The state of traditional foods in rural Imbabura diets
Traditional foods remain a part of daily life for farmers in our study population, but there is no bar to gauge how much traditional food consumption is “enough” to curb TF displacement and mitigate the nutrition transition toward foods that contribute to a double burden of over- and undernutrition. Most farmers consume at least half of the indicator TFs assessed, and they consume them often: agroecological farmers report consuming indicator TFs 260 times a year, and reference farmers do so 144 times a year. All farmers continue to practice wild harvest to some extent, and most do so on a weekly basis. TF consumption appears more alive in this farming population than in other spaces in the country; for example, a recent representative study in three Ecuadorian highland cities found that only 19% of participants consumed either quinoa, amaranth or Andean lupine more than three times per month (49). The comparable figures in our study population would be 60% of reference farmers and 85% of agroecological farmers. Even the indicator TFs that we selected because they are locally recognized as underutilized (amaranth, yacón, oca, mashua, quinoa leaf, amaranth leaf) are all still present to some extent in our study population’s diets. Some of these products are receiving attention for their potential to support dietary health. For example, amaranth seed is recognized for its protein and lipid profiles(50), and amaranth and quinoa leaves are green leafy vegetables with high concentrations of nutrients that are of special concern in the Ecuadorian rural population, notably vitamin A, iron, calcium, zinc and vitamin C (19,50,51). Even though some of these products are only marginally alive in the diets of reference farmers (i.e., with median consumption of only once yearly), they point to opportunities to strengthen the use of endogenous foods to support nutritional health.
Opportunities for traditional food promotion
Our analysis suggests that TF consumption is associated with TF production. This is no surprise in light of the expanding literature on the pathways between production and consumption, and namely production diversity and dietary diversity (52). Indeed, we find that farmers that grow a given TF are not only more likely to consume it, but they also consume it more frequently. Some underutilized TFs are exclusively obtained from own harvest. For other TFs, farmers who do not produce them obtain them from farmers who do, relying on social economy transactions such as barter or direct purchase. That these underutilized products are never purchased at markets is likely a consequence of their reduced availability (53), and signals the importance of the social economy in filling supply gaps.
The diversity of TF products grown on the farm is associated with higher overall farm production diversity of edible products. Nevertheless, it is unlikely that increasing agrobiodiversity alone would guarantee an increase in TF production diversity. Instead, the association we detected may reflect adherence to more traditional cropping systems, which depend on relatively high agrobiodiversity (1), or it may be a reflection of the diversity supported by the ecological niche. While there may not be a direct causal relationship between overall farm production diversity and TF production diversity, the two may be mutually reinforced as farmers and organizations aim to increase farm production diversity for ecological, productive and nutritional reasons (32). Doing so by targeting TF production diversity may be particularly relevant for nutrition-sensitive agriculture initiatives, given that TFs are shown to simultaneously contribute to agricultural resilience, food access (54–57) and to dietary intake of key macronutrients, micronutrients and phytochemicals (56–61), and they further play a protective role against chronic diseases (6,8,59).
We further find that farmers whose diets rely less on conventional markets and more on own harvest or the social economy maintain stronger TF practices. Other scholars similarly discuss the importance of non-market subsistence practices such as own production and local trade in conserving traditional crops (5,23). In contrast to other studies (4,26,28,29), market distance, income and age did not emerge as strongly or consistently associated with TF practices among our study population. This means that in this context, TF practices are not merely a relic of the most isolated, impoverished and aging—or in short, marginalized—people, as public opinion has long perceived them to be (7). In the development literature, practices that are the purview of the most marginalized people, and especially of subsistence-oriented farmers, tend to be discussed as “coping” or “adaptive” strategies driven by reactive necessity rather than proactive agency (62). In contrast, the fact that we detected an association with reliance on non-market food sources but did not detect a strong association with marginalization implies that TF practices in our study population are not merely a reaction to adverse conditions. Possibly, farmers may be participating in a globalized cultural shift toward re-valorization of TFs, as has been described in Europe (63). Doing so, some may even perceive TF practices as active agents in strengthening cultural identity and food sovereignty (36).
Agroecology as an incubator for traditional food promotion
Agroecological farmers unambiguously perform better than their reference neighbors on three of the four TF practices assessed. They produce twice as much TF diversity, consume 40% more TF diversity and consume TFs 80% more often compared to their reference counterparts. In our path analysis, participation in agroecology was directly associated with both TF production diversity and TF consumption diversity, leading to a downstream association with TF consumption frequency. While we did not measure changes over time, agroecological farmers emphatically identify their participation in agroecological markets as the drivers of increased TF production and consumption, pointing to agroecology as a means to strengthen TF practices. Moreover, the strongest differences in consumption of specific TFs appear precisely in those that are locally recognized as underutilized. Agroecology may thus be key for reclaiming at-risk TFs in this region and re-inserting them into healthy dietary patterns.
