Categories
Pedagogy UVA Collaboration

My Experience Leading a Workshop on Text Analysis at Washington and Lee University

[Enjoy this guest post by Sarah McEleney, doctoral candidate in the Slavic Languages and Literature Department at University Virginia. She came to W&L to give a workshop in Prof. Mackenzie Brooks’s DH 102: Data in the Humanities course through a Mellon-funded collaboration with the Scholars’ Lab at UVA. More information about this initiative can be found here. This post is cross-listed on the Scholars’ Lab blog.]

As a graduate student participating in the University of Virginia and Washington & Lee University digital humanities collaboration, during the fall 2017 I led a guest workshop on text analysis in Mackenzie Brooks’ course DH 102: Data in the Humanities. This workshop was an exploration of approaches to text analysis in the digital humanities, which concurrently introduced students to basic programming concepts. For humanities students and scholars, the question of how to begin to conduct text analysis can be tricky because platforms do exist that allow one to perform basic text analyses without any programming knowledge. However, the ability to write one’s own scripts for text analysis purposes allows for the fine-tuning and tailoring of one’s work in highly-individualized ways that goes beyond the capabilities of popular tools like Voyant. Additionally, the existence of a multitude of Python libraries allows for numerous approaches for understanding the subtleties of a given text of a corpus of them. As the possibilities and directions for text analysis that Python enables are countless, the goal of this workshop was to introduce students to basic programming concepts in Python through the completion of simple text analysis tasks.

At the start of workshop, we discussed how humanities scholars have used text analysis techniques to create some groundbreaking research, such as Matthew Jockers’ research into the language of bestselling novels, as well as the different ways that text analysis can be approached, briefly looking the online text analysis tool, Voyant.

For this workshop students downloaded Python3 and used the simple text editor that is automatically installed with it, IDLE. This way we didn’t have to spend time downloading multiple programs. While IDLE is rather barebones, its functionality as a text editor is fine for learning the basics of Python, especially if one doesn’t want to install other software. From here, by using a script provided to the students, we explored the concepts of variables, lists, functions, loops, and conditional statements, and their syntax in Python. Using these concepts, we were able to track the frequency of chosen words throughout different sections of a story read by the script.

The workshop then delved into a discussion of libraries and how work can be enhanced and made to better suit one’s needs by using specific Python libraries. As the focus of the workshop was on text analysis, the Python library that we looked at was NLTK (Natural Language Toolkit), which has a vast variety of functions that aid in natural language processing work, such as word_tokenize() and sent_tokenize(), which break up a text into individual parts, as words or sentences, respectively. The NLTK function FreqDist() simplifies the task of getting a count of all the individual words in a text, which we had done with Python alone in the prior script before working with NLTK. The inclusion of NLTK in the workshop was meant to briefly show students how important and useful libraries can be when working with Python.

While only so much can be covered over the course of a single workshop, the premise of the workshop was to show students that you can do some very interesting things with text analysis with basic Python knowledge, and to dive into Python programming headfirst while learning about general concepts fundamental to programming. As digital humanities methods for humanities research are becoming more and more common, working with Python’s capability for natural language processing is a useful tool for humanists, and in an introductory class, the goal of my workshop was to spark students’ interest and curiosity and provide a stepping stone for learning more, and at the end of the workshop, further resources for students to turn to in learning more about Python and text analysis were discussed.

Categories
Announcement

We’re hiring a Digital Scholarship Librarian!

Yes that’s right, we’re looking for a Digital Scholarship Librarian to join DH @ WLU. As a member of a library team emphasizing digital humanities, data, and collections in the liberal arts, the Digital Scholarship Librarian will have a role in advancing digital scholarship writ large at Washington and Lee. This is a position for an individual with wide-ranging interests in digital initiatives.

Please share with anyone who might be interested!

Learn more and apply on Interfolio.

