Non-public welfare in Turkey: new and old forms of religiously-motivated associations

ABSTRACT

By studying the contemporary functioning and historic roots of a variety of religiously-motivated associations (RMAs), this paper explores the blurred boundaries between state institutions, political parties, and religious groups in the realm of welfare, in Turkey. Empirical findings are based on semi-structured interviews conducted with the volunteers and administrators of 27 RMAs, in three cities. These interviews focus on the RMAs’ motivations, administrative and organisational functioning and connections, presenting a comprehensive picture of RMAs as non-public welfare providers. Analysis of both new and old associations provides an understanding of how their histories and proximity to political power differ; it also reveals clues about the politicization of social provisioning. The findings of this study show that these associations have been some of the most important actors in non-public provisioning over the past two decades; they also shed light on the political economy of welfare provisioning.

KEYWORDS:

• Non-public welfare
• religiously motivated associations
• faith-based organisations
• social policy
• Turkey

Introduction

Recent decades in Turkey have seen a proliferation of religiously-motivated associations in the welfare realm.¹ Their numbers have increased dramatically since the late 1990s, as these associations have become major players in social assistance provision. Although this rise is in line with that of faith-based organisations in many parts of the world including Europe, the US, and the Mediterranean (see Atia 2012; Dierckx, Vranken, and Kerstens 2009; Göçmen 2013; Dinham, Furbey, and Lowndes 2009; Harris, Halfpenny, and Rochester 2003; Jawad 2009; Tuğal 2013), in Turkey there are case-specific reasons that are strongly related to major shifts in its political sphere over the past decade (Göçmen 2014). In contrast to developments in many European countries and in the USA, the rise of RMAs has not happened in Turkey at a time of welfare retrenchment. Recent decades (which have seen these associations become increasingly active in public provisioning) were also those in which state provision of social assistance was on the rise (Buğra and Candaş 2011; Göçmen 2014; Buğra 2015). Since the mid-1990s, a new social policy approach, based on a combination of Islamic ideals and neo-liberal principles, has dominated the social policy environment in Turkey. I see the subsequent rise of RMAs as an extension of this new approach to social provision, developed by political parties such as the Welfare Party and the Justice and Development Party (AKP), which have their roots in the Islamist National Outlook movement (Milli Görüş). As the empirical section of this paper will show, the significant increase in the number of RMAs happened during AKP rule – yet the foundations on which these RMAs were built were laid in preceding decades.

In recent decades, although media attention has focused on the increasing presence of RMAs in the welfare realm and their discreet connections to certain political parties, municipalities and business groups, research on RMAs remains limited. This paper presents the findings of in-depth research conducted with 27 RMAs in three Turkish cities, in the period between January 2008 and June 2009. Semi-structured interviews were conducted with the volunteers and administrators of 27 associations in the three cities (Bursa, Istanbul and Konya). Having created a list of associations operating in these cities, I investigated their web pages to see whether explicit or implicit references were made to Islam, God, the Prophet Muhammad, or other divine issues. This initial step allowed me to make a preliminary decision as to whether an association could be considered an RMA. Once I had gained access to a couple of RMAs within a city, I used snowball sampling. The following data was collected: (1) History of the organisation: founding date, professional and educational backgrounds of founders, and main motivations in founding the organisation. (2) Organisational features: number of employees and volunteers, scope of action, main activities, decision-making, and resources. (3) Connections: relations to institutions, other civil society organisations, municipalities, political parties, media and business. All information was collected in Turkish, and the direct quotations used in the text are my own translation. The empirical research reveals a comprehensive picture of RMAs as non-public policy providers.

An important contribution made by this research is the distinction it makes between new and old associations. By revealing specific characteristics of the interactions between these two groups of RMAs and various religious groups, political parties and state institutions, this distinction makes clear that there is no homogenous field of RMAs in Turkey. Analysis of the new and old associations provides an understanding of the existing differences in terms of both their history and their proximity to political power. It also sheds light on the political economy of current welfare provisioning.

The paper is structured as follows: First, I delve into the literatures of welfare mixes and governance, to contextualize the rise of RMAs in various European countries and in Turkey. Second, I present my general findings on the characteristics of new and old RMAs. Third, I discuss the positions of old and new RMAs vis-a-vis the state, describing the differences in their relations to it.

