|

Caleb Elfenbein, Farah Omer and Julia Schafer: Mapping Anti-Muslim Hostility and its Effects

Introduction: Confessions of a Neophyte

1 Leave a comment on paragraph 1 0 In September of 2015, I (Caleb) was teaching my introduction to Islam course when the story of Ahmed Mohamed became national news. Mohamed, then 14, was arrested for allegedly bringing a hoax bomb to school. It turns out that it was a digital clock that he had built for a school assignment. Nonetheless, Mohamed was led out of school in handcuffs, a hauntingly terrified look on his face. I remember teaching that morning. These are the conditions in which we are studying Islam in the contemporary United States, I explained as I projected an image of officers leading him out of the principal’s office in handcuffs. Suspicion of Muslims is sufficient to lead to the arrest of a 14-year-old student fulfilling an assignment. I did not know it at the time, but in retrospect, this was the moment that what became Mapping Islamophobia began to percolate in my mind.

2 Leave a comment on paragraph 2 0 Just two months after Mohamed’s arrest, in the context of an increasingly incendiary, anti-Muslim political environment, coordinated November attacks in Paris appear to have unleashed a remarkable wave of anti-Muslim hostility across the United States. While obviously troubling, this in itself was not unprecedented. The attacks of September 11 led to a spectacular rise in anti-Muslim activity, including hate crimes and government sweeps and detentions. Yet political discourses at the time were very different than those circulating in 2015. While government surveillance programs continued well beyond the months immediately following September 11, rates of hate crimes and open harassment of Muslims abated considerably in the space of a few months (even if remaining higher than pre-September 11 levels).

3 Leave a comment on paragraph 3 0 In contrast, in 2015, openly anti-Muslim political discourse appears to have created conditions in which public anti-Muslim hostility thrived, resulting in significant increases in rates of hate crimes and other anti-Muslim activity. It is in the context of these conditions that I happened to attend a workshop in postcolonial digital humanities, which, quite unexpectedly, transformed the way that I think about research and the role that scholars play in public life.

4 Leave a comment on paragraph 4 0 I must admit that prior to this workshop I did not have a particularly good sense of the remarkable diversity of work that falls within the category of digital humanities. In fact, I was attending the workshop largely because I was the incoming director of Grinnell College’s Center for the Humanities and I thought that I ought to develop a better understanding of this dynamic, growing field of work. It just so happens that Roopika Rasim, who led the workshop, is a leading voice in postcolonial digital humanities. This subfield of digital humanities emphasizes the social justice possibilities of wedding technology and humanistic inquiry. Among the project examples Dr. Rasim presented was Mapping Police Violence.[1] I am not exaggerating when I say that encountering this website changed the trajectory of my career.

5 Leave a comment on paragraph 5 0 The rise of Black Lives Matter beginning in 2014 and increasing media coverage had made me more aware of police violence against African Americans, but seeing the website, developed and maintained by a research collective that includes Samuel Sinyangwe, Deray McKesson, and Brittany Packnett, gave me an entirely different understanding of this civil rights travesty. The website’s landing page presents an interactive animated map with a pin for every death at the hands of law enforcement. A flash highlights individual data points as the map moves through time. Seeing these flashes accumulate over time had an effect on me. It communicated the horror of police violence in a way that news stories alone could not. I have since come to appreciate my response to seeing this website as developing affective understanding—an emotional response to a phenomenon that exists in fundamental relation to our intellectual understanding of that phenomenon.

6 Leave a comment on paragraph 6 0 Presenting data in a way that elicits an emotional response is one way that Mapping Police Violence is humanistic in nature. We cannot reduce this kind of engagement with data as mere “tugging at the heart strings,” which suggests a kind of duplicitous tactic. As Sara Ahmed argues in The Politics of Human Emotion, emotion is a central component of human sociality, a crucial aspect of how we experience and engage the world around us. Deliberately working at an emotional register draws attention to the limits of rational understanding in capturing the human experience. In the case of Mapping Police Violence, seeing the data points flash across the screen gave me an opportunity to appreciate the cumulative weight of incident after incident of police violence against African Americans in a way that statistics never had.

7 Leave a comment on paragraph 7 0 A second humanistic element of Mapping Police Violence relates to data accessibility. Each data point on the map is clickable, opening a pop-up that contains information about the person behind the pin. These stories underscore that the weight of police violence against African Americans, which the animated map communicates so effectively, accumulates in and through individual lives. For those who are not directly affected by police violence against African Americans, these two humanistic elements of Mapping Police Violence have the potential to create empathy of a kind that other methods of presenting and engaging data might not.

8 Leave a comment on paragraph 8 0 Mapping Police Violence had a profound effect on me. In addition to helping me understand police violence against African Americans in an entirely new way, I knew right away that the site could serve as a model for a similar project on Islamophobia and anti-Muslim hostility. The resulting project, Mapping Islamophobia, has been the most rewarding work I have done as a scholar.

