Marine Bourgeoisa lecturer in political science at Sciences Po Grenoble and the Pacte laboratory, @Marine_Bgs
How would you define objectivity in the social sciences?
The quest for objectivity is consubstantial with scientific work. It is the horizon towards which all researchers, whatever their disciplinary background, must strive. Objectivity presupposes a form of detachment or even rupture between the researcher and his or her prejudices, beliefs and moral or political preferences, in order to limit subjective bias and go beyond common sense. Objectivity is a professional requirement - a skill, some would say - that can never be totally acquired. It is based on a process punctuated by stages that researchers learn to follow and master throughout their careers (from the construction of the object to writing, via the state of the art, the formulation of hypotheses, the construction of the investigative device, the use of methods, data collection and analysis). Academic institutions also offer guarantees and safeguards for objectivity, through collective and critical discussion of scientific production, and peer review.
The requirement for objectivity implies that researchers explain their position, justify their bias, specify the methods used to collect and analyze their data, and present their results without dismissing those that could weaken their argument. This criterion of transparency also implies the identification of possible biases in methodology. It is a form of intellectual honesty and professional ethics. In its absence, the conditions for scientific debate are not met: the administration of proof is flawed; the quality of the research is not guaranteed; the validity of the results cannot be discussed. Conversely, providing access to the "black box" of research enables us to compare results, revisit certain fields or re-analyze a survey. Analytical rigor, combined with empirical work, characterizes the scientific approach and distinguishes it from other registers of expression of the social world, such as journalistic writing.
Of course, there's never just one "right" way to analyze a social phenomenon. The researcher always has a choice of several "glasses": some allow a close-up view (micro), others a distant one (macro). The description and explanation of a situation are always situated, and therefore partial and biased. In so doing, the academic environment must be sufficiently heterogeneous to allow all ideas to flourish.
I'm very attached to theoretical and methodologicaleclecticism. In my opinion, it's the best way to consolidate, qualify or refute results, generate controversy and advance knowledge. When researchers using different approaches and methodologies establish converging facts on the same subject, on the basis of abundant and diversified evidence, then this can be seen as a guarantee of scientific solidity. The porosity between disciplines and specialties also enables great contributions to be made. Conversely, the homogenization and standardization of research, through the assertion of a dominant perspective or method, is a sign of scientific impoverishment.
Is researcher neutrality possible and desirable?
The neutrality of the researcher is often understood as the absence of any subjectivity: a "neutral" researcher would be a detached, uninvolved researcher, without opinion or conviction. This conception of neutrality, as the "neutralization" of the researcher, seems to me impossible to maintain, and even less desirable.
Indeed, it is unrealistic to think that the political scientist, sociologist or historian can be free of any personal orientation. Social scientists are always involved in the social world they study. Moreover, at the origin of a project, there is almost always an intellectual interest or a personal concern. The normative dimension is therefore not absent at this stage. Max Weber himself did not dispute this when he called for axiological neutrality. He recognized that preferences can influence choices of object, question and method, but called for a strict separation of facts and values in the construction of scientific discourse. This posture implies a kind of "suspension of judgment", or, more likely, forces reflexivity. Failing to be completely neutral, the researcher must specify "where he's talking from".
Establishing the facts does not preclude taking individual and collective positions. While intellectual work must not be guided by ideology, it does enable us to deconstruct the obvious, to refine our worldview: in some cases, the investigation confirms previous beliefs; in others, it leads the researcher to revise his or her initial judgment. For example, when I first became interested in the issue of social housing allocations as a means of better understanding the mechanisms of segregation, I was fairly convinced of the validity of the principle of social diversity as an objective of urban policies. Nevertheless, my work showed that this category was one of the sources of the institutional production of discrimination in access to housing. These results enabled me to put forward proposals for public action to fuel the debate on housing policies. What may seem like " academic activism " takes place after and outside scientific discussion.
There is a strong social demand for researchers to take part in debates. They are constantly called upon by the media and other social players (ministries, local authorities, consultancies, associations, schools) to share their expertise. In fact, the knowledge produced by researchers circulates and produces effects. This dimension of transforming the social world is part of the DNA of the social sciences. It is valued in the academic world as long as it does not abstract from the elementary rules of scientific debate. Initiatives such asOuscipo and Politeia offer researchers the opportunity to promote their work, making it more accessible and enriching the questioning of civil society players.
