The Prolonged Inclusion of Roma Groups in Swedish Society

Inclusion policies focusing on Roma groups started in Sweden during the 1950s, when the Swedish government recognized the formal citizen status of the so called “Swedish Gypsies”, a group consisting of approximately 740 people. As the Roma were perceived as people living outside the boundaries of normal society, the challenge facing the Swedish authorities was how to outline and organize the new policies. In our analyses we focus on the taken-for-granted premises of these policies. We discuss the “entry process” of these Roma into Swedish society. People-processing organizations classified Roma as “socially disabled” in different administrative contexts. In the early 1960s adult male Roma were classified as socially disabled on the labor market. Later during the same decade, experts and professionals increasingly focused attention on the Roma family as a problematic institution. In this context, Roma adults were classified as disabled in relation to the normative representations of parental capacities during that time, while Roma children of school age were defined as children with difficulties and put in special groups for children with problems. The related interventions were justified by a discourse on social inclusion, but in reality produced a web of measures, practices and yet further interventions, which in the long run have contributed to perpetuate the social marginality of Roma groups.


Introduction
In this article we focus on the period 1950 to 1970, i.e. the time when new Roma policies emerged in Sweden and were established within the welfare system.In our analyses we discuss the taken-for-granted premises of these policies and investigate the perceptions and actions of the authorities once the 740 Roma were recognized as part of the Swedish population (SOU, 1956: 43, p. 145).It is important to combine the analyses of the prescribed model role of policies with an analyses of the local and practical implementation of these, since the informal implementation of formal policies interacts and shapes social politics (Lipsky, 1980).
Brodkin conceptualizes this as the politics of practice (Brodkin, 2010).Methodologically, we are inspired by the concept people processing organizations, i.e. organizations that shape a person's life by processing them and conferring them a public status (Hasenfeld, 1972;Prottas, 1978), and thus become the site for the politics of practice.A precondition for the initiation of such a process for the Roma in Sweden was to formally recognize them as members of the nation-state.The citizen status is one of the most important entry categories to be considered part of the nation-state organization (Sainsbury, 2012).Hence, in 1952, the Roma were granted Swedish citizenship.However, this formal status was followed by a long series of steps that peo-ple processing organizations developed in order to manage the entry of the Roma into the Swedish welfare state.The entry was thus turned into a comprehensive, expanding and continuous process involving an increasing number of interventions and knowledge producing activities (Kaminski, 1980;Marta, 1979;Montesino, 2002;Ohlsson Al Fakir, 2015).These interventions and activities were justified with reference to the old narrative on Roma as a problematic group, which in the first place was used to legitimize the entry of Roma into the welfare system.The processing of the Roma into this system followed the classical steps in the work of people processing organizations (Hasenfeld, 1972(Hasenfeld, , 2010;;Prottas, 1978):  A detailed evaluation of the current situation in order to determine the legitimacy and extent of public intervention. The identification of the attributes that make citizens potential clients in order to identify the appropriate interventions. To the above-mentioned steps followed the monitoring of the relocation process of the target groups.
Professionals in the social field and other experts assumed the role of gatekeepers in these organizations (Iacovetta, 2006).People processing work occurs at various places in society, thus gatekeepers are active at different bureaucratic levels and in different institutions.
Assuming the role of institutional gatekeepers were, among others, officials, who in their professional every day practice serve as "cultural interpreters" dedicated to facilitate adaptation.In the Canadian post-war example discussed by Iacovetta (2006), this involved convincing "old Canadians" of the valuable contributions that the newcomers were making to their society and culture, mainly in terms of music, foods, clothing, etc.In Sweden during the post-war decades, there were not as many social agencies as in Canada dealing with newcomers.Instead, institutional gatekeepers were in most cases civil service officials at departments like the National Board of Health and Welfare and, notably, the National Labor Market Board.The later institution was an important agency in the Social-Democratic post-war reform programs aiming at full employment and welfare for everybody (Ohlsson Al Fakir, 2015;Rothstein, 1996;Tydén, 2002).Academic experts worked in close cooperation with officials at the different civil service departments to provide the scientific material on which interventions were to be based.Among these experts were those who defined the organizational boundaries for the processing of the new citizens in the social order.In their evaluation research they constructed Roma citizens as welfare clients.In doing this, they employed the administrative taxonomy of the welfare services, which they also refined using medical and social arguments.Hence, during the 1950s and the 1960s these experts and professionals formulated the intellectual foundation for the institutional responses to the entry of Roma citizens in the welfare system.In practice this implied departing from established categories and identifying yet new dimensions in the classification map that justified new activities in which Roma citizens became targets of further evaluations.In these studies the expert added characteristics that confirmed the representation of Roma as deviants, that is, deviants who belonged to the national deserving poor.This gatekeeper elite held what Prottas has defined as an "organizational role as a boundary actor" (Prottas, 1978, p. 290).They defined the institutional boundaries, within which the street-level bureaucracy acted.Gatekeepers at the local level such as social workers, public health staff and schoolteachers, i.e. street levels bureaucrats (in Lipsky's terms), or frontline caseworkers (in Iacovetta's terms), performed the routine work of ordering and processing Swedish Roma as welfare clients classified according to established criteria (age, health, family situations, etc.).In this way gatekeepers at both organizational/scientific and local levels were involved in shaping the politics of practice in processing Swedish Roma citizens.