9 Leave a comment on paragraph 9 0 Mapping Islamophobia draws together a variety of my interests and goals as a scholar. Since I began graduate school, I have tried to imagine how I could produce scholarship meant for a broad audience. As a professor, I see the immense pedagogical value of mapping and data visualization in helping students begin to engage information in creative and accessible ways. As a scholar of Islamic studies, I have long felt a responsibility to Muslim communities. I am well aware that my career is made possible by the lives of those for whom Islam has been and is important in some way. As a scholar of religion, I am fascinated by the conditions in which societies engage in debates and negotiations about the values and goals animating community life, or what I have come to call the conditions of public life.

10 Leave a comment on paragraph 10 0 Of course, I am able to see how the project draws these things together only after having worked on it for almost two years. When I began Mapping Islamophobia, all I had was an idea for a map and a sense that this could be a good way to communicate to a broader public audience. I had no idea how to create a map like the one I imagined. The workshop motivated me to approach Grinnell’s curricular technology specialist at the time, Mike Conner, explaining what I hoped I could do. He introduced me to Carto, a very user-friendly web-based mapping platform.[2] He showed me how to create a dataset that would provide Carto with the information it needed to do what I had in mind and helped me begin to think about basic design questions. It was so much more straightforward than I had imagined. This was an empowering discovery. From there I began to think about how to collect data on anti-Muslim hostility. I worked with a research librarian, Dr. Phil Jones, to identify the ideal methods of harvesting data from media sources and experimented with different search strings until I began getting good results. After a brief initial period collecting data, I turned to data coding. What patterns was I seeing? How could I translate those patterns into categories? I had also begun collaborating with student researchers, bringing my first student aboard within a few months of getting the project off the ground.

11 Leave a comment on paragraph 11 0 By this time, the process of data collection and entry was in place, making it possible to move beyond a few initial entries in creating the first dataset on anti-Muslim hostility. Although we have refined things over time, the basic process has remained more or less the same. Each row of data begins with a news article, typically gathered through news alerts, from which we derive information about the location of an incident, the people involved, the nature of the incident, and relevant details about what happened. We then use Google maps to get geolocation information (down to a building or intersection whenever possible). At times we have to search for and consult sources beyond our initial article, but it is surprising how much you can pull from one article. Once we have completed the entry we file the article so that we have a record of all sources. Periodically, we upload an updated dataset to Carto, which in turn updates the maps on Mapping Islamophobia.

12 Leave a comment on paragraph 12 0 It is a pretty straightforward though time consuming process—each entry takes about fifteen to twenty minutes, sometimes more when it is necessary to draw on additional resources to identify a precise date or location. With well over 3000 rows of data as of this writing, you can see why it is necessary to have many people working on the project at once. Even with multiple people it can be difficult to keep up with current developments while also collecting data on past years.

13 Leave a comment on paragraph 13 0 There is a pattern in this brief account of how I began Mapping Islamophobia: This project has been a collaboration from the start. Digital humanities projects require bringing together people with different skillsets to solve a problem. With Mapping Islamophobia, the problem was how to represent and communicate in accessible fashion a social justice problem I saw unfolding before my eyes. I was able to go live with the project within the space of a few months in large part because of the group of people who helped me think through how to solve this problem.

14 Leave a comment on paragraph 14 0 Figure 1:The project’s first map, which uses animation to show changes in anti-Muslim hostility over time.

15 Leave a comment on paragraph 15 0 From its very first map (Figure 1), which uses Carto’s animation function to show incidences of anti-Muslim hostility as they unfolded over time, the project has grown to include over a dozen maps drawing on two distinct datasets. Most of these maps are interactive. Users can click on each point on the map to learn more about the stories behind the data.

16 Leave a comment on paragraph 16 0 Mapping Islamophobia has continued to be a fundamentally collaborative project beyond the creation of this first map. The addition of a second dataset, which tracks American Muslim participation in public life through various kinds of community outreach and political activity, emerged out of a question that arose in conversation with the very first student collaborator, Chloe Briney. Early on in the initial stages of data collection, she asked what we should do with the “good stories,” examples of cooperation between Muslim and non-Muslim communities in the United States. Having started the project to highlight the place of anti-Muslim hostility in the conditions of public life for American Muslims, I did not have a good answer to her excellent question. It took months of thinking—and further questions about where the “good stories” fit into my work from audience members in early public presentation of the project’s findings—to formulate an answer.

17 Leave a comment on paragraph 17 0 These “good stories” now sit beside incidences of anti-Muslim hostility to create a picture of the complexity of public life for American Muslims. Visual representations on Mapping Islamophobia present this picture in an accessible fashion. Users can see trends in anti-Muslim hostility as well as in American Muslim outreach efforts. Having these maps side by side on the site provides an opportunity to see that we cannot understand the heartwarming stories of cooperation and dialogue without reference to the place of anti-Muslim hostility in the conditions of public life for American Muslims.

