Paradoxes of Media and Information Literacy

Front matter

“A remarkable book. By clearing conceptual ground, synthesizing policies

and debates, formulating challenges and first-hand reflections, the book will stand as a key book for researchers and students dealing with MIL.”

Ulla Carlsson

Contents

List of figures Preface

The crisis of information 2 Does alcohol protect you against a virus? 2 Crisis, co-constitution, and digital culture 4 Facts and opinions 6 Erosion of trust 9

Paradoxes 14

Preface

Having worked on information literacy for many years, and more recently also on media and information literacy, we have struggled with the many paradoxes we have encountered. Even our research seemed to be built on a paradox, caught between being asked for advice on how to develop and teach media and information literacy and studying this extremely heterogeneous field from a critical distance. As a researcher, one would have the obvious choice of either accepting the contradictions and moving forward in research to make a difference or changing focus and examining the field as such. Since we sometimes overcomplicate things, we decided to try and do both at the same time.To paraphrase a famous saying: We want to have the apple and eat it, comment on how the symbol-laden fruit of knowledge and temptation is rotting as we speak and provide means to intervene.

The book contains questions and ideas that we have been developing and carrying around for a long time, but the book is mainly an outcome of the project “Algorithms and Literacies: Young People’s Understanding and Society’s Expectations” (The Swedish Research Council, 2017-03631). We also build on material created in two other research projects, although to a lesser extent. In “Search critique and the critical assessment information in compulsory school” (funded by The Swedish Internet Foundation) Hanna Carlsson was a co-researcher and in “Algorithm awareness between generations” (funded by the National Library of Sweden) Lisa Ohlsson Dahlquist was a co-researcher. Thank you very much, Hanna and Lisa! Your contributions were very important, and without your insights this book would not have been the same. We also want to thank Johan Frid at the Humanities Laboratory at Lund University for his help.

Some of the ideas and empirical material we draw on and elaborate in the book have been previously presented in earlier shorter versions in the following publications. “Information literacy as a site for anticipation: Temporal tactics for infrastructural meaning-making and algo-rhythm awareness” in Journal of Documentation, 2021;

xii Preface

“Information literacy challenges in digital culture: Conflicting engagements of trust and doubt” in Information, Communication & Society, published online 2020; “Where is Search in Information Literacy:A Theoretical Note on Infrastructure and Community of Practice” in Sustainable Digital Communities: 15" International Conference, iConference 2020, Boras, Sweden, March 23—26, 2020, Proceedings; Algoritmmedvetenhet i mötet mellan generationer: En forskningsrapport inom ramen for Digitalt först med användaren i focus, 2020 (with Lisa Ohlsson Dahlquist); “Skolan och biblioteken som aktörer for demokratiska samtal” in Det demokratiska samtalet i en digital tid: En antologi om desinformation, propaganda och näthat, 2020;“How Do you Trust? On Infrastructural Meaning-making and the Need for Self-reflection” in Understanding Media and Information Literacy (MIL) in the Digital Age: A question of Democracy, 2019; Sök — och källkritik i grundskolan: En forskningsrapport, 2018 (with Hanna Carlsson).

We have also presented some of the ideas at seminars and conferences in Sweden and internationally. We would like to mention the keynote “Algorithms and Platforms: Is there a Place for Information Literacy?” that Olof Sundin gave at the ECIL European Conference in Information Literacy in 2021. Olof Sundin was also invited to give the talk “Literacy, Open data and the Crisis of Information” at the Data Literacy in Context — DaLiCo.Virtual Summer School on Open Governmental Data (2020) where he presented some initial thoughts on the crisis of information. These two talks developed ideas that became part of the first and second chapters. Jutta Haider presented earlier versions of Chapters 2 and 6 at higher seminars at the Department of Culture and Media Studies, Umea University, and at the Centre for Nordic Media Research at Gothenburg University (Nordicom). Jutta Haider also gave a talk entitled “From what to when: On representations of information inequality and the role of time” at the Symposium on Theories of Information Disparities in Society (2017), Nankai University, China. This was the first time the ideas that are presented in Chapter 4 were tested.

