had to be disposed of; burning was cheap, rapid, and immediately as satisfactory. The ultimate effects were not contemplated in the grim struggle for existence.45In the early twentieth century, however, the promotion of enclosure and a progressive reduction in the size of farms affected established farming methods. The Fencing Act of 1912, amended in 1922, provided white farmers with subsidies for fencing.46 Farmer M. J. Beukes noted that “in the good old days,” when farmers had huge farms, it was acceptable to burn because it was the only way to get rid of “rank and coarse” grass and control ticks and other insects; in more modern and “civilized” times of smaller farms, however, “other methods of farming must be adopted, whether one likes them or not.”47 Farmer C. R. Prance worried that while it was feasible for his contemporaries to send their cattle to bushveld farms (in savanna, at lower elevations) while they waited for the “brand” (green grass after a burn) to come up on their home farm, “the next generation will have to share our farms—and then what?”48 Such fears over the future of farming in South Africa were shared by the government of the newly formed Union of South Africa (1910), led by two Afrikaner farmers and statesmen, Louis Botha and Jan Smuts, the latter of whom was a grass aficionado and keen botanist. Into the 1940s, the government feared the effects of unprofessional land management in the face of the closing of the South African frontier.49 BOTANY AND ECOLOGY Developments in scientific botany influenced agricultural experts’ and drought commissioners’ views on the effects of fire on vegetation. During the first decades of the twentieth century, botany was professionalized in South Africa, and ecological ideas became influential in the discipline, informing ideas about natural resource management more generally. A key figure, John William Bews, traveled from Scotland to take up the first professorship of botany at Natal University College in 1910. Bews’s “An Account of the Chief Types of Vegetation in South Africa, with Notes on the Plant Succession” (1916) was the first survey to apply an overtly ecological approach to South African vegetation and was informed by Frederic Clements’s ideas about vegetation succession.50 Clements argued that vegetation progresses linearly through a series of increasingly complex stages toward a stable climax community that is in equilibrium with the prevailing environmental conditions. Climate drives a linear unidirectional succession toward a single vegetation type, but a variety of disturbances (notably fire) may inhibit or temporarily reverse this progression.51 This theoretical framework had two important implications for thinking on grasslands and fire. First, Bews included in his vegetation classifications “changed” or “false” formations that were not in equilibrium with prevailing climatic conditions and which he believed were created and maintained by overgrazing and human use of fire. This included extensive grassland areas east of the Drakensberg Mountains in what is now KwaZulu-Natal Province. He thought that the destruction of the bush by wildfires and human burning practices was contributing to the desiccation of South Africa, particularly on the Highveld.52 Second, Bews argued that burning sets back the plant succession. He theorized that in “primitive” or semi-open grassland, burning prevented the plant succession from progressing to a more nutritious type and should be banned.53 The idea that anthropogenic burning had converted large forested areas to grassland, and thus that much of the country’s grassland was a secondary and somehow unnatural vegetation type, remained influential in South African scientific thinking until the 1980s when pollen records showed that the grasslands of the moist eastern region were of great antiquity.54 Bews developed his understanding of both the local flora in Natal and the region’s agricultural needs. Rather than adopt a purely anti-burning stance, he made an exception for situations where the succession appeared to be progressing toward scrub or forest. Where such vegetation encroached on grassland, Bews encouraged veld burning, should grazing be the desired land use. Bews rated investigations into the plant succession and principles underlying it to be “one of the most important economic results of our study of ecology in Natal” because based on this knowledge accurate guidance on the use of veld burning could be given, tailored to specific conditions and kinds of veld.55 Other major writers on veld burning in this period cited Bews’s interpretation of the South African flora through the ecological concept of succession. From 1927 to 1931, John Frederick Vicars Phillips worked on tsetse eradication in Kondoa Irangi in Tanganyika under Charles Swynnerton, where he observed the use of fire to combat tsetse flies and the practices of indigenous pastoralists and settler farmers. In 1931 he returned to South Africa to begin work as professor of botany at the University of the Witwatersrand in Johannesburg (a post he held until 1948). There he worked with T. D. Hall, agricultural adviser to African Explosives & Chemical Industry (which produced fertilizers), and J. D. (Hamish) Scott to develop Frankenwald, the estate bequeathed by the gold and diamond magnate Alfred Beit. Phillips made Frankenwald a center of ecological research, notably of problems relevant to pasture management and conservation. In the postwar period he trained nearly two hundred returning veterans in soil conservation on his “Donga Doctors” course, and his students dispersed all over the British territories in Africa.56 Phillips had just published the first review of scientific research on fire in southern and eastern Africa. He noted the efforts of “scientific workers” to educate the public “of the manifold evils following in the wake of fire,” pointing out that “little scientific experimentation [had] been brought to bear” upon these “problems.” He believed that both natural lightning fires and anthropogenic fires had swept the regions’ grasslands and savannas since prehistoric times, and he stressed the need to understand local conditions including climate, soil, and vegetation type before deciding on regional fire policies.57 Historian Peder Anker has portrayed Phillips as an ideologue, particularly as he perceives a link between Phillips’s championing of the ecosystem as a holistic concept and the political holism advocated by Phillips’s mentor Jan Smuts.58 However, this alleged political idealism should not overly color portrayals of Phillips as a scientist. Influenced by Clements, with whom he corresponded, Phillips also stressed the importance of contingent factors including the successional stage of the plant community and the nature of the season in terms of moisture and temperature. He was in fact wary of trying to generalize from basic scientific principles or from localized findings in an unsuitable way. He wrote cautiously, “I feel it desirable to urge the need of our carefully considering all regional circumstances in the light of scientific experience, before we definitely decry the practice of firing. Possibly I may be criticised for this statement, but in making it I consider I have but done my duty” [emphasis in original].59 As noted, the government was concerned over the future of farming in South Africa, and the Great Depression exacerbated this as the economy suffered a dip in mining returns and economists worried over the small internal market for manufacturing. The government looked to farming as a promising avenue for improvement through modernization. Scientific research and management was deemed required to save the land and improve agricultural output, and pasture research was institutionalized. The first pasture research stations were opened in 1934.60 The University of Pretoria and the African Explosives and Chemical Industry convened a Grassland Research Committee that issued a report in 1932. They argued that overgrazing and burning were resulting in the impoverishment of the naturally nutritious grasslands characterized by grasses like Themeda triandra (“rooigras”). Burning also destroyed the humus, and the soil became less able to absorb water, resulting in increased runoff during heavy rains and consequent soil erosion.61 By the mid-1930s, veld-burning experiments, notably those at Potchefstroom in Transvaal Province, were suggesting that annual burning in the Highveld—in sweetveld and semiarid regions—created good grazing in the short term. However, over the long term, burning resulted in reduced grass cover and soil erosion.62 In contrast, experiments at Cedara Agricultural College in the Natal Midlands showed that in high rainfall areas, in sourveld, burning was an acceptable management practice.63 Despite the positive findings for burning in certain conditions and regions, in 1936 Phillips published a major article on fire, “Fire in Vegetation: A Bad Master, a Good Servant, and a National Problem.” His choice of title reflected his assertion that “the man in the street has been instructed in print and by lecture that firing of vegetation is always highly destructive, and is to be considered as an act of vandalism against the national heritage.” Phillips agreed that burning in the wrong season could allow in undesirable weeds and encourage inferior grasses, but he argued that burning at the right time could actually reduce undesirable grasses and encourage more nutritious ones. Not burning at all could lead to a deterioration of the veld and scrub encroachment. Thus burning could be good farming practice in grasslands or savannas. Phillips was at pains to point out that “the management of grazing after fire is so important that investigation of influences of firing, not taking cognisance of this, and not providing for the control of this process, must largely prove abortive.”64 This was seemingly confirmed by experiments set up following severe droughts in 1932 and 1933 that drastically affected water supplies in the Natal midlands. It was popularly believed at the time that veld burning in the mountains was the cause of the water shortage, and that if the Drakensberg Mountain Range was protected from fire, this would solve the problem. In 1936 land for two research stations was acquired, one traversing savanna, thornveld, and grassland on commonage leased from the Borough of Estcourt, and the other in highland sourveld at Tabamhlope. Two experiments were set up in tall grassveld, the first to measure the effects of burning at different times, excluding grazing, the second to test the optimum burning treatments, with grazing. The main conclusions drawn from the first experiment, as interpreted by Hamish Scott in his 1948 doctoral thesis, were that total protection from fire led to a deterioration of the grass sward, increased runoff, and increased risk of accidental fires that would damage the grass cover and soil. Burning in the wrong season could be harmful, however, and the least harmful time to burn was in spring, after the first good rains.65 The second experiment somewhat modified the earlier findings. It showed that the grass sward would decline in quality if old vegetation was not removed. However, the average weekly increase in live weight of grazing animals was higher on mowed than on burned veld. This was a more critical view of burning, but the management implications were that, where mowing was not possible, burning was necessary to remove old ungrazed grass. Rather than specify a rigid “correct” interval between annual spring burns, Scott advocated judging how often to burn by observing the accumulation of dead grass.66 PRAGMATISM, IDEOLOGY, AND DESICCATION Bews’s and Phillips’s Clementsian theoretical framework of vegetation succession did not blind them to local conditions. Phillips advocated an ecological approach influenced by succession theory but advised against generalizations from basic scientific principles or from localized findings. He and his protégé Hamish Scott recommended experimental studies that took into account climatic conditions including seasonal fluctuations, soils, the state of the vegetation, and grazing and farming practices. They did not regard fire as an unnatural disturbance to African grassland and savanna ecosystems, but did criticize what they regarded as harmful burning practices. Although experimental studies at Cedara, Estcourt, and Tabamhlope showed the positive effects of properly applied veld burning for sourveld pastures, many officials remained opposed to the practice. Advice published in the Department of Agriculture’s journal Farming in South Africa grudgingly acknowledged the potential advantages of appropriate veld burning but presented the practice as a “necessary evil.”67 In 1932 Illtyd Buller Pole Evans, chief of the Union’s Department of Botany and founder of the Pasture Research and Veld Management Section in the Division of Plant Industry, had advocated the adoption of the “ingenious system, known as Deferred Grazing” employed in the United States, citing Arthur Sampson’s 1923 book Range and Pasture Management.68 In South Africa, this system developed into ever more sophisticated methods of rotational grazing, designed to make veld burning unnecessary, despite a lack of consensus over the superiority of well-managed rotational grazing versus well-managed continuous grazing.69 Soil conservation officials in particular resisted the findings of early experimental studies on fire that suggested it could be a good management practice. As David Anderson has shown, the American dust bowl experience, along with the Depression of the early 1930s, raised concerns over the environmental impacts of rapid increases in human and livestock populations in African Reserves in the context of the apparently more frequent droughts afflicting east Africa. These factors meant that soil conservation became a central concern of the British colonial authorities and spurred them to intervene directly in African farming practices. They drew on both South African and American recommendations and expertise.70 Colonial experts attributed desiccation and soil degradation to veld burning in this period. Foresters were particularly opposed to fire and veld burning, and in the 1930s this was the subject of a bitter dispute between agricultural and forestry experts in South Africa.71 At the fourth British Empire Forestry Conference, convened in South Africa in 1935, agriculturalist R. W. Thornton asserted that “burning on a gigantic scale … throughout the grass veld and savannah areas of this continent … has denuded the country of vegetation.” He argued that such burning removed trees that could restrict winds and thus contributed to wind erosion and high evaporation. “This is not only a South African question: it is an international African question,” he maintained.72 An expert narrative linking the drying up of the country to veld burning thus developed in South Africa during the 1920s and 1930s, exacerbated by droughts. The Drought Investigation Commission report of 1923 warned that failure to take action would result in “‘The Great South African Desert’ uninhabitable by Man.” In a 1924 symposium focused exclusively on veld burning, prominent botanists criticized the practice.73 In 1932–33, at a time when the American dust bowl was much in the news, there was a severe drought in Natal Province, and in 1934 Parliament passed a resolution to investigate the effects of veld burning on water supplies and soil erosion in mountain catchment areas. This was debated in the parliamentary capital, Cape Town, which experienced very high fire incidence in 1934. As a result, the state began acquiring private properties in mountain catchments, to be managed by the Division of Forestry. Foresters advocated prohibiting grazing in these regions because “wherever grazing is allowed the veld is burned as almost all owners of stock consider that annual burning of the veld is necessary to provide adequate pasturage”; thus they began to impound cattle that farmers grazed in these areas, particularly in times of drought. This considerably inflamed the controversy over burning.74 During World War II, soil conservation officials discouraged all burning. Jan Smuts set up a National Veld Trust in 1943, and Hugh Hammond Bennett, chief of the Soil Conservation Service of the US Department of Agriculture, toured the Union of South Africa in August 1944, noting widespread soil erosion and condemning “indiscriminate veld burning.”75 Grassland scientists worked within a highly charged political and policy environment that linked poor land management practices to treason.76 In the postwar period, forestry and agricultural researchers gradually came to believe that burning could be safely applied. However, they deemed this permissible only under stringent restrictions, mainly relating to the presumed least ecologically harmful time of year to burn. The Soil Conservation Act of 1946 restricted burning on private mountain catchment land by creating Fire Protection Districts, managed by Fire Protection Committees. An interdepartmental committee appointed by the Soil Conservation Board surveyed the country’s catchments from December 1952. Their report, published as the Ross Commission, carefully considered arguments for and against veld burning. They concluded that, while burning was a major factor contributing to catchment degradation, and should be forbidden in drier regions with vulnerable soils, in some humid catchments it was an acceptable farming practice.77 Following high fire incidence in the Cape mountain catchments in 