From adoption claims to understanding farmers and contexts - cimmyt

Agriculture, Ecosystems and Environment 187 (2014) 116–132

Contents lists available at ScienceDirect

Agriculture, Ecosystems and Environment journal homepage: www.elsevier.com/locate/agee

From adoption claims to understanding farmers and contexts: A literature review of Conservation Agriculture (CA) adoption among smallholder farmers in southern Africa Jens A. Andersson a,∗ , Shereen D’Souza b a b

CIMMYT-Southern Africa, 12.5 km peg Mazoe road, Harare, Zimbabwe Independent scholar

a r t i c l e

i n f o

Article history: Received 5 February 2013 Received in revised form 12 July 2013 Accepted 21 August 2013 Available online 21 September 2013 Keywords: Conservation agriculture Southern Africa Adoption Malawi Zimbabwe Zambia Farming systems

a b s t r a c t This literature review of Conservation Agriculture (CA) adoption among smallholder farmers in southern Africa (Malawi, Zambia and Zimbabwe) analyses the historical background of the upsurge in CA promotion, the various definitions of CA that have emerged since the 1990s, the barriers to its adoption, as well as uptake figures and adoption studies. First tested as soil and water conservation measures, large-scale promotion followed a reframing of CA as a production-enhancing set of practices. Different definitions of what constitutes and is promoted as CA in southern Africa complicates the assessment of adoption across the region, while a commonly used, reductionist notion of CA adoption – as the uptake of minimum tillage – in adoption data collection, casts doubts on the validity of adoption figures. As CA uptake is often also incentivized by means of input support (fertilizers, seeds, herbicides) provided by promotional projects, adoption claims have limited value. Current CA adoption studies are methodologically weak as they are biased by the promotional project context in which are carried out, and build on farm-scale analyses of standard household surveys. A more thorough analysis of farming households and their resource allocation strategies is required to understand the farm-level adoption constraints different types of farmers face. As contextual factors appear key influences on smallholders’ farming practices, studies focusing on the wider market, institutional and policy context are also needed if we are to understand (limited) CA adoption in southern Africa. © 2013 Elsevier B.V. All rights reserved.

1. Introduction Over the past decade Conservation Agriculture (CA) has, arguably, become a hegemonic paradigm in scientific and policy thinking about sustainable agricultural development. Numerous policy documents and development projects have been dedicated to CA, and specialized units researching and promoting CA have emerged in leading international research and policy institutes such as the CGIAR and FAO.1 CA’s prominence in recent debates on sustainable intensification, climate change and as a form of Climate Smart Agriculture, is further evidence of the paradigm’s prominence in global agricultural development policy (FAO, 2011a; McCarthy et al., 2011). At the same time questions and controversies have emerged regarding the ability of CA to achieve the many virtues that proponents assert it embodies. For instance, claims regarding the role of CA in carbon sequestration (Lal, 2004;

∗ Corresponding author. Tel.: +263 772 469211; fax: +263 4301327 E-mail address: [email protected] (J.A. Andersson). 1 See: http://www.fao.org/ag/ca; http://www.cimmyt.org/en/programs-andunits/conservation-agriculture-program (accessed 21/12/2012). 0167-8809/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.agee.2013.08.008

FAO-REOSA, 2011; Kassam et al., 2009) have been questioned as evidence is lacking or inconclusive (Baker et al., 2007; Govaerts et al., 2009; Chan et al., 2011; Chivenge et al., 2007; Luo et al., 2010; Paul et al., 2013). Simultaneously, the universal applicability of CA’s three main principles – (1) minimal soil disturbance, (2) permanent soil cover and, (3) crop rotation (and crop diversification) – both individually and in combination, has come under scrutiny. Some scholars and practitioners favour more practical and context-specific approaches over the strict implementation of CA principles such as no-till (Kirkegaard et al., 2011). Others question the applicability of CA principles in the context of diverse, smallholder farms and farming systems (Giller et al., 2009; Guto et al., 2011a). These contrasts and controversies are also pertinent to the issue of CA adoption among smallholder farmers in sub-Saharan Africa. While it is often acknowledged that uptake has been (s)low, views on and estimates of adoption vary greatly (Derpsch et al., 2010, p. 15; Erenstein et al., 2012, p. 198; Kassam et al., 2009; Bwalya and Friedrich, 2009, p. 7,11; see also below). Assertions in the literature of incremental or even exponential uptake in some areas (WorldBank, 2012; Bunderson et al., 2009) are juxtaposed with evidence of dis-adoption and, limited or partial uptake elsewhere

J.A. Andersson, S. D’Souza / Agriculture, Ecosystems and Environment 187 (2014) 116–132

(Mazvimavi et al., 2011; Mazvimavi and Nyamangara, 2012; Arslan et al., 2013). This paper reviews a limited academic literature and a more substantial body of policy and project documents on CA adoption among African smallholder farmers. It aims to assess the extend of CA adoption and the barriers that limit its uptake, as well as how CA adoption has been studied and what can be learned from this. The focus is on Zambia, Zimbabwe and Malawi, as these southern African countries have been at the forefront of smallholder CA promotion efforts since the late 1990s. They also represent a range of agro-ecological circumstances, different demographic, socio-economic, and institutional contexts. In order to understand CA adoption, and how it has been promoted and studied in southern Africa, it is necessary to understand the historical context in which CA for African smallholder farmers emerged. The paper therefore first discusses how the large-scale promotion of CA in southern Africa followed a reframing of CA as a production and food security enhancing set of practices. It is shown how the distinct socio-economic and (donor)political contexts of the three countries in which this reframing occurred, gave rise to different CA promotional packages and sets of agronomic practices promoted under the banner of CA. The diversity in what is considered CA, complicates the assessment of CA uptake and barriers to adoption. While the literature has identified a number of adoption constraints at farm level, contextual factors influencing CA (non) adoption have generally received less attention. A review of CA of adoption figures, in Section 3, underscores the importance of the development project context for our understanding of CA adoption in southern Africa; donor-supported development and humanitarian aid projects have profoundly influenced adoption figures and adoption studies. As an adopting farmer is usually taken as someone applying one or more CA principles (usually minimum tillage) on some part of his/her land in a particular season, figures on CA adoption in southern Africa need to be treated cautiously. In the discussion section, we scrutinize the methodology of current CA adoption studies in southern Africa. It is argued that current adoption studies are inadequate in defining exactly the extent of CA uptake, and also biased. The CA promotional project context in which these studies are usually carried out, influences the selection of respondents as well as these studies’ findings. A heavy reliance on econometric analyses of standard farm household survey data further limits our understanding of CA adoption. The interpretative framework of these CA adoption studies appears weak as the functioning of smallholder farming households and their production systems appears ill-understood. Consequently, these econometric analyses reveal general characteristics of CA (component) adopters, rather than revealing farmers’ resource allocation strategies that underpin adoption and non-adoption. Household economic analyses highlighting the viability and feasibility of CA for smallholders despite the higher capital investment needs of CA, appear equally limited in their understanding of the realities of smallholder farming in southern Africa. CA adoption studies may therefore benefit from adopting a – empirically grounded – systems perspective and the use of a wider set of quantitative and qualitative research methods. Complementary analyses of wider socio-economic, institutional and policy factors are needed as farm-level practices are often constrained or enabled by forces beyond the farm. 2. CA in southern Africa: Drivers, definitions, and adoption constraints 2.1. The development of smallholder CA in southern Africa Minimum tillage systems have been around in southern Africa for many decades. While some traditional African land use systems

117

may be regarded as such (Page and Page, 1991), minimum or conservation tillage usually refers to reduced tillage systems that were developed after the widespread adoption of plough agriculture during colonial times.2 In the countries considered in this review, such systems were first introduced on large-scale commercial farms in Rhodesia (now Zimbabwe) in the late 1960s. Escalating fuel, machinery and maintenance costs triggered interest in these techniques and prompted research (Smith, 1988). Cost considerations also drove initial interest among large-scale commercial farmers in Zambia in the 1980s (Haggblade and Tembo, 2003a, p. 13). The interest of large-scale farmers was matched by that of researchers and agricultural officers who considered the erosion reducing effects of these practices as most relevant for smallholder farming, particularly in Rhodesia’s Tribal Trust Lands (now Communal Areas) (Andersson and Giller, 2012, p. 24). Soil erosion had been a preoccupation in colonial policy circles in southern Africa at least since the 1930s (Trapnell, 1943; Baudron et al., 2012; Beinart, 1984; McCracken, 2012, p. 207–209), and soil conserving CA practices well-fitted that tradition. As with other soil conservation technologies, research on, and promotion of smallholder-specific CA technologies thus began as an adaptation of the technologies developed for large-scale, commercial farms interested in cost savings (Elwell, 1993, p. 8; Haggblade and Tembo, 2003a, p. 7). 2.1.1. Zimbabwe: from soil conservation to rural livelihood crisis In Zimbabwe, research on conservation tillage for smallholder farmers expanded at government research institutes in the 1980s (Norton, 1988; Twomlow et al., 1995). Extensive onstation and on-farm experimentation with different conservation tillage techniques was conducted in different agro-ecological zones (Andersson and Giller, 2012, p. 24). The preoccupation of colonial policies with soil degradation was reinforced by recurrent and severe droughts in the first two decades after Zimbabwe’s independence in 1980. Researchers and development practitioners valued conservation tillage techniques for their contribution to soil and water conservation in the often densely populated, but agriculturally marginal Communal Areas (Witoshynsky and Phiri, 2000; Andersson and Giller, 2012); areas where African farmers had become concentrated during the colonial era. Yet, the largescale promotion of CA principles such as minimum soil disturbance and permanent soil cover to smallholder farmers remained limited. Increases in production and yields were largely realized through African smallholders’ expansion and intensification strategies following the removal of colonial blockades on production and marketing (Andersson, 2007, p. 685). Market incorporation, government (price) subsidies and credit, and the extension of green revolution technologies resulted in a smallholder production revolution (Rukuni and Eicher, 1987). A mounting economic crisis and dwindling government support for smallholder agriculture in the 1990s ended this smallholder production boom. Increasing rural poverty and food insecurity formed the backdrop for the large-scale promotion of CA techniques as part of donor-funded humanitarian aid programmes in 2003. For political reasons, funding for such programmes that combined seed and fertilizer distributions with CA promotion, was largely channelled through NGO’s rather than the Zimbabwean government (Andersson and Giller, 2012). Changes in the political economy of Zimbabwe’s smallholder farming sector thus led to a reframing of CA; ‘CA for soil and water management’

2 The use of minimum, reduced, conservation, and zero tillage suggests that even the CA principle of minimum soil disturbance has divergent definitions. Whether some practices, such as the planting basin-technology, the appliance of zero tillage to only one crop in the rotation (see: Erenstein, 2012, p. 52), or the harvesting of root and tuber crops qualify as minimal soil disturbance is a moot point. For convenience, this paper takes all such practices as minimal soil disturbance.

118


lost eminence as the urgency of poverty reduction, food security and productivity growth became the new drivers of CA promotion. A booklet of the Zimbabwe CA taskforce–a collaboration of the FAO, donors, international agricultural research and government organizations–is illustrative of this shift, CA “can significantly boost production, and improve the food security and livelihoods of farming households” (ZCATF, 2009, p. ii). 2.1.2. Zambia: reversing productivity decline As in Zimbabwe, concerns over soil degradation raised interest in extending conservation tillage techniques to smallholder farmers in Zambia. Such concerns became increasingly acute in the early 1990s, when government subsidies for smallholder agriculture were abruptly discontinued. At the same time, drought and livestock diseases drastically reduced cattle numbers—i.e. draught power. Consequently, the incidence of late planting increased, which impacted negatively on yields in plough-based smallholder farming. Since independence, the Zambian government had sought to expand and intensify land use and food production in this relatively sparsely populated country through the introduction of high-input agriculture (hybrid maize and fertilizers) and animal traction subsidies to smallholders (van Donge, 1984). Government support for maize production and marketing was further expanded in the 1980s, when the country’s declining copper industry prompted government to develop rural alternatives for an increasingly unemployed urban population (Ferguson, 1999; Gould, 1997). As government-subsidized maize production proved financially unsustainable, subsidies had to be abolished under structural adjustment in 1991 (Baudron et al., 2007, p. 7). According to Haggblade and Tembo (2003a, p. 10) ‘Farmers quickly responded by diversifying out of maize production and by reducing fertilizer use by over two-thirds as availability diminished and input prices jumped’. Maize productivity decline does, however, appear to have been a more gradual process, set in already in the late 1980s, after rapid yield growth in the earlier part of that decade (Howard and Mungoma, 1996, p. 10). The crop’s share in farmers’ cropped land area also decreased more gradually, and this decrease occurred in the latter part of the 1990s. Haggblade and Tembo’s (2003a, p. 10–12) own data suggest that drought and corridor disease were more important shocks in Zambian agriculture in the early 1990s, causing a rapid decline in cattle numbers. Whereas the impact of declining government support for maize production and marketing on smallholder farmers productivity may be contested, this support features prominently in interpretations of soil degradation in Zambia’s low to moderate rainfall zones of the Southern, Central and Eastern Provinces: ‘The continuous high-input maize mono-cropping left Zambian soils seriously degraded. . . Heavy application of nitrogen fertilizers, coupled with little attention to organic material, led to serious soil degradation—erosion, acidification, reduction in soil organic material and the build-up of plow pans across much of Zambia’s maize belt’ (Haggblade and Tembo, 2003b, p. 8; Baudron et al., 2007, p. 7). Evidence supporting these claims is, however, not often provided. An exception is Baudron et al. (2007, p. 7) who observed that in ‘Southern Province, the widespread plough pan problem . . . has contributed to a significant number of households migrating to other lands.’ Yet, a study in the three provinces where CA promotion has been concentrated (Southern Eastern and Central), found little evidence of plough pans (Umar, 2012). Claims regarding excessive use of fertilizers (Saasa, 2003, p. 9) and fertilizer-induced soil acidification are also disputed. Many soils in Zambia are inherently acidic and as Aagaard (2010, p. 4) has argued, ‘67% of smallholders use none [fertilizers] at all.’ The decline of smallholders’ productivity in Zambia increased rural poverty and food insecurity. With it, Aagaard (2010, p. 1)