18 Leave a comment on paragraph 18 0 Mapping Islamophobia is, at base, an educational resource. The site provides users with an opportunity to explore anti-Muslim hostility and American Muslim participation in public life, presenting information in a way that allows them to draw their own conclusions. This is asking a lot of users and may in some ways limit our audience. Nonetheless, we hope this decision leads to quality of engagement with the data for those who are willing to put in the time to exploring Mapping Islamophobia, including teachers and students in secondary school and higher education.

19 Leave a comment on paragraph 19 0 As the project has continued to unfold and grow over time, student collaborators have taken on an increasingly significant role. My co-authors, Julia Schafer and Farah Omer, have each taken primary responsibility for one of the two datasets, allowing me to take a broader view of the project and to develop related scholarly work that analyzes the significant data we have collected. Along the way, they have been essential conversation partners, exploring the implications of decisions we make about how to code data, what data to include and why, and the findings about the conditions of public life for American Muslims that present themselves in the course of our work. In addition to opening space for their immense intellectual contributions to the project, thinking of student contributors as collaborators ensures that Mapping Islamophobia meets or exceeds best practices in digital humanities work.

20 Leave a comment on paragraph 20 0 Article 1 of the Student Collaborators’ Bill of Rights, created by staff and affiliates of the UCLA Center for Digital Humanities, states, “As a general principle, a student must be paid for his or her time if he or she is not empowered to make critical decisions about the intellectual design of a project or a portion of a project (and credited accordingly). Students should not perform mechanical labor, such as data-entry or scanning, without pay.”[3] I have been fortunate enough to have access to resources to pay all of my student contributors an hourly wage, using a combination of my institutional research funds and resources available through Grinnell’s recent Mellon Foundation digital humanities grant. This is especially important for those who are somewhat less involved than Julia and Farah in big-picture decision-making conversations. Still, whatever the nature of their contributions, being able to compensate all student collaborators has helped make Mapping Islamophobia an educational and professional experience for them.

21 Leave a comment on paragraph 21 0 In what follows, Julia and Farah will present their own experiences with Mapping Islamophobia, offering important insight into student collaborator experience in the development of a digital humanities project. They present their contributions and experiences in the framework of their respective datasets. In so doing, they will provide additional detail about the project and its ongoing work.

Reflections of a Student Collaborator: Julia Schafer

22 Leave a comment on paragraph 22 0 Our first dataset records instances of anti-Muslim hostility across the United States. Anti-Muslim hostility takes a multitude of forms in public life, from anti-Islam graffiti on mosques, anti-Muslim protests, and public campaigns targeting Muslims to local, state, and national political discourse and state-level anti-shari‘a legislation. A critical assumption of our project is that anti-Muslim hostility creates conditions that are harmful to free and voluntary participation in public life. We use a series of interactive maps to visualize the ways in which different expressions of public hate manifest, both geographically and temporally, in the lives of Muslim American communities and across the United States. I have been intimately involved in creating this and managing this dataset. In the process, I have come to understand the value and role of digital humanities in creating scholarship for the public good, some key facets of what goes into creating work meant for a public audience, and the terrain of collaboration in digital humanities work.

23 Leave a comment on paragraph 23 0 As Caleb noted, our goal is to provide data in a way that empowers viewers to develop their own understanding of anti-Muslim hostility in the United States. Toward this end we have carefully curated information about the nature of the events that anti-Muslim hostility inspires, made this data available to users in an accessible way, and made a deliberate decision to refrain from providing in-depth analysis of our data.

24 Leave a comment on paragraph 24 0 A core value of Mapping Islamophobia is public accessibility. The desire to make data more legible to broader publics has presented us as scholars with an opportunity to consider what motivates publicly-facing work. For example, as a student contributor, this was the first time I thought about why and how it might be useful to purposefully omit explicit analysis in scholarly work. We certainly have our own analysis that reflects a deep engagement with the thousands of rows of data we have collected, but we have decided to create a resources that shows data in as accessible a form as possible rather than telling people what the data means. In this sense, Mapping Islamophobia is deeply informed by pedagogical considerations. We want to make sure that students, and others, have the chance to work with the data and come to their own conclusions. These decisions about legibility and access were central to our mission of creating a digital humanities project that contributes to public education for social good.

25 Leave a comment on paragraph 25 0 To facilitate users’ direct engagement with data, we created a careful system of collecting, organizing, and coding the information we include in our datasets. Each point of data is recorded and curated to provide the location, date, and nature of an event driven by hostility. Our categories were developed through a series of collaborative discussions over time as our team began collecting and parsing data. The fact of deciding how to code an event is its own analysis, requiring in-depth discussions to make sense of the stories we have found and continue to find through our research process. As contributors we have each been able to contribute to the key questions of what data to include, why we should include it, and what implications its inclusion and presentation will have for how users understand the conditions of public life for American Muslims.

