Article Land and Contestations over Autochthony and Local Citizenship in Agrarian Ghana Gertrude Dzifa Torvikey1 Abstract This article examines the reimagination of communities in an industrial cassava frontier of Ghana in the wake of a contested land grab supported by state and community institutions. Qualitative and survey data were used to construct the existing social relations in the communities through the lens of earlier processes of agrarian change that have transformed the social base of the communities. It is argued that the expansion of capitalist production systems into agrarian areas results in local citizenship contestations centered on land, and redefinition and reclassification of people and their access to land. The multiple claims and contestations that arose from the land grab and the political reactions from below are highlighted. It is further argued that differentiated dispossession and class differences determine the strategies used by affected people. While some farmers demonstrated agency by holding on to a “little pie” to enjoy greater community social cohesion, others, drawing from their local citizenship status, although contested, fought the land grab. Keywords Ghana, land grab, local citizenship, cassava, social relations Agrarian South: Journal of Political Economy 11(1) 134–157, 2022 © 2022 Centre for Agrarian Research and Education for South (CARES) Reprints and permissions: in.sagepub.com/journals-permissions-india DOI: 10.1177/22779760211068791 journals.sagepub.com/home/ags 1 Institute of African Studies, University of Ghana, Accra, Ghana Corresponding author: Gertrude Dzifa Torvikey, Feminist Africa, Institute of African Studies, University of Ghana, Box LG 73, Accra, Ghana. E-mail: torvikeydzifa@gmail.com http://crossmark.crossref.org/dialog/?doi=10.1177%2F22779760211068791&domain=pdf&date_stamp=2021-12-26 Torvikey 135 Introduction Land access, ownership, and control are at the center of identity poli- tics and local citizenship struggles. Since the early 2000s, when land grabbing by foreign actors re-emerged and intensified in its character and form, debates have emerged detailing the ramifications for gender, class, and generational inequalities, and local citizenship and social con- flict in recent times. In rural areas in Africa, natural resource control remains at the center of many of these struggles and transformations in social formation. Underlined by the reclassification of social groups and access rules, claims of belonging exacerbate already existing class differentiation (Moyo et al., 2015). While the agricultural land grabs are both systematized and abrupt, exclusion can be gradual and subtle. The multiple claims to land and differentiated access are invoked when land commodification and commercialization intensify. Changes in the benefits people derive from land, scarcity, and competition deepen class struggles and introduce new vulnerabilities (Borras et al., 2010). In Africa, notably, the catalytic effect of the scramble for agricultural lands by foreign actors merged with the longer-term processes of changes in land access, ownership rules, and governance structures (Yaro, 2010). In a plantation economy like Côte d’Ivoire, the tensions and internal strife that have occurred since 1993 have been attributed to land, migration, and citizenship questions. The demography of that economy has changed historically due to coercive colonial policies and attractive postcolonial immigration policies that brought migrants from other parts of Francophone West Africa, particularly Upper Volta (now Burkina Faso), to work in cocoa and coffee fields. In the immediate postcolonial period, attractive landownership programs and the abolition of customary land laws incentivized migrants to plant cocoa in their own right (Chauveau & Richards, 2008). Historically, commodity production areas in peripheral economic enclaves have migrant inflows. Through coercive processes, peasants are compelled into market relations. They move to areas of commodity production, where competition for land is already intense. Over time, migrant inflows into the areas change the social relations and differentiation (Greene, 1996). The existing social structure is again reconstructed, especially in the definition of local citizenship, to exclude some social groups from benefiting from land resources. In peasant economies in West Africa, particularly in Ghana, several studies have drawn attention to the demographic shifts and migrant inflows into commodity frontiers (Addo-Fening, 1997; Hill, 1961, 1963). Cocoa, oil palm, and other export commodities trigger migration into rural areas, 136 Agrarian South: Journal of Political Economy 11(1) both seasonal and permanent. The processes transform agrarian structures of land access and ownership rules of communities. The constitutive parts of agrarian change situated in its historical epoch are essential in clarifying agrarian social structures and how contemporary land grabs produce new forms of change. This study reflects on the dimensions of change in local citizenship and determinants of exclusion in five communities, namely Hodzoga, Aveafe, Alavanyo, Achianse, and Hodzove in the Ho Municipality of the Volta Region of Ghana. The area has become an industrial cassava production frontier since 2005. The expansion of cassava production in rural areas, mainly for industrial use, is an essential feature of agrarian change, which has triggered tensions in social relations in frontier communities in Ghana. The study focuses on how land is centered in local citizenship claims through the prism of the historical processes underlying the founding of communities, the ramifications of the land grab for social relations, and the political reactions of affected people. In the current conjuncture, tensions about local citizenship and belonging in agrarian areas demand full attention, particularly to unpack the determinants of inclusion and exclusion and how they are related to the transformation in local citizenship and local hierarchies. To underscore the relevance of cassava in the land grab debates, the second section provides a brief but detailed historical trajectory of industrial cassava production in Ghana. The third section discusses the research methodology and describes the research communities and the context of the land acquisition in 2005. The theoretical framing of the article is discussed in the fourth section, through the lens of the materiality of agrarian change as a contested field and as having multiple effects. The fifth section provides an empirical analysis of the history of the settlement patterns of the communities and the cumulative processes that underline the agrarian change in the area. The sixth section provides the context of land loss and reactions from different social groups. In the conclusion, finally, it is argued that agrarian change has multiple trajectories, and that the centering of land in the construction of local citizenship provides a lens to examine modern agricultural projects in rural areas. Contextualizing Industrial Cassava Production in Ghana Since the 1970s, the Ghanaian state has been keenly interested in devel- oping cassava for industrial use and export. The Five-Year Development Torvikey 137 Plan (from 1975–1976 to 1979–1980) under the regime of General Igna- tius Kutu Acheampong was framed within food self-sufficiency and import substitution ideologies. It identified cassava as an essential crop for domes- tic consumption and industrial use (Ministry of Economic Planning, 1977). In 2001, the New