Materials and methods

Study Area. ST live in hundreds of scattered settlements organized in seven ancestral, socially and politically independent communities, each one composed of a religious and political center, several medium to large settlements (anexos) and a large number of small settlements.

The territory inhabited by ST is among the largest (9,380 km²) and rugged areas occupied by indigenous peoples in Mexico; it is mainly in Durango state with a few settlements in the states of Zacatecas and Nayarit. Two landscape units (ecoregions) of the SMO are present: Madrean and Tropical (^{González-Elizondo et al. 2012}); the first generally above 2,000 m, with temperate and semi-cold climates, and mostly pine and pine-oak forest; the later enters the SMO through the deep canyons on the western flanks, where the elevation ranges from 540 to 2,200 m (Figure 1).

Figure 1 Southern Tepehuan territory in Durango, Mexico. 

Data collection. The field work was carried out from September 2017 to November 2018. We worked with 162 young Tepehuan informants from three different educational contexts: 1) Rural students (with formal education received within their communities (Figure 2); both high school and undergraduate, all of them studying in La Guajolota settlement); 2) Urban students (with formal education; high school and higher education, received outside their communities in an urban context, in various school centers in Durango city); 3) Non students (young people dedicated exclusively to field activities such as agriculture, within their communities). The interviews and the questionaries were conducted in Spanish, always with the company of an O'dam translator who helped with details about Tepehuan terminology.

Figure 2 Some students of the Technological University of Mezquital and the first author during the fieldwork in Santa María de Ocotán, Durango. 

Each informant provided personal data (name, age, community of origin and current grade or school situation); answered a ten questions questionnaire (Table 1) about 20 selected species of wild edible plants of the ST ethnoflora. The species selection criteria were: to represent different environments in the region and, according to previous field studies, to include some which TEK seems to be well spreading among people as well as some species previously recorded based on one or a few informants (Table 2). These were shown to informants using a catalog with images of each species. Once the ten questions per plant section was complete, we asked each informant to write a free list of other edible plants of the region that they knew of.

Taxonomic identification of the species mentioned, by common name, in the free-list instrument, was done combining two main procedures: 1) search for the common local names in the CIIDIR herbarium database, and 2) key informant interviews and botanical collections with their participation, specifically aimed at finding the listed species. Plants collected were processed through conventional techniques, identified using specialized literature and deposited at CIIDIR herbarium.

Data analysis. Based on answers to six questions (1, 2, 3, 4, 5 and 9) about each of the 20 plants of the visual catalog, plus those of the free list of each informant, a traditional knowledge grade per person was determined. This grade was calculated from the sum of one point for each correct answer, the “correctness” based on the ethnobotanical and ecological information on each species available in the CIIDIR Herbarium database and in previous ethnobotanical studies. To analyze how the degree of traditional knowledge varies depending on the socioeconomic characteristics of the informants, using the STATISTICA 7 software, an analysis of covariance (ANCOVA) was carried out; with a significance level of 95 % (α = 0.05) and taking the traditional knowledge grade as the dependent variable, age as a covariate, and the educational context and gender as factors. The residuals were previously calculated and based on these, the assumptions of normality and homoscedasticity of the variances were confirmed through the Kolmogorov-Smirnov (P > 0.05) and Levene (P > 0.05) tests respectively. Also, a Tukey test (α = 0.05) was performed to determine which means differ among the factors.

Moreover, homogeneity group Chi-square tests (α = 0.05) were carried out for answers to questions 2, 7 and 8, to analyze whether there is an association between the educational contexts and: 1) the knowledge of the Tepehuan names of the plants; 2) a positive perception of the consumption of each plant; 3) the knowledge of the season in which each plant can be used for edible purposes.

Answers to questions 6 and 10 were recorded and analyzed only through percentages, being useful to know aspects such as the form of consumption, as well as anecdotes or comments to know some aspects about the dynamics of the cultural transmission of traditional knowledge, which were classified according to the three categories proposed by ^{Hewlett & Cavalli-Sforza (1986)}: vertical, horizontal and oblique transmission.

