In learning about eighteenth-century travel and working with the Grand Tour Explorer, students have developed ideas and projects in many directions. They have worked with a variety of methods and approaches—some choosing to annotate texts by hand to create new data, others writing code to run on large corpora. Mapping and network analysis remain enticing for many who are excited to use them on historical material. With these approaches, topics toward which students have gravitated include influence and networks and trying to understand better the logistics, the choices, and the experiences of travelers. Women travelers and matters related to gender have also interested many students. And in the wake of the coronavirus pandemic, so has the topic of disease.

The projects featured here are a selection chosen as representative of the broader body of student work produced in these classes. They do not represent fully realized research, but rather forays into unfamiliar subjects under the time constraints of an academic quarter. Yet the students’ creativity shines through in them, as does their general comfort with technology and data. For most students, an important discovery was the complexity of the records of the past and the thrill and courage of putting forward interpretation when faced with sources from a different, incomplete world. There was also a widespread desire among them to think about the limitations imposed by data categories, and to close the divide separating open from closed access, as well as digitized from forgotten material.

Here is a gallery of work. (Please note the embedded images are screenshots, so definition might not be as clear as in the original.)

The Influence of “The Dilettanti”

Matthew Tan, Winter 2019

Matthew Tan uses Explorer data to discover new evidence and context on the influence of the Society of Dilettanti, a society that served as quasi authorities in the infant fields of European archaeology and classical arts in the eighteenth and nineteenth centuries. Tan charts the influence of the Dilettanti by analyzing its Explorer data in order to gain a macro perspective through the actions of its individual members, extrapolating on the larger data points of the group as a whole. Specifically, Tan compiles information on each of the fifty-five society members who are recorded to have been affiliated with societies beyond the Dilettanti, such as the Royal Society, Society of Antiquarians, Royal Academy, Royal Irish Academy, and the Free Society of Artists. Tan develops multiple data visualizations that explore the chronology of when each member joined various societies, the quantity of overlapping memberships, and the gaps in the data set to argue that the ability of the society to attract such valued people presents strong indicators pointing to the increasing influence that the society had during the eighteenth century. Incompleteness in the data set, however, such as missing information on important society leaders, presents a larger motivation for more complete and accurate data sets in the study of the digital history.

Fig 1. Left: Waffle plot indicating the relative number of recorded Dilettanti in the data set who are part of a single or more societies. Right: Visualization of the completeness of data surrounding the Dilettanti. Each row corresponds to a single entry. Black indicates that the data is known; white indicates unknown data.
Fig. 2. Timeline showing when various members of Dilettanti started and ended membership in a society. The names are shown on the left. A bar indicates that the person joined the society. Color indicates which society. We filtered only by entries that had start dates. End dates when not known are filled in using death dates. The count of how much is known in start and end dates is shown in the bottom and right respectively.
Fig. 3. Age first traveled by Dilettanti members against the number of years from the time they joined. Notice that majority of the Dilettanti first traveled in their twenties and joined within ten years. An analysis of the outliers would be interesting.

“The Women Kiss Me, & the Men Write Verses about Me:” Hester Piozzi, Gender, and Cicisbeo on the Grand Tour

Amy Hoemeke, Winter 2019

Amy Hoemeke analyzes the writings in Hester Piozzi’s Observations and Reflections Made in the Course of a Journey through France, Italy, and Germany (1789) to explore how Grand Tour travelers wrote about and discussed gender. Hoemeke does this by identifying gendered words and topics such as cicisbeism—a term for an attendant who served a woman, sometimes sexually—or female education, which was possibly more accessible in Italy than in England. Hoemeke compiled a list of gendered words found in Piozzi’s writing and used this to extract data from other tourists’ entries using the Grand Tour Explorer. Hoemeke records the city the traveler was in when they wrote the word or topic, the traveler’s gender, and the date of travel to create a visual representation of how, when, and where gendered topics were discussed. Furthermore, she spatially mapped the frequency of each gendered topic written about any city to reveal that Florence and Rome were, in her data, overwhelmingly represented as cities where travelers wrote about gender and gendered experiences. The number of mentions of cicisbeo in Florence supports the hypothesis that while cicisbeo might have been the most distinct feature of gender difference in Italy, it also acted as a vehicle for further discussion and analysis of the topic. Ultimately, this data shows that the experience of gender for travelers on the Grand Tour was one of exposure to new ideas of women’s freedoms.

