Christian Ilbury, Rianna Walcott
8 min
This study investigates the origins and public perception of what is frequently labeled in media and online discourse as TikTok language or Gen Z language. Using a corpus of 178 TikTok videos, the authors employ Critical Technocultural Discourse Analysis (CTDA) to examine how users define, contest, and claim ownership of these linguistic features. By centering the perspectives of everyday users, the researchers explore how social media platforms serve as sites of linguistic debate and recontextualization.
The researchers demonstrate that the vast majority of terms identified as TikTok language are, in fact, lexical features of African American Vernacular English (AAVE). They identify a clear tension in online discourse: while some users propagate the label of TikTok language, others actively reject it as a form of cultural appropriation. This conflict is analyzed through the lens of White ignorance, where the erasure of AAVE's Black roots allows users to adopt the language for social capital without acknowledging the community of origin.
Furthermore, the authors introduce the concept of linguistic context collapse to describe how social media platforms blur geographically and socially bounded speech communities. They argue that TikTok's algorithm, which prioritizes viral content and creates a feedback loop of engagement, further accelerates this erasure by detaching AAVE features from their historical and social indexicality, reifying them as a novel digital register.
This paper contributes to the fields of sociolinguistics and digital culture by showing how social media platforms act as mechanisms for linguistic appropriation. It provides empirical evidence of how labels like TikTok language function to displace Black varieties of English. By highlighting how users resist this process through hyper-citation and the correction of linguistic misinformation, the study underscores the active, political nature of digital communication and the ongoing struggle for the recognition of Black linguistic agency.
ABSTRACT Sociolinguistic research has long documented the appropriation of African American Vernacular English (AAVE) across media including film, music and advertising. In this article, we add to this body of work by exploring the digital recontextualisation of a subset of AAVE features as ‘TikTok/internet language’. To do this, we analyse metalinguistic discourses of linguistic appropriation in a corpus of 178 TikTok videos. We identify two main competing discourses: On the one hand, a concern regarding the indexical erasure of AAVE as a variety of English spoken by Black Americans; and on the other, claims of a new register of ‘internet language’. Concluding, we argue that, in the participatory context of social media, the circulation of the label ‘TikTok language’ poses an issue for the raciolinguistic enregisterment as a ‘Black variety’ of English.
Sam: And were they measuring ordinary speech, or what people said about speech? That distinction changes what the counts mean.
Alex: Mainly what people said about speech. They counted features explicitly identified in the videos as TikTok language, then traced their origins using linguistic scholarship. They also analysed the videos’ arguments, considering captions, comments, and other visual elements. Machine-generated transcripts were checked manually by both authors.
Sam: What is the methodological idea that makes that more than researchers correcting other people’s terminology?
Alex: They centre everyday users as producers of knowledge about language. That approach connects to citizen sociolinguistics: ordinary people investigating and debating language. Their critical framework also asks how users’ beliefs interact with platform design, rather than treating either in isolation.
Sam: Give me a concrete finding. Which words were people putting under the TikTok label?
Alex: The most frequently cited was “periodt,” an emphatic interjection, mentioned forty times. Next came “bussin,” meaning excellent or delicious, with twenty-seven mentions. These are counts within the collected videos, not estimates of how commonly TikTok users say those words.
Sam: And the comparison is against earlier evidence of those features in Black American speech, rather than against another platform?
Alex: Yes. The authors trace all the most commonly cited lexical features to AAVE. Their third most frequent example is “finna,” a contraction of “fixing to,” cited eighteen times. Their finding concerns the origins of features people name, not proof that TikTok never produces linguistic innovation.
Sam: Words are easy to recognise and list. Does that risk reducing a whole language variety to slang?
Alex: The study finds that very imbalance. Users overwhelmingly cite vocabulary, while grammatical features receive little explicit attention. The authors stress that AAVE has systematic grammar. Their example is habitual “be”: “he be running” describes recurring activity, not simply what he is doing now.
Sam: That helps explain why collecting fashionable words is not the same as speaking the variety. What objections did users themselves raise?
Alex: The authors identify connected objections about erasure, inaccurate grammar, and exploitation. Start with erasure: one creator says she assumed a phrase was TikTok language because she had heard it only there. The authors interpret that as evidence that the platform can become mistaken for the source.
Sam: So the phrase keeps its conversational use, but loses its association with Black speakers.
Alex: They call that indexical bleaching: language features losing their social associations as they spread. They argue that relabelling them as TikTok language can further erase recognition of their Black American origins. Users rejecting the label connect this to “White ignorance,” understood as socially structured misunderstanding, not merely an individual knowledge gap.
Sam: What does the exploitation objection add? It sounds stronger than saying somebody got the history wrong.
Alex: It concerns unequal benefits and burdens. One creator describes outsiders adopting long-established Black slang, dropping it when it stops seeming fashionable, then mocking Black people who continue using it. The authors connect that complaint to earlier research: outsiders can gain social value without bearing the same racial discrimination.
Sam: But does the article accept every ownership claim made in response? Protecting a community’s language could also produce overly rigid boundaries.
Alex: It explicitly examines that tension under “Overdetermination”: excessive links between linguistic forms and racial categories. Some creators treat Blackness alone as determining authentic use. The authors contrast that with research documenting non-Black speakers whose habitual use is accepted within close Black social networks.
Sam: So someone’s locally accepted speech might be judged differently when a video reaches strangers.
Alex: That is central to the argument. Local recognition does not necessarily travel with the speaker. The authors also find mistaken claims about individual features, including treating “naur,” an online spelling associated with Australian pronunciation of “no,” as AAVE. They interpret such overclaiming as resistance to repeated erasure, without endorsing its accuracy.
Sam: Where does the platform enter the explanation? Showing these disputes does not itself show that an algorithm caused them.
Alex: The authors propose “linguistic context collapse”: distinct speech communities brought into a shared digital setting. People encounter features outside their usual social networks, then associate them with where they encountered them. They argue that personalised recommendations can reinforce the TikTok-language interpretation while making the originating community less visible.
Sam: How far should we trust that mechanism, given the way the corpus was assembled?
Alex: My reading is that the evidence is strongest for the naming dispute and the documented origins of frequently cited features. Discovery followed recommendations and hashtags from a shared starting point, so this is not a representative survey. Nor did the study experimentally compare recommendation systems or track language change over time. The algorithmic explanation remains an argument supported by the qualitative analysis.
Sam: Who should read the full article, and where should they begin?
Alex: Researchers working on digital discourse, language contact, or racialised language should read it. Start with “Metadiscourses of TikTok Language,” then “Overdetermination,” which tests the limits of the ownership claims. Read “TikTok Language and Linguistic Context Collapse” for the broader theoretical proposal. For everyone else: where you first hear a word is not necessarily where that word began.
Sam: A distinction to keep when the next slang explainer appears.