Abstract Drawing together recent scholarship in journalism studies, the philosophy of science, and science studies, this article offers a new descriptive and normative model of the epistemology of science journalism that acknowledges how the reduction of technical complexity affords the expansion of public meaning. To explicate the model, this article looks at how journalists construct and justify truth claims in coverage of direct detection of dark matter experiments. Ultimately, this article argues that good journalism—science or otherwise—must shepherd the relationships and connections that allow truth to circulate across time, space, and reference, while simultaneously working to open content for public discussion, consideration, and meaning-making. Introduction When a spokesperson for the White House trumpets “alternative facts” and political leaders voice expedient, yet entirely unsupported claims, it is not surprising that scholars, politicians, and even late-night talk show hosts have begun to wonder if we are living in a “post-truth era” (Tanz, 2017). Amid debates concerning climate change, biotechnology, and vaccines, scientific facts and findings are increasingly at the center of public discussion and negotiation. Yet, as our hold on truth becomes more tenuous, it is not just science that is at stake, but also the epistemological foundations of contemporary public life. Recognizing that science journalism is one of the main sources of public information about science, communication scholars have given consistent attention to news reporting on scientific research (e.g., Dunwoody, 2008; Nelkin, 1995). Many studies have looked at how the public press treats scientific issues (e.g., Boykoff & Boykoff, 2004), how the practice of science journalism is changing along with new technologies (Allan, 2011; Trench, 2007), or the wider functions that science journalism plays in public life and democracy (e.g., Brossard & Lewenstein, 2010; Secko, Amend, & Friday, 2013). Few, however, have asked about the underlying epistemology of science journalism: how it is, exactly, that reporters construct and justify knowledge claims. However, if journalism is public knowledge, understanding the epistemology of (science) journalism can help us better understand the changing knowledge infrastructures of the current age. Existing models treat only part of the story of the epistemology of science journalism. Accounts such as the “continuity model” of science communication (Bucchi, 1996, 2008) tend to highlight either the simplification of scientific detail accomplished in translating content for lay publics or the continuity of scientific fact maintained in doing so. These models, however, miss the productive work that science journalism accomplishes in generating meaning by reducing technical detail and adding content and context. Extending and building on Latour’s (2010, p. 222; 1999) account of science as “chains of reference,” this article offers a new model of science journalism epistemology. Grounding science journalism epistemology in a balance of stasis and change, the model attends to the complex interrelations between fact and meaning. For Latour, science works through the production of successive representations. Each pares away the complexity and detail in a way that preserves key relationships while also allowing scientists to recognize and produce new insights and relationships. Truth is not something that inheres in any one representation, but rather something that circulates across the whole chain, like electricity through a circuit. Although Latour is somewhat ambiguous about meaning, this article offers a reading of Latour’s account of articulation, which leaves meaning generating like a magnetic field around that circuit. There can be different sorts of relationships between truth and meaning, yet the field of potential meanings is constantly changing along with each new constituting relation among people, things, and ideas. As a result, science journalists are like magnetologists, attempting to build operational circuits of truth that link together diverse people, things, and ideas and to generate meaning-fields. In order to articulate and investigate this model, this article presents a case study involving one major initiative in astro-particle physics: the direct detection of dark matter. Combining semi-structured interviews with both the scientists behind these experiments and the science journalists who have covered them with a thematic textual analysis of a large collection of news articles about these experiments, the case grounds this new model in contemporary journalistic practice. Ultimately, this article offers not only a descriptive, but also a normative account of science journalism. This model provides a way to recognize that good science journalism must balance between maintaining the important connections from antecedent representations, while also adding and arranging content to help produce new perspectives and new meanings for articles. Good journalism, science or otherwise, can escape neither its democratic responsibility to shepherd the relationships and connections that allow truth to circulate across time and space, nor its role in opening up content for public discussion, consideration, and meaning-making—in making things public. Literature review Models of science journalism There are many different forms of public-directed science communication, from museums, to education, to popular culture (Perrault, 2013). This article narrowly considers science journalism, which it defines pragmatically as articles in periodicals that describe timely scientific research and are meant for non-expert audiences. Scholars have begun to recognize that science journalism achieves a range of social and democratic functions beyond simply communicating news about timely research (Fahy & Nisbet, 2011; Secko et al., 2013). Yet, arguing that journalism’s role as a unique form of public knowledge is central to many of these broader social functions, this article focusses on the epistemology of science journalism, the assumptions and practices through which public information about timely research is produced. One journalism textbook describes science writers as, “first of all, bridging the jargon gulf, acting as translators between the sciencespeak of the researcher and the short attention spans of the public at large” (Blum, Knudson, & Henig, 2005, p. vii). This textbook is not unique in describing science journalism as a simplification or a reduction in complexity—this is a thread that runs through many accounts of science journalism. A number of scholars have offered models that speak to the roles or functions that science journalism plays in society. Drawing on Brossard and Lewenstein’s (2010) models of the public understanding of science, Secko et al. (2013) recognize four models of science journalism in the literature. Two models address “information delivery” and two “public engagement” (p. 67). Similarly, Fahy and Nisbet (2011) offer a nine-part typology of the “roles” of science journalist, from “conduit” to “watchdog” to “advocate” (p. 780). In addressing the functions that it plays in society, both accounts, however, bypass the underlying epistemological processes through which science journalism operates. The “continuity model” of science communication includes an implicit epistemological treatment of science journalism (Bucchi, 2008; Cloître & Shinn, 1985). This framework: […] presents science communication as a continuity of texts with differences in degree, not in kind, across levels, [and] invites us to imagine a sort of trajectory for scientific ideas that leads from the intraspecialist expository context to the popular one, passing though the intermediate levels. (Bucchi, 