2. Social Security and Weak States

Rather than limiting the concept to state-provided protection, as is the usual view, I take a more integrated stance to the ways in which societies organize care and protection. A broad conception of social security includes all mechanisms by which a society intends to ensure that people who cannot take care of themselves can have an appropriate standard of living. While, in principle, every society has such mechanisms, the corresponding resources and mechanisms actually devoted to facilitating equitable access and distribution show great variation. Social security is always a matter of degree. It may be more or less institutionalized and institutions may play a stronger or a weaker role in providing social security and may provide a higher or lower level of security. And some societies are in such disarray that social security is hardly discernable at all. How much self-care is expected and what is considered neediness, and what kind and degree of care is appropriate, who is entitled to it, and who has the obligation to provide care—the degree of security that existing mechanisms are expected to provide—vary across societies and over time. An integrated and relational approach allows us to study care and protection under conditions in which the state is not the primary or most important provider of protection and to make comparisons including countries that do not have a welfare state, as well as other polities. It enables understanding how different forms of protection are organized and affect each other and ultimately affect personhood.

Social security is constituted in relationships that form part of circles or networks of support. With changing moral conceptions and under changing social and economic conditions, circles of support and care relations within those circles change in character, and obligations and rights are redefined. One result of current globalization is that global economic relations affect local forms of social security that are situated widely apart. Social security is an important motivation for migration that in turn has a deep impact on networks of solidarity. Shifts in identification with kin groups, religious communities, as well as new social circles that are being developed in the diaspora have important implications for conceptions of selfhood and the construction of significant relations.Footnote ¹² Shifts in conceptions of need often indicate changes in gender, age, social status, and class relations. Studying the organization of support therefore generates insights into changes in social organization, including kinship structures and intergenerational links. But it also throws light on processes of identification, on religious organization, and on broader governance patterns beyond the national borders.

State-provided social security has greatly contributed to social peace and the reduction of poverty in Western welfare states. This was a reason why poor countries were pressed in the twentieth century to introduce similar social security systems. However, as Midgley,Footnote ¹³ Mesa-Lago,Footnote ¹⁴ and recently TownsendFootnote ¹⁵ have shown, such systems tend to have counter-productive effects. Under certain conditions, being limited to civil servants, lacking administrative capabilities among the poor rural and urban population, and a failing taxation system, state-provided social security has a redistributive effect from the poor to the wealthier. These conditions are typical for poor, weak, and predatory states.

Indonesia, though officially no longer a poor country, still has a large poor population. The poverty rate is falling, but 11.4% of Indonesia’s population and 14.3% of its rural population still live under the poverty line.Footnote ¹⁶ Yet, it spends only 1.2% of its gross national product (GNP) on social insurance and social assistance. And although a large percentage of the population is covered by some health insurance, the system functions so poorly that additional payments are necessary and many have to rely on relatives to incur their health expenses. In principle, Indonesia is wealthy enough to raise the level of social expenses, but its malfunctioning taxing system prohibits such a rise in the state budget, because it is fragmented, inconsistent, and lacking proper administration and monitoring functions for its implementation. Besides, environmental disasters pose a serious threat to the maintenance of Indonesia’s natural resources that often are an important resource for social security. There is a general lack of trust towards a government that has been thoroughly corrupt.Footnote ¹⁷ Project-based governance had been the hallmark of Suharto’s regime, who used infrastructure and social security projects to garner political allegiance. Under President Megawati, the government launched structural programmes to eliminate poverty, but the presidency of Yudhoyono returned to project-based funding.Footnote ¹⁸ Such projects have been notoriously unsustainable and erratic, and fraught with corruption. For most people in rural areas, a sense of security of any kind is therefore hardly associated with the state. It is too early to know what direction the current President Widodo will take, whether tax collection will improve, and how successful his anti-corruption policies will be.

Many countries in Southeast Asia are in a similar position, though not all have the vast amount of natural resources that Indonesia has and the specific constellations may differ. What they share is that the state is only a minor provider of social security. For the rural population, other sources, mobilized on the basis of local norms within kinship and local relations as well as support organized by institutionalized religion, are key providers of social security.

