As the digital humanities continue to grow, it’s important to acknowledge where the discipline has been, where it’s going, and why its methods are of use to all humanists, digital and non.
These are questions Martyn Jessop and Joanna Guldi tackle in their articles. They’re interested in the use of digital visualization and textual analysis in humanities scholarship.
In “Digital Visualization as Scholarly Activity,” Jessup describes digital visualization with three characteristics. It is:
- Allows the manipulation of both graphical representations and the data it’s derived from.
- And finally, it acts as the primary medium of communication.
Today, digital visualization comes in various forms, but Jessup looks specifically at the types of data being visualized and identified the following:
- Space: While Geographical Information Systems (GIS) software tends to dominate the study of spatial relationships, digital dynamic maps are more often utilized by humanists.
- Quantitative Data: Quantitate Analysis as visualization can be found in generic statistical analysis software or embedded in specialized applications like those used in text analysis.
- Text: A lot of text visualization methods are analytical quantitative methods borrowed from the sciences. These can include tables or graphs, or more recently, unidimensional or two-dimensional physical objects, abstract objects showing relations among words or between words and annotation, and animations.
- Time: These usually look like timelines that study complex historical events, their buildup, and the interrelatedness of earlier events.
- 3D Visualization: This is typically in the form of visualizations of built environments, like a recreation of Ancient Roman Cities.
While these forms of digital visualization in the humanities appear to be something new, Jessup shows how they fall in the wider context of visual sources in the humanities that scholars have utilized for some time. These include galleries of images, single images, museums and collections of objects, moving images like film or television, dramatic recreations, maps and atlases, and visualizations of data. Each of these visual secondary sources organize thoughts about the past and communicate them to others. They represent a continuum of visualization within Humanities scholarship that have always constituted valid scholarly activity. Digital visualization is simply the latest development.
Jessup does however find problematic qualities in digital visualization because it presents viewers with a complete and convincing image that is formed from research that is often incomplete. In addition, only few users practice digital visualization or are even visually literate which results in a lack of historical background regarding theoretical issues and methodology among practitioners.
To ensure the integrity of digital visualization as a scholarly activity Jessup argues that standards of the field need to be established in order to ensure intellectual rigor. He looks to the London Charter, a document created for and by the digital heritage community, as a model for the creation of such standards. It addresses issues of method, sourcing, transparency, and documentation standards that should be applied to digital visualizations in order to reach its full potential application in the humanities.
The issues Jessup highlights with digital visualization are personified in Guldi’s study of walking in London during the first half of the nineteenth century.
Guldi uses electronic databases like Eighteenth Century Collections Online (ECCO), The Making of the Modern World (MMW), Google Book Search, the Oxford English Dictionary (OED) and others to understand how walking symbolized social change in nineteenth century London. She ultimately found that during the rise of middle-class political power that occurred around the time of the 1832 Reform Act, walking became a means of identity making on the streets of London where power was vied for by middle-class ‘striders’ and aristocratic ‘loungers.’
Guldi’s work shows that these type of textual digital visualization methods can be extremely beneficial to historical inquiry. However, this research was not without its complications.
As Jessup notes above, problems of method and incomplete data pose an issue to visualization, issues that Guldi encountered during her research. She found many constraints in the databases she was using. They were often incomplete, reflected institutional biases, or had questionable accuracy rates.
Despite these shortcomings, Guldi did in fact find serious benefit to using these resources –they just required some cross checking, contextual readings, selection of terms, iterations, and careful periodization. This approach is how Guldi was able to perform a nuanced series of keyword searches across these platforms to identity how the term ‘walking’ in its various forms appeared in secondary publications from 1800-1850. Applying contextual readings based on these searches, Guldi was able to identify how the ‘stride’ of the middle-class and the ‘lounge’ of aristocrats were constructed and symbolized class differences and conflict during this era.
While Guldi found that walking in nineteenth century London was a subject particularly well suited for this type of textual analysis because of its substantial use of nuanced terminology, it’s likely that other areas of inquiry would benefit from this type of analysis as well. As digital analysis methods develop into concrete standards and practices like Jessup hopes, and visual literacy expands, humanists will be able to utilize these methods on a greater scale. This will open up new avenues for research that might have always existed, but perhaps we didn’t know were there or didn’t know how to use them.
Has Guldi’s research inspired you to try out textual analysis in your own research? Can you think of a project that would benefit from digital visualization? What kind of standards should methods and tools of digital analysis have?
2 Replies to “Digital Analysis Is Becoming a Walk in the Park.”
Thank you, Alex! I was super entertained by Joanna Guldi’s article, fascinating! It confirmed my suspicion that strolling was crucially affected by class one belonged to. As I grew up reading Russian classics of the nineteenth century where everyone insisted on walking all the time.
It also has helpful reminders for us as scholars to cross-check by way of multiple databases and to make sure we understand the form words in each context in which they appear. One curious thing would be to interrogate our own presentist biases. If I were to conduct research on what constituted “womanhood” in Russia by using digitized newspapers from World War II, what words would I search for? This is quite exciting. How about you, did you think of a way you may apply this knowledge in your research?
Thank you, Alex, for this great synthesis of Jessop’s and Guldi’s main ideas. I thought Jessop’s concern over the image literacy of humanities scholars particularly interesting. I’m wondering how facial recognition technology could be implemented in the same way to identify patterns in photographs in the same way that Guldi is using it to examine texts—and what that would entail.
Like Laura, I’m fascinated by the process of selecting which words to focus the data analysis on, particularly with words that have come to assume different meanings over time. I believe this has already been done to some degree, but it would be interesting to see a full study on the evolution of what “queer” has meant over the ages.