claims, ‘land degradation and deforestation are accelerating, and millions of farmers are busy depleting the soil upon which they and future generations depend’ (see also: Milder et al., 2011). Smallholders’ ‘inappropriate farming practices’ became singled out as the cause of Zambia’s declining land productivity (INESOR, 1999, quoted in Haggblade and Tembo, 2003b, p. 8). That such farming practices may have been an appropriate response – albeit with undesirable effects – to a particular wider socio-economic environment and incentive structure received little attention. Focusing on the immediate causes – inappropriate farming practices – of land degradation and productivity decline, a technology-centred agricultural development narrative gave rise to the promotion of CA for Zambian smallholders in the 1990s. The Zambian government, which has been supportive of CA in numerous policy documents (Umar, 2012; Aune et al., 2012), has not been at the forefront of CA research and extension. As large parts of its agricultural budget go to incentives for maize production – for instance through its Fertilizer Support Programme (FSP) (WorldBank, 2010) – it has been suggested government may actually impede the spread of CA practices such as crop rotation (Umar et al., 2012, p. 923). Rather than government, it was the Zambian National Farmers Union (ZNFU) that set-up the Conservation Farming Unit (CFU) which drove the promotion of CA at scale (Aagaard, 2012). The importance of cotton production and (previously) commercial maize production in the areas where it was first promoted, has situated CA in Zambia firmly in a narrative of commercial smallholder production. Cotton companies provided inputs through contract farming arrangements thus enabling farmers to overcome the initial larger capital requirements. Earlier studies of CF uptake and spread in Zambia reflect this commercial farming orientation; there is a strong focus on yield gains, income gains, returns to land and labour, labour productivity, margins, etc. (Langmead, 2006; Haggblade and Tembo, 2003a, 2003b). The stress on precision input application in the CF package promoted by CFU may be understood similarly, as an attempt to reduce input costs. In the 2000s, CA uptake in Zambia has become increasingly incentivized by means of direct support to farmers. In large CA projects funded by the FAO and the Norwegian government such support has ranged from providing Faidherbia albida seedlings and CA farm implements (Aune et al., 2012), to fertilizer and hybrid seeds (Ndiyoi et al., 2012; FAO, 2011b; FAO-OED, 2012; see also Section 3.3.4).

2.1.3. Malawi: the imperative of land use intensification While the development of CA for African smallholders in southern Africa has its roots in Zimbabwe and was first extended at scale in Zambia, its emergence in Malawi appears to have been largely disconnected from these developments. Characterized by far higher rural population densities than Zimbabwe and Zambia3 and, in many areas, very small land holdings (Ellis et al., 2003), the need for land use intensification strategies, such as CA, is self-evident in Malawi. Low livestock densities in many areas, which facilitate crop residue retention as mulch instead of use as fodder (CA principle 2), may seem to predispose Malawian smallholder farmers towards CA-based land use intensification. Nevertheless, it is important to understand the emergence of CA for Malawian smallholders in relation to wider socio-economic developments, such as the deregulation and liberalization of the

3 Population densities in Malawi, Zimbabwe and Zambia in 2010 were 158, 32 and 17 persons/km2 (http://data.worldbank.org/ respectively: indicator/EN.POP.DNST, accessed 1 January 2013) In large parts of Zambia (as well as parts of northern Malawi), forms of shifting cultivation continue to be practiced, whereas in Malawi and Zimbabwe such practices have become rare.


country’s agriculture-dominated economy and the accompanying high poverty levels and recurrent food shortages of the 1990s.4 Trying to reduce the need for expensive maize imports and food aid to rural populations lacking the cash to either invest in food production or to buy food, donors such as DfID assisted the Malawian government with non-market interventions aimed at increasing food security and alleviating poverty. The ‘Starter Pack’ and ‘Targeted Input’ Programmes (TIP) (1998–2004) both revolved around the distribution of free inputs (Levy, 2005). These programmes ‘implied a recognition that market forces alone cannot increase poor farmers’ maize production, as input constraints are binding’ (van Donge et al., 2001, p. 15). It is therefore not surprising that the first large-scale project promoting CA practices to smallholders in Malawi, Sasakawa Global 2000, offered smallholders at a nominal price, an input package similar to the Malawi government’s starter packs, including hybrid or OPV maize, NPK fertilizer and urea. Farmers were to buy herbicides themselves (Ito et al., 2007). Incentives in the form of input packages, credit or subsidies have since become a salient feature of CA promotion projects in Malawi, influencing not only uptake, but raising questions as to the sustainability of such uptake. As in Zimbabwe and Zambia, it was the persistent concern over land degradation among intervening agencies that raised interest in CA for smallholders in Malawi. Yet, as in the other countries, such concern did not trigger the large-scale CA promotion in the region in the late 1990s and early 2000s. Nor did increasing land scarcity and recurrent food shortages provide sufficient incentives for smallholders to intensify their land use using CA (e.g. Erenstein, 2006; Baudron et al., 2012). Wider processes of socio-economic decline, as well as the withdrawal of government support to smallholder production and marketing (following the implementation of structural adjustment policies) appear to be central for our understanding of productivity decline in smallholder agriculture in southern Africa. Consecutively affecting Zambia, Malawi and Zimbabwe, increased rural poverty and food insecurity opened the door for agricultural interventions aimed at reversing the trend of declining smallholder production levels. Such interventions built on a renewed attention for the role of agriculture in development (WorldBank, 2008), and global trends in development policy stressing poverty reduction—the Millennium Development Goals (UNDP, 2012). Thus, we may understand how narratives used by CA proponents in southern Africa have come to stress the production and productivity benefits of CA adoption, rather than the cost and resource saving characteristics of CA that motivated farmers elsewhere in the world to restructure their farming operations. Although the development of input support programmes in Malawi and Zimbabwe evidences an acknowledgement of the adverse socio-economic conditions impacting smallholder farming, proponents of CA in southern Africa have not focused much on such wider forces structuring smallholder farming practices. Even where CA has been promoted as an add-on of humanitarian programmes, its proponents have tended to situate the cause of low and declining smallholder productivity at the plot level, blaming ‘inappropriate farming practices’ for causing soil degradation. This ‘narrowing-down’ of the causes of soil degradation to inappropriate farming practices may make CA a logical, and for some the only appropriate, response:

4 Both deregulation and liberalization were, at the time, ‘considered to be propoor, as regulation and state control were believed to protect monopoly interests. The question of whether liberalization. . . benefited smallholder farmers, especially the poorer ones, is a key theme in the debate on agricultural policy making’ (van Donge et al., 2001, p. 2).

119

Zambian Conservation Farming Unit (CFU): (CFU, 2003; 2007a, 2007b) Improved Reduced Tillage (IRT) involves (1) permanent, hand-hoed planting basins or ripped rows (with an ox-drawn ripper), established before the rains; (2) early planting of crops; (3) early weeding. Conservation tillage (CT) is IRT plus retention of crop residues. (>30%, see: Baudron et al, 2007:5) Conservation Farming (CF) is CT plus crop rotation with a minimum of 30% legumes in the system. (CF thus encompasses all three CA principles) Conservation Agriculture (CA) is superimposed onto the accepted CF system and introduces the establishment of cassava, Faidherbia albida, and fruit trees, in order to addr ess issues of food security and external reliance on inputs. In addition, farmers are encouraged to establish Jatropha curacas as a live fence, in order to demarcate their land, ke ep out unwanted livestock, and ready themselves for… biofuel markets.’ (Aagaard, 2010; www.conservationagriculture.org/cfu, accessed 4/5/2011). Zimbabwe CA Task Force (ZCATF): Conservation Farming (CF) ‘refers to the particular practice of using planting basins and soil cover’ (ZCATF, 2009:3) (CF does not encompasses all three CA principles, as crop rotation is not mentioned) Conservation Agriculture (CA) ‘is a broader term that encompasses activities such as minimum tillage and zero tillage, tractor powered, animal powered and manual methods, integrated pest management, integrated soil and water management, and includes CF’ (ZCATF, 2008:3). Precision Conservation Agriculture (PCA) is a term used by ICRISAT to refer to the hand-hoe, basinsbased package as promoted through humanitarian relief and recovery programmes in Zimbabwe. It stresses the precision application of small doses of nitrogen fertilizer, but unlike: 3), But unlike CF, it does not take permanent soil cover as a requirement (and is therefore not always considered as CA) (Twomlow et al 2008a: 41; 2008b). National CA Taskforce Secretariat (NCATFS), Malawi: Conservation Agriculture is ‘adoption of all the three key principles of the technology and adopting a minimum figure of 30% as the requirement for permanent organic soil cover’ (MAFS 2012: 15)

Fig. 1. Definitions of Conservation Agriculture concepts used in southern Africa.

In Malawi, soil degradation threatens the attainment of household food sufficiency for smallholder farmers. The natural approach to this is to reduce tillage and adopt technologies that promote maximum cover and control weeds in ways that comply with CA (IFAD, 2011, p. 19) (Italics by authors). 2.2. CA packaging: Different circumstances, different definitions, different practices With the reframing of CA from a resource saving to a production and productivity increasing concept, it is perhaps not surprising that what is taken as CA is not uniform either. For instance, the Zambian CFU uses the term Conservation Farming (CF) to refer to different practices that befit the three CA principles, and has started to associate CA with agroforestry and reduced reliance on external inputs. The Zimbabwean CA Taskforce (ZCATF) on the other hand, uses the term CF to refer to a particular practice–the manual digging of permanent basins in which crops are planted5 –in combination with CA principle 2, permanent soil cover. It defines CA much broader than the three main CA principles as defined by FAO. In Malawi, CF is not a commonly used term (see Fig. 1). The country-specific overviews below elaborate the different CA definitions and associated practices used in southern Africa, and highlight the different socio-economic and agronomic circumstances in which they have emerged. The discussion starts with Zambia, as the large-scale promotion and uptake of CA for smallholders farmers started in this country. 2.2.1. Zambia: Towards low external input dependence Large-scale CA promotion to smallholder farmers in southern Africa started in Zambia, where a hand hoe, planting basin-based

5 The spacing and dimensions of the planting basins promoted in Zambia and Zimbabwe differ. In Zambia’s agro-ecological zones I and II (see note, permanent basins are recommended as wide as a hoe blade, 20 cm deep (to break plough pans), 30 cm long and dug every 70 cm in rows that are 90 cm apart – resulting in 15,850 basins/ha (CFU, 2012). In Zimbabwe, recommended basin dimensions are 15 cm × 15 cm × 15 cm, spaced 60–75 cm within rows that are 75–90 cm apart – resulting in 17,777–22,222 basins/ha). Spacing depends on the type of crop, rainfall (natural region), and use of draught animals for weeding (ZCATF, 2009, p. 37)

120


system known as Conservation Farming (CF) was promoted from the mid 1990s onwards. Introduced by the newly formed Conservation Farming Unit (CFU) and the Golden Valley Agricultural Research Trust (GART)6 , CF built to a large degree on the experiences of Brian Oldreive, a Zimbabwean farm manager who had experimented with minimum tillage technology in the 1980s (Langmead, 2006; Aagaard, 2012; see also: Oldreive, 1993; 2013). The CF package was especially promoted to smallholders in the moderate to low rainfall agro-ecological zones I and IIa where cotton is often an important cash crop.7 The focus on low to moderate rainfall areas has shaped the definition of CF in Zambia and the practices promoted under that that banner at least in three ways. First, dry-season land preparation became regarded as a key practice, because late planting is often found to have a negative effect on yields (Aagaard, 2010; Tittonell et al., 2008). The locally developed, ox-drawn, Magoye ripper fits this stress, enabling dry-season preparation of large land areas. Yet, it has been suggested that dry season land preparation may not be the most attractive aspect of CF; hard soils prevent many farmers from basin digging and weak animals make ripping during the dry season difficult. Postponing these activities to the onset of the rains, when the soil is less hard, takes away ‘the expected benefits associated with early planting and water harvesting’ (Baudron et al., 2007, p. 17). Second, the focus on low to moderate rainfall areas has resulted in the use of larger planting basins – as a water harvesting technology – in Zambia as compared to Zimbabwe, where the technology was introduced from (Mazvimavi, 2011, p. 13; see also note 5). It has been argued that this water-harvesting technology is inappropriate for Zambia’s high rainfall cassava-dominated areas of Zone III where mounds, rather than basins, are needed (IFAD, 2011, p. 26).8 Third, soil degradation in the form of hard pans, understood as resulting from years of ploughing, resulted in the adaptation of Brian Oldrieve’s planting basin technology; in Zambia deeper planting basins, sufficient to break plough or hoe pans are recommended (Chikowo, 2011, p. 10; CFU-Zambia, 2012). The centrality of the CFU in CA promotion in Zambia has resulted in relatively well-defined, diversified CF packages for both hand hoe farmers and farmers using animal draught power (CFU, 2007a, 2007b). The CFU appears to have adopted an incremental approach to CA adoption, moving from improved reduced tillage (IRT), via conservation tillage (CT) and CF, to CA, in which the latter is more broadly defined as a low-external input form of CA. 2.2.2. Zimbabwe: Efficient use of humanitarian input support The distinct socio-economic and institutional context in the early 2000s, gave rise to a different emphasis, definition and targeting of CA in Zimbabwe. Here CA promotion was latched on to a coordinated humanitarian relief effort by international donors

6 CFU and GART continue to play the leading roles in CA extension to smallholders. Major donors supporting their work include the World Bank, Canadian International Development Agency, European Union and the governments of Norway and Finland. Additionally, a number of foreign NGOs, have provided support (Umar et al., 2011; Mazvimavi, 2011). Other actors promoting CA to smallholders in Zambia have included the Government of Zambia’s Ministry of Agriculture and Cooperatives (MACO), FAO, IFAD and USAID (partnering with Catholic Relief Services, CARE, World Vision and Land o’Lakes). 7 Zambia has four agroecological zones, I, IIa, IIb, and III. Zone I, in the southernmost part of the country, is a low rainfall region, receiving less than 800 mm of rain annually, with low agricultural potential soils. Zone IIa receives between 800 and 1000 mm of annual rainfall and has high agricultural potential soils (SiacinjiMusiwa, 1999). Over 95% of Zambia’s cotton farmers operate in Zone IIa because of the good soils, sufficient amounts of rain and proximity to markets (Haggblade et al., 2011). 8 Associated with the Sahelian Zaï pits, from which they are said to be derived, Zambian planting basins are smaller and deeper than Zaï pits, and their position along the slope is usually undefined (Mando et al., 2006; Chikowo, 2011, p. 6; Twomlow et al., 2008a,b).

