In Hot Water. A study on sociotechnical intervention models and practices of water use in smallholder agriculture, Nyanyadzi catchment, Zimbabwe

This study focuses on intervention processes in smallholder agriculture in the Nyanyadzi river catchment, located in Chimanimani district, Manicaland Province Zimbabwe. In particular it concerns itself with sociotechnical interventions that were implemented by Agritex, the local extension and irrigation service, in the mid-1990s. Despite a flurry of interventions and agrarian policies directed at the intensification of agricultural production and promotion of commercial agriculture in communal and resettlement areas, agricultural production has neither raised sufficiently nor in a sustainable manner.In this study intervention is taken as a measure to evoke a change in ordering practices of social actors, artefacts and natural elements by pursuing a model of how these three categories of actors might interrelate in a new way. Three models are researched in detail: the model of the smallholder commercial farmer as propagated in the master farmer training programme; the model of intensive smallholder irrigated agriculture in the case of a government managed smallholder irrigation scheme; the model of controlling water flows by means of conservation works and state management at catchment level of the Nyanyadziriver. In all three cases results have been disappointing to date: the master farmer programme has been ineffective in evoking widespread innovations in smallholder agriculture, the performance of the smallholder irrigation scheme has been low at high cost and finally siltation, land degradation and an increasingly fierce struggle over scarce river water have emerged in the catchment.The history of state interventions in the smallholder sector of Zimbabwe can be characterised by the presence of a strong state and a number of persistent themes and dichotomies, such as the prominence of the Land Question; recurrent swings in emphasis between development and control, modernity and tradition, and voluntary change and force; and a persistent belief in the potential ofthe mixed farming model to intensify agrarian production in the communal areas. Thus the post-Independence state adopted the very same technocratic policies that limited the possibilities to pursue the peasant option and had elicited widespread support for African nationalism and the ensuing liberation war. This study seeks to understand and explain these continuities in agrarian modernisation policies; qualify the impact of state interventions by paying explicit attention to alternative pathways of agrarian development pursued by smallholder farmers; develop an interdisciplinary understanding of water management and use at three hydraulic levels; and explore the possibilities and room for manoeuvre for reforms that calibrate policy discourse with local practice. It does so by engaging with the relationship between technology and society, informed by the actor­oriented approach, social construction oftechnology approach, and actor-network theory. The central research question guiding this study is:

'How did state engineered intervention models for agricultural modernisation of smallholder farming emerge, and which continuities and outcomes did these models produce at three hydraulic levels(field, scheme, catchment) in Nyanyadzi river catchment?'

A Native Commissioner, by the name of Keigwin, gave the first push towards the establishment of a govemment agency concerned with the segregated industrial and agricultural development of the African population in its own area. His plan in 1920 led to a training scheme for African instructors who were to demonstrate improved agricultural practices in the Reserves. With the appointment of E.D. Alvord, an American missionary, in the post of Agriculturist for instruction of natives in 1926, a more encompassing development scheme was embarked on, relying for its propagation on the concept of 'seeing is believing'The agricultural improvement package that Alvord developed provided a radical break with existing African agricultural practices, by maximising production per unit of land rather than per unit of labour. Whilst this break was considered necessary to allow the squeeze of the African population into the Reserves, facilitate administrative control and introduce the benefits of modem life to Africans, it may be questioned whether the resultant mixed smallholder farming model suited the social and ecological fabric of the Reserves. The elaboration of Alvord's philosophy of improving the livelihood of Africans led to the appointment of instructors, called demonstrators, in the fields of agriculture, community, development, home industries, forestry, irrigation, livestock and conservation. These demonstrators were tasked to develop African households and Reserves along an evolutionary path of modernisation.The global economic depression and rise of conservationist concerns in the 1930s shifted the emphasis of the demonstrator programme from livehood improvement to the prevention of destruction of natural resources. Disappointed with the limited spread of his voluntary change programme. Alvord in the 1940s succumbed to Conservationist pressures to enforce agricultural modernisation in the Reserves. The expanded agricultural Bureaucracy, left after Alvord's retirement in 1950, set out to impose the modernisation package by means of the Native Land Husbandry Act. However, African nationalist protests stopped its implementation in 1961,revertingthe initiative over African development to tradionalist Administrators that sought to implement the Rhodesian Front's strict segregationist policies.Thus African modernist aspirations that had been created by alvord, where squashed