Introduction
Participatory research, in a form, which has become known as the ‘Farmers First’ concept, has been first proposed in two classic papers at the beginning and in the mid-1980s (Rhoades and Booth Reference Rhoades and Booth1982; Rhoades et al., Reference Rhoades, Horton, Booth, Jones and Wallace1986).
Ashby and Sperling (Reference Ashby and Sperling1995) covered extensively the origin and the diffusion of the concept of involving users and clients in research and development as a principle of successful innovation. Within plant breeding, participatory research has had a very uneven growth in the scientific community, and even within CGIAR, from where it originated, only few breeders, defined as the scientists responsible for the breeding program(s) of the Center, used it despite its greater demonstrated efficiency (Ceccarelli, Reference Ceccarelli2015).
Participatory research has been implemented in plant breeding as participatory plant breeding (PPB) that we define as the participation of clients (more often, but not only, farmers) in all the most important decisions during all the stages of a plant breeding program as shown in Figure 1. Depending on when the participation starts, a distinction is made between PPB and participatory variety selection (PVS). The latter is when farmers’ participation begins during the testing of experimental varieties (Weltzien et al., Reference Weltzien, Smith, Meitzner and Sperling2003). While, on one side, PVS is technically easier than PPB to organize, because farmers are only involved in expressing their opinion on the limited number of lines that usually reach that stage, on the other side it leaves to them a limited number of choices to make. Furthermore, with PVS there is a risk for breeding material potentially desirable to farmers to be discarded before it is even seen by them.
Main stages of a breeding program.
However, because it is simpler to organize, PVS can be a useful entry point to start experimenting with the participation of farmers, assuming that PVS is fully decentralized, namely assuming that selection is done in the target environment.
PPB is a socially inclusive process, which brings with it farmer empowerment and increased agrobiodiversity, and can improve gender equitability (Tufan et al., Reference Tufan, Grando and Meola2018). Additional advantages of PPB/PVS are the production of varieties more resilient, with higher yield (Ceccarelli et al., Reference Ceccarelli, Grando, Singh, Michael, Shikho, Al Issa, Al Saleh, Kaleonjy, Al Ghanem, Al Hasan, Dalla, Basha and Basha2003; Gibson et al., Reference Gibson, Mpembe and Mwanga2011), greater adoption rate (Galluzzi et al., Reference Galluzzi, Estrada, Apaza, Gamarra, Pérez, Gamarra, Altamirano, Cáceres, Gonza, Sevilla, López Noriega and Jäger2014), increase genetic diversity (Joshi et al., Reference Joshi, Subedi, Rana, Kadayat and Sthapit1997), and household food security (Joshi et al., Reference Joshi, Devkota, Harris, Khanal, Paudyal, Sapkota and Witcombe2012).
PPB has been very attractive to social scientists who have been actually the first practitioners. This has, on one side, created a gap between social scientists and breeders (Ceccarelli and Grando, Reference Ceccarelli, Grando, Westengen and Winge2019), and on the other, has generated the belief that PPB is a methodology better suited to the developing world where many poor farmers in marginal areas have not benefitted of the conventional (non-participatory) approach (Bellon, Reference Bellon2006). Because of the gap mentioned above, a number of scientists, not only breeders, have used participatory approaches in plant improvement projects using various methodologies in several countries and in several crops. As a result, the literature on PPB (inclusive of PVS) is extremely varied, ranging from technical papers on the actual implementation of a PPB program, describing breeding methods, experimental designs, statistical analysis, and results, to more socially oriented papers with emphasis on the participants, to exclusively methodological and conceptual papers.
