Emojis, Hashtags, and Code-Switching: A Literary-Linguistic Analysis of Multimodal Digital Textuality

Linguistic Forum Published by MARS Publishers Published by Licensee MARS Publishers. Copyright: © the author(s). This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/). Emojis, Hashtags, and Code A Literary Multimodal Digital Textuality Corresponding Author: Sehar Ja been < [email protected] Sehar Khalid < [email protected] Zakra Nadeem < [email protected] Received: September 10, 2025 Abstract This article examines emojis, hashtags, and code firstcentury digital communication rather than peripheral embellishments. Drawing on a qualitative multimodal discourse analysis, we analyze a purposefully sam posts from Twitter/X and Instagram (2021 WorkLifeBalance, and StudentLife ; posts were included if they contained at least one emoji and evidence of a hashtag or codeswitching. Analysis proce linguistic/pragmatic functions of each resource; (2) interpreting their aesthetic and literary affordances; and (3) situating their interaction within broader sociocultural frameworks. Findings indicate that emojis primarily index stance and tone irony— while hashtags operate as indexical “refrains” that structure participation and intertextual linkage; intrasentential and taglevel code ISSN (Online) 2707 Volume http://doi.org/ 10.5281/zenodo.17308022 Published by Licensee MARS Publishers. Copyright: © the author(s). This article is an open access article conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/). Emojis, Hashtags, and Code - Switching: A Literary - Linguistic Analysis of Multimodal Digital Textuality Research Article been seharja[email protected] > PhD Scholar, Department of Applied Linguistics, Government College University, Faisalabad, Punjab, Pakistan. Sehar Khalid khalidse[email protected] > MPhil Scholar, Department of Applied Linguistics, Government College University, Faisalabad, Punjab, Pakistan. Zakra Nadeem [email protected] > MPhil Scholar, Department of English, Riphah International University Pakistan. Publication Details Accepted: October 5, 2025 Published: This article examines emojis, hashtags, and code - switching as core semiotic resources in twenty century digital communication rather than peripheral embellishments. Drawing on a qualitative multimodal discourse analysis, we analyze a purposefully sam pled corpus of 300 public posts from Twitter/X and Instagram (2021 – 2023) selected via the keywords ; posts were included if they contained at least one emoji and switching. Analysis proce eded in three stages: (1) identifying the linguistic/pragmatic functions of each resource; (2) interpreting their aesthetic and literary affordances; and (3) situating their interaction within broader sociocultural frameworks. Findings primarily index stance and tone — including patterned uses of affect and while hashtags operate as indexical “refrains” that structure participation and intertextual level code - switching, in turn, performs identity wor ISSN (Online) 2707 -5273 Volume 7, Issue 1, 2025 10.5281/zenodo.17308022 Pages 38-51 Published by Licensee MARS Publishers. Copyright: © the author(s). This article is an open access article conditions of the Creative Commons Attribution (CC BY) license Switching: Linguistic Analysis of Multimodal Digital Textuality Scholar, Department of Applied Linguistics, Government College University, Faisalabad, Punjab, Pakistan. Department of Applied Linguistics, Government College University, Pakistan. MPhil Scholar, Department of English, Riphah International University , Faisalabad, Published: October 10, 2025 switching as core semiotic resources in twenty - century digital communication rather than peripheral embellishments. Drawing on a pled corpus of 300 public 2023) selected via the keywords MeToo, ; posts were included if they contained at least one emoji and eded in three stages: (1) identifying the linguistic/pragmatic functions of each resource; (2) interpreting their aesthetic and literary affordances; and (3) situating their interaction within broader sociocultural frameworks. Findings including patterned uses of affect and while hashtags operate as indexical “refrains” that structure participation and intertextual switching, in turn, performs identity wor k and audience Emojis, Hashtags, and Code-Switching. . . LinFo www.linguisticforum.com 39 Linguistic Forum 7(1), 2025 design, producing polyphonic, hybrid utterances. Taken together, these resources foreground the multimodal, participatory character of contemporary textuality, complicating canonical distinctions between speech and writing. The study argues for treating