Sonja Zmerli, professeure de science politique à Sciences Po Grenoble et au laboratoire Pacte, @SonjaZmerli
Is the researcher’s neutrality possible and desirable?
One of the major privileges that ‘doing scientific research’ confers is exemplified by the uncompromisable freedom of research and teaching. Or in other words, scientists and by the same token social scientists as well, in their professional pursuit of ‘doing research’ are ideally driven by their own curiosities, observations, and interpretations, all embedded in and instilled by their scientific knowledge and expertise.
To take it even one step further, individual value orientations and belief systems might also partially explain why the interest in and pursuit of some research objects appear to be more pertinent than that of others. From this vantage point, it is questionable whether a social scientist can be truly neutral in determining her paths of research while exercising her granted scientific freedom.
Such a relativisation may be disconcerting at first, since the public generally expects value-free scientific findings that can be objectified and are based solely on empirical evidence. Should this idea of neutrality thus be revised, since it seems to be based on a premise that cannot be fulfilled? Or should the requirements for a scientist’s neutrality towards her research object be defined more broadly, so that they are oriented less towards the scientist’s underlying motivation and more towards the value-free and knowledgeable exercise of her scientific ‘craft’, which should guarantee the objectifiability of the research results?
This includes, who could deny it, extensive knowledge and conscientious implementation of the principles of scientific work as well as the methods of the social sciences, without which scientific findings as such simply cannot be produced. Against the background of these requirements, it is not without reason that the IEP Grenoble places one of its teaching foci on imparting solid theoretical as well as application-oriented methodological knowledge.
What role do methods play in your research approach?
Comparable to learning a new language, which first enables one to formulate meaningful sentences then statements in order to subsequently communicate them to members of this newly emerging ‘language community’ and to engage with them in an exchange about these statements, social science methods form the analogous grammatical backbone that defines clear prerequisites for the ‘syntax’ of the scientific process.
Often though it seems as if social scientists in ‘doing research’ have a preference for some methods over others. There are some of us, for instance, who indulge in analyzing quantitative data, such as population surveys, media contents or so-called big data, which comes with the alleged advantage of producing statistically verifiable findings on causal relationships. Others, by contrast, are more concerned with acquiring a deep understanding of research objects that present themselves as single cases, thus choosing qualitative methods as adequate tools, such as individual or group interviews, ethnographies or hermeneutics. Irrespective of the method chosen though, it is indisputable that each comes with its limitations in terms of scientific achievement and advances. The question then is to what extent each social scientist is ready and willing to leave her familiar ‘territory’ and embark on a new scientific journey using a new ‘compass’.
As a quantitative political scientist, I have been experiencing these limitations myself in pursuing my topics of scientific interest: Having broadly conducted quantitative research on the causes and consequences of political trust in democratic societies, I have increasingly been ‘alerted’ by findings that demonstrate how detrimental economic inequality is to citizens’ confidence in political institutions. What initially started out as a rather ‘neutral’ acknowledgment of this occurrence, turned into a conviction and a broader empirically-driven desire to better understand how citizens view, experience, think and feel about these inequalities.
Clearly, my familiar ‘toolbox’ of quantitative methods can only take me so far in advancing an in-depth knowledge. For this matter, I will have to turn to new and different data sources that need to be explored by qualitative methods. I am confident that this new territory will not only provide instructive insights on the subject matter but enhance and broaden more generally my own views on how to do research in the future.
Yet, regardless of whether qualitative or quantitative methods are used during this process of acquiring scientific knowledge, it is indispensable that the so-called intersubjective verifiability of scientific statements and their origins remains guaranteed. Accordingly, scientific findings are only to be recognised as such if, firstly, defined research objects are actually observable, secondly, the methods applied and primary or secondary sources used are disclosed and, thirdly, on this basis, the scientific findings obtained can be replicated or, according to Carl Popper falsified, by anyone.
Certainly, these requirements set a high standard for the practice of social science research and thus constrain each scientist to continuously practice applying this ‘grammar of science’. Whether this succeeds is decided not least by the members of the scientific community who, mastering the same language, take note of published study results, critically scrutinise them and, if possible, continue to advance or refute the knowledge.
At the same time, however, these preconditions for the acquisition of scientific knowledge expose a central weakness that may have contributed to the polarization of public discourse, especially in recent times: the communication of social science studies to the general public requires honest brokers of knowledge who master the ‘grammar of science’ and recognize and appreciate it. Drawing on the mediatized discourses that have recently brought the IEP into the national limelight, it becomes clear, in my view, that most media pundits were neither familiar with the grammatical backbone of the social sciences nor felt the desire to appropriate this body of knowledge. One possible outcome is that trust in social scientists and their expertise may have been attenuated, who, in return, may be even more called upon to communicate their scientific knowledge and practice to the public – a type of communication, i.e. ‘vulgarisation’, that seems already to be more entrenched in some French leading media than in other countries.