Results

Informants. A total of 162 people in an age range of 15 to 30 years participated as informants in this study. According to the three educational contexts considered, they were distributed as follows: rural students (n = 83; 32 men and 51 women; mean age: 18 years old); urban students (n = 46; 22 men and 24 women; mean age: 17 years old); and non-students (n = 33; 15 men and 18 women; mean age: 22 years old).

Comparison of traditional knowledge grade. The ANCOVA analysis indicate that the variables age (F _{1, 155} = 23.73, P = 0.000003), gender (F _{1, 155} = 5.06, P = 0.025756) and educational context (F _{2, 155} = 8.84, P = 0.000231) were statistically significant; not so the interaction between the educational context and gender (F _{2, 155} = 1.98, P = 0.141267). Moreover, a Tukey test found that the traditional knowledge grade between rural students (x̅ = 49.434) and urban students (x̅ = 44.152) does not present statistically significant differences, in contrast with non-students, whom had a higher level of knowledge (x̅ = 69.765) (Figure 3); in addition, women (x̅ = 53.827) had a higher traditional knowledge grade than men (x̅ = 49.376).

Figure 3 Comparison of traditional knowledge grade between three educational contexts of Southern Tepehuan informants. 

Knowledge about plant names in Tepehuan language. Answers to question 2, based on the Chi-square test, show significant differences between the three groups of youngsters with respect to the knowledge of common names in Tepehuan for 13 of the 20 visual catalog species (Supplementary material Table S1). No significant differences were found for seven species: Agave shrevei, Agave vilmoriniana, Arctostaphylos pungens, Leucaena leucocephala, Prosopis laevigata, Sinclairia palmeri and Tigridia multiflora. For the taxa that did differ, the group of non-students cited the Tepehuan common names of most of the species in a higher percentage, except for those of Dahlia coccinea and Begonia tapatia, cited in a higher proportion by rural and urban students respectively.

Form of consumption. The percentage of informants who showed knowledge about the form of consumption was higher among non-students for 19 of the 20 plants included in the visual catalog (except for Agave vilmoriniana), followed by the group of rural students (Figure 4). The forms of consumption of five species (Arctostaphylos pungens, Leucaena leucocephala, Lippia graveolens, Prunus serotina and Stenocereus queretaroensis) were cited by 100 % of the group of non-students, while there were reported by 63.9 to 90.4 % of rural students and by 20.5 to 48 % of urban students (Supplementary material Table S2). On the other hand, the plants whose form of consumption was less known were: Salvia hispanica, Tigridia multiflora and Tripsacum dactyloides; the first and the last were only cited by very few informants of the non-students group.

Figure 4 Comparison of the percentage of informants from the three educational context groups that cited traditional knowledge about the form of consumption of 20 plants. 

Perception about the consumption of each plant (Do you like to eat this?). Statistically significant differences were found about the perception of consumption for 17 of the 20 plants included in the visual catalog; being the group of non-student informants the one with the highest percentage of people with a positive perception of plant consumption (Supplementary material Table S3). The three species in which no significant consumption perception difference was found were Agave shrevei and Arctostaphylos pungens, with high, and Agave vilmoriniana with low positive perception by the three groups.

Knowledge about season of availability of each plant. Regarding question 8 (In what season of the year can you get it?) there are statistically significant differences in the knowledge of the three groups about the availability of edible parts of the plants through seasons of the year for 16 of the 20 species, with non-students group 1knowing better about the seasons for 14 of those (Supplementary material Table S4). No significant differences were found for Agave vilmoriniana, Arctostaphylos pungens, Prosopis laevigata andTripsacum dactyloides.

Dynamics of the cultural transmission of traditional knowledge. Regarding question 10, a high percentage of responses referring to the vertical transmission of traditional knowledge were recorded in the three groups (Supplementary material Table S5), except for Stenocereus queretaroensis and Lippia graveolens in the case of non-students; for the first, a high proportion of people gave responses allusive to oblique transmission (78.7 %), while for the latter, both oblique and vertical transmission resulted in equal percentages.

