The aim of this paper is to explore and mitigate barriers to full adoption of banana Xanthomonas wilt (BXW) disease mitigation measures in eastern Africa (and Uganda specifically). Although officials in Uganda declared  BXW   under   control   by    2016,   subsequent decreases in funding and support for continuing mitigation have led to resurgences of the disease and farmers finding themselves almost entirely on their own (Gotor et al., 2020). This issue has further given that BXW  is  a  trans-boundary  (regional)  issue   capable  of crossing international borders, such that mitigation failures in one country can potentially spill over into other parts of the region (Goodman et al.,  2021).

To address this aim, the authors turn in this paper from the well-documented economic impacts of BXW in eastern Africa to investigate and add to the picture of the challenges around the social pillar of sustainable development-solutions for BXW eradication. In particular, they explore how the current framing of recommendations for mitigating BXW in Uganda can fail to resonate practically and culturally with the broader social world and discourse that banana occupies in the region, leading to decreased (only partial) BXW mitigation compliance. To better secure mitigation (in light of the fact that large-scale mass commitments are unlikely to re-occur anytime soon), it can specifically be the social pillar of sustainability - more than the economic or environmental pillars - that best stands to address the challenges arising from both from efforts to change cultural banana farming practices in Uganda as well as political impacts from BXW trans-boundary issues.

Given that the lack of practical and cultural resonance with BXW mitigation solutions can result (and has resulted) in farmer noncompliance with anti-BXW measures in Uganda, such a situation decreases the effectiveness of past and on-going efforts to control it locally while increasing the likelihood of the disease’s spread both within and across the Ugandan border. In particular, following the generally successful mass-scale mitigation of BXW in Uganda, which reduced the percentage of farm infestation from 50% in 2013 to 1.9% in 2015 (Gotor et al., 2020), the subsequent lack of on-going support has left banana farmers on their own to face subsequent resurgences of BXW (Gotor et al., 2020).

In as much as it seems unlikely that there will be any adequate provision of such support going forward, it becomes critical to find culturally feasible and cost-effective channels for reaching banana farmers such that fuller compliance with BXW mitigation protocols is achieved. As such, through this paper’s analysis, the findings not only disclose potential alternative policies (or at least strategic wordings of policies) that better align with the broader discourse and social importance of banana for local banana farmers in Uganda but also increase the likelihood of policy compliance by people for improved control (if not elimination) of BXW both locally and therefore regionally.

Achieving the triple bottom line of sustainable development

The analytical lens of this paper is the social pillar of sustainability. Although sustainable solution-development calls for triple bottom-lining of the economic, social, and environmental gains arising  from  any  proffered  solution  (Luetz and Walid, 2019), Afful et al. (2019) have correctly noted that the social pillar of sustainable development tends to remain underemphasized and vague. For example, a meta-analysis of key terms in human development reports by Luetz and Walid (2019) finds that “‘economic’ perspectives dominate ‘sustainability’ and ‘social’ perspectives by a factor of 2 and 4.6,7 respectively” (301). Afful et al. (2019) also highlight a continuing tendency to assume that if the economic pillar of sustainable development is addressed or supported, then somehow the social pillar of sustainable development “automatically” takes care of itself.

Because the three pillars of sustainable development are intertwined or may be difficult to distinguish from one another at times (Luetz and Walid, 2019), this complicates their measurement. For example, while economic success measures can be interpreted as evidence for social success measures (Keyes and Shapiro, 2004), ample research also demonstrates how economically motivated replacement of traditional social farming habits with industrial or monoculture farming can have not only direct, negative environmental impacts (for instance, the degrowth literature, c.f., Martínez- et al., 2010; Schneider et al., 2010), but also social ones as well (Bello-Bravo, 2020b; Mather et al., 1999; Rudel and Horowitz, 1993). As Corntassel (2008) notes, for example, “Unfortunately, what is considered sustainable practice by states comes at a high price for indigenous communities, often leading to the further degradation of their homelands and natural resources” (108).

Rather than relying on indirect economic measures to gauge social sustainability, the authors draw on the notion of social wellbeing to highlight a direct and immediate element of the social pillar of sustainability. Social wellbeing is a person’s “self-report of the quality of his or her relationship with other people, the neighborhood, and the community” (Keyes and Shapiro, 2004:350). Given that such social wellbeing includes the goals of “social integration, social contribution, social coherence, social actualization, and social acceptance” (Keyes, 1998:121), then various means for achieving these goals must necessarily be sustainably present within a community to ensure their attainment.

