Embracing risky refuge: women, land laws and livelihood vulnerabilities in rural Kenya

There has been a growing interest in laws governing resources particularly land in reference to gender in Africa. Law reforms in relation to land have produced potentially useful regulations and espoused egalitarian land rights. However, the backdrop to these reforms contains a scene of land disputes, resistance to laws, violence against women and poor enforcement leading to injustices to women with a pervasive effect on families in vulnerable communities. Using focused ethnographic research methods, the writer investigated women’s land rights between November 2015 and August 2016. In-depth interviews, focus group discussions, review of archival records and observations were utilised. The study found that in contexts of prolonged LAW DEMOCRACY & DEVELOPMENT LAW DEMOCRACY & DEVELOPMENT VOLUME 25 (2021) Special Edition DOI: http://dx.doi.org/10.17159/20774907/2020/ldd.v25.spe4 ISSN: 2077-4907 CC-BY 4.0 LAW, DEMOCRACY & DEVELOPMENT/ VOL 25 (2021) SPEC ED Page | 72 livelihood vulnerabilities, as in the case of the Luo people of south-western Kenya, women seeking refuge from livelihood difficulties employ two strategies to anchor their security. They migrate from marital homes to fishing villages and also lay claim to marital land, which is held by men according to customary laws. These strategies produce social dilemmas and risky manoeuvering. Statutory land laws that are enacted to mitigate land related conflicts undermine the existing customary land laws that advantage men. Therefore, women’s land claims, and statutory land laws that espouse equality in land ownership, destabilise men’s sense of masculinity. By drawing on the experiences of women, I show the intersection between land laws, enduring injustices and gender relations in a context of strained livelihoods.


INTRODUCTION
Land issues have been discussed through a range of different prisms in sub-Saharan Africa. The scholarship has focused on the historical evolution of land from community tenure towards individualised rights, 1 population growth and ensuing land conflicts, 2 gendered land management 3 , and the enactment and reform of land laws. 4 Verma's research shows how gender norms shape land control particularly soil management for agricultural purposes in western Kenya. My analysis of the Luo women and land recognises the critical significance of land and relationships between people through land. 5 Land is an important form of natural capital, in which people anchor their livelihood opportunities not only for farming but also as a safety net for women who migrate to other places. In the case of the Luo women in rural Kenya, a home on marital land is important as a place where they root themselves. In addition, the value of a home on marital land also constitutes acceptable femininity -being a dhako moromo (a complete woman), which frames women's aspirations.
This article explores contestations about land, which has material and symbolic value for both women and men. I discuss the context of women's land claims as anchored in their rights as married women, by looking at rights that are sanctioned by customary land laws and statutory laws. In an environment characterised by the marginalisation of women, the dispossession of widows, and familial disputes over land, I show how women's land claims and the provisions of statutory laws that espouse equality destabilise the accepted idea that men control land. In a setting of multiple instabilities, men's self-understanding as proper men hinges on their sole right to own and distribute land especially when the other marker of manhood, that of providing for family, is limited. Faced with this threat to acceptable manhood, they respond through physical violence towards "their" women.
Women's varying social positions influence their security by shaping their access to land, however mediated by men. This variation depends on their performance of acceptable femininity by fulfilling marital and reproductive obligations. Marriage acts as the only opportunity through which a woman's residence arrangement is defined and secured. Unmarried women are regarded as incomplete, lacking the characteristics of acceptable femininity and therefore not recognised insofar as land use or residence are concerned. Yet, by discussing the lived realities of widows, I demonstrate that counting on rights of married women according to customary laws may also not offer security. Widows still face physical abuse on, and dispossession of, their marital land. I argue that despite Kenya's constitutional provision, which outlaws gender discrimination on matters such as land, there is a lack of awareness of these laws and resistance to them. Steep gender inequalities, justified in customary terms but equally the result of recent change, also persist. The overlapping effects of customary land arrangements, statutory land laws, and widespread vulnerabilities are crucial in understanding the gendered nature of livelihoods in the fishing villages and in rural Kenya more generally.
The analysis in the article is limited to the lived experiences of women in their efforts to lay claim to land using customary sanctions. This article does not analyse the jurisdictive processes or formal court cases in regard to women's land claims. To set the scene within which women lay claim to land, a livelihood context and legal pluralism context in regard to land laws will be offered.

