If you want to know how memes came to dominate the web, how they may evolve into "temes," and why they may hold the secret to the future of life on earth, then you better read this.
When Jay Maynard turned up at the Penguicon Linux-meets-sci-fi conference in 2003, he was just an average guy—an IT professional with a homemade costume. By the time he left, he was immortal.
He had become a meme.
Maynard—now better known as Tron Guy—had submitted pictures of himself wearing his homemade Tron suit at the conference to Slashdot. His images caught the imagination of the site's users, were passed around, appeared on other internet humour sites, took on a life of their own, and drew all the trolling and barbs you'd expect from the snarkier online commentators.
"Honestly, I did not expect the negative reaction I got. It was vicious, and juvenile. I drove home from the conference, and that 14-hour drive was no fun at all. I was thinking the entire time, 'Damn, what did I do?' That entire week was pretty terrible," Maynard said.
Maynard's hard time continued after radio show DJs sought him out for interviews, seemingly only wanting to take turns to bully him live on-air. Maynard later got himself invited onto Jimmy Kimmel Live. He was initially invited on the talk show on as a web curiosity, and he went on to parlay his first appearance into a regular roving reporter slot, gathering vox pops from the public at events including the Gay Hollywood Parade. That's when Maynard began to feel things were turning around for him and his newly-minted meme. Tron Guy did things Jay Maynard never would have.
Looking back on the highs and lows of being a meme, would Maynard have done anything differently.
"I wasn't wearing a dance belt [a type of thong favoured by male ballet dancers to hide their modesty], so it wasn't just 'look at the fat guy in spandex', it was 'oh my god, there's a fat guy in spandex and you can see his balls!'," he laughs. "Looking back, I don't know if I'd have worn any underwear, what effect that would have had. Would it have completely destroyed [Tron Guy] as a subject of memehood?"
Underwear or no underwear, the Tron Guy meme persists. Around a decade later, new variants are still appearing, often remixed with memes of an altogether newer vintage. It's a good thing Maynard's down with his memehood, because chances are there'll never be any getting rid of it at this point.
Memes are, at their most basic, a unit of culture, an idea that passes from person to person, getting copied, altered, or adapted along the way. As such, the term meme covers a huge variety of things—songs, art, books, architectural styles, fashions, religious tenets, the combustion engine, a chemical formula, the shape of a wok, the practice of ear piercing, and even words and languages.
However, meme is now mostly used as shorthand for the myriad of web memes, from the Gangnam Style video to Lolcatz—those nagging tweets, images, videos, and ideas that make us laugh, think, or react and then share them with our friends and followers online.
Jokes for the in-crowd: So knowing, much smug
Tron Guy became popular in 2003, not long after meme life on the internet had begun flourishing in earnest, fed by the likes of Slashdot, Fark, Something Awful, and 4chan.
Memes first hit the public consciousness properly in the 1990s and early 2000s: the Hampster Dance [sic], All Your Base, Badger Badger Badger, Dancing Baby, Star Wars Kid, Domo-kun's threat to kittens, 419s, and lottery scams, all arrived on our PCs in the space of a few years.
Hanging out on the internet in the late 1990s and early 2000s was far less common than it is now. The digital memes of those days reflected the sort of things that web users at the time found funny: sci-fi, broken English, animation, animals doing daft stuff. Humour was paramount for memes, and mostly in-jokes at that.
Memes became a sort of code—if you don't get it, you're not an insider, you're not part of the subculture. If you laugh at this stuff though, you're one of us, and creating and passing on internet memes is your secret handshake to show you're in the club. Over a decade later, all those elements that found favour with earlier internet users are still very much in evidence in today's memes.
Take Doge, for example: a dog wearing a bandana, with semi-coherent messages written over the top in perkily-coloured Comic Sans font. That's Comic Sans, a font so loathed and ridiculed as to have hate sites devoted to it. In Doge, Comic Sans plays the part of the in-joke. Whoever first slapped the font over the top of a picture of a Shiba Inu did so precisely because it was such a cheesy, reviled font, and because they knew that you knew that too.
"For a lot of those memes, people who make them or pass them are like, 'I'm going to layer a whole bunch of subtexts into it and it's funny for me because I understand it', but it's largely unintelligible for people outside of that specific group," said Kate Miltner, a Microsoft social media research consultant and holder of an MSc in Lolcatz.
