Evaluation and the tension between generalization and particularity: The negotiation of supplementary child allowance in a collective firm
This paper follows the negotiations of a model to calculate supplementary child allowance inside a radical democratic organization – the Radical Cola Collective (RCC) – that pays equal hourly wages to all. With equal wages, the RCC originally refrains from evaluating differences of worth, but also differences of need. The inclusion of new members with children leads the collective to reconsider this decision. The RCC embarks on a process of discovery and negotiation to find a model for calculating fair supplementary child allowance.
Analyzing this intra-organizational process of valuation directs attention to the always already present ‘highly complex socio-technical orderings involving several actors and instruments’ which together perform valuation (Helgession and Muniesa, 2013: 3). Brighenti (2017) has recently proposed to conceptualize these socio-technical orderings as ‘measure-value environments’, which highlight that ‘measure and value exist in an entangled relation’ (ibid.: 16f). Organizations are one such ‘measure-value environment’ where evaluation devices have to conform with existing values and norms. The organization this article is concerned with is based on values of radical equality and inclusivity. Theoretically, everyone who feels affected by the RCC is eligible to take part in their decision-making processes. This creates a setting in which there is ample opportunity for competing moral considerations and values to emerge. Negotiation processes in this context are ‘hot situations’ in which everything can become controversial (Callon, 1998: 260). This makes the study especially interesting to understand valuation as a social practice (Doganova et al., 2018).
The situation in which the collective realizes that it has to find a way to pay supplementary allowance constitutes a critical moment (Boltanski and Thévenot, 1999) in which there is radical uncertainty about how to move on. The collective engages in a process in which it tries to move from a shared appreciation of children, to the appraisal of the share of the costs of raising children that should be paid by it. The challenge is how to translate the appreciation of something into an organizational, monetary valuation practice. This task is especially complex since the object to be evaluated is ambiguous. Zelizer (1994), in her work on the changing social value of children in the U.S., traces how different conceptions of childhood influence the monetary evaluation of children. In particular, she shows that, when children became economically ‘useless’ and emotionally ‘priceless’, profound interpretative challenges emerged around children. There is a common understanding at the RCC that children are valuable. However, this understanding lacks any explicit formulation to clarify for whom exactly children are valuable, why children are valuable, and who is responsible for ensuring the good life of children. While children may be priceless, raising them is certainly not costless. But what are legitimate costs of raising children? Can some children be more expensive than others?
The task of finding a model for supplementary child allowance is made more complex since a connection has to be made between the value of children and the amount of costs an employer should take care of in addition to the state. Processes of evaluation are fundamentally processes in which relations and their meaning are negotiated and defined (Espeland and Stevens, 1998; Fourcade, 2011; Zelizer, 1994). By negotiating a model for fair supplementary child allowance, the collective is negotiating the employer-employee relationship. Are employers responsible for the costs of employees children? The RCC, which is concerned with being a social collective, has to come up with a shared understanding of how much of the costs of raising children of employees a social collective has to take care of.
The language around valuation is not always clear, and meanings and definitions are blurry in both ordinary language use and academic disciplines. To clarify the different layers of valuation as a social practice, the paper follows suggestions from the field of Valuation Studies, where Beljean and Lamont have proposed to differentiate between evaluation as the process of assessing worth and valuation as the process of giving worth (Kjellberg et al., 2013: 20). Vatin (2013), meanwhile, differentiates between evaluation as assessment and valorization as the production of value. The paper draws on three meanings that are often encompassed by the term ‘valuation’ in ordinary language: (1) The RCC is valuing children, in the sense of appreciating them; (2) As a consequence, the RCC is looking for a way to evaluate the costs of raising children and, furthermore, the share of these costs a good employer should pay; (3) By considering implementing supplementary child allowance, they are valorizing children, as such a practice ‘gives worth’ to children in the performative sense in which bonuses, prizes or medals signify value. An evaluation device, then, structures a process of assessment or calculation (Callon and Muniesa, 2005). It determines ex-ante how people and objects have to be evaluated, and it treats all according to the same rules, regardless of their specific characteristics. It achieves what Callon calls ‘framing’, establishing ‘a boundary within which interactions take place more or less independently of their surrounding context’ (Callon, 1998: 248). Furthermore, an evaluation device is used to standardize evaluation, to repeat the same process of assessment for different objects over time. Evaluation is a complex, critical, and highly moral task, especially when ambiguous objects and multiple values are at stake. In evaluation processes, the moral sense of people cannot always be satisfied with the application of universal principles, it might require to consider what is ethical in context (Reinecke and Ansari, 2015). An attempt at standardized evaluation thus entails the moral tension between using general principles and considering what is just in a particular context.
Standardized evaluation is linked to the core function of organizations: it ensures ongoing coordination by limiting uncertainty through coordinating forms that standardize and thus create calculability. Organizations are compromising devices: by combining different values and rationalities, organizations organize (moral) complexity (Thévenot, 1984; 2001). This paper deals with the moral tension between generalized evaluation procedures and their implied sacrifices. This is the tension between what is generalizable and can be measured and what is considered incommensurable and immeasurable. By following the process of negotiating supplementary child allowance, the paper contributes to an understanding of organizational practices and processes through which things get constituted as valuable (Kornberger, 2017), highlighting the role of an organization’s self-image in valuation practices, as well as how the tension between generalization and particularity can lead to the establishment of compromises.