Fig. 4. Chart of gendered terms used by Grand Tourists and Hester Piozzi and the cities where the terms were used, based on data from the Grand Tour Explorer and Hester Piozzi’s Observations and Reflections Made in the Course of a Journey through France, Italy, and Germany.
Fig. 5. Chart of gendered terms used by Grand Tourists and Hester Piozzi and the frequency with which the terms were used, based on data from A Dictionary of British and Irish Travellers in Italy, 1701–1800 compiled from the Brinsley Ford Archive by John Ingamells, and Hester Piozzi’s Observations and Reflections Made in the Course of a Journey through France, Italy, and Germany.
Fig. 6. This figure depicts the search terms themselves and the frequency by which they were mentioned in the Grand Tour Explorer entries. This further depicts the prevalence of cicisbeo in discussions of gender, which, given that it is a relationship so present and noncontroversial in Italian culture, reveals once again the distinct differences between women’s freedoms in Italy and those in England and other cultures these travelers may have been exposed to. Other terms with larger frequencies include girls, sex, lover, mistress, and unmarried.

Understanding Piozzi’s Classicism

Jacob Kaplan-Lipkin, Winter 2020

Jacob Kaplan-Lipkin created and categorized new data to set out to analyze how Hester Piozzi’s classical descriptions differed or compared to other traveler’s writings such as Louis Dutens and John Moore. Kaplan-Lipkin asks, To what extent are classical references manifested in Piozzi’s writing, and how does she interact with the classical world on the Grand Tour? Is this different from her male peers? To pursue this question, Kaplan-Lipkin created a spreadsheet of every allusion Piozzi makes to a classical topic in her writings on Florence, noting the relevant quotation, a summary of what is being referenced, a category for the type of inspiration (e.g., art, text, etc.), whether the reference is mappable to a specific location, the location of the reference, and whether the reference is Greek or Roman. Once this information was compiled, Kaplan-Lipkin repeated the process with Dutens and Moore’s writings on Florence. In total, twenty-six mentions or allusions to classical antiquity were found across the three texts. Kaplan-Lipkin uploaded this data into Palladio to map these data points onto Florence. The visualizations helped clarify that while Dutens and Moore’s classical references were mostly inspired by their visits to the Uffizi gallery, Piozzi sees allusions to the classical world all across the city and through personal encounters. Piozzi’s writings point to a worldview influenced heavily by an orientation toward the past and references Hellenism much more than those of her peers. Piozzi is also the most diverse in her allusions, referencing art, mythology, literature, and religion while her peers mostly allude to classical art.

Fig. 7. Map of classical allusions in the city of Florence by three different tourists. Red represents Hester Piozzi; blue represents Dr. John Moore; green represents Louis Dutens.
Fig. 8. Bar charts that represent, in two different ways, the breakdown for references to Greek antiquity, Roman antiquity, or both, among the three travelers studied: Piozzi, Dutens, and Moore. The graph shows Piozzi to have the most Greek references.

The Grand Tour Influence on the Founding of the Royal Academy of London and the Genre of British Landscape Painting