Patriotic Party (NPP) administration launched the presi- dential special initiative (PSI-Cassava) under a comprehensive industrial policy that included three other commodities—oil palm, salt, and garments and textiles. The four PSIs were to generate a revenue of US$4.4 billion in 4 years of their implementation (Tonah, 2006). Additionally, the program envisaged the creation of 100,000 direct jobs. PSI-Cassava was designed to expand the export and industrial foundations of the country. Within the policy framework, the state established the Ayensu Starch Company Limited (ASCO) in 2004, at Awutu Bawjiase in the Central Region of Ghana, to process cassava starch for export. The funding for the PSIs included the United Nations Development Programme (UNDP), Enterprise Africa, Danish International Development Agency (DANIDA), International Financial Consortium (lFC), Oiko Credit of the Netherlands, as well as domestic banks like the National Investment Bank (NIB) and the Ghana Commercial Bank (GCB) (Tonah, 2006). Although ASCO collapsed, some agribusinesses started investing in the industrial cassava sector henceforth. In 2012, when the state instituted tax waivers for manufacturers who used local raw materials instead of imported ones, cassava became one of the crops to see rapid industrial use. Between 2012 and 2013, Guinness Ghana Breweries PLC, a local subsidiary of Diageo, and Accra Breweries Limited (ABL), a local subsidiary of SABMiller, introduced new beer brands, namely Ruut Premium Beer and Eagle Lager beer, respectively, made mainly from cassava. These companies sourced cassava from smallholders and intermediaries who sourced from outgrowers. As a result, cassava production for export and domestic processing continues to expand in rural areas. In 2019, 22,713,014 metric tonnes of cassava were produced in Ghana, while the demand was 15,897,544 metric tonnes. In 2020, both production (24,368,693 metric tonnes) and demand (18,641,395 metric tonnes) increased tremendously (Statistics, Research and Information Directorate of Ministry of Food and Agriculture, 2021). Cassava is an essential food commodity that occupies an integral part of the socioeconomic development in Ghana (Bayitse et al., 2017). This makes its rising use for industrial production a new addition to the older agrarian questions raised about the commodity in Ghana, particularly in rural areas. According to Kleih et al. (2013), an estimated 1000 metric tonnes of high-quality cassava flour (HQCF) was produced in Ghana in 138 Agrarian South: Journal of Political Economy 11(1) 2011. HQCF has multiple uses. It is used in the bakery industry for producing biscuits and in the construction industry for plywood manufacturing. The authors predicted that market demands for HQCF would reach 10,000 metric tonnes in a short time. Other cassava products like industrial flour (IF) and cassava chips are essential raw materials in plywood and animal feed production, respectively. As a result of the industrial importance of cassava, some exporters have taken a keen interest in the crop. The exports of cassava products have also increased over the years. While, in 2010, the net weight of exported cassava products was 0.24 metric tonnes, the figure increased to 1998 metric tonnes in 2019 (Ministry of Trade and Industry, 2021). According to the Ghana Export Promotion Authority (2017, 2020), Ghana’s cassava starch export has increased from US$850,000 in 2017 to US$1 million in 2018. Export markets include the USA, Italy, China, Belgium, Germany, and Finland. Nigeria, Poland, Portugal, and Malaysia are identified as emerging markets. The growing demand for cassava starch for export, estimated to be US$1.77 billion, is driving investment globally. In Ghana, the domestic need for cassava products for industrial use provides an additional impetus, driving investment in large-scale production, contributing to land grabbing in rural areas. Theoretical Framing Rural anthropologists and agrarian political economists have identi- fied two core transformations in agrarian areas linked to external pres- sures on land. First, the development of land markets and promotion of neoliberal capitalism create different contestations over resources linked to the hierarchies of power relations, exclusion, and local citizenship struggles. The direction of the changes determines who is included or excluded in capitalist circuits of production. In addition, community and household relations in rural areas are redefined when land tenure regimes, rules, and practices change (Boone, 2007). Since land is an essential socioeconomic resource, it exerts a particular form of social order, hierarchy, and legitimacy (Lund, 2006; Lund & Boone, 2013). The introduction of new rules is accompanied by acceptance or rejection of the local institutions that have introduced the regulations. Peluso and Lund (2011) argue that land control has political and institutional dimensions through the reimposition of new practices and changes in older ones. Second, when forces external to the community instigate transformations in how farmers access land resources, the introduction of rules of access, Torvikey 139 ownership, and control allows the governing elite to change existing social hierarchies, leading to dispossession and exclusion of existing land uses. In areas with migrant farmers, their relationship to land and how they are viewed as local citizens change in tandem with land pressures and scarcity, which create tensions within the communities. Anthropologists have identified multiple meanings of local citizenship, identity politics and claims of belonging, which change with time and geography. Claims of autochthony, for example, are used commonly to connote the order of settlement in an area (Greene, 1996; Pelican, 2009; Geschiere, 2009; Geschiere & Jackson, 2006). This is mainly anchored in the time and sequence of arrival at a place. Social groups claiming this type of belonging argue that they, or their forebears, arrived in a particular area first. Besides, they confer cultural and spiritual bonds to the land that other groups either accept or contest. In addition, first-comer groups assume political supremacy and hierarchy in the social organization of an area. This work situates microlevel changes in land tenure practices and rules occasioned by land grabbing in the broader debates about the centering of land in citizenship in the rural areas and how this reshapes social relations in a deeply hierarchical society. Research Methodology The article draws on quantitative and qualitative datasets collected between 2014 and 2018 in five communities, namely Hodzoga, Aveafe, Alavanyo, Achianse, and Hodzove in the Ho Municipality of the Volta Region of Ghana. The communities are at the center of a booming indus- trial cassava production that started in 2005, with the establishment of an industrial cassava company (referred to here as Agro Cassava Company) that acquired 3000 ha of land for cassava production and processing. A combination of key informant interviews and observation was used during the first phase of data collection to understand the context of the company’s establishment and the social relations in the community. At this stage, the five communities were identified as the most directly affected by the land grab and other production models and processes of the company. Thus, the criterion for selecting the communities was mainly based on their relationship with Agro Cassava Company. Different categories of knowledgeable people from the communities were carefully sampled for key informant interviews. Traditional leaders, women, migrants, and religious leaders were interviewed to get various dimensions of the agrarian process, structure, and relations. 