RMAs as hybrid organisations: an important part of the welfare mix in Turkey

A welfare mix is a social provisioning arena shared by public and private actors and institutions, such as family, state, market and voluntary sector (Wuthnow 1991; Evers 1993; Abrahamson 1995; Evers and Olk 1996; Abrahamson 1999; Ascoli and Ranci 2002). Welfare mix configurations have always been in a making and remaking process as part of dynamic changes in societies’ political, economic and social spheres. Following welfare state transformations in the post-1980s period, governments and political parties in various countries were taking a close look at the possible roles and responsibilities of the voluntary sector, having become aware of its potential. One group of actors that came to the foreground in many countries, including Turkey, was faith-based or RMAs.

In addition to the growing presence in the welfare realm of the voluntary sector in general and RMAs in particular, relations between the various actors and institutions of welfare have also been changing over the past few decades. The recent emphasis on the issue of governance in the social policy arena is an indicator of these changing relationships (Jessop 1999; Daly 2003; Bode 2006). Governance mainly refers to increasingly decentralized and diffused forms of policymaking and provision, as an outcome of coordination between private and public bodies. As will be shown in the empirical section of the paper, RMAs’ operational modes and connections to state institutions, political parties and public welfare providers in these Turkish cities are in line with contemporary developments in the governance of European welfare states. As in these European welfare states, a growing multiplicity of actors is engaged in social provisioning in Turkey, yet this very multiplicity is restricted, since those associations that are not a good fit with the current neo-liberal and Islamic agenda of the governing party are either restricted – or placed in disadvantaged positions.

The RMAs’ relations with other actors in the welfare mix (such as state social provisioning institutions, business associations, and other market and voluntary sector actors) are generally vague in nature. This is due to the intermediary nature of a third sector (Anheier and Seibel 1990; Seibel 1990; Gidron and Kramer 1992; Evers and Laville 2003; Kendall 2009; Evers 1990) in which ‘there is no clear line of demarcation between, on the one hand, the marketplace, the political arena, communities and state organisations, and on the other the third sector’. (Evers and Laville 2003, 36). Forms of cooperation and partnerships – both within these organisations, and between them and the state – provide a further defining characteristic for RMAs. Some scholars have interpreted new forms of cooperation such as networks and partnerships as creating win-win situations for all participants – including markets, civil society, community-based initiatives and the state (Evers and Laville 2003, 37). Others point to the potential issues of dependency creation and instrumentality that can arise between governments and civil society organisations (Lewis 1999; Dahrendorf 2001; Lewis 2004). Current social service organisation diversity – in terms of resources, goals, steering mechanisms and corporate identity – are generally defined using the analytical concept of hybrid organisations (Evers and Laville 2003, 237–252). My claim is that the nature of the relationships between hybrid organisations and the state becomes even more interesting when the organisations focused on are the religious or faith-based ones. This is especially true in the case of Turkey, where the rise of political Islam has been a major defining characteristic of the political sphere in recent decades; the relationships these associations build with state institutions are crucial to understanding the new forms of governance that are established.

RMAs as non-public actors of social provisioning in Turkey

RMAs are non-public welfare associations, unconnected to institutionalized forms of religion or places of worship. The main reasons for conceptualizing them as ‘religiously-motivated’ are the grounds on which they provide welfare and the largely veiled connections they have with religious groups, sects, and tarikats (Sufi orders and communities). The organisational scope of these associations varies hugely. To represent their current variety, the sample includes RMAs that are active at local, national and international levels. Their budgets range from USD 2000 to USD 100 million (Göçmen 2014). The total budget of the 13 associations willing to disclose figures is about USD 66 million (excluding international humanitarian aid) (Göçmen 2014). When compared with the budget of Turkey’s largest state institution for social assistance,² these figures show that the RMAs’ share of the total social assistance arena is significant, arguably placing them among the main providers.