These times are challenging times for liberal democracies worldwide. Overtly or covertly, their core principles and achievements have come under attack from public figures, politicians, political parties, interest groups, social movements with anti-democratic sentiments or even from their own governments or adversary states with strategic anti-democratic agendas. Social media platforms with their alternative communicative space, creating so-called echo chambers of like-minded others, certainly facilitate and amplify these endeavors. To be sure, liberal democracies offer ample ground for criticism, demands for renewal or transformation. Yet, European history of the 20th century reminds us of the dangers inherent in lightheartedly dismissing the ‘rules of the game’ of liberal democracies or in other words violating their constitutive normative and legal framework.
As a political scientist but maybe even more importantly as a democratic citizen I observe these multilayered and sometimes only subtle anti-democratic dynamics in Europe, the U.S., Asia and Latin America with grave concern. Admittedly, these attempts come in many forms and are particularly effective when they succeed in undermining the perceived trustworthiness of agents of the free press, scientists or other experts for that matter.
Which other recent event could be more emblematic of the politicized ‘crisis of scientific expertise’ than the pandemic caused by Covid-19? Public opinion surveys in the highly politically polarized U.S. on this topic suggest, for instance, that trust in scientists and attitudes towards the Covid crisis stand in stark contrast with trust in former President Trump. Yet, the breadth of scientific disdain is ever growing and appears nowadays to successfully target, in France and elsewhere, the social sciences, too. Some of these critics may be disingenuous in as much as they aim to instrumentalise and exacerbate legitimate criticism of the social sciences for political or ideological gains, others, by contrast, may raise sincere concerns about the foundations of social sciences that could, at least, be partly addressed by scientists being transparent about their motivations, objectives and modus operandi.
How would you define objectivity in the social sciences?
While the knowledge and application of the ‘grammar of science’ opens up a certain scope for the interpretation of a scientist’s neutrality, the requirement of objectivity of social science research findings appears to be more difficult to fulfil, especially since these findings can only ever make visible partial sections of social science contexts, and can thus only ever result in approximations to an object of investigation, but never its complete comprehension and explanation.
Depending on the formulation of the research question, the underlying temporal horizon or the analysed data, different findings can result with regard to one and the same object of investigation, which can even contradict each other. Only the continuous critical-constructive exchange based on scientific criteria, which, however, presupposes the objectifiability of social science findings, is most likely to do justice to the claim of objectivity, even if it is ultimately not able to redeem it.
To be sure, this apparent ambiguity of social science research sometimes causes irritation and frustration among university students. In their initial expectation of a systematic acquisition of knowledge that, similar to the natural sciences, teaches, if not laws then at least regularities about the coherence of the world, they often find themselves confronted with probabilistic statements that may raise more questions than they provide reliable knowledge. Sensitising students to this peculiarity of social science research and sharpening their critical view is consequently one of the core tasks of methods training, as it has already been established at the IEP Grenoble.
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 would like to address this question in a more general manner by outlining recent developments in the social sciences that may negatively affect social scientists’ underlying motivations to advance and increase scientific knowledge and, as a consequence, would lend criticism to their trustworthiness and to the social sciences writ large. More specifically, I refer to the changes in criteria for the recognition of scientific achievements which usually reflect institutional guidelines and, undisputedly, impact scientists’ most profound decisions.
What has already found its way into the university evaluation schemes of scientific achievements in a number of European countries for some years now seems to increasingly take roots in French universities as well: scientific career opportunities increase with publication achievements, preferably in international quality peer-reviewed journals. There are undoubtedly numerous advantages associated with this kind of cross-border and cross-linguistic exchanges of scientific knowledge. And yet, an evaluation of scientific achievements which mainly focuses on bibliometric criteria can lead to undesirable side effects with detrimental repercussions for the scientific community more generally.
To put it more bluntly, a researcher’s desire to advance her academic career and thus to publish in certain prestigious journals can influence her choice of methods, data, time horizon or even the units of analysis. In the worst of all cases, research findings are even manipulated in the hope of increasing the chances for publication, or, conversely, so-called non-results of studies do not find their way into the scientific community because they do not appear as reportable as others from the perspective of journal editors.
To counter these aberrations, an increasing number of journals have started to publish scientific articles only in conjunction with the study’s underlying data basis and calculation methods, so that intersubjective verifiability is readily possible. By the same token, it is being discussed to give the publication of replication studies greater importance than before. Moreover, the advancements of Open Science, which is taking place with extensive national and European support, promises to cut back the powerful roles played by key international publishing houses and thus to promote greater diversity and independence in social science research.
The appeal of the social sciences, this could be stated in conclusion, lies in its continuous advancements, in respectful and critical-constructive exchanges, in reaching boundaries of knowledge and crossing them with new means. Its dangers, on the other hand, lurk predominantly where political or market-oriented forces subtly or not so subtly impair scientists’ scope of decision-making or where scientific findings are abridged or falsified and instrumentalised for political or ideological motives.
Democratic societies cannot do without the freedom of science and teaching. Maintaining it and filling it with life is therefore a societal task. By the same token, scientists have a duty to meet the requirements of their profession and to communicate their findings to society in a comprehensible way. The underlying causes and intentions of the polarisation of societally ‘sensitive’ topics could thus be more easily unmasked and their breeding ground removed.