by
L. Loewen-Rudgers, E. Rempel, J. Harder and K. Klassen Harder
Mbeya Oxenization Project, Mbeya, Tanzania
Abstract
The Mbeya Region in Tanzania is relatively fertile and contributes significantly to the national production of maize, the major food crop. Inadequate weeding is considered the main factor limiting maize yields. Animal-drawn weeders could do much to alleviate this. At present only 15% of the 200,000 smallholder farmers in the Region own cattle, and the majority of these do not use draft animals for weeding. One objective of the Mbeya Oxenization Project is to increase the use of animal weeding by farmers already plowing with cattle.
Following a literature review, baseline survey and discussion with farmers, nine major constraints to the adoption of animal traction weeding have been identified. In descending order of importance these are:
non-availability of implements;
poor implement quality;
inadequate repair services at village level;
previous emphasis on inter-row rather than over-the-row cultivators;
poor extension;
lack of communication between manufacturers and farmers;
poor timeliness of operations;
inadequate training of animals;
fear of crop damage by animals and implements.
Since the specific target group comprises ox-using farmers, it is considered that four possible constraints are not significant. These are: limited capital, overriding risk avoidance, gender division of labour and too complicated technical packages. However, these could be constraints to farmers adopting animal traction for the first time.
The project hopes to overcome the problems relating to implement supply, design and quality by working with local manufacturers. The problems of repairs may be solved with initiatives designed to support and train village artisans. Solutions to the constraints relating to on-farm practices are probably known by some progressive farmers. Working in villages with "contact farmers" should identify appropriate solutions and may stimulate the interest of other farmers. This extension approach, based on innovative farmers and inter-farmer information flows, should be effective and sustainable.
Background
The population of the Mbeya Region is 1.1 million, 5% of Tanzania's population of 22 million. The Region's land area is six million hectares or 7% of Tanzania's land area of 89 million hectares. Approximately 200,000 predominantly smallholder farm families cultivate 385,000 ha out of the total 2.8 million arable hectares. Corresponding figures for Tanzania are 2.25 million farm families cultivating 6.2 million hectares out of the total of 39.5 million arable hectares (Croon, 1982; EIU, 1987).
Soils in the Region vary considerably. There is a high proportion, perhaps one third, of Inceptisols (USDA) containing recent volcanic ash (in the FAO system of classification they are known as Cambisols and Andosols). These have higher fertility, water-holding capacity and pH levels than the more typical "tropical soils" such as Oxisols (Ferrasols) which occupy perhaps one third of the land area of the Region. About one third of the area is occupied by sandy Entisols (Arensols and Regosols). Average annual rainfall in the Region is rela tively high (1,300 mm) but again varies considerably (600 to 3,600 mm). The region is relatively high (1,300 m) but this also varies considerably (500 to 2,800 m) (Rombulow-Pearse and Kamasho, 1982).
Variation in precipitation, elevation and soil type enables the Region to produce a wide variety of food and cash crops including maize, beans, bananas, coffee, tea, rice, groundnuts, cotton and wheat (Rain, 1984a). The relatively high precipitation and better soils enable the Mbeya Region to export food to other regions in Tanzania and occasionally to other countries (Croon, Deutsch and Temu, 1984). Tanzania is frequently self-sufficient in maize, the major food crop, and sometimes exports it as in 1987 (EIU, 1987). In 1987 the Mbeya Region, with only 5% of Tanzania's population, produced 324,000 tonnes of maize, that is 14% of Tanzania's total production of 2.36 million tonnes. Smallholder farmers grow most of their maize in rows with little intercropping.
Animal traction status
Ox-drawn mouldboard plows were introduced into the Mbeya Region of Tanzania during the 1930s (Kjaerby, 1983). Although there have been few animal traction development programmes in the Region (or in the rest of Tanzania), conditions seem favourable. There are high cattle numbers, surplus land for area expansion and relatively high production levels of cash crops of both food and nonfood types. Nevertheless, only between 10 and 20% of farming households own oxen, the most common draft animals (MRIDEP, 1987). This is a figure similar to that for all of Tanzania (ILO, 1987b; Kjaerby, 1983; Starkey, 1988b). The proportion of farming households owning work cattle varies greatly within the Mbeya Region from 0% in areas producing no cash crops to 66% in areas producing cash crops. Overall use of draft animals (including those households which borrow or rent cattle) varies from 0% to 93%.