3. Social Security in a Moslem Village on the Island of Ambon

The Island of Ambon, in the East Moluccas, has a population that consists largely of ethnic Ambonese, half of whom are Christian and the other half Moslem. The villages are largely religiously segregated; only a handful have a mixed Moslem and Christian population. Before the violent revolts around the end of Suharto’s regime,Footnote ¹⁹ there was a sizeable group of Butonese migrants from the south-eastern tip of Sulawesi in the rural areas. In Ambon town, the capital, there are old Indonesian–Chinese and Arab communities and some smaller more recent migrant communities from other areas of Indonesia. Typically, Moslems relied on a combination of customary law and Islamic law, as well as on state law for their social security.Footnote ²⁰ Each of these normative orders entails specific conceptions of personhood, namely how persons were expected to take care of themselves, on what was defined as need, and the social responses to need. These rules were, however, interpreted in specific ways. The state had adopted the International Labour Organisation (ILO) approach to social security and need. For example, the ILO, premising its policies on the assumption of a nuclear family consisting of a male breadwinner and a female carer for dependent children, defines widows in principle as deserving financial support. In Ambonese villages, a gendered division of labour prevailed, but this did not follow the ILO logic. Households usually contained more than a nuclear family, and women did participate in economic life. Therefore, not all widows were needy according to customary law. Only those without land and without close relatives were considered needy. This difference had important implications. In the 1980s, the Indonesian state began to launch social projects in rural areas, targeting disabled persons, orphans, and widows, following the ILO logic. These social projects were aimed at all widows, but in the villages they were taken to be meant for poor widows only. Wealthier widows were uncomfortable because they did not want to be thrown into one basket with poor widows. They did participate in such projects, but mainly to support with administrative tasks that the poorer and poorly educated widows could not perform.

Zakat fithra was an important religious form of social security, by which more money was redistributed than by state-provided social security. Here, too, adjustments were made to the local situation. The categories of deserving persons were redefined as to include persons who, according to customary law but not to Islamic law, needed support. The state was attempting hard to capture the zakat funds to use them for state-provided social security, and has intensified its effort since. So far, only civil servants, for whom the zakat mal is deducted from their salary, are paying into the zakat institutions. The state sees this as an excellent opportunity to raise funds for its social security programmes, and to demonstrate that they are taking this obligation seriously while also showing that they take religion seriously. But many people have a different view. In the 1980s, people in Ambon were adamant that it was not the business of the state to collect and redistribute their zakat, which they regarded as a personal obligation towards Allah.

Kinship was and still is the main source of social security. The strongest obligations are among a small circle of close relatives of different generations. Beyond relations of parents, children, and grandparents, it is not taken for granted that support is automatically extended, though this is not always followed in practice. It is for this reason that the poorest and neediest are those who lack the material basis for a living such as land or an income as well as close relatives. They have to rely on relationships in which the provision of care is a weaker, more conditional obligation. Wider circles of support need careful maintenance. Much of social cohesion consists of creating and maintaining—or severing—social exchange relations meant to secure future relations of support. Similarly to what Thelen, Thiemann, and RothFootnote ²¹ have shown for kinship-based social security in Serbia, here, too, kinship not only is a mechanism to provide care; care relations often become defined in terms of kinship. In other words, care and support generate kinship, even among relative strangers. On Ambon, higher-ranking civil servants would use their position within the state to provide support to needy relatives or co-villagers, acting as mediators and patrons to secure state-provided support for those who lacked administrative skills. But they also provided support with what we have called “off-state” activities,Footnote ²² setting up social projects quite similar to those of the state, but with their private funding. They could do this on the basis of the authority they derived from their position within the state administration, as judges, as teachers, etc. The official village co-operative also functioned as a mechanism of social security, both in its official and its unofficial incarnation. This generated an interesting attitude towards the co-operative. Seen from the perspective of a formal state-like institution and measured against the law of co-operatives, it was a thoroughly corrupt enterprise in which considerable revenues were siphoned off to the exclusive use of those members that had close relations to the leaders, instead of distributing them to all members as the law prescribed. Those who were excluded from these advantages complained bitterly about the corruption and, as a formal institution, the co-operative had a very bad reputation because of the resources used for private purposes against the official regulations. But seen as a system of kinship obligations, the head of the co-operative received much approval, because he used the revenues not only for himself—which he did—but also to support his poorer relatives within and beyond the co-operative. It served as a successful and respected kinship system of patronage in which kin had a very broad meaning. Many became members because of their kinship relationships with the leaders, and had to perform additional services for them in order to remain within the patronage system. Very few had become a member because they believed in the co-operative in its official form, and they were disappointed when they discovered that it did not function according to the official rules. Others had become a member as a relative of persons that were not elected as leaders and therefore could not deploy the system to strengthen their network. They were the ones who complained most bitterly about the corruption.

