Digital Humanities and the computational revolution: Four takeaways from our webinar with Prof. Myke Cuthbert | MIT J-WEL

You are here

Digital Humanities and the computational revolution: Four takeaways from our webinar with Prof. Myke Cuthbert

Myke Cuthbert and Kate Stringer

 

On August 6, J-WEL Higher Education held a webinar on "Digital Humanities and CS Education” with Michael Cuthbert, Associate Professor of Music and Faculty Director of Digital Humanities. Cuthbert led a discussion on how the field of digital humanities is blossoming under the computational revolution. Check out four takeaways from the webinar below! 

 

1)   Humanism and the sciences have more in common than many people realize.

Great storytelling is essential to being a good humanist. To tell great stories, humanists act a bit like scientists, gathering data and analyzing it. The humanities and sciences are also united by their “love of experimenting with new technologies.” Many humanists use technology to improve their research and pedagogical skills.

 

2)   Computers are adept at finding patterns and identifying outliers, but humans are the ones who need to tell the stories of why and how.

Computers can retrieve information, find patterns, and identify outliers in datasets. But humans are the ones who need to tell the stories—to interpret the data to identify the “why’s” and the “how’s,” and to come up with new and creative ways to tell them. For example, a human might write code that tells a computer to do something a single person cannot do, like scan one million twitter posts for far-right political rhetoric—and identify trends for support of far-right candidates in a particular country. But a human is needed to explain why support for far-right politicians is increasing or how the movement is spreading.

 

3)   Making the leap from machine learning to analysis and storytelling is not intuitive—it is a skill that needs to be learned.

Professor Cuthbert explained that students need to not only learn not only how to use computers to retrieve information and find patterns, but also how to analyze this data and tell stories with it. He explains how MIT’s new Programs in Digital Humanities, which connect programming and digital technology with research and teaching in the humanities, are accomplishing this:

"What we're planning on doing is having a mission to use code to encourage communication across humanities, computer science, and tech world divides and build a community of practitioners who are fluent in the language of the humanities and also of programming and computation.

"At MIT, one of the things that we thought about trying was how to take computer science students and people who are interested in programming and other technical skills, and what happens if we give them humanistic problems to apply these skills to. How would they understand how to read data within the humanities? And how would we understand how to answer a question or a narrative across computers-- across a humanistic discipline? So our aim over three years is to really start affecting how humanities research can be conducted at MIT but also what types of problems computer scientists would want to work on."

 

4)   Bringing in younger students can be a great way to gain a fresh perspective on problems considered by many to be “too difficult to solve.”

Cuthbert explained how he selected students to assist with a digital humanities research project:

"We thought the best way...to start was to grab a group of people who didn't realize that the problems that we're going to be working on were considered too hard to solve. That is to say, we wanted a group of people who are smart and talented but also naive enough to not know that what they were going to be doing was considered just impossible to succeed.

"So how did we do this? We started by getting a bunch of [first years]."

Thirty undergraduate students in Cuthbert’s program collaborated on a project analyzing the use of gendered language across 4,200 19th century novels and 327 million words in two months. They collaborated to write programs that could identify commonalities and differences in gendered writing and compare male and female authors’ use of language. The students found that in 19th century novels written in English:

"Male authors will [often] write books that just simply exclude women. Among the top 10 books by male authors in our data set, the average distance, the median distance between mentions of a female character, is 20,000 words [about 30 or 40 pages]…whereas even female authors writing primarily about female characters had at most 40 or 50 words between mentions of a male character. So there are very, very gendered distinctions."

Cuthbert noted that using computation in this way allows researchers to identify commonalities like these, as they involve examining millions of words together. The research conducted by the mostly first-year students was accepted for a presentation only open to graduate students. “They were able through this work to produce results that were of interest at the level, and of sophistication at the level that graduate students generally do,” noted Cuthbert. The students created a website called gendernovels.digitalhumanitiesmit.org. Their work was also covered in Ms. Magazine.
 

J-WEL members and registered users of the J-WEL site can watch the full webinar (login required).

 

Top photo: Prof. Cuthbert and Kate Stringer, resident J-WEL musicologist, after the webinar