A number of reviews reported several cases of PPB/PVS. A first (to our knowledge) systematic review of PPB/PVS was done in 2003 (Weltzien et al., Reference Weltzien, Smith, Meitzner and Sperling2003) reporting 48 cases of which 11 presented in detail. A second one was presented in the form of a paper in 2004 (Ashby and Lilja, Reference Ashby and Lilja2004) and mentioned, without listing them, more than 200 projects using participatory approaches in 15 countries. More recently, an inventory of cases of participatory research was done as part of the SOLIBAM project (http://www.solibam.eu/) and covered 22 cases in 17 countries (Ceccarelli et al., Reference Ceccarelli, Al Yassin, Goldringer, Mendes Moreira and Chable2013).
In this paper we have updated the information on participatory research in plant breeding worldwide, including all the cases described in previous inventories. The aim is to provide as much quantitative information as possible on the extent of participatory research applied to plant breeding, which is still the subject of considerable scientific debate. In the paper we also presented some analyses of the information provided in the Supplementary appendixes (available online at http://doi.org/10.1017/S0014479719000127) as non-exhaustive examples of how that information can be used.
Methodology
We conducted a literature search using the following search engines: https://scholar.google.com/, www.getcited.org, and http://academic.research.microsoft.com. The strings used included ‘participatory,’ ‘participation,’ ‘participatory research,’ ‘farmers’ preferences,’ ‘plant breeding,’ ‘evolutionary plant breeding,’ and their combinations. We searched English, Spanish, and French sources. As sources, we used referee journals, conference proceeding, book chapters, working documents, and informal reports for which printed and accessible documents were available. In all cases, we also used the reference lists as additional sources. We also shared a preliminary list of publications with 31 scientists known to be active in either PPB or PVS, and 15 of them responded with additional references. The search ended on May 31, 2018.
For the reasons discussed earlier, we included publications on both PPB and PVS, as well as more recent examples of evolutionary-participatory plant breeding (EPPB) because we consider it as a development that amplifies the benefits of PPB. Evolutionary plant breeding (EPB) was first proposed by Suneson (Reference Suneson1956), although actually started much earlier (Harlan and Martini, Reference Harlan and Martini1929); it is based on letting early segregating populations evolve under the effects of natural selection. EPB becomes participatory when an evolutionary population is used by both breeders and farmers together as a source for artificial selection (Murphy et al., Reference Murphy, Lammer, Lyon, Carter and Jones2005).
We classified the publications in two broad categories: those reflecting mostly experimental work (Supplementary Appendix 1) and those predominantly addressing methodological, organizational, institutional, policy, or conceptual aspects of participatory research (Supplementary Appendix II). In some cases, the distinction is not well defined. In selecting the publications to include in the inventory, our priority was to be as inclusive as possible in terms of crops, countries, objectives, and institutions. In doing so, in some cases we have not included a few publications that we felt did not add to those four criteria over and beyond those already included for the same author(s), country(ies), crop(s), and institution (s).
We included publications on adoption related to PPB and/or PVS. Bench mark adoption rates were extracted from publications on broader adoption. Papers on seed systems were also included as participatory programs considerably affect the dynamics of seed diffusion.
The trend of publications over time was graphically represented using 3 years moving means, that is, the means of subsets of 3 years in which each subset is shifted forward by 1 year at the time. For example, if the first is the mean number of publications in 1982, 1983, and 1984, the second will be the mean number of publications in 1983, 1984, and 1985, the third will be the mean number of publications in 1984, 1985, and 1986, and so on.
Although our vision of participatory research is that of a highly gender-inclusive process, we are aware that a number of social scientists prefer keeping the gender dimension separate. Therefore, we also included in the survey gender-related publications when related to plant breeding.
Eventually, when not otherwise specified, we assumed that ‘wheat’ was bread wheat; we kept corn separate from sweet corn and the various millets distinct from each other, when species and genus were specified. In one case when only the term ‘millet’ was used, we assumed it was pearl millet. We also kept papers dealing with broccoli, cauliflower, and kale as separate cases, and we did the same with papers dealing with squash (= zucchini), winter squash, and butternut squash.