26 Leave a comment on paragraph 26 0 Our map presenting anti-Muslim hostility data by the gender of those affected is illustrative of the implicit analysis in much of our decision-making.[4] When we created this map, which we display below, we noticed that the green dots representing incidents affecting Muslim women were obscured by the other data points depicting incidents that affected Muslim men or Muslims more generally. We felt that this reproduced what we had learned anecdotally—that Muslim women, especially Muslim women wearing head covering of some kind, seem to be less likely to report everyday experiences of anti-Muslim hostility. We wanted our map to counter this trend, and pulling that layer of data forward on the map is one subtle way of doing so. Although we don’t talk about that decision on the website, it is an example of how choices we make behind the scenes are in themselves analytical work. Having said that, however, we have made every effort to be as transparent as possible in all that we do.

27 Leave a comment on paragraph 27 0 In fact, one of the first decisions made about the project was to be completely transparent in how we collect and record data. Our entire data set is published online and available to the public. We have also provided direct access to every source used to compile the data in the form of an attached URL. This serves two purposes. First, a key value of publicly-facing work is transparency. If we are asking our users to engage our data, we need to be clear about how we have put together the information before them. Second, given some public skepticism around the extent of anti-Muslim hostility in the United States we wanted our data to be as close to beyond question as possible.

28 Leave a comment on paragraph 28 0 Another way we prioritize transparency is through careful recordkeeping. We have documented the major—and a good number of seemingly minor—decisions we have made in the process of developing this project. If anyone has questions about decisions we have made or specific elements of the project, we can answer them in fairly detailed fashion. This has influenced me to be very intentional as I work on the dataset. As I make decisions and keep records of my search process I am constantly aware of the fact that my actions are public. I have come to see transparency as one element of creating a project with deep integrity. Further elements of creating a project with integrity is care in the data collection and coding process.

29 Leave a comment on paragraph 29 0 To ensure the integrity of the data behind our project, we use data sourced from media outlets with clear editorial oversight.[5] Additionally, when creating entries, we privilege articles that emerge out of reports to civil rights organizations and/or law enforcement agencies. Our team uses online databases to access archives of news media sources that report on instances anti-Muslim hostility. We chose to restrict our data to these types of sources because they provide the most transparency and authority in reporting on cases of hostility. The project is ongoing, and through individual work on the data set we both maintain up-to-date reporting on the current year and work backwards through time. The work of the project is distributed by calendar year of analysis across multiple research team members, but decisions regarding visualization, coding, and language are made as a team.

30 Leave a comment on paragraph 30 0 Coding the data with consistency is another important element of maintaining the integrity of the project. Compiling and presenting data is essential to the interactive nature of the site, but ensuring that the data is well- and consistently-organized requires a challenging analysis and decision-making process for the researcher. The majority of our collaborative group meetings are dedicated to discussion about questions of coding events and the use of language in our project. In this process both students and faculty have worked closely together to make substantial decisions regarding the (implicit) analysis entailed in coding data and writing descriptions. These discussions are an ongoing feature of the project as new and challenging questions and problems arise from the research process.

31 Leave a comment on paragraph 31 0 In order to fully represent the many expressions of anti-Muslim hostility, we decided to code each event as either a crime against a person, a crime against a place, a bias-related incident, legislation, public speech by a political figure, and public campaigns.[6] We also record the gender of those affected by an incident, the URL or bibliographic record of the media news source, the exact location (using latitude and longitude), and a brief, carefully crafted description of the event. All of this information makes its way directly into how we present our data, from the range of maps on anti-Muslim hostility on our website to the pop-up windows that provide information on the stories behind the data. When developing the descriptions that appear in these windows, we pay careful attention to writing formally and objectively, specifically refraining from any editorial comment. Our aim is to report the data clearly and factually and to promote the legibility and clarity of our project.

32 Leave a comment on paragraph 32 0 Figure 2: Users can select one year or multiple years of data.

33 Leave a comment on paragraph 33 0 Figure 3: Users can sort the data by the gender of those affected. Many incidents, including vandalism, political discourse, and legislation, affect the entire community and so we code it accordingly.

34 Leave a comment on paragraph 34 0 Figure 4: Users can sort data by type of activity. The layers correspond to the codes we assign each data point. In this map, we display an example of a pop-up window that provides access to stories behind data points, an essential part of the project.

35 Leave a comment on paragraph 35 0 Beyond presenting the data in accessible fashion, there are a number of steps we take to support legibility through quality control, or maintaining uniform standards in coding and writing. As an editor of the anti-Muslim hostility dataset, I have worked to maintain an unbiased and formal writing style as a means of ensuring continuity as other student contributors rotate on and off the project. Different writing and working styles are a reality of collaboration, making the role of editor absolutely essential in a project such as ours. We want our users to engage data as seamlessly as possible. Having as close to a uniform voice as is feasible is an important way to work toward this goal.