In spite of the wide range of organisational scope, these associations do share common areas of activity – such as regular support for poor families via the provision of basic supplies, temporary relief provision in times of emergency, and informal recipient education on moral and religious issues. The RMAs’ target populations are citizens living below the poverty or extreme poverty line, and not covered by any forms of state provision (such as social security or social assistance). The associations work on a voluntary basis, and their main financial resource is donations. Local RMAs are generally dependent on personal and neighbourhood connections, whereas the national-scale organisations use all kinds of media (television shows, radio and newspaper advertisements) to increase donations. Of 26 RMAs interviewed, nine currently had weekly television shows and radio programmes to support their activities, and more than half had advertisements on TV, on radio and in newspapers. By increasing RMA visibility, media connections build support for the activities they undertake. In addition to allowing associations to publicize their activities, these television and radio show also include advertisements for the business holdings and companies that support them financially.

The way RMAs operate is marked by a charitable logic, lining up perfectly with the existing historic trajectory in the realm of social assistance. Historically, social provisioning has always been an area of benevolence in Turkey (Buğra 2008), and both informal relations between providers and recipients, and a lack of professionalization in the social assistance arena have always been the rule rather than the exception. Yet in the periods following the 1999 earthquake, and the 2001 economic crisis during the Bulent Ecevit governments, attempts were made to make social assistance more systematic and professional. For example, the growing presence of state vakifs (Sosyal Yardimlasma Dayanisma Vakiflari-SYDV) as main actors in social assistance, in the post-2002 crises, did bring a certain amount of professionalization to the field. In 2004 SYDV was given the institutional structure of a general directorate called the Prime Ministry General Directorate of Social Assistance and Solidarity, mainly financed by off-budget funding from income-tax revenues, traffic fines and voluntary donations. Through programs such as conditional cash transfers to the poor, social transfers to the elderly and disabled, assistance programs for education and health, and development projects such as micro-credits main restructuring of the welfare administration continued in the AKP period (see Göçmen 2016). Policy instruments such as means testing and targeting started to be used in a more sophisticated way which resulted in further bureaucratization and professionalization of social assistance. Nevertheless, this professionalization did not necessarily bring with it policies that are right based, the social assistance mechanisms developed both at central, local and civil level in the period of AKP rule continued to be marked by an ethos of charity (Buğra and Candaş 2011; Buğra and Keyder 2006; Göçmen 2011). When the practices of RMAs is under focus, it is possible to see that practices of provision by RMAs lack objective eligibility criteria; and tighter surveillance mechanisms exist, to control the use of the help provided.

New and old RMAs

Empirical research presented in this paper has shown that RMAs in Turkey are not a coherent group; they differ in many respects, such as in their relations to central and local state institutions, their political and ideological standpoints, and their perception of their own role in the realm of social welfare. Analysis of the individual associations made clear that the main dividing line in the sample is between what I have labelled the ‘new’ and ‘old’ associations. This distinction is based on characteristics such as date of foundation, main aims of founders and main areas of activity at the time of establishment. Although it is not easy to set strict dividing lines in any of these areas, it is still possible to discern the common trends that divide the two groups (see Table 1 below).

Table 1. New and old RMAs.

	New RMAs	Old RMAs
Foundation dates	Late 1990s, post 2000s	Pre-1990s
Activity scope	National and international	Mainly local
Activity areas	Limited to social provision	Broader than social provision
Formulation of motivations	Humanitarian	Islamic
Social/educational background of administrators	University graduates, young professionals	People with formal or informal religious education
Sources of funding	International campaigns, TV programmes, advertisements	Personal and local community connections

The associations’ foundation dates, which corresponded to different eras of political Islam in Turkey, define their main motivations and characteristics. It is possible to say that both groups have been important organising forces for the rise of political Islam. Yet, they have used different strategies to achieve this aim. The old RMAs were mainly established in order to keep the practices and teachings of Islam alive. In the period preceding the 1980s, most of these associations were organising places for cemaats (religious communities) and tarikats (Sufi orders) – some of which have been major actors in the rise of political Islam. Old RMAs are revised versions of previously-existing institutions; their foundation dates are in the 1970s, 1980s and 1990s. Nearly all the RMAs included within this category are foundations (vakifs) – which are also civil associations but have some legal differences from associations. One important difference is that associations work on a membership basis, whereas foundations are founded on charitable endowments so that their main source of funding is drawn from sustainable resources. Vakifs have a history dating back centuries.³