Results
At the moment of writing, we have identified a total of 254 publications (including one manuscript) dealing with participatory research in plant breeding. Of these, 172 publications (Supplementary Appendix I) deal mostly with experimental work (as defined earlier), and 82 (Supplementary Appendix II) deal predominantly with methodological and/or conceptual aspects of participatory research in plant breeding,
The first group of publications covers 47 crops (Table 1) including self-pollinated, cross-pollinated, and vegetatively propagated crops. The crops more often addressed in PPB or PVS projects or programs include some of the most important food crops such as rice, maize, bean, sorghum, barley, potato, bread wheat, and pearl millet. Interestingly, self-pollinated, cross-pollinated, and vegetatively propagated crops are included among the most frequently used crops, showing that it is not necessarily true that PPB or PVS is easier to implement with self-pollinated crops, as commonly believed.
Crops used in PPB or PVS programs and projects, classified according to the number of times they were cited in the 172 publications (Supplementary Appendix I)
* The total does not add to 172 because several papers deal with more than one crop.
Countries where most work on either PPB or PVS was (or is being) done are India, Ethiopia, USA, Nepal, Nicaragua, Mali, Syria, and Burkina Faso (Table 2). On one hand, this reflects the work done by few institutions, such as the International Center for Agricultural Research in Dry Areas (ICARDA) in Syria (on barley), the International Crop Research Institute for Semi-Arid Tropics (ICRISAT) in West Africa (on sorghum and pearl millet), and the CAZS Natural Resources (University of Wales, Bangor) in India and Nepal with rice. Therefore, these data fit well with the crops’ data given in Table 1. USA is an interesting case, where the majority of PPB and PVS work is done on horticultural crops and some on wheat and oats mostly addressing organic agriculture (Zystro et al., Reference Zystro, Colley and Dawson2019). This proves that, contrary to a common belief, PPB and PVS are methodologies not necessarily restricted to developing countries as shown also by an additional 24 publications from other 8 developed countries addressing organic agriculture and/or local adaptation.
Countries that hosted one or more PPB or PVS projects or programs with the number of times they were cited in the 172 publications (Supplementary Appendix I)
Since the idea of participatory research originated from within the CGIAR, one might wonder to what an extent the CGIAR contributed to the implementation and documentation of PPB/PVS after the original papers cited earlier.
Nearly all CGIAR centers (11) contributed to a different extent to PPB/PVS programs with the majority of contributions coming from ICARDA, ICRISAT, CIAT, CIMMYT, BIOVERSITY INTERNATIONAL (including those of IBPGR), CIP, and IRRI, with 10 or more publications each, followed by other 5 centers with less publications (Table 3). Eventually nine publications were attributed to the CGIAR System Wide Program for Participatory Research and Gender Analysis (PRGA), established in 1996 and closed in 2011, and, more recently, to the CGIAR Gender & Breeding Initiative (http://www.rtb.cgiar.org/gender-breeding-initiative/).
Using 3 years moving means, the temporal distribution of the CGIAR publications on PPB/PVS shows a steady increase until 2001–2003 (Figure 2), followed by a fluctuating trend with a tendency to decline. On the other hand, while the contribution of non-CGIAR institutions, with some exceptions, was lower than the contribution of CGIAR until the mid-2000, it become higher afterword. The decline of PPB/PVS programs in the CGIAR is likely to be associated with the shift toward more upstream research.
As many as 60 universities contributed publications to either PPB or PVS (Table 4), the majority of which (44 or 75% of the total) from 11 developed countries. Among those, the majority were from USA, UK, and Italy. The interest in PPB in these countries could be associated with the fact that PPB is mainly implemented in organic agriculture (Shelton and Tracy, Reference Shelton and Tracy2016), which is rapidly expanding. In the case of USA, the interest in PPB could also be associated with a less rigid seed system as compared to Europe.
There is no relationship between the data of Table 4 and those of Table 2, because Table 4 shows only the affiliation of the author(s) who actually did their work in a different country(ies).
