This paper presents a study of patterns in the distribution and transmission of medicinal plant knowledge in rural Andean communities in Peru and Bolivia. Interviews and freelisting exercises were conducted with 18 households at each study site. The amount of medicinal plant knowledge of households was compared in relation to their socioeconomic characteristics. Cluster analysis was applied to identify households that possessed similar knowledge. The different modes of knowledge transmission were also assessed. Our study shows that while the
The social processes of acquisition and transmission of knowledge, which are unique to each culture, shape local (environmental) knowledge (hereafter LK) defined as “a cumulative body of knowledge, practices, and beliefs, evolving by adaptive processes and handed down through generations by cultural transmission” [
Intracultural variation is another key contemporary field of inquiry related to LK and ethnobotanical research [
In the Andes, LK has been described as being highly diverse and place specific, and transmitted from one generation to the next through cultural transmission [
As part of a larger research project on the processes of transformation in Andean medicinal knowledge, this paper deals with the patterns of distribution and transmission of medicinal plant knowledge in rural communities of the Peruvian and Bolivian Andes. Research questions to be answered include (1) what are the socioeconomic factors and personal variables that account for intracultural variation of medicinal plant knowledge in the Andean highlands? and (2) what are the dominant modes of knowledge transmission? Our first working hypothesis was that schooling, migratory activity, and market integration—all important factors of social change in the Andes-influence the distribution of medicinal plant knowledge among rural Andean communities. Our second hypothesis was that medicinal plant knowledge is mainly transmitted vertically from parents or grandparents to offspring—a transmission mode typical of conservative knowledge systems with high intracultural variation [
Research for the present paper was part of BioAndes, a regional program of the Swiss Agency for Development and Cooperation that aimed to conserve biocultural diversity in Andean areas of Peru, Bolivia, and Ecuador. BioAndes was executed by a consortium of local institutions and their network of local partners from November, 2005 to March, 2011 and included conservation projects and action-research activities (for a synthesis of BioAndes experiences see [
For the purpose of this study, two of the seven implementation areas of BioAndes were selected as case study sites owing to their similarity in terms of ecological and cultural settings: the district of Pitumarca in the Department of Cusco in Peru and the subcentral of Waca Playa in the Department of Cochabamba in Bolivia (Figure
Map of the study sites. Research was conducted in two communities from Pitumarca District in the Department of Cusco, Peru, (a) and in two communities from Waca Playa Subcentral in the Province of Tapacari, Department of Cochabamba, Bolivia, (b).
Pitumarca District is located in Canchis Province, 87 km south-west of the city of Cusco, in the Southern Peruvian highlands. Altitude ranges from 3,400 meters above sea level in the valley to 6,372 meters above sea level at the Ausangate summit. The climate ranges from semihumid, temperate cold, to humid and frigid, with mean annual temperatures and precipitation varying according to altitude, from 12°C and 650 mm in Pitumarca (3,600 meters above sea level) to 8°C and 910 mm in Phinaya (situated at 4,500 meters above sea level) [
Waca Playa Subcentral belongs to Tunari National Park in Tapacari Province, 65 km east of the city of Cochabamba, in the Eastern cordillera of the Bolivian Andes. The altitude ranges from 2,760 to 4,100 meters above sea level. The climate is semiarid to semihumid and temperate cold, with a mean annual temperature of 11°C and mean average precipitation of approximately 500 mm that is distributed throughout the rainy season from November to March [
Field work was performed by the first author during numerous visits to the two case study sites between June, 2006 and April, 2010. Most data collection took place between March, 2007 and December, 2008. Logistic support was given by partners of BioAndes who already had a long working presence in these areas, namely, AGRUCO in Waca Playa and CEPROSI, IMAGEN, and IMAPI in Pitumarca. In each of the four communities, initial meetings were facilitated by these organizations in the local language to present the research aims and select the participating households. At these community meetings, it was also jointly decided that research results would be returned to the participants by means of the elaboration of medicinal plant booklets at each research site. Except in Huito, where there was no community consensus about authorizing the research because a few individuals expressed concern about the possible economic motivation of the study, community members from Huasapampa, Tres Cruces, and Lambramani all gave their collective verbal consent for carrying out the proposed research activities. In Huito, individual verbal consent was granted by the participants, who either voluntarily approached the first author to contribute to the research, or were approached by her and selected through snowball sampling.
