Book Reviews

In Data Theory, Simon Lindgren offers the “modest proposal” that the “datafication” of society requires those who study and analyse it to adopt methods that are appropriate to a data-saturated social environment. What follows over eight chapters is a series of studies in how this proposal plays out in encounters between interpretive sociological concepts and case studies where computational methods are applied to datasets such as Reddit posts, tweets from the 2018 Swedish elections, and the #deletefacebook movement. Borrowing the concept of “deep mediatisation” from Couldry and Hepp (2017), Lindgren is careful to note that media are no longer channels of centralised content; instead, they function as “platforms for enacting social life” (p. 8). Hence the need to take data seriously as both raw material and lived experience.

Therefore, Lindgren's book takes a welcome step back from the concept of Big Data; rather, he provides a valuable corrective to the fuzziness of this term. As boyd and Crawford (2012) point out, it refers to a combination of technology, analysis, and mythology. What Lindgren offers instead is a conceptually informed engagement with specific, limited datasets – some of them quite large. In these case studies, Lindgren provides a valuable set of illustrations of how to “do” data science as a social researcher, equipped both with the conceptual apparatus of social theory and the tools of contemporary data science. This is an iterative process, which presents some challenges for the linear form of a book in which one thing comes after another. Lindgren demonstrates a continuous back-and-forth between theoretical concepts, data, and research questions, illustrating how each informs the other.

The Reddit case study in Chapter 5 is a good example, in which Lindgren applies a sociology of knowledge perspective to the analysis of large-scale unstructured text data. Or, as he might prefer to put it, the case study is “theoretically sensitised” by the sociology of knowledge. The focus in the chapter is on “collective representations”, using a dataset of 1.2 billion Reddit posts. Lindgren uses text mining, machine learning, and word embedding models to conduct ethnography in a networked social setting. Machine learning as a tool for social analysis is already “part of the architecture of sociality in today's world” (p. 99). Therefore, he argues, it is imperative for social researchers to understand it and use it, given that we encounter machine learning any time we open a social media newsfeed or browse through recommendations on a streaming service.

There is an impressive historical depth to Lindgren's approach to sociological concepts. On the first pass through the book, the case studies monopolised my attention as a reader; however, on the second read-through, I appreciated more the theoretical richness and range of his references. Data Theory amounts to a revisionist intellectual history of social theory, an argument built around the centrality of data in its various forms for social research. For example, noting the difficulties that researchers have had in applying Bourdieu's theory of forms of capital to large-scale empirical explorations, Lindgren shows the theoretical framework in a fresh light: Applying Bourdieu's concepts relies on finding quantitative proxy metrics for qualitative comparisons within a taxonomy of forms of capital (social, cultural, economic, and symbolic). Therefore, Lindgren presents us with a modification of Bourdieu's taxonomy of capital forms “in order to make them measurable through social media data” (p. 5).

The focus on data and how it can be used as raw material is another distinctive feature of Data Theory. In sharing the technical details of his methods, Lindgren makes explicit which Python libraries he uses and how for each case study. Therefore, this book serves as a demo of sorts, a hands-on show-and-tell of how to use open-source libraries and coding tools in practice. This represents a valuable teaching resource: Students, especially those who have some fundamental training in Python for data analysis and machine learning, will be able to follow the connections that Lindgren makes from theoretical concepts, through the choices of case study and datasets, and down to the details of code in Jupyter notebooks available on GitHub. In the style of a cookbook author, Lindgren shares his code and makes explicit how he uses Python libraries such as word2vec as well as specific software packages like Gephi to collect, analyse, and represent the data featured in the case studies. This commitment to sharing the practice as well as the theory makes Data Theory an encouraging, open-ended book for students and researchers to use as a starting point for their own projects.

This is emblematic of Lindgren's overarching commitment to an ethos of hacking and theoretical bricolage, to opening up a new field of possibilities “in the name of simply doing the best we can with the tools we have at hand, to get research results that make sense” (p. 126). In this, Data Theory sits in close network-proximity among other works dedicated to “mess” in social research, along the lines of John Law's After Method (2004), embraces a methodological inclusivity in the style of Richard Rogers's Digital Methods (2015), and demonstrates a commitment to locating data and artificial intelligence within wider social and economic structures, as in Kate Crawford's Atlas of AI (2021). Moreover, with its explicit open-source commitment to sharing, alteration, and transformation, Data Theory gestures towards technical documentation by making explicit how and why certain libraries are used and by sharing the GitHub code repositories for reviewing and repurposing.