While increased wealth sometimes affords such goals for some individuals, they can also erode and negatively impact social life generally (Errington and Gewertz, 1989). Unsurprisingly—given not only the platitude that “money can’t buy happiness” but also the insight that the social is logically prior to the economic—the role that wealth plays for social wellbeing generally appears to have more to do with the social forms of activity that arise from or around it rather than the amount of it (Miñarro et al., 2021; Zelizer, 1996, 2000). For these reasons, one cannot automatically assume that the provision of financial wealth supports the social pillar of sustainability—not only because wealth can actually erode social life and communities (Desai, 2017; Errington and Gewertz, 1989; Sandel, 2012) but also because it may be not available or is actually unnecessary in some contexts for achieving social wellbeing goals (Miñarro et al., 2021). Instead, if any proposed or implemented developmental solution fosters the key social wellbeing goals of “social integration, social contribution, social coherence, social actualization, and social acceptance” (Keyes, 1998:121), then that solution will better reflect support for the social pillar of sustainability.

Recovering the social pillar of sustainable development for addressing BXW

Understanding banana through the lens of the social pillar of sustainably requires a broader look. Economically, banana is the world’s fourth largest staple food and provides a livelihood and enhanced food security to more than 100 million people south of the Sahara (Nannyonga et al., 2015). In Uganda specifically, banana production previously accounted for 10 million tons annually, approximately 80% (or $440 million) of the total economic output at the crops’ average of farm price, and 22% of the national agricultural rural revenue (Embrechts et al., 1996). More recently (from 2008-2010), only 4 million tons were grown, likely in part due to BXW devastation (Kikulwe and Asindu, 2020). Per capita, the region has the highest consumption in the world and most of the crops are consumed locally (Kalyebara, Nkuba, Ramadhan, Byabachwezi, and Edmeades, 2007), including by more than seven million people (65% of the population) in urban areas (Karamura et al., 1998).

However, these economic measures in eastern regions of Africa do not occur in isolation but interrelate with social and environmental aspects as well. Banana was originally domesticated in southeast Asia, possibly as long ago as seven millennia in Papua New Guinea, with evidence for banana cultivation in Africa dating from the first millennium CE (Carney and Rosomoff, 2009). Importantly, because domestication of a plant species requires direct human intervention for the species’ successful reproduction and cultivation—in the case of banana, the deliberate removal and replanting of the plant’s suckers that develop from its corm (Carney and Rosomoff, 2009)—this repeated human activity also tends to coalesce gradually into a series of socio-environmental norms and behaviors (indigenous knowledge and practice), including specific crop timings, seasonally repeated activities or rituals, and a cultural superstructure for supporting these activities (Carney and Rosomoff, 2009). This also includes networks of mutual aid groups, markets for selling the crop, and local specializations and specialists who provide technological, physical, or magical expertise to enable successful crop growth (Carney and Rosomoff, 2009). These are elements in the social pillar of sustainability for banana in the region.

Supporting this sociocultural pillar for sustainable development is not only necessary but also typically results in beneficial outcomes (often measured economically). For instance, in eastern Africa, several biotic threats to banana (including weevils, nematodes, fungal, and viral diseases) have been managed using cultural methods that included mulching, spacing, using acidic plant mixes, and intercropping with other plants, e.g., coffee and cassava (Ouma, 2009).

While this had a net-reduced economic productivity from banana per se, the gains from these social and environmental interventions maintained sufficient levels of food and income security that would likely have been (and were already facing threats of) decreased sufficiency (Ouma, 2009). Moreover, the social interventions in these cases did not especially introduce or risk the introduction of chemical, mono-cultural, or industrial farming means that have been shown to have pronounced environmental impacts (Martínez-Alier et al., 2010; Schneider et al., 2010). This integrating of the economic, social, and environmental “bottom lines'' for interventions affecting banana production exemplify a sustainable developmental approach (Afful et al., 2019).

However, the BXW epidemic has since triggered a shift to other crops (especially coffee), as well as attempts to introduce genetically modified BXW-resistant banana - specifically, the banana suckers needed for planting - to Ugandan farmers (Kikulwe and Asindu, 2020). Without intending any comment on the beneficial or harmful effects from these shifts, the authors would simply note that they introduce potential/new environmental and social impacts through changes within the agricultural sector. For example, since 2015 - when BXW was declared under control (Gotor et al., 2020) - drastic decline in public investment to mitigate BXW followed and led to further resurgences of the disease (Gotor et al., 2020). While it is doubtless premature to say that banana farmers have been “abandoned,” it is also clear that Ugandan agriculture in general is also attempting to heavily focus on alternatives. That banana is a major staple for in-country consumption (Kalyebara et al., 2007) makes any such redirection of agricultural attention to other crops socially precarious.