The social provision was not a crucial aspect of the Islamist movement prior to the 1990s; the possible political benefits of social welfare provision were only discovered in the mid-1990s. The Welfare Party was a pioneer in exploring the most efficient ways of using grass-roots social provision as a way of building its vote share. This marked an important shift in the rise of political Islam. Following this shift, the old associations and foundations, whose establishment had been driven by various motivations (such as Qur’an courses, teaching Islamic ways of living, or supporting the international Islamic movement) began prioritising provision activities in the post-mid-1990s period. Most of the old associations have been involved in activities supporting the Islamization of both society and the political system. Some have experienced times of political pressure, being prohibited or closed down for periods of time. Yet, echoing the history of pro-Islamic political parties, following every closure, they re-opened with new names. The military intervention of February 28th, 1997 was an important turning point for most associations that were undertaking action aimed primarily at the Islamization of society and the political system. The more radical among them, whose aims included changing the political system, for example, were closed down. What I define as the old RMAs are those associations that redefined their identities and areas of action in the period following the February 28 military memorandum. Some of those associations that had been prohibited just re-opened, with different aims; others that had not been closed down simply became more aware of the possibility of closure and took action to avoid it. As mentioned in many interviews, the proportion of social provision in the associations’ action repertoires increased after this point. Below is one example:

We started with educational activities; the main aim of the association was to educate people in matters of morality, and religion. We were active as a cemaat (religious community) even before the association was legally established. Three years later, we also started undertaking activities in the area of infak (social provision) to help the needy (Istanbul, Chair of RMA21).

Realising the impossibility of changing the political system in the short-run, and digesting the failures of the past, a group of Islamists redefined their main aim as the Islamization of daily life⁴ – and social provision was an excellent instrument for achieving this aim. The activities of many associations followed the example set by the Welfare Party in organising social provision. The Welfare Party’s main political program, ‘Just Order,’ was the first attempt to bring Islamic ethics into the capitalist economy and create a community around these ideals. With its emphasis on achieving equality and social justice, the ideals of the ‘Just Order’ programme were similar in some respects to those of a welfare state. Nevertheless, the road chosen to achieve this ‘just order’ was via the establishment of communities and networks founded on Islamic ideas of morality and trust, rather than via state redistribution or a broadening of the scope of welfare policy. As Buğra stated, ‘Only in the cultural setting of Islam could justice acquire a meaning, a meaning derived from reciprocal obligations of trust, loyalty and solidarity that bind the community of believers’ (Buğra 2002, 129). The main instruments of creating the ‘just order’, then, were private benevolence and voluntary initiatives organised around communal forms of belonging and solidarity. The Welfare Party was the first political party to invest significant effort in establishing networks to support the poor that were based on religious membership. As is widely discussed in the literature, these networks have been where intra-community solidarity to Anatolian capital arose, providing their members with mutual support and cooperation: ‘Religious groups or communities, while building mosques, Qur’an courses, schools and student dormitories with the money collected from members or friends, who give to charity out of religious duty, also prepared a customer and capital base for the schools, business and enterprises of their members’ (Demir, Acar, and Toprak 2004). Understanding the success of the Welfare Party in the 1994 local elections and 1995 general elections demands an awareness of the support of ‘a complex network of Islamist economic ventures which included companies, chambers of commerce and industry, trade unions, women and youth groups, some 50 publishers, 45 radio stations, 19 television channels and hundreds of video and cassette producers’ (ibid.). It is my contention that these Welfare Party networking activities paved the way for both the transformation of old RMAs and the rise of new RMAs in the post-1990s period. Nevertheless, further institutionalization and expansion of new RMAs in the social provision arena have occurred under the rule of the Justice and Development Party (AKP) (2002 to present).