by a community development policy that was modelled on a reinvented role for tradional leaders.In chapter three it is shown how the persistence of the model was achieved by assessing in more detail the methods, ideas and practices that made the master farmer programme the pivotal element of state intervention at field level. Whether the model of modernisation was successful in creating a vibrant, modernist c1ass of African smallholders mimicking their large scale European counterparts is also assessed. The agricultural success and upward mobility of master farmers and early generations of demonstrators suggests Alvord's modernisation strategy paid off initially. However, rather than improving and developing the (communal) area they originated from, Christian master farmers and demonstrators used their agricultural wealth to escape the Reserves, investing in either purchase farms, irrigation or agri-businesses and the education of their children.Thus amodem elite emerged that was actively supported by both the pre- and post-independence state, but was oflimited value in terms of the productive or political support rendered. After independence the master farmer programme expanded in scope to facilitate a rapid transfer of technology revolutionise and commercialise the previously neglected smallholder sector. However, the standardised programme capitalised on a farming style that did not fit the ecological and socio-economic fabric of the communal areas and continued to cater for a limited number of wealthy smallholders (6% ofthe total) that had access to labour, cattle, capital and water.The second part ofthis chapter tums to Chimanimani district during the mid-1990s, assessing the meaning and effects of master farmer training, field days and agricultural shows in the different agro-ecological settings of Nyanyadzi river catchment. These three tenets of agricultural extension still formed the lynch pin of Agritex' mission to commercialise smallholder farmers using the image of their large-scale commercial counterparts. Despite their importance as heroes of progress and proof of Agritex' relevance, very little master farmer training was actually done by Chimanimani's extension workers. It is c1aimed that the model of the master fanner represents a mode of ordering (both material, social and ritual) that acts as an icon of modernisation. However, the trickle down effect (extension) to the masses of communal area dwellers did not occur. In two subsequent case studies involving master fanners the exclusionary features of the programme are highlighted. Thefustcase study assesses the relationship between field days and master farmers. Master fanners prove to be crucial actors for the reproduction and performance assessment of individual extension agents and their extension practices. Agricultural shows in Chimanimani, both at area and district level, have a highly ritual character where hardly any leaming takes place. The second case study demonstrates the active use that master fanners make of their relationship with extension workers as weil as their image as good fanners, in order to get access to state mediated resources, like land, water rights, loans and cattle. Aspiring to be a master fanner has less to do with wanting to practise agricultural methods that Agritex propagates, than with a desire to expand one's security and wealth by means of a master farmer badge.Thge latter serves as a ticket to state resources such as the heifer loan scheme aimed at restocking cattle heds in communal areas.The second part of the thesis (chapters 4 to 8) assesses the fTuits and fallacies of Nyanyadzi irrigation scheme as a model for intensive agricultural modemisation. In the intermezzo a new methodology is presented (technography) for the analysis of the life of an irrigated settlement scheme and the various actors involved in (re)shaping it. It is proposed to treat the social and technical aspects of settlement schemes as different but intemally related dimensions of a single object. Furthermore a strict ontological separation of the design of the irrigation scheme from its use is not helpful in analysing its success or failure, since 'closure' hardly occurs, i.e. the exact shape and use that is made of the water-network is subject to change. On the one hand settlement schemes are subject to iterative design processes informed by the disciplinary interests of its navigators (Engineers,Agricolas.and Administrators) and shifts in policy discourse or engineering paradigm. On the other hand,attempts to stabilise the scheme as a bounded entity are informed by the technologies of control that are devised by the management to control the behaviour of water and settlers, and counter-discours es and strategies of appropriation on the part of the users.The technography of Nyanyadzi irrigation scheme is further informed by three paradoxes conceming the role of irrigation in the emergence of sustained opposition against the govemment. Whilst Nyanyadzi irrigation scheme produced agric).lltural wealth for its modem users (in contrast to dry land fanners in the Reserves, who could not pursue the 'peasant option '), Nyanyadzi also became the scene of the most violent outbreaks of African opposition during the period of open Nationalism (1957-64). Thirty years later the agricultural success of the scheme had evaporated and the majority of users depended on drought relief hand-outs. Yet, the intemally divided Nyanyadzi irrigators persisted in their unified opposition against the govemment.Chapter four presents a first go at the base material, Nyanyadzi scheme, as conceived in a dream that Alvord had in 1926. A close look is taken at the construction and emergence of Nyanyadzi irrigation scheme as one of the most successful state engineered