The Past is a Foreign Country: Teaching as Tour Guiding

Rachel Midura

When I walk into a classroom or conference gathering, I introduce myself as a digital historian of early modern Europe. That statement is often greeted by polite confusion. “Isn’t that a contradiction in terms?” a student or colleague will ask. They imagine a mill-powered computer or knights moonlighting as information technology consultants.

I have learned to love these questions because they open a portal to many others: why does it seem like a contradiction? What does that say about our assumptions regarding premodern history or digital technologies? These conversations bring assumptions into the open air like so much dirty laundry. The truth is that history and the “digital” are often perceived as opposite ends of the trajectory of progress. History describes eons of stability, scarce information, and largely inaccurate knowledge. The “digital” age on the other hand is both our present and future: ever-changing, so overloaded with information it often defies individual comprehension, and only navigable with the aid of technology smarter than humans. These characterizations may seem like parodies, but as any educator knows, that is the nature of unexamined assumptions that have been left to molder.

Our hope is that the lesson plans here will not only teach how to use the Grand Tour Explorer with a variety of tools but also structure conversations about the nature of digital history. There is no substitute for direct experience of navigating premodern data. To this end, we have sought to flatten barriers to entry. The following practicums do not require any prior knowledge of coding or installation of tools. The only prerequisite is access to the Explorer, Google Sheets, and an internet connection. Slowly, but surely, readers practice many of the basic principles and best practices of data manipulation and visualization. We seek to level the playing field for both dyed-in-the-wool coders and self-described novices.

While there is no substitute for learning by doing, I will briefly address three overarching themes here: the first is the value of starting with questions and quickly flipping them to consider the validity of the hypotheses they represent or inspire. The second is the merit of teaching researchers digital approaches to historical sources. The third is the importance of historicizing approaches to data. I see the third category as providing some of the most important lessons that students will take from the classroom into their daily lives.

Over the course of the 2018–19 school year, I collaborated with Giovanna Ceserani to design and teach a course at Stanford University in tandem with the development of the Grand Tour Explorer. In this iteration of “Virtual Italy: Methods for Historical Data Science,” at that time called “Mapping the Grand Tour: Digital Methods for Historical Data,” each week focused on an essential question, such as “Who were the Grand Tourists?,” and “Why did tourists travel?” These open-ended questions allowed us to move fluidly between traditional source analysis and the Grand Tour Explorer’s database of thousands of British travelers. We modeled for students that being a historian meant having a full toolbelt of approaches. This included finding and reading primary sources but also selecting a program or metric that could produce the most useful quantitative or visual analysis.

Scholars are skilled artisans: it can be difficult to spot the cracks and absences in the literature. Our question-based approach meant that students quickly learned to see holes in the historiography. We deliberately chose many questions that scholars had long declared “settled.” It was only in attempting to recreate the results using different sources (hypothesis testing) that students began to see different results and to wonder why that was the case. Was it true that “tourists travelled as rapidly as possible between major cities” (Black 2003, 3)? If you mapped the data from the Explorer, you would indeed see very little representation of the countryside. Looking at the chronology of travels instead (practicum 5) revealed the weak plaster. This was our entry point for discussions about the nature of sources, which were more likely to record a date in general terms than to the day (“a few hours,” or “some months back”), and more likely to explicitly name a destination city than a humble way station. It was also the start of discussions about the compounding impact of data interpretation. “Rome” might be easy to match to coordinates and stick on a spreadsheet and map, but what about day trips or time “in the country” or “on the road”? If we “cleaned up” this messiness, how did it change our overall vision, and who would no longer be a part of the picture?

Timing is also tricky in other ways. The truth is that in a single semester or ten-week quarter, students are unlikely to arrive at radical new answers. Historians often spend years in the exploratory stages of research before attempting publication. But the research paper as a genre can still teach students that the right stance is a definitive statement. The paper, duly argued and submitted, is then swiftly forgotten. Our goal for Mapping the Grand Tour was an approach in keeping with a history “lab.” Students knew from the beginning that they would be a part of an ongoing project with its own quirks. Platform outages at inopportune moments may have kept us up at night, but they vividly illustrated that history, too, is always “under construction.” Our hope was that students might be empowered, in turn, to make their own contributions and corrections.

Take this as a warning: the result is often frustration—at least, at first. What seems remarkably smooth in the classroom (outages and all) can seem insurmountable when revisited at home, especially when applied in entirely self-directed inquiries. I met with our students after they submitted their proposals for independent research. I found that they had learned to ask the big questions but were stymied by what felt like the impossibility of answering them. Assurances that final projects would be exploratory provided only a modicum of comfort. Fired up by the possibilities, however, students wanted to accomplish all or nothing.

I am a firm believer that any coursework should lead to independent practice, but moving from the structured to unstructured application is a significant hurdle. This, to me, is one of the most valuable lessons we can teach: how to grapple with limitations. Meeting individually with students is the invisible labor of these lesson plans. Students will need a capable cicerone helping them to break down big questions into small steps: if you don’t have the information you need, how can you get it? What might serve as a proxy, or how can you integrate another source, or tool? How much time will it take you to do so, and is that feasible given your life? We have written these lesson plans with many such pauses, where we encourage you to pull back the curtain whenever possible, being clear about not just what you are doing but why.