To our colleagues at Lund University and at the Swedish School of Library and Information Science in Boras: rather than risking forgetting anyone, we would like to thank you all. The research seminars at our respective universities have been particularly important: Research Seminar for Mediated Culture and Information at Lund University and Research Seminar in Library and Information Science as well as the Information Practices and Digital Cultures research group at the University of Boras. We would also like to thank all the students, teachers, and librarians we met in lectures, workshops, and conversations inside and outside the university. As we prepare for these activities, many of our ideas take on an understandable format.

A special thanks to Mathias Cederholm, whose tireless work mediate everything important concerning media and information literacy in the Swedish Facebook group “Källkritik, fake news och faktagranskning” [Source criticism, fake news, and fact checking]. The efforts that Mathias makes on a daily basis, together with the other administrators and members, have been absolutely indispensable for our environmental scanning. We also want to extend our gratitude to the many participants in the different research projects this book draws on. You talked to us about

Preface xiii everything from literacy, to Google, to algorithms and different aspects of your everyday or professional lives and provided many thought-provoking perspectives on the challenges society faces.

During the last months’ work with the book, we asked some colleagues to read chapters at short notice. Thank you for reading and commenting parts of the book: Fredrik Bertilsson, Hanna Carlsson, Ulf Dalquist, Alison Gerber, Joacim Hansson, Heidi Julien, Dirk Lewandowski, Karin Linder, and Lisa Olsson Dahlquist. You have generously shared your expertise with us, for which we are very grateful. Special thanks go to Ola Pilerot and Louise Limberg, who read and commented on most of the chapters in the book. With your long and extensive experience in the field, your insightful comments were extremely valuable. Getting a “thumbs up” from you was more important to us than you can imagine. At the same time, any weaknesses, mistakes, or problems in the book are definitively not the fault of any of our colleagues.

Taylor & Francis Taylor & Francis Group http://taylorandfrancis.com

Introduction To The Literacy Paradoxes

Can injecting disinfectant help you recover from a viral infection? Obviously, it cannot, but it was what former US President Donald Trump recommended scientists investigate, live at a press conference in April 2020. In the pandemic years of 2020 and 2021, the world exploded with information about a new respiratory disease, COVID-19, that was caused by a novel coronavirus. We encountered highly complex scientific information, advice issued by public authorities, rules and regulations provided by various authorities, different types of health information, speculation about prevention and treatment, a flurry of statistics, in addition to rumours and numerous conspiracy theories and purposefully spread, potentially harmful content. We also encountered tweets by concerned citizens and Facebook updates with interpretations of information they had read, as well as TikToks from all around the world about life under lockdown. Datasets from different countries and international organisations were made available. Soon, anyone with an internet connection and a spreadsheet program could act as an amateur epidemiologist, seemingly without any need for a contextual understanding of the production of data. Information on this new disease and the virus causing it was everywhere. The various strategies different countries chose, as well as the responses to them, came to be considered not merely through the lenses of epidemiology and public health, or even international relations and policymaking, but also as symbols of ideological or even national identity. Counting and comparing the number of cases, ICU admissions, and even the death toll across different countries developed into a competitive sport, a cynical new activity for some to engage in when all else was shut down.

This book is not about COVID-19, but many information challenges became acutely apparent in the wake of this pandemic. These are brought together with examples from other areas and used to locate what we propose to call an ongoing crisis of information in contemporary society. This crisis, we argue, is characterised by a volatility of information that manifests differently across various domains, but which taken together presents a number of challenges to the precariously balanced networks of institutions, knowledge systems, and practices that have traditionally — but of course, imperfectly — constituted democracies in the West. Much is to be said, and has been said, about this crisis, and there are diverse disciplinary contributions to be made to the analysis and understanding of it. Our contribution to this conversation is a modest one, but one which we hope adds a unique dimension. We focus our attention on one specific attempt at containing this crisis of information, namely media and information literacy, a multifaceted and malleable concept whose plasticity makes it an ideal device for interrogating the current information landscape. It appears in different guises in various fields, and — as we will explore in due course — it is loaded with different, often contradictory values, norms, and ideals. At the most basic level, what unites the otherwise profoundly different understandings of media and information literacy is the assumption that certain understandings, competences and resources enable people to proficiently engage with information and media and that these are involved in arranging an individual’s relation to society.

