The problem of asymmetric representation: The marginalisation, racialisation, and deservedness of Roma in Slovenia

Special rights (such as a right to representation) are often used to address under-representation of minority and marginalised groups in political institutions. The case of Roma community in Slovenia is no different. Using a framework of social and political marginalisation, our article outlines asymmetry and exceptionality in the existing minority protection provisions for Slovenian Roma. We argue that the existing provisions are a consequence of majority’s perception of Roma as racialised, marginalised, and undeserving group. We show that the limited access to special rights for Roma is not an oversight in the Slovenian democratic institutional design but a result of deep-rooted stereotypes and racialisation that this marginalised group is subjected to. More broadly, the case of Slovenian Roma demonstrates how principles of racialisation and marginalisation can help us understand absences of different marginalised minority groups from frameworks of special protection.


Introduction
Roma are probably the largest minority group in Slovenia. The need to protect this group was recognised in the 1991 Constitution -the founding document of the new independent Republic. Yet Roma political rights are inferior to the other two named national minorities (Italians and Hungarians). This is particularly significant, we argue, because Roma are not only a minority but also a marginalised group. By marginalisation, we mean a group who to 'varying degrees, exist politically, socially, or economically "outside" of dominant norms and institutions' (Cohen, 1999: 37) and unless special provisions are provided for such groups, they remain outside political realm subjected to racial abuse and deprivation.
The Slovenian Constitution grants special status to three minority communities demonstrating a willingness to acknowledge issues of under-representation. However, one of the three communities lacks equal access to political institutions. This article aims to determine how Roma were excluded from the 'special access' provided for the other two minorities, and how this exclusion is still reflected in Slovenia's governing structures today. We argue that Roma's democratic disadvantage stems not from their minority status per se but from marginality and racialised discourses which underpin institutional design for minority protection. The founding constitution is grounded in liberal notions of minority representation which recognised the need to grant special rights to minority populations (Bešter et al., 2017). The absence of Roma from the list of minorities would seem less strange than the peculiar halfway house solution that Slovenia manufactured. We argue that this case presents a particularly intense problem for a community known to lack political power, suffering from stigmatisation and racialisation and, as McGarry (2017) shows, confronted by discriminatory and inflammatory media representation.
Some have drawn attention to inadequate representation of Roma in Slovenia (Bešter et al., 2017;Žagar, 2017) but while their analysis aids the understanding of the political status of Slovenian Roma within the existing framework of minority protection laws, it ignores the marginalisation of the Roma community and how 'minority provisions' perpetuate marginalisation.
Our approach is grounded in critical theory and unstructured modes of discourse analysis which loosely aligns with Foucauldian critical praxis. For Foucault (1981: 48), discourse or praxis comprises of rules, system, rituals, processes, and procedures which constitute and are constituted by power (truth). Discourse is not a bearer of truth/meaning but a violent imposition through which 'subjectivities are created and bodies acted upon' (Manchanda, 2020: 18). While ideas of marginalisation, racialisation, and deservedness build the article's theoretical framework, they also guide analysis of the selected material. They highlight moments of contention, and reveal biases and stereotypes which discursively construct Slovenian Roma as racialised subjects. Using collage or montage-as-method (Manchanda, 2020;Shapiro, 2012), materials for the analysis were selected: following the established literature on minority protection, we selected key legal documents concerning Slovenian Roma (Bešter et al., 2017;Žagar, 2017); the minutes of the Commission for Constitutional Questions (CCQ, 2001) provide insight into how Roma were perceived at a point when minority protection laws were discussed, while publicly accessible interviews with Roma politicians and key political institutions provide insight of the continuing racialised marginalisation of the group. We analysed the selected material using unstructured content analysis and applied Foucauldian praxis to determine the contexts (including biases and stereotypes) in which Roma are discussed and the intended outcomes. This is not an exhaustive account of how the discourse about Roma is constructed. However, it is an important insight into how Roma are racialised at a politically significant moment and how that racialisation translates into political institutions and continues today.
The article opens with an outline of analytical framework which builds on theories of marginalisation, racialisation, and deservedness, and sets out approaches to representation (wide and narrow) for minority and marginalised groups. Second, we outline representative legislation concerning Roma in Slovenia, and compare it to the political rights of other minorities. Third, using critical praxis, we analyse the selected materials to determine how constitutional discourses construct and represent Roma. Finally, drawing on the theoretical framework, we outline the asymmetric political design that creates Roma as an exceptional group and explore how this exceptionalism is a product of marginalisation and racialisation that permeate political mechanisms governing 'Roma issues'. The case of Slovenian Roma demonstrates how the minority framework can precipitate political and social marginalisation of the community the framework is supposed to protect and why an analysis grounded in racialisation and marginalisation is essential for understanding the institutional set up and political rights of marginalised groups in general.