that provided maize seed and fertilizer support to rural households hit hard by Zimbabwe’s deep economic crisis (Twomlow et al., 2008b). The focus on vulnerable households and CA promotion within the context of humanitarian relief has structured both the definition and practices promoted. Unlike Zambia, where Brian Oldreive’s hand-hoe planting basins constituted one possible practice for minimal soil disturbance (CA principle 1) in the CF package extended to commercially-oriented smallholder farmers, in Zimbabwe CF became strongly associated with the planting basins technology and subsistence-oriented production. Here, CF obtained meaning as a practice enabling resource-poor, vulnerable households to plant their food crops early. Planting basins thus increased the returns on donor-provided input packages. In many instances, NGOs made input support conditional on observable basins in farmers’ fields. Like in Zambia, CF promotion was not a government initiative. Numerous international and local NGOs implemented the combined CF/humanitarian relief projects of the Protacted Relief Programme, a multi-donor initiative to assist crisis-struck Zimbabwe (Andersson and Giller, 2012, p. 32–34). CA promotion targeting mechanized and commercially oriented smallholders was developed in the shadow of this humanitarian aid programme, but mechanized CA has played a minor role in the claimed success of CA in Zimbabwe (Marongwe et al., 2011). Shaped by the humanitarian relief context in which it was promoted, the distinct definition of CF in Zimbabwe also emphasizes ‘good management’, which refers to timely implementation, precise operations, and the efficient use of inputs (Fig. 2). This emphasis is also reflected in the concept of Precision Conservation Agriculture (PCA) (Twomlow et al., 2008a; Figs. 1 and 2). Again, this stress on good or precise management can be understood in the context of the country-wide humanitarian relief effort and the agro-ecological circumstances prevailing in Zimbabwe’s Communal Areas (formerly known as Native Reserves). First, input packages (seed and fertilizers) were sufficient for 0.2 ha only, which allowed for precise operations. Second, on the generally poor soils of Zimbabwe’s Communal Areas, concentrating effort and resources on a small piece of land may appear a sensible strategy. In situations of crop residue scarcity and competing uses for such residues as prevailing in Zimbabwe’s Communal Areas, concentrating such scarce resources is also a necessity for CA implementation. Yet, it needs to be noted that while the particular CF package of input support, planting basins and precise management sought to intensify land use of Zimbabwe’s rural poor, smallholder farmers in southern Africa often extensify their land use in the face of input scarcity and the risk of crop failure (Baudron et al., 2012). 2.2.3. Malawi Despite government involvement since the 1990s, NGOs and private sector companies have often taken the lead in CA promotion in Malawi (Mloza-Banda and Nanthambwe, 2010). Introduced in an era of input support to smallholder farmers, CA has also often been promoted as one element in larger agricultural development programmes. The Ministry of Agriculture and Food Security (MAFS), for example, implemented a Farm Income Diversification Programme promoting practices such as composting and contour ridge realignment, in addition to minimum tillage, crop residue mulching and intercropping. Similarly, Sasakawa Global 2000 promoted new spacing techniques in addition to common CA tenets (Ito et al., 2007). It is therefore perhaps not surprising that among CA promoting organizations in Malawi, the understanding of CA as consisting of three principles was found to be limited (MAFS, 2012, p. 10). As in Zambia and Zimbabwe, Malawi’s specific socio-economic and agronomic context has shaped the definition of CA and the specific practices promoted. While different kind of resource conserving practices may be associated with CA, other important


121

Fig. 2. Contexts and key practices promoted in Zambia, Zimbabwe and Malawi (Abbreviations of CA concepts explained in Fig. 1, * basin dimensions differ in Zambia and Zimbabwe; see note 5).

features of CA promotion in Malawi include: herbicide use, a focus on manual minimum tillage technologies, a relative lack of focus on crop rotation, and the add-on of agro-forestry technologies. First, unlike the Zimbabwean CA-based relief programmes, nearly threequarters of CA promotion projects in Malawi promote the use of labour-saving herbicides, despite their high cost and limited availability (MAFS, 2012). The role of partnering herbicide companies in projects needs to be considered here (Ito et al., 2007). Second, because of the limited number of draught animals in many parts of Malawi, ripping is rarely promoted. Instead, minimal soil disturbance is achieved by encouraging farmers to abandon the hand hoe made, and annually split, planting ridges introduced in colonial days (Green, 2009; McCracken, 2012). As an alternative, planting on the flat, or on permanent ridges using a dibble stick are promoted (Ngwira et al., 2012a, p. 2). Besides dibble stick planting, also planting basins are promoted in Malawian CA projects, albeit not usually as a water harvesting technology, but as a way to maximize the impact on yields of limited manure and compost supplies (MAFS, 2012, p. 22). Low livestock numbers in many areas (FAO, 2005), are seen as necessitating the precise application of such organic inputs. Third, because of the extremely small landholdings in many parts of the country (Ellis et al., 2003) and the importance people attribute to producing one’s own maize (van Donge, 2005), smallholders are seen as disinclined to adopt crop rotations. Consequently, CA promotion in Malawi usually stresses inter-cropping of maize and a legume (MAFS, 2012). This practice is already common in southern Malawi where farmers intercrop maize with pigeon pea (Cajanus cajun). A fourth practice often promoted in combination with CA in Malawi is agroforestry (Bunderson et al., 2009). A document of the World Agroforestry Centre (ICRAF) even suggests that agroforestry is more promoted as a CA practice than is crop rotation (Sosola et al., 2011, p. 2). Commonly promoted tree species include Tephrosia vogelli, Gliricidia sepium and Faidherbia albida, as well as the grain legume, Cajanus cajan (MAFS, 2012). While the soil fertility enhancing and pest-repelling properties of the tree species have

been widely established in the literature, competition between trees and food crops may be an important barrier to the adoption of agroforestry on the small land holdings of Malawian farmers. Widespread adoption of agroforestry in Malawi is contested (de Wolf, 2010). To conclude, what constitutes and is promoted as CA is diverse and not merely agronomically or agro-ecologically defined. The diverse socio-economic and institutional contexts of smallholder farming in southern Africa, have also shaped different definitions of what constitutes CA beyond the three principles as defined by FAO. As CA became increasingly important as a brand, to be differentiated from conventional tillage-based agriculture, creating meaning with this brand within a donor-funded agricultural development community has probably further increased the range of practices associated with it (cf. Sumberg et al., 2012). As a consequence, alternative plant spacing, fertilizer micro-dosing, agroforestry, etc. are now associated with, and/or promoted under the banner of CA. Thus, there is a tendency of CA becoming a ‘container concept’, in which all kind of practices can be sold to donors as CA interventions (Andersson and Giller, 2012, p. 35). 2.3. Barriers to CA adoption in southern African smallholder agriculture As an increasing set of practices has become associated with, and promoted as CA in southern Africa, assessing the literature on CA adoption and barriers to adoption becomes a difficult task. Such an assessment is further complicated by two additional factors. First, the integrated nature of the CA concept – the three CA principles are generally seen as highly interdependent – raises questions regarding the meaning and benefits of partial adoption. Second, contextual factors such as input support, subsidies, agricultural policies, and markets often shape the adoptability of new technologies and practices by farmers, including CA. Before discussing in more detail the findings regarding adoption figures in

122


the literature and the way in which CA adoption studies have been conducted (in section 3), this section therefore briefly explores the identified barriers to adoption that have caused CA adoption in southern Africa to be considered (s)low (Derpsch et al., 2010, p. 15). For our understanding of the factors influencing CA adoption, it may be useful, following Sumberg (2005), to distinguish adoption constraints from prerequisite conditions; the former referring to the ‘goodness-of-fit’ between the innovation and the potential users (innovation x potential users), while the latter focuses on contextual factors that cannot be influenced by the innovationdevelopment process. Thus, constraints recurring in the literature, such as limited availability and competing uses for crop residues, weed pressure (Umar et al., 2012, p. 923; Marongwe et al., 2011, p. 156; Aune et al., 2012), capital requirements for additional fertilizer, herbicides, implements (hoes, rippers, sprayers) and, in some situations, labour requirements (Baudron et al., 2012; Mazvimavi, 2011), fall into the first category. Factors such as relative land abundance, communal tenure arrangements (Baudron et al., 2012), absent or dysfunctional markets for legumes (Thierfelder et al., 2013a) and limited access to financial capital (see Wall, 2007), relate to the prerequisites for adoption. The sections below review respectively, commonly cited adoption constraints and prerequisite conditions in the CA literature on southern Africa. They make clear that barriers to CA adoption identified in the literature tend to be related to specific CA practices (the innovation) and agro-ecological circumstances, rather than to the type of farmers (potential users) or socio-economic circumstances in which these practices are to be adopted. 2.3.1. Limited availability and competing uses for crop residues Numerous reports and studies have pointed to the problems of crop residue retention and the trade-offs between different uses in crop-livestock farming systems in southern Africa (Chivenge et al., 2007; Giller et al., 2011; Hove, 2011; Wall, 2007; Umar, 2012; Hove, 2011; Rusinamhodzi et al., 2013). Since biomass production in many parts of southern Africa is relatively low, cattle keepers ‘rely on crop residues to feed their livestock during the long dry seasons, implying substantial opportunity costs to their use as mulch’ (Valbuena et al., 2012). But even where farmers are able to make additional investments in fencing or livestock repellents to retain crop residues in the field (Mutsamba et al., 2012), crop production levels in these maize-dominated systems may still be insufficient to attain the required soil cover of 30% as outlined in many CA definitions and manuals (Giller et al., 2009). With grain yields generally ranging between 0.5 and 2.0 t/ha, maize productivity in southern Africa often falls short of generating the stover required to attain the required level of soil cover (CA principle 2). Analyses of this adoption constraint in southern Africa do not usually establish for what type (or proportion) of farmers the required soil cover is attainable, or what the possibilities and consequences are of alternative uses of crop residues for these farmers. 2.3.2. Basins and weeds: labour constraints to adoption It is generally acknowledged that reduced tillage reduces labour needs for land preparation. This is particularly true for areas in Zambia and Malawi where hand hoe made ridges are a common practice. Yet, it has also been argued that the area of land prepared with planting basins has remained stagnant (ca. 0.2 ha per farm) mainly due to labour constraints (Giller et al., 2011, p. 470). Higher weed pressure under CA increases peak labour demand at weeding, especially where no herbicides are used (Baudron et al., 2012), but the limited CA area of 0.2 ha may also relate to the size of the input packages provided by NGO’s promoting CA among vulnerable households, notably in Zimbabwe. Studies focusing on either input support beneficiaries (that is, the potential users) and/or farmers’ CA area in relation to their total cultivated land area may establish

for which farmers and under what agro-ecological circumstances labour constraints and/or soil fertility investment constraints limit the expansion of the CA area beyond 0.2 ha. Weed problems, which may prevent expansion of the CA area, may be further aggravated by insufficient crop residue levels, as common in drier areas and crop-livestock systems. On the other hand, high levels of crop residues can reduce weed growth. But where weeds are persistent, crop residues may also ‘intercept 15–80% of the applied herbicides and this may result in reduced efficacy of herbicides in CA systems’ (Chauhan et al., 2012, p. 62), thus requiring higher levels of capital demanding herbicide use. In addition, the contextual factor of herbicide availability limits its use. For instance, in Zimbabwe, very few smallholders have access to herbicides, its use being mostly confined to CA demonstration plots. In Zambia and Malawi, herbicide use generally depends on input support or contract farming arrangements (Chikowo, 2011, p. 17). For which farmers investment in herbicides makes economic sense and what type of farmers will actually invest in herbicides therefore largely remain hypothetical questions as prerequisite conditions for adoption are not met. 2.3.3. The performance of CA Studies of basin-based CF in both Zambia and Zimbabwe, have pointed to significant increases in maize (and cotton) yields in comparison with tillage based agriculture, due to ‘earlier sowing in CA, water harvesting effects, improved infiltration of water in the basins, and more efficient use of nutrients as the nutrients are concentrated adjacent to the plants’ (Aune et al., 2012, p. 10; Umar et al., 2011). Yet, a two-year study in semi-arid southern Zimbabwe (Natural region IV; annual rainfall 450–650 mm) reports less clear yield advantages; ‘On both soil types [clay and sand], neither mulching nor tillage method had a significant effect on maize grain yield, irrespective of rainfall received’ (Mupangwa et al., 2007, p. 1133). A meta analysis of the long term effects of CA on maize yields revealed that with crop rotations, soil cover and high input use, maize yields under CA generally increase over time in low rainfall areas (Rusinamhodzi et al., 2011). Yet, in years of high rainfall or in high rainfall areas, mulch cover may lead to lower yields due to water logging (Chikowo, 2011; Rusinamhodzi et al., 2011). Benefits of adoption are thus season and agro-ecology specific and dependent on sustained investments. Given the high inter-annual rainfall fluctuations in southern Africa, yield stability thus becomes an important criteria for assessing CA’s suitability for different types of farmers. But as only wealthier farmers or farmers with off-farm income sources are able to generate and sustain, without external assistance, the investments that are required for long-term yield growth, the promotion of CA as a production growth strategy for the poor may be questioned. 2.3.4. Returns on investment and costs Higher land productivity (yield) is a major factor in the profitability of CA in southern Africa (Ito et al., 2007; Mazvimavi, 2011), although some argue that area expansion, made possible by reducing labour bottlenecks, also contributes to higher farm output (Haggblade et al., 2011, p. 2). Net margins and returns to labour (per day) are found to increase when minimum tillage is adopted, not only in machine-based CA systems, but also in animal draft power and manual CA systems. For instance, in Zambia it was found that the planting basin based CF system out-performs conventional ploughing of the whole field by hand hoe or ox-drawn plough, as well as ripping (Aune et al., 2012, p. 12-13). In Zimbabwe, where manual ploughing of whole fields is far less common, economic analyses also show higher returns under (partial) CA (Mazvimavi and Twomlow, 2009, p. 27; Mazvimavi, 2011). Besides returns, also costs tend to be higher in CA systems as new tools (rippers, sprayers, etc.), additional fertilizer and more