The history of new RMAs is somewhat different from that of the old RMAs. New RMAs are those established as a result of the success of the rise of political Islam; all were established during the period when either the Welfare Party or the AKP were in power. They differ from the old RMAs in that their connections to political parties, business groups and Islamic networks formed the basis on which they could rise. They have thus always been political actors – right from the very day of their establishment. Most have, or have had, close relations to the political party in power and those that lost that proximity also lost their privileges, or closed down. Examination of how the biggest new RMAs were established gives some idea of the blurred boundaries between these associations, political parties, and certain business groups. For example, the history of the Deniz Feneri Association – an RMA identified with the rise of moderate political Islam – dates back to a television program Sehir ve Ramazan (City and Ramadan) that began during the religious month of Ramadan, in 1996. The channel that produced and broadcast the program was a municipal channel in Istanbul when Recep Tayyip Erdogan was Istanbul’s mayor, and the Welfare Party was the city’s governing party. This channel was subsequently privatized and sold to an Islamic capital business group, changing its name to Kanal 7. The founding of the Kimse Yok Mu Association, an RMA sounding important nationalistic tones, is similar to Deniz Feneri in significant ways. In 2004, Kimse Yok Mu also started as a follow-up to a television programme broadcast by a TV channel – Samanyolu – having an Islamic identity. Both the association and the television channel were known for their close connection with the Gülen Community.⁵ By the time of interviews Kimse Yok Mu, with branches in 28 Turkish cities, was the biggest RMA in Turkey – and as the quote below illustrates, this very substantial growth was due to its connection to an Islamic social movement headed by Fethullah Gülen.

When our association was no more than two rooms, an earthquake happened in another country (Indonesia – 2004). Then the leader of one of the largest cemaats (religious communities) in Turkey, who is highly respected, stated that it was important to help the people in that country. Over the next month, we received USD 12 million. It is safe to say that this association was established at his behest. Whenever he briefly mentions the importance of helping our Muslim brothers, millions gush into the organisation. So it is beyond our control, we are expanding every day, even if we do not want to. Samanyolu television was founded at his suggestion, as was Zaman, a daily newspaper. The Cihan News Agency was also set up at his initiative (İstanbul, RMA3 Administrator).

This quote demonstrates the benefits of the RMAs’ interconnectedness to both the religious community and the networks established around it; these communities’ shared social and cultural capital can thus easily be transformed into economic resources.

Cansuyu is another new RMA, founded in Ankara in 2005. It has offices in six major cities in Turkey. The Cansuyu administrative staff of consists of former bureaucrats, mayors and activists – many of whom hail from the Saadet Partisi (Felicity Party), which was another political party with roots in Nationalist Outlook Movement. Cansuyu administrators talked openly about its connections to the Felicity Party:

You probably know that we were established in connection with the Felicity Party. IHH was also established on the advice of Necmettin Erbakan, but our close connection with them ended at a point. Now we have Cansuyu as our welfare association. We also have a television channel. We have a programme about social help and solidarity, on the TV5 channel. This is how the idea of founding a solidarity association came up. […] When you start as part of a pre-existing structure or political party, it is much easier to expand. Having political connections means having a connection to the whole world. […] Another important benefit of these connections has to do with finding volunteers and working with them. When people in need come to the political party to ask for help, we tell them we have an association distributing social assistance, and send them to that association. The state cannot achieve everything; we should also fulfil our responsibilities (Bursa, chair of RMA9).

Close relations between political parties, business networks and RMAs are thus mutually beneficial to political parties and welfare associations. Associations use the political parties’ networks to source volunteers and increase donations, while parties use RMAs to provide services to their supporters. The foundation histories of the new RMAs demonstrate that these associations are a direct product of the rise of political Islam. The political movement created its new associations rather than only relying on the old ones. As the examples in the next section will also demonstrate in most of the cases the new associations were established as top-down organizations in order to assist the political movement.

Positions of the new and old RMAs vis-a-vis the state

Blurred boundaries between religion, politics and social assistance are characteristics of both new and old RMAs, yet patterns of RMA-public sector interactions differ between the two groups of associations. Two important defining characteristics are: how new and old RMAs perceive the state,⁶ and their own position vis-a-vis the state, in the arena of social provision. New RMAs see state institutions as potential partners, defining their own role in the social provision as supporting the state. They gave quite positive answers to my questions on how they see their role in the provision arena, and whether they are open to working with state institutions. Partnerships between a civil society association and a state institution in local, national or international projects were still rare at the time of my interviews, although there were some examples.⁷ In addition to partnerships, the most common interaction between welfare administrators and RMAs at local level concerns exchanges of information. Within a locality, there are generally three types of institutions providing social assistance, namely: state vakıfs (SYDV), the municipality and non-public welfare actors (RMAs and others, if any). One of the problems of this institutional multiplicity is that there is a lack of organisation and communication between these institutions. In some cases, where there are positive relationships between the RMAs and the SYDV and/or municipality, they share their lists of beneficiaries with each other to prevent multiple provisions – sometimes even conducting means testing jointly. At the time, this was still a very new practice, mentioned by just a couple of associations.