The criteria for selection of participants were age and residence: each community assembly was asked to make a selection representing young (recently married and/or with small children), middle-aged (with grown-up children that participate in family tasks), and elderly households (couples or widows whose children had already left home). In Huito the same criteria were applied to selection of volunteers. Furthermore, in Waca Playa where the population is scattered over the territory, an additional criterion was to include an equal number of participants who lived in the communities’ upper, middle, and lower zones.
Semistructured interviews and freelisting exercises were conducted with 18 households from Pitumarca (9 households each from the Tres Cruces and Lambramani community, or 18% and 23% of all households, resp.) and 18 households from Waca Playa (10 households from Huasapampa and 8 households from the Huito community, or 16 and 13% of all households, resp.). The data were collected in three to six visits to each household, depending on the participants’ availability, meaning that it was often impossible to complete the interview during one visit and that it continued during subsequent visits.
Interviews were conducted in Quechua, Spanish, or in both languages, according to the participants’ preference and language ability, with the help of a native Quechua-speaking translator. The husband and/or the wife—and in some cases the children also—were asked questions about the household’s characteristics, history, and livelihood strategies. In addition, the adult most knowledgeable about medicinal plants (husband or wife), according to the household members’ own perceptions, was asked to list all the medicinal plants (s)he knew and/or used and explain how (s)he acquired his/her knowledge about natural remedies and Andean medicine in general. (S)he was also asked to include the medicinal plants that grew outside the research site, such as dry plants from the Amazon that they bought at the local markets or plants from the highlands that they exchanged with members of other communities. Interviews were conducted individually with the most knowledgeable member, but in about 50% of the cases, other members of the household unit (spouse and/or children) were present and allowed to intervene. This happened because interviews were usually conducted during the household’s daily activities, during which other members were present; these activities included cooking and eating, washing clothes, resting during a day work in the fields, or grazing livestock. It was thus difficult to either systematically isolate only the most knowledgeable individual from the other household members, or to systematically ensure that all household members would be present.
Most interviews were recorded and translated into Spanish when they were conducted in Quechua and then transcribed. In the few cases when participants’ authorization was not granted for recording or when it was technically not possible to do this, detailed notes were taken during the interviews in Spanish and eventually transcribed.
Voucher specimens of most medicinal plants were collected and photographed at the two research sites during walks with participants. Information was recorded about the collection number, date, locality, informant, and the plants’ local names. At the end of the day, specimens were pressed and dried according to standard botanical practices. In cases of contradictory information from different participants about a plant’s name, additional specimens were collected in order to double-check the information. Cultivated and broadly distributed medicinal plants (such as
Voucher specimens were identified by specialists at the Herbarium Vargas from the
Data were recorded at the household level, because preliminary field trips and observation showed that the household was the basic unit for health care practices at both research sites. When a household member is ill, the most knowledgeable parent usually diagnoses the ailment and decides on the treatment that will be administered—in some cases alone but frequently in consultation with the other parent. If this treatment includes the use of medicinal plants, which is frequently the case, then the children are often asked to collect these plants. As a consequence, knowledge about medicinal plants is widely shared and discussed at the household level, which also explains why the interviews were sometimes conducted with several household members.