1961, and a huge fire in July 1962, the Department of Forestry decided to investigate controlled burning in earnest. Fire prevention was simply not possible in these upland grasslands. The department also faced rising criticism in this period from farmers who alleged that its plantations (expanding eastward into the country’s inflammable grasslands) were drying up water supplies. An interdepartmental inquiry was convened, reporting in 1968. The commissioners argued that protecting natural vegetation from fire in fact reduced stream flow. This was because fires lower average veld age, reducing water use and evapotranspiration.78 South African experts were not developing their thinking on fire either exclusively in association with the colonial metropoles or American experts. The Southern African Regional Committee for the Conservation and Utilisation of the Soil (SARCCUS) was born out of postwar European colonial efforts to coordinate technical cooperation in sub-Saharan Africa, notably an inter-African conference at Goma in Belgian Congo in 1948 focused on soil conservation and land utilization. It worked in collaboration with the Commission for Technical Co-operation in Africa South of the Sahara, active from 1950, which was a joint venture of the Belgian, French, Portuguese, and British colonial authorities together with Southern Rhodesia (Zimbabwe) and the Union of South Africa. A 1974 review of the SARCCUS program noted the importance of training and expertise to this enterprise, emphasizing the value of “expert personnel with a lifetime of local experience.”79 The 1948 Goma conference called for the experimental study of fire, lamenting that “running fires annually devastated the whole of tropical Africa, causing a deterioration of the vegetal protecting mantle of the soil and exposing them to all forms of erosion.” Aware that fire was central to many “primitive” African land use practices, however, it was judged pragmatic to “temporarily tolerate” such fires but regulate them on areas of particular value. Further, given the difficulty of combating bush fires, they advocated “the palliative of early fires … at the commencement of the dry season, under the control of qualified services,” but “insisted that these tolerant measures … must not obscure the final goal, most vital for Africa, of the total suppression of bush fires.”80 A decade later, however, SARCCUS experts accepted that the use of fire is necessary in some systems of veld management for grazing animals. The question then is, at what time of year and under what ecological circumstances is it best to burn? The total exclusion of fire or even restriction of burning to defined seasons was deemed “extremely difficult.” Propaganda and education about good burning practices, rather than control through legislation, was advocated. The recommendation for motivating African farmers to practice better fire control was to encourage tree planting and improve their pastures. Experts acknowledged that “in the use of fire there is a conflict between the aims of the forester and the grazier … reflected in legislation and in official recommendations.” In countries where grazing interests were paramount, farmers practiced late burning at the end of the dry period just before spring growth. Where forest growth was important (West Africa and Malawi, for instance), land managers practiced early burning, at the beginning of the dry period when forests were still moist.81 By the 1970s, fire was believed to “[minimize] the evils of selective grazing” in sourveld, prevent “the accumulation of old grass and … [improve] the quality of the herbage and [reduce] the hazard of accidental fires.” In South Africa, sheep farmers in the Orange Free State, Natal, and Transvaal Highveld burned in late summer to stimulate green leafy herbage and in late autumn (this was frowned on) and winter to stimulate growth in early spring (and provide enhanced protein and phosphate content of the herbage). SARCCUS experts acknowledged that “the most desirable time for burning [in sourveld] is still a matter for controversy.”82 The difficulty of delaying burning through the dry months was that it was hard to prevent runaway fires. The summary of advice on burning (1974) noted irritably that “it seems to be impossible to generalise from region to region about burning,” but it concluded hopefully that increases in the price of land and more intensified farming practices would lead to the abandonment of large-scale burning and autumn burning on South Africa’s Highveld.83 THE NATAL SCHOOL When Hamish Scott became professor of pasture science at the University of Natal in 1948, he took the helm of an institution founded to tackle soil erosion and improve veld management. In 1950 Scott set up the now longest running fire experiments in South Africa, at Ukulinga, designed to work out the best approach to burning and grazing the veld (figure 2). At the same time, he worked on rotational farming methods to replace the necessity for burning. He had to combine scientific experiment with the requirements of state and provincial agricultural bureaucracies, who wanted tight restrictions on when fires could be lit, and ultimately (and fruitlessly) to replace fire altogether.84 Figure 2. View largeDownload slide The Ukulinga notice board showing details of treatments, as amended (1956). Credit: Simon Pooley, 2007. Figure 2. View largeDownload slide The Ukulinga notice board showing details of treatments, as amended (1956). Credit: Simon Pooley, 2007. In South Africa, veld burning had become widely accepted as a management strategy in higher rainfall regions by 1970. Administration of the Fire Protection Committees, which had resolutely opposed burning, was transferred to the Department of Forestry in 1970. The department adopted a policy of controlled burning, formalized with the passing of the Mountain Catchment Areas Act 63 of 1970.85 However, tight guidelines were issued, for example burning was only allowed between late July and October in Natal. Burning in August was permitted only after at least 15 mm of rain in twenty-four hours, but for the remainder of the period of legal burning, no rain was required beforehand. Permission from the local Soil Conservation Committee was required for any burning outside of this period.86 In 1977 a South African program on the ecological effects of fire was initiated as part of a wider program of the international Scientific Committee on Problems of the Environment (SCOPE), an initiative of the International Council of Scientific Unions. This South African SCOPE program stimulated both broad reviews and specific studies of key questions about ecological processes and management strategies. The first phase synthesized ongoing work on long-term experiments. For the grassland biome in South Africa, the surviving long-term experiments drawn on were at Ukulinga and in the Drakensberg catchment areas.87 Neil Tainton, who grew up in the Eastern Cape, trained at the University of Natal, and became a lecturer in the Department of Pasture Management and Soil Conservation there in 1959, oversaw the publication of a comparative review of the results of twenty-eight years of burning treatments in grassland at Ukulinga.88 The results were striking. Analysis of plots burned biennially or triennially showed no long-term influences of the preceding burning season. The report concluded, In the long term the effect of veld burning per se is little different irrespective of whether the veld is burnt in the dormant season and whether it is applied annually, biennially or triennially. In practice, however, different burning times and frequencies are associated with different grazing practices and it is this difference in post-burning management which is likely, in practice, to be the main contributory factor to differences in the response of veld to differing burning schedules.89This was revolutionary, upending theories on when to burn, all the shorter term experimental findings, and the persistent prescription to farmers to burn after the first spring rains. Grassland research in South Africa’s KwaZulu-Natal Province during the twentieth century generated a cogent and coherent approach to understanding and managing humid grassland, one that integrated fire as a useful tool. The continuity of ecological and management thinking on fire is remarkable, certainly when it comes to when to burn. Recommendations first recorded in the Natal Colony’s agricultural journal in 1899 are essentially those laid down by the Department of Agriculture in 1999. Ecological principles laid down by Bews and Phillips and related conceptions of rangeland succession still form, with due modification, the basis of research and management thinking. However, considerable progress was made in understanding why different burning treatments affect grass composition and productivity in the ways they do. This knowledge was disseminated throughout South Africa in a series of influential books and the Journal of the Grassland Society of Southern Africa.90 Despite this, Kirkman and Morris could still write in 1999 that “the role of fire and grazing procedure … as management tools are probably the most contentious issues regarding grazing management recommendation in South Africa.”91 Winston Trollope (figure 3), the son of an Eastern Cape farmer, graduated in grassland science after being inspired by Hamish Scott at the University of Natal in the 1970s. He landed the position of senior lecturer in pasture science at the University of Fort Hare (formerly the South African Native College) in 1970. His aim at Fort Hare was to develop farming methods appropriate for African farmers who had few resources: “it’s no good coming along with some weedicide or mechanical techniques; you’ve got to come along with something that doesn’t cost them any money.” Bush encroachment (notably by Acacia karoo) was a serious problem in the region, and in his experiments on grassland he had found that fire was the most practical means of dealing with it—and it was cheap to apply. Figure 3. View largeDownload slide Winston Trollope in one of his burn plots near Fort Beaufort, Eastern Cape. Credit: Simon Pooley, 2007. Figure 3. View largeDownload slide Winston Trollope in one of his burn plots near Fort Beaufort, Eastern Cape. Credit: Simon Pooley, 2007. Perhaps Trollope’s major contribution to South African fire science was his realization of the importance of fire intensity, previously ignored by South African researchers.92 Based for his entire academic career (1970–2005) at the University of Fort Hare in the former Homeland of the Transkei (now Eastern Cape Province), Trollope drew on Australian field-based research on fire intensity, US laboratory studies, his own fire experiments, and African communal burning practices and goat keeping to devise an effective and affordable means of veld management, in particular for controlling bush encroachment. Still active in fire research and management, Trollope concluded that research has proven that “indigenous fire management has demonstrated the advantage of using fire to remove moribund grass material and stimulate the growth of highly nutritious grass forage that should be grazed as soon as possible after the fire … provided the veld is given periodic extended rest periods.” He observed that he and other range scientists had condemned this practice for decades.93 Setting aside theorists, experts, and officials, then, what of the farmers? In 1999 Kevin Kirkman, professor of grassland science at the University of KwaZulu-Natal, admitted that most sourveld farmers had ignored the rotational grazing system for so long advocated by the Department of Agriculture. Part of the reason for this was the tension between the ecologically sensitive flexible management approach pioneered by Scott and policy requirements for fixed dates and rules. However, financial constraints were found to be farmers’ chief consideration when choosing a veld management strategy. By the end of the century, many of South Africa’s large-scale commercial farmers were in financial trouble, and small-scale commercial farmers and communal farmers had always had limited access to capital. Burning is cheap and labor efficient.94 CONCLUSION This history of the use of fire in South Africa’s grasslands reveals a chronology complicated by interactions between a range of intellectual, institutional, and managerial factors and shaped in important ways by a series of individuals who adapted their thinking as they moved between different modes of expertise. They developed their thinking across a range of professional and ecological contexts and in response to sometimes conflicting sectoral traditions and priorities. Attempts to fit them into a Procrustes’ bed of colonial or settler expertise obscures the diversity and range of their professional and personal trajectories and influence. It should be clear, for instance, that neither Bews’s Clementsian account of the structure of vegetation in South Africa nor Phillips’s theoretical publications on the ecosystem concept or his Clementsian influences meant that either man was incapable of adapting his thinking to local ecological circumstances and particular practical land management challenges. Bews acknowledged that different land management goals required different fire management strategies, and Phillips was explicit about the need to be careful when generalizing from basic scientific principles, and to always consider local ecological context including land use. South African farmers and officials had well developed ideas about veld burning before the advent of experimental work. When the first results of scientific research on burning became available, they were both unsuitably generalized from sweetveld to sourveld, and also resolutely ignored by agricultural and soil conservation officials who feared the desiccating effects of burning. Researchers were well aware they were operating in a highly charged political context. They worked in parallel on schemes to eradicate fire and on projects to manage grasslands with fire. The predominant focus in grassland fire research on water supplies, soil conservation, and grazing management for commercial farming precluded research on African fire management and also meant little work was done on grassland biodiversity. This latter omission is part of a global conservation science culture that has ignored tropical grasslands in favor of forests, some assuming that grasslands are a successional stage toward forest, and others that many types of grassland are of anthropogenic origin.95 Whatever the advice and policies of state and provincial officials, in many (probably most) cases, farmers ignored official attempts to restrict and replace burning with other techniques. For many farmers, the alternatives were simply impractical (on stony or steep land) and too expensive or labor intensive. If history is to speak to policy (and it need not), then it seems beneficial to explore the actual entanglements that inform or hinder policy development and implementation. It is not sufficient to study separately the histories of scientific theory, research, sectoral institutions and their policy and management recommendations, or the practices of the land managers they are meant to influence. Simon Pooley is the Lambert Lecturer in Environment (Applied Herpetology) at Birkbeck University of London. He has published a monograph, Burning Table Mountain (2014), on the fire history of South Africa’s Cape Peninsula, and articles on invasive plants and animals. He has taught conservation science at Imperial College London and is a visiting researcher at WildCRU, University of Oxford, where he brings a historical perspective to human wildlife conflict research. He serves on the IUCN’s Crocodile Specialist Group and Task Force on Human Wildlife Conflict. Notes This article was stimulated by a request for a talk for a festschrift for William Beinart, for which my thanks are due to Sandra Swart. I thank William for his guidance during my early explorations of this material. I thank the reviewers and editors for their constructive suggestions. 1. K. B. Showers, Imperial Gulleys: Soil Erosion and Conservation in Lesotho (Athens: Ohio University Press, 2005); M. Leach and R. Mearns, eds., The Lie of the Land: Challenging Received Wisdom of the African Environment (Oxford: James Curry, 1996); J. C. Scott, Seeing Like a State: How Certain Schemes to Improve the Human Condition Have Failed (New Haven: Yale University Press, 1998). 2. J. Carruthers, “Tracking in Game Trails: Looking Afresh at the Politics of Environmental History in South Africa,” Environmental History 11 (2006): 804–29. For the ecosocial trend, see Farieda Kahn, “Rewriting South Africa’s Conservation History—the Role of the Native Farmers Association,” Journal of Southern African Studies 20, no. 4 (1994): 499–516; and South Africa's Environmental History: Cases and Comparisons, ed. Stephen Dovers, Ruth Edgecombe, and Bill Guest (Athens: Ohio University Press, 2003). 3. W. Beinart, K. Brown, and D. Gilfoyle, “Experts and Expertise in Colonial Africa Reconsidered: Science and the Interpenetration of Knowledge,” African Affairs 108/432 (2009): 413–33. 4. P. Alagona, J. Sandlos, and Y. Wiersma, “Past Imperfect: Using Historical Ecology and Baseline Data for Conservation and Restoration Projects in North America,” Environmental Philosophy 9, no. 1 (2012): 49–70; M. M. Steen-Adams, N. Langston, M. D. O. Adams, and D. J. Mladenoff, “Historical Framework to Explain Long-Term Coupled Human and Natural System Feedbacks: Application to a Multiple-Ownership Forest Landscape in the Northern Great Lakes Region, USA,” Ecology and Society 20 (2015), http://dx.doi.org/10.5751/ES-06930-200128. 