In this webtext, I have experimented with various visualizations to explore how visual rhetoric might benefit from putting digital visualization techniques to work. Given my little prior experience in digital visualization and computational analysis, I am extremely grateful to Scott Weaver and Aaron Beveridge, who were able to listen and translate my data needs into visualizations that could not only advance my own research but also communicate the findings of that research. Such collaboration enabled us to produce visualizations that prove useful in enhancing visual rhetoric research and that are more sophisticated than visualizations I could ever have produced on my own. Yet, perhaps even more importantly, such collaboration enabled me to gain confidence throughout the project not only in analyzing and interpreting visual data but also in producing visualizations of my own, such as the actor-network map, which I produced via the data visualization platform Kumu. Altogether, then, this data visualization project proves to be useful in multiple ways. In the analysis portions of this webtext, I tried to touch upon many of the nuanced affordances digital visualization techniques have in relation to studying an image's circulation, genre diffusion and absorption, and collective activity. To conclude here, I discuss three important benefits with the hope that other visual rhetoric and circulation studies scholars will delve into their own data visualization projects to harness this research strategy's immense potential.
Before I delve into those benefits, however, I want to touch upon some of this experimental project's global limitations. As I articulated in the discussions about circulation, genre, and actor-network mapping, each digital visualization technique undertaken for this project has limitations. The dynamic spatiotemporal heat map fails to offer an image's exact circulatory path. The genre maps fail to account for all the nuanced genres in which Obama Hope surfaced. The actor-network maps fail to account for the full complexity of human and nonhuman actors involved in Obama Hope's collective activities. Together, they also risk misrepresenting Obama Hope's circulation, genre diffusion, and collective activities in that the data is limited to only 1000 pictures: this limited data set is especially obvious in that over 1000 Obamicons alone have been produced since 2008. In addition, and I think this is especially important to note, these digital visualization techniques are limited by my own lack of expertise in both coding visual data and data visualization. I am convinced that more experience in both these regards would elicit different findings, an acknowledgement that presses me to further educate myself in coding visual data and producing and interpreting data visualizations as well as to advocate for other visual rhetoric scholars to do so as well.
Digital visualization techniques are useful for deepening our data sets and complicating the rhetorical stories that we aim to tell.
Yet, despite such limitations, this webtext does illustrate that there is much to be gained by indeed pressing on to further explore what digital visualizations techniques can do for visual rhetorics and circulation studies. First, as I have tried to demonstrate and articulate, digital visualization techniques are useful for helping researchers become more intimate with their objects of study and extending data sets. Such intimacy is necessary for the design process. As Nathan Yau (2011) has argued, "you need to know your source material to tell good stories with data," because useful visualizations can only be designed if you "get to know your data and learn the context of the numbers" (p. 328). Additionally, becoming intimate with our objects of study can help deepen our research and ultimately our knowledge of an image's rhetorical impact, especially when a runaway object's story involves ongoing transnational circulation. While I had already spent six years tracking Obama Hope before embarking on this project, I discovered many previously unknown rhetorical experiences and genre activities during this research project. I was surprised to learn, for instance, just how many political and grassroots campaigns Obama Hope has participated in across the world. From fighting for gender equality in Singapore and Belgium to helping politicians in Spain win new votes to catalyzing support for both sides of the Israeli-Palestinian conflict, Obama Hope's rhetorical contributions simply get more intense with more in depth research. Setting the goal of coding 1000 pictures forced me to search for new data, new stories that could complicate my understanding of Obama Hope's ongoing rhetorical life. Digital visualization techniques are useful, then, for deepening our data sets and complicating the rhetorical stories that we aim to tell. In pointing to new places for qualitative research, of course, they can also help generate new insights and arguments.
Digital visualization techniques can help generate quantifiable evidence to substantiate new claims and either confirm or revise past claims based on qualitative research.
Second, digital visualization techniques can help generate quantifiable evidence to substantiate new claims and either confirm or revise past claims based on qualitative research. Too often scholars wait until after their research has been completed to use visualizations to present their research findings. While useful, such research limits the potential of digital visualization techniques, which if implemented during the research process can help triangulate our data and support our own rhetorical claims. Such use of digital visualization techniques is essential when trying to account for phenomenon such as visual art movements. As I learned in generating the actor-network map (Figure 6), qualitative research does not always uncover the interconnectedness of actants embroiled in collective activities. Actor-network maps can expose such intra-actions and thus provide more accurate and detailed accounts of the phenomenon we wish to expose.
Digital visualization techniques are especially essential to transnational or global visual studies. In one sense, digital visualization techniques facilitate a comparative approach to visual studies. For scholars interested in how various cultures interact with a single image and/or how images function differently across cultures, digital visualization techniques such as geographical and genre mapping prove especially useful in helping to identify both common trends and anomalies. In another sense, digital visualization techniques can help to substantiate claims about a runaway object's widespread circulation and participation in collective world-making. When we are trying to make claims about a single image's transnational impact, we ought to be able to verify our claims. When it comes to tracing the circulation of an image such as Obama Hope, for instance, it is one thing to say something has gone viral and made a global impact, and another to actually demonstrate this to be true. Dynamic visualizations with spatio-temporal dimensions coupled with actor-network maps can help map out this complex process. Such visualizations, as previously noted, may not accurately represent an image's exact circulatory paths, but they can nonetheless help show the speed of and widespread locations to which an image travels as well as the diverse multiplicity of collective activities in which it has participated.