The village also had a private fishing co-operative set up by an individual fisherman in which he provided work for needy relatives of his crew. This fisherman had in the past been a civil servant, but had returned to the village and became a fisherman. He saw it primarily as his religious obligation to support poor relatives. This co-operative was not subject to the law on co-operatives. It was clear who was the boss, and the co-operative did not suffer from the kind of exclusion that we saw in the official co-operative with the Janus face of being a state agency and a private patronage system.

These social projects and private co-operatives are fragile and tend to depend on individual persons who take the initiative. Once these move away or pass away, the network easily falls apart. In this respect, such off-state projects differ little from state-provided social security, which people experience as at least as fragile. Most villagers have long been accustomed to the fact that all support and care, whether from the side of the government, villagers, or religious institutions, is short-lived and can be withdrawn at any time. Villagers generate a certain degree of security by diversifying, by engaging in a diverse range of networks and care relationships that characteristically include unequal power relationships of patronage, like that of the village co-operatives and the projects of the higher state officials. By doing so, they develop an ambivalent attitude towards the state. On one hand, they resent that the state does not offer enough support, but they take without qualms what they can get for as long as it lasts. Relying on patrons within the state-mediated services, some manage to profit from the state in an indirect manner, but this is often seen as kinship support and not state support. The state in its depersonalized form is deemed unreliable and corrupt, and is approached with deep suspicion and distrust. But, to the extent to which it can be redefined and mobilized in terms of kinship, it becomes acceptable and is more reliable than an impersonal state institution.

The examples also show that the same institution that serves social security is often viewed from the perspective of different normative orders. The village co-operative was a state-like institution regulated by the Law on the Cooperatives and at the same time it was a kinship system following the customary rules of “kinning,” of generating and maintaining meaningful relationships in terms of kinship.Footnote ²³ Likewise, zakat fithra was considered a religious obligation, to be paid to categories as prescribed by the sharia, but in practice it was preferably given to a poor relative. It was rare that someone paid zakat to a person with whom he or she had no kin relations. And, if this happened, it was part of a process of kinning, of creating kinship relations with an unrelated person, or it concerned the customary midwife who customarily received the zakat of a child during the first three years of its life. The state itself was also keen on mixing normative conceptions of support by establishing zakat institutions. It could demonstrate taking religious obligations of the Sharia seriously by extending state-provided social services according to state law. An important reason why there was so much opposition against official zakat institutions was that people would lose control over the decision as to who deserved support and would no longer be able to use it for kinning purposes to maintain their networks of support.

The main material basis for social security on rural Ambon, as it is in rural areas elsewhere in Southeast Asia, is land. Land grabbing, expropriation, and land degradation all have a deep impact on social security in many regions, but these were not pressing issues on Ambon. Among Ambonese citizens in rural areas, there are few who have no access to land at all, though some have very little and not all can actually use the land they theoretically have access to. This may be because the land is too far away from where they live, or because they are too old and weak to work their land. These people need support in terms of work, but not in terms of land rights. Private property as well as different kinds of access to family land or clan land is inherited bilaterally. Here, again, there is an interesting amalgamation of different legal systems. When asked about the rules of inheritance, people will usually answer that they apply Islamic law, explaining that all children inherit equal portions from the father’s and the mother’s side. And they would add that this is their customary law—their adat. The majority of the population has no idea that their interpretation does not follow even the basic principles of scholarly legal doctrine. To them, it is Islamic law and at the same time their adat.