Knowledge about other edible plants (free listings). Once the free listings provided by each of the 162 informants were compiled and analyzed, a general list composed by a total of 57 taxa was obtained. Particular free listings for each of the three informant groups included 42, 31 and 34 taxa for non-students, rural students and urban students, respectively (Supplementary material Table S6). The general free list includes 48 plants, 7 fungi, 1 insect and 1 indeterminate taxon.

Discussion

Comparison of knowledge among educational contexts. The results of this study indicate that the group of non-students, youngsters engaged in field activities, has a higher traditional knowledge about edible plants than students, both rural and urban (Figure 3). These results are consistent with other studies that find that TEK has a negative association with socioeconomic factors such as formal education. The main explanations for this fact are: 1) the decrease in the time that students spend in the field to handle academic activities and, consequently, the decrease in learning from nature; and 2) the marginalization of content related to TEK in educational curricula (^{Wester & Yongvanit 1995}, ^{Sternberg et al. 2001}, ^{Zent 2001}, ^{Voeks & Leony 2004}, ^{Quinlan & Quinlan 2007}, ^{Srithi et al. 2009}, ^{Reyes-García et al. 2010}, ^{Saynes-Vásquez et al. 2013}, ^{Bruyere et al. 2016}).

It is notorious the lack of statistically significant difference in the grade of traditional knowledge between rural and urban students. It was expected that the former would have a higher grade of traditional knowledge, since they live in an environment where these wild edible plants are available. Urban students, on the other hand, are immersed in an environment where deforestation and loss of biodiversity negatively impact the transmission of TEK (^{Monroy & Ayala 2003}, ^{Barreau et al. 2016}). The unexpected similarity in the grade of knowledge among the two student groups may be related to the fact that ST diet has incorporated in recent years a considerable number of industrialized foods that have displaced the traditional ones (^{Reyes-Valdez 2007}).

Regarding the knowledge about the plant names in Tepehuan language, significant differences were found for 13 of the 20 plants in the visual catalog, 11 of which were cited in a higher proportion among the non-students (Table S1). The other two groups of youngsters spend less time in nature and thus, know less about it. ^{Berlin (1972)} points out that words of the traditional classifications on biodiversity disappear when activities related to the environment decline. There was no statistically significant difference in the knowledge of seven of the 20 plants among the three groups, in three cases (Arctostaphylos pungens, Agave shrevei and Leucaena leucocephala) it obeys to the fact that such plants are widely known among all the informants; for the other four species, there is a general lack of knowledge, and less than 50 % of the informants in all groups was able to cite the Tepehuan plant names. Among the species for which the knowledge differs significantly among the groups, the less known are Salvia hispanica and Tripsacum dactyloides, cited by less than 30 % of people in each group. Interestingly, Salvia hispanica, a sage, was highly appreciated in antique Mesoamerican cultures and is at present recognized as a promising food in Mexico and the European Union (^{Muñoz et al. 2013}, ^{Xingú-López et al. 2017}). Lack of knowledge about it among ST may be related to the fact that it is scarce in the wild, where it may be a relic of old crops now in disuse in the Tepehuan zone (^{Narváez-Elizondo et al. 2020}).

According to the linguistic research of ^{Torres-Sánchez (2018)}, Tepehuan children of Santa María de Ocotán begin to stay in contact with the Spanish language early, when they enter school and, as they keep studying, the use of Spanish becomes more frequent than Tepehuan for them.

The educational curricula are carried out almost entirely in Spanish; additionally, there are some teachers that do not speak Tepehuan; this situation is accentuated in the case of students living in the city of Durango, even more immersed in mestizo culture. Furthermore, younger generations use Tepehuan and Spanish interchangeably in different situations, which could lead to a state of balanced bilingualism or Tepehuan may be displaced by Spanish. These scenarios documented by ^{Torres-Sánchez (2018)} can partially explain the low knowledge of the Tepehuan names of some plants of the visual catalog, and even the low grade of traditional knowledge about edible plants among the rural and urban students. Some other studies indicate that the loss or disuse of indigenous languages ​​is associated to the decrease in traditional knowledge about biodiversity (^{Benz et al. 2000}, ^{Zent 2001}, ^{Martínez-López et al. 2016}).