The intended audience of Data Theory is students and researchers in specialisms across social research who want to expand their standard disciplinary toolkit into data science. However, the book can also be profitably read from the other direction, as an invitation to add the conceptual richness of interpretive sociological concepts to data science and machine learning. As the sociology of knowledge teaches us, there is no such thing as a tool or technique disconnected from its social context. Drawing on Barney Glaser's notion of “theoretical sensitivity”, Lindgren's leitmotif is the question, “What is the relationship between data and theory?” (p. 145). One of the valuable contributions this expansive book makes is to “reboot” this question for the era of databases, analytics, and data science. This is data science as ethnography in its expansive sense, the writing of culture and society, which “can be ‘written’ in many ways” (p. 148). What makes Data Theory successful is that it simultaneously invites and enables its readers to do that for themselves.

From time to time, a community must take a moment to take a critical look at itself, especially in light of major events changing the world. This is also the case with the journalism profession. In a time when journalists and newsrooms face increasing pressure, including questions of trust and distrust, fake news, and financial difficulties, a critical assessment of the profession both from within and outside is of importance. This book offers this.

How do events such as the Covid-19 pandemic, the Rwandan genocide, or the war in Syria influence journalism? What about a Danish media scandal? In this book, four leading international journalism scholars have taken on the important and complicated task of assessing how critical incidents have made and continue to make journalists reconsider their routines, roles, and rules.

So, this is a most welcome and timely book. At the same time, the issue of critical incidents and journalism is complicated and comprehensive. There are many possible connections, and it is an ambitious task. In Critical Incidents in Journalism, an impressive field of 33 authors contribute to this task in 18 chapters. As a first impression, this looks overwhelming; however, the book is organised in a way that helps the reader navigate through the various sections and chapters.

The book is organised into four sections addressing various aspects of critical incidents: conceptualizing (Section I), characteristics (Section II), communities (Section III), and finally, consequences (Section IV). Combined, these sections take us through the issues from the journalists’ perspectives, interpretive communities inside and outside journalism, as well as looking at the issues through various case studies.

Section I (Conceptualizing Critical Incidents), through two chapters, discusses the concept of critical incidents, theoretical foundations, as well as methodological aspects of applying the concept to empirical work, how it has been used historically in journalism studies, and how journalism studies has contributed to the understanding of critical incidents.

Section II (Characteristics of Journalistic Work) features case studies from around the world, in order “to illuminate the role of critical incidents in discursively negotiating various characteristics of journalistic work” (p. 6), and this is done from an “inside” point of view. This section is based on no less than seven case studies, which can be grouped in two parts.

In the first part, consisting of three chapters, the voices of the journalists themselves are vital. Chapter 2.1 discusses a case of how accusations of plagiarism in Denmark instigated a debate about the boundaries of journalism practice. Chapter 2.2 deals with children's role in news narratives about the war in Syria. Finally, Chapter 2.3 looks at how news media in Switzerland covered a broadcasting licence fee case that affected the journalists themselves, and how self-interest can shape news narratives.

The second part of this section discusses how journalists deal with issues of state repression, injustice, and violent attacks. Here, the aspects of journalist's roles in society are addressed. Chapter 2.4 is based on a case study from Kenya, where the authors show how journalists publicly renegotiated their roles amid a nationwide effort toward healing and reconciliation. Chapter 2.5 deals with security issues among journalists covering Mexico's drug war – including protection from the cartel, politicians, and public officials – which led to self-censorship, avoiding reporting, and anonymity. Chapter 2.6 discusses insights from Rwanda and genocide as a critical incident, including insights from both foreign journalists and Rwandan journalists covering the genocide. The last chapter in this section focuses on a case from Brazil where journalists got it wrong, and people were falsely accused: a case that led to self-assessment of journalistic working methods.

