Suddenly, colleges and universities have become arbiters of life and death. American institutions of higher education have acknowledged — implicitly or explicitly — that there is “no way to reopen safely” during the uncontained novel coronavirus pandemic. Those who have returned to campus have faced the clear and present danger of contagion. In a dramatic reversal of its reopening policy, the flagship public university University of North Carolina at Chapel Hill pivoted to online education for its undergraduates on August 17, after about 130 students tested positive for COVID-19 within the first week of classes. A day later, the University of Notre Dame temporarily suspended all in-person classes due to an even larger outbreak on its residential campus during week one of the semester.

Yet despite the rapidly escalating numbers of COVID-19 cases on campus, this elite private Catholic research university has kept its undergraduates in the dormitories at full capacity, and expressed its intention to reopen its campus again by Labor Day weekend for in-person classes, work, and other activities including Division I football games.

Despite the glaring risks to the public health of their local communities, the majority of American colleges and universities (60%) announced by early July their intention to bring tens of thousands of students back to their campuses to resume in-person classes.The minority (33%) were then planning for either online or hybrid models of instruction. Many factors played into the making of these two contrasting policies, including financial imperatives and administrative confidence in measures to manage risk (such as testing, contact tracing and isolation, syndromic surveillance, physical distancing, and mask wearing).

The choice between in-person and online/hybrid instruction, however, ultimately rotated around how the risk of the pandemic has been characterized and evaluated by different actors. While the vast majority (77%) of private institutions deemed, by early July, that in-person instruction was worth the risk given the mitigation and containment measures proposed, 45% of public institutions judged that the risk of fully resuming in-person instruction was too high and therefore unacceptable. By early August, however, the overall trend had reversed, with online and hybrid modes of education prevailing in campus reopening plans.

It is crucial for faculty, staff, students, and the local communities surrounding our colleges and universities to understand and evaluate the rationales behind these divergent risk appraisals so that they are assured of the reasonableness of the decisions taken.

There are two features of risk governance for health that are immediately relevant for judging the reasonableness and adequacy of the two contrasting risk management positions that presently characterize American higher education. The first is the extent of scientific understanding of the pandemic; the second is the values and the stakes involved in making judgements about reopening by colleges and universities.

The emerging consensus within the scientific community is that the SARS-CoV-2 pandemic represents a “complex adaptive interconnected infection system.” The transmission of the novel coronavirus exhibits all of the features of such a complex system: deep uncertainty, high unpredictability, complex interdependencies, non-linearity, emergent properties, and “fat-tailed” risk. Such systems present particular challenges for risk appraisal because of their propensity for causing low probability, high consequence, events resulting in extreme unexpected surprises, especially within the tail of the infection distribution.

These characteristics make conventional risk-management approaches — based on naïve empiricism and “best estimates”  — inadequate for handling infection risk.This is because none of these risk assessments can be made by simple “scientific” exercises alone, as the uncertainties of the pandemic are too large to make accurate estimates and predictions. Nor can they be made by simple empirical determinations of the right balance between over- and under-protection. They are far too complicated to be prescribed by some algorithm that assumes regularity, simplicity, and certainty in the phenomenon and in our interventions. The situation is further exacerbated when stakes are high, values are in dispute, and decisions are urgent.

If the scientific evidence is uncertain or contested in a campus’s COVID-19 risk assessment, then the decisions taken are primarily driven by three subjective assessments of value: (1) divergent institutional values that underpin individual perceptions of risk, (2) the leadership’s degree of confidence in their ability to control campus outbreaks in the context of their culture and behavioral norms and their available financial resources, and (3) varying interpretations of the imperative to safeguard local communities during an ongoing pandemic.

In fact, in these circumstances, these diverse value-loadings also dictate who is likely to be included within the risk appraisal process and whether the decisions made are in accordance with the preferences of all relevant actors. Broader concerns for the public good generally allow for wider representation of all germane perspectives in the deliberative process, in contrast to top-down hierarchical approaches that favor a narrower set of values. This inclusion of wide representation is even more important when decision stakes reflect conflicting purposes among stakeholders. An extended peer community is also important for evaluating the “scientific” inputs used in decision making.

Members of colleges and universities and their surrounding communities should apply the “precautionary principle” in their ongoing decision making concerning the reopening of American campuses.

Put simply: if the consequences of an action can be serious and are subject to scientific uncertainties, then the risk cannot be tolerated. Precautionary measures should be taken — that is, avoiding extreme or “fat tail” risks through caution — or the action should not be carried out. Planning for online, or possibly hybrid, education this fall is precautionary in this light. Such decisions reflect an appropriate respect for uncertainty in the face of possible catastrophic outcomes for both the campus and the surrounding communities. Online teaching and learning options stand out for avoiding the fatalistic fallacy that nothing prudent can be done in the face of an impending public health disaster.

The July decision of the Harvard Faculty of Arts and Sciences to offer online courses exclusively in 2020–21 suggests that the protection of academic prestige and elite branding can be compatible with prioritizing the lives and health of the people in the local community. If the most prestigious university in the country sets a trend in quality online teaching and learning, then we may see a wider reassessment of the driving values of American higher education.

Before COVID-19, public colleges and universities led the way in offering online and other remote educational options in the U.S. After COVID-19, the temporary movement of Harvard and other elite private institutions to largely online instruction might create a new norm of scholarly excellence that is more compatible with public health, if the predictions of the near future proliferation of zoonotic viral pandemics like the avian flu and SARS-CoV-2 are even remotely realized.

Keeping in mind the possibility of catastrophic consequences for public health, local citizens and members of college communities ought to push academic administrators to ask: Is any educational mission “worth the risk” of causing the otherwise preventable death of a human being from complications of COVID-19? Is it justifiable to fully reopen campuses this fall, given that a vaccine will likely be available by year’s end?

Our answers to these questions reveal our own deepest values, just as they uncover the real, driving values of colleges and universities nationwide. The time for reconsidering the weights we place on those academic values is now, before the lives and health of tens of thousands of students, faculty, staff, and community members are put at risk by opening up campuses too quickly during the pandemic.

Eileen Hunt Botting is a professor of political science at Notre Dame and a political theorist; her next book is Artificial Life After Frankenstein (Penn Press, December 2020). 

Edwin Michael is a professor of epidemiology at the University of South Florida and an infectious disease modeler; he leads the development of the SEIRcast webportal for facilitating the forecasting and management of COVID-19 transmission at the county level in the United States.