online discourse as both a linguistic phenomenon and a vernacular literary form, and outlines implications for sociolinguistics (integrating translanguaging and multimodality into analytic units) and literary studies (extending close reading to networked, ephemeral texts). Keywords: emojis, hashtags, code-switching, multimodality, translanguaging, digital discourse, vernacular literature 1. Introduction Digital communication increasingly relies on multimodal resources that combine written language with visual and paralinguistic cues. Emojis, hashtags, and code-switching are no longer peripheral embellishments but routine means of shaping stance, organizing participation, and performing identity across platforms such as Twitter/X, Instagram, and WhatsApp (Kress, 2010; Lyons, 2024; Konrad et al., 2020). Linguistic research has documented how these resources operate pragmatically—indexing tone, alignment, and audience—while cultural and literary perspectives highlight their aesthetic and narratological effects in everyday networked writing (Kress, 2010; Lyons, 2024). For instance, empirical work shows that emojis efficiently signal intention and affect, though their interpretation depends on context and interactional norms (Thompson & Filik, 2016; Stark & Crawford, 2015; Lim, 2015; Yang, 2024; Aljasir et al., 2024; Cavalheiro et al., 2024). Hashtags, meanwhile, function as organizational and connective devices that tie utterances into searchable publics and professional or political discourse (Enli, 2018; Sinpeng et al., 2021). In parallel, translanguaging and code-switching practices online perform identity and audience design, producing hybrid, polyphonic utterances (Hua, 2022; Montes-Alcalá, 2024). Two gaps motivate the present inquiry. First, existing scholarship often treats emojis, hashtags, and code-switching in isolation, offering resource-specific typologies rather than an integrated account of how these semiotic cues co-operate within single posts (cf. Konrad et al., 2020; Thompson & Filik, 2016; Enli, 2018). Second, despite recognition that platform logics structure visibility and uptake, there remains limited cross-resource analysis grounded in authentic posts that foregrounds both communicative and aesthetic effects (Enli, 2018; Sinpeng et al., 2021). Addressing these gaps, the article adopts an interdisciplinary lens that brings together sociolinguistic concepts— indexicality, stance, audience design, and translanguaging—with literary-critical notions of voice, refrain, intertextuality, and polyphony to examine how meaning is co-produced in everyday digital texts (Hua, 2022; Lyons, 2024; Sinpeng et al., 2021). Methodologically, the study employs qualitative multimodal discourse analysis of a purposefully sampled corpus of public posts from Twitter/X and Instagram (2021–2023). Posts were included if they contained at least one emoji alongside either a hashtag or evidence of code-switching. Analysis proceeded in three stages: (1) identifying the pragmatic functions of each resource within posts; (2) interpreting their aesthetic affordances and relations (e.g., how hashtags scaffold thematic cohesion while emojis inflect tone); and (3) situating these relations within wider sociocultural frames, including participation in issue publics and the performance of identity (Enli, 2018; Emojis, Hashtags, and Code-Switching. . . LinFo www.linguisticforum.com 40 Linguistic Forum 7(1), 2025 Sinpeng et al., 2021). Ethical procedures for working with public social-media data and anonymization guidelines were observed. This design enables the article to articulate both linguistic and literary payoffs. Linguistically, it demonstrates how emojis routinely index stance and tonal nuance, how hashtags coordinate alignment and uptake by linking utterances across networks, and how code-switching— intrasentential and tag-level—indexes audience, expertise, and affiliation (Thompson & Filik, 2016; Stark & Crawford, 2015; Lim, 2015; Enli, 2018; Hua, 2022; Montes-Alcalá, 2024). From a literary perspective, the analysis shows how these resources structure rhythm and voice, create refrains and motifs, and generate polyphonic textures that complicate boundaries between speech and writing, private expression and public circulation (Kress, 2010; Lyons, 2024). The contribution is therefore twofold. Empirically, the article offers an integrated account of three high-frequency resources as they co-occur in naturally occurring posts, moving beyond singleresource typologies (Konrad et al., 2020; Enli, 2018). Conceptually, it advances an interdisciplinary framework that treats social-media discourse as a vernacular literary form while retaining the analytical precision of linguistic pragmatics and translanguaging research (Hua, 2022; Lyons, 2024; Montes-Alcalá, 2024). This integrated perspective clarifies how meaning is made in platformed environments and provides tools for close reading of everyday digital texts. 