It is important to make the invisible labor visible for students, both so that they can accomplish their goals and so that they can value what scholars spend the bulk of their time doing. Yes, their decisions and priorities will impact the work they produce, much as it has for thousands before them. This is the constant thread that ties together history and the present age: humans can be data, but data is invariably human.

Opening Up the World: Teaching Data-Intensive History with OpenData

Annie. K. Lamar

In 2014, a group of researchers introduced what is now one of the largest consortiums of ancient open-data practitioners, curators, and developers. In their introduction, they write, “Freeing datasets from their tools is another crucial step for re-use as well as sustainability.”¹ The Grand Tour Explorer is a prime example of what it means to free a dataset. By turning a large, inaccessible reference text into an expert-curated, searchable, open-access database, the Grand Tour Project team has not only freed data from its source, but also unlocked this data for researchers, scholars, and students.

The advantages of a large, curated, transdisciplinary, open-access dataset are numerous. As an interdisciplinary scholar whose research focuses on using machine-learning methods to promote and understand ancient languages, I strive to instill the values of open research in my students and my own work. When I taught the course “Virtual Italy: Methods for Historical Data Science,” I was a PhD candidate at Stanford splitting my time between the Data Science and Classics departments, and concurrently completing an MA in Education. Issues of data sustainability and pedagogy were at the forefront of mind at that moment.

In this brief reflection, I focus on the advantages specific to the Grand Tour Explorer and its data, rather than the overall benefits of normalizing the production of open-access data, which include bolstering democratic ideals through the educational and political empowerment of individuals, stimulating scientific discovery and innovation, and driving forward a global initiative to confront climate change and promote climate justice (to name a few). The Grand Tour Explorer, by the context of its subject, the intentionality of its design, and the presentation of its content exemplifies how powerful open-access data can be in education.

As a teacher for “Virtual Italy: Methods for Historical Data Science,” it was apparent that the project’s dataset provides a rich foundation for personalized education. When students are interested in what they study, they are more motivated, confident, and involved in the class. The Grand Tour Explorer provides a varied context from which students can choose to pursue the ancient past by exploring the ancient Mediterranean monuments and places the travelers went to see, the recent past by examining patterns of the travelers, the present moment by asking what socio-cultural patterns have persisted from the Grand Tourists to our society, or the future by thinking about what parallel cultural fascinations and movements might exist in relation to our world. Students can also dive into questions about labor, politics, religion, race, socio-economic status, gender, and all the intersections between. The sheer amount of expert-curated metadata in the Grand Tour Explorer makes it a starting point for a research project in nearly any discipline.

The Grand Tour courses we taught consisted of first-year students and students about to graduate. We had students with advanced levels of programming, and students who had to be taught how to right-click. Students with all levels of academic and technical experience were able to form high-quality research questions that they found compelling, and use the available data to address these questions. I did in fact teach one of my students how to right-click. That student was able to find a research question so personally compelling that by the end of the course, they were learning Python and eventually landed a role as a digital humanities intern. The power of curated, diverse, and open-access data can be immensely transformative in students’ academic careers because of the opportunities it provides.

At the same time this dataset offers opportunities for individually-designed learning experiences, it also fosters an excellent environment for collaboration. The framing of the course attracts students from a wide variety of disciplines, and the intersectional content of the data invites students to seek out others with knowledge bases different than their own. Furthermore, the project-based character of the course means that students need to develop not only technical skills, but also academic and meta-cognitive skills to succeed. Teams of students with different strengths naturally form in a classroom environment that rewards collaboration, encourages academic risk-taking, and demands questioning.

Rachel Midura already summarized well the value of learning through inquiry and how digital methods promote such learning. I’ll only comment further to emphasize that the Grand Tour Explorer itself is designed to promote varied and original inquiry. The search engine allows learners to add and remove filters from numerous categories, to search across all text-based fields, and to view travelers both individually and in relational contexts. The advanced search page presents categories equally as important, allowing the learner space to formulate their own ideas about what avenues of inquiry they might pursue.

On a final note, it was an incredible boon that this data was available for teaching during the COVID-19 pandemic. I taught “Virtual Italy: Methods for Historical Data Science” alongside Giovanna Ceserani during the summer quarter of 2021, when Stanford University was still largely virtual, and during the fall quarter of 2021 as Stanford transitioned back into in-person classrooms. During the pandemic, it was not possible for our students to participate in many of the activities that are generally available to history classes; in particular, the library archives were closed. Archival research is a powerful way to help students connect with specific people and their lived experiences, and we are fortunate that Stanford Library has several travel manuscripts and journals.

The Grand Tour Explorer allowed us to simulate an archival experience because so much care was put into the presentation of individual Grand Tourists. Students connected with people’s past experiences by reading carefully transcribed and formatted descriptions of their travels, which often feature quotes, mentions of other people in the database, and personal details. Even in a (relatively) post-COVID world, many students and institutions do not have access to archives to enable this unique type of learning. Databases like the Grand Tour Explorer give millions of students the chance to engage in archival-like, micro-history-informed research by practicing open science and embracing open data.

Notes