The socioeconomic characteristics of participating households were compiled from interview transcriptions. The following variables were taken into account: community of residence, age, sex, kinship ties, level of education, migratory activity, degree of market integration, and health specialization. For the purpose of analysis, we distinguished the following three age categories according to the age of the main household member interviewed: (1) young households (20 to 34 years old), (2) middle-aged households (35 to 49 years old), and (3) old households (50 years old and above). We recorded the sex of the main household member interviewed. Where they existed, we recorded the first degree kinship ties between the main interviewee in each household (parents-child, grandparents-grandchild, aunt/uncle-niece/nephew, and between siblings, including in-laws). The level of education was measured as the level achieved by the main household member interviewed: (1) none (no formal education), (2) primary level (1 to 6 years of schooling), (3) secondary level (7 to 12 years of schooling), and (4) superior level (13 years of schooling and above). We distinguished three categories according to household migratory activity: (1) households with no migratory activity (who only made short trips to neighboring towns or communities to visit relatives, sell and buy products, or go to a health center), (2) seasonal migrants (who spend or in previous years had spent up to three months per year either in the lowlands as agricultural workers or merchants, or in neighboring urban centers to pursue off-farm activities); and (3) semipermanent migrants (who had lived for over one year outside the research area). We also classified participating households into three categories according to their market integration: (1) low (agricultural production mainly for own consumption, only surplus and, when necessary, one sheep or head of cattle sold at the local weekly market), (2) moderate (regular sale of agricultural products, sheep, own made wool, cheese or textiles on the local market and, in some cases, also at important commercial centers in the area), (3) and high (besides farm activities, these households traded cattle at important commercial centers in the area, made regular trading trips to the Amazon lowlands, and/or had a small store in the community). When households identified themselves and were identified by other community members as specialists (healers practicing Andean medicine or health workers collaborating with the health center) this information was also recorded.
On the basis of freelisting exercises, the number of medicinal plants mentioned by each household was recorded and taken as an indicator of plant knowledge. When several household members were interviewed, the plants that were cited by the various participants were summed. The number of medicinal plants mentioned thus corresponds to the household’s total knowledge (at both study sites, the most knowledgeable household member contributed approximately 85% of this information as opposed to other household members). Furthermore, at each study site, an inventory of medicinal plants locally known and/or used by the 18 participating households was compiled. To identify the best-known medicinal plants, the number of households that mentioned each plant was tallied, and plants were ranked accordingly.
Comparison of the average or median medicinal plant knowledge among households was carried out in two ways. First, the amount of plant knowledge, calculated as the average or median number of plants mentioned per household, was compared at each research site according to the household’s socioeconomic characteristics. The correlation between age and the total number of medicinal plants mentioned per household was statistically tested by means of the Spearman Rank Order Correlation in Pitumarca (because normality of data failed) and the Pearson Product Moment Correlation in Waca Playa (because the data were normally distributed). The categories of community of residence, sex of the main interviewee, level of education of main interviewee, migratory activity, and market integration were compared statistically by means of a
Second, the type of medicinal plant knowledge was compared using cluster analysis (NTSYpc21 version 2.10 L) as used by [
The household interview transcriptions were analyzed qualitatively to assess the processes of knowledge transmission and acquisition. Answers about the source of acquisition of medicinal knowledge were summed according to the following categories, corresponding to the modes of cultural transmission described by Hewlett and Cavalli-Sforza [
The socioeconomic characteristics and medicinal plant knowledge of the participating households are summarized in Tables
Socioeconomic characteristics and medicinal plant knowledge of participating households from Pitumarca, Peru. Medicinal plants: total number of medicinal plants mentioned.
Residence | Age (years) | Age category | Sex | First degree kinship ties | Education level | Migratory activity | Market integration | Health specialization | Medicinal plants | |
---|---|---|---|---|---|---|---|---|---|---|
Household 1 | Huasapampa | 26 | young | male | grandson of 8 and brother-in-law of 2 | primary | seasonal | low | none | 37 |
Household 2 | Huasapampa | 25 | young | female | sister-in-law of 1 | primary | seasonal | low | none | 36 |
Household 3 | Huasapampa | 21 | young | female | daughter of 5 | superior | none | low | none | 57 |
Household 4 | Huasapampa | 49 | middle | male | primary | seasonal | high | none | 22 | |
Household 5 | Huasapampa | 52 | middle | female | mother of 3 | primary | seasonal | moderate | none | 99 |
Household 6 | Huasapampa | 45 | middle | female | primary | seasonal | high | none | 50 | |
Household 7 | Huasapampa | 44 | middle | male | primary | seasonal | low | health worker | 12 | |
Household 8 | Huasapampa | 63 | old | male | grandfather of 1 and brother of 9 | primary | seasonal | low | Andean healer | 82 |
Household 9 | Huasapampa | 76 | old | male | brother of 8 | primary | seasonal | low | Andean healer | 44 |
Household 10 | Huasapampa | 62 | old | male | primary | semipermanent | moderate | health worker | 52 | |
Household 11 | Huito | 32 | young | male | secondary | semipermanent | low | none | 28 | |
Household 12 | Huito | 24 | young | female | secondary | none | low | none | 47 | |
Household 13 | Huito | 38 | middle | male | secondary | semipermanent | high | none | 53 | |
Household 14 | Huito | 43 | middle | female | primary | seasonal | low | none | 50 | |
Household 15 | Huito | 42 | young | male | primary | semipermanent | moderate | health worker | 53 | |
Household 16 | Huito | 47 | middle | male | secondary | semipermanent | moderate | none | 46 | |
Household 17 | Huito | 68 | old | male | none | seasonal | low | none | 29 | |
Household 18 | Huito | 65 | old | female | none | none | low | none | 47 |
Socioeconomic characteristics and medicinal plant knowledge of participating households from Waca Playa, Bolivia. Medicinal plants: total number of medicinal plants mentioned.