2.1 Research Questions 1. How do emojis, hashtags, and code-switching function individually and jointly to index stance, organize participation, and perform identity in social-media posts? 2. In what ways do these resources interact to produce aesthetic and narratological effects (e.g., voice, refrain, polyphony) that warrant reading digital discourse as vernacular literature? 3. What implications follow for sociolinguistic analysis (e.g., units of analysis that include multimodal cues and translanguaging) and for literary studies (e.g., extending close reading to ephemeral, networked texts)? 2. Literature Review This review brings together rigorous studies on three widely used semiotic resources in digital discourse—emojis, hashtags, and code-switching/translanguaging. We proceed from foundational accounts of multimodality to focused strands on each resource, then integrate these strands to surface precise gaps that motivate our three research questions. The guiding premise is that platformed discourse is both linguistic interaction and vernacular textuality, requiring an analysis that is at once pragmatic and aesthetic (Kress, 2010; Lyons, 2024; Konrad et al., 2020; Enli, 2018; Sinpeng et al., 2021; Thompson & Filik, 2016; Stark & Crawford, 2015; Lim, 2015; Yang, 2024; Cavalheiro et al., 2024; Aljasir et al., 2024; Hua, 2022; Montes-Alcalá, 2024). 2.1 Multimodality and Vernacular Textuality Foundational multimodality research shows that meaning in digital environments is co-produced by textual, visual, and paralinguistic resources rather than alphabetic language alone (Kress, 2010). Emojis, Hashtags, and Code-Switching. . . LinFo www.linguisticforum.com 41 Linguistic Forum 7(1), 2025 Recent applied-linguistic work on “post-digital” life foregrounds how platform affordances and offline practices jointly shape production and interpretation (Lyons, 2024). Together, these perspectives justify reading social-media posts as vernacular textual artefacts that warrant both linguistic analysis and forms of close reading—framing research question 2 (how resource interactions yield aesthetic/narratological effects) and research question 3 (implications for analytic units and interpretive procedures). 2.2 Emojis as Indexical, Affective, and Interactional Cues Empirical studies show that emojis efficiently signal intention and stance, aiding tonal disambiguation in written interaction (Thompson & Filik, 2016). Scholarship also cautions that emoji meanings are negotiated and community-dependent, not fixed lexical substitutes (Stark & Crawford, 2015; Lim, 2015). Newer SAGE work deepens this account: smiling-face variants as socially situated cues (Yang, 2024), relational outcomes of reciprocal emoji use (Cavalheiro et al., 2024), and social presence among Arab users (Aljasir et al., 2024). These studies specify what emojis do individually—a base for research question 1—and indicate how they texture voice and rhythm when combined with other resources—feeding research question 2. 2.3 Hashtags, Searchability, and Publics Hashtags function simultaneously as metadata and rhetoric: they render posts discoverable, bundle dispersed utterances into conversations, and signal participation frames (Enli, 2018). Work on hashtag activism shows how tags recruit attention, scaffold collective narration, and align identities in contentious publics (Sinpeng et al., 2021). These findings clarify individual hashtag functions pertinent to research question 1 and suggest how hashtags operate as refrains or motifs across posts, shaping aesthetic patterning relevant to research question 2. 2.4 Translanguaging, Code-Switching, and Audience Design Language choice online performs identity and audience design. Hua (2022) theorizes translanguaging as performative, unsettling code-bounded views of multilingualism. Complementing this, MDPI research on Spanish–English texting in emoji-rich contexts shows how intrasentential and tag-level switching cooperates with paralinguistic markers to accomplish nuanced social work (Montes-Alcalá, 2024). These accounts position code-mixing as integral—not ancillary—to how users choreograph voice, negotiate expertise, and signal affiliation within single posts, directly informing research question 1 and bridging to research question 2. 