D.L. PLUCKNETT and N.J.H. SMITH
Network typology
Collaborative research networks
Benefits
Start-up and operation
Tapping the unrealized potential
References
Major international agricultural research networks, essentially a phenomenon of the last two decades, now penetrate most facets of agricultural research. Agricultural research networks are increasing in number and scope, particularly in the Third World. While new networks are being formed, existing ones often take on a range of fresh initiatives and are thus becoming more complex and comprehensive. More networking activity, however, does not necessarily mean better and more efficient research.
To assess the current status of networking in international agricultural research and gauge whether the potential of networks is being realized, we first define and categorize networks. Expectations of the performance and impact of cooperative and collaborative activities should vary according to the type of network. We then explore the benefits, both tangible and qualitative, of networking to participants in the Third World. The start-up and operation of networks are examined in order to pinpoint key ingredients in launching and sustaining collaborative research efforts. Finally, we discuss evolution of networks, stress the need to increase support for collaborative research efforts and emphasize the importance of patience while waiting for networks to help solve problems associated with boosting agricultural production.
Any number of topologies can be drawn up to classify agricultural research networks, depending on the purpose and analytical goals. Networks can be grouped according to subject matter, such as soil fertility or farming systems, or according to their structure and operational mode (Plucknett and Smith, 1984). For example, one can identify networks that operate in a hub-and-spoke manner, where information flows mostly from a coordinator to cooperators and those which operate in a wheel manner where information flows in a rim fashion between satellite nodes as well as along spokes to and from the coordinating hub.
Ralph Cummings, Jr. and Calvin Martin of the U.S. Agency for International Development (USAID) categorized agricultural research networks according to their operational style in a report to the Special Programme for African Agricultural Research (SPAAR, 1986). Cummings and Martin proposed three main types of networks, commonly referred to as the SPAAR typology: information networks (type 1), scientific consultation networks (type 2) and collaborative research networks (type 3). We adopt this typology with one modification: the addition of material exchange networks.
Information networks are essentially newsletter or information dissemination operations in which information is collated by a coordinator and sent to individuals on the mailing list (Table 1). Material exchange networks are primarily concerned with testing crop proto-varieties and agricultural machinery designs at various sites. International nurseries and agricultural machinery networks fall under the material exchange category. In scientific consultation networks, individuals in ongoing research programmes agree to share information and ideas. Periodic workshops and meetings are organized to discuss problems and compare results; some methodological or strategic adjustments to autonomous research programmes may be made as a consequence. Scientific consultation networks are cooperative rather than truly collaborative undertakings. In collaborative research networks, the entire research enterprise may be jointly planned and executed, usually from the ground up. The joint planning and execution characteristics of collaborative research networks, along with the development of a major conceptual or methodological document, are the major distinguishing features separating them from scientific consultation networks. Participants are actively involved in establishing research priorities, agreeing to a division of research responsibilities and following common methodologies.
Table 1. Some characteristics of information exchange, material exchange, scientific consultation and collaborative research networks.a
|
Trait |
Information Exchange |
Material Exchange |
Scientific Consultation |
Collaborative Research |
|
Coordinator |
Yes |
Yes |
Yes |
Yes |
|
Publications |
Yes |
Yes |
Yes |
Yes |
|
Advisory board |
No |
Yes |
Yes |
Yes |
|
Study tours |
No |
Yes |
Yes |
Yes |
|
Training |
No |
Yes |
Yes |
Yes |
|
Workshops |
No |
No |
Yes |
Yes |
|
Common methodology |
No |
Yes |
No |
Yes |
|
Joint planning |
No |
No |
No |
Yes |
a There are, of course, exceptions in some cases.
As described in our earlier paper (Plucknett and Smith, 1984), successful networks are grounded on seven main principles: a clearly defined problem and a realistic research agenda; a problem that is widely shared and in which there is strong self-interest involved; a willingness on the part of participants to commit resources, such as personnel and facilities; a possibility for outside funding to facilitate the birth of the proposed network and to ensure its functioning for the first few years; sufficient scientific caliber of the participants to ensure that contributions will be made and finally, strong and efficient leaders who have the confidence of the participants.
Collaborative research networks are the focus of this paper. While all types of networks have an important role to play in furthering research and improving agricultural productivity, collaborative research networks have the greatest potential to upgrade skills of participants and to fortify the ability of Third World nations to conduct research designed to overcome constraints to increased agricultural production. In a sense, collaborative research networks represent networking in its purest form since individuals in jointly planned research ventures pool their resources and talents, participate in planning and policy formulation and pull in unison. We do not wish to imply that all networks can, or should, become collaborative research efforts; for some purposes, information exchange, material exchange or scientific consultation networks suffice. But for other research problems, particularly if they are narrowly focussed, collaboration can be a powerful approach.