In the following sections, we first present a background to this inclusion process.Then we go on describing the taken-for-granted premises of the authorities' evaluation of the situation of the Roma.The subsequent section discusses disability as a time specific content of these premises.Thereafter we examine the contents of the activities used to incorporate the new citizens.The article ends with some concluding remarks.
The basis of the article is empirical data collected in two research projects conducted in the fields of Social Work and History respectively.The first project (Montesino, 2002) makes an analytical description of the Swedish government's Roma policy from 1880 to 1970.The study is primarily based on the analysis of public documents in the Swedish National Archives (Riksarkivet, 1950(Riksarkivet, −1960) ) and includes government reports and accompanying background material.The second project (Ohlsson Al Fakir, 2015) deals with one particular Public Health study concerning the Swedish Roma carried out in the 1960s and it concludes that the study in question resulted in increased and expanded activities of experts working with Roma in the medical and social fields.The empirical sources of this research consists of primary sources from various scientific and political-administrative contexts, including correspondence, working material and other written sources that reflect the daily work of experts and professionals in different scientific, institutional and administrative fields.

Background
The Roma, who became the focus of the Swedish au-thorities in the early 1950s, were a demographically insignificant group, at the time estimated to include only some 740 individuals.They descended from Roma groups that had arrived in the country at the beginning of the 20th century (SOU, 1956: 43).The granting of citizenship to this group formed part of the new social policy, which questioned all earlier social policies in the years following WWII.This policy change by the Swedish authorities resulted in a new approach vis-à-vis the Roma and other groups perceived as social deviants (Lind, 2000;Lindqvist, 2007).How to incorporate these groups into to Swedish society became a specific field of social intervention (Montesino, 2002;Ohlsson Al Fakir, 2015).
The public debate the years following the end of WWII was strongly influenced by social policy experts and social workers arguing for the inclusion of new groups.They criticized previous policies for being repressive and argued for a social approach to the understanding of deviance.In this debate, they collaborated with medical experts and added a social dimension to the medical definition of disability, stating that social conditions could also create disabilities (Kerz, Werner, & Wesser, 1995;Montesino, 2012;Ohlsson Al Fakir, 2015).This new approach created the conceptual space for drawing up new policies directed at groups previously defined as undeserving and/or unwanted.
Hence, the inclusion of the Roma in Swedish society is a process linked to many different levels and aspects of Swedish social and political history.It includes questions about citizenship, the requirements for being considered part of Swedish society, and the imaginary and constantly re-constituted distances between old and new members of Swedish society and policy practices.These policies and their implementation are taking place against the background of the old and deeply embedded narratives about the Roma within public services (Cf.Montesino, 2002;Willems, 1997).In the construction of these narratives public health and social experts have played a crucial role.The taken-forgranted premises of the dominant narrative in policy and practices have their origin in the formulation of the Gypsy Question, where we find the basic premises for the 20th century interpretation of the situation of Roma citizens.

Taken-for-Granted Premises: The Gypsy Question
The structure of the evaluation of the Roma situation after WWII has to be understood in relation to the general content and structure that we find in other "social questions" (e.g.The Social Question of the late 19 th century).Basic to this structure are the identification of a situation as unwanted, i.e. a social problem, and the subsequent demands for its solution.In order to solve the problem a number of strategies are elaborated and planned as progressive steps pointing to-wards the final stage, when finally the problem should be solved.The contents of the "questions" represent a common attitude to social problems based on a plethora of taken-for-granted assumptions about what constitute normality and deviance.