Founders of new RMAs conveyed the idea that the state cannot be a remedy for all needs in the society. They believe that associations should also play a role in social provision. Therefore, as the quotations below exemplify, they have decided to fulfil what they interpret as their own responsibilities:

We see what we do as a part of our social responsibility. I believe that is how things should work. I am very happy that the number of civil society organisations working in this area has been increasing. Of course, the state has to provide the social provision, but civil society organisations also fulfil very important roles (Istanbul, chair of RMA14).

The state has to be more active in the area of social policy. We see our role as guiding and stimulating the state. We do not dream of taking over the role of the state in the area of welfare. We just want to be a role model for the state activities. Civil society has to function as a role model in this area (Istanbul, chair of RMA11).

There are some things you cannot expect the state to achieve; they are not a part of the state’s responsibility. The state has the responsibility in the areas of education, health and infrastructure; however, it cannot be expected to create solidarity. Creating bonds of solidarity can only be a function of civil society organisations such as ours. In a way, what we do is complementary to state activities (Konya, chair of RMA19).

This positive approach, of believing in a necessary cooperation between state and RMAs in the area of provision, is common to the associations in this group.

The AKP government has been fairly supportive of RMAs, facilitating their growing presence in the welfare realm through legal changes in multiple areas. One legal change worthy of attention in this context was the introduction to income tax legislation of a food bank system, in 2004. This change to tax legislation made it possible for companies to donate their products to civil organisations and deduct the amount of donations from their income taxes. A minor revision of the law in 2005 extended the food banking to include clothing, home cleaning materials, and fuel – which in turn increased the amount of donations RMAs get from different businesses. Another indicator of the positive relations between new RMAs and the AKP is the discursive support accorded to certain associations in the media.⁸ The changes introduced in the associational law and tax legislation also show that the ruling government sees RMAs as potential partners and seeks to increase their involvement in the welfare arena. One important indicator of this was the awarding of Association for Public Interest status to some of the largest RMAs in this group. This status, which can only be decided by the Council of Ministers, guarantees important rights such as the right to collect donations without prior permission, and significant tax exemptions. Three of the new RMAs (Deniz Feneri, Kimse Yok Mu and Dost Eli) have been awarded Association for Public Interest status – in 2004, 2006 and 2014 respectively. However, Kimse Yok Mu’s status of Association for Public Interest was revoked by the Council of Ministers in 2014, following increasingly conflictual relations between the Turkish government and the Gülen movement. Both the granting of this status to the abovementioned RMAs, and it is being withheld from Kimse Yok Mu are excellent indicators, showing that RMA involvement in welfare provisioning has resulted in politicization, and that the relationship between RMAs and the ruling government is a dynamic one, capable of shifting from one day to the next following changes in the political sphere.

Unlike new RMAs, old RMAs see state institutions as control mechanisms, defining their actions in the provision arena as either totally free of state action (an outcome of religious aims) or as fixing state failures (such as the failure to institutionalize zakat – distribution of the alms tax). In contrast to the way in which new RMAs have been completely incorporated to the political system, old RMAs (notwithstanding the crucial transformation they have undergone in recent years) continued, at the time of field research, to include traces of Islamist opposition to the existing system. Dissatisfied with the current system, the administrators and the volunteers in this group yearn for a return to the Prophet Mohammed’s time and/or the Ottoman Empire period. They believe that zakat – the Islamic duty to give a set proportion of your wealth to the needy – should be practised independently of the existence of a welfare state. To their minds, in an Islamic state, it should be the state’s responsibility to collect and distribute zakat, as alms tax. They claim that since the Turkish state does not practice this rule, civil society organisations should assist citizens in fulfilling their religious duties. The following quote provides an example of this:

We started this association as five small-scale tradesmen. You know, there is an institution called zakat in Islam. Our aim was to create a pool – from the zakats of we five – from which we can distribute to the poor. In the meantime, however, our friends and acquaintances heard about our organisation and offered to give their zakat to us. We therefore officially established the association (Bursa, chair of RMA22).

