Most network analyses of narrative texts have focused on character interactions, often limiting their scope to social relationships as envisioned by social network analysis. This paper, however, presents a network analysis of the narrator and the main character in F. Scott Fitzgerald's The Great Gatsby, expanding on my previous research that examined characters as networks of words within a dramatic narrative. I conceptualize the narrator and characters as lexical networks derived from the novel's dialogues and narration. A "symptomatic reading" of a character's speech network uncovers hidden aspects of that character, such as Gatsby's obsessive desire for Daisy and fixation on the lost past. Furthermore, analyzing a character's ego network within the narrator's narration reveals how the narrative voice understands and portrays that character. Specifically, Gatsby's ego network exposes the narrator's preoccupation with physical appearances, his subtle male gaze, his speculation about Gatsby's mysterious past, and his narrative strategy to mythologize Gatsby through temporal and spatial movements. Finally, the bipartite network between the narrator and the character, mediated through shared words, illustrates points of convergence and divergence, emphasizing the stark contrast between Gatsby as a character and Nick as the narrator. This study demonstrates how computational literary criticism can contribute to digital humanities by providing a refined examination of literary texts while creatively employing digital methodologies.
Most network analyses of narrative texts have focused on character interactions, often limiting their scope to social relationships as envisioned by social network analysis. This paper, however, presents a network analysis of the narrator and the main character in F. Scott Fitzgerald's The Great Gatsby, expanding on my previous research that examined characters as networks of words within a dramatic narrative. I conceptualize the narrator and characters as lexical networks derived from the novel's dialogues and narration. A "symptomatic reading" of a character's speech network uncovers hidden aspects of that character, such as Gatsby's obsessive desire for Daisy and fixation on the lost past. Furthermore, analyzing a character's ego network within the narrator's narration reveals how the narrative voice understands and portrays that character. Specifically, Gatsby's ego network exposes the narrator's preoccupation with physical appearances, his subtle male gaze, his speculation about Gatsby's mysterious past, and his narrative strategy to mythologize Gatsby through temporal and spatial movements. Finally, the bipartite network between the narrator and the character, mediated through shared words, illustrates points of convergence and divergence, emphasizing the stark contrast between Gatsby as a character and Nick as the narrator. This study demonstrates how computational literary criticism can contribute to digital humanities by providing a refined examination of literary texts while creatively employing digital methodologies.
Network analysis of fictional narratives has been mostly focusing on character interactions. Characters are treated as minimum units of social relationships in literary worlds they inhabit. Their networks then reveal the synchronic and diachronic structures, and the social dynamics and temporal plot of a narrative. Labatut and Bost (2019) thoroughly examine almost every research on this topic over the past 20 years in “Extraction and Analysis of Fictional Character Networks: A Survey.” They show that character interactions have been extracted in several different ways including co-occurrence, conversations, mentions, actions, and affiliations. What is striking is that all these approaches share the same premise that a character is the singular entity of analysis as in social network analysis, hampering a richer understanding of the complexities of a literary character. Korean scholars are also in the same vein as these studies theoretically and methodologically (Won 2023; H. Kim 2024). My previous research, however, has recently argued that structuralist narratology provides a way out of this impasse, suggesting a literary character as a lexical network in “Character as a Web of Words: Towards a Network Theory of Narrative” (2024).
On the other hand, the question of the narrator has not been seriously raised in terms of network analysis of narratives. Elson, Dames, and McKeown's (2010) essay “Extracting Social Networks from Literary Fiction”, for instance, only slightly touches on this issue. They are primarily concerned with extracting social networks of characters based on their conversations, only to find that the first person narrator, “I” occupies the overwhelming portion in character networks (section 5.2). Their limitations are clear. They simply view the narrator in terms of conversational character interactions. As long as it is regarded as an indivisible unit of social relationship like a character, network analysis cannot help but impoverish the multi-dimensional depth of the narrative voice. I argue that the narrator should be considered as a network of words extracted from its narrations, not merely as a monolithic agent of social interaction.