Weeding with draft animals
Starkey (1986 and 1988b) reports that although animal-drawn weeders are available in most African countries, only 5% of those farmers utilizing animal traction for plowing use animal-drawn weeders in row crops. Figures cited vary from around 0% in Botswana, Mozambique, Uganda and Zambia, to between 10% and 20% in Cameroon and Mali, and finally to as much as 20% to 40% in South Africa and Zimbabwe. Numerous other references report low adoption of animal traction weeding in Africa by farmers utilizing animal traction for plowing (Anderson, 1985; Barrett, Lassiter, Wilcock, Baker and Crawford, 1982; EFSAIP, 1984; Francis, 1986 and 1988; ILO, 1987a; Jaeger, 1984; Kjaerby, 1983; Smid, 1982). The percentage of farmers using animals for both plowing and crop weeding is nearly zero in the Mbeya Region (Croon et al., 1984; Rain, 1984a) and the situation is similar for Tanzania as a whole (ILO, 1987b; Kjaerby, 1983).
Acland (1971) and Terry (1984) report that maize in East Africa should be kept free of weeds for the first month after emergence, should be weeded three times: when plants are 5-10, 45 and 90 cm high. They also say that if maize growth is checked by weeds shortly after emergence, it never fully recovers. In five experiments in the Southern Highlands of Tanzania, Croon et al. (1984) found that one weeding of maize at 10 cm resulted in an average yield of 4.2 tonnes per hectare compared to 2.3 tonnes per hectare when maize was not weeded at all. They also reported that poor weeding of maize is the biggest constraint to maize production in southern Tanzania. They suggested that timely weeding was itself more important than use of improved varieties, fertilizers, insecticides or timely planting. In a survey of 320 farmers in 20 villages in the Mbeya Region, over 50 constraints to crop production were listed, and the four most important were insufficient credit, lack of hand hoes, damage by wild animals and late weeding. Lack of good animal-drawn weeders was listed as the main constraint to crop production by those farmers who plowed with animals (Rain, 1984b).
Numerous researchers report that introduction of animal traction plowing without animal traction weeding increases labour productivity through expansion of area planted but decreases yield per hectare because of insufficient labour for the timely weeding of the larger crop area. The same researchers report that animal plowing without animal weeding limits the effectiveness of animal traction farming and slows the overall rate of adoption of animal traction technology (Anderson, 1985; Francis, 1986; Jaeger, 1984; Kemp, 1987; Kjaerby, 1983; Rain, 1984a; Smid, 1982; Starkey, 1981).
Considering the apparent benefits of timely weeding through use of animal traction, particularly by those farmers plowing with animals, it is difficult to understand the low level of adoption of animal traction weeding technology.
Sources of information
In the following sections, nine of the most crucial current constraints to the adoption of animal weeding technology in the Mbeya Region of Tanzania are reviewed, and ways in which those constraints might be lessened are discussed. There have been three main sources of information for this analysis.
· First, a review has been made of the available literature on animal traction in Africa.· Second, use has been made of the results of a highly structured survey of 511 smallholder farmers in areas of the Mbeya Region having the highest utilization of animal traction. This survey collected baseline data to guide the implementation of the Mbeya Oxenization Project (MOP). This is the most recent programme in the Region designed to increase smallholder use of animal traction, and it is supported by the Government of Tanzania, the Canadian International Development Agency (CIDA) and the Mennonite Economic Development Associates (MEDA).
· Third, information has come from the informal questioning of 300 farmers during 18 visits to 11 visages for the preliminary on-farm testing of prototype animal-drawn implements.
Possible limitations and dangers
Three important factors may limit the accuracy and applicability of the prioritized list of constraints. Thus the list must be considered as "tentative".
Non-typical farmers
Although it is intended that later phases of the Mbeya Oxenization Project will have broader objectives, the first four-year phase of the Project (begun in July 1987) is specifically attempting to increase the use of animal-drawn equipment by those farmers who are now using animal traction for plowing. Emphasis is placed on maize weeders and carts. The MOP is therefore working with farmers already using animal traction and with the existing public and private infrastructure. MOP is facilitating the manufacture and marketing of implements that have minimal imported components, and hopes to extend animal traction technology to more farmers. However, since farmers participating in the initial survey and in the preliminary testing of implements are likely to be wealthier than typical smallholder farmers, they may not be subject to the same constraints as average farmers. For example, the cost of implements and the availability of credit may be less crucial to them.
Farmer courtesy
The true intentions and feelings of smallholder farmers are reflected far more in their actions than in their words. When demonstrating a new implement or discussing whether a new practice will be adopted, the development worker, whether a national or an expatriate, is usually surrounded by friendly enthusiasm, co-operation, courtesy and affirmation and receives the answers he or she wants to hear. These may actually be the exact opposite of the farmers' true intentions or feelings. This may hinder the development worker from assessing accurately constraints to the adoption of a technology. Moreover, it may become a constraint in itself through introduction of inappropriate methods. Starkey (1988a) points out that although animal-drawn wheeled tool-carriers functioned well and were nearly always well-received when demonstrated to farmers, very few of the hundreds given or sold to farmers in Africa in recent years are still in use today.