In addition to being critical of the state, the associations in this group interpret their role as being totally independent of its welfare actions. In comparison to new RMAs, old RMAs prefer to minimize their interaction with the state. Respondents were fairly negative when discussing potential relations with state agencies:

We do not see the kind of work we do as a state responsibility. This is our duty. There are things that are promised to us by God and the Prophet. There are rules we have to obey, such as sadaka and zakat. That is how we started. In time, we realised that we are doing the sort of work that the state has to do (Konya, administrator of RMA13).

In Turkey, you are only acceptable if you do everything behind a secular display. If you do social assistance with religious motivations, you are labelled as fundamentalist and/or reactionary. The state expects me not to have any religious motivations. It is impossible not to have any religious motivation and still take part in these kinds of activities. Inspiration can only come from a spiritual source (Bursa, chair of RMA23).

In their view, the only kind of interaction state agencies have with the RMAs is aimed at controlling them. A statement from one respondent sums up the general attitude of this group: ‘We are glad if they just stand back and do not interfere with what we do’ (Bursa, chair of RMA23). Some RMAs that are reorganisations of previously banned associations by the state are especially negative towards the idea of working together with the state institutions. As the excerpt below demonstrates, the associations do not see a common ground on which partnerships or projects may develop between voluntary organisations and the state:

Civil servants working in state social provision departments do not perceive civil society organisations in a positive light. They try to classify us as belonging to this or that fundamentalist or reactionary group (Istanbul, chair of RMA8).

I interpret these two clearly defined positions of the new and old RMAs vis-a-vis the state as a reflection of shifts in the political sphere. Whether the associations were established in a time period in which the political environment was welcoming or not has been an important defining factor in shaping state-RMA relations. Nevertheless, it is also crucial to stress that the picture presented here is quite a dynamic one and is closely related to the establishment or disestablishment of power alliances in the political arena. At the time of conducting this empirical research, Kimse Yok Mu was the biggest RMA in Turkey – due to good relationships between the ruling party and Fethullah Gülen. Yet in 2014, the association’s Public Interest status was revoked by the Council of Ministers, following growing discord between the Erdogan government and the Gülen movement. It was then closed down following the failed coup attempt of July 15^th and its administrators arrested and taken to court for taking part in terrorist activities. This example shows that relations between political parties and RMAs can be liable to sudden change.

In analysing relations with local government and local welfare offices, it is impossible to make a clear distinction between new and old RMAs. The differences here are dependent on the political party in charge at the local offices. In cases where the city governors, mayors of different municipalities and civil servants in the social assistance offices are supporters of the same political parties, relations between these institutions are quite positive. This positive relationship generally takes the shape of a structural advantage for associations that are close to the AKP at the local level. In most cases, even the old RMAs would work with AKP municipalities. As one of my respondents in an old RMA openly declared: ‘Currently, the local municipality lets us use the conference rooms in the municipality for free, but if the municipality were not represented by party x – which is an Islamic one – then they would not let us use it’ (Konya, RMA2). Or as another respondent mentions: ‘We always had good relations with the municipality, sometimes they sent us needy people so that we could help – in a way, that can also be seen as a form of advertisement for us. Imagine, as a citizen, you go to the municipality to ask for help and the workers in the municipal office send you to us, this makes us feel that our work matters. It means we’re doing important work.’ (Bursa, RMA7). In contrast, one interviewee from a municipality governed by the Republican Party stated, in response to a question about their relationship to that municipality: ‘Everyone is minding their own business’ (Istanbul, RMA27). When I asked him to elaborate on what he meant he said that their perception of who needs help and how it should be provided is different from that of the municipality and that therefore, they do not have any contact.

