Intermittent Mechanism

Ian here—

Fourth episode of the Shape Up! series is here, this time on a game I’ve taught several times in classes and long wanted to do a video about: P.T.

Due to the busyness of this academic quarter, I won’t be able to keep up with the one-video-per-month schedule I’ve set for myself since October. Hoping to get at least one more episode of the Shape Up! series out between now and summer, and then return to the one-video-per-month schedule during the summer.

The script that follows below the jump is the shooting script for the video. (It differs from the actual transcript in one key respect.)

Hello everyone, and welcome to another episode of Shape Up!—an ongoing series about form, structure, and pacing in games. This episode will revolve around one case study, Kojima Productions’ P.T., following a rather twisty thread.

If you weren’t following horror games back in 2014, here’s a basic rundown of what P.T. is, or was, and why people remember it. The Silent Hill franchise was in a pretty dire state in the early 2010s: the original creative team that had made the first four games had disbanded, and the franchise had been passed off to a revolving door of developers making hit-or-miss entries. To turn things around, the franchise’s owner Konami gave the next entry in the series to Hideo Kojima, best known for Konami’s Metal Gear franchise, partnering with the film director Guillermo del Toro. The planned game, Silent Hills, was announced via a mysterious playable teaser for the PS4, referred to only as “P.T.” on its PlayStation Store page.

Players who downloaded and booted up this playable teaser were greeted with a hallway. The hallway has one prominent 90º turn in it, one door opening onto a bathroom that is sometimes open but usually not, and a door leading to small stairway down, with an exit at its end. When you go through the exit, you arrive at the start of the hallway again. The geography loops from here. On some cycles through, you simply walk from one end of the hallway, around the bend, and down through the exit. On other cycles, the door at the end is closed, and players have to find some way of interacting with the environment to get it unlocked. The demo’s controls are very basic—you’re mostly limited to just zooming-in the view—so these interactions are never elaborate, but they are obscure. Along the way, players catch glimpses of a woman’s ghost, Lisa, who can kill you if you’re not careful, re-spawning you at the hallway’s entrance but otherwise not disrupting the demo’s overall progress.

After thirteen distinct loops of going from one end to the hallway to the other, sometimes needing to unlock the door with an action and sometimes not, the demo gets much more confusing. The hallway will now loop infinitely, and the only way to break the cycle is to collect six photo scraps hidden around the environment, and then enact a sequence of steps that remains enigmatic and contested to this day, a decade later. If this is done correctly, the player will receive a phone call—in the game, not, like, on your cell phone; it’s not that complicated—and when they walk through the exit a final time they will then, and only then, see the announcement that what they’ve just been playing has been a teaser for the upcoming game Silent Hills.

Konami ultimately cancelled Silent Hills in 2015, and they pulled P.T. from the PlayStation Store. On the one hand, this is a understandable move. It was a demo for a product that was no longer ever going to exist, so why keep it out there? But on the other hand, in the nine months that it had been downloadable, P.T. had unexpectedly become one of the most buzzed-about and influential horror games of the 2010s, just by itself, as a stand-alone object, regardless of any connection to a larger promised project. And Konami didn’t just pull P.T.’s listing from the PlayStation Store, so that no one new could play it: they also made it impossible to re-download if you had previously owned it, meaning that the demo would be forever stuck on a finite number of PS4 hard drives. And then in 2020 they announced that it would be impossible to transfer a copy of P.T. over to a PS5 if you owned it on PS4, further trapping it on aging hardware. The fate of P.T. became a cause célèbre around matters of game preservation and archiving in the era of digital distribution, with Konami’s unyielding position illuminating just how much power I.P.-holders have in the current era to wipe things from existence.

But—at the risk of repeating myself—this is a series about game form and structure, not about game preservation. So while I’m certainly distressed by the fate of P.T., it won’t be my main focus here. There’s plenty of opinions in the matter that have circulated in the past decade. What I want to do in this video is talk about P.T.’s shape.