5. Richard H. Grove, Green Imperialism: Colonial Expansion, Tropical Island Edens and the Origins of Environmentalism, 1600–1860 (Cambridge: Cambridge University Press, 1995). 6. William Beinart, The Rise of Conservation in South Africa: Settlers, Livestock, and the Environment 1770–1950 (Oxford: Oxford University Press, 2003). 7. Brett Bennett and Frederick J. Kruger, Forestry and Water Conservation in South Africa: History, Science and Policy (Canberra: ANU Press, 2015); Brett Bennett, “Margaret Levyns and the Decline of Ecological Liberalism in the Southwest Cape, 1890–1975,” South African Historical Journal 67, no. 1 (2015): 64–84; Simon Pooley, Burning Table Mountain: An Environmental History of Fire on the Cape Peninsula (London: Palgrave Macmillan, 2014); “Recovering the Lost History of Fire in South Africa’s Fynbos, c. 1910–90,” Environmental History 17 (January): 55–83; and “Pressed Flowers: Ideas About Alien and Indigenous Plants at the Cape, c. 1902–45,” Journal of Southern African Studies 36, no. 3 (2010): 599–618. 8. See Joseph Hodge, “Science and Empire: An Overview of the Historical Scholarship,” in Science and Empire: Knowledge and Networks of Science across the British Empire, 1800–1970, ed. Brett Bennett and Joseph Hodge (London: Palgrave Macmillan, 2011), 3–29. 9. Peder Anker, Imperial Ecology: Environmental Order in the British Empire, 1895–1945 (Cambridge: Harvard University Press, 2001); Saul Dubow, A Commonwealth of Knowledge: Science, Sensibility, and White South Africa 1820–2000 (Oxford: Oxford University Press, 2006). 10. See S. J. Pyne et al., “Introduction,” in Wildland Fire Management: Handbook for Sub-Sahara Africa, ed. J. G. Goldammer and C. De Ronde (Freiburg: Global Fire Monitoring Centre, 2004), 1–10; Pooley, Burning Table Mountain. 11. L. Scott, H. M. Anderson, and J. M. Anderson, “Vegetation History,” in Vegetation of Southern Africa, ed. R. M. Cowling, D. M. Richardson, and S. M. Pierce (Cambridge: Cambridge University Press, 1997), 62–84. 12. Perennial grasses can have C3 or C4 pathways for capturing carbon dioxide during photosynthesis. All have C3 pathways, but in the tropics some species evolved the C4 pathway, which makes them better adapted to growing in warm or hot seasonal conditions, specifically through losing less water through transpiration. C3 species can tolerate frost and grow at higher attitudes. 13. T. G. O’Connor and G. J. Bredenkamp, “Grassland,” in Vegetation of Southern Africa, ed. R. M. Cowling et al., 215–57, 215–18, 222, 232. 14. R. G. Uys et al., “The Effect of Different Fire Regimes on Plant Diversity in Southern African Grasslands,” Biological Conservation 118 (2004): 489–99. 15. William J. Bond, “Fire,” in Vegetation of Southern Africa, ed. R. M. Cowling et al., 421–46, 440; L. Mucina, D. B. Hoare, L. Mervyn, et al., “Grassland Biome,” in The Vegetation of South Africa, Lesotho and Swaziland, ed. L. Mucina and M. C. Rutherford (Pretoria: SANBI, 2006), 348–436, 355–56; L. Mucina et al., “Indian Ocean Coastal Belt,” in The Vegetation of South Africa, 569–83, 577–80. 16. Pooley, Burning Table Mountain. 17. M. T. Hoffman, “Human Impacts on Vegetation,” in Vegetation of Southern Africa, ed. R. M. Cowling et al., 507–34. 18. Brett Bennett and Fred. J. Kruger, Forestry and Water Conservation in South Africa; W. Beinart and P. Coates, Environment and History: The Taming of Nature in the USA and South Africa (London: Routledge, 1995). 19. H. Witt, “‘Clothing the Once Bare Brown Hills of Natal’: The Origin and Development of Wattle Growing in Natal, 1860–1960,” South African Historical Journal 53, no. 1 (2003): 99–122. 20. Pooley, Burning Table Mountain, 29–44, 47–95. 21. E. P. Phillips, “A Preliminary Report on the Veld-Burning Experiments at Groenkloof, Pretoria,” South African Journal of Science [SAJS] 16, no. 4 (1919): 285–99; and “Veld Burning Experiments at Groenkloof: Second Report,” Department of Agriculture Scientific Bulletin [DASB] 17 (1920): 7; R. R. Staples, “Experiments in Veld Management: First Report,” DASB 49 (1926): 5. 22. In 1913 J. P. H. Wink criticized sourveld farmers for burning in December/January to get winter grazing “with the least labour and expense.” Agricultural Journal of the Union of South Africa [AJUSA] 6, no. 2 (August 1913): 378. 23. For a farmer’s agreement on this point, see “Floreat Agricultura,” Agricultural Journal of the Cape of Good Hope [AJCGH] 16, no. 4 (July–December 1899): 247. 24. W. Beinart, The Rise of Conservation, 247. 25. H. D. Leppan, J. M. Hector, W. R. Thompson, T. D. Hall, R. A. Lindsay, and D. Moses, “The Grasslands of South Africa: Problems and Possibilities,” University of Pretoria Series 1, no. 23 (Pretoria: University of Pretoria, 1932), 15, 30. 26. For example, AJCGH 14, no. 5 (March 1899): 274; and 34, no. 6 (June 1909): 609; 16, no. 4 (February 1900): 246; AJUSA 5, no. 4 (April 1913): 515. On getting rid of old grass to avoid a fire hazard, see ‘Hippias,’ “Adventure by Fire,” Natal Department of Agriculture and Mines Journal [NAJ] 2, no. 14 (September 1899): 429–33. 27. See Choles in AJUSA 5, no. 4 (April 1913): 515, supported by farmers J. J. Claase and J. H. Vorster in AJUSA 6, no. 2 (August 1913): 378, and 6, no. 6 (December 1913): 987, and R. von Gernet, “Is South Africa Drying Up?” AJUSA 8, no. 1 (January 1914): 49–50. See also F. B. Smith, “Veld Burning,” NAJ 8, no. 6 (June 1905): 783. 28. See J. L. Baker, “Burning-off the Veld,” AJCGH 17, no. 3 (August 1900): 171; Choles in AJUSA 6, no. 3 (September 1913): 428; and R. Von Gernet, “Is South Africa Drying Up?”: 47–50; and K. A. Carlson, “Forestry in Relation to Irrigation in South Africa,” AJUSA 5, no. 2 (February 1913): 229. 29. See J. L. Baker in AJCGH 15, no. 11 (November 1899): 714–15; F. B. Smith, “Veld Burning”: 563–64; “Editorial Notes,” AJUSA 7, no. 3 (September 1918): 428. See also W. Torrance, Observations on Soil Erosion, Union of South Africa Department of Agriculture Bulletin 4 (Pretoria: Department of Agriculture, 1919), 4. 30. Beinart, The Rise of Conservation; Pooley, Burning Table Mountain. 31. Journal of the Department of Agriculture [JDA] 11, no. 5 (November 1925): 391. 32. Beinart, The Rise of Conservation; Pooley, Burning Table Mountain. 33. On mineral deficiencies caused by burning and the link with lamziekte, see “District Reports,” NAJ 2, no. 21 (January 1900): 646; AJUSA 6, no. 2 (August 1913): 378; 6, no. 6 (December 1913): 986, 987–89; and Alex Pardy, “Grass Burning,” NAJ 9, no. 7 (July 1903): 686–7. 34. H. D. Leppan, “The Grasslands of South Africa,” 20. 35. A. O. D. Mogg, “A Method of Veld Estimation,” SAJS 17, no. 2 (April 1921): 222–26, 223, 225. On links between changes in botanical composition and dunziekte, see C. J. Botha and M.-L. Penrith, “Poisonous Plants of Veterinary and Human Importance in Southern Africa,” Journal of Ethnopharmacology 119, no. 3 (October 2008): 549–58, 550–51; A. J. Mills and M. V. Fey, “Declining Soil Quality in South Africa: Effects of Land Use on Soil Organic Matter and Surface Crusting,” SAJS 99 (September/October 2003): 429–36, 433–34. 36. ‘Observor,’ “Another Natal View on Burning,” in “The Veld Burning Question: Some More Readers’ Opinions,” AJUSA 6, no. 6 (1913): 982–91, 985; see also M. J. Beukes from the Orange Free State, in the same edition, 987. 37. “District Reports,” in NAJ 2, no. 7 (June 1899): 202, and NAJ 2, no. 23 (February 1900): 712. 38. “District Reports,” NAJ 9, no. 7 (July 1906): 724. 39. E. de Beer and M. Mekhelas, “Looking Beyond Borders: Communal Perspectives on Fire in the Maloti-Drakensberg Transfrontier Project Area,” Maloti-Drakensberg Transfrontier Project Report (November 2004): 1–10, 3; T. Kepe, “Grasslands Ablaze: Vegetation Burning by Rural People in Pondoland, South Africa,” South African Geographical Journal 87 (2005): 10–17, 13; L. C. C. Liebenberg, “Veld Burning: How It Affects the Farmer as Well as the Country,” Farming in South Africa [FSA] (June 1934): 213–15, 214. 40. “Rural Notes,” Transvaal Agricultural Journal [TAJ] 1, no. 3 (April 1903): 82; and J. J. Claase’s letter in AJUSA 6, no. 2 (August 1913): 378. 41. For example, E. Baker, “Methods of Fire Protection: With Special Reference to Fires Caused by Sparks from Railway Engines,” in which he also lists veld burning by farmers as a problem, JDA 1, no. 5 (August 1920): 15; N. L. King in JDA 11, no. 3 (September 1925): 220–32, 227. 42. AJCGH 15, no. 11 (November 1899): 715–17. 43. AJUSA 6, no. 3 (September 1913): 426–27. See also AJUSA 5, no. 4 (April 1913), 513–15. 44. S. J. Pyne, “Problems, Paradoxes, Paradigms: Triangulating Fire Research,” International Journal of Wildland Fire 16 (2007), 271–76. 45. AJUSA 5, no. 4 (April 1913), 513–15. W. Muir supported this view in AJUSA, 6, no. 6 (December 1913): 986. 46. See W. Beinart, The Rise of Conservation, 224; L. Van Sittert, ‘Holding the Line: The Rural Enclosure Movement in the Cape Colony, c.1865–1910,” Journal of African History 43 (2002): 95–118. 47. AJUSA 6, no. 6 (December 1913): 986–87. 48. AJUSA 7, no. 2 (February 1913): 256–59. 49. Pooley, Burning Table Mountain, 57–60. 50. John W. Bews, “An Account of the Chief Types of Vegetation in South Africa, with Notes on the Plant Succession,” Journal of Ecology 4 (1916): 129–59. 51. Frederick E. Clements, Plant Succession: An Analysis of the Development of Vegetation (Washington, DC: Carnegie Institution of Washington, 1916). 52. J. W. Bews, “An Œcological Survey of the Midlands of Natal,” Annals of the Natal Government Museum 2, no. 4 (August 1913): 496; and “Some Aspects of Botany in South Africa and Plant Ecology in Natal,” SAJS 18, nos. 1 and 2 (1921): 63–80. 53. J. W. Bews, “Account of the Chief Types of Vegetation,” 147. 54. Notably as advocated by John Acocks in his influential Veld Types of South Africa (Pretoria: Government Printer, 1953). See J. McAllister, “Fire and the South African Grassland Biome,” in South Africa’s Environmental History, ed. S. Dovers et al., 160–73. On pollen evidence, see L. Scott and J. C. Vogel, “Late Quaternary Pollen Profile from the Transvaal Highveld, South Africa,” SAJS 79 (1983): 266–72; M. E. Meadows and H. P. Linder, “A Palaeoecological Perspective on the Origin of Afromontane Grasslands,” Journal of Biogeography 20 (1993): 345–55. 55. J. W. Bews, “Some Aspects of Botany,” 63–80, 70. 56. H. D. Leppan et al., “The Grasslands of South Africa: Problems and Possibilities,” University of Pretoria Series 1, no. 23 (Pretoria, 1932); H. B. Gilliland, “On the History of Plant Study upon the Witwatersrand,” Journal of South African Botany 19, no. 3 (1953), 93–104. 57. J. F. V. Phillips, “Fire: Its Influence on Biotic Communities and Physical Factors in South and East Africa,” SAJS 27 (1930): 352–67, 352. 58. Peder Anker, “The Context of Ecosystem Theory,” Ecosystems 5 (2002): 611–13. 59. Phillips, “Fire: Its Influence,” 364. 60. Pooley, Burning Table Mountain, 57–60. 61. H. D. Leppan et al., “The Grasslands of South Africa,” 11. 62. G. C. Theron, “Veld-Burning in the Western Transvaal,” FSA (September 1932): 244, 254. 63. R. R. Staples, “Studies in Veld Management: A Second Report on Certain Veld-Burning and Grazing Experiments,” DASB 91 (1930): 10; Anonymous, “Judicious Veldburning,” FSA 8, (1933): 247–48; G. A. Gill, “Veld-Burning Experiments,” FSA 11 (1936): 134. 64. J. F. V. Phillips, “Fire in Vegetation: A Bad Master, A Good Servant, and a National Problem,” Journal of South African Botany 2, Part 1 (1936): 35–45. 65. J. D. Scott, “A Contribution to the Study of the Problems of the Drakensberg Conservation Area” (PhD diss., University of the Witwatersrand, 1948). 66. Ibid. 67. K. E. W. Penzhorn, “Burning Veld out of Season,” FSA 17 (1942): 453–54. 68. I. B. Pole Evans, “The Union as a Pastoral Country,” FSA 7 (September 1932): 235–36, 263. 69. K. P. Kirkman and C. D. Morris, “Grazing and Fire Management of Humid Grasslands in South Africa: Empirical Basis of Current Recommendations,” National Department of Agriculture Report (August 1999): 10; D. D. Briske et al., “Rotational Grazing on Rangelands: Reconciliation of Perception and Experimental Evidence,” Rangeland Ecology & Management 61 (2007): 3–17. 70. D. Anderson, “Depression, Dust Bowl, Demography, and Drought: The Colonial State and Soil Conservation in East Africa during the 1930s,” African Affairs 83 (1984): 321–43. 71. Pooley, Burning Table Mountain, 65–70; B. M. Bennett and F. J. Kruger, “Ecology, Forestry and the Debate over Exotic Trees in South Africa,” Journal of Historical Geography 42 (2013): 100–9. 72. R. W. Thornton, statement in Sixth Session, Fourth British Empire Forestry Conference, South Africa (Pretoria: Government Printer, 1936): 128–29. 73. H. S. D du Toit, S. M. Gadd, G. A. Kolbe, A. Stead, and R. J. van Reenen, 1923, “Final Report of the Drought Investigation Commission,” U.G.49-’23 (Cape Town: Government of the Union of South Africa), 3, 5, 62. On the veld burning symposium, see SAJS 21, 1924. 74. On the resolution, see Report of the Chief Conservator of Forests (ARFD) Government of the Union of South Africa (U.G.) 1936/37 U.G.53/1937, 7; on fire incidence: ARFD1935/36 U.G.53—1936, 16; on catchments, and catchment surveys, ARFD1935/36 U.G.53–1936, 13; ARFD1936/37 U.G.53–1937, 7. On prohibiting burning, see W. E. Watt, “Forest Protection from Fire in South Africa,” in “Papers and Statements on Exotics,” Proceedings of the Fourth British Empire Forestry Conference, Union of South Africa (Pretoria: 1936): 1. 75. H. H. Bennett, Soil Erosion and Land Use in the Union of South Africa (Pretoria: Department of Agriculture, 1945), 9, 22. 76. Pooley, Burning Table Mountain, 80–83; Bennett, Soil Erosion and Land Use. 77. J. C. Ross, “Report of the Interdepartmental Committee on the Conservation of Mountain Catchments in South Africa” (Pretoria: Department of Agriculture, 1961). 78. Pooley, Burning Table Mountain, 89–91; H. L. Malherbe, E. R. March, U. W. Nänni, C. E. M. Tidmarsh, F. S. Grevenstein, J. C. Cox, and J. S. Whitmore, “Report of the Interdepartmental Committee of Investigation into Afforestation and Water Supplies in South Africa” (Pretoria: Department of Forestry, 1968). 79. J. W. Rowland, The Conservation Ideal: Being the SARCCUS Record for the Period 1952–1970 (Pretoria: SARCCUS, 1974), 3–9. 80. Ibid., 101. 81. Ibid., 101, 102. 82. Ibid., 102. 83. Ibid., 102. 84. Simon Pooley, “An Environmental History of Fire in South Africa in the Twentieth Century” (PhD diss., University of Oxford, 2010). 85. Pooley, Burning Table Mountain, 91. 86. P. J. Edwards, “The Use of Fire as a Management Tool,” in Ecological Effects of Fire in South African Ecosystems, ed. P. de V. Booysen and N. M. Tainton (Berlin: Springer, 1984), 349–62. 87. B. J. Huntley, ed., “South African Program for the SCOPE Mid-Term Project on the Ecological Effects of Fire,” South African National Scientific Programs Report 32 (September 1978), iv, 13, 16, 20. On the Drakensberg, see U. W. Nänni, “Veld Management in the Natal Drakensberg,” South African Forestry Journal 68 (1969), 5–15. 88. N. M. Tainton, P. de V. Booysen, D. J. Bransby, and R. Nash, “Long Term Effects of Burning and Mowing on Tall Grassveld in Natal: Dry Matter Production,” in Thirteenth Congress of the Grassland Society of Southern Africa, ed. J. M. L. C. Rhind (Bloemfontein, 1978): 41–44. 89. Ibid. 90. D. Meredith, ed., The Grasses and Pastures of South Africa (Johannesburg: Central News Agency, 1955); N. M. Tainton, ed., Veld and Pasture Management in South Africa (Pietermaritzburg: University of Natal Press, 1981); P. de V. Booysen and N. M. Tainton, eds., Ecological Effects of Fire in South African Ecosystems. 91. K. P. Kirkman and C. D. Morris, “Grazing and Fire Management of Humid Grasslands in South Africa: Empirical Basis of Current Recommendations,” National Department of Agriculture Report (August 1999): 4. 92. On Trollope’s career, interview with Winston Trollope, Fort Beaufort, Eastern Cape, South Africa, September 20, 2007. On Ed Komarek and Tall Timbers, see J. T. Hiers, “Birdsong Nature Centre” (2005), accessed August 2, 2016, http://www.georgiaencyclopedia.org/nge/Article.jsp?id=h-326. On the Tall Timbers conference in honor of Phillips: E. V. Komarek, “Proceedings: Annual Tall Timbers Fire Ecology Conference: Fire in Africa,” 11 (Tallahassee, 1972), unnumbered page, prelims. 93. Reassessment of indigenous farming practices by some British colonial experts noted by Hodge (in “The Hybridity of Colonial Knowledge: British Tropical Agricultural Science and African Farming Practices at the End of Empire,” in Bennet and Hodge, eds., Science and Empire) is a recent development in South Africa. On Trollope’s research, see W. S. W. Trollope, “Characteristics of Fire Behaviour,” in Ecological Effects of Fire, ed. P. de V. Booysen and N. M. Tainton, 200–17; and on African fire management, “Fire—a Key Factor in the Ecology and Management of African Grasslands and Savannas,” Seventh E.V. Komarek, Sr. memorial fire ecology lecture, in Proceedings of the 23rd Tall Timbers Fire Ecology Conference, ed. R. E. Masters and K. E. M. Galley (2007): 2–14. 94. K. P. Kirkman and C. D. Morris, “Grazing and Fire Management,” 4, 15. 95. W. J. Bond and C. L. Parr, “Beyond the Forest Edge: Ecology, Diversity and Conservation of the Grassy Biomes,” Biological Conservation 143 (2010): 2395–2404. © The Author 2017. Published by Oxford University Press on behalf of the American Society for Environmental History and the Forest History Society. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Environmental History Oxford University Press