Livelihoods and law reforms context
Southwestern Kenya has undergone considerable livelihoods stress due to high levels of unemployment, 6 environmental degradation leading to poor farm outputs, 7 and reduced fish yields from Lake Victoria. 8 Moreover, high HIV/AIDS prevalence has also produced social and economic strain for the community. 9 Unemployment has had the effect of reducing remittances to rural areas, leaving them dependent on unreliable lake resources. 6 Francis E "Gender and rural livelihoods in Kenya" (1998) 35 (2) Journal of Development Studies 72. 7 Conelly WT "Population pressure, labour availability, and agricultural disintegration: the decline of farming on Rusinga island, Kenya" (1994) 22(2) Human Ecology 145 at 145 Due to this scenario, rural settings have been associated with a struggle to survive, an outlook illustrated by a common phrase amanya manya ("am just searching and searching"). While the older generation thrived on the promise of modernity and managed to secure jobs in the urban centres, at present jobs are not readily available even for those who have some education. Indeed, other research in rural western Kenya has noted the discouragement among the youth who have skills for work but lack paid jobs to sustain their lives. 10 These youth spend their time amanya manya-ing through unreliable income activities or clutching their school certificates as they move about searching for work.
The problems of unemployment, environmental degradation, and reduced natural resources are not unique to rural Kenya. However, the instability of southwestern Kenya has been aggravated by high HIV/AIDS prevalence. This has reduced productive labour capacity due to high morbidity and mortality. Further, the disease has also destabilised family arrangements and heightened social and economic insecurity in a broader sense. For instance, women who are widowed as a result of HIV/AIDS have attempted to seek basic security through land ownership claims against the kin of their deceased husbands, but this has led to new forms of family dispute. Of particular importance for such women are the provisions of customary and statutory laws, which protect their land rights, but whose application is uneven as I show below.
The context of persistent multiple social and economic instabilities has shaped land issues. For instance, while customary laws entrench married women's entitlements to use land, their rights to own and manage are limited when they are widowed. This article explores women's options for seeking redress in land disputes, and how these options influence men's understanding of their own masculinities.
The historical law reform efforts towards equality in land access and control in Kenya are buttressed in terms of various laws, treaties and conventions: The UN's endorsement of the Convention on the Elimination of All Forms of Discrimination against Women (CEDAW) 11 ; the Universal Declaration of Human Rights (UDHR); the Africa Charter on Human and People's Rights 12 ; Kenya's Constitution Articles 60 (1) (a), (b), (f) and (g) 13 ; and Kenya's Land Registration Act and Matrimonial Property Act 14 .
While these land law reforms aimed at mitigating inequalities have offered solutions, some problems presented by legal pluralism are apparent.

Legal pluralism in context
Legal pluralism refers to cases where more than one set of laws is enforced among a group of people. 15 This may include the application of customary laws and statutory laws. Both statutory laws and customary laws are applied by people who are socially recognised, and enforced through well-defined processes. Customary laws include norms, rules and regulations that are entrenched within the social context of a particular community and serve the role of controlling and regulating behaviour, relationships and the use of resources, such as land. 16 Woodman argues that laws are enforced through control wielded by people who are socially sanctioned to do so, and they carry specific consequences in case of deviation. 17 A range of forms of legal pluralism has been explored worldwide, revealing their prevalence and influence on day-to-day social, economic and political environments. In Africa, the background to legal pluralism can partly be traced to colonial history through which foreign laws were introduced to organise colonies where traditional arrangements already existed. 18 Whereas there existed widespread customary arrangements of order and control, the colonial regime and consequently newly independent States brought a parallel set of laws to which the populace was subjected. Therefore, competing claims over land, according to parallel sources of authority, are not new. Such contexts are made still more complex by different actors and their divergent interests as the case of Kenya attests. 19 The existence of multiple sets of laws may be seen as harmful, 20 but they may also be regarded positively. Negative views of legal pluralism in relation to land laws argue that it produces insecurity among interested parties, creating a "tragedy of contested access" in which confusion about rights results from land users not being sure about which laws to adhere to. 21 Ansoms et al observe how interested parties make use of what they call "open moments", such as, confusion between legal codes, or between lawenforcement institutions, to advance their interests when overlapping legal regimes are in force. 22 These "open moments" come about as a result of intersecting events or situations: sudden demands for the implementation of some laws; inadequate governance structures, such as in conflict prone communities; and the emergence of powerful commercial land users who change the prevailing land use arrangements. These factors, as Ansoms et al argue, provide opportunities for people to renegotiate norms, which in turn either provide space for the rights of the marginalised or entrench prevailing power relations. 23 This is certainly the case among the Luo, where confusion between the provisions of the statutory laws and customary laws prevail, creating "open moments". These "open moments" create the ground for the dispossession of widowed women.
By contrast, some theorists see legal pluralism as advantageous. They argue that it allows adaptability to changes within a community. 24 Krueger's study in the Nyambene Hills in eastern Kenya found that although State law weakened the customary institutions of elders, it supported the establishment of some of their interests, such as protection of forested land. 25 Turner also observes that plurality means that different norms or ideas are available as people encounter change, offering alternative arrangements if one legal framework falls short. 26 These studies reveal that multiple sets of laws enable people to be more flexible in the face of instability and insecurity than single systems of laws.
In my research I found that, under customary law, lineage land is protected, while statutory law protects widows' rights to land. Furthermore, in customary law, married women have full rights to live on their husband's land even if they do not own it. As my study found, although such rights grant security to women, mechanisms for enforcing such entitlements have been weakening. Yet, multiple legal codes cause confusion and conflict. In what follows, I thus show how this state of affairs shapes gender relations and how social injustices are produced more generally. I argue that the multiplicity of legal codes presents more problems than advantages partly because of the extent of everyday livelihood vulnerabilities.

METHODOLOGY
This article is part of a study on women's negotiation of gender relations in times of prolonged livelihood instability in rural Kenya. The research was carried out using focused ethnographic methods among the inhabitants of five fishing villages along Lake Victoria. In-depth interviews, focus group discussions, and participant observations were utilised to collect data. The ethnography focused on how women cope with land claims in the context of customary and statutory land laws. Stories about women and their land claims were collected during field research, and these provided the basis for conversations in the focus group discussions. The stories were read aloud to the participants, followed by discussions exploring the options presented by customary arrangements for solving the problems raised in the stories. Relevant statutory laws were then read, prompting discussions about whether these laws would mitigate the problems presented in the cases. As part of discussing the options, the participants first outlined customary land laws as they saw them -both as abstract principles enshrined in Luo traditions and as practically applicable rules.