Specifically, about a structural element that I’m going to be circling back to throughout this series, across numerous examples: the off-ramp. Off-ramps are the mechanism by which games give their players closure, and in effect give them permission to stop playing. Of course, you can stop playing a game at any time, just like with any other medium. There are plenty of books on my shelf that still have some bookmark in from where I stopped reading and never picked them back up. People leave plays in the middle of performances. People decide a TV show has jumped the shark, and stop watching it. You can always lose interest in a story, or decide you no longer have faith in the author’s ability to continue it in a satisfying way. The type of off-ramp I’m talking about here is different, though, because it’s officially prescribed by the work’s creator as a stopping point. Think Marvel movies: they typically have an ending, where the credits roll and casual moviegoers can feel free to leave, then a mid-credits sequence teasing the next MCU film for the fans, and finally a post-credits sequence for the super-fans. It’s a tiered system that gives casual filmgoers permission to leave, while conditioning the most dedicated fans to stick around to see what comes after the credits. Marvel movies have normalized this practice in cinema, but it’s even more widespread in games. Single-player games will have their basic ending that grants enough closure for casual players to feel like they’ve gotten enough. That’s the game’s main off-ramp. But developers know that there will be completionist players who want to keep collecting things, hunting achievements, seeking out undiscovered content and hidden endings. And there will often be a cascade of different off-ramps for those players.

In future videos I’ll talk about some games that use off-ramps in innovative ways. But the main innovation of P.T. is that it refuses to have tiered off-ramps for different players, in the way you might expect.

We can easily imagine a version of P.T. that comes to a close and lets casual curious players see the announcement for Silent Hills after those initial thirteen loops, without making everyone figure out that you need to take exactly ten steps after the clock strikes midnight and listen for a baby laughing. Then there could be additional easter eggs, for those who wanted to spend more time poking around with the demo. In fact, we don’t even have to imagine this—because it’s exactly the route that Capcom took with their Resident Evil 7: Beginning Hour demo two years later. That demo has a standard ending that most players will see. Then there’s a half-dozen hidden endings you can unlock through puzzle-solving, including a super-obscure “true” ending. It feels like suits at Capcom made an executive call: yeah, okay, we’ll include all sorts of abstruse puzzles in our demo, in direct imitation of P.T. But that will be for the hardcore fans. For the normies, the demo will end after an hour with a clear announcement of the product they’re supposed to buy. And I can’t say that didn’t work. Resident Evil 7 sold a lot of copies. But, if Resident Evil 7 had been cancelled, I doubt people would still be playing the Beginning Hour demo, that it would have spawned imitators and fan remakes, and that PS4s with it installed would be passed around like holy artifacts, or sold on eBay at inflated prices. P.T. became a zeitgeist-defining game precisely because it didn’t coddle its players with an easy and satisfying off-ramp. It taunted its players with how hard it was just to see its end credits—and in doing so, it fostered a community.

So in this video I want to talk about the ending of P.T. Not its ending in the sense of it being pulled from its online store, but in terms of the work that needs to be put in to get to the point in the game at which the credits roll, the way in which the game denies players an easy out, and leaves us stuck in a frustrating loop.

Let’s frame this in broad, historical terms.

In Poetics, Aristotle posits that a story that is whole consists of a beginning, a middle, and an end. This is such an easily-graspable idea that it’s taught in elementary school—and it’s often taken as an unexamined truism, so banal it’s not even worth pointing out. But once you start looking, there are a ton of counter-examples throughout history, not only in games, but in a whole host of other media. In the 19th century, it was common to publish novels as serials, made up of regular installments that would appear weekly or monthly, depending on the journal’s publication schedule. Most of Charles Dickens’ best-know novels were published serially. Authors might start writing a serial with a clear ending in mind, but the economics of serial publication actively disincentivizes wrapping up a story, from both the perspective of the writer and the journal. Instead, the sweet spot is a sort of endless middle, with a cliffhanger at the end of every installment, satisfyingly wrapped up and then led into another cliffhanger in the subsequent installment.

In the early 20th century, the serial format spread into other media. French filmmaker Louis Feuillade took the lessons learned about cliffhanger endings in serialized literature and applied them in his film serials Fantômas and Les Vampires. Winsor McCay’s Little Nemo in Slumberland brought some light serialization to the comics. By the mid-20th-century, you had the emergence of the television series, which remains the most popular mode of serial storytelling today.

Serialized storytelling is a mode of writing stories that don’t necessarily have endings—even as if they do have beginnings and endless-middles. But there are also modes of exhibition that disrupt the Aristotelian norm of beginning-middle-end.

In the first half of the 20th century, film exhibition tended to be much more catch-as-catch can than it is today. Movie theaters would show features as part of a program together with newsreels and cartoons or other comic shorts, and there was no expectation that audiences would arrive at the theater exactly at the program’s start time. People were free to wander in after the shorts had started if they wanted to see the feature, and even if you came to the theater halfway through the feature, the ticket-taker would still let you in—your money was just as good as anyone else’s. The programs played on a continuous loop, and it was a cultural norm that if you came in during the middle of a feature, you could stay in the theater after it ended, watch the shorts and newsreel as they came up again, and then watch the first half of the movie. Then, when you started to recognize that things were repeating themselves, you’d turn to your date, and say, “this is where we came in.”