A novel poses a significant challenge for network analysis of literature. A drama is almost entirely composed of dialogues between characters. However, the predominant presence and saturation of the narrator in non-dramatic narratives, especially novels, demand a much more refined appreciation of the narrative voice than a drama. My essay tackles the question of the narrator as well as that of characters in a novel, building on the above mentioned study that has explored the possibility of looking at characters as a network of words in a dramatic narrative. In other words, I read characters and the narrator as lexical networks extracted from dialogues and narrations of a fiction. As it turns out, a "symptomatic reading" of a character's speech network is quite an effective strategy to uncover the hidden aspects of the character. I also suggest that an ego network of the narrator's narration illustrates very well how the narrative voice understands and presents a character. And finally, the bipartite network between the narrator and a character mediated through the shared words shows how they converge and diverge with each other, revealing the narrator's interpretive strategies to construct the character. F. Scott Fitzgerald's The Great Gatsby (1925) is chosen to illuminate the dynamics of the narrator and a character, exemplifying network analysis of the first-person narrative.
In a novel, the text comprises both the narrator's narration and the characters' dialogues. Dialogues are conveyed indirectly through the narrator's voice or presented in an unaltered form as direct speech. However, even direct speech is not entirely free from the narrator's mediation, as the narrator retains the authority to selectively offer such speech. In certain instances, the narrator may even reshape direct speech to serve their own interests. When the narrator is also a character, as in The Great Gatsby, their choice of words and phrasing may subtly reflect a subjective interpretation of other characters. Despite this potential influence, direct speech remains the least mediated and most reliable source of insight into a character. Accordingly, this study extracts each quoted segment and attributes it to its respective speaker to better understand character traits through network analysis.1
I propose "symptomatic reading" as a hermeneutic method to uncover the various, and even hidden, aspects of a character's speech. Characters do not always reveal their innermost thoughts when conversing with others. At times, they conceal their true selves and present a social persona a mask worn in social interactions. While moments of honesty may occur, readers must remain open to the possibility that what is evident in speech may not fully align with the character's complex nature. Therefore, it is essential to adopt a reading strategy that accounts for the discrepancy between what is shown on the surface and what lies beneath. Symptomatic reading, a concept rooted in the psychoanalytic tradition, enables a nuanced understanding of the character by identifying latent symptoms within network graphs of words.2 This approach is uncommon in quantitative analysis. Computational literary criticism, I argue, should complement its focus on quantitative aspects with a more refined interpretive method, such as symptomatic reading, in order to fully appreciate the subtleties of literary works.
To identify the narrator's voice and analyze its traits in a novel, it is necessary to isolate the narrations, specifically the parts of the text that belong to the narrator rather than characters. Narration includes all text outside of dialogue, encompassing descriptions and commentary that are not directly voiced by characters.3 The most straightforward approach is to extract the text while excluding the dialogues. This method works even for the sentences that simply report direct speech by a character. For example, in the line, "'How do you get to West Egg village?' he asked helplessly," the question itself is part of the character's dialogue, whereas "he asked helplessly" constitutes the narrator's words. In such cases, the reporting phrase "he asked helplessly" reflects the narrator's perspective, adding a layer of description and evaluative judgment. These narrative remarks often carry the narrator's subjective views, subtly projecting their perspective onto characters and events. Thus, extracting the parts of the text that exclude dialogues can be justified to identify and analyze the narrative voice.
The whole process of creating datasets for the analysis of both characters and the narrator, as well as conducting network analysis and visualization, is as follows.4 I began by retrieving the text of The Great Gatsby from Project Gutenberg, removing extraneous elements such as headings and endnotes. Dialogue segments were then extracted based on quotation marks, with manual review to include any overlooked segments and exclude quotations that are not speech, as explained in footnote 2. Each piece of dialogue was further examined to identify both the speaker and the primary listener, the direct recipient of the speech. The remaining text, with dialogue excluded, was designated as narration. Using this segmented data, I created adjacency matrices on the assumption that words appearing within the same sentence have a relationship to each other. Then, I analyzed and visualized the lexical networks of character dialogues and narration using Gephi software.5
This paper examines character traits as they emerge through dialogue, with a particular emphasis on Gatsby. Additionally, by analyzing both Nick's dialogue and narration, I explore Nick's dual role as both character and narrator. The ego network of Gatsby constructed from Nick's narration makes it possible to highlight the narrator's own understanding of the particular character in the context of narrative strategy of interpretation. By comparing the character's network of spoken dialogue with that of Nick's narration, then, we can clearly recognize the differences between the two. Finally, the bipartite network of narrator-word and character-word relationships reveals the strong contrast between them by showing the shared and unshared words, and thus the lexical convergence and divergence between the narrator and a character.
