This paper investigates the multivocal nature of accountability in the context of grand challenges (GCs) by examining how diverse perspectives from science and politics combined and shaped the justifications provided by the Italian government in its response to the COVID-19 crisis.
Drawing on Latour’s concepts of purification and mediation as processes to frame, separate or negotiate between science and politics, we thematically analyse the Italian government’s press conference transcripts and official public documents. This analysis enables us to identify the patterns of purification and mediation through which accountability was discharged and evolved over 15 months through the pandemic challenge.
Our findings reveal how processes of purification and mediation of scientific and political aspects of the COVID-19 response contributed to the construction of a multivocal form of accountability by the Italian government. The processes are found to support the continuous redefinition and repositioning of multifaceted interpretations that are necessary when responding to a highly complex challenge such as COVID-19 pandemic.
In making tough decisions in the context of GCs and crises, governments must continuously strike a balance between scientific evidence and political and societal concerns. A multivocal approach to accountability is invoked to mediate among multiple considerations from scientific communities, politicians and citizens.
This work contributes to accountability research by adopting the concept of multivocality to characterise the adaptive and multi-dimensional nature of accountability in the context of GCs, where science and politics continuously intertwine.
1. Introduction
Grand challenges (GCs), such as the COVID-19 pandemic, represent some of the most complex crises societies face, often involving significant uncertainty, high stakes, and multifaceted impacts on health, the economy and society. These challenges are subject to a variety of interpretations and meanings (Ferraro et al., 2015) and demand coordinated responses that bring together a variety of actors, including governments, scientific experts, health organisations and citizens, all characterised by different views on the issue (George et al., 2016). In such high-pressure situations, accountability is paramount—not only as a mechanism for maintaining transparency but also as a means of ensuring that decision-making is responsive to the evolving nature of the crisis (Rinaldi et al., 2020; Leoni et al., 2022; Rinaldi, 2023). However, accountability in the context of GCs is a complicated and dynamic concept (Andreaus et al., 2021; Columbano et al., 2024). It cannot be reduced to a straightforward assessment of actions against fixed benchmarks (Contrafatto et al., 2024); instead, it must account for a range of diverse and sometimes conflicting considerations, including scientific data, political imperatives and societal concerns. In other words, accountability should be “multivocal” (Ferraro et al., 2015), allowing diverse perspectives to tackle the challenge and making sense of decisions. Multivocal accountability acknowledges that different stakeholders—such as scientists, citizens, professionals, etc—may have different criteria and expectations and that whoever is in charge of responding to CGs must accommodate these diverse perspectives.
This paper explores the multivocal nature of accountability in the specific context of the Italian government’s response to the COVID-19 pandemic throughout 15 months by examining how diverse perspectives from science and politics emerged and converged in shaping the government’s accountability. Unlike typical policy areas, where decisions might be based on relatively stable evidence or predictable conditions, the pandemic created an environment where rapid, often unprecedented decisions with profound and wide-reaching consequences were required. Drawing on Bruno Latour’s theoretical concepts of purification and mediation (Latour, 1993, 2004), this study investigates how the Italian government navigated the tensions between science and politics while discharging accountability. Purification, in this case, refers to efforts to present decisions as based purely on scientific facts or only on political concerns, while mediation involves combining both scientific and political considerations (Latour, 1993). By thematically analysing official public documents and press releases, the study explores how these processes of purification and mediation shaped the justification of decisions at key moments during the pandemic.
The findings of this study highlight that the Italian government’s accountability was not a static or singular response; rather, it evolved over time, with different justifications being employed as the crisis progressed. In some instances, accountability was framed through scientific data, such as infection rates, death tolls and hospital capacity, which were used to justify the imposition of tough restrictions. At other times, political considerations were brought to the forefront, particularly as the economic and social consequences of the pandemic became more pronounced. This selective framing of accountability, where different sets of considerations were foregrounded depending on the context, reflects a multivocal approach to decision-making. This multivocal approach is here interpreted through Latour’s view of science and politics, their apparently different ontologies and the processes that shape such ontologies. The multivocal nature of accountability in GCs is found to be constructed through recurring purification and mediation processes that are leveraged to reconcile multiple – but equally valid – perspectives, based on scientific and/or political concerns raised during the GC response.
The concept of multivocal accountability adopted in this study contributes to the growing body of research on accountability in (public) governance (Columbano et al., 2024; Andreaus et al., 2021), offering a nuanced understanding of how accountability can be adaptive and multi-dimensional in the context of crises. Further, by grounding the concept of multivocality in Latour’s thought, this work shows the relevance of Latour’s concepts to extend our understanding of accountability (Yates et al., 2025). The paper shows that accountability processes in GCs are not linear but can shift over time, depending on the evolving circumstances and the range of actors involved. Importantly, these shifts can be legitimate if they contribute to transparency, build trust and enhance the overall legitimacy of decision-making. The practical implications of this research are significant for governments and decision-makers faced with complex challenges. It demonstrates the importance of striking a delicate balance between scientific evidence, political decisions and societal needs. As the pandemic has shown, governments must continuously navigate these complex considerations, ensuring that their accountability mechanisms reflect the dynamic, multivocal nature of their responses, which is essential for fostering public trust and ensuring effective crisis governance. In other words, accountability should acknowledge the intertwining of various perspectives.
The remainder of the article is structured as follows. The next section introduces the theme of GCs and highlights the role of multivocal accountability in such contexts. The third section draws on Bruno Latour’s concepts of purification, mediation and variable ontology (Latour, 1993, 2004) and provides a theoretical ground for multivocal accountability. The fourth section positions our study within the accountability literature on the COVID-19 challenge, while the fifth section presents the whole research design. The main findings are presented in the sixth section. Finally, the last section discusses the results considering the theoretical frame and prior studies and offers conclusions, highlighting the study’s contributions and suggesting directions for future research.
2. Grand challenges and multivocal accountability
The term “grand challenges” (GCs) increasingly describes major global problems arising from—and sometimes exacerbated by—technological, social and economic progress. GCs are often viewed as wicked problems: complex, difficult to solve and resistant to conventional methods (Wexler, 2009). Growing concern has spurred scholarly attention across disciplines, aiming to raise awareness, assess implications and explore solutions. Examples include climate change, poverty, natural disasters, inequality and modern slavery (Brammer et al., 2019). Definitions vary, but Eisenhardt et al. (2016, p. 1113) describe GCs as “highly significant yet potentially solvable problems … affect[ing] vast numbers … with complex, intertwined technical and social elements.” Like wicked problems, GCs lack clear definitions and involve conflicting stakeholders, ambiguous data and unpredictable outcomes (Churchman, 1967; Wexler, 2009). In a globalised world, their effects transcend communities, threatening human welfare (Howard-Grenville, 2021). Ferraro et al. (2015) further characterise GCs by complexity, uncertainty and evaluativeness. In detail:
First, grand challenges are complex, entailing many interactions and associations, emergent understandings, and nonlinear dynamics. Second, grand challenges confront organisations with radical uncertainty, by which we mean that actors cannot define the possible future states of the world and therefore cannot forecast the consequences of their present actions, or whether future others will appreciate them. And third, grand challenges are evaluative, cutting across jurisdictional boundaries, implicating multiple criteria of worth, and revealing new concerns even as they are being tackled. Taken together, these three facets pose formidable organisational challenges (Ferraro et al., 2015, p. 36 – emphasis added).
The complexity, uncertainty and evaluative nature of GCs shape how they are addressed. Their complexity can obscure a full understanding of the system, leading to varied interpretations (Ferraro et al., 2015). GCs also evolve unpredictably, often creating new, unforeseen issues. Uncertainty limits decision-makers’ ability to foresee outcomes, making traditional models ineffective. As evaluative problems, GCs lack one “right” solution—multiple, equally valid responses may exist. These traits also complicate identifying who should tackle GCs. Research has examined responses at various levels—businesses, governments and NGOs—through different lenses. In accounting, wicked problems and GCs have mostly been studied in public sector contexts (Jacobs and Cuganesan, 2014; Thomasson et al., 2020), highlighting the complex role of governments. Public sector tools like budgets and financial systems have been key in managing these problems and ensuring accountability.