Marginalisation, racialisation, and deservedness in the discourse of representation
The under-representation of minority and marginalised communities is not uncommon in democracies. While representational design may be used to address both conditions, the significance of protective representational rights for marginalised groups is particularly important. Marginalisation adds extra complexity to representation and protective rights. Cohen's operationalisation of marginalisation is central here. She defines a marginal group as one that exists in a state of socio-political deficiency without the resources to reverse it. Such groups are '(d)enied access to the resources and skills that allow for substantial participation in decisions about the quality of life' and 'often find that their members lack access to resources such as political and social capital' (Cohen, 1999: 37).
The experience of marginalisation is common; the marginal is often excluded, 'othered', or denied the everyday capital required to participate in the political sphere on 'equal terms' and refused 'access to rights and privileges enjoyed by the majority group' (Cohen, 1999: 38). Marginalisation commonly impacts minority ethnic and racial groups or precarious social-economic classes. It leads to particularly acute experiences of disempowerment (Cohen, 1999;Shilliam, 2018). As a concept, it has been used to understand the disempowerment of Black Africans in South Africa, African Americans, Dalit and Muslim groups in India (Cohen, 1999: 37;Ahuja, 2019) the undeserving working class in the United Kingdom and the industrial West (Shilliam, 2018), and Roma/Travelling communities across Europe (McGarry, 2017).
Social identity -how a marginal group is perceived by the majority -is a significant indicator of marginalisation. Often social identities and representations of a marginalised group are portrayed as illegitimate, abnormal, unhealthy, and even criminal (Cohen, 1999: 38). Marginalised groups are deemed to deviate from standard norms upheld by the majority, so their everyday practices or ways of life are considered inferior or lacking in authority and organisation. Such construction of communal social identity is dependent on the outside gaze; how the majority sees the group, and perceives its difference as foreign, abnormal, or inferior. In other words, social identity is a clear expression of dominance and superiority that stigmatises and racialises the marginal group.
The effect of 'majority gaze' upon marginal groups has been identified in Black and Indigenous communities globally but also of Roma and Traveller communities across Europe (McGarry, 2017). We adopt Shilliam's (2018: 4) operationalisation of racialisation, which 'refers to the way racist attributes and hierarchies come to determine the everyday meaning and common sense valuation of an entity or phenomenon'. In that sense, a normative framework in which a marginal group gains its social identity, its 'deviant' and other characteristics, is racialised as White and superior. A marginalised group is always seen as somehow lacking in relation to 'whiteness' or dominant White ideals. Furthermore, Shilliam (2018: 4) discusses the link between racialisation and otherness in his work Race and the Undeserving Poor claiming that social identities or representations of marginalised groups are always racialised as either those who are considered to be deserving or others who remain undeserving. The deserving and undeserving dichotomy mirrors the concept of belonging. Those who are perceived as belonging to the wider social polity are often racialised as deserving; whereas outsider others are often racialised as undeserving. Crucially belonging is determined by perception and not conditioned on legal documents or rights. For example, being a holder of Slovenian citizenship does not protect Roma from being perceived as not belonging to Slovenian civil society. As Cohen (1999: 38) states, belonging is primarily dependent on being perceived as sharing the same normative ideals. Shilliam (2018: 15, 34, 61, 71) distinctly identifies normative values such as self-organisation, self-help, prudence, industriousness, settledness, and patriarchy as deserving. The undeserving not only lack these values, but threaten the very organisation of the given society. As we show in the final section of the article, the asymmetric institutional design in Slovenia distinctly racialises the Slovenian Roma as undeserving.
Marginalisation can be upheld by law, institutions, and violence. The role of the institutions and the law is particularly key for our concern. Drawing on Fanon, Cohen (1999: 44) states that laws and institutions exclude and regulate groups. While marginalised groups can indeed begin to receive institutional recognition and rights, those rights -as we show in the case of Slovenian Roma -mirror White/dominant imaginaries about how the marginalised can and should organise and stem from perceived inhibitions around the inherent limits of acceptable political organisation. Institutional design through which marginal groups receive some rights of recognition can also suppress unwanted political organisation. Leaders within marginal groups can be co-opted into the dominant system of representation, be socialised in the hegemonic normative frame, and become 'agents' of the dominant order. They begin to see themselves in the same way as the majority sees them. Cohen (1999: 52-59) calls this internal marginalisation, where 'indigenous elites' capture available resources and further marginalise their own communities, extending the division between those deserving and undeserving of support (and resources). Internal marginalisation relies on personal relationships between the elite and marginal sub-groups but also on the acceptance of the dominant ideals that the marginal political elite internalised. A capacity or desire to 'become white' (Seshadri-Crooks, 2000) can emerge as an important factor dividing and uniting different components of the marginalised group.
