Drawing on the work of Michel Foucault, governmentality scholars (Barry, Osborne, and Rose 1996) have argued that those present-day societies usually designated as ‘advanced liberal’ or ‘neoliberal’ govern their subjects by inciting them to act in specific ways and holding them individually responsible for their own wellbeing (Lemke 2002). Thus the incentives to work, study, care, consume or enjoy become the ultimate means of control. In an ‘advanced liberal’ society, one is governed not through restrictions, but through freedom (Rose [1989] 1999). Rather than exploiting or suppressing pre-given subjective content (interests, needs, motivations, desires), contemporary ‘conduct of conduct’ (Foucault 1982) creates or shapes the content of the self. This is the solution to the problem of government posed by the neoliberal delegitimation of state intervention in the market sphere, coupled with a wholesale marketization of all areas of life (of which numerous lucid examples can be found in Sandel 2012).

The result is a subject ‘who may be more desirous of its own subjection and complicit in its subordination than any democratic subject could be said to be’ (Brown 2006, 702). Indeed, it is much more difficult to resist subordination that comes from the ‘inside’, grounded in experiences of individual moral responsibility to study or work (against all structural odds), than one that is clearly identifiable as imposed from the ‘outside’. Accordingly, the techniques of government through shaping of subjectivity (‘subjectification’) draw their strength from the mere fact of remaining covered-up, black-boxed, ‘outside the formal control of the “public powers”’ (Rose 1996, 58). What facilitates their black-boxing is that they are not centralized but dispersed, embedded in diverse discourses, institutions, procedures and material artefacts. They work through complex ‘assemblages’ (Deleuze and Guattari [1980] 2004; Latour 2005) of humans and non-human entities. Accordingly, the tracing of the networks and processes that shape subjectivity is an effective tool for immanent critique by showing that the neoliberal society ‘does not live up to its self-portrayal [of taking subjectivity as it is] because it relies on molding subjects to an extent that remains unacknowledged by most neoliberal thinkers’ (Biebricher and Johnson 2012, 211).

In the domain of social policy, neoliberal rationality underpins a transition from welfare to workfare (Dean 2007, 577–578). The attendant approach of ‘active’ labour market policies makes social support (benefits, care) for working age adults conditional on preparing for (e.g. through appropriate education or training) or engaging in paid employment. At present, social policy measures that render welfare recipients individually responsible for their wellbeing enjoy global acclaim:

the notion that the state should play a reduced role in the provision of services and that individuals should take on greater responsibility for their own lives is, of course, the central tenet of the neo-liberal common sense which now underpins the programmes of most political parties the world over. (Ferguson 2007, 394)

The desires to be active, to work or to study are not reducible to functions of neoliberal assemblages – rather, neoliberal ‘activation’ shapes such desires to better serve the ends of responsibilization and welfare state retrenchment. Workfare programmes tend to disregard structural barriers to personal fulfilment, including labour market exploitation, impeded social mobility, and lack of adequate social support. Underprivileged groups such as disabled people are particularly vulnerable to this increasingly hegemonic approach of workfare (Grover and Soldatic 2013). It has been incorporated in different disability support mechanisms, including personal assistance schemes – for example, in their analysis of personal assistance in Scandinavian countries, Bonfils and Askheim (2014, 73) report that until 2009, ‘users in Denmark had to document a certain activity level to be qualified to get PA [personal assistance]’. This reflects a more general tendency to utilize personal assistance as a technique of governing: ‘the user must prove that the assistance is used in proper ways. Such conditions are internalized in the user as self-management, realizing that, if PA is not used as intended, the service will be withdrawn’ (Bonfils and Askheim 2014, 74).

The rest of this paper will focus on a Bulgarian case of governing through personal assistance. It will thus contribute to critical disability scholarship that follows Foucault in conceptualizing modern power as productive and dispersed rather than repressive and centralized in order to understand how practices of disability support govern people (Tremain 2005). The analysis will look at an assemblage of activities and material artefacts (documents) that constitute an assessment procedure designed to establish eligibility and need for personal assistance of people with different impairments (physical, sensory, mental). It will be argued that disability assessment tools and practices do not only measure people’s needs, but also contribute to the government of individuals by shaping their self-understanding, motivation and conduct. To this end, the analysis will attend critically to the effects of the disability assessment on the subjectivity of those undergoing it. Of particular concern will be the moral aspects of this subjectification. The conclusion will suggest practical policy alternatives.

