Reconstituting family transitions of Sahelian western Niger 1950-2000: an agent-based modelling approach in a low data context
Résumés
This research analyses the impacts of the family organization on the diversity of income sources and the sustainability and the dynamics of rainfed farming systems of Sahelian Niger, through an individual-centred agent-based model which variables were defined through anthropological investigation. Results show that family organisation has strong effects on wealth levels and distribution and on demographic growth. They also suggest a historic shift from the patriarchal mode to a mono-nuclear mode in the 70's in this specific area, due to a higher resilience of the latter thanks to a broader diversification and a better adequacy between wealth and family demography.
Cet article présente la méthode et les résultats d’une analyse des impacts des modes d'organisation de la famille sahélienne sur la diversité des sources de revenus ainsi que la durabilité et la dynamique des systèmes de cultures pluviales sahéliens au Niger, et ce, au travers d’un modèle multi-agents individu-centré dont les variables ont été définies par une enquête socio-anthropologique approfondie. Les résultats montrent que l'organisation familiale a des effets importants sur les niveaux et la distribution de richesse individuelle et familiale ainsi que sur la croissance démographique. Ces résultats suggèrent également un tournant historique de basculement de la société sahélienne nigérienne dans les années 70, d’un mode familial patriarcal à un mode familial mononucléaire, en raison d'une plus grande résilience de ce dernier mode, du fait d’une plus large diversification et une meilleure adéquation entre richesse et démographie familiale.
Entrées d’index
Mots-clés :
Niger, Sahel, structures familiales, modélisation multi-agent, société rurale, dynamiques socio-environnementalesKeywords :
Niger, Sahel, family organisations, Agent-based modelling, rural society, socio-environmental dynamicsPlan
Texte intégral
Introduction
1Among all of the gaps that stand between social science results and development operators’ guidelines regarding rural areas in West Africa, that of the perceptions of the family structure has a crucial influence on the future evolutions of these societies and on the connection between development operators and the rural areas in which they work. This gap has both a historical and a sociological origin.
- 1 Nine villages out of ten in Sahelian Niger are less than 150 years old. Most were settled between 1 (...)
- 2 Olivier de Sardan (2003) recalls testimonies for such an evolution as far back as 1914!
2As discussed by Rain (1999) and Olivier de Sardan (2003), the enlarged and patriarchal family archetype was ultra-dominant in the beginning of the 20th century; however, this archetype is not inherent to the African tradition. Rather, it emerged from the need for strong clans and links due to the generalised insecurity and several epizooties and famines that affected Western Niger in the 19th century until the Pax Francia (Ki-Zerbo 1972; Gado 1993; Olivier de Sardan 2003). The colonisation era, with the opening of free land (thanks to the new security), the “indigenous code” and the military service, was a period of strong incentive for new settlements1 as well as for the expansion of the existing multi-rationality, multi-activity and multi-finality within families2 (Raynaut et al. 1997). In the post-colonisation era, with the demographic explosion and improvement in transportation, new economic activities such as seasonal migration arose, seasonal gardening was eased and farming colonisation began in the territory. The evolution of economic activities does not occur between different types of variously specialised Sahelian families because they are not adequately specialised (Coquery-Vidrovitch and Moniot 1974). We hypothesize that such an evolution may occur within families, according to the balance of power and orientations between the different members of each family. Following Marks and Rathbone (1983) and Hansen and Strobel (1985), we consider that the external factors that have affected and continue to affect rural populations in this area may have several and important effects. To analyse family-related socio-economic changes, a first deductive methodology is analysing external factors and their impacts on family structures (Reinwald 1997) by comparing data on these external factors and family structures. However, in a low-data context, one may encounter difficulties accessing sufficient data on external factors for this a purpose. Moreover, one can only gain access to present-time family structures, leading to difficulties comparing family structures and external factors from a diachronic perspective. There is thereby a need for a tool that can consider the impact of family structures on external factors, such as economic, spatial and demographic elements, for which historical data are more available in order to test the existence of various family structures along history.
- 3 For instance, the social obligation to contact the male family head as the representative of one “h (...)
- 4 See Quin’ones et al. (1997), Buresh et al. (1997), Ahmed et al. (2000) or Breman et al. (2001) as i (...)
- 5 For instance, in order to observe inheritance procedure evolutions, generations are required. The i (...)