Marginalisation with embedded tendencies to racialise groups as un/deserving is a complex political problem, which cannot be resolved solely through policy or institutional design changes. In democratic theory and practice, representation has been established as an effective method of addressing political visibility of certain groups, be they minorities, marginalised, or both (Reynolds, 2005). We adopt one of Pitkin's wider definitions of representation: [P]olitical representation is primarily a public, institutionalised arrangement involving many people and groups, and operating in the complex ways of large-scale social arrangements. What makes it a representation is not a single action by any one participant but the over-all structure and functioning of the system. (Pitkin, 1967: 221) This allows a broader understanding of representation which does not focus solely on who represents and who is represented, or on the institutional design in the representative bodies. It also considers non-representative political bodies and structures open to, or dealing with, marginalised groups such as interest groups or media.
A narrower understanding of representation focuses on institutional design and increased participation of under-represented groups in the political system. Hard-to-reach groups can be encouraged to participate in political institutions in many ways. Political parties and political bodies might use quotas to give under-represented groups increased presence in organisations (Krook, 2010), or communal tensions might be addressed through reserved seats in an institution on ethnic, linguistic, or cultural grounds (Bird, 2014). Parliaments in India, New Zealand, Niger, Romania, and Venezuela all grant reserved representation in this way (Htun, 2016), usually to better reflect the shape of society and improve legitimacy. As Reynolds (2005: 303) asserts, 'lack of descriptive representation indicates the exclusion of important minority interests from government'. Young (1989) even suggested that granting representatives the power of veto over policies that impact on their communities could provide extra security for the interests of the marginalised.
While Slovenian institutional design, which reserves seats and special institutions for minority groups aligns with a narrow definition of representation, the distinction between wide and narrow representation is in fact blurred. Phillips' (1995) politics of presence asks if formal equality can provide for equal representation in public life. In particular, if minorities are invisible in political decision-making 'their very presence [would] make it more likely that members of dominant groups will recognise and speak to their concerns' (Phillips, 2001: 26). Equally, the benefits of special access for minority/marginalised groups are contested. McGarry andAgarin (2014: 1987) outline a 'participation puzzle' for European Roma -a dilemma between gaining political visibility and highlighting the group's difference. Studies on political representation of Roma in particular point to the dangers and ineffectiveness of institutional representation for Roma groups (Vermeersch, 2003).
A resolution of a participation puzzle for marginalised minority groups might be elusive; however, this lens allows us to acknowledge that 'inclusion is the antidote for invisibilisation', as Htun (2016: 171) argues in her work on Indigenous peoples in Brazil, even if such presence strengthens the difference of the marginalised community or creates a marginalised political elite. As Jensenius (2017) shows in her study of Scheduled Caste (SC) representation in India, inclusion normalises the presence of a marginalised group in the polity. While it does not eliminate discrimination, it legitimises group presence. This resonates with a broad definition of representation -what McGarry and Agarin (2014) might call participation -whereby the visibility in the institutional design impacts how the group is publicly perceived, how it is represented by the media and majority perceptions.
While commonly minority status rather than marginalisation drives the call for special access, the case of Slovenia highlights why marginalisation is significant in the context of political representation and demonstrates how Roma in Slovenia are politically racialised and marginalised.

Reserved representation and Roma: Slovenian legislation and the problem of asymmetry
In the last census permitting questions about ethnicity, 3246 Slovenian citizens declared themselves as Roma (Census [Popis], 2002). More recent unofficial estimates suggest that the number of Slovenian Roma is closer to 10,000 (about 0.5% of the entire population) (Amnesty International, 2017) -a small number compared to neighbouring countries but significant nevertheless. So much so that Slovenia's constitution recognises Roma as one of only three minorities with special status -Hungarians, Italians, and Roma. Drawing on the historical context and key legal documents, this section demonstrates how political rights of Slovenian Roma were created and how they remain undeveloped.
