Scary Stories on the Web: How does Digital Folklore Trick its Readers?

***Harper Arnold argues that the Creepypasta phenomenon shows how internet narratives blur the line between fiction and reality. It is reproduced here with the author’s permission as part of our CityLIS Writes initiative. Harper is on Twitter @HarperArnold2***


““Darkness after all, is the unseen and therefore the unknown…And what could inspire fear more than the terror of uncertainty?” – Dr. Hill, Until Dawn” – The quote displayed on the Creepypasta Wiki (April 17, 2020)

There’s a multitude of information available on the Internet, and some of it is true, some of it is not, and some of it is partly true and partly not. In this paper, we’ll examine the way authorless narratives can manipulate their readers into believing the unbelievable. By examining how readers trust these narratives, we can also examine how and why certain information is believed or beloved in the digital era. We’ll be examining a very popular type of Internet narrative: Creepypastas, which are original short horror stories posted to forums on the Internet. The term “Creepypasta” comes from the term copypasta, which is a portmanteau of copy and paste (Balanzategui, 2019), and reflects the viral and sharing potential of these horror stories. Although fictional, these stories often purposely blend elements of reality and the uncanny to create a twilight narrative. While often rooted in the real world, or having various aspects that make it appear so, the main plot elements of these stories are almost always completely fictional. Oftentimes the stories are told in a way that preys on the anonymity of the Internet. This is why scholars have dubbed Creepypastas “digital folklore”: “Creepypasta stories deliberately embed such informal circuits of production, consumption, and dissemination into their generic form as a marker of folkloric authenticity. Initially, Creepypasta stories tended to constellate around particular images and consist of anecdotes or rituals with roots in pre-digital urban legends, echoing chain emails and early memes…They now tend to be more complex, often collaboratively authored narratives that resonate with some of the most significant cultural anxieties of the contemporary digital moment (Balanzategui, 2019)”. Creepypastas are also reminiscent of urban legends for the same reason they’re like folklore: the stories are presented in such a way to purposely make the reader unsure about what is true. This type of internet narrative can often subvert every traditional way of valuing information, simply because it’s anonymous roots and various iterations leave it up to the reader to decide what is real and what is not. By examining and deconstructing a particular Creepypasta story, we’ll begin to see how digital folklore purposely tricks its readers. Then we’ll examine the real-life effects of Creepypasta stories, as well as more modern iterations of digital folklore, and finish by discussing how and why these stories are so pervasive and believable.

The story we’ll be examining is titled “Lavender Town Syndrome”, a story whose origins are so murky that there are competing narratives about the first website it was even published on: it might be Pastebin (Hernandez, 2016) or 99chan (Hathaway, 2015). Either way, it’s agreed that the story was published sometime around summer of 2010 and that the author is titled under “anonymous” or “guest”. The basics of the story are constructed around a fictional narrative surrounding the game Pokémon “Red” and “Green”, released in in the mid-90’s. The story, in its most “original” form, is about children who, after playing the game, commit atrocities or kill themselves. This violence is linked to the song used in a section of the game titled “Lavender Town”, which uses tonal beats that only children can hear, and which instructs them to act horrifically. This is how the story gained the title “Lavender Town Syndrome”, even though that’s not the original title of the story (the first being Decoding the mystery of “Lavender Town Syndrome” and the “Lavender Tone” (2010)). The writing of the story is convincingly real, with the narrator acting as a sort of detective or journalist and using reports and interviews to piece together the mystery surrounding Lavender Town. There is a character called Ms. Harue, who is the interviewee, and who has an update dedicated to her at the end of the story stating that she died from Leukemia on “May 3rd, 2001” (Decoding the mystery of “Lavender Town Syndrome” and the “Lavender Tone”, 2010). All of this, of course, is fictional. However, the story itself is not taken to be fictional even within online communities: it’s been called a “conspiracy” in a Black Bag article (Hathaway, 2015), an “urban legend” in a Life Wire article (Oxford, 2019), a “myth” in a Kotaku article (Hernandez, 2016), and a “theory” in a Youtube video published by TheGamerFromMars (and has, as of writing, over 2.3 million views) (2011). Yet, the story, although with murky origins and authorless, is proven to be fictional in basically every account one can find. So why does it persist as something more concrete than just a horror story posted on a forum board?