It wasn’t until the second half of the twentieth century that a new norm developed in which theaters publicly advertised specific start times of movies, and refused entry if you arrived at the theater too late. One factor that moved the needle was the publicity stunts Alfred Hitchcock undertook during the initial theatrical run of Psycho. Hitchcock insisted that any theater showing Psycho refused to admit any prospective moviegoer that arrived after the film’s start time, to preserve the surprise of Janet Leigh’s character being killed off so early in the film. Paramount’s publicity materials advised theaters to hire Pinkerton private security to forcibly keep people out of the theater. To alleviate the inconvenience of the then-unfamiliar act of lining up outside a theater and waiting to be let in, Hitchcock recorded messages to be played on loudspeakers to patrons, assuring them that “Psycho is most enjoyable when viewed beginning at the beginning and proceeding to the end. I realize this is a revolutionary concept, but we have discovered that Psycho is unlike most motion pictures, and does not improve when run backwards.”[i]

These days you’re likely to get kicked out of a movie theater if you try to loiter between screenings. But it’s not like the experience of coming into the middle of a movie on a loop disappeared entirely. Today if you wander into a corner of a museum that’s showing a short film or video piece, you’ll have the same experience. Filmmakers who know that their pieces are going to show on a loop in a museum have even leveraged this fact to play with narrative form. The Canadian video artist Rodney Graham’s piece Vexation Island is based around a causal loop of a man getting hit in the head by a coconut, getting knocked out, getting up, and getting hit in the head again, so you can never come in at the “wrong” time—it will always make sense as a narrative, though the longer you stay watching it the more you’ll realize how nonlinear it is.

I know this is an odd litany of examples, but want to illustrate that Aristotle’s insistence that stories have a beginning, middle, and end isn’t always true in practice. Other historical norms of storytelling have existed. Storytellers have told stories in loops before—sometimes inadvertently, due to an accident of film exhibition, but other times purposefully. And not every story necessarily has an ending.

And yet …

There is a difference between passive media, which drop their readers or viewers into a story that moves in a loop, and more active media such as games that have the power to trap their players within a looping scenario, with no end in sight. If I visit a museum and sit in on Vexation Island, I’m experiencing a loop, but I’m not trapped in one. A game can take a loop and transform it from just a quirk of exhibition into a hostile form.

What does this mean—when does a loop become hostile? I would say, when it frustrates a desire for closure. Closure is different in games than in other media, because it’s won. If I sit around in a gallery watching Vexation Island, eventually I can decide for myself that I get it, and I’ve had enough of the experience. Nothing’s keeping me from leaving that corner of the gallery. And, technically speaking, nothing’s preventing me from putting down the controller and walking away from a game, either—except for the cultural understanding that games are problems to be solved. We’re supposed to be able to beat them.

“Beating” a game: think of the violence of that phrase. We say that we “beat” people in competitions, so it makes sense to say that you “beat” your opponents in a competitive multiplayer game. But why do we call finishing a single-player game “beating” it? The language we use indicates a struggle of some sort, a test between adversaries, in a way that isn’t true of other narrative media.

Espen Aarseth has this wonderful phrase where he describes games—text games, he’s talking about specifically, or works of electronic literature, Choose Your Own Adventure books, and certain puzzle-like works of modernist literature—that unlike tradition written works, they require nontrivial effort to traverse, and they “put the would-be reader at risk of rejection.”[ii] We’re put “at risk.” When we play a horror game, our character is at risk of being killed by a monster. But we are also at risk in P.T. It’s not just that we can fail in the traditional horror game sense—getting our character’s neck snapped by a ghost, say—but we can also fail to traverse the path properly, fail to understand what to do. We can be rejected by the game. There’s a risk that we won’t be among the chosen. And in that case, the game will grind to a halt.

Or, rather, not grind to a halt—that would imply stopping—but threaten to go on forever. Threaten to be endless, a hallway repeating until we die of boredom. When the desire for closure is stymied by the threat of infinity: that is a hostile loop.

And that’s P.T.—P.T.’s the game I’m talking about. Went awhile there without saying its name.

Anyway, where was I? P.T.’s resistance to being beaten. It’s really quite remarkable.