2008, p. 61) This model, however, ultimately, follows Fleck ( 1981) in seeing that as scientific knowledge is translated for wider audiences, it loses its history, complexity, and contingency, and instead “becomes incarnated as an immediately perceptible object of reality” (p. 125). This means “the communicative path from specialist to popular science can thus be illustrated as being like a funnel that removes subtleties and shades of meaning from the knowledge that passes through it, reducing it to simple facts attributed with certainty and incontrovertibility” (Bucchi, 2008, p. 62). This model highlights the reduction or simplification of content—both in terms of scientific complexity and “shades of meaning.” That journalism reduces the level of scientific detail is undeniable. Yet, as discussed in more detail below, science journalists add something too: they can make new connections, bring in new ideas, and add new perspectives. That is, science journalists generate “shades of meaning.” It is not enough to recognize the reduction inherent in science journalism—we must attend to its complexification as well. Also, Bucchi’s assertion that science journalism brings with it a “certainty and incontrovertibility” is incongruous with wide-scale attacks on (journalistic coverage of) science (Bucchi, 2008, p. 62). Journalistic epistemology For more than 75 years, scholars have recognized news as a distinct “form of knowledge” (Park, 1940). Influenced by early work in the sociology of knowledge, Park’s approach continues to be influential in the recognition that journalistic practice is constrained and enabled by disparate social and structural forces, from economic pressures to technological changes (e.g., Bourdieu, 1999). These dynamics have become increasingly visible amid radical structural changes in traditional (see Boczkowski & Anderson, 2017) and science and health journalism (e.g., Allan, 2011; Trench, 2007). Similarly, over the past several decades, epistemologists have been increasingly considering the “epistemic properties of individuals that arise from their relations to others, as well as epistemic properties of groups or social systems” (Goldman, 2010, p. 1), as part of a “social epistemology.” Ettema and Glasser (1984) adopt aspects of Park’s sociological approach to analyze the epistemology of investigative journalism through a phenomenological “sociology of epistemology” (p. 5). Rather than attempt to “determine whether journalists’ knowledge claims are valid assertions,” they ask, “(i) what counts as empirical evidence and (ii) how that evidence becomes justified empirical belief” (p. 6). By inquiring into the justification of knowledge, Ettema and Glasser actually align themselves with what was for a long time the dominant approach in analytic epistemology. Going back to Plato’s Theaetetus, scholars adopted a three-part definition of knowledge as “justified, true belief.” As a result, analytic philosophers originally defined epistemology as the study of justification (Pollock & Cruz, 1999, p. 11). This three-part understanding of knowledge as justification was famously undermined by the publication of Edmund Gettier’s three-page “Is Justified True Belief Knowledge?” in 1963. Gettier offered two counter-examples that show how someone can have justified belief of a true proposition, but it still should not count as knowledge.1 Gettier’s article touched off a search for an additional, fourth condition of knowledge (e.g., Creath, 1992), but also shifted the focus in analytic philosophy away from looking for better understandings of epistemic justification (Pollock & Cruz, 1999, p. 14), to producing a more rigorous treatment of a range of diverse epistemological problems spanning from the ontology to the “value” of knowledge (Williams, 2001, p. 2). Put in terms of science journalists, rather than asking like Ettema and Glasser (1984), “what journalists regard as acceptable knowledge claims,” it is to ask how journalists produce “valid assertions” in the first place (p. 5). Importantly, this concern has a necessary ethical or normative dimension (see Bok, 2011; Maras, 2013). Epistemological problems “are not just about how what we do believe but what (in some sense) we must, ought, or are entitled to believe; not just with how we in fact conduct our inquiries but how we should or may conduct them” (Williams, 2001, p. 11). Epistemology therefore requires that we ask not only how science journalists produce public knowledge about science, but also what practices and approaches best achieve the outcomes we desire. Recently, a number of journalism scholars have also moved away from questions of justification to give more consideration to if and how journalists produce valid knowledge claims (Goldstein, 2007; Maras, 2013). In his recent book Journalism and the Philosophy of Truth, Hearns-Branaman (2016) identifies four models of truth that are found in or relevant to journalism. Yet, none of these models do justice to the complexity and specificity of science journalism. Hearns-Branaman suggests that the “normative epistemology of Anglo-American journalism lies in the dialectical relationship between two different epistemic practices, Realism and Pragmatism” (p. 66). Realism, associated with Positivism, is grounded in the idea that truth inheres in a proposition’s correspondence to reality. Realism, however, provides neither a descriptive nor a normative account of truth in science journalism. Science journalists are not attempting to align their reporting to the complexity of the natural world but rather to what (expert) sources tell them. Yet even then, journalists do not necessarily attempt to treat complex science with the same level of detail as scientists. This is to say, whether we use the real or the science as the metric against which to judge science journalism, it is always going to come up short. Rather than focussing on the correspondence to reality, pragmatist conceptions of truth turn on the difference a proposition makes. For Hearns-Branaman, American journalism embodies pragmatist ideas in the continued invocation for “balance” in reporting, in which the news acts as a “marketplace of ideas” (United States, 1953). Here, the goal is not to discover a transcendental truth, but rather to understand “what best serves our pragmatic needs now” (Hearns-Branaman, 2016, p. 54). While science journalism does seem to reject transcendental truth in favor of the best current description—always subject to revision with new data—science journalism rarely acts like a “marketplace of ideas.” Instead, science reporters attempt to offer clear, accurate descriptions and explanations of scientific research. Hearns-Branaman identifies two additional frameworks more associated with academic scholarship of journalism, what he calls the “anti-realist” and the “hyper-realist” models. Hearns-Branaman associates these epistemologies with a wide range of (post)structuralist and social constructivist thinkers. However, these approaches offer little purchase on a science journalism that fundamentally holds to some sort of notion of knowable external truth. A new model Latour and scientific representations Following the “continuity model” (e.g Bucchi, 1996, 2008) discussed above, this article assumes that models of science journalism epistemology must in some sense contend with scientific truth. However, rather than draw on accounts in the analytic philosophy of science, this article looks to recent work in science and technology studies for a productive account of truth in science. Unlike the philosophy of science, which has worked for centuries to ground and understand the epistemology of science, science studies scholars have been far more willing to engage with the empirical reality