The Gypsy Question as part of the wider discussion of the Social Question tells us a lot about the basic conditions for membership in a national community.The Social Question deals with how to make citizens of the poor, while our analysis of the Gypsy Question illustrates how authorities have envisaged this citizen making process in practice, when considering the admission of poor strangers in the nation-state organization.
The Gypsy Question epitomizes views and strategies for the social inclusion of Roma groups in different national contexts.It was originally formulated at the end of the 18 th century by Heinrich Grellmann (1753Grellmann ( -1804)), who published the book Dissertation on the Gipsies, being an historical enquiry, concerning the manner of life, economy, customs, and conditions of these people in Europe, and their origin (Grellmann, 1787).This study and its thematic framework have had a strong influence on later policy and research concerning the Roma in different European countries.In Grellmann's study Roma people were for the first time considered potential members of society, that is, members who had to be "transformed" before they could be fully accepted.The contents of the Gypsy Question can be summarized in six basic statements (Marsh & Montesino, 2013): 1. Gypsies are one people.2. The Gypsies are outsiders.3. Gypsies should be transformed into "useful" citizens.4.This transformation demands "special measures" and requires a long time. 5. Education/schooling is the key instrument in this process.6. Gypsy children should be the principal objects of these measures.
These statements sustain that there is a question/problem, namely the Roma and their status as outsiders.This approach leads to a limitation in both the search for, and the production of knowledge.For instance any problems that occur in the process of policy implementation are automatically interpreted as caused by the Roma.The statements have later been reproduced and adapted to national and local contexts whenever authorities of different kinds have taken an interest in the situation of Roma groups.The uncritical reproduction of century old statements have contributed to the still dominant perception of Roma as one people with common attributes (Lucassen, Willems, & Cottaar, 1998;Tervonen, 2010).Examples of this in Scandinavia are the social reports by Eilert Sundt (Sundt, 1859(Sundt, , 1862) ) in Norway and later those by Arthur Thesleff (Thesleff, 1898(Thesleff, , 1911) ) in Finland and Sweden.Sundt's reports obtained mandatory status in the philanthropic organizations, initiating activities in Scandinavia at the start of the twentieth century, as well as in social reports arguing for public intervention (Ohlsson Al Fakir, 2013).The statements of the Gypsy Question have in these reports been the taken-for-granted starting point (SOU, 1923(SOU, : 2, 1956: 43): 43).
In the following sections we elaborate on the consequences of this dominant narrative for the Swedish authorities' evaluation of and interventions towards Roma groups in the post-war decades.We argue that, in spite of changing official discourses, there is a strong continuity of policies in the practices elaborated by the welfare authorities during this period.The official Swedish policy discourse towards Roma (and other groups considered culturally and/or socially different) underwent a radical change in the 1980s when more or less enunciated assimilation policies were abandoned in favor for integration policies towards migrants and ethnic minorities.However, the Swedish political scientist Carl Dahlström (2004) differentiates between the rhetoric on immigrant policies and the practice of it; policy rhetoric has changed but practical policies show a remarkable continuity between 1964−2000 (Dahlström, 2004).This is confirmed when considering the practices developed since WWII to make newcomers "fit in"; migrant and ethnic minority groups have been seen as people with special needs in these processes (Montesino, 2012), "requiring special treatment to enable them to adapt to the Swedish society" (Eastmond, 2011, p. 280).
Despite changes over time in policy objectives, from assimilation, meaning total adaptation to majority language and culture, to integration, denoting individual cultural, linguistic and religious freedom, adaptation of immigrants and minorities has been the main goal.Furthermore, the practical adaptation programs has remained more or less the same, focusing mainly on language training in Swedish, information about Swedish society, support for language and culture, special labor market and adult education programs and support for "troubled urban areas" (Dahlström, 2004).It is thus clear that changing rhetoric does not necessarily correspond to changing practices in the area of immigration and minority policies.The question of what the notions of assimilation and integration actually implicate, and how the two differ, must in other words be answered in the following: it depends on which level the analysis aims at.In our analysis of social policies towards Roma in Sweden, we have focused on the underlying premises of these policies, that is, the administrative welfare routines and categories that were established in the development and implementation of policies.Welfare services developed practices based on perceptions of Roma as unable to manage their own incorporation to the Swedish society.Independently of policy frame, this approach induces to practices that reproduce a view on Roma and immigrants as people who need special social support in their entry process.