3These evolutions were perceived with a large delay (and typically through various ideological prisms) by rural development operators, whose backgrounds are mainly in the biophysical sciences. These operators have little genuine connection with social science literature and are not encouraged to because of their background, as they build “interpretative models oriented by their discipline without acknowledging the subjective aspect of this interpretation” (Olivier de Sardan et al. 1995; Bonnal et al. 1997; Lavigne-Delville 1999). Therefore, the majority of development projects present families as a stable structure, based on the enlarged family archetype. Multi-activity, multi-finality and multi-decision are considered by these operators; however, in terms of practical usage, development programs rarely include such points3. The fact is that “the main weakness of anthropology [and other social sciences such as history] is that it produces knowledge and not decision for action. How can one organise action based on the often very subtle and multiform framework of the [social scientist]?” (Bonnal et al. 1997). Many present-time biophysical scientists working in West African rural areas continue to justify research proposals on assumptions that deny the three points of multi-activity, multi-finality and multi-decision4. However, these factors are a structural characteristic of these societies (Milleville 1989; Paul et al. 1994). How can one demonstrate the effects of social factors beyond a qualitative description in low-data contexts? To do this, a tool that can gather environmental, social and economic aspects together and balance their impacts at the relevant decision unit (i.e., the individual within his/her household) is necessary (Saqalli et al., 2010)5.
4Therefore, the current research has two objectives: (i) By combining all economic activities at the relevant level, we propose to show that family types, ceteris paribus, have a strong impact on Sahelian village populations and territories. More precisely, we investigate the impacts of family organisation on family economic activities and sustainability, explicitly considering multiple sources of income. This is accomplished through a modelling approach that can assess the consequences of the interactions between individual strategies within a family, as affected by family structure. These interactions are neither necessarily coherent nor necessarily antagonistic. (ii) In our study context, family changes may have occurred but no data are available. Therefore, we propose to analyse the family types’ evolution along the post-decolonisation era through a diachronic comparison between simulation outputs and the rare data available as well as through an analysis of the family types’ effect on population resilience.
- 6 Unitary family models consider the household or the family as a homogenous decision-maker for all e (...)
5As previously mentioned, the enlarged family was dominant and development operators continue to consider it the archetype. Meanwhile, field investigations, as assessed by Saqalli et al. (2010), demonstrate the present-time dominance in Sahelian Western Niger of a different family type, i.e., mononuclear families. We position our work in the economics debate between unitary structures6 and collective non-cooperative structures (Donni 2004; Radja 2004). The latter correspond to a multi-decision family organisation in which there is a collective organisation but that is not based on cooperation efficiency, as defined by Pareto (Meignel 1993).
6Family type proportions are known only for the present-time. Therefore, the only relevant methodology is to compare two simulated villages, each populated with one family type, with (i) the external data available and (ii) each other. We compare one-family-type villages rather than villages with a mixture of both family types (as is observed in reality) because this is the best first step in explicitly demonstrating the impact of family types, all other things fixed and defined. A family type mixture analysis is a further step after the demonstration of the impact of family types. Two family structures are then compared, these include:
-
The collective non-cooperative family. This is currently the most common family type in the study region (Tahirou 2002 p.23), with a Non Cooperative Family Structure (NCFS). This structure constitutes 76% of the families of the sample collected by Saqalli et al. (2010).
-
The unitary concept of the enlarged family. This is development agencies’ main concept of African families, hereafter referred to as the Unitary Family Structure (UFS).
7We used the SimSahel model of Saqalli et al. (2010). These authors built an empirical individually-based multi-agent model of a Sahelian village on the commune of Dantiandou, also called the Fakara (region of Tillabery, Western Niger). We chose Agent Based Models (ABMs) because they are particularly useful for simulating the multi-disciplinary approach of a multi-sectorial reality (Berger 2001). This is due to their capacity to integrate and formalise information and data of various forms, origins and disciplines (Rouchier and Requier-Desjardins 1998) and thereby, their power for allowing the emergence of global phenomena due to apparently minor and micro-level dynamics (Bousquet and Le Page 2004). From a historical point of view, because ABMs combine biophysical and social elements at the relevant level and can integrate temporal evolutions of social phenomena, it is useful to integrate them into a methodology in which hypotheses concerning micro human individual behaviour are tested and compared with external macro diachronic data. The model results provide sufficient confidence for this type of comparison. The model can be considered as adequately reliable and as providing enough confidence in terms of consistency between global population-level results and field observations on the same Fakara site from other sources (Loireau-Delabre 1998; La Rovere 2001; Tahirou 2002), adequacy regarding micro-behaviours as observed during field anthropological research assessed in this field site and meaningful reactions to a sensitivity analysis (Saqalli et al. 2010).
Methodology
The overall field and modelling methodological approach
8The entire work of model construction can be seen as an iteration of ‘there and back’s’ between field working and modelling, as discussed by Rouchier and Requier-Desjardins (1998). We based our analysis on a previously built empirical and KIDS agent-based model that describes a village archetype (Moss and Edmonds 2005; Janssen and Ostrom 2006). The selected ABM platform is CORMAS (Common Resources Management Agent-based System), developed by CIRAD (Bousquet et al. 2001).