To understand how members of the collective negotiate supplementary child allowance, Boltanski and Thévenot’s sociology of critical capacities (1999; 2006), and specifically the notions of the test and compromise, will be used as heuristic framework. These concepts have been developed in ‘On justification’ (2006), henceforth OJ, as well as in further work of Boltanski (2011) and Thévenot (2001). In organization studies, the test and the compromise have been used to explain the maintenance of legitimacy in public discourses (Patriotta, Gond and Schulz, 2011; Taupin, 2012), as well as the compromising of conflicting values in the contexts of public management (Oldenhof, Postma and Putters, 2013), entrepreneurs in biotechnology and sustainability markets (Kaplan and Murray, 2010; Suckert, 2014) and knowledge commercialization (Mailhot and Langley, 2017). Of particular interest for the purposes of this paper are studies influenced by OJ that are concerned with processes of commensuration and analyze the emergence, critique and legitimization of calculative or evaluation devices (Annisette et al., 2017; Fourcade, 2011; Huault and Rainelli-Weiss, 2011; Reinecke, 2010). Using the notions of the test and compromise will illuminate the complex entanglements between values and measures in organizational valuation practices.
In the following sections, the paper will give an overview of the economies of worth framework developed by Boltanski and Thévenot (2006), outline the concepts of test and compromise, and discuss studies of evaluation devices that utilize Boltanski and Thévenot’s framework, before turning to the case of the RCC.
The economies of worth framework: Tests and compromises
Boltanski and Thévenot‘s ‘On justification’ (2006) develops a framework to explain the competences that enable actors to make critiques or to justify themselves in the face of critique. In a dispute, actors use principles of equivalence that make it possible to assess the relative value of the people and things engaged in a dispute, or their worth. OJ introduces six orders of worth that can serve as frame of reference in a dispute, and each of these orders specifies a form of common good. These orders of worth construct a model of a society that is just because it is a meritocratic society in which members are ordered according to their worth, determined by their contribution to the common good. In the civic polity, for instance, the worthiest people are the ones that are concerned with the interest of all and can embody the general will. In the market polity, the worthiest people are the wealthy, who maintain competition in a marketplace. Based on the polities, OJ develops ‘common worlds’ (ibid.: 130ff), which are historically developed, socio-material instantiations of orders of worth, inhabited by qualified persons and objects. It is a necessary step for the framework to include the modeling of critical competences since it allows us to pay attention to material devices and objects that can be used to demonstrate, test, criticize or legitimize worth in a situation. In the civic world, for instance, the highest states of worth are attributed to collective persons and their representations; important subjects are parties, public collectives and elected officials; important objects are rights, legislation, order, program; and the state of worth is tested through mobilization or democratic votes. Cloutier et al. (2017) note that OJ has been a catalyst for sociological developments in valuation studies and offers a useful conceptual apparatus to study valuation and evaluation both within and across organizations. In order to understand the process by which the collective tries to come to a shared understanding of a fair model to pay supplementary child allowance, this paper uses Boltanski and Thévenot’s notions of the test and the compromise.
Boltanski and Thévenot developed the notion of the test to explain how people can move from a critical moment in which there is a disagreement regarding the appropriate order of beings in a situation to a moment in which the dispute has been resolved. A central principle of equivalence, the higher common good of each of the common worlds, determines how test formats can be established. Each world thus entails its own standards for proving the value of any object or idea. In each world, different objects and ideas are relevant for the testing of worth.
OJ defines two different tests, as a result of two different critiques of a social order: a ‘radical critique’ and a ‘reformative critique’. A radical critique leads to a ‘clash of worlds’ in which the adequate way to evaluate a situation is not certain anymore. For instance, if payment in a firm is primarily based on formal qualification, an exceptional salesman without any higher education might criticize the way that wages are determined from the point of view of the market world, in which a university degree is not of relevance, unless it leads to more sales. In order to come to an understanding regarding the adequate state of the world, the involved participants will have to decide on one order of worth from which to evaluate the situation at hand. Thus, they have to conduct a ‘test of order of worth’ (Dansou and Langley, 2012: 511). A ‘reformative critique’, meanwhile, does not radically criticize the order of worth that is underlying an situation, but rather criticizes the correct execution of evaluation. For instance, a person with a PhD claims that he has been falsely put into a category together with colleagues who only have a master’s degree. A reformative critique may lead to a ‘test of state of worth’ (ibid.) in which people move from a dispute to a new agreement by bringing people and objects in their appropriate order, according to a central principle of equivalence that is related to one world.
Figure 1: Ideal-typical movement from critical moment to test of state of worth
Dansou and Langley (2012) argue that the notion of the test allows us to examine three key dimensions related to how actors question or reproduce constitutive value frameworks: agency, relationality, and temporality. In order to mobilize the test for empirical studies, and to make use of the potential of the conceptual apparatus of Boltanski and Thévenot’s pragmatic sociology of critique, tests should not be seen as linear endpoints but as ‘temporary truces’ with varying degrees of stability (Reinecke et al., 2017).