Challenges for BXW mitigation

For BXW specifically in eastern Africa, the social aspect plays a large role in mitigating or eliminating the disease. Since 2001, BXW reached epidemic proportions in Africa, with production and income losses estimated at USD$35 million and USD$100 million in 2005 and 2006 alone (Nannyonga et al., 2015) and consequent negative impacts on people’s social livelihood and community wellbeing. Previous regional efforts (in Uganda and Tanzania) to control the disease met with only limited success (Tushemereiwe et al., 2006), and resurgences of the disease in previously controlled area have been reported (Gotor et al., 2020; Ongu, 2014). Factors involved in these suboptimal outcomes include, but are not limited to:

1) Shortages of resources at every level, including the infeasibility of farmers applying control measures due to high intervention costs (Gotor et al., 2020), shortages of extension personnel, shortages of funds for creating and disseminating educational materials in farmer-useable formats (non-print, and translated into local dialects), and shortages of trainers for that material.

2) A lack of access by farmers to best-practices approaches for mitigating or preventing the disease (including often conflicting sources of BXW mitigation information when shared farmer-to-farmer rather than through extension services, or in the absence of extension officers or other knowledgeable people to provide “real-time” diagnoses of BXW in the field) (Tushemereiwe et al., 2006).

3) Farmer costs (of time and money) associated with BXW mitigation mandate compliance (McCampbell et al., 2018), which could often be offset by traditionally social intra-network materials and tool sharing farmer habits in Uganda that potentially increased the risk of BXW spreading (Bagamba et al., 2006; Blomme et al., 2014; Magala et al., 2019; Staver et al., 2010; Tinzaara et al., 2011).

In addition to the above, one study found that BXW mitigation compliance yielded a net $35 per acre per annum benefit for full adopters, a $30 benefit for partial adopters, and a net $30 per acre per annum loss for non-adopters (Gotor et al., 2020). Given the relatively modest per acre gain for full adoption compared to partial adoption, Gotor et al. (2020) reasonably propose that onerous front-end costs explain why only 33% of farmers fully adopted the mitigation measures, and 62% only partially adopted them. Using an average Ugandan banana farm size of 0.8 ha (Jogo et al., 2011), this yields an approximately net $10 per year for full adoption over partial adoption. Larger farms would reap larger grains; smaller farms smaller.

From 2013-2016, when BXW was most comprehensively funded for intervention in Uganda, the most effective approach involved demonstrations to farmers about (1) exclusively using only clean planting materials and (2) monitoring strategies to identify BXW, which ultimately reduced the percentage of affected farms from 50% in 2013 to 1.9% in 2015 (Gotor et al., 2020). These technique-interventions were perceived by solution providers as highly effective (Gotor et al., 2020). Since that success, however, “control of the BXW was entirely entrusted to rural households” (Gotor et al., 2020) and has resulted in resurgences of BXW.

Because BXW is highly infectious and spreads rapidly (Jogo et al., 2011), five principal  control  measures  were proposed to mitigate its propagation:

1) Advising farmers to conduct pruning after three months once they identify plants that are infected

2) Cutting infected suckers before infesting the top fruiting parts

3) Uprooting and destroying any sucker identified as wilting (yellowish appearance)

4) Removing the male bud once the banana has fruited and formed fingers (using a forked stick), and

5) Exclusively using only disinfected tools, particularly with emphasis on pangas [a specialized cutting knife] (Jogo et al., 2011).

These efforts aim to reduce the major identified vectors of disease transmission (namely: insects, plant material exchange, and infected tools). However, the role for each of these factors or their specific combination in disease spread is still not well-characterized (Kilwinger et al., 2019; ProMusa, 2021; Tinzaara et al., 2011). For example, although BXW is transmissible by insects, it is less susceptible to this compared to other banana diseases (ProMusa, 2021). Similarly, while sharing plant materials among farmers readily transmits BXW, limited plant material sharing has been observed among farmers, which likely diminishes the role of that vector at present (Kilwinger et al., 2019). Lastly, infected tools have been shown play a larger role in the area considered in this study (Mbarara district) than insect transmission (Jogo, Karamura, Tinzaara, Kubiriba, and Rietveld, 2013). Whatever the exact roles of these established vectors, an only 62% partial adoption of BXW mitigation measures will play key role in any resurgence of the disease (Gotor et al., 2020).