models for African modernisation under the vigorous leadership of its patron, Alvord. The technography starts with the rise and demise of the small MuNyanyadzi project that Alvord and his staff initiated in 1934 as a famine protection work and was washed away in a storm flood in 1942. This provided Alvord with a valuable leaming experience, which came in handy during the construction and settlement of the big Nyanyadzi project. Nyanyadzi provided Alvord with a key opportunity to realise his vision of agricultural modemisation. Whilst the resulting water­- network succeeded in eradicating famine and increasing agricultural production, the rural industrialisation policy, designed to cater for the unemployed agricultural workers and skilled professionals trom an emergent Atrican middle class, floundered. In the conclusion an assessment is given of the model of modemisation that Nyanyadzi scheme provided, strongly reflecting the central tenets of an irrigation factory. It is also shown how Alvord and his growing staff of Agricolas succeeded in crafting a successful water-network through the (re)alignment of various key elements, i.e. plot holders, crops, water and the market. The resulting water-network transforrned the existing Ndau society and physical landscape, by transferring wealth trom rain-fed cultivators to irrigators, whilst closing off opportunities for wet land cultivation in the Nyanyadzi catchment.During the next life-phase of the network (1950-67), treated in chapter five, the Agricolas focused on intensifYing and diversifYing irrigated production resulting in tangible increases in both the welfare and production attained by the plot holders. Yet, the aspirations of Nyanyadzi's wealthy plot holders went beyond that ofthe contented agriculturist, resulting in violent opposition during the period of open Atrican nationalist politics (1958-64). Instead of withdrawing, the Smith regime under the aegis of its Administrators tightened its hold over the Nyanyadzi water-network and its users, turning the scheme into a highly productive irrigation factory, yet failing to secure a dam to solve the network's persistent water woes. Events during the war of independence ultimately led to closure of the scheme, and sowed the seeds of future splits in the Nyanyadzi community. In the conclusion the life and transforrnation ofNyanyadzi scheme trom a famine protection work to an irrigation factory is reviewed by highlighting three recurrent eddies of thwarted hand-over of the scheme to its users within the general flow of increased state control. A pre-liminary answer to the Atrican nationalist paradox is provided. Both Agricolas and Administrators blamed the violent outbreaks of resistance on the unemployed and uneducated members of Nyanyadzi community. The latter had been instigated by urban politicians prying on feelings on insecurity that came with the threat of eviction, according to Agricolas. The Administrators blamed the umest on a lack of tribal cohesion and discipline. Yet Sithole (1970) and the Nyanyadzi nationalists themselves stress the leading role of Alvord's mission-educated modem men and women, who had acquired new wealth through irrigation and aspired better education, better wages and more business opportunities. Atrican nationalism provided both the voice and means to articulate these desires.Chapter six assesses the durability and demise of Nyanyadzi scheme as a state run venture, by investigating the intricacies of day-to-day management after Independence. The chapter opens with a review of govemment and donor agency initiated attempts to forrnulate a new irrigation management policy, stressing the benefits of increased cost recovery through user involvement in their management. The resulting (neo-liberal) policy discourse proved quite at odds with (futile) attempts by govemment staff on the ground to revive the scheme as a state run irrigation factory. The lack of a clear-cut policy statement on irrigation management allowed local govemment staff to push for such a revival, whilst head office staff engaged in lucrative donor mediated consultancies that sought to address this lacuna. Attempts to revive the Nyanyadzi factory failed in the face of natural and political opposition. It is argued that policy practice and discourse were enacted in two separate, though unconnected, policy arenas. Whilst massive donor support to and involvement with the smallholder irrigation sector produced an increase in both irrigated commands and managerial capacity on the part of Agritex, it did not contribute to a viabie model for handing over the management of these schemes to the users.The deadlock which emerged between the users and the management, and the complexity of the Nyanyadzi water-network is succinctly demonstrated by assessing attempts of both plot holders and the scheme's management to make the most of the dry winter season in 1995. Whilst a new water rotation schedule provided some reprieve, measures to reduce the crop acreage (one-acre rule) were ineffective, whilst continued break-downs and ultimate closure of the pump station seriously affected yields of the bean crop. Further complications were presented by the ineffectiveness of the irrigation management committee in regulating water distribution and the juggling act of gate keepers (bailiffs) that tried to negotiate their way between the list of official plot holders and de facto users of the scheme. The abrupt hand­over of the Nyanyadzi scheme to its politically divided users after 1995 was a direct result of the confluence of thnie currents, viz. the enactrnent of neo-liberal policies, a cash-strapped government, and the refusal of the plot holders to mobilise the resources required for the scheme's continued operation as an expensive