However, scholars argue that traditional accounting and accountability models may fall short (Jacobs and Cuganesan, 2014; Thomasson et al., 2020). There is a growing call for research into how accounting can better support GC responses—especially in areas like measurement, control systems and accountability frameworks (Pimentel et al., 2023). When considering accountability in the context of GCs, it is crucial to recognise that those responsible for addressing GCs “operate at the intersection of conflicting demands” (Ferraro et al., 2015, p. 368), arising from the various stakeholders affected by the issue, each with different perspectives on potential solutions. The inherently conflicting nature of wicked problems such as GCs means that no single solution exists; rather, solutions are negotiable and subject to multiple interpretations (Thomasson et al., 2020). The treatment of a GC and its outcomes must accommodate diverse interpretations due to the differing perspectives of those affected (Ferraro et al., 2015). Addressing and incorporating these varying viewpoints is associated with multivocality, defined as “discursive and material activity that sustains different interpretations among various audiences with different evaluative criteria” (Ferraro et al., 2015, p. 375). Thus, multivocality describes the activities involved in responding to a GC, which “can be interpreted coherently from multiple perspectives simultaneously” (Padgett and Ansell, 1993, p. 1263). While those responsible for tackling GCs are expected to provide an account of their actions, justify their decisions and demonstrate accountability for their attempted solutions, such accounts are inevitably shaped by multivocality, that is, the multiple interpretations that characterise GCs and their solutions.
Accountability is also shaped by the complex nature of GCs. The concept of multivocal accountability—emphasising multiple voices, expectations and meanings—offers a useful lens for understanding how accountability is negotiated during crises (Ferraro et al., 2015). Studying how actors account for their responses to GCs and the multivocality involved supports the call to position accounting scholars more centrally in GC and wicked problem debates (Thomasson et al., 2020; Jacobs and Cuganesan, 2014; Pimentel et al., 2023). Further, this may contribute to reducing the lag ascribed to accounting scholarship in engaging in GCs, providing the accounting and accountability perspective in a field where management and organisational research have already produced relevant outputs. Indeed, much of the existing management and organisational literature focuses on “actions” in response to GCs, exploring how different organisations have (un)successfully addressed these challenges or proposing methodological approaches to tackling them (Ferraro et al., 2015; George et al., 2016). However, by exploring GCs from an accounting or accountability perspective, further insights can be gained into managing these challenges through accounting mechanisms, as well as the “discursive activity” (Ferraro et al., 2015, p. 375) associated with accountability to justify the “actions” of those in charge of tackling GCs, whether they be governmental, non-governmental or business organisations.
3. A Latourian perspective on multivocal accountability
The concept of multivocal accountability derived from management studies (Ferraro et al., 2015) offers a valuable lens for advancing accountability research by foregrounding the multiplicity of voices, rationalities and expectations that must be mediated in times of crisis. We rely on Bruno Latour’s concepts of purification and mediation (Latour, 1993, 2004) to theoretically frame how multivocal accountability encapsulates scientific and political considerations in the context of GCs.
Latour (1993, 2004) argues that scientists, moralists, politicians and other decision-makers deceive themselves by representing complex phenomena through contrasting dualisms—such as science and politics, body and mind, human and nature, or objective and subjective. For instance, from a “scientific” perspective, a complex issue may be framed in terms of “pure facts,” with related decisions shaped by “(apparently pure) science producing facts with reason and without politics” (Christensen and Skærbæk, 2010, p. 527). Conversely, approaching a complex issue from a political standpoint involves interpreting reality as a process of social construction (Latour and Woolgar, 1979), where the origins of facts emerge from intricate processes of negotiation. Politicians, in turn, “agree to seek their legitimacy very far from the scene of the facts, in another land, that of the universal or formal foundations of ethics” (Latour, 2004, p. 98).
Latour claims that modernists seek to keep these forms of reality representation as separate as possible, employing purification strategies to uphold such dualisms (Latour, 1993). This leads to the portrayal of scientific discoveries as “purified” of their deeper social motivations, as if scientific facts were produced without human intervention. Attempts of purification have been detected in various cases where accounting was used to justify decisions. For instance, Young (2014) found that to reject the accusation of a politicised decision regarding the accounting for stock compensation, the Financial Accounting Standards Board (FASB) attempted to demonstrate that the technical accounting process was separate from politics. Young (2014) suggests that instead of neglecting the political nature of accounting regulation, the due process should have changed to include more constituents’ instances to mediate between politics and science rather than separate the two. Skærbæk and Christensen (2015) also consider the process of purification within the context of audit by analysing the case of an accounting system failure and the actions undertaken to “purify” the malfunctioning system, such as the dismissal of a police commissioner found responsible for such malfunctioning. Purification processes are also studied in public healthcare systems, where they are found to support the “accountingisation” of healthcare activities and patients (Sjögren and Fernler, 2019).
However, as a practice theorist, Latour rejects the notion of dualisms in favour of dualities, which acknowledge the mutually constitutive relationship between science and politics, body and mind, human and nature, and so forth. He critiques the idea that science generates objective knowledge while politics deals with subjective matters, instead asserting that scientific and political processes are inherently intertwined. Latour warns against the illusion of maintaining a strict separation between science and politics, arguing that purification inevitably leads to processes of mediation, in which supposedly distinct domains become entangled, forming “hybrids”—that is, “new types of beings, hybrids of nature and culture” (Latour, 1993, p. 10). Within this complex arrangement, as a postmodern perspective suggests, science and politics—as well as facts and values—are inextricably linked.
Latour argues that the processes of purification and mediation—though seemingly distinct and irreconcilable, much like political and scientific perspectives—are destined to coexist and, in fact, depend on each other to advance the understanding, representation and governance of complexity (Latour, 1993). The ability to connect science with politics relies on mediation processes that transport, redefine, reposition and even distort the original content of seemingly separate political and scientific explanations, ultimately enabling a joint interpretation (Latour, 1993). It is through the integration of the essence of both science and politics that discourse becomes truly meaningful. Mediation enables different degrees of proximity to nature and society, interpreting natural facts and societal values along a spectrum and exhibiting varying levels of stability—what Latour (1993, p. 85) refers to as a “variable ontology”. As mediation and purification processes are temporal and never assured, evolve over time, and allow the interpretation and re-interpretation of scientific and political explanations, they can be associated with multivocal activities, as they support coherent interpretation of reality from multiple perspectives (Padgett and Ansell, 1993).
Latour’s concepts of variable ontology, purification and mediation between science and politics (Latour, 1987, 2004) align with the multivocal approach that is required of those in charge of a GC (Ferraro et al., 2015). Those in charge, whether in business or governmental organisations, do not merely apply science as an independent, objective force; rather, they negotiate its integration into policy decisions, which are deeply intertwined with societal and political values. Extant literature on GCs shows that those in charge are faced with conflicting demands and multiple interpretations of the possible responses, thereby needing to continuously balance between the scientific aspects of the response and their political role, that is, between tackling the issue and maintaining their political legitimacy. The intersection of science and politics in GCs introduces tensions, as scientific recommendations may conflict with political feasibility or public acceptability (Jasanoff, 2013; Lai et al., 2019). It follows that we should accept that the accountability discharged by those in charge of the issue—i.e. governments—may have a nonlinear process and multiple meanings, in line with the nature of the challenge for which an account must be provided. If governments must continuously negotiate and strike a balance among multiple interpretations of a GC, this will inevitably affect their accountability. Multivocality recognises the presence of multiple interpretations—all valid—of the challenge and the need to balance the technical rigour of scientific recommendations with the contextual and social realities of political decision-making. Latour’s perspective helps to explain how, during a GC response, accountability can embrace multivocality, thereby enabling narratives with different interpretations of the challenge, as the scientific discourse cannot exist in isolation but shapes and is shaped by political imperatives. Indeed, GCs are inherently complex, uncertain, and evaluative issues that demand new forms of engagement and collaboration among diverse actors (Gehman et al., 2022), including governments, corporations, citizens, scientists and NGOs, as well as non-human entities such as forests, oceans, lakes and cities (Latour, 2017). Providing effective responses to GCs calls for the co-operation of a broad range of actors beyond business firms and governmental organisations. However, these multi-actor initiatives often present difficulties in collective sensemaking, particularly in balancing the need to establish common ground among diverse stakeholders while preserving the multivocality of their perspectives (Ferraro et al., 2015).
