In his debut book Algospeak, Adam Aleksic, an American linguist who discusses the origins of words as the fast-talking @etymologynerd, argues social media is fundamentally transforming language. Not just the words we use, but how we speak and write, and how we think of ourselves and present our identities to the world.
Detouring around governmental censorship and corporate content moderation, younger generations are increasingly resorting to coded language. This can take many forms, including “Leetspeak” that uses character replacements (5U1C1D3), bowdlerisation ($#*!), emojis, “minced oaths” (seggs for sex), initialisms (SA for sexual assault) and taboo words (orange man for Trump). The Listener quizzed Aleksic about the new euphemisms, the fabled Gen Z stare, and what to do when your grandmother starts using the skull emoji.
You begin the book at a museum exhibition which had a caption: “Kurt Cobain un-alived himself at 27.” Unalive began as the first word noted that was invented to escape online censorship, but for younger people can now be almost a straight synonym, an easier way to talk about suicide, or used ironically. We have been euphemising death since forever, but is there a downside to this?
I have talked with dozens of educators and guidance counsellors about the word “unalive” and it seems like they’re pretty split. On one hand, it could trivialise the topic of suicide. On the other, it could give kids an opportunity to express themselves when they’re uncomfortable talking about such a serious thing. “Unalive” sounds less scary, which is why it’s taken on a genuine euphemistic function like other phrases – “kicked the bucket”, “passed away”, “deceased”. I would argue the same question applies to those euphemisms as well.
Will words like unalive last? How long will Gen Alpha slang like rizz, gyat, sigma, skibidi and delulu hang around?
What makes a word “stick” is a combination of several different factors. It needs to fill a lexical gap –a need in our language, outcompete potential synonyms, be adaptable to new contexts – and not stick out too much. In the case of “brainrot words” like gyat [dang] and skibidi [cool], we already see those fading in usage because they are tied to the lifespan of a meme and are dying out with that meme.
Your book highlights some fascinating language change, but isn’t it also true that humans always figure out ways to communicate information, tone, subtext, socially prohibited ideas, etc, and generations have always misunderstood each other, accidentally and deliberately? Isn’t part of the rapid change the sheer amount of written text being communicated?
One of the things I try to emphasise in this book is that everything we see online is a new iteration of social patterns that humans have been exhibiting for millennia. We’ve always had censorship avoidance and memes and in-groups. What is different is the medium. The algorithm is a new infrastructure underlying how we communicate, and as such is reshaping our language with it.
You write that online influencers have been changing word stress, speed of delivery, lengthening words and using “uptalk” – rising inflexion at the end of a sentence. Is this influencing offline speech?
The “influencer accent” is not influencing offline speech as much because it’s understood to belong to a specific social context. A good analogy is the “broadcaster accent” used on television: a “presenter voice” that someone can code-switch into and out of, depending on the situation. Influencer cadences are, however, affecting other people when they talk online, because now there’s an expected way they’re supposed to speak. It makes more sense for words to bleed into new contexts, not accents as much.
Social media has homogenised mass culture, but has also hugely revived niche interests, you note. But while New Zealand kids can, for example, indulge their endless fascination with insect-eating plants, it also means they are now increasingly saying candy instead of lollies or sweets, and even changing their pronunciations.
Language was homogenising before social media, due to globalisation. Regional accents and linguistic variation have been dying out throughout all of the 20th century, and algorithms are just accelerating that. Paradoxically, there is, in fact, more variation starting to take place online since we’re spending more of our time there.
These new language trends add meaning and creativity, but are we losing complexity and nuance communicating in one-minute clips or 280 characters? Do we always have to chase virality over substance? One rule seems to be that story remains prime.
Each new medium will uniquely affect how we communicate. Physical books constrain the author to write in formal English and segment stories into introductions and chapters and conclusions. Yes, algorithmic media is restrictive in its own way, and as such, shapes the way we tell stories. No matter what, though, we still are telling stories, because that’s the point of why we use language.
You write, “The old boundary between slang and ‘proper speech’ has been dissolved and redefined.” What is the role of language prescription through dictionaries and grammar guides now? There are always cycles predicting the death of the semi-colon or “whom” – do you foresee a swing back to more prescriptive modes?
Institutional gatekeepers like language academies and dictionaries have no say over what people do online. There will always be a need for prescriptive speech in schools and traditional publishing, but we are seeing a simultaneous flourishing of informal speech online driving language change faster than ever before. Both can exist at the same time, because we’re code-switching for different contexts.
In a piece for The Washington Post in May, you noted how the skull emoji is a tone tag conveying humour or irony to avoid misinterpretation. But why not just a winky emoji? Less-online people are often confused about tone and “semantic bleaching”, where a word’s meaning is softened over time. Is there any sign of pushback on this continual change?
It is precisely because of the less-online people that humour tone tags are constantly evolving. Now, it’s the wilting rose and heartbreak emojis. These tone tags are used to identify a sociolinguistic in-group: the people who are online and in the know. If your grandmother starts using the skull emoji, it’s over. So, no, you can’t stop the cycle of language innovation. That’s been happening forever in some form, but now faster in the algorithmic linguistic landscape.
Days after your book came out in the US, The New York Times moved aside four of its traditional critics. “Our readers are hungry for trusted guides to help them make sense of this complicated landscape, not only through traditional reviews but also with essays, new story forms, videos and experimentation with other platforms,” a memo to staff said. It struck me as part of the wider change you’re describing.
There is definitely a cultural undercurrent back towards human curation and tastemakers, especially in the age of AI slop and oversaturated content.
There’s the “Gen Z stare”, a vacant expression sometimes given to a customer in relation to a question. One explanation is that social media has left them unable to engage with others in mundane interactions until they’re interested. Is it real?
The “Gen Z stare” is mostly another fake generational label that I describe in my book. It works as viral trendbait and thus spreads online. That being said, it likely does identify an actual social pattern in how young people engage in the workforce.
Algospeak, by Adam Aleksic (Ebury Press, $40), is out in ebook and audiobook now and in paperback from Tuesday, October 14.