In theory, for instance as expressed in formal policy documents, the difference between assimilation and integration is crucial as the former accepts no deviation from mainstream norms while the latter seems to support cultural difference.In practice, however, cultural difference has often been considered an obstacle (Cf.Bauböck, 1996;Westin, 1996).This contradiction is for instance expressed in the interventions targeting Roma (like those targeting immigrants).The contents of these interventions have remained much the same from the 1960s to the 1990s.The welfare services that work for the integration of Roma and other groups reproduce perceptions in which cultural difference is perceived as a social deviation.In our research, we draw the conclusion that the difference between assimilation and integration lays on the rhetoric rather than the operative level.
In this line of argument, we maintain that there is continuity in Swedish official policies towards Roma groups; the rhetoric has changed but the practices have been developed within the same policy paradigm formulated in the Gypsy Question.The six statements formulated in this question continue to shape the entire policy field where even newly created activities are reproducing and refining the contents of the old Gypsy Question (e.g.SOU, 2010).In the 1950s the incorporation of Swedish Roma was a part of the general policy towards groups identified as social deviants.The question of cultural difference was subordinated to this approach; cultural assimilation was considered the only strategy to overcome social disability.
The classification of Roma as socially disabled was later extended to newly arrived Roma groups and other migrants and refugees, who were incorporated into the already existing (and expanding) welfare practices (Iverstam Lindblom, Johansson, & Wall, 1978;Marta, 1979).In the 1980s when the official rhetoric focused on the integration of ethnic minorities and migrants, the contents of welfare practices remained (Dahlström, 2004).There is in other words a discrepancy between rhetoric and practice, which only becomes visible when juxtaposing normative with practical levels of policy work, and adding a historical perspective to the analysis of the politics of practice (Brodkin, 2010).Nowadays the term "Gypsy Question" is not used in policy documents at national level, but the statements are repeated in local policy projects, even in activities where Roma are represented among the staff.In the next sections, we elaborate on the main contents of the "question".

The Roma as One People
The assumption of the Roma being one people is based on the belief that the Roma consists of a group with a common history and culture (Lucassen et al., 1998).This assumption is repeated in different studies, confirming the homogeneity and identity of the Roma as either a social or an ethnic group (Bunescu, 2014;Willems, 1997).In the Swedish context, this assumption was used at the end of the 19th century when the authorities started to distinguish between Resande (Travelers)-allegedly a result of a racial mix of Roma and native Swedes-and Roma as an ethnic group.The authorities at the time viewed the Roma as strangers to be expelled from the country, while Resande were seen as a specific category of Swedish vagrants and hence were exposed to compulsory and repressive policies.Roma groups arriving before the turn of the 19 th century were described in public reports as undesirable outsiders (Montesino, 2002).When their entitlement to citizenship was recognized in the 1950s, the assumption of their identity as a homogenous group was reproduced, but at the same time the importance of the ethnical dimension was reduced and the Roma's social marginality in the Swedish welfare society was emphasized.As new citizens, they were registered as "one people"-the Swedish Roma ("svenska zigenare")despite the fact that ethnic registration was officially abolished after WWII (Axelsson, 2011).
After proposals from local authorities for the need to establish a central register, the National Board of Health and Welfare (Socialstyrelsen), created a small official national register (Socialstyrelsen, 1960).This national register was created using the information that the Board's in-house expert of the period, Carl-Herman Tillhagen, had gathered during the 1940s and supplemented while working on the 1954 inventory (SOU, 1956).The Board later recommended that the social authorities at the local level should use this national register for background information in all decisions involving the Swedish Roma.Later this register was used in a detailed and comprehensive socio-medical examination of Swedish Roma (Ohlsson Al Fakir, 2015).
The practice of registering the Roma in Sweden thus became institutionalized within different governmental contexts.A register was also accessible to a researcher, who, much later, reproduced the classification constructed by the authorities during the post-war decades (Cf Arnstberg, 1998).Such practices are still in use; as recently as 2013, the existence of an "unofficial" police Roma register was discovered, and later denounced, in Scania in the South of Sweden.This register can be seen as a consequence of the internalized and taken-for-granted premises in the Swedish authorities' evaluation of Roma groups as belonging to a certain problematic category.Registration was perceived in this process as a prerequisite to relocation.Hence, the police register must be interpreted as a continuation of the process of registering that was initiated at the national level after WWII, when the Swedish Roma became the target of author-ity intervention (Westin, Wallengren, Dimiter-Taikon, & Westin, 2014).