Reflections of a Student Collaborator: Farah Omer

36 Leave a comment on paragraph 36 0 Our second dataset documents outreach efforts that Muslim Americans do to humanize themselves in the face of anti-Muslim hostility. These humanizing efforts take a multitude of forms, such as mosque open houses, interfaith dialogue, and “Meet a Muslim” and “Ask a Muslim Anything” events. Islamic organizations, local mosques, families, and college students organize them. This dataset serves a critical role in the Mapping Islamophobia project because it illustrates effects of public hate on Muslim American lives, particularly the form and the extent of their participation in public life. I (Farah) have been primarily focused on creating and managing this dataset. In the process, I have come to understand that deciding on the parameters of our datasets, coding the data that drives Mapping Islamophobia, and making decisions about data visualization are themselves forms of analysis, showing that digital humanities work, even when it does not include much explicit analysis, as with our site, is a scholarly endeavor. Moreover, as with Julia and Caleb, I have come to see the unique possibilities that the field of digital humanities provides scholars who want to create publicly-facing scholarship that addresses pressing social justice issues.

37 Leave a comment on paragraph 37 0 Prior to working on this project, I did not fully understand the collective burden on Muslim Americans to constantly distance themselves from extremist ideologies and terrorist attacks. Of course, we have all become well acquainted with the script that follows each terrorist attack. Muslim religious leaders, private citizens, and Muslim advocacy organizations alike publicly denounce the attacks, calling them un-Islamic. Conservative media outlets and politicians and anti-Muslim activists typically ignore these statements, publicly demanding that Muslims prove that Islam does not condone such violence and extremist ideologies. Collecting data on the many ways that Muslims are reaching out to their non-Muslim neighbors across the country has shown us how hollow these calls really are, and our Mapping Islamophobia data visualizations present this evidence in an extremely accessible and compelling way.

38 Leave a comment on paragraph 38 0 Studying the outreach efforts of American Muslim communities in the context of anti-Muslim hostility provides incredible insight into the effects of public hate on the ability of American Muslims to partake in a core value of democratic life in the United States: the ability to freely and voluntarily participate in public life and discourse. Although, as Julia mentioned, we largely refrain from explicit analysis on our website, the user can observe these relationships in the maps. For instance, data visualization allows us to see the correlation between spikes in anti-Muslim hostility and increases in the humanizing outreach efforts of American Muslim communities. Our site explores such correlation and connection in a variety of ways, which Figures 5 and 6 demonstrate.

39 Leave a comment on paragraph 39 0 Figure 6: Users can engage data about anti-Muslim hostility and American Muslim humanizing efforts on the same map. Mapping Islamophobia includes a series of maps presenting data from both datasets by year. As with other maps on the site, users can access information about the stories behind the data.

40 Leave a comment on paragraph 40 0 Mapping Islamophobia aims to shed light on the incredible pressure on Muslim Americans to exonerate themselves from public suspicion and uses digital tools to make this pressure legible and accessible to a wider audience. Analyzing the data shows that there is a definite correlation between anti-Muslim hostility and Muslim public outreach and engagement, in terms of both when we see ebbs and flows in outreach work and what people doing the work have to say about the impetus behind it.

41 Leave a comment on paragraph 41 0 By drawing attention to the relationship between anti-Muslim activity and Muslim public outreach and engagement we attempt to bridge the divide between scholarship and activism, creating a space for the two to inform and enrich each other. After all, our data consist of many American Muslims who, through community engagement, want to actively change the perception of Muslims in their own communities. We are indebted to them because their hard work makes our scholarship possible. In return, we hope that some of these same activists can use our interactive, user-friendly maps to inform their work by engaging their audiences in visualizations of the rise of anti-Muslim hostility and efforts of Muslim Americans to counter this phenomenon.

42 Leave a comment on paragraph 42 0 In addition to our scholarly pursuits, we hope that our publicly-engaged humanistic work will become a tool for activists who are fighting for this vulnerable population. We are keenly aware of the complexity and the burden of the communities that make our scholarship possible. The technological tools of mapping and data visualization help us make a small contribution to their efforts.

43 Leave a comment on paragraph 43 0 I have found that creating the parameters of the dataset documenting humanizing efforts of American Muslims has been one of the hardest and most rewarding aspects of creating these tools. Caleb, Julia, and I have spent a tremendous amount of time discussing what and who to include and as well as the ethical and analytical implications of the collaborative decisions we make. Do we include professional speakers who advocate for interfaith dialogue? Or is their status as professional advocates at odds with the project’s goal of showcasing the burden of everyday Muslim Americans? If we opt to not include such actors are we erasing important work, skewing the reality of American Muslim outreach efforts? Do we include individuals who engage in outreach efforts not to humanize Muslims, but simply to practice the tenants of their faith that encourage charity and generosity? These are some of the questions we have continued to wrestle with in our work. These decisions are in themselves acts of analysis and, as such, are a foundational element of our scholarly work.