Fire, Smoke, and Expertise in South Africa’s Grasslands

Loading next page...
 
/lp/ou_press/fire-smoke-and-expertise-in-south-africa-s-grasslands-wbXakHM0tu
Publisher
Oxford University Press
Copyright
© The Author 2017. Published by Oxford University Press on behalf of the American Society for Environmental History and the Forest History Society. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com
ISSN
1084-5453
eISSN
1930-8892
D.O.I.
10.1093/envhis/emx094
Publisher site
See Article on Publisher Site

Abstract

Abstract Fire as a management practice in South Africa’s grasslands inflamed heated debate throughout the twentieth century. Imported ecological ideas meshed with homegrown sectoral land management traditions to reinforce a powerful anti-burning narrative among experts. Farmers, however, developed their own theories on burning, and the history of fire research, policy, and management reveals a series of entanglements between ecological theory, management policies and recommendations, and practice that complicate narratives grounded in historiographical traditions focused on critiquing settler and colonial expertise. This article recommends three distinctions to make when thinking about the history of scientific expertise: first, between an individual’s abstract theorizing and his or her “thinking in the field”; second, between the influence of accepted scientific findings and the thinking guiding official policies of land management sectors; and third, between official policies and actual land management practices. This article provides overviews of long-term fire use in the country’s grasslands, the ecology of grasslands and fire in South Africa, of early debates over fire, and scientific fire research and management in the country. INTRODUCTION Cambridge-educated Scottish botanist John Bews introduced American ideas about vegetation, succession, and fire into South Africa in the 1910s, thereby providing a theoretical framework for an anti-burning narrative developed by local experts. Bews and several South African researchers including John Phillips, James (Hamish) Scott, and Winston Trollope tested and modified these ideas over the next half century in local grasslands. While Phillips in particular had concluded that fire was a natural and not necessarily destructive ecological agent in grasslands by the 1930s, foresters and agricultural officials—influenced by colonial traditions of forestry and botany—opposed burning. They linked it to vegetation degradation, soil erosion, and desiccation. The idea that anthropogenic burning had converted large areas of forest to grasslands or savanna, and that many of the country’s grasslands were human artifacts and did not constitute a so-called natural biome, persisted into the 1980s. From the 1920s, this anti-burning narrative was reinforced by concern over desiccation during extended dry periods, and later by fears that a natural disaster on the scale of the American dust bowl might occur in South Africa, a fear shared by British colonial experts in other parts of Africa. Farmers hotly debated the pros and cons of burning, and most continued to burn despite expert disapproval. Experimental studies began in the 1920s and included testing different approaches to burning as a grazing management practice. By the late 1930s, some of these showed that in some grasslands burning was an acceptable management practice, but it took another forty years for this to be officially adopted as policy rather than a sometimes “necessary evil.” A study of the evolution of grassland science, policy, and management in South Africa over the twentieth century reveals the complex interactions of scientific theories, experimental investigations, land management traditions, sectoral goals, and the interactions of land management bureaucracies, experts, and the farmers they were meant to be influencing. Influential experts worked in several contexts, as theoreticians, policy and management advisers, and researchers. Narrating this history requires engaging with the contradictions and accommodations these experts made as they moved between these operational modes and contexts and provides an opportunity to reflect on how historians write about experts and expertise in general. The history of the importation and misapplication of Western scientific ideas in colonial territories figures prominently in environmental historiography. Many of these narratives focus on foresters, commercial farmers, and agricultural experts and criticize their imposition of inappropriate land management techniques or their uncritical adoption in the Southern Hemisphere of ideas developed in the Northern Hemisphere. These studies frame science and expertise primarily in terms of relations of power between colonial authorities and the oppressed and have their roots in critiques of modernist and nationalist narratives emerging from African studies and South Asian subaltern studies in the 1970s and 1980s.1 As Jane Carruthers has shown, the environmental history of South Africa follows this pattern, with roots in a radical Marxist history that emerged in the 1970s, using class relations as a lens to reveal the subjugation and dispossession of Africans. As the country approached the transition to democracy in the early 1990s, the analytical emphasis shifted toward ecosocial themes, notably studies of how colonial and apartheid-era interferences in, and regulations of, space and natural resource use resulted in inter- and intraracial social conflicts and disruption.2 Both historiographical strands, however, contributed to a focus on environmental justice, largely presenting experts and expertise as the instruments of oppression. William Beinart has since challenged the assumptions underlying such portrayals. In a 2009 article with Karen Brown and Daniel Gilfoyle, he sought to broaden debates over colonial (and postcolonial) science and expertise by moving beyond narratives of the short-sightedness of experts mobilized by authoritarian states, instead encouraging an interdisciplinary effort to understand the science and document the production of scientific knowledge and ideas.3 This is a much more useful approach for understanding historical interactions between scientific research, expertise, policy, and management and the many other social and ecological factors that shape how land is actually managed. Reappraisals of the complicated chronologies that have shaped present-day land use can provide a useful context for developing future environmental policies, even when they do not aim to extract “lessons learned.”4 Models for such reappraisals preceded Beinart’s call. Richard Grove already advocated in 1995 for a reconceptualization of environmental expertise, introducing the notion of “green imperialism” and questioning the centrality of the metropole in the development of global environmentalism and environmental science.5 Since 2003, Beinart has reappraised the careers of South African veterinarians, agricultural officials, and modernizing farmers, focusing on the positive contributions they made to land and livestock management, based on long careers spent developing expertise grounded in local knowledge.6 Later work by Bennett, Kruger, and Pooley has further explored the sophistication and diversity—and also limitations—of the intellectual engagements of scientific foresters, botanists, and agricultural experts in the region.7 Collectively, this research reveals the ways in which scientific theory, policy, and management were imbricated with social, cultural, environmental, and economic factors. It does so by focusing on the activities, influence, and circulation of individual experts in specific contexts and within polycentric networks. This more nuanced approach chimes with developments in the history of science characteristic of the “new imperial history.”8 One way to capture this complexity is to think about three distinctions when reconstructing histories of scientific expertise. First, we must not assume that an individual’s abstract theorizing (as expressed in academic journals) necessarily chimes with what might be termed their “thinking in the field.” Second, we should not confuse or overstate the influence of accepted or published contemporary scientific findings with the thinking that guided official policies of land management sectors. Third, we ought to distinguish between official policies and actual land management practices. This article explores these distinctions in the context of the history of grasslands fire management in South Africa in the twentieth century. It reconsiders the contributions of some key figures in the history of grassland research. Of these, only Phillips is now relatively well known in the historiography, figuring prominently in books by Anker, Bennett and Kruger, and Pooley. Phillips played an important role in the history of ecological thinking in South Africa and has been discussed in terms of both his commitments to ecological science and also to what Saul Dubow has called the South Africanization of science. While a history of ideas approach is valuable here, it also has its limitations when writing about researchers who got their boots dirty in the field and routinely considered both policy and practice in their work.9 This article is also a contribution to the historical literature on fire in Africa, which—Stephen Pyne’s summary comments on fire in Africa and the growing literature on fire in the southwestern Cape’s fynbos biome, notwithstanding—is woefully small considering the scale of burning that occurs across Africa’s grasslands and its rich history of fire management by indigenous people and settlers.10 GRASSLANDS AND FIRE Pollen data indicate that grasslands have been present in South Africa throughout the Holocene, occupying roughly the same region as they do today, and they were even more extensive during the Pleistocene.11 The grassland biome (figure 1) falls mainly east of 25°E and between 25°S and 33°S in South Africa. With the exception of the high Drakensberg Mountains, it is dominated by C4 grass species (more efficient at higher temperatures and light intensities than C3 species12), with the characteristic species being Themeda triandra, which is primarily adapted to defoliation by fire.13 The region’s grasslands may be divided into semiarid regions, with 550 mm mean annual precipitation (MAP), mesic (735 mm MAP), and montane (1380 mm MAP). This biome is very rich in species diversity at 1000 m2 scale, but research has previously focused on grass species valuable for livestock production, rather than the plants that constitute this species diversity.14 Farmers and agriculturalists have customarily divided the grasslands into “sweet” (dry) and “sour” (moist) grasslands. As a rule, sweet grasslands occur in areas with MAP below 650 mm, often at lower elevations, where growing conditions are highly variable, linked to less predictable rainfall. The grasses are more palatable to livestock all year round because they have low fiber content and maintain aboveground nutrient content in winter. The offtake (by grazing, fire, or other means) removes a large percentage of production, except in years of exceptionally high rainfall. Changes in biomass are largely driven by rainfall rather than fire, which is relatively infrequent under natural conditions and has little effect on grass composition. Sweet grasslands are thus climate dependent. By contrast, sour grasslands (with a mean annual rainfall above 650 mm) are often found at higher elevations and along the Indian Ocean coastal belt where growing conditions are seasonally relatively fixed, making the grass unpalatable in the dry dormant season. The offtake removes only a small percentage of production, with a natural fire frequency of one to three years. Fire exclusion results in changes to grass composition and invasion by trees and shrubs. Sour grassveld is thus fire dependent.15 Figure 1. View largeDownload slide Map of South Africa’s grassland biome. Faint lines show country and provincial borders; the dark line outlines the grassland biome. The Dry Highveld grassland bioregion occurs on the interior central plateau of the country, where rainfall is below 600 mm per annum and is known as “sweet” grassland. Rainfall increases west to east across the Highveld, the vegetation transitioning from sweetveld into the sourveld of the Mesic Highveld grassland bioregion. The Drakensberg grassland bioregion occurs in the high Drakensberg Mountains, extending south through Lesotho as far as the Amathole Mountains in the eastern Cape, characterized by steep slopes and high rainfall. On the eastern foothills of the Drakensberg are the subescarpment grasslands, which are moist grasslands, in the mist belt. The letters refer to provinces as follows: A = Mpumalanga (formerly in Transvaal Province), B = Free State (formerly Orange Free State), C = KwaZulu-Natal (formerly Natal Province), D = Eastern Cape, E = Western Cape, and F = Kingdom of Lesotho. Credit: Simon Pooley, redrawn from Muscina and Rutherford, eds., The Vegetation of South Africa, Lesotho and Swaziland, Strelitzia 19 (Pretoria: South African National Biodiversity Institute, 2006), 44–45. Figure 1. View largeDownload slide Map of South Africa’s grassland biome. Faint lines show country and provincial borders; the dark line outlines the grassland biome. The Dry Highveld grassland bioregion occurs on the interior central plateau of the country, where rainfall is below 600 mm per annum and is known as “sweet” grassland. Rainfall increases west to east across the Highveld, the vegetation transitioning from sweetveld into the sourveld of the Mesic Highveld grassland bioregion. The Drakensberg grassland bioregion occurs in the high Drakensberg Mountains, extending south through Lesotho as far as the Amathole Mountains in the eastern Cape, characterized by steep slopes and high rainfall. On the eastern foothills of the Drakensberg are the subescarpment grasslands, which are moist grasslands, in the mist belt. The letters refer to provinces as follows: A = Mpumalanga (formerly in Transvaal Province), B = Free State (formerly Orange Free State), C = KwaZulu-Natal (formerly Natal Province), D = Eastern Cape, E = Western Cape, and F = Kingdom of Lesotho. Credit: Simon Pooley, redrawn from Muscina and Rutherford, eds., The Vegetation of South Africa, Lesotho and Swaziland, Strelitzia 19 (Pretoria: South African National Biodiversity Institute, 2006), 44–45. Hominins have been present in southern Africa for about 3 million years, but the available evidence suggests that although modern human hunter-gatherers used fire for hunting and fire-stick farming for at least 100,000 years, the practice had little impact on vegetation.16 Early Iron Age Bantu agropastoralists moved down the east coast from ca. 1700 BP, along the humid low-lying coastal plain (east of the 400-mm isohyet) into what are now KwaZulu-Natal Province and the Eastern Cape. They used fire for hunting, clearing land for agriculture, and iron smelting, and burned grassland and savanna to create fresh grazing for their livestock. The Highveld in the interior plateau of South Africa (above 1200 m above sea level) and its montane grasslands were settled later, from around 700 years BP, possibly as a result of increasing domestic stock numbers and resource depletion on the moister coastal lowlands. The key impacts during the early Iron Age in the eastern region were clearing and cultivation in the rich soils of valley bottoms, with lasting impacts on vegetation structure and composition, notably patchiness in savanna regions. By the late Iron Age, settlements shifted to the hilltops and the interior treeless grasslands, and regional economies developed where political power and wealth were linked with cattle. This resulted in overgrazing in some (notably northern) regions of what is now South Africa.17 Under colonial rule, livestock farming with sheep, cattle, and goats became the main land use practice in regions too dry for agriculture, resulting in land degradation in some areas. In grassland regions suitable for agriculture, extensive areas were converted to maize and cereal crops, particularly in the former Orange Free State and Transvaal provinces to supply the burgeoning urban centers developing around diamond and gold mines. From the mid-1800s, sugarcane was planted in the coastal lowlands of KwaZulu-Natal on the east coast. Plantations of introduced trees (mostly pines and eucalypts) were planted to supplement the country’s meager timber resources, first by state foresters in the Cape Colony in the 1860s, and then, following the creation of a national forestry department in 1910, in the moist upland grassland regions of what are now KwaZulu-Natal, Mpumalanga, and Northern Provinces.18 Private wattle (Acacia) plantations were established in Natal from the 1860s.19 FIRE, FARMERS, AND EXPERTISE For European settlers in the region, the cultural legacies of the anthropogenic transformations of the grassland and savanna biomes included both extensive use of burning by farmers and colonial experts’ hostility to burning practices, in particular botanists, foresters, and soil conservationists, all of whom regarded burning as a primitive and destructive practice. These experts believed that most bush fires were anthropogenic and resulted in the desiccation of the land, destruction of the indigenous flora in particular forests, favored less palatable grasses and scrub, and resulted in soil erosion.20 Analysis of the agricultural journals published in the Cape, Natal, and Transvaal colonies and territories (1898–1910) of what became the Union of South Africa in 1910, up to 1926, reveals the major debates about veld burning before prolonged scientific studies were set up. There were many rhetorical appeals to the authority of science, and concepts borrowed from the emerging science of ecology were used to justify a variety of opinions. Only after World War I, however, did South African scientists begin to publish some preliminary findings on fire and pasture management derived from experimental research.21 Controversy was fierce among farmers over the costs and benefits of using fire to manage veld, with most farmer correspondents to the journals in favor of controlled burning.22 Department of Agriculture officials and editors of the journals advocated new and improved farming methods to replace the need for veld burning, but they acknowledged that most of these were simply too expensive for the average farmer.23 South African farming mechanized late with tractors widely adopted by white farmers only after World War II.24 A 1932 report on the country’s grasslands acknowledged that research into “artificial pastures” and “grass silage” were in their infancy.25 In the meantime, judicious veld burning was a cheap and effective, if imperfect, method of veld management requiring no special equipment and no sustained use of labor. The main reason given by farmers in the agricultural journals for burning, especially in sourveld, was to get rid of old dry grass, which was useless as fodder, and force up new grass, which was edible and nutritious for livestock.26 They also argued that dry grass was a hazard when wildfires swept through their land. In contrast, the anti-burning camp (mostly experts, with some farmers, often from sweetveld regions) asserted that veld burning led to the “coarsening of veld grasses,” resulting in reduced stock-carrying capacity.27 Furthermore, they argued that destruction of the vegetation and humus layer by fire denuded and hardened the soil, thereby reducing its capacity to both absorb and hold moisture and leaving it prone to erosion.28 They maintained that these evil effects of burning contributed to the general desiccation of the country, negatively impacting climate, and rainfall in particular.29 Expert opposition to fire informed by the rhetoric of science began in the mid-1800s with concern over its desiccating effects. The key figures were the Cape Colony’s official botanists Ludwig Pappe and John Croumbie Brown, both active during periods of prolonged drought in the colony. Key Afrikaner agricultural bureaucrats trained in the United States following the South African (or Boer) War (1899–1902) became influential after union in 1910. They drew on ideas about vegetation and pasture science developed in the prairies of the American Midwest. The period from 1918 throughout the 1920s was very dry in South Africa, and a Drought Investigation Commission was convened, reporting in 1923. They concluded that the ongoing desiccation of the country was the result of bad farming practices, including veld burning, rather than declining rainfall.30 The commission drew on current ideas about successions to climax vegetation, influenced primarily by climate, to argue that “[g]rass-veld which grows so vigorously that it cannot be grazed down by full stocking is naturally tree-veld rather than grass-veld,” and that farms that could only be rendered fit for stock farming by burning should possibly be given over to timber production.31 By transposing arguments about (climate-driven) sweetveld onto (fire-driven) sourveld areas, they obscured the natural dynamics of fire-regulated sourveld areas. While it is true that fire exclusion would result in a shift to scrub and trees, fire exclusion was not “natural” in most sourveld grassland areas. As noted earlier, grasslands have been present in South Africa since the Pleistocene and thus fires must have been present before humans arrived on the scene, and further, colonial records indicate that settler farmers have been burning grass since the seventeenth century and learned the practice from indigenous herders.32 Between 1900 and 1913, farmers and officials frequently speculated that impoverishment of the soils and thus the veld by burning caused animal diseases including lamsiekte (which was actually botulism in phosphorus-deficient cattle from ingesting carrion and bones infected with clostridial toxins) and gallsickness (in reality a tick-borne disease).33 In the Transvaal, it was believed that burning could eradicate growths of “moulds and bacteria … one of which, Bacillus botulinus, may cause deaths amongst cattle.”34 Albert Moggs of the Division of Veterinary Services concluded that dunziekte (emaciation, caused by ingesting plants of the genus Senecio) and stijfziekte (an unidentified viral disease) were associated with burning and overstocking.35 Many African and white farmers saw burning as a valuable way of purifying or sanitizing the veld through controlling “vermin, snakes, ticks, etc., by removing the cover.”36 In 1899 Magistrate F. Y. Gibson of Nongoma District in Zululand advocated not burning grass routinely because grass burning could then be deployed against locusts.37 Fire was used to destroy cichlid moth larvae, believed to poison cattle.38 By the 1930s, most scientific authorities had discounted veld burning as an effective means of eradicating ticks and instead favored dipping. This is more expensive, however, and African farmers in the Maluti/Drakensberg area and in eastern Pondoland still list pest control (especially ticks) as a reason for controlled burning.39 Some correspondents to the agricultural journals argued that burning encouraged weeds and invasive plants, to the detriment of nutritious grasses.40 Several stressed the destructive effects of veld fires on trees, offering suggestions on prevention and firefighting.41 Agricultural experts acknowledged the problem of dealing with weeds with inadequate labor on terrain unsuitable for mowing. It was no use trying to change “the cultural custom of four or five generations” without offering “a sufficient substitute for the [condemned] practices,” wrote Charles Eustace Pillans of the Cape Department of Agriculture. He proposed paddocking combined with systematic weed eradication and cultivation of useful stock-food plants.42 H. J. Choles, an editor for Agricultural Journal of the Union of South Africa, recommended slowly phasing out burning, replacing it with heavy stocking and grazing as a means of veld management.43 Fire historian Stephen J. Pyne has critiqued fire research in general for focusing on the technical study of fire behavior (led by foresters) and neglecting the social dimension.44 There is much truth in this. However, agricultural researchers in South Africa in the early 1900s were well aware of the socioeconomic constraints on good farming practice, and therefore why farmers resorted to veld burning. In 1913 Choles observed, The firing of the veld is a time-honoured custom in South Africa, and dates from the first days of great, poorly stocked farms when, as is still sometimes the case, holdings were beyond the capacity of the limited number of stock they carried to keep the grass short; and as the grass shot up in the late summer months and dried out into coarse, unpalatable herbage, it was necessary either to mow it or burn it off to provide for succulent young shoots for the impoverished stock in the spring. Mowing was quite out of the question over the extensive areas that had to be disposed of; burning was cheap, rapid, and immediately as satisfactory. The ultimate effects were not contemplated in the grim struggle for existence.45In the early twentieth century, however, the promotion of enclosure and a progressive reduction in the size of farms affected established farming methods. The Fencing Act of 1912, amended in 1922, provided white farmers with subsidies for fencing.46 Farmer M. J. Beukes noted that “in the good old days,” when farmers had huge farms, it was acceptable to burn because it was the only way to get rid of “rank and coarse” grass and control ticks and other insects; in more modern and “civilized” times of smaller farms, however, “other methods of farming must be adopted, whether one likes them or not.”47 Farmer C. R. Prance worried that while it was feasible for his contemporaries to send their cattle to bushveld farms (in savanna, at lower elevations) while they waited for the “brand” (green grass after a burn) to come up on their home farm, “the next generation will have to share our farms—and then what?”48 Such fears over the future of farming in South Africa were shared by the government of the newly formed Union of South Africa (1910), led by two Afrikaner farmers and statesmen, Louis Botha and Jan Smuts, the latter of whom was a grass aficionado and keen botanist. Into the 1940s, the government feared the effects of unprofessional land management in the face of the closing of the South African frontier.49 BOTANY AND ECOLOGY Developments in scientific botany influenced agricultural experts’ and drought commissioners’ views on the effects of fire on vegetation. During the first decades of the twentieth century, botany was professionalized in South Africa, and ecological ideas became influential in the discipline, informing ideas about natural resource management more generally. A key figure, John William Bews, traveled from Scotland to take up the first professorship of botany at Natal University College in 1910. Bews’s “An Account of the Chief Types of Vegetation in South Africa, with Notes on the Plant Succession” (1916) was the first survey to apply an overtly ecological approach to South African vegetation and was informed by Frederic Clements’s ideas about vegetation succession.50 Clements argued that vegetation progresses linearly through a series of increasingly complex stages toward a stable climax community that is in equilibrium with the prevailing environmental conditions. Climate drives a linear unidirectional succession toward a single vegetation type, but a variety of disturbances (notably fire) may inhibit or temporarily reverse this progression.51 This theoretical framework had two important implications for thinking on grasslands and fire. First, Bews included in his vegetation classifications “changed” or “false” formations that were not in equilibrium with prevailing climatic conditions and which he believed were created and maintained by overgrazing and human use of fire. This included extensive grassland areas east of the Drakensberg Mountains in what is now KwaZulu-Natal Province. He thought that the destruction of the bush by wildfires and human burning practices was contributing to the desiccation of South Africa, particularly on the Highveld.52 Second, Bews argued that burning sets back the plant succession. He theorized that in “primitive” or semi-open grassland, burning prevented the plant succession from progressing to a more nutritious type and should be banned.53 The idea that anthropogenic burning had converted large forested areas to grassland, and thus that much of the country’s grassland was a secondary and somehow unnatural vegetation type, remained influential in South African scientific thinking until the 1980s when pollen records showed that the grasslands of the moist eastern region were of great antiquity.54 Bews developed his understanding of both the local flora in Natal and the region’s agricultural needs. Rather than adopt a purely anti-burning stance, he made an exception for situations where the succession appeared to be progressing toward scrub or forest. Where such vegetation encroached on grassland, Bews encouraged veld burning, should grazing be the desired land use. Bews rated investigations into the plant succession and principles underlying it to be “one of the most important economic results of our study of ecology in Natal” because based on this knowledge accurate guidance on the use of veld burning could be given, tailored to specific conditions and kinds of veld.55 Other major writers on veld burning in this period cited Bews’s interpretation of the South African flora through the ecological concept of succession. From 1927 to 1931, John Frederick Vicars Phillips worked on tsetse eradication in Kondoa Irangi in Tanganyika under Charles Swynnerton, where he observed the use of fire to combat tsetse flies and the practices of indigenous pastoralists and settler farmers. In 1931 he returned to South Africa to begin work as professor of botany at the University of the Witwatersrand in Johannesburg (a post he held until 1948). There he worked with T. D. Hall, agricultural adviser to African Explosives & Chemical Industry (which produced fertilizers), and J. D. (Hamish) Scott to develop Frankenwald, the estate bequeathed by the gold and diamond magnate Alfred Beit. Phillips made Frankenwald a center of ecological research, notably of problems relevant to pasture management and conservation. In the postwar period he trained nearly two hundred returning veterans in soil conservation on his “Donga Doctors” course, and his students dispersed all over the British territories in Africa.56 Phillips had just published the first review of scientific research on fire in southern and eastern Africa. He noted the efforts of “scientific workers” to educate the public “of the manifold evils following in the wake of fire,” pointing out that “little scientific experimentation [had] been brought to bear” upon these “problems.” He believed that both natural lightning fires and anthropogenic fires had swept the regions’ grasslands and savannas since prehistoric times, and he stressed the need to understand local conditions including climate, soil, and vegetation type before deciding on regional fire policies.57 Historian Peder Anker has portrayed Phillips as an ideologue, particularly as he perceives a link between Phillips’s championing of the ecosystem as a holistic concept and the political holism advocated by Phillips’s mentor Jan Smuts.58 However, this alleged political idealism should