Digital Humanities Lab 1: From Machine Poetry to Mining Dickens

Bots, Books, and Big Data: My First Foray into Digital Humanities

Welcome to my academic blog! This post documents my practical exploration of Digital Humanities (DH) as part of my coursework at the Department of English at Maharaja Krishnakumarsinhji Bhavnagar University.

Traditionally, our literary studies rely heavily on manual close reading and theoretical frameworks. However, this recent Lab Session, guided by Dilip Barad Sir, introduced a completely different methodology: using computational tools and big data to analyze literature.

Here is the detailed infograph of this blog:

In this post, I will walk you through my first hands-on experience bridging the gap between technology and the humanities. Below, you will find:

• The AI Poetry Debate: My experience taking an interactive test to see if algorithms can truly replicate human poetics and emotional resonance.

• Corpus Stylistics with CLiC: A step-by-step visual analysis detailing how I used the CLiC database to uncover structural and psychological traits of the character Mr. Dick in Charles Dickens's David Copperfield.

• Pedagogical Outcomes: The key learning outcomes my group members and I discussed regarding how these digital methodologies complement our traditional literary frameworks.

Let’s dive into how algorithms and databases are reshaping the way we read and analyze texts.

Here is the Mind Map : Click Here

Part 1: The Machine Poetry Debate – Aesthetics vs. Algorithms

Our initial thinking activity centered around a provocative question inspired by Dilip Barad Sir’s blog: Can machines write poetry? This immediately brings up complex questions regarding aesthetics, emotion, and authorship. Can an algorithm truly replicate the emotional resonance or the complex poetics crafted by a human mind?

Can a computer write poetry? | Oscar Schwartz

To test this, I took the interactive NPR "Bot or Not" poetry quiz.

The Experience: I approached the interactive quiz expecting the mechanical verses to be distinctly obvious perhaps lacking in semantic depth or emotional warmth. Instead, the AI-generated poems successfully mimicked structural rhythms, classical forms, and stylistic tropes, making the quiz surprisingly difficult.

The Understanding: While I eventually managed to identify the human authors, the sheer structural competence of the bots was a humbling experience.

It forces a re-evaluation of what constitutes 'creativity' in the digital age. It proves that while machines might not experience the existential crises of the authors we study, they are highly capable of coding its vocabulary and mimicking its structure.

Part 2: Corpus Stylistics with CLiC – Tracing Mr. Dick in David Copperfield

The most rigorous and revealing part of the lab involved applying Corpus Linguistics in Context (CLiC) to Charles Dickens’s David Copperfield. CLiC allows us to perform macro-level analysis, shifting from reading a single page to querying an entire textual database. I chose to track the character of Mr. Dick, whose unique narrative quirks offer perfect material for computational stylistics.

Here is the step-by-step breakdown of my methodology and findings:

1. Establishing a Baseline Frequency 

I initiated a concordance search in the "All text" subset. Searching the root "Dick" yielded 229 entries. 

which I then refined to the specific proper noun "Mr. Dick," returning a targeted 208 entries. 

This established the sheer volume of his narrative footprint within the novel.

2. Isolating the Narrative Voice 

To understand how he is presented, I filtered the search for "Non-quotes," which narrowed the results to 178 entries.

• Analysis: This data point is crucial. The high frequency of non-quotes reveals that the vast majority of Mr. Dick's presence is mediated through the narrator's external descriptions rather than his own dialogue, positioning him as a highly observed character.

3. Analyzing Textual Pacing and Pauses 

Running the search within "Long suspensions" returned 12 entries 

• Analysis: In structural narrative theory, these suspensions often indicate deliberate pacing choices. They highlight moments where Dickens pauses the conversational flow to emphasize a physical gesture, a psychological hesitation, or a lingering silence surrounding the character.

4. Quantifying Psychological Motifs 

The true analytical power of CLiC became evident when I filtered the concordance rows by specific anatomical keywords to test character traits.

• Filtering by the word "head" produced 16 entries. 
  
  
  
  This data perfectly maps onto Mr. Dick's defining psychological obsession with King Charles I's severed head.
  
  

• Conversely, filtering by "mouth" yielded just 1 entry. 

By contrasting these two filters, the tool allows us to visually graph a character's defining psychological motifs using hard textual data.

Part 3: Group Discussions and Collaborative Learning Outcomes

Because this CLiC activity was a collaborative effort, the three members of our group sat down to discuss the implications of our findings. Comparing our traditional reading experiences with this new computational data led us to several unified learning outcomes:

• Empirical Evidence for Character Studies: We realized that digital tools offer hard, objective data to back up subjective literary theories. Finding exactly 16 references to Mr. Dick's "head" transforms a casual observation about a character motif into a solid, quantifiable academic argument.

• The Learning Curve of Distant Reading: We all agreed that shifting from traditional "close reading" (turning the pages of a physical book) to "distant reading" (manipulating databases and filters) requires a completely new analytical skill set. Navigating the exact parameters of "short" versus "long" suspensions took collaborative trial and error to fully grasp.

• Complementary Methodologies: Our group concluded that databases like CLiC do not replace traditional reading; they act as a magnifying glass. While human intuition is required to understand the emotional weight or comedy in Dickens's work, the software is unmatched at quickly locating the structural patterns that build those themes.
  
  

Personal Reflection: My First Foray into Digital Humanities

Reflecting on this entire lab session from grappling with the philosophical implications of AI poetry to tracking character data in CLiC my biggest takeaway is how technology expands our analytical horizons. I have always been deeply interested in structural narrative theories, and seeing those abstract textual structures mapped out in real-time through concordances and frequencies was incredibly validating. It bridges the gap between the art of literature and the science of data.

Conclusion:

Ultimately, this first step into Digital Humanities has proven that algorithms, corpora, and big data have a vital role to play in modern academic study. Whether it is an algorithm attempting to write a sonnet or a database highlighting Dickens's narrative habits, these digital tools force us to ask better, more precise questions about the texts we read. I am excited to see what other invisible literary patterns we can bring to light as we continue our coursework.

Here is the Slide Deck of this blog:

Here is the Video Overview of this blog:

References:

• Barad, Dilip. "What if Machines Write Poems." Dilip Barad | Teacher Blog, 21 Mar. 2017, blog.dilipbarad.com/2017/03/what-if-machines-write-poems.html.
• Mahlberg, Michaela, et al. CLiC: Corpus Linguistics in Context. University of Birmingham, clic.bham.ac.uk. Accessed 3 Aug. 2026.
• Schwartz, Oscar. "Can a computer write poetry? | Oscar Schwartz." YouTube, uploaded by TED, 10 Feb. 2016, youtu.be/UpkAqPEcMyE?si=YUhKv1pXK9xl9sbh.