Although fictional, almost every major plot point of Decoding the mystery of “Lavender Town Syndrome” and the “Lavender Tone” is based on a real life instance. There was indeed a “Lavender Town” level in a Pokémon game, but it wasn’t the “Red” and “Green” version, but rather, the “Red” and “Blue” versions (Strategy Wiki, 2020). Most obviously borrowed from real life, the “Lavender Tone” is music played while in the Lavender Town level, and is considered quite unsettling. The music was written about online even before the story came out, back in 2005, as a prime example of “depressing game music (Hathaway, 2015)”. The music reflects the storyline of Lavender Town, which is a place where Pokémon go to be buried (Hernandez, 2016). This is quite a jarring storyline, and is the first mention in of Pokémon actually being able to die. The level then centers on the player climbing a Tower, in which a crazed spirit is haunting, and also where a bunch of Trainers are mourning their recently deceased Pokémon. And although the “Lavender Tone” has been proven to not actually give kids headaches or drive them to suicide, the weird music coupled with the dark story elements has given a basis to the half-believable story that surrounds it. This use of childhood nostalgia and subsequent trauma is not a new element in Creepypastas: as Jessica Balanzategui writes about another Creeypasta ‘Candle Cove’, which centers on recollection of a children’s television program in the 70’s, “The gradual revelation of the sinister underside to childhood experience exposes the uncanniness inherent in nostalgia, by suggesting that lurking within fondly recalled childhood pasts are technological ghosts in the nursery that only fully reveal themselves to conscious thought in adulthood (2019).” This use of a childhood memory, i.e. playing or remembering Pokémon, is often fodder for these internet-based stories and conspiracies simply because memory is so faulty, and some childhood memories are only understood once adulthood is reached. Who’s to say that the ‘Candle Cove’ television show didn’t exist? Or that one level, ‘Lavender Town’, didn’t cause prolonged traumatic effects in the children who heard its music? This is another reason these stories are digital folklore: they take a commonly held experience and question its validity. These stories also whisper questions to the reader, asking whether they really remember the past like it was. As Balanzategui also points out, the use of analogue technology better enhances the mystique of the Creepypasta, since who still has an original Game Boy laying around with the Pokémon game mentioned in the story to test it out?

In an example of reality is stranger than fiction, another element of the story taken from real-life instances are the harm the game caused kids. However, it was the Pokémon television show, and not the game, that caused health issues:

“On the evening of December 16, 1997, millions of people all over Japan gathered in their homes to watch Pokemon (episode #38). About 20 minutes into the program there was a scene of a rocket explosion that flashed red and blue lights at a rate of about 12 times per second. This explosion scene was mixed with about five seconds of flashing lights from the eyes of “Pikachu,” a popular Pokemon character. Suddenly, viewers started to complain of blurred vision, headaches, dizziness and nausea. Some people even had seizures, convulsions and lost consciousness. A total of 685 children…were taken to hospitals by ambulances. Although many children recovered during the ambulance trip, more than 150 of them were admitted to hospitals (Chudler, 2000).”

The flashing red and blue lights, coupled with the fast speed, triggered kids with photosensitive epilepsy. Interestingly enough, 12,000 kids actually reported the same symptoms, but the majority of those cases have been chocked down to mass hysteria (Rogers, 2017). With just a stretch of imagination it’s easy to see how seizures and headaches caused by an innocent TV show could turn into something far more sinister in its accompanying video games.

There are other misleading elements to the story to make it seem more realistic: the investigative journalist tone, the interviews conducted in Japanese that make it difficult to fact check for those who can’t read Kanji (Hernandez, 2016), and the time gap of the incident happening over 20 years ago leading to many elements being unverifiable. But even though the story has certain believable narrative devices, at the heart of it all, it is just a story. So why do online communities persist in calling labeling it as something more than it really is? As we’ve seen, the story is written is such a way to make it half-believable, including the well known Pokémon franchise games and coupling real-life elements caused by the television show, but there are other reasons horror stories and Creepypastas are so hard to disbelieve.