Part of the reason why agroecological farmers in our population perform so much better on TF practices may be because agroecology explicitly promotes farm production diversity and reliance on non-market food sources (34,36), which are correlates of TF practices. Yet even when these are held constant, agroecology participation still shows an association, suggesting that other forces are at play. Focus group discussions help clarify these unknowns, identifying two additional potential drivers that may motivate agroecological farmers to increase their TF practices.
First, the social environment of the agroecological market association may drive farmers to produce and consume TFs for their nutritional properties, taste, agricultural resilience, cultural value and even aesthetics. While such convictions around TFs are also found among other farmers in Northern Ecuador (64), the social encounters in agroecological spaces appear to further concentrate these convictions by inserting TF practices into social norms that strengthen a shared cultural identity. Further, they seem to embed TF consumption into the moral impetus of feeding healthy food to the family. The importance of these socially-driven elements in guiding TF practices is consistent with dietary behavior models that find food decisions to be informed by “affective” components, including feelings and emotions, moral obligations, and social norms and pressures (65).
Second, focus group discussions also identified the specialized consumer demand for TFs in agroecological markets as a potential driver of TF practices among farmers. Other studies on TFs similarly find that consumer demand-driven value chains influence TF production (66,67). However, discussion participants further sustained that when they grow TFs for sale, they also increase their own consumption. These flows of influence are probably bidirectional, given that agroecological farmers’ associations played an important role in the emergence of a nation-wide campaign to form “responsible” consumers that seek out traditional Andean crops as well as nutritious, socially just and ecologically sustainable food (68,69). As such, there appears to be a feedback loop between agroecological market farmers and clients in forming affective spaces (36) that support traditional foods.
Wild harvest
Wild harvest appears to be a common practice among our study population, even though most of the harvested foods are consumed in small amounts as condiments or snacks. As far as we can tell, wild food consumption diversity among our study population is not associated with age, income, distance to markets, food acquisition strategies or agroecology participation. This relative democratization is compelling given that wild foods can be remarkably nutritious, but also remarkably neglected and underutilized (7,70,71). This combination often relegates wild foods to coping strategies for the poorest of the poor and erroneously dismisses them as “famine foods” (7,70). While we were unable to detect plausible pathways promoting wild food consumption, we find that people who consume a greater diversity of TFs in general also consume a greater diversity of wild harvested products, potentially signaling similar drivers for these two dietary outcomes. While our findings suggest that wild foods have not been prioritized by the local agroecology movement, its unique affective and commercial spaces may hold the enabling conditions to effectively promote wild foods.
Internal and external validity of findings
We believe a word of caution is warranted regarding our data on TF consumption frequency, given the cognitive recall difficulties that beleaguer FFQs (72) as well as the added complexity of seasonality (73)[1]. However, farmers participating in focus group discussions corroborated the detected pathways between TF production and consumption. This triangulation between qualitative and quantitative methods gives us more confidence in our findings, despite the relatively small sample assessed in the survey. Nevertheless, we only conducted FGDs with agroecological farmers and we are uncertain of the subjective biases at play. FGDs were also key for identifying farmers' perceptions of causality between agroecology and TF practices. Moreover, path analysis has the advantage that it can identify likely chains of influence, even with cross-sectional data (74). While neither the subjective experiences of farmers nor the results of path analysis are sufficient to definitively establish causality, the triangulation of the two strengthens the internal validity of our results. Nevertheless, our study is limited to a single region, and we recognize that many contextual factors could affect external validity. Not only is agroecology a term that embraces many local expressions (30), but other factors that are subject to broad variation include the cultural presence of TFs, ecological context, food acquisition patterns, gender norms around food and many more. Rather than providing a proscriptive formula for strengthening TF practices, it is our hope that we shed light on how these possible paths can play out, recognizing that they will likely be different in other localities.
[1] Farmers in our study population would often report that they ate a given product “every day while it’s available”, which, for a product that is available for two months, would lead to an unrealistic estimated frequency of 60 during that time period. While we believe this inflation would be equally distributed across both agroecological and reference farmers, we do not have a reliable means of correction in order to obtain a more accurate TF consumption frequency.