What the Young Are Actually Saying: A Defence of a Language Its Speakers Did Not Ask to Have Defended
The embarrassment is instructive. It is the sensation you get when you realise that the thing you dismissed as noise was, in fact, a signal you were not equipped to receive.
This position has the advantage of being comfortable and the disadvantage of being wrong. The more carefully you read Gen-Alpha language—not in the spirit of an anthropologist cataloguing a curiosity but in the spirit of a linguist paying the language the respect of close attention—the harder the degradation thesis becomes to maintain. What you find, when you read carefully, is not the absence of complexity. You find complexity operating at a different register and at a different speed, organised by principles that conventional literacy assessment was not designed to measure and consequently cannot see.
The language works in layers, and the layers do not replace each other. They accumulate. Any given utterance in Gen-Alpha might draw simultaneously from inherited internet slang, gaming vocabulary, meme culture, African American Vernacular English, anime and Korean and Japanese borrowings, psychology, philosophy, abbreviation, and irony deployed over sincerity in ratios that shift without announcement. These are not separate languages deployed on separate occasions. They coexist in the same sentence, sometimes in the same word, and the receiver is expected to be navigating all of them at once. This is not what degraded language looks like. Degraded language loses layers. This language adds them.
Consider what happens with npc. To an outsider—and the outsider status is, as I will argue, part of the mechanism—it is gaming terminology: a non-player character, one of the background figures in a video game whose behaviour is scripted, whose responses are predetermined, who populates the world the player moves through without exercising any of the agency that makes the player’s experience meaningful. This is what the acronym means in its technical context.
It is not what it means in use. In use, npc has become a compact vehicle for a set of ideas about human agency, conformity, social performance, and self-awareness that would take several paragraphs of careful philosophical prose to deploy with equivalent precision. To call someone’s behaviour npc is not to say they are boring. It is to say that they appear to be executing a programme rather than making choices, that their responses to social situations are so thoroughly scripted by convention that the script has become invisible to them, that they have confused the performance of a social role with the exercise of genuine agency. Whether one agrees with any particular application of this label is a separate question. The label itself is doing far more conceptual work than three letters have any conventional right to do.
Or consider based, which began as a term within a specific internet context and has since developed into something considerably richer and harder to pin down: a word that carries simultaneously ideas of independence from social approval, authenticity in some sense the language perpetually resists defining too precisely, irony in relation to the speaker’s awareness of their own position, and a quality of admiration for someone who has arrived at a view or a position through their own reckoning rather than by adopting whatever the surrounding culture considers correct.
The instability of based—its resistance to a fixed definition, its context-dependence, the way its meaning shifts with the speaker and the target and the ironic distance between them—is not a weakness of the term. It is the term doing what sophisticated vocabulary does: carrying a web of associations that the receiver reconstructs from context, from prior exposure to similar uses, from the relationship between this instance and the hundred instances that preceded it. A word that means exactly one thing all the time is a word that cannot carry more than one thing. Based can carry several simultaneously, and the receiver knows, in the moment of reception, which combination is active, because they have traversed this territory before.
Delulu is perhaps the most interesting case in the current vocabulary. On the surface it is playful—the sound of it is playful, the spelling is playful, it bounces off the tongue in a way that signals levity. Underneath the playfulness it has become a shared shorthand for a specific and recognisable human condition: the state of believing something about your situation that you probably know, at some level, is not quite true, and persisting in the belief because the persistence serves you emotionally, and being aware, at a further level, of the persistence and the service it provides, without this awareness being quite enough to dissolve the belief. This is not a simple concept.
Philosophers of self-deception have written at length about it without producing three syllables that capture it this efficiently. Delulu captures it efficiently because it carries the tone of its subject—it is itself a little delulu, cheerful and self-aware about the nature of its own cheerfulness—and the tone is part of the meaning, not decoration applied over it.
These are not random noises. They are compressed conceptual packages, and the compression is the achievement, not the limitation.
The abbreviations deserve separate attention, because their history illuminates what Gen-Alpha has done with a long tradition. Abbreviations have historically arisen from material constraints: the cost of telegraph signals, the character limits of early SMS, the procedural brevity of military radio communication, the economy of printed space before cheap paper.