The following section shows what previous studies have already illustrated about the accountability practices implemented by national governments in the context of the GC constituted by the COVID-19 pandemic.
4. Accountability for the COVID-19 challenge: prior studies
Although COVID-19 may seem a temporary issue that has been apparently solved, it has been associated with a GC (Howard-Grenville, 2021). Indeed, following from the various definitions of GCs, the COVID-19 pandemic can be registered as a GC and an urgent and unprecedented one (Howard-Grenville, 2021). The pandemic developed in a nonlinear way (“complex”), its consequences were often unexpected (“uncertain”) and could be tackled with different approaches and according to multiple interpretations (“evaluative”) (Ferraro et al., 2015, p. 36). Further, the emergency evolved quickly at a global scale, affecting vast numbers of individuals (Eisenhardt et al., 2016) and having multiple and intertwined implications at economic, social, institutional and health levels. While before the COVID-19 crisis, examples of GCs were usually associated with climate change, poverty, global hunger, natural disasters, inequalities and diseases (George et al., 2016), the pandemic is now included in the list. First, scientists warned about this issue for decades; second, it developed and reached a global scale, affecting a vast number of people and generating an emergency that required an immediate governmental response.
While the accounting research community has lagged in engaging in the debate about GCs, when faced with the outbreak of COVID-19, it responded promptly to the number of issues raised by the pandemic (Leoni et al., 2021), turning its focus to the role of accounting and accountability in the crisis, with the view to understanding impacts and learning lessons for the future (Parker and Troshani, 2022). The effort of the accounting community has generated a massive research output on the accounting, accountability, governance and control dimensions of how COVID-19 has impacted society, governments, businesses and the profession (Leoni et al., 2022; Carnegie et al., 2023; Rinaldi, 2023; Leoni, 2024). Among the numerous works, the studies on the accounting and accountability dimension during or in response to the crisis are relevant to the objectives of the present investigation. These studies are grounded on the concept of accountability as entailing a “relationship in which people are required to explain and take responsibility for their actions” (Sinclair, 1995, pp. 220–221), which “is subjectively constructed and changes with context” (Sinclair, 1995, p. 219). Thus, the study of accountability in major crises provides insights on the forms, modes and contents of accountability in these specific contexts that coherently can be extended to GCs and wicked problems.
Facing the COVID-19 challenge, governments were the main actors expected to react promptly and sustain public health, as well as social and economic activities, with an interventionist approach that was long abandoned in public sector policies (Parker and Troshani, 2022; Ahrens and Ferry, 2021). They were required to tackle the major issue and, as previous literature on governance accountability has shown, were also expected to provide an account of their decisions and actions (Lai et al., 2014; Sargiacomo, 2015). Further, within the specific COVID-19 literature, studies showed that governmental leaders responded to the crisis by relying on accounting to manage, understand and control the pandemic (Leoni et al., 2021; Parisi and Bekier, 2022; Sargiacomo et al., 2021; Huber et al., 2021; Ahmad et al., 2021) and by discharging accountability to account for and justify their decisions and strategies against the emergency (Ahrens and Ferry, 2021; Andreaus et al., 2021; Leoni et al., 2021; De Villiers and Molinari, 2022; Columbano et al., 2024; Bigoni and Occhipinti, 2024; Ferry et al., 2024).
Studies on public accountability in times of crises have shown that governmental institutions or non-governmental organisations in charge of tackling the crisis are expected to discharge accountability (Lai et al., 2014; Taylor et al., 2014), but they are found to often fail in taking actual responsibility (Baker, 2014; Perkiss and Moerman, 2018) due to the disruption crises can cause to the neutrality and objectivity required for accountability (Gallhofer and Haslam, 1991). In these extraordinary contexts, those in charge of tackling a challenge often rely on accounting to support a transparent and objective accountability through data (Sargiacomo et al., 2021; Contrafatto et al., 2024; Mazzola et al., 2024). However, in these extreme circumstances, research has shown that what and how is accounted may vary from the traditional forms of financial accounting, as we know it. Indeed, during the COVID-19 pandemic, different objects were accounted for to sustain the accountability on the tackling of the pandemic, such as extraordinary budgetary and financial data at the state level (Ahrens and Ferry, 2021), health data on disease, deaths and contagion trends (Ahmad et al., 2021; Andreaus et al., 2021; Antonelli et al., 2022; Bigoni and Occhipinti, 2024; Columbano et al., 2024), and finally social data regarding citizens’ activities, engagement and social media habits (Ahn and Wickramasinghe, 2021; Antonelli et al., 2022; La Torre et al., 2022).
Due to the extraordinary situation and the different data that sustained it, the accountability discharged by governmental institutions was found to take new or different styles, with the main goal to shift responsibilities. In Italy, for instance, the government was found to attempt to shift the responsibility of contrasting the pandemic from itself to the citizens throughout the initial phases of the pandemic (Andreaus et al., 2021). In a similar vein and context, the Italian government was found to discharge three different configurations of accountability (paternalistic, political and communal) to again shift responsibility from the pandemic itself to the government and, finally, to the citizens (Columbano et al., 2024). Other studies have highlighted how the use of social data, that are individuals’ activities, such as movements, habits or social interaction, to manage and provide an account of the crisis may threaten the freedom of individuals (Ahn and Wickramasinghe, 2021; Antonelli et al., 2022; Ferry et al., 2024), thereby undermining the beneficial role usually attributed to accountability.
While accountability is deemed crucial for governments to pursue crisis resolution goals in challenging times (Lai et al., 2014; Sargiacomo and Walker, 2022), previous studies on COVID-19 demonstrate that the accountability discharged in times of crisis is far from ideal; rather, it serves the governments’ need to shift responsibility (Andreaus et al., 2021; Columbano et al., 2024). Prior accounting research on the COVID-19 pandemic has unveiled the contingent styles (Andreaus et al., 2021), configurations (Columbano et al., 2024) and accounts (Contrafatto et al., 2024) through which accountability is discharged. These works focus on understanding the responsibility shifts underlying accountability, but more investigation is needed to understand how those in charge solve (or not) the tension between technical/scientific interpretations and the political/social readings (Latour, 1993) of the issue in their accountability. In other words, more research is needed on the multivocal nature of accountability, recognising that we should not expect a static or univocal form of accountability in a context inherently characterised by complexity, uncertainty and evaluative judgement.
While other accounting scholars have explored the role of different accounts in addressing governmental response to extraordinary events (Contrafatto et al., 2024), we are not aware of prior studies that try to unveil how accountability discourses “mediate” science and politics worldviews according to Latour’s interpretation of these apparently distinct ontologies. More specifically, accountability has not been explored for its capability to “mediate” scientific and political arguments in representing complex phenomena like GCs. Studying how the Italian government drew on science and politics to represent its response to the COVID-19 pandemic may contribute to extending our knowledge on accountability by highlighting the multivocal nature of accountability in GCs, including pandemics. Furthermore, while previous works tend to focus on the immediate moments after the virus outbreak and the role of accountability in the turbulent immediate response to the emergency (e.g. Broadbent, 2020; Demirag et al., 2020; Contrafatto et al., 2024), no work has yet considered a longer time span, where COVID-19 is configured as a GC, evolving unpredictably in time and generating new and unexpected issues. Indeed, the study of accountability throughout a longer period, not limited to the most frenetic moments, may help unpack the nature of accountability discharge during the response to a complex, uncertain and evaluative issue.