The co-occurrence network graph of Gatsby's dialogue displays the overall distribution of his interconnected character traits. It is no surprise that his signature phrase, "old" "sport," takes central positions, reflecting defining elements of his speech style.6 Other words, such as "well," "see," and "thought," also illustrate his habitual way of speaking; for example, "see" frequently appears in phrases like "you see," while "thought" often takes the form "I thought." These recurrent expressions dominate Gatsby's dialogue network. Notably, however, terms that reflect Gatsby's deeper, underlying traits appear not in the network's center but in peripheral clusters. For instance, the cluster around "thing," which connects terms like "capital," "Venice," and "ruby," subtly evokes Gatsby's preoccupation with wealth and a cultivated image of luxury. Another cluster, organized around "infantry" and "gun," hints at associations with authority or military honor, indicating Gatsby's need to assert social prestige and legitimacy. These clusters imply that Gatsby's language is layered, where the periphery terms implicitly express his strong desire for status and honor while the central ones cover his being with superficial styles.
What is interesting about this graph is the presence of the small group of words organized around "car" in the central part of the graph. It is not conspicuously visible with no strong hub, but it still belongs to the main cluster, positioned close to the most prominent nodes in the network. It contains terms related to the pivotal car accident involving Daisy, with words like "nervous" and "rushed" positioned nearby. It is almost hidden, but ostensibly visible right next to the big nodes that are in some sense insignificant as habitual expressions of Gatsby's speech. It is symptomatic. The proximity of "Daisy" to this cluster is very revealing in this regard. It has to be noted that the node "Daisy' is the only character name present in Gatsby's dialogue network although it appears smaller than other major terms. Daisy is the one who is deeply seated in Gatsby's mind as she lurks in his vocabulary. The ultimate cause of Gatsby's desire reveals itself as a small node on the outskirts of the central cluster.
To better understand the status and meaning of Daisy in Gatsby's speech, we need to take a closer look at this symptomatic node, "Daisy." Figure 2 is the ego network of "Daisy" in the co-occurrence network of Gatsby's dialogue. The first thing that stands out is that words related to the car accident gather distinctly to the upper right of "Daisy." Daisy is closely associated with the accident. Their uniform node sizes within this group suggest that these terms tend to co-occur within a relatively long sentence. More significant, then, is the node "want," which is linked to "Daisy" by a thicker edge, implying a particularly strong connection. This link leads to two possible interpretations: it might indicate Gatsby's desire for Daisy, or it could represent Daisy's. Closer examination of sentences where both "Daisy" and "want" appear reveals that the longing consistently originates from Gatsby himself. This pattern is especially evident in Chapter 7, when he says, "I want to speak to Daisy alone," emphasizing his possessive desire. Gatsby's own desire for Daisy is much more important to him than what she actually wants. Her response, "Oh, you want too much!" succinetly captures the overwhelming nature of Gatsby's need, exposing an unbridgeable gap between his obsessive fixation and her autonomy.7
In this context, Gatsby's frequent use of the verb "thought" invites further scrutiny. While "thought" commonly functions as an idiomatic expression in his speech, it turns out to contain more meanings than it seems, especially in relation to Daisy. It is quite striking that Daisy and Tom use the verb "think" rather than "thought." The term "thought" appears almost exclusively in Gatsby's conversations with Nick, often in phrases like "I thought" or "she [Daisy] thought," implying that he reserves his reflective sentiments for this confidant. The use of "thought" in the past tense also suggests a retrospective distance, reflecting Gatsby's intense preoccupation with past memories. This sentiment is particularly vivid in Chapter 8, where Gatsby says, "she was in love with me too. She thought I knew a lot because I knew different things from her." The past tense verbs in this quote reveals his obsession with the past love with Daisy.