By passing on a Doge picture, you're telling those you send it to that you're in the know, and you know they are too: so meme, much subculture.
As well as signalling to friends and co-workers that we're part of the clique, memes also serve to confirm how we think of ourselves.
The link between memes and identity is highlighted by Jonnie Hughes in his 2011 book on memes, On the Origin of Tepees, where the author cites Coca-Cola as an example. The company flourished under its first CEO, Asa Candler, who promoted the ubiquity of the brand, making sure it appeared anywhere and everywhere across the US, embedding it into the country's national identity.
According to Hughes writing in the The Independent, the power of this association is still with us, and can be demonstrated by MRI scanners.
"Measure the activity of the brain's pleasure centre as people drink different colas, and Coke comes way down the list. But measure the prefrontal cortex—the centre for self-identity—and it lights up like a candle. Coke remains number one not because it's more pleasurable, but because Candler's brand is the one we associate with our personality."
All you meme are belong to us—not the corporations
Jay Maynard was the second man to submit pictures of himself in a homemade Tron suit to Slashdot. So why did that initial Slashdot user not become Tron Guy, and yet Maynard, a decade on, is still known as his online alter-ego? What makes a meme successful?
While the subjects of memes are often baffled by their popularity and why they went viral, marketing companies are nonetheless trying to find a way to mimic the popularity Tron Guy found: trying to find the secret sauce that will turn their content, or their client's content, into memes.
Memes are cheap, easy to create, and spread like wildfire—all characteristics to excite marketing departments keen to getting their brand in front of young people. If you're passing on a corporate meme to your friends, you're acting as an advocate for it, and the product behind it. You're marketing it to your friends at no cost to the company behind it.
And yet, only the tiniest proportion of successful memes appear to be created by companies—only Will It Blend?, a series of videos by blender maker Blendtec, and the Man Your Man Could Smell Like, a campaign created by ad agency Wieden+Kennedy for the range of Old Spice male grooming products, have made it into memehood proper.
Most attempts by companies to create memes have crashed and burned for those reasons of identity—they either don't get memes, or they're perceived as outsiders trying to butt in on a joke-party they weren't invited to.
It's a phenomenon that's writ large in the political sphere. Take textsfromhillary—photos of Hillary Clinton bossing public figures with pithy putdowns. Should the meme have come from Clinton's PR people, it would doubtless have been met with stinging sarcasm. As it was the textsfromhillary Tumblr was created by two humourists and went onto enjoy moderate success until it was closed by its creators in 2012.
When memes are made by politicians, rather than average Joes, "it cuts against the ethic, the aesthetics, the way we think about participatory media—YouTube, Twitter, Tumblr, Reddit—we kind of envision them as grassroots, open, by the people, for the people, outlets, and so overt attempts at subversion are met with resistance. It's, 'this is our space'," Ryan Milner, assistant professor of communications at Charleston University, said.
Most memes are created or passed on by ordinary people, often hoping to give their friends a quick laugh. As they get passed from user to user, their origins become lost. While sometimes the subjects of memes are tracked down by the media, keen to discover the person behind the image, their creators rarely are.
Oh hai, I am in your politics, expressing subtle dissent
In some countries, authorlessness has an additional benefit—it allows people living under regimes where other forms of protest are strangled to express political dissent without the fear of being tracked down and punished.
China, for example, has a thriving meme culture. In a country where overt political censorship and the subsequent self-censorship it inspires is rife online, Chinese web users have turned to memes to express their opinions.
One example is the Grass Mud Horse, one of a series of fictional animals that were added to China's equivalent of Wikipedia—Baidu Baike—following a crackdown by Chinese authorities on online freedom of expression.
In Chinese, the name of the Grass Mud Horse resembles the characters for "fuck your mother" and it spawned a number of related species. The sworn enemy of the Grass Mud Horse, for example, turned out to be the river crab, an oblique reference to online censorship and the government officials in charge of it.
"It is quite obvious that the dirty pun of Grass Mud Horse represents the average internet users' anger and frustration at censorship, and with the help of digital technology this unique form of expression is evolving into a collective attempt at resistance," Bingchun Meng, a lecturer in the department of Media and Communications at the London School of Economics, wrote in a paper on "e gao", Chinese satirical political memes.
The Grass Mud Horse spawned soft toys, music videos, mockumentaries, and more, and eventually led to the bizarre situation where online platforms were warned by the authorities not to promote anything involving horses or crabs due to their political implications.