Another aspect to highlight is the greater knowledge of non-students about the form of consumption as well as the seasons of the year in which the edible parts of most plants can be obtained (Tables S2 and S4). An explanation for this is the fact that these youngers are engaged in activities of the primary economic sector (agriculture, extraction of forest resources, etc.), as has been also found in other studies (^{Reyes-García et al. 2007}, ^{Saynes-Vásquez et al. 2013}). It should be mentioned that the consumption of wild plants, mainly those consumed raw, is common during activities related to nature (^{Pardo-de Santayana et al. 2005}, ^{Maseko et al. 2017}); while plants that involve preparation prior to its consumption, can be collected in the field to be eaten after work or stored to be available even in later times, through processes as drying (^{Peredo & Barrera 2017}), as is frequently the case among ST for Lippia graveolens and Tagetes lucida.

The lack of knowledge about the harvest season of wild plants may represent a problem related to food security issues. Some studies have documented the importance of wild plants as complement or substitutes of other foods in the face of shortages during social conflicts (^{Caballero & Mapes 1985}, ^{Tardío et al. 2006}, ^{Theien 2009}, ^{Redžić 2010}, ^{Vorstenbosch et al. 2017}) or due to climatic issues (^{Laferrière 1992}).

Comparison of knowledge by age. The group of non-students with the highest mean age (22 years old) turned out to hold the highest traditional knowledge grade, suggesting that age may influence TEK, despite the narrow age range of this study (15-30 years old). A positive association between age and TEK has been found in studies with broader age ranges, e.g., that of ^{Bortolotto et al. (2015)} who worked with a range of 21-86 years old. However, there are also reports of a null association between age and TEK (^{Arango-Caro 2004}, ^{Martínez-López et al. 2016}), and therefore the effects of other socioeconomic factors must be considered together with age to explain the ethnobotanical phenomena of each society. It is important to note that the three study groups in our work were within an age range that, according to some authors (^{Zent 2001}, ^{Case et al. 2005}) corresponds to part of a very important phase of TEK learning, which begins from childhood and can culminate between 18-30 years old. So, exposure to the factors negatively associated with traditional knowledge as those discussed above, could induce additional processes of cultural erosion on other aspects linked to the useful flora.

Knowledge grade by gender. It was found that women have a greater traditional knowledge on edible plants than men. This coincides with reports in other regions, e.g., Salento, Colombia (^{Arango-Caro 2004}), Boboye, Niger (^{Dan Guimbo et al. 2011}), and El Jadida, Morocco (^{Tbatou et al. 2016}). It should be mentioned that this does not necessarily reflects a constant pattern, since some studies report not finding differences about the knowledge of this kind of food resources between men and women (^{Camou-Guerrero et al. 2008}, ^{Bortolotto et al. 2015}); or even, in some others, a greater knowledge among men is reported (^{Kujawska & Łuczaj 2015}, ^{Pío-León et al. 2017}). The pattern found among young ST can be explained through gender role differences in domestic activities, as food preparation is a task carried out largely by women, especially during some traditional ceremonies (^{Cramaussel 2013}).

Perception of wild plants consumption. The perception of the consumption of wild plants was significantly different by educational context for 17 of the 20 plants considered. The non-students group shows the highest tendency towards a positive perception of plant consumption. Good taste was the main reason associated with a positive perception for all taxa in the three informant groups. This is important if we consider what has been pointed out in other studies (^{Sõukand 2016}, ^{Thakur et al. 2017}), in which factors such as taste and good aromas are reasons that encourage the consumption of wild edible plants, contributing to the conservation of this type of biocultural heritage.