As a result, the present work aims to explore the multivocal nature of accountability in the context of GCs by examining how different perspectives from science to politics emerged, coexisted and shaped the construction of the Italian government’s accountability. To achieve this, the research undertakes an exploratory analysis of the Italian government’s response to the COVID-19 crisis, focusing on the nature of the accountability discharged during the emergency, which was shaped by both scientific data and political considerations. The next section introduces this research context and outlines the research design, highlighting the methods adopted to address the aim of the study.
5. Research design
5.1 Context of the investigation
Drawing on Bruno Latour’s concepts of science and politics, as well as purification and mediation, the study employs thematic analysis to investigate how scientific and political considerations emerged and evolved throughout the Italian government’s long-term response to the crisis. The first case of COVID-19 in Italy was recorded on 21 February 2020, and a state of national emergency was declared in early March 2020. However, it was not until April 30, 2020, that the Italian government officially introduced a structured system to monitor the evolution of the pandemic that was named the key emergency indicators (KEIs) system and constructed through a coordinated effort between scientific experts and policymakers. This study examines government accountability from 30 April 2020 to 31 July 2021, when the most critical phase of the challenge was considered over. By focusing on an extended period and excluding the immediate, traumatic phase following the outbreak, this research concentrates on the response to a large-scale challenge, as COVID-19 later came to be understood, rather than on the urgent and turbulent actions taken at the onset of the emergency (Andreaus et al., 2021; Columbano et al., 2024). Analysing the pandemic over a longer timeframe allows it to be framed more clearly as a GC rather than a short-term crisis.
Indeed, during the 15 months under investigation, four pivotal moments are identified when the Italian government made key decisions in response to the crisis. Each moment marks the beginning of a different phase in managing the challenge (Table 1) and was legally supported by the issuance of one or more urgent government decrees (DPCM) [1]. The phases are outlined below.
Timeline of the decrees issued by the President of Council of Ministers (Presidente del Consiglio dei Ministri) and by the President of the Republic (Presidente della Repubblica) and related decisions throughout the four phases under investigation
| Phase | Dates | Decrees of reference | Decision |
|---|---|---|---|
| Phase 1 | 30 Apr 2020 / 11 Oct 2020 | Initial lift of Covid-19 restrictions and implementation of the monitoring system based on KEIs | |
| Phase 2 | 12 Oct 2020 / 1 Dec 2020 | Second wave of Covid-19 – initial re-establishment of restrictions | |
| Phase 3 | 4 Dec 2020 / 25 Apr 2021 | Strengthened restrictions for the holidays | |
| Phase 4 | 26 Apr 2021 / 31 Jul 2021 | Reopenings and progressive lift of restrictions |
In Phase 1 (30 April 2020–11 October 2020), the government decided to lift some COVID-19 restrictions in response to fewer cases and reduced pressure on healthcare services and thanks to the introduction of the monitoring system (Presidente del Consiglio dei Ministri, 2020a). Phase 2 (12 October–1 December 2020) is characterised by the reintroduction of COVID restrictions, established through three different government decrees (Presidente del Consiglio dei Ministri, 2020b, 2020c, 2020d), with varying levels of severity in different regions based on the data from the KEIs. Phase 3 (2 December 2020–25 April 2021) is marked by even stricter restrictions due to the higher risk of infections during both the Christmas and Easter holidays (Presidente del Consiglio dei Ministri, 2020e; Presidente della Repubblica, 2021a). Finally, Phase 4 (26 April 2021–31 July 2021) signifies the government’s decision to begin progressively lifting all restrictions (Presidente della Repubblica, 2021b), thanks to the reduced number of cases and the increasing vaccination rate, while continuing to monitor the situation through the KEIs system.
This study begins with the introduction of the KEIs system on 30 April 2020 by the Italian government, following a coordinated effort between scientific experts and policymakers responsible for the COVID-19 response. After the stringent lockdown measures implemented in the initial turbulent phase of the pandemic—during which the government adopted various accountability styles and shifted responsibility (Andreaus et al., 2021; Columbano et al., 2024)—the government introduced the KEIs system. This system aimed to reduce uncertainty by monitoring the progression of the pandemic and informing decisions based on scientific and technical data.
The KEIs generated weekly indicators derived from scientific and healthcare system data, which were compiled into weekly reports and used to guide decision-making in response to the emergency. This research examines reports from 30 April 2020 to 31 July 2021 (Table 2) [2], with a particular focus on those published around the pivotal moments of the government’s response to the crisis (Table 1).
List and dates of KEIs monitoring reports (in italic those that were consulted for the triangulation with the thematic analysis) released by the Ministry of Health (Ministero della Salute)
| Report name | Month, days | Year | Rt index | Incidencea | Reference |
|---|---|---|---|---|---|
| Monitoraggiob 0 | May, 4–10 | 2020 | 0.55 | 16 | Ministero della Salute (2020a) |
| Monitoraggio 1 | May, 11–17 | 2020 | 0.52 | 17 | Ministero della Salute (2020b) |
| Monitoraggio 2 | May, 18–24 | 2020 | |||
| Monitoraggio 3 | May, 25–31 | 2020 | |||
| Monitoraggio 4 | June, 1–7 | 2020 | |||
| Monitoraggio 5 | June, 8–14 | 2020 | |||
| Monitoraggio 6 | June, 15–21 | 2020 | |||
| Monitoraggio 7 | June, 22–28 | 2020 | |||
| Monitoraggio 8 | June, 29–July, 5 | 2020 | |||
| Monitoraggio 9 | July, 6–12 | 2020 | |||
| Monitoraggio 10 | July, 13–19 | 2020 | |||
| Monitoraggio 11 | July, 20–26 | 2020 | |||
| Monitoraggio 12 | July, 27–August, 2 | 2020 | |||
| Monitoraggio 13 | August, 3–9 | 2020 | |||
| Monitoraggio 14 | August, 10–16 | 2020 | |||
| Monitoraggio 15 | August, 17–23 | 2020 | |||
| Monitoraggio 16 | August, 24–30 | 2020 | |||
| Monitoraggio 17 | August, 31–September, 6 | 2020 | |||
| Monitoraggio 18 | September, 7–13 | 2020 | |||
| Monitoraggio 19 | September, 14–20 | 2020 | |||
| Monitoraggio 20 | September, 21–27 | 2020 | |||
| Monitoraggio 21 | September, 28–October, 4 | 2020 | |||
| Monitoraggio 22 | October, 5–11 | 2020 | 1.17 | 75 | Ministero della Salute (2020c) |
| Monitoraggio 23 | October, 12–18 | 2020 | 1.34 | 210 | Ministero della Salute (2020d) |
| Monitoraggio 24 | October, 19–25 | 2020 | 1.60 | 280 | Ministero della Salute (2020e) |
| Monitoraggio 25 | October, 26–November, 1 | 2020 | 1.72 | 451 | Ministero della Salute (2020f) |
| Monitoraggio 26 | November, 2–8 | 2020 | 1.43 | 648 | Ministero della Salute (2020g) |
| Monitoraggio 27 | November, 9–15 | 2020 | |||
| Monitoraggio 28 | November, 16–22 | 2020 | |||
| Monitoraggio 29 | November, 23–29 | 2020 | 0.91 | 591 | Ministero della Salute (2020h) |
| Monitoraggio 30 | November, 30–December, 6 | 2020 | 0.82 | 455 | Ministero della Salute (2020i) |
| Monitoraggio 31 | December, 7–13 | 2020 | |||
| Monitoraggio 32 | December, 14–20 | 2020 | 0.90 | 329 | Ministero della Salute (2020j) |
| Monitoraggio 33 | December, 21–27 | 2020 | 0.90 | 305 | Ministero della Salute (2020k) |
| Monitoraggio 34 | December, 28–January, 3 | 2021 | |||
| Monitoraggio 35 | January, 4–10 | 2021 | 1.09 | 368 | Ministero della Salute (2021a) |
| Monitoraggio 36 | January, 11–17 | 2021 | 0.97 | 339 | Ministero della Salute (2021b) |
| Monitoraggio 37 | January, 18–24 | 2021 | |||
| Monitoraggio 38 | January, 25–31 | 2021 | |||
| Monitoraggio 39 | February, 1–7 | 2021 | |||
| Monitoraggio 40 | February, 8–14 | 2021 | |||
| Monitoraggio 41 | February, 15–21 | 2021 | |||