Disability policy in Bulgaria and the personal assistance scheme of the Sofia municipality

Bulgaria is a former state socialist country located in Southeastern Europe. After the fall of the socialist regime in 1989, Bulgaria experienced a ‘transition’ from one-party rule to liberal democracy and from centrally planned to free-market economy; along this way, on 1 January 2007 the country become a member of the European Union. Many of the economic, social, political and cultural transformations in the aftermath of 1989 had a neoliberal flavour and followed the pattern of Eastern European ‘shock therapy’ (Murrell 1993), including deregulation, privatization of public assets, liberalization of foreign trade, restrictive fiscal policies, and welfare state retrenchment. Within this general context, the Bulgarian disability policy of the past two and a half decades has been moulded by the twin forces of the country’s state socialist legacy and postsocialist neoliberalization (Mladenov 2015a). State socialism has bequeathed paternalism, medicalization of service provision and segregation of disabled people in residential institutions (Mladenov 2011, 2015a, 2015b). On its behalf, postsocialist neoliberalization has conditioned low levels of funding for assistive technology and housing adaptations, cuts to disability benefits, decentralization of service provision (leading to unsustainability and unequal geographical distribution of services), weak monitoring, control and enforcement of disability regulations, and the gradual incorporation within disability policy of workfare conditionality (for an extended discussion, see Mladenov 2015b).

Workfare has affected disabled Bulgarians in different ways. An example of its indirect impact is provided by the national programme ‘Welfare to Work’ that was launched by the Bulgarian government in 2002 (CIL 2009a, 5) with the aim to take unemployed people off benefits and bring them back to work. A component of this programme mandated to long-term unemployed to work as personal assistants under the threat of benefit sanctions. As a result, assistance users were faced with ‘demoralised assistants whose selection was beyond disabled people’s control’ (Mladenov 2015b, 450). The focus of this paper is on a case of workfare policy that has affected disabled Bulgarians directly by being embedded in a scheme for the provision of personal assistance titled ‘Assistants for Independent Living’ (AIL) that has been implemented by the Sofia Municipality since 2007.1

The first round of applications took place at the end of 2007, with subsequent rounds held at the end of each calendar year. Although it is not a ‘direct payments’ scheme – no funds are transferred to the user and the assistants can only be employed by the service provider – the user nevertheless enjoys considerable liberty in choosing his/her assistants (currently, up to five per user), determining their tasks and times of working (currently, up to 300 hours per month), evaluating their work and dismissing them, if deemed necessary. These elements match the personal assistance model promoted by Independent Living advocates such as Ratzka (2004, 3), where:

users are free to choose their preferred degree of personal control over service delivery according to their needs, capabilities, current life circumstances, preferences and aspirations. Their range of options includes the right to custom-design their own services, which requires that the user decides who is to work, with which tasks, at which times, where and how.

The genealogy of the AIL scheme is also related to the Independent Living philosophy and practice (DeJong 1979; Morris 2004) and, particularly, to Ratzka’s model (for a discussion and critical evaluation, see Mladenov 2012). In its original draft version, the AIL Ordinance was proposed by the Bulgarian disabled people’s organization Center for Independent Living – Sofia ( on the basis of the organization’s previous experience with administering personal assistance for disabled people under a foreign-funded project (Dakova 2004) and following the guidelines provided by Ratzka, whose work the organization translated into Bulgarian and published as a separate booklet in 2005 (available online at: After years of advocacy, the AIL Ordinance was adopted by the Municipal Council on 26 July 2007. Since then, the Sofia Municipality has been promoting the AIL scheme as ‘unique’ in the sense that there is no other service in Bulgaria that provides disabled people with such a degree of control over their own assistance, and, consequently, with the chance ‘to lead an active social life and feel full citizens’ (Sofia Municipality 2013, n.p.).

Nevertheless, over the years of its existence the AIL scheme has been plagued by a number of problems. To begin with, personal assistance under the scheme is available to the residents of the Sofia Municipality only, which contributes to the uneven geographical distribution of disability services in Bulgaria. In reaction to this, for a number of years now the Center for Independent Living – Sofia has been advocating for the adoption of a Law on Personal Assistance that would make the service available on a national level. Further, the scheme is funded by the municipal budget and administered by municipal service providers, which considerably limits the available resources. Consequently, users are required to reapply each year, competing for funding with all other applicants on the basis of a controversial ranking system (discussed in detail below). At that, the demand for the service clearly exceeds its supply (see Sofia Municipality 2015). Those who apply but do not qualify for assistance under the AIL scheme are forced to recourse to more limiting assistance options, informal care, and/or segregated alternatives such as daycare or long-term residential care (see Panayotova 2014, 18–19, 25–26). The sustainability of the scheme is also an issue, considering that the amount allocated from the municipal budget to be spent on AIL is revised yearly by the Municipal Council on the basis of needs forecasts and the available resources.