Kikulwe et al. (2018) underscores how the five BXW mandates above must be specifically and necessarily cultural; “The only disease management strategy for crop protection against BXW in Uganda is the use of one or a combination of cultural BXW control practices” (88), which when implemented correctly can result (and have resulted) in complete control of the disease (Karamura et al., 2008). Among the five recommendations, number 5 (“exclusively using only disinfected tools, particularly with emphasis on pangas”) adds overhead to farmer budgets by requiring them to purchase, or otherwise freely acquire, disinfectant for their tools and up to three different pangas for cutting banana leaves when cooking (because using one knife increases the risk of contaminating other plants). Without support for these additional costs, BXW mitigation efforts can suffer due to decreased solution-uptake or compliance (Gotor et al., 2020; Kikulwe et al., 2018).

In one study, Kikulwe et al. (2019) provided Ugandan banana farmers with the resources needed for the adoption and practice of three cultural BXW control measures. While the majority of participants (80%+) assented   to   the    effectiveness    of   these  measures, subsequent non-adoption of the BXW control measures among participants associated with farm size, i.e., smaller, non-commercial farms were less able to afford the resources required for the tested control measures. Kikulwe et al. (2019) also specifically observed that “women-inclusive training and farmer-preferred information dissemination pathways” (13) would enhance adoption; Gotor et al. (2020) echoed this point.

In particular, this emphasis on women and female-owned farm households not only raises the important social issue of gender with respect to sustainably securing full BXW mitigation compliance (Kikulwe et al., 2018) but also bears directly on the fifth recommendation above, since its requirement to use three pangas for cooking applies explicitly (if not exclusively) to women (and female-headed farms). Beyond differential perceptions of BXW control by men and women (Kikulwe et al., 2018), delivering BXW control messaging to Ugandan women will face the usual constraints found in a majority of rural situations, i.e., decreased access to education, literacy, extension services, and agricultural resources generally (Bello-Bravo and Lutomia, 2016; Bello-Bravo et al., 2017, 2019). As Kikulwe et al. (2018) stresses, “It is therefore critical that existing gender-based and underlying perception constraints are addressed” (87).

Messaging

How BXW mitigation messaging is delivered matters as much as what is delivered (Bello-Bravo, 2020a; McLuhan, 2009), whether in women-inclusive ways, through farmer-preferred dissemination channels, or other means. This specifically cultural element of delivery has redoubled significance when addressing BXW mitigation regionally (across Uganda and Tanzania specifically, but including Rwanda, Burundi, and eastern Democratic Republic of Congo as well).

As a trans-boundary problem (Ansell et al., 2010), it risks cultural incompetence and irrelevance to address BXW control problems regionally using only one cultural form of the message. The most obvious point where this risk occurs involves the requirement to translate any messaging materials into locally most comfortably spoken dialects (Rodríguez-Domenech et al., 2019), but this translation is as much culturally behavioral as linguistic (Braçaj, 2015; Kikulwe et al., 2018).

Although effective communication and message delivery always becomes more challenging during a crisis (Reynolds and Seeger, 2005), self-organization responses can often complement or stand in place of more top-down responses (Kendra and Wachtendorf, 2001; Miresmailli et al., 2015; Solnit, 2010). One advantage of such self-organized responses includes their often culturally competent and relevant character. Nonetheless, as Kikulwe et al. (2019) makes clear in the differential  adoption  rates  by  farm  size,  even  within  a culture certain variations can exist that generate obstacles to BXW mitigation messaging. In a trans-boundary context like BXW - holding all other “logistical” needs for coordination equal - attention to cultures (in the plural) becomes paramount for more effectively ensuring that BXW mitigation messaging gets received in all contexts delivered. This is not only because “one size fits all” (or even regional-level) BXW mitigation will likely not be effective in all cultural contexts, but also because differences in national-level commitment or mitigation approaches can vary in a way that makes one nation susceptible to less effective mitigation by their neighbors. Locally in Uganda, as the extension agent interviewed for this study noted: “The current way of delivering [BXW mitigation] messages to local farmers is efficient because of the language, message, visuals, and extension materials, but one of the challenge is to reach all the farmers.” Achieving fuller compliance involves precisely reaching more farmers, especially female farmers.

Study area

In Uganda, Mbarara district is the country’s second largest banana growing region and one of the hardest hit for banana production due to BXW and declining soil fertility (Namara, 2018).  For that reason, it was selected for this study.

Participants

Participants from both small-scale and large-scale/commercial banana farms in Kashaka parish (in Bubaare subcounty, Mbarara) were recruited on the criteria of farm accessibility, willingness of farmers to take part in the study, and the prevalence and past mitigation efforts against BXW in the area. Previously, Kashaka parish had had the highest number of BXW-affected farmers (during the highest-impact period of BXW from 2013-2016) but in the last five years have registered fewer cases of the disease at the District Agricultural Office. In all, 17 participants (detailed in the findings below) were identified and formally consented to participate in the study.