Acceptable femininity as a limitation to land claims
In southwestern Kenya, cultural norms and arrangements are crucial for understanding land issues. These norms are sharply gendered, and have particular significance for migrants in the fishing villages. Although some people temporarily leave their marital or ancestral land for fishing villages, its significance remains central to how they imagine their futures. The fishing villages are temporary abodes to which men and women migrate to eke out a living from fishing related activities. Ancestral and marital homes, on the other hand, represent permanent residence because this is where men own land as part of lineage inheritance, even if they have to move back and forth from the fishing villages. Communal ownership of lineage land has implications for widows' land claims, but a widow may have relative ease with regard to a land claim if it is self-acquired. Reference to self-acquired land among the research participants was minimal.
Marital status and residence in a marital home is one of the markers of acceptable femininity in dholuo, a dhako moromo (a complete woman). Residing in the marital dala (home) confers prestige on a married woman as a feature of a dhako moromo , and every woman desires to achieve this goal. Conversely, remaining unmarried, migrating to a fishing village or being dispossessed of land increases the chances of chronic insecurity. Within the customary land laws, land for unmarried women is unheard of, because it is expected that every woman should marry and lay claim to her husband's land. Before marriage, the status of a Luo woman is amorphous and transient and she has no rights over any property or land in her natal home. It is marriage that gives a woman permanent identity through belonging to her marital family and gaining permanent residence on her husband's land. Once married, women are required to remain in the marital home to retain land use rights, while men maintain the overall control.
Women who bank on the acceptable femininity particularly of widows, face a dilemma: exercising their rights of land control as sanctioned in statutory laws, or retaining a harmonious relationship with their marital families. Migrating to live in other places, such as in the case of women who migrate to the fishing villages for livelihood opportunities, destabilises their marital land claims despite their marriage status. While the arrangements that ensure women's access to land through marriage are well stipulated in customary arrangements, emerging changes have influenced land relations significantly leading to cases of dispossession and perpetual insecurity.

Restricted rights: how could she do that without consulting the family?
Discussions of illustrative stories of women, as outlined below, revealed a scene of rights to land that are restricted.

Case 1: Akinyi
Akinyi was a widowed mother of two, a son and a daughter. Before her husband died, she had been shown the piece of land that he owned. When Akinyi's son was admitted to Rongo University, she decided to sell part of the land to raise money for the university fees. Since her farm was adjacent to an all-weather road, she quickly found a buyer, who began fencing his land after completing the payment. When Akinyi's brothers-in-law found out, they attacked the buyer, vowed that he would not occupy the land, and demanded that he get his money back from Akinyi, as she had no right to sell their deceased brother's land.