These and other related issues precipitated a number of changes in the way the needs assessment and resource allocation were conducted. Since 2007, the scheme underwent several revisions. The last major change was introduced in 2012, when the scheme was formally split in two – one for children aged 5–18 years, and one for working age adults. The two components remained codified by the AIL Ordinance. Most important for the purposes of the present analysis, since 2012 the scheme has incorporated a significant emphasis on ‘social activity’ (sotsialna aktivnost), defined in the AIL Ordinance (Additional Provisions) through a focus on paid work and formal education:

The social activity of the user includes his [sic] labour activity under employment, business and equivalent contractual relationships (contracts for services, freelancers, sole traders, etc.), educational activity for obtaining an educational degree (primary, secondary, tertiary – vocational Bachelor, Bachelor, Master, doctoral degree) or vocational qualification in accordance with the Law on Vocational Education and Training, and attending kindergarten.

The emphasis on ‘social activity’ changed the eligibility criteria. At present, the eligibility of the adult applicants is effectively conditional on their engagement in paid employment and/or formal education, and the eligibility of children – on their already being in formal education. Such workfare conditionality is not explicitly stated in the AIL Ordinance. Rather, it is embedded in the needs assessment procedure, thus exercising a considerable albeit implicit productivist pressure on the applicants. As already suggested in the introduction, such pressure is characteristic of neoliberal social policy regimes where ‘[c]itizens receiving welfare are forced to earn their social entitlements through participation in the labour market’ (Soldatic and Chapman 2010, 141; the impact of productivism on disabled people is discussed extensively in Mladenov 2015a).

Workfare responsibilization has victimizing and disciplining effects. It tends to blame individuals for failings of structural origins, and it produces a disciplined workforce by imposing productivist moral standards. Thus workfare is a governmental technique that shapes subjects by making them internalize a specific productivist morality. The critics of neoliberalism have regarded this as ‘the economy’s colonization of the Freudian superego’ (Lazzarato [2011] 2012, 95), whereby an external injunction for labour market productivity is transformed into an internal demand. Workfare responsibilization is especially damaging when applied to disabled people who encounter additional structural barriers in their attempts to be productive in the open labour market. Consider the employment of disabled Bulgarians of working age – according to the latest available data from the National Statistical Institute, their employment rate in 2011 was 22.5%, while the employment rate of non-disabled people was 59.1% (Panayotova 2014, 7). Furthermore, drawing on data provided by Eurostat, Panayotova (2014, 5) has reported that in 2012, the employment rate of Bulgarians with ‘severe disability’ (who comprise the target group of the AIL scheme) was merely 12.4%.

In this regard, it is important to note that the disability movement and disability scholars have resisted and criticized workfare policies and practices. For example, according to the ‘UK Disabled People’s Manifesto: Reclaiming our Futures’, welfare support ‘must not be restricted or limited to those that are viewed as “deserving” or productive’ (Inclusion London 2013, 7). The Independent Living model for the provision of personal assistance, referred to above, explicitly states that eligibility should be granted regardless of the user’s ‘employment or insurance situation’ (Ratzka 2004, 3). The UK organization Disabled People Against Cuts (DPAC, actively supports campaigns against workfare.

In the case of the AIL scheme after its 2012 revision, workfare conditionality and the attendant responsibilization mechanism have been covered up or black-boxed through their dispersal in the assemblage of the scheme’s needs assessment procedure. To unpack this assemblage will be the main aim of the next section.

The needs assessment procedure

The AIL scheme elaborates its own needs assessment procedure that is intended to measure the eligibility of the applicants for participating in the scheme and the degree of their need for personal assistance. The procedure is complex and involves a considerable amount of documentation. The AIL Ordinance is accompanied by 30 annexes – 16 for adult applicants and 14 for children. These annexes are forms to be filled in by the applicants, their relatives, prospective assistants, social workers (assessors) and other decision-makers. In addition, a number of evidential documentation is also required from each applicant in support of his/her application – proof of identity, proof of address, disability certificate, proof of employment and/or enrolment in formal education (if any), as well as other documents evidencing ‘social activity’.