Similarly, the verb "love" in Gatsby's dialogue reveals his fixation on the past. While Daisy often uses "love" in the present tense, highlighting her focus on present relationships and realities, Gatsby's use of "love" overwhelmingly appears in the past tense, underscoring his persistent attachment to their lost past. This contrast is most apparent in Chapter 7, when Daisy says, "I love you now isn't that enough? I can't help what's past." The present outweighs the past. For Gatsby, however, the past of Daisy's exclusive love for him is crucial as his emphatic assertion to Tom-"she never loved anyone except me!" exhibits. Daisy's simple reply, "I loved you too" (Chapter 7), shatters Gatsby, dismantling his carefully constructed fantasy of a perfect past and destabilizing his sense of identity.
Nick Carraway is the narrator and a character of the novel. He is within and without the story. To fully understand him, it is necessary to take into account both aspects together. Nick as a character can be analyzed through his dialogue; Nick as the narrator, through his narration. Figure 3 presents the co-occurrence network of Nick's speech. Central terms in this network include "want," "come," "know," "tell," "say," and "got." Significantly, these are primarily verbs rather than concrete nouns, suggesting that Nick's speech often lacks substantive content and instead emphasizes actions related to gathering and conveying information. This verbal emphasis aligns with his role as an observer, functioning less as an active character and more as a narrative channel, offering readers insight into the other characters. For instance, terms like "know," "tell," and "say" support his role in both receiving and relaying information. Another prominent term, "want," mostly appears in relation to other characters desires rather than Nick's own, underscoring his passive role and limited self-expression throughout the novel.8
What is interesting about this graph is that it contains all major character names in varying degrees, including "Gatsby," "Daisy," "Thomas," "Buchanan," "Miss," "Baker," and "Wilson."9 It shows that Nick forms relationships with all major characters of the novel as an observer and as a potential narrator. Significantly, Nick's own last name, "Carraway," appears adjacent to "Gatsby," showing that, while Nick often maintains an observational stance, he is still closely connected to Gatsby. It is true that there are large, interconnected nodes on the network's right side, but their more or less equal node sizes suggest that these terms belong to unusually long sentences. "Gatsby" is by far the most prominent node among character names, the only one to rival Nick's verbs in degree centrality. This prominence highlights Nick's primary narrative focus on Gatsby throughout the novel, reinforcing the idea that Nick's attention is disproportionately centered on Gatsby, even as he strives to retain a neutral role within the story.
The ego network of Gatsby in the narration (modularity resolution: 1.0); nodes' size: degree

Figure 4 is the ego network of "Gatsby" in the co-occurrence network of Nick's narration, colors indicating the subgroups of the whole graph.10 This ego network shows the connections between words in sentences that specifically include "Gatsby," offering insight into how Nick as the narrator understands and shapes Gatsby's world. Interestingly enough, terms associated with Gatsby's material wealth and honor-such as references to his luxurious possessions or military background, which are prominent in his own dialogue are absent here. This omission suggests that Nick's perspective either downplays or dismisses these elements of Gatsby's self-image, choosing instead to focus on other aspects of his character. The characters most prominently linked to "Gatsby" are "Daisy" and "Tom," indicating that Nick perceives Daisy as central to Gatsby's life while simultaneously recognizing the inescapable reality that Tom, as her husband, is beside her.
The green group includes the main characters such as "Daisy," "Tom," "Buchanan," "Wilson," "Jordan," and "Baker." It reflects Nick's perception of social relationships and interactions between characters surrounding Gatsby. It is noteworthy that nouns related to the body-such as "face," "hand," "eye," "foot," "voice," "body," "arm," and "hair" appear alongside sensory or movement-related verbs like "looked," "turned," "moved," "followed," "stared," "glanced," and "leaned." These terms highlight the narrator's keen interest in observing physical appearances and behaviors of the characters. Significantly, these words are largely, though not exclusively, associated with women. They are positioned around female characters like "Daisy" and "Jordan." Moreover, such terms as "woman," "girl," and "shoulder" appear nearby, though they are categorized into a different group. This suggests the narrator's male gaze toward women.11
The purple cluster centers on spatial vocabulary, with "house" serving as a key node. On the left are terms like "west," "egg," "village," and "island," while below "house" appear nodes such as "new," "york," "station," "beach," "street," and "place," all representing geographical spaces. Words related to domestic settings such as "hall," "garden," "lawn," "upstairs," and "floor" are also present, alongside boundary-related words like "door," "window," "step," "porch," "front," "outside," "beside," and "side." Color and light descriptors, including "night," "white," "light," "dark," "coloured," "green," "grey," "blue," and "black," further enhance Nick's visual depiction of space. This spatial vocabulary implies that Nick views Gatsby as a figure who navigates multiple worlds and social boundaries. In Nick's perspective, Gatsby represents movement and transformation, crossing boundaries from West to East, West Egg to East Egg, and poverty to wealth, echoing his transformation from James Gatz to Jay Gatsby. This spatial movement emerges as a key motif in building a myth surrounding Gatsby.