Transmission of traditional knowledge. Concerning the forms of transmission of traditional knowledge for each plant, vertical transmission was recorded in a higher percentage into the three groups of informants for most plants, except for Stenocereus queretaroensis and Lippia graveolens among the non-students, who commented on oblique transmission for the first, and oblique and vertical transmission for the second plant. Generally, a higher frequency of vertical transmission coincides with other studies that indicate the importance of parents and grandparents as the main diffusers of knowledge about wild edible plants (^{Yates & Ramírez-Sosa 2004}, ^{Eyssartier et al. 2008}, ^{Mosquera-Mena et al. 2015}, ^{Ochoa & Ladio 2015}). It is worth mentioning that this form of transmission of traditional knowledge is considered the most conservative, and very few innovations occur in this type of knowledge when it is presented alone or in a dominant way (^{Cavalli-Sforza & Feldman 1981}, ^{Hewlett & Cavalli-Sforza 1986}).

The importance of the oblique transmission on the knowledge of Stenocereus queretaroensis and Lippia graveolens for non-students, can be related to the fact that the collection of their fruits and leaves, respectively, are important economic activities for the inhabitants of the ST region. Therefore, youngsters may be learning aspects of these plants, either through people who market them or even during the same collection work, interacting with other people. Similar results have been found in cases such as that of a medicinal plant market in Tanzania, in which vendors exchange TEK with each other (^{McMillen 2012}); or among Austrian migrants in Australia, Brazil and Peru, who by working as chefs in restaurants or as kitchen instructors, transmit knowledge about food to other people, which they previously learned through vertical transmission (^{Haselmair et al. 2014}). According to ^{Cavalli-Sforza & Feldman (1981}) and ^{Hewlett & Cavalli-Sforza (1986)}, oblique transmission, which involves agents external to the family, mainly when it occurs from one person to a group, is an efficient communication process where knowledge can begin to show greater variation in less time, since it is spread intergenerationally.

Knowledge about other edible plants shown in free listings. Non-students provided the most extensive list of edible taxa (74 % of the total), followed by urban students (60 %) and, finally, rural students (54 %). It is important to note that, even though during the information gathering, participants were asked to provide information about uncultivated edible plants, in all three study groups, there were some informants who gave data regarding cultivated plants, as well as on fungi, and even on an insect. This may be related to the fact that some people do not distinguish the differences between fungi and plants, being fungi one of the most appreciated edible wild resources by ST during the rainy season (^{González-Elizondo 1991}). Further, the three informant groups mention jocolish, a wild edible resource related with an insect (family Cynipidae) that induces the development of galls on oaks (Quercus spp.) as part of their life cycle, and are precisely these galls the edible structures, therefore associated as part of these trees.

Of the total of 57 taxa included in the general free list (Table S6), 17 (30 %) are cultivated species. Four of them (Amaranthus sp., Cucurbita sp., Cucurbita ficifolia and Zea mays) are traditional crops that have been used by ST since pre-Hispanic times (^{González-Elizondo et al. 2017}). The other 13 correspond to plants that are traded or cultivated on a small scale and can also be considered exotic in the Tepehuan territory. Although the report of exotic plants is usually related to a process of cultural erosion, these taxa were considered to calculate the traditional knowledge grade of each informant since, according to ^{Albuquerque (2006)}, this situation could also be interpreted under the hypothesis of diversification, which propose that the stock of traditional knowledge about useful flora is enriched by the incorporation of exotic species. A comparative study is necessary on aspects such as the frequency of use and the proportion of useful native and exotic species among the ST to thoroughly analyze possible processes of cultural erosion or scenarios such as that proposed by the diversification hypothesis.

According to the frequency of mentions of each taxon included in the free list, the most important for each group are as follows: for non-students, Opuntia sp. and Valeriana pratensis (36 % each), Pithecellobium dulce and Spondias purpurea (30 % each), and Amanita basii (21 %); for rural students, Amanita basii (30 %), Cynipidae galls (25 %), and Opuntia sp. (19 %); while for urban students, Opuntia sp. (43 %), Cynipidae galls (24 %), and Spondias purpurea (13 %). The presence of Amanita basii and Opuntia sp. in the free list of the three informant groups is notorious; the former, as mentioned above, is one of the most appreciated food resources during the rainy season, while from the latter are eaten both the fruit and the tender stem or cladode, which is consumed plain, as a side vegetable or as an ingredient of various stews, mainly before the rainy season. The mention by informants of the three groups of an unidentified ethnotaxon called sombiadam, is a detail that indicates that it is worth continuing with ethnobotanical studies in the area to rescue TEK not yet documented.

