Exploring Corpus Stylistics through CLiC: A Digital Humanities Learning Experience
Digital Humanities has transformed the way we study literature by combining traditional literary analysis with digital technologies. Instead of relying only on close reading, digital tools allow us to explore language patterns, character interactions, and narrative styles in a more systematic and evidence-based way. As part of this activity, I worked with the CLiC Activity Book, which introduces students to Corpus Stylistics through hands-on exercises using the CLiC (Corpus Linguistics in Context) web application. I also explored Voyant Tools, a web-based text analysis platform that provides visual representations of textual data. Through Activities 17.1–17.5, I analysed speech clusters, compared fictional and real spoken language, examined Jane Austen's dialogue and narration, and gained practical experience in using digital tools for literary research. This blog presents my observations, interpretations, and learning outcomes from these activities. CLICK HERE FOR TEACHER BLOG
My Experience with the Human vs. Machine Poetry Test
To understand whether I could distinguish between human-written and AI-generated poetry, I took an online poetry test. The test included six poems, and I had to decide whether each one was written by a human or by a machine. Before checking the answers, I read every poem carefully. I focused on the emotions, word choice, theme, and the overall meaning instead of making quick guessesYOU CAN GIVE THE TEST
(1)(1).png)
Poem 1 ✅ Correct
I identified this poem as human-written, and my answer was correct. The poem expressed grief, memories, and personal emotions in a very natural way. References to the hospital, the mother, and loss made it feel like someone was sharing a real experience.
(1).png)
Poem 2 ❌ Wrong
I thought this poem was written by a human, but it was actually generated by a machine. The unusual images and emotional atmosphere made me believe it came from human imagination. This surprised me because the AI successfully created a poem that felt creative and meaningful.
Poem 3 ✅ Correct
I correctly identified this poem as human-written. The poem describes the growth of a small plant with gentle emotions and beautiful observations about nature. It had a smooth emotional flow that made it feel authentic.
Poem 4 ✅ Correct
This was another poem that I correctly identified as human-written. The language, rhythm, and ideas about love and hope seemed thoughtful and emotionally balanced, which made me think it was written by a person.
Poem 5 ✅ Correct
I guessed that this poem was written by a machine, and I was right. While reading it, I noticed that the words followed a pattern, but many lines did not connect naturally. The poem had the style of older poetry, but the ideas felt less meaningful and more mechanically arranged.
Poem 6 ❌ Wrong
For the last poem, I believed it was written by a machine, but it was actually written by a human. I thought the structure was too organized and predictable, which reminded me of AI-generated writing. However, the result showed that humans can also write in a clear and structured way. This taught me not to judge a poem only by its structure.
What I Learned
I scored 4 out of 6 in the test. This experience changed my understanding of AI-generated poetry. Before taking the test, I believed that emotional poems could only be written by humans. However, some AI-generated poems were so convincing that they fooled me. At the same time, I noticed that some AI poems still have patterns that make them feel more systematic than naturally expressive.
The activity showed me that it is becoming increasingly difficult to distinguish between human creativity and machine-generated writing. Instead of simply asking whether machines can write poems, we should now ask how closely AI can imitate human creativity and whether readers can still recognize the difference.
The CLiC Activity Book & Corpus Stylistics
The CLiC Activity Book introduces students to corpus stylistics through the CLiC (Corpus Linguistics in Context) web application. It demonstrates how digital tools can be used to analyse literary texts by exploring patterns of language, dialogue, and narration. These activities encourage students to combine close reading with corpus-based analysis.
For our group activity, we worked on "Fictional and Real Speech in Jane Austen and the 19C." Using the CLiC website, we explored speech patterns in Jane Austen's novels and compared them with the 19th Century Reference Corpus (19C). By analysing recurring phrases, quotations, and concordance lines, we observed how fictional dialogue differs from real language use and gained a better understanding of Jane Austen's distinctive style of writing.
What We Did:
Our group used the CLiC website to search for frequent speech clusters such as “I should like to know,” “What do you think of,” and “It seems to me that.” We examined concordance lines and compared fictional dialogue in Jane Austen’s novels with real spoken language represented in the BNC corpus.
Activity 17.1: Speech in the 19th Century Reference Corpus (19C)
The 19C corpus in CLiC contains twenty-nine nineteenth-century novels. By examining the most frequent speech clusters, we can understand how authors such as Jane Austen, Charles Dickens, George Eliot, Thomas Hardy, and others represented spoken language in fiction. Although these dialogues sound natural, they are more organized and polite than real-life conversations. Writers use speech not only for communication but also for characterisation, plot development, and expressing social relationships.
Interpretation
The list of the most frequent speech clusters in the 19th Century Reference Corpus (19C) shows that conversations in nineteenth-century novels were generally polite, formal, and carefully structured. Many clusters such as "I should like to know," "I should like to see," and "I should like to have" express requests or wishes in a respectful way. Clusters like "What do you think of" and "It seems to me that" are used to ask for opinions or express personal views politely. Others, such as "I am not going to", show determination or refusal, while "What am I to do" expresses worry or helplessness. The cluster "A quarter of an hour" is used to refer to time in the story.