labour may be required. Studies identifying which farmers may be able to make such investments and at what scale they need to operate in order to benefit from such additional investments, are scarce. In Zambia and Zimbabwe, labour costs were found to be higher in basins based CF (Aune et al., 2012, p. 12; Mazvimavi, 2011, p. 6). Such costs are likely to inhibit the expansion of the CA area beyond what can be planted and weeded by (unpaid) family labour, and may explain why growing numbers of farmers take up CA components while the ‘CA area’ per farmer remains stagnant (see also above). In Malawi on the other hand, labour costs are reported to decrease under CA. Not only are planting basins less commonly used here, annual ridge making is abandoned under CA, and the use of (subsidized) herbicides further reduces labour costs (Mloza-Banda et al., 2011, p. 34). 2.3.5. Adoption constraints and pre-requisite conditions: Crop rotation Crop rotation with legumes has been promoted and acknowledged by both researchers and farmers as a pest and disease control and soil fertility enhancing technology in southern Africa for many decades (Alvord, 1958; Giller, 2001). Nevertheless, in many southern African smallholder farming systems rotation with legumes, both in conventional and CA systems, has remained limited to less than 1/3 of the cultivated area. Contextual factors are usually identified as the main explanation, notably dysfunctional markets and the unavailability of seeds (Mazvimavi and Twomlow, 2009, p. 28). Where legume markets do function, as appears to be the case in central Mozambique, maize-pigeon pea intercropping may provide not only legumes, but simultaneously yield benefits for maize (Rusinamhodzi et al., 2012). Besides prerequisite conditions beyond the farm, farm-level constraints also limit crop rotation and the area smallholders dedicate to legume production. First, farmers may not grow a large variety of crops. Second, farmers may have access to different types of soils that differ in their suitability for different crops (Baudron et al., 2012, p. 47). Third, higher labour demands and a preference for staple food crop production among farmers with very small land areas may cause legume areas to be small (Thierfelder et al., 2013a; Giller et al., 2011). In planting basin based CA systems, differences in plant spacing further constrain legume production, while harvesting of legumes like groundnuts ‘make it difficult to avoid soil disturbance as groundnuts have to be pulled out of the soil, which will compromise the first principle of ‘minimum soil disturbance’ to a certain extent’ (Thierfelder et al., 2013a, p. 15). Legume production is also likely to compromise CA principle 2 (permanent soil cover) as legume residues are often preferred animal feed or, when retained, disaggregate very quickly. 2.3.6. Mindset of the plough In the literature on CA, mindset is another oft mentioned constraint to, and pre-requisite condition for CA adoption. Regularly mentioned as a most important change necessary to adopt the minimum tillage component of CA, the need for mindset change not only refers to farmers, but also to extension agents and researchers (Derpsch et al., 2010; Wagstaff and Harty, 2010, p. 71; Thiombiano and Meshack, 2009, p. 18; Erenstein et al., 2008, p. 269). General observations and farmer testimonies are used to suggest that tradition and prejudice are pervasive and that concerted support at all levels is necessary to overcome the mindset of ploughing as a necessity (CIMMYT, 2012, p. 4; FAO, 2008, p. 11; Hobbs et al., 2008, p. 551; Vaneph and Benites, 2001, p. 14). Yet, there are no systematic analyses and the literature seems equivocal about this adoption barrier, as it is also suggested that, unlike researchers and extension workers, ‘Farmers generally undergo this change in mindset relatively quickly when they experience, or are exposed to, the benefits of CA’ (Wall, 2007, p. 142). This suggests that the

123

‘mindset of the plough’ is foremost a problem of supporting agencies and researchers rather than of smallholder farmers themselves. Yet, as all three countries under study have organizations and government policies promoting CA to smallholders the importance of this factor in limiting the promotion of CA in southern Africa may be questioned. Besides limited evidence of an unsupportive institutional context for ‘plough-less’ agriculture in southern Africa, a psychological reductionism seems inherent to the notion of ‘mindset of the plough’. This is problematic as it causes historical and contextual factors to be ignored. The common association of the ‘mindset of the plough’ with traditionalism, disregards the fact that animaldrawn ploughs were only introduced to smallholder farmers in southern Africa about a century ago, and that their uptake among ‘traditional’ smallholder farmers was often fast. For instance, in colonial Zimbabwe the number of ploughs used by smallholders grew exponentially during the first part of the 20th century (Palmer, 1977, p. 201). To understand such rapid adoption, historians have pointed to the importance of contextual factors such as low population densities and the abundance of relatively fertile land in many areas, expanding markets, and good prices for agricultural produce (Phimister, 1988). Such contextual factors appear still relevant for our understanding of smallholder farmers’ preference for ploughbased agriculture today. For instance, Baudron et al. (2012, p. 405) found that in Zimbabwe’s Zambezi Valley, where land is still relatively abundant and fertile, the plough continues to be considered the hallmark of good farming as it enables extensification-based production growth. Nevertheless, in the literature on CA in southern Africa the ‘mindset of the plough’ is not commonly linked to such contextual factors of adoption, nor systematically analysed.

2.3.7. Institutional and policy issues: prices and maize subsidies Above it was elaborated how the large scale promotion of CA to smallholders in Zambia, Zimbabwe and Malawi emerged in the context of production crises and increasing food insecurity in these countries. The need to increase smallholder production not only shaped CA projects, definitions and adoption, but also other agricultural policies. Over the past decade, input support and fertilizer subsidy programmes have become a salient feature of agricultural policy in southern Africa and the wider context in which CA promotion takes shape. Such policies may not always be aligned to CA promotional efforts. For instance, in Zambia, large-scale CA promotion had already started when a Fertilizer Support Programme (FSP) targeting maize production was introduced in 2002. Providing subsidized fertilizers to mono-crop maize as well as guaranteed, pan-territorial maize prices, it has been suggested these policies undermine the legume rotation as promoted under CA (Umar, 2012, p. 81). In Malawi, it was found that the inclusion of legumes in the Starter Pack was modestly successful (Barahona and Cromwell, 2005, p. 166), while its hailed fertilizer subsidy programme that followed it, distributed mostly vouchers for fertilizer and maize seeds and has impacted mostly maize production (Dorward et al., 2008). Input packages provided in Zimbabwe as part of the humanitarian aid effort can equally be seen as oriented towards maize production. As the above discussion of barriers to CA adoption already suggests, the understanding of constraints and prerequisite conditions require different kind of analyses. Whereas the former may be addressed in farm or household-level adoption surveys, the latter call for wider socio-institutional, market and policy analyses. In absence of the latter type of analyses, the few studies that do systematically analyse CA adoption in southern Africa, are of the survey-based, farm-level analysis type. These will be discussed in the following sections.

124


Table 1 National level estimates of adoption of CA or CA components, Zambia. Source

Estimated no. of adopters/hectares (ha)

Method for estimating adoption

Thiombiano and Meshack(2009, p. 14) Derpsch et al. (2010)

100,000 smallholders 110,000 ha 40,000 ha

No details

Baudron et al. (2007)

78,000 hand hoe farmers (2002/3 season)

Aagaard (2010, p. 5) (CFU estimates) Friedrich et al. (2011)

160,000-180,000 smallholder households 4% of farmland (2009/10 season) 270,000 smallholders (2006/7 season) 20,000-60,000 hand hoe + 4000 ox-drawn rippers (2001/2 season) 200,000 (In 2011) approx. 5.8% of cultivated area

Aagaard (2009) in Simasiku et al. (2010) Umar et al. (2011)

FAO AQUASTAT Accessed 23/6/2013

Definition of adoption

No details. Based on data from ‘a network coordinated by FAO with qualified informants’. No details. Citing CFU estimates

no-tillage

No details

‘Basic forms of CF on portions of their land.’ ‘Some form of CA.’

No details

35,000 had adopted IRT; 25,000 had adopted CT; 18,000 had adopted CF (for definitions, see Figure 1)

Based on a baseline study of 93,000 smallholders by CFU No details. Cites Haggblade and Tembo (2003b).

CF on some portion of land

No details

Disturbed area 30%. ‘Rotation is not a requirement for CA at this time.’

3. Findings: CA adoption figures If defining what constitutes and is promoted as CA in the diverse smallholder farming contexts in southern Africa appears to be difficult, so does delineating what is considered ‘CA adoption’ in adoption figures. Considerable variation exists in adoption measurements and estimates. For instance, Simasiku et al. (2010, p. 14) estimate CA adoption in Zambia at 270,000 smallholder farmers (in 2010?) who practice CA ‘on portions of their land.’ A 2010 CFU document, states that only 160,000–180,000 families apply ‘the basic forms of CF on portions of their land’ (Aagaard, 2010, p. 5). Baudron et al. (2007), drawing on CFU data from 2003, estimate that only around 25% of the smallholders who have adopted CF in Zambia have adopted the full CF package. Nyanga’s study of 415 randomly selected beneficiaries of the 120,000 smallholders targeted by CFU’s CA Project (CAP) in Zambia considered ‘a household that had area under minimum tillage. . . as practicing CA in the 2009/2010 farming season’ (Nyanga, 2012a, p. 30). It found that 71% of the project beneficiaries had adopted minimum tillage on 25% (approx. 0.8 ha) of their cropped land.9 Virtually no farmers in her sample had adopted mulching with crop residues but the use of crop rotation had increased since the introduction of CA (Nyanga, 2012a, 2012b). The FAO’s Aquastat database, on the other hand, adopts an area-approach, taking the extent of CA adoption as areas with >30% soil cover and minimum soil disturbance as less than 25% the cropped area. It appears to disregard the time dimension of adoption altogether as rotation is not yet seen as a requirement. 10

These examples not only illustrate that estimates vary considerably, but also that what CA adoption entails in practice, is defined differently or remains opaque, and needs to consider what is already practised (for instance, crop rotation). Obviously, illdefined notions of what constitutes ‘CA’ and ‘adoption’ compromise

9 For farmers who engaged exclusively in hand hoe farming, digging planting basins, the average area under minimum tillage was 0.5 ha (Nyanga, 2012a, p. 34). 10 See: http://www.fao.org/nr/water/aquastat/data/popups/itemDefn.html?id= 4454, accessed 22 June 2013.

Disaggregated data

Hand hoe farmers and farmers using ox-drawn rippers

aggregation and comparisons of CA adoption figures as envisaged with the FAO’s Aquastat database. Furthermore, if it is recognized that CA comprises of different practices that are suitable to different types of farmers and environments, one would expect farm household typologies or environmental characterisations to feature prominently in adoption estimates, yet they do so only to a limited degree (see Tables 1–3).

3.1. CA adoption estimates and project bias Data collection on CA adoption usually takes place within the context of on-going agricultural development projects promoting CA. This has three important consequences. First, a numerical or ‘accountancy’ approach to CA adoption predominates. With donor demands for development impact at scale, the success of CA projects has predominantly become measured in terms of the number of ‘CA adopters’, with scant attention being paid to other dimensions of adoption, such as duration and the area on which it is practiced (see: Nyanga, 2012, p. 30). Second, research institutes that collect adoption data are often simultaneously involved in the promotion of CA within the same projects. Consequently, adoption figures and studies may be biased towards adopters and project beneficiaries (see below). Third, CA promoting projects often provide incentives, notably free or subsidized fertilizers, seeds, and herbicides to farmers. This raises questions regarding the nature of the adoption claimed, such as whether CA uptake within projects can be taken as adoption or whether true adoption can only be assessed much later, after a project has ended (Andersson and Giller, 2012). Adoption figures thus need to be treated cautiously. National level estimates of CA adoption in southern Africa suffer from the same ills. As Tables 1–3 illustrate for respectively Zambia, Zimbabwe and Malawi, such estimates are fundamentally uninformative as adoption is generally ill-defined, methods for estimating adoption are usually not elaborated, and the data is generally neither disaggregated by socio-economic status or production orientation of adopting farmers, nor by agro-ecological zone. In addition, it is unclear if the national estimates are indicative of long-term adoption.


125

Table 2 National level estimates of adoption of CA or CA components, Zimbabwe. Source




Derpsch et al. (2010)

7500 ha under CA

No tillage

Mazvimavi (2011)

130,000 HH (2009/10 season); 40,000 HH (2008/09 season); 50,000 HH (2007/08 season) 50,000 adopters

No details. Based on data from ‘a network coordinated by FAO with qualified informants’. No details. Extracted from ZCATF data. No details provided on dynamics of adoption figures. No details.

FAO (2008? 2009?) (Accessed from FAO project server)

Mazvimavi et al. (2011)

89% (n = 401) basin planting; 56% (n = 252) applied mulch; 19% (n = 86) did crop rotation (2008/09 season)

Kassam et al. (2009)

15,000 ha (2008/09 season) (or 0.4% of cultivated area) 100,000 smallholders 15,000 ha (in 2009)

IFAD (2011) Marongwe et al. (2011, p. 156,157)

Nyagumbo et al. (2012)

100,000 farmers in 2010 (up from 2000 in 2003)


139,300 (in 2011). approx. 3.3% of cultivated area

Additional information

Extracted from Grant application to European Commission Delegation proposing the‘Promotion of Conservation Agriculture (CA) and Coordination of Agricultural Activities in Zimbabwe’ project Authors also present adoption data from previous years and suggest trends of dis-adoption of basins and of provided fertilizer.

Based on panel data of 30 households with at least 2 years of experience with CA practices in 15 districts (N = 450)a . No details. Cite FAO AQUASTAT data for 2009 basins Authors cite FAO for 2009. ‘The actual area under CA for this season (2009/10?) has not established as yet, but indications suggest that it may reach close to 100,000 ha nationally.’ Cites pers. communication with Marongwe

No details

CF

Authors suggest area is mostly farmed by resource poor HH with limited access to draft animals, each working about 0.25 ha


HH: households a The districts had CA promotional activities associated with the multi-donor Protracted Relief Programme, the European Commission Humanitarian Aid Office (ECHO) or other EU programmes.

