Investing in scientific diversity: Why theoretical arguments aren’t enough

MONDAY, 15 FEBRUARY 2021

A sensible investor always hedges their bets. The same applies to good scientists — rather than exclusively concentrating on one novel method, theory, or idea, those who are keen to maximise their returns will spread their energy across a range of alternatives. Scientists should diversify their assets, especially when exploring scientific unknowns.

A primary argument often made in support of diversifying the scientific workforce is closely connected to this rationale of “hedging our bets”. Individuals who have had different social and intellectual experiences tend to think differently, hence, a diverse community is more likely to offer a wider range of theories and methods^[1]. This reasoning is part of a broader set of arguments that constitutes the “diversity-promotes-excellence” framework. In short, this framework suggests that it is in the scientific community’s best interests to include people of diverse geographies, languages, races, genders, abilities, and socioeconomic backgrounds among its researchers.

While these arguments are theoretically sound, and provide an important impetus to challenge the status quo, they are by no means sufficient to make scientific communities more welcoming to those who currently constitute its minority. To illustrate this point, I present several examples from my own discipline, Social Epistemology, a field dedicated to understanding how the nature of our scientific communities impacts the knowledge they produce. These examples show how scientific social structures themselves must change if we are to reap the benefits that diversity has to offer. The key lesson to be learned from the dynamics of scientific social practices is that theoretical arguments in favour of diversity simply aren’t enough to drive the changes that science so desperately needs.

Why diversify science?

The recognition that non-diverse, homogeneous thinking is no better than dogmatism goes at least as far back as John Stuart Mill. In his canonical work, On Liberty, he emphasises the “tyranny of the majority”, whereby those who constitute the majority create an oppressive environment for those in the minority, coercing them to abandon their unique opinions and instead join the “wisdom of the crowds”^[2]. Mill encouraged a society where individuals are granted the freedom to think as they please, so long as they do not harm one another. He argued that in such a free society, a diversity of ideas will flourish, generating a better society which has the ability to select the best ideas to address each problem encountered.

Mill’s image of a free society was incorporated by the radical philosopher, Paul Feyerabend, into a new approach to science in the twentieth century. Feyerabend argued that a plurality of methods and ideas is necessary for science to progress^[3]. Given that a diverse community is more likely to produce a plurality of ideas, diversity is necessary for progressive science.

Philosophers of science today credit diversity with at least two more virtues: overcoming bias, and achieving objectivity. Often, it is hard to identify the problematic aspects of one’s own theory. This is especially true when the problems come from implicit, often subconscious, biases that we may unknowingly possess. Only those who do not have those same biases as us are able to identify these pesky predispositions^[4]. For example, many theories of the evolution of human behavior credit the hunting practices of early men with the development of human social skills. It was only when questioning why the same cannot be said for early women’s gatherer and childcare practices that the possibility that behavioral theorists were biased towards male practices became evident. Importantly, it was only once a significant number of women had entered the field of evolutionary behavioral science that these biases were uncovered^[5].

Closely related to the notion of uncovering bias is the idea that for a theory to truly be objective, it must be vetted from a diversity of perspectives. While we tend to think of an objective statement as something that is simply True, and made evident from a God’s eye perspective, it is more practical to think of a statement as objective when those with a diverse range of perspectives agree that it is true. If a statement only seems valid from a particular group’s perspective, then the objectivity of the statement should be questioned^[6]. Therefore, encompassing as many perspectives as possible in science is necessary to endorse scientific claims as truly objective.

While the arguments presented here offer undeniable support to the notion that a diverse scientific community is a better scientific community, achieving these results requires science to be conducive to the flourishing of alternative viewpoints. For the reasons I outline below, we have serious reason to doubt that this is the case at the moment.

The cultural Red King effect:

Science is inherently social — scientific work is enabled by ongoing interactions between researchers in laboratories, academic institutions, societies, and conferences. In particular, scientists interact by collaborating and sharing resources. Despite the fact that the theoretical arguments described above indicate that increasing the diversity of scientific communities leads to better outcomes, such virtues will only manifest if minority viewpoints are allowed to flourish. However, minorities often lose out when engaging in these interactive practices, conforming to the norm rather than making a difference. This phenomenon has come to be known as the cultural Red King Effect, and it appears in all interactive communities, not just in science^[7].