Clearly, in many ways, this is a valid assumption. At the same time, information in contemporary society is characterised by great instability, carrying with it an increasingly profound uncertainty regarding the control over public knowledge. Where does this leave us? We want to bring to the fore effects and manifestations of this volatility. This crisis of information has many shapes, forms, and facets, but through the prism of media and information literacy we can identify a number of information-related challenges that constitute this situation. These include fragmentation, individualisation, emotionalisation, and the erosion of the collective basis for trust, all with implications for not just the individual, but also for democracy, and ultimately the organisation of society. We will provide some clarification on our use of these concepts in the next section. Of course, different aspects of the state of information in contemporary society have been studied, argued over, and theorised in myriad ways, from different disciplinary, political, legal, or economic perspectives. The flow of publications is unlikely to ebb any time soon. This is an expression of the urgency of the concern being felt; an urgency that is amplified by the realisation that the crisis of information cannot be divorced from its social and experiential foundations and effects: it is thus — simultaneously a health crisis, a climate crisis, a crisis of trust, a crisis of democracy and participation, and even a crisis of the future.

The crisis of information Does alcohol protect you against a virus?

The English-language Wikipedia article now named “COVID-19 misinformation” has a version history going back to early February 2020, a good month before the World Health Organisation (WHO) even declared the outbreak of a pandemic (Wikipedia 9-9-2021). Since then, the article has grown long and rich in content, and similar articles exist in over 30 other languages on Wikipedia. It includes well-known and widespread conspiracy theories such as that the spread of the virus is caused by electromagnetic fields from 5G mobile networks, or less popular ones such as the idea that the virus arrived in the Wuhan region of China with a meteorite.

The article also contains a section on the misinformation that drinking pure alcohol supposedly protects people against the virus. This false claim has led to the deaths of hundreds in Iran as a result of drinking methanol (Islam et al., 2020). CNN reports how the former US President Donald Trump alone made 654 false claims about the virus in just 14 weeks during spring 2020 (Dale and Subramaniam, 2020). Some of the false information circulating about the virus is easy to debunk; for example, the danger of 5G mobile communications masts, an idea that has resulted in sabotage of such masts and threats against people working with the network. The idea that cocaine use can prevent COVID-19 was so widely spread at one point that the French Ministry of Health found itself officially disparaging the claim. Other theories or false claims arise from a lack of context. This complicates comparison between different countries and leads to certain explanations or statements changing their meaning when transferred from one country or context to another. Some theories are based on unsubstantiated correlation, while yet others might even start from actual science-based observations which are then interpreted and developed further in ways that are unsupported by the scientific study in question. What Wikipedia’s taxonomy of COVID-19 misinformation and similar taxonomies show is less how many and which different types of misinformation and conspiracy theories exist about COVID-19, but rather that what we call misinformation is by no means a homogenous entity, neither are so-called conspiracy theories. We will return to this important point in due course.

Before the WHO classified the outbreak of the then-novel coronavirus as a pandemic and even before said virus and the disease it causes had been assigned their official names, SARS-CoV-2 and COVID-19, the organisation issued a warning about the new coronavirus being “accompanied by a massive ‘infodemic’” (WHO, 2020). This, they explained, is “an over-abundance of information-some accurate and some not-that makes it hard for people to find trustworthy sources and reliable guidance when they need it” (ibid.). Since then, talk of an infodemic has joined the older image of information overload (Bawden and Robinson, 2020) in public debate and media reporting. Infodemic makes for a powerful metaphor, yet it invites problematic oversimplification of a complex social phenomenon by biologising it (Simon and Camargo, 2021). The image fuses well with already established analogies and images, such as those of computer viruses and memes or other content going viral on social media, and engenders new ones, including the notion that people can inoculate themselves against fact resistance, that psychological vaccinations are possible, or that so-called ‘fake news’ can be stopped by creating herd immunity to it. The term “fake news”, in particular, has come to be used by politicians and other public persons to describe news they reject or discredit positions they disagree with. It is in the current debate best understood as a “floating signifier”, as Farkas and Schou (2018) suggest, which is also why we refrain from using the notion as an analytical concept.