2.5 Integrated Accounts: the Specific Gap Across strands, consensus holds that (i) emojis are flexible indexical cues (Thompson & Filik, 2016; Stark & Crawford, 2015; Lim, 2015; Yang, 2024; Cavalheiro et al., 2024; Aljasir et al., 2024); (ii) hashtags organize participation and publics (Enli, 2018; Sinpeng et al., 2021); and (iii) translanguaging/code-switching performs identity and audience design (Hua, 2022; MontesAlcalá, 2024). What remains scarce is integrated, within-post analyses tracing how these resources co-operate in sequence to yield both communicative and aesthetic effects. Much existing work privileges single-resource typologies or decontextualized illustrations, leaving the interplay among Emojis, Hashtags, and Code-Switching. . . LinFo www.linguisticforum.com 42 Linguistic Forum 7(1), 2025 emojis, hashtags, and code-switching under-specified. This gap motivates research question 1 (a joint functional account) and research question 2 (their interactional–aesthetic payoffs). 2.6 From Evidence to Analytical Payoffs Because multimodality and post-digital accounts imply that analytic units must match how users actually compose posts, there is a methodological imperative to integrate pragmatic and literary lenses. Doing so has implications for sociolinguistics (units of analysis that include multimodal cues and translanguaging) and for literary studies (extending close reading to networked microtexts). Making these implications explicit and operational motivates research question 3. 2.7 Positioning the Present Study To address these aligned gaps, we adopt qualitative multimodal discourse analysis of authentic posts (2021–2023) to:  Model how emojis, hashtags, and code-switching function individually and jointly to index stance, organize participation, and perform identity—integrating insights from Thompson & Filik (2016), Enli (2018), and Hua (2022).  Explain how co-deployment yields aesthetic/narratological effects—voice, refrain, polyphony—connecting platformed discourse (Enli, 2018; Sinpeng et al., 2021) with multimodality and post-digital theory (Kress, 2010; Lyons, 2024).  Articulate implications for sociolinguistic and literary analysis by proposing integrative analytic units and interpretive procedures aligned with actual composition practices (Konrad et al., 2020; Lyons, 2024). The section provides robust resource-specific findings but under-theorizes their joint work in naturally occurring posts. By mapping research question 1, 2 and 3 directly onto these lacunae, this study offers a theoretically coherent and empirically transparent account of how patterned combinations of emojis, hashtags, and code-switching yield communicative effects (stance, alignment, audience design) and vernacular literary effects (motif, rhythm, polyphony) within the same micro-texts—clarifying how meaning is made, and made legible, in platformed discourse. 3. Methodology 3.1 Research Design and Analytical Stance This study employs a qualitative multimodal discourse analysis with an interpretive–abductive logic. The focus is on how emojis, hashtags, and code-switching operate as semiotic resources within naturally occurring social-media posts and how their co-deployment yields communicative (pragmatic) and aesthetic (narratological) effects. The analytical lens integrates sociolinguistic constructs (indexicality, stance, audience design, translanguaging) with literary concepts (voice, refrain, polyphony). Insights are developed iteratively through movement between excerpts and theory. Emojis, Hashtags, and Code-Switching. . . LinFo www.linguisticforum.com 43 Linguistic Forum 7(1), 2025 3.2 Data and Sampling The corpus comprises publicly accessible posts from Twitter/X and Instagram, selected through purposive maximum-variation sampling centered on three high-signal entry points that consistently elicit multimodal expression—#MeToo, #WorkLifeBalance, and #StudentLife. Eligible posts met all inclusion criteria: (a) public availability; (b) at least one emoji; and (c) either at least one hashtag or clear evidence of code-switching (intraor inter-sentential, or tag-level). We excluded content from private/protected accounts, media-only posts without accompanying text, commercial spam/advertorials, and exact duplicates. The final dataset totals N = 300 posts—150 from Twitter/X and 150 from Instagram—balanced across topics, with no more than three posts contributed by any single account to minimize user-level clustering. For each post, we stored the textual content and light metadata (platform, language[s], and counts of emojis/hashtags) and archived screenshots to ensure auditability; all identifiers (handles, URLs, and precise timestamps) were removed or masked at the point of analysis. 