In their study of the evolution of biotechnology, Kaplan and Murray (2008) argue that the shape of this field, initially uncertain and equivocal, emerged through the resolution of contests around multiple interpretations of the value of its technology. The central task of entrepreneurs in biotechnology was to actively construct and reconstruct justifications for the value of their firms by arguing for particular tests of value and mobilizing evidence to satisfy those tests (ibid.: 12). Kaplan and Murray describe a discovery process spanning three eras, in which the predominant test formats and the respective interpretations of value and formats of evidence were contested and changed. Only after thirty years in which test arrangements were contested, changed and adapted, a stable definition of the field of biotechnology emerged. The stabilized field is a compromise constituted through a complex network of interactions between organizational actors with different rationalities (ibid.: 36).
Compromises between different rationalities are a central concept in Boltanski and Thévenot‘s sociology of critical capacities. A compromise suspends a dispute that involves more than one world without settling the dispute in only one of these worlds. The resulting setup is a composite arrangement that involves persons and things that can be identified in different worlds. A compromise is less stable than an agreement based on a test in one world. Since in a compromise entities are not unambiguously ordered according to one principle of equivalence, it is easy to point out inconsistencies in the evaluation. A compromise needs the involved participants to be ‘favorably disposed toward the notion of a common good’, while at the same time they ‘do not attempt to clarify the principle of their agreement’ (ibid.: 326). Compromises can be made more stable by the creation of objects composed of elements stemming from different worlds which are endowed with their own identity ‘in such a way that their form will no longer be recognizable if one of the disparate elements of which they are formed is removed’. An example, here, is the compromise object of a ‘competitive public service’ that entails the higher common principles of the civic world (public service) and the market world (competition) (ibid.: 278).
According to Thévenot (2001), all organizations are fundamentally arrays of compromises, ‘compromising devices’ between different repertoires of evaluation. Organizations achieve a compromise and coupling between different practices and their respective rationalities by combining devices (ibid.). With this, organizations create structures that are able to deal with (evaluative) ambiguity (Knoll, 2014) and allow them to overcome the uncertainty and tension that emerges from the simultaneous presence of different values.
Evaluation devices as compromises that achieve generalization over time and space
In this paper, a model for supplementary child allowance is conceptualized as an evaluation device. An evaluation device standardizes how people and things are evaluated. It is what Thévenot (1984) calls an investment in forms, a rule that supports a stable connection and the establishment of equivalency and calculability between different entities. Form investments generalize relations between actors and their environment over time and space, specifying what has to be done in a given situation and thus making coordination less uncertain and less costly. Form investments thus sacrifice ‘particularization or characterizations of entities’ (Thévenot, 2011: 41) in order to facilitate such coordination. A wage payment scheme, for instance, can specify that workers will be remunerated by hours worked, which means that their actual performance is not evaluated, or by piece work, which means that their time spent working for the company is not evaluated. Every evaluation device sustains a certain form of evaluation which is based on specific principles of equivalence (time, piece) and, at the same time, excludes other possible principles. In this sense, evaluation devices can be either test arrangements based on one order of worth or, conversely, they can be compromises based on different principles of equivalence.
In most empirical situations, evaluation devices have to achieve a complex commensuration between different principles of equivalence in order to sustain standardized ways of evaluation. The compromise is a useful concept to understand commensuration, which is ‘the expression or measurement of characteristics normally represented by different units according to a common metric’ (Espeland and Stevens, 1998: 315). For example, Reinecke (2010) analyses the determination of Fairtrade minimum prices for coffee as the establishment of a compromise between different orders of worth. In the beginning, when Fairtrade products where not marketed to a mainstream audience and significantly smaller, fair prices were established in face-to-face negotiations between producers and independent fair trading organizations, which created personal, long-term relationships based on trust and mutual recognition. These personal negotiations were later substituted with a formalized price determination based on the Cost of Sustainable Production methodology (CoSP), which calculates the cost of production. The CoSP methodology is an evaluation device, in the sense that is sustains a standardized way of evaluating the situation of different coffee producers.
Since evaluation devices create categories and standards, they are never just pragmatic, but also ethical and political choices since ‘each standard and each category valorizes some point of view and silences another’ (Bowker and Star, 2000: 5). An evaluation device renders certain characteristics measurable and others immeasurable, or not worth measuring. An evaluation device thus entails tradeoffs between different values, and between generalization and particularities in a given situation. By abstracting from particularities and creating general categories, evaluation ‘flattens’ the world (Kornberger, 2017: 19). Regardless of the underlying principles of evaluation, all evaluation devices express a specific idea of justice based on the notion of equality, where all people/objects/situations have to be treated equal if they share certain characteristics. To come to justifiable agreements and orderings, people have to ‘divest themselves of their singularity and converge towards a form of generality transcending persons and the situations in which they interrelate’ (Boltanski and Thévenot, 1999: 363). However, by considering only that what is or can be made general, people lose the ability to consider what is particular and might be incommensurable or immeasurable. Considering particularities and context might be a part of an ordinary sense of justice. From this viewpoint, it may seem unjust to abstract from the particularities of something in order to measure only specific characteristics. Thus, the tension between two different moral orientations – focused on either generalization or particularity – might be an obstacle to the development and application of an evaluation device.