The group of blue nodes is particularly remarkable because, despite appearing to contain insignificant words, it uncovers a crucial aspect of the narrator's relationship with Gatsby.12 Apart from "girl" and "woman," which align more closely with the green cluster, the blue group mostly contains words whose meanings are hard to determine. These include numeric terms like "one," "two," and "first," time-related words such as "day," "year," and "minute," and auxiliary or cognitive verbs like "would," "could," "might," "found," "knew," and "thought." The predominance of past-tense auxiliary and verbs with unclear meanings suggests that this cluster is primarily concerned with Gatsby's past and Nick's speculations about him. In this regard, certain smaller nodes turn out more important. "Perhaps," located between "would" and "might" is a telling word that condenses the guesses and conjectures about Gatsby's past. Peripheral terms like "remember" and "forgotten" also have to do with Gatsby's past, while "Wolfshiem" alludes to his darker, concealed affiliations. Collectively, these terms emphasize how Nick's narration is imbued with speculation and hints of secrecy regarding Gatsby's history and identity.
Bipartite network of Gatsby, Nick as the narrator, and their words; nodes' size: in-degree

Figure 5 is a network graph illustrating connections between speakers and their words, with Gatsby and his words on one side, and Nick, the narrator, on the other. This bipartite network shows the relationship between a character and the narrator through the shared and unshared nodes. The most striking feature is the substantial disparity in vocabulary size between the two. The narrative voice has an overwhelming amount of words compared to a character since it covers all the characters and leads the story. Nodes positioned between Nick and Gatsby in the middle of the network represent the terms that they share. A character shares most of his words with the narrator, whereas the narrator, only a small portion of their vocabulary.
In Nick's narration, "said" emerges as the most salient one among the nodes other than character names, almost exclusively tied to him, underscoring his position as the primary voice conveying the words and perspectives of others, both directly and indirectly. The narrator is the one who uses "said" most frequently in the narrative. The network also includes the major character names such as "Gatsby," "Daisy," "Tom," "Jordan," and "Wilson," with Gatsby's name standing out as the largest character node. This prominence emphasizes Nick's narrative focus on Gatsby, who becomes his central subject. What is interesting is that "Daisy" is the only character name that they share and that it is positioned closer to the narrator. Nick recognizes the importance of Daisy in Gatsby's existence and pays close attention to her.
The nodes that the narrator does not share with Gatsby are distinguished by body-related words like "eye," "foot," "body," "head," "shoulder," "arm," "hand," and "face," terms he uses almost exclusively in narration to describe characters' physical appearance and behavior. This emphasis on corporeal features reflects Nick's observational focus and subtly incorporates a male gaze, especially in his descriptions of female characters. It is interesting to note that Gatsby's vocabulary does not display this obsession with physicality. The nodes that the narrator does share with, but uses far more than Gatsby include time-related nouns such as "night," "day," "minute," "year," "afternoon," "time," and "hour," as well as numbers like "one," "two," "first," "five," and "three." Spatial terms like "house," "room," "door," and "window" are also shared, though they are more characteristic of Nick's narrative voice. Additional spatial verbs such as "went," "toward," "back," "away," "came," "car," "way," "around," and "along" highlight Nick's focus on Gatsby's temporal and spatial movements. These elements reveal Nick's narrative strategy to mythologize Gatsby, drawing symbolic meaning from Gatsby's transitions across both literal and figurative boundaries.