Unlike many other Roma/Traveller communities, Slovenian Roma have largely abandoned nomadic lifestyles and settled in permanent neighbourhoods that can vary massively in terms of quality of life. Some settlements -typically on the outskirts of urban centres -create their own small districts with a self-governing body that integrates into local political infrastructures. Others resemble camps consisting of part-built wooden huts without electricity and running water. These settlements are often constructed illegally on municipal or private land, and disputes over territory and access to utilities are common. Such living conditions underscore why Roma remain by far the poorest and socially vulnerable community in Slovenia (Amnesty International, 2017). Following Cohen's (1999: 37) definition, Roma in Slovenia are the epitome of a marginalised group -highlighted by every indicator of exclusion and disadvantage, including economic inactivity, living standards, incarcerations rates, youth pregnancies, and school drop-outsand are subjected to negative media representation and everyday experience of racism and xenophobia (Bračič, 2016). The everyday experience of Roma marginalisation can be measured through low levels of formal education and high rates of unemployment (National Programme of Measures for Roma of the Government of the Republic of Slovenia for the Period 2010Period -2015Period , 2010. Commonly pushed to the periphery of political and social life, Roma lack the resources and political and social capital to effectively challenge and change their status. Politically, Slovenia differentiates between three different groups of Roma -autochthonous Roma, 'new Roma' (those who came to Slovenia in times of socialist Yugoslavia to work or during the 1990s wars), and a small group of Sinti (Office for National Minorities, n.d.). Laws concerning minority protection and political rights are chiefly concerned with the so-called autochthonous Roma only (said to have historic ties to Slovenia, their presence can be traced back to the 13th century). Claims for political rights of autochthonous Roma are driven by the existing minority protection laws. Considering that minority protection laws for Italian and Hungarian communities are written in the Slovenian constitution, this is unsurprising. Article 64 for Italian and Hungarian minorities and Article 65 for the Roma community address minority protection explicitly. Article 64 guarantees Hungarian and Italian reserved political representation at a national level, the use of national symbols, the use of their mother tongue in public office and in education, and support for economic, scientific, and cultural development. In addition, Article 5 makes a declaration to 'uphold and guarantee' the rights of autochthonous Italian and Hungarian ethnic communities, while Article 11 states Italian and Hungarian languages as one of the official languages in regions with minority population (Slovenian Constitution, 1991). In contrast, Article 65 acknowledges special status for the Roma community, but does not specify the nature or extent of their rights. It simply states that a legal framework, the Roma Community Act (2007), will further define Roma rights. This is the only constitutional acknowledgement of the Roma community.
Despite the 1991 constitutional requirement for special protection, a comprehensive law regulating special rights for Roma was only passed in 2007, full 17 years after its need was constitutionally recognised. Meanwhile, in 2002, Local Government Law established the role of Roma reserved representatives in local/municipal councils. These councillors are elected in 20 municipalities with historical Roma settlements. Thus, Slovenian Roma gained seemingly similar local political rights to the other two minorities (Bešter et al., 2017). This includes a dual voting right, whereby members of Roma community in a municipality with a Roma councillor can cast two ballots -one vote for a local councillor representing their community and one for their preferred candidate in elections for other local councillors (ZRomS-1). However, there are important limits to the status of Roma councillors. Unlike Italian and Hungarian councillors, they do not have veto rights -they cannot stop decisions that could affect their communities. Furthermore, Hungarian and Italian minorities can maintain a cultural, ethnic, and linguistic identity in public roles and through political participation, but there is no such allowance for the maintenance of 'Roma identity'. Roma language is unrecognised in public institutions; in fact, the knowledge of Slovenian is highlighted as the primary benchmark of Roma's literacy (Bešter et al., 2016).
While all three communities are recognised as autochthonous, only Italian and the Hungarian minorities are national communities, whereas Roma are classified as a special community (Office for National Minorities, n.d.). This distinction permits different levels of protection and leads to asymmetric political representation. The comprehensive effects of the asymmetric rights afforded to Roma compared to Hungarian and Italian Slovenians are presented in Table 1.
This overview of minority provisions shows how political rights constitutionally granted to selected minorities in the Slovenian constitution are asymmetric. Only two of the three communities (Hungarians and Italians) enjoy a complete set of rights. They each have a minority member of National Assembly and minority bodies effective on local, regional, and national levels. Neither Hungarian nor Italian communities are marginalised in economic or social terms to the extent of Slovenian Roma. Thus, despite being named in the independence constitution as a minority needing protection, protective rights of Slovenian Roma remain limited. We now move on to analyse how this peculiar institutional arrangement came to be.

The constitutional grounding of asymmetric representation: Roma as an exceptional community
In drafting the inaugural constitution in 1990-1991, the political establishment carefully considered the legal status of national minorities, and Roma in particular. Two considerations framed the outlined organisation of political rights and differences between Roma, and Italian and Hungarian minorities. First, the international context: between 1986 and 1990, Conference on Security and Co-operation in Europe (CSCE) (Organization for Security and Co-operation in Europe (OSCE)) meetings in Vienna and Copenhagen set the foundations for minority (including Roma) protection (CSCE, 1990). In 1993, the Council of Europe famously referred to Roma as a true European minority, 'one that does not fit into the definitions of national or linguistic minorities' (Council of Europe, 1993: Recommendation 1203). As Zevnik and Russell (2020: 52) argue, 'the outcome of these debates was to accept that Roma need to be recognised as political agents in the states in which they live'. States with Roma populations, including Slovenia, felt compelled to make progress towards inclusion of Roma in the decision-making processes at least at the local level (OSCE, High Commissioner on National Minorities, 2000). Second, the historical context: Italian and Hungarian minorities were mentioned in 1974 Constitution of the Socialist Federative Republic of Slovenia (1974). Articles 250 and 251 granted them linguistic and cultural rights and rights to self-organisation in municipalities in which they Yes -elected municipal minority council.