Realistic time scales
An early assessment of constraints to adoption of animal traction weeding technology is necessary so that the more appropriate methods for alleviating the constraints can be introduced at the outset of the programme. However, early assessment is probably not as accurate as later assessment. Unfortunately, it may be five years before it can be seen whether animal-drawn weeders are being purchased in significant numbers, and it may be ten years before it can be seen whether those weeders are still being used. The current methods for transferring animal traction weeding technology will certainly have to be judged in the long-term. Nevertheless, decisions have to be taken now which will influence the success of the technology transfers envisaged.
Nine crucial constraints
The constraints to adoption of animal traction weeding technology in the past and those anticipated in the future are discussed below, in descending order of severity. The subjective nature of much of the information and the relatively small differences in the severity of some of the constraints, means that the ranking is only qualitative and tentative. The discussion focuses on animal-drawn tine cutivators for maize. However, other animal-drawn implements may also be of relevance for weeding and soil conservation, particularly in the hillier areas of the Region. Ridgers, instead of tine cultivators, are used by some farmers for the second or third weeding of maize at 60-90 cm. They simultaneously weed and cover the second application of nitrogen fertilizer. Implements for making tied ridges may also have potential in hillier areas for simultaneous weeding and water and soil conservation at the time of the cultivation (at 90 cm). However, the following constraints refer primarily to the potential for the adoption of tine-cultivators.
Equipment availability
Weeding implements have not been distributed to stores close to farmers. In the literature consulted, there is little evidence to suggest that availability of implements at village level is a problem elsewhere in Africa. Indeed Starkey (1986) implied that it is not a serious problem when he stated that weeders were available in most countries. However, during informal questioning during prototype testing in villages in Mbeya Region, many farmers indicated that they did not have weeders because they had never seen any and did not know where to buy them. In the highly structured survey of 511 farmers in the Mbeya Region, the most popular place for purchasing animal-drawn implements (usually plows) was the regional capital Mbeya, often over 50 km and sometimes over 100 km away. Farmers may be willing to travel such distances for the essential, and more familiar, plow but not be willing to travel for the more unfamiliar weeder. Informal questioning of farmers indicates that if weeders were available in village stores, they would be more inclined to adopt animal traction weeding technology. Greater demand at the village level would then encourage the distribution of weeders from larger centres which then might in turn encourage increased supply through local manufacture or importation. The poor supply of animal-drawn implements in Tanzania is referred to several times in the literature (Croon et al., 1984; ILO, 1987b; Kjaerby, 1983; Rain, 1984b). However, it may be that this is primarily a problem of local distribution rather than national supply. For the past 20 years, the Government of Tanzania has been largely responsible for the marketing of animal-drawn implements. The distribution of weeders might be improved by helping the private sector manufacturers and retailers. This may be more possible now than it would have been a few years ago as the Government now has a more liberal attitude towards free enterprise (EIU, 1987).
Equipment quality
The quality of the available cultivator is poor. Most cultivators available in Tanzania are of the Cossul model from India or a similar model from Zambia. Several reports indicate that using such cultivators on stony or stumpy land leads to the breakage of cast iron parts and the bending of soft steel components (ILO, 1987b; Kjaerby, 1984). This has been the case in several cultivators observed during village visits by the MOP. About half of the 300 Cossul cultivators provided by the EEC in the early 1980s for purchase by farmers at 14 ox-training centres in the neighbouring Iringa Region remain unsold. This is partly because farmers became quickly aware of the poor quality of the cultivators (Massunga, 1988).
The introduction of higher quality cultivators would seem the most logical method to alleviate this constraint, but these are likely to be more expensive. Other approaches might include educating farmers not to use the cultivators on stony or stumpy land, facilitating establishment of village-repair services (blacksmiths), and/or making spare parts available in villages. Few, if any, such repair services exist in the villages and spare parts can be obtained only by cannibalizing unsold new cultivators located far away.
Repair services deficiency
Village-level repair services and spare parts are almost totally lacking in the Mbeya Region. The need for local craftsmen who can repair cultivators and other animal-drawn implements, was emphasized in the previous section. Several literature references suggest that the adoption of animal traction technology in Africa depends to a large degree on village artisans who can repair and even manufacture implements (Anderson, 1985; Haug and Gerner-Haug, 1982; ILO, 1987a; Pingali, Bigot and Binswanger, 1987). Such craftsmen working with smallholder farmers could become enthusiastic research and development teams. They would be more in touch with the needs and wants of the farmers than could any parastatal manufacturer, government research institution, extension organization or donor-sponsored development project. There would be a need for development projects to facilitate the establishment of village craftsmen-farmer teams through the provision of credit and technical knowledge. This type of development is what Bunch (1982) refers to as "participatory" rather than the more typical "paternalistic" (giving ad and doing all) development. This latter has seldom resulted in a sustainable transfer of technology.