On Gatsby's side of the graph, nodes such as "want" and "loved" stand out among the words that he rarely shares with the narrator, excluding his signature phrase "old sport." These verbs signal his desire for Daisy and attachment to the past. A distinct cluster on the far left contains Gatsby's own words he uses exclusively. Words like "capital," "Venice," "infantry," and "insignia" suggest wealth and honor, reflecting his efforts to project a prestigious self-image. Nick does not share these terms, indicating his narrative distance from Gatsby's materialistic values and skepticism toward his claims of wealth and honor. By selectively omitting certain elements of Gatsby's self-presentation, the narrator reconfigures Gatsby's story around his enduring attachment to Daisy and the past. Overall, the narrator-character network captures Nick's interpretive role in shaping Gatsby's myth. Nick emphasizes themes of temporal and spatial movement, casting Gatsby as a liminal figure whose aspirations and desires transcend the immediate social reality.
This paper attempts a network analysis of the narrator and the main character in F. Scott Fitzgerald's The Great Gatsby. In essence, I view the narrator and characters as lexical networks derived from the novel's dialogues and narration. A "symptomatic reading" of a character's speech network, as it turns out, proves effective in uncovering hidden aspects of that character, revealing Gatsby's obsessive longing for Daisy and fixation on the lost past. I also argue that analyzing a character's ego network within the narrator's narration demonstrates how the narrative voice understands and portrays that character. Gatsby's ego network reveals the narrator's keen interest in physical appearances along with his subtle male gaze toward women, his endeavor to speculate about Gatsby's secret past, and his narrative strategy to recreate Gatsby's myth in terms of temporal and spatial movement. Finally, the bipartite network between the narrator and a character mediated through the shared words illustrates how they converge and diverge with each other, highlighting the stark contrast between Gatsby as a character and Nick as the narrator of the novel.
In some sense, my approach goes against a dominant trend in digital humanities scholarship which puts much emphasis on large data sets and macroanalysis. Computational literary criticism, I argue, contributes to digital studies of culture and history by being more attentive to the details of literary texts and the texture of narrative art while creatively adopting digital methodologies. It is on the middle ground between distant reading and close reading, or in the continuous movement between the two. Most network analyses of narrative texts have focused on character interactions so far, limiting themselves to the social relationships within a diegetic world as envisioned by social network analysis. The paper, “Character as a Network of Words: Towards a Network Theory of Narrative,” was an attempt to get over this persistent tradition by recuperating the structuralist narratologist, Seymour Chatman's theory of character as “a paradigm of traits” (126). The present study extends the same frame of interpretation to the analysis of a non-dramatic narrative, regarding both the narrator and a character as lexical networks. In so doing, symptomatic reading, which is alien in digital humanities, turns out to be an effective way of uncovering hidden aspects of a character. The ego network of a particular character in the network graph of narration also helps reveal how the narrative voice understands and constructs the character.
More work lies ahead in building network theories of narrative. This paper has focused on the first-person narrative in which the narrator is also a character inside the diegetic world of a novel. The first-person narrator has a limited knowledge of other characters while still controlling the whole narrative. The third-person narration may have its own distinct characteristics. It may describe the thinking process, inner psychological state and stream of consciousness occurring within a character's mind. It may have much more control over characters and the story. Plot is another daunting challenge for network analysis of literature. A network is a paradigmatic representation of interconnections between multiple entities, whereas a plot is a syntagmatic chain of events along the temporal axis. It remains to be seen whether network analysis can resolve this problem and successfully capture the diachronic aspect of narratives. These are exciting challenges for digital humanists as well as literary scholars.
Jiwoo Choi, a graduate student at the School of Digital Humanities and Computational Social Sciences, Korea Advanced Institute of Science and Technology, contributed to many aspects of this research, especially in preparing datasets for the network analysis of The Great Gatsby. All relevant files for this research paper are available in the GitHub repository at https://github.com/vadoro/TheGreatGastby. ChatGPT, a generative AI platform, is partially used to proofread and improve the English sentences in this paper. However, the content and scholarly work presented in this paper are entirely my own.
1. Some sentences inside the quotation marks are excluded when they are not direct speech by a character: the song lyrics, "I'm the Sheik of Araby / Your love belongs to me / At night when you're asleep / Into your tent I'll creep" (Chapter 4) and the imaginary lines that are not actually uttered, "I never loved you" (Chapter 6); "Look here, old sport, you've got to get somebody for me. You've got to try hard. I can't go through this alone" (Chapter 9).
