Aarhus University Seal


Research stream 1: Definition and taxonomy

CEPDISC will develop a clear definition of discrimination and subspecies thereof grounded in philosophical analysis and experimental methods.  

WP1: Mapping different conceptions of discrimination in the humanities and social science

WP1 employs conceptual analysis to identify the concepts of discrimination used in different academic disciplines, e.g., philosophy, sociology, and law. The aim is to distinguish between a number of different conceptions of what discrimination is and to define them in terms of necessary and sufficient conditions. Differential treatment is a necessary condition of discrimination, but it is not a sufficient one. Different suggestions exist in the literature as to what the further conditions are – e.g., wrongfulness, irrationality, demeaningness, or differential treatment on the basis of morally irrelevant properties – and an important task is to identify those properties and their interrelations. The importance of this endeavor is easy to see once one considers that to determine, say, whether discrimination is increasing, we need to be able to measure it and for that purpose we need a precise definition. (As a working definition, we will understand discrimination as differential treatment based on membership of socially salient groups, e.g., race groups, women (or men) and disabled people.)

Leader: Kasper Lippert-Rasmussen
Members: Marion Godman, Søren Flinch Midtgaard, Simone Sommer Degn, Diana Galos

WP2: Empirical studies of citizens’ conception of discrimination

WP2 utilizes experimental approaches to investigate what citizens characterize as discrimination, the specific content of their discriminatory stereotypes, and variations in classificatory dispositions across sub-groups (based on, e.g., gender, ethnicity etc.). First, we test the former topic using conjoint experiments and vignette-based survey experiments. Conjoint experiments help us test specific features and criteria of discrimination (identified by WP1) against each other with respect to what citizens understand and classify as discrimination. Next, vignette-based survey experiments allow testing if citizens also classify specific scenarios differently with respect to what counts as discrimination depending on whether these scenarios contain key theoretical features of discrimination or not (cf. WP1&3). Second, discrimination often is associated with negative stereotypes about discriminated groups. The content of these stereotypes is hard to determine, because individuals might merely implicitly stereotype certain social groups. Yet, these implicit stereotypes might guide behavior and attitudes related to discrimination. To illuminate these stereotypes WP2 utilizes “the reverse-correlation image classification technique.” This experimental set-up permits evoking visual depictions of citizens’ mental representations of potential discriminatees.

Leader:  Lasse Laustsen     
Members: Kim Mannemar Sønderskov, Nicholas Hass, Claire Gothreau, Benedicte Oman Grønhøj

WP3: Taxonomy of discrimination

Based on the conceptions of the genus of discrimination developed under WP1, WP3 develops a taxonomy of different sub-species of discrimination – a taxonomy which possibly will vary across different conceptions of discrimination and which in many cases will involve explicative definitions that aim at improving present conceptual apparatus in the interest of clarity, fruitfulness etc. It is common to distinguish between, inter alia, direct and indirect discrimination (roughly, the distinction between differential treatment rooted in representations of the discriminatees such as in prejudicial beliefs, and those that do not), statistical and nonstatistical discrimination, structural and non-structural discrimination (roughly, the distinction between differential treatment rooted in institutions, social practices and differential treatment that is not so rooted). WP3 will 1) identify the many ways in which these distinctions are drawn in various academic disciplines and in official guidelines, and 2) critically assesses these in light of the standard virtues of definition generally accepted in the philosophical literature on definitions, e.g., are the distinctions exhaustive (could there be discrimination that is neither direct, nor indirect?), mutually exclusive (could there be discrimination that is both structural and non-structural?), revisionist in their implications relative to ordinary language etc.

Leader: Kasper Lippert-Rasmussen 
Members: Marion Godman, Andreas Bengtson, Hugo Cossette-Lefebvre

Research stream 2: Objections to discrimination

CEPDISC will map the reasons people object to discrimination and the logical relations between them.  