| Monitoraggio 42 | February, 22–28 | 2021 | |||
| Monitoraggio 43 | March, 1–7 | 2021 | |||
| Monitoraggio 44 | March, 8–14 | 2021 | 1.16 | 250 | Ministero della Salute (2021c) |
| Monitoraggio 45 | March, 15–21 | 2021 | 1.08 | 240 | Ministero della Salute (2021d) |
| Monitoraggio 46 | March, 22–28 | 2021 | |||
| Monitoraggio 47 | March, 29–April, 4 | 2021 | |||
| Monitoraggio 48 | April, 5–11 | 2021 | 0.85 | 160 | Ministero della Salute (2021e) |
| Monitoraggio 49 | April, 12–18 | 2021 | 0.81 | 157 | Ministero della Salute (2021f) |
| Monitoraggio 50 | April, 19–25 | 2021 | |||
| Monitoraggio 51 | April, 26–May, 2 | 2021 | |||
| Monitoraggio 52 | May, 3–9 | 2021 | |||
| Monitoraggio 53 | May, 10–16 | 2021 | |||
| Monitoraggio 54 | May, 17–23 | 2021 | |||
| Monitoraggio 55 | May, 24–30 | 2021 | |||
| Monitoraggio 56 | May, 31–June, 6 | 2021 | |||
| Monitoraggio 57 | June, 7–13 | 2021 | |||
| Monitoraggio 58 | June, 14–20 | 2021 | |||
| Monitoraggio 59 | June, 21–27 | 2021 | |||
| Monitoraggio 60 | June, 28–July 4 | 2021 | |||
| Monitoraggio 61 | July, 5–11 | 2021 | |||
| Monitoraggio 62 | July, 12–18 | 2021 | |||
| Monitoraggio 63 | July, 19–25 | 2021 | |||
| Monitoraggio 64 | July, 26–August, 1 | 2021 |
Incidence is calculated as the weekly number of infections in 100,000 inhabitants
Translation: monitoring
5.2 Research method
To achieve its aim, the research conducts a thematic analysis of the government’s accountability, which was presented through various official press conferences held by the Prime Minister and other government representatives to explain the various decisions made by the government. A total of 19 press conferences and official discourses that took place around the pivotal moments that mark the end and the start of the four phases are analysed via thematic analysis throughout the 15 months of the challenge response (Table 3). The videos and related transcriptions are publicly available on the Italian government’s official channel on YouTube (https://www.youtube.com/@palazzochigi). The official transcripts were used to perform the thematic analysis.
List of transcribed videos released by the Presidency of the Council of Ministers (Presidenza del Consiglio dei Ministri) analysed with thematic analysis (available at: https://www.youtube.com/palazzochigi)
Thematic analysis is “a method for identifying, analysing and reporting patterns (themes) within data” in qualitative research (Braun and Clarke, 2006, p. 79). There are two possible approaches: inductive and deductive. In this work, we adopted the deductive approach, which entails a theory-driven thematic analysis to deliberately explore the data within the bounds of a theoretical framework (Braun and Clarke, 2006, p. 79). Consequently, Latour’s view on politics and science and his reflections on purification and mediation strategies inform the deductive thematic analysis (Latour, 1993, 1996, 2004). This approach allows for a structured investigation using Latour’s theoretical concepts to interpret the data, thereby providing a deeper understanding of the government’s accountability during the COVID-19 crisis through these established ideas. By applying a deductive approach, we ensure that the analysis is rooted in Latour’s framework, offering insights into how scientific and political considerations were integrated into government accountability.
The thematic analysis involves six steps: (1) familiarisation with the data; (2) generating initial codes; (3) searching for themes; (4) reviewing themes; (5) defining themes; (6) writing-up (Braun and Clarke, 2006; Maguire and Delahunt, 2017; Naeem et al., 2023). First, two of the authors independently familiarised themselves with the texts (1) and the relevant theoretical framework. Second, guided by Latour’s theoretical concepts, they manually generated initial codes (2) in the texts (Braun and Clarke, 2006). This step entailed identifying segments of texts about the government’s justification that were relevant to Latour’s concepts of science and politics. References to scientific reasons and data (e.g. the Rt index) were coded as scientific justification (“science”), whereas references to social and economic needs and concerns were labelled as “politics”. Third, following the coding, the researchers independently looked for patterns in the data that captured Latour’s strategies of purification, or processes of mediation, and identified preliminary themes (3). As outlined above, “purification” refers to efforts to present decisions as based solely on scientific evidence (or, alternatively, exclusively on political considerations), whereas “mediation” denotes instances in which both scientific and political factors were jointly invoked by the Italian government to justify its decisions. Themes were also reviewed (4) in light of the weekly reports from the KEI system. Discussions between the two researchers were undertaken to resolve any disagreements (Parker and Roffey, 1997) and define themes (5). Through thematic analysis of official public documents and press releases, this study examines how these processes of purification and mediation shaped the justification of decisions at key moments during the pandemic, thereby supporting a multivocal form of accountability. The identified themes were also reviewed. The main themes that were identified are presented in the following findings section (6) and are related to the relevance of the scientific KEIs system, the purification attempt, the mediation between science and politics, and the multivocality of the government’s justification.
6. Findings
This section presents the findings, based on the thematic analysis. First, the system of scientific and healthcare data included in the KEIs system is described to identify the data that constitutes the scientific reference for the challenge response. Then, the results from the thematic analysis of the press conferences related to the four phases, along with the triangulation with reports from the KEIs system, are presented. The findings indicate that the government initially pursued a purification strategy (Latour, 1993) by referring solely to scientific data to justify its decisions. However, in the subsequent phases, the government openly incorporated political concerns (i.e. social and economic), thereby engaging in mediation processes to balance science and politics in the challenge response. This shows that the government’s accountability reflects the multiple interpretations of the challenge that emerged throughout the phases, in line with the need for a multivocal approach to GCs (Ferraro et al., 2015).
6.1 The construction of key emergency indicators system
After the first turbulent months of crisis management that followed the COVID-19 outbreak (February 21, 2020) and the declaration of the national emergency in Italy (March 9, 2020), the government officially introduced the KEIs monitoring system on April 30, 2020. This system was said to “keep the epidemic curve under control and to allow [the government] to intervene when necessary” (Presidenza del Consiglio dei Ministri, 2020a), as well as to publicly report on the evolution of the pandemic. The system entailed 17 KEIs calculated on a weekly and regional level. Information was gathered from healthcare regional institutions and collated to depict the situation at national and regional levels. The weekly KEIs were complemented with a report by the Ministry of National Health commenting on the most important indicators. The 17 KEIs were grouped into three dimensions (Table 4), as follows: (1) monitoring system capability; (2) transmissibility and impact; (3) resilience of healthcare services. The first dimension regarded the data completeness and intends to assess the monitoring ability of the pandemic data collection system. The second dimension comprised both the virus transmissibility and the hospital logistic situation using two sub-groups of indicators: (a) virus transmissibility and (b) hospital bed occupancy. Finally, the third dimension measures the resilience of the healthcare services and includes two subgroups: (a) timely testing capacity and (b) contact-tracing coverage.