140 Agrarian South: Journal of Political Economy 11(1) The observation from the communities, especially from the differences in the Ewe language spoken, gave more significant meaning to the social structures in the communities, as, through this interaction, it was clear that the communities are a mix of Ewe people from Ghana and Togo. This stage of the data collection was iterative due to its context-setting nature. Key informant interviews were used to understand social relations in the communities and how they structured the agrarian processes. More particularly, the social history of the people and settlement patterns were linked to land and labor relations. These questions also revealed the cropping patterns. During the interview, one of the traditional male leaders confirmed that settler farmers from Togo introduced groundnut farming in the area. Another linked maize production to Avenor migrants from the Volta Region. The interviews also centered on the commodity production trajectories and how these shaped migration, land, labor, and gender relations in the communities. For example, cassava production was linked to cocoa and oil palm production in the early twentieth century. Cassava was only cultivated as a hunger crop as yam production became tedious. As a result, households committed most of their labor and fertile lands to produce export commodities. Interviews were conducted with officials of the Ministry of Food and Agriculture (MoFA) at the Ho Municipality and Volta Regional Office to map the state’s role in the agrarian transformations unfolding in the communities. In-depth interviews were conducted with 33 individuals in 19 households. The questions focused on household characteristics such as ethnicity, migration status, relationship, past and present land tenure practices, labor relations, and economic activities. The cassava company officials were also interviewed. The company’s factory and household farms were also observed. Intra-household cropping patterns and land relations emerged visibly. Five focus group discussion (FGD) sessions comprising two groups of men, two groups of women, and one mixed youth group were held in the five communities. The questions centered on the impact of industrial cassava production on land, gender, labor, and class relations. A survey of 400 households, constituting 70% of total households listed, was also conducted, as a second component of the research. A prior screener question was administered to identify households directly affected by the company, migrant households, and households headed by women. All households affected by the company’s land acquisition listed were sampled in addition to a prioritization of migrant- and female-headed households. Torvikey 141 The Socioeconomic Structure of Study Communities The study communities are in the Ho Municipality of the Volta Region of Ghana. Mainly agrarian, crop farming, including cassava, maize, rice, groundnut, pepper, and tomatoes, is the main livelihood activity for peo- ple in the area. In addition, a few farmers cultivate oil palm and cocoa. Most of the communities belong to the Ewe ethnic group, mainly from Ghana and Togo. The communities have a high migrant population, which are associated with agrarian transformation, especially commodity production. A survey of the households shows that 42% of 1235 adult household members enumerated were migrants (Table 1). The operational definition of permanent migration for this work is migrants who had stayed in the community for at least 12 months, which is the standard definition used by the Ghana Statistical Service (2019). The communities receive a high number of migrants. A total of 68% of migrants enumerated in the households were from the Volta Region. Another 17% were from other parts of Ghana and the rest from Togo, which shares both physical boundary and ethnolinguistic affinity with the Volta Region. The household survey showed that the top three reasons for the migration into the area were marriage and family-related motivations (45%), search for agricultural land (15%), and search for work (15%). Of course, these reasons were not exclusive as motivation for migration could change at the destination—however, migration for marriage and family-related reasons and land highlighted the agrarian structure in the communities. The communities were categorized into two groups, mainly for analytical purposes. Hodzoga, Aveafe, Alavanyo, and Achianse were categorized as autochthonous, while Hodzove was categorized as migrants. Table 1. Distribution of Migration Status of Household Members in Study Communities Migration Status Percentage Indigene 58 Permanent migrant 25 Children of migrants 17 Total 100 Source: Household survey (2017). 142 Agrarian South: Journal of Political Economy 11(1) The four autochthonous communities had some variations. Families in Hodzoga and Aveafe controlled the lands in the area, made possible by their early arrival and settlement status. They also controlled the traditional governance structures. However, it was families in Hodzoga who allocated lands to the other three communities. As the research progressed, it became clear that the communities were heterogeneous in many other ways that revealed the ramifications of the area’s development as a cassava frontier. Hodzove, which claimed a migrant origin and continued to have familial relations with their place of origin, was categorized as a migrant community. The hierarchy of control of land and its access became an essential factor in awakening a latent question of belonging and power relations, translated into a language of exclusion and rent seeking. Land Politics and Changing Local Hierarchies In 2005, a wholly Ghanaian-owned Agro Cassava Company acquired 3000 ha of land for industrial cassava production in the communities, creating tensions and reactions. The land acquired belonged to five land-owning families and a few individuals. It was the most signifi- cant acquisition in the area at the time. The land acquisition coincided with the global land rush, which peaked in the 2000s. The discourses mimicked discussions at the global level that justified the land grabs, namely job creation, farmers’ market access, and revitalization of rural economies. The company estimated that the land grabs dispossessed 100 farmers. However, the affected farmers gave a counter-estimate of about 300 farming household members. The contradictions in the number of land users prior to the land acquisition shed light on who is counted as a farmer and from what perspectives the counting is done. While the company may count individuals to abridge the impacts, farmers counted household members whose livelihoods depended on that land. The lease term was pegged at 10 years. However, the agreement was not renewed in 2017, when the tenure had long expired. The landowners blamed the nonrenewal on production difficulties that the company faced. The losses were uneven. Many dispossessed farmers were migrants across the communities and families in the late-arriving communities. Lands in Hodzoga and Aveafe belong to families like the land ownership structure among Ewe people in the Volta Region (Kludze, 1972, 1974; Kumekpor, 1974; Nukunya, 1969). A few individuals own lands that they acquired through purchase, gifts, and other means. Torvikey 143 It is important to note that, in these communities, a chief does not have control