9To understand the complexity of a farming system, our first assumption is that taking into account all its components and simplifying them is more relevant than neglecting some activities by focusing on the one which seems to be the most important. Our second assumption is that taking into account the strategies of all the members of a family and simplifying them induces a lower gap than simplifying it to a unique “household” strategy. Finally, our third assumption is that, as villages should be considered as open systems, it means that one may found more consistency in local but imprecise values, figures and relations than ones coming from literature describing equivalent but not local situations. Thereby, our modeling approach was supported by a long-term field investigation period (1.5 years) required to define the local context, the main differentiation factors between villagers and a villager-based typology of access to economic activities.
10The field and modelling methodology, including the parameter functions and related sources for the agro-ecological and village socio-economic modules, is fully described in Saqalli et al. (2010). This paper also contains the individual-centred model, with relationships and dependencies between villagers (gender & rank as main factors of hierarchy in the family; lineage & individual and family wealth as the main factors at the village level) as well as their differentiated access to economic activities (agriculture, livestock keeping, seasonal migration, dry season gardening). All of the field (assumptions & steps) and modelling methodology (assumptions, parameters & rules) as well as the validation process (comparison with external data, sensitivity analysis) are thereby described in the Saqalli et al. (2010) article.
11The temporal scale is defined as one week for each time step. The model is spatialized through a dynamic 100*100 pixels map as the main interface for the visualization of interactions, each pixel corresponding to one hectare.
Field researches
12The behaviour rules are based upon the translation of the investigations that were conducted on the Fakara, according to an interpretation process similar to that of Gladwin (1989). The area (approximately 2000 km² and 40 villages and hamlets) is located 70 km northeast of the populated valley of the Niger River and Niamey, the capital city of Niger (800,000 inhab.). The eastern border is the populated fossil valley (56.36 inhab.km-²), called "dallol Bosso". The Fakara is crossed by a network of dry and thin talwegs (less than 5% of the total surface) where dry season gardening can be practiced. Wide sandy plains extend from these talwegs, representing 75% of the area and the major part of the arable land. The remaining 20% is covered by hardened lateritic plateaus, which can be used only for pasture and wood gathering (Loireau-Delabre 1998). The dominant ethnic group is the Zarma, who are mainly farmers who increasingly keep livestock (Olivier de Sardan 2003; Saqalli et al. 2009).
13Village and regional interviewing tools (Saqalli et al. 2009) helped to select the relevant villages in which individual interviews have to be done. Four villages were then chosen based on size, access to dry season cropping and roads. The villagers' sample was defined upon a two-criteria stratification. Gender is the fundamental criterion of discrimination with respect to the access to activities. The second criterion is the level of responsibility, i.e. the number of dependents. We used semi-direct interviews focused on the main activity carried out by the interviewed person at the time of the interview: we have looked after the adequacy between the timing of the interviews and critical dates for each activity. Each person was interviewed once or twice depending on the number of activities he/she manages. 126 persons were interviewed, not including investigations on village history with elders and village chiefs. Individual interviews provided social rules that condition the differential access and power of each individual, which should be parameterized by field location- relevant figures, coming from our own interviews and other "grey" sources", i.e. reports from NGOs and other institutions working on site. Because of data uncertainty, these parameters are defined as relative to compare and hierarchize elements and factors for each village individual, thereby avoiding artificial threshold effects due to gaps between values coming from different data sources. Finally, we used theoretical hypotheses found in the socio-anthropological literature that describe social and environmental dynamics as supports for developing hypotheses and testing evolution scenarios.
A field & modelling process
14The focus of the current work is the impact of family organisations on population developments. Therefore, the main principle we followed was to implement field observations as sequences of behaviour without introducing external postulated rationality.
15For the entire biophysical module (climatology, pedology and phyto-ecology) as well as for all functions that are not related to socio-anthropological logics of production means management (demography, price evolutions of non-local products), rules and parameters are based upon the available published or unpublished literature (Reports & documents from development or research agencies, M. Sc & Ph. D. dissertations). The Fakara research site was selected because of the existence of extensive literature and data sets resulting from 20 years of agricultural research that was completed by several institutions (mainly ICRISAT, ILRI, INRAN, and IRD). The site can be considered as representative of the average situation of the rainfed Sahelian Niger. For the agro-ecological portion of the model, the villager-based criteria were defined through the available literature data (Gérard et al. 2001; Turner and Williams 2002; La Rovere and Hiernaux 2005).
16The model is split into an agro-ecological part and a socio-economic part.
17Climate and land dimensions were simulated through the cellular automata matrix of the model, where each pixel is a parcel of the village territory. Livestock agent behaviour rules and biophysical constraint hierarchy are defined from criteria identified by the villagers and from literature: for instance, the list of the main factors to describe a piece of land (fertility, gardening capacity, reaction to rainfall, weeding, grazing and weed growth) were defined through interviews, but the description of these factors are defined through literature and unpublished research data.