Creepypastas have a history of subverting traditional ways of valuing information, especially since their anonymity makes for easier collaboration, and by inserting half-truths (as seen in the Lavender Town Syndrome) as a way to play with believability. As Balanzategui has stated, Creepypastas had their origin in simple images usually echoing chain emails or early memes, an example being ‘Smile Dog’ [2008] which was a ‘cursed image’ that readers were encouraged to circulate to avoid the curse (2019). Although chain emails and texts may have little bearing on a person’s day-to-day life, these are prime examples of how stories begin to take root in a non-traditional narrative form, i.e., an often work-related communication chain such as an email, or personal communication chains such as a text message. These non-traditional ways of communicating information can be misleading at best and tragic at worst: in 2014, two 12-year-old girls, Anissa and Morgan, lured their friend, Payton, into the woods near their Wisconsin homes and stabbed her 19 times in the hopes of appeasing Slender Man, another Creepypasta character. Payton survived, and although the other two girls have since been committed to staying in a mental health facilities, and they clearly weren’t in their right mind, they both strongly believed that Slender Man wanted them to kill their friend (Janes, 2019). Morgan and Anissa later reported that, after the stabbing, they ran off into the woods to try to meet Slender Man and find his ‘mansion’. It’s easy to dismiss this case as an extension of deteriorating mental health in these two tweens (Morgan was later diagnosed with schizophrenia (Janes, 2019)), but it points to an interesting, and perhaps problematic, way of story dissemination. As discussed before, stories like Lavender Town are often purposely acting like the narratives are real, but the format of Creepypastas are often framed like they’re literally real. Many read like a Reddit post or like a Yahoo Answers question, with some tagging ‘based on a true story’ (which is always, conveniently, unverifiable). Even looking in the comments section of the Creepypasta Wiki there are always users questioning how true certain Creepypasta stories are (Creepypasta Wiki, 2020), and many of these comments are certainly written by children. There is rarely any disclaimer on the Creepypasta Wiki that states everything is fictional, and this became a point of contention in the moral dilemma following the Slender Man stabbing attack: “…in response to the extensive public outcry, the Creepypasta Wiki (2017)…released a statement on its homepage declaring in capitals that “ALL WORKS PRESENTED ON THIS WIKI…ARE FICTIONAL STORIES AND CHARACTERS (Plunkett, 2014) (as cited by Balanzategui, 2019).” The statement went further on to read, “Of course, only a small minority of people (mostly newcomers) on the wiki (and the Internet) truly believe what they read here. And for most people, they will not attempt replicate atrocities presented in some of the literature on the wiki (as cited by Plunkett, 2014)”. The Creepypasta Wiki is right in the sense that they are a literature site, but they are not correct in thinking that digital folklore (especially without warnings) isn’t convincing enough for children and the uninitiated to believe. Also, this statement is no longer on the homepage, and the current statement can be found at the start of this paper.

Moreover, with a lack of authorship comes lack of responsibility. Even the Slender Man story has an “author” of sorts: Eric Knudsen (alias “Victor Surge”), who posted a photo of Slender Man on a “Create Paranormal Images” thread on website Something Awful (Mar, 2019). However, the difference between authorship in analogue documents and the infosphere becomes self-evident, as Knudsen, who does technically own the rights to Slender Man, wasn’t actually much involved with spreading the tale. It was other artists, writers, screenwriters, video game developers, and fans, which really spread the legend. Creepypastas almost always tend to be collaborative, with some stories being controversial in origin (i.e., multiple people have claimed be the originators of the monster ‘Jeff the Killer’ (Creepypasta Wiki, 2020). This makes it very difficult for any one story to be attributed to an author, and often lessens the ‘authenticity’ of the story if it can be traced to a singular source. For this reason, lack of authorship and collaborative endeavors are often the norm of digital folklore, and this is why they can be believed easier than a novel or article.

