Discussion – Key Debates, Viewpoints & Interviews
Beliefs and knowledges – The Largest Possible Experiment: The Corona Pandemic as Nonknowledge Transfer
Matthias Gross, Department of Urban and Environmental Sociology, Helmholtz Centre for Environmental Research – UFZ, Leipzig, Germany, and Institute of Sociology, University of Jena, Germany.
In 1986, after the Chernobyl nuclear disaster, sociological observers labelled the accident the “largest thinkable experiment” . They did so since the origin of the accident started during a standard experiment on the nuclear reactor that subsequently went into the wild and became a real world experiment with an unknown outcome. My argument is that the Coronavirus pandemic now has by far topped Chernobyl and has become the largest possible experiment since it involves the population of the whole planet , not only as passive recipients, but also as active participants, or even (knowingly or unknowingly) as co-experimenters. I will explain why that is, what it implies, and why it may be sociologically wise further to think along the notion of “experiment” in and with society when we analyse the spread of a virus that is shrouded in uncertainties and ignorance.
The Coronavirus was first debated in the media in the final days of December 2019. After the disease was discovered in the Chinese city of Wuhan, it became clear that not too much was known about the way it spreads, the way it affects people, and how it should be medically treated. All actors involved – I.e. nations, governments, scientists, and basically all citizens – were challenged since the world had been turned into a laboratory, and everybody became part of the experiment whether they wanted to or not. Whereas the ideal natural science laboratory shuts out risks from the society outside, conducting a real world pandemic experiment loads all risks, dangers, and uncertainties, but also new possibilities, hopes, and learning processes, onto society at large.
One could say that the boundary conditions of the laboratory now are the ends of the world (as yet). It became the largest thinkable experiment running in real time, and the experimenter and the object of the experiment had not been decided upon clearly yet. In addition, this real-world experiment has many side-lines and many – for lack of a better term – “sub-experiments”, where scientists computed, correlated and combined data from people who felt sick, who had died, or who tested positive with the virus, etc. However, up until now no one knows what the right path was or is, or how many people actually have died from the virus, or merely with the virus. Is relying on multivariate statistical analysis of epidemiologists the best basis for decision-making, or is it better to rely on empirical research, which obviously will take time? What would be a measure for success at all? Would it be minimizing infection rates, or effectively treating new cases in well prepared hospitals at a certain rate? Or perhaps there is a tipping-point as to when to choose one or the other? Very often decisions must be made well before “official” scientific evidence, and sometimes decision-makers do not seem to know clearly what questions need to be answered. Consequently, in the end all decisions have to be political decisions. Is learning from success and failure the right path, as it is sometimes in laboratory experiments? But at what costs? A comparison between the strategies in most European countries (lockdown, stay at home orders, social distancing) and the Swedish case seemed a good opportunity in this regard . But how reliable and meaningful could comparisons on such scales be? How can such an experiment been managed responsibly, if at all?
If we as sociologists want to frame the 2020 Corona pandemic as a world or global experiment, several conceptual issues should be addressed. One could surely argue that the largest possible experimental process is going on worldwide right now. But based on these processes, geographically smaller laboratories are set up as results of the large experiment in order to compare different national experiments, such as the treatments in China (Wuhan region) with that of the USA, for instance. Furthermore, the definition of what constitutes a laboratory is then determined after worldwide processes have unfolded and have shown impacts on national or even regional levels (northern Italy versus central Spain, for example). It thus seems that before any laboratory has been consulted, findings need to be translated “back” from the real-world experiment to corresponding local or regional laboratory settings . Such laboratories are raised, for instance, by testing stakeholders in laboratory-like settings (e.g., the German case of the town of Heinsberg as an early centre for the pandemic, which was selected for an intensive study in late‑March 2020 to become a “Coronavirus laboratory”, as The Guardian labelled it ), in order subsequently to develop and test medicines by academic and industry researchers in controlled spaces. In this view the public can be understood as a place to test the efficacy of political change programmes in the face of unexpected and surprising changes. This assumes that modern society should be ready for the application of the experimental method to public policy and reform.
In that sense perhaps most important in conceptualizing experiments for understanding and analysing phenomena such as the spread of the Coronavirus is, I believe, accepting the relevance of ignorance and nonknowledge. After all, abrupt social changes such as the arrival of the Corona pandemic and experiments share some crucial similarities. An experiment in the most general sense can be defined as a cautiously observed venture into the unknown. Even more so, an experiment is arranged to generate unexpected events, so that the surprising effects derived out of the experimental set up can be seen as the motor force for producing new knowledge, since surprises help scientists become aware of what they did not know . A hypothesis, in this stream of thinking, can be rendered as a specification of what is unknown or what can be called nonknowledge – knowledge about the exact realm of what is unknown, but which can be used for decision-making and further planning. Following the Popperian encouragement only to test hypotheses if one seeks disconfirmation , then in order to falsify a hypothesis, the experiment needs to fail. If one knows that the hypotheses have been unusable, then one could say that the experiment itself was useful. Thus understood, only experiments that fail are successful experiments or, put more generally, sudden unexpected changes that make the experimenters aware of their own nonknowledge (hypothesis falsified) are the impulse for new knowledge.
Classical risk assessments assume that the probabilities of occurrence of the relevant events in a certain area under consideration are known. In these terms, dealing with nonknowledge in experimental settings clearly differs from taking or limiting risks, since the risk of a certain event occurring presupposes knowledge of both the character of events that may occur and of the probability that they will do so. In the case of the Coronavirus, it is clear that we know more about what we do not know than what we know. Thus, probabilities and risks are not available in a reliable manner, and they should not lead to a false sense of precision . At least from a sociological perspective, however, this should not lead to despair. After all, science itself is part of that large-scale experiment in society. The basic logic here is that such an experimenting society is not only a society that allows and helps the conducting of real-world experiments, but that society is itself an experiment. An experimental society thus would mean to put into practice Friedrich Nietzsche’s  well-known aphorism: “We are experiments: let us also want to be them!”. The problem is, we cannot get out of the experiment, and we are part of it, whether we want that or not.