Conclusion

In line with developments in other parts of the world, recent decades have seen a growing presence of voluntary actors in the welfare realm in Turkey. In accordance with the country’s existing institutional structure (which has always side-lined social assistance as a matter of benevolence) and political climate (which has been marked by the rise of political Islam and neo-liberalism), RMAs have become one of the main actors in social assistance provisioning. The scale of social provisioning by RMAs shows that though they are not the only non-public actor in the realm of social assistance, they are the main actor with the greatest share – and this situation also results in the increased politicization of social provisioning. The lack of plurality in the associational sphere and the biased position of the governing party towards RMAs are problems of the governance regime that has been established in the AKP period. There is evidence showing that some RMAs work as an extension of political parties.

Closer analysis of the associations demonstrates that RMAs do not form a homogeneous group. By focusing in particular on the political climate in which the associations were established, I have categorized RMAs as being ‘old’ or ‘new’. Both groups are non-public actors in the welfare arena and associations belonging to both groups have indistinct connections to political parties, religious groups and state institutions. New RMAs rose on the networks of reciprocity established during the rise of Islam in the economic, political, and social spheres, and these have been political actors from the outset. On the other hand, the study of old RMAs demonstrates how these associations evolved in parallel to such contextual factors as transformations in the political and socioeconomic arenas. Due to their time of establishment, and their proximity to current political power, new RMAs have a more positive perception of state institutions and see their own role as contributory to the state. Collaboration between state vakifs, municipal governments and RMAs mostly takes the form of everyday information exchange, and the creation of facilities; it rarely happens in project form. In comparison to this, old RMAs’ relationship to the state is less positive. Old RMAs perceive their role in the social welfare arena as being independent of what the state does, and generally, they prefer not to work with state institutions. Moreover, they are either critical of state institutions or believe state institutions to be critical of them.

One important finding of this study is the presentation of the amount of dynamism in the area. The establishment and disestablishment of power alliances in the political sphere is continuous, and this seems to be the most important criterion in determining the future presence and share of an association in the social welfare arena. Within the political context at the time the study was conducted, it was new RMAs that were closer to the AKP government and thus had the biggest share, yet when the growing conservatism in the period following the study is taken into consideration, we would expect the position of the old RMAs to also have changed, in between times. Further research is needed to discover the current state of relations.

Disclosure statement

No potential conflict of interest was reported by the author.

Additional information

Notes on contributors

İpek Göçmen

İpek Göçmen is a postdoctoral researcher at the Max Planck Institute for the Study of Societies, Cologne. She holds a PhD in Political Science from the International Max Planck Research School and University of Cologne. Her research interests include comparative social policy, religion and welfare, gender and child protection. She has published in journals such as Journal of European Social Policy, European Journal of Women's Studies, Social Science & Medicine; and Gender, Work and Organisation

Notes

1. RMAs are not the only actors in the welfare provisioning arena; there are also secular organizations, mainly involved in services such as education. Moreover, because these secular associations are generally also critical of the government’s activities, they face marginalisation, being closed down or seeing their administrators taken to court (Buğra 2015). Cagdas Yasami Destekleme Dernegi and Sarmasik Dernegi are just two examples among many.

2. The total SYDGM (General Directorate of Social Assistance and Solidarity) budget for 2009 was around USD 395 million (see http://www.sosyal- yardimlar.gov.tr/tr).

3. For an in-depth study of historical transformations of Vakıfs see Zencirci, G. (2015) .

4. For a study on this subject, see Tugal, C. (2009).

5. Following the failed coup attempt of 15 July 2016, this community (FETO/PYD) was added to the list of terrorist organizations in Turkey.

6. Although all interviews were conducted when the AKP had been in office for almost a decade, I do not think that respondents’ answers reflect their relations to the state under the AKP government alone. That may be the case in some examples, but in many interviews, respondents were careful to distinguish between a more general perception of state institutions, and the political party in power.

7. Such as a joint Cansuyu – Ministry of Education project distributing free wheelchairs to disabled children.

8. For Recep Tayyip Erdogan’s positive comments on Kimse Yok Mu for their activities during religious festivals, see the related article in the Zaman daily newspaper, 22.12.2007). The talks Erdogan held after the Gaza Flotilla incident of May 2010 were an example of support for IHH.