Unweaving the Web: 3 Lessons from Inside the Internet
A deep dive into the gross wonders of the Internet and 21st century journalism, featuring snakes, generative AI bots, a secretive mukbanger called Kate Yup, and a man who makes Pepto-Bismol pancakes.
Who makes the Internet?
If you, like tens of thousands of others, watched Glass Onion: A Knives Out Mystery and, noticing the unusual wrinkles on actor Dave Bautista’s head, turned to Google to find out how they got there, or if you saw a TikTok video claiming that the Large Hadron Collider at CERN was actually summoning vortexes and producing Mandela effects and therefore represented an existential threat to human existence, or if you’ve ever found yourself down an Ultimate Fighting Championship rabbit hole, hunting tidbits on Dagestani fighter Khabib Nurmagomedov’s elusive wife, then you may have read my work.
Writers like me, at least when we’re writing in the capacity I’m talking about here, tend to be invisible. We don’t write for a regular audience, and least of all for our friends. We write for a nebulous mass of Internet users.
We make the Internet.
And yet, no fewer than two (2) of my friends stumbled onto articles I had written in this capacity and, noticing my author bio, contacted me to say, ‘so this is what you do’. The website I wrote these articles for plays host to lots of pieces like this, answering some of the scrolling public’s weirder, more urgent, and less important questions. Along with innumerable other sites, it operates on the basis of tracking, in real time, what people type into Google, and publishing articles that provide the relevant information to satisfy those queries and, they hope, rank first in Google search results.
For anyone that doesn’t already know, this is how a lot of the Internet works. A lot.
I spent more or less two years writing for The Focus, from its infancy through its boon months and to its eventual petering out and temporary dormancy, under the employ of a medium size British digital publisher. I worked on a freelance basis but was effectively full time for much of it. The website was launched during the first of the UK’s COVID lockdowns in 2020, and positioned itself as a safe space for furloughed and out-of-work journalists to keep their writing brains active, and freelancers like me to get onto the ladder’s first rung. The site didn’t pay the writers, nor was it very selective about what it published. It was just a small, benign corner of the Internet.
But it quickly grew into a profitable 24-hour operation, and the publishing company at its helm started to pay writers on a per-article basis. Those of us who went full-time (I didn’t) received a pro rata salary plus perks for good performance. The company’s flagship website had already been running for 20 years at this time and was, and still is, its primary earner, but for a while our fledgling website was a key part of the publisher’s portfolio, some of our more ‘successful’ articles racking up views comfortably in the six figures and generating, one presumes, considerable income.
And it taught me some important lessons about the role Google, and the Internet at large, plays in the lives of ordinary people – or, rather, Americans, since they were 80% of our readers.
Journalist and editor Alexandra Ciufudean has summed up some of what she learned from these weird front lines in an article titled ‘Lessons from Google Trends about the American psyche’, published recently on her personal website. Her experiences of trawling and monitoring Google Trends for months on end gave her the impression of a United States plagued by loneliness and ignorance, and preoccupied with questions about ethnicity, sexuality and gender. And, I would add based on my experiences of doing the same, borderline obsessed with finding pictures of celebrities’ feet.
Ciufudean’s summary of how Google functions, and the way people use it, is as instructive as any I would be able to write, so it bears repeating here. She says:
The way Google’s search engine works is based on keywords – words, phrases or fragments of speech that users type into the search bar to find what they’re looking for. If you’re looking to buy something for Mother’s Day you might use a keyword like mothers day gift ideas or, if you’re trying to refine the results a bit, you can add an adjective – cute mothers day gift ideas – or the year – mothers day gift ideas 2023 – to get recommended the latest trends.
Note: this is what people already do, en masse, not what we’re recommending you start doing. Personally I didn’t do many pieces on gift ideas. Websites like Good Housekeeping sort of have that covered. Ciufudean continues:
Or, if you’ve just read a disconcerting headline on X (FKA Twitter) that, say, Republicans are rooting for a “red Caesar”, and you don’t know what that means, you might type in the keyword red caesar meaning.
Keywords are your best bet for telling Google what you’re interested in. The more detailed the keyword, the narrower the results, and the likelier they are to be exactly what you’re looking for. Type in sylvester stallone and you’ll get a sea of articles, interviews and artist profiles. But if you type in why does sylvester stallone own an abandoned mansion because of that one viral TikTok, this narrows down to a handful of articles exploring one singularly narrow rabbit hole.