Case 2: Rita
Rita, John's wife, discovered from her neighbour, Ruth, that John had sold land to Ruth's husband. Rita had gone to Sienga fishing village, where she worked as a fish trader, with a new kanga cloth that her husband had bought with part of the proceeds from the land sale. However, she had not known about the sale until Ruth teased her over her new dress: "I can see you are enjoying the land money." This comment naturally caught Rita by surprise.
These stories provoked animated discussions in the focus groups, particularly Akinyi's case, with some respondents offering similar examples that they had encountered. Rita's case was dismissed as normal; they explained that she did not need to be upset about her husband's decision to sell land. Going by their responses, cases like Akinyi's, where widows sell land secretly, are common. The discussion participants agreed that it was the responsibility of the brothers-in-law to take charge of widows like Akinyi, including paying school fees for their children. While noting that male in-laws might fail to shoulder such responsibilities, the respondents felt that it is unacceptable for a woman to sell land without consulting relatives, even if for a good reason, such as school fees. "How could she do that without consulting the family?", one male respondent asked. "She ought to tell the brothers-in-law", another female participant opined. One man angrily said: "Usually, according to our tradition, it's the men that own the land. And if a woman comes with a suggestion that they want to sell the land then you will know who the man is!" In much the same spirit, another respondent asserted that "it's the man who has the authority".
Although it was noted during the discussion that the brothers-in-law were obligated to support her in paying the school fees, scorn was soon directed towards Akinyi herself and not towards the neglect she suffered. The serious implications of Akinyi's choice could be read in the physical expressions of the discussion participants. Men initially shook their heads in silent disapproval; the few who did make statements to the group did so after conferring with their neighbours in hushed tones. Since Akinyi's husband had acquired his land from the lineage, it was understood to remain the property of the lineage, whatever the Matrimonial Property Act says. That piece of legislation protects the individual rights of women over land as equal to those of men. It constitutes a challenge to customary law, which is premised on the collective ownership of land.
Certainly, when the land remains within the lineage, with restrictions on selling it to outsiders, this protects the security of members' resident arrangements. While customary law in cases such as that of Akinyi seeks to protect wider kinship rights, statutory laws seek to protect individual women's rights to own land and use it as they see fit. In cases like this, women are aware that the marital family will resist the idea of selling land, and so they do not disclose their intentions. Even with full knowledge of the provisions of the Constitution, which encourages the elimination of discriminatory laws, women in reality still have to abide by customary law. Women like Akinyi are faced with hard choices, including between conflicts with the marital family and an inability to meet basic responsibilities like a child's education.
The Matrimonial Property Act, which was passed by the Kenyan Parliament in 2013, was to provide specifically for the rights and responsibilities of spouses. It stipulates : "A married woman has the same rights as a married man to acquire, administer, hold, control, use and dispose of property whether movable or immovable property jointly owned and acquired during the subsistence of the marriage according to the contribution of either spouse towards its acquisition, and shall be divided between the spouses if they divorce or their marriage is otherwise dissolved." 27 According to this Act, "contribution means monetary and non-monetary contribution and includes domestic work and management of the matrimonial home, child care, companionship, management of family business or property; and farm work". 28 But there is a substantial gap between this law and popular opinion on the matter in Luoland. In cases like Akinyi's and Rita's, the focus group respondents felt that men generally have the sole right to decide on issues of land. Some respondents nevertheless agreed that there are problems associated with a man disposing of land without the wife's consent. One male respondent stated: "You may get these men selling the land without the woman knowing. He just comes with money and you see life continues, you enjoy life but you don't know where the money has come from and how you will live afterwards. But then you know too late, from here he goes [to sell the] next." [Male focus group discussion (FGD) participant].
Although the general feeling from the discussion was that there is no need to consult one's wife on land sale decisions, a negotiated solution was proposed. "They should sit together and agree", a respondent asserted. "Maybe a man may use up all the money on changaa (alcohol) and other women if the wife is not aware", a woman added, citing a case where a man sold land and ran off with another woman only to return later, broke and sickly. Yet such negotiation was seen more as a solution to avoid disputes, rather than as a concession to the requirements of national law.
Matters are made more complicated by the fact that Article 60(1)(g) of the Constitution of Kenya requires "an encouragement of communities to settle land disputes through recognised local community initiatives". 29 Among the Luo, the local community initiatives include the village elders and chiefs. The elders are appointed from every village and constitute a council representing various clans. The chief and sub-chiefs are administrative officials appointed by the Office of the President. They have physical offices that are situated in specific locations and sub-locations. The elders, chiefs and sub-chiefs hold regular barazas (meetings) where land disputes are heard and arbitrated.
Focus group participants themselves pointed out that women in cases like that of Akinyi should report the matter to the chief or the elders. However, other respondents noted that since the chiefs and elders operate within the terms of customary law, women like Akinyi would be found to have dishonoured the accepted arrangements and would therefore face disciplinary action. Female respondents, particularly, felt that the barazas are not favourable for them as the majority of the chiefs and elders are men. One woman stated that "as women, we are only collected to go and put signatures at the chief's office but sometimes we are not". She was referring to the ways in which women are used to endorse decisions that have already been made, at best, and sometimes not even that, but are not involved in the actual process of decision-making.
Women have a range of attitudes to customary norms and institutions, but they still have to accommodate themselves to them. Of the six focus group discussions that I conducted, one was a women-only group in which 13 of the 17 women were widows. After I read out the stories of Akinyi and Rita, there was a long silence, and then the women began sharing their own related stories. From where I sat at the end of the semicircle during the meeting, I noted that one of respondents appeared impatient as others were telling their stories. When I asked her to say something, she rose and began to speak: "If you are a good lady who has been taught well, after your husband dies you have your brothers-in-law whom you should consult if you want to do something. Because the time you are selling that land you have to explain to them the reason why you want to sell that land. If I were Rita, I would prepare him chai mopoto (thick tea with a lot of milk) so he can tell me more about that matter of selling land. I will ask him why he sold the land." (Female FGD participant).
"A well-taught lady", according to her, would not be like Akinyi. The lady who spoke appeared to be older than the rest in the group, and it was with the resulting status that she adopted a reprimanding or disciplining tone in her response. Giving her own example, another woman pointed out that women do not speak out because the land belongs to men: "My sister's husband borrowed money from a sugar businessman. My sister was not aware about this matter. He couldn't pay so the businessman came for the land and my sister was given three days' notice to vacate. My sister had nowhere to live but found refuge in a church. She did not ask him anything because it was his land. So there is nothing we can say or do." (Female FGD participant).
Still another woman said : "Even if you are not happy, no one thinks you have a right to ask. Just keep quiet." The tone of this focus group discussion changed from discussing the cases that I had shared to the participants' own cases and those of other women whom they knew. It was clear that women, and especially widows, face many constraints when it comes to accessing and controlling land. It was also evident that the causes of these constraints are interpreted differently depending on the status of the women. Widows who have adult sons have more authority and are likely to have more negotiating power compared to those who have no children or have younger sons, or those who have only female children. This is why one of the respondents opined that some women face problems because they are "not taught well". Her reprimanding tone had to do with her own position, as an older woman who had sons and daughters-inlaw, as I came to learn. Her position was more secure, and she thus acted as a defender of her sons' rights. By doing so, she also guaranteed her own security. Her opinion on the woes of younger widows was grounded in notions of "well-taught" women, and involved rebuking those she considered as deviants. For women with status, customary laws that disadvantage women overall may nevertheless be seen as proper.

Corrupt arbiters: "When I went to the chief, he asked for a bribe"
The discussions of the following two stories in the focus group meetings revealed the frequency with which widows are dispossessed of land and how local arbitration is compromised by corrupt authorities.

Case 3: Aoko
Aoko, was a 26-year-old widowed mother of four. Upon the death of her husband, Otieno, Otieno's brother, Onyango, took in Aoko through the levirate custom and she bore him one child. Onyango had been a fisherman in one of the fishing villages but had returned to the ancestral home due to unreliable and perilous fishing work conditions. Onyango and Aoko had frequent fights as he complained that she did not cook him good food. The fights were frequent but she would nurse her injuries and continue with life. But one day Onyango beat her up seriously and threatened to strangle her. Seeing that her life was in danger, Aoko moved to live in a fishing village in a rented room. While there, she met Noaz, a fisherman with whom she lived for a short time, but life became difficult in the fishing village, prompting her to return to her late husband's home. To her shock, she found that Onyango had sold all the land including even the portion where her house stood.