Section III (Communities Engaging in Interpretation) brings together journalistic and non-journalistic communities of interpretation to reflect about varied aspects of journalism. It aims to demonstrate “how critical incidents can activate internalizations and discourses from a range of interpretive communities that can extend beyond those of journalists” (p. 8). Chapter 3.1 shows how photojournalists covering President Rodrigo Duterte's drug war in the Philippines engaged in public self-reflection, also questioning their own norms. Chapter 3.2 covers a case from the UK and the scandal around the News of the World phone hacking, where members of the news media disagreed on press regulation versus press freedom as a response to the scandal. Chapter 3.3 explores journalistic reflections about coverage of the pro-democracy protests in Hong Kong and how attacks on journalists were framed by both Chinese- and English-language press.

The two final chapters of Section III show how more peripheral actors engage in journalism. Chapter 3.4 discusses how Save the Children in the UK and others were able to facilitate citizen-produced content, like blogs, to catch the interest of journalists – and how this broke with the traditional coverage of humanitarian crises. The final case is from New Zealand, where the authors examined how social media users reflected on the livestreaming of the 2019 Christchurch shooting, including the sharing of the content. The chapter discusses the roles of social media users as possible gatekeepers of content, a role traditionally kept by journalists.

The final section of the book, Section IV (Consequences of Critical Incidents), discusses the outcomes of critical incidents in four chapters. Chapter 4.1 focuses on how the Charlie Hebdo terror attack led to a cross-border collaboration and cooperation among fellow investigative journalists and how collaboration and solidarity can come out of a critical incident. Furthermore, Chapter 4.2 discusses how the conduct of the German newspaper Der Spiegel changed state-press relations in Germany after World War II. Chapter 4.3 addresses the case study of the Tahrir Revolution as a critical incident and the how it continues to shape politics and media in Egypt. Finally, Chapter 4.4 revisits the concept of a critical incident and asks: What if an incident ought to be treated as critical but ultimately is not? Here, the potential limitation of the concept of a critical incident is discussed.

In the book's final chapter, the editors draw together the various threads and complete four tasks: First, they situate the concept within journalism studies. Second, they identify two main approaches to the study of critical incidents: a discourse-centric approach that emphasises what journalists say and a consequence-centric approach that emphasises what journalists do. Third, they identify critical considerations of the concept, like temporality and contextuality. Finally, they provide recommendations and advice to journalism researchers for further studies of critical incidents.

After this overview of the content, and after having assessed the logic and structure of the book, my overall conclusion is that this is a timely, important, and impressive contribution to journalism studies. The main issues of critical incidents in journalism have been studied before, and the editors refer to works by Barbie Zelizer and others who have both inspired and influenced this book. Critical Incidents in Journalism combines an impeccable scholarly level with a rich collection of case studies. The strong team of authors from around the world also reflect their cases in critical literature, which again gives a rich discussion of journalism studies and the state of journalism in light of critical incidents. And the editors manage to reflect on the interplay between critical incidents and journalism in light of the case studies. This shows that the book has needed this strong team of four experienced editors to handle the massive amount of material and to assist the team of authors in this anthology.

The above mentioned strengths can also be a challenge for the reader, however. The material is so massive and comprehensive that the book is probably best used as a resource where selected relevant cases are read and applied. The conceptual discussion of the main concepts in the beginning is recommended reading; here, not much is missed. I had to take a break several times when reading, since crucial thoughts are on almost every page.

Critical Incidents in Journalism has a global profile, and it is comprehensive and wide-reaching, ambitious, and covering so many issues that one might be a bit apprehensive as a reader. However, the outline of the four c's of critical incidents in journalism (conceptualizing, characteristics, communities, and consequences) is a guiding light though the material.

Finally, I want to quote the beginning of the book, where the editors encapsulate the importance and potential of critical incidents in journalism:

In helping to illuminate what norms, functions, and practices journalists cherish and defend, studies of critical incidents in journalism contribute to a more nuanced understanding of the sensemaking that journalists engage in, as well as to a growing cache of collective memory and lessons learned that can guide current and future stakeholders of journalism in navigating what shapes to be a challenging future for the field.