RONEPP - Roma Networks and Political Participation


Many scholars underline that in Western Europe some Roma groups are coordinating themselves, but their collective action has not the characters of an ethnic movement, and it is far from being a social movement. They do not engage in contentious politics. The situation is quite different in Eastern European Countries where Roma developed ethnic political parties. Also, in Eastern Europe these political parties are quite fragmented, and do not find a common identity or a form of coordinated action.

How to explain the lack of contentious politics among Roma?

The RONEPP project has funded by Sciences Po for a two years’ duration. It has followed another project: Repin. One of the unexpected result of the Repin project has been to show that most of the Roma migrants experiencing upward social mobility in France does not claim against discrimination and does not commit for Roma rights.

Departing from these results, the puzzle of the RONEPP project is the following:

  • Roma are highly discriminated in all European Countries, but with different intensitie
  • In Western Europe, Roma middle classes mostly pursue a strategy of invisibility, and do not engage in collective action for identity recognitio
  • Roma NGOs mobilisation is focused on grant seeking and fund raising, but some forms or Rights claim-making are emerging

RONEPP will analyse Roma collective action in two European western countries, France and Italy. It will be a pilot project, to test a preliminary hypothesis about the role played by Roma personal networks in sustaining or avoiding collective action.

To answer the main research question three analytical challenges has to be considered.

  • The first one is to describe the emerging forms of Roma mobilisation and campaigning for their rights.
  • The second one is to understand the meanings Roma activists attribute to collective action.
  • The third one is to test two contradictory hypothesis, one emphasizing the lack of resources (education, economic capital, legitimacy, political rights), and the other one linked to the form of personal networks in endogamic groups. 


In particular the data collection will allow us to test the hypothesis that subgroup segregation within personal network reduce the odds of engaging in collective action. Briefly we gauge the subgroup structure of the Roma's personal network, and how Ego's contacts form cohesive subgroups, and Ego's social network is partitioned in separate circles and cliques. The analysis of network structures will be mostly based on the Girvan-Newman algorithm, to find cohesive subgroups in a network (coherently with the main results of the previous SAB project “Repin”). We will dispose of a typology of personal networks that accounts for factional structures, core-periphery structures, tight-knit communities, and hybrid forms. So we will test the hypothesis that In factional and diverse networks, segregation is correlated to higher level of political participation; in core-centered and homogeneos networks, segregation means cultural differentiation, close ethnic identity and a strategy of invisibility.

Having read the literature, and having done qualitative preliminary analysis in the past, in the RONEPP project we will be able finally to test this hypothesis. In factional and diverse Roma personal networks, segregation means that Ego is the broker, the bridge, the mediator between different cliques, social circuits, social environments etc, each one homogeneous within. This position makes Ego more adaptive and less attached to a strategy of invisibility. In core-centered and homogenous networks, segregation means that not only there is significant homogeneity in the network, but there is still more homogeneity in the central core of the network. Not only has Ego a personal network dominated by co-ethnics and endogamy (homogeneity in the whole network); but his/her network revolves around a central, very cohesive and closed core, which is even more homogenous than the rest of the network (subgroup segregation). Our hypothesis is that these network characteristics make Ego “less” engaged in collective action and “more” committed to an invisibility strategy. At the moment no one has ever tested this hypothesis for Roma, or more in general for highly stigmatised minorities. Comparing Roma individuals in two different Countries (France and Italy) will make our test even stronger.

Methodology and work plan

The research agenda is organised in three steps.

The first step consists in generating knowledge on all the national and local Roma mobilisations in France and in Italy. These mobilizations will be listed from 2010 to the end of 2014, so to cover a period of 5 years, before and after the elaboration of the National Strategies for Roma inclusion. The dynamics of the institutional contexts will be taken in into account. Classifying Roma political collective action and engagements thus will be the first milestone. The description of these mobilisations will be possible analysing media, Roma NGOs documentation, and through 11 interviews in France, and 11 in Italy to Roma leaders (the 18 main leaders of Roma organisations in the two countries). Additional source will be the “Roma Pilot Survey” (EU Fundamental Rights Agency 2012).

The second step of the project is the gathering of a large collection of interviews. Half of the sample will be composed of Roma activists involved in Roma NGOs, while the other 50% of the respondent will be non active Roma. These interviews will explore two main analytical question:

  • the meaning of collective action, identity politics, and rights claiming.
  • the form of the network of the respondent (via a technique de personal network analysis, already used for the Repin project).

Obviously socio-demographic variables will be collected. The sample will not include only individuals living in extremely precarious condition, but it will be stratified by class, gender and age.

The third step will finally analyse the data collected and proceed to develop a deeper description of motives for and against collective action among Roma groups. In this WP3 will be analysed eventual differences in the personal network of activists and non-activists, as well as the meaning of collective action, and the level generality of their claim making, from funds lobbying to participation in local policy making, to contentious politics too. The analysis of personal networks will focus on 3 dimensions: network composition, network structure in terms of cohesive subgroups, structural segregation in the network. The analysis will proceed in 3 steps: 1. We will identify typologies of personal networks along each of the four dimensions mentioned above. 2. We will study the association between network types in different dimensions. 3. We will use the typologies in the three dimensions as three categorical independent variables in predictive models of collective action and political participation (we will estimate linear regression models for collective action and political participation, as dependent on both Roma's individual and network characteristics). We will control for standard individual variables: Ego's nationality, Gender, Age, Time since migration, Educational status.