It is interesting to note that in Europe and North America farmers themselves, in conjunction with local craftsmen, manufactured hand tools and animal-drawn implements. Even after some local craftsmen developed larger factories for animal-drawn (and later tractor-drawn) implements, farmers went to local blacksmiths who made spare parts and repaired implements. Only within the last 30 years have most local blacksmiths disappeared. Only recently have large implement manufacturers together with local implement dealers provided most new implements, spare parts and repair services. It should also be noted that most of the "development" in European and North American agriculture occurred within the private sector. Governments provided increasing levels of assistance in terms of research, extension and transportation infrastructure, but they seldom became involved in implement manufacture or marketing.
Emphasis on inter-row cultivation
Emphasis has been placed on inter-row as opposed to over-the-row cultivation. Inter-row cultivators, as introduced by many animal traction development programmes in Africa, are somewhat difficult to operate and often do not kill all weeds within the crop row. Thus weeding with a hand hoe is necessary after animal cultivation. Difficulties in operation arise from the wide weeding yoke which prevents the animals from working as a team and makes them difficult to steer. Successful operation of the inter-row cultivator also requires planting in parallel rows which necessitates either the use of relatively expensive animal-drawn planters or time-consuming systems of accurate spacing during hand planting. Even when rows are parallel, steering is difficult because the operator must look at two crop rows simultaneously. Finally, to avoid crop injury, particularly when rows are not exactly parallel, the cultivator is often kept too narrow to throw soil on top of small weeds growing within the crops rows.
Roosenberg (1987) suggests that over-the-row cultivators might be adopted more quickly by farmers because they are easier to operate and throw enough soil into the crop rows so that later weeding by hand is seldom necessary. Over-the-row cultivators are easier to steer because the same narrow yoke can be used as for plowing or carting. The operator only has to look at one row at a time. Further, costly or time-consuming planting of crops in exactly parallel rows is not necessary. Nevertheless, it is more difficult to design an affordable over-the-row cultivator since they generally require wheels and a heavier construction.
Poor extension
Extension of animal traction weeding technology has been almost nonexistent. While a few small local and donor-assisted development programmes have been attempting to extend animal traction plowing and transport technology to Mbeya Region farmers, any efforts to extend animal weeding technology have been minimal. Lack of such extension efforts may have resulted from the great expense of traditional extension methods which require the training of numerous extension workers to go into villages to train farmers. On the other hand, the very immensity of the task of transferring the difficult-to-learn technology of inter-row cultivation may also have contributed to the lack of extension in this field.
During preliminary testing of prototype over-the-row cultivators, reaction was more favourable in villages where one or more progressive farmers were successfully using inter-row cultivators. Farmers were less enthusiastic about weeding with oxen in the villages where farmers had never seen a cultivator in use. This not only demonstrated the necessity of good extension, but also that progressive farmers themselves may very well be the most effective extension personnel. The most desirable extension approach might be one in which animal traction weeding technology is transferred as inexpensively and as quickly as possible to a few progressive "contact farmers" in each village who in turn could extend the technology to other farmers with little further involvement of expensive projects or government extension personnel.
Attempts to transfer animal traction technology to "contact farmers" at 14 oxen-training centres in the neighbouring Iringa Region are failing because progressive "contact farmers" do not want to leave villages to attend sessions in oxen-training centres. These training sessions by necessity must be held during the very times when farmers need to remain on the farm to plow, plant and weed. In Iringa, attempts are also failing because the oxen-training centre buildings and personnel cannot be maintained now that the donor agency has withdrawn support (Massunga, 1988). A less ambitious oxen-training centre approach in the Mbeya Region is also failing.
It would seem that animal traction weeding technology could be transferred more effectively by externally supported development programmes if less money were spent on building institutions such as oxen-training centres. Instead, more money should be spent on frequent village visits by expatriate development workers and their national counterparts from established extension institutions. They should endeavour to train "contact farmers" on their own farms. In this way, the "contact farmers" should be able to continue extension efforts with some help from Tanzanian extension personnel after the end of the expatriate-assisted development project.