2. It is true that literary criticism has always been attentive to uncovering hidden meanings. However, "symptomatic reading" proposed here is different from the traditional literary criticism in that it does not read the literary text itself, but its abstract network representation. I use the term, "symptomatic," to differentiate my reading strategy of network graphs from the dominant approaches in digital humanities to literary texts.
3. The imaginary line in Chapter 9"I wanted to get somebody for him. I wanted to go into the room where he lay and reassure him: 'I'll get somebody for you, Gatsby. Don't worry. Just trust me and I'll get somebody for you" is considered part of the narration as it reveals Nick's inner thoughts.
4. The Natural Language Toolkit (NLTK) is utilized for text processing tasks within Python programming. Stop words including pronouns are excluded. For comprehensive details and guidance, refer to its official documentation available at the NLTK website: https://www.nltk.org/.
5. Gephi is an open-source software for network analysis and visualization. It is available on the official Gephi website: https://gephi.org/.
6. The ForceAtlas2 algorithm is used to draw the network graphs in this paper. It is "a force directed layout" where "[n]odes repulse each other like charged particles, while edges attract their nodes, like springs" (Jacomy et al. 2014, 2). For more technical details, see Jacomy et al. (2014).
7. Certain words and proper nouns have been capitalized for grammatical accuracy, despite their appearance in lowercase in the graphs.
8. Last names like "Baker" appear in his dialogue but do not hold significant representation in the network.
9. The sentence, "You're sure you want me to come?" is a typical example of Nick's use of the word "want" (Chapter 5).
10. Gephi uses the Louvain algorithm for calculating modularity. The outcomes of modularity analysis may vary with each iteration, particularly for nodes positioned on the boundaries between communities, which may be assigned to different groups in subsequent analyses. This inherent ambiguity necessitates cautious consideration during interpretation to ensure accurate conclusions.
11. One of the most exemplary passage that reveals Nick's "male gaze" is as follows: "I enjoyed looking at her [Baker]. She was a slender, small-breasted girl, with an erect carriage, which she accentuated by throwing her body backward at the shoulders like a young cadet. Her grey sun-strained eyes looked back at me with polite reciprocal curiosity out of a wan, charming, discontented face" (Chapter 1).
12. The fact that "Gatsby" belongs to this cluster should not be given too much importance. Since this graph is the ego network of "Gatsby," it has to align with any cluster of words. But the modularity analysis algorithm forces all nodes into specific clusters despite possible overlaps. This limitation should always be considered when interpreting any networks of modular groups.
Elson, David K.; Dames, Nicholas; and McKeown, Kathleen R. (2010). "Extracting Social Networks from Literary Fiction." Proceedings of the 48th Annual Meeting of the Association for Computational Linguistics, Association for Computational Linguistics. https://aclanthology.org/P10-1015/
Fitzgerald, F. Scott. (1925). The Great Gatsby. Project Gutenberg. https://www.gutenberg.org/ebooks/64317
Jacomy. Mathieu; Venturini, Tommas; Heymann, Sebastien; and Bastian, Mathieu. (2014). "ForceAtlas2, a Continuous Graph Layout Algorithm for Handy Network Visualization Designed for the Gephi Software." PLOS ONE. https://doi.org/10.1371/journal.pone.0098679
Kim, Heejin. (2024). "Shakespearean Character Network Analysis: Residual Dramatic Structures of Medieval Touring Companies and Neoclassical Unity." In/Outside: English Studies in Korea. https://doi.org/10.46645/inoutsesk.56.1
Kim, Yongsoo. (2024) "Character as a Network of Words: Towards a Network Theory of Narrative." The Journal of Criticism and Theory. https://doi.org/10.19116/theory.2024.29.2.125
Labatut, Vincent, and Bost, Xavier. (2019). "Extraction and Analysis of Fictional Character Networks: A Survey." ACM Computing Surveys. https://doi.org/10.1145/3344548
Won, Young Seon. (2023). "Exploring Digital Literary Study, an Application: Social Network Analysis of Emma." Nineteenth Century Literature in English. https://doi.org/10.24152/NCLE.2023.9.27.2.39