not overly color portrayals of Phillips as a scientist. Influenced by Clements, with whom he corresponded, Phillips also stressed the importance of contingent factors including the successional stage of the plant community and the nature of the season in terms of moisture and temperature. He was in fact wary of trying to generalize from basic scientific principles or from localized findings in an unsuitable way. He wrote cautiously, “I feel it desirable to urge the need of our carefully considering all regional circumstances in the light of scientific experience, before we definitely decry the practice of firing. Possibly I may be criticised for this statement, but in making it I consider I have but done my duty” [emphasis in original].59 As noted, the government was concerned over the future of farming in South Africa, and the Great Depression exacerbated this as the economy suffered a dip in mining returns and economists worried over the small internal market for manufacturing. The government looked to farming as a promising avenue for improvement through modernization. Scientific research and management was deemed required to save the land and improve agricultural output, and pasture research was institutionalized. The first pasture research stations were opened in 1934.60 The University of Pretoria and the African Explosives and Chemical Industry convened a Grassland Research Committee that issued a report in 1932. They argued that overgrazing and burning were resulting in the impoverishment of the naturally nutritious grasslands characterized by grasses like Themeda triandra (“rooigras”). Burning also destroyed the humus, and the soil became less able to absorb water, resulting in increased runoff during heavy rains and consequent soil erosion.61 By the mid-1930s, veld-burning experiments, notably those at Potchefstroom in Transvaal Province, were suggesting that annual burning in the Highveld—in sweetveld and semiarid regions—created good grazing in the short term. However, over the long term, burning resulted in reduced grass cover and soil erosion.62 In contrast, experiments at Cedara Agricultural College in the Natal Midlands showed that in high rainfall areas, in sourveld, burning was an acceptable management practice.63 Despite the positive findings for burning in certain conditions and regions, in 1936 Phillips published a major article on fire, “Fire in Vegetation: A Bad Master, a Good Servant, and a National Problem.” His choice of title reflected his assertion that “the man in the street has been instructed in print and by lecture that firing of vegetation is always highly destructive, and is to be considered as an act of vandalism against the national heritage.” Phillips agreed that burning in the wrong season could allow in undesirable weeds and encourage inferior grasses, but he argued that burning at the right time could actually reduce undesirable grasses and encourage more nutritious ones. Not burning at all could lead to a deterioration of the veld and scrub encroachment. Thus burning could be good farming practice in grasslands or savannas. Phillips was at pains to point out that “the management of grazing after fire is so important that investigation of influences of firing, not taking cognisance of this, and not providing for the control of this process, must largely prove abortive.”64 This was seemingly confirmed by experiments set up following severe droughts in 1932 and 1933 that drastically affected water supplies in the Natal midlands. It was popularly believed at the time that veld burning in the mountains was the cause of the water shortage, and that if the Drakensberg Mountain Range was protected from fire, this would solve the problem. In 1936 land for two research stations was acquired, one traversing savanna, thornveld, and grassland on commonage leased from the Borough of Estcourt, and the other in highland sourveld at Tabamhlope. Two experiments were set up in tall grassveld, the first to measure the effects of burning at different times, excluding grazing, the second to test the optimum burning treatments, with grazing. The main conclusions drawn from the first experiment, as interpreted by Hamish Scott in his 1948 doctoral thesis, were that total protection from fire led to a deterioration of the grass sward, increased runoff, and increased risk of accidental fires that would damage the grass cover and soil. Burning in the wrong season could be harmful, however, and the least harmful time to burn was in spring, after the first good rains.65 The second experiment somewhat modified the earlier findings. It showed that the grass sward would decline in quality if old vegetation was not removed. However, the average weekly increase in live weight of grazing animals was higher on mowed than on burned veld. This was a more critical view of burning, but the management implications were that, where mowing was not possible, burning was necessary to remove old ungrazed grass. Rather than specify a rigid “correct” interval between annual spring burns, Scott advocated judging how often to burn by observing the accumulation of dead grass.66 PRAGMATISM, IDEOLOGY, AND DESICCATION Bews’s and Phillips’s Clementsian theoretical framework of vegetation succession did not blind them to local conditions. Phillips advocated an ecological approach influenced by succession theory but advised against generalizations from basic scientific principles or from localized findings. He and his protégé Hamish Scott recommended experimental studies that took into account climatic conditions including seasonal fluctuations, soils, the state of the vegetation, and grazing and farming practices. They did not regard fire as an unnatural disturbance to African grassland and savanna ecosystems, but did criticize what they regarded as harmful burning practices. Although experimental studies at Cedara, Estcourt, and Tabamhlope showed the positive effects of properly applied veld burning for sourveld pastures, many officials remained opposed to the practice. Advice published in the Department of Agriculture’s journal Farming in South Africa grudgingly acknowledged the potential advantages of appropriate veld burning but presented the practice as a “necessary evil.”67 In 1932 Illtyd Buller Pole Evans, chief of the Union’s Department of Botany and founder of the Pasture Research and Veld Management Section in the Division of Plant Industry, had advocated the adoption of the “ingenious system, known as Deferred Grazing” employed in the United States, citing Arthur Sampson’s 1923 book Range and Pasture Management.68 In South Africa, this system developed into ever more sophisticated methods of rotational grazing, designed to make veld burning unnecessary, despite a lack of consensus over the superiority of well-managed rotational grazing versus well-managed continuous grazing.69 Soil conservation officials in particular resisted the findings of early experimental studies on fire that suggested it could be a good management practice. As David Anderson has shown, the American dust bowl experience, along with the Depression of the early 1930s, raised concerns over the environmental impacts of rapid increases in human and livestock populations in African Reserves in the context of the apparently more frequent droughts afflicting east Africa. These factors meant that soil conservation became a central concern of the British colonial authorities and spurred them to intervene directly in African farming practices. They drew on both South African and American recommendations and expertise.70 Colonial experts attributed desiccation and soil degradation to veld burning in this period. Foresters were particularly opposed to fire and veld burning, and in the 1930s this was the subject of a bitter dispute between agricultural and forestry experts in South Africa.71 At the fourth British Empire Forestry Conference, convened in South Africa in 1935, agriculturalist R. W. Thornton asserted that “burning on a gigantic scale … throughout the grass veld and savannah areas of this continent … has denuded the country of vegetation.” He argued that such burning removed trees that could restrict winds and thus contributed to wind erosion and high evaporation. “This is not only a South African question: it is an international African question,” he maintained.72 An expert narrative linking the drying up of the country to veld burning thus developed in South Africa during the 1920s and 1930s, exacerbated by droughts. The Drought Investigation Commission report of 1923 warned that failure to take action would result in “‘The Great South African Desert’ uninhabitable by Man.” In a 1924 symposium focused exclusively on veld burning, prominent botanists criticized the practice.73 In 1932–33, at a time when the American dust bowl was much in the news, there was a severe drought in Natal Province, and in 1934 Parliament passed a resolution to investigate the effects of veld burning on water supplies and soil erosion in mountain catchment areas. This was debated in the parliamentary capital, Cape Town, which experienced very high fire incidence in 1934. As a result, the state began acquiring private properties in mountain catchments, to be managed by the Division of Forestry. Foresters advocated prohibiting grazing in these regions because “wherever grazing is allowed the veld is burned as almost all owners of stock consider that annual burning of the veld is necessary to provide adequate pasturage”; thus they began to impound cattle that farmers grazed in these areas, particularly in times of drought. This considerably inflamed the controversy over burning.74 During World War II, soil conservation officials discouraged all burning. Jan Smuts set up a National Veld Trust in 1943, and Hugh Hammond Bennett, chief of the Soil Conservation Service of the US Department of Agriculture, toured the Union of South Africa in August 1944, noting widespread soil erosion and condemning “indiscriminate veld burning.”75 Grassland scientists worked within a highly charged political and policy environment that linked poor land management practices to treason.76 In the postwar period, forestry and agricultural researchers gradually came to believe that burning could be safely applied. However, they deemed this permissible only under stringent restrictions, mainly relating to the presumed least ecologically harmful time of year to burn. The Soil Conservation Act of 1946 restricted burning on private mountain catchment land by creating Fire Protection Districts, managed by Fire Protection Committees. An interdepartmental committee appointed by the Soil Conservation Board surveyed the country’s catchments from December 1952. Their report, published as the Ross Commission, carefully considered arguments for and against veld burning. They concluded that, while burning was a major factor contributing to catchment degradation, and should be forbidden in drier regions with vulnerable soils, in some humid catchments it was an acceptable farming practice.77 Following high fire incidence in the Cape mountain catchments in 1961, and a huge fire in July 1962, the Department of Forestry decided to investigate controlled burning in earnest. Fire prevention was simply not possible in these upland grasslands. The department also faced rising criticism in this period from farmers who alleged that its plantations (expanding eastward into the country’s inflammable grasslands) were drying up water supplies. An interdepartmental inquiry was convened, reporting in 1968. The commissioners argued that protecting natural vegetation from fire in fact reduced stream flow. This was because fires lower average veld age, reducing water use and evapotranspiration.78 South African experts were not developing their thinking on fire either exclusively in association with the colonial metropoles or American experts. The Southern African Regional Committee for the Conservation and Utilisation of the Soil (SARCCUS) was born out of postwar European colonial efforts to coordinate technical cooperation in sub-Saharan Africa, notably an inter-African conference at Goma in Belgian Congo in 1948 focused on soil conservation and land utilization. It worked in collaboration with the Commission for Technical Co-operation in Africa South of the Sahara, active from 1950, which was a joint venture of the Belgian, French, Portuguese, and British colonial authorities together with Southern Rhodesia (Zimbabwe) and the Union of South Africa. A 1974 review of the SARCCUS program noted the importance of training and expertise to this enterprise, emphasizing the value of “expert personnel with a lifetime of local experience.”79 The 1948 Goma conference called for the experimental study of fire, lamenting that “running fires annually devastated the whole of tropical Africa, causing a deterioration of the vegetal protecting mantle of the soil and exposing them to all forms of erosion.” Aware that fire was central to many “primitive” African land use practices, however, it was judged pragmatic to “temporarily tolerate” such fires but regulate them on areas of particular value. Further, given the difficulty of combating bush fires, they advocated “the palliative of early fires … at the commencement of the dry season, under the control of qualified services,” but “insisted that these tolerant measures … must not obscure the final goal, most vital for Africa, of the total suppression of bush fires.”80 A decade later, however, SARCCUS experts accepted that the use of fire is necessary in some systems of veld management for grazing animals. The question then is, at what time of year and under what ecological circumstances is it best to burn? The total exclusion of fire or even restriction of burning to defined seasons was deemed “extremely difficult.” Propaganda and education about good burning practices, rather than control through legislation, was advocated. The recommendation for motivating African farmers to practice better fire control was to encourage tree planting and improve their pastures. Experts acknowledged that “in the use of fire there is a conflict between the aims of the forester and the grazier … reflected in legislation and in official recommendations.” In countries where grazing interests were paramount, farmers practiced late burning at the end of the dry period just before spring growth. Where forest growth was important (West Africa and Malawi, for instance), land managers practiced early burning, at the beginning of the dry period when forests were still moist.81 By the 1970s, fire was believed to “[minimize] the evils of selective grazing” in sourveld, prevent “the accumulation of old grass and … [improve] the quality of the herbage and [reduce] the hazard of accidental fires.” In South Africa, sheep farmers in the Orange Free State, Natal, and Transvaal Highveld burned in late summer to stimulate green leafy herbage and in late autumn (this was frowned on) and winter to stimulate growth in early spring (and provide enhanced protein and phosphate content of the herbage). SARCCUS experts acknowledged that “the most desirable time for burning [in sourveld] is still a matter for controversy.”82 The difficulty of delaying burning through the dry months was that it was hard to prevent runaway fires. The summary of advice on burning (1974) noted irritably that “it seems to be impossible to generalise from region to region about burning,” but it concluded hopefully that increases in the price of land and more intensified farming practices would lead to the abandonment of large-scale burning and autumn burning on South Africa’s Highveld.83 THE NATAL SCHOOL When Hamish Scott became professor of pasture science at the University of Natal in 1948, he took the helm of an institution founded to tackle soil erosion and improve veld management. In 1950 Scott set up