Case 4: Atieno
Atieno, a mother of three, lived with her husband, Okinyi, in a village near Nyagina fishing village. Their house was known for frequent fights and Atieno's screams, especially on the nights when Okinyi was on a break from night-time fishing duties. He would drink in the nearby local breweries and then Atieno's yells would follow. She could not bear his beatings and, fearing for her life, she decided to migrate to another fishing village away from her marital home.
The focus group discussions about these stories quickly moved away from the stories themselves, as participants brought in a wide range of their own experiences. In the discussions three solutions were offered as ways to mitigate problems of this nature, although they were not simply agreed upon. The respondents offered: "If I were Aoko, I would take the case to the chief "; "I would convene a family meeting to arbitrate"; "Let her go to the elders as they should follow up her case to get her land back". From these responses, it was clear that the option of seeking arbitration of the elders and the chief was well-known as being available. As is customary, arbitration by elders and chiefs involves reporting the matter to them, followed by a hearing in a village baraza (gathering). However, the difficulties of women's participation in village barazas were pointed out, as one man said: "They don't listen to women actually. It becomes very difficult for a woman to appear before the baraza." Besides, as I had noted during a focus group discussion that I held with village elders, women were a minority and often quiet. For instance, in one of the focus groups comprised of village elders, there were two women out of a total of eleven in the group. I noticed that whenever one of the ladies spoke, the chairman would summarise her opinions, sometimes completing her sentences. Eventually, they did not contribute, choosing to nod as the conversations progressed.
From discussion of Aoko's and Atieno's stories, key issues emerged: violence towards women, threats of dispossession or eviction, and the corrupt nature of local arbitration mechanisms. It was reported that sometimes these elders and chiefs act as conduits in dispossessing women by allying themselves with wealthier land buyers. They also ask for bribes to listen to cases, thereby making it difficult for women who cannot afford to pay. One widowed woman offered her story: "Aoko's story is the same as my case. My shemeji (in-law) sold land while I was still living there. When I went to the chief, he asked for a bribe. I didn't have it so I had to move away and live in a small rented house in the fishing village. Then one day I decided to go to the councillor who persuaded the chief to intervene. Shemeji was threatened and he allowed me to stay even if he had sold most of the land." (Female FGD participant).
Corruption thus adds to the frustrations of women who seek help from chiefs. Another woman reinforced the point by underlining the indifference of the authorities: "As a mother of ten children I had gone to the chief because my husband said he did not want me anymore. Then one day he moved away with the second wife and left me on the land that he had already sold. The buyer came and evicted me forcefully, destroying my house. I had nowhere to go. The person who bought my land is a rich man and he does not care. He has not even used the land he bought even if he evicted me forcefully and in a hurry. The chief's offices are not helpful. Even the other in-laws do not care about my children and me." (Female FGD participant).