over community land, since the chief belongs to a family and can only exercise power over lands in that family, but it is only through the family structure. Various tenure regimes exist, mainly through maintenance of first clearance and adherence to social functions that allow migrants to use the land if they settle in the community. The initial negotiation processes started with some chiefs. One of the chiefs was a member of one of the landowning families. Another chief was the regent of the late-arriving autochthonous communities and a Municipal Assembly member of the local area at the time of the acquisition. Both played multiple roles, including serving as a liaison between the company and the landowning families and convincing the aggrieved families to accept the land deal. Lanz et al. (2019) found similar interference of chiefs in family land areas in the Volta Region of Ghana. The role of the chiefs in the family land issues elided the family’s power of self-determination. The acquisition process was met with initial resistance from some landowning families. The issue was reported at the Municipal Assembly as families threatened to go to court. Land users, on the other hand, did not take any such actions. The local government officials at the Municipal Assembly mediated between the landowners and the company and encouraged the former to release the land for the project, which the representative promised will create jobs for the youth in the communities. Some chiefs justified the dispossession of farmers by reimagining and recreation of the boundaries of citizenship in the communities. Kopytoff (1987) describes the political and social organization of many agrarian communities in Africa as a combination of settlement order and its historical importance, resource control, and migration patterns. Within this is the bountiful nature of land at the time the movements were occurring. These mobility patterns mostly lead to a bifurcated order of local citizenship where first-comer and latecomer status, although fluid, become the basis for the contest for and delineation of land resources. The difference between communities shows diversity in discourse about local citizenship of the area based on what anthropologists call “priority in time” and first coming. This classification emanates from the sequence of arrival with the community that labeled itself authentic citizens, which in Ewe language translates as aku aku. The expression aku aku ran through the interviews with some community elders. However, this language is commonly used only when land questions are raised, especially concerning the acquisition by the cassava company. 144 Agrarian South: Journal of Political Economy 11(1) The question of autochthony arose due to the people continuously negotiating long-term migrancy and asserting a hierarchy of power over land resources and local politics. The hierarchy of belonging was woven around allodial land ownership tied to the order of settlement. Greene (1996, p. 183) found a similar definition and redefinition of who belonged and who was the “other” in a study of the social history of the Anloga community in the Volta Region. She reiterated the issue with a very significant quote by an elder in a clan stigmatized as the “other” and who, frustrated with the label of “stranger,” retorted: “Can we be strangers after 100 years?” The latecomer autochthonous communities agreed in principle that the land was allocated to them. Nonetheless, members of this community are confident about the foundations of their local citizenship. Before the land acquisition by the cassava company, members of this community were not landless, nor were they landowners. Their land access was constructed within the customary rules, which permit land use perpetually. However, the acquisition threatened the inclusive land practices in the community. The company and the landowning families justified this claim to allodial ownership and nonrecognition of lesser interests, such as user rights, sharecropping, and annual rentals. This tension became the basis for local citizenship struggles and contentions. Migrants and Differentiated Resource Access Two waves of recent migration are recorded in the area, explaining the nature of resource access and control among migrants. The migration and settlement periods are based on oral histories but have remained significant in the people’s self-definition. This history is also at the base of differentiated access to natural resources. These are classified as land- induced and labor-driven migrants. The pattern of settlement and reason for migration differ significantly. However, the two groups of migrants do not have access to the same resources. Similarly, the resource access regimes and practices differ. The older migrants and recent migrants are differentiated as this section will show; migrants who arrived in the 1930s had more lands than their counterparts who arrived in the 1950s. Table 2 summarizes the features of the two groups of migrants in the area. The first wave of permanent migrants, who were land-induced, came to the area to escape ecological stress in their place of origin at Asafotu in the Volta Region. They left their communities due to a prolonged Torvikey 145 Table 2. Features of Migration Waves Migration Wave Date of Arrival Features First The 1930s • Mass settlement • Ecologically induced • Land search for agriculture • Wholesale land grant Second The 1950s • Seasonal labor migration • Gradual and individual settlement • Individual land access Source: Author’s fieldwork (2014–2018). drought that made farming impossible. This group arrived from a town further south of the Volta Region in search of agricultural land. They arrived in the 1930s and were allocated lands by landowning families already settled there. Tracing the trajectory of migration into the area, a 70-year-old migrant whose father helped in founding the migrant community estimated the year of the arrival of his people to the area as 1933. According to him, his people came seeking land for agricultural production. The land allocation was based on the logic of integration. Settlers could use the land if they continued to live in the area. There was also an instrumental reason for the allocation of the land in addition to its integrative features. Elders of the Hodzoga community affirmed that lands allocated to the migrants are at the boundary with neighbouring communities. Therefore, inherently, the settler farmers were used to secure land boundaries with other communities and to prevent encroachment and invasion. The settler farmers established their community, and families had some degree of control over land. The founding members of the community were responsible for land redistribution to the different families. Due to the undeveloped nature of the land markets, there were no monetary exchanges in the land tenure system, which was based on allegiance and contribution to and participation in social events like festivals and funerals. Additionally, after every harvest season, the community gave part of its crops to the paramountcy. However, the exact quantity of the gift was left to the discretion of the leaders of the migrant community. This notwithstanding, there was still class- and gender-based land concentration and accumulation: families that founded the village owned more significant lands than others. The migrant chief of the community owned over 70 acres of land. This land size was more extensive than 146 Agrarian South: Journal of Political Economy 11(1) lands controlled by some