A handful that includes ours, I should add. The trick of this kind of work is to catch a keyword that’s on an upswing. If it’s already climbed too much, larger and more authoritative websites will probably already have put out an article responding to it. Longer keyword phrases like the one above are good because their length makes them ultra-specific. The amount of words in the search term is enough to weed out competition from more generic (but also more authoritative, to Google’s all-seeing eye) information websites like Wikipedia and Britannica.
Using Google Trends to find out what people had been typing into their search bars formed the backbone of our practice. ‘Practice’ as opposed to, say, journalism… I understand that eight people picked at random will have eight different definitions of the word journalism – half will probably shiver when they hear it – and I spent a lot of time thinking about whether or not I could justify calling myself a journalist when performing this kind of work. On the one hand, we (almost) never asked anyone any questions – beyond requests for comment, which were more about avoiding litigation than seeking truth (this may also be true of many news organisations). We were not adversarial (although, these days, few ‘news’ organisations can afford to be). And the process we worked by, almost by definition, gives audiences what they want rather than what they need: it’s hard to shoehorn meaningful information about the nakba or Laos’ Secret War into a piece on late rapper XXXTentacion; one cannot plough one’s own furrow if content is dictated by what people are already searching.
On the other hand, our whole mode of content production placed us in the wake of the news cycle. We were beholden to what was playing out in the actual news media, and that made us feel like we were part of the journalism ecosystem, even if we weren’t anywhere near its centre. And the absence of any discernible editorial line gave us writers a degree of freedom to explore and portray issues according to our own journalistic principles. So… some of us were more journalists than others. Which opens up a whole other can of worms. But suffice it to say for now that from my perspective the apoliticalness of the whole operation precluded it from being journalistically bad in at least that particular way. There wasn’t an editorial line to be (mis)guided by.
And we weren’t alone! There are countless websites running very similar operations, from sleazy ‘content farms’ (on which more later), the ultimate bottom feeders of the Internet food web, to more substantial organisations like Distractify and AmoMama, some of which can be seen to have ‘graduated’ from marginal SEO ops to semi-legitimate news-ish websites with a sense of identity, readers who recognise the brand, a bit of money available to take chances on more investigative pieces, etc. BuzzFeed and HuffPost also started in small corners of the Internet. And because such a large proportion of us are using Google to find news and information, even legacy media companies (e.g. The Sun) have teams of SEO journalists monitoring real-time Google searches and pumping out 350-word answer pieces to get users onto their websites.
Many things led to this state of affairs. I will describe two. One: fewer and fewer people buy newspapers. They’re on the Internet instead, reading their news for free, meaning organisations have to hunt the money, and the easiest way for such organisations to do that is to scrabble around for advertising revenue generated by your attention. Two: it’s because of operations like ours, siphoning advertising revenue away from (once, arguably) reputable organisations whose actual (stated) aims resemble those identified by journalism scholars as the true purpose of the fourth estate.
Explanatory sidenote: SEO stands for search engine optimisation. Because of how people use the Internet, SEO is fundamental to the success of most websites.
When you want to find out a particular piece of information, what do you do? You probably don’t type the URL of an official, reputable website that’s likely to have that information, and use its search bar to filter through its pages/posts until you find it. Nor do you use Google’s various search tools to narrow your Google search to filter out all but the most creditable websites. You type your question directly into the search bar at the top, or a few relevant words that you hope Google will use to dredge up the information you desire.
Google then does its biz, trawls the Internet, and ranks its results according to how useful it thinks they’ll be to you.
SEO can take many forms, all of which relate to how Google works and how people use Google. (Google dominates the world of search engines. Internet writers do not optimise their webpages for DuckDuckGo users.) You, the user, want reliable information that’s well sourced. If other websites link to it, even better, because that shows that it’s authoritative. You also want a webpage that loads quickly, because you’re impatient and can’t stand (or sit) waiting around. One of Google’s criteria, therefore, is how fast the webpage will load. A webpage filled with large (in terms of file size) images, graphics and videos will likely rank lower in Google’s search results for this reason. So SEO can take the form of compressing images to increase load speed. It can also take the form of writing content for other websites that links back to your website, so that Google will think the original website is more authoritative.
Basic SEO techniques for Internet writers include things like weaving the keyword into the intro of the article text, inserting it into one subheading, breaking the article up into sections each comprising no more than 300 words, and so on. Doing these things helps Google understand that you are writing with readability in mind.