Dispossession and limits of family obligations
The discussions on the Atieno and Aoko stories further revealed cases of dispossession and inability of families to intervene. A widow whose only son had died lamented: "I had only one son who died and then the family pushed me away from the land. The elders knew this and no one did anything." In one case, a widow still faced dispossession even after she had saved her own money to buy the land that she occupied: "My land was taken away when my husband died. When my husband's grandfather gave me land it was also taken away when he died. Then I bought a piece of land with my savings. There is a problem because they want me to tell them to whom I will give that land when I die because my only son died. I want to give it to my daughters but seeing how hostile my husband's relatives are, I will just sell it." (Female FGD participant).
In Aoko's case, the respondents generally felt that according to the custom, Onyango had erred and that Aoko had full rights to live on her late husband's land. This is an indication of the limitations of customary law and the failure of arrangements that should grant security to widows like Aoko. Although customary law protects Aoko's rights to live and hold the land for her sons, it is clear that the mechanisms for enforcing such entitlements are weak in cases where a widow leaves the land. Migrating from the land leaves the widow at greater risk of losing it to other interested members of the lineage. Migration therefore becomes an "open moment", 30 a point of confusion or flexibility in practice, which here leads to dispossession. This scenario is not unusual. Speaking of women and succession in Sub-Saharan Africa, Kuenyehia observes : "In the past there existed certain sound underlying assumptions for what now seems to be the lack of protection for widows. By custom it was the legal responsibility of the customary successor to maintain the surviving widow and children out of the estate of the deceased. This obligation was discharged with all seriousness by customary successors. In this present individualistic age, however, the interests of the widow and children are more often than not subjugated to the personal interests of the customary successors. It is not uncommon to find widows who have been thrown out of their matrimonial homes upon the intestate death of a husband." 31 While a somewhat generalising view, and one that paints a rather idyllic picture of the past, Kuenyehia's observations fit with what we see among widows today. In Zimbabwe, Matondi observes something similar: "There were cases recorded by the Ministry of Gender in Mazowe where men have 'chased' their deceased brothers' widows with the intention of grabbing the land and other assets left by the deceased. Widows who had not been married long and those who have young children are particularly vulnerable to being 'chased', whereas those with adult sons are in a stronger position to defend their right to the land." 32 What we see in Uganda, Zambia, Ghana and Zimbabwe, in the works of Kuenyehia and Matondi, is the effect of weakened or compromised arbitration systems. This is also true in Luoland, but in addition, women's marginalisation is exacerbated by acute economic instability. The Constitution of Kenya endorses the use of local arbitration mechanisms to resolve land disputes. 33 Clarification of the role that the elders and chief as custodians play when deciding upon land subdivision is still needed. However, I found that corrupt dealings, such as those reported by dispossessed widows, tarnish their authority as women's hopes for effective mediation are dashed. Furthermore, livelihood difficulties put greater pressure not only on land, but also on how these local arbitration mechanisms actually operate. For instance, bribes are one of the means through which local authorities attempt to meet their own needs, albeit in ways that disadvantage others.
Another significant finding relates to the limits of family obligations. While a brother-in-law is in theory obligated to care for his brother's widow, the reality is that this ideal is often not met. Even when it is, the resulting arrangement may be burdensome for the widow. The inability to fulfil kinship obligations is largely a result of social breakdown and livelihood difficulties. For instance, Onyango subjected Aoko to physical violence due to her inability to provide him with "good food". This is a typical example of a failed obligation that in fact reveals the difficulties faced by both parties: Aoko was unable to cook "good food", while Onyango was unable to provide the means of acquiring it. Onyango's own work as a fisherman had ended, after having lived in one of the fishing villages but later returning to his ancestral home when fishing work became difficult.
During the discussions, it was generally agreed that Aoko had the right to use and live on her late husband's land and that Onyango should be questioned for selling this land. "That man had no rights over the land. The land belonged to her husband and her 31 Kuenyehia, Akua "Women, marriage, and intestate succession in the context of legal pluralism in Africa". children so that even when this man sold the land, he just did it blindly" one respondent argued, noting Aoko's rights as a mother of sons. Notably, Aoko's hope of repossessing her land was linked to her male children, as another male respondent made clear: "People also tend to fear male children. Another thing that Aoko could have done is to wait until her children come of age, then when they go back there, those people will automatically fear the boys, and they will render themselves weak, and that's how Aoko will possess the land." (Male FDG participant).
Another respondent added: "She should wait for the children to become men because these children are owners of land and a woman just comes back because the children are the owners of land. So, you know that should she go to the government, Onyango will be in trouble." (Male FGD participant).
The discussions underlined that Aoko's chance of getting justice was augmented by the fact that she had sons. On further inquiry as to what would happen if a woman did not have children, a respondent said that "if she did not have any children, she should go elsewhere". "Going elsewhere", as I learnt from more probing, meant remarrying. The respondents proposed that women in Aoko's or Atieno's situation could find a solution by marrying again, a sure way of accessing land from another man. "She should look for another husband. Marry and marry again until you get land through a husband; even women know that they will be married somewhere", a male respondent suggested and others agreed in unison. This may explain why women who find themselves in situations like that of Aoko often move on to other relationships with the hope that they will endure and lead to marriage, and subsequent access to land and a dala (residential home).
Beyond sharing Aoko's story in the focus groups, her case became an important one for me to follow during the fieldwork. It revealed that a friend had temporarily accommodated her together with her four children. When I asked her about the option of going back to her father to seek refuge, Aoko looked at me blankly and just shook her head. According to the Luo customs, it is shameful for a married daughter to return to her father's home. "I would not go because I would even be a bad omen if I died in that place", she explained. It is expected that a female child only temporarily lives in her father's home, waiting to go to her own husband from whom she receives land. When women like Aoko are rendered landless despite having been married, they are faced with a choice about whether to forge other relationships, despite the risk of further physical abuse.
Article 60 (1) of the Kenya Constitution requires non-discrimination along gender lines in matters such as land, as noted above. This provision particularly speaks to customary laws that hold that female children cannot receive a share of their father's property, as they are expected to marry and access land elsewhere. The spirit and intention of this constitutional provision is to protect vulnerable women who find themselves evicted from homes and from their sources of livelihood. One implication is that it holds out the option of women getting land from their natal homes. In reality, that option is in conflict with customary laws and norms. During the discussions, one male respondent suggested that "Atieno's father should call her and give her land to settle", amidst shouts from others who retorted that "returning to her father to seek refuge is unheard of ". One man irritably explained: "If my sister is given land to settle by my father, her children will take away the luck of my children. So, I won't allow that to happen. I can only help her get land elsewhere." This was followed by laughter from the rest of the respondents. "If a woman dies in her father's compound, she will bring bad omens to her brothers", a male respondent insisted, to murmured agreement. The association of returnee women with bad omens, which may have adverse effects on a brother's children, leads women to dread seeking refuge with their fathers when they are dispossessed. No wonder Aoko responded with vehement disapproval when I asked her whether she would consider returning to her father's home. Moreover, as the participants indicated, such a return signals failure and shame on the part of the family of the woman. They are perceived to have raised a stubborn girl who fails to sustain a marriage. This explains why women would rather choose to migrate to the crowded fishing villages or to urban slums. Since a returnee woman's presence is associated with shame and with bad omens, what natal families often do is rent a house elsewhere for her or support her in paying school fees for her children. In the discussions, most men said they would offer to contribute to school fees for a sister's children, or pay her rent, rather than allow her to live near their land.