That said, if acted on, certain types of incorrect health information can be directly lethal. For example, the American Center for Disease Control (CDC) reported deaths that had resulted from people drinking hand sanitiser in attempts to treat or prevent COVID-19. Drinking urine, on the other hand, may not be lethal or even dangerous by itself, but using it to treat COVID-19 can still set in motion chains of events leading to deadly outcomes and add to an already ongoing destabilisation of trust in healthcare professions and government institutions more broadly (Islam et al., 2020).

Crisis, co-constitution, and digital culture

Paradoxes of Media and Information Literacy: The Crisis of Information is a book about media and information literacy in digital culture. The word crisis marks a turning point, and in its original Greek meaning it denotes the turning point in a disease. While we want to be careful not to simplify complex, multi-layered social phenomena using poorly understood medical metaphors, there is a certain appeal in describing the extreme volatility of information that characterises contemporary society as just that: a crisis and a potential turning point. The appeal lies in the way that crisis, in such an understanding, conjures up the possibility of hope on the one hand, but also leaves open the possibility of different outcomes on the other. And further, that it is not a sudden event that comes out of nowhere, but one that has multiple causes, different developments and, regardless of how it unfolds, inevitably leaves traces or even scars, and opens up new paths. The crisis of information cuts through numerous social arrangements and reveals (and also challenges) their interdependence in new and profound ways. In the process, many taken-for-granted assumptions — not only about what counts as information, but also about how information should be produced, questioned, organised, and sustained — are being challenged. But what are the most tangible facets of this crisis, and how can they be traced out for our exploration of media and information literacy?

During the COVID-19 pandemic, the increased volatility of information has become ever more palpable. Rumours, governmental information, research reports, pre-prints, policy documents, official statistics, research data, journal articles, and conspiracy theories, all amalgamate on the same platforms, detached from their original context and imbued with different meanings. Often, it is near impossible to confidently establish the origin or status of the fast-changing, continuously updated information that is aggregated in social media feeds, in search engine results, or even in spreadsheets. Fragmentation describes how a complex body of knowledge is arranged into a continuously shifting shape provided by networks of ever fewer corporate information platforms. Not only is information becoming more fragmented, but access to information is also becoming more individualised, depending on, for example, who you are, who you follow, and who you interact with. The extent of personalisations and their societal effects are invisible to the individual user. Nobody knows for sure what others encounter, which unique combination of apps, search engines, and social media make up other people’s information ecosystem, how they arrive at which terms to enter into which search engine, or how the feed will reconfigure itself in the next reload, in response to the next swipe, after the next search or engagement. Positioning the fragmentation of information at the level of experience highlights how intimately the ongoing fragmentation of the collective understanding of society’s knowledge base relates to the information infrastructure within which everything from everyday life to politics plays out.

Both the foundations of trust and the possibilities for the creation of shared meaning are called into question. This points to a number of interrelated challenges that contribute to the destabilisation of information — fragmentation, individualisation, emotionalisation, and the erosion of the collective basis for trust — all of which are, clearly not created, yet exacerbated by society’s commercial, algorithmic information infrastructure, and a qualitatively new form of politicisation of information, where information is increasingly tailored to the form provided by multi-sided platforms and their specific logic of amplification. The importance of algorithms, user data, and increasingly Al-based systems for contemporary culture, more specifically for multi-sided platforms such as search engines, recommender systems, intelligent household assistants, streaming services, or dating apps, cannot be overstated. It is becoming increasingly important to understand how algorithmic systems work and how they are trained to perform in specific situations, while at the same time they are becoming ever more elusive and embedded in society and everyday life at all levels. One way to approach this is to use the term “digital culture”, not as a historical period with a clear definition at the expense of other understandings of culture and society, but as a perspective that allows us to foreground the prevalence of certain socialities and ways of knowing and being in the world that are embedded in the organisation of society at different levels.