"Although e gao activities may appear to be rather incoherent or even chaotic, they represent innovative strategies for articulating social critique and fostering societal dialogue in a heavily controlled speech environment," Meng writes.
"The playful and often a-rational texts of e gao are partially born out of the difficulty of conducting serious rational debate on the Chinese internet. Meanwhile, the comic effect produced by these spoofs can have political connotations as it satirizes those who possess the power to define the parameters of appropriate forms of speech."
Equally, repressive governments have also worked out the power of memes, as in the case of Azerbaijan's current presidential administration, which recently created and disseminated its own "fail" image macros and "y you no" guy memes featuring opposition figureheads.
Political memes are of course equally big business in non-authoritarian regimes—image-based memes were common currency during the 2010 UK elections, 2011's Occupy Wall Street Protest, and the 2012 US elections. President Obama even embraced them personally: he wrapped up his AMA on Reddit by describing the whole experience as "not bad", a reference to a meme involving himself—Obama Rage Face—that sprung up in 2011.
While humour is one of the strongest traits of online memes of all stripes, memes with a more serious point to make are far more common in the field of politics.
In most cases, we pass on memes, simply because we think the recipient will enjoy them. After all, who doesn't like a picture of a stingray photobombing some holidaymakers? But clearly, there's more to it than that. Recipients may enjoy a well formed "damnyouautocorrect" but anyone posting a status message on the US healthcare system is probably not just in it for the lolz.
Again, it's easy to see memes as a way of identifying which tribe you belong to—anyone posting such a message is more than likely choosing to display their political tendencies as well as just passing on the meme.
Frozen in carbonite for writing about memes
A study of a meme that appeared on Facebook, put together by the company's data scientists, shows the phenomenon in action. Facebook tracked the progress of the following status message, posted and reposted across the social networking site from September 2009: "No one should die because they cannot afford health care, and no one should go broke because they get sick. If you agree, post this as your status for the rest of the day". Later, the political content became more overt: Facebook users began to repost a status message saying "no one should go broke because the government taxes and spends" or "no one should die because Obamacare rations their healthcare".
Some Facebook users who saw the message in their friends' status bar went on to repost it (over 470,000 people did so, according to Facebook), and it developed variations as it travelled: names were inserted at the start of the message ("no one should die because' became 'Dave believes no one should die because"), the time changed a little ("the rest of the day' became '24 hours"), and so on.
Some of these changes were deliberate, others occurred because people had made slips when copying and pasting the message from a friend's status bar to their own—unexpected mutations.
Eventually, other types of variation occurred, either the more overtly political types, or in the form of parodies: "No one should be frozen in carbonite because they cannot afford to pay Jabba the Hut...", "no one should go thirsty because they can't afford wine..." etc. Those parodies, too, were passed on, with some making it further along than others.
The selfish meme
The word "meme" was coined in a 1976 work, The Selfish Gene, by the scientist Richard Dawkins as a unit of cultural transmission. It is actually derived from "gene". Like a gene, a meme exists purely to get itself copied, and must battle evolutionary pressures to do so.
The similarities between memes and genes are clearly shown in the Facebook example. The lifecycle of genes is a process of selection, variation, and transmission. It's a process that happens within living organisms—a gene gets passed from one organism to another, occasionally changes or mutates, and is passed on again by the organism's descendants.
These three processes characterise memes as well: first, memes are selected - we pick the ones we like; perhaps we tweak them (the variation phase); then we pass them onto people we know, transmitting them by email or social networking.
The "no-one should die because..." meme demonstrates the selection, variation, and replication hallmarks in action. Facebook users saw the meme, made their own changes, and reposted it.
While memes' similarity to genes is encoded into its name, Facebook's data scientists—who were able to draw from a vast anonymised pool of data—found new areas where the two overlapped. Like genes, textual memes fared better when smaller, as did genes. They also found it was the areas at the start and end of both that were most likely to see a mutation—a result of sloppy copying and pasting, the data scientists wrote in their study (PDF).
Like genes, memes must battle to survive and be passed on. They must compete with all the other ideas fighting for our attention online and off—those we pass on survive for a little longer, those that don't fall into obscurity. And, of course, there are memes that confer no benefit to us, and yet we pass them on nonetheless.
But the relationship between memes and genes isn't a perfect fit, something that those who study viral internet phenomena and those that study the science of memes (known as memetics) are quick to point out.