From Outsider to Socially Disabled
The second statement of the Gypsy Question-Roma as outsiders-has likewise been the starting point for a great variety of studies in Sweden demanding public intervention.Among others Tillhagen (1965) substantiated this statement by arguing that the "traditional" Roma were victims of an unavoidable development in which they had become unable to survive without the support of the authorities (SOU, 1956: 43;Takman, 1976;Tillhagen, 1965).In Tillhagen's view, the Swedish Roma were left in a hopeless situation and lacked the appropriate resources to overcome this helplessness.Tillhagen and other researchers (e.g.Takman, 1966;Trankell & Trankell, 1968a) included Roma in the specific category, the socially disabled, developed in the 1950s and that provided a large number of arguments for intervention.In practice the classification of Swedish Roma as socially disabled permitted their entry in the administrative welfare systems (Ohlsson Al Fakir, 2015).
Consequently the introduction of this category into the conceptualization of social problems signified a further expansion of the social area; it justified new interventions and permitted the expansion of old practices.This expansion-the knowledge and technologies it drew upon and reproduced-was partly driven by actors in the medical field and/or in medical institutions (Berg, 2009;Montesino & Thor, 2009).Hence, the expansion of the social area was at the same time an expansion of the medical field into the social area, which indeed should not necessarily be interpreted as a medicalization of social problems but rather as part of the establishment of a new conceptual (and practical) space where the social and the medical fields were inseparably intertwined.
This development presupposed a reconceptualization of social problems, from repressive and openly excluding strategies to differentiated strategies developed to handle-i.e.socially relocate-citizens classified as deviants.During the decades following WWII the category of disability thus provided arguments for the admission of new citizens, as well as arguments for interventions towards "old" citizens considered socially deviant (Montesino, 2012).The scientific justification and legitimacy of such arguments was provided by both researchers and practitioners in the social field, hence giving way to the establishment of new dimensions in the understanding of social problems.
The political scientist Deborah Stone has elaborated on the decisive role played by the disability category in the development and expansion of welfare policies (Stone, 1984).In the early 1900s, disability became the administrative category that "entitles its members to particular privileges in the form of social aid and ex-emptions from certain obligations of citizenship'' (Stone, 1984, p. 4).In this way, categorizing individuals as disabled provided a solution to the dilemma of redistribution.At the same time, the new approach established a temporal dimension in the politicaladministrative handling/management of the population; disability was a relative and "treatable" category, hence social relocation became possible.Originally borrowed from the area of clinical medicine, treatment and rehabilitation thus became the key instruments that would transform disabled individuals and groups into useful citizens and relocate them to productive areas of society.
This implied, in the words of Stone, that the authorities at the end of WWII "pushed new 'undeserving' to the side of 'deserving poor'" (Stone, 1984, p. 10), thus expanding the category of "deserving" and diminishing the category of "undeserving".The process added new social and medical dimensions to categorizations and classifications; from the previously dominant focus on medical aspects of the disability condition of individuals, disability was redefined to include also mental and social aspects.Social disability was related to different criteria: age (children and elderly poor), family situation (families with several children, single mothers), education (illiteracy), cultural belonging and cultural practices (minorities or non-European poor), etc.It was in this process of blurring boundaries that the incorporation of the new Roma citizens occurred in Sweden.Concretely, Swedish Roma were moved from the status of unwanted strangers to citizens' status and administratively relocated to the social services created to manage social deviance, i.e. the organizations formed to manage disabled citizens.In Sweden during the early post-war decades, such policies were primarily developed and implemented as part of the expansive labor market policies, including extensive vocational training programs (Ohlsson Al Fakir, 2015).
To be classified as socially disabled implied that the Roma (and other groups) were temporally entitled to social aid and to some other kinds of welfare support.At the same time it made them a target of interventions aiming at full incorporation into the obligations of citizenship, i.e. in the first place paid employment for adult men and compulsory schooling for children.
As disabled citizens the Roma thus became targets of rehabilitation activities, originally created by people processing organizations (e.g.sanatoriums and psychiatric institutions).In the decades following WWII, it was the expanding labor market authorities that organized most rehabilitation activities, notably within the rehabilitation treatment institution (arbetsvården) (Takman, 1962).Experts from the health and social area were engaged in these institutions to determine the extension and degree of the disability of individual adult Roma.The national authorities soon considered these experts to be the "Gypsy Experts" ("zigenarex-perter") par excellence.Hence, social and medical experts became key persons in formulating contemporary and future policy plans concerning socially disabled citizens such as the Roma in the post-war decades (Montesino, 2002(Montesino, , 2012;;Ohlsson Al Fakir, 2015).