To borrow from media researcher Ted Striphas, “algorithmic culture” is one where “human beings have been delegating the work of culture — the sorting, classifying and hierarchizing of people, places, objects and ideas — to data-intensive computational processes” (Striphas, 2015, p. 396; see also Lloyd, 2019). Digital culture as a perspective takes this into account, but also how algorithms, along with the platforms in which they are embedded, and the data extracted from users, exist within wider rationalities and programmes of social change (see also Beer, 2017). Like all knowledge organisation and information systems, multi-sided platforms, algorithms, and data are always contingent and never impartial. In other words, algorithmic configurations, and multi-sided platforms are active and co-constitutive of the wider social fabric. Co-constitutive implies not only that this is not a one-way relation, but also that people, society, algorithms, data, platforms, regulators and so on are all constitutive of each other (Barad, 2003; 2007; Orlikowski and Scott, 2015). For the purpose of understanding society from the vantage point of digital culture, they cannot be meaningfully separated. People create meaning from platforms and algorithms in the various practices they are involved in, but they also resist them in more or less consequential ways. Platforms and their algorithms not just influence people in one direction, but through their interactions with these systems people also change them. Fragmentation, individualisation, and emotionalisation are integral to how commercial algorithmic information systems operate, at least in the sense that the categorisations applied to cluster, target, and extract data from people is intentionally invisible to those subjected to it. Trust in public knowledge, on the other hand, requires collectively shared and societally accepted methods for producing, challenging, and vetting knowledge, which fragmentation obscures and undercuts.

Facts and opinions

In a seminal text published in The New Yorker in the 1960s, Philosopher Hanna Arendt describes how “factual truth, if it happens to oppose a given group’s profit or pleasure, is greeted today with greater hostility than ever before” (Arendt, 2000[1967], p. 552). As a result, she maintains, facts are often ideologised, and this way turned into opinions, or to use a contemporary term, into “alternative facts”, a notion popularised by Donald Trump’s former senior counsellor. These opinions can then be legitimately disagreed upon.Therefore, the strategic blurring of the line between facts and opinions is a political move that makes sense. It allows the transformation of facts into opinion, which can then be represented as negotiable information and circulated in ways that are easily scaled up and politicised. In Arendt’s words, “[u]nwelcome opinion can be argued with, rejected, or compromised upon, but unwelcome facts possess an infuriating stubbornness that nothing can move except plain lies” (Arendt, 2000 [1967], p. 556). Opinions are not only easier to dispute; such dispute is legitimate and expected in political deliberation (Davies, 2018, p. 159-165). Opinions cannot be deemed true or false; they can only be more or less popular, more or less shared between people or groups.

While not originating it, the commercial logic of algorithmic information systems certainly contributes to escalating the strategic transformation of facts into opinions, and thus of what can be called the networked mass production of the tactical politicisation of facts. Donald Trump, whose presidency coincided with the start of the COVID-19 pandemic, has become the epitome of this move. Everything became a matter of opinion, or at least ambiguous, and thus, also easily dismissible by others: climate change, sexual integrity, medical treatments, and so on. However, when Trump attempted to delegitimise the result of the 2020 presidential election, he went further. The legitimately vetted numbers that confirmed him losing the US presidential election were not only challenged in the way intended by the system, which they also were, but more importantly they were transformed to support a different fact, namely the ‘fact’ of vast fraud and conspiracy. Only then, the accuracy of the voting results could be questioned, and a new fact could be established through its circulation and repetition in various social media. This lie — at odds with all empirical evidence — was so difficult to uphold, that, to borrow Arendt’s words, expressed more than 50 years earlier:

if the modern political lies are so big that they require a complete rearrangement of the whole factual texture — to the making of another reality, as it were, into which they will fit without seam, crack, or fissure, exactly as the facts fitted into their own original context — what prevents these new stories, images, and non-facts from becoming an adequate substitute for reality and factuality?

The presidency of Donald Trump — covered in detail, as US politics often are, by media across the globe — made this rearrangement of the factual texture more visible and tangible. It became a part of political commentary in the media and in everyday life. Lies and semi-lies, or alternative facts and ambiguities, were constantly cabled out through social media, in particular Twitter. There, at the end of his four years as US president and before he was banned from the platform, Trump’s account had almost 90 million followers. Professional journalists with their ethos to “seek truth and report it” (Society of Professional Journalists, n.d.) found themselves passing on tweets and commenting their outrageousness.