Description of the KEIs by dimension and their related thresholds/targets and alerts
| KEI description | Threshold | Alert | |
|---|---|---|---|
| Dimension 1: monitoring system capability | |||
| 1 | Notified symptomatic COVID cases with details on the start of the symptoms per month/total notified symptomatic cases in the same period | ≧ 60% and increasing | <60% |
| 2 | Notified hospitalisation cases with details on hospitalisation dates per month/total notified hospitalisation cases in the same period | ||
| 3 | Notified intensive care hospitalisation cases with details on hospitalisation dates per month/total notified ICU hospitalisation cases in the same period | ||
| 4 | Notified cases with details on residential address/total notified cases of the period | ||
| Dimension 2: transmissibility and impact | |||
| 2a: virus transmissibility | |||
| 5 | Declared number of cases in the last 14 days | Decreasing trend or stable | Increasing within the last 5 days |
| 6 | Weekly average transmission index (Rt Index) | ≤1 | Rt > 1 or not calculable |
| 7 | Weekly average of the daily number of diagnosed cases | Decreasing trend or stable | Increasing within the last week |
| 8 | Weekly average number of new outbreaks (within known transmission chains) | Lack or stable number of new outbreaks | Proof of new outbreaks in the last 7 days |
| 9 | Weekly average number of new cases (outside known transmission chains) | Stable number in presence of outbreaks | Increasing number in presence of outbreaks |
| 2b: hospital beds occupancy | |||
| 10 | Weekly average of the occupancy rate of intensive care (IC) beds by COVID-19 patients | ≤30% | >30% |
| 11 | Weekly average of the occupancy rate of no IC – hospital beds by COVID-19 patients | ≤40% | >40% |
| Dimension 3: resilience of healthcare services | |||
| 3a: timely testing capacity | |||
| 12 | Weekly average positive COVID tests rate | Decreasing trend | Increasing trend |
| 13 | Average days between symptoms and diagnosis | Weekly median ≤5 days | Weekly median >5 days |
| 14 | Average days between symptoms and isolation | Weekly median ≤3 days | Weekly median >3 days |
| 3b: contact-tracing coverage | |||
| 15 | Weekly average time/person allocated to contact tracing activities | Aligned to EU recommendations | Not adequate compared to EU recommendations |
| 16 | Weekly average time/person allocated to testing and quarantine monitoring activities | ||
| 17 | Weekly average confirmed cases determined via contact-tracing/weekly average total confirmed cases | Improving trend (target 100%) | |
The indicators were the result of a coordinated effort among scientific experts (i.e. epidemiologists) and healthcare system experts (i.e. hospital managers) and were used by the Italian government to inform its decisions and to issue the corresponding decrees introducing or easing restrictions. In this phase, the government only refers to scientific data to justify its decisions:
Since the beginning [of the pandemic], the government has immediately adopted a scientific approach that entails a continuous discussion with scientific experts to provide a scientific basis to the various decisions taken (Presidenza del Consiglio dei Ministri, 2020b).
Such a strong reference to “science” suggests an initial purification strategy (Latour, 1993) by the Italian Government and an attempt to separate “pure facts” (Christensen and Skærbæk, 2010, p. 527) from political reasoning. From its introduction on April 30th, the KEIs system has produced weekly reports regarding the spreading of COVID-19 at the regional and national levels that were used by the government to control the changes in the spread of the virus and respond to it by implementing or lifting restrictions. Indeed, the government underlined how the data and information from the KEIs system would have made the pandemic visible, thereby providing a basis for the political intervention:
These whole initiatives will allow us to have a wider information dataset, to assess more efficiently the diffusion of the virus (Presidenza del Consiglio dei Ministri, 2020b).
6.2 The use of key emergency indicators and purification strategies in the government’s accountability
After the introduction of the KEIs system, particularly during Phase 1 and the first part of Phase 2, the government’s accountability appeared to be based on scientific facts only. Indeed, the government states that the challenge-response should be grounded in scientific knowledge and be led by scientific information rather than political concerns:
A government’s duty is to protect the health and lives of the citizens when facing such a lethal threat, it is not to ground its decisions on unstructured opinions, rather on the recommendations proceeding from the research of qualified scientific experts (Presidenza del Consiglio dei Ministri, 2020b).
Even in the face of a clear social interest in the reopening of schools, the Prime Minister maintained that scientific considerations must take precedence. While social concerns began to be acknowledged, their role remained secondary. For instance, in deliberations over the potential reopening of schools, scientific concerns were prioritised above all other social and political considerations.
It is very complicated to make the right decision to allow the education and the health safety to coexist (Presidenza del Consiglio dei Ministri, 2020a).
The analysis suggests that, in prioritising science and scientific facts, the government attempted to separate its decisions from any political influence, that is, any subjectivity of value judgement (Latour, 1993). In so doing, the government was employing a purification strategy to justify its harsh restrictions by means of scientific reasons (Latour, 1993). Indeed, the message that is conveyed is that scientific—and objective—considerations were prioritised in the response to the emergency, rather than any subjective considerations based on political concerns.
Later, but still during Phase 1 (Table 1), in June 2020, again the government justifies its decisions on purely scientific considerations by referencing the KEIs system:
It is a decision that we have taken with full awareness because the scientific data of the epidemic curve show that the monitoring system is working, as well as our decision to proceed with a gradual lifting of the restrictions, according to the results of our monitoring activities (Presidenza del Consiglio dei Ministri, 2020c).
The governmental decisions are said to derive from the scientific data only, purified from any other subjective considerations. Indeed, from the end of April 2020 the restrictions were eased (Phase 1), thanks to the relevant decrease of cases (KEI 5) after two months of strict lockdown measures. The KEI 5 was 12.20, which is lower than the alert threshold of 50 (Ministero della Salute, 2020a). Moreover, the transmissibility power of the virus (also known as the Rt Index – KEI 6 – Table 4) was below 1 in every Italian region, which meant that the infection was spreading very slowly in comparison to March and April 2020 (Ministero della Salute, 2020a).
This strategy of purification continued also in the first part of Phase 2 of the investigated period, when the government was forced by the (scientific) data to re-establish severe restrictions upon the citizenship (Ministero della Salute, 2020d). Again, the monitoring system was depicted as producing pure and objective facts guiding the governmental decision-making:
The management [of the pandemic] is always the same: constantly follow and monitor the curve, elaborate all the data, study them with the scientific experts, who then suggest which restrictive measures to adopt (Presidenza del Consiglio dei Ministri, 2020d).
In October and November 2020, scientific data were again presented as the sole considerations of the political decision to reintroduce the restrictions upon the population. Following the scientific data of the KEIs system that was showing a worsening of the situation, the government had to re-establish the lockdown that was announced on November 3, 2020 (Phase 2). In the days preceding the re-introduction of the lockdown, the government justified its decision by referring again solely to the scientific data as the only consideration influencing its decisions:
The data from the last weeks show that the epidemic curve is rapidly increasing with the virus spreading in the entire Country. The transmission index has reached the alert threshold of 1.5, and the number of new cases has increased, making it difficult to conduct contact tracing. There is a significant increment in the hospitalised cases, both in intensive and non-intensive care units. From all this follows the need to adopt measures that allow to mitigate at best the growing curve of the infections (Presidenza del Consiglio dei Ministri, 2020h).