Digital visualization techniques can boost data literacy.
Third, and lastly, digital visualization techniques can help boost our own data literacy. As many have noted, data literacy is becoming one of the most important critical skills to have in the 21st century (Harris, 2012; Carlson et al., 2011). While different stakeholders are still debating over how to help faculty and student develop data literacy, one vital skill we should make concerted efforts to cultivate regards data visualization. As visual rhetoricians, we certainly need to be able to read and analyze digital visualizations; as evident in any major news source being published today, data visualizations have become a mainstay in contemporary communication as means to inform, persuade, direct, and mobilize people. Visual communication scholars such as BanuInanc Uyan Dur (2014) and Helen Kennedy et al. (2016), in addition to writing and rhetoric scholars such as Aaron Beveridge (2016), are making strong headway into deepening our understanding of how visualizations are generated at the level of design, data processing, and statistical analysis. Such scholarship is especially important in that data literacy requires understanding not only the visual conventions (Kennedy et al., 2016) commonly at work in visualizations but also, as Beveridge (2016) notes, "understanding … the common exploratory methods that lead to polished infographics and effective data analytics."
When we take risks to be vulnerable with our scholarly production by engaging in distant reading and experimenting with digital visualization techniques, our confidence increases, and we become more likely to experiment and innovate.
Yet, especially in light of its ability to enhance our research process and communicate research findings, it would also behoove us to figure out ways to design and produce visualizations not only for our own scholarly sake but for our students' as well. This is especially the case as visual rhetoricians will not be excluded from the responsibility of "making sense of content at new orders of magnitude" (Losh, 2015, p. 286), a responsibility that comes with the ever increasing amount of pictorial data produced by an ever increasing amount of cultural producers. Such undertaking with larger data sets and computational analysis might make us uncomfortable in our scholarly shoes, as it demands developing new methods of reading, new literacies, and taking on new kinds of risks with our scholarship. Thanks to the work of Franco Moretti, Matthew Jockers, and others, distant reading has become a particularly useful practice, but the focus of such work also often often unduly associates its practice with literature studies and the digital humanities, making us hesitant to take it up for our own rhetorical study needs. But, testifying from my own experience with this project, when we take risks with our scholarship by engaging in distant reading and experimenting with digital visualization techniques, our confidence increases, and we become more likely to experiment and innovate. We also reap the rewards of diversifying our digital research practices. In thinking more deeply about how to cultivate data literacy and harness it for studies of visual rhetoric, then, we ought to be working harder to develop and hone digital visualization techniques. My hope is that the digital visualization techniques offered here, which include data collection, computational analysis, and data visualization, will act as a catalyst for developing the kinds of data literacies that can enhance our research practices.
I want to conclude by noting that as visual rhetoricians begin to increasingly find ourselves in a digital humanities situation (Alvarado, 2011), a situation in which we realize that digital technologies can play a crucial and transformative role in our interpretive process, we especially need to be proactive in producing digital visualization techniques that can help us address our most pressing research needs. In many regards, such proactive measures require intense ambition, especially when it comes to circulation studies. As Aaron Beveridge (forthcoming) has persuasively argued, as we continue to develop what he calls circulation analytics (a methodology for tracing the location, movement, and impact of digital artifacts), we must be willing to move beyond retrofitting already existing software toward designing software that is custom designed for our specific research needs. Software design such as PikTrack, which I briefly discussed in the introduction, is especially necessary for supplementing digital research methods such as iconographic tracking that we invent for ourselves but want others to adopt and adapt for their own research projects.
While we should be building teams that can collaborate on highly sophisticated digital visualization techniques, we can also start out simply and slowly, playing with computational analysis and visualizations in ways that are productive yet still within our reach as scholars-on-the-way-to-becoming-data-literate.
As this webtext demonstrates, while we should be building teams that can collaborate on highly sophisticated digital visualizations techniques, we can also start out simply and slowly, playing with computational analysis and visualizations in ways that are productive yet still within our reach as scholars-on-the-way-to-becoming-data-literate. Again, such experimentation may put many of us who lack extensive experience with computational analysis in a vulnerable place, as it requires entering into a process of digital experimentation with hope for success yet also with openness toward failure. But such experimentation may also not be as daunting as we surmise. The more difficult trick is to figure out where we want to venture as visual rhetoric begins to adjust to the "new orders of magnitude" relating to visual and media production and just how we want to contribute to the humanities at large in the face of our contemporary data deluge. My hope is that this webtext will point toward productive paths as we collaborate on large-scale digital research projects and learn together how to become more data literate.