A total of other 140 research institutions, NGOs and other associations contributed significantly to PPB or PVS with a number of publications during the 36 years period covered by this survey, ranging from 1 to 17: those with 2 or more publications are shown in Table 5.
Although the papers listed in Supplementary Appendixes I and II come from a wide range of publications, 160 (63.2%) were found in referee journals. Of these, 112 (44% of the 254 papers) were found in 14 referee journals (Table 6) with only two journals accounting for 67 papers, that is, slightly more than a quarter of all the publications.
Discussion and Conclusions
One of the most interesting results of this survey is that there is no correspondence between the frequencies with which PPB and/or PVS has been practiced and its institutionalization as the example of Ethiopia and India, but also Syria, Burkina Faso, and Mali shows. In all these countries, despite the evidence of its benefits, institutional (public) plant breeding is still centralized and non-participatory. In these countries, one of the reasons for the lack of adoption of PPB is the reward system for public breeders, which is still largely based on the variety release. Involving other partners in the selection process implies sharing the merit of obtaining a new variety. In many countries, one of the reasons often quoted is the seed legislation as an obstacle to PPB variety registration and release although there are examples – not many – of varieties released from PPB such as potato (Laurie and Magoro, Reference Laurie and Magoro2008), rice (Gyawali et al., Reference Gyawali, Sthapit, Bhandari, Bajracharya, Shrestha, Upadhyay and Jarvis2010), and sweet potatoes (Gibson et al., Reference Gibson, Mpembe and Mwanga2011).
The involvement of several universities, including many from developed countries and of other research organizations (such as CIRAD, INRA, LI-BIRD, and CIPRES among others), is a strong indication that PPB and PVS have solid scientific basis, although even in these countries, most of the public plant breeding is non-participatory.
PPB has been often considered to be more expensive than conventional plant breeding, and in fact, a cost analysis of PPB shows that its costs are higher than conventional plant breeding, particularly because of the incidence of the travel component (Mangione et al., Reference Mangione, Senni, Puccioni, Grando and Ceccarelli2006). However, we have argued (Ceccarelli, Reference Ceccarelli2015) together with others (Walker, Reference Walker2006; Witcombe and Yadavendra, Reference Witcombe and Yadavendra2014) that what needs to be considered is the benefit/cost ratio which is considerably higher in PPB due to the higher rate of variety adoption (Witcombe and Yadavendra, Reference Witcombe and Yadavendra2014). One of the main obstacles to a wider adoption of PPB could be the difficulty to perceive those advantages but also the breeding curricula in several universities, which is almost entirely based on formal plant breeding. Another likely obstacle to a wider adoption of PPB has to be found in the reluctance to accept the paradigm shift that PPB inevitably implies in what might be called ‘seed sovereignty’ and, consequently, ‘food sovereignty.’ In fact, PPB is a reversal of the model defined ‘delegative’ (from the French délégatif) by Bonneuil and Demeulenaere (Reference Bonneuil, Demeulenaere, Charvolin, Micoud and Nyhart2007) and by Thomas et al. (Reference Thomas, Dawson, Goldringer and Bonneuil2011). In that model, agricultural production, seed production, varietal innovation, and conservation of genetic resources are functionally separated and delegated to specialized scientists. Farmers lost the responsibilities for innovation and conservation that they had for millennia in a process defined as farmers’ deskilling (Fitzgerald, Reference Fitzgerald1993) or ‘dispossession’ (Kloppenburg, Reference Kloppenburg2010). The model that resulted in a system of power, authority, and control was well established at the time participatory research was first proposed. PPB is reversing this model and is seen as very radical and perhaps even subversive (Crane, Reference Crane2014), because it may eventually lead that what Kloppenburg (Reference Kloppenburg2010) has defined as a ‘repossession.’
Author ORCIDs
Salvatore Ceccarelli 0000-0003-3063-9836
Supplementary materials
For supplementary material for this article, please visit http://doi.org/10.1017/S0014479719000127