Residence | Age (years) | Age category | Sex | First degree kinship ties | Education level | Migratory activity | Market integration | Health specialization | Medicinal plants | |
---|---|---|---|---|---|---|---|---|---|---|
Household 1 | Tres Cruces | 27 | young | female | daughter of 9 | primary | seasonal | low | none | 27 |
Household 2 | Tres Cruces | 38 | middle | female | primary | none | moderate | none | 21 | |
Household 3 | Tres Cruces | 51 | old | male | primary | seasonal | low | none | 23 | |
Household 4 | Tres Cruces | 29 | young | male | primary | seasonal | moderate | none | 39 | |
Household 5 | Tres Cruces | 39 | middle | male | brother-in-law of 7, nephew of 6 | primary | seasonal | moderate | none | 17 |
Household 6 | Tres Cruces | 59 | old | female | sister-in-law of 8, aunt of 5 | primary | none | low | none | 25 |
Household 7 | Tres Cruces | 38 | middle | male | brother-in-law of 5 | secondary | seasonal | low | none | 15 |
Household 8 | Tres Cruces | 52 | old | male | brother-in-law of 6 | primary | semipermanent | moderate | none | 16 |
Household 9 | Tres Cruces | 71 | old | male | father of 1 | primary | none | low | Andean healer | 42 |
Household 10 | Lambramani | 43 | middle | male | primary | semipermanent | low | none | 27 | |
Household 11 | Lambramani | 38 | middle | female | none | semipermanent | moderate | none | 32 | |
Household 12 | Lambramani | 57 | old | female | none | seasonal | high | none | 24 | |
Household 13 | Lambramani | 50 | old | female | none | semipermanent | moderate | none | 26 | |
Household 14 | Lambramani | 50 | old | female | niece of 15 | none | seasonal | low | Andean healer | 50 |
Household 15 | Lambramani | 63 | old | female | aunt of 14 | none | none | low | none | 22 |
Household 16 | Lambramani | 38 | middle | male | primary | none | high | none | 23 | |
Household 17 | Lambramani | 44 | middle | male | brother-in-law of 18 | primary | semipermanent | low | none | 33 |
Household 18 | Lambramani | 54 | old | female | sister-in-law of 17 | none | semipermanent | moderate | none | 47 |
No significant differences were found between the average or median number of medicinal plants mentioned by households according to the community of residence in either Pitumarca (Huasampa versus Huito communities; Mann-Whitney Rank Sum Test,
In Pitumarca, participants mentioned a total of 249 medicinal plants that they knew and/or used in the free listing exercises. The 47 best-known plants (those that were mentioned by at least 6 of the 18 participating households) were used to compare households’ medicinal plant knowledge with cluster analysis and are listed in Table Young households from the Huasapampa community, with seasonal or no migratory activity and low market integration; the cluster includes two siblings (in-laws). Old households from the Huasapampa community, with seasonal migratory activity and low market integration (apart from household 6, which is middle-aged and showed a high level of market integration); the two traditional healers, who are also siblings, are part of this cluster. Households from all age categories from the Huito community, with seasonal or no migratory activity and low market integration. Two middle-aged households, both from the Huito community, whose main interviewee was male, with semipermanent migratory activity and moderate market integration.