Natalie Francis and Brennan Megregian, Winter 2020

Natalie Francis and Brennan Megregian explore the influence of the Grand Tour on the development of the Royal Academy and on the development of British landscape painting through natural language processing; primary source documents; and spatial, textual, and network analysis. The Royal Academy of London, established in 1768, sought to promote the advancement of visual arts education and exhibition; and equal numbers of historical, portrait, and landscape painters among its founding members reflects the rising prominence of the classically resonant genre in the British arts scene. This academy counted many former Grand Tourists among its thirty-six founding members, most notably the first president and portrait painter Sir Joshua Reynolds (1723–92). To assess “influence,” Francis and Megregian focus on evidence of landscape-painting interest from the social networks and linguistic trends of select Royal Academy Grand Tourists, as well as broader geospatial and network trends among all Royal Academy painters identified through the Grand Tour Explorer database. Megregian employs the NLP technique of word2vec on Joshua Reynolds’s Archive.org digitized corpus of thirty documents that range from his Italian sketchbooks to Royal Academy presidential addresses and private correspondences to identify contextual trends in his writing. She also performed geospatial analysis in Palladio and Python to identify what cities future Royal Academy members traveled to with the highest frequency. Francis investigated network analysis through producing Python visualizations of the relationship between the twenty-three Royal Academy/Grand Tour painters and their Grand Tour Explorer (GTE) Matched Mentions. She also cross-referenced Richard Wilson’s GTE entry with his Matched and Unmatched Mentions to generate a set of Palladio graphs of his social network developed during the Grand Tour. Using breadth of digital humanities methods, Francis and Megregian identify positive relationships between the Grand Tour and rise of British landscape painting, while also encountering limits, both in terms of the Explorer data—wishing similar records existed for artists of different nationalities—and of digitized corpora—wishing more digitized sources existed for artists beyond Reynolds.

Fig. 9. A “spring” network visualization that expresses the relationship between Royal Academy affiliation (purple), Royal Academy members (red) and their Matched Mentions (blue).
Fig. 10. A word2vec visualization of a corpus of thirty Sir Joshua Reynolds documents on art (Brennan). The purple nodes for “landscape” with words such as composition, exhibition, capital, historical, and practical align well with scholarship on Reynolds, which portrays him as more a didactic portrait painter than someone like Wilson, who was unconventional in his landscapes.

Using Topic Modeling to Compare Travel Accounts from Florence by Male and Female Authors

Sarah Pincus, Summer 2020

Sarah Pincus utilizes topic modeling to explore how gender influenced four Grand Tour writers’ choice of words to describe their experiences in Florence. Focusing on primary sources written by Hester Piozzi, Anne Miller, Thomas Watkins, and Tobias Smollett, Pincus analyzes the narratives each wrote about their time in Florence through Latent Dirichlet Allocation (LDA), detecting word and phrase patterns within each text. Interested in differences between female and male authors, Pincus assigned Hester Piozzi and Anne Miller to one file and Thomas Watkins and Tobias Smollett to another and then used Orange3 to identify the top five topics within each file. Some differences were immediately apparent. Only the results from the “Female Author” group attest a gendered term that does not refer to men. Furthermore, the word us appears in the results of the “Female Author” group three times, whereas it appears zero times in the results of the “Male Author” group. Also, an MDS diagram visualizes how the female authors were much more likely to write about a variety of topics, whereas the men had a more limited range. Pincus theorizes that this might be because men had a more scripted travel experience on the Grand Tour, while women, remarkably, had less social pressure to study classical antiquities and, thus, more freedom to consider multiple topics. Pincus notes the irony, however, that this authorial freedom for women originates from their exclusion; the concept of a proper Grand Tour was encouraged for young men, not young women. The data supports this idea, as the Grand Tour Explorer contains database logs that 1,242 men traveled to Florence, whereas only 283 women traveled there. Pincus suggests that further research could be done by enlarging the data set to include more writers and to explore how rather than treating gender as binary, a more in-depth research project could assume gender as a spectrum instead.

Fig. 11. Results of topic modeling analysis run on “female authors” versus “male authors,” showing the most relevant and the most frequently used words.
Fig. 12. Comparing the MDS diagrams makes it evident that the female authors were significantly more likely to write about a wider variety of topics than the male authors, who, for the most part, wrote only about Topic 3.