WP4: Philosophical objections to discrimination

The objections to discrimination articulated in philosophy and legal theory fall into three broad categories:

1) mental state-based accounts according to which discrimination is wrongful in virtue of the disrespectful attitudes of the discriminator;

2) meaning-based accounts according to which discrimination is wrong because of what it expresses about the discriminatee; and

3) harm-based accounts according to which discrimination is wrong because of how it affects discriminatees negatively.

WP4 identifies the implications of these accounts and their main strengths and weaknesses. Also, it assesses whether the accounts are individually sufficient explanations of the wrongness of discrimination or rather parts of an adequate overall account thereof and how these fit with the empirical explanations of the sources of psychological discrimination identified in WP7, e.g., if much direct discrimination is rooted in implicit bias this might motivate revision of standard mental state-based accounts, which typically focus on occurrent mental states. Finally, WP4 draws on the taxonomy developed in WP3 to see how the different objections apply to different types of discrimination.

Leader: Søren Flinch Midtgaard
Members: Kasper Lippert-Rasmussen, Andreas Brøgger Albertsen, Viki Møller Lyngby Pedersen, Marion Godman, Andreas Bengtson, Miklós Zala 

WP5: Empirical studies of citizens’ objections to discrimination

WP5 examines citizens’ objections to discrimination by conducting population-based survey experiments, in which representative samples of citizens are presented with short vignettes describing different cases of putative discrimination as well as condensed versions of philosophical arguments for why the cases are (or are not) wrongful discrimination. They will then express their views on whether they (dis)agree with the arguments in question. The findings of this large-scale study will be supplemented with structured in-depth interviews with a subsample of the people who participated in the survey experiment. This will allow for a richer understanding of citizens’ thinking concerning discrimination’s wrongness, which is of special theoretical interest with regard to (a) attributions of moral responsibility for discrimination (Who do citizens perceive to be at fault for discriminatory behavior, and do their responsibility-attributions differ from philosophers’ responsibility-attributions?), as well as with regard to (b) cases where it is more controversial whether these really are forms of discrimination (Do citizens consider objectionable cases that, e.g., on key philosophical accounts of discrimination, would not count as discrimination, and if so why?)

Leader: Andreas Brøgger Albertsen
Members: Lotte Thomsen, Marion Godman, Lasse Laustsen, Bjørn Gunnar Hallsson, Ida Bruun Nørregaard, Lydia Tsiakiri  

WP6: The relation between philosophical and citizens’ objections to discrimination

WP6 charts the interrelation between 1) the main philosophical objections to discrimination (cf. WP 4) and 2) those that are identified as the most important citizen objections (cf. WP 5). The main objections have different and sometimes conflicting implications. Take age discrimination against the elderly. Such discrimination arguably increases lifetime equality of opportunity, since dying young means lesser opportunities. Yet it might involve disrespectful meanings, e.g., stigmatization. By way of further illustration: a significant divide in the philosophical literature exists between those who think considerations about disrespect explains the wrongness of discrimination and those who think considerations about harm do so. Yet, arguably treated disrespectfully cannot be understood as a form of harm, i.e., dignitary harm. If so, the two objections will motivate different policies regarding discrimination. Also, WP6 assesses the significance of discrepancies between the objections of citizens and philosophers, e.g., do philosophers have a (moral) expertise when it comes to objections to discrimination – and, if so, how does that fit with their common self-understanding as simply articulating citizens’ commonsense views? Part of the answer to this questions depends on how citizens deliberate in the light of arguments pertaining to the wrongness of discrimination (WP5), e.g., the degree to which their objections are sensitive to framing. Along these lines, WP6 will address the issue of the limits of the philosophical-experimental approach.

Leader: Marion Godman
Members: Kasper Lippert-Rasmussen, Lotte Thomsen, Andreas Brøgger Albertsen, Bjørn Hallsson, Søren Flinch Midtgaard, Maryana Balezina

Research stream 3: Causes and mitigation

CEPDISC will identify the most effective means of counteracting discrimination which are compatible with the reasons mentioned in Research stream 2.  