government venture.Chapter seven focuses on the actual objects of irrigation based modernisation, the users. How did they experience the successive attempts to remould their livelihood? And in what ways did they actively re-shape, re-mould, and re-craft parts of the water-network to suit their own interests? First it is shown that irrigated agriculture in indigenous African schemes in East Africa is only a complementary activity to livestock or rain-fed production, whilst initial investments in the network's construction moderate its operation and maintenance (hydraulic property). The second section re-looks at the emergence of nationalist inspired resistance to the scheme's management in Nyanyadzi, and unravels how different farming strategies produced two types of plot holders that each had their own reasons to support the struggle for independence. By highlighting the effects of inheritance and investrnent patterns across different generations of Nyanyadzi plot holders, it is shown how political splits (n' the Nyanyadzi community after independence came to reflect different livelihood orien-tations and vice versa. An extensive plot survey in blocks A and C of the Nyanyadzi water-network reveals present-day differences in livelihood strategies pursued by plot holders of different religious and political orientation. Two cross-generational accumulation patterns and one recent politically informed accumulation pattem come to light, that produce a different understanding of how plot holders over the years have appropriated the Nyanyadzi water­network. These differences throw a new light on contemporary debates on small holder irrigation and livelihoods, impinging on the relationship between plot sizes and economic viability, the gender effects of irrigated production, the cultural dimension of risk avoidance, and tbe interplay between the sociotechnical environment of an irrigation scheme and the development of livelihood options by its users.Chapter eight is devoted to emergent modes of (re- )organisation that various users have developed to respond to and cope with the vagaries imposed by the crumbling official water­network. By means of three case studies it is shown how Nyanyadzi plot holders have re­appropriated and re-aligned some of the constituent elements of the scheme in order to make it work. The modalities of this re-alignment are informed by the two pattems of accumulation described in the previous chapter, and a third strategy that derives its strength from a combination of violence and political patronage (providing a prism to future events in Zimbabwe as a whole, after the start of the land occupations in 1999). The first case study is situated in block C where a traditionalist mode of organisation enables the successful implementation of main canal maintenance and crop marketing activities. Tbe second case study highlights the attempts of modernist plot holders and their urban kin to improve the water .supply situation of the scheme as a whole, through the mobilisation of donor funds to line the badly leaking main canal. Finally, a land conflict in block B highlights the failure of the traditionalleadership, Agritex, Rural District Council, and civil court to fend off a family of land claimants, who violently occupied the plot previously held by their evicted father. Ultimately, the heirs of the evictee succeed in chasing the legitimate plot holder off his land through the mobilisation of a politically inspired network ofZANU(PF) politicians.In the conclusion the three different modes of organisation are compared with emergent forms of organisation in other (irrigated) contexts in Zimbabwe and sub-Saharan Africa. It is shown that original inhabitants generally operate within a traditionalist mode of appropriation perceiving irrigated production as complementary to other livelihood pursuits, whilst the modernist strategy of investing in the education of sons and daughters that in turn re-invest resources through urban networks is quite common in West Africa amongst capita list commodity producers. Finally some lessons are presented that could inform the modalities of successful irrigation management transfer from government to water users. The Nyanyadzi case study stresses the need for a careful appreciation of locally available modes of organisation that go beyond formal organisational models (i.e. Water User Associations) currently promoted by (inter)national policy makers. Characteristic of such local modes of organisation is their simplicity and fluidity; kin mediated accountability mechanisms that drive the execution of essential tasks (i.e. water distribution, maintenance and marketing); propensity to minimise on energy costs and general avoidance of complex dependency relationships. User ownership of (parts of) the water-network is claimed, not on the basis of an official hand-over, but through the investrnent of various mixes of social (labour), cultural (chiefly authority), economic (money) and political capital.The third and final part of the thesis looks at water management at catchment level. The official model regulating water abstraction and use in times of scarcity has been fundamentally changed as a result of the Zimbabwe< water reform process (1994-98) that incorj>orated both local concerns (redress of historically grown inequities) asweIlas three giobally endorsed shifts in water governance. Thus the new Water Act (1998) and institutional constellation are based on river basins, rather than administrative units; decentralised management by (sub) catchment councils that .are manned by stakeholder representatives; and the introduction of water pricing as an allocative mechanism. Furthermore an integrated water resource management strategy guides the operational activities ofthe