5:15 PM - 6:45 PM
[HGG02-P07] A co-occurrence network analysis of environmental discourse: the case of Maasai residents in Laikipia County, Kenya
Keywords:Environment, Discourse, Erosion, Cactus invasion, Sand harvest, Kenya
1. Objective
This research presents findings from the field activities on the environmental discourse of more or less sedentary pastoral Maasai people in Il Polei area, Laikipia County in Kenya. The local knowledge idea is frequently missing in natural scientific studies on human-environmental interactions. One useful approach to the local knowledge is the text mining approach to people’s views on their environment, and it is still not very common in the geographical studies and earth scientific research in semi-arid areas of East Africa to focus on local/indigenous knowledge and environmental discourse using this approach. This study is an attempt to fill the academic gap, employing co-occurrence network analysis.
2. Informants and fieldwork
The fieldwork was conducted between November 27 and December 1, 2023. After selecting seven local-born informants in Il Polei and administering a small questionnaire on their basic individual/household characteristics, that may have differential effects on their environmental perception and recognition, their environmental utterance/discourses were recorded. At the beginning of the interview, photos of gully erosion, invasive cactus, and sand harvest in the research area were shown to the informants to set the context in which they expressed their environmental view: at the initial stage, we avoided in the conversation the words erosion, invasive cactus, or sand harvest, so as not to predetermine their utterances.
Then we asked the informant to state what kind of events were taken by the photos, take a longitudinal perspective and explain/describe what has caused the events, and explain/describe any linkages of the events to some item(s) present in the photo and to some item(s) having been present in the past but absent when the photos were taken. We also asked the informant to explain/describe any linkages of the events to the informant’s economic struggle (livelihood strategies) and explain/describe any linkages of the events to traditional and/or Christian divinity or belief system. We then stimulated discussion on what was the most important event among the three that needed immediate change and asked the informant to link all thinking derived from the three sets of photos and explain/describe pastoral life in the research area.
Two FGDs (focus group discussions) were also organised to acquire supplementary information to be used for interpreting the text mining results. Their environmental utterance/discourses in Maa language were recorded, transcribed, and translated to both Swahili and English for the application of co-occurrence network analysis.
3. Visualisation and interpretation of word co-occurrence in environmental discourse
Alongside with word cloud diagrams (e.g. Fue et al. 2016), co-occurrence network diagrams (e.g. Nikiema et al 2023) of the Maa words (and their Swahili and English translation) employed in the environmental discourse of the informants are a means for visualisation of the words that come after a specified word, as well as of more general co-occurrence relations. Separate co-occurrence network diagrams are produced, for instance, for different male and female informants to compare and examine any gender difference. The comparison may be done using diagrammatic indicators such as the total number of extracted words, the number of words used, percentages of the number of “nodes”, the numbers of “co-occurrence community”, among others. The presentation shows some of the key findings.
The environmental discourse is examined especially focusing on: how the people understand, describe, and explain the mechanism of the three events of erosion mechanism, mechanism of cactus invasion, and mechanism of sand accumulation and harvest, how and to what extent their understanding, description, and explanation of the three events are conditioned and influenced by the “global environmental change”, and how and to what extent their divine notions (traditional and Christian) condition and influence their understanding, description, and explanation of the three events.
References
Fue, K., Geoffrey, A., Mlozi, M. R. S., Tumbo, S., Haug, R., & Sanga, C. (2016). Analyzing usage of crowdsourcing platform Ushaurikilimo'by pastoral and agro-pastoral communities in Tanzania. International Journal of Instructional Technology and Distance Learning, 13 (12): 3-21.
Nikiema, T.; Ezin, E.C.; Kpenavoun Chogou, S. 2023. Bibliometric analysis of the state of research on agroecology adoption and methods used for its assessment. Sustainability, 15, 15616.
This research presents findings from the field activities on the environmental discourse of more or less sedentary pastoral Maasai people in Il Polei area, Laikipia County in Kenya. The local knowledge idea is frequently missing in natural scientific studies on human-environmental interactions. One useful approach to the local knowledge is the text mining approach to people’s views on their environment, and it is still not very common in the geographical studies and earth scientific research in semi-arid areas of East Africa to focus on local/indigenous knowledge and environmental discourse using this approach. This study is an attempt to fill the academic gap, employing co-occurrence network analysis.
2. Informants and fieldwork
The fieldwork was conducted between November 27 and December 1, 2023. After selecting seven local-born informants in Il Polei and administering a small questionnaire on their basic individual/household characteristics, that may have differential effects on their environmental perception and recognition, their environmental utterance/discourses were recorded. At the beginning of the interview, photos of gully erosion, invasive cactus, and sand harvest in the research area were shown to the informants to set the context in which they expressed their environmental view: at the initial stage, we avoided in the conversation the words erosion, invasive cactus, or sand harvest, so as not to predetermine their utterances.
Then we asked the informant to state what kind of events were taken by the photos, take a longitudinal perspective and explain/describe what has caused the events, and explain/describe any linkages of the events to some item(s) present in the photo and to some item(s) having been present in the past but absent when the photos were taken. We also asked the informant to explain/describe any linkages of the events to the informant’s economic struggle (livelihood strategies) and explain/describe any linkages of the events to traditional and/or Christian divinity or belief system. We then stimulated discussion on what was the most important event among the three that needed immediate change and asked the informant to link all thinking derived from the three sets of photos and explain/describe pastoral life in the research area.
Two FGDs (focus group discussions) were also organised to acquire supplementary information to be used for interpreting the text mining results. Their environmental utterance/discourses in Maa language were recorded, transcribed, and translated to both Swahili and English for the application of co-occurrence network analysis.
3. Visualisation and interpretation of word co-occurrence in environmental discourse
Alongside with word cloud diagrams (e.g. Fue et al. 2016), co-occurrence network diagrams (e.g. Nikiema et al 2023) of the Maa words (and their Swahili and English translation) employed in the environmental discourse of the informants are a means for visualisation of the words that come after a specified word, as well as of more general co-occurrence relations. Separate co-occurrence network diagrams are produced, for instance, for different male and female informants to compare and examine any gender difference. The comparison may be done using diagrammatic indicators such as the total number of extracted words, the number of words used, percentages of the number of “nodes”, the numbers of “co-occurrence community”, among others. The presentation shows some of the key findings.
The environmental discourse is examined especially focusing on: how the people understand, describe, and explain the mechanism of the three events of erosion mechanism, mechanism of cactus invasion, and mechanism of sand accumulation and harvest, how and to what extent their understanding, description, and explanation of the three events are conditioned and influenced by the “global environmental change”, and how and to what extent their divine notions (traditional and Christian) condition and influence their understanding, description, and explanation of the three events.
References
Fue, K., Geoffrey, A., Mlozi, M. R. S., Tumbo, S., Haug, R., & Sanga, C. (2016). Analyzing usage of crowdsourcing platform Ushaurikilimo'by pastoral and agro-pastoral communities in Tanzania. International Journal of Instructional Technology and Distance Learning, 13 (12): 3-21.
Nikiema, T.; Ezin, E.C.; Kpenavoun Chogou, S. 2023. Bibliometric analysis of the state of research on agroecology adoption and methods used for its assessment. Sustainability, 15, 15616.