Categories
Announcement Event on campus Speaker Series UVA Collaboration

Day of DH @ Fall Academy 2017

While it’s still a little scary to admit that school will be starting in a month, we’re excited about this year’s Day of DH! Join us for a morning of pedagogy and digital scholarship discussion from some of your favorite faculty members. We’re thrilled to have Amanda Visconti (Managing Director of Scholars’ Lab) joining us for the lunch time talk. And don’t forget, there’s the third annual Library/ITS mixer in the afternoon.

Sign up for these sessions and check out all the great offerings in Fall Academy event manager.

9:00-10:00am Breakfast and Mellon and You: Graduate Student Teaching Fellows
Interested in the latest updates on the Digital Humanities grant from Mellon, including pedagogical and research opportunities? Paul Youngman (Professor of German and Chair of the Digital Humanities Committee) explains! Curious about the Graduate Student Teaching Fellows and how you could leverage a UVA graduate student in your class? Hear from Caleb Dance (Assistant Professor of Classics), Suzanne Keen (Dean of the College, Professor of English), and Taylor Walle (Assistant Professor of English) on their experiences with teaching fellows, what worked, what didn’t, and what students and faculty learned.
10:15-11:45am Incentive Grant Winners: WRIT 100
Last year’s Mellon DH incentive grant winners focused on multi-modal composition in their WRIT 100 courses. This group faced unique challenges and opportunities in working with first-year students in a variety of topics. Hear about their experiences, with time to discuss ideas further. Speakers: Sydney Bufkin (Mellon Digital Humanities Fellow, Visiting Assistant Professor of English, Topic: Romantic Comedy), Genelle Gertz (Professor of English, Topic: Faith & Doubt), Sascha Golubuff (Professor of Anthropology, Topic: Terror & Violence), Wan-Chuan Kao (Assistant Professor of English, Topic: The Good Wife), and Kary Smout (Associate Professor of English, Topic: Whole New World).
12:00-1:30pm Digital Humanities Guest Speaker Amanda Visconti: Community Design Takes Time
Amanda Visconti avatar
As we experiment with virtual ways of connecting digital humanities practitioners, what kind of human effort must we invest? I’ll share two recent projects I’ve worked on to explore the (sometimes hidden) work of designing DH communities and supporting the very real humans who make up these groups. Infinite Ulysses is a participatory digital edition of James Joyce’s novel Ulysses. A variety of scholarly methods—design, coding, usertesting, blogging, and social science analysis—combined to try out a virtual space for conversations among readers from within and without the academy. The Digital Humanities Slack is an online forum for chatting about all aspects of DH. Over 1,500 people interested in DH from around the world are members, and anyone can join, regardless of experience or affiliation. I’ll use these projects to discuss what has and hasn’t worked for me in audience-nurturing DH projects, and how those experiences are shaping my part in the trajectory of the Scholars’ Lab.
4-5:30pm ITS and Library Mixer for Faculty and Staff
Attend the ITS & library reception for a fun and informative kick-off to the academic year. Meet the librarians and ITS staff and learn about our many resources and services. A wide variety of refreshments will be served. Eat, drink, and be merry with us! Hors d’oeuvres and tea will begin at 4:00. Beer and wine will be served from 4:30-5:30. This event will take place on the main level of Leyburn Library.
Categories
Announcement

Now Hiring: Director of Data Education

We’re hiring a Director of Data Education! Please see the complete job ad for more details. Review of applications will begin July 5 and continue until the position is filled.


Washington and Lee University Library invites applications for Director of Data Education, a non-tenure track faculty appointment, to establish a program to:

• engage faculty and students in incorporating statistical and computational data analysis methods (data science) into the undergraduate curriculum and scholarship;
• provide guidance for students in finding data sets as well as teaching students how to clean and manipulate data for use in analytical and statistical applications;
• offer peer tutoring to students who need assistance with data and statistical applications.

The program will make a significant contribution toward meeting the increasing demand for data-intensive courses as well as promoting data and statistical literacy across the curriculum. We are interested in candidates from all areas of statistical sciences and analytics, in the broadest possible sense, to teach skills with broad, real-world application.