of scientific practice (Fuller & Collier, 2004). Science journalism traffics in representations of scientific information or content. Bruno Latour, one of the founders of science studies, offers a treatment of scientific practice and truth that recognizes the epistemological function of representations in “Circulating Reference” a chapter in his book Pandora’s Hope (1999). For Latour, science is a set of practices through which successive representations of the natural world are produced. However, these representations must be recognized as complex “actor-networks” that are simultaneously social, material, and discursive. It is the job of the scientist to ensure that the important relationships are preserved across successive representations. Each representation is necessarily a simplification of the complexity of the one before. Yet, because representations pare away certain information, they not only make it possible to observe relationships that might otherwise have been hidden, they also permit the extension of new connections and relationships. For example, dark matter physicists build instruments that can detect particles far too small to see. These instruments supply huge volumes of data about the particles they encounter. These data do not do justice to the utter complexity of the world, but they produce reliable information about some of the relationships of interest to physicists. To make sense of this data, physicists might produce a graph that shows different characteristics of detected particles. This graph captures neither the complexity of the data, nor that of the world. However, if well made, the graph permits physicists to recognize something about particles that couldn’t otherwise be seen—perhaps that some of these particles are dark matter. Since each representation is undeniably a simplification, truth cannot be seen as cohering in the correspondence of a given representation to the real world. Instead, “Truth-value circulates here like electricity through a wire, so long as this circuit is not interrupted” (Latour, 1999, p. 69). The “circuit” is the whole chain of representations, stretching back to the real. What matters are the connections between representations—connections that, like the wires in a circuit, allow truth-value to move between representations. If necessary, scientists can follow the circuit back to the real. Ultimately, this means that truth is an ongoing and active process—one that depends on an entire assemblage of different people, things, situations, and ideas. Science journalists as magnetologists This article posits that good science journalism can be seen as a continuation of this scientific chain of reference. While a piece of science journalism will always be a simplification of the complexity of those representations before it, simplification not only serves a functional role in helping make clear certain relationships and ideas, but can also preserve the key relationships between people, findings, and propositions to permit truth-value to circulate. When described in these terms, it becomes necessary to recognize the role that meaning plays both in “circulating reference” and in science journalism. Each representation in the chain is intentionally produced to reduce complexity and reveal certain relationships—ultimately, to help elicit certain meanings. Latour is far less clear about meaning than he is about truth in actor-networks. Perhaps the clearest discussion of meaning for Latour is through his concept of articulation. As for others, articulation for Latour extends beyond the linguistic sense of an enunciation. For Latour, articulation is broadly construed as the way in which propositions relate to each other. Drawing on Whitehead, Latour (1999) defines propositions “in the ontological sense of what an actor offers to other actors” (p. 309)—propositions can be discursive, material, human, or some combination of the three. Articulation is what replaces correspondence when we give up necessary ontological differences between language and the world, body and mind, or things and people. In a sense, meaning is the product, result, or content of articulation.2 Three things stand out in Latour’s articulation-based sense of meaning. First, articulation is heterogeneous. This is an idea that goes back to one of the founders of American Pragmatism, C. S. Peirce (1878), who associates meaning with semiosis, a signifying practice broader than linguistic semiotics. Second, articulation is active. It is no mistake that the term also refers to enunciation. Again, this is also the case for Peirce, for whom Floyd Merrell (1997) observes, “meaning is not in the signs, the things, or the head; it is in the processual rush of semiosis” a “translation” or “becoming” of signs (p. xi, p. xiv). Finally, articulation is multiple yet contingent. This is an idea that brings Latour in-line with Stuart Hall’s notion of articulation. In an interview, Hall observes about religion: [I]ts meaning—political and ideological—comes precisely from its position within a formation. It comes with what else it is articulated to. Since those articulations are not inevitable, not necessary, they can potentially be transformed, so that religion can be articulated in more than one way. (Grossberg, 1986, p. 54) Ultimately, if truth is something that runs like electricity through an assemblage of people, ideas, and things (Latour, 1999, p. 69), meaning can be seen like the magnetic field generated around that electrical circuit. This is not to say that meaning is super-structural or less real—indeed it is as real as a magnetic field. Yet meanings, as products of articulations of propositions, can mutate and change with each new relation. Meanings, therefore, are complex and shifting: overlapping, conflicting, and metamorphosing.3 In science, chains of references must articulate consistent meanings—leaving science as a project where truth and meaning often align. Latour (1999) observes of chains of references, “[W]hat a beautiful move, apparently sacrificing resemblance at each stage only to settle again on the same meaning, which remains intact through sets of rapid transformations” (p. 58). Yet, to extend this continuity between meaning and truth beyond science would be a mistake. Meaning, especially for nonscientist publics, is far more complex, fickle, and mutable. It is the job of science journalists, in producing public knowledge about science, to extend articulation beyond scientific truth and bring in disparate connections and possibilities. Understanding truth and meaning in this way casts journalists as magnetologists, who, balancing stasis and change, attempt to produce a representation of science that can fit into a larger system through which truth can circulate and around which new meanings can be generated. More concretely, this model of science journalists as magnetologists offers an account that, like Bucchi’s (1996, 2008) continuity model, recognizes relations between different forms of science communication. Yet, this model brings to the fore two balancing tendencies that structure the production of science journalism: stasis and change. On one hand, translations between formats, whether in the production of scientific results, articles, or pieces of science journalism, must maintain some connection to the real. As discussed in detail below, pieces of science journalism can accomplish this in a number of ways, yet this means ultimately preserving both traceable connections to antecedent representations and key relationships as other material is pared away. This maintenance or stasis is what allows a science news article to preserve its hold on the truth, to allow its “truth-value” to circulate back across the whole chain. At the same time, however, these