No Sub-national/regional level Autonomous Organisations Yes -a self-governing regional minority council.
State funding for activities within the remit of the minority Council including cultural events, language and bilingualism, and also economic and social support for the development of the region (Self-Governing Ethnic Community Act, 1994).
Yes -a self-governing regional minority council. State funding for activities within the remit of the minority Council including cultural events, language and bilingualism, and also economic and social support for the development of the region (Self-Governing Ethnic Community Act, 1994). No -there is no self-governing Roma minority region.
CRC is a national and state-funded advisory body for Roma with limited powers -it may only offer opinions or start initiatives concerning specific matters that pertain to the status of Roma.
CRC: Council of the Roma Community.
officially reside. The 1974 Constitution however has no mention of special political rights such as reserved representation in the Parliament or municipal council, only the 1991 constitution introduced those rights (Zevnik and Russell, 2020). A very different picture emerges in debates over whether the same rights should be extended to Roma. The minutes of the Parliamentary Commission for the Constitutional Questions (a body tasked with writing the new independence constitution throughout 1990) reveal ongoing communication between members of the Commission and representatives of Roma. The Italian and Hungarian minorities were thus not the only groups lobbying for a more visible role in the new Republic (CCQ, 2001: 716-720). While they succeeded, Roma did not. The process of writing the first constitution of the independent Republic is very well documented in the minutes of the CCQ. Particularly striking is a discussion about the mechanisms for minority protection set out in Articles 64 and 65. The Commission first considered the rights of Hungarian and Italian minorities. Supported by the rights inherited from the 1974 Constitution and the emerging regimes of international minority protection, the full extent of rights for Italian and Hungarian communities was uncontested (CCQ, 2001: 687). Instead, the focus was on defining and essentialising the differences between minority groups. National community and autochthoneity emerge as two key criteria in determining the extent of minority rights. At the very outset, the term national community is proposed to draw a distinction between the Italian and Hungarian communities and others, including Roma. In doing so, they highlighted the state's obligation to protect the two groups, secure their rights (including political and legal subjectivity), and safeguard from assimilation. This level of protection should not be afforded to other minorities, immigrants, or ethnic groups (CCQ, 2001: 667-669). The justification explicitly says that the term 'national community' allows to gradate between different minority groups residing in Slovenia affording them different rights (CCQ, 2001: 687). Justifications for replacing the term minority with national community are grounded with references to the existing literature in sociology (CCQ, 2001: 670-671), the examples of African American communities in the United States (CCQ, 2001: 670), perceived differencescultural and historic -between the two minorities and other ethnic groups (CCQ, 2001: 674-675), as well as acknowledgement that 'the term minority is derivative of the nation state ideology', which is considered as something that should be avoided (CCQ, 2001: 669). Considering these justifications for the use of the term national community -especially the objection to nation-state ideology -it is thus even more important to note that the proximity of the 'mother state' becomes a factor when defining the rights afforded to Italian and Hungarian minorities and those of Roma (CCQ, 2001: 679). Roma cannot claim the status of a national community because they do not have a mother state (CCQ, 2001: 679), despite fitting with other explanations (sociological, historical) that the Commission used to justify the use of the term.
The second criteria concerns autochthoneity. The key effect here was to limit the rights of minority groups (national communities) to those with long-standing ties to the territory and existing political presence (CCQ, 2001: 673-675). Italian and Hungarian communities are automatically perceived as autochthonous (CCQ, 2001: 674-675), whereas other minority groups are not. This distinction enables differential protection of Italian and Hungarian communities on one hand and other minority groups (mostly from the former Yugoslav republics) on the other. Questions were raised about the status of small groups of autochthonous Serbs and Germans, but the Commission deemed these groups lacked the necessary existing political subjectivity to legitimately claim rights (CCQ, 2001: 670, 679). Autochthoneity, however, applies to Roma communities as well, as some have acknowledged (CCQ, 2001: 695), as does emerging political subjectivity (CCQ, 2001: 680, 716). Some even said that Roma's relative political invisibility was down to direct and indirect suppression (CCQ, 2001: 718) but despite that, it seems that maintaining the distinction between the Italian and Hungarian communities and Roma was politically expedient and crucial. The Commission underscored its wish to 'gradate' the rights of these groups, whereby Italian and Hungarian communities were afforded full protection and Roma 'something else' (CCQ, 2001: 670). The possibility of Roma gaining the same rights as Italian and Hungarian communities was not excluded completely but the protection framework for those rights was not provided by the constitution (CCQ, 2001: 679, 681). Despite fitting into both categories set up by the Commission on which political rights were afforded to minorities, Roma remained without the same level of protection.