In her study, Reinecke analyses tensions regarding the calculation of minimum prices for coffee and the adequacy of CoSP as an evaluation method for a large number of coffee farmers with differing production and cost structures:
What level of labor costs should be used? What was a decent wage in different local contexts? Whose labour should be taken into account? The labour of the farmer, or the labour of his entire family? What did sustainability mean? And what was sustainable – compared to inefficient production? (2010: 574)
The resulting compromise combines the CoSP methodology with democratic decision making of all stakeholders. Reinecke interprets the result as a compromise between the industrial world on the one hand, in which productivity and efficiency are values measured by standardized criteria, and the civic world on the other, where the collective interest expressed through formal and democratic procedures is valued. Considering the inherent tension between generalization and particularity in evaluation devices, the result can also be interpreted as a compromise between generalization and leaving space for the negotiation of particularities. It therefore combines an evaluation device that is based on generalization with a procedural rules for decision making, which integrates the particular voices of different stakeholders and leaves room for considering particular circumstances in situ.
Annisette et al. (2017) describe the development of an evaluation device for large-scale capital investment projects of a water utility in Western Australia as a ‘test of worths in compromise’ that calculates the financial, social and environmental impact of a project. This compromise device represents a site for ongoing critique, targeted at the mechanics of calculation, i.e. the inclusion of appropriate objects and measures. One of these critiques targets the fact that Aboriginal cultural and heritage issues were not included in the valuation. In response, the developers of the evaluation device argue that certain types of values ‘are not appropriate values to be monetized’ (ibid.: 231). Instead of trying to integrate everything that is deemed valuable in the evaluation device, the developers acknowledge that some environmental and social costs and benefits cannot be monetized or may be inappropriate to monetize. The developers therefore argue that the evaluation device itself should play an important, but not exclusive, role in the decision making process and should be supplemented by other tools, such as multi criteria analysis (ibid.). This study is another example of the combination of an evaluation device with a less standardized method that is better able to include incommensurable values in the evaluation.
Both studies by Reinecke (2010) and Annisette et al. (2017) analyze a context in which there is a high potential for contestation due to the need to come to an agreement which is considered fair by a variety of different stakeholders. In both cases, an evaluation device is contested and has to be supplemented by methods that allow one to consider particular situations and characteristics that cannot easily be generalized. In the following, this paper analyses a context that bears similarities to Reinecke and Annisette et al.’s studies. The RCC is a collective that has to come up with an evaluation device for calculating supplementary child allowance that is considered fair by a variety of people. The collective, however, fails to develop such a generalizing device, and instead settles on a device that signifies worth but refrains from evaluation. It thus achieves a compromise between generalization and particularity.
Methods and case
The Radical Cola Collective (RCC) is a virtual organization that emerged out of a collective of loosely coupled people, connected through the internet. These people wanted to create an economic organization which satisfied their ideas of a just economy. One of their central ideas is that the economy is a collective endeavor and therefore all people affected by a company should have the possibility to directly influence production, including suppliers, subcontractors and wholesalers, as well as end-consumers. At the center of the RCC is a compromise between the civic, the industrial, and the market world. On one hand, it is a cooperative in which equality and solidarity are important (civic worth), but it is also an organization that uses planning methods for coordination and efficiency (industrial worth) and an organization that cares about its survival on the market and considers prices and costs (market worth).
The RCC organizes the supply-chain for their signature product, a cola with less sugar and more caffeine than usual, as well as beer and lemonade. It is based in Germany but also sell their products in two neighboring countries. It has been very slowly but steadily growing from selling a 1000 bottles to its members and friends in 2001, to selling more than one million bottles in 2015. The RCC has a very specific system of distribution that is based on so-called ‘ambassadors’. These are people who want to be part of the RCC and act as semi-professional, semi-activist, local salesmen. In every city in which the RCC’s products are distributed, this is due to an ambassador who started talking to small retailers, restaurants and bars about the RCC project. Apart from the ambassadors, there are six people working part-time for the RCC on day-to-day operations. All the other people involved in the RCC network are business partners, like the bottler and the manufacturer, as well as the wholesalers and retailers. In keeping with its conviction that the economy is a collective endeavor, the RCC organizes itself in large part through an online board where decisions are taken according to consensus. This means that decisions are taken as long as there is no strong rejection in the form of a veto. Theoretically, everyone who feels affected by the RCC can access the online board and take part in the decision making process, although it is mostly people who directly work with or for the RCC who take part in these online discussions.
The RCC combines characteristics of a business with a social collective as well as a social movement. While it has been selling cola, lemonade and beer for more than 15 years now, throughout these years the RCC distinguishes itself from conventional businesses. Membership is open, in principle, to anyone; community members are geographically dispersed; and decisions are taken according to consensus and adhere to a broad agenda of anti-corporate activism and sustainability. The RCC pays equal hourly wages to all, that is, the members of the organization team and the members of the collective who do project-based work, like updating their website. The RCC decided that everyone is equally worthy, a decision that is – although the word ‘equal’ suggests otherwise – not based on an actual belief in equivalence, but a sensibility towards the particular, individual and incommensurable contributions of everyone working for the RCC. According to Espeland and Stevens, ‘incommensurables can be vital expressions of core values, signaling to people how they should act toward those things. Identities and crucial roles are often defined with incommensurable categories. Believing that something is incommensurable can qualify one for some kinds of relationships’ (1998: 327). By paying equal wages, the RCC signals their core values and secures an identity that is based on collective solidarity. With equal wages, the RCC originally refrained from setting up differences of worth, but also of need, between them. This decision was questioned, however, when a father and potential new ambassador for the RCC asked if he would get the same pay as people without children.