connections mean that structural dynamics and pressures of science, public relations (PR), policy, as well as journalism itself constrain and enable the production of science news (Bauer & Bucchi, 2007; Gandy, 1980). On the other hand, translations are transformations: processes of change. In producing articles, journalists, like scientists, must strip away detail to reveal otherwise unseen connections and relationships. The reduction in technical complexity, like that which occurs in science itself, helps a journalistic article reveal otherwise occluded relationships. Finally, in producing an article, journalists add something as well. A piece of journalism can bring new ideas, details, and voices, introducing and revealing new relationships, understandings, and perspectives. All together, this balance of stasis and change, effected in the preservation, removal, and addition of content, not only preserves truth but also helps to produce new meaning possibilities for public science. Importantly, journalistic articles are rarely produced in isolation; not only do articles influence each other, journalists routinely swap stories, sources, and content (Boczkowski, 2010). Similarly, readers are increasingly encountering multiple articles about a single topic or finding (Su, Akin, Brossard, Scheufele, & Xenos, 2015). As a result, it is important to recognize that the meaning-field surrounding different articles can overlap and the complex interplay of constructive and destructive interference reshapes the meaning field further. Ultimately, the model adds four things to our understanding of science journalism. First, as Latour (1999) acknowledges for scientific research, this model describes science journalism as being composed of a series of translations between a succession of different forms and formats. Second, the model acknowledges that good journalism rests on maintaining those connections such that truth-values can circulate up and down the chain. Third, the model holds that each addition, subtraction, or translation modifies the meaning field generated across the circuit. Finally, the model asserts a normative position. Existing models focus on the way journalists reduce the complexity of science; at their best, journalists are seen to hold fact and truth constant. In contrast, this model understands journalists as also playing an important generative role in public knowledge and therefore democracy—not only bringing science to the public, but making it public (Latour & Weibel, 2005) by facilitating the articulation of diverse public meanings to science. Recognizing this endows science journalists with a democratic and normative responsibility to (re)produce accurate facts while simultaneously opening science to diverse publics and meanings. Case study: methods and background In order to better explicate and defend this model of science journalism epistemology, this article offers a detailed look into journalistic coverage of dark matter direct detection experiments. Since the early 1930s, astronomers have calculated that as much as 27% of the mass in the universe cannot be directly seen (NASA, 2017; Zwicky, 1933). One of the most prominent hypotheses holds that this dark matter is composed of hard-to-detect particles, descriptively called Weakly Interacting Massive Particles (WIMPS). Since the 1980s (see Ahlen et al., 1987) dozens of collaborations of physicists have built instruments to attempt to detect these particles. However, despite decades and millions of dollars, physicists have not yet seen convincing evidence of WIMPS in their detectors. Direct detection of dark matter experiments provide a strong case to study contemporary science journalism. Although these experiments are not the center of broad public debate, they have received consistent media attention over the past 30 years. Somewhat counter-intuitively, much of the recent scholarship on science journalism has addressed either coverage of research initiatives at the center of public debate, such as climate change (e.g., Boykoff & Boykoff, 2004), or coverage of major and successful scientific topics, like the human genome project (e.g., Hilgartner, 2012). Far less research has studied coverage of more “normal” (Kuhn, 2012) science. Direct detection of dark matter experiments can, however, provide a look into the ways that science that is not heavily politicized—the vast majority of scientific research—is covered by journalists. This case study is based on data collected for a larger project that explores changes in science production and communication in the contemporary media environment through a detailed analysis of direct detection of dark matter experiments. This article draws on 60 semi-structured interviews with journalists, physicists, and public information officers who have been involved with or covered dark matter research. After collecting a large corpus of journalistic articles about direct detection experiments (see below), articles were coded for basic information including publication, author, date, and main subject. Sources of direct quotations were also identified and coded in each article. After the data was consolidated, journalists and communication specialists who had written several articles were identified and then contacted. Interviews addressed both the day-to-day work of science journalism as well as the specific work of covering direct detection experiments. Informants were given a choice to be named or be provided with pseudonyms. Every informant cited below gave explicit permission to be identified by name. This article also employs a thematic textual analysis of 479 English-language news articles about direct detection experiments from August 1991 to July 2016. Rather than constructing a sample, this article attempts to collect, catalogue, and analyze every available article produced about these experiments through 2016. Stories were collected through searches of a variety of archives, including Lexus Nexus, Web, News Wire, and individual news organizations. Searches used the names of each collaboration along with more generic terms like “direct detection,” “dark matter,” or “weakly interacting massive particles.” Texts were also collected through a modified snowball approach. Every time an article from a new news site was identified, that site’s archives were searched for additional articles about other direct detection searches. Collaborations themselves also archived news articles on their websites. Articles derive from a range of publications, 113 in total, including the New York Times, Popular Science, Gizmodo, and Futurism.com. Texts were analyzed for recurrent themes, structural components, and approaches. Codes were generated both inductively, arising through immersion “in the texts and let[ing] the themes of analysis slowly emerge” (Brennen, 2017, p. 208), as well as deductively from the model offered above. Specifically, the model directed analysis to consider the ways that journalists modified content and meanings in producing articles. Overall, following Kracauer (1952), analysis focussed on both “the surface meanings and the underlying intentions of a text” in order to “bring out the entire range of potential meanings in texts” (Brennen, 2017, p. 205). The epistemology of dark matter journalism The model of science journalism epistemology introduced above asserts a balance of stasis and change: the way journalists, constrained by specific structures and pressures, maintain elements of antecedent representations to allow “truth-values” to circulate while also removing and adding content to reveal hard-to-see relations and produce new potential meanings. But what does this balance actually look like? How do journalists actually maintain a hold on truth, even as they introduce