WP7: Experimental studies of the sources of prejudice and discrimination

WP7 investigates the psychological mechanisms behind direct discrimination. Despite the decline in explicit bias, recent decades have not seen a substantial reduction in differential outcomes across groups, suggesting that implicit biases, i.e., relatively unconscious and automatic prejudiced judgment and social behavior, may still operate. WP7 conjectures that many psychological mechanisms underpinning the expression of bias are motivated and mediated by social dominance orientation (SDO). This suggests that bias interventions should address these underlying relational processes. SDO motives respond to societal macro-structure, e.g., economic inequality, and context, e.g., the extent of competition over scarce resources, and have been found especially predictive members of high-status groups, especially if their position is threatened. WP7 uses implicit association tests (selective pairing of positive and negative valence with particular social groups), reverse correlation studies (cf. WP2), and biased evaluations of blame and intention when varying the group membership of the target to assess levels of prejudice and discrimination. WP7 expands the study of the effect of general relational motives on discrimination to also encompass relational motives for meritocracy and effort-based reward systems and for communal sharing between and within groups. WP7 also explores the extent to which hierarchical and egalitarian intergroup motives reflect personal or shared critical childhood and concurrent experiences, e.g., relative deprivation and marginalization. Throughout, WP7 draws on the conceptual and taxonomical work in WP1&3.

Leaders: Lotte Thomsen
Members: Lasse Laustsen, PostDoc TBA

WP8: Combining governmental data registers and experimental data

WP8 explores the role of social context in discrimination. Specifically, it addresses whether people’s social contexts, e.g., the volume of their interactions with out-groups, matter for their propensity to discriminate and for which groups they discriminate against. Also, WP8 analyzes the role of long-term and/or childhood exposure to certain groups in relation to discrimination, e.g., are adults who had classmates with a low socioeconomic background more or less likely to discriminate against socially marginalized groups? Theoretically, WP8 draws on two social theories – conflict theory and contact theory – that have opposing predictions regarding the role of social context in relation to behavior towards other social groups. Empirically, it combines experimental data and official population registers. To characterize people's past and present social context it uses longitudinal registry data on subjects’ school affiliation, workplace affiliation and place of residency along with registry data on the social composition of those contexts, e.g., share of immigrants etc. These data will be combined with measures of discrimination using similar methods as those employed in WP2&7. The combination of the detailed Danish registers, which are unique in an international perspective, and state-of-the-art experimental data on discrimination provides us ample opportunities to generate new and relevant insights on when discrimination occurs and if certain social settings and experiences are fostering/preventing discriminatory acts.

Leader: Kim Sønderskov
Members: Julian Schüssler, Diana Galos, Mathias Kruse, 

WP9: Feasible and desirable instruments of mitigation

WP9 has two parts. First, it builds on the results in WP7 and WP8 to assess which interventions to mitigate discrimination are feasible. Most interventions focus on making people aware of their implicit biases. However, such interventions only work to the extent that people are motivated and capable of exercising self-control, something that research has shown is difficult. Accordingly, WP9 will test a range of psychological interventions related to, e.g., mindfulness, which have been found to increase impulse-control. Furthermore, WP9 will compare these interventions to empowerment techniques that target the threats and resource deficiencies that cause biases in the first place. Second, on the basis of the findings in the first part of WP9, it will explore how a range of different motivations, both related to values, personality and self-interest, intersect in support for different kinds of interventions, e.g., in relation to trigger warnings, safe spaces, and similar increasingly common anti-discrimination interventions. In this regard, WP9 will also draw on the work of WP4 to assess how different understandings of the wrongness of discrimination leads to support for different antidiscrimination methods. For example, it is likely that empowerment techniques will not resonate with awareness-oriented accounts.

Leader: Kim Mannemar Sønderskov
Members: Nicholas Hass