Drawing on the analysis of online discussions, this paper will illustrate how the tension between generalization and considering particularities is negotiated during a ‘critical moment’ in which there is uncertainty over an adequate evaluation device for supplementary child allowance. The negotiation of supplementary child allowance has two parts. The first part of the negotiation is an email conversation with 30 emails over 7 days with 9 participants. During this discussion, people decide on an evaluation device to calculate supplementary child allowance, although no one ends up claiming child allowance. The second part of the negotiation happens a year later, when a freelancer who is supposed to do project-based work for the RCC asks for child allowance. During this second part, the evaluation device is tested by calculating its concrete outcomes, which leads to a radical critique and eventually a new proposal. Since the RCC has in the meantime switched to using an online board, the second phase lasts for 41 days, with 32 posts and 6 participants. Of this 6 participants 3 have been involved in the first phase as well.
The material of both discussions has been analyzed with MaxQDA using a coding scheme that included codes for three categories: (1) codes that were related to discursive movements and inspired by conversation analysis (i.e. explanatory introduction, apologizing), (2) codes that were related to justification work and testing (i.e. critique, industrial worth, compromise), and (3) codes that were related to content (i.e. payment, good collective member).
Testing for supplementary child allowance
The use of Boltanski and Thévenot’s framework for conducting the analysis of the discussions showed that people did not follow the ideal-typical route from uncertainty to deciding on relevant worlds and eventually to a test of state of worth. Instead, their negotiation resembles a discovery process: at stake was not only who should be evaluated and how, but also if it was possible to develop a fair model for all situations in the first place.
Phase One: Questioning the justness of generalization, but still adopting an evaluation device from family law
Shortly before this discussion happens, it was decided that ambassadors who work for the RCC should get their working hours paid in order for them to establish a network of customers. A father of three children asks one of the core-members (Udo) if he would get the same hourly wage as all the others if he would start working as ambassador. Udo in turn asks the collective whether they ought to think about adding supplementary child allowance to their salary scheme and how much it should be:
We decided that we will pay new ambassadors the hours they have to work in the beginning to set up a network. So now we have the first ambassador who asks (rightfully so) if he will get paid as much as everyone else, although he has three children. Without really knowing what would be fair, I told him I would say something like ‘additional five Euros per child‘, but this is just me guessing. Should it be more, should it be less, should it decrease with every additional child? And can we as the RCC afford to pay this? These are complex questions, but we have to provide an answer to this potential ambassador.
This open question sets up a critical moment in which there is uncertainty over the correct amount of supplementary child allowance. The question if the RCC should pay extra to parents was already decided at an annual meeting a few months before. Even though Udo does not question the necessity to pay, he seems to be inclined to find a solution that weighs the survival of the company (‘can we afford to pay this?’) with the importance of some idea, though not explicitly stated, of justice (‘without really knowing what would be fair’).
The first response to Udo’s question comes from Theodore, who thinks that it is in principle important to have a ‘social component’ in the payment scheme. However, he highlights that the RCC will never be able to create a scheme that does justice to every individual case, not least because there are also public child benefits:
On one hand, there is a system of public child benefits, but we also have injustices in exactly this system (single parents have to pay more taxes than couples, people who are in the highest income class receive child benefits as well). We can‘t account for all of these differences and compensate for injustices. We would have to collect so much information for each individual case to achieve justice for each individual case.
Thus, the tradeoff between treating particular cases and using general formulas is present already from the start of the discussion. Theodore continues with a proposal for the amount of child allowance: 30 percent supplement for the first child, 20 percent for the second child and 10 percent for the third child. Even though Theodore thinks it is not fair to use a general solution for each particular case, he weighs the coordination savings of an evaluation device against the costs – in terms of time and resources – and introduces a compromise between his awareness of particular circumstances and the concession that the RCC needs a general solution.
Udo in turn asks if someone with children could confirm that more children are related to decreasing marginal costs. The father of three answers that there is a decrease of costs to some extent and that he would be OK with the 30%, 20%, 10% model. At this point. Karin adds that the father of three is not the only person with children at the RCC, since she has been part-time-mother of her partner‘s children for over 2 years now and also contributes to the payment for these children‘s expenses for vacation, education and hobbies. Next, Udo wants to draw a line between one’s own children and children that someone feels responsible for. This is countered by Anna, who criticizes Udo for sticking to an outdated idea of family:
I think it would be extremely unfair, in a time of patchwork families and alternative family concepts, to punish people who can’t have children of their own, or fall in love with someone who already has children, or who are homosexual and adopt children.
This exchange between Karin, Anne and Udo shows that the question of the relevant beings that should be included in the test has not yet been exhausted. While the idea seems to have been to pay child allowance for parents, it is not clear who counts as a parent. Udo responds that he did not know that Karin is actually living with her partner’s children part-time and adds that the supplementary child allowance should of course apply to patchwork parents as well. He says that he has learned that there is another member of the RCC with four children, so when they decide on supplementary child allowance it will be very expensive, but he still thinks it is important.
If we want to be a socially minded project, that is a part of it. In this case, the ambassadors will just have to become even more reliable and efficient.