Marketing difficulties
The failure of farmers to express disquiet may result in the marketing of unwanted cultivators. To take an example: the animal-drawn wheeled toolcarrier was engineered well and farmers said they liked it. However, obviously they did not really like it, because despite being distributed to farmers, few were ever used for any length of time (Starkey, 1988a). The Tanzanian smallholder farmer is reluctant even to express displeasure when a Tanzanian extension worker attempts to introduce a new practice or piece of machinery the farmer does not like. The farmer is even more courteous when foreign development workers do so. A good implement may function well, increase the farmers' productivity and fulfil their needs. Nevertheless, it would be unwise to invest time and money in manufacturing and marketing large numbers of such an implement if the farmers do not like the implement well enough to use it. To avoid this mistake with new weeding implements, it might be advisable to demonstrate and test each implement several times in each of 10 to 20 villages. If this appears successful each implement could be test-marketed in small numbers (100). Finally, it would be important to return to the villages to see if farmers are actually using the implement before the production and marketing of large numbers should be contemplated. Although the main object of this exercise would be to determine farmers' true demands, occasionally it might be possible to change farmers' wants (extend a more appropriate technology) so that their wants are more compatible with their needs.
The MOP should be prepared to facilitate the production and marketing of a "less desirable" implement (an inter-row cultivator) as opposed to a "more desirable" one (an over-the-row cultivator) in the event that farmers do not like the "more desirable" option. The adoption of "less desirable" cultivators would seem better than total rejection of animal traction weeding technology.
Poor timing in cultivation practices
Farmers seem to weed late, so that animal-drawn cultivators do not function well and potential yield increases are not realized. For example, from the informal questioning of 300 farmers in 11 villages of the Mbeya Region it was concluded that only 10% of farmers begin weeding maize by hand-hoe when the crop is 15 cm tall (just slightly after the recommended time). Approximately 60% begin weeding when the crop is 30 cm while the remaining 30% begin when the crop is 45 cm or taller. Only a few farmers felt it necessary to begin weeding earlier. About half indicated they would purchase a cultivator (at the current price of 2,400 Tanzanian shillings) if available, but primarily to relieve labour constraints rather than to begin weeding earlier. In fact, some farmers indicated they would not buy the inter-row or over-the-row cultivators being demonstrated because those cultivators would not remove weeds that were 30 cm high. It is likely that if animal weeding technology were introduced to farmers without emphasizing the importance of early weeding, the technology would be rejected in the long run. This is because few cultivators function well when weeds are 30 cm high and overall yields would be no higher than those resulting from the usual untimely hand weeding. Relieving labour constraints without increasing yields would justify only a low level of adoption of animal weeding technology.
It is evident that to facilitate adoption of an acceptable level of animal-drawn weeders, timely weeding will have to be an important component of the "contact farmer" extension approach. In demonstration plots on their own fields, farmers will have to be shown that cultivators function better when weeding begins early. It will also have to be demonstrated that timely weeding with, or without cultivators, results in higher yields than weeding at the usual time (30-45 cm).
Training difficulties
The difficulty of training oxen to follow crop rows may limit adoption of animal-drawn weeders. There has been considerable disagreement as to the best approach to use in training and guiding oxen, particularly for use in row crop cultivation. Farmers in Tanzania using oxen for plowing use one person to control a yoked pair using voice commands initially taught through varying degrees of whipping and encouragement. Several oxenization programmes have attempted to improve upon this, particularly for row crop cultivation, by using nose rings or halters with some kind of rope connection back to an operator. Initially, an additional person may have to lead the animals in front, particularly for row crop cultivation. This approach is being advocated by the Uyole Agricultural Centre near Mbeya and is taught at the various oxen-training centres in the Iringa and Mbeya Regions (Massunga, 1988; Shetto, 1988).
From the informal questioning of farmers in 11 villages it was found that those farmers who were unenthusiastic about animal traction cultivation felt that it would be difficult to train oxen to follow crop rows. In contrast, those who wanted to buy cultivators felt that oxen could be trained to follow crop rows in several days using the voice command system. The views of Conroy (1988) agree with those of farmers in the Mbeya Region who are enthusiastic about oxen row crop cultivation. Conroy reports that the use of nose rings, ropes, halters and bridles is unnecessarily complicated and that the most effective approach is the use of a simple yoke, voice commands and a goading stick. Unfortunately, he does not mention whether this method is suitable for row crop cultivation.
Obviously, a successful animal weeding technology extension package must include one or more effective methods for training animals to follow crop rows. Since it is not clear just what those effective methods might be, it may be best to go to the animal traction farmers themselves for the answers. By observing how the "contact farmers" who are testing prototype cultivators presently train their oxen to follow crop rows, it is hoped that a successful guidance system will be found. This could be incorporated into the animal weeding extension package. It is to be hoped that farmers will want over-the-row cultivators since animals pulling these should be able to learn more easily to follow crop rows. It is possible, however, that this constraint will turn out to be far more severe than is now envisaged.