The three criteria most relevant to us as writers were clarity, originality and trustworthiness. For our operation to work, we had to show Google that we could structure our articles clearly and without waffle, that our content was original (in a clinical sense, i.e. not plagiarised), and that Internet users could trust us for up to date and relevant information. In this sense, we were forced by 1) our end goal (generating advertising revenue) and 2) the path we had chosen to get there (appeasing Google’s algorithm to make it place us first in its search results) to write articles that at the very least were… not garbage.
Last paragraph of this explanatory sidenote: there’s a company called Moz that predicts how likely a website is to rank highly in search engine results pages (another acronym: SERP). Moz gives websites and individual webpages a domain authority (DA) score out of 100, based on several factors. Key among these is how well linked it is. For example, if The Guardian, Wikipedia, The Boston Globe and the South China Morning Post all link to your website, Moz will give it a better DA score, which is a prediction that Google will recognise this and consider it a more reliable result. It may be unsurprising that www.google.com has a DA score of 100!
My day-to-day involved opening Trends and typing in what we called ‘seed’ keywords – generic words or phrases that commonly appear as part of more specific searches. Then I would look over the related keywords to see what specific search queries were popular during the previous four or 12 hours. Ciufudean gives the example of ‘meaning’, along with several that relate to people, such as ‘age’, ‘ethnicity’, ‘is black’ and ‘is jewish’. For those of us covering the movies and TV ‘beat’, regular seed keywords included ‘ending’, ‘explained’, ‘does x die’, ‘how many episodes’ and ‘was y cancelled/renewed’.
There were periods when my go-to seed keywords included ‘elon’, ‘kanye’, ‘republican’, ‘democrat’, ‘trump’ and ‘biden’, but fear of litigation eventually forced us to steer clear of discussing Trump and Kanye.
We were supposed to publish six articles per day. The actual writing, editing and formatting of an article would usually take somewhere between 30 and 75 minutes, and depended on a multitude of factors not worth describing in detail here; they’re boring and miniscule. Many made a difference of only a few seconds. Writing could take as little as 10 and as many as 60 minutes. Sometimes I’d get lucky and find the information I needed in a few minutes. Sometimes I’d spend half an hour trawling Instagram for proof of a minor celebrity’s age, or when they started dating such and such. When it was decided that we should write articles with four subheadings of four paragraphs each (rather than three/three), writing time increased by as much as 50%. This was a bummer, as the daily article quota and per-article fee remained the same. But it also meant the articles themselves were (mostly) better – more detailed, more discursive – so, you win some you lose some. Factors affecting speed also included which site the article was for and how recently that site had changed its rules for how to structure and format articles.
But the single factor with the biggest impact on how long the whole process took from beginning to end was how quickly you could find your keyword.
Without a keyword, there is no article. When I started out, my editor provided all my keywords. (A keyword usually consists of multiple words.) As I became more experienced as a writer, I started looking for my own, and eventually I became fully independent. Because we each had a daily article quota, and we received remuneration on a per-article basis, the surest way to guarantee a stress-free day of work as an independent contractor-freelance journalist was to increase the speed and reliability of keyword searches, so as to minimise time spent not writing. Or to put it another way, to decrease the time wasted on finding keywords.
One seed keyword that I found to be particularly reliable (fertile?) was ‘death’, because usually, without too much searching, it put me on the trail of at least one writable story per day. Often it was two. These were most often incidents where students, major and minor celebrities, and obscure personalities from the sitcoms of yesteryear had died in the last 12 or so hours. A lot of them were actually death rumours that needed debunking.
Death articles had a pretty safe go-to format that was easily replicable. And because of the detached, mathematical eye I developed while trawling the backwaters of Google Trends for search terms with sufficient percentage increases and amply buoyant graphs etc. etc., the ease with which ‘death’ led me to Facebook posts by American colleges mourning this or that basketball prodigy, or tweets (as they were known then) from the spouses of 1980s TV actors who had succumbed to chronic disease, or TikTok videos explaining ‘we need to talk about mental health’ because a mukbang YouTuber hadn’t upheld their regular posting pattern and must therefore have passed... the ease of writing those stories, which were straightforward and formulaic (except, actually, in the case of the mukbanger – more on her later) and involved finding only the most basic biographical information about the deceased and relaying some words of condolence… it preyed on a cold, calculating part of me, and made me relax gratefully whenever I found such a story, and sigh with disappointment when I didn’t, since then I’d have to do some actual work.
It was a bit morbid. It was peak SEO journalism.