Data collection

During February, 2019, researchers collected data from group and open-ended, in-depth personal interviews with participants, a technology usage questionnaire, and feedback on a video animation format for delivering information and communication technology (ICT) educational content on BXW mitigation. Researchers also conducted farm walk-throughs of banana plantations to collect observational data on farm acreage, plant appearance, application of control measures, modifications to the five (government-recommended) control mechanisms suggested for the control of BXW (Jogo et al., 2011), and inter-cropping strategies.

Group and individual interviews explored topics around farmer experiences with BXW, methods used to control the disease, where they had learned of those methods, any modifications to the methods they applied, if any, as well as success stories and identification of further support information participants desired about the symptoms, transfer, and mitigation methods to control, prevent,  or  eliminate  the  disease. As part of the group interviews,

qualitative feedback from participants was elicited after presenting an educational animation on BXW mitigation (translated into the participants’ most comfortably spoken local dialect). All group and individual interview data were recorded, transcribed, and member-checked with participants to enhance validity (Birt et al., 2016), coded independently by two researchers using a framework of social wellbeing around themes and subthemes related to farmer experiences with BXW (Saldaña, 2016), and then conferenced by both researchers to resolve any coding discrepancies and achieve full inter-rater agreement (Gwet, 2014).

Participants also answered a questionnaire about ownership of technology channels (e.g., radio, TV, mobile phones, and Internet), how they received communications about BXW (e.g., from media, including newspaper, agricultural regulation authorities, other farmers, along with the kinds of information shared, frequency of sharing, sharing-media, and the feasibility of sharing information using different media when relevant to BXW), and socioeconomic impacts of BXW control methods.

Data analysis

Interview data were coded by two researchers and conferenced to resolve any discrepancies until an inter-rater agreement of 100% was achieved (Gwet, 2014). Codes were then clumped into themes and both member-checked and triangulated against documentary data (Mathison, 1988; Olsen, 2004). For questionnaires, dichotomous answers were tallied and more open-ended question items were coded as above to triangulate group and individual interview data (Mathison, 1988; Olsen, 2004).

In total, the 17 participants included 1 extension agent (in-depth interviewed), and 16 farmers (12 group-interviewed and 4 open-endedly interviewed personally in-depth). Demographically, 10 farmer-participants (63%) were male and 6 (37%) were female. Educationally, 4 (25%) had no formal education, while 4 (25%) had a primary-level (Grade 1 to 7) and 8 (50%) had a secondary-level (Grade 8 to 13) formal education. Thematic analysis of all collected data yielded two main themes related to social wellbeing (as part of the social pillar of sustainability): namely (1) social contexts generally (including farm coverage, habits, and trans-boundary issues) and (2) key issues affecting full compliance with BXW mitigation measures (including challenges related to disinfecting tools and the limited reach of BXW mitigation messaging). There are addressed in turn below.

Socio-demographic contexts

Farm coverage and practices

Out of four farms visited, few cases of BXW were observed, with an infection rate of approximately 1% of the total banana plantation crops in the study. Despite this low prevalence, all participants expressed concern about the disease, in part for having been made acutely aware  of  the  problem  during  the  2013-2016  height  of BXW (Gotor et al., 2020). The most-reported methods of control included the use of JIK (a disinfectant), iodized salt, rock salt (locally referred to as kisula), and wood ash to sanitize farm equipment during and after use. Wood ash is also poured onto infected cut banana stumps and stems to control transmission from one plant to another. The majority of participants suggested that more mitigation training, as well as lowering the cost of disinfectants like JIK, was needed, echoing other research (Gotor et al., 2020; Kikulwe et al., 2018; Kilwinger et al., 2019). Although 90% reported using JIK as an effective disinfectant for farm equipment during and after pruning banana plants, they also highlighted the issue of counterfeit JIK, suggesting that authorities should better ensure a more reliable supply of authentic products.

Trans-boundary themes

BXW is a trans-boundary problem that readily crosses international nation-state boundaries (McCampbell et al., 2018). While this requires regional coordination, solutions must still be made locally and culturally practical and feasible (Schmidt, Stiefel, and Hürlimann, 1997). In the present study, only the extension agent explicitly addressed the trans-boundary aspect of BXW.

This is a cross-border disease. Communication with Uganda (from other regional actors) in terms of research and findings is taking place, and implementation of any solution strives to be sustainable. Different stakeholders are involved in the research so that it becomes sustainable. The extension agriculture office is involved at the district level and at the regional level. They can share some data, but this problem has affected the livelihoods of communities in the affected areas. Coordination is difficult. There are seven districts and eight local governments. And data from Tanzania  (Mapinda et al., 2019) is suggesting that contaminated soil may be more important for transmission than thought previously.