The analysis of the CCQ (2001: 712-723) minutes reveal that discussions about the status of Roma oscillated wildly. The final compromise was a constitutional acknowledgement of Roma as a special community without constitutionally safeguarded rights (CCQ, 2001: 713). The minutes demonstrate that the driver of this compromise was grounded in Roma's difference from the majority, the other two minorities, and other ethnic groups. In other words, Slovenian Roma were racialised into a separate category all of their own. For example, the Commission accepted that Roma are organised (CCQ, 2001: 690-692), but in a way that did not fit within the existing majoritarian understanding of group organisation or political subjectivity (CCQ, 2001: 712-713). Despite fitting the categories created to privilege Italian and Hungarian minorities, Roma were deemed incapable of enjoying the same political rights. Commission noted that granting Roma equal access as the other two groups would be politically unacceptable (CCQ, 2001: 712-713). Scholars even coined the phrase 'political adolescence' to describe the perceived inability of Roma to properly participate in national politics (Komac, 2007: 38). In other words, the Commission tasked to write the new Constitution crafted Roma as an exception. They sought ways in which they could delineate between the characteristics of different minority groups and attach to them different mechanisms of special protection with Italian and Hungarian communities deserving full set of minority rights and Roma a halfway house. Such asymmetry depended on claims about Roma's inability to fully participate in politics (CCQ, 2001: 712-713), their apparent lack of organisation or expressed political will (CCQ, 2001: 718), and, bluntly, their otherness (CCQ, 2001: 719). While some did propose equal protection for Roma and Italian and Hungarian minorities, they also acknowledged that existing power relations were not supportive of this outcome (CCQ, 2001: 713). These arguments are strongly anchored in logics of marginalisation and racialisation.
The marginalised logics saturating constitutional design translated into the peculiar institutional arrangements for Roma, most notably the aforementioned laws from 2002 and 2007. Returning to Table 1, we note four examples of how minority rights secured subsequently to the drafting of the Constitution were applied asymmetrically to perpetuate Roma's exceptional status. Each example demonstrates that Schattschneider's (1960) famous mobilisation of bias consistently disadvantages Roma.
The first example is the lack of a clear chain of command for local Roma representatives which weakens their effectiveness and makes the organisation of Roma beyond municipalities difficult. Roma have no overarching minority governing body bringing together the elected Roma councillors. The Council of the Roma Community (CRC) resembles Italian and Hungarian minority councils but with fewer powers. While the CRC can open communication with key governing bodies, it is limited to offering opinions, or starting initiatives concerning specific matters that pertain to the status of Roma (Article 12, ZRomS-1). The gap between the non-elected CRC that can act on a national platform and elected Roma councillors working locally remains unexplained and unaccounted for. Second, there is top-down interference in Roma issues. Despite significant Roma populations spread throughout the country only 20 of Slovenia's 212 municipalities have a Roma Councillor (Act Amending Local Self-Government Act, 2002). The criteria for the selection of such municipalities is unclear, it is supposed to overlap with the areas with autochthonous Roma population but there are several obvious omissions where significant Roma presence does not result in Council seats (Bešter et al., 2017). Outside these 20 municipalities, local governments can create their own working bodies to deal with Roma issues -but such working bodies need not even include Roma representation. Third, Roma representatives have restricted autonomy. Roma councillors have no Young-style veto over matters affecting their own community unlike Italian and Hungarian representatives who can exercise veto power at local and national levels. Fourth, the very method of providing for the Roma community in Slovenia is asymmetric. A sizable category of Roma in Slovenia is excluded altogether as legislation concentrates on autochthonous Roma at the expense of new Roma who will be represented at the local level only if they live among or are approved by autochthonous Roma. Sinti remain completely invisible. Slovenian constitutional law essentialises autochthonous Roma and creates a clear hierarchy of minority rights not only between constitutionally recognised minorities but also within the most marginalised community (ZRomS-1).
The political system of minority protection in Slovenia is commonly recognised as progressive. Yet, as our analysis of the institutional design shows, Roma were excluded from equal access to representation by the architects of Slovenian democracy and they continue to be locked out of participatory repertoires by the predominant culture of 'politics as usual'. Returning to the theoretical frameworks of marginalisation, racialisation, and deservedness, the next section unpacks the underlying logic of the set out institutional design. Bringing to the fore language, rights, and structures, our analysis shows how asymmetry is symptomatic of enduring marginalisation and racialisation. Cohen (1999: 37) asserts, '[marginalised] groups often find that their members lack access to resources such as political and social capital'. The analysis of representative structures for Slovenian Roma as a marginalised group reveals limited protective mechanisms and possibilities to interact with mainstream political institutions. Following McGarry (2017), we did not just look at the representation of Roma but rather at the representation for Roma. These forms of representation are interdependent, so the asymmetric structures of political representation mirror the prejudicial stereotypes that the majority tend to characterise Roma with. Thus, the four asymmetries embedded in the Slovenian institutional design reflect un/ conscious everyday stereotypes about Roma. The lack of legitimacy, transparency, or autonomy and a top-down approach align with the perceived lack of organisation, political agency, group's difference, and adolescence that curtailed the development of Roma rights already in the Constitution. These examples are symptoms of racialised bias and remain firmly within the logics of marginalisation and racialisation.