Here we can see that it is not only the question of who parents are and how much children cost that is put to a test in this discussion. The question of child allowance is also a test of the nature of the employer. If the RCC wants to be a ‘socially minded project’, it has to pay an adequate amount of child allowance. Next, Udo actually calculates what the 30:20:10 model would mean in absolute terms and does not think it would be a good model:
The 30:20:10 percent model was just an estimation that Theodore threw in, and it seemed to make sense for the potential new ambassador. However, did any of you actually calculate what this means in absolute numbers? For someone working full-time and calculated with 15 Euro per hour, 30 percent means 720 per month and 10 percent 240 per month. This is a complicated topic because there are also public child benefits and reliefs and, therefore, children don’t have to be supported by their parent’s income alone. But still: To me, 720 per month seems too much and 240 and not enough. What do you think?
Udo puts the proposed evaluation device to a new test by calculating what it would actually mean. He doesn’t like the outcome and, though he can’t actually explain why 720 is too much and 240 not enough, he tries to delegitimize the device by pointing out that it was just something someone threw in. Again we can see that he is worried about the status of the RCC as a good employer who, it seems, has to make sure that their employees earn enough to cover the cost of their children. Martin, not responding to Udo, but asking another fundamental question, wonders why there should be a specific consideration of the situation of people with children and not also other situations where people have more need. He thinks that parents are not necessarily more entitled to additional remuneration than people who are engaged in other activities that are for the common good, such as voluntary work. Martin, like Karin before, criticizes the determination of relevant beings for the test. To him, it is not only parents who deserve supplementary allowance, but all people who contribute to the common good, which is framed according to the civic order of worth.
Udo counters that voluntary work is voluntary and can be done flexibly and to varying degrees, whereas people cannot decide how much care they want to give to their children, or whether they want to care for ill relatives. He asks if they should include care work for relatives in their deliberations. Beppo, a father of two children, wants to clarify that not everyone has increased costs due to children because, for him, the child benefits paid by the state are already sufficient. He proposes to use the Düsseldorfer Tabelle (Düsseldorfer Table), developed by judges specialized in family law in order to calculate alimony payments, as orientation for the rates of supplementary child allowance and only pay supplementary child allowance if people explicitly ask for it. But Beppo’s contribution also contains a critique of generalization: while he thinks he does not need supplementary child allowance, he does not want to speak for all other parents. His proposed compromise between generalization and particularity is to pay child allowance only when people explicitly ask for it and to use an evaluation device from family law, thus combining civic, domestic and market orders of worth.
Udo likes the proposal and formulates a possible resolution, namely, to use the Düsseldorfer Tabelle only on the condition that people ask for child allowance. As a reaction, three people raise concerns because to them it would be fairer if people had to individually propose their own wages, since one solution for all is by definition unfair to individual cases. For instance, Arno says:
My ideal would be that everyone who works for the RCC and wants money for their work can negotiate their wages freely with the collective […] the person can ask for what they actually need, not some approximate value derived from some table. I think it is kind of awkward to say that we all get equal wages, but because we know this is not fair under this or that circumstance we pay +X percent and actually we know, that this is still not really fair.
Udo reacts by drawing a line between a decision for the same wages for everyone and a decision for paying higher wages to people with children. Regarding the same wages for everyone, he recalls the discussions they’ve already had where, in the end, they decided that all people who work for the RCC are equally valuable. In the end, there is no veto against Udo’s proposal for resolution.
Figure 2: Negotiation of supplementary child allowance, phase 1
This first phase delivers three insights that are relevant for a sociology of critique and valuation. First, inside a critical moment, there might be the need to negotiate two different test arrangements at the same time: the search for fair child allowance involves a test of the RCC as good employer, as well as the costs of children. Second, while trying to figure out a test arrangement, people might realize that they have to clarify or change the relevant beings that should be tested. In the beginning it is neither clear if only parents deserve additional pay nor what exactly qualifies one as a parent. Third, people might not always consider ‘a rise towards generality’ as just. In this discussion, many people voice a critique of generalization as unjust. However, most are also willing to settle with a compromise between generalization and particularity. These compromises involve a standardized model which is supplemented with methods that open up a space for considering particularities in the evaluation.
Phase Two: Fundamental critique of calculation device, questioning the justness of generalization and resorting to a compromise
Phase 2 starts when Kate, the accountant, opens a discussion thread in the board in which she reminds everyone that the decision has been taken to pay supplementary child allowance. She then calculates what this would mean in absolute terms and asks if people are still OK with this. She adds that this has not been applied yet, but that a potential contractor has asked for child allowance. Frank, the freelancer with children who will be doing project work for the RCC in the future, then asks if everyone was aware of these sums when they decided on the Düsseldorfer Tabelle as a reference point for the supplementary child allowance. While the amounts seem reasonable to him (he would get approx. €8.50 per hour more than the others), they are also quite high. Kate responds that for someone working full time, the Düsseldorfer Tabelle would mean an additional income of €522 per month for a small child and she thinks this is too much. Instead, she proposes a pay raise for everyone to €20 per hour:
In general, I think we should pay everyone €20 per hour, regardless if they have children or not. This will save us individual administration efforts and freelancers don’t have to tell us how many children they have. :-)
Here, we can see something that already happened before: an evaluation device is tested by looking at its actual outcomes and then criticized, not for the principles of methods of calculation, but because the outcome does not ‘feel’ right. Furthermore, Kate proposes a new evaluation formula that would mean higher, but equal, pay for everyone, a proposal that is accompanied by a reference to efficiency and transparency. Next, Theodore criticizes the Düsseldorfer Tabelle because it is not intended to calculate supplementary child allowance:
I’m asking myself if the Düsseldorfer Tabelle is the right measure. To remind you: the function of the table is to secure the livelihood of a child through payments by the separated parent. The table is originally NOT intended to calculate supplementary child allowances by the employer.