Importantly, Mapinda et al. (2019) also emphasize infected tools as an important transmission vector in Tanzania. Because JIK is not effective when tools are dipped for less than three minutes (Mwangi et al., 2007), this impacts the fifth recommendation above - namely, to exclusively use only disinfected tools and at least three pangas for cooking. Dipping and waiting for more than three minutes per use for a single tool to disinfect when cooking and working in the field is prohibitively time-intensive. Fire also can disinfect field and cooking tools (Shimwela et al., 2016), but is also typically too time intensive to ensure effective sterilization or compliance (Jogo et al., 2013).

Key problems

Given  labor-availability  differences  (e.g.,  farm  or family size) (Jogo et al., 2013) and drastic reductions in national-level support such that “control of the BXW [has been] entirely entrusted to rural households” (Gotor et al., 2020), a do-it-yourself (DIY) ethos may be needed to achieve the necessary diffusion of “cultural BXW control practices” (Kikulwe et al., 2018:88), especially among less labor-equipped or poorer farms. Full BXW control means persuading partial adopters to become full adopters. Accordingly, two main themes emerged as constraints on full BXW mitigation compliance: (1) issues around disinfection of tools, including cost, counterfeit JIK, and the cultural infeasibility of requiring the purchase of additional tools to ensure sterilization, and (2) insufficient reach of BXW mitigation information to all potential message recipients, particularly women.

Disinfecting tools

Participants noted that full adoption of recommended BXW mitigation measures adds considerable physically taxing work to farmers’ everyday lives, all the more so on family or small farms with limited labor, echoing previous research (Jogo et al., 2013). As such, farmer decision-making about compliance will involve perceptions about the value of BXW mitigation practices (Bagamba et al., 2006) and weighing any additional time-consuming and physically exhausting compliance.

Social wellbeing, as a major component of the social pillar of sustainability, involves a person’s “relationship with other people, the neighborhood, and the community” (Keyes and Shapiro, 2004:350) and their “social integration, social contribution, social coherence, social actualization, and social acceptance” (Keyes, 1998:121). Not being able to realize these goals due to exhaustion or a lack of time ruinously impacts social sustainability. Requiring mitigation compliance without supporting these social goals risks decreased BXW control, physical and mental health, and decreased productivity (Bentein et al., 2017; Ferreira et al., 2019; Kloutsiniotis and Mihail, 2020; Molia et al., 2015). As such, in addition to the economic (financial) impacts involved tool disinfection compliance (Kilwinger et al., 2019; Langyintuo and Mekuria, 2008), the prohibitive and additional social costs involved in the time-intensiveness of disinfecting tools, whether by JIK or fire, affect compliance and BXW control as well (Jogo et al., 2013; Mapinda et al., 2019; Mwangi et al., 2007).

The most ideal outcome would be for farmers to adopt all BXW mitigation recommendations (Bagamba et al., 2006; Gotor et al., 2020; Jogo et al., 2013; Kikulwe et al., 2019). In fact, tool disinfection mandates are not the most noncompliant item. At the beginning of the most recent BXW crisis in 2013, Jogo et al. (2013) measured the percentage of tool disinfection compliance as third and second-best (of six items) for East African highland banana (68.6%) and Kayinja ‘beer’ banana (58.4%) production systems and high-altitude (78.2%) and medium-to-low-altitude (59.2%) farms,  respectively,  and third-best overall (63.5%). By 2019, however, approximately 11% of farmers perceived tool disinfection not effective (Kikulwe et al., 2019). More research would be needed to determine why Ugandan farmers complying with this mitigation item compared to others is not (or no longer) a good use of their time within the daily calculus and allocation of their farm activities.

Limited reach of BXW mitigation messaging

With a highly infectious disease like BXW, ensuring disease-mitigation knowledge delivery to the widest possible number of message recipients is needed (Bello-Bravo, 2020a; Biruma et al., 2007). Participants reported receiving BXW mitigation messaging through traditional channels (e.g., government/extension workers, other farmers, and media, including radio and televisions) and digital ICT channels. The extension agent observed that:

Farmers have mobile phones and can receive SMS messages and also have access to radio, posters, calendar, cartoon books to convey information. There are radio talks twice a week; farmers call and ask questions and the feedback is very positive. Also, some radio programs offer agricultural educational programs for farmers. Almost every farmer has a radio. TV depends on the locality but TV is more effective. Farmers ask questions on agronomic practices. [Our institution] organizes outreach through face to face, visiting farmers and asking them questions. Sometimes they organize training; this is done in Swahili.