Collaborative research networks in agriculture are few in number among regional or global networks. Probably no more than a dozen genuine collaborative research networks operate today on an international scale (Table 2). The African Trypanotolerant Livestock Network (ATLN) exemplifies true collaborative research and has many lessons to offer. Other jointly planned research networks with a track record to guide the establishment of future collaborative research programmes include the International Benchmark Soils Network for Agrotechnology Transfer (IBSNAT), the Asian Rice Farming Systems Network (ARFSN) and PRECODEPA (Programa Regional Cooperativo de Papa), a multi-purpose commodity network focussing on potato production (Plucknett et al., in preparation).
Although few in number, collaborative research networks have been established in a wide variety of subject areas. The ATLN, coordinated and supported by ILCA and ILRAD, focusses on a widespread livestock disease, trypanosomiasis. A sister international agricultural research centre, the International Rice Research Institute (IRRI), helps administer ARFSN, a farming systems network that has grown spectacularly in its dozen years of existence and now spans sixteen-Asian nations. ARFSN embraces several research thrusts, including integration of livestock in rice farming systems, fertilizer trials and varietal testing. IBSNAT, supported by USAID and coordinated by the University of Hawaii, has built on the foundation of a predecessor research project, the Benchmark Soils Project. IBSNAT, created in 1982, explores the transferability of agronomic practices and cropping systems within three major soil families of the tropics.
PRECODEPA, one of five regional potato research networks organized or supported by the Lima-based Centro International de la Papa (CIP), addresses ecological and socio-economic constraints to increased potato production in nine Central American and Caribbean nations. PRECODEPA has divided its research responsibilities as follows: late blight (Mexico); seed production (Mexico, Guatemala, Costa Rica); tuber moth (Costa Rica, Guatemala); golden nematode (Panama, Mexico); bacterial diseases (Costa Rica); rustic storage (Guatemala); socio-economics (Guatemala); processing (Guatemala).
Table 2. Some international collaborative research networks in the agricultural sciences.
|
Network |
Focus |
Countries |
Began |
|
ARFSN (Asian Rice Farming Systems Network) |
Rice farming systems in South, S.E. Asia |
16 |
1975 |
|
PRECODEPA (Programa Regional Cooperativo de Papa) |
Multipurpose potato research in Central America and the Caribbean |
9 |
1978 |
|
IBSNAT (International Bench-mark Soils Network for Agrotechnology Transfer) |
Transferability of crop management practices within same soil families in the tropics |
16 |
1982 |
|
ATLN (African Trypanotolerant Livestock Network) |
Control of African trypanosomiasis in livestock |
10 |
1983 |
Recently, some networks with international centre-like characteristics have been established. Their operational style is networking while their governance structure is like some of the international agricultural research centres, including a director and a board of trustees. Examples of these are the International Board for Soil Research and Management (IBSRAM) and the International Network for the Improvement of Banana and Plantain (INIBAP). Topics for other centre-like networks currently being discussed include aquaculture and tropical vegetables.
Agricultural scientists around the world are probably involved in more than a hundred networks ranging from information exchange to collaborative research. The proliferation of networks in the last two decades has been fuelled by the many perceived benefits of participating in information and material exchange networks as well as in scientific consultation and collaborative research networks.
We now concentrate on six professional advantages provided by viable collaborative research networks: concentration on important, widespread problems; providing access to new information and methodologies; avoiding redundancy; saving time; increasing training possibilities; and creating opportunities to develop leadership skills.
A significant advantage of research networks is that they coalesce around important problems. The number of important research problems facing agricultural scientists is almost limitless and there is always the danger that researchers may select less important topics for research. While all research has the potential to contribute useful information, in a world with limited resources and a shortage of qualified personnel it is imperative that the major issue be identified and tackled first and not be left unresolved because resources have been drained away on less important projects. When a sizeable number of scientists agree to work together on a project, usually the problem or issue they are grappling with is a pressing one. In the case of ATLN, for example, scientists have banded together to study control measures against a disease that strikes some 50 million cattle, 30 million sheep and 40 million goats in a broad zone stretching from West to East Africa (ILRAD, 1987). Similarly, IBSNAT focusses on three soil families that are widely distributed in the humid tropics and on which hundreds of millions of persons depend for livelihood and sustenance.