3.3 Operationalization In this study, emoji are treated as Unicode pictographs embedded in the textual stream to modulate tone, stance, and alignment; a hashtag is defined as any token beginning with # that indexes topic, signals stance, or marks a participation frame; and code-switching refers to the alternation of linguistic codes within or across clauses and/or hashtags, identified through orthographic, script, and lexical cues (including stylized loan forms). 3.4 Analytic Procedure Analysis proceeded in three stages; each stage explicitly addresses the research questions. Stage 1 — Resource-Specific Coding Each post was annotated for the functions of each resource using non-exclusive labels:  Emoji: stance/affect, irony, emphasis, mitigation, alignment cue.  Hashtag: topical index, participation frame, alignment signal, refrain/motif, metacommentary.  Code-switching: audience design, identity display, expertise/indexicality, play/quotation. Stage 2 — Relational Analysis Within-post sequencing and co-occurrence were analyzed to model how resources interact. Aesthetic/narratological outcomes—voice (tonal texturing), refrain (repeating hashtag motifs), and polyphony (multivoicedness via switching)—were linked back to the pragmatic functions identified in Stage 1. Emojis, Hashtags, and Code-Switching. . . LinFo www.linguisticforum.com 44 Linguistic Forum 7(1), 2025 Stage 3 — Pattern Synthesis and Theorization Cross-case matrices (resource × function × platform/topic) surfaced recurrent configurations (e.g., emoji-as-mitigation + hashtag-as-refrain + tag-level switching). From these, the study articulates implications for (a) units of analysis in sociolinguistics (multimodal cue bundles) and (b) literary close reading (motif/rhythm at micro-text scale). 3.5 Reliability and Analytic Rigor First, to establish shared interpretive norms, we initiated coder training, during which two coders independently annotated a 20% pilot subset (n = 60). After two calibration rounds, Krippendorff’s α (nominal) reached ≥0.75 for resource identification and ≥0.70 for function labels—figures that we regard as acceptable for interpretive qualitative work. Where disagreements persisted, they were adjudicated, and the codebook was refined accordingly, ensuring that subsequent coding proceeded on a more stable footing. Next, to enhance credibility, we implemented analyst triangulation (two coders plus the author), maintained iterative memos to make interpretive moves explicit, and conducted negative-case analysis to probe and, where necessary, revise emergent claims. Importantly, these strategies were applied throughout the analysis rather than confined to an initial pilot, thereby sustaining rigor across the corpus. Finally, to secure auditability, we maintained versioned codebooks, decision logs, and an anonymized excerpt bank (text with redacted screenshots). Taken together, these materials constitute a reproducible audit trail that is available to editors upon request. In sum, this stepwise protocol—training, credibility checks, and audit documentation—systematically aligns our analytic procedures with established standards of reliability and rigor. 3.6 Ethics Only public posts were analyzed. Identifiers (handles, profile images, URLs, precise timestamps) were removed or blurred. Quotations are lightly paraphrased when necessary to avoid back-tracing while preserving meaning; such instances are marked. No contact with users occurred, and platform terms were observed. 3.7 Reporting Conventions Findings are presented as (a) exemplar vignettes with minimal metadata to support inference, and (b) cross-case tables summarizing recurrent configurations. The subsections explicitly signpost how evidence addresses research questions 1 to 3. 3.8 Limitations and Scope Conditions The corpus is restricted to two platforms and three topical entry points; claims are analytically (theory-building), not statistically, generalizable. Visual-only posts and private content fall outside Emojis, Hashtags, and Code-Switching. . . LinFo www.linguisticforum.com 45 Linguistic Forum 7(1), 2025 scope. Platform affordances evolve; interpretations reflect the interface conditions at the time of capture. 4. Discussion Findings of this study are organized around the three research questions. We first report resource‐specific functions, then examine within-post interactions and their aesthetic/narratological effects, and finally draw out theoretical and methodological implications. Illustrative excerpts are anonymized and lightly paraphrased to protect user identity while preserving meaning. 