If ever there was a story about something happening in the UK, either because it was important enough to American search engine users to come up in our efforts or because an editor reckoned it would work well at home, it led to a peculiar feeling of suddenly being back where I belonged, writing from experience rather than research.
And it put all the American news into perspective. There I was, writing article after article about American politics, people, brands and trends, for an American audience, using American spellings and grammar rules and including references to other American ephemera, and all the while doing so as someone who has never been to America and, for much of the time, didn’t care to. The whole charade made me feel somehow like a fake, a pretender. I was ingratiating myself, literally adopting the linguistic conventions of another people, while almost completely ignoring ‘my own’, and chances are, very few of them ever even considered the notion that I might not know very much at all about any of the things I was writing about so authoritatively.
The added detail that we were instructed to log our employment on our LinkedIn profiles under the job title ‘freelance journalist’ is a minor one but worth noting, since, as I understand it, only a very small number of the writing (or any) contingent (as opposed to HR, backend, etc.) were actually trained journalists, had an NCTJ qualification, or came from a reporting background. I didn’t. And that was OK because what we were doing wasn’t journalism per se (depending on how you… you know), and of course people can be journalists without training – no doubt there are people out there with good arguments that journalism training kills the journalistic instinct – and if we’d been instructed (or even encouraged) to refer to ourselves professionally as digital content creators, Internet text producers or even by the actual job title of content writer, a lot of us would have balked. So to call us journalists wasn’t wrong-footed, but it’s interesting to talk about, for me anyway. For what it’s worth, outside of LinkedIn I call(ed) myself a ‘writer’ because it’s the most generic and therefore usually true, and also admittedly because it sounds more romantic and sophisticated.
All of which leads me to the first of three important lessons I’ve learned about the Internet, during my time inside it, working for and within it.
A lot of what you read there – many of the words that make it up – was and is written by people like me: untrained, constantly making small personal decisions that end up coming across as larger editorial ones, far from expert in anything except possibly in the art of collating and presenting textual information in a way that appeals and is legible to average Internet users, and finding stuff out not by going to places and talking with real people but by trawling the rest of the Internet.
The Internet makes itself
There is another job I had, and in particular a thing that happened after I had finished the job, that helped me understand the machinations of the Internet, and it involves frogs, otters and an almost certainly made-up man called Sunny.
I used to write about animal husbandry – how to keep frogs, caracals, otters, snakes and other ‘exotic’ pets. It was my biggest source of income for a few months. I found the job on Upwork, a website that connects freelance writers (and others) to people and companies that want to pay for (usually textual and digital) stuff/content and don’t have anyone in-house to do it (or, I guess, any regulars that will do it for cheap enough).
The first job I found on Upwork was to write an advertising slogan for an ethical candle company, and earned me $100 minus fees. They sent me a sample slogan. I spent an hour writing punny alternatives. They made suggestions. Eventually, they chose one that was basically the same as the sample. Work smart, not hard!
Overall I didn’t find Upwork to be a satisfying place to find work. But there was this one job, writing guides on how to keep exotic pets at home, that gave me just about enough work with little enough chat and nag to be sustainable, at least for a while.
My contact’s name wasn’t Singh, but let’s call him that. His written English was limited. He would provide me with a brief, which included a title, something like Are Grapes Safe For Bearded Dragons? or Where Does The Fennec Fox Live? or Vietnamese Mossy Frog Pet Care, a word count of up to 4,000, and a quote for $0.025 per word. I would deliver the goods, learning about terrariums, vivariums, substrates, cloacas and all manner of rosy boa morphs as I went. Over a period of seven months, I made [checks notes] nearly $3,000 writing these guides, and not once did I check the website where they were all being published to see how they looked once they were online.
Fast-forward a year. I’m curious. I’m an Internet user. I go onto the website to see how it’s doing.
My articles are there among a plethora of other guides. Together they form a chorus celebrating the myriad joys of keeping tomato frogs, finger monkeys (aka pygmy marmosets) and marble foxes, and warning against the perils of keeping otters and caracals captive. And they look much as I remember them, edited for SEO down to the last tittle and crossbar, and made fun and engaging with the not-so-judicious and possibly overzealous deployment of exclamation marks!!!, overly short sentences, wildly unnecessary adjectives and adverbs, and instances of second person address. It’s awesome. They’re the cutest. You love it. Everything is there. That is, except for one thing: my name.