indigenes. Differentiation was based on whose parents participated in founding the community. The founder and non- founder dichotomy and relations are other layers of local citizenship within the migrant community, which determined resource access and control. There were also gender dimensions to land access and control. Men owned more land than women. The nature of one’s citizenship determined the kinds of resources that one could access and how they could be used. Therefore, by the nature of their land access and control, women in this community were treated as sub-citizens, as male land ownership was the norm and prioritized. The processes that set the second wave of permanent migration started in the nineteenth century, where commodity production peaked. During this period, land abundance caused labor scarcity in the area. As a result, wealthy peasants brought labor groups from communities in Togo to work seasonally on farms. As a result, the seasonal migrants began to settle gradually. The exact date of permanent settlement of these groups of migrants was estimated to be the 1950s. In an all-female FGD, some respondents in a Sukukpota, micro-migrant settlement within the Hodzoga community confirmed the period of their migration by saying: “we are told our settlement is about 50 years old. Many of us are from Ando in Togo.” The date was corroborated by a religious leader and an indigene whose father hosted some of the then labor migrants until they started their permanent settlement. He indicated that this latter group from Togo settled in the area more recently than their counterparts from the southern part of the Volta Region. One could easily see a demarcation of Togolese settler farmer areas, which demonstrated both integration and separation of migrants and indigenes. Other migrants who arrived later established residences within various parts of the communities. Compared to their first wave migrants, both farm and residential lands of the second wave migrants were contiguous, but their second-wave counterparts were fragmented and spread across various parts of the community. These differences in migration and settlement patterns affected their socioeconomic position in the community. The pattern of migration by the migrants from Togo into the area showed labor migration–based mobility. The laborers worked mainly on cotton, cocoa, oil palm, and yam farms and were accommodated by the farmers who brought them. The accounts further noted that as payment for labor became difficult, some farmers gave parcels of their lands to the migrants to cultivate food crops to meet their food needs. This practice of farmers granting free lands to laborers for food crop cultivation has Torvikey 147 been described as a healthy way of keeping social and agrarian relations going (Hill, 1970, 1986). The laborers only went back to their communities in Togo at the end of the agricultural season. The contracting farmers accommodated the laborers until the end of the farming season. Due to land abundance and less population pressure at the time, migrants from Togo settled permanently in the area gradually. With the availability of food crops, many of the laborers, mainly males, brought their wives, children, and other relatives. As their population increased, they sought lands on their account to cultivate. Contrary to the large land allocation to the Hodzove community, the migrants from Ando in Togo acquired lands on mixed-tenure arrangement terms. Individual farmers acquired lands through outright allocation, sharecropping, yearly rentals, inheritance from settler grandparents, and gifts. Some of the migrants also accessed lands through marriage to indigenous women from landowning families. Like their southern counterparts, participation in social events and contribution toward the same was a requirement for sustaining the land relations. This has been practiced over the years. The differences in the landholding structure of these two types of migrants also showed in their differential landholding structure and the size of land controlled by its members. The first wave of migrants who established the Hodzove community controlled larger tracts of lands than their second-wave counterparts from Ando in Togo. Land concentration and accumulation were minimal among the second-wave permanent migrants as compared to their first-wave counterparts. For example, whereas the chief in the first-wave migrant community controlled 70 acres of land, his second-wave counterpart owned a paltry 10. The second-wave migrant women mainly accessed lands through husbands and fathers. Due to land scarcity, the women were allocated fallow lands. They planted leguminous crops to revitalize the soil or cassava since it was the only crop that could do well on such marginalized lands. These migration patterns shaped women’s access to land and their differentiated dispossession when the cassava company acquired it. While the first-wave migrants accessed lands as members of their more extensive lineage, their second-wave counterparts did so through their nuclear household units. In both spaces, kinship determined the land access rules for women and the basis for their incorporation into capitalist production and, at the same time, their proletarianization. Kea (2012) proposes that agrarian clientelism and the incorporation of migrants into capitalist production relations reinforce migrants’ conception of local citizenship. At the same time, this can provoke contradictory questions of belonging. For pragmatic and instrumental 148 Agrarian South: Journal of Political Economy 11(1) reasons, indigenes can create spaces for migrants for specific resources and not for others. The relationship of this group to the landowning community also changed considerably since their inclusion was based on their migration status and, therefore, a weak temporary land control regime. Farmers in this community lost crops on the land, and as a company official had said, “they are not the landowners, so we did not consult them.” As argued elsewhere (Torvikey, 2021), companies feign ignorance of landholding structures in communities to absolve themselves from blame in case of conflicts. Apart from the community-based classifications linked to land relations, migrants and indigenes could be found in any community. Nonetheless, these features predominated and highlighted the class-based features of differentiation in the communities. Triggers of Identity Formation The agrarian production processes discussed earlier created multiple ways of conceptualizing land-centered politics of belonging and assumed dif- ferent functions for the autochthons and migrants alike. The stark changes in social relations and agricultural production intensified with the land acquisition by the Agro Cassava Company in 2005. In acquiring the land, the company introduced a new language of land relations into the area. The acquisition process invoked legitimation of power hierarchy, which justified land acquisition without the users’ due consultation. Akaateba (2019) argues that, in Ghana, the development of land markets that have increased land competition trigger a redefinition of customary land prac- tices and often results in the exclusion of some people. The deliberate use of “landlords,” a term not known in the region’s communities, is a pre- text for the exclusion and to justify discrimination of migrants and late- arriving autochthonous farming households whose land access depended on allodial title holders. The following statements by interviewees highlight the ramifications of the land grab