In Zambia, where national estimates of CA adoption range from 160,000 to 270,000 smallholders and between 40,000 and 110,000 hectares (see Table 1), estimates of adoption generally refer to the uptake of the minimum tillage component only; very few estimates include crop residue retention in their definition of adoption. In Zimbabwe, an independent basis for national estimates appears to be lacking, as adoption data collection has (exclusively?) taken place in the context of humanitarian aid programmes promoting CA and providing input support in a selected (yet large) number of districts. As the overview of studies presenting national estimates shows, definitions of adoption and methods of data collection used are not explained (Table 2). In Malawi, where national level estimates also diverge substantially, the National Conservation Agriculture Taskforce Secretariat (NCATFS) indicates that in absence of standardized monitoring tools, ‘critical statistics such as land area under CA are difficult to estimate’ (MAFS, 2012, p. 14). Without common or elaborate definitions in the CA literature and adoption estimates, it is perhaps useful to distinguish at least four basic dimensions of CA adoption: (1) the number of farmers taking up; (2) the technologies or practices they take on; (3) the land areas to which these practices are applied, and; (4) the number of cropping seasons in which these are applied. The paragraphs

below discuss these dimensions, underscoring the limitations of current figures and data collection on CA adoption in southern Africa. 3.1.1. CA adoption as the number of farmers practising minimum tillage As Tables 1–3 Malawi show, such estimates are often based on the number of farmers taking on CA components, usually only minimum tillage. This rather narrow definition of CA adoption undoubtedly relates to the problems associated with crop residue retention and crop rotations in the crop-livestock and predominantly maize-based farming systems of southern Africa (Umar et al., 2011, 2012; Ndiyoi et al., 2012; Baudron et al., 2007; Marongwe, 2012; Thierfelder et al., 2013b, p. 2). At the same time, the production of such narrowly defined adoption figures is driven by the politics of donor-funded, impact-oriented development projects in which CA promotion takes place. Reducing CA adoption to the implementation of minimum tillage is not merely a convenient measure of impact in development projects, where implementing NGO’s sometimes have made the provision of inputs conditional upon the digging of planting basins–as has been reported for Zambia and Zimbabwe (Mazvimavi, 2011, p. 16; Andersson and Giller, 2012, p. 33). Such

126


Table 3 National level estimates of adoption of CA or CA components, Malawi. Source



Thiombiano and Meshack(2009, p. 14) IFAD (2011: 6)

5407 smallholders 47,000 ha

No details.

Over 54,000 groups of farmers 47,000 ha (in 2009)

No details. Refers to FAO report but does not offer cite.

Mloza-Banda and Nanthambwe (2010)a

37,594 farmers 16,028 ha (2009/10 season)

Based on extension efforts reports of Land Resources Conservation Department

MAFS (2012, p. 31)

Estimates are not yet possible


16,000 (In 2011) approx. 0.4% of cultivated area

‘Estimates of area under CA were available from most projects consulted but they did not disaggregate by different CA practices that were adopted by farmers. . .. With lack of systematically developed monitoring and evaluation tools, it is difficult to compareachievements in CA from different projects.’ No details


Additional information

‘Some form of CA’

‘Of this area, an estimated 1000 ha could be considered as being under true CA’ Area figure based on sum of: reduced tillage: 1,356 ha; use of herbicides: 1,139 ha; crop residue management: 6,387 ha; basin planting: 609 ha. Note! These figures don’t add up to 16,028


a While not specified in Mloza-Banda and Nanthambwe (2010), it is possible that the adoption figures presented are actually the number of farmers and corresponding number of hectares targeted, as opposed to adoption.

reductionism, it has been argued, undermines the fundamentals on which CA’s positive benefits rest. Minimum tillage without residue retention, ‘can be more harmful to agro-ecosystem productivity and resource quality than a continuation of conventional practices’ (Erenstein et al., 2012, p. 182; Wall, 1999, p. 316; Govaerts et al., 2005; Guto et al., 2011b, p. 93). A recent CA component omission study in multiple sites confirms such findings for southern Africa: ‘Retention of crop harvest residues increased yield in no-tillage systems [but] no-tillage without residues depressed yield by 50% when compared with yields of conventional tillage’ (Thierfelder et al., 2013a,b, p. 1) (italics by authors).

3.1.2. CA practices taken up: Partial and incremental uptake As CA is considered to be a complex set of crop management practices, adoption is often believed to be incremental in nature (Giller et al., 2009, p. 30). When farmers take on only some CA components this is generally referred to as partial uptake. Farmers may either gradually expand their area under CA, and/or adopt different CA components in a stepwise manner. Partial and incremental uptake may thus either be measured on an area basis or a time scale–with more components taken on over time–and there is no univocal use of the terms in the literature. Since the sequence in which different CA components are taken up may vary, some studies use the notion of diverse ‘adoption pathways’ (Baudron et al., 2007, p. xi). The uptake of minimum tillage–the most commonly practiced CA component in southern Africa–may thus be seen as an initial step towards full CA adoption. However, as shown above, such partial uptake may not necessarily be beneficial in terms of productivity and resource saving when soil cover is not simultaneously practiced, rendering a component-by-component incremental approach to CA adoption problematic.

Systematic study into ‘adoption pathways’ in southern Africa is probably limited to an ICRISAT panel study implemented in the context of Zimbabwe’s combined humanitarian aid and input-supported CF promotion programme, the Protracted Relief programme (PRP). Using ‘adoption intensity’ as a measure of partial uptake, the study did not find a unidirectional adoption pathway. While the use of planting basins has remained high over the years (because of the nature of CA promotion in Zimbabwe), the ‘. . .incremental uptake of the various components of the CF technology’ in earlier years (Mazvimavi and Twomlow, 2009, p. 21), was followed by a downward trend in farmers’ implementation of CF related practices such as winter weeding, soil cover maintenance, and the application of manure and fertilizers. Only the percentage of farmers practicing crop rotation on their generally small CA plots increased; from 13% of farmers in 2004/5 to 30% in 2009/10 (Mazvimavi and Nyamangara, 2012). Besides diverse trends in the adoption of CA components, implementation practices may further complicate the study of CA component uptake. For instance, farmers may adopt lower levels of crop residue retention, or change the field in which minimal tillage is practiced (Milder et al., 2011, p. 19). 3.1.3. The land area farmers dedicate to CA Viewing incremental uptake as the expansion of the CA area per farmer is a less commonly used measure of adoption. When used, mostly in household surveys conducted within the context of CA projects, there appears to be no clear trend towards increased uptake. For instance, in Zimbabwe it is reported the number of farmers practicing CF on some part of their land has increased, but their area has often not increased beyond the initial 0.2 ha for which inputs were provided by humanitarian aid programmes. ‘Farmers have cited labour constraints for basin preparation and


127

weeding as the major constraint for not expanding the area under CA’ (Marongwe et al., 2011, p. 157).11 In Zambia, the reported average CA area per farmer tends to be larger. In the CFU’s CAP project targeting some 120,000 farmers, it was found that hand-hoe farmers using basins had an average CA area of respectively 0.8 ha per farmer, representing about a quarter of their cropped land in 2009/10 (Nyanga, 2012a). Farmers using animal drawn rippers had on average 2.1 ha under CA (Aune et al., 2012, p. 14). But whereas it was found that ripper ownership also had a significant positive effect on the area under CA because of its labour saving properties, ownership of a chaka hoe had no such relation with the area under CA (Nyanga, 2012a, p. 34); while reducing the labour peak for land preparation as basins can be made over a longer period during the dry season, basin digging (with a chaka hoe) was found to be more labour demanding than hand hoe-ing of fields (Umar et al., 2012, p. 914). For Malawi, figures on the average CA area per farmer could not be found, possibly because the earlier noted lack of critical statistics on CA adoption in Malawi (MAFS, 2012, p. 14).

It is widely acknowledged that artificial incentives for CA adoption influence uptake and jeopardize its sustainability (CARWG, 2011, p. 74; Umar et al., 2011, p. 51; Giller et al., 2009, p. 29; FAO, 2011b, p. 17; Andersson and Giller, 2012). With funding mainly provided by international donors, Umar’s observation that ‘CA promotion in Zambia has been and remains donor dependent’ (Umar, 2012, p. 82) therefore also applies to other southern African countries, including Zimbabwe and Malawi. As was noted above, evidence of uptake of CA components thus does not necessarily imply incremental uptake of CA components or increasing number of farmers, as assumed by some interventionists.13 There is evidence of substantial dis-adoption after the promoting project or incentives end (Haggblade and Tembo, 2003a; Mazvimavi and Nyamangara, 2012; Norton et al., 2012). Figures on CA uptake generated within the context of promotional projects are therefore fundamentally uninformative with regard to long-term or sustained CA adoption.

3.1.4. The multi-season dimension: Adoption or uptake? Adoption estimates usually disregard the fact that crop rotation (CA principle 3) requires a multi-seasonal definition of CA adoption. Although CA definitions used in the different countries commonly define crop rotation as a sequence of three crops (see section 2), the number of cropping seasons in which CA practices have been applied hardly feature in adoption figures other than in the ICRISAT panel study of CF farmers in selected districts of Zimbabwe (Mazvimavi and Nyamangara, 2012). Undoubtedly influenced by the project context in which CA is promoted, shortterm uptake or evaluation by farmers is thus usually equated with longer-term adoption. Although studies of CA adoption preferably consider whether and what farmers continue to practice after a number of years after support has ended, current adoption figures are typically based on uptake levels at the end of the project.

CF promotion in Zambia initially took place within the context of strong synergies between the CFU and private sector entities such as agro-dealers and cotton companies in drier agro-ecological zones (Haggblade and Tembo, 2003a). Input provision through contract farming arrangements undoubtedly created an enabling environment for cash crop production and successful CF adoption. Yet, adoption rates of the planting basin and ripping technologies have shown dramatic differences across agro-ecological regions, both in early studies and later ones that report dis-adoption (Arslan et al., 2013). District-level rainfall variability appears strongly related to adoption in Zambia, suggesting ‘that farmers are using minimum tillage/planting basins as a strategy to mitigate the risk of rainfall variability’ (Arslan et al., 2013, p. 25). Yet, also between provinces and even within districts, large differences in adoption have been observed. In Zimbabwe, by contrast, CF uptake appeared higher in high rainfall areas (Natural Region II; annual rainfall 7501000 mm), but a confounding factor may have been that ‘NGOs operating in these areas tended to supply more basal and topdressing fertilizer than NGOs operating in the drier Natural Regions III, IV, and V’ (Mazvimavi and Twomlow, 2009, p. 24). For Malawi, such comparisons in adoption are yet difficult to make, but there are indications that CA adoption has already positive effects on yields in the short-run in drier agro-ecologies (Ngwira et al., 2012a). An important conclusion can be drawn from these findings on smallholder CA adoption. Sensu stricto, CA adoption in southern Africa is more limited than most adoption figures suggest, especially if we accept that ‘[M]inimal soil disturbance alone is an insufficient condition for CA’ (Erenstein et al., 2012, p. 181). Since CA constitutes a package of interrelated practices, it would be logical that definitions of adoption reflect this. Current definitions underpinning CA adoption figures neither incorporate all three components nor the necessary time frame.

3.2. Incentivized adoption Unlike early CA uptake in Zambia, which occurred in the context of contract farming arrangements and revolving loan schemes, CA uptake in southern Africa has increasingly taken place within the context of humanitarian and development projects providing incentives in the form of farm inputs. This is most apparent in the case of Zimbabwe’s Protracted Relief Programme (PRP) where input provision boosted both adoption rates and cereal yields of CF adopters, thus creating a ‘CA policy success’ (Andersson and Giller, 2012, p. 35). But also in Malawi in Zambia, incentives in the form of (subsidized) agricultural inputs (seed, fertilizer), farm implements and food aid have become common (Nyanga, 2012a, p. 37; MlozaBanda and Nanthambwe, 2010; Mazvimavi, 2011, p. 16).12 Common justifications for providing incentives for CA adoption include: the initial investments required, the delayed benefits associated with CA adoption, and higher levels of risk associated with the innovation, enabling farmers ‘to progress in CA’ (Nyanga, 2012a, p. 37).

11 This limitation of planting basin based CF has contributed to a growing interest in mechanized CA systems using animal drawn rippers, direct seeders and herbicides in Zimbabwe. Obviously, the interest in mechanized CA also means a shift from resource poor, vulnerable households towards more endowed and commercially oriented farmers (Nyagumbo et al., 2012; IFAD, 2011, p. 6; Marongwe et al., 2011, p. 154–155; Twomlow et al., 2008a,b, p. 3). 12 For instance, the FAO Emergency Programme in Zambia provided ‘food security packs’ consisting of seeds, fertilizer, herbicides and tools to farmers training in CF (Mona, 2011). FAO’s Farmer Input Support Response Initiative (FISRI) and CA Scaling-up for increased Productivity and Production Project (CASPP) provided input redeemable e-vouchers (FAO, 2011b; FAO-OED, 2012).