the now longest running fire experiments in South Africa, at Ukulinga, designed to work out the best approach to burning and grazing the veld (figure 2). At the same time, he worked on rotational farming methods to replace the necessity for burning. He had to combine scientific experiment with the requirements of state and provincial agricultural bureaucracies, who wanted tight restrictions on when fires could be lit, and ultimately (and fruitlessly) to replace fire altogether.84 Figure 2. View largeDownload slide The Ukulinga notice board showing details of treatments, as amended (1956). Credit: Simon Pooley, 2007. Figure 2. View largeDownload slide The Ukulinga notice board showing details of treatments, as amended (1956). Credit: Simon Pooley, 2007. In South Africa, veld burning had become widely accepted as a management strategy in higher rainfall regions by 1970. Administration of the Fire Protection Committees, which had resolutely opposed burning, was transferred to the Department of Forestry in 1970. The department adopted a policy of controlled burning, formalized with the passing of the Mountain Catchment Areas Act 63 of 1970.85 However, tight guidelines were issued, for example burning was only allowed between late July and October in Natal. Burning in August was permitted only after at least 15 mm of rain in twenty-four hours, but for the remainder of the period of legal burning, no rain was required beforehand. Permission from the local Soil Conservation Committee was required for any burning outside of this period.86 In 1977 a South African program on the ecological effects of fire was initiated as part of a wider program of the international Scientific Committee on Problems of the Environment (SCOPE), an initiative of the International Council of Scientific Unions. This South African SCOPE program stimulated both broad reviews and specific studies of key questions about ecological processes and management strategies. The first phase synthesized ongoing work on long-term experiments. For the grassland biome in South Africa, the surviving long-term experiments drawn on were at Ukulinga and in the Drakensberg catchment areas.87 Neil Tainton, who grew up in the Eastern Cape, trained at the University of Natal, and became a lecturer in the Department of Pasture Management and Soil Conservation there in 1959, oversaw the publication of a comparative review of the results of twenty-eight years of burning treatments in grassland at Ukulinga.88 The results were striking. Analysis of plots burned biennially or triennially showed no long-term influences of the preceding burning season. The report concluded, In the long term the effect of veld burning per se is little different irrespective of whether the veld is burnt in the dormant season and whether it is applied annually, biennially or triennially. In practice, however, different burning times and frequencies are associated with different grazing practices and it is this difference in post-burning management which is likely, in practice, to be the main contributory factor to differences in the response of veld to differing burning schedules.89This was revolutionary, upending theories on when to burn, all the shorter term experimental findings, and the persistent prescription to farmers to burn after the first spring rains. Grassland research in South Africa’s KwaZulu-Natal Province during the twentieth century generated a cogent and coherent approach to understanding and managing humid grassland, one that integrated fire as a useful tool. The continuity of ecological and management thinking on fire is remarkable, certainly when it comes to when to burn. Recommendations first recorded in the Natal Colony’s agricultural journal in 1899 are essentially those laid down by the Department of Agriculture in 1999. Ecological principles laid down by Bews and Phillips and related conceptions of rangeland succession still form, with due modification, the basis of research and management thinking. However, considerable progress was made in understanding why different burning treatments affect grass composition and productivity in the ways they do. This knowledge was disseminated throughout South Africa in a series of influential books and the Journal of the Grassland Society of Southern Africa.90 Despite this, Kirkman and Morris could still write in 1999 that “the role of fire and grazing procedure … as management tools are probably the most contentious issues regarding grazing management recommendation in South Africa.”91 Winston Trollope (figure 3), the son of an Eastern Cape farmer, graduated in grassland science after being inspired by Hamish Scott at the University of Natal in the 1970s. He landed the position of senior lecturer in pasture science at the University of Fort Hare (formerly the South African Native College) in 1970. His aim at Fort Hare was to develop farming methods appropriate for African farmers who had few resources: “it’s no good coming along with some weedicide or mechanical techniques; you’ve got to come along with something that doesn’t cost them any money.” Bush encroachment (notably by Acacia karoo) was a serious problem in the region, and in his experiments on grassland he had found that fire was the most practical means of dealing with it—and it was cheap to apply. Figure 3. View largeDownload slide Winston Trollope in one of his burn plots near Fort Beaufort, Eastern Cape. Credit: Simon Pooley, 2007. Figure 3. View largeDownload slide Winston Trollope in one of his burn plots near Fort Beaufort, Eastern Cape. Credit: Simon Pooley, 2007. Perhaps Trollope’s major contribution to South African fire science was his realization of the importance of fire intensity, previously ignored by South African researchers.92 Based for his entire academic career (1970–2005) at the University of Fort Hare in the former Homeland of the Transkei (now Eastern Cape Province), Trollope drew on Australian field-based research on fire intensity, US laboratory studies, his own fire experiments, and African communal burning practices and goat keeping to devise an effective and affordable means of veld management, in particular for controlling bush encroachment. Still active in fire research and management, Trollope concluded that research has proven that “indigenous fire management has demonstrated the advantage of using fire to remove moribund grass material and stimulate the growth of highly nutritious grass forage that should be grazed as soon as possible after the fire … provided the veld is given periodic extended rest periods.” He observed that he and other range scientists had condemned this practice for decades.93 Setting aside theorists, experts, and officials, then, what of the farmers? In 1999 Kevin Kirkman, professor of grassland science at the University of KwaZulu-Natal, admitted that most sourveld farmers had ignored the rotational grazing system for so long advocated by the Department of Agriculture. Part of the reason for this was the tension between the ecologically sensitive flexible management approach pioneered by Scott and policy requirements for fixed dates and rules. However, financial constraints were found to be farmers’ chief consideration when choosing a veld management strategy. By the end of the century, many of South Africa’s large-scale commercial farmers were in financial trouble, and small-scale commercial farmers and communal farmers had always had limited access to capital. Burning is cheap and labor efficient.94 CONCLUSION This history of the use of fire in South Africa’s grasslands reveals a chronology complicated by interactions between a range of intellectual, institutional, and managerial factors and shaped in important ways by a series of individuals who adapted their thinking as they moved between different modes of expertise. They developed their thinking across a range of professional and ecological contexts and in response to sometimes conflicting sectoral traditions and priorities. Attempts to fit them into a Procrustes’ bed of colonial or settler expertise obscures the diversity and range of their professional and personal trajectories and influence. It should be clear, for instance, that neither Bews’s Clementsian account of the structure of vegetation in South Africa nor Phillips’s theoretical publications on the ecosystem concept or his Clementsian influences meant that either man was incapable of adapting his thinking to local ecological circumstances and particular practical land management challenges. Bews acknowledged that different land management goals required different fire management strategies, and Phillips was explicit about the need to be careful when generalizing from basic scientific principles, and to always consider local ecological context including land use. South African farmers and officials had well developed ideas about veld burning before the advent of experimental work. When the first results of scientific research on burning became available, they were both unsuitably generalized from sweetveld to sourveld, and also resolutely ignored by agricultural and soil conservation officials who feared the desiccating effects of burning. Researchers were well aware they were operating in a highly charged political context. They worked in parallel on schemes to eradicate fire and on projects to manage grasslands with fire. The predominant focus in grassland fire research on water supplies, soil conservation, and grazing management for commercial farming precluded research on African fire management and also meant little work was done on grassland biodiversity. This latter omission is part of a global conservation science culture that has ignored tropical grasslands in favor of forests, some assuming that grasslands are a successional stage toward forest, and others that many types of grassland are of anthropogenic origin.95 Whatever the advice and policies of state and provincial officials, in many (probably most) cases, farmers ignored official attempts to restrict and replace burning with other techniques. For many farmers, the alternatives were simply impractical (on stony or steep land) and too expensive or labor intensive. If history is to speak to policy (and it need not), then it seems beneficial to explore the actual entanglements that inform or hinder policy development and implementation. It is not sufficient to study separately the histories of scientific theory, research, sectoral institutions and their policy and management recommendations, or the practices of the land managers they are meant to influence. Simon Pooley is the Lambert Lecturer in Environment (Applied Herpetology) at Birkbeck University of London. He has published a monograph, Burning Table Mountain (2014), on the fire history of South Africa’s Cape Peninsula, and articles on invasive plants and animals. He has taught conservation science at Imperial College London and is a visiting researcher at WildCRU, University of Oxford, where he brings a historical perspective to human wildlife conflict research. He serves on the IUCN’s Crocodile Specialist Group and Task Force on Human Wildlife Conflict. Notes This article was stimulated by a request for a talk for a festschrift for William Beinart, for which my thanks are due to Sandra Swart. I thank William for his guidance during my early explorations of this material. I thank the reviewers and editors for their constructive suggestions. 1. K. B. Showers, Imperial Gulleys: Soil Erosion and Conservation in Lesotho (Athens: Ohio University Press, 2005); M. Leach and R. Mearns, eds., The Lie of the Land: Challenging Received Wisdom of the African Environment (Oxford: James Curry, 1996); J. C. Scott, Seeing Like a State: How Certain Schemes to Improve the Human Condition Have Failed (New Haven: Yale University Press, 1998). 2. J. Carruthers, “Tracking in Game Trails: Looking Afresh at the Politics of Environmental History in South Africa,” Environmental History 11 (2006): 804–29. For the ecosocial trend, see Farieda Kahn, “Rewriting South Africa’s Conservation History—the Role of the Native Farmers Association,” Journal of Southern African Studies 20, no. 4 (1994): 499–516; and South Africa's Environmental History: Cases and Comparisons, ed. Stephen Dovers, Ruth Edgecombe, and Bill Guest (Athens: Ohio University Press, 2003). 3. W. Beinart, K. Brown, and D. Gilfoyle, “Experts and Expertise in Colonial Africa Reconsidered: Science and the Interpenetration of Knowledge,” African Affairs 108/432 (2009): 413–33. 4. P. Alagona, J. Sandlos, and Y. Wiersma, “Past Imperfect: Using Historical Ecology and Baseline Data for Conservation and Restoration Projects in North America,” Environmental Philosophy 9, no. 1 (2012): 49–70; M. M. Steen-Adams, N. Langston, M. D. O. Adams, and D. J. Mladenoff, “Historical Framework to Explain Long-Term Coupled Human and Natural System Feedbacks: Application to a Multiple-Ownership Forest Landscape in the Northern Great Lakes Region, USA,” Ecology and Society 20 (2015), http://dx.doi.org/10.5751/ES-06930-200128. 5. Richard H. Grove, Green Imperialism: Colonial Expansion, Tropical Island Edens and the Origins of Environmentalism, 1600–1860 (Cambridge: Cambridge University Press, 1995). 6. William Beinart, The Rise of Conservation in South Africa: Settlers, Livestock, and the Environment 1770–1950 (Oxford: Oxford University Press, 2003). 7. Brett Bennett and Frederick J. Kruger, Forestry and Water Conservation in South Africa: History, Science and Policy (Canberra: ANU Press, 2015); Brett Bennett, “Margaret Levyns and the Decline of Ecological Liberalism in the Southwest Cape, 1890–1975,” South African Historical Journal 67, no. 1 (2015): 64–84; Simon Pooley, Burning Table Mountain: An Environmental History of Fire on the Cape Peninsula (London: Palgrave Macmillan, 2014); “Recovering the Lost History of Fire in South Africa’s Fynbos, c. 1910–90,” Environmental History 17 (January): 55–83; and “Pressed Flowers: Ideas About Alien and Indigenous Plants at the Cape, c. 1902–45,” Journal of Southern African Studies 36, no. 3 (2010): 599–618. 8. See Joseph Hodge, “Science and Empire: An Overview of the Historical Scholarship,” in Science and Empire: Knowledge and Networks of Science across the British Empire, 1800–1970, ed. Brett Bennett and Joseph Hodge (London: Palgrave Macmillan, 2011), 3–29. 9. Peder Anker, Imperial Ecology: Environmental Order in the British Empire, 1895–1945 (Cambridge: Harvard University Press, 2001); Saul Dubow, A Commonwealth of Knowledge: Science, Sensibility, and White South Africa 1820–2000 (Oxford: Oxford University Press, 2006). 10. See S. J. Pyne et al., “Introduction,” in Wildland Fire Management: Handbook for Sub-Sahara Africa, ed. J. G. Goldammer and C. De Ronde (Freiburg: Global Fire Monitoring Centre, 2004), 1–10; Pooley, Burning Table Mountain. 11. L. Scott, H. M. Anderson, and J. M. Anderson, “Vegetation History,” in Vegetation of Southern Africa, ed. R. M. Cowling, D. M. Richardson, and S. M. Pierce (Cambridge: Cambridge University Press, 1997), 62–84. 12. Perennial grasses can have C3 or C4 pathways for capturing carbon dioxide during photosynthesis. All have C3 pathways, but in the tropics some species evolved the C4 pathway, which makes them better adapted to growing in warm or hot seasonal conditions, specifically through losing less water through transpiration. C3 species can tolerate frost and grow at higher attitudes. 13. T. G. O’Connor and G. J. Bredenkamp, “Grassland,” in Vegetation of Southern Africa, ed. R. M. Cowling et al., 215–57, 215–18, 222, 232. 14. R. G. Uys et al., “The Effect of Different Fire Regimes on Plant Diversity in Southern African Grasslands,” Biological Conservation 118 (2004): 489–99. 