Fear of crop damage
The great reluctance of farmers to tolerate visible damage to their crops caused by animals may limit the adoption of animal-drawn weeders. Farmers do not seem to appreciate the extent to which weeds can cause invisible crop damage. Farmers are extremely fearful that animals will eat and trample crops during cultivation, or that the cultivator may uproot crop plants or cover them with soil. Many of these fears can be decreased to an acceptable level through teaching farmers to place muzzle baskets over the mouths of animals and ensuring that animals are properly guided to follow crop rows. However, great extension efforts will be required to convince farmers that although over-the-row cultivators may cover a few crop plants, the increased yield and/or the decreased need for hand weeding resulting from covering weeds should more than compensate for the visible crop damage. Nevertheless, it would seem that the extreme fear of crop damage rules out the possibility of introducing the use of harrows to remove small weeds just after crop emergence. This was a common weed control method in maize and beans in North America prior to the use of herbicides.
Less critical constraints
Several constraints to the adoption of animal traction weeding technology commonly discussed in other situations are not considered to be crucial constraints within MOP. This is because MOP is presently attempting to transfer the technology to moderately wealthy smallholder farmers who are already using animals for plowing. However, while the following four constraints may seem unimportant to the present target group, they may become critical when attempts are made to transfer animal-drawn weeding technology to farmers who are adopting draft animals for the first time.
Capital and credit
Numerous papers list expensive implements or inadequate credit as important constraints (Anderson, 1985; Barrett et al., 1982; ILO, 1987b; Jaeger, 1984; Kjaerby, 1983). However, this should not be an important constraint for MOP as the cultivators being considered are inexpensive and the current target farmers are comparatively wealthy. (During village visits several farmers even asked if they could purchase the project vehicle!) However, credit assistance may be necessary for establishment of small private workshops and repair services.
Risk avoidance
Several sources suggest that improved technology is very difficult to transfer to typical smallholder farmers because they operate mainly in a non-market economy. In such cases, maximizing the security of mediocre yields is most important. Risking the possibility of little or no yield because of some innovation which might result in a high profit is seldom considered (Hyden, 1980; Lappe and Collins, 1977). However, most farmers using animals for plowing in the Mbeya Region are operating very much within the market economy and should be willing to take risks such as adopting animal weeding in order to increase profits by decreasing their labour costs and/or increasing their crop yields.
Gender issues
Several literature references suggest that animal weeding technology might not be adopted for maize because although it is a "men's crop", hand weeding is "women's work" and driving oxen is "men's work" (EFSAIP, 1984; ILO, 1987a; Kalb, 1982). Certainly among farmers using animals for plowing in the Mbeya Region, driving oxen is "men's work". However, results from the highly structured survey of 511 farmers in 18 villages suggested that nearly 50% of the work force for hand hoeing maize is male. It would therefore seem that men would want to adopt animal traction weeding technology if for no other reason than to decrease their own labour burdens.
Complicated packages
Many oxenization programmes have attempted to introduce at one time "complete packages". These packages have included the animals, the training of animals, the training of farmers, all soil-moving implements (often complicated multipurpose implements) and animal-drawn carts. It is often argued that adoption of "complete packages" is necessary in order to realize the full potential benefits of animal traction technology. This may be true, but simultaneous introduction of all components of a complicated package seems too expensive and too complex for easy adoption by smallholder farmers. In such complicated projects, the adoption of any one component, such as weeding technology, might not be achieved simply because it "gets lost in the shuffle". Croon et al. (1984) in their description of what has become a very successful maize improvement programme in the Mbeya area of Tanzania point out that farmers do not and cannot adopt complete improvement packages at one time. Instead, the components of the package must be prioritized and then introduced step by step. Starkey (1986; 1988a) indicates that in designing animal traction development programmes, the most limiting factor should be tackled first and that introduction of complicated multipurpose implements should be minimized. This approach is being used in the Mbeya Oxenization Project. Farmers using animal-drawn plows have identified inadequate weeding as the key limiting factor in the production of maize and some other crops. The animal-drawn cultivators being introduced are single-purpose implements mainly because the target farmers already have single-purpose plows.
Conclusion
During its initial survey work and discussions with farmers, the Mbeya Oxenization Project identified weed control in maize as a primary limiting factor. The project then considered nine factors that might be restricting the adoption of animal-drawn weeders. The project hopes to overcome the problems relating to implement supply, design and quality by working with local manufacturers. The problems of repairs may be solved with initiatives designed to support and train village artisans. Solutions to the various constraints relating to on-farm practices are probably already known by some of the more progressive farmers in the area. Working with such "contact farmers" in the villages will not only identify appropriate means of overcoming the constraints, but may also stimulate other farmers. Farmer training should be village-based and frequent village visits by development workers and extension personnel are therefore required. This extension approach, based on innovative farmers and inter-farmer information flows, should be effective and stimulating. It should also be sustainable in the long term, when external assistance to the project ceases.