It turns out, many of the articles I wrote were actually written by someone called ‘Sunny’. He ‘loves working out’ and ‘beat [sic] everyone at games’. ‘Lou’ wrote some of them. Among other things, she’s ‘the right person to go to’ if you want to learn more about snakes and reptiles. I should hope so too! These people have profile pictures and author bios. Are they real? Did they write anything? Are they Singh in disguise? Did the website’s slippery editors borrow their visages from LinkedIn? This is not uncommon practice, as I’ll explain later. Do the real people whose faces are papered all over this website have any idea?
Sidestep: One of our peripheral duties as journalists writing for the website I spoke about before was to type in the keyword for which we’d just written and published an article into Google and see what came up.
This way we’d see if we’d succeeded in picking a keyword that was specific enough for our website to rank for, and if the article we’d written was optimised enough for Google to place it first in its search results. If our article came up as the first result, we’d ticked one of the essential boxes of post-SEO journalism. The other reason we did this was to search for ‘copies’: articles based on ours, and published within minutes or even seconds of ours, whose purpose was to siphon off our advertising revenue.
Sometimes these were transcribed verbatim, cannibalised and churned out with algorithmic celerity, a hungry, parthenogenetic Internet chewing on its own tail. Sometimes someone (or something) had rewritten (but usually not restructured) them so as to (attempt to) fool Google into thinking it was unique text – original, authoritative, trustworthy, Googlable. These ones took a little longer, perhaps a few minutes.
The Internet is a busy place populated by both ingenuous and disingenuous actors. Copies are the work of the latter. If any copies came up, we were supposed to flag them with the tech team, who would submit a removal request to Google. When Google detects plagiarised material, or even duplicate material, it punishes websites by making them rank more weakly in its search results. This meant there were two forces compelling us to find and report them: first, the website plagiarising our content was stealing our revenue; second, Google might incorrectly presume we had copied them, and punish us.
In the example above, I think it was a mixture of humans and bots performing the task of copying our material. The humans did it better, the machines faster. But the reason I’m bringing this up in the context of ‘Sunny’ and ‘Lou’, the possibly made-up writers credited with writing my snake articles, is because the copies had to have real people listed as writers in order for them to look at all reliable, to readers. Real people need names and faces. And once, I Googled the name of one of these writers, and I found her name and face on LinkedIn, and she was a human rights lawyer from Woking with a big smile and a glint in her eye. Another was a photographer based in Northern Ireland with a chin like a shovel.
I felt a biting temptation to reach out to these people, to tell them some faceless content farm was using them for cheap profits.
I never did, but this is the second lesson about the Internet that I learned during this time.
Much of it is produced by machines, or humans acting as machines, and consists of overt plagiarism. It is a morass, and constantly feeding on itself, self-reproducing, consuming, birthing, an ouroboros whose tentacles sprout tentacles of their own, which they then eat: autophagic, insatiable, unstoppable.
A perfect metaphor
Remember earlier I mentioned a mukbanger?
I was reminded of mukbangers and how they relate to Internet culture at large by an article written by a friend of mine for Off-Chance, titled ‘Eating With Our Eyes – The Obsession With Food Content’. Liam Murphy starts by talking about an article the Boston Globe published in 1901, in which reporter Thomas F Anderson imagined what Boston might look like in 100 years.
A central facet of Anderson’s vision of the future was a series of pneumatic tubes that would deliver all kinds of things to households across the city. Newspapers, communications from friends and correspondents, even food.
Over 100 years later, Murphy writes, not many of Anderson’s predictions have come true. The Internet renders physical tubes obsolete, shattering many of our perceived dependencies on physical objects and throwing vast swathes of the human experience into the realm of ephemera. The newspaper has already been delivered, everywhere. The digital world is all at once. And while food itself still requires transportation, everything else to do with food exists in abundance, in the Internet.
A mukbang, also spelled meokbang, is an eating show, or ‘eatcast’. Its name comes from the Korean words for ‘eating’ and ‘broadcast’. Mukbangers record themselves eating large amounts of food, usually while interacting with their audience. Those that pre-record their shows obviously don’t interact meaningfully, but that’s the gist. In turn, audiences get voyeuristic pleasure out of watching bizarre and implausible feats of consumption.