44 Leave a comment on paragraph 44 0 Decisions about the parameters of this dataset—about who and what to include—are directly tied to questions about coding the data, which is itself also a deeply analytical process. For instance, in the humanizing public life dataset we have a category for political activity. This covers electoral work, such as Muslim Americans running for office, organizing voter-registration campaigns, and sponsoring fundraisers for political candidates. When considering adding this category, Caleb and I had a long discussion about the potential benefits and consequences of including Muslim American candidates in our dataset and what we hoped the category would convey. We agreed that including such candidacies is important for two central reasons.

45 Leave a comment on paragraph 45 0 First, we wanted to include American Muslim candidates for public office because running for office is such a core element of participatory democracy. After all, the dataset is about American Muslim participation in public life. Second, including Muslim candidates in the dataset enables us to show how their rates of candidacy have changed over time, plunging after 2001, remaining quite low for over fifteen years, and entering a potentially new phase of increased rates of candidacy in response to the incredible growth of public anti-Muslim hostility since 2015.

46 Leave a comment on paragraph 46 0 Yet we are also aware of the added burden of being a Muslim candidate for office amidst this apparent increase. We found that Muslim candidates often struggle to be seen as equal citizens who are running for office simply because they want to improve the lives of their potential constituents. Instead, coverage tends to emphasize their Muslim identity and they are seemingly required to prove their allegiance to “American values.” Therefore, after many discussions, we agreed to include them in the dataset only when their religion is a part of their media coverage. We have found that at present we have not had to exclude more than a few candidates as a result of this decision, certainly not enough to skew results. This is an important finding in our project. Running for office is itself an instance of humanizing outreach for American Muslims. Thus, although we code such candidacies as “political activity,” the stories behind the data show that the nature of American Muslim candidacies is shaped by the conditions of public life, much like the broader story that Mapping Islamophobia tells. Public anti-Muslim hostility affects what voters expect to hear from Muslim candidates, raising important questions about their capacity to freely choose how to present themselves and discuss their platforms.

47 Leave a comment on paragraph 47 0 The data we have collected for this dataset shows quite clearly that humanizing outreach efforts far outnumber instances of political activity. While the humanizing nature of American Muslim candidacies for public office illustrates the often blurry boundaries of the codes that datasets require, coding them as “political activity” helps us show how anti-Muslim hate affects the nature of American Muslim participation in public life. Data visualization illustrates the relative frequency of humanizing outreach and political activity with a force that simply presenting percentages would not. Hopefully this helps generate affective understanding for the viewer. Highlighting political activity as a separate category, represented by a distinct color in our maps, makes this possible. The coding decisions we make do not occur in a vacuum. They are tied to what we are trying to communicate to users and, importantly, how we hope to communicate it.

48 Leave a comment on paragraph 48 0 Figure 7: Users can select layers of data to explore the different kinds of humanizing work in which American Muslims have engaged over time.

49 Leave a comment on paragraph 49 0 Figure 8: Sortable data allows users to see the incredible discrepancy between the scope of humanizing outreach (evident in Figure 7) and the numbers of American Muslims running for political office over time.

Reflections Together

50 Leave a comment on paragraph 50 0 As you can see, each of us brings different skills, experience, and sensibilities to this project. Our collaboration has made this project much richer than it would be otherwise. We have all learned a lot about digital humanities as the project has unfolded. This process has also occasioned important conversations about what it means to do public scholarship. The project’s emphasis on a pressing civil rights issue is a significant element of how we understand the purpose of public scholarship. But equally important is our recognition of our debt to the community we are seeking to serve, our attempt to be as transparent as possible in all that we do, and our decision to create an educational resource. These are some of the core values of digital humanities as we have come to understand them. It has been a wonderful process to discover these values in and through creating a digital humanities project, moving from neophyte status to advocates of the value of this kind of work as a mode of accessible publicly-engaged scholarship.

51 Leave a comment on paragraph 51 0 We have already seen one example of how our publicly-engaged digital humanities work can reach audiences that more traditional forms of scholarship might not. Jason Harshman, a professor at the University of Iowa College of Education, has integrated Mapping Islamophobia into his social studies teacher-training courses and professional development workshops, and Alisa Meggitt, a teacher at North Central Junior High in Iowa City, Iowa, draws on the project to open conversations about global citizenship with her students. In both cases, the content and the form/medium of Mapping Islamophobia have been important to the way these educators have drawn on the project in their teaching.

52 Leave a comment on paragraph 52 0 Throughout this essay, we have emphasized that Mapping Islamophobia presents information in a way that encourages users to explore the data and generate their own insights and conclusions. Clearly, from deciding on dataset parameters to coding and making decisions about data visualization, we are making a range of analytical decisions behind the scenes. It is our hope that these decisions leave sufficient room for users of Mapping Islamophobia to do their own work. At the same time, we are also very aware and self-conscious of the fact that we are telling a story, a story that we think is a pressing social justice concern. This is a difficult balance. We acknowledge this balance by including a small measure of explicit analysis accompanying one map. Users would have to spend a fair amount of time exploring the Mapping Islamophobia website before discovering this particular map, “Hate’s Effects: An Analysis of Collected Data.” Even still, we preface the map with a prompt to the user, asking them to consider the data for a few moments before moving on to our brief analysis.