18For the socio-economic part, we developed the entity-centered ABM concept by choosing individuals as the main model agent entity and not households or families, without introducing external postulated rationality, following Janssen and Ostrom (2006). No ultimate goal is defined, as all agents are reactive. Agents are defined as equivalent, which means that they have the same attributes: only the values of these attributes vary, defining each agent and its characteristics.
Building an agro ecological module
19Rainfall can be considered as the unique climate factor to impact on local systems. The challenge was to simulate its high spatial and temporal variability. Seasonal rain variability comes from ICRISAT rainfall data (La Rovere and Hiernaux 2005, Gérard et al. 2007), as well as the spatial correlation of rain events at the village level. Climate simulation is done through a 5*5 pixel "Climate Blocks" matrix. For each time step, each block receives a rainfall probability, and if it rains, a probability for the rainfall quantity. Both probabilities are then affected by a drought factor, randomly defined at the beginning of each year, varying from 80% to 130% of the average rainfall.
20The village territory is spatially described through a layer of pixels (i.e. “parcels”), where each cell has an initial agro-pedological fertility potential P: the plain (P = 1), the valley (P = 2), the village and the plateau (P = 0) are simplifications of the soils’ qualitative description of Glättli (2005). Each pixel fertility value is initially equal to the parcel’s potential and then varies according to the vegetative growth of grass and shrubs in the pixel, defined by local pixel rainfalls, fertility, cropping impact for owned and cultivated parcels and manure.
21The fields’ manuring and grazing impact is defined along a ratio between surface and the number of animals of each species. At each step, a parcel may be grazed / browsed depending on the species (i.e. preferentially shrubs for goats, preferentially grass for the others). Cattle are differentiated between a calf and milking cow group, staying on the village territory while weaning and milking, and a “dry” group, leaving for transhumance during dry seasons. The manure from the remnant herd is then defined for each animal specie for each pixel: half of one animal manure is spread in grazed areas (at day time) and the other half in corrals at night (Ayantunde et al. 2000; Ayantunde et al. 2002): this last part of the manure is then spread solely on the fields of the animal owner, which means that this action leads to a potential differentiation between fields according to the field owner’s herd.
Simulating the villagers production activities
22Agriculture. Rainfed agriculture is the main activity in terms of land use and manpower requirement during the rainy season, from June to October (Tahirou, 2002). Sowing (sometimes several times if it remains possible for seeds to grow), weeding (at least twice according to weed levels) and harvesting are the most labour-requiring production steps. Only male adult agents can crop millet fields. Only household heads can extend the family property according to family needs and manpower.
23Therefore, one parcel produces only if it has either been appropriated or borrowed by a villager agent. Each week, the villager agent observes his parcels’ status and calculates the available manpower he can use (his own and his dependent relatives who are available, according to the family structure). He then classifies priorities and chooses the field so as to minimize the loss of production over all his fields. Crop growth is then governed by the available manpower at the right time, rainfall and the fertility rate. Weeds evolve according to parcel fertility, rainfall and weeding actions (Lassina, 1992; Montagne and Housseini, 1998; Sangaré et al., 2001; Schlect et al., 2006) and affect millet production until millet is higher than weed. Crops yields result thereby from a combination of the crop growth rate, weeds and fertility effect rates.
24Dry season vegetable gardening follows an equivalent sequence of actions. Dry season vegetable gardening can be practiced only in the villages that have access through wells to shallow groundwater in “valley” fields. Only married women/widows with daughter-in-laws or unmarried daughters can garden, by mobilizing the available female manpower from then on in a similar fashion than the family head millet farming. This activity can generate around 52.50 € per woman per year, at the end of the hot dry season, roughly April.
25Migration. Seasonal migration occurs during the dry season, from October to May. Most of the male adults move to the Gulf of Guinea basin to find jobs (Timera, 2001). The money they earn is a major source of income for the entire Sahelian zone (Reardon, 1994). Furthermore, the money they bring home, approximately 290 € per family per year or 30 € per inhabitant per year, is an important contribution to the remaining family members (Mounkaïla, 2003; Saqalli et al., 2011). Racket of custom services often occur on migrants while they cross the border coming back. Migrants use the money for food, religious and/or social ceremony spending purposes, small ruminants, savings and/or the next migration travel expenses. Departure time is typically after crop harvest but many young bachelors leave earlier. Only male adult agents more than 16 years old have access to the migration activity. The price for the roundtrip travel is the major constraint for leaving. The date of departure depends on the family rank, i.e. the number of dependent relatives. To come back, the migrant should overcome two constraints: (i) for prestige reasons, a male adult agent cannot come back in the village without at least the roundtrip ticket for the next migration, i.e. 2* 45 €.