Resilience: A useful concept for science to cope with digital transformation

In our knowledge-based society, hardly any development has changed scientific knowledge production and scholarly communication as profoundly as digitalisation. In our eyes, the concept of resilience is useful to inform change in science governance in the digital age. Starting with a discussion of the term, our essay discusses the potential of the concept of resilience for science in digital transformation.

Defining resilience

The notion of resilience stems from psychology, where it is conceived as a person's ability to successfully adapt to difficult or challenging life experiences. In recent years, the concept has been adapted to describe the ability of organisations, social systems or technical infrastructures to withstand and adapt to external disruptions. Broad use of the term flatters a zeitgeist that perceives the present as challenging and crisis-laden. Lately, resilience has also been described as a desirable condition for science, for instance, by the German Science Council (Wissenschaftsrat, 2021). Although it is tempting to use resilience in the context of science, the conception of resilience comes with two flaws that call for sound reflection.

First, in these conceptions the change (or crisis) to which one resiliently responds is externalised. In psychological cases, it might be adequate to understand change as an external disruption when dealing with reactions to personal fate (such as the death of a close relative). In organisational or societal contexts, however, it is often inadequate to externalise the disruption. For example, digital transformation is a fundamental and ongoing change that science helped create. In accordance, to see anthropogenic climate change as an external disruption, even a force of nature, would negate our responsibility for the crisis, and thus constrain accountability. Arguably, that would not be a good basis for sustainable change. Second, an understanding of resilience that merely allows for reaction or adaptation (i.e. incremental change) can hardly be a suitable change paradigm, because it excludes the possibility of transformative change. A purely incremental logic would imply, for example, that in light of rising sea levels due to climate change, we should build higher dams instead of changing action in politics, the economy and society to slow down global warming. In terms of science, incremental change could mean that we simply adapt to changes in digital infrastructure rather than address them.

From our perspective, resilience can only be a meaningful change paradigm, if it a) sees change context-bound and b) allows for assigning accountability. In accordance with this narrow conception, there is a flaw in science governance in dealing with digitalisation. Too often, digital change has been regarded as a force of nature to which science must adjust. In the following, we elaborate on this thesis with the help of two examples that relate to scholarly publishing and public science communication. These examples are helpful to illustrate the fundamental changes that the digital transformation has brought for science.

Transformation as label fraud: The case of scholarly publishing

The Internet has long been seen as a solution to the problem of the so-called serials crisis, a shorthand for the staggering subscription costs of scholarly journals. These hopes were well-founded, as the possibilities of online archiving made the need for a printed journal superfluous. Early proponents of open access saw this as a way to free science from dependence on a handful of commercial publishers who dominated (and still dominate) the market of scholarly publishing. This hope was articulated early on in the Open Access Declarations of Budapest (Budapest Initiative, 2002), Berlin (Berlin Declaration, 2003) and Bethesda (Bethesda Statement, 2002). Similar to what is known from other contexts, the Internet, at the time, was seen to play an emancipating and democratising role, which ultimately did not prove to be true.

Science policy has embraced open access with the aim of flipping the journal market. Indeed, more and more articles are being published as open access today (Piwowar et al., 2019). However, the market for scientific publishing remains an oligopoly where the top five publishers account for half of all the published papers (70% in the social sciences) (Kim & Park, 2020; Larivière et al., 2015). Golden open access models are being introduced, meaning that authors pay a fee (over 2,500 US dollars on average per article) for their papers to be published. In many countries, the major scientific organisations have ‘transformative open access agreements’ with leading commercial publishers (Borrego et al., 2021). These agreements are essentially a large-scale version of the gold model, where scientific organisations pay large lump sums in consortia to publishers so that the articles from its members appear open access (Farley et al., 2021).

Has this science policy been shown to contribute to resilience? Arguably not, because the so-called ‘transformative agreements’ are in fact reproducing the market power of commercial publishers in the digital age and, even worse, excluding researchers whose institutions cannot afford to buy into the deceptive transformation. They can read the articles from researchers covered by these agreements, but cannot publish open access themselves, which is a blatant violation of established scientific norms such as universalism and the idea that scientific validity is independent of the sociopolitical status of a researcher (Merton, 1973). The implied “transformation” is in fact a labelling scam; a short-sighted and uninformed measure that exacerbates the commercialisation of science. Transformative agreements mark a historical mistake in science policy.

For us, developments in scholarly publishing are an example where transformative change is needed, but incremental change is preferred.

Into the battle? Researchers attacked on social media

In 2018 the RAND Corporation released its report ‘Truth Decay’, which stated that ‘in the national political and civil discourse, disagreement over facts seems to be greater than ever before’ (Kavanagh & Rich, 2018). Researchers are (rightly) seen as playing an important role in the ‘fight against disinformation’ (Fleming, 2020). In the USA, studies by Dudo & Besley (2016) show that researchers are highly motivated to engage in public communication, in particular ‘to defend science from misinformation and educate the public about science’. However, it appears that researchers are being sent into this battle relatively unprotected.

In a 2021 survey from the journal Nature, among researchers who have spoken publicly about COVID-19, 15% reported receiving death threats as a result. More than two-thirds of respondents reported negative experiences as a result of their media appearances or social media comments (Nogrady, 2021). This problem is particularly pronounced in research areas with immediate sociopolitical implications (Valiverronen & Saikkonen, 2021). In the natural and technical sciences, cases of attacks against research on climate change (Anderson & Huntington, 2017), vaccination (Grimes, 2019), nanotechnology (Toumey, 2013), genetically modified foods and stem cell research (Nisbet et al., 2015) have been described in recent years. Humanities scholars and social scientists have been attacked for their research on multiculturalism, gender or migration (Valiverronen & Saikkonen, 2021).

Of course, not every negative experience is an illegitimate attack. However, if it is expected that researchers participate in the public discourse, it seems reasonable to protect them from attacks as a measure of occupational health and safety. Although the problem is seen among decision-makers in science governance, support structures for researchers under attack are sparse (O’Grady & Errington, 2022). In scientific training, science communication and especially its adverse effects, only play a marginal role (Fähnrich et al., 2021). Instead of strengthening decentral capacities, central communication departments at universities are being extensively expanded. These, however, provide little support for researchers' communication activities (Fecher & Hebing, 2021; Vogler & Schäfer, 2020).

Science management to date fails to meet its responsibility to support and protect researchers in cases of attacks, and thus protect science at large.

Resilience as a strategy to cope with digital transformation and its effects

The above examples are indicative of a general approach to science governance that often appears passive in relation to digitalisation; that underestimates risks and misses opportunities for change. Resilience as a broad concept can be a sensible paradigm for change if it relies on the scientific community's responsibility to actively exploit the opportunities of digitisation, and at the same time protect science from its risks.

In the first case, for example, resilient action could imply investing in innovative public infrastructure for open access, instead of subsidising publishers' transition to digital markets. This would be transformative because it removes players from the game who are not necessarily known for fair play, and would help science to regain autonomy over its critical infrastructure. In the second case, resilient action would imply an investment in decentralised capacities at scientific institutions and supporting researchers in their communication activities, including, for instance, offering legal and psychological help.

Resilience, in that understanding, implies not only the ability to cope with or adapt to an exogenously generated irritation, but aims at renewal (Wissenschaftsrat, 2021). This notion of resilience is useful for understanding science in times of digital transformation.