Asymmetric representation: A tale of marginalisation, racialisation, and undeservedness
Looking at the four examples, the lack of legitimacy and transparency concerns the disconnect between the different structures created for Roma representation. While political structures for Roma representation exist, each particular institution (the CRC or a local Roma councillor) lacks the power to produce real change. Unless institutions where marginalised groups can voice their demands have significant/transformative power, more often than not they only deepen the problem (Young, 1990: 41). The only official political body for Roma is the unelected and advisory CRC. No Roma political institution possesses a right to veto any decision that would adversely impact Roma. Such structures are ineffective in their communication with the state and create a non-elected 'political elite' with significant financial powers (distributing state funds among different Roma societies and projects) and very little accountability. Cohen (1999: 60) suggests that 'indigenous political elites' are given power because they demonstrated a level of amiability to accept and work within the existing political structures. Integrating these individuals into existing power structures creates an example of 'successful' integration and internalisation of majoritarian values and removes the potential threat these individuals might pose. Instead of leading a revolt against ongoing marginalisation and discrimination, an indigenous political elite turns into an example of successful integration into the majoritarian society.
Restricted autonomy and the top-down approach to Roma governance stall the capacity of Roma to organise politically and stratify the marginalised community further. Restricting Roma political organisation to 20 municipalities (Bešter et al., 2017) and the lack of a veto are two examples which promote cleavages and hostilities within the community and perpetuate tropes of political adolescent (manifested as a perceived inability to exert political agency in the expected way) and lack of agency. Shilliam (2018: 4) states racialisation frames the capacity of a marginalised community to organise politically. The President of the Forum of Roma Councillors asserts that racism and anti-Roma sentiment are frequent occurrences in official political chambers (Petrovčič, 2010), the experience which is also echoed by Jensenius' (2017: 179) study of SC representation in India. She concludes that despite overall improvement in their status, SC Representatives find it difficult to overcome the notion of 'pollution acquired by birth'. Just like SC Representatives, Roma political elite report feeling belittled and disrespected in official forums (Petrovčič, 2010).
While Shilliam uses racialisation to describe the distinction between deserving and undeserving, racialisation in the case of Roma follows two main pathways -external and internal racialisation. External racialisation shows how Roma are racialised in relation to other minority groups as well as the majority, whereas internal racialisation overlaps with intra-group marginalisation and hierarchies of deservedness between different Roma communities.
Asymmetric representation reflects external racialisation as Roma lack key political mechanisms to participate in politics because they are seen as departing from normative whiteness and less deserving than other groups to qualify for equal political participation. The architects of the Slovenian constitution referred to the Roma community as adolescent, lacking the political capital, organisation, and consciousness necessary to fully appreciate a full set of minority rights. Existing Roma organisation were criticised for inferior 'indigenous methods' (CCQ, 2001: 712-713) lacking legitimacy in the normative framework of Slovenian politics. A sense of paternalism was revealed in the inclination that 'we should not impose rights Roma did not ask for and force them to organise in a way uncommon to them' (CCQ, 2001: 713). This external racialisation bears clear hallmarks of undeservedness and racial stereotyping -the 'indigenous' community is immature and incapable of self-representation or incapable of effective interaction with the existing political institutions.
In contrast, internal racialisation de-legitimises political organisation of all but autochthonous Roma. If the constitution used autochthoneity to differentiate between the level of protection rights afforded to Italian and Hungarian minorities and those of Roma and others, autochthoneity in internal racialisation distinguishes between different Roma groups. While Roma remain undeserving of equal rights overall, the category of autochthoneity privileges one section of the group. It creates distinct hierarchies of deservedness. The rights of autochthonous Roma dominate the rights of new Roma and Sinti. Such distinction grants some Roma a limited privilege, a bargaining chip to level the existing distribution of power slightly. The result of internal racialisation can be seen in the proliferation of different Roma societies with similar goals serving either new or autochthonous Roma only and in Roma political elites who might support the existing division of rights or see the plea for political rights of new Roma as inconvenient. Marginalisation and racialisation of Roma are thus intrinsically interrelated -they create a Roma political elite who with the promise of upwards mobility internalise majoritarian notions of normality and view 'Roma social identity' as problematic, atypical, and requiring correction.