Gen-Alpha inherits some of these abbreviations—the ones that migrated from SMS culture and early internet culture—but the inheritance has been transformed. Many of the abbreviations persist and proliferate long after the technical constraints that might have generated them have disappeared. On a modern platform with no character limit, with voice input available, with every technological affordance that should make brevity unnecessary, the language continues to compress. This tells you something important: the abbreviations are no longer about saving characters. They are about something else.
What they are about, specifically, is the performance of a boundary. Iykyk—if you know you know—does not save typing in any meaningful sense. Four letters do not represent a significant economy over the full phrase they compress. What iykyk does is mark the distinction between people inside a reference and people outside it, perform that distinction in the act of utterance, and make the marking the entire communicative content of the phrase.
The phrase is not saying: I will save you the explanation because explanation would take too long. It is saying: I am not going to explain, and the not-explaining is the point, and the fact that you know what I am not explaining proves that you are the kind of person for whom no explanation is needed. This is a gate, as noted earlier. The interesting thing about a gate is that its function is performed by its existence, not by anything on either side of it.
What all of this requires, from the people who speak the language fluently, is not what conventional educational frameworks tend to measure. It does not require encyclopaedic knowledge organised into stable categories. It requires what might best be called cultural navigation: the capacity to move rapidly and accurately across a densely interconnected symbolic environment—games, YouTube, TikTok, Reddit, Discord, anime, memes, psychology, politics, fandoms, AI—in which references accumulate and evolve and illuminate each other and drift in meaning, and in which the person who cannot follow the drift is not merely uninformed but, more fundamentally, unequipped to read the territory at all.
A conversation fully fluent in this language might contain, without any explanatory apparatus, a gaming mechanic, a psychological concept, a Korean loanword, a physics metaphor, an anime reference, a meme from six months ago that has already begun to shift in meaning, an AI term, and then an abbreviation that collapses the whole accumulation into four letters and assumes the receiver can reconstruct it. To someone outside the ecology, this looks chaotic. To someone inside it, it feels economical, in the precise sense of doing the most work with the least apparatus.
That economy—the sign carrying not one piece of information but an entire web of associations, the receiver reconstructing the web almost instantly because they have traversed similar paths hundreds of times before—is the economy of every sophisticated sign system. It is the economy of the Victorian noblewoman’s fan vocabulary, of the Polari and body-language codes of mid-twentieth-century gay communities, of the Pacific navigator reading the ocean.
The sign rewards the initiated receiver with the full weight of its associations and withholds that weight from the uninitiated one, not as a punishment but as a necessary consequence of the fact that the associations exist only in the context of prior immersion. You cannot receive the meaning without the history of exposure that makes the meaning available to receive. This is not obscurantism. It is how sophisticated communication has always worked.
What Gen-Alpha has done with this structure is to build it at a speed and at a scale of cross-domain synthesis that has no obvious precedent. The symbolic environment from which the language draws is broader than any previous vernacular tradition, in the sense that it spans cultures, disciplines, media formats, and temporal registers simultaneously. This breadth produces a literacy that is genuinely polymathic in a specific and modern sense: not the encyclopaedic command of a single extended tradition, but the high-bandwidth synthesis of many overlapping ones.
The person who can navigate this environment fluidly is not demonstrating intellectual emptiness. They are demonstrating exactly the cognitive capacities that the most demanding intellectual work requires: rapid pattern recognition, contextual rather than literal interpretation, comfort with ambiguity, continual updating, and an expectation that meanings drift rather than stay fixed—that language evolves through use rather than through definition, and that this is a feature rather than a failure.
The older observer who mistakes this for degradation has committed the error that Bassett-Wren committed in 1905 when he explained Mallard’s findings as evidence of the condition of their author. He looked at something correct and unfamiliar and read the unfamiliarity as error. The error was his. The finding was not wrong. It was early, and it was in a form that the receiving apparatus of the discipline could not process, and it sat in a file for a hundred and fifteen years while the world continued to be exactly as the finding described it.
Gen-Alpha language is not degraded English. It is English that has been put under a new kind of pressure and has responded, as language does under pressure, with a new kind of precision. The precision is not visible to instruments calibrated for the old kind. That is a calibration problem. It is not a language problem. The language knows exactly what it is doing. The remarkable thing is not the language. The remarkable thing is how confidently the people who cannot read it assume that there is nothing in it worth reading.
The author acknowledges an initial response to Gen-Alpha language that was less generous than the language deserved, and regards the revision of that response as more interesting than the initial response itself. Both responses are noted here as data.
You are here: DUX