on land tenure practices: If your ancestors did not own land but you want to farm, you would contact landowning families and individuals to allocate land to you. Then, if the crops did well, you would want to thank them with some crops. However, now, because of civilisation, people are renting the lands and getting money, so if you want land for cultivation, it is difficult. So they give it to those who can afford to pay for it. (Delali, male, farmer, 32 years, indigene, 30 March 2017) Those days there were no strict rules for an indigene who wanted land to farm. One could walk to the landowner and ask for land freely given for production. It was only in the case of tree crops such as oil palm and cocoa Torvikey 149 that sharecropping applied. Even that, when the crops matured, the farm, rather than the crop, was shared. So food crops did not attract any fees. However, people are taking annual rent for all types of production now. What I have realised is that the landowners do not want to farm. They now rely on rent from the land. This is what is happening in this area now. People now see cassava as money-making venture as more people are investing in the sector. So, cassava has also caused changes in land tenure practices. If you produce more, you know it is an investment towards the future, so you will need to pay for the land. (Unit Committee member, male, indigene, 30 March 2017) Two types of claims were identified. The first was the claim of early set- tlement, conquest, and land accumulation. An elder in Hodzoga used the concept of aku aku, which translates as “proper,” “authentic,” and other variants of purity. This was directed at the community members who were considered not migrant but also not authentic enough due to second arrival status. With this, the landowning community used its access to land to question the citizenship of other community members. The second claim to autochthony is derived from the level of control over land. The claim is distinguished by allodial titleholder versus land uses. This claim challenges the social basis for resource access, control, and inclusion. This type is directed at migrants, particularly because autochthony and migrancy become fluid and only used when land becomes highly commoditized. Nevertheless, how long can a person remain a migrant? In Ghana, even at the national level, how long a person can remain a migrant lingers and is raised with the ebb and flow of the country’s economic fortunes. The migration discourse does not apply equally to specific migrants who are seen as threats at a particular moment in history. Akyeampong (2006) makes this case for the Lebanese population accepted and rejected in equal measure at different periods. At the same time, Asante (2019) notes that similar questions of belonging are equally being raised at the national level. The third element of exclusion is its unevenness and its socially differentiated nature. Sapana (2012) argues that displacement and dispossession are often class based. Land dispossession affected migrant and late-arriving autonomous communities differently. As Hall et al. (2011) argue, exclusion should be seen in land access shifts. It is noted that while migrants are dispossessed, younger migrants are disproportionately affected as older migrants control more considerable land sizes and, subsequently, many of them later moved to virgin lands when they were dispossessed. Again, dispossessed older migrants were able to acquire more lands in frontier villages. The reactions of two migrants with generational and resource differentials exemplify these arguments. 150 Agrarian South: Journal of Political Economy 11(1) The 72-year-old wealthy migrant farmer whose father was a founding member of the earlier arriving migrant community reacted to the acquisition and dispossession of farmers from that community in this statement: nothing comes out of litigation. After all, nobody is fair; if someone does not understand me, another person will. When I got there and realised that my farm was part of the land sold to the Company, I returned home. I had just 11 acres of land there, and I lost all of that land. In the early 2000s, when the land was acquired, I owned a tractor, and I had money, so I did not worry about that small land lost. I had seventy acres of land at the other side of the com- munity, so I did not worry because you will soon die if you continue litigation of land matters. I did not do anything about my land loss. I was cool about it and continued producing and successfully too. (Togbe Akoto, male, migrant community, 30 March 2017) According to Berry (1993), the structure of agrarian areas in Africa makes peasants’ decision not to confront landowners on land-loss issues an enhancer of social relations rather than a hindrance. However, while the author emphasized that the ability to mobilize in rural Africa made exclusion disadvantageous, it appears in this case that the effect of migrant exclusion has depended much on the gender, class, and status of the migrant. Many peasants who lost lands emphasized their subordinate position as tenants who had no right to question how and to whom the land was devolved. The generational differential impacts of the dispossession had many ramifications for migrant youth and their livelihood pathways. There were several outcomes: proletarianization, where some youth took jobs in the company as an alternative to landlessness; de-agrarianization, where affected youth out-migrated to engage in nonfarm work; acquisition of another plot of land for production; and a mix of all three. A 31-year-old, born in the community to Togolese migrant parents, lost land and took employment in the company but then quit due to low wages. He explained the circumstances leading to the land loss as follows: my father came to ask for land over here, and they gave it to him for farm- ing. My parents had been cultivating the land for a very long time before we were born. One day we were on the farm harvesting maize for my father, and a tractor came and destroyed the crops on the land. The tractor was just busily uprooting and gathering the debris, so we asked what happened, and they said they leased the land and were going to produce cassava on the land. I had my farms there prior to the arrival of the cassava company. I had three acres of maize, two acres of groundnut, two acres of beans and an acre of okro. My parents gave the lands to me. We lost all the lands without Torvikey 151 knowing. We only realised that anytime we attempted to cultivate part of the land, the Company stopped us. When that happened, I stopped farming and got a job on the Company’s farm. I worked there for three years and stopped due to low wages. I migrated to Ho where I did menial jobs. I returned to the community when I acquired another land to start farming. However, the land was near the communities, so livestock disturbed the crops. That is why I have stopped farming and now operate this rickety tricycle commercially, carting goods from community to community. (Xornam, son of a Togolese migrant farmer, 29 March 2017) In terms of the gender-based outcomes, women complained about the distance to lands in the frontier villages, which would have severe impli- cations for combining reproductive and productive work. In addition, they could not afford new lands to enable them to continue their agricul- tural production. The dispossession triggered further limits on land access within the family. Land grabbing from within the family was rife. A total of 77% of lands cultivated by migrant women were for