Medicinal plants from Pitumarca mentioned by 6 or more households (
Local name(s) | Scientific name | Plant family | Households | Voucher numbers [or source] |
---|---|---|---|---|
ajinco | Asteraceae | 7 | SM73(P) | |
alqo kiska | Asteraceae | 10 | SM10(P) | |
cáncer qhora | Lamiaceae | 8 | SM123(P)-SM48(P) | |
cebada | Poaceae | 6 | SM122(P) | |
chiqchi | Berberidaceae | 6 | SM35(P) | |
chirichiri | Asteraceae | 15 | SM91(P) | |
clavel, clavel rojo, clavel negro | Caryophyllaceae | 7 | SM83(P) | |
coca | Erythroxylaceae | 15 | [b] | |
eucalipto | Myrtaceae | 16 | SM78(P) | |
hierbabuena | Lamiaceae | 9 | SM5(P) | |
hinojo | Apiaceae | 6 | SM51(P) | |
kanlli | Rosaceae | 8 | SM89(P) | |
kharo | Rhamnaceae | 12 | SM21(P) | |
limón | Rutaceae | 6 | [b] | |
llantén | Plantaginaceae | 6 | SM96(P) | |
llauli, china llauli | Asteraceae | 14 | SM26(P) | |
malva | Malvaceae | 11 | SM18(P) | |
manka phaki | Asteraceae | 10 | SM27(P)-SM109 (P) | |
manzanilla | Asteraceae | 15 | SM82(P) | |
matapalo | Loranthaceae | 6 | [c] | |
muña | Lamiaceae | 13 | SM33(P)-SM120(P) | |
muñak’a | Polygonaceae | 8 | SM31(P) | |
mutuy | Caesalpinaceae | 13 | SM20(P) | |
nabo, yuyo | Brassicaceae | 11 | SM4(P) | |
oq’e thurpa | Malvaceae | 6 | [c] | |
oqororo, alqo oqororo | Scrophulariaceae | 10 | SM19(P), SM38(P) | |
orqo llauli | Asteraceae | 7 | SM77(P) | |
patakiska | Cactaceae | 12 | [a] | |
perejil | Apiaceae | 9 | SM9(P) | |
pilipili, diente de leõn | Asteraceae | 11 | SM6(P) | |
pimpinilla | Rosaceae | 7 | SM79(P) | |
puka thurpa | Malvaceae | 8 | [c] | |
puka t’ikaq kisa | Loasaceae | 10 | SM25(P) | |
p’uku p’uku | Convolvulaceae | 7 | SM39(P) | |
pupusa | Asteraceae | 9 | SM93(P) | |
salvia | Lamiaceae | 8 | SM88(P), SM106(P) | |
sangre de grado | Euphorbiaceae | 7 | [b] | |
sasawi | Asteraceae | 16 | SM94(P) | |
thurpa | Malvaceae | 9 | [c] | |
toronjil | Lamiaceae | 8 | [b,c] | |
uña de gato | Undet. | 6 | ||
wamanlipa | Asteraceae | 16 | SM95(P) | |
waraqo | Cactaceae | 10 | [a] | |
wichullu | Undet. | 14 | ||
yana kisa | Urticaceae | 6 | SM50(P) | |
yawarch’unka | Onagraceae | 17 | SM63(P) | |
zaptilla, pucucho pucucho | Calceolariaceae | 9 | SM1(P)-SM111(P)-SM110(P) |
Tree resulting from cluster analysis of 18 households’ knowledge about 47 medicinal plants in Pitumarca, Peru. HH no.: household number. Correlation coefficient
A fifth cluster of two households was identified even though the association between them is weak. One middle-aged and one old household, whose main interviewee was male, both with semipermanent migratory activity and moderate to high market integration.
Households were not grouped according to the sex of the main person interviewed (with the exception of two households from cluster 4 and the two households from cluster 5), nor were households that had a parents-offspring or grandparents-grandchild kinship relationship. Moreover, the household’s levels of education showed great heterogeneity in the described clusters.
In summary, the analysis revealed the following patterns in the data: the factors that influence the distribution of medicinal plant knowledge in Pitumarca are community of residence, age category, intragenerational kinship ties, migratory activity, market integration, and health specialization in the case of Andean healers. The factors that did not have an influence on medicinal plant knowledge distribution were sex, intergenerational kinship ties, level of education, and health specialization in the case of health workers.