Résumé
La région Mbeya en Tanzanie est relativement fertile et produit une part importante de la production nationale de maïs, l'une des cultures principales du pays. L'inefficacité des techniques de désherbage est perçue comme la contrainte principale de la culture du mais à haut rendement. Des sarcleuses attelées pourraient grandement palier ce problème. Mais parmi les 200.000 petits fermiers de la région, seulement 15% possèdent du bétail, et la plupart sarclent d la main. L'un des objectifs principaux du Mbeya Oxenization Project est de développer la pratique du désherbage mécanisé sur les exploitations utilisant des animaux de trait pour le labour.
Des recherches bibliographiques, des enquêtes et des discussions avec les paysans, ont permis d'identifier neuf contraintes principales d l'adoption de la traction animale. Classés par ordre décroissant d'importance, ce sont:
le manque d'équipements;
la pauvre qualité des matériels;
la faiblesse des services d'entretien au niveau des villages;
l'utilisation de cultivateurs inter-sillons;
la faiblesse de la vulgarisation;
le manque de communications entre fabricants et fermiers;
la mauvaise périodicité des opérations;
un dressage lacunaire des animaux;
la conviction des fermiers que les équipements et les animaux endommagent les récoltes.
La présence d'exploitations en culture attelée dans le groupe cible permet de considérer quatre contraintes possibles comme négligeables. Ce sont: la faiblesse du capital d'investissement, le haut niveau de risque, la division sexuelle du travail, la complexité excessive de la technologie. Ces contraintes demeurent toutefois réelles pour les fermiers postulant d la traction animale.
Le projet espère résoudre les problèmes concernant la fourniture des équipements, leur conception et leur qualité en coopérant avec les fabricants locaux La question de l'entretien et des réparations peut être résolue par le soutien et la formation des artisans villageois. Certains fermiers plus avancés connaissent probablement les solutions aux contraintes affectant les pratiques culturales. Un travail coopératif au niveau des villages avec des fermiers sélectionnés contribuera d'identifier les solutions viables et à stimuler l'intérêt des autres paysans. Les stratégies de vulgarisation soutenues par des fermiers novateurs et le partage des connaissances entre les exploitations devraient permettre un développement efficace et continu.
References
Acland J. D. 1971. East Africa crops. Food and Agriculture Organization of the United Nations and Longman Group, London, U.K. 252p. (E).
Anderson F. M. 1985. Draught animal power systems in Sub-Saharan Africa: their production impact and research needs. pp. 26-31 in: Copland, J.W. (ed) Draught animal power for production. Proceedings of an international workshop held at James Cook University, Townsville, Qld, Australia, 10-16 July 1985. ACIAR Proceedings Series 10, Australian Centre for International Agricultural Research, Canberra. 170p (E).
Barrett V., Lassiter G., Wilcock D., Baker D. and Crawford E. 1982. Animal traction in Eastern Upper Volta: a technical, economic and institutional analysis. Michigan State University Development Paper. East Lansing, Michigan, USA. 118p. (E).
Bunch R 1982. Two ears of corn. A guide to people-centered agricultural improvement. World Neighbours, Oklahoma City, Oklahoma, USA. 251p. (E).
Conroy D. 1988. The traditional ox team and its yoke.. Tillers Report 8, 1: 1-5. (Spring 1988). Tillers Small Farm Program, Kalamazoo, Michigan, USA. (E).
Croon I. 1982. Agricultural development strategy and area based projects by agro-ecological zone in Mbeya Region. RIDEP Report No. 42. Mbeya Rural Integrated Development Project (RIDEP), Mbeya, Tanzania. 64p. (E).
Croon I., Deutsch J. and Temu A.E.M. 1984. Maize production in Tanzania's Southern Highlands: current status and recommendations for the future. Centro Internacional de Mejoramiento de Maíz y Trigo (CIMMYT), Londres, Mexico. 110p. (E).
EFSAIP 1984. Final report No. 8 1976-1984. Evaluation of farming systems and agricultural implements project (EFSAIP). Ministry of Agriculture, Gaborone, Botswana. 403p. (E).
EIU 1987. Tanzania: country profile, 1987-1988. The Economist Intelligence Unit, London, UK. 34p. (E).
Francis P. 1986. The impact of ox draft power on small-scale agriculture in Mpika District, Northern Zambia. Occasional Paper No.7, Integrated Rural Development Programme (IRDP), Mpika, Zambia. 39p. (mimeo). (E).
Francis P. A. 1988. Ox draught power and agricultural transformations in Northern Zambia. A Agricultural Systems 27 (1): 35-49. (E).