new elements? To understand better the way science journalists manage this dynamic of stasis and change in producing public knowledge about science, this section asks three questions of the data collected through interviews and textual analysis: What do science journalists preserve? What do science journalists remove? What do science journalists add? It is important to note that many actual journalistic practices could be placed in several of these categories. For example, in selecting article subjects, journalists both preserve topics selected by scientific experts4 as notable stories and also reject many others (e.g., Shoemaker, Vos, & Reese, 2009). That being said, these three questions help identify key elements of the epistemology of science journalism that might otherwise remain hidden. What do science journalists preserve? Even as they translate complex scientific content into something more widely understandable, journalistic articles should preserve key informational, personal, and material relationships. The model highlights two elements to this preservation: providing references to antecedent content and maintaining key scientific relationships. Writing about scientific practice, Latour recognizes that it is essential that scientists can trace representations along chains of references so they can know how exactly each translation has been produced. Arguably, the same is true for science journalism, where references to antecedent content should be, and often are, preserved across translations. This can be as simple as providing a reference or link to an existing scientific article, press release, or other piece of news content. When there is not simply a text that can be referred to, journalists find other ways of referencing source content. For example, in one piece, Dennis Overbye writes: The team, known as the Cryogenic Dark Matter Search, announced its results in a pair of simultaneous talks by Jodi Cooley from Southern Methodist University the SLAC National Accelerator Laboratory in California and by Lauren Hsu of the Fermi National Accelerator Laboratory in Illinois at Fermilab, and they say they plan to post a paper on the Internet. (2009) Overbye carefully—if somewhat awkwardly—includes each institutional affiliation, while also specifying where the results were already released, and where they will be released in the future. Second, simply put, journalists have to get the science right. Good journalism does this by preserving the general relationships, even while dropping much of the fine-grained detail. This is most clear in what science journalism textbooks call an “explainer” graph (Blum et al., 2005), a paragraph (or more) that provides the reader with some of the background necessary to make sense of a particular scientific result or event. For example: A dark matter particle striking a xenon nucleus causes it to recoil, prompting the emission of light and ionization. The ratio of the amount of light emitted to the amount of ionization indicates whether a particle of dark matter has been found” (Wired, 2011). While many of the specific technical details that a scientist would consider necessary to making these statements “true” are missing, nearly every phrase here elides a great deal of complex science. Just as an interested reader could trace back to an antecedent piece of science through references, here, she could trace from these statements to more technical descriptions. In this sense, while a scientist might rely on tacit knowledge (Polanyi, 1998) to fill in the gaps, here gaps are left as potential connections—virtually validated truth claims. Including direct quotations from knowledgeable sources is another important strategy that journalists employ to maintain continuity in reporting both science news and feature articles. As with other forms of journalism, quotations often serve as a key currency of articles. Usually journalists will reach out to lead authors of scientific articles or collaboration leaders, usually called “spokespersons.” Within the articles collected for this article, eight of the 10 most commonly quoted scientists in journalist articles were spokespersons for experiments. Adopting public relations strategies that are increasingly common in other scientific fields (Bauer & Bucchi, 2007; Gandy, 1980), dark matter collaborations have been working with press offices to produce press or media releases. These releases almost always include quotations from collaboration spokespersons or principal investigators. There are varying norms about using quotations included in press releases or institutional stories. While working at Space.com and writing short news articles, Clara Moskowitz (personal communication, August 15, 2016) remembered: I wouldn’t talk to anybody, it would be kind of a straight-forward story, so I’d read the press release, and I’d read the paper, and then I’d just use the quotes from the press release from the story and say: “so and so said in a statement,” so then, some times you did no reporting. In contrast, Davide Castelvecchi (personal communication, August 24, 2016) strongly rejected the implication that he would ever use institutionally supplied quotes. In maintaining these elements from scientific articles or results, journalists help to preserve the meanings around articles. In many ways, there remains a tight coupling between meaning and truth in scientific research—the meaning or relevance of a finding has a close connection to its scientific import, functionality, or success against “trials of strength” (Latour, 1993, p. 37). Direct detection experiments provide a telling example of this. Although they have failed to find dark matter, many physicists would argue that experiments have held great value and meaning in helping to incrementally narrow the range of possible dark matter candidates (R. Gaitskell, personal communication, September 22, 2016). In maintaining connections to and elements of antecedent representations or texts, journalists help to preserve these scientific meanings. For example, one news article from New Scientist begins, “One of the world’s leading dark matter detectors has wrapped up a nearly two-year-long search for the mysterious particles, without finding a single whiff. The results suggest that the days may be numbered for the dominant model of dark matter” (Aron, 2016, np). The rise and fall of technical models of dark matter are usually more important to scientists than to laypersons. Yet, this piece asserts that this finding is meaningful to its audience, the lay public, because of its scientific import. What do science journalists remove? As noted above, Latour recognizes that in scientific practice, the paring away of detail that happens across successive representations allows scientists to reveal hard-to-see relations. The same is true of science journalism where, by removing much of the technical detail, journalists can help readers better understand complex material and ideas. In interviews, journalists frequently referenced the work they put into making their articles “clear.” Adrian Cho (personal communication, March 3, 2016) observed, “I try my level best to really understand what’s going on and explain it as clearly as I can.” Speaking about her readers, Clara Moskowitz (personal communication, August 15, 2016) noted, “I think they appreciate a clear story that defines its terms and explains everything well.” As Peirce (1878) recognized long ago, clarity and meaning are inextricably linked. He famously observed that understanding “how to make our ideas clear” is “to know what we think, to be masters of our own meaning.” For science journalists, the quantity of technical detail is less important than clear and understandable explanations. Aside from writing simply, journalists