Creepypastas are also evolving into narratives that aren’t even intended to be Creepypastas. For example, one of the most famous, and maybe the first, Twitter thread that turned into a horror story was the ‘Dear David’ thread posted by writer Adam Ellis. The thread consisted of Ellis tweeting about a ghost boy named David who was haunting him. The interesting thing about this thread is that Ellis still maintains it was completely real, even with the nearly unbelievable “photos” he caught of David. Ellis archived this entire thread on Wakelet to make for easy reading (Ellis, 2018), and, after selling the movie rights, has since stopped giving regular updates to the Dear David story (Fishbein, 2018). Although Ellis maintains the veracity of his story, the narrative has switched and commentators are now taking the whole thread to just be a fictional story (Fishbein, 2018). This is one story that went from the “potentially real” realm to a place where others thought it was at least a hoax, if not an outright lie to gain followers (Ellis gained around one million followers after his Dear David story gained traction (Fishbein, 2018)). If it was just a story, Ellis was exploiting the vulnerability of believable Twitter users who took his story at face value, and in the process demonstrating how easy and beneficial it can be to make a story seem real, even if it has every element that should detract believers.
So how are Creepypasta stories valued by their readers? The real conclusion is that it depends on who’s reading. Especially as stories become more ambiguous and anonymous on the Internet, and sources all across the board become less trusted, these stories are giving the reader more power to believe or disbelieve them. Creepypasta authors, and other forum board members, don’t usually take any responsibility for the effects of their stories, and the Creepypasta Wiki seems to rarely make a concerted effort to warn young readers from believing in these fictional stories. By looking at Internet horror stories and digital folklore, we can begin to see how much people put their trust into online discourses written by strangers. Trusting internet information blindly is becoming the norm, with Kalev Leetaru writing for Forbes, “Instead of browsing a small carefully curated set of high quality informational streams, our online citizenry are thrown into an ocean of almost limitless low-quality information, forced to expend considerable effort to forage for the rare bit of insight (2019).” Although fictional narratives published on the Internet wouldn’t traditionally be discussed in a digital literacy conversation, as we’ve seen before with the Adam Ellis Twitter thread, there are times when even the most unbelievable information is asked to be believed. Especially as a collaborative endeavor, and an oft-anonymous endeavor, these types of digital folklore create a world of information with no attempt to clarify the truth. These types of narratives obscure their sources and origins, and displace the decision-making from the originator to the reader. Who’s to say what’s real and what’s not? Well, in the digital age, it’s all left up to who’s reading.


Balanzategui, J. (2019) ‘Creepypasta, ‘Candle Cove’, and the digital gothic’, Journal of Visual Culture, 18(2), pp. 187-208.

Chudler, E. H. (2000) Pokemon on the Brain. Available at: (Accessed 10 March 2020).

Creepypasta Wiki (2020) Creepypasta Wiki Home. Available at: (Accessed 17 April 2020).

Decoding the mystery of “Lavender Town Syndrome” and the “Lavender Tone” (2010) Available at: (Accessed 28 March 2020).

Ellis, D. (2018) Dear David. Available at: (Accessed 19 April 2020).

Fishbein, R. (2018) How Did “Dear David” End? The Movie About Adam Ellis’s Infamous Ghost Boy Has A Lot Of Questions To Answer. Available at: (Accessed 19 April 2020).

Hathaway, J. (2015) Lavender Town: The Bizarre Pokémon Child-Suicide Conspiracy Theory. Available at: (Accessed 6 April 2020).

Hernandez, P. (2016) Pokémon’s Creepy Lavender Town Myth, Explained. Available at: (Accessed 20 March 2020).

Janes, D. (2019) The Complete Timeline and True Story Behind the Slender Man Stabbing. Available at: (Accessed 19 April 2020).

Leetaru, K. (2019) Why Do We Believe What We Read On The Internet? Available at: (Accessed 27 April 2020).

Mar, A. (2017) Into the woods: how online urban legend Slender Man inspired children to kill. Available at: (Accessed: 19 April 2020).

Oxford, N. (2019) What’s Pokémon’s Lavender Town Syndrome? Available at: (Accessed 6 April 2020).

Plunkett, L. (2014) Creepypasta Wiki issues statement saying Slender Man isn’t real. Available at: (Accessed 4 May 2020).

Pokémon Red and Blue/Lavender Town (2020) Available at:émon_Red_and_Blue/Lavender_Town (Accessed 21 April 2020).

Rogers, K. (2017) Did ‘Pokemon’ Actually Give Kids Seizures In the 90s?. Available at: (Accessed 12 April 2020).

TheGamerFromMars (2011) The Lavender Town Theory. 12 October. Available at: (Accessed 6 April 2020).

About Joseph Dunne-Howrie

I am artist in residence in the MA/MSc Library and Information Science department at City, University of London and module year coordinator for MA/MFA Performative Writing/Vade Mecum at Rose Bruford College of Theatre and Performance.My research interests include intermediality, live performance in digital culture, participatory and immersive theatre, performance documentation, archives, and performative writing.
This entry was posted in CityLIS Writes and tagged , , , , , , . Bookmark the permalink.