In the first episode of this series, I discussed an overall aesthetic trend in early-to-mid-2010s game design, one that emphasized pure discovery, even at the risk of players feeling lost. If you describe them on paper, games like Kairo and Antichamber might sound like straightforward first-person puzzle games. But that’s not what the experience of playing them is like: at least in their opening hours, they’re all about exploring an alien landscape and learning to navigate it purely trough trial-and-error. They’re better described as secret box games than they are traditional puzzle games, because they prize novelty over a coherent ruleset.

And when it released, P.T. was very much a part of this aesthetic trend. There’s a certain historical narrative that’s calcified around it: that it was an important milestone in first-person survival horror, a stepping-stone between the indie hit Amnesia: The Dark Descent in 2010 and Resident Evil 7 in 2017. It spawned a legion of imitators, including Layers of Fear, which was such a success for its developers Bloober Team that Konami hired them to remake Silent Hill 2. These things are all true, but they don’t describe the experience of playing P.T. very well. Playing P.T. is weird and frustrating, in a way that feels, let’s say, distinctly 2014. The game remains resolutely obscure, every step of the way. Its puzzles don’t feel like puzzles, they feel like odd discoveries. The collection of the photo scraps—which could have been totally straightforward—is unfair several times over. It’s terribly hard to see some of the scraps, and hard to pick them up when you don’t have a cursor or a reticle, just a zoom function. The sequence of button presses you need to do to successfully collect the one scrap from the pause screen is so unintuitive that I couldn’t remember how to do it when capturing this, even though I did remember the twist that there was one on the pause screen. This is a game that is truly unafraid to frustrate you, and to be really unfair. It’s a game that seemingly wants to be so inexplicable that it forces you into magical thinking, adopting little rituals as you try to make your way through it.

Take the matter of triggering the credits. The official sequence, generally agreed upon in guides, is this: When the clock chimes midnight, walk exactly ten steps. You will trigger the sound of a baby laughing. Then, as Lisa starts appearing on the map and the soundtrack starts getting noisier, you should speak into your microphone. People disagree on what you should actually say into your microphone—I’ve seen “Hello Lisa” as a possible phrase. Others insist the correct option is “Jay” or “Jarith” or “Jack.” I’ve seen one person insisting that “bebe bebe bebe” works. While recording this, I decided to cover multiple bases, and repeated “Hello Lisa, it’s Jay, short for Jarith” in three-second intervals until I got the second baby laugh. Then, my controller vibrated and I froze and waited until a baby laughed a third time, and the phone rang.

There was already a lot of magical thinking going on when I mashed together the various phrases I had heard worked for people, crafting a sort of ecumenical compromise. But that’s nothing compared to my attempts to trigger the ending without the help of a microphone. This is a process for which there are no good guides, only a swirling haze of rumors in the form of 10-year old Reddit and YouTube and GameFaq comments. Some people insist you can do it by zooming in on the messages on the walls. Others have proposed a complex system of sound cues, where the place you should be positioned in the hallway is determined exactly by the part of the sound loop that’s playing. For every decade-old proposed solution you find, you’re going to find find a decade-old complaint insisting it doesn’t work—and these comment threads are all dead, so these disagreements remain unresolved.

It’s such a common thing these days to have all of the mysteries of a game mapped—especially a game that’s a decade old, so that walkthroughs and wikis have had a chance to be refined and perfected through careful crowd-sourced testing. P.T. is mostly mapped in this way, but triggering the second laugh without a microphone remains mysterious all these years later. The demo being pulled from the PlayStation Store means the total amount of people who has ever played P.T. remains finite. And although I’m disgusted by Konami’s actions from a game preservation standpoint, I have to admit that in some ways the experience of actually playing P.T. is enhanced by how many secrets it still holds, precisely because so few people ever had it installed on their PS4s. It will never get the sort of exhaustive wiki treatment that other games get—and there are benefits to that. It means its inner workings remain arcane, all the better to promote superstitions.

Anyway, when I was recording footage for this video I attempted to get the second baby laugh without plugging in a mic, using the sound/location method described by Krizzy in this YouTube video. But I couldn’t get it to work. And this is especially frustrating because I have previously triggered the ending without the help of a mic. I know for a fact that it’s not a myth that you can do so. It happened during a play session I was holding for a class in October 2022, and it was one of the most joyous gaming experiences I’ve ever had, with everyone in the class glued to the screen, super excited about what was happening. But that only happened after my students and I all passed the controller around for awhile, trying different things—so I couldn’t tell you what the actual action was that triggered it. All I can say is that: it happened. And there was a room full of witnesses. But the game wasn’t cooperating for me during this capture session.