Haug H. and Gerner-Haug I. 1982. Promotion of draft-animal traction in Mali through the supply of simple agricultural equipment. pp. 403-423 in: Munzinger P. (ed), Animal traction in Africa. GTZ, Eschborn, Germany. 490p. (E,F,G).
Hyden G. 1980. Beyond Ujamaa in Tanzania. Under development and an uncaptured peasantry. Heinemann Educational Books, London, UK 270p. (E).
ILO 1987a. Structural readjustments and Zambia's self-reliance in farm equipment. Proceedings of a workshop on farm tools in Lusaka in 1985. International Labour Organisation (ILO), Geneva, Switzerland. 135p. (E).
ILO 1987b. In quest of agricultural mechanisation policy and strategies in the United Republic of Tanzania. Proceedings of a workshop on farm tools in Dar es Salaam in 1985. International Labour Organisation (ILO), Geneva, Switzerland. 69p. (E).
Jaeger W. K. 1984. Agricultural mechanization: the economics of animal traction in Burkina Faso. Ph.D. dissertation, Stanford University, Stanford, California USA. 242p. (E).
Kalb D. 1982. Sociological aspects of the use of draft animal on African smallholdings. pp. 339-372 in: Munzinger P. (Editor). Animal traction in Africa. GTZ, Eschborn, Germany. 490p. (E,F,G).
Kemp D. C 1987. Draught animal power some recent and current work. FAO World Animal Review 63: 7-14. (E,F,S).
Kjaerby F. 1983. Problems and contradictions in the development of ox-cultivation in Tanzania. Research Report 66, Scandinavian Institute of African Studies Uppsala, Sweden and Centre for Development Research, Copenhagen, Denmark. 163p. (E).
Kjaerby F. 1934. The demand for ox-plows, spare parts and animal-drawn equipment. Annex 8.2 in: B. Mothander, F. Kjaerby and K. Havnevik (eds), Farm implements for small scale farmers in Tanzania. Swedish International Development Agency, SIDA, Uppsala, Sweden. (E).
Lappe F. M. and Collins J. 1977. Food first: the myth of scarcity. Institute of Food and Development Policy Souvenir Press, London, UK. 416p. (E).
Massunga M. 1988. Oxenization Officer, Iringa Regional Officer, Ministry of Agriculture and Livestock Development, Tanzania. Personal Communication.
MRIDEP 1987. Mbeya Rural Integrated Development Project Office, Government of the United Republic of Tanzania. Persona. Communication.
Pingali P., Bigot Y. and Binswanger H. P. 1987. Agricultural mechanisation and the evolution of farming systems in Sub-Saharan Africa. Johns Hopkins University Press, Baltimore, Mary and, USA. 216p. (E).
Rain D. K. 1984a. Background study on the production and use of ox-drawn implements in Iringa and Mbeya. Report to Canadian International Development Agency, Hull, Quebec, Canada. 31p. (E).
Rain D. K. 1984b. The constraints to smallholder peasant agriculture. production in Mbeya and Mbozi Districts, Mbeya Region. Report to Canadian international Development Agency, Hull, Quebec, Canada. 43p. (E).
Rombulow-Pearse C W. and Kamasho J. A. M. 1982 Land resources of the Mbeya Region. Research Report No. 35. Mbeya Rural Integrated Development Project (RIDEP), Mbeya, Tanzania. 112p. (E).
Roosenberg R 1987. In search of better animal drawn weeding systems. The Tillers Report, Summer 1987, pp.1-3. Tillers Small Farm Program, Kalamazoo, Michigan, USA. (E).
Shetto R 1 988. Scientific Officer, Oxenization Section of Agricultural Engineering Department, Uyole Agricultural Centre, Mbeya, Tanzania. Personal Communication.
Smid J. 1982. Use of draft oxen in Northern Ghana. pp.451-476 in: Munzinger P. (Editor), Animal traction in Africa. GTZ, Eschborn, Germany. 490p. (E, F. G).
Starkey P. H. 1981. Farming with work oxen in Sierra Leone. Ministry of Agriculture and Forestry, Sierra Leone. 81p. (E).
Starkey P. H. 1986. Draft animal power in Africa: priorities for development, research and liaison. Networking Paper No.14, Farming Systems Support Project, University of Florida, Gainesville, Florida USA. 40p. (E).
Starkey P. H. 1988a. animal-drawn wheeled toolcarriers: perfected yet rejected. Vieweg for German Appropriate Technology Exchange, GTZ, Eschborn, Federal Republic of Germany. 161p. (E).
Starkey P. H. 1988b. Animal traction directory Africa. Vieweg for German Appropriate Technology Exchange, GTZ, Eschborn, Federal Republic of Germany. 151p. (E).
Terry P. J. 1984. A guide to weed control in East African crops. Kenya Literature Bureau, Nairobi. 186p. (E).