male relations, namely husbands (44%) and fathers (33%). Since many migrant women derive land rights from men, male dispossession impacts women’s land access and is the cause of micro-dispossession in the household unit. Kinship relations determine the terms of women’s access to resources. Meillassoux (1981) linked kinship with the social and economic organization of the household. In this framing, marriage and men’s control of political power are linked to the productive and reproductive spheres where land is also central. The interweaving of reproductive and productive spheres in an agricultural household makes recognition of women’s work difficult, and therefore their productive resource access is subsumed under that of men. Internal land grabbing of women’s lands by men occurred as they laid claims to the available household lands for production. In this case, in households where men allocate lands to women, displacement of men created shifts in land use and the types of lands that the women can access. Migrant women described as associative migrants also relied on men for land allocations. Men cultivated the few plots left, and women who continued agricultural production had to cultivate smaller plots repeatedly. This continuous land use had deleterious effects on their output and yields. Many affected migrant women obtained jobs in the company, although these were casual jobs with little security. Many withdrew to resist low wages and strict working hours. Other women left farming altogether to trade. Affected male farmers in Achianse and Alavanyo went to frontier villages to acquire lands, while women in those communities occupied part of the company’s land. The differentiated 152 Agrarian South: Journal of Political Economy 11(1) features of the exclusion and prior land relations shaped the responses of the various social groups affected by the land acquisition. Resistance Strategies The exclusion does not go unnoticed and without reactions, and as Mamdani (1998) argues, the encounters of belonging and struggles for resources are class based. The affected communities used both evasive and confrontational strategies to resist their exclusion and to correct it. Dispossessed women from the autochthonous communities occupied part of the company’s land. They also demanded that landowners allocate new agricultural lands to them. For some, the situation demanded clandes- tine land-use strategies. The development of land markets also affected women’s land-use rights, which became difficult, especially with tree crops. Some women planted cocoa, considered a lifetime asset crop at night, to circumvent land scarcity and changing access rules. Others planted the young tree crops within the food crops to hide them until they matured. They feared that the land would be taken away from them should landowners realize that tree crops rather than food crops are planted on the farm. These strategies are also resistant to the increasing rent-seeking behavior of landowners. If women and men had to compete for land for sharecropping, women usually lost out as farm owners looked for farmers with resources that could bring maximum output for sharing. In this case, dispossessed male farmers in Achianse, for example, acquired lands in other communities. It was not surprising that it was women in this community who led the resistance against the acquisition. In a household where both husband and wife were dispossessed, the woman occupied the company’s land. She continued cultivating there, while the man went further afield to frontier villages to acquire land. Deere (1976) argues that women’s economic functions in peripheral capitalist production systems are the reason they remain at the frontlines of land occupations historically. The renegotiation of power relations between the customary owners of the land and women led to new land allocations to the latter. The land acquisition exposed the class structure, which places women under men in the communities, especially regarding land access. As men and women related to resources differently, dispossession and exclusion became differentiated as well. Women became the worst affected and were also those who mounted the fiercest resistance to the acquisition. The women used resistance as a counter to claims of autochthony. One woman exclaimed: “we agree that our Torvikey 153 ancestors were allocated lands here; nevertheless, we are citizens of the community too.” In the migrant community, men boycotted the contract farming scheme that the company integrated into its production system. Farmers in this community did not make demands on the landowners for more lands, since families still had reserve lands, which men in the family quickly began to cultivate. The exclusionary strategies affected women disproportionately in the affected communities, which revealed the power hierarchies that shaped resource access and control. Conclusion Land-centered claims of belonging and identity politics are rife in rural areas. External factors, like the remarkably abrupt land dispossession, are an essential element of the current processes. Historically, export and industrial commodity production have transformed the class structure of communities due to migrant inflows into those areas. This adds to the current struggles in already differentiated communities. In this study, it has been argued that land grabbing has created the reimagination of communities through land-centered politics of belonging and local citizenship claims. The reordering of land access rules defines who qualifies to benefit from land and its resources and whose interests are legitimized in land deal negotiations. The co-optation of traditional institutions into capitalist economic systems through their multiple roles as liaison between the state, capital, and communities created new forms of identity politics and exclusion and shifts in political and social inclusion. In creating the history of settlement within the current processes of land dispossession, the study carefully detailed the political reactions from below linked to the differentiated resource access and asset base. It is argued that the political strategies deployed by affected people are linked to their differentiated dispossession; thus, for instance, wealthy settler farmers with considerable land sizes avoided confrontation in the context of a history of already accumulated land and the quest to maintain social ties with the landowning authorities. The dimensions of the land dispossession and differentiation were analyzed using cassava frontier communities in agrarian Ghana. The framework elucidated the materiality of land to people. The study centered land in local citizenship claims and politics of belonging and exclusion through the prism of the historical processes underlying 154 Agrarian South: Journal of Political Economy 11(1) the founding of communities and the ramifications of the land grabs. These processes have been made more apparent through the changes in social relations and the political reactions by affected people. It is argued that, in the current conjuncture, tensions about local citizenship and belonging in agrarian areas demand full scholarly attention. Researchers must render a complete account of the determinants of autochthonous claims and how these are related to the transformation in local citizenship and local hierarchies. The history of the settlement is vital in unpacking the nature of differentiated access and resource base, which fit into differentiated dispossession and reactions. Land, an essential part of the