While all participants reported having radios and the ability to access local radio stations, only 60% reported having TVs. Despite these available communication channels, participants did not report currently receiving any BXW messaging. Several noted that BXW mitigation programs had been previously available on these channels but were funded by projects that had since gone off the air.

Although all farmers reported access to at least one information and communication technology (ICT) channel -including radio, television, mobile phones, and the Internet—rural Internet access could vary widely, particularly given poor wireless infrastructures. This echoes similar findings on challenges facing rural Internet and mobile phone usage in Uganda (Boulahya et al., 2005; Eilu, 2018). For example while all participants reported having mobile phones, only 40% had video-enabled smart phones, with the remainder having medium- to low-technology phones usable only for receiving and sending texts and calls. Several also acknowledged not using the text feature on their phones, while the majority reported a preference for video messages and audio. More specifically, while 100% of participants reported using mobile phones as a flashlight and for personal and business calls, only 80% used them for texting, and only 10% used them as a learning tool (e.g., to access information about BXW mitigation).

Mobile phones are being underutilized for educational purposes. Often, the costs of mobile phones, data plans, or limited coverage affect this and can necessitate phone sharing (Chiumbu, 2012; Wesolowski et al., 2012). For example, all participants reported having access to neighbors’ phones (including for sending and receiving money and watching videos). In terms of usage, 75% reported keeping their phones on at all times, while 25% reported not being able to do so due to a lack of, or no access, to electricity for recharging or interruptions of (upstream) network connections. Moreover, with respect to receiving messages, 100% of participants reported text-capabilities on their phones, but 20% reported difficulties or barriers to receiving them, including technological limitations or malfunctions and non-print literacy in the dialect used for the text. Among the 80% of participants who reported understanding text messages, 50% stated that text messages in the local dialect were easier to understand; the other 50% reported being comfortable with text messages in English and Luganda

BXW mitigation animation

With respect to feedback on the BXW mitigation educational animation shown to participants, 100% of participants with smartphones expressed a willingness to follow such agricultural problem-solving videos if sent to their phones.

Participants were also very enthusiastic about the practice of the method illustrated in the video, which echoes the 80%+ positive perception of existing BXW mitigation recommendations (Kikulwe et al., 2019). However, many participants added that the steps or required materials for the method would have to be within their means (accessible and affordable); perhaps relatedly, the recommendation to disinfect tools also had the highest “not effective” perception (~11%) by farmers in Kikulwe et al. (2019). This perception of ineffectiveness almost certainly makes any time-calculation about the value of BXW mitigation compliance less convincing.

Study participants agreed that it was easier for them to follow and utilize a video or a message with a local dialect voiceover. Relatedly, Kikulwe et al. (2019) found that farmers perceived non-print forms of communication, including radio shows, farmer-to-farmer interactions, and extension visits, as more effective than print forms, which likely relates to print literacy and educational levels generally. However, audio (non-print) delivery includes the social aspect of spoken-word communications (whether in person, by radio, or through the use of voiceovers in the video animations); for example, Tata and McNamara (2016) found that female learners in Kenya preferred face-to-face rather than computer-mediated learning. Similarly, animated educational videos can elicit participant interest simply through  hearing  their

mother-tongue in (animated) recorded media for the first time in their lives (Bello-Bravo et al., 2019; Bello-Bravo et al., 2011; Rodríguez-Domenech et al., 2019).

Besides the personal appeal of the animation presented in this study, participants also reported a willingness to share the video, when means permitted, with others (e.g., neighbors, friends, and relatives), either electronically or by phone-sharing. Electronic means can include Bluetooth, if the person is close by, apps like WhatsApp if access to the Internet is available, or simply by replaying the video on an owned or borrowed phone (Bello-Bravo and Pittendrigh, 2018). An additional 20% reported being able to use a flash drive to share files by transferring them. This form of sharing not only ensures the (recorded) message’s informational integrity when the video is replayed (Bello-Bravo and Pittendrigh, 2018) but also affords reaching community members, especially women, not otherwise likely to be reached (Bello-Bravo et al., 2020).

Gender

Although reaching female farmers is recognized as a critical need in Ugandan banana farming (Gotor et al., 2020; Kikulwe et al., 2018), in part because women more often implement actual BXW mitigation measures (even on male-owned farms) and “are more engaged in day-to-day management of banana plantations” than men (Kikulwe et al., 2019, p. 11), social and educational barriers alike generally constrain their access to BXW mitigation-type messaging and can decrease adoption (Kikulwe et al., 2018). That linguistically localized animations have demonstrated statistically significant knowledge transfer without regard to gender, education, or socioeconomic level (Bello-Bravo et al., 2018) makes them well-suited for bypassing these social constraints, while also enabling women’s neighbors, friends, and relatives to informally share the video on mobile phones where they work or live (Bello-Bravo et al., 2019).