Categories
DH Pedagogy UVA Collaboration

Why To Teach Students to Not-Read Novels

[Enjoy this guest post by James Ascher, doctoral candidate in the English Department at University Virginia. He came to W&L to give a workshop in Prof. Taylor Walle’s ENGL 335 course through a Mellon-funded collaboration with the Scholars’ Lab at UVA. More information about this initiative can be found here. His post is cross-listed on the Scholars’ Lab blog.]

This post has a simple argument: if you teach novels, you should teach students to not-read novels. Now, before you get concerned, I’m not arguing against teaching literature or avoiding novels altogether; the hyphen in “not-read” means a method, not a rejection of reading. Indeed, my whole argument is based on the idea that students need help returning careful attention to texts, but faced with a deluge of texts, we teachers ought to show them how some professional literary critics not-read, or as others call it “distant read.”

What is a novel anyhow? A reasonable question to ask your students as a semester goes along, since it seems to be a long form of prose that comes to dominate what we now consider literature. If one is to believe Amazon Rankings, then the most read forms of electronic books and the most purchased books remain novels (at least when I wrote this, but I’d be surprised if it changes). This wasn’t always the case and part of the history of the novel is its commercial success. Franco Moretti makes the case clear in his Graphs, Maps, and Trees where he argues that the rise of the novel can be studied using charts, as he does in his previous work Atlas of the European Novel (Moretti’s response, which is more useful.) Building on the idea of charting large-scale phenomena over time, he notices that novels rise (fig. 1).

Fig. 1 rises of novels from Graphs, Maps, and Trees

Fig. 1 rises of novels from Graphs, Maps, and Trees

Something about the form, the markets, the people, or something else means that this particular literary form comes to dominate, so whatever time period you consider, it’s worth considering what was going on.

Its well established that the English novel rose first—the form seems to have been invented there—so even in a non-English classroom, it’s worth considering how those novels were imported into a broader literary discourse. Luckily, the text of most eighteenth-century English novels are freely available online. We have a fortuitous confluence of the intellectually important materials that have become technologically available (a careful reader will note that this is one possible explanation of the novel’s rise as well). But, how can you study the features of a whole genre? An easy way is to read them by putting them on your syllabus, which I encourage, but after reading them you can look back at the whole syllabus and chart the topics that come up.

Fig. 2 Novel topics

Fig. 2 Novel topics

To test this idea, I presented for Taylor Walle in her English 335 “Radical Jane: the Politics of Class, Gender, and Race in Austen’s ‘Polite’ Fiction.” Her course asked students to think about how English novels formed identities and related to the growing issues of British society. It seemed like a great chance to try some topic-modeling across her entire syllabus, the chart was produced with ten topics across the whole class (fig. 2). You can see the lesson here if you want to reproduce the work.

After demonstrating how the method works, we turned to this chart and looked for topics that crossed the texts. The works read for the class and on the chart—from left to right—are Emma, Northanger Abbey, Pride and Prejudice, Sense and Sensibility, A Sentimental Journey, A Sicilian Romance, and A Vindication of the Rights of Woman. You can see, and the class immediately saw, that the topics broke at the boundaries of books. You can see that many of the topics to detect specific books, but a few cross boundaries. Now, the topic used in topic modeling isn’t quite the normal sense of the word “topic.” It means a list of words with probabilities that, when they occur, signal the topic that is that list of words. The topics by their top words are,

time made heart letter moment feelings mind spirits happiness
present long felt thought affection left hope return day love
situation ...

elizabeth darcy bennet jane bingley wickham collins mrs sister
lydia catherine lady lizzy longbourn gardiner father family
netherfield kitty charlotte ...

miss mrs good great dear young make time room house give day
thought friend heard man home replied pleasure hear ...

julia marquis door ferdinand madame hippolitus castle heart duke
heard marchioness length appeared light night discovered time part
count scene ...