According to Mikaela Helane of Pattern magazine, the phenomenon started in 2009 when, in order to stave off mealtime loneliness, South Koreans began streaming themselves eating on the peer-to-peer online platform AfreecaTV. Fans would send messages and donations to their favourite streamers, and soon enough, for better or worse, the whole world was mukbanging. South Korea first emerged as a major exporter of popular culture in the late 1990s – Taiwanese journalists first coined the term ‘Korean Wave’ (aka Hallyu) in 1998. Success stories include BTS (formed in 2010), Psy’s earworm ‘Gangnam Style’ (2012), Blackpink (formed in 2016), Bong Joon-ho’s inimitable Parasite and, in case you were living under a rock for the entirety of 2021, Squid Game. Hallyu catapulted mukbangs onto the global stage. By 2014, they were all over YouTube.
I once spent longer than I should have reading about, and watching, a mukbanger called Kate Yup. No one is sure where she’s from, but she speaks French and some English. She pronounces the L in ‘salmon’ – accent experts, eat your heart out. Someone on Reddit thought she might be European female UFC fighter Joanna Jędrzejczyk, but they don’t look that alike. And she’s known for consuming ungodly amounts of seafood. One of her videos is titled, ‘I EAT 3KG OF MEAT in 3 SLABS OF BUTTER AND OIL’. Yes, all in caps except the word ‘in’. In another she substitutes butter for bread to create (and eat, obviously) the ‘world’s fattest sandwich’. She also eats a lot of raw meat, and makes a lot of mess. One time, she crushed her meal with her feet before eating it. Becoming successful online is partly down to fate and luck, but at least three things worked in Kate Yup’s favour: she performs feats of consumption many miles beyond the realm of possibility for most of us (see it to believe it…); she is slender, elegant and attractive; and she wears a mask covering the upper part of her face, adding an element of mystery and intrigue. What put her onto my radar, however, was the torrent of conspiracies surrounding her.
In one video, she has a bruise on her left hand and a cut lip. In another, if you listen really hard, you might make out a voice whispering words like ‘fast’, ‘just eat’, ‘hurry up’, and ‘I’m gonna kill you’. Red flags for the eagle-eared. A third video contains the text: ‘the meat Sooo deliciOuS .... soft and tender’. Notice the capitalisation of the letters S, O and S? Well, someone did! And in the caption of a fourth video, the first letters of four consecutive sentences are H, E, L and P. Some reckoned she was tapping secret messages in Morse code while chowing on, I don’t know, a cow’s udder. And in another video, she appears to lose two teeth. Explanations for her behaviour ran the gamut from kidnapping to eating disorders. In the sole post on her channel’s Community board, she reassures her fans that ‘EVERYTHING is OK for me’, and that no one is forcing her to eat; the Sun caused bruise-like welts to appear on her arm; and the wound on her lip is herpes liabilis. Her ‘disappearance’, which lasted from November 2019 until October 2022, only fuelled the fire.
When she returned in 2022, all hell broke loose. She posted a video with a thumbnail bearing the words ‘I’AM ALIVE’ [sic] – fishy – in which she claims via on-screen text to be the ‘only person in this place’, yet the camera moves without her touching it. The descriptions of the foods she eats are all in lower case except for (again) the letters S, O and S in ‘raw SalmOn’ and ‘fries Shrimp’. And the hands that load the food into containers at the beginning of the video look nothing like hers, which would be totally normal if not for her assurance that she’s alone. Is she? Isn’t she?
It all made for riveting reading for someone like me when I wrote about it for The Focus, and catching up with all the latest developments has been engrossing, but get this: the consensus among many of her former fans is she has been playing us all along. It was, and is, all a controlled lie, they say, and the SOS/HELP messages, arm bruises and inexplicable camera movements are there to provoke concern – concern that, if you’re an online content creator, is monetisable. Which, if true, adds a whole other dimension. Is it totally morbid to consider the falseness of it all a disappointment? That she wasn’t, in fact, a victim of domestic abuse ought to come as a huge relief, but the idea that she took her fans for fools, and made a lot of money in the process, is hardly appetising. About a year ago, Kate Yup – or whoever manages her channel! – stopped enabling comments on her uploads. I like to think this is in response to one commenter writing, ‘I can’t believe this woman managed to gaslight the entire internet’.
The whole charade proved instructive, and it plays into the narrative I’m trying to build about the Internet and dishonesty. Plus, there’s something poetically awful, or awfully poetic, about content creators feeding their audiences lies while literally overfeeding themselves.
It’s tempting to think of mukbangs as existing in the same cultural space as hot dog eating competitions, but they’re very different, for reasons I’ll explain.