Social Networks of Grand Tour Tutors

Regina Ta, Winter 2020

Tutors played an influential role in shaping the Grand Tour experience for the British aristocracy, Regina Ta argues, and developed their own social networks, which grew as they traveled more extensively. Building on previous research by Sánchez-Jáuregui, Ta hypothesizes that with information from the Grand Tour Explorer database, it is possible to map the edges of tutor’s social networks and discover how extensive these networks really were. Ta searched the database for employment identifiers and occupations, compiling sixteen entries for individual tutors. She defined an extensive network by the number of tours they led, as well as the number of nonunique locations visited on their travels. In developing a scatterplot and other Palladio visualizations, Ta found there is a positive correlation between traveling extensively and being well-connected. For example, Ta writes, Richard Phelps, who led five tours, became acquainted with architects, painters, and statesmen, eventually gaining membership into the Society of Dilettanti. And in the visualization of tutors’ social networks, there are some shared connections between tutors, showing that their social circles overlapped. The social networks of tutors offered the possibility of advancement as many tutors went on to become diplomats or secretaries.

Fig. 13. Scatterplot graph showing positive correlation between traveling on multiple tours and extent of social contacts.
Fig. 14. Palladio graph visualization of sixteen tutors’ social networks revealing some shared connections between them and showing that their social circles overlapped.

Travel Patterns to Venice and Rome, 1710–1730

Peter Benitez, Autumn 2021

Peter Benitez investigates how the Great Plague of Marseille (1720–23) affected Grand Tour travel patterns to Venice and Rome through collecting textual evidence, travel data, and creating detailed maps of the density of travel in the period before the plague (1710–19), the period of plague (1720–23), and the postplague years (1724–33). Benitez gathers the textual evidence recorded in the Explorer entries from travelers such as George Parker and Jonathan Richardson the Younger, both of whom write about how the plague impacted their travels. Parker documented how Venice required testimonials from a doctor to enter during the years of the disease, and Richardson detailed how he did not go to Venice, traveling to Rome instead, owing to disease. Benitez then turns to data, compiling information on 329 Grand Tour travelers during the 1710–33 period. Using Palladio, Benitez creates three maps to highlight travel patterns of the three periods before, during, and after the plague. These maps visually depict how during the time period of the plague, travel decreased to Venice and increased to Rome. This decrease, Benitez argues, can be attributed to the fear of the plague and the quarantine measures put into place by the city of Venice, making it difficult for foreigners to enter without the proper medical paperwork. This information can be used to better understand the history of modern health practices and their impacts on people moving across borders.

Fig. 15. Maps detailing the density of travel in the period before the plague (1710–19), the period during plague (1720–23), and the post plague years (1724–33).
Fig. 16. This figure depicts all quantitative information discussed in the study in the form of a table.

Networks of Social Influence in Edward Gibbon’s Grand Tour, 1764–1765

Coen Armstrong, Autumn 2021

Coen Armstrong asks how we can formalize and measure the notion of a “general intellectual scene” during Edward Gibbon’s Grand Tour (1764–65) when descriptions of the Grand Tour often omit who travelers encountered and why. Many travelers on the Grand Tour had few contacts with other travelers, and for many, gaps between social groups were often bridged by art collectors, dealers, and to a lesser extent merchants, scholars, and soldiers. Gibbons, however, offers a meaningful account of social influence in his letters, which establish that those of the same social class, over the weeks they would be in the same location, were brought together by mutual acquaintance, letters of introduction, or presentation at court. This suggests that institutional affiliation played a central role in who spent time with whom. To create a more sophisticated lens on who bridged social groups, Armstrong constructs a visual network that reflects the influence of certain travelers on others, based on how many life experiences they share. To measure this, Armstrong designs a metric of similarity called affinity. Two people have an affinity if they share a life event: being married, going to the same school, being a general, or being part of the same social circles, for example. One affinity is represented as one edge between two nodes, where each node is a person. Armstrong utilized the full Grand Tour Explorer database and filtered only for dates from 1764 to 1765, the dates of Gibbon’s travel, which included twelve hundred entries. The affinity graph reveals that many travelers have very low affinity scores, and few have extremely large affinities. Armstrong concludes that there are deep qualitative differences in travelers Grand Tour experiences and social networks.

Fig. 17. Chart of the local degree of each node on the affinity graph, for Gibbon’s contemporaries. The spike at 1 is explained by people who were married but had no other affiliation, while the spike at 17 is explained by those members being merchants. Most striking is the inequality in distribution: most people have quite low affinity scores and very few connections at all, but some people have an extremely large number of connections.
Fig. 18. Palladio graph visualization showing connections between some of Gibbon’s contemporaries based on shared education data points, that is, whether these travelers had gone to the same school—one of Armstrong’s measures of affinity among travelers.
Fig. 19. Table showing betweenness centrality score of a few of Gibbon’s contemporaries, alongside some comments to characterize these travelers.