Expert knowledge has been fundamental in the development of Swedish policies concerning social problems during the twentieth century (Lundqvist & Petersen, 2010;Montesino, 2001;Ohlsson Al Fakir, 2015).The experts engaged in the Gypsy Question after WWII were focusing on identifying the specific characteristics that made the Roma into disabled citizens, hence their alleged social disability was examined within different fields of expertise.The work of these experts resulted in a large number of professional opinions and detailed reports to inform authorities at local and national levels about the situation of Swedish Roma, individuals as well as families.Social workers focused on the economic situation and stated that Roma families were no longer self-supporting; social disability could also be the result of inadequate economic and social support.In socio-medical studies, researchers and social workers tried to calculate the degree of this disability, considering both medical and social factors (e.g.undernourishment, physical condition, disease, family situation, educational level, and housing situation) (Takman, 1962(Takman, , 1976)).Classifying Swedish Roma as socially disabled thus led to new, more detailed and comprehensive, social and medical evaluations, which would be used to develop public activities aiming at a social relocation of Roma citizens.
According to these socio-medical evaluations, adult Roma lacked the basic knowledge that was required for incorporation into the labor market.Illiteracy was identified as the most fundamental problem, which made the authorities develop interventions that would adjust these deficiencies.In line with this, experts proposed "rehabilitation measures", which included schooling and training to prepare adult Roma for the incorporation into the labor market (Takman, 1962).Professionals and experts from the educational and psychological fields subsequently added new aspects to the supposed disability of the Swedish Roma families, including parental capacities, children's learning capacities and women's maternal relationship to their children (Trankell & Trankell, 1968b).These reports conveyed the message that Swedish Roma were in need of external expert support: as parents they needed psychological supervision, while the Roma children were in need of special assistance to manage school as well as the family situation.This support should be provided by professionals, who supposedly would facilitate the entry of the Roma into the normality of Swedish society.How this evolved in the daily practices of local gatekeepers working in people processing organizations is the focus of the next section.

Managing the New Citizens
The incorporation of the category of disability into the administration of welfare led to the development of new institutions, e.g. the rehabilitation treatment institution discussed above.These institutions were seen as transitional places where the disabled had to be socially relocated into a new social context.Roma citizens classified as socially disabled had to adapt to the established routines, in many cases, developed for the treatment of other allegedly disabled groups, such as refugees (Montesino, 2012).In these institutionalized contexts Roma citizens would be prepared to fulfill the obligations of citizenship, i.e. to be educated in following the laws and adjusted to the mainstream norms concerning the settled population.Hence, citizenship obligations in this context spanned a settled lifestyle, employment within the regular labor market (mainly for the male adult Roma) and schooling for the children.To achieve these goals regarding the Swedish Roma, the authorities focused mainly on three different areas: housing, work training for adults and schooling for children.The Gypsy Question was thus reproduced in activities related to these specific intervention areas, which we discuss in the following sections

Housing
The housing situation of the group played a key role in the problematization of Swedish Roma after WWII.
If the Gypsies are provided with permanent, adequate housing and the children are brought up in an understanding way from the very first years of life, one may presume that the 'Gypsy problem' will cease to be a problem per se within a couple of decades.(Takman, 1952(Takman, , 1976, p. 11) , p. 11) Housing was perceived as a precondition for both regular work and school attendance; hence the housing "problem" was the first to be formulated.Before the war Roma were not included in Swedish housing policies.Harsh local and national regulations of mobility and local populations' discriminatory attitudes against Roma contributed to the situation described in the social debate of the 1950s (Montesino, 2002;Westin et al., 2014).The permanent settlement of Roma was identified as a necessary condition to solve the "problem".
In line with this argument the housing situation of Swedish Roma was object of both social and medical studies during the 1950s and 1960s, and it also became the main area of interventions during these decades.In the 1950s, Roma families got permission by some local authorities to establish themselves in stationary camps.Later, in 1960, the state began to reimburse local welfare authorities for certain kinds of welfare support to Swedish Roma families, notably such support that would improve the housing situation (Montesino, 2002).
