You’re in a nondescript hallway at the beginning of The Witness. There’s zero exposition, and the only explicit tutorializing comes in the form of some on-screen prompts, explaining which button does what. The real tutorial, though, is a puzzle so simple that it doesn’t even qualify as a puzzle: You have to draw a line from the specified start point to the specified end point. That’s it.
This straightforward, abstract line-drawing stuff is pretty much your sole means of interacting with The Witness, and it’s infinitely easier to explain with basic examples than with words. So the game categorically refuses to explain its rules using anything other than the rules themselves.
What if there were two symmetrical lines, drawn in parallel? What if the line had to pass through or avoid certain points en route to its destination? What if you were the line? You learn a new type of puzzle by solving the simplest possible version of it, and then the complexity escalates.
The Witness spelunks fanatically for every possible variation on its core conceit, leaving no implication or possibility untapped. In the same way that Divekick aspires to be a fighting game that’s 100% clutch moments by volume, The Witness aspires to be Ohhhh, I Get It: The Game. The core, catch-all mechanic allows for inventive wrinkles, rewards equally inventive solutions, and makes you feel like a transcendent being of pure white-hot genius-light when you come up with those solutions.
It’s also a mechanic that, again because it’s so inhospitable to verbal conveyance, resists spoilers fairly fiercely. To truly spoil yourself on a puzzle in The Witness generally means looking at an image and tracing it in your own game, thereby removing any pretense that you’re still, like, playing the game in any meaningful sense. If you’re not just copying a pattern over by rote, then what you’re getting isn’t a spoiler so much as a hint, a fragment of a Rosetta Stone.
Case in point: The following image contains a spoiler. But to understand why it’s a spoiler is part and parcel to understanding the puzzle itself, as well as the series of puzzles that follows it.
If you’re not tuned in to what’s going on in this image, then it doesn’t mean much. Conversely, if you do see the relationship between the puzzle at hand and the rest of the composition, then you’ve just learned a rule with implications beyond the screen at hand. Either way, even if it’s making one tiny part of the game less mysterious, it’s not precisely spoiling anything. Neat, right?
But let’s take a step back.
Jonathan Blow’s previous game, Braid, set the template for what we were briefly comfortable calling “indie games.” It existed (mostly) outside the traditional studio system of videogame-making, it sold cheap, and it aimed at arthouse profundity, riffing all the while on the Marios of old. Reading the game required literacy in the history of videogames, and a willingness to read systems, aesthetics, and narrative—and most importantly, a willingness to think about how all of those elements fit together, or didn’t, as the case may be.
When the elements of a game didn’t quite gel, we called the result “ludonarrative dissonance,” back then. (We were young and foolish). As I’ve said before, the phenomenon was an important one to observe, even if the phrase itself is ungainly beyond redemption.
It’s possible that, in Robert Yang’s words, “Ludonarrative dissonance doesn’t exist because it isn’t dissonant and no one cares anyway.” He goes on to say:
Game logic is frequently illogical. When is that okay and not okay? When should we do better?
I’d reckon that AAA designers, either wisely or obliviously, rarely have these existential crises. It’s not just the players, but also the developers, who have to train themselves to ignore the gameisms inherent in their games.
Eight years after Braid, it’s clear that the rarer thing is ludonarrative assonance. For the love of God, let’s not call it that, but maybe we should think of it as a rare virtue when a game’s multitudinous moving parts cohere, rather than thinking of dissonance as a risible anomaly.
The Witness, Blow’s first release since Braid, presents a lush simulated environment, and a single minimalistic gameism—and the relationship of that game stuff to the island you’re exploring is frequently and inventively assonant. (I mean it, though: Let’s probably not call it that).
Like I said, you start out in a hallway, and you emerge onto an island of castle grounds and alluring vistas and big-ass trees. It’s an intricate bounded space of consummate, stylized beauty, and of thoroughly explored variations on themes.
In other words, the island is doing a version of what the puzzles are doing—that is, inviting contemplation, exploration, consideration and lingering and rethinking. It’s a space meant to look and sound the way solving the puzzles feels.
The Witness argues that there are forms of wonder that only systems and rules and interactivity can provide. That’s not to say that the game diminishes or dismisses the aesthetic impact or technological wizardry at play in videogame audiovisuals—on the contrary, it revels in these things, celebrates them—but the environments are the left jab, and the rules are the right hook. Or the former are the harmonies, the latter the melodies.
The island is a delivery system for discovery. The puzzles are a delivery system for epiphany.
And The Witness is aiming at a confluence of discovery and epiphany, repeated over hundreds of puzzles and sustained over dozens of hours of play. The result is placid unto zen, yet ambitious unto hubris.
I had no idea that this was the follow-up to Braid that I wanted, but here it is: a philosophical first-person puzzle game that’s every bit as singular in 2016 as Braid was in 2007. Let’s see if it can set the template for independent games all over again.