suggested two additional approaches they employ to produce clear explanations. First, Clara Moskowitz observed: some things can only be understood by math, and so I run up against this problem, where you kinda just have to wave your hands and hint at things that are really only clear when you look at the equations. (personal communication, August 15, 2016) Moskowitz’s strategy of “waving your hands” to elide or bypass complexity, can be seen across the corpus of texts. For example, one article from the BBC’s online platform stated: In short, if you do the maths on the universe, something strange happens. No matter how many times you check the figures, the answer always comes out the same […] the Universe should weigh a lot more than it does. The best explanation scientists can come up with is that there is a lot of “stuff” in the universe that we can’t see or hear or touch, but which makes up for that extra weight. They call this stuff “dark matter” (BBC, 2010). “Waving your hands” isn’t about ignoring the detail; it’s more about being comfortable making large leaps over the complex math or detail through which scientists originally discovered or demonstrated connections. Mathew Francis suggested an alternative approach while describing writing an article about an “esoteric” paper on neutrino masses: [S]o I’m not going to get into the mathematical structures and grand unified theories and why this all matters. I’m just going to talk about what are the implication of this, why would people consider this, what’s cool about it. (personal communication, March 4, 2016) While Francis might similarly try to condense down the key relationships, he suggests it can be useful to focus on what is “cool” or most interesting about a story. Other journalists also adopt this approach. In one article from the L.A. Times, Amina Khan describes dark matter this way: “Dark matter outnumbers normal matter in the universe 5 to 1, yet remains one of physics’ ultimate mysteries. It can’t be seen or felt, and passes through Earth like a phantom” (Khan, 2013, np). Rather than getting bogged down in complex description of the science behind this fact, Khan picks out what is most compelling or interesting about dark matter. What do science journalists add? In addition to removing or simplifying content, science journalists also add detail and context to scientific stories in order to expand the scope and meaning of scientific research. To return to Latour’s (1999) material semiotics (see also Lenoir, 1994), journalists align textual signs, but also quotations, objects, actors, and ideas to produce fertile and complex ground for meanings. As discussed above, quotes from expert sources are key components of science journalism articles. While some journalists regularly incorporate quotes from statements or press materials, others refuse. Tushna Comissariat explained her hesitation: It just sounds so boring as compared to what researchers actually say to us, which is a lot more exciting. We recently had an actual quote in a news story where [a scientists] said they were so excited they punched the wall, I don’t think you’ll find that in the press release. (personal communication, August 31, 2016) For Comissariat the concern is less an ethical prohibition against canned statements, and more the worry that they do not add anything to a story. For her, quotes not only serve as a key form of evidence to maintain scientific integrity, they also add detail, color, and perspective to articles. These details help readers grasp fundamental ideas or relationships while also expanding the scope and meaning of scientific research. There are many ways that journalists add this sort of detail or perspective. Matthew Francis explained that he often tries to interview less senior collaborations members such as PhD students and postdoctoral researchers, who are rarely given voice in academic papers or institutional press releases, but who: […] are the ones who actually know how it [the experiment] works […] The people who are the spokespeople, part of their job is PR, they’re going to tell me things about how they’re thinking, they’ve always got one eye pointed at the funded agency. (personal communication, March 4, 2016) Another way journalists seek out new connections and relations for their stories is by securing quotes from scientists not associated with collaborations. For example, across the journalistic coverage of the LUX experiment’s 30 October 2013 release, journalists cited six different members of the LUX collaboration, and 13 physicists—more than twice as many—who were not members of LUX.5 Set within a journalistic article, these voices can help situate findings in larger fields or disciplinary contexts. Judging by interviews and collected articles, theorists are one of the more common types of outside sources.6 Peter Graham, a theorist at Stanford University, described serving as a source for journalists in a similar way to how he works with experimentalists: […] kind of putting it all together and trying to see kind of where each piece fits it, I would say that’s what theorists should be doing, what the use of a theorist is, also what I think is useful to a journalist in an article. (personal communication, August 23, 2016) There are a number of other strategies, beyond including outside voices, that journalists employ to add context to a current piece of research. Some articles situate direct detection experiments within the larger universe of scientific studies of dark matter (see Morelle, 2013). Other articles provide historical detail of our understanding of dark matter (e.g., The Seeker, 2011), or of the locations central to dark matter research. For example, a 2015 piece by the freelance writer and novelist Kent Meyers in Harper’s Magazine details the long history of the Homestake gold mine in South Dakota, the site of the Sanford Underground Research Faculty. In other instances, journalists highlight the philosophical, or even metaphysical aspects of dark matter research. One article in The Guardian quotes a Cambridge astronomer: “Dark matter is what created the structure of the universe and is essentially what holds it together […] Without it, we wouldn’t be here” (Sample, 2009). Finally, rather than going broad, some go deep—providing a behind-the-scenes look at the actual practice of science (Overbye, 2009; Wired, 2011). Conclusion Amid declining public trust in media and an outright attack on the credibility of mainstream news organizations, it is all the more important that we understand what is it that journalism does—how exactly it produces its unique form of knowledge. The model of science journalists as magnetologists developed here describes a set of practices composed of distinct tendencies: stasis and change, fact and meaning. Good pieces of science journalism must maintain connections to antecedent references to allow truth-values and facts to circulate through them, even while they expand the scope of possible meanings. In this account, these two tendencies work in concert: meanings are generated, in part, out of the same connections that allow truth to circulate, yet journalists also maintain and build truth-carrying circuits to produce meanings. Understanding the productive work of science journalism can help us celebrate rather than lament the differences between scientific and journalistic articles. Rather than providing another reason to distrust (science) journalism, we should see these differences as part of journalism’s power and promise. Unfortunately, some of the loudest critics of science media are scientists themselves, who often lack a complex understanding of or vocabulary for how science journalism works and what it is trying to accomplish. Part of the issue might rest with