What are media? What do they do? Are these two questions the same? And is the process we have in mind in asking them that of mediation or of mediality? In debating the issue, we came to realize that we are also dealing here with a translation problem. When a German and an Israeli speak in English about media, mediation, and mediality, it is only natural, given the different cultural and academic contexts, that there will be differences in understanding. Whereas in German the difference between mediation as Vermittlung and mediality as Medialität is quite clear, this hardly is the case in English. The Anglo media discourse often uses mediation and mediality interchangeably, and largely as related to the Aristotelian metaxy, a third that is in-between or in medio: [a]–[M]–[b]. But in German this refers to Vermittlung: the mediator (der Vermittler) by virtue of being in the middle mediates (vermittelt) something to be mediated (das Vermittelte). Importantly, the task of the mediator is tantamount to its ability to dissolve in the process, to bring things together while receding to the background.1 In this sense, media designate the conditions for mediation so as to put the relata in relation in the first place. Conversely, Medialität refers to the specific structure of a specific medium, i.e., its aesthetic obstinacies, technical operations, or semiotic significations. Thus, in English, discussion on Medialität actually takes place under “mediation,” which in German is normally translated as Vermittlung. However, at stake is more than conceptual inconsistencies. The translingual comparison suggests something essential to the understating of media.

Iconic photographs are symbolically dense images characterized by broad circulation over time and recognition by large publics. Following this definition, we track the republication and reframing, over nearly 70 years, of 15 news photographs previously identified as most recognized by the Israeli public. Distinguishing between “discrete icons” (singular photographs of particular scenes) and “aggregate icons” (where several variants of the same event are continually reproduced), our quantitative and qualitative analyses show that iconic images are both resistant to the passing of time and kept in motion through renewed media use. We identify four “iconic modalities,” corresponding to different ways in which iconic status and meanings are achieved, transformed, or denied through republishing and reframing. This concept improves our understanding of iconicity as a fluctuating material and symbolic process, whereby the circulation of images not only produces shifts in meaning, but constructs powerful aggregative frameworks of collective visual memory.

The task of topical segmentation is well studied, but previous work has mostly addressed it in the context of structured, well-defined segments, such as segmentation into paragraphs, chapters, or segmenting text that originated from multiple sources. We tackle the task of segmenting running (spoken) narratives, which poses hitherto unaddressed challenges. As a test case, we address Holocaust survivor testimonies, given in English. Other than the importance of studying these testimonies for Holocaust research, we argue that they provide an interesting test case for topical segmentation, due to their unstructured surface level, relative abundance (tens of thousands of such testimonies were collected), and the relatively confined domain that they cover. We hypothesize that boundary points between segments correspond to low mutual information between the sentences proceeding and following the boundary. Based on this hypothesis, we explore a range of algorithmic approaches to the task, building on previous work on segmentation that uses generative Bayesian modeling and state-of-the-art neural machinery. Compared to manually annotated references, we find that the developed approaches show considerable improvements over previous work.

Automatic extraction of narrative elements from text, combining narrative theories with computational models, has been receiving increasing attention over the last few years. Previous works have utilized the oral narrative theory by Labov and Waletzky to identify various narrative elements in personal stories texts. Instead, we direct our focus to informational texts, specifically news stories. We introduce NEAT (Narrative Elements AnnoTation) – a novel NLP task for detecting narrative elements in raw text. For this purpose, we designed a new multi-label narrative annotation scheme, better suited for informational text (e.g. news media), by adapting elements from the narrative theory of Labov and Waletzky (Complication and Resolution) and adding a new narrative element of our own (Success). We then used this scheme to annotate a new dataset of 2,209 sentences, compiled from 46 news articles from various category domains. We trained a number of supervised models in several different setups over the annotated dataset to identify the different narrative elements, achieving an average F1 score of up to 0.77. The results demonstrate the holistic nature of our annotation scheme as well as its robustness to domain category.

For voters, estimating beforehand which candidates will receive many or few votes in an upcoming election is valuable information. Probabilistic election forecasts help voters brace themselves for adverse outcomes or mobilize themselves to bring about favored ones, and constitute a critical prerequisite for strategic voting: By estimating which candidates stand a chance, voters can adjust their vote choices between multiple acceptable options, trying to maximize the impact of their vote. Especially in two-round voting systems such as the French presidential elections, tactical estimations of candidates’ chances are critical for making one’s vote count (Plutowski et al., 2020). While strategic voting has been widely studied (e.g., Meffert et al., 2011), we know little about how voters rely on the news and other sources of information to gauge candidates’ differential chances at receiving a sufficient share of votes. In this study, we draw upon a four-wave panel survey coupled with a large-scale news content analysis to examine how French voters in the 2022 presidential elections predict candidates’ respective chances and adjust them over time to accommodate new developments. Using a longitudinal linkage-study design, which links voters’ expectations to the media contents that they are exposed to, we distinguish two main mechanisms that may explain voters’ differential probability estimates (Blais & Bodet, 2006): On the one hand, ongoing news coverage informs voters’ expectations, as journalists, pollsters, pundits and other commentators give visibility to those candidates deemed most relevant and expressly discuss their respective chances. On the other hand, voters have relatively stable, intrinsic reasons for believing in the viability of candidates’ bids based on their political party preference, para-social sympathies and other forms of motivated reasoning. Expanding the ongoing scholarly debate about the electoral effects of public opinion polling (e.g., Daoust et al., 2020), we investigate how both wishful thinking and different forms of current information inform the formation and updating of voters’ probabilistic estimates, eroding or reinforcing confidence in their intended vote choice.