Indeed, looking at the KEIs and the monitoring reports (Table 2), the number of cases had increased throughout October 2020 at a great speed, with the case incidence moving from 75 cases for every 100,000 inhabitants (Week October 5–11, 2020) to 280 for every 100,000 inhabitants two weeks later (Ministero della Salute, 2020e), up to 648 for every 100,000 inhabitants at the beginning of November (Ministero della Salute, 2020g). The weekly Rt index progressively increased from 1.17 (Ministero della Salute, 2020c), to 1.60 (Ministero della Salute, 2020e), up to 1.72 at the beginning of November (Ministero della Salute, 2020g). The pressure on the health system was also increasingly problematic, having reached and overcome the alert level both in the intensive care units (ICUs) and in other medical units (KEIs 10 and 11 from Table 4) by mid-October 2020. Such data were referred to by the Italian government when it announced new restrictions at the beginning of Phase 2. Again, the scientific facts from the KEIs system were presented as the sole consideration by the Italian government:
The last epidemic data that we have analysed from Friday (October 23, 2020) with the support of the scientific experts cannot leave us unresponsive. The analysis of the epidemic curve signals a rapid increase, with the result that the virus is spreading throughout the Country. This puts under pressure the healthcare system, which has reached concerning alert levels. The Rt index has reached the threshold of 1.5, with more than twenty thousand cases only yesterday. This makes it difficult for health agencies to provide full contact tracing. The government wants to achieve a clear aim: we want to keep the epidemic curve under control. Only this way we can oversee the pandemic without being defeated by it (Presidenza del Consiglio dei Ministri, 2020g).
In brief, the analysis of Phase 1 and most of Phase 2 shows that the Italian government initially engaged in a purification strategy (Latour, 1993) by arguing that its decisions were led by scientific considerations and not contaminated by political or subjective matters. Scientific considerations are applied to make decisions about the challenge in various moments of the two phases. However, over the course of the period investigated, political considerations increasingly came to be incorporated into governmental decisions and justifications, signalling a shift from a purely technical role to a more political one. Following a relatively calm summer with low case numbers, schools were reopened in September 2020. Despite a resurgence of cases in autumn 2020, the government adopted a clear political stance in keeping schools open, contrary to the recommendations provided by scientific experts:
The school is an untouchable point of the country system that absolutely cannot take steps backwards … even in a situation of evident health and social difficulty, we thought that the school should be a privileged asset to be protected […] School activities will continue in presence; we cannot afford that one of the main pillars of the country where the best guarantees of a better future are placed could suffer further sacrifices (Presidenza del Consiglio dei Ministri, 2020e).
The following section provides insights into the evolution of the government’s accountability, which started to combine science and politics (Latour, 1993) more visibly in the justifications of its decisions.
6.3 Mediation processes between science and politics in the government’s accountability
With the evolution of the pandemic, the government’s accountability started to include more and more political considerations when justifying COVID-19 restrictions. With time and experience of the challenge, valid and non-scientific considerations made their appearance in the justifications provided by the government. At the end of Phase 2, the data from the KEIs system depicted an improved situation and suggested that the challenge was under control. At the end of November 2020, from the monitoring reports, the challenge was under control, with an Rt-index of 0.91 (Ministero della Salute, 2020f), but there were many infected people among the population (590 in 100,000 inhabitants). With a similar Rt-index, in Phase 1, the government had decided to lift the restrictions, sustained by scientific evidence of an improved situation. However, in this instance, the government made a joint interpretation of both scientific and political considerations:
Within just a month, we have bent the infection curve, bringing it back to an Rt-index of 0.91, and the number of emergency room accesses and intensive care presence has reduced greatly. We are avoiding a general lockdown, like the one in the Spring, which would have penalised the Country at both economic and social levels. These results comfort us, but there is one aspect that we cannot underestimate that the Christmas festivities are around the corner. If we deal with the festivities by easing the restrictions, we will face an increase in the infection rate. Thereby, we are forced to introduce new measures that will strengthen the restrictions during the Christmas holiday (Presidenza del Consiglio dei Ministri, 2020j).
In contrast to earlier purification strategies, the government began to move away from a strict reliance on scientific considerations, thereby revealing the political dimensions of its decision-making. Indeed, the government justified its actions not only with reference to scientific data but also by drawing on prior experiences with challenge response and the lessons learnt. These experiences were “integrated” with scientific evidence in an effort to manage the complexity of the COVID-19 emergency and justify choices. By invoking past experience alongside economic and social concerns, the government’s approach to accountability served to mediate between politics and science, accommodating evolving interpretations of the crisis (Ferraro et al., 2015) that were rooted in distinct ontologies (Latour, 1993). In the case of the Italian government, political considerations were prioritised, even as scientific data indicated an improvement in the pandemic situation. Consequently, the form of accountability exercised during Phase 3 mediated between the political and scientific considerations emerging in response to the crisis:
[Our decision] offers a balance between the restrictions we must necessarily implement and the exemptions that remain necessary, given the social and, I would even say, ideal importance that these upcoming festivities hold within our national community (Presidenza del Consiglio dei Ministri, 2020k).
Scientific data remained valid but were “merged” with social and economic concerns, as well as with lessons learnt from previous experience. This balance among various considerations is further emphasised in the same discourse:
The pandemic is teaching us valuable lessons—namely, that we must always be ready to intervene with a method based on utmost caution, while ensuring that measures are proportionate, scientifically grounded, and communicated with transparency. This includes speaking openly to citizens and making data readily available (Presidenza del Consiglio dei Ministri, 2020k).
The accountability discharged by the government was not only sustained by scientific data, as in Phase 1 and the first part of Phase 2, but it is now shaped by political considerations (i.e. social and economic) that were presented as equally valid in constructing the challenge response. This shows the inevitable coexistence of purification and mediation processes (Latour, 1993) in interpreting the COVID challenge, constructing responses and justifying decisions.
Also, throughout Phase 4, the government continues to mediate between scientific and political interpretations. This phase was characterised by a progressive lifting of the pandemic restrictions, based on several considerations that were not limited to scientific data only. Indeed, in the month preceding the start of Phase 4, the scientific data were not as positive, as the country was still registering an Rt Index of 1.08 (Ministero della Salute, 2021d) and an incidence of 240 infections per 100,000 inhabitants. The government’s accountability still reflects the relevance of scientific considerations:
The ease of the restrictions will have to happen. I don’t have a date yet because we are currently considering it, and I don’t have a date because it depends on the trend of infections. Earlier, I was asked, “Which data?”—this is one: the trend of infections. Then, Professor Locatelli will also discuss other data, including one that hasn’t been considered so far: the progress of vaccinations among at-risk groups. That will certainly be another key factor in the decision on lifting the restrictions (Presidenza del Consiglio dei Ministri, 2021e).
Nevertheless, the government also underlines the need to mediate the scientific data with political considerations when, on April 16, 2021 it announced a lift of the restrictions from April 26, 2021 (Phase 4). The decision is presented as a risky one from the point of view of the scientific data but a necessary one from a social and economic point of view (i.e. from a political perspective).
With today’s decision, the government has taken a risk—but a calculated one. A risk based on improving data, which certainly meets the expectations of citizens, but one that rests on a crucial premise: that the measures governing behaviour in reopened activities are strictly followed. I am referring to the use of masks and adherence to social distancing. There will need to be a particular effort from the authorities—regional governments, local bodies, and law enforcement—to ensure these behaviours are observed. By doing so, this calculated risk can become an opportunity—an extraordinary opportunity not only for the economy but for our social life itself. At the same time, the vaccination campaign continues to progress well, with both positive and negative surprises along the way. This was also one of the key considerations in making these decisions (Presidenza del Consiglio dei Ministri, 2021f).
Further, the government accounts for the need to strike a balance between different instances: the scientific concerns regarding the infections and the political considerations stemming from the growing difficulties encountered by the Italian population under tough restrictions:
These decisions respond to the distress of various ranks, of many professionals, families and children. They bring more serenity in the Country and to a further extent they provide the basis for revamping the economy (Presidenza del Consiglio dei Ministri, 2021f).
In brief, when justifying its decisions in response to COVID-19, the Italian government employed both purification and mediation processes, reflecting multiple interpretations of the crisis over the 15 months under investigation. The analysis reveals that the government’s decisions were not based solely on scientific or political considerations but rather emerged through processes of mediation (Latour, 1993). Throughout the COVID-19 crisis, the scientific data from the KEIs system exhibited a “variable ontology” (Latour, 1993, p. 86), ranging from instances in which they served solely to foreground the scientific perspective—rendering the political dimension invisible—to others in which they supported a connection between science and politics. In the first two phases, data from the KEIs system were employed in a mechanistic manner to support accountability, primarily through data initially presented by the Italian government. However, in Phases 3 and 4, additional subjective considerations—including political, economic and social concerns—began to be incorporated into the government’s accountability.