From a sociological perspective, such an experimental approach can be conceived of as a way of coordinating the contingent activities of diverse actors (e.g. political parties, virologists, economic players, industries), which are continued despite an acknowledged awareness of ignorance, so that processes do not have to be interrupted. It is decision-making based on not knowing. This is the biggest challenge the world is facing with the Coronavirus. From a sociological point of view it should be clear that instead of glossing over knowledge gaps with risk assessments or sidestepping them by rhetorics of certainty, what is needed is a way to describe nonknowledge so that, for instance, policy-makers have an alternative to risk assessments based on limited data and figures when communicating with the public. In general, given the unavoidable uncertainties of the situation with the worldwide pandemic, decision-making may necessitate an open acknowledgement that nonknowledge cannot be avoided. In this way it can help to make the unknown more transparently explicit, in order to build trust and collaboration among concerned citizens, researchers, planners, and policy-makers. Georg Simmel already reminded us that trust can serve as a bridge between knowledge and nonknowledge . As in the case of the Corona pandemic, trust in dealing with unknowns has become a reality not by choice but by necessity. When virologists reveal their nonknowledge, they need to be clear about why it is impossible to have more certainty right now. The aim thus is not to overcome ignorance, but to develop possibilities for decision-making in spite of not knowing. After all, openly admitting and acknowledging that nonknowledge can be rendered part of “good science” might foster new trust in science and its organizations. This in turn should also help to make nonknowledge a subject of open, democratic debate.
Consequently, in addition to the oft-quoted knowledge transfer between science and society, in face of the Corona crisis we should think about strategies to develop further successful ways to communicate and transfer knowledge gaps and unavoidable nonknowledge to a concerned public and policy-makers, but also between different cultures, such as between virologists and economists, doctors, nurses, police, supermarkets, teachers, and manufacturers of masks and other medical equipment. In the same way that knowledge transfer is meant to disseminate scientific knowledge and to provide inputs to problem-solving for policy-makers, it needs to be completed with nonknowledge transfer to capture, clarify, and clearly communicate what is unknown, and make it available and understandable for the public. Nonknowledge needs to be understood as part of problem solving. Thus understood, nonknowledge transfer is even more difficult than knowledge transfer, because it is often rendered a regrettable and inferior state, even if it is clear that more (reliable) knowledge cannot be generated in a given time. Worse, much nonknowledge is difficult to clearly communicate, since it is tacit or simply uncommon to articulate in a world where knowledge is rendered a most important value and its (sociologically) natural flipside – nonknowledge – is to be avoided at all costs.
However, the practice of an experimental procedure in the interaction between humans and a non-human virus should be nothing unusual from a sociological point of view . To this end, unavoidable unknowns and uncertainties from the largest thinkable experiment and its many sub-systemic experimental settings, are relocated to corresponding laboratories, to cope successfully with what is growing by the minute: nonknowledge. Depending on the type of uncertainties and unknowns involved, these may be either traditional science labs or laboratory-like involvements of stakeholders outside the realm of science – or both. In a way, such a notion of the “Corona Experiment” may help us to imagine sociological possibilities to real-world experimentation, and to make visible what has always been there: knowledge and its natural counterpart. Thus from a sociological viewpoint, Corona can help us to accept that as much as not knowing can precede knowledge, at the same time nonknowledge can be a result of unavoidable experimental processes of knowledge generation in society. And this can be a good thing, if it is possible to foster learning processes from real-world experiments; if nothing is learned then it should not be rendered as an experiment, but rather as a directionless trial-and-error process. However, in order to learn successfully from experiments, all stakeholders involved need to find ways to cope with nonknowledge.
 Krohn, W. and P. Weingart (1986): Tschernobyl: Das größte anzunehmende Experiment. Kursbuch 85: 1-25.
 As of September 12th, 2020 only a handful of countries have not reported any Covid-19 cases. These were either island nations in the South Pacific, or countries such as Turkmenistan or North Korea, which are suspected of simply denying that the pandemic has spread within their borders. See different statistics at: https://www.statista.com (accessed September 12, 2020).
 “Maybe the Experts Were Right About Covid-19 the First Time.” See: https://www.wsj.com/articles/maybe-the-experts-were-right-about-covid-1… (accessed April 29, 2020)
 Gross, M. (2016): Give me an Experiment and I will raise a Laboratory. Science, Technology & Human Values 41 (4): 613-634.
 “Worst-hit German District to Become Coronavirus ‘Laboratory.”’ See: https://www.theguardian.com/world/2020/mar/31/virologists-to-turn-germa… (accessed April 11, 2020).
 Gross, M. (2010): Ignorance and Surprise: Science, Society, and Ecological Design. Cambridge, MA: MIT Press.
 Milne, P. (1995): A Bayesian Defence of Popperian Science. Analysis 55(3): 231-215, p. 214.
 Kay, J. and M. King (2020): Radical Uncertainty: Decision-Making Beyond the Numbers. New York: Norton.
 Nietzsche, F. (1982 ): Daybreak. Cambridge, UK: Cambridge University Press, p. 453.
 Simmel, G. (1906): The Sociology of Secrecy and of Secret Societies. American Journal of Sociology 11(4): 441-498.
 Latour, B. (2011): From Multiculturalism to Multinaturalism: What Rules of Method for the New Socio-Scientific Experiments? Nature + Culture 6(1): 1-17.
Comment on this article – log in with your ESA username and password: a comment field will appear.