All networks, from information exchange to collaborative research, provide opportunities for participants to learn about new methodologies and research approaches. In collaborative research programmes however, contact between participants is more systematic than in other networks and they tend to adopt a common methodology. In this manner, new research techniques are more readily disseminated and adopted. Information on or access to the latest laboratory techniques and field research methods are particularly valuable to Third World scientists who often find it difficult to gain access to up-to-date library facilities, to obtain new equipment and to travel to international meetings.
One of the most important benefits of networking, particularly with collaborative research, is that unnecessary repetition in research is avoided. Some replication of experiments and trials to verify results is a standard and necessary procedure in science, but redundancy is a luxury that few countries can afford.
A corollary to avoiding redundancy is the time savings generated by quality collaborative research. Because the task is divided up among participants, either by region or research topic, results can be achieved more quickly than if the scientists were working independently of each other. A wider array of environments are sampled, cross-fertilization of ideas is encouraged and a broader dimension of research effort is achieved. Collaborative research does not always lead to immediate dramatic breakthroughs or payoffs for citizens of the host countries, but jointly planned research has a better chance of achieving success with widespread problems than isolated efforts. Even if a research thrust in a collaborative programme proves to be less effective than planned because more basic research is required or because appropriate methodologies have yet to be devised, at least that conclusion is reached faster than in the case of more isolated research efforts.
Many networks offer short training courses to meet specific needs of participants. Depending on the nature of the network, such courses offer a combination of classroom, laboratory and field activities. In addition to network training courses, participants may also have opportunities to take other courses offered by international agricultural research centres. Networks offer a great opportunity to train young scientists in conduct and management of research.
Some networks rotate coordinators to spread the administrative burden more evenly, to reduce any paternalism of a dominant institution and to provide leadership opportunities for participants. However, frequent changes in network leadership may not be desirable, and, in some cases, may cause instability and loss of effectiveness for the larger network operation. Even if coordination remains in the hands of one institution during the life of a network, leadership skills are still developed when local scientists assume responsibility for their portion of the research task, help organize co-workers and report results at "networkshops" and conferences. Furthermore, some research networks have catalyzed the establishment of small sub-networks on a national or regional scale. This trend is particularly noticeable with ARFSN and IBSNAT. Experience gained in a larger, international network is invaluable when a national programme scientist takes on the job of coordinating a sub-network.
Third World countries clearly benefit from international networks, but participation has a price. Developing countries must be willing to contribute resources-human, financial and the use of buildings, equipment and land-if networks are to flourish. Next we consider some of the ingredients for successfully launching and maintaining collaborative research networks.
Jointly planned research networks generally require more time and consideration to launch than other networks because common working procedures have to be formulated and agreed to by the participants. A founding document must also be available; if such a document is not available, the first task of the network may be to prepare it. Research methods may have to be changed and fresh approaches adopted in collaborative research ventures. For scientific consultation networks, less time for establishment may be required, in part because only minor adjustments may be necessary to ongoing research. Start-up for information networks is relatively straightforward since much less consultation is done, at least at first. Materials exchange networks may also require less time to establish, because in many cases the initial consultation and planning is carried out by international agricultural research centres that initiate them and assemble the materials to be exchanged, often using national scientists and former trainees as their first contacts.
A believable founding document is an essential ingredient for assembling a collaborative research network. Without a solid proposal or conceptual framework, based on careful analysis and preferably some initial survey work, prospective collaborators will have a hard time convincing administrators and funding agencies, both within their national governments and at international levels, of the validity and viability of their cause. Groundwork for the ATLN began in 1977 when the United Nations Environment Programme (UNEP) and FAO commissioned a survey of trypanosomiasis in Africa and the distribution of trypanotolerant breeds of cattle, sheep and goats (ILCA/FAO/UNEP, 1979). This baseline study carried out in collaboration with ILCA concluded that trypanotolerant livestock could play a larger role in increasing meat and milk yields in Africa and urged further studies. Later, the Network issued a training manual which provides a methodology framework for the collaborative research (Murray et al., 1983).
Although national programmes should contribute the major share of resources to start and operate collaborative research networks, some external funding is also usually required. Outside support is often crucial in bringing together interested parties for an initial consultation meeting. The idea of starting a network often occurs to scientists at annual professional conferences, but a follow-up meeting is required to discuss strategies and draw up a proposal. Scientists in developing countries often find it impossible to obtain funding from national agencies for travel abroad to any meeting. By providing air tickets and living expenses for prospective participants, external donors play a vital role in inaugurating networks.