It will be analytically useful to regard the forms included in the AIL needs assessment procedure as mediators that not only carry but also co-construct or translate meaning between the elements that they connect such as people (disabled applicants, personal assistants, social workers, decision-makers), other forms, funding and assistance. The concepts of ‘mediators’ and ‘translation’ are borrowed from Actor-Network Theory (Latour 2005) in order to highlight the crucial role of the non-human entities in the constitution of human meanings. Latour (2005, 39) distinguishes ‘mediators’ from ‘intermediaries’ – the latter are entities that transport meaning without transforming it, while the former transport meaning by transforming, co-constructing or translating it. To regard an entity as an intermediary is to black-box the work of translation, whereas to regard it as a mediator is to unpack translation by illuminating and exploring the entity’s role in producing or shaping of meaning.

The concepts of ‘mediators’ and ‘translation’ make traceable the diverse and locally dispersed ways in which internal, subjective content (needs, affects, desires, aspirations, expectations, motivations) is externally or objectively produced, maintained and/or modified. The micro-level sociological investigations informed by these concepts could effectively complement macro-level critiques of contemporary regimes of power that govern individuals by shaping their subjectivity. In the domain of social policy, the concepts of ‘mediators’ and ‘translation’ enable the concrete exploration of the mechanisms for embedding specific moral imperatives into pivotal practices for the administration of social support such as needs assessment procedures. Such practices are conventionally regarded as neutral – in other words, they get black-boxed as mere intermediaries. Yet in a social policy context that promotes workfare strategies for addressing welfare issues, they tend to articulate joblessness as a ‘private moral failure’ (Soldatic and Chapman 2010, 142). The AIL scheme is an instance of such an articulation, where the multiplication of needs assessment documentation significantly augments the work of mediation and translation, thus making it a fertile ground for exploring the social constitution of individual morality (on the micro level) and the attendant mechanisms of governing individuals through freedom (on the macro level). Below, I will focus on the needs assessment of adult applicants. There, the productivist pressure is more pronounced than in the case of children, given that adults are subjected to injunctions to work and study, whereas children are only pressurized to study.

Four stages of translation

The choice of annexes to be analysed here is informed by their significance for the needs assessment procedure – accordingly, the analysis will focus on those documents that do most of the work of translating the applicants’ needs into assistance hours. For the purposes of the present analysis, four stages will be distinguished in this work of translation:

  • the applicant’s needs are translated into degrees of ability and activity (Annex 1.7);
  • the degrees of ability and activity are translated into points (Annex 1.12);
  • the points are translated into ranking (Annexes 1.13 and 1.14);
  • and the points of those who are successful in the ranking are translated into assistance hours (Annex 1.15).

This elaborate work of translation involves a number of actors – the applicant, his/her assistant(s), the two social workers conducting the assessment, and the decision-makers that comprise the Commission under Art. 15 of the AIL Ordinance. It should be noted that this assemblage is not horizontally but hierarchically structured, with the Commission at the top. In particular, the latter is empowered to make final admission decisions, as well as final decisions on the amount of assistance hours to be provided to the individual users who have been admitted (Art. 6 of the AIL Ordinance). The members of the Commission include municipal councilors, physicians and other experts. The Commission functions as a focal point, a centre of attraction that gathers the completed forms and supporting documents (bottom-up movement) and sanctions the distribution of resources (top-down movement). Notwithstanding this centralization and hierarchization though, the effects of power and subjectification are also – and significantly – distributed within the assemblage that constitutes the AIL needs assessment procedure. This will become clear by looking closely at each of the four stages of translation that are at work in the procedure.

The applicant’s needs are translated into degrees of ability and activity (Annex 1.7)

The items included in Annex 1.7 – the ‘Needs assessment questionnaire’ – request information about the applicant’s health condition, family situation, abilities and activities. The first section of the annexe inquires about the health condition of the applicant, reflecting the medicalization of the eligibility criteria and, more broadly, the hegemony of the ‘medical model’ of disability in Bulgarian disability policy (Mladenov 2011) – in order to be eligible for using the AIL scheme, adults need to be certified (prior to applying for the scheme) with ‘90% and over 90% permanently decreased ability to work with assigned assistance, [and be in possession of] a valid territorial expert medical commission (TEMC) or national expert medical commission (NEMC) decision’ (Art. 4 of the AIL Ordinance). This medical-productivist translation of disability into a number is a function of the way in which the Bulgarian state frames disability for welfare purposes in general – as a percentage of ‘decreased ability to work’ (namalena rabotosposobnost) derived from a medical diagnosis and/or medically identified functional limitation established by a Territorial/National Expert Medical Commission (TEMC/NEMC) through a procedure conducted by medical professionals, in a medical setting, using medical criteria (Mladenov 2011). The ‘expert decision’ issued by TEMC/NEMC in the form of a disability certificate functions on a meta-level, as a meta-decision – it is a decision that opens up the possibility for the individual to be subjected to other decisions. In other words, it is the standard basis for getting access to all disability-related welfare structures and processes in the country. This excessive medicalization of disability is problematic in itself (Mladenov 2011), but it is particularly troubling with regard to assessing eligibility for personal assistance because many assistance needs are lost in this translation of one’s disability into a medical-productivist number (i.e. percentage of ‘decreased ability to work’).