The National Labor Market Board also decided to create a special housing improvement loan intended for persons belonging to groups living in deficient housing conditions, such as Swedish Roma, refugees or immigrants.The authorities thus "solved" the housing problem mainly by establishing financial incentives for local authorities to include the Roma on local housing markets.The local officials who distributed these loans based their decisions on professional opinions made by socio-medical experts, who had evaluated the situation of individual Roma and Roma families (Ohlsson Al Fakir, 2015).Social and medical expertise thus contributed to defining the contents of the authorities' activities intended to make citizens out of the "disabled" Roma.

Work Training for Adults
Once most Roma could be defined as settled, a new problem emerged concerning their maintenance; as citizens without employment they lacked the financial means to pay for their housing.In line with this problematization, the National Labor Market Board assumed responsibility for Roma policies from 1958.Among the activities developed by the Board were the settlement loans discussed above.Paid labor was another mainstay in these activities during the 1960s.Authority interventions must, again, be based on knowledge.Hence, social and medical experts also evaluated the health of adult Roma in order to certify them as either able-bodied or physically disabled.These experts also recommended suitable interventions on the individual level, e.g.medical treatments or rehabilitation, literacy classes, driving instruction, professional training or other vocational education.Scientifically certified interventions like those mentioned, were thought to solve the "problems" for both the individual, the family as well as for society at large.
During the 1970s, the authorities' activities expanded and became even more detailed and invasive.The 1960s interventions had mainly concentrated on making adult Roma employable through education and not on their family and personal life.However, as these interventions were deemed unsuccessful, new psychological and pedagogical experts as well as other professionals claimed that the core of the problem was the Roma's lifestyle and ("dysfunctional") habits at home (Trankell & Trankell, 1968a).As a consequence, interventions during the 1970s focused partially on changing the Roma's private behavior as parents, wage earners and home makers (Ohlsson Al Fakir, 2015).
One of the instruments used to achieve these kinds of personal and familial changes was Adaption to the daily life (ADL).This was a method originally developed by occupational therapists in the 1940s in order to take care of patients with chronic diseases (e.g.physical disabilities) and cognitive disabilities (e.g."mentally retarded"), etc.In health institutions, ADL training had the aim of making the users competent in most simply daily matters (Marta, 1979, p. 15).The method was expanded during the 1950s to include social work with poor families.In such ADL work, social workers visited families and taught them the accepted routines for organizing family life, like getting up early in the morning, following established routines for mealtimes, sending children to school, etc. (Liljeroth & Niméus, 1971).The Swedish welfare authorities recommended that the practices developed in ADL training for individuals with cognitive disabilities should also be used in social work with Roma families (SOU, 1956: 43).
The intervention plan that was formulated in the 1970s took the form of an educational project intended to make Roma adults more attractive for the labor market.The educational plan included literacy classes and ADL courses for the adults, which were later imparted to Finnish Roma and other newly arrived Roma groups (Marta, 1979).To achieve the authorities' objectives, specialized social workers, so-called family therapists, worked with Roma families in their homes (Iverstam Lindblom et al., 1978).The family therapists were expected to have a mediating function between the Roma families and the authorities, focusing on transmitting the routines of what was considered a normal daily life.This mediating function was later extended to school teachers, neighbors and other social workers (Turunen, 1984).Today this function has been conferred to a new mediating category, socalled "brobyggare" (bridge builders), that is, Roma adults who are supposed to have the "cultural competence" required to mediate between members of their own ethnic group and welfare staff in different contexts.The notions of helpless and less competent, i.e. handicapped, Roma is obviously still prevailing in the politics of practice concerning Roma in Sweden.However, mediating strategies does not only apply to the (allegedly dysfunctional) relationship between Roma and local professionals, but also to the relationship between parents from different migrant groups and the teaching staff (Alfakir & Lindberg, 2004).
From the above related activities developed in the areas of housing and work, we draw the conclusion that gatekeepers, i.e. professionals and experts that worked with "deviant" (potential) citizens, assumed an authoritative role in the determination of how the entry process of Roma into Swedish welfare society initiated in the 1950s should be organized.They also contributed to define the content of the practical interventions.As a consequence multiple activities emerged and made the inclusion operation into a prolonged processing of Roma as potential citizens.Roma children were considered as strategically the most important target groups in these activities.