And maybe that’s for the best. I wouldn’t want to post any definitive solutions on the internet, ruining that mystique I was just ruminating about.

There’s a line from the Jorge Luis Borges short story “Death and the Compass” that occurs after a detective has fallen into a villain’s trap. The story’s coming to an end, and he’s about to be killed, but he can’t help but channel his inner YouTuber and voice one last complaint about the game design of the trap. “There are three lines too many in your labyrinth,” the detective says to his nemesis. “I know of a Greek labyrinth that is but one straight line. So many philosophers have been lost upon that line that a mere detective might be pardoned if he became lost as well.”[iii]

I’m thinking of this dialogue because I once asked—twice asked, actually—my students if P.T. is a labyrinth. And they kind of went through the checklist of maze design in games: multicursal design, with forked paths, most of which lead to dead ends or are loopbacks, and concluded that P.T. was, in fact, not a maze. Fair enough. But the conclusion “P.T. is not a maze” doesn’t actually answer the question of whether P.T. is a labyrinth. And, truth be told, I think it is.

Historically, many labyrinths have been unicursal, meaning they don’t have branching paths. Unicursal labyrinths aren’t exactly “one straight line,” like the one in the Borges story. Instead, the route you take is windy. But it inevitably leads to the center if you keep going. As Penelope Reed Doob points out in her book The Idea of the Labyrinth, there’s no actual danger of getting lost when wandering a unicursal labyrinth, but there is a danger of stopping short of your desired goal. Labyrinths like these require “constitutional perseverance” to complete.[iv] They don’t trap you, but they can make you feel trapped, by limiting your view, forcing you to look at a series of curved or angled walls and denying you a clear view of your ultimate destination.[v] And this tension between feeling overwhelming while also being completely extricable is part of the reason unicursal labyrinths often popped up in medieval church architecture. They work well as architectural metaphors for the idea that although one can be entangled within the deceitful world, if you accept God as your guide, and follow His plan, divine order shines through the chaos of the worldly.[vi] Drawing from sources dating from the middle ages, Doob describes Easter Monday festivities that would take place within cathedrals in France, in which the dean stood at the center of a labyrinth and danced, tossing a ball back and forth with other clergy members who were also dancing, either remaining along the edges of the labyrinth or carefully tracing its path as they moved and tossed the ball back and forth.[vii]

And this is what I think P.T. is, or was. Not literally in its specifics, of course. But as a cultural practice, there are a surprising amount of parallels. P.T. is not a multicursal maze that you can get lost in. But it is a unicursal spiral that you can get frustrated with because of the lack of a visible goal, and lose your will to complete, give up on the desire for closure. And much like the dean of the medieval cathedral who danced and tossed a ball and generally guided his followers through the path of the labyrinth, P.T. had its own class of spiritual guides who demonstrated that there was an end to the path that could be successfully reached, if we all kept the faith and persevered. And people danced the dance of these spiritual guides, and in the end, they saw the divine order behind the apparent randomness. They were chosen. And then the demo was pulled from the PlayStation servers, and this site of cultural exchange disappeared. The Easter Monday labyrinth dance that Doob describes is lost to time, and can only be vaguely guessed at by scholars pouring over old incomplete manuscripts discovered in a vault. Likewise, I’m sure that decades from now, when the last remaining drives on the last remaining PS4s have failed, people will very little idea of what to make of the strange ramblings found in old comment threads. Or my strange ramblings in this video.

And with that, I think we’ve reached the center of this labyrinth. As always, thanks for watching, and stay tuned—there will be further discussion of loops coming soon.

[i] Stephen Rebello, Alfred Hitchcock and the Making of Psycho (Berkeley, CA: Soft Skull Press, 2012), 195–196

[ii] Espen J. Aarseth, Cybertext: Perspectives on Ergodic Literature (Baltimore, MD: Johns Hopkins University Press, 1997), 4

[iii] Jorge Luis Borges, “Death and the Compass,” in Collected Fictions. Translated by Andrew Hurley. (New York, NY: Penguin Books, 1999), 156

[iv] Penelope Reed Doob, The Idea of the Labyrinth from Classical Antiquity through the Middle Ages (Ithaca, NY: Cornell University Press, 1990), 50

[v] Ibid., 55

[vi] Ibid., 130

[vii] Ibid., 123–127