Once a network is operating, external support is still frequently needed to fund monitoring tours to research sites, for convocations and planning meetings, to purchase some equipment and supplies and to organize training courses. Monitoring tours are important mechanisms for participants to keep in touch with colleagues, discuss common problems and to observe new methodologies in action. Workshops generally serve two vital functions that help in network operation: they are used to assess results of surveys, experiments and trials and to chart future directions. Training, as indicated earlier, helps upgrade the performance of participants. Since all the above activities require international travel and living costs, external support is an essential ingredient for most large-scale networks.
Although outside funding is usually crucial for international research networks, money alone will not buy success. Intellectual capital and motivation are just as important as adequate funding. If the network's objective has been thoroughly considered and the research agenda carefully drawn up, the goal should be feasible and the research approach and methodology should be conceptually sound. If the research programme is badly flawed, no amount of funds poured into the network will generate viable products. When the research venture is on scientifically sound footing and participants are well qualified to do the work, a flow of new insights and discoveries usually ensues. Fresh ideas revitalize research and help maintain the vigor of networks (Greenland et al., 1987).
Thus far, most networking activities have centred on information exchange, material exchange or scientific consultation networks and these have proliferated. Indeed, there are some indications that networking may be reaching the saturation point in some cases. The scramble to boost food supplies in Africa is creating conditions ripe for proliferation and redundancy among information exchange, material exchange and scientific consultation networks. As of 1986, thirty-five networks focussing on improving commodities were operating in Africa (SPAAR, 1986). Most of these are information exchange or scientific consultation networks; material exchange networks are not included in that figure. It is not possible to know whether the African research establishment can absorb all of the networks that are operating or planned. Donors are partly responsible for the proliferation of networks; a potent motivating force in establishing networks is identification of a donor with a particular activity. Donors are naturally eager to receive credit for helping find solutions to food production problems, but excessive donor wishes for identification can be a negative force in network effectiveness.
While competition may have emerged between some scientific consultation networks, particularly in Africa, there is room for more collaborative research networks. Collaborative research networks could increase in two ways: evolution from other kinds of networks, particularly scientific consultation networks, or the creation of new ventures. While it is always possible to restructure cooperative networks so that they become truly collaborative, few scientific consultation networks seem to be heading in this direction. Some scientists have vested interests in adhering to ongoing research activities and are reluctant to drop or alter radically their programmes to accommodate requirements of a collaborative research enterprise.
The potential payoffs are greater with collaborative research, so consideration about how this transformation from scientific consultation to collaborative research can be effected is warranted. Donors and research leaders in Third World countries could profitably devote more thought to how some scientific consultation networks can be upgraded to collaborative research networks. The potential payoffs from such a process appear greater than simply creating more scientific consultation networks.
In some situations it may be easier to forge collaborative research networks from scratch and indeed, this is how most of them have formed. Existing personnel and facilities are used in collaborative research networks, just as with other networks, but participants devise a workplan from the start based on the comparative advantage of each member. And participants are constantly involved in minor adjustments and revamping, as necessary. Based on the experience of the few international collaborative research programmes now operating, the donor community might want to consider nurturing the birth and development of more jointly planned research efforts. As with all research projects, however, some patience is required. Quality research programmes take several years before they bear fruit. Networks will not offer an immediate cure to the many problems facing agriculture, but by improving research productivity they can greatly facilitate the search for solutions.
Greenland, D.J., E.T. Craswell and M. Dagg. 1977. International networks and their potential contribution to crop and soil management research. Outlook on Agricul. 16 (1):42-50.
ILCA/FAO/UNEP. 1979. Trypanotolerant Livestock in West and Central Africa. 2 vols. Addis Ababa: ILCA.
ILRAD. 1987. Improved trypanosomiasis control: studies on drug treatment. ILRAD Reports 5 (1):1-6.
Murray, M., J.C.M. Trail, D.A. Turner and Y. Wissocq. 1983. Livestock Productivity and Trypanotolerance. Network Training Manual. Addis Ababa: ILCA.
Plucknett, D.L. and N.J.H. Smith. 1984. Networking in international agricultural research. Science 225:989-993.
Plucknett, D.L., N.J.H. Smith and S. Ozgediz. Networking: principles and concepts from agricultural research. In preparation.
SPAAR. 1986. African Agricultural Research Networks: Summary Papers and Tables. Meeting of the Technical Working Group on Networking of the Special Programme for African Agricultural Research (SPAAR), Brussels.