3.3. Where is CA taken up?

4. Discussion: The limitations of CA adoption studies In a review of 31 CA adoption studies worldwide, Knowler and Bradshaw (2007, p. 44) aptly depict the common focus of such studies as attempts by ‘social scientists to understand the farm-level adoption of CA, with the ultimate aim of offering refined policy prescriptions for augmenting adoption’. Yet, despite their finding of nearly 170 significant variables and conclusion that ‘there are few if any influences on adoption that apply universally’ (Knowler

13 For an example, agriculture-zambia.

see:

wbi.worldbank.org/wbi/multimedia/conservation-

128


and Bradshaw, 2007, p. 44), CA adoption studies continue to be dominated by econometric approaches that link farm-household characteristics to adoption (Ngwira et al., 2012c; Erenstein and Farooq, 2009; Mazvimavi et al., 2009; Mlamba, 2010; Nyanga, 2012a; Chiputwa et al., 2011; Arslan et al., 2013). While this review has highlighted the importance of socio-economic and institutional factors beyond the farm level for our understanding of smallholder farmers’ uptake of CA practices, there are hardly any systematic analyses of such higher scale issues for southern Africa (an exception is Mazvimavi and Twomlow, 2009). At the same time, the few available CA adoption studies for southern Africa that seek to identify–through household surveys–the characteristics that ‘determine’ adoption, suffer from serious methodological issues. Below, it is suggested that these shortcomings relate to the project context in which survey data is gathered and a limited understanding of (diverse) smallholder production systems in southern Africa. As a consequence, it is suggested, these regression-based analyses yield on their own a rather poor understanding of smallholders’ CA adoption.

discussion groups should vary. Yet, the tool largely ignores the institutional politics, power dynamics and vested interests surrounding CA adoption (Ndah et al., 2012; Corbeels et al., this issue). A similar project bias is apparent in survey-based analyses of adoption that find a significant relationship between farming households’ adoption of CA and (government or NGO provided) agricultural extension advice–often operationalized as the frequency of extensionists’ visits, farmer group membership or farmer field school participation. While the extend of such farmer–extensionist contact is frequently used in the CA literature to underscore that CA is ‘knowledge intensive’, the double role of (local) extension agents as both knowledge providers on CA and gatekeepers to (subsidized) farm inputs is either ignored (see: Ngwira et al., 2013; Mlamba, 2010, p. 30; Chiputwa et al., 2011; Nyanga, 2012), or the indivisibility of these dual roles is not be addressed in the study (Arslan et al., 2013, p. 25). As a consequence, the importance of knowledge provision by extension agents in the uptake of CA cannot be properly assessed.

4.1. CA adoption studies: common focus and biases

4.2. CA adoption studies and the understanding smallholder farming systems

As with data collection regarding CA (component) uptake, adoption studies in southern Africa are often conducted within the context of development projects extending CA to smallholders. This results in CA adoption figures that cannot be taken as true indicators of CA success in the form of long-term adoption of the three CA principles because CA is usually reduced to practicing minimum tillage on some part of land in a particular season (see section 3). Adoption studies also tend to be biased in their choice of development project participants as respondents, limiting both their wider applicability and our understanding of CA adoption. For instance, ICRISAT’s panel study of smallholder CA uptake was conducted in the context of the Zimbabwe’s Protected Relief Programme (PRP) (Mazvimavi et al., 2008, 2009, 2010; Mazvimavi and Twomlow, 2009; Mazvimavi and Nyamangara, 2012). It covered 12 districts and over 230 households, purposely sampling experienced (>1 year) CF practitioners, and actively searched for spontaneous adopters to include in the survey (Mazvimavi and Twomlow, 2009, p. 20). Thus, the sampling method precludes statements about adoption rates in the selected districts. Like other studies executed in the context of CA development projects, there is a bias towards adopters, and no systematic analyses are made of the reasons for dis-adoption and non-adoption (see also: Aune et al., 2012; Nyanga, 2011). Another type of bias resulting from CA adoption studies being carried out in the context of promotional projects is not limited to the commonly used survey methodology or its selection of respondents, but relates to the construction of knowledge in public social events such as focus group discussions. As Mosse has argued for the practice of Participatory Rural Appraisal (PRA), in focus group discussions–but also interviews with individuals–the answers people give are ‘likely to be strongly influenced by expectations people have of the project and its particular interest in them’ (Mosse, 1994, p. 517). Adoption research in the context of CA promotional projects that simultaneously provide (subsidized) input support–often carried out by researchers associated with the project–is thus likely to be biased by farmers’ expectations of (continued) support. A recently developed Qualitative Assessment Tool for CA adoption in Africa (QAToCA), which is meant for experts, research teams and managers of development projects also appears to fall victim to this bias. Focusing on the ‘adoption potential’ for CA, the tool uses a systematic list of questions and criteria to guide focus group discussions with CA farmers, extensionists, CA experts and other stakeholders in the promotion of CA. To guarantee unbiased and reliable answers it is suggested the composition of the

The interpretation of regression-based analyses of household characteristics in CA adoption studies is not merely hampered by a lack of understanding or ignorance of the promotional project context. A reliance on standard survey instruments using variables such as age, sex, education level, household size, farm implement ownership, and some CA related variables such a implement ownership and herbicide use, also causes our understanding of CA adoption by smallholder farmers to be limited. Such instruments may merely reify the effects of CA promotion, for instance, when it is found that ‘CA trainings, previous experience in minimum tillage, membership in farmer organizations, and ownership of CA tillage equipment significantly increased the likelihood of CA adoption’ (Nyanga, 2012). Alternatively, the underlying mechanisms of found relationships between survey-derived variables remain underexplored and ill-understood. For instance, Chiputwa et al. (2011, p. 14) found ‘a significant positive relationship between cattle ownership and adoption and use of zero-tillage’ in northern Zimbabwe. Assuming that cattle ownership should be taken here as an indicator of wealth, they postulate that therefore ‘farmers with cattle might be able to raise the initial investment capital required. . .’ or that ‘the bigger the herd the more the labor and capital requirements for management purposes and hence the need to explore labor saving technologies (e.g. zero-tillage)’. By contrast, a study in Zambia found that cattle ownership was negatively related to CA component uptake (Nyanga, 2012; Arslan, 2013). However, a monitoring and evaluation report of Zambia’s CA Project (CAP) found that relatively poorer farmers were more likely to adopt (basin based) CF, stating ‘adopters have 24% less income. . . 20% fewer animals. . . and 34% fewer oxen per household than the non-CA adopters’ (Aune et al., 2012, p. 14). Again, cattle ownership is interpreted as an indicator of wealth, but the relationship between wealth and CF adoption is implied to be the reverse, in the Zambia CAP case, of the aforementioned Northern Zimbabwe case. Another study in Zambia, using a more elaborate set of wealth indicators, distinguishing agricultural (farm implements) from non-agricultural wealth (household durables), and using also the number of oxen as an independent variable, provides more insight: ‘. . .betteroff households with higher wealth indices (i.e. more household durables) are more likely to adopt.’ Yet, ‘households with more oxen are significantly less likely to adopt CF’ (Arslan et al., 2013, p. 18). While the use of more specific indicators thus produces more clues as to the relevant characteristics of CA adopters, the actual mechanisms (or resource allocation strategies) leading households with more oxen to adopt or not adopt CA remain opaque.


A recent study in Malawi (Ngwira et al., 2013), where few farmers use animal draught power, provides another example of a lack of understanding of the mechanisms underpinning the statistical relationships found. The study suggests that wealthier farmers, here defined as those who (are able to) hire labour, are more likely to adopt CA practices. Although not elaborated upon in the study, it appears that wealthier farmers substitute the labour they hire from their poorer counterparts for (subsidized) herbicides, thus turning CA uptake into a mechanism of advanced social differentiation. But as with the previous examples, additional analyses of existing data or new research is required to understand whether this is indeed the underlying mechanism of the statistical relation found. These findings suggest that current analyses of survey-derived data are hampered by their reliance on general indices and a limited interpretative framework. It appears that the understanding of CA adoption could benefit from a stronger focus on the mechanisms that shape different smallholder farmers’ adoption decisions, rather than a singular focus on the characteristics of adopters. Such a shift requires a more thorough understanding of the functioning of diverse smallholder farming households, their resource allocation strategies, and labour relations. The use of a wider set of both quantitative and qualitative research methodologies thus appears a necessity, not only for our understanding of prerequisite conditions, but also to comprehend better CA adoption at farm-scale.

4.3. Cash needs and net returns: the limitations of household economic analyses A limited understanding of smallholder farming households and their production systems not only appears to limit the value of survey-based adoption studies, but restricts household economic analyses of CA adoption as well. These analyses usually suggest that capital requirements and variable costs are higher under CA (for CA implements, additional fertilizer, herbicides). For instance, in manual CA systems in Malawi, it was found that ‘[T]otal variable costs were higher in CA systems compared to conventional practice’ (Ngwira et al., 2012b, p. 154). The commonly used argument in the CA literature that reduced labour costs and higher yields under CA can compensate for such higher costs (Aune et al., 2012, p. 12; Mazvimazi 2011, p. 6; Ito et al., 2007, p. 421) and result in higher returns under CA is, however, problematic in the context of smallholder agriculture in southern Africa. First, mobilizing cash for farm inputs at the beginning of the season is, for many poor smallholder farmers, a daunting task, as evidenced by low mineral fertilizer application rates–estimated at an average 8 kg ha−1 in sub-Saharan Africa (SSA) (Groot, 2009). CA further increases such cash requirements for farming, as additional fertilizer and herbicides and/or additional labour for weeding need to be purchased. Second, although net returns to labour may be higher under CA, food insecurity is common among smallholders. The increased production that may result from practicing CA is therefore unlikely to sustain the additional investment requirements of CA. In many situations increased production is more likely to be consumed by the household. The common assumption in household economic analyses that such additional consumption releases funds–to sustain the required investment in CA–otherwise spend on food purchases is, however, problematic. As seasonal hunger is a recurrent phenomenon in many rural African households, additional household consumption resulting from increased yields is more likely to reduce hunger, than to increase household income or release other funds for investment in CA. This is particularly true in many parts of southern Africa where land holdings are small (sometimes even < 0.5 ha); here, increased yields are both unlikely to generate a marketable surplus, or substantially reduce food purchases in order to leave sufficient cash for agricultural investment.

129

Third, the argument of lower labour costs under CA due to the adoption of reduced tillage is problematic as it is based on the theoretical assumption that farming households cost labour against a fixed, monetized, price. In reality, most smallholder farms rely on unpaid family labour and operate in farming communities where the availability of labour may fluctuate sharply within the agricultural season due to peaks in labour demand (Baudron et al., 2012; Haggblade et al., 2011). Hence, farm labour is, in reality, a far more complicated production factor than household economic analyses of CA assume, and these analyses are therefore of limited value for our understanding of–the likelihood of–CA adoption by smallholder farmers. In short, the reliance on family labour as well as the chronic lack of cash that characterises smallholder farming in southern Africa, make it unlikely that many smallholders can sustain the capital investment levels required for CA. Thus, we may understand why smallholders’ CF-basin area does not expand beyond the area that can be manually weeded with family labour or why smallholders won’t reduce the drudgery of manual weeding and buy herbicides just because ‘the additional cost of [herbicide] inputs [i]s balanced out in the gross margins’ (Muoni et al., 2013, p. 7).

5. Conclusion Whereas CA for smallholder farmers in southern Africa was first tested as a collection of soil and water conservation measures, large-scale promotion followed a reframing of CA as a production, food security and livelihood enhancing set of practices. This reframing was primarily a response to the wider socio-economic, institutional and (donor)-political context of smallholder farming in southern Africa at the end of the 20th century. Yet, extensive promotion of CA for smallholder farmers meant a concentration on ‘inappropriate farming practices’ in need of change, while largely ignoring the structural factors shaping these practices. Consequently, CA promotional practice has become technology and plot-level focused, while higher-level contextual factors, or prerequisite conditions (Sumberg, 2005) for adoption are largely ignored. In addition to the various framings used to justify CA promotion, we examined the different definitions of CA that have emerged in Zambia, Zimbabwe and Malawi. It was shown these are not merely agro-ecologically defined, but fit the socio-economic and institutional contexts in which they work. The diverging definitions of what constitutes and is promoted as CA complicate the assessment of adoption across the region. Nevertheless, a recurrent set of farm-level constraints and prerequisite (contextual) conditions for smallholder CA adoption has been identified in scientific articles and project documents. These barriers are generally regarded as the cause of limited CA adoption among smallholder farmers in Africa. Yet, the identified adoption barriers are usually related to specific CA practices or the agro-ecological circumstances in which they need to be applied, while their adoptability by different types of farmers and in different socio-economic circumstances tends to be understudied. Amidst definitional diversity regarding what constitutes CA, a lack of clarity about what adoption entails, and the widespread use of incentives by donor-supported CA promoting projects, adoption figures are usually more obscuring than revealing. Basic dimensions of adoption, such as the specific CA practices taken up, the areas covered, and number of cropping seasons in which these practices are applied, are often ignored in favour of a commonly used–but often implicit–definition of the CA adopter as a farmer who practices the minimum tillage component of CA on some part of his/her land in a given season. Sensu stricto, the adoption of CA in southern Africa as consisting of three interrelated components is thus more

130


limited than claimed, while the use of incentives probably makes long-term adoption less than what studies suggest. Methodological weaknesses of the current–limited number of–adoption studies further limit our understanding of smallholder CA adoption in southern Africa. Usually carried out in the context of promotional projects that provide input support or subsidies, respondent selection and findings are project biased. A strong reliance on standard household surveys in CA adoption studies does not help either. Econometric analyses of survey-derived data yields general characteristics of CA adopting households, but leaves farmers’ resource allocation strategies that underpin the adoption or non-adoption of CA by different types of farming households largely unexplored. Household economic analyses showing the viability of CA adoption appear equally limited in understanding the realities of smallholder farming that result in low CA adoption rates in southern Africa. Current farm-scale oriented studies of CA adoption in southern Africa are in need of improvement. The adoption of a systems perspective and the use of a wider range of quantitative and qualitative research methods in the analyses of smallholder farming households and their resource allocation strategies may go a long way in understanding better the farm-level adoption constraints different types of farmers face. In addition, different type of studies focusing on the wider market, institutional and policy context are needed. As this article has shown, contextual factors appear key for our understanding of both the current drive to extend CA among African smallholder farmers, as well as these farmers’ real adoption of CA. Acknowledgements The authors wish to thank Bruno Gerard, Ken Giller, Christian Thierfelder, Carl Wahl and the journal’s anonymous reviewers for comments and suggestions. References Aagaard, P.J., 2010. Conservation farming, productivity and climate change [WWW Document]. Conservation Farming Unit, Zambia. Aagaard, P.J., 2012. History of the Conservation Farming Unit–Zambia [WWW Document]. Alvord, E.D., 1958. The Development of Native Agriculture and Land Tenure in Southern Rhodesia. University of Zimbabwe Library, Harare (unpublished manuscript). Andersson, J.A., 2007. How much did property rights matter? Understanding Food insecurity in Zimbabwe: A critique of Richardson. African Affairs 106, 681–690. Andersson, J.A., Giller, K.E., 2012. On heretics and God’s blanket salesmen: Contested claims for Conservation Agriculture and the politics of its promotion in African smallholder farming. In: Sumberg, J., Thompson, J. (Eds.), Contested Agronomy: Agricultural Research in a Changing World. Routledge, London, pp. 22–46. Arslan, A., McCarthy, N., Lipper, L., Asfaw, S., Catteneo, A., 2013. Adoption and intensity of adoption of conservation farming practices in Zambia, ESA Working paper 13-01. FAO, Rome. Aune, J.B., Nyanga, P., Johnsen, F.H., 2012. A monitoring and evaluation report of the Conservation Agriculture Project 1 (CAP1) in Zambia. Department of International Environment and Development Studies, Noragric, Norwegian University of Life Sciences. Baker, J.M., Ochsner, T.E., Venterea, R.T., Griffis, T.J., 2007. Tillage and soil carbon sequestration—What do we really know? Agriculture, Ecosystems and Environment 118, 1–5. Baudron, F., Andersson, J.A., Corbeels, M., Giller, K.E., 2012. Failing to Yield? Ploughs, Conservation Agriculture and the Problem of Agricultural Intensification: An Example from the Zambezi Valley. Journal of Development Studies 48, 393–412. Baudron, F., Mwanza, H., Triomphe, B., Bwalya, M., 2007. Conservation agriculture in Zambia: a case study of Southern Province. African Conservation Tillage Network. Centre de Coopération Internationale de Recherche Agronomique pour le Développement, Food and Agriculture Organization of the United Nations, pp. 1–57. Barahona, C., Cromwell, E., 2005. Starter Pack and sustainable agriculture. In: Levy, S. (Ed.), Starter Packs: a Strategy to Fight Hunger in Developing Countries? Cabi Publishing, Walingford, pp. 155–174. Beinart, W., 1984. Soil erosion, conservationism and ideas about development: A southern African exploration, 1900–1960. Journal of Southern African Studies 11, 52–83.