man life love woman character mind world society sense opinion
great beauty good present taste nature understanding husband
degree subject ...

emma harriet weston mrs knightley elton thing jane woodhouse miss
fairfax frank churchill body hartfield bates highbury father sort
harriet's ...

fleur paris monsieur poor hand count thou man set told madame good
french thy heart lady tis put made nature ...

elinor marianne mrs dashwood edward jennings sister willoughby
colonel lucy john mother thing brandon ferrars barton middleton
marianne's lady town ...

catherine tilney isabella thorpe morland allen general henry bath
eleanor catherine's brother james father street hour northanger
abbey john captain ...

women men reason virtue sex respect mind duties affection make
heart children power render human virtues true allowed till duty
...

As far as the course goes, the first topic seems to cover what all these texts have in common, but notice everything isn’t perfectly lined up by novel. The topic beginning “julia marquis door” clearly comes from Sicilian Romance, but also hits on some later chapters of Northanger Abbey. Why would that be? Well, if you know the texts, you realize that some of the same Gothic themes occur in both texts and they use the same words.—“Dear students, can anyone bring us to where in the text this happens?” And, we enter the realm of the normal literature classroom.

By presenting a broad view of the texts, built by a computer algorithm, but out of the words of the text, we invite students to re-read works. Not-reading becomes re-reading and presenting words across the entire corpus puts students into partnership with technology to ask what it is about the form of novelistic prose that makes it popular and speak to social issues. Furthermore, we also encourage students to be critical of the results of computerized analysis. Several students noted that these topics were obvious, having read the works, and that they could have come up with them by hand, which is—of course—true. It’s easy to scale these up beyond what you could do by hand, but seeing how they reflect what is accurately in the text shows that they provide some purchase on truth and also suggests what might be going on with other computerized analyses. One way we imagined it was that the computer applied an obvious rule at a fine level of detail. If we follow the same method, but only for Emma by paragraph, we get a much messier chart (fig. 3), but seeing that chart, students can begin to engage with both literary texts and computers that help them to not read—to ask what it means and what can be done with it.

Emma by chapter

Fig. 3 Emma by paragraph

Categories
Announcement DH

Announcing Our New Mellon DH Fellow: Sydney Bufkin

Sydney Bufkin Headshot
We are happy to announce our new Mellon DH Fellow, Sydney Bufkin. Sydney is a familiar face for many of us at W&L as she’s been teaching in the English Department for the last few years. Sydney received her PhD in English from the University of Texas at Austin with a specialization in nineteenth-century American literature and reception studies. At W&L, Sydney has taught a range of courses in the English Department and Writing Program. She specializes in digital approaches to pedagogy and has received a DH incentive grant for her multi-modal writing assignments. Her interest in computational approaches to literature manifests itself in her own research on a corpus of reviews of nineteenth-century purpose fiction and hopefully in future DH courses!

Sydney will being on June 12. Please welcome her to her new role in Leyburn Library.

Categories
Event on campus Speaker Series

DH Speaker Series: Jeffrey Witt

Join us for the final DH Speaker of the 2016-2017 academic year. Jeffrey Witt, Assistant Professor of Philosophy at Loyola University Maryland, will be speaking on “The Scholastic Commentaries and Texts Archive: Reconceiving the Medieval Corpus in a Linked Data World.”

Monday, May 8, 2017
12:15-1:15pm
Hillel 101
Lunch provided, please register.


photo of jeffrey witt
Jeffrey Witt is an assistant professor of philosophy at Loyola University Maryland. He is the founder, designer, and developer of the Scholastic Commentaries and Texts Archive and the LombardPress-Web publication system. He is working on several editions of previously unedited Latin texts, aiming to make them freely available and searchable on the web. He sits on the advisory board of the Digital Latin Library and is co-chair the IIIF Manuscript Community Group. In 2016, he was awarded a Visiting Research Fellowship at the Schoenberg Institute for Manuscript Studies at the University of Pennsylvania to develop TEI transcriptions of the Sentences commentary of William de Rothwell and to incorporate those transcriptions into the Scholastics Commentaries and Texts Archive. Jeffrey Witt completed his graduate work in the philosophy department at Boston College in the spring of 2012. His dissertation focused on issues of faith, reason, and theological knowledge in the late medieval Sentences commentaries. He is the co-editor of The Theology of John Mair (Brill 2015) and the co-author of a monograph on the 14th century philosopher and theologian Robert Holcot (Oxford University Press, 2016).