53 Leave a comment on paragraph 53 0 Figure 9: We present what we think are our most significant findings in a map that is nestled in the site rather than highlighted as a key element.

In Retrospect: Reflections of a Neophyte Project Director

54 Leave a comment on paragraph 54 0 Nearly four years of working on Mapping Islamophobia has provided plenty of opportunity to consider what I might have done differently. There are so many decisions that go into getting a project off the ground. Very few of these decisions are set in stone, yet changes do become more complicated and time consuming over time. I’ve considered on a number of occasions, for example, whether the site’s data management and visualization platform, Carto, is the best fit for the project over the long term. I’ve also reconsidered some of the categories I’ve put in place to code and organize data.

55 Leave a comment on paragraph 55 0 Reflecting on the big picture, however, these aren’t the things that rise to the top of the list of things I would do differently, particularly if the question is what, on the basis of my own experience, I would advise others to keep in mind as they consider starting a digital humanities project. What rises to the top of the list are considerations about project sustainability. A senior colleague warned me as I began working on Mapping Islamophobia that the web is (among many other things) a graveyard of once-promising digital humanities projects. The best way to avoid this fate is to consider sustainability from the first moments of your project, which in retrospect I wish I had done more intentionally. Thinking about sustainability includes being clear-eyed about how the project relates to your overall research agenda, developing a good sense of what kinds of support and collaborations you’ll need (and have access to), and being open to a mode of doing research that departs from the “solo researcher” model that remains the norm in many fields.

56 Leave a comment on paragraph 56 0 If you are considering a digital humanities project, one of the first steps to take is to check to see what technical support your institution can provide over the long term. As colleges and universities encourage more and more graduate students and faculty to pursue work in digital humanities, it is essential to ask what kind of support your project will receive once you have moved beyond the initial, and often very exciting, start-up moments. The more successful your institution is at encouraging digital humanities, the more demands will be made on your staff collaborators as they support an ever-expanding roster of projects. This means that technical changes, like moving data management and visualization platforms, may feel like insurmountable challenges—very few of the technical or content-related decisions are set in stone, but as the project moves along, you will likely need technical expertise and other kinds of support to make them.

57 Leave a comment on paragraph 57 0 Ongoing project support is particularly important if your initiative, like Mapping Islamophobia, is an open-ended endeavor.[7] Mapping Islamophobia doesn’t have an obvious end point. When I started the project I didn’t really understand what it meant to embark on an open-ended digital humanities project, especially one whose goal was, at least in part, to serve as a resource for activists and policymakers. I started it because it felt important to do so. However, I didn’t truly stop to think where this project fit into my long-term research horizon. About two years into the project, a senior colleague advised me that in order for Mapping Islamophobia to “count” as fully as possible in salary and promotion reviews, I would need to publish about the project and its data in peer-reviewed settings. This seemed reasonable enough at the time. Now that I have committed significant time to doing so, however, I have real questions about the project’s sustainability.

58 Leave a comment on paragraph 58 0 My undergraduate student collaborators have been a welcome and essential element of Mapping Islamophobia. Still, the time they have to contribute has its limits, not least of which is how quickly cohorts move through their undergraduate careers. When you consider the time it takes to train students, even with significant support from staff collaborators in the digital humanities, undergraduate collaborators may cycle through the project more quickly than you can replenish the pipeline of those ready to contribute. Project sustainability requires that you consider such factors. As I focused more time on publishing around Mapping Islamophobia, I began to depend more on student collaborators to collect and enter data and for project management. Managing and editing this work still takes time, and so balancing the role of project director with other parts of your research agenda—not to mention teaching and service—can be quite challenging. Having student collaborators, and being a good mentor to them, requires a significant investment in time.

59 Leave a comment on paragraph 59 0 Moreover, sustaining an ongoing project according to best (ethical) practices requires that you compensate your student collaborators. Just as you need to inquire into what kind of ongoing technical support you will have, long-term sustainability requires that you be clear about what kind of financial support your institution can provide for student compensation that probably exceeds what would normally be available for research assistants. If your institution doesn’t or can’t provide this kind of support, identifying external sources of funding becomes all the more important. Of course, this takes significant time as well.

60 Leave a comment on paragraph 60 0 I have enjoyed, and am enjoying, developing and maintaining Mapping Islamophobia. As I reflect on the past years, I wish that I had taken more time to talk to people already doing digital humanities work to ask for their advice. Gathering information more intentionally from experienced digital humanities practitioners would have helped me consider big questions about where the project fits into my research agenda— and professional goals more generally—and how to sustain the project over time even when I turned my attention to other things, as inevitably happens in our professional lives. Digital humanities projects, like many other kinds of research projects, will change and grow, and having a clear sense of your goals, project scope, and long-term sustainability, even around things as basic as site maintenance if your project has a more delimited temporal and/or substantive scope than Mapping Islamophobia, will help make the experience all the more fulfilling.