Citizens turn increasingly to social media to get their political information. However, it is currently unclear whether using these platforms actually makes them more politically knowledgeable. While some researchers claim that social media play a critical role in the learning of political information within the modern media environment, others posit that the great potential for learning about politics on social media is rarely fulfilled. The current study tests which of these conflicting theoretical claims is supported by the existing empirical literature. A preregistered meta-analysis of 76 studies (N = 442,136) reveals no evidence of any political learning on social media in observational studies, and statistically significant but substantively small increases in knowledge in experiments. These small-to-nonexistent knowledge gains are observed across social media platforms, types of knowledge, countries, and periods. Our findings suggest that the contribution of social media toward a more politically informed citizenry is minimal.

Modern democracies are built around the idea of representation: the electorate is represented, and perhaps more importantly, should feel represented. Yet recent decades have been marked by a growing gap between citizens and the political system. The emergence of new, more direct communication channels such as social media gave rise to expectations that the two sides can be brought closer together. This study investigates whether a connection with politicians and parties via social media platforms can enhance citizens’ feelings of being represented by a certain politician or party (dyadic representation), or by the entire parliament (collective representation). Our data come from a two-year panel study, covering four Israeli elections. Findings demonstrate that passive (receiving political information), but not active (producing political information) connection increases dyadic representation perceptions. Collective representation perceptions, by contrast, were not affected by either type of connection. Furthermore, our analysis does not demonstrate a reverse causal order; we did not find evidence supporting the idea that representation perceptions influence connection with politicians or parties. The upshot of our findings is that unidirectional communication between politicians and citizens on social media can foster citizens’ feelings of being represented in the political arena and of closeness with the political elites. That said, such effects were obtained only with regard to dyadic representation and are therefore limited in scope. The study contributes to the ongoing theoretical debate concerning the complex relations between new technologies, citizens, and political elites.

This paper takes recent PTSD claims by content moderators working for Microsoft and Google as a starting point to discuss the changing nature of trauma in the context of social media and algorithmic culture. Placing these claims in the longer history of how media came to be regarded by clinicians as potentially traumatic, it considers content moderation as a form of immaterial labor, which brings the possibility to be traumatized into the cycle of digital labor. Therefore, to the extent that content moderators’ trauma exists as a clinical condition, it cannot be taken as an incidental side-effect but as a built-in potentiality. It is about the commodification of traumatic vulnerability itself. The discussion then proceeds to speculate about the possibility of using algorithms to identify potentially traumatic content and what would that mean for the understanding of trauma, especially as a mediated experience.

Associations between trauma and media theory are longstanding, going back at least to Walter Benjamin’s observations on technology and modernity, which were themselves informed by Freud’s 1920 speculations on war trauma following WWI. A century later, and in the wake of numerous conflicts, catastrophes, and far-reaching technological transformations—and of course the COVID pandemic—it is time to reconsider the relation between trauma and media, digital platforms in particular. While some significant scholarship has noted the intersections of modern media technologies such as photography, film, radio, television, and recently digital and algorithmic media, with the conception and experience of trauma, a more systematic theoretical consideration of the relation between media and trauma remains to be developed. And with the intensifying reliance on new and old media in these pandemic times the question of these relations is increasingly urgent. Moving beyond conceptions of media as representing or inducing trauma, this special section of Crosscurrents explores how (digital) media and trauma shape one another.

Focus group methods specialize in the analysis of interactive discourse, but are only rarely employed as a stand-alone method to study such phenomena, owing to inherent limitations concerning the comparability and generalizability of findings. In this paper, we argue that focus groups undergo three kinds of transformations, involving changes in participants’ cognitive states, social ties, and discursive behavior, which raise both analytic challenges and valuable opportunities for the study of shared meanings and interactive negotiation processes in society. Introducing Serial Focus Groups, we extend familiar focus group designs as a method for studying interactive discourse in a longitudinal perspective, capitalizing on the analytic potentials raised by these transformations. Reviewing the methodological literature and drawing upon two large-scale focus group studies of socially interactive sense-making, we argue that serial focus groups can help overcome some of the limitations of cross-sectional focus groups and offer valuable new opportunities for analysis and validation.