He also adds that in order to find a fair solution, the RCC would have to deal with every case individually which it cannot do due to data protection rights and administrative overhead. Theodore would like to make the supplementary child allowance lower than the amounts calculated by the Düsseldorfer Tabelle and instead increase the hourly wage for everyone. Furthermore, while he is for paying supplementary child allowance, it is normal that expenses for children are paid for by the wages of the parents, as well as the public child benefits. Theodore provides a new, radical critique of the evaluation device: The Düsseldorfer Tabelle’s intention is not to calculate supplementary child allowance in the first place. This critique is radical in the sense of Boltanski and Thévenot, since it argues, that the formula at hand is employing an inappropriate evaluative frame, one that belongs to a different ‘world’. Furthermore, by stating that it is ‘normal’ that parents pay for their children with their wages, he ‘relativizes’ (Boltanski and Thévenot, 2006: 336f) the character of supplementary child allowance: It is not a right of parents and not an obligation for a good employer, it is just a nice gift, a private arrangement between the RCC and its members that does not refer to a general good. Frank and Theodore then both discuss if children are more expensive when the parents are separated, due to higher rent and child care costs. Frank also notes that the RCC will probably not find a ready-made solution out there and asks what exactly the supplementary child allowance should achieve: a gradual improvement of the financial situation of people with children, or a coverage of the costs of children, minus public child benefits? Theodore responds by once more stating that what the Düsseldorfer Tabelle wants to achieve is not what the RCC wants to do with supplementary child allowance. Then he proposes to think again what exactly they want to achieve with supplementary child allowance and if maybe they could just agree on an additional €2 or €2.50 per child and hour. With this, Theodore is the first person that proposes to discuss what exactly the purpose of paying child allowance should be. However, instead of providing possible answers for his question Theodore immediately proposes a relatively simple solution, which is not further legitimized. Frank agrees with the proposal:
So I’ve thought about this again and I’ve realized that my more complex considerations on the calculation might drive me to insanity so I’ve just calculated my income if I would work for 40 hours, and would get additional €2 or €2.50 and this would be fine for me. Therefore, I have no objection against the proposal.
Frank’s argument that complex considerations might lead to insanity suggests the relevance of using a kind of ‘satisficing’ logic (Simon, 1956) for legitimization: although there is a desire to achieve true justice here, it seems that, as people realize how complex the question is, at some point finding a totally fair and adequate evaluation device does not seem important anymore. Instead, a simple solution, with no complex calculations attached, is deemed to be legitimate. Anton joins the discussion to add that people should just say how much additional pay they need. He adds that this makes more sense since people could have higher financial needs due to various reasons – not only children. He proposes a compromise between generalization and particularity:
We will never be able to discuss and consider all potential situations! But we can develop a system which will allow us in the future, if needed, to incorporate different situations. I am thinking of a system in which higher financial needs of people could be described as prototypical cases in order to allow orientation. These prototypical cases could be extended whenever needed.
We have seen this before, Anton both criticizes attempts to find general solutions for particular cases and the beings chosen for this evaluation test. To him, it is not only the situations of parents that should be considered. Frank responds to this and says that a solution like this would end in overly complex calculations and he thinks the +€2 proposal makes sense. Several people agree with Frank. In the end, there are no critical concerns and they decide on €2 per child and hour, provided that people ask for it.
In the second phase of the discussion there are three insights for a sociology of critique and evaluation. First, a calculation device is criticized due to its outcome, not the principles that are used for calculation. Second, the complexity of trying to find a general solution that can do justice to plenty of different and particular cases can lead to a point where the objective of finding a just model does not seem enough to justify complexity. On the contrary, it can ‘lead to insanity’! Third, the compromise that the collective eventually agrees on can be understood as the outcome of a discussion in which the tension between generalization and particularity has been discussed exhaustively to the point where people agreed on a ‘satisficing’ solution.
Figure 3: Negotiation of supplementary child allowance, phase 2
The process by which the RCC tries to collectively find a model for paying fair supplementary child allowance has been analyzed by utilizing Boltanski and Thévenot’s notion of the test as a heuristic framework. By following the justification work of members of the collective, the analysis showed that actors in a situation of radical uncertainty do not necessarily follow a linear path from giving worth to its assessment. Rather, actors conduct a collective negotiation process by which values and their appropriate translation into measures are simultaneously discovered and created. In a testing situation like this, together with the prevalent ideas of justice, the aims, subjects and objects of a test may change. The test is more a discovery process than an implementation of established ideas or values; it is more exploration than exploitation. In this process, people discover what they actually mean when they value something and the possibilities of assessing worth influence the ways of valuing. When the RCC starts discussing supplementary child allowance, there is initially no certainty about which beings are put to a test. It appears that it is not only parents that are (financially) responsible for the up-bringing of children. Problematizing the concept of parents in turn brings up the consideration of other relations of care, posing the question of whether indeed every kind of responsibility for people in need of care should be considered for supplementary allowances. Even when these questions are answered, the RCC has to develop an evaluation device that can simultaneously assess the costs of raising children and what it means to be a good employer with regard to these costs. The paper further illustrates that in order to mobilize the test for empirical studies and to make use of the potential of the conceptual apparatus of Boltanski and Thévenot’s pragmatic sociology of critique, tests should not be seen as linear endpoints, but rather as ‘temporary truces’ with varying degrees of stability (Reinecke et al., 2017). While, in a first phase, the RCC agrees on an evaluation device for calculating supplementary child allowance, this is criticized and subsequently replaced in the second phase.