Schooling for Children
Towards the end of the 1960s, local authorities and experts maintained that, despite settled families and ongoing work training for adults, Roma children remained outside the school system.The identification of children as a special target group reflects a general perception of childhood as a period for investment in the future (Trankell & Trankell, 1967).This approach has justified the institutional violence (compulsory care and other maltreatment) against children from poor families and/or children from ethnic minorities (e.g.Westin et al., 2014).However, in the period under study, compulsory childcare started to be questioned and different methods of "care in the family" were development by the social authorities.Focus now moved to the schooling of the children and the compulsory elements assumed other forms.
The schooling of Roma children was part of a prolonged administrative process involving pedagogic expertise, teachers and social workers.In these processes, some Roma children were put in small groups, in so called "educational clinics", for children with learning disabilities.Other Roma children attended ordinary classes but they also had to attend the "educational clinics" (Trankell & Trankell, 1968a).In general, Roma children were considered children with special needs having enormous difficulties compared to most other children.At the same time the school authorities saw the school and education as the key strategy in solving the Gypsy Question.
To support the children in their schoolwork, a mobile school clinic was established in 1966 in Stockholm, with 35 children participating in the project (Román, 1992, p. 32).The aim of the mobile school was to help the children to adapt to the school routines and prepare them for the integration into normal school activities.
The work in the mobile school consists to a large extent nowadays of repairing the effects of the Gypsy parents' inability to prepare their children for school (Trankell & Trankell, 1968b, p. 14) As the authorities concluded in the 1970s that a project only aiming at school age children was not enough, kindergarten activities for Roma children were put on the agenda and outreach activities started to be implemented.These consisted of school staff from educational clinics for the older children working at home with both the children and their parents.Hence the parents also became targets of the educational work, which involved the initiation of yet new projects.
Managing the social relocation of the Roma in Swedish society resulted in an extensive administration.
New measures led to further activities in which the contents of the Gypsy Question were reproduced.Housing, work training and schooling of Roma became institutionalized activities where the assumed social disabilities of Roma were differentiated and reproduced.These activities were extended to the Finnish Roma immigrants in the 1970s and later to Roma refugees from Eastern Europe and former Yugoslavia; literacy teaching for adults, special support for school children, social projects for Roma youths, family guidance activities, etc., were established on a routine basis (Cf. Iverstam Lindblom et al., 1978;Kaminski, 1980;Marta, 1979).

Concluding Remarks
This article concerns the process were the Roma became a field of intervention for the Swedish Welfare State after the Second World War.In this process, individuals identified as Swedish Roma were constituted as a special category of citizens, whom had all aspects of their lives examined and used as the basis for intervention by the authorities.In their respective work areas, experts and professionals produced knowledge that confirmed that the Roma were a deviant social group, and their presupposed deviance was constructed in terms of social disabilities.In the process of knowledge production, the statements of the Gypsy Question were repeated and adapted to the contemporary understanding of social problems.The incorporation of the new citizens was thus integrated into the Social Question as expressed in the Gypsy Question.This was first described in local and national reports, later it was repeated in the specific practices developed to find the right methods to solve the question.The inclusion of the Swedish Roma was thus conceptualized as a social problem to be solved; the group's inclusion in the category of disability provided justification for the practices initiated during this period.The inclusion of the Swedish Roma turned out to be a long process that had to be planned and supervised by professional experts working at different levels and presenting arguments that had already been repeated for many decades.Their activities contributed to the expansion of the social area, and in this expansion the Roma were processed in different administrative contexts.We have explained this in terms of the logics of people processing organizations, which follow stepwise procedures.The aim of the authorities was the social inclusion of the Swedish Roma, but the inclusion process took another course: Swedish Roma were continuously labeled and treated as socially deviant citizens.When analyzing the history of Swedish Roma policies it is necessary to take into account how these policies contributed to the construction of excluding spaces within the organizational boundaries of the nation state.Focusing on these spaces gives a more nuanced view of the Swedish welfare state.With reference to the Gypsy Question, the period 1950 to 1970 is a good example of how a discourse may change while the practical contents of welfare policies remain the same.The prolonged process of Roma inclusion is primarily visible in the practices developed at the local level.We have focused on a period after WWII, but we see indications that the same ideas are reproduced in problematizations and local activities regarding the Roma and education today.From 1999, when the Roma were recognized as a Swedish national minority, policies and practices regarding their inclusion have been formulated in terms of rights.However, at the same time Roma continues in focus of interventions very similar to those of previous periods.