These expressions reveal that dialogue in nineteenth-century fiction was written not only to sound natural but also to reflect good manners, social relationships, and the characters' emotions. Compared with modern spoken English, many of these expressions are still understood today, but they are usually used in shorter and more informal forms. For example, "I should like to know" becomes "I'd like to know," "It seems to me that" becomes "I think," and "I am not going to" is often shortened to "I'm not going to" or "I'm not gonna."
Comparison with Present-day Spoken English
| 19th-Century Expression | Modern Spoken English |
|---|---|
| I should like to know | I'd like to know |
| I should like to see | I'd like to see |
| I am sorry to say | Sorry to say... |
| What do you think of...? | What do you think of...? (still common) |
| It seems to me that... | I think... / I feel... |
| I am not going to... | I'm not gonna... / I'm not going to... |
| What am I to do? | What should I do? |
Many of these expressions are still understood today, but modern English is generally shorter and more informal.
Observations from Concordance
After examining the concordance lines in CLiC, I observed that these speech clusters are used in different situations such as asking questions, expressing opinions, making polite requests, refusing something, and showing emotions. The dialogues are generally formal and well-structured, reflecting the manners and social values of nineteenth-century society. Unlike real conversations today, the fictional speech is more polished and grammatically complete.
Activity 17.2 Comparing it seems to me that then and now
Comparison of the Concordance Lines of "it seems to me that" in the 19C Novels and the British National Corpus (BNC)
The cluster "it seems to me that" appears in both the 19th Century Reference Corpus (19C) and the British National Corpus (BNC) Spoken Corpus. By comparing the concordance lines, I observed both similarities and differences in the way this expression is used in fictional dialogue and real spoken English.
Similarities
- In both corpora, the expression is used to introduce the speaker's personal opinion or viewpoint.
- It helps the speaker sound polite and less direct.
- In both fictional and real conversations, the phrase expresses an opinion rather than a definite fact.
Differences in Form
- In the 19C corpus, the phrase appears in fictional dialogue. The sentences are longer, grammatically complete, and more formal.
- In the BNC Spoken Corpus, the phrase occurs in real conversations. The speech is more natural and includes pauses, repetitions, hesitation markers such as "erm", and incomplete sentences.
Differences in Function
- In the 19C corpus, the phrase is mainly used by characters to express opinions politely during conversations.
- In the BNC Spoken Corpus, the phrase is used in everyday discussions to explain ideas, express uncertainty, or share personal opinions in a natural conversational setting.
My Observation
After comparing the concordance lines, I noticed that the expression has the same basic meaning in both corpora. However, the fictional speech in the 19C corpus is more formal and carefully written, while the BNC examples reflect spontaneous and informal spoken English. This comparison helped me understand the difference between literary dialogue and authentic everyday speech.
The activity showed that "it seems to me that" functions as a polite way of expressing an opinion in both fictional and real speech. Although its purpose remains the same, its presentation changes according to the type of text. The 19C corpus contains polished literary dialogue, whereas the BNC Spoken Corpus reflects natural, everyday conversation
Activity 17.3: Speech in Jane Austen's Novels
1. The 15 Most Frequent Clusters in Quotations (from CLiC)
| No. | Cluster | Frequency |
|---|---|---|
| 1 | I do not know what | 41 |
| 2 | I am sure you will | 33 |
| 3 | I do not know that | 25 |
| 4 | I do not mean to | 19 |
| 5 | and I am sure I | 20 |
| 6 | I have no doubt of | 21 |
| 7 | I do not think I | 9 |
| 8 | I am sure I should | 14 |
| 9 | I am sure I do | 14 |
| 10 | I do not pretend to | 15 |
| 11 | quite out of the question | 13 |
| 12 | I do not know how | 17 |
| 13 | I am sure I do not | Not found |
| 14 | very much obliged to you | 24 |
| 15 | and I do not know | 12 |
2. Compare these clusters with those from the general corpus of 19th-century novels (19C). Do the lists overlap?
Yes, there is some overlap between Jane Austen's novels and the general 19th-century (19C) corpus.
Both corpora include conversational clusters such as "I do not know what" and "I do not know how," which express uncertainty and personal opinion. However, Jane Austen's novels contain more polite and formal expressions, such as "very much obliged to you," "I do not mean to," and several clusters beginning with "I am sure." These reflect the polite social conversations and manners that are typical of Austen's writing.
The 19C corpus, on the other hand, contains a wider variety of expressions because it includes novels by many different authors and genres.
3. In what contexts and with what function do you expect them to be used?
These clusters are mainly used in:
- Conversations between characters
- Polite social interactions
- Giving opinions
- Expressing uncertainty
- Disagreeing politely
- Refusing requests or suggestions
- Thanking other people
Their functions are:
- Showing politeness and good manners
- Softening criticism or disagreement
- Expressing doubt or uncertainty
- Giving reassurance
- Rejecting an idea politely
- Making conversations sound natural and realistic
4. Read through the concordances in Figures 43 and 44. Do the examples confirm your expectations for these two clusters?
Cluster 1: "I do not mean to"
Context
This cluster is used when a character wants to express an opinion or criticism without sounding rude.