The presupposition informing the other sections of Annex 1.7 will be explored in the next part of this text, when looking at the translation of the degrees of ability and activity into points. Here, I would like to make some preliminary comments on sections III (‘Physical condition’) and IV (‘Psycho-emotional and sensory condition’) of Annex 1.7 that are dedicated to assessing the applicant’s abilities. The items included in these two sections are descriptions of activities that are assessed on a scale of three degrees of ability: full ability – partial ability – inability (can do … – partially can do … – cannot do … ), for example: ‘1. Locomotive activity: can move alone – partly can move alone – cannot move alone’. The translation of the applicant’s needs into degrees of ability effected by this formulation relies on a crude and decontextualized understanding of ‘ability’. It is based on a binary model, where the organizing dichotomy is ‘presence vs. absence’ – an ability is either fully present, partially present or fully absent. This tripartite distinction is crude in the sense that it does not allow the identification of other degrees of ability or of any dynamics in ability’s presence/absence – thus, for example, fluctuating conditions, including fatigue and pain, are left out of the equation. The insertion of a middle term – ‘partial presence’ – does little to expand or deconstruct the binary. Rather, the middle term reinforces its inherent logic, according to which one’s abilities are defined with respect to an imagined fully autonomous state of complete self-sufficiency. The items in sections III and IV follow the formula ‘Can/partially can/cannot do X alone’ – it is assumed that one is properly able when one can do something on one’s own. Thus it is individual agency rather than social-material structure that confers ontological status to an ability. In other words, an ability is real (full, present) only as far as it issues from an isolated agent, rather than being maintained by an en-abling social-material structure.2 This liberal-individualist rendering of agency and, by extension, ability underpins the next stage of translation as well.

The degrees of ability and activity are translated into points (Annex 1.12)

At this stage of translation, the inputs solicited by the ‘Needs assessment questionnaire’ (Annex 1.7) are translated into points by Annex 1.12. Thus each applicant is assigned points that are meant to designate his/her level of ‘ability’ and ‘activity’. The resultant number will then be used to rank the applicant in the next stage of translation. The maximum amount of points that an applicant can get in each of the sections reflects the significance attributed to the section for the purposes of the needs assessment. Therefore, by looking at the distribution of the maximum points among the sections, it is possible to elicit the assumptions about social support that are embedded in the needs assessment procedure.

Echoing Annex 1.7, Annex 1.12 contains two sections that focus on the applicant’s ‘abilities’ and two sections that cover the applicant’s employment, education and other ‘activities’. The maximum overall number of points that the applicant can get for his/her perceived lack of abilities is 47, while the maximum overall number of points in the section on ‘social activity’ is 80, where up to 40 points are assigned for being in full-time employment and another 40 points – for being a full-time student at the time of application. If we add to this the maximum of 20 points assigned additionally for ‘other activities for active social inclusion’, including activities for improving one’s physical and mental health, participation in artistic and sports events and competitions, and voluntary work, it turns out that the share of the ‘activity’ points in the overall maximum number of points (157) is 64% (100 points), while the share of the ‘ability’ points is 30% (47 points), with up to 10 additional points assigned in case the applicant lives with his/her minor(s).

This simple arithmetic shows that the translation of the degrees of ability and activity into points mediated by Annex 1.12 renders current engagement in employment and education activities the dominant consideration for granting access to personal assistance. The applicant has to be in some kind of paid employment or formal education in order to have a chance to compete for funding for personal assistance under the scheme, given the weight of these two ‘activity’ categories relative to all other categories, including the ‘ability’ categories. Thus Annex 1.12 contains implicit workfare conditionality. Through the mediation of this document, the scheme effectively enforces ‘social inclusion’ and ‘active living’ by demanding engagement in paid employment and formal education. This generates a paradox or a contradiction, because in that way the provision of assistance is made conditional on the applicant’s ability to cope without assistance – note that the applicant is denied access to the service if s/he is using similar services at the time of application (Art. 4 of the AIL Ordinance).