Green economy efforts rest on and rely upon incorporating economically, socially, and environmentally sustainable practices. While acknowledging the tendency to predominate or prioritize economic costs and benefits within sustainability efforts, this paper also highlighted the social impacts of time-costs on Ugandan farmers’ social lives and their social wellbeing as a factor for BXW mitigation  partial compliance. Future research might apply themes and insights from the literature on burnout, emotional exhaustion and work productivity, and their impacts on social wellbeing for Ugandan banana farmers (if not African farmers generally) to better support the social pillar of sustainability. Currently, it seems little to no such research has been done on this issue.

Moreover, as investments in female farmers’ knowledge typically have greater returns than similar investments with men, a participatory design process to (1) identify a cost- and time-effective local tool disinfection technique, (2) proof-concept it, and then (3) present its steps in a clear, simple, and locally translated animated educational protocol sharable on video-enabled phones in Uganda will not only potentially increase BXW mitigation perceptions and compliance for this item but also afford an opening for reaching underserved and overlooked members of the Ugandan banana production value-chain generally. Such enhanced BXW mitigation compliance and better food security overall for Uganda secure broader trans-boundary benefits regionally.

While the above clearly lays out an argument for recommending the use of animated educational videos to readily extend the reach of current BXW mitigation measures and potentially increase mitigation compliance, other barriers intrude. For example, only 40% of participants reported having video-enabled smartphones, while 80% reported that the majority of the people they would want to send such an animation to would likely not have smartphones and thus could not receive or retrieve such videos. This is a problem of cost, not technological literacy; among the 60% of participants without smartphones (and thus unable to receive or send videos); the cost of smartphones was cited as a barrier, with only 20% reporting a lack of education or skills as a barrier. Of those able to receive and send videos with smart phones, only 25% reported being able to access that service all day and all the time; the remaining 75% cited the high price of internet services or inadequate access to air time as a barrier - mobile internet costs approximately 2.00-3.99% of monthly income (DW, 2021). Participants also agreed that having to walk long distances (or pay transport fees) to buy air time or access airtime loading kiosks limited any ability to receive or send videos. Accordingly, participants proposed the use of printed airtime vouchers over electronic systems, since one could then buy several vouchers at once and keep re-loading data plans at their convenience.

Even assuming a financial feasibility to providing no-cost video-enabled smartphones with free data plans for viewing and sharing BXW mitigation videos that emphasize tool disinfection, this would still not address the social time-cost of the requirement. Rather, we recommend - echoing Jogo et al. (2013) - a collaborative process with Ugandan smallholder banana farmers and global and local BXW mitigation experts to:

1) Identify a readily available, low- to no-cost, and socially sustainable tool disinfectant solution-method,

2) Perform a proof of concept of easily replicated and learned steps for applying that method, and then

3) Produce an educational animation translated into local dialects that demonstrate those method-steps for farmer training and dissemination through the existing smartphone resources in Uganda.

Notably, a similar approach by Mocumbe (2016) measured a 97.9% knowledge retention and 89% solution-adoption at a two-year follow-up post-training for an improved postharvest seed-storing approach using jerrycans (Bello-Bravo et al., 2020).

Self-evidently, the constraints imposed by limited smartphone ownership among Ugandan banana farmers mandate creative local solutions for enabling the full realization of the participatory process above. However, whatever protocol is determined for a specific, actionable, and socially sustainable method for disinfecting tools, such protocol content and any steps for full compliance with BXW mitigation measures could be included in the educational animation, thus extending the reach of BXW messaging to demographics currently missed or overlooked, especially women. Moreover, the DIY context that now places responsibility for BXW mitigation predominantly on farmers and farms means that grassroots efforts and volunteerism may be the only (or best) resource for realizing a socially sustainable, time-effective solution for securing tool disinfection compliance. By no means the most ideal route - given that better funded, or actually funded, efforts would more readily enable the participatory project above - if this is currently the only visible pathway, then one must take it (pending still-better options) given that securing some additional compliance for BXW mitigation is more desirable than achieving none or leaving it to chance. Through such an approach, the social wellbeing of farmers - in terms of more access to daily life by having more time away from work during the day - affords critical support for social sustainability in general.

Funding for this research was provided by Michigan State University’s International Studies and Program’s Alliance for African Partnership.

The authors have not declared any conflict of interests.