15. William J. Bond, “Fire,” in Vegetation of Southern Africa, ed. R. M. Cowling et al., 421–46, 440; L. Mucina, D. B. Hoare, L. Mervyn, et al., “Grassland Biome,” in The Vegetation of South Africa, Lesotho and Swaziland, ed. L. Mucina and M. C. Rutherford (Pretoria: SANBI, 2006), 348–436, 355–56; L. Mucina et al., “Indian Ocean Coastal Belt,” in The Vegetation of South Africa, 569–83, 577–80. 16. Pooley, Burning Table Mountain. 17. M. T. Hoffman, “Human Impacts on Vegetation,” in Vegetation of Southern Africa, ed. R. M. Cowling et al., 507–34. 18. Brett Bennett and Fred. J. Kruger, Forestry and Water Conservation in South Africa; W. Beinart and P. Coates, Environment and History: The Taming of Nature in the USA and South Africa (London: Routledge, 1995). 19. H. Witt, “‘Clothing the Once Bare Brown Hills of Natal’: The Origin and Development of Wattle Growing in Natal, 1860–1960,” South African Historical Journal 53, no. 1 (2003): 99–122. 20. Pooley, Burning Table Mountain, 29–44, 47–95. 21. E. P. Phillips, “A Preliminary Report on the Veld-Burning Experiments at Groenkloof, Pretoria,” South African Journal of Science [SAJS] 16, no. 4 (1919): 285–99; and “Veld Burning Experiments at Groenkloof: Second Report,” Department of Agriculture Scientific Bulletin [DASB] 17 (1920): 7; R. R. Staples, “Experiments in Veld Management: First Report,” DASB 49 (1926): 5. 22. In 1913 J. P. H. Wink criticized sourveld farmers for burning in December/January to get winter grazing “with the least labour and expense.” Agricultural Journal of the Union of South Africa [AJUSA] 6, no. 2 (August 1913): 378. 23. For a farmer’s agreement on this point, see “Floreat Agricultura,” Agricultural Journal of the Cape of Good Hope [AJCGH] 16, no. 4 (July–December 1899): 247. 24. W. Beinart, The Rise of Conservation, 247. 25. H. D. Leppan, J. M. Hector, W. R. Thompson, T. D. Hall, R. A. Lindsay, and D. Moses, “The Grasslands of South Africa: Problems and Possibilities,” University of Pretoria Series 1, no. 23 (Pretoria: University of Pretoria, 1932), 15, 30. 26. For example, AJCGH 14, no. 5 (March 1899): 274; and 34, no. 6 (June 1909): 609; 16, no. 4 (February 1900): 246; AJUSA 5, no. 4 (April 1913): 515. On getting rid of old grass to avoid a fire hazard, see ‘Hippias,’ “Adventure by Fire,” Natal Department of Agriculture and Mines Journal [NAJ] 2, no. 14 (September 1899): 429–33. 27. See Choles in AJUSA 5, no. 4 (April 1913): 515, supported by farmers J. J. Claase and J. H. Vorster in AJUSA 6, no. 2 (August 1913): 378, and 6, no. 6 (December 1913): 987, and R. von Gernet, “Is South Africa Drying Up?” AJUSA 8, no. 1 (January 1914): 49–50. See also F. B. Smith, “Veld Burning,” NAJ 8, no. 6 (June 1905): 783. 28. See J. L. Baker, “Burning-off the Veld,” AJCGH 17, no. 3 (August 1900): 171; Choles in AJUSA 6, no. 3 (September 1913): 428; and R. Von Gernet, “Is South Africa Drying Up?”: 47–50; and K. A. Carlson, “Forestry in Relation to Irrigation in South Africa,” AJUSA 5, no. 2 (February 1913): 229. 29. See J. L. Baker in AJCGH 15, no. 11 (November 1899): 714–15; F. B. Smith, “Veld Burning”: 563–64; “Editorial Notes,” AJUSA 7, no. 3 (September 1918): 428. See also W. Torrance, Observations on Soil Erosion, Union of South Africa Department of Agriculture Bulletin 4 (Pretoria: Department of Agriculture, 1919), 4. 30. Beinart, The Rise of Conservation; Pooley, Burning Table Mountain. 31. Journal of the Department of Agriculture [JDA] 11, no. 5 (November 1925): 391. 32. Beinart, The Rise of Conservation; Pooley, Burning Table Mountain. 33. On mineral deficiencies caused by burning and the link with lamziekte, see “District Reports,” NAJ 2, no. 21 (January 1900): 646; AJUSA 6, no. 2 (August 1913): 378; 6, no. 6 (December 1913): 986, 987–89; and Alex Pardy, “Grass Burning,” NAJ 9, no. 7 (July 1903): 686–7. 34. H. D. Leppan, “The Grasslands of South Africa,” 20. 35. A. O. D. Mogg, “A Method of Veld Estimation,” SAJS 17, no. 2 (April 1921): 222–26, 223, 225. On links between changes in botanical composition and dunziekte, see C. J. Botha and M.-L. Penrith, “Poisonous Plants of Veterinary and Human Importance in Southern Africa,” Journal of Ethnopharmacology 119, no. 3 (October 2008): 549–58, 550–51; A. J. Mills and M. V. Fey, “Declining Soil Quality in South Africa: Effects of Land Use on Soil Organic Matter and Surface Crusting,” SAJS 99 (September/October 2003): 429–36, 433–34. 36. ‘Observor,’ “Another Natal View on Burning,” in “The Veld Burning Question: Some More Readers’ Opinions,” AJUSA 6, no. 6 (1913): 982–91, 985; see also M. J. Beukes from the Orange Free State, in the same edition, 987. 37. “District Reports,” in NAJ 2, no. 7 (June 1899): 202, and NAJ 2, no. 23 (February 1900): 712. 38. “District Reports,” NAJ 9, no. 7 (July 1906): 724. 39. E. de Beer and M. Mekhelas, “Looking Beyond Borders: Communal Perspectives on Fire in the Maloti-Drakensberg Transfrontier Project Area,” Maloti-Drakensberg Transfrontier Project Report (November 2004): 1–10, 3; T. Kepe, “Grasslands Ablaze: Vegetation Burning by Rural People in Pondoland, South Africa,” South African Geographical Journal 87 (2005): 10–17, 13; L. C. C. Liebenberg, “Veld Burning: How It Affects the Farmer as Well as the Country,” Farming in South Africa [FSA] (June 1934): 213–15, 214. 40. “Rural Notes,” Transvaal Agricultural Journal [TAJ] 1, no. 3 (April 1903): 82; and J. J. Claase’s letter in AJUSA 6, no. 2 (August 1913): 378. 41. For example, E. Baker, “Methods of Fire Protection: With Special Reference to Fires Caused by Sparks from Railway Engines,” in which he also lists veld burning by farmers as a problem, JDA 1, no. 5 (August 1920): 15; N. L. King in JDA 11, no. 3 (September 1925): 220–32, 227. 42. AJCGH 15, no. 11 (November 1899): 715–17. 43. AJUSA 6, no. 3 (September 1913): 426–27. See also AJUSA 5, no. 4 (April 1913), 513–15. 44. S. J. Pyne, “Problems, Paradoxes, Paradigms: Triangulating Fire Research,” International Journal of Wildland Fire 16 (2007), 271–76. 45. AJUSA 5, no. 4 (April 1913), 513–15. W. Muir supported this view in AJUSA, 6, no. 6 (December 1913): 986. 46. See W. Beinart, The Rise of Conservation, 224; L. Van Sittert, ‘Holding the Line: The Rural Enclosure Movement in the Cape Colony, c.1865–1910,” Journal of African History 43 (2002): 95–118. 47. AJUSA 6, no. 6 (December 1913): 986–87. 48. AJUSA 7, no. 2 (February 1913): 256–59. 49. Pooley, Burning Table Mountain, 57–60. 50. John W. Bews, “An Account of the Chief Types of Vegetation in South Africa, with Notes on the Plant Succession,” Journal of Ecology 4 (1916): 129–59. 51. Frederick E. Clements, Plant Succession: An Analysis of the Development of Vegetation (Washington, DC: Carnegie Institution of Washington, 1916). 52. J. W. Bews, “An Œcological Survey of the Midlands of Natal,” Annals of the Natal Government Museum 2, no. 4 (August 1913): 496; and “Some Aspects of Botany in South Africa and Plant Ecology in Natal,” SAJS 18, nos. 1 and 2 (1921): 63–80. 53. J. W. Bews, “Account of the Chief Types of Vegetation,” 147. 54. Notably as advocated by John Acocks in his influential Veld Types of South Africa (Pretoria: Government Printer, 1953). See J. McAllister, “Fire and the South African Grassland Biome,” in South Africa’s Environmental History, ed. S. Dovers et al., 160–73. On pollen evidence, see L. Scott and J. C. Vogel, “Late Quaternary Pollen Profile from the Transvaal Highveld, South Africa,” SAJS 79 (1983): 266–72; M. E. Meadows and H. P. Linder, “A Palaeoecological Perspective on the Origin of Afromontane Grasslands,” Journal of Biogeography 20 (1993): 345–55. 55. J. W. Bews, “Some Aspects of Botany,” 63–80, 70. 56. H. D. Leppan et al., “The Grasslands of South Africa: Problems and Possibilities,” University of Pretoria Series 1, no. 23 (Pretoria, 1932); H. B. Gilliland, “On the History of Plant Study upon the Witwatersrand,” Journal of South African Botany 19, no. 3 (1953), 93–104. 57. J. F. V. Phillips, “Fire: Its Influence on Biotic Communities and Physical Factors in South and East Africa,” SAJS 27 (1930): 352–67, 352. 58. Peder Anker, “The Context of Ecosystem Theory,” Ecosystems 5 (2002): 611–13. 59. Phillips, “Fire: Its Influence,” 364. 60. Pooley, Burning Table Mountain, 57–60. 61. H. D. Leppan et al., “The Grasslands of South Africa,” 11. 62. G. C. Theron, “Veld-Burning in the Western Transvaal,” FSA (September 1932): 244, 254. 63. R. R. Staples, “Studies in Veld Management: A Second Report on Certain Veld-Burning and Grazing Experiments,” DASB 91 (1930): 10; Anonymous, “Judicious Veldburning,” FSA 8, (1933): 247–48; G. A. Gill, “Veld-Burning Experiments,” FSA 11 (1936): 134. 64. J. F. V. Phillips, “Fire in Vegetation: A Bad Master, A Good Servant, and a National Problem,” Journal of South African Botany 2, Part 1 (1936): 35–45. 65. J. D. Scott, “A Contribution to the Study of the Problems of the Drakensberg Conservation Area” (PhD diss., University of the Witwatersrand, 1948). 66. Ibid. 67. K. E. W. Penzhorn, “Burning Veld out of Season,” FSA 17 (1942): 453–54. 68. I. B. Pole Evans, “The Union as a Pastoral Country,” FSA 7 (September 1932): 235–36, 263. 69. K. P. Kirkman and C. D. Morris, “Grazing and Fire Management of Humid Grasslands in South Africa: Empirical Basis of Current Recommendations,” National Department of Agriculture Report (August 1999): 10; D. D. Briske et al., “Rotational Grazing on Rangelands: Reconciliation of Perception and Experimental Evidence,” Rangeland Ecology & Management 61 (2007): 3–17. 70. D. Anderson, “Depression, Dust Bowl, Demography, and Drought: The Colonial State and Soil Conservation in East Africa during the 1930s,” African Affairs 83 (1984): 321–43. 71. Pooley, Burning Table Mountain, 65–70; B. M. Bennett and F. J. Kruger, “Ecology, Forestry and the Debate over Exotic Trees in South Africa,” Journal of Historical Geography 42 (2013): 100–9. 72. R. W. Thornton, statement in Sixth Session, Fourth British Empire Forestry Conference, South Africa (Pretoria: Government Printer, 1936): 128–29. 73. H. S. D du Toit, S. M. Gadd, G. A. Kolbe, A. Stead, and R. J. van Reenen, 1923, “Final Report of the Drought Investigation Commission,” U.G.49-’23 (Cape Town: Government of the Union of South Africa), 3, 5, 62. On the veld burning symposium, see SAJS 21, 1924. 74. On the resolution, see Report of the Chief Conservator of Forests (ARFD) Government of the Union of South Africa (U.G.) 1936/37 U.G.53/1937, 7; on fire incidence: ARFD1935/36 U.G.53—1936, 16; on catchments, and catchment surveys, ARFD1935/36 U.G.53–1936, 13; ARFD1936/37 U.G.53–1937, 7. On prohibiting burning, see W. E. Watt, “Forest Protection from Fire in South Africa,” in “Papers and Statements on Exotics,” Proceedings of the Fourth British Empire Forestry Conference, Union of South Africa (Pretoria: 1936): 1. 75. H. H. Bennett, Soil Erosion and Land Use in the Union of South Africa (Pretoria: Department of Agriculture, 1945), 9, 22. 76. Pooley, Burning Table Mountain, 80–83; Bennett, Soil Erosion and Land Use. 77. J. C. Ross, “Report of the Interdepartmental Committee on the Conservation of Mountain Catchments in South Africa” (Pretoria: Department of Agriculture, 1961). 78. Pooley, Burning Table Mountain, 89–91; H. L. Malherbe, E. R. March, U. W. Nänni, C. E. M. Tidmarsh, F. S. Grevenstein, J. C. Cox, and J. S. Whitmore, “Report of the Interdepartmental Committee of Investigation into Afforestation and Water Supplies in South Africa” (Pretoria: Department of Forestry, 1968). 79. J. W. Rowland, The Conservation Ideal: Being the SARCCUS Record for the Period 1952–1970 (Pretoria: SARCCUS, 1974), 3–9. 80. Ibid., 101. 81. Ibid., 101, 102. 82. Ibid., 102. 83. Ibid., 102. 84. Simon Pooley, “An Environmental History of Fire in South Africa in the Twentieth Century” (PhD diss., University of Oxford, 2010). 85. Pooley, Burning Table Mountain, 91. 86. P. J. Edwards, “The Use of Fire as a Management Tool,” in Ecological Effects of Fire in South African Ecosystems, ed. P. de V. Booysen and N. M. Tainton (Berlin: Springer, 1984), 349–62. 87. B. J. Huntley, ed., “South African Program for the SCOPE Mid-Term Project on the Ecological Effects of Fire,” South African National Scientific Programs Report 32 (September 1978), iv, 13, 16, 20. On the Drakensberg, see U. W. Nänni, “Veld Management in the Natal Drakensberg,” South African Forestry Journal 68 (1969), 5–15. 88. N. M. Tainton, P. de V. Booysen, D. J. Bransby, and R. Nash, “Long Term Effects of Burning and Mowing on Tall Grassveld in Natal: Dry Matter Production,” in Thirteenth Congress of the Grassland Society of Southern Africa, ed. J. M. L. C. Rhind (Bloemfontein, 1978): 41–44. 89. Ibid. 90. D. Meredith, ed., The Grasses and Pastures of South Africa (Johannesburg: Central News Agency, 1955); N. M. Tainton, ed., Veld and Pasture Management in South Africa (Pietermaritzburg: University of Natal Press, 1981); P. de V. Booysen and N. M. Tainton, eds., Ecological Effects of Fire in South African Ecosystems. 91. K. P. Kirkman and C. D. Morris, “Grazing and Fire Management of Humid Grasslands in South Africa: Empirical Basis of Current Recommendations,” National Department of Agriculture Report (August 1999): 4. 92. On Trollope’s career, interview with Winston Trollope, Fort Beaufort, Eastern Cape, South Africa, September 20, 2007. On Ed Komarek and Tall Timbers, see J. T. Hiers, “Birdsong Nature Centre” (2005), accessed August 2, 2016, http://www.georgiaencyclopedia.org/nge/Article.jsp?id=h-326. On the Tall Timbers conference in honor of Phillips: E. V. Komarek, “Proceedings: Annual Tall Timbers Fire Ecology Conference: Fire in Africa,” 11 (Tallahassee, 1972), unnumbered page, prelims. 93. Reassessment of indigenous farming practices by some British colonial experts noted by Hodge (in “The Hybridity of Colonial Knowledge: British Tropical Agricultural Science and African Farming Practices at the End of Empire,” in Bennet and Hodge, eds., Science and Empire) is a recent development in South Africa. On Trollope’s research, see W. S. W. Trollope, “Characteristics of Fire Behaviour,” in Ecological Effects of Fire, ed. P. de V. Booysen and N. M. Tainton, 200–17; and on African fire management, “Fire—a Key Factor in the Ecology and Management of African Grasslands and Savannas,” Seventh E.V. Komarek, Sr. memorial fire ecology lecture, in Proceedings of the 23rd Tall Timbers Fire Ecology Conference, ed. R. E. Masters and K. E. M. Galley (2007): 2–14. 94. K. P. Kirkman and C. D. Morris, “Grazing and Fire Management,” 4, 15. 95. W. J. Bond and C. L. Parr, “Beyond the Forest Edge: Ecology, Diversity and Conservation of the Grassy Biomes,” Biological Conservation 143 (2010): 2395–2404. © The Author 2017. Published by Oxford University Press on behalf of the American Society for Environmental History and the Forest History Society. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com

Journal

Environmental HistoryOxford University Press

Published: Jan 1, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 12 million articles from more than
10,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Unlimited reading

Read as many articles as you need. Full articles with original layout, charts and figures. Read online, from anywhere.

Stay up to date

Keep up with your field with Personalized Recommendations and Follow Journals to get automatic updates.

Organize your research

It’s easy to organize your research with our built-in tools.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

Monthly Plan

  • Read unlimited articles
  • Personalized recommendations
  • No expiration
  • Print 20 pages per month
  • 20% off on PDF purchases
  • Organize your research
  • Get updates on your journals and topic searches

$49/month

Start Free Trial

14-day Free Trial

Best Deal — 39% off

Annual Plan

  • All the features of the Professional Plan, but for 39% off!
  • Billed annually
  • No expiration
  • For the normal price of 10 articles elsewhere, you get one full year of unlimited access to articles.

$588

$360/year

billed annually
Start Free Trial

14-day Free Trial