Memeticists say the common understanding of memes as purely a cat-infested internet phenomenon is too narrow. Web meme watchers say that under the scientists' definition we are purely hosts to the memes, we pass them on unthinkingly—a scenario that they find out of whack with the rivers of social content that flow past us everyday.
From the outside, however, it's easy to see how the two ideas of memes became interchangeable in the public's lexicon: internet memes display the three characteristics of overall memehood-in-general writ large. They are selected by the internet audience, copied onto Twitter, email, Tumblr, and Kik, and passed on, perhaps first having been rewritten or adapted in some other way.
However uneasily they may sit together, the two disciplines of study—memetics and the study of viral internet phenomena—can also inform each other.
According to Daniel C Dennett, philosopher and co-director of the Center for Cognitive Studies at Tufts University, the internet has certainly sped up many of the transmission and adoption processes for memes, and this in turn has probably led to some novel patterns.
"More important for those of us who want to study memes scientifically, the internet is our Drosophila, our C. elegans, our laboratory rat—a 'model organism' that comes already wired up to record behavioral data. Computer search and data mining are the ideal hi-tech investigative tools of memetics. Of course you can't use it to study the dispersal of hunting techniques in Amazonian tribes, or pronunciation shifts among nursery school children, but then, inconvenience looms large in other areas of science as well."
Google's stats show how our interest in the word meme grew gradually from 2004 onwards, jumping up massively in 2011. Our interest in memes, and our tendency to see them as purely viral funnies, seems to be growing as the lifecycle of internet memes appears to be speeding up.
Take a finger-in-the-air comparison of two well-known internet memes, almost 10 years apart: "All your base are belong to us", a badly-translated intro sequence from a Japanese game that surfaced in the early 2000s, and "overly attached girlfriend", a 2012 YouTube video of a woman singing stalkerish lyrics to the tune of Justin Bieber's Girlfriend.
Like many internet memes, they were created and discovered by web users who spread them online. Their popularity snowballed, they were adapted, and passed on again, before interest in them dwindled over time.
Compared to their meme ancestors, for modern memes the time that cycle takes is speeding up. For all-your-base, the cycle took years; for OAG, months.
There are good reasons for this, of course: internet speeds and availability have mushroomed over the last decade, the internet population has risen threefold, and new services ideally suited to propagating memes—hey there, Twitter—have arrived.
Do memes ever die? Centuries ago, before the printing press, they certainly would have. Words, fashions, and practices would have fallen out of use and been forgotten. Since then, printed material have helped extend the lifespan of our information, but before the internet and online storage, they too would go out of print, be forgotten, lost, or destroyed.
For every folk song that travelled the centuries to become a modern nursery rhyme, there are those that fell out of favour and disappeared from our collective memory. Historically, it was the same for online memes. For every ceiling cat, there's unloved image macros that just didn't become a thing. Now, as the cloud companies and analysts are so fond of telling us, Moore's Law means that storage is getting cheaper and cheaper, leading us to save more and more, leading in turn to storage getting cheaper and cheaper, and so on.
Whereas in the past, there would have been a natural attrition of memes because of the relatively high cost and low availability of online storage, now they are seemingly with us forever. Indeed, there are several sites dedicated specifically to cataloguing popular viral internet memes. Memes are far less likely therefore to be forgotten, and so more likely to be rediscovered in the future, perhaps rejoining the cultural stream, in much the same way Victoriana has been given a new lease of life by the steampunk movement.
With the number of memes growing alongside our tendency to store them, and the mechanisms to pass them on being more widespread and instantaneous, memes have never had it so good. There's no doubt that the invention of computing in general and the web in particular has changed the nature of memes.
While memes were once a wholly human pursuit—while animals feature in many memes, they don't, as far as we know, create or adapt them—we can send high-fidelity copies of them around the world in seconds, and thanks to 3D printing, memes don't have to be digital alone.
While neither memes nor genes clearly aren't sentient they both have agency of a sort—they impel us to build huge infrastructures to create, store, adapt, and transmit them. Much of what we do is designed to ensure their survival, in the same way as procreation ensures genes carry on, whether we're aware of it happening or not.
Halp! The future of memes—without humans
In memetics, genes are known as the first replicator. They were the first thing to ensure their own propagation, using plants, animals, and humans as vehicles to help carry on. When the genes gave rise to humans that could copy and imitate, the second replicator was let loose—memes.