Supplementary child allowance is conceptualized as an evaluation device. The concept of the evaluation device captures the double moral complexity of both developing a formula for assessing worth and using this formula to standardize evaluation over time and space. In both Reinecke’s (2010) study on Fairtrade minimum prices for coffee, as well as Annisette et al.’s (2017) study on evaluating large-scale capital investment projects of a water utility in Western Australia, an evaluation device is contested on the grounds that it does not allow to consider particular situations and characteristics that cannot easily be generalized. While Reinecke and Annisette et al. discuss their findings as compromises between different worlds, this paper contributes a different possibility for interpretation: compromises between generalization and particularity. The tension between generalization and particularity in evaluation and, especially, evaluation devices is illustrated and further elaborated by the analysis of the (failed) development of an evaluation device in the RCC. In the end, the RCC settles for a compromise between the need to find a generalizable agreement and the hesitation to use a standardized format to evaluate the situation of particular parents and their children. The final solution is a compromise, since the +€2 model answers to the need of making a generalized difference between people with children and people without children. It is also a formula that offers a clear answer to the question of how much supplementary child allowance a person with one, two or three children should get. In that sense, the formula can be seen as a form investment that stabilizes relations. However, it is a very specific device in that the +€2 model lacks any concrete foundation of legitimization. No one actually knows how these +€2 are supposed to be related to the situation of parents with children; it is not clear what this €2 should achieve and why it has to be exactly this amount. It is an arrangement that allows to calculate the future, but in itself does not calculate, i.e. it does not specify a principle of equivalence that could explain itself. The +€2 model is a valuation and valorizing device that signifies and remunerates value, but abstains from evaluation.
There are, then, at least two coordinating forms that accomplish a compromise between the general and the particular. On one hand, the combination of evaluation devices with methods that can consider particularities and, on the other hand, valorizing devices that signify and remunerate value, but are void of concrete evaluation based on principles of equivalence.
Boltanski and Thévenot’s economies of worth framework allows us to view organizations as structures that compromise multiple orders of worth. In this paper, the framework has been successfully applied to deal with normative tensions in organizations. However, the tensions that the framework is able to capture are limited to those that can be related to competing higher common goods. In the economies of worth framework, legitimacy is always related to a ‘rise to generality’. Considering particularity, in contrast, problematizes all attempts to establish equivalency between different particularities, which is the basis for any generalization. By analyzing the justification work accompanying the failed attempt to develop an evaluation device, this paper points to a different tension that organizations have to deal with, as well as a different compromise. The tension between the general and the particular is especially relevant to consider when studying valuation in organizations, since organizations have to create generalization and standardization in order to guarantee certainty and stability for coordination (Thévenot, 2001). This need for generalization can come into tension with considerations of particularity, as has been demonstrated in the case of the RCC negotiating supplementary child allowance.
Evaluation can be a highly uncertain process: the final outcome as well as what exactly might constitute worth can be unclear. Evaluation can be made less uncertain and less costly if established evaluation devices are used. This paper analyzed the attempt to develop an evaluation device that was used to calculate fair supplementary child allowance in a collective based on radical democracy. To understand the process by which the collective tries to come to a shared understanding of a fair model to pay supplementary child allowance, the paper utilized Boltanski and Thévenot’s notion of the test. By applying the test as heuristic framework, the analysis offers two insights relevant for the literature on test. First, the analysis shows that actors in situations of radical uncertainty do not necessarily follow a linear path from giving worth to its assessment. It is rather a discovery process in which people learn what their values imply and how they can be realized in practice. Second, in following the justification work to develop a fair model to pay supplementary child allowance, the analysis furthermore shows that even in a situation that is characterized by an imperative to justification, people might settle for solutions that are good enough, or ‘satisficing’. By bringing in a consideration of the tension between the general and the particular in evaluation practices, the paper thus contributes to studies of evaluation in contexts of moral complexity.
I am grateful to Karin Lohr and two anonymous reviewers for comments and suggestions. I also received valuable feedback from other speakers at the ephemera ‘Beyond measure’ conference, Stockholm, June 2017. I thank the members of RCC for being courageous and inspiring, and for letting me spend time with them.
 While Zelizer’s study only looks at the US, similar processes related to a ‘sacralization of children’ have been taken place in Europe as well, i.e. a ban on child labor, and universal compulsory education.
 For a general overview of research inspired by OJ in Organization Studies, see the introduction to the Research in the Sociology of Organizations special issue on ‘Justification, evaluation and critiques in the study of organizations’ (Cloutier et al., 2017).
 Practices that merely signify and remunerate value, without actually evaluating value, are probably most often non-monetary, like honorary titles or ceremonial rituals. But one could think of certain bonuses or concessions as valorizing devices, if they combine symbolic valuation with monetary remuneration, while abstaining from evaluation.