The participating households from Waca Playa cited a total of 150 medicinal plants. The 25 best known plants of this list (those mentioned by at least 6 of the 18 households) were used to run the cluster analysis (see Table Households from the Tres Cruces community. Young and middle-aged households, with seasonal or no migratory activity and low to moderate market integration. Two old households with seasonal or no migratory activity and low market integration. Households from the Lambramani community (apart from households 8 and 9) One middle-aged and one old household, whose main interviewee was male, both with semipermanent migratory activity and low to moderate market integration. One middle-aged and one old household, with seasonal or no migratory activity, both with high market integration. Two old households with seasonal or no migratory activity and low market integration; both are traditional healers. Two old households, whose main interviewee was female, both with semipermanent migratory activity and moderate market integration. Two middle-aged households with semipermanent migratory activity and low to moderate market integration.
Medicinal plants from Waca Playa, Bolivia, mentioned by 6 or more households (
Local name(s) | Scientific name | Plant family | Households | Voucher numbers [or source] |
---|---|---|---|---|
andres huaylla | Solanaceae | 17 | SM9(B), SM132(B), SM88(B) | |
chini muña, muña | Lamiaceae | 11 | SM15(B), SM93(B) | |
coca | Erythroxylaceae | 16 | [a] | |
durazno | Rosaceae | 7 | SM55(B) | |
kalisto, eucalipto | Myrtaceae | 16 | SM6(B), SM14(B), SM87(B) | |
khara malva, malva | Malvaceae | 8 | SM32(B), SM33(B), SM73(B), SM83(B) | |
khara sapi, kharasa, leche leche | Asteraceae | 8 | SM58(B), SM66(B), SM134(B) | |
k’oa muña, haya muña, muña | Lamiaceae | 11 | SM10(B), SM76(B), SM57(B) | |
lanti lanti | Plantaginaceae | 6 | SM61(B), SM103(B)-SM125(B), SM129(B) | |
llave | Asteraceae | 9 | [b] | |
manzanilla | Asteraceae | 10 | SM62(B) | |
molle | Anacardiaceae | 15 | SM16(B), SM89(B) | |
paiqo | Amaranthaceae | 7 | SM29(B), SM60(B), SM67(B), SM91(B) | |
raqacho, raqa raqa | Lamiaceae | 9 | SM18(B) | |
romansa, lanti lanti | Polygonaceae | 10 | SM22(B), SM70(B) | |
salvia | Undet. | 8 | — | |
sauco | Adoxaceae | 7 | SM75(B) | |
sira ch’ilka | Undet. | 6 | — | |
sira paiqo, ch’ini paiqo | Amaranthaceae | 14 | SM128(B) | |
t’ola | Asteraceae | 7 | SM13(B), SM100(B) | |
uri uri | Asteraceae | 7 | RB47.14 | |
verbena | Verbenaceae | 8 | SM72(B) | |
wacanwayo | Amaranthaceae | 8 | SM113(B) | |
wira wira | Asteraceae | 8 | SM94(B), SM127(B) | |
zapatilla | Calceolariaceae | 6 | SM21(B), SM78(B) |
Tree resulting from cluster analysis of 18 households’ knowledge about 25 medicinal plants in Waca Playa, Bolivia. HH no.: household number. Correlation coefficient
According to this interpretation of the cluster analysis results, the factors that influence the distribution of medicinal plant knowledge in Waca Playa are the community of residence, age category, migratory activity, market integration, and health specialization. Sex, kinship ties (intra- and intergenerational) and education level do not influence medicinal plant knowledge distribution. Interestingly, these results confirm the patterns identified in Pitumarca.
Figure
Sources of medicinal plant knowledge for 18 households from Pitumarca and 18 households from Waca Playa. Different sources were sometimes mentioned by one household, which explains why the sum is greater than 18 at each study site.
Six households from Pitumarca and five from Waca Playa learned about medicinal plants from a traditional healer or specialist (one-to-many mode of cultural transmission). But in the case of four of the households from Pitumarca, this specialist happened to be the parent or grandparent of the participant and was thus pooled out of the analysis, because it also accounts for a vertical mode of cultural transmission. Elders from the community were also reported as a source of knowledge (many-to-one transmission).