When we first learned how to copy something—a picture, a skill, a sound—we brought memes into existence. How to make fire, for example, would have passed from early human to early human, with each civilisation tweaking the practice to suit its environment and reaping the benefits, ensuring the fire-making meme would be passed on.
While genes and their evolution extends to all life on earth, memes have been a uniquely human phenomenon. Until now. As genes led to humans which in turn led to memes, memeticists believe that a third replicator could now have sprung forth, a product of the human mind but existing outside of it.
The third replicator is known as temes (a contraction of "technological memes"). That is, memes are created, adapted, and copied using digital means. In the same way genes let loose biological evolution and memes let loose cultural evolution, temes could cause a third wave of evolution that could, under the worst case scenario, do away with humanity entirely.
Rather than having people select, copy, adapt, and pass on memes, machines or networks could do the same job without us.
"We constructed all that machinery, the servers, the PCs, for our benefit, we thought. In fact, you can look at it from the other side: it was the memes taking the opportunity when it arose to get copied varied and selected more easily," psychologist and memeticist Sue Blackmore said.
"The assumption is that... the reason they're building more servers is that more people are sending emails, they're sending photos, they're using social media, they're sending videos, all of which take up more capacity. As they add capacity, they're doing it for the humans. The obvious implication of memetics is that it's not at all for the sake of the humans—the humans are the meme machines, they're the copying machinery for all this stuff, and it's actually happening for the benefit for the memes," she added.
For some reason, for me this brings to mind IBM's cognitive computing effort Watson, and the system's recent attempts to create its own recipes—perhaps the first entirely machine-made dish in history. Watson's first effort was a Swiss-Thai fusion quiche, so perhaps the teme-led apocalypse is some distance off yet. The system may be smart enough to come up with a recipe, but it's just not smart enough to know it's an unappetising one.
Yet, in passing this onto you, I'm ensuring the success of the Swiss-Thai fusion quiche meme. If you share this article, so will you be. Should Watson have done the transmission part itself—shared the recipe via update to a Twitter feed for example, and had that retweeted by Twitter bots all over the place—the human-free meme chain would already be in operation.
Of course, that would involve the simplest of memes—there's no doubt computers today could create and share image macros easily without any human intervention, and perhaps already do.
However, memes can vary in complexity and scale, right from ceiling cat up to an entire language or a design for a power plant or a universal scientific theory. Assuming that our technology infrastructure becomes so powerful it could devise even these higher-level memes and share them with other systems and networks without human intervention, the age of the third replicator would have arrived.
"In principle, yes, memes could exist without humans, if only because in principle, robots could gradually replace us humans as the top 'predator'/consumer species on the planet. It would require a long transformational period, and casting off human hosts would require major social and economic changes.
"We already have computer systems that can analyze and even evaluate music to some degree. I can imagine a day in the future when a competition between new songs became an evolutionary knockout tournament, from which a winning song eventually emerged the winner without ever having been heard by any human being!" Dennett said.
But if the age of the teme had arrived, would we even be aware of it? Assuming we weren't involved at any level of the meme lifecycle, would we be able to see the signs that we'd been displaced? Blackmore's theory is that we'd only notice some kind of computational black hole—we'd throw more and more processing power, bandwidth, and storage at our datacentres, and watch it swallowed up without really understanding where it was all going. Think of it as the computing equivalent of dark matter, says Blackmore.
If the temes follow the same pattern as the memes, it doesn't bode well for us flesh-composed types. The theory goes that we're using up an increasing amount of the earth's resources and causing damage to the environment because we're being impelled to do so to a large extent by the memes.
If the temes can persuade machines to do the same, the effect could be even more noticeable —more datacentres means more power which means more fossil fuel use. And you know where that ends up—a lifeless planet, with all its resources used to keep a highly intelligent teme-transmitting infrastructure running.
While that would be devastating for humanity, from the perspective of the top lifeform that succeeds it, it would be a necessary step on its evolutionary journey.
"Maybe that's what evolution is about—[the infrastructure] will be super hyper intelligent compared to us, it will be amazing and able to contact other planets, span time, perhaps understand the entire universe of which it's a part, who knows? You could say from the point of view of a super-intelligent being in the year 3000 that our fantastic role as a very short-lived species that survived only a couple of million years was to let loose the best thing that could come out of evolution," Blackmore said.
Do computers laugh at electric sheep? If so, we could be in trouble. Pass it on.