61 Leave a comment on paragraph 61 0 Our description of Mapping Islamophobia shows that the project unfolded through an organic process. We learned as we went. It’s one of the things that has made working on Mapping Islamophobia such a joy. I’ve learned more than I could have imagined about a huge range of things—data collection, data coding, data visualization, not to mention all that working on the project has helped me see about anti-Muslim activity in the United States and the tremendous work that Muslims communities have undertaken to provide a gracious model of public life. Much of this learning has resulted from the collaborative nature of digital humanities work. This has helped me see that while project sustainability does depend on clarity regarding your own goals and horizons, it also depends on embracing a way of learning and researching fundamentally rooted in community.

62 Leave a comment on paragraph 62 0 It’s not surprising that some of the projects that have most influenced the way I think about digital humanities are sustained by collaboratives. Mapping Police Violence, which I discuss earlier in the chapter, has a multi-person planning committee in place to run the project. Another amazing example of social justice-oriented digital humanities, the Anti-Eviction Mapping Project, also depends on an extensive leadership and contributor collective. Had I studied other projects more intentionally I may very well have created a very different model for sustaining Mapping Islamophobia. While very few of us will embark on projects as ambitious as these amazing examples, their structure helps me reflect on my own experience and brings something into clear relief. Passion will help get your project off the ground. Your collaborative relationships will be what truly makes them possible in the fullest sense, hopefully in a way that will sustain them over the long term.


63 Leave a comment on paragraph 63 0 [1] You can find Mapping Police Violence at https://mappingpoliceviolence.org/. It inspired Mapping Islamophobia both in its goals and modes of data visualization.

64 Leave a comment on paragraph 64 0 [2] Over time, I’ve also come to recognize Carto’s limitations. Relative to other platforms, it has limited functionality and options for different kinds of data animation and interactivity. Tableau, for example, offers more options for data visualization of data interactivity, but requires much more training to use and maintain.

65 Leave a comment on paragraph 65 0 [3] http://cdh.ucla.edu/news/a-student-collaborators-bill-of-rights/. Accessed 05/24/18.

66 Leave a comment on paragraph 66 0 [4] There is certainly a good argument to be made for more explicit analysis, especially when creating a resource of possible value to policymakers. TellMAMA: Measuring Anti-Muslim Attacks, based in the United Kingdom, teams with a range of partners to produce reports drawing on and illuminating its data.

67 Leave a comment on paragraph 67 0 [5] We recognize that opting to include only those incidents that receive media attention from outlets with clear editorial processes and oversight limits our dataset. We feel strongly, however, that building on the careful editorial work of local, regional, and national “print” news outlets ensures that we can be very confident in all of our data. In an era of claims around “fake news” and “hate crimes hoaxes,” we want our data to be as immune from criticism as possible. We also want to make sure that people using our site can access all of our sources themselves, a key element of transparency. Other projects have made less restrictive choices about data collection. For example, a project out of the New America foundation that came online in 2018, “Anti-Muslim Activities in the United States,” includes data from what their site describes as a wide range of publicly available sources, including news reports and legislation. TellMama: Measuring Anti-Muslim Attacks uses crowd-sourcing and testimonial methods for gathering data. Pro-Publica’s “Documenting Hate,” which seeks to build a general dataset of hateful activity across the United States, uses an approach that combines these methods, including data from publicly available sources as well as a crowdsourcing component, such as reports by those targeted, witnesses, and others with knowledge of incidents.

68 Leave a comment on paragraph 68 0 [6] Coding data is a central feature of digital humanities work. It forms the basis for data organization and presentation, providing categories to make it possible for users to navigate the information you’re providing. The Southern Poverty Law Center’s Hate Map provides a nice model of how data coding provides a way to filter information. They’ve organized things according to the ideology of the organizations they’re tracking and offer an easily navigable menu for exploring the data accordingly. We were able to establish our six main coding categories fairly early on in the project, in part because patterns in the data quickly began to appear. We also use a second level of coding, which we call “Event (short name).” This coding is slightly more descriptive of individual incidents. Over time, the terms in this second level of data coding began to proliferate, creating inconsistency across the dataset, leading us to consolidate them while maintaining sufficient specificity to serve the intended descriptive purpose. We use drop down menus in our Excel spreadsheets to maintain data entry uniformity in many categories. This is especially useful when many different people serve as contributors.

69 Leave a comment on paragraph 69 0 [7] Not all, or even most, digital humanities projects are ongoing in this way. A wonderful example of a discrete digital humanities project is Kayla Wheeler’s Mapping Malcolm’s Boston: Exploring the City that Made Malcolm X. Aspects of the project may continue to develop over time. Nevertheless, because it draws on what is ultimately a limited dataset project sustainability looks different than for projects with ever-growing datasets.

Source: https://opr.degruyter.com/digital-humanities-and-research-methods-in-religious-studies/caleb-elfenbein-farah-omer-and-julia-schafer-mapping-anti-muslim-hostility-and-its-effects/