Bunderson, W.T., Jere, Z.D., Chisui, J.L., Museka, R., Wall, P.C., Ngwira, A., Mwale, C., 2009. Conservation Agriculture in Malawi: Integrating agroforestry to enhance productivity and sustainability. http://conservationagriculture.org/ external-research (accessed 10 May 2013). Bwalya, M., Friedrich, T., 2009. Conservation Agriculture in Development: The Case of Africa., pp. 1–14. CARWG, 2011. Regional Conservation Agriculture Symposium-Workshop documentation, Presented at the Conservation Agriculture Regional Working Group (CARWG), Johannesburg, pp. 1–113. CFU-Zambia, 2007a. Conservation Farming and Conservation Agriculture Handbook for Hoe Farmers in Agro-Ecological Regions I and IIa-Flat Culture, Zambia National Farmers Union, Lusaka. CFU-Zambia, 2007b. Conservation Farming and Conservation Agriculture Handbook for Ox Farmers in Agro-ecological Regions I and IIa, Zambia National Farmers Union, Lusaka. CFU-Zambia, 2012. Hoe CF: Land Preparation and Basal Dressing. Zambia National Farmers Union, Lusaka. Chan, K.Y., Conyers, M.K., Li, G.D., Helyar, K.R., Poile, G., Oates, A., Barchia, I.M., 2011. Soil carbon dynamics under different cropping and pasture management in temperate Australia: results of three long-term experiments. Soil Research 49, 320–328. Chauhan, B.S., Singh, R.G., Mahajan, G., 2012. Ecology and management of weeds under conservation agriculture: A review. Crop Protection 38, 57–65. Chikowo, R., 2011. Climatic Risk Analysis in Conservation Agriculture in Varied Biophysical and Socio-economic Settings of Southern Africa. FAO Regional Emergency Office for Southern Africa (REOSA), Johannesburg. Chiputwa, B., Langyintuo, A.S., Wall, P.C, 2011. Adoption of conservation agriculture technologies by smallholder farmers in the Shamva District of Zimbabwe: A Tobit application, Presented at the Southern Agricultural Economics Association (SAEA) in Texas, USA, February 5–8, 2011, Texas, USA, pp. 1–28. Chivenge, P.P., Murwira, H.K., Giller, K.E., Mapfumo, P., Six, J., 2007. Long-term impact of reduced tillage and residue management on soil carbon stabilization: Implications for conservation agriculture on contrasting soils. Soil Tillage Research 94, 328–337. CIMMYT, 2012. Conservation agriculture in Malawi: “we always have problems with rain here” [WWW Document]. cimmyt.org. URL http://www.cimmyt.org/en/ front-page-tems/aboutmediaresources/1308-conservation-agriculture-africanfarmers-dodge-hunger-from-drought de Wolf, J.J., 2010. Innovative farmers non-adapting institutions: a case study of the organization of agroforestry research in Malawi. In: German, L.A., Ramisch, J.J., Verma, R. (Eds.), Beyond the Biophysical: Knowledge, Culture, and Power in Agriculture and Natural Resource Management. Springer, Dordrecht, Netherlands, pp. 217–239. Derpsch, R., Friedrich, T., Kassam, A., Hongwen, L., 2010. Current status of adoption of no-till farming in the world and some of its main benefits. International Journal of Agricultural and Biological Engineering 3, 1–25. Dorward, A., Chirwa, E., Kelly, V.A., Jayne, T.S., Slater, R., Boughton, D., 2008. Evaluation of the 2006/7 agricultural input subsidy programme, Malawi. In: Final Report. Michigan State University, Department of Agricultural, Food, and Resource Economics. Ellis, Kutengule, F.,M., Nyasulu, A., 2003. Livelihoods and rural poverty reduction in Malawi. World Development 31, 1495–1510. Elwell, H.A., 1993. Development and adoption of conservation tillage practices in Zimbabwe. In: Soil Tillage in Africa: Needs and Challenges (FAO Soils Bulletin 69). FAO, Rome. Erenstein, O., 2006. Intensification or extensification? Factors affecting technology use in peri-urban lowlands along an agro-ecological gradient in West Africa. Agricultural Systems 90, 132–158. Erenstein, O., 2012. Conservation Agriculture-based technologies and the political economy: Lessons from south Asia. In: Sumberg, J., Thompson, J. (Eds.), Contested Agronomy: Agricultural Research in a Changing World, Contested Agronomy. Agricultural Research in a Changing World. Routledge, London, pp. 47–63. Erenstein, O., Farooq, U., 2009. A survey of factors associated with the adoption of zero tillage wheat in the irrgated plains of south Asia. Experimental Agriculture 45, 133–147. Erenstein, O., Sayre, K., Wall, P.C., Dixon, J., Hellin, J., 2008. Adapting No-Tillage Agriculture to the Conditions of Smallholder Maize and Wheat Farmers in the Tropics and Sub-Tropics, World Association of Soil and Water Conservation (WASWC)., pp. 253–278. Erenstein, O., Sayre, K., Wall, P.C., Hellin, J., Dixon, J., 2012. Conservation agriculture in maize- and wheat-based systems in the (sub)tropics: lessons from adaptation initiatives in South Asia, Mexico, and Southern Africa. Journal of Sustainable Agriculture 36, 180–206. FAO, 2005. Livestock Sector Brief: Malawi, fao.org. FAO, Rome. FAO, 2008. Investing in Sustainable Agricultural Intensification The Role of Conservation Agriculture: A Framework for Action, p. 1–24. FAO, 2011a. Save and Grow: A policymaker’s guide to the sustainable intensification of smallholder crop production. FAO, Rome. FAO, 2011b. Terminal report Conservation Agriculture Scaling Up for Increased Productivity and Production (CASPP) Project. FAO/Norwegian Embassy/MInistry of Agriculture and Cooperatives, Lusaka. FAO-OED, 2012. Final evaluation of the Farmer Input Support Response Initiative (FISRI) to rising prices of agricultural commodities in Zambia: final evaluation report (draft). Office of Evaluation (OED), Food and Agriculture Organisation of the United Nations, Rome, Italy.

J.A. Andersson, S. D’Souza / Agriculture, Ecosystems and Environment 187 (2014) 116–132 FAO-REOSA, 2011. Farming for the Future: Stories from the Field. FAO, Johannesburg. Ferguson, J., 1999. Expectations of Modernity: Myths and Meanings of Urban Life on the Zambian Copperbelt. University of California Press. Friedrich, T., Derpsch, R., Kassam, A., 2011. Global Overview of the Spread of Conservation Agriculture. Presented at the 5th World Congress of Conservation Agriculture, September 201, Brisbane. Giller, K.E., 2001. Nitrogen Fixation in Tropical Cropping Systems. Cabi, Wallingford. Giller, K.E., Corbeels, M., Nyamangara, J., Triomphe, B., Affholder, F., Scopel, E., Tittonell, P., 2011. A research agenda to explore the role of conservation agriculture in African smallholder farming systems. Field Crops Research 124, 468–472. Giller, K.E., Witter, E., Corbeels, M., Tittonell, P., 2009. Conservation agriculture and smallholder farming in Africa: The heretics’ view. Field Crops Research 114, 23–34. Green, E., 2009. A lasting story: Conservation and agricultural extension serivices in colonial Malawi. Journal of African History 50, 247–267. Gould, J., 1997. Localizing Modernity: Action, Interests and Association in Rural Zambia. Transactions of the Finnish Anthropological Society, Helsinki. Govaerts, B., Sayre, K.D., Deckers, J., 2005. Stable high yields with zero tillage and permanent bed planting? Field Crops Research 94, 33–42. Govaerts, B., Verhulst, N., Castellanos-Navarrete, A., Sayre, K.D., Dixon, J., Dendooven, L., 2009. Conservation agriculture and soil carbon sequestration—between myth and farmer reality. Critical Reviews in Plant Sciences 28, 97–122. Groot, J.R., 2009. Update of fertilizer supply and demand: Sub-Saharan Africa, Presented at the IFA Africa Forum. IDFC, Egypt. Guto, S., Pypers, P., Vanlauwe, B., de Ridder, N., Giller, K.E., 2011a. Socio-ecological niches for minimum tillage and crop-residue retention in continuous maize. Soil Tillage, Conservation and Management 103, 1–11. Guto, S.N., Pypers, P., Vanlauwe, B., de Ridder, N., Giller, K.E., 2011b. Tillage and vegetative barrier effects on soil conservation and short-term economic benefits in the Central Kenya highlands. Field Crops Research 122, 85–94. Haggblade, S., Kabwe, S., Plerhoples, C., 2011. Productivity impact of conservation farming on smallholder cotton farmers in Zambia. In: FSRP Working Paper No. 47. Food Security Research Project, Lusaka, Zambia. Haggblade, S., Tembo, G., 2003a. Conservation Farming in Zambia EPTD Discussion Paper No. 108. International Food Policy Research Institute, Washington and East Lansing. Haggblade, S., Tembo, G., 2003b. Development, Diffusion and Impact of Conservation Farming in Zambia. Food security Research Project, Lusaka, Zambia. Hobbs, P.R., Sayre, K.D., Gupta, R., 2008. The role of conservation agriculture in sustainable agriculture. Philosophical Transactions of the Royal Society B: Biological Sciences 363, 543–555. Hove, L., 2011. Current priorities and constraints in using crop residues by smallholder farmers in semi-arid areas of Zimbabwe: implications for Conservation Agriculture. Howard, J.A., Mungoma, C., 1996. Zambia’s Stop-and-go Revolution: The Impact of Policies and Organizations on the Development and Spread of Maize Technology. Michigan State University, Department of Agricultural, Food, and Resource Economics, East Lansing. IFAD, 2011. Smallholder conservation agriculture: rationale for IFAD involvement and relevance to the East and Southern Africa region. International Fund for Agricultural Development, Rome. Ito, M., Matsumoto, T., Quinones, M.A., 2007. Conservation tillage practice in subSaharan Africa: The experience of Sasakawa Global 2000. Crop Protection, 26, 417–423. Kassam, A., Friedrich, T., Shaxson, F., Pretty, J., 2009. The spread of Conservation Agriculture: justification, sustainability and uptake. International Journal of Agricultural Sustainability 7, 292–320. Kirkegaard, J., Conyers, M.K., Hunt, J., Kirkby, C., Watt, M., Rebetzke, G., 2011. Sense and nonsense in conservation agriculture: principles, pragmatism and productivity in Australian mixed farming systems. aciar.gov.au., 1–12. Knowler, D., Bradshaw, B., 2007. Farmers’ adoption of conservation agriculture: A review and synthesis of recent research. Food Policy 32, 25–48. Lal, R., 2004. Soil carbon sequestration impacts on global climate change and food security. Science 3, 1623–1627. Langmead, P., 2006. Hoe conservation farming of maize in Zambia. Levy, S., 2005. Starter Packs: A Strategy to Fight Hunger in Developing Countries? Cabi Publishing, Wallingford. Luo, Z., Wang, E., Sun, O.J., 2010. Can no-tillage stimulate carbon sequestration in agricultural soils? A meta-analysis of paired experiments. Agriculture, Ecosystems and Environment 139, 224–231. MAFS, 2012. Approaches to the Implemetation of CA among Promoters in Malawi. National Conservation Agriculture Task Force Secretariat Land Resources Conservation Department, Lilongwe, Malawi. Mando, A., Fatondji, D., Zougmore, R., Brussaard, L., Bielders, C.L., Martius, C., 2006. Restoring soil fertility in semi-arid West Africa: Assessment of an indigenous technology. In: Uphoff, N., Ball, A.S., Fernandes, E., Herren, H., Husson, O., Laing, M., Pretty, J., Sanchez, P., Sanginga, N., Thies, J. (Eds.), Biological Appraoches to Sustainable Soil Systems. Taylor and Francis, London, pp. 391–399. Marongwe, S., 2012. Conservation agriculture implementation in the FISRI project. Draft Final Report: Final Evaluation of the Farmer Input Support Response Initiative (FISRI). Marongwe, S., Kwazira, K., Jenrich, M., Thierfelder, C., Kassam, A., Friedrich, T., 2011. An African success: the case of conservation agriculture in Zimbabwe. International Journal of Agricultural Sustainability 9, 153–161. Mazvimavi, K., 2011. Socio-economic Analysis of Conservation Agriculture in Southern Africa, second ed. FAO-REOSA, Johannesburg.