The 45" president of the United States is definitely not the only person to have politicised facts in this way, and the USA is not the only country where this occurs. Yet, US politics have global reach and ramifications that are normative beyond their national borders. Many other examples, and this list could be made very long, can be identified, regarding issues such as climate science and the politics of climate change, reproductive and sexual health, structural discrimination of historically oppressed and excluded communities, public health responses to diseases, and how slave trade, colonisation, the Holocaust, and other genocides are presented and also questioned. Today’s platformised information infrastructure — such as it exists through social media, search engines, or various recommender systems — tends to increase and scale up this politicising of information through algorithmic reinforcement, allowing for amplification of content that attracts emotions, engagement, and, not least, habitual enragement. This in itself does not necessarily imply that only lies or counterfactual information have this effect. Evidently, this is not always the case and the use of social media platforms for organising in productive, community-building ways, clearly also happens, yet we argue that this happens despite, not because of the means for interaction these services provide.

Still, it has been shown that social media have, as for instance media studies researcher Richard Rogers (2021) suggests, a tendency to systematically marginalise the mainstream and mainstream the marginal. The prevailing rationale shaping society’s dominant information infrastructures is guided by market principles on almost all levels. As this increasingly applies to all types of information and forms of knowledge, it has far-reaching implications for how information is produced, judged, and vetted in the first place. Political economist William Davies (2018) makes the connection to neoliberal economic theorist and Noble prize winner (1974) Friedrich Hayek who, in a move away from the enlightenment understanding of knowledge and expertise, suggested that the credibility and trustworthiness of knowledge should be tested against consumers, and not against any internal scientific criteria. In such a perspective, Davies sums up, “[i]deas and knowledge should only be deemed trustworthy if they were tested in a competitive arena” (Davies, 2018, p. 163).

This view on how to establish what constitutes trustworthy knowledge, while at odds with the traditional values of civic debate and libraries (Tanner and Andersen, 2018), chimes well with how the algorithms of, for instance, social media platforms, search engines, recommender systems or rating sites work.

To simplify somewhat, the leading principle for visibility and reach of social media, recommender systems, and search engines is popularity. Further dimensions are involved, yet since monetisation through both data extraction and the placing of adverts depends on achieving high levels of exposure and interaction, it makes sense to assign weight to popularity. While visibility does not necessarily directly translate into trustworthiness, as its equivalent, the intimate association between the two factors still establishes what may be called a trustworthiness chimaera. This trustworthiness chimaera underlies a number of contradictions in media and information literacy expressions. Specifically, corroboration of information by establishing whether it is widely spread or just reported by one source is one of the basic recommendations for how to establish the credibility of a statement. Yet, if visibility is primarily and intentionally created for other reasons, the validity of the strategy of corroborating information by establishing how widely spread a claim is, if not entirely invalidated, at least compromised.

Visibility in social media is contingent on interactions and or engagements. Albeit slightly differently, this principle also applies to search engines. In a comparison between public agency websites and websites that repeatedly spread misinformation, the latter were shown to attract several times as many interactions as those of the WHO, the European Centre for Disease Control (ECDC) and the American Center for Disease Control (CDC) (Avaaz, 2020). Using the language of epidemiology, the same study identifies 42 “‘superspreaders” of misinformation on Facebook. It appears that misinformation and conspiracy theories are generally better suited to the creation of interactions, or at least they are more effectively produced and adapted for the purpose of creating interaction. Read with Hayek, as Davies does in his analysis, these can be said to fare better at the market of ideas. Yet, as Davies also remarks, there are ramifications to such an understanding, since here “markets are not so much tools for producing facts but for gauging our feelings” (Davies, 2018, p. 167). The emotionalisation of information, which we identify as an important facet of the crisis of information, needs to be understood through this idea of markets as shaping the value of information and knowledge, positioning them against each other. The means for gauging facts, trustworthiness, and credibility — which is precisely what media and information literacy is usually regarded as — are increasingly put to work on a terrain of feelings and, we should add, of ambiguities. In the course, they are being rendered increasingly ineffective and dysfunctional. This is intensified by what sociologist Noortje Marres (2018, p. 433) calls “a crisis of public evidence” and the “truth-less” (p. 435) character of social media, which produce knowledge primarily for data engineers to optimise the various models needed to monetise content and users of these platforms. Thus, she points out, “[d]igitization is challenging not just the status of facts in public discourse, it is undermining institutional architectures that ground accountability in empirical evidence across societal domains” (Marres, 2018, p. 343).