The village Hila, the village where we did research in the 1980s, had a small Christian enclave among an otherwise Moslem population. This Christian community had been established during the early colonial period of the Dutch East Indies Company VOC, to defend the Dutch fort. Relationships among Christians and Moslems were somewhat distant but cordial, and many were part of some cross-religious caring networks, though most support was extended among people of the same faith. With the revolts at the turn of the century, this changed dramatically. Bloodshed was avoided because the Christians were warned by their Moslem friends of planned attacks, but their homes were destroyed and they had to flee to a Christian village on the other side of the mountain, where they were given temporary shelter with relatives. Eventually, they settled in that village and built houses on land of their relatives, with state support. When the revolts were over, Moslem leaders of Hila invited them to return to Hila. However, they could not afford to move back because the state would not pay a second time to build a house. Ambon now has become more thoroughly segregated along religious lines than it had been for centuries. In this Ambon, there is far less room for cross-religious significant relationships of care.

For the Butonese immigrants who settled in Hila since the late nineteenth century mainly in rural areas, access to land was more of a problem. They were not regarded as full citizens according to adat, and relied on Ambonese clans and families who gave them permission to grow vegetables. They were not allowed to own trees, because that could be interpreted as a claim to the land on which the trees grow. This means that they were not allowed to own clove, nutmeg, coconut, and cocoa trees, cash crops that were vital material resources for social security of the Ambonese population. Though there were some wealthy Butonese, they all had to live under the precarious conditions of dependence on Ambonese patron families. During the 1980s, relationships with Ambonese were improving and the degree of citizenship was rising. They successfully claimed the right to participate in the mayor elections by reference to state law. However, social change had not reached the level that Butonese could claim full citizen rights in terms of access to land according to adat. They had to strike a careful balance between relying on state law to press for equal rights, yet observing adat in order not to be excluded completely from access to land. Their resource basis for social security was much more precarious that that of the Ambonese. And, in as far as they entertained relations of support with Ambonese, these were characterized by inequality and dependency, while the security of support was precarious. The violent revolts at the beginning of the century, when many Butonese were forced to flee, made it painfully clear just how fragile their position was. It disrupted the carefully built up relationships with Ambonese families, and left those Butonese that stayed with a sense of greater insecurity than they had known before. They had learned that they could not rely on their Ambonese patrons, nor on the Indonesian state. Instead of protecting its citizens, an utterly divided state, where police, military, and navy were set up against each other, played an incinerating rather than dampening role in the revolts. For their social security, they had to rely almost exclusively on their Butonese co-villagers. This situation reinforced their sense of personhood in terms of being a Butonese rather than an Ambonese resident or an Indonesian citizen.

4. Migration to the Netherlands

Lacking security of life is the number one motivation for rural–urban migration, for international labour migration, or asylum seeking. Generally, persons migrate not only to find better opportunities for themselves, but also to be able to support their families that stay behind. At the same time, migration is for many only possible with family support. The return for this support is rarely immediate, because it usually takes considerable time before a migrant is capable of beginning to support his or her family. A migrant moves typically through stages from being a recipient to one who can just about take care of herself but no more than that, to one of provider of support—if she is successful at all. We have seen that it is senior, higher-ranking civil servants to whom relatives turn for support. Students and younger civil servants characteristically need support. Young teachers in Indonesia, for example, are often living with families who are regarded as foster parents. Through this process of kinning, these foster parents provide social security to these young teachers, while at the same time supporting the state that pays its young civil servants wages that are so low that they cannot live on their own, should they wish to. Though their status as teacher and state representative is recognized, by putting it in terms of kinship, the support is acceptable for both sides, and foster parents and foster children can live comfortably together embedded in social relationships. The fact that the relationship is couched in terms of kinship does not preclude that calculation is involved in the set-up. For the teacher, the immediate advantage is that she does not have to live in isolation; the foster family may calculate that at a later stage the teacher may mediate services of the government. Thus, solidarity circles of persons responsible for supporting each other change when persons migrate. For those back at home, it may mean fewer mouths to fill, but it may also mean lacking labour, for it is the young and able persons who move away. The family invests in this kind of support in the hope that in the future they will reap the fruits by being supported by their migrated relative. When settling down, the migrant over time will develop relationships in the place of migration and, if successful, will manage to create or enter into new networks of support. These relationships are often more horizontal in character than the kinship relationships they left behind, though this is not necessarily the case, as the Butonese experienced. They co-exist with the network of support with the people back home.Footnote ²⁴

With international migration, migrants have to adjust not only to new social and economic conditions, but also to the new legal order of the country of settlement. Living in new social and economic environments also affects their notions of neediness, of obligation and responsibility, even more than internal migration. Immigration itself often stimulates changes in the system of state-provided social security in the country of settlement. These processes not only have an impact on their personhood and their relationship of care and support within the country of settlement; it also affects their long-distance support relations. I will first describe changes in notions of personhood and how these worked out for state-provided social security in the Netherlands. This is followed by an analysis of how Moluccan migrants absorbed these changes, and what that meant for their relations of care in the long-distance networks with Ambon.