Threatened masculinity, violence and land
"If I tell her, go, get out of here, get out of my compound, she will just cry and come back to plead but if she has land somewhere else, she will just go." These were the words of a male participant describing how land ownership by women may change relationships between men and their wives. In the discussions it emerged that wife beating is one of the means by which men exercise their authority and control. Research in the region attributes violence against women to a crisis of masculinity. Izugbara, 34 Amuyunzu-Nyamongo, 35 Francis, 36 and Silberschmidt 37 all allude to men's tendency to compensate for threatened manhood through physical violence towards their partners. Actually, in an environment where men's authority as providers is challenged, land ownership remains a significant way to assert male control. The prevalence of wife beating was well articulated by participants during one of the focus group discussions, which was conducted among male and female traders in the fishing village: "Researcher: I want to hear more about ngoyo dhako (wife beating).
Male respondent: In the past men used to beat their wives. That's their right. I mean, it happens.
Another male respondent: Even these days they are beaten. On further inquiry about wife beating, I learnt that this is one of the means of maintaining control. Ownership of land by a woman is unimaginable as men generally see their exclusive ownership of land as key to the authority they have as men over their women. The ownership of land by women threatens men's existing control, hence the insistence during the discussion of wife beating as a man's right : "They are beaten, yes they are", asserted the male participants. One man further underlined why it would be unacceptable for women to own land: "It is not acceptable for a woman to own land because if I beat her, she will just leave and go to live where her land is situated. If she knows she has somewhere else to go, she will not respect me." (Male FGD participant).
Men's respectability is attached to their ability to remain as the sole owners of land and for the women to remain dependants in that regard. Many echoed the fear of losing respect and manliness. On whether as a man one would allow a wife to own land, animated debate ensued, with some men stating: "I can't chase her if I want"; "It is tormenting for men to imagine that wife has some property"; "If wife owns land, you are not somebody". Another one angrily asked : "[W]hat kind of man would that be?" Men are aware that even when they beat their wives, it is unlikely that they will run away. One man stated: "[I]f I beat her and chase her away, she will just cry and plead to stay but if she has land elsewhere, she will just go." Another one added: "Traditionally, whenever there is a dispute between a man and a woman, a man will chase his wife away, they will tell them to move out of the compound, or move out of the house. But according to that land law [the Matrimonial Property Act], it is clear that I cannot chase my wife out of that land." (Male FGD participant).
To women, on the other hand, in the context of poverty and the prevailing ideal of femininity, dhako moromo, which defines their value according to the ability to stay married, the thought of eviction by a husband is terrifying. A woman would rather endure violence than face the reality of homelessness and shame. Even women respondents in the discussions were of the opinion that owning land may bring problems with their husbands. One offered: "If a woman owns land it might make them refuse marriage or bring problems to their husbands so this matter for me is hard." The existence of customary law that excludes women from the ownership and control of land is not the only reason for the poor implementation of new land laws like the Matrimonial Property Act. The preservation of gendered power through threat and violence points to another reason why the new laws are not taking effect in ensuring equity in land access. As other studies have noted, customary law does not essentially hinder statutory land law, but the latter does threaten or interrupt existing power structures. 38 The Matrimonial Property Act, for instance, amounts to the end of sole male ownership of land by stipulating joint spousal ownership. When it comes to land issues, customary law affirms men's authority and sole ownership of land while statutory laws do the opposite.
From the discussions, women's land rights were overwhelmingly linked to marriage. Regarding the options available to women who are not married or who separate from their husbands, focus group participants proposed resorting to marriage and remarriage. "Marry and marry again" was the refrain. On whether the statutory laws would be helpful to support widows who are dispossessed, the respondents insisted that this avenue is not feasible. "These laws deny sons their rights and once a woman is married, she has a right to enjoy everything including land where she is married", a female respondent opined. On fathers giving land to daughters who are not married or if they separate from their husbands, there were mixed reactions. One response by a male participant was: "No, not easy unless she fails to marry or maybe I would buy her land elsewhere instead." Others felt that they would think differently about their own daughters. One man asserted: "When it comes to matters of my daughters, it becomes very difficult. Sisters may be a bit far, but now my daughter, it's very difficult, it's very personal and I think even if I can't find a piece of land for her in the ancestral land, then I can even decide to look for a way to settle her. If I have land, as a father, I'll have full rights or full control over that land and I'll be the key decision maker over my property. So, when I tell my children that, OK, so and so you'll have this portion, then no one will question that." (Male FGD participant).
Men in Luo patrilines still largely own land collectively. Indeed, in instances where land is not yet subdivided among the male relatives, it is more difficult for individual men to make decisions regarding the land, including decisions about co-ownership with their spouses or bequeathing it to their daughters. However, even that idea of men thinking differently about their own daughters provokes a threat to other men, as one male respondent stressed: "Usually, according to our tradition, it's the men that own the land. And if a woman comes with a suggestion that they want to sell the land then you will know who the man is!" (Male FGD participant).
"Know who the man is" is an assertion of the man's position. "It's the man to give the authority", another respondent asserted. Land ownership affirms "who is the man" and therefore stands as an expression of authority and definition of proper manhood. This is particularly the case in a setting where there has been a general decline in livelihood means such as was demonstrated by men like Onyango and other fishermen whose work is uncertain in their villages of origin and also in the fishing villages.
Men's efforts to safeguard their authority and control over land by resisting women's claims was revealed by a final case discussion. This story was about a married woman whose father had given her a piece of land.