Moreover, the conditionality implied in the second stage of translation and embedded in Annex 1.12 makes the needs assessment to contradict the aim of the scheme, which is to ‘compensate the deficit of people with permanent disabilities and difficulties in their active social inclusion and in everyday care’ (Art. 2 of the AIL Ordinance). This internal contradiction stems from the attempt to combine a welfare policy of providing social support (in order to ‘compensate the deficit’ of impairment) with a workfare policy of holding people individually responsible for their own support. From a macro-level perspective, this contradiction has structural origins – capitalism needs the welfare state in order to reproduce and discipline its workforce, but also in order to legitimize itself by mitigating the consequences of its mode of production, for example, intensification of work, industrial incidents, inequality, environmental degradation, and so forth. With neoliberalism, the former task takes precedence over the latter:

The fear is that social security measures starve enterprises of labour, meaning that those who are willing to work are more expensive to employ, and that in the longer-term labour will become ill-disciplined and may even raise future generations with similar negative attitudes towards paid work. However, not to provide social security support for workless people is equally problematic because not to do so has the potential, as analysts in the 1970s argued, to erode the legitimacy of capitalism. Such dilemmas and tensions are arguably exacerbated by neoliberalism as an accumulation regime, for although it still has the longer-term strategic needs of social reproduction it is framed, as we have seen, by the disdain for state intervention because of its potential to stifle the efficiency of the free market. (Grover and Soldatic 2013, 225)

As already pointed out, the neoliberal solution to the problem of how to intervene without intervening – a problem emerging from the classical liberal political rationality that frames people as autonomous, self-driven entities, and freedom as non-interference – is to ‘govern at a distance’ (Rose 1996), to incite rather than impede action, and to shape rather than repress subjectivity. Accordingly, welfare is transformed into workfare. In the case of AIL, the solution to the problem of how to support without supporting is to implicitly make assistance conditional on self-assistance, that is, on the applicant’s ability to cope on his/her own. The needs assessment procedure structures the possibilities for action of the applicants so that they make use of their freedom by engaging in paid employment and formal education. This workfare conditionality is not explicitly stated – on the contrary, the AIL Ordinance presents the scheme as enabling unqualified freedom or ‘independent living’. Conditionality is embedded in the needs assessment procedure by orchestrating the translation of needs into assistance hours according to a workfare logic. As a result, disabled people find themselves responsibilized into pursuing paid employment and formal education, notwithstanding the structural barriers they encounter, including the lack of personal assistance. On the contrary – being responsible in the ways prescribed by the procedure is elevated to a major condition for getting assistance.

The points are translated into ranking (Annexes 1.13 and 1.14)

At this stage of translation, the two social workers conducting the assessment submit to the Commission under Art. 15 a ‘Proposal for ranking of the applicant’ (Annex 1.13). On the basis of this proposal, the Commission decides on the final number of points and the assistance hours the applicant will get (Annex 1.14). Only those applicants who receive enough points in order to rank above a certain level determined by the locally available funds receive funding for personal assistance (Art. 17 of the AIL Ordinance) – the rest are included in a waiting list. Thus through the mediation of ranking, applicants find themselves in competition for funding with all other applicants – the more points one gets, the higher one’s position in the overall ranking and the bigger one’s chances of getting funding for personal assistance. The applicants’ ‘assets’ are comprised of their lack of abilities and their registered level of ‘activity’, meaning above all their involvement in paid employment and formal education.

The introduction of the principle of competition at this stage of translation is a case of a market-based solution to a welfare austerity created by the limitations imposed by the municipality on the AIL budget. The attendant incentive to maximize one’s capacity to compete with others – even when it comes to getting access to social assistance – is yet another technique of neoliberal government:

the generic disposition induced by neoliberalism is an organising principle of the self, of the self’s relation to the self, and of its relation to others, articulated towards the maximisation of the self in a world perceived in terms of competition. (Hilgers 2013, 83)

The conjunction of the techniques of responsibilization and competition creates a self-perpetuating, self-feeding cycle of subjectification – the applicant seeks to maximize his/her productive activity so that s/he could more effectively compete with others for funding, a substantial amount of which will be targeted at maximizing the applicant’s productive activity, as will become clear in the analysis of the next stage of translation. An ‘entrepreneurial self’ (Peters 2001) is summoned to secure access to assistance that will feed into its entrepreneurial capacity. In a circular movement characteristic of the regimes that govern through freedom, the AIL scheme produces subjects fit for the specific type of assistance that it provides.3 At that, the pressure to compete with other people in need of personal assistance ‘crowds out’ (Sandel 2012) the value of solidarity and erodes the possibility for collective action that are among the pillars of the Independent Living philosophy and practice (Mladenov 2012).