Function
- Softens disagreement
- Shows politeness
- Gives an explanation before criticism
- Protects the listener's feelings
Example
"I do not mean to be impertinent."
The speaker politely introduces a criticism or opinion instead of speaking harshly.
Cluster 2: "Quite out of the question"
Context
This cluster is used when a character firmly refuses an idea or proposal.
Function
- Expresses strong refusal
- Rejects a suggestion
- Shows that something is impossible
- Ends the discussion clearly
Example
"Quite out of the question."
The speaker clearly states that the suggestion cannot be accepted.
Activity 17.4 The speech of Austen’s characters
The aim of this activity was to explore how Jane Austen's characters use recurring speech patterns in their conversations. Using the CLiC Concordance tool, I searched for some of the most frequent speech clusters from the previous activities and examined how they are used in different contexts.
"I do not know what"
After reading the concordance lines, I noticed that this expression is mostly used when characters express uncertainty, confusion, or hesitation. The speakers often use it when they are unsure about a person, situation, or decision. Rather than making direct statements, the characters speak carefully and politely, which reflects the conversational style of Jane Austen's novels
"I am sure"
The cluster "I am sure" appears frequently in conversations where characters express confidence, reassurance, or personal opinions. It is often used to support an idea or reassure another character without sounding rude or forceful. This expression also reflects the polite and respectful communication that is common in Austen's fiction.
My Experience
This activity helped me understand that the meaning of a phrase depends on its context. Before using the CLiC tool, I usually focused only on the storyline while reading novels. However, by examining the concordance lines, I could observe how Jane Austen repeatedly used certain expressions to develop conversations and relationships between characters. Seeing all the examples together made it easier to recognize language patterns that are difficult to notice during normal reading.
Activity 17.4 showed me that corpus stylistics is a useful way to study literary dialogue. By exploring the concordance lines in CLiC, I was able to understand how Jane Austen's characters use recurring expressions to express uncertainty, confidence, and politeness. This activity improved my understanding of both Jane Austen's writing style and the practical use of digital tools in literary analysis
The Non-quotes cluster list is clearly different from the Quotes cluster list. The most frequent clusters, such as "as soon as they were," "a quarter of an hour," "in the course of the," "she had been used to," and "the rest of the family," are descriptive rather than conversational. These expressions are used by the narrator to describe actions, characters, settings, and the sequence of events in the story. Many of the clusters contain third-person pronouns such as she, her, and they, showing that the narrator is telling the story from an outside perspective.
In contrast, the Quote clusters from the previous activity contain expressions such as "I do not know what," "I am sure," and "I do not mean to," which are used by the characters during conversations. These expressions include first- and second-person pronouns like I and you and are used to express opinions, uncertainty, reassurance, or polite disagreement.
This comparison shows that Jane Austen uses two different styles of language in her novels. The narrator's language is descriptive and helps explain events and characters, while the characters' speech is interactive and reflects everyday social conversations. By comparing the Quotes and Non-quotes subsets in CLiC, I was able to clearly see the stylistic difference between narration and dialogue in Austen's novels.
My Observation
After exploring the Non-quotes subset, I noticed that the recurring clusters mainly describe actions, time, and characters instead of direct conversations. Comparing these results with the Quotes subset helped me understand how Jane Austen separates the narrator's voice from the characters' speech. This activity showed me that corpus stylistics can reveal writing patterns that are not easy to notice through normal reading.
Another useful digital tool in the field of Digital Humanities is Voyant Tools. It is a free web-based text analysis platform that helps students, teachers, and researchers explore literary texts through visualizations and statistical analysis. By uploading one or more texts, users can examine word frequency, keywords, trends, and relationships, making it easier to identify patterns that may not be noticed through close reading alone.
Learning Outcomes
Through these activities, I achieved the following learning outcomes:
I learned the basics of corpus stylistics and how it combines literary analysis with digital tools.
I became familiar with the CLiC web application and learned how to search for speech clusters, concordance lines, and quotations in literary texts.
I understood the difference between fictional speech in nineteenth-century novels and authentic spoken English by comparing the 19C Corpus with the British National Corpus (BNC).
I observed how Jane Austen's characters use polite and formal expressions to express opinions, uncertainty, reassurance, and disagreement.
I learned to distinguish between the language of the narrator and the language of the characters by comparing the Quotes and Non-quotes subsets in CLiC.
I developed the ability to interpret concordance lines and identify recurring language patterns instead of relying only on close reading.
These activities improved my analytical thinking and showed me how Digital Humanities can make the study of literature more systematic, interactive, and evidence-based.
Works Cited
Barad, Dilip. "What if Machines Write Poems." Dilip Barad | Teacher Blog, 21 Mar. 2017, https://blog.dilipbarad.com/2017/03/what-if-machines-write-poems.html. Accessed 7 Aug. 2026.
Mahlberg, Micaela, et al. CLiC 2.1: Corpus Linguistics in Context. University of Birmingham and University of Nottingham, 2020. https://clic.bham.ac.uk/. Accessed 3 Aug. 2026.
.png)
.png)
No comments:
Post a Comment