Joey Chestnut is an American competitive eater I know too much about. Since the mid 2000s, Chestnut has been the most famous and successful competitive eater to grace American stages (and television screens). Now 40, and from Kentucky originally, he is a white man with a large-ish jaw and high forehead. In 2007, he ate 182 chicken wings in half an hour, 108 burgers in eight minutes, 45 pulled pork sandwiches in 10 minutes, and 56 sausage and cheese kolaches in eight minutes. A kolach is a Czech/Slovakian ring-shaped sweetbread usually filled with cottage cheese, fruit jams and/or poppy seeds.
Joey Chestnut is not fat.
In 2011, at an event in New York, he ate almost 6kg of potatoes. Two years later he ate 141 hard-boiled eggs in eight minutes. The list goes on, and includes a frightening amount of deep-fried asparagus. To watch Joey work is almost as hard as it is to be Joey. Beads of sweat roll down his pinking forehead. Residual mush and sludge clump on his chin. Nausea awaits anyone who keeps their eyes open.
For modern audiences, Chestnut put eating competitions on the map. But the venerable title of ‘godfather of competitive eating’ goes to a Japanese man, Takeru Kobayashi. He’s 45 and 5ft 8in tall, and cut his proverbial teeth eating more bowls of ramen than you could possibly shake a chopstick at. When he first entered the (US) ring in 2001, at the July 4th Nathan’s Coney Island Hot Dog Eating Contest, he doubled the previous record (of 25⅛).
In 12 hot dog-fuelled minutes, he launched competitive eating into the professional realm. America didn’t know what had hit it. The show’s organisers started scribbling numbers onto placards because they didn’t have signs big enough to track his achievement in real time. And no one could catch up with him. He won six consecutive competitions and broke his own record three times. In 2006, he ate 97 Krystal Burgers at the (then-) official World Hamburger Eating Championship in Chattanooga, Tennessee, beating the world record by 28. Krystal is a fast food chain known for square-shaped slider-burgers that are pretty insubstantial. But still, you know, 97 small things equal a very big thing.
Kobayashi (aka ‘The Prince’) and Chestnut (aka ‘Jaws’) are American emblems, even if one of them is Japanese. Their endeavours strike out from the promontory of the New American Dream, an ideal of which is to exist in a state of such surplus as to be able to performatively consume too much. In staging these spectacles of consumption, American society performs an autophagy of its own. Members sicken themselves – stretch and warp their physical bodies with gruelling training regimens – in order to triumph at over-consuming. Because participants medically self-sabotage, and because the competitions themselves serve as promotional events for corporations that, in turn, sicken their customers, these events can be called festivals of (self-)consumption. They create a world in which consumerism meets actual consumption.
And they are analogue. They happen in real life and real time, in locations people make real by their presence. Even though they are broadcast and watchable on TVs and on the Internet, they are – at their core – analogue events because people go to them and are physically in the same space as each other, joined in active, physical participation.
Not so with mukbang. Mukbangers do what they do only in front of a camera. Digital mediation is intrinsic to the phenomenon. The Internet is the location as well as the means of communication.
In his piece, Liam Murphy dives into the eco-system of ‘online food’. First he encounters recipe developers and dieticians who create content so that their audiences can emulate it. Then comes a second demographic: people who have no intention of ever making the dish but ‘enjoy it as content’. He calls them ‘wayfaring scrollers’ and ‘passive gourmands’ who ‘eat with their eyes’. They (or, admittedly, we) scroll in order to satisfy their ‘reality-independent’ gustatory imaginings, consuming content instead of the food it depicts.
Then, Murphy writes, there are content/culinary creators who make deliberately weird things that satisfy a different impulse altogether. Welcome to the world of ‘post-food’ content.
Consumers of this type of online content don’t even consider re-creating the dishes offered to them. He offers Pepto-Bismol pancakes and spaghetti ice-lollies as examples. People watch it ‘because they don’t like it’ (emphasis added).
As I see it, the post-food consumer-creator relationship is interesting because it involves an element of sadomasochism: creators make content that it is uncomfortable, on some level, to watch; audiences watch willingly, cringing and grimacing with delight. Post-food content steps outside the slipstream of creating good things that are good for people, just as postmodern literature moved beyond the modernist goal of refining and improving upon the process of writing fiction, and postmodern philosophy questioned traditionalist notions of progress towards an ever brighter future.
Murphy moves on to observe post-food content creators whose culinary processes become cyclical, producing the ingredients they started out with. Like someone turning water into ice and melting it again, except… more so.