Furthermore, the implicit distinction between the ‘deserving’ and the ‘undeserving’ embedded in the ranking system is imposed on top of a prior, meta-level distinction between the ‘deserving’ and the ‘undeserving’ brought about by the medical expert assessment, as expressed in the TEMC/NEMC decision discussed above. At this stage of translation, disabled people are effectively re-classified and a new group of ‘truly disabled’ – from the perspective of the AIL scheme – is circumscribed. This operation repeats, at a smaller scale, a process of reclassification that is currently underway in OECD countries such as the UK. The process aims to restrict the access to the category of ‘disability’ and associated social benefits. It is part of the effort to reduce public spending in times of austerity and it has been associated with the neoliberal turn in social policy, where ‘the relief of the financial needs of disabled people are being subverted to productivist concerns with labour flexibility, growth in part-time, casualised labour markets and low wages related to international economic competition’ (Grover and Soldatic 2013, 228). Disabled Bulgarians in need of personal assistance have opposed the re-classifying operation of the AIL scheme, interpreting it as a case of disability-based discrimination. In 2012, petitions signed by more than 40 users of the scheme were submitted to the Bulgarian Commission for Protection against Discrimination and to the Ombudsman of the Republic of Bulgaria, stating that the AIL Ordinance contains discriminatory texts that bring about unfavourable treatment of some people with disabilities in comparison to others (Новини/190.html). So far, these criticisms have remained ineffective.

The points of those who are successful in the ranking are translated into assistance hours (Annex 1.15)

The translation of the degrees of ability and activity into assistance hours is mediated by Annex 1.15. Notwithstanding the availability of this form, the mechanism of making the final decision on the precise number of assistance hours to be granted to the successful applicant remains black-boxed. First, each degree of ability is assigned an upper limit rather than a specific amount of assistance hours, for example: ‘Partially can prepare food alone – up to 15 hours’. Thus the way in which the members of the Commission under Art. 15 decide exactly how many hours (up to the prescribed limit) to fund in each particular case remains a mystery for the outsiders. Second, the rationale for associating specific degrees of ability with specific upper limits of assistance hours is nowhere made explicit. The mechanism of this standardization is black-boxed as well. Third, only selected ‘physical’ (in)abilities get translated into assistance hours through Annex 1.15, while ‘psycho-emotional and sensory’ (in)abilities are inexplicably lost at this stage of translation, although they have contributed to the ranking of the applicant (which means that they have been regarded as signifying need). This loss undermines the communicative aspects of assistance – the latter gets reduced to help with locomotion or manual tasks that excludes support with perception, expression or understanding. The consequence is an implicit restriction of the opportunities for independent living of people with sensory, intellectual or psycho-social impairments. Most important, the black-boxing of the mechanisms underpinning the fourth stage of translation enhances the power asymmetry between those who apply for assistance and those who make admissions decisions. Thus it sustains and reinforces the traditional hierarchy of disability-related service provision (Morris 2004).

This critique applies not only to the fourth stage of translation, but also to the needs assessment procedure as a whole. In itself, the multiplication of the stages of translation and their corresponding mediators has the effect of disempowering the applicant – an ironic corollary, considering that the scheme is meant to enable disabled people to lead ‘independent and active lives’. The irony is further intensified by the fact that the scheme has originally been devised by following the Independent Living philosophy and practice (Morris 2004; Ratzka 2004). Yet in the model for the provision of personal assistance described by Ratzka (2004), the translation of the applicant’s needs into assistance hours happens on the spot, in the initial contact between the individual and his/her assessors. More precisely, needs are expressed in assistance hours from the outset: ‘The need of personal assistance is expressed in the average number of assistance hours per month that a person needs and not in terms of one of several need categories’ (Ratzka 2004, 5). This puts the applicants in a much better position to negotiate the hours in a regime of dialogue with their assessors. Thus the model described by Ratzka presupposes a single stage of translation. By multiplying the stage of translation and their corresponding mediators, the scheme of the Sofia Municipality turns the needs assessment process into a ‘machine’ (Deleuze and Guattari [1980] 2004) for the production of compliant subjects – ‘objects and targets of a power that disciplines them in order to maximise production’ (Hilgers 2013, 83).

The last point is supported by an analysis of the distribution of hours among ability and activity categories (see Table 1 that presents the distribution of assistance hours by category as codified in Annex 1.15). Overall, the biggest share of hours is provided for each of the three entries comprising the ‘Social activity’ category – paid work (16.7%), formal education (16.7%), and ‘other activities’ (16.7%). Thus the fourth stage of translation implicitly reinforces the already discussed logic of responsibilization in the context of workfare conditionality.

Table 1.

Distribution of assistance hours in Annex 1.15.