increasingly ubiquitous media trainings that teach scientists how to sell or promote their work, but say little about the value of science journalism. At the same time, we need science journalists to be better—and louder—public advocates for the work that they do. This article has argued that science journalism is, in many ways, distinct from other forms of journalism. Even still, there is reason to suspect this model may provide insight into our understanding of truth and meaning in journalism more broadly. There have been many models and accounts of journalism over the past century—many that highlight how journalism is much more than just a source of information (e.g., Deuze, 2005; Zelizer, 2004). Yet, despite its multivalent complexity, journalism remains a unique form of knowledge. As such, better understanding the epistemology of science journalism raises a number of key questions about how that form of knowledge works. How can we think of truth in journalism such that it is not left necessarily deficient against expert knowledge? What is it, exactly, that journalism as a particular type of knowledge produces? How does journalism cover other forms of knowledge? At the same time, in describing the unique and relational epistemology of science journalism, this article suggests the importance of understanding better the ways that changing structural dynamics of the contemporary media system, including technological, economic, and cultural changes, are affecting the epistemological foundations of science news. Equipped with a more rigorous account of science journalism epistemology, future work will address these questions. Democracy hinges on “making things public” (Latour & Weibel, 2005). There is always a cost in doing so: at a minimum, a sacrifice of technical complexity. Yet it is a price we pay because what we lose in technical detail, we gain in symbolic complexity. Bringing a story into the public—whether it is about science, politics, or a new TV show—opens it to a near-infinite reserve of perspectives, ideas, people, things, and, ultimately, meanings. Some might argue such an opening is a good in itself, others might see it more practically as a source of innovation (Benkler, 2006). This is the part that journalism must play in democracy: to hold strongly enough to its antecedents so that truth can circulate widely, but also to open the truth to new connections and relationships. Ultimately, we look to journalism to help build the infrastructures that give meaning to public life. Footnotes 1 For example: a graduate student looks out his window to see if his cat is in the backyard. He sees a round grey shape in the distance in a spot known to be the cat’s favorite place to sun herself. Given this visual evidence, he concludes that the cat is in the backyard. Yet, what he saw was not actually his cat, but a dirty tarp that had been blown onto the fence. However, the cat, was, in fact, sitting behind a tree in a different corner of the backyard. Thus, the graduate student had justified belief the cat was in his yard—owing to his usually trust worthy visual perception, and his statement was true. However, given that he mistook a tarp for his cat, this hardly can be seen as knowledge. 2 While Latour rejects scientific translations as metonymy (1999, p. 63), his account shares similarities with Jakobson’s account of meaning as occurring through the intersection of metaphor and metonymy if one accepts more expansive accounts of metonymy (e.g., Bredin, 1984). 3 Importantly, this metaphor overstates a causal determinism. As has been well documented and theorized, readers of journalistic texts will ultimately decode articles in complex and often unexpected ways. 4 Journal editors (Nelkin, 1995) and public relations professionals (Gandy, 1980) also play a notable role in story selection. 5 Forty-four stories concerning this release were identified. Of the six collaboration scientists quoted, five were also quoted in press materials (9 different releases). The sixth was a graduate student. 6 Direct detection collaborations do not usually include theorists. References Ahlen , S. P. , Avignone , F. T. , Brodzinski , R. L. , Drukier , A. K. , Gelmini , G. , & Spergel , D. N. ( 1987 ). Limits on cold dark matter candidates from an ultralow background germanium spectrometer . Physics Letters B , 195 ( 4 ), 603 – 608 . doi:10.1016/0370-2693(87)91581-4 . Google Scholar Crossref Search ADS Allan , S. ( 2011 ). Introduction: Science journalism in a digital age . Journalism: Theory, Practice & Criticism , 12 ( 7 ), 771 – 777 . doi:10.1177/1464884911412688 . Google Scholar Crossref Search ADS Aron , J. ( 2016 , July 21). Dark matter no-show puts favoured particles on death row. New Scientists. Retrieved from https://www.newscientist.com/article/2098217-dark-matter-no-show-puts-favoured-particles-on-death-row/ Bauer , M. , & Bucchi , M. (Eds.). ( 2007 ). Journalism, science and society: Science communication between news and public relations . New York : Routledge . BBC ( 2010 , January 29). “Dark matter at Boubly Potash mine?” BBC. Retrieved from http://news.bbc.co.uk/local/tees/hi/people_and_places/history/newsid_8488000/8488294.stm Benkler , Y. ( 2006 ). The wealth of networks: How social production transforms markets and freedom . New Haven, CT : Yale University Press . Blum , D. , Knudson , M. , & Henig , R. M. (Eds.). ( 2005 ). A field guide for science writers: The official guide of the National Association of Science Writers . New York : Oxford University Press . Boczkowski , P. ( 2010 ). News at work: Imitation in an age of information abundance . Chicago, IL : University of Chicago Press . Boczkowski , P. J. , & Anderson , C. W. ( 2017 ). Remaking the news: Essays on the future of journalism scholarship in the digital age . Cambridge, MA : MIT Press . Bok , S. ( 2011 ). Lying: Moral choice in public and private life . New York : Vintage . Bourdieu , P. ( 1999 9). On television . New York, NY : The New Press . Boykoff , M. T. , & Boykoff , J. M. ( 2004 ). Balance as bias: Global warming and the US prestige press . Global Environmental Change , 14 ( 2 ), 125 – 136 . doi:10.1016/j.gloenvcha.2003.10.001 . Google Scholar Crossref Search ADS Bredin , H. ( 1984 ). Metonymy . Poetics Today , 5 ( 1 ), 45 . doi:10.2307/1772425 . Google Scholar Crossref Search ADS Brennen , B. ( 2017 ). Qualitative research methods for media studies ( 2nd ed. ). New York : Routledge . Brossard , D. , & Lewenstein , B. ( 2010 ). Communicating science: New agendas in communication . New York : Routledge . Bucchi , M. ( 1996 ). When scientists turn to the public: Alternative routes in science communication . Public Understanding of Science , 5 ( 4 ), 375 – 394 . doi:10.1088/0963-6625/5/4/005 . Google Scholar Crossref Search ADS Bucchi , M. ( 2008 ). Of deficits, deviations and dialogues: Theories of public communication of science. In M. Bucchi & B. Trench (Eds.), Handbook of public communication of science and technology (pp. 57 – 76 ). New York : Routledge . Cloître , M. , & Shinn , T. ( 1985 ). Expository practice. In expository science: Forms and functions of popularization . New York : Springer . Creath , R. ( 1992 ). Induction and the Gettier problem . Philosophy and Phenomenological Research , 52 ( 2 ), 401 – 404 . doi:10.2307/2107943 . Google Scholar Crossref Search ADS Deuze , M. ( 2005 ). What is journalism? Professional identity and ideology of journalists reconsidered . Journalism: Theory, Practice & Criticism , 6 ( 4 ), 442 – 464 . doi:10.1177/1464884905056815 . Google Scholar Crossref Search ADS Dunwoody , S. ( 2008 ). Science journalism. In M. Bucchi & B. Trench (Eds.), Handbook of public communication of science and technology (pp. 15 – 26 ). New York : Routledge . Ettema , J. S. , & Glasser , T. L. ( 1984 , August). On the epistemology of investigative journalism. Presented at the AEJMC, Gainesville, FL. Fahy , D. , & Nisbet , M. C. ( 2011 ). The science journalist online: Shifting roles and emerging practices . Journalism: Theory, Practice & Criticism , 12 ( 7 ), 778 – 793 . doi:10.1177/1464884911412697 . Google Scholar Crossref Search ADS Fleck , L. ([ 1935 ] 1981). Genesis and development of a scientific fact . Chicago, IL : University of Chicago Press . Fuller , S. , & Collier , J. H. ( 2004 ). Philosophy, rhetoric, and the end of knowledge ( 2nd ed. ). Mahwah, NJ : Lawrence Erlbaum Associates . Gandy , O. ( 1980 ). Information in health: Subsidized news . Media, Culture & Society , 2 , 103 – 115 . Google Scholar Crossref Search ADS Gettier , E. L. ( 1963 ). Is justified true belief knowledge? Analysis , 23 ( 6 ), 121 – 123 . doi:10.1093/analys/23.6.121 . Google Scholar Crossref Search ADS Goldman , A. I. ( 2010 ). Why social epistemology is real epistemology. In A. Haddock , A. Millar & D. Pritchard (Eds.) Social epistemology (pp. 1 – 28 ). New York : Oxford University Press . doi:10.1093/acprof:oso/9780199577477.003.0001 . Goldstein , T. ( 2007 ). Journalism and truth: Strange bedfellows . Evanston, IL : Northwestern University Press . Grossberg , L. ( 1986 ). On postmodernism and articulation: An interview with Stuart Hall . Journal of Communication Inquiry , 10 ( 2 ), 45 – 60 . doi:10.1177/019685998601000204 . Google Scholar Crossref Search ADS Hearns-Branaman , J. O. ( 2016 ). Journalism and the philosophy of truth: Beyond objectivity and balance . New York : Routledge . Hilgartner , S. ( 2012 ). Staging high-visibility science: Media orientation in genome research. In S. Rödder , M. Franzen & P. Weingart (Eds.), The sciences’ media connection—Public communication and its repercussions (pp. 189 – 216 ). New York : Springer . Khan , A. ( 2013 , April 8). A hunt for dark matter in a former gold mine. L.A. Times. Retrieved from http://www.latimes.com/science/la-sci-dark-matter-detector-20130408-dto-htmlstory.html Kracauer , S. ( 1952 ). The challenge of qualitative content analysis . Public Opinion Quarterly , 16 ( 4 ), 631 – 642 . doi:10.1086/266427 . Google Scholar Crossref Search ADS Kuhn , T. S. ( 2012 ). The structure of scientific revolutions . Chicago, IL : University of Chicago Press . Latour , B. ( 1993 ). The pasteurization of France (A. Sheridan & J. Law, Trans.). Cambridge, MA : Harvard University Press . Latour , B. ( 1999 ). Pandora’s hope: Essays on the reality of science studies . Cambridge, MA : Harvard University Press . Latour , B. ( 2010 ). The Making of Law: An Ethnography of the Conseil d'Etat (M. Brilman & A. Pottage, Trans.). Malden, MA : Polity Press . Latour , B. , & Weibel , P. ( 2005 ). Making things public . Boston, MA : MIT Press . Lenoir , T. ( 1994 ). Was the last turn the right turn? The semiotic turn and AJ Greimas . Configurations , 2 ( 1 ), 119 – 136 . doi:10.1353.con.1994.0014 . Google Scholar Crossref Search ADS Maras , S. ( 2013 ). Objectivity in journalism . New York : John Wiley & Sons . Merrell , F. ( 1997 ). Peirce, signs, and meaning . Toronto, ON : University of Toronto Press . Morelle , R. ( 2013 , October 30). LUX results: Dark matter hunt nears final phase. BBC. Retrieved from http://www.bbc.com/news/science-environment-24733131 NASA ( 2017 ). “Dark energy, dark matter.” NASA Science. Retrieved from https://science.nasa.gov/astrophysics/focus-areas/what-is-dark-energy Nelkin , D. ( 1995 ). Selling science . New York : Freeman and Company . Overbye , D. ( 2009 , December 17). At a mine’s bottom, hints of dark matter. The New York Times. Retrieved from http://www.nytimes.com/2009/12/18/science/space/18dark.htm Park , R. E. ( 1940 ). News as a form of knowledge: A chapter in the sociology of knowledge . American Journal of Sociology , 45 ( 5 ), 669 – 686 . Google Scholar Crossref Search ADS Peirce , C. S. ( 1878 ). Deduction, induction, and hypothesis . Popular Science Monthly , 13 , 470 – 482 . Perrault , S. ( 2013 ). Communicating popular science: From deficit to democracy . New York : Springer . Pollock , J. L. , & Cruz , J. ( 1999 ). Contemporary theories of knowledge . New York : Rowman & Littlefield . Polanyi , M. ( 1998 ). Personal knowledge: Towards a post-critical philosophy . London : Psychology Press . Sample , I. ( 2009 , December 17). Has dark matter finally been detected? The Guardian. Retrieved from https://www.theguardian.com/science/2009/dec/17/dark-matter-detected Secko , D. M. , Amend , E. , & Friday , T. ( 2013 ). Four models of science journalism: A synthesis and practical assessment . Journalism Practice , 7 ( 1 ), 62 – 80 . doi:10.1080/17512786.2012.691351 . Google Scholar Crossref Search ADS Seeker ( 2011 , May 4). Dark matter debate heats up…again. Seeker. Retrieved from https://www.seeker.com/dark-matter-debate-heats-up-again-1765237840.html Shoemaker , P. J. , Vos , T. P. , & Reese , S. D. ( 2009 ). Journalists as gatekeepers. In K. Wahl-Jorgensen & T. Hanitzsch (Eds.), The handbook of journalism studies (pp. 73 – 87 ). New York : Routledge . Su , L. Y. F. , Akin , H. , Brossard , D. , Scheufele , D. A. , & Xenos , M. A. ( 2015 ). Science news consumption patterns and their implications for public understanding of science. Journalism & Mass Communication Quarterly , 92 ( 3 ), 597 – 616 . doi:10.1177/1077699015586415 . Crossref Search ADS Tanz , J. ( 2017 , February 14). From Fake News to disappearing ad revenue, journalism fights for survival. WIRED. Retrieved from https://www.wired.com/2017/02/journalism-fights-survival-post-truth-era/ Trench , B. ( 2007 ). How the Internet changed science journalism. In M. Bauer & M. Bucchi (Eds.), Journalism, science and society (pp. 133 – 142 ). New York : Routledge . United States v. Rumely , 345 U.S. 41 ( 1953 ). Vu , H. T. ( 2014 ). The online audience as gatekeeper: The influence of reader metrics on news editorial selection . Journalism: Theory, Practice & Criticism , 15 ( 8 ), 1094 – 1110 . doi:10.1177/1464884913504259 . Google Scholar Crossref Search ADS Williams , M. ( 2001 ). Problems of knowledge: A critical introduction to epistemology . New York : Oxford University Press . Wired ( 2011 , April 4). Underground experiment fails to find dark matter. Wired. Retrieved from http://www.wired.com/2011/04/xenon100-dark-matter/ Zelizer , B. ( 2004 ). Taking journalism seriously: news and the academy . Thousand Oaks, CA : SAGE Publications, Inc . Zwicky , F. ( 1933 ). The redshift of extragalactic nebulae . Helvetica Physica Acta , 6 , 110 . © The Author(s) 2018. Published by Oxford University Press on behalf of International Communication Association. All rights reserved. For permissions, please e-mail: email@example.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
Communication Theory – Oxford University Press
Published: Nov 1, 2018
It’s your single place to instantly
discover and read the research
that matters to you.
Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.
All for just $49/month
Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly
Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.
Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.
Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.
All the latest content is available, no embargo periods.
“Hi guys, I cannot tell you how much I love this resource. Incredible. I really believe you've hit the nail on the head with this site in regards to solving the research-purchase issue.”Daniel C.
“Whoa! It’s like Spotify but for academic articles.”@Phil_Robichaud
“I must say, @deepdyve is a fabulous solution to the independent researcher's problem of #access to #information.”@deepthiw
“My last article couldn't be possible without the platform @deepdyve that makes journal papers cheaper.”@JoseServera