ABSTRACT
The global spread of Covid-19 led to travel and market restrictions that impacted Guatemala’s rural food systems. This distinct economic shock directly affected campesinos, or small-scale farmers who depend on subsistence and commercial food production. Some Guatemalan farmer organizations have been promoting agroecology for decades in efforts to strengthen rural livelihoods and food sovereignty, defend Indigenous rights, and adapt to climate change, and agroecology is positioned as a tool for resilience to various shocks. We consider the neoliberal cooptation of the concept of resilience, and its usefulness in preserving alternative and previous (Indigenous) practices. Data from surveys and semi-structured interviews with farmers and leaders at eight organizations that promote agroecology suggests that prior engagement with a farmer organization, enacted through both agroecological practices and social networks, contributes to campesino resilience to the pandemic’s economic shock at the farm level, with regards to production and consumption. This study illustrates the range and diversity of strategies taken up by campesinos during the pandemic, and considers the importance of social networks for collective actions that increase current and future economic solidarity in campesino communities.
Acknowledgments
We are deeply grateful to the numerous farmer participants, research assistants, staff and farmers from eight organizations across Guatemala. Without your collaboration, kindness and commitment to local food system development, none of this would have been possible. We acknowledge generous support on statistical analysis and sharpening points about campesinos’ constraints and resilience from Matthew Turner and Lisa Naughton. We also thank three anonymous reviewers for their suggestions and improvements.
Disclosure statement
No potential conflict of interest was reported by the author(s).
Notes
1. Campesino, or “person of the country”, refers to a rural Latin Americans who subsists off of diverse economic activities (including farming), centers horizontal social relationships and collectively shares the burdens of resource poverty. Campesino situates study participants in post-modern Latin America (Loker Citation1996) and reflects rural farmer movements’ terminology (FUNDEBASE Citation2012). It is important to note that not all campesinos are farming agroecologically. Campesinos are generally considered smallholders, or small-scale farmers.
2. Due to variable capacity and changing pandemic-related restrictions, some research assistants conducted fewer than 12 interviews.
3. For Mancomunidad Copanch’orti’, Jocotán and Camotán were combined as participants frequent both markets. For Red Kuchub’al, a weighted average was taken between Sibinal, Tacaná, and San Marcos municipalities, accounting for participants with dispersed locations.
4. For ordered logistic models, standard ancillary parameters called cutpoints (or thresholds) are presented in which are used to differentiate between the adjacent values of the dependent variable. Cutpoint 1 is used to differentiate lowest ordinal value from the middle and high values and cutpoint 2 is used to differentiate between low and middle values from highest value. These cutpoints are presented in order to include all details in the models, but are not values of particular interest for our analysis.
5. The suitability of two-level logistic regression analyses (random intercept) with organization serving as the level 2 random-effects variable was assessed by evaluating the significance of random-intercept null models (with no fixed factors) of all dependent variables on the organization random effect. Two-level logistic regressions were only performed if the null model was significant. This was true for all dependent variables except for the lack of agrochemical inputs variable, which was thus analyzed using a simple logistic regression. STATA software with the multi-level regression commands of meologit and melogit were used for ordered logistic and simple logistic regressions respectively. For all two-level models, random intercept rather than random coefficient models were determined to be the most appropriate.
6. Level 2 statistics include the intraclass correlation coefficient (ICC) (also referred to as rho), which is a measure of the variation in the independent variable that can be attributed to differences among the level 2 clusters (organization).
7. Given low sample sizes and their geographic proximity, two of the organizations were combined for the statistical analyses only. ACPC and Qachuu Aloom organizations were combined, leading to a total of 7 groups in the statistical analyses.
8. Measurements originally given in cuerdas. 1 cuerda = 441 m2.