privatised Zimbabwe Water Authority (ZINWA).The basic idea for this thesis, the linking and integration of water use and management at three different hydraulic levels, was inspired by an upstream raid organised by the irrigation manager of Nyanyadzi scheme in May 1994. The raid represented an attempt to forcefully bring Nyanyadzi river water to the scheme's intake, leaving a swath of destroyed irrigation furrows in its wake. As shown in chapter nine, the raid and its destructive nature was legitimised in terms ofthe Water Act of 1927, which provided the officially endorsed mode of ordering water use at a catchment scale. This mode of ordering was initially geared towards capture of the resource water for Rhodesian settler farmers, through the application of the prior appropriation doctrine. This network produced discriminatory features during times of water scarcity and depends for its implementation on a strong monitoring agency that lacked in Nyanyadzi catchment. In contrast, the 'illegal' farmer initiated network of irrigation furrows builds on indigenous water and land use practices that are informed by a wide array of available cultural dispositions, that emphasise the public nature of river water. Whilst actively suppressed during settler mie, the network was quick to re-emerge and establish itself after Independence, presenting a formidable competing force with official water users in the catchment in the 1980s and 1990s. The upstream raids reflected attempts by the official water-network to get to grips with increasing competition over scarce Nyanyadzi river water. However, the post-independence politica I dispensation asweIlas the nature of flowing water worked against the emergence of an effective mechanism to share Nyanyadzi's water amongst its different users. The only tangible effects produced by the raids were the creation of awareness amongst the various water users of their hydraulic interdependency, and the gradual inclusion of furrow irrigators in the official network (through the allocation of more water rights).By means of a case study on the development and management of a network of irrigation furrows in the remote Ruwedza valley, located in the upper end of Nyanyadzi catchment, the strengths and weaknesses of the indigenous paradigm on catchment water management arehighlighted. The paucity of state interventions in Ruwedza have allowed the emergence of water use principles and practices that come close to the professed ideals of integrated land and water management pursued by international policy actors. Various social, material, and agro-ecological characteristics of the furrow network pay tribute to the capricious behaviour of water. The temporary and permeable nature of the used infrastructure helps to distribute water equitably, whilst minimising flood damage. The flexible and distributed spatial set-up of furrow intakes along the river helps to maximise capture of available water, whilst minimising conflicts amongst different furrow irrigators, partially through the practice of staggered planting in the valley in response to altitude related frost risks. During water scarcities the limited supply is distributed equitably through the operation of time-based water rotas that are moderated by the headman, whose furrow is situated at the bottom of the valley. Vet the network is limited in span and subject to collapse during periods of duress, triggered by the extreme water scarcities experienced in 1992 and 1995 and the increased population densities caused by ongoing settlement of new corners. In conclusion, sh!!tPcontrasts can be observed between the official network at catchment level that operates on notions of state ownership of water, prior application, and a hard system conception of water use, and the farmer furrow network that is informed by principles of riparian appropriation of a God-given resource, proportional allocation, and network conceptions of flowing water. It is observed that the new Water Act (1998) and institutional constellation that were produced by the Zimbabwe water reforms are iII-equipped to bridge the gap between legality and reality of water use in Nyanyadzi catchment.In the overall conclusion, first the emergence, continuities and outcomes produced by the three state engineered mode Is for agricultural modemisation of small holder agriculture are presented. It is observed that the emergence of nationalist opposition in Nyanyadzi was not unique but resembles the opposition movements that sprang forth in the Mwea scheme in Kenya and the Office du Niger in Mali. Thus rather than irrigation works feeding the emergence of despotic regimes (Wittfogel 1957), irrigation schemes on the African continent have provided the impetus for the overthrow of despotic regimes. Secondly, some pointers are provided for an alternative way of analysing and crafting water-networks drawing on the strengths of an interdisciplinary focus engrained in technography. It is argued that new approaches should focus on flow rather than control, favour network over hard systems conceptions and draw on existing practices and modalities through a process of in situ prototyping and negotiation. Finally the likely impact of the Zimbabwe crisis on smallholder agriculture is provided by presenting the winners and losers of the crisis and assessing the future of agrarian modemisation in Zimbabwe. It is observed that the present crisis spelIs doom for the state modemisation project and those associated with it, whilst it negatively affects the neo-liberal and democratie foundations of the water reforms. Vet the very ineffectiveness of the new forms of water govemance may open alternative avenues for user based modes of organising the use and distribution of water, as described for the Ruwedza case.