Sarah van Walsum traces the intimate connection between immigration, social security, and nation building.Footnote ²⁵ Taking a historical perspective, she demonstrates how state-provided social security changes under changing economic conditions in the Netherlands. It also reflects a shift from a nation that is conceptualized in terms of family relations and religion towards one of individualism. Citizenship in the Netherlands used to be mediated through the family, with a male breadwinner and a dependent female carer for the family and children, and through hierarchically organized religious communities. Social security and care had primarily been a matter of the religious communities before the state increasingly took over responsibilities. With the development of the welfare state, the state came to stand in direct relationship with its individual citizens, while family responsibilities for care were relegated to the private sphere. Thus, social security was employed by the state to maintain a moral order, for state protection was only extended to those who had done all that was deemed reasonable to provide for the necessary protection and security.

What was deemed reasonable, however, changed over time. The changes were a result of two quite distinct developments. Growing emancipation had put pressure on the state to decrease the dependency of (married) women, which resulted in a more individualistic attitude. This development became entangled with a second development: the convergence of a fully fledged state social welfare system with a rising number of immigrants from the former colonies, from southern Europe, and from Turkey and Morocco, where the majority of migrants came from, and asylum seekers. As it became increasingly clear that they would stay and not, as originally intended, return to their country of origin, the government became concerned that the possibility of family reunion would put too great a burden on the social security system. This enhanced the tendency to define citizens and residents in individualistic terms rather than in terms of family affiliation. The new individualistic conception of citizen became the foundation of a moral order to which migrants in search of permanent residency or citizenship were subjected. It was also used to keep out family members who had been allowed to enter the country under the previous moral order of the family. Close family ties came to be associated with backward immigrants who could not let go their culture of origin. Those claiming state support had to subject themselves to a high degree of control in the intimate sphere. Immigrants in particular suffered from this form of intrusion, from which they could not escape at the risk of expulsion. With reference to Stoler, Van WalsumFootnote ²⁶ argues that the distinctions made by means of these legal techniques did not simply define existing differences, but were constitutive of the distinctions. As in the Dutch Indies, genealogy, gender, and education together formed the powerful web of narratives with stereotype representations of the excluded, generating a sharp borderline between the in- and excluded.

Of late, the Dutch government has discovered that the demographic developments force adjustments of the state social security system and that they do need the very kind of networks and institutions that they had ruined by their over emphasis on the autonomous individual. Today, the discussion is about the “participation society,” in which persons are obliged to rely less on the state and more on relatives, neighbours, and friends. But there are two fundamental differences with the past. Social security in this participation society is not founded on an image of a male breadwinner flanked by a female carer at home. Second, the population is much more mobile. Men and women alike are forced to participate on the labour market and to travel long distances to wherever they may find a job. Relatives and friends no longer live closely together and neighbours come and go. Creating and maintaining care relationships are problematic and the networks are fragile. It is entirely unclear who actually will take on this “privatized” care. On the other hand, younger generations are more used to a withdrawing state and might more readily turn to friends and relatives for help than the older generations.

The Dutch Moluccans have followed these social changes of Dutch society to a large extent, be it with some delay and more pronounced in the younger generations. They entertained the very stereotype of the closely knit migrant families for which they were criticized so much. But, in contrast to that ideology, for them, it held a positive connotation and not one of backwardness, but of a high moral quality, which formed the ideological backbone of the kind of support they were giving to their families and communities back home, but also to their relatives in the Netherlands. While most migrant communities emphasize family and ethnic relations, this was particularly strong among Moluccans because of their special migration history.Footnote ²⁷ As members of the KNIL, they had long left their places of origin before they arrived in the Netherlands in 1951. They were accommodated in barracks in the understanding that they would return soon to an independent Moluccan state. When that turned out not feasible, the state built from 1960 onwards neighbourhoods for the Moluccan population at the outskirts of medium-sized towns spread over the country. During the first years, they were not allowed to enter the labour market and the state took total responsibility but, in 1956, it suddenly changed its policy and forced them on the labour market.