Category Maximum assistance hours Share of total assistance hours (%)
Everyday care
1. Help with feeding, intake of liquids 25 8.3
2. Help with the preparation of food 20 6.6
3. Help with shopping – provision of necessary goods 20 6.6
4. Help with maintaining personal hygiene 40 13.3
5. Help with maintaining hygiene in the inhabited premises 10 3.3
6. Help with the intake of medication, therapeutic manipulations 10 3.3
7. Help with visiting a physician, therapeutic procedures (hemodialysis, chemotherapy, etc.), hospitalization 25 8.3
Social activity and other activities for active social inclusion
1. Accompanying to the workplace and back, and assistance with labour activity 50 16.7
2. Accompanying to the educational facility/organization for professional qualification and back, and assistance during the educational process 50 16.7
3. Other activities for: 50 16.7
 • improving one’s health and psycho-emotional condition;
 • personal improvement, activities at home and outside (participation in contests, competitions, choruses, exhibitions, sports events and similar activities – to be described), volunteering, caring for one’s minor child or adopted child
Total 300 100

Concluding remarks

There are practical as well as methodological corollaries to be drawn from the analysis of the AIL scheme presented in this paper. On the practical level, the analysis suggests that the AIL scheme, as it is presently codified in the AIL Ordinance and provided by the Sofia Municipality, requires an overhaul. There are many problematic elements – for example, availability is limited to local residents, funding is restricted and unsustainable, users are forced to reapply each year, demand exceeds supply (for a comprehensive overview of these and other problems with the scheme see CIL 2009b). This analysis focused on the needs assessment procedure as a key element in need of revision. The procedure should be simplified, reducing the mediators and the stages of translation involved, and redistributing decision-making power from the service provider towards the service users. Most important, eligibility should be decoupled from workfare conditionality. In addition, applicants should not be forced to compete with each other for social support – instead, funding should be made available to everyone in need of assistance. These suggestions support the demands for changes voiced by the Center for Independent Living – Sofia (Новини/190.html). In 2009, the Center produced a comprehensive assessment of the AIL scheme (the resultant report is available online in Bulgarian – CIL 2009b) that included interviews with users of the scheme. One of the key issues raised by the users was the heavily bureaucratized, non-transparent and resource-led (rather than user-centred) needs assessment process (CIL 2009b, 11–14). Since then, the scheme has changed, but not for the better.

In a longer-term perspective, the inequality generated by the sheer lack and/or the uneven geographical distribution of personal assistance for disabled people in Bulgaria (Mladenov 2015b) will only be overcome by adopting a Law on Personal Assistance that will take into account the lessons learned from the AIL scheme and will follow the guidelines of the Independent Living philosophy and practice (Morris 2004; Ratzka 2004). As has been argued elsewhere (Mladenov 2012), only a strong disabled people’s movement, underpinning a nationwide network of user-led and user-controlled Centres for Independent Living that provide self-help, peer-support, advocacy and watchdog activities, could guarantee the proper development and enforcement of such a policy. Such a network represents the collective dimension of personal assistance that is often overlooked in the liberal-individualist promotion of consumerism in disability policy, but without which individual empowerment is unsustainable – not to say impossible – because it lacks socio-political grounds (Mladenov 2012). The proposed legislation would also enable the country to comply with Article 19 of the UN Convention on the Rights of Persons with Disabilities, ratified by Bulgaria in 2012.

As far as methodology is concerned, the present analysis supports the view espoused by governmentality scholars that macro-level critique of contemporary power is in need of micro-level investigations of the constitution of subjectivity (Lemke 2002). Applied to the domain of disability policy, this means to regard workfare conditionality, responsibilization, and competition for public support as techniques for governing people by shaping their subjectivity. As argued by Bonfils and Askheim (2014), to represent personal assistance schemes as unequivocally liberating is to cover up their disciplining effects. In this paper, I argued that, against its promise to free welfare recipients for ‘independent and active living’ by undoing the paternalistic grip of top-down care, the AIL scheme ‘smuggles’ back unfreedom through implicit and dispersed workfare conditionality, responsibilization and a pressure to compete. This brings about an even stricter regulation that works by imposing a productivist morality (Mladenov 2015a) of self-maximization on the individual applicant, eroding at that the possibility for collective identification and action. This way of governing through subjectification is characteristic of contemporary neoliberal regimes (Lazzarato [2011] 2012; Peters 2001) that seek to retrench the welfare state and render welfare recipients as ‘free consumers’ while maintaining a disciplined, compliant and productive workforce. The specific techniques of such a mode of government gain in strength by remaining black-boxed and, accordingly, lose strength when highlighted and unpacked.