Mapping Perception

Joshua Francis, Autumn 2021

Joshua Francis asks how combining Explorer travel data with textual data from travelers’ accounts could reveal how Grand Tourists perceived different locations across different time periods. Utilizing data on how long travelers stayed in various locations across Italy and cross-referencing it with written accounts of those places by the travelers provides a glimpse into the perceptual landscape of Italy in the eighteenth century. To compare the shifting perceptions of place, Francis divided his analysis into four time intervals: before 1725, 1725 to 1750, 1750 to 1775, and 1775 onward. To correlate travelers’ length of stay with appreciation, he developed a metric based on the proportion of time spent in each location in their itinerary and the number of standard deviations of each proportion from the mean of all proportions. To analyze travel accounts in Adam Matthew Digital’s (AMD) Grand Tour database, Francis designed an algorithm to search transcripts for place names and then pull the twenty words preceding and following the location name for analysis. Francis looked for Grand Tourists who were both present in the Grand Tour Explorer and had text present in the AMD database. After scraping the databases and finding five matches, Francis used logistic regression, support vector machines, and an MLP neural network to find the data results. With results accuracy at 70 percent, more research is needed. Yet Francis’s data analysis of the perceptual landscape of eighteenth-century Italy confirms scholarly analysis that the perceptions of locations on the Grand Tour shifted during the long eighteenth century.

Fig. 20. Graphic visualization of itinerary metric for perception over four time intervals (before 1725, 1725 to 1750, 1750 to 1775, and 1775 onward). The size of the circle represents the perception score; red stands for negative and blue for positive. Only positive perceptions are labeled and many of these are not in the top twenty most-visited locations.
Fig. 21. Natural Language Toolkit’s Synset analysis of perception of locations in five travelers’ accounts: the graph representing changes in score (blue is positive and red is negative) in first to second period, second to third, and third to fourth.

The Hamilton Network: Quantifying the Influence of Eighteenth-Century Antiquities Dealers

Eliot Jones, Autumn 2021

Eliot Jones utilizes data from the Grand Tour Explorer database to chart a network of four British antiquities dealers: Sir William Hamilton, James Byres, Thomas Jenkins, and Gavin Hamilton. Employing network theory to understand the dealers’ sphere of influence over the art market in the eighteenth century, Jones questions how these men shaped not only the British taste in art but also how we think about the origins and provenance of antiquities today. The network operates through three main “layers.” The first layer of the network is formed by finding all the connection points between the four dealers and the database, followed by a search within the more than six thousand entries of the Grand Tour database for any connection points based on names mentioned in their biographical entries. The third layer repeats that process again for the second layer to attempt to prevent bias. Focusing on degree of network and betweenness centrality, the visual data reveals that W. Hamilton, Jenkins, Byres, and G. Hamilton appear in the top four nodes with highest betweenness centrality. By these metrics, it is clear that these four individuals played a major part not only in the network but also in the art and antiquities market during the Grand Tour. Furthermore, Jones focuses on Sir William Hamilton to explore his political and cultural relations. For example, the sale of Hamilton’s collection formed the basis of the British Museum’s original antiquities collection. By amassing such an important collection, Hamilton ransacked archaeological sites and burial grounds, smuggling vases and other artifacts of importance across borders. Still today, Hamilton’s methods of collecting have left scars as his practice set a precedent for other dealers to steal and smuggle as seen most recently by billionaire Michael Steinhardt’s trial, which seized more than 180 stolen antiquities.

Fig. 22. The four British antiquities dealers object of this study—all connected, and their nodes sized by their overall network metrics.
Fig. 23. Network graph showing all connection points between the four dealers and the rest of the Explorer’s database of travelers, revealing that W. Hamilton, Jenkins, Byres, and G. Hamilton appear in the top four nodes with highest betweenness centrality.
Par. X format_quote link