Case 6: Mara
Mara was married to George and they lived together with one of George's sisters, Lily. One day, Mara's father phoned, informing her that he was planning to give her some land as he was sharing his property among his other children. When Mara excitedly shared the news, Lily said that accepting such a gift of land would shame her brother: "why do you want to shame my brother like this?" she asked Mara.
The reason for Lily's response was discussed. This story elicited heated discussions that alluded to the undesirability of a married woman's access to land in her natal home. After all, proper masculinity is anchored in sole ownership over land. One respondent stated: "It's tormenting for men to imagine a wife has some land", expressing what he saw as an abnormal situation. This feeling, as I learnt from the discussion, did not imply that women cannot own land, but it did reveal how men might feel with regard to such rare cases. As one man said, referring to a man who would allow his wife to own land: "That can't happen unless he [the husband] is crazy, is he really a Luo man?" These sentiments were also shared by women. Indeed, in the story itself, Lily reasoned that Mara's ownership of land would affect her brother's reputation as a man. Most focus group participants dwelt on Lily's reference to shame, suggesting that a woman's acceptance of land from her natal home would minimize her husband's respectability. Furthermore, the perceived threat to masculinity was also attached to women's access to some of the power which men hold through land ownership. As one man asserted: "If a woman owns land, she will become superior and have a say on certain issues that they are not traditionally supposed to." The second significant marker of proper masculinity that was revealed through the discussion was control and authority over one's wife. This is anchored in bridewealth payment, which transfers a woman from her natal home to her marital family. For a man, bridewealth payment justifies a sense of ownership over his wife, and authority over his home. It is for this reason that, in cases like that of Mara, a woman who accepts a gift of land from her natal family weakens the husband's authority. This is because, as one focus group participant argued and others agreed, "a woman's family expects that a man is to give them wealth not to take it from them". Partly, paying bride wealth is supposed to demonstrate the man's ability to care for his family. It was because of this that the participants felt that a man allowing his wife to take a gift would show his inability and thus his inadequacy, and hence shame him.
In addition, the respondents were of the opinion that when women own land, their marriages would be unstable due to lack of respect that may arise from such a woman: "You know there will be problems in that marriage because of lack of respect from the lady towards her husband", one male respondent explained. This is important with regard to men's sense of authority because marriages are considered a measure of proper masculinity and femininity.
In an environment where people are faced with economic and social fragmentation, a significant marker of proper masculinity, that of providing for families, is now hard to achieve. Men are therefore left with the sole ownership of land according to customary laws as a pre-eminent mark of proper masculinity. When women's land claims, and the laws that sanction egalitarian values, threaten this land ownership, hegemonic masculinity is placed under further pressure.

CONCLUSION AND RELEVANCE
In a context of livelihood instability and legal pluralism, access to justice is threatened leading to marked vulnerabilities. These problems are exacerbated by an already difficult socio-cultural environment and by family fragmentation. As this plays out, a lack of awareness of, and resistance to, laws that protect rights, corrupt local arbitration mechanisms, and threats to notions of masculinity complicate the situation for women. This article reveals a series of strategies of women positioning themselves for relative security in a context of multiple vulnerabilities.
These strategies are what I call options of refuge, and as illustrated in the discussions, they turn to be risky options of refuge. Migration to fishing villages, customary rights of married women, and banking on egalitarian statutory laws present the options of refuge. Women who face violence migrate to fishing villages away from their marital homes. This migration exposes their land and homes in the marital place to possession by interested relatives. Migration is construed as deviation from acceptable norms of femininity, which attracts denigration of female migrants who settle in fishing villages. Indeed, the dispossession of marital land is justified on the unacceptable notions which women's migration signals. While the unstable livelihood conditions require women to migrate to the fishing villages, this move places them at greater risks.
Women also anchor their security on customary rights of married women that only act as ideas of refuge with no enforceable benefits. The research showed that rights of residence as a married woman can be threatened and used as grounds for men's assertion of control. This is particularly so in a context where men's sense of acceptable masculinity is threatened. Violence on women remains one of the means of asserting authority. Married women, particularly widows, are dispossessed of land even though they have rights of residence in customary terms. Although customary laws protect women's rights to live on their marital land, the mechanism for enforcing such entitlements are weak in many cases especially for women who temporarily migrate to fishing villages. Arbitration in case of disputes disadvantages them due to corrupt processes and marked patriarchal leanings.
Similarly, provisions of the egalitarian statutory laws have not succeeded in offering women security of tenure, access, or control of land. These women face a dilemma: exercising their rights of land control as sanctioned in statutory laws, or retaining a harmonious relationship with their marital families. Yet multiple legal codes cause confusion and conflict. The multiplicity of legal codes presents more problems than advantages partly because of the extent of everyday livelihood vulnerabilities. One aspect of this vulnerability is the response of men to women's land claims, which are seen as challenges to an already precarious hegemonic masculinity as men struggle to meet their obligations as providers. While statutory laws outlaw discrimination on matters of land, resistance to these measures is prevalent partly because of how egalitarian ideals destabilise men's self-understandings. Marked gender inequalities anchored in customary arrangements are further given a new lease of life by the livelihood vulnerabilities that have affected both men and women. Women's hope for security in land is dimmed, and as injustices, violence and social dilemmas follow, a scene of "risky refuge" takes root.
The strategies towards realisation of Strategic Development Goal 5 on gender equality and the broad Goal 3 on ending poverty in all its forms everywhere will be dependent on a continuous review of how people's agency mutually reinforces vulnerability or progress. As a natural capital upon which many people are dependent in a poverty-stricken place, land plays a crucial role as a significant resource. As such, women's livelihood strategies, such as land claims, cannot be understood outside of other significant actors, particularly men. The legal frameworks that shape and regulate land use, rights and control operate within a social context where other sanctions exist. These intersecting factors produce particular implications that require scrutiny before the rolling out of any interventions towards gender equality or the fight against poverty.
The development sector operates in contexts of multiple players. Empowerment interventions need to factor in the effects of other players, such as the social structures, particularly, gendering practices. This study contributes to an in-depth understanding of the intricate context within which people set their strategies of survival. Furthermore, the realities of enforcement of already existing egalitarian land laws are brought to light with possible implications on what needs to be done for these laws to actually achieve their intended positive outcomes. Processes of enforcement of egalitarian laws are critical for positive outcomes of those laws. Furthermore, poverty mitigation and justice programs in sub-Saharan Africa require current empirical research on how land laws, gender relations and livelihood vulnerabilities produce mutually reinforcing effects on development.