The findings overall suggest how, in engaging in purification and mediation, the government’s accountability kept redefining and repositioning the considerations that sustained its decisions, at times only displaying scientific or political reasons, at other times mixing scientific data with political considerations, such as economic and social concerns, as well as lessons learnt throughout the emergency.
7. The multivocal nature of accountability in grand challenges: discussion and conclusions
This study aimed to explore the multivocal nature of accountability in the context of GCs, using the Italian government’s response to the COVID-19 pandemic as a case study. Drawing on Bruno Latour’s concepts of purification and mediation, the paper reveals how accountability in the context of GCs is sustained through multiple interpretations, shaped by a combination of scientific and political considerations. The findings of this study highlight how processes of purification and mediation contributed to the construction of multivocal accountability by the Italian government, through the redefinition and repositioning of multiple—and at times competing—interpretations (Latour, 1993) that emerged in response to a complex, uncertain and evaluative challenge (Ferraro et al., 2015) such as the COVID-19 pandemic. The Italian government’s approach demonstrates that accountability is not a static process but rather a dynamic construct that adapts to shifting socio-political and scientific landscapes. Initially, the government sought to justify its decisions as purely scientific, relying heavily on the KEIs to sustain an objective, data-driven response. In doing so, the government enacted a purification strategy (Latour, 1993) that positioned its COVID-19 response as based solely on scientific explanations. However, as the challenge response progressed, the government’s accountability discourse began to incorporate political considerations, including economic and social concerns, as well as lessons learnt from its ongoing experience in managing the pandemic. This shift signalled the emergence of a mediation process, through which diverse voices and perspectives were integrated to justify governmental decisions (Latour, 1993).
In summary, the analysis demonstrates that the multivocal nature of accountability is jointly constructed through interrelated processes of purification and mediation, which necessarily combine to address the complexity inherent in GCs. In the context of GCs, the interplay between purification and mediation processes varies according to the degree of emergency and the level of experience accumulated by those responsible for managing the response. In the initial stages of a crisis, purification strategies tend to prevail, serving as a means of delivering a rapid response by concentrating on a single dimension of the multifaceted challenge. However, as experience is gained over time and through practice, mediation processes become more viable, allowing for the inclusion of multiple aspects of the challenge and fostering a more nuanced and integrative approach. Indeed, the combination of purification and mediation processes in the government’s accountability was particularly evident in the selective framing of the challenge, where scientific data (such as infection rates and hospital capacity) were used to justify restrictive measures, while political considerations (i.e. economic or social aspects or lessons learnt) were brought to the fore at times when scientific data were suggesting a different response.
These processes are found to characterise the Italian government’s multivocal approach to accountability, enabling it to engage with the multifaceted nature of a GC by accommodating multiple interpretations of the crisis. Indeed, throughout the response, the government’s multivocal accountability reflects the multiple possible interpretations of a GC, shifting between scientific and political considerations, all worthy of consideration in building a possible solution to the challenge (Ferraro et al., 2015). The coexistence of purification and mediation processes supports the continuous redefinition and repositioning of explanations that are necessary within the context of GCs (Ferraro et al., 2015). As Latour (1993) argues, “science” and “politics” coexist along a “continuous gradient” (p. 88), constantly intermingling and “alternating the ‘symbolic dimension’ of things with the ‘natural dimension’ of societies” (p. 85).
As a result, this study argues that, by adopting a multivocal form of accountability, governments can integrate diverse concerns into their responses to GCs, thereby enhancing their capacity to adapt to the complex and continuously evolving nature of such phenomena. The findings underscore that governmental leaders can more effectively address GCs by engaging with their inherent complexity, uncertainty and evaluative dimensions, and by resisting the tendency to separate scientific and political considerations. Drawing on Latour’s (1993, 2004) conceptualisation of the interrelation between politics and science, the study demonstrates that the accountability of those responsible for tackling GCs emerges from the combination of purification and mediation processes. This form of accountability is inherently multivocal, as it must accommodate diverse—and at times conflicting—interpretations of the challenge, each of which offers a legitimate basis for action.
This study contributes to the literature on accountability in the context of GCs in several ways. First, by examining how governments navigate complexity, uncertainty and competing evaluative frameworks, it introduces the concept of multivocal accountability to characterise the form of accountability discharged by those responsible for addressing GCs and wicked problems. In doing so, the paper extends current understandings of accountability during the COVID-19 pandemic by exploring its multivocal nature beyond the contingent styles (Andreaus et al., 2021), configurations (Columbano et al., 2024) and accounts (Contrafatto et al., 2024) through which it has previously been studied. The findings suggest that, irrespective of the specific styles, configurations, or forms of accounts employed, accountability in such contexts involves a continuous interplay between science and politics. This interplay renders implausible any claim that decisions are made on the basis of purely scientific or purely political grounds. In the context of a GC, scientific and political rationales are inherently and inextricably interwoven in both decision-making and its justification. Recognising this entanglement represents a first step towards developing a form of multivocal accountability capable of reconciling the diverse and often conflicting interests at stake. This line of inquiry may enrich our understanding of accountability dynamics in atypical yet increasingly relevant contexts (Granà et al., 2024).
Secondly, by examining the COVID-19 pandemic over a 15-month period beyond its initial critical stages, this study extends the existing literature by capturing the evolution of accountability practices over time, rather than focusing solely on the early, turbulent moments of the emergency (Broadbent, 2020; Demirag et al., 2020; Contrafatto et al., 2024). This extended temporal scope reveals that data associated with specific categories of accountability—such as scientific indicators, including infection rates and hospital capacity—are not consistently utilised in a uniform manner. Rather, the interpretation and use of such scientific data made by the government are mediated by political aspects, allowing for multiple, and sometimes contrasting, readings of the same information.
Thirdly, by drawing on Latour’s concepts of purification and mediation, this study affirms the relevance and applicability of Latourian thought for investigating the dynamics of GCs, responding to recent calls in the literature (Danner-Schröder et al., 2025). The study further demonstrates the potential of Latour’s conceptual framework for advancing accountability research, extending beyond prior work that has predominantly focused on accounting tools rather than on accountability discourses (e.g. Skærbæk and Christensen, 2015). Specifically, the research underscores the value of analysing purification and mediation in tandem as a means of understanding how organisations shape accountability through a multivocal lens. Future research could build on these insights by engaging with additional Latourian constructs to further enrich our understanding of accountability, in line with recent scholarly initiatives such as the call for papers issued by Qualitative Research in Accounting and Management (Yates et al., 2025).
While this study provides valuable insights into the multivocal nature of government’s accountability during GCs, it is important to acknowledge some limitations. The analysis is based on a single case study. Future research could extend this investigation to other national contexts, enabling comparative analyses of how accountability is constructed and performed across different settings. Moreover, the concept of multivocal accountability could be further explored in relation to other pressing global challenges—such as climate change or poverty—thereby broadening our understanding of how governments navigate complexity, uncertainty and competing demands in varied domains.
The authors are grateful to the Guest Editors Cristiano Busco, Charl de Villiers, Ruth Dimes and Fabrizio Granà for the support during the review process, and the reviewers for their insightful comments.
Notes
DPCM stands for Decreto del Presidente del Consiglio dei Ministri (Decree of the President of the Council of Ministers) which is a form of urgent legislative intervention directly issued by the Prime Minister and also known as “Emergency Decree”.
The monitoring reports are all publicly available on the Ministry of Health official website (available at: https://www.salute.gov.it/portale/nuovocoronavirus/archivioMonitoraggiNuovoCoronavirus.jsp).