This program made possible by the Andrew W. Mellon Foundation, Italian Studies, and Medieval and Renaissance Studies

Categories
Announcement Project Update Undergraduate Fellows

Work in Progress: Updates from Our DH Fellows

Join us on April 4th from 2-4pm to hear project updates from our current Mellon DH Undergraduate Fellows. If you’re interested in becoming a fellow next year, this is the perfect chance to learn what it’s all about!

Applications are now open for the 2017-2018 academic year.

We’ll be in the DH Workspace (Leyburn 218). There will be snacks.

Categories
Announcement DH

DH Courses Coming This Fall

Though most of us can’t see beyond winter term, fall registration is almost here! We have two DH courses on offer this fall. Check them out!


DH 102: Data in the Humanities

This course introduces students to the creation and visualization of data in humanities research. The course is predicated on the fact that the digital turn of the last several decades has drastically changed the nature of knowledge production and distribution. The community and set of practices that is digital humanities (DH) encourages fluency in media beyond the printed word such as text mining, digital curation, data visualization, and spatial analysis. Readings and discussions of theory complement hands-on application of digital methods and computational thinking. While the objects of our study come primarily from the humanities, the methods of analysis are widely applicable to the social and natural sciences. Three unit-long collaborative projects explore the creation, structure, and visualization of humanities data. This course meets in two-hour blocks to accommodate a lab component.

Prof. Mackenzie Brooks // TR 2:30-4:30pm // SC FDR // course website

DH 110: Programming for Non-Programmers

Computer science and IT graduates are no longer the only people expected to have some knowledge of how to program. Humanities and social science majors can greatly increase their job prospects by understanding the fundamentals of writing computer code, not only through the ability itself but also being better able to communicate with programming professionals and comprehending the software development and design process as a whole. The most centralized and simple platform for learning is the Web. This course starts with a brief introduction to/review of HTML and CSS and then focuses on using JavaScript to write basic code and implement preexisting libraries to analyze and visualize data. Students become familiar with building a complete Web page that showcases all three languages.

Prof. Jason Mickel // MW 2:30p-3:55p, R 8:35a-10:00a // SC FDR // course website

Categories
Announcement DH Summer Research

Mellon Summer Research Grant Recipients Announcement

We are pleased to announce the recipients of our Mellon Summer Research Grants. These faculty members will work with one or more students on DH research projects during the the summer of 2017. We are thrilled to have such an amazing group of newcomers and veterans and look forward to tracking their progess.


“The Lost Clew Archives at the Institute for Clew Studies: A Digital Research Facility" – Clover Archer, Director of the Staniar Gallery, and MC Greenleaf ’19

“Florence As It Was: the Digital Reconstruction of a Medieval City" – George Bent, The Sidney Gause Childress Professor of Art, Aidan Valente ’19, and Sam Joseph ’19

"The Evolving Language of Conscious Capitalism" – Drew Hess, Associate Professor of Business Administration, and Cassidy Fuller ’18

"The Social Context of Ethical Leadership" – Megan Hess, Assistant Professor of Accounting, and Alex Farley ’19

"Steinheil: Sex, Scandal and Politics in Belle Époque France" – Sarah Horowitz, Associate Professor of History, and Megan Doherty 19

Additionally, we are funding the summer research of the following students:

  • Nathan Brewer ’19 and Skylar Prichard ’19 – "Ancient Graffiti Project" (Benefiel)
  • Arlette Hernandez ’18 – "Quantifying Best American Comics" (Gavalar)