In addition to these interpersonal modes of cultural transmission, other types of knowledge sources were cited by the participants. The specialized knowledge of Andean healers, for instance, is usually believed to have a supernatural origin. Three (two from Pitumarca and one from Waca Playa) of the four specialists interviewed mentioned that their knowledge about traditional medicine could be explained by a
Regarding the time of acquisition of participants’ medicinal plant knowledge, ten households from Pitumarca (56%) and fourteen from Waca Playa (78%) mentioned that they had learned as children, while twelve (67%) and eleven (61%) households from Pitumarca and Waca Playa, respectively, said that they acquired their skills as adults. Both times were sometimes mentioned by one household, which explains why the sum is greater than 18 at each study site. Here, also there seem to be equal portions of horizontally versus vertically acquired plant knowledge Knowledge about plant remedies is typically first acquired during childhood within the family (24 households in total), but an important part of this knowledge is also gained in adulthood, usually from people other than parents or grandparents (23 households).
As the quotations below demonstrate, there is a strong perception, widely shared among the participants, that medicinal plant knowledge is not “taught” as such, but it is the result of one’s own personal active quest. In order to acquire medicinal knowledge, one needs to ask people in one’s immediate environment, listen, look, try, and practice. And the capacity to learn will depend on interest, curiosity, intelligence, and in the end on personal ability.
Medicinal plant knowledge is freely shared among community members out of solidarity and in a relationship of reciprocity, either through informal exchanges and comments during daily activities or when someone is ill and needs help from his/her neighbors or kin (see quotations below). In fact, the acquisition of medicinal plant knowledge is often the result of concrete illnesses that occur within the family, “as we walk through life”, and that oblige one to look for a specific remedy within the communal body of therapeutic knowledge. According to the households interviewed, this also explains why women are usually more knowledgeable about plant remedies than men, since they need to care for the children when they are ill.
“
The following narrative from a woman who grew up in town before coming back to the rural area where her parents lived illustrates that knowing how to use medicinal plants is considered to be among the skills needed to live in the community, skills which are not shared by people who live in urban centers.
“
Our findings show that there is an important intracultural variation of medicinal plant knowledge in terms of the number of plants known at both study sites, but no clear patterns could be identified to explain this variation. Factors generally reported in the literature to account for differences in ethnobotanical knowledge and LK in general such as age [
Contrary to our findings on the
Another important finding from our study is that age peers, including siblings in the case of Pitumarca, have similar knowledge about medicinal plants, whereas this is not the case for kin from different generations, namely parents-children and grandparents-grandchildren. This result contradicts the hypothesis of a mainly vertical knowledge transmission. We suggest that life experience may be an important determinant of medicinal plant knowledge. People from the same generation are exposed to the same processes of change in the socioecological context throughout their lives (for instance, variable levels of migratory activity of the population, changes in the composition of vegetation or the status of natural resources, the presence of NGOs, etc.), and thus also to the same new sources of medicinal plant knowledge. (for instance, actors and contexts encountered during the migratory experience, new plants available locally, a workshop on medicinal plants conducted in the community, etc.). We believe that the prevalence of migration processes that affect all households at the study sites, but at different levels throughout the last decades, may account for the homogenization of plant knowledge among age peers. Another explanation is that age peers may face similar health problems and thus have the same need to know about natural remedies for treating them. Based on a survey conducted in Oaxaca, Mexico, Giovannini et al. [
The role of life experiences in individual medicinal plant knowledge can also explain our findings regarding the influence of migration and market integration. In Pitumarca and Waca Playa, households were often clustered according to their degree of migration and market activities. Both processes imply different degrees of mobility, since higher market integration meant traveling to the commercial centers of the region or to the Amazon lowlands. Consequently, both migration and market activities imply different degrees of interaction with noncommunity members, and thus access to new sources of medicinal plant knowledge. We postulate that the mobility of Andean households places them in situations of “encounters at multiple [social] interfaces” that stimulate the emergence of new knowledge as a product of the interaction between different actors [