4.1 Emojis, Hashtags, and Code-Switching 4.1.1 Emojis: stance, tone, and alignment Across topics and platforms, emojis primarily indexed stance and tonal contouring (e.g., empathy, irony, softening). In #StudentLife, facial and sweat-face emojis frequently mitigated complaint sequences, converting potential face-threats into affiliative bids. In #WorkLifeBalance, briefcase/clock combinations intensified time-pressure claims while maintaining a collegial register. Irony tended to be marked by smiling-with-sweat or winking faces adjacent to otherwise negative propositional content. These uses align with a “tone-as-metadata” function: emojis do not replace words so much as layer evaluative cues on them. 4.1.2 Hashtags: participation frames and thematic scaffolding Hashtags served two recurrent functions. First, they indexed topical alignment and discoverability (e.g., #MeToo, #StudentLife), situating posts within issue publics and routinizing uptake through searchable labels. Second, users deployed secondary or creative tags (e.g., #NotAllHeroes, #DeadlineMode) as metapragmatic glosses, compressing stance into compact refrains that readers could echo. In multi-tag posts, one “anchor” hashtag typically established topic, while shorter, inventive tags signaled evaluation or community norms. 4.1.3 Code-switching: audience design and identity work Intrasentential switching (e.g., English–Urdu; English–Spanish) signaled audience calibration and identity display. Tag-level switching (e.g., #KhairChalo alongside English body text) recurrently flagged cultural intimacy or insider status. In #MeToo posts, short switches (address terms, idioms) intensified alignment without displacing the global “anchor” language of the thread. Code choice was thus a resource for who is being addressed and how solidarity or expertise is being claimed. 4.1.4 Joint functions When co-present, the three resources tended to bundle into stable configurations: (a) Mitigated complaint (emoji-as-softener + topical hashtag + brief tag-level switch); (b) Call to witness (emojias-emphasis + anchor/issue hashtag + creative stance hashtag); and (c) Insider aside (emoji-aswink + tag-level switch + minimal topical tagging). Emojis, Hashtags, and Code-Switching. . . LinFo www.linguisticforum.com 46 Linguistic Forum 7(1), 2025 4.2 Aesthetic and Narratological Effects of Resource Interaction 4.2.1 Voice (tonal texturing) Emoji placement—especially adjacency to evaluative lexemes—produced a recognizable voice: a post could sound wry, earnest, or conspiratorial without lexical change. In #StudentLife, sequences like “finals week… again” yielded a self-deprecating persona that invited affiliative replies. Hashtags occasionally extended this voice across posts; recurring creative tags functioned as authorial idiosyncrasies (micro-style markers). 4.2.2 Refrain (motif via hashtags) Anchor hashtags operated as refrains at two scales. Within posts, they punctuated clauses (topic → evaluation → topic), creating rhythmic recurrence. Across posts, their repetition threaded dispersed utterances into a serial composition, allowing readers to scan a feed as a chorus with variations. Creative stance tags (e.g., #StillHere, #DoBetter) acted as motivic cells, accruing meaning through reuse across contexts. 4.2.3 Polyphony (multivoicedness via code-switching) Switching recruited additional “voices”: a quoted proverb or address term inserted an alternate perspective without overt reporting clauses. In #MeToo, short switches invoked familial or communal frames, juxtaposed against a public, global tag—producing layered voicing in a compact form. Polyphony intensified when switches co-occurred with echo-hashtags (e.g., #BelieveHer / #HumBhi): the post simultaneously inhabited global and local discursive worlds. 4.2.4 Interactional choreography (sequence matters) Function and form were sensitive to order. Emoji-before-proposition primed the reading (e.g., wink → irony), emoji-after softened closure; anchor-hashtag-first framed uptake, while creative tags in post-final position acted as punchlines. Tag-level switching placed at the end read as an aside; embedded switches read as identity claims integral to the proposition. These patterned placements constitute a tacit compositional grammar. 4.3 Representative Configurations Example 1 — Mitigated complaint (#StudentLife) “Four deadlines this week… #StudentLife #CoffeeIsMyPlan” Reading: emoji mitigates complaint; anchor hashtag situates audience; creative tag crystallizes motif (fatigue-as-fuel). Payoff: Voice (self-deprecating), Refrain (anchor tag), no switching needed. Example B — Call to witness (#MeToo)