We see a bowl of cereal made, chucked into a blender, become batter that is then fried into a pancake. That pancake is then chopped up into chunks and put in the oven. Those chunks are then baked, becoming cereal (again), and the process starts again, this time making a brownie… which is then made into cereal.
But when it comes to analysing the relationship between content creators and content consumers – or, Internet producers and Internet users – I think we can go even further.
We have:
1. People consuming food-related content without any intention of eating, i.e., dissociating from the act of consuming actual food, or consuming content as food;
2. Sadistic creators serving masochist consumers, whose relationship integrates a complete dissociation of food content from food consumption;
3. Post-food content creators performing repeated processes on ingredients that eventually become what they were in the beginning, i.e., process without result, or creation without product.
Mukbangers enter the picture as creators who have to consume to produce, whose audiences actively suffer while watching them eat, and who themselves actively suffer while eating. In the world of mukbang, food loses all connection from its traditional meaning. It doesn’t sustain, as it traditionally has – it doesn’t even educate or entertain, because the consumers who feed on it experience nausea. Instead of satisfying hunger, the role of food in a mukbang is to produce hunger’s opposite. It is a complete subversion.
The press agent credited with ‘cooking up’ the July 4 Nathan's Hot Dog Eating Contest is Mortimer ‘Morty’ Matz. He’s nearly 100, and by all accounts still going strong. His version of the competition’s origin story reveals much about the lens through which America views itself. On 4 July 1916, Matz claims, four immigrants held a hot dog eating contest at a Nathan’s Famous stand on Coney Island to settle an argument – about who was the most patriotic. A man called Jim Mullen won, and a precedent was set: overeating and patriotism were indelibly etched onto opposite sides of the same coin. However, in Horsemen of the Esophagus, author Jason Fagone explains that this story is likely untrue, which to my mind reveals even more. This is the lens through which America desperately wants to view itself. And it will invent lies to write its own history.
Competitive eating contests held in the US are an ideal metaphor for understanding American culture, colourfully combining consumer capitalism, the literal overconsumption of ultra-processed foods, mass media, national pride in the Declaration of Independence (many are held on the Fourth of July), American exceptionalism and competitiveness, and the American Dream. Gluttony, advertising and cash prizes.
Mukbang presents a correlate when it comes to Internet culture, even if it feels like an altogether darker phenomenon. Some of the same elements are there: overconsumption, the subversion of food-as-nourishment, and the attainment of fame via acts of heinous gluttony. But mukbangs do without the patriotism, lurid sponsorships and, crucially, the realness of space. The absence of competition renders them meaningless. There is no one to cheer for, no flag to wave in celebration. There is no communion, no jostling, no sharing of sweat. Audiences are stripped of their role; they do not yell. Mukbangers perform nauseating tasks in no one’s company but their own – and the camera that transmits their performance to their faceless voyeurs. Digital mediation inheres in the mukbang, whereas it is secondary to the world of Joey Chestnut and Takeru Kobayashi. And digital mediation, as we saw in Kate Yup’s alleged exploitation of the compassion of her audience, allows for deceptive stagecraft and yarn weaving. Just as in-person eating contests take the act of eating and turn it into something destructive and unnatural, mukbangs take the act of using the Internet and perform a similar transformation.
They are post-food content par excellence, subverting the consumption of video and Internet content. At one end of the spectrum, we consume video content for the purposes of entertainment (cinema), education (documentary) and enrichment (art film), and in many cases we do so communally, with all the benefits that entails. Meanwhile, the Internet can help people find community. It can bring us closer in new and interesting ways. It allows for the instant communication of huge amounts of useful information, and presents opportunities for collaboration across cultures and entrepreneurship. Mukbangs, on the other hand, are typically watched alone, on smartphones; they neither entertain nor teach nor enrich. They are atomising, impoverishing, and uninstructive. They exist at the unenlightening end of the spectrum, exploiting the medium they rely on and, in the case described above, the compassion of their audiences.
And so here comes lesson three.
A slim, attractive, anonymous woman sits alone in front of a camera eating 3kg of meat. Her audiences, sitting alone in their bedrooms or tapping away on the tube, speculate that she has been abducted and that she performs her feats in a state of captivity. This brings them together. It brings communion. The woman encourages this narrative, all the while making substantial profits from the compassionate speculation of countless fans. When the project becomes too unwieldy, she simply stops posting. The balloon implodes, rendering the communion meaningless and fake. It is the perfect metaphor.
Welcome to the Internet.
Fascinating read. And I also learned too much about eating.