131

Mazvimavi, K., Ndlovu, P.V., Nyagumbo, I., Dimes, J., 2009. Trends in Adoption of Conservation Farming Practices Among Smallholder Farmers in Zimbabwe. ICRISAT, Bulawayo, Zimbabwe. Mazvimavi, K., Nyamangara, J., 2012. Dynamics in conservation farming adoption intensity, 2007–2010. Presented at the Presentation to the CA Task Force, 21 January 2011, Harare, Zimbabwe, pp. 1–16. Mazvimavi, K., Nyathi, P., Murendo, C., 2011. Conservation Agriculture practices and challenges in Zimbabwe. Presented at the 5th World Congress of Conservation Agriculture, September 2011, Brisbane, Australia, pp. 295–296. Mazvimavi, K., Twomlow, S.J., 2009. Socioeconomic and institutional factors influencing adoption of conservation farming by vulnerable households in Zimbabwe. Agricultural Systems 101, 20–29. Mazvimavi, K., Twomlow, S.J., Belder, P., Hove, L., 2008. An assessment of the sustainable uptake of conservation farming in Zimbabwe, [report]. ICRISAT, Bulawayo, Zimbabwe. Mazvimavi, K., Ndlovu, P.V., Nyathi, P., Minde, I.J., 2010. Conservation Agriculture Practices and Adoption by Smallholder Farmers in Zimbabwe. Presented at the 3rd African Association of Agricultural Economists (AAAE) and 48th Agricultural Economists Association of South Africa (AEASA) Conference, September 19–23, 2010, Cape Town, pp. 1–20. McCracken, J., 2012. A History of Malawi, 1859-1966. James Currey Publishers, Woodbridge, UK. McCarthy, N., Lipper, L., Branca, G., 2011. Climate-Smart Agriculture: Smallholder Adoption and Implications for Climate Change Adaptation and Mitigation. Milder, J.C., Majanen, T., Scherr, S.J., 2011. Performance and potential of conservation agriculture for climate change adaptation and mitigation in Sub-Saharan Africa, An assessment of WWF and CARE projects in support of the WWF-CARE Alliance’s Rural Futures Initiative. Ecoagriculture Partners, Washington, DC. Mlamba, J.L., 2010. Factors Affecting Adoption of Conservation Agriculture in Malawi (A Case Study of Salima District). Mloza-Banda, H.R., Nanthambwe, S.J., 2010. Conservation Agriculture Programmes and Projects in Malawi: Impacts and Lessons. National Conservation Agriculture Task Force Secretariat Land Resources Conservation Department, Lilongwe. Mloza-Banda, H.R., Njoloma, H.M., Sikwese, M.M.G., 2011. Conservation Agriculture Research Study Report. Concern Universal/Land Resources Conservation Department, Ministry of Agriculture and Food Security, Lilongwe, Malawi. Mona, N., 2011. FAO-Zambia achievements. Food and Agriculture Organisation of the United Nations. Mosse, D., 1994. Authority, gender and knowledge: theoretical reflections on the practice of participatory rural appraisal. Development and Change 25, 497–526. Mupangwa, W., Twomlow, S.J., Walker, S., Hove, L., 2007. Effect of minimum tillage and mulching on maize (Zea mays L.) yield and water content of clayey and sandy soils. Physics and Chemistry of the Earth, Parts A/B/C 32, 1127–1134. Mutsamba, E., Nyagumbo, I., Mafongoya, P., 2012. Dry season crop residue management using organic livestock repellents under conservation agriculture in Zimbabwe. Journal of Organic Systems 7, 5–13. Ndah, H.T., Schuler, J., Uthes, S., Zander, P., Triomphe, B., Mkomwa, S., Corbeels, M., 2012. Adoption potential of conservation agriculture in Africa: a newly developed assessment approach (QAToCA) applied in Kenya and Tanzania. Land Degradation and Development. Ndiyoi, M., Mwala, M., Mupindu, S., Marongwe, S., 2012. Beneficiary assessment report on capacity building for CA: Expansion of the Farmer Input Support Response Initiative (FISRI) to rising prices of agricultural commodities in Zambia. In: Draft Final Report: Final Evaluation of the Farmer Input Support Response Initiative (FISRI), Zambia. Ngwira, A.R., Aune, J.B., Mkwinda, S., 2012a. On-farm evaluation of yield and economic benefit of short term maize legume intercropping systems under conservation agriculture in Malawi. Field Crops Research 132, 149–157. Ngwira, A.R., Thierfelder, C., Lambert, D.M., 2012b. Conservation agriculture systems for Malawian smallholder farmers: long-term effects on crop productivity, profitability and soil quality. Renewable Agriculture & Food Systems, 1–14. Ngwira, A.R., Johnsen, F.H., Mulugetta, M., Aune, J.B., Thierfelder, C., 2012c. Understanding Determinants of Adoption of Conservation Agriculture by Smallholder Farmers in Some Selected Areas of Central and Southern Malawi (unpublished manuscript). Norton, A., 1988. No-till tied-ridging: a recommended sustained crop production system. agris.fao.org. Norton, A., Chaora, B., Mullens, D., 2012. An evaluation of Operation Joseph. In: River of Life Church Harare. Nyagumbo, I., Mutsamba, E., Barrett, T., Dengu, E., Thierfelder, C., 2012. Prospects for Up-scaling Conservation Agriculture in Zimbabwe using Animal Traction Mechanization Technologies. Nyanga, P.H., 2012a. Food security, conservation agriculture and pulses: evidence from smallholder farmers in Zambia. Journal of Food Research 1, 120–138. Nyanga, P.H., 2012b. Factors Influencing Adoption and Area under Conservation Agriculture: A Mixed Methods Approach. SAR 1, 27–40. Oldreive, B., 1993. Conservation farming for communal, small-scale, resettlement and co-operative farmers in Zimbabwe. In: A Farm Management Handbook. Page, S., Page, H., 1991. Western hegemony over African agriculture in Southern Rhodesia and its continuing threat to food security in independent Zimbabwe. Agriculture and Human Values 8, 3–18. Palmer, R., 1977. Land and Racial Domination in Rhodesia. Heinemann, London. Paul, B.K., Vanlauwe, B., Ayuke, F., Gassner, A., Hoogmoed, M., Hurisso, T.T., Koala, S., Lelei, D., Ndabamenye, T., Six, J., Pulleman, M.M., 2013. Medium-term impact of tillage and residue management on soil aggregate stability, soil carbon and crop productivity. Agriculture, Ecosystems and Environment 164, 14–22.

132


Phimister, I., 1988. An Economic and Social History of Zimbabwe 1890–1948: Capital Accumulation and Class Struggle. Longman, London. Rukuni, M., Eicher, C.K., 1987. Zimbabwe’s Agricultural Revolution. University of Zimbabwe Publications, Harare. Rusinamhodzi, L., Corbeels, M., Zingore, S., Nyamangara, J., Giller, K.E., 2013. Pushing the envelope? Maize production intensification and the role of cattle manure in recovery of degraded soils in smallholder farming areas of Zimbabwe. Field Crops Research 147, 40–53. Rusinamhodzi, L., Corbeels, M., Nyamangara, J., Giller, K.E., 2012. Maize–grain legume intercropping is an attractive option for ecological intensification that reduces climatic risk for smallholder farmers in central Mozambique. Field Crops Research 136, 12–22. Rusinamhodzi, L., Corbeels, M., van Wijk, M.T., Rufino, M.C., Nyamangara, J., Giller, K.E., 2011. A meta-analysis of long-term effects of conservation agriculture on maize grain yield under rain-fed conditions. Agronomy for Sustainable Development 31, 657–673. Saasa, O.S., 2003. Agricultural Intensification in Zambia: The role of policies and policy processes. Institute of Economics and Social Research University of Zambia. Siacinji-Musiwa, J.M., 1999. Conservation tillage in Zambia: some technologies, indigenous methods and environmental issues. In: Kaumbutho, P.G., Simalenga, T.E. (Eds.), Conservation Tillage with Animal Traction. A Resource Book of the Animal Traction Network for Eastern and Southern Africa (ATNESA), Harare, Zimbabwe, pp. 27–33. Simasiku, P., Chapoto, A., Richardson, R., Sichilongo, M., Tembo, G., Weber, M., Zulu, A., 2010. Natural resource management, food security and rural development in Zambia: Moving from research evidence to action (No 44). Food Security Research Project, Lusaka. Smith, R.D., 1988. Tillage trials in Zimbabwe: 1957-1988: Report commissioned for the Contill project. Institute for Agricultural Engineering, Harare. Sosola, B., Gudeta, S., Akinnifesi, F.K., Ajayi, O.C., World Agroforestry Centre, Lilongwe, Malawi, 2011. Conservation Agriculture Practices in Malawi. Opportunities and Challenges, 1–7. Sumberg, J., 2005. Constraints to the adoption of agricultural innovations: Is it time for a re-think? Outlook on Agriculture 34, 7–10. Sumberg, J., Keeney, D., Dempsey, B., 2012. Public Agronomy: Norman Borlaug as “Brand Hero” for the Green Revolution. Journal of Development Studies 48, 1–14. Thierfelder, C., Cheesman, S., Rusinamhodzi, L., 2013a. Benefits and challenges of crop rotations in maize-based conservation agriculture (CA) cropping systems of southern Africa. International Journal of Agricultural Sustainability, 1–17. Thierfelder, C., Mombeyarara, T., Mango, N., Rusinamhodzi, L., 2013b. Integration of conservation agriculture in smallholder farming systems of southern Africa: identification of key entry points. International Journal of Agricultural Sustainability, 1–14. Thiombiano, L., Meshack, M., 2009. Scaling-up Conservation Agriculture in Africa: Strategy and Approaches. FAO Sub Regional Office for Eastern Africa, Addis Ababa. Tittonell, P., Shepherd, K.D., Vanlauwe, B., Giller, K.E., 2008. Unravelling the effects of soil and crop management on maize productivity in smallholder agricultural systems of western Kenya—An application of classification and regression tree analysis. Agriculture, Ecosystems and Environment 123, 137–150. Trapnell, C.G., 1943. The Soils, Vegetation and Agriculture of North-Eastern Rhodesia. Government Printer, Lusaka. Twomlow, S.J., Ellis-Jones, J., Hagmann, J., 1995. Soil and Water Conservation for Smallholder Farmers in Semi-arid Zimbabwe: Transfer between Research and Extension. Presented at the Proceedings of a Technical Workshop, 3–7 April 1995, Masvingo, Zimbabwe.

Twomlow, S.J., Hove, L., Mupangwa, W., Masikati, P., Mashingaidze, N., 2008a. Precision conservation agriculture for vulnerable farmers in low-potential zones Increasing the productivity and sustainability of rainfed cropping systems of poor, smallholder farmers. In: Proceedings of the CGIAR Challenge Program on Water and Food workshop on Rainfed Cropping Systems, pp. 22–25. Twomlow, S.J., Urolov, J., Jenrich, M., 2008b. Lessons from the field: Zimbabwe’s Conservation Agriculture Task Force. Journal of SAT Agricultural Research 6, 1–11. Umar, B.B., 2012. Reversing agro-based land degradation through conservation agriculture: emerging experiences from Zambia’s smallholder farming sector. SAR 1, 72–87. Umar, B.B., Aune, J.B., Johnsen, Fred H., Lungu, I. Obed, 2012. Are smallholder Zambian farmers economists? A dual-analysis of farmers’ expenditure in Conservation and Conventional Agriculture Systems. Journal of Sustainable Agriculture 36, 908–929. Umar, B.B., Aune, J.B., Johnsen, F.H., Lungu, O.I., 2011. Options for improving smallholder conservation agriculture in Zambia. Journal of Agricultural Science 3, 50–62. UNDP, 2012. The Millennium Development Goals Report 2012. United Nations, New York. Valbuena, D., Erenstein, O., Tui, S.H.-K., Abdoulaye, T., Claessens, L., Duncan, A.J., Gérard, B., Rufino, M.C., Teufel, N., van Rooyen, A., van Wijk, M.T., 2012. Conservation Agriculture in mixed crop–livestock systems: Scoping crop residue trade-offs in Sub-Saharan Africa and South Asia. Field Crops Research 132, 175–184. van Donge, J.K., 1984. Rural-Urban Migration and the Rural Alternative in Mwase Lundazi, Eastern Province, Zambia. African Studies Review 27, 83–96. van Donge, J.K., 2005. The Farmer’s perspective: Values, incentives and constraints. In: Levy, S. (Ed.), Starter Packs: a Strategy to Fight Hunger in Developing Countries? Cabi Publishing, Wallingford, pp. 117–127. van Donge, J.K., Chivwaile, M., Kasapila, W., Kapongamgaga, P., Mgemezulu, O., Sangore, N., Thawani, E., 2001. A Qualitative Study of Markets and Livelihood Security in Rural Malawi: Module 2 part 2 of the evaluation programme for the 2000-2001 Targeted Inputs Programme (TIP), DFID/Ministry of Agriculture, Malawi, pp.1-142. Vaneph, S., Benites, J., 2001. First World Congress on Conservation Agriculture: A World-Wide Challenge , Madrid, October 1–5, pp. 1–21. Wagstaff, P., Harty, M., 2010. The impact of conservation agriculture on food security in three low veldt districts of Zimbabwe. Tro´ıcaire Development Review, 67–84. Wall, P.C., 1999. Experiences with crop residue cover and direct seeding in the Bolivian highlands. Mountain Research and Development 19, 313–317. Wall, P.C., 2007. Tailoring conservation agriculture to the needs of small farmers in developing countries. Journal of Crop Improvement 19, 137–155. Witoshynsky, M., Phiri, Z., 2000. The Water Harvester. Weaver Press, Harare. WorldBank, 2008. Agriculture for development, World development report 2008. WorldBank, 2010. Zambia: Impact Assessment of the Fertilizer Support Program, Analysis of Effectiveness and Efficiency, Report No. 54864-ZM, Sustainable Development Department Agriculture and Rural Development Africa Region, WorldBank. WorldBank, 2012. Conservation Agriculture: Zambia’s experience. Worldbank, Washington [FILM]. ZCATF, 2008. Conservation Agriculture toolbox for Zimbabwe (2nd version), pp. 1–32. ZCATF, 2009. Farming for the Future: A Guide to Conservation Agriculture in Zimbabwe, Zimbabwe Conservation Agriculture Task Force, Harare.