The institutions set up in the aftermath of the constitutional design continue to reflect the same tropes of marginalisation and racialisation. For example, in 2019, a new parliamentary Sub-committee for Roma Affairs (SRA) was set up to improve 'Roma lives' and the majority-minority relationship. The minutes of the meetings reveal that familiar stereotypes tend to fuel the committee's concerns (SRA, 2019(SRA, , 2020. The SRA is charged with how the state can ensure that Roma are law-abiding and do not disturb the majority population, while solutions are framed with regard to assimilation enforcing desirable changes in Roma communities (SRA, 2019(SRA, , 2020. This majority-led preoccupation is reflected in issues discussed -unemployment, petty crime, education, early pregnancyand in SRA's membership. Particularly striking is the presence of police (SRA, 2020), suggesting the primary concern is law enforcement and security in Roma settlements (Committee for Home Affairs, 2020) -a sentiment which can also be noted in the European Commission report (2019a: 203-204) on the integration of Roma in Slovenia. In contrast, suggestions that would allow Roma communities and their leadership autonomy to organise tend to be dismissed as idealistic -despite being a favoured tactic in European Union (EU) documents on Roma Inclusion (European Commission, 2019b). Meanwhile proposals to empower Roma through political and civic education delivered by the community tend to be dismissed due to lack of funding (SRA, 2019). The community itself is expected to invest in programmes for political empowerment that closely match the vision of the majority. Young (1990: 41) acknowledges this conundrum whereby marginalised groups have some structures through which they can make their voice heard, and yet their status remains unchanged. She calls this 'the normal process of everyday life'.
In her analysis, Cohen (1999: 38) asserts that illegitimate or stigmatised social identity of the group perpetuates and extends marginalisation even when more resources are available. Characterising activities as illegitimate or undesirable closely follows the logic of otherness as majoritarian society considers them a deviation from the norm. Shilliam (2018) shows how ideas of normality and respectability are racialised -and further, how 'the normal' is racialised as White and how that which falls outside normative whiteness is racialised as 'other/foreign'. Following Shilliam, we suggest that pre-conceived social identities of Slovenian Roma are ascribed by the majority population in racialised terms as undesirable and not-White (or at least insufficiently Slovenian). Such stigmatisation curtails any rights afforded to the community and helps to solidify the group's secondary status (Cohen, 1999: 38). The stigmatisation of identity and absence of desirable characteristics determine which groups deserve protection and which do not. As Shilliam (2018: 4) states, deservedness is a racialised category and rhetorics, which is used by elites to racialise and re-racialise the historical distinction between deserving and undeserving people and communities.
The asymmetry in political rights of Roma was the result of racialised preconceptions about the Roma community when the constitution was drafted and which persist today. The laws of 2002 and 2007 could have corrected the asymmetry but instead entrenched it. The unwillingness of the Slovenian political elite to address this imbalance can be understood through the lens of racialisation and deservedness. Roma in Slovenia are seen as a social problem, irresponsible outsiders that cannot be yet trusted with equal representational rights.

Conclusion
Our contribution demonstrates how the hierarchy of political rights can be racialised distinguishing between groups deserving and undeserving of protection. Full special rights are afforded to members of Hungarian and Italian communities in Slovenia, autochthonous Roma are granted partial rights, and other Roma remain invisible. Drawing on marginalisation and racialisation, we have demonstrated how special rights racialise the Roma community and reproduce their marginal status, rather than address it.
Reserved representation is the preferred technique of the Slovenian state to deal with minority under-representation. Yet it is not applied equally. The case of Slovenia has shown that the extension of rights in its current form cannot afford equal status for Roma. On the contrary, the existing provisions are part of the problem -they create and perpetuate the notion of Roma as problematic, exceptional, and undeserving. The existing theory of representation highlights limited transformative potential of special provisions (Jensenius, 2017;Pietsch, 2018), and creates possible dangers like the identity trap (Ahuja, 2019) or the affirmation of difference (McGarry and Agarin, 2014). Nevertheless, representation can also be a positive step. It normalises and legitimises the presence of a group in the national polity (Htun, 2016;Phillips, 1995). In societies where special provisions exist, the absence of a group from such provisions speaks to a deeper structural problem that de-legitimises the group's very existence and opens it to continued stereotyping and abuse. As we have shown in the case of Slovenia, the absence of Roma is not an oversight in the institutional design but a result of deep-rooted stereotypes and racialisation towards Roma as a marginalised group. The case of Roma in Slovenia demonstrates how principles of racialisation and marginalisation can help us understand absences of marginalised minority groups from frameworks of special protection and why halfway house solutions within those frameworks (even if well-intentioned) can stifle the group's scope for political agency for years to come.

Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.