The fact that Andean healers at the study site share similar medicinal plant knowledge with their professional peers could be expected, since the distinction between specialist and lay knowledge is widely recognized in ethnomedicinal research around the world. In their paper on healers’ knowledge in Bolivia, Vandebroek et al. [
Level of education had no influence on the distribution of the amount or type of medicinal plant knowledge among participating households, which contradicts previous findings about the negative correlation between schooling and LK [
Our findings about the sources of acquisition of medicinal plant knowledge indicate that there are equal proportions of vertical versus horizontal modes of transmission at both study sites, which is a rejection of our working hypothesis and contradicts the results from other studies that state that LK is mainly transmitted by parents to offspring [
Local healers represent another source of medicinal plant knowledge at our study sites (reported by almost one third of all participating households). Furthermore, the results from the cluster analysis indicate that Andean healers have a certain amount of medicinal plant knowledge in common with lay people from their communities. This observation suggests that healers are important actors in the cultural transmission of generalist knowledge, which represents a one-to-many mode of cultural transmission and thus of a highly dynamic knowledge system [
The methodological tool chosen for this research was freelisting. The advantage of freelisting and other types of open-ended questions frequently used in the social sciences is that responses from participants are not directed in any way by the researcher, thus allowing in our case for an unbiased inventory of the most significant plants known to and/or used by each participant. Freelists are useful for assessing who in a community knows more (or less) about medicinal plants [
A second methodological constraint of the present study was the limited sample size. A considerable amount of time was required to obtain qualitative data that provided in-depth overall knowledge about the research context and made it possible to establish relationships of trust with participants and conduct long open-ended interviews and informal exchanges. As a consequence, we collaborated with a total of 36 households at the two study sites over a total of 22 months. The same number of households was interviewed at each study site (18), in order to allow for comparison of the medicinal plant inventories from both sites. A larger sample size would be recommendable for a stronger statistical comparison of plant knowledge according to a set of differing socioeconomic variables.
Another aspect worth noting is the poor fit of the cluster analysis results from the Waca Playa case study. The correlation coefficient from the cluster analysis tree from Waca Playa was low (
A final constraint of this study is that voucher specimens of some plants could not be collected because of the study’s time limitations, either because they grew far from the research site and were not easily available during the field work season, or because the specimens collected were deteriorated due to adverse climatic conditions. In these cases, the corresponding plants could only be determined indirectly through literature, which does not ensure the accuracy of this information and thus limits its use for future comparative studies with the same plants.
This study shows that LK in culturally homogeneous groups presents certain patterns of variation and distribution that result from differences at the socioeconomic and personal levels, including individual life experiences. Our findings demonstrate the usefulness of triangulating the results from several methods to assess intracultural variation, namely, by comparison of both the
The authors are very grateful to the participants from the communities of Tres Cruces and Lambramani in Bolivia and Huasapampa and Huito in Peru for sharing their time and knowledge. AGRUCO in Cochabamba, CEPROSI in Cusco, and ETC Andes in Lima provided invaluable logistic support during field work. Their warm thanks go to their respective directors, Freddy Delgado, Elena Pardo, and Teobaldo Pinzás. The following persons assisted with interview interpretation and/or translation: Martin Huamán, Sonia Medrano, Sandra Acuña, Roger Juárez, Rolando Sánchez, and Deicy Mejía. Fructuosa De La Torre, director of the Herbarium Vargas in Cusco, and Magaly Mercado, director of the Herbarium Martin Cardenas in Cochabamba, are acknowledged for their support with plant identification. Cecilia Gianella elaborated the research sites map, and Ted Wachs did language editing. They also wish to thank two anonymous reviewers for interesting comments that helped improve this paper. This research was conducted within the framework of the Swiss National Centre of Competence in Research (NCCR) North-South, cofunded by the Swiss National Science Foundation (SNSF), the Swiss Agency for Development and Cooperation (SDC), and participating institutions. Support from the SDC program “Jeunes Chercheurs,” managed by the Swiss Commission for Research Partnerships with Developing Countries (KFPE), is also gratefully acknowledged. S. L. Mathez-Stiefel wishes to thank Stephan Rist and Urs Wiesmann at the Centre for Development and Environment (CDE) of the University of Berne for supervising the research.