Within the communities, care for kinship was a strict obligation, hierarchically arranged along gender and generation lines. Specific forms of kinship, such as “boat kinship” among persons who had become very close on the boat from Indonesia to the Netherlands, adoption, and pela,Footnote ²⁸ ritual kinship between members of two villages that obtained an extended interpretation in the Netherlands, were important elements of Moluccan personhood. It was also characterized by a high level of community work and institutionalized support for life-cycle rituals.Footnote ²⁹ As long as nearly all Moluccans lived together in the barracks and neighbourhoods, contacts with the Dutch environment mainly occurred through their social workers and at the workplace. For young people, schools were the main place of contact. The attitudes towards the state were ambivalent. Moluccans were highly critical for its betrayal, but they had no qualms to demand support from state institutions.

Over time, Moluccan norms for support changed from strict obligation towards one’s parents, one’s pela, and the elder generation in general, in which the oldest generation decided what support was needed, to a form of support that emphasized that the obligation was taken seriously, but at the same time the fact that it was voluntarily given. The most important shift was perhaps that care for children began to mean more than providing food and clothing, and to include mental care and support in education. As a result, care for children began to take priority over care for one’s ageing parents. Family relations and support are still very much part of Dutch-Moluccan personhood. But the scope for negotiation has become much larger and the ways of extending support have become more diverse and are no longer defined by the oldest generation. Moluccans agree that a certain degree of individualism has become part of Dutch-Moluccan selfhood. This is a point of concern for many who fear that they will lose their distinct Moluccan kind of personhood. However, this concern is immediately qualified, for people are quick to point out that they are not individualistic in the same liberalist way as they ascribe to their Dutch environment. When they talk about individualism, this means that they no longer subscribe to the strict demands of family relations of the past and the circles of support for which these strong obligations apply have narrowed. It is too early to assess how the latest social security policies of the Dutch government that lays responsibility for care with family, neighbours, and friends will affect Moluccans. The fact that family ties are still so much part of Moluccan selfhood may help them to accommodate to the latest turn towards the participatory society, especially among those still living in Moluccan neighbourhoods.

The question we have to turn to now is how the relations with their relatives “back home” developed. At first, Dutch Moluccans focused on support for close relatives in the Moluccas such as parents, siblings, and first-grade cousins. This was considered a strict adat obligation. From the 1970s onwards, in a period in which Moluccan incomes were still low, an enormous amount of money and support in kind was nevertheless transferred to the Moluccas. This may indicate that the support relationships were one-sided, but Moluccans emphasize that they got spiritual support in return, that tied Moluccan migrants to their places of origin. Later on, when their economic position had improved, they also began to support “the community” in the form of reconstructing a mosque or church. In the 1990s, when more Moluccans had received a higher education, and under the influence of a booming development business largely funded by the state but implemented by private organizations, they began to extend more “structural support” in the form of developing projects. They claimed to be more suitable to carry out such projects than Dutch development workers, because they shared a common adat with the Moluccas. But it turned out to be a miscalculation, for their Dutch Moluccan adat appeared to be quite different from Indonesian Moluccan adat. In each of these types of support, they came to realize that they had developed different ideas of need, of ways of organizing projects, and of claiming cultures—that is, of Moluccan personhood. They had entertained an idealized image of the circles of solidarity on the Moluccas which was not ideal at all in real encounters. Unwittingly, they had absorbed the very changes in attitude towards social security and conceptions of personhood that Dutch society had undergone, and of which they were so critical. The recipients of their support recognized these differences in personhood and were quite critical of the patronizing way of the Dutch Moluccan development workers. For these, it was a painful experience that not everybody was willing to admit. Even more painful was the period of the revolts, because it put them before a dilemma. Dutch Moluccans on one hand felt even more obliged to support their relatives in their dire situation, but they knew the money could well be used on weapons. And, though many thought a person was entitled to defend himself and his family, they did not support the use of arms. Only a handful were in favour of the use of arms. Many did send money or even went to the Moluccas to bring it in person, in the hope that it would be used for living expenses and not for weapons. Quite a few were also involved in organizing attempts at mediation. Since the end of the revolts, relationships have been resumed.