To be credible, the committed researcher must rely on the knowledge produced. If necessary, he or she leaves the academic register and becomes an expert or a columnist. The development of social networks and the emergence of new, very short formats are helping to blur these roles: opinion, indignation and provocation are often preferred to scientific discourse; researchers invited onto television sets and into radio studios are called upon to simplify their words, sometimes at the risk of shortcuts and misunderstandings. This mixing of genres, combined with the instrumentalization and hysterization of certain debates, feeds the attacks launched against the University, and the social sciences in particular, which are criticized for their increasing politicization. In the face of this, researchers need to be ever more attentive to the spaces in which they intervene, and to the ways in which the actors involved seize upon their words.
How important are methods to you as a researcher?
Methods are essential in that they lend credibility to the scientific approach. They refer not only to the use of techniques for collecting and analyzing data, but also to the definition of a relationship to research, a posture with regard to the objectives pursued, a strategy, and the rigorous implementation of a demonstration based on facts and observations.
The methodological choices that guide my research stem from the questions I ask myself. In my thesis, I used mainly qualitative methods. I carried out ethnographic surveys, combining interviews and observations, in order to get a close-up view of the work of attribution. This combination of methods proved particularly fruitful in ensuring that the study of professional practices was not based on discourse alone, and in highlighting the distortions between what actors say and what they do. When it comes to ethnicity, discrimination and racism, the risks of the unspoken and of concealment are also considerable.
The main difficulty lies in the social desirability bias, which consists in interviewees saying what they think is appropriate. Observation made it possible to reduce this bias by establishing a relationship of trust with the interviewees and by making it easier to gather their views. For this survey, I also used a comparative approach to identify the conditions under which discrimination occurs and to explain its recurrence. The emergence of new questions then led me to diversify my methods: press reviews, archives, descriptive statistics, content analysis. Career-long training in methods is an essential aspect of the job, enabling me to tackle new questions and/or re-investigate subjects in a new light.
The question of methods is far from being confined to purely technical considerations. The link between epistemological posture, research objectives and data collection and analysis methods is at the heart of what I try to convey in my teaching at Sciences Po Grenoble. Research design and the comparative approach always play a central role. Teaching methods also means becoming aware of the place they now occupy in scientific production. We have to admit that methodological aspects are still largely neglected in publications, particularly when it comes to data analysis.
Beyond the description of the material and the presentation of the protocol, access to the "kitchen of research", and the reflexive posture that this implies, are still too often relegated to second place in French-language journals, even if the Open Science approach is leading more and more journals to ask academics to make their data available and explain their methodologies, as is the norm in a large proportion of international journals. Enhancing the value of methods, both in training courses and in publications, would also be a better way of responding to some of the spurious accusations levelled at the University.
Could you present an example of research, ideally from your own work, to illustrate the issues and tensions surrounding objectivity and neutrality in the social sciences?
I have been working for nearly ten years on the processes of sorting and selecting populations in public action, on issues of access to rights, inequalities and discrimination, with a particular interest in urban and housing policies. I'm part of a long tradition of research that rigorously documents the social logics of discrimination. In particular, I show that, for social housing managers, racialized groups constitute "at-risk" groups whose presence must be limited in certain stairwells, buildings or residential groups, since tenants may react with rejection towards these groups. Some of these actors themselves develop a belief in the cultural maladjustment of racialized people, and deploy simplifications and routines that associate ethno-racial belonging, lifestyles and ways of living. These strategies and rationales lead to the institutional production of discrimination, the sources of which are now well known.
The persistence of ethno-racial discrimination has been scientifically established in several areas of public action (housing, employment, education) by numerous empirical surveys, based on qualitative and/or quantitative methods (testing and statistical analyses). What's striking, however, is how little non-academic actors know about this work, or even reject it. The controversies surrounding race and Islamophobia are a case in point. While the concepts used to describe these phenomena are debatable, their existence is no longer in doubt. In public debate, as in the academic world, the choice of terms is never neutral and is open to criticism. I experienced this first-hand in my thesis, when I used the categories of race and ethnicity to describe what I was observing empirically. The use of these categories led me to clarify the conceptual issues they cover and to position myself in a particularly heated race/ethnicity controversy. Once again, the important thing was to make my biases explicit and to argue them!