Opportunities for revitalizing traditional knowledge. Education is a human right closely linked to well-being. However, it has also been documented that it may induce a trend towards the homogenization and disappearance of some components of cultural diversity throughout the world, by imposing visions and practices from the Western culture above local customs and knowledge (^{Aikman 1999}, ^{Bensasson 2013}, ^{Hickling-Hudson & Ahlquist 2003}). As mentioned previously, a process of cultural erosion of the TEK may be occurring in all three groups of informants, but particularly on those exposed to formal education.

A possible solution to this is the implementation and strengthening of intercultural education, which should include content related to TEK in academic activities and programs. For example, some studies have documented that people of different ages who had received or were taking educational programs with contextualized material in their indigenous languages, and whose teachers belonged to the same ethnic groups (who better understood their culture), had a higher TEK compared to people whose educational system did not have these particularities (^{Reyes-García et al. 2010}, ^{Wyndham 2010}). In this sense, it is worth noting the observations reported by ^{Torres-Sánchez (2018)}, about a greater influence of the Spanish language as Tepehuan students advances and concludes their academic formation within their territory.

Although formal education contextualized according to each local culture helps to transmit TEK, some of its elements, such as practical knowledge, may continue to be affected, since they are often learned outside the classrooms, unlike other theoretical issues (for instance: names of plants) (^{Reyes-García et al. 2010}). Therefore, in addition to the theoretical ethnobotanical content and the participation of teachers, scientists and policy makers, the participation of TEK holders from the same communities is necessary in the design and implementation of activities that promote the transmission of TEK (^{Reyes-García et al. 2010}, ^{McCarter & Gavin 2011}, ^{Sandoval-Rivera & Mendoza-Zuany 2017}, ^{Nigh & Bertely 2018}); examples of those activities are ethnobotanical field trips, ethnogastronomic exhibitions, school garden projects, and storytelling, among others. Traditional knowledge plays a crucial role to outreach a healthy interaction with the environment and provides a channel for sustainable development (^{Mazzocchi 2006}, ^{Segger & Phillips 2015}). To pursue social justice and sustainability, a collaborative intercultural dialogue and a respectful and symmetric interaction among the distinct actors of knowledge is essential (^{Argueta-Villamar & Pérez-Ruiz 2019}).

A particularity of the education system in the study area (La Guajolota) is the presence of a high school specialized in natural resources. Thus, the promotion of citizen science projects that involve the TEK from students and other community members, can be a valuable tool in the pursuit for a sustainable development in ST territory.

The results of this study suggest that among the three studied groups of young Tepehuans, non-students have a higher traditional knowledge grade about edible plants than both rural and urban students. This knowledge includes issues related to vernacular names, harvesting season, recognition of edible uses, parts of the plant that are eaten, and habitat of the plants. Results support the notion of a better TEK by people engaged in activities in the primary economic sector, and a negative association between socioeconomic factors such as formal education and TEK, probably related to the decrease in time spent living and learning in nature for attending academic activities. Furthermore, as activities related to nature lose importance, the lexicon related to biodiversity tends to decline or disappear in indigenous languages.

It was also found that women have a higher degree of traditional knowledge than men, which can be explained by the differences in gender role in this group, where women are more rooted in food preparation. This highlights the need to always involve women in environmental management processes aimed at the conservation and sustainable use of the biodiversity of the ST territory.

The socioeconomic changes that are occurring around Southern Tepehuan youth (such as migration to urban areas, less use of their indigenous language, and a formal education received almost entirely in Spanish language), is causing cultural erosion. Thus, it is important to develop strategies favoring alternative ways of transmission of the traditional knowledge, e.g., its inclusion into the educational curricula and the promotion of a sense of pride for the traditional cultures and languages, in order to conserve and revitalize this valuable biocultural heritage.

Supplementary material

Supplemental data for this article can be accessed here: https://doi.org/10.17129/botsci.2792

Supplementary material