bundles of rural citizens’ material conditions, has undergone rapid tenure changes. The contestations that change land and social relations also redefine, reconfigure, and reclassify land and social groups. Land grabbing and enclosures have changed the very social base of rural areas, and new conceptual tools are needed to understand them. Declaration of Conflicting Interests The author declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article. Funding The research received funding from the DEMETER (Droits et Egalité pour une Meilleure Economie de la Terre) Project funded by the Swiss National Science Foundation (SNSF) and Swiss Programme for Research on Global Issues for Development. ORCID iD Gertrude Dzifa Torvikey https://orcid.org/0000-0001-8848-5994 References Addo-Fening, R. (1997). Akyem Abuakwa 1700–1943: From Ofori Panin to Sir Ofori Atta (No. 19, Trondheim Studies in History, African Series No. 1). Norwegian University of Science and Technology. Akaateba, M. A. (2019). The politics of customary land rights transformation in peri-urban Ghana: Powers of exclusion in the era of land commodification. Land Use Policy, 88(2019), 104–197. Akyeampong, E. K. (2006). Race, identity and citizenship in black Africa: The case of the Lebanese in Ghana. Africa, 76(3), 297–323. Asante, K. T. (2019). Imagining citizenship and belonging in Ghana. Development, 63(1), 90–94. Torvikey 155 Bayitse, R., Tornyie, F., & Bjerre, A. (2017). Cassava cultivation, processing and potential uses in Ghana. In C. Klein, (Ed.), Handbook on cassava: Production, potential uses and recent advances (pp. 313–333). Nova Science Publishers. Berry, S. (1993). No condition is permanent: The social dynamics of agrarian change in sub-Saharan Africa. The University of Wisconsin Press. Boone, C (2007). Property and constitutional order: Land tenure reform and the future of the African state. African Affairs, 106(425), 557–586. Borras, J. S., McMichael, P., & Scoones, I. (2010). The politics of biofuels, land and agrarian change: Editors’ introduction. The Journal of Peasant Studies, 37(4), 575–592. Chauveau, J., & Richards, P. (2008). West African insurgencies in agrarian perspective: Côte d’Ivoire and Sierra Leone compared. Journal of Agrarian Change, 8(4), 515–552. Deere, C. D. (1976). Rural women’s subsistence production in the capitalist periphery. Review of Radical Political Economics, 8(9), 9–17. Geschiere, P., & Jackson, S. (2006). Autochthony and the crisis of citizenship: Democratisation, decentralization, and the politics of belonging. African Studies Review, 49(2), 1–7. Geschiere, P. (2009). The perils of belonging: Autochthony, citizenship, and exclusion in Africa and Europe. University of Chicago Press. Ghana Export Promotion Authority. (2020). Potential markets-cassava markets. Ghana Export Promotion Authority. Ghana Export Promotion Authority. (2017). Competitor analysis: Cassava in China. Ghana Export Promotion Authority. Ghana Statistical Service. (2019). Ghana living standards survey round 7: Main report. GSS. Greene, S. E. (1996). Gender, ethnicity and social change on the upper slave coast. James Currey. Hall, D., Hirsch, P., & Li, T. M. (2011). Powers of exclusion: Land dilemmas in Southeast Asia (Challenges of the agrarian transition in Southeast Asia (Chatsea)) Illustrated edition. University of Hawaii Press. Hill, P. (1961). The migrant cocoa farmers of southern Ghana. Journal of the International African Institute, 31(3), 209–230. Hill, P. (1963). The migrant cocoa- farmers of Southern Ghana: A study in rural capitalism. Cambridge University Press. Hill, P. (1970). The occupations of migrants in Ghana. Anthropological Papers. University of Michigan Press. Hill, P. (1986). Development economics on trial: The anthropological case for a prosecution. Cambridge University Press. Kea, P. (2012). Becoming local citizens: Senegalese female migrants and agrarian clientelism in the Gambia. African Studies Quarterly, 13(3), 1–12. 156 Agrarian South: Journal of Political Economy 11(1) Kleih, U., Philips, D., Wordey, M. T., & Komlaga, G. (2013). Cassava market and value chain analysis: Ghana Case Study. University of Greenwich, Natural Resources Institute and Accra, Ghana, Food Research Institute. Kludze, A. K. P. (1974). Family property and inheritance among the Northern Ewe. In C. Oppong (Ed.), Legon family research papers, No. 1: Domestic rights and duties in Southern Ghana. University of Ghana, Institute of African Studies. Kludze, A. K. P. (1972). Problems of intestate succession in Ghana. University of Ghana Law Journal, 9(2), 89–122. Kopytoff, I. (1987). The internal African frontier: The making of African political culture. In I. Kopytoff (Ed.), The African frontier. The reproduction of traditional African societies (pp. 3–84). Indiana University Press. Kumekpor, T. (1974). The position of maternal relatives in the kinship system of the Ewe. In C. Oppong (Ed.), Legon family research papers, No. 1: Domestic rights and duties in Southern Ghana. University of Ghana, Institute of African Studies. Lanz, K., Prügl, E., & Gerber, J. D. (2019). The poverty of neoliberalized feminism: Gender equality in a ‘best practice’ large-scale land investment in Ghana. The Journal of Peasant Studies, 47(3), 525–543. Lund, C., & Boone, C. (2013). Introduction: Land politics in Africa—constituting authority over territory, property and persons. Africa, 83, 1–13 Lund, C. (2006). Twilight institutions: An introduction. Development and Change, 37(4), 673–684. Mamdani, M. (1998). When does a settler become a native? Reflections of the colonial roots of citizenship in equatorial and South Africa. Text of inaugural lecture as A C Jordan Professor of African Studies, University of Cape Town, Lecture Theatre 1, Education Building, Middle Campus, Wednesday 13 May 1998, 8.15 p.m. https://citizenshiprightsafrica.org/wp-content/ uploads/1998/05/mamdani-1998-inaugural-lecture.pdf Meillassoux, C. (1981). Maidens, meal and money: Capitalism and the domestic community. Cambridge University Press. Ministry of Economic Planning. (1977). Five-year development plan-1975/76- 1979/1980. Ministry of Economic Planning. Ministry of Trade and Industry (unpublished data obtained in 2021). Export and import commodities. MOTI. Moyo, S., Tsikata, D., & Diop, Y. (2015). Introduction: Africa’s diverse and changing land questions. In S. Moyo, D. Tsikata & Y. Diop (Eds.), Land in the struggles for citizenship in Africa (pp. 1–34). CODESRIA Publications. Nukunya, G. K. (1969). Kinship and marriage among the Anlo Ewe (Monographs No. 37). Athlone Press. Pelican, M. (2009). Source complexities of indigeneity and autochthony: An African example. American Ethnologist, 36(1), 52–65. Peluso, N. L., & Lund, C. (2011). New frontiers of land control: Introduction. Journal of Peasant Studies, 38(4), 667–681. Torvikey 157 Sapana, D. (2012). The politics of the evicted: Redevelopment, subjectivity, and difference in Mumbai’s slum frontier. Antipode, 00(0), 1–22. Statistics, Research and Information Directorate of Ministry of Food and Agriculture (unpublished data obtained in 2021). Crop production and demand statistics. SRID-MoFA. Tonah, S. (2006). The presidential special initiative on cassava: A bane or blessing to Ghana’s smallholder farmers. Ghana Journal of Development Studies, 3(1), 66–84. Torvikey, G. D. (2021). Land, gender, and class relations in Ghana’s cassava frontier. Agrarian South: Journal of Political Economy, 10(1), 139–154. Yaro, J. A. (2010). Customary tenure systems under siege: Contemporary access to land in Northern Ghana. Geo Journal, 75(2), 199–214.