The philosophers Cailin O’Connor and Justin Bruner have developed an ingenious way to illustrate why minorities tend to lose out when engaging with peers, rather than encouraging the diversification of ideas worth investing in^[8]. They have used an evolutionary game model to show that it is the very fact that they are minorities that leads to them conforming to the majority view. Evolutionary game models simulate group learning, or adaptation, brought about by exchanges with peers. The model was originally developed to test Darwinian evolutionary claims regarding the adoption of new traits brought about by competition between organisms^[9]. Minorities are represented in the model as a group with a particular set of behaviors, while the majority group has a different set of behaviors. When minority and majority members interact, the majority member could either adopt the behavior of the minority, or vice versa. For the minority’s novel ideas to be invested in, and to achieve the theoretical benefits of diversity, a critical number of majority members must engage in minority behavior. However, since it is more likely that a minority member will encounter a majority member, simply due to there being more majority members, it is probable that, over a set number of interactions across a period of time, minorities will switch their behavior to match the majority. Hence, over time, the novel views offered by the minorities get wiped out, rather than adopted as a viable alternative by the community. What the cultural Red King effect highlights is that performatively hiring a couple of “token diversity representatives” offers no substantial benefit to the department, nor to the token-hires. The mere presence of a handful of researchers who do not conform to the norm makes minimal difference.

Diversity free-riding:

A limited amount of the virtues of diversity discussed above can be extracted without any of the associated benefits to minorities. Other faculty members can “free-ride” off the insights of their co-workers, meaning that the originators of those ideas tend to miss out on receiving the credit^[10]. To understand how this process takes place, we need to understand the difference between formal and informal scientific communities. Formal communities refer to conferences, peer review, and laboratory teams. In formal communities, there are rigid procedures in place for who does what work,and who gets the credit. Informal communities include break room conversations, casual chats, and extracurricular seminars and lectures. These informal communities are important for the day-to-day conduct of science, but they pose a threat to minorities. Those in the minority can provide novel insights for their peers in the majority through these informal interactions, which do not always translate into increased, fairer collaborations in the more formal aspects of scientific interaction. In relation to the Red King effect, diversity free-riding highlights that, even when those in the majority do learn from the minority, they do so in a way that does not benefit the minority, is in limited amounts, and is done on their own terms.

Furthermore, the “diversity work” carried out in informal communities is often done by junior, early career, or even graduate faculty members. This is significant in the light of the power dynamics at play between senior and junior researchers. For an alternative position to be viewed as a viable alternative, worth investing energy into, the proposer must be viewed by the community as someone who is likely to have put forward an idea worth considering. In other words, equally viable options must be put forward by equal practitioners, namely, peers. Hence, the theoretical arguments for increased diversity in the workplace requires diversity at the highest level. This means alternative positions must be given enough legitimacy to be viewed by the majority as being worthy of pursuit. Relegating diversity work to free-riding off junior researchers fails to fully, and effectively, diversify the scientific workforce.

The deeper issue at play is the existence of several barriers preventing those who currently constitute the minority from rising up the scientific hierarchy. Limited child support and parental leave, publish-or-perish culture, and implicit biases in peer review all act to prevent the scientific community from adequately diversifying.

How to move forward?

One way to overcome the cultural Red King effect is to increase the number of people in the minority, i.e. remove the existence of a hegemony, or “majority”. This can be achieved by not only diversifying, but doing so in a way that is proportional: having a “token” scientist from a different race, gender, or nationality is inadequate. To overcome the free-riding issue, we need to establish platforms for those in the minority to offer their contributions on their own terms. We must ensure that colleagues receive fair credit. To do this, we can concentrate on establishing cultural practices that protect the intellectual interests of those in the minority.

The key point here is that the theoretical virtues of increasing diversity can only be obtained through systemic change in the community structure of science, rather than at an individual level. The abstract, theoretical arguments often used by those who wish to encourage institutions to invest in diverse faculty members can only get us so far. If we wish to improve science for its own sake, and make it a more welcoming place to those who have historically been excluded, we must change the way our scientific communities are structured.

Charlotte Zemmel, Section Editor at Bluesci and co-editor of the Minorities in Science Series, explores why theoretical arguments for diversifying science aren't enough

References

[BACK] Kitcher P. The division of cognitive labor. The journal of philosophy. 1990 Jan 1;87(1):5-22.
[BACK] Jacobs S. John Stuart Mill on the tyranny of the majority. Australian Journal of Political Science. 1993 Jul 1;28(2):306-21.
[BACK] Feyerabend P. Against method. Verso; 1993.
[BACK] Longino HE. Science as social knowledge: Values and objectivity in scientific inquiry. Princeton University Press; 1990 Feb 21.
[BACK] Asquith, Pamela J. A woman of science: sorting fact and illusion in gender and primatology. Primate Life Histories, Sex Roles, and Adaptability. Springer, Cham, 2018. 79-89.
[BACK] Harding S. “Strong objectivity”: A response to the new objectivity question. Synthese. 1995 Sep;104(3):331-49.
[BACK] O’Connor C. The cultural red king effect. The Journal of Mathematical Sociology. 2017 Jul 3;41(3):155-71.
[BACK] Bruner JP, O’Connor C. Dynamics and Diversity in Epistemic Communities. Erkenntnis. 2017.
[BACK] Alexander JM. Evolutionary game theory.
[BACK] Fehr C. What is in it for me? The benefits of diversity in scientific communities. InFeminist epistemology and philosophy of science 2011 (pp. 133-155). Springer, Dordrecht.