2001's ICO was not the best-selling PlayStation 2 title, but the groundbreaking work of interactive art gradually came to define Sony's console as its most critically lauded first-party release. The system had other commercially successful and well-reviewed exclusives, but ICO was the one title that PS2 owners could always point to as a unique experience that could not be found on any past or contemporary machines. So as the PS2 approached the end of its life cycle, fans were eager to see what else Team Ico and director Fumito Ueda could give them. In 2005, the answer arrived in the form of Shadow of the Colossus (known in Japan as Wander and the Colossus), another transcendent game to redefine and essentially close the book on the story of the PS2, much as ICO had begun it.
Bearing more than a few similarities to Ueda's previous game, Shadow of the Colossus again tossed players into the middle of a foreign, seemingly post-human world with little exposition. The protagonist this time was Wander, a young warrior bearing a magical sword. His only living companion in the game's desolate setting would be his horse, Agro. This time around, the girl in the story was already dead before it began, so Wander's quest was not to protect her but to resurrect her. He would attempt this by striking a bargain with a disembodied entity, which would instruct him to slay sixteen colossi spread across the land. That was about all that the player would be told, and then it was off to hunt these behemoths.
Although it contained much more dialogue and also included a traditional heads-up display for gameplay purposes, Shadow of the Colossus was even more minimalist in its narrative than ICO. ICO told the player almost nothing about the characters or the setting, but, while the dreamlike world inspired many questions about the backstory, the action and events within the game were not too hard to follow, and the journey was largely self-contained. Ico and Yorda's partnership was one born out of immediacy. Perhaps they had lives before their meeting in the game, but the pressing concern that they shared, which the player also shared throughout the adventure, was to escape the castle that held them both prisoner.
In Shadow of the Colossus, the player and Wander were clearly not on the same page. For Wander, this was a deeply personal quest, but he would disclose little of what he knew to the player. As a result, the game would not grip the player as tightly, as instead of identifying with Ico, I began initially unable to empathize with Wander's circumstances, and consequently I felt a little lost while having to share in his arbitrarily herculean trials. Nor would there be any developments along the way within the game to really alter the narrative from those beginnings; Wander was basically just repeating the same task over and over, and although there were obviously a finite number of colossi to conquer, there was not much besides the decreasing number itself to indicate a progression. Most likely that was why the developers chose to include some tedious exposition at the beginning of the game. Like the voiceover narration in the theatrical cut of Blade Runner, this text did not feel very native to the whole of the work, which was so deliberately spare in its story, but it may have been necessary because otherwise players would never have had any idea what was happening.
Unlike ICO, which proceeded continuously along a linear path, Shadow of the Colossus was structured almost like an open-world game, although its massive environment offered no quests to undertake, no enemies to slay, other than the sixteen colossi. After every battle, Wander would restart back at the central structure, and his magical sword would emit a beam of light to direct him to his next foe. The sword drew only a straight path, however, offering no guidance as to what turns to take en route, and the crudely drawn in-game map was almost no help. The farther away and more hidden in twisted geography the destination was, the more tiring and aimless a journey could begin to feel. Frankly, this was one of my biggest complaints with the design of the game, as I wasted too many hours just trying to find the next fight. But at its most effective, the design gave the player time to pause and reflect as Agro carried Wander across the expanse.
As in ICO, the ruined yet still magnificent architecture littering the landscape would inspire questions about the world. More than that, while wandering to the next destination, the lengthy breaks between battles would provide time enough to wonder about Wander and his quest. In Gungrave, this was the sort of mandatory interlude that would give way to a revelatory flashback. In lieu of any such cut scenes in Shadow of the Colossus, it was my mind that wandered to consider theories for the backstory. What would motivate a man, armed with almost nothing, to scale these living mountains? This was not about saving the world, or protecting yourself or anyone else, or even pursuing revenge. The colossi were terrifying, but they were not terrorizing the land; it was Wander who would actively seek them out to pick a fight. Why? Of course it was to resurrect the girl, but what was she to him anyway? The game would not tell the player much, so the question eventually turned inward as I asked myself, what would I do it for?
The trial-and-error mechanics of many video games could be reduced to sequences of simple questions, but when I arrived at this one, what astonished me was the realization that I had never asked myself that question before. Other objective-based games would tell me my motivation, and even if it was silly or stupid, I would go along with it. This was clearly a game that needed a motivation but did not supply it, asking instead that the player find one by looking within. So the question became, not what the girl was to Wander, but what she might represent to me. A player who could provide an answer to justify the suicidal mission would quite likely find Wander suddenly a character easier to empathize with and more human than any of the nobler heroes in other games.
Shadow of the Colossus may have been pitched as a game composed only of boss battles, but the battles actually played out as puzzles not altogether unlike those that made up most of ICO. The challenge was never in dodging an enemy's attacks and landing your own, but in figuring out how to get on top of the towering colossus. Once you figured out the tricks, the actual sword action was usually very simple, as you would just stab their weak points until they fell. Seemingly testing your wits and resourcefulness, rather than your manual skill, the best of these were probably the most rewarding boss fights I'd ever experienced.
The problem was that many of the puzzles were highly unintuitive. ICO's puzzles also followed some peculiar logic that drew less from common sense and more from a player's video game experience--blocks are to be pushed, levers pulled, and bombs exploded--but at least both the tools' basic functions and the eventual objective would always be clear, and the fine challenge would be usually in how to properly order or assemble the pieces. At its heart, Shadow of the Colossus relied on even more arcane game logic--if it glows, whack it, or if anything seems otherwise conspicuous, it is probably the key to it all, even if you have no idea why. Thus would I find myself standing on a brown disc in the middle of a dirt field, waiting for something to happen, all the while having no idea what the actual plan was until after the result came to pass. Ueda and his team must have even realized at some point that it was just too esoteric; whereas ICO laid the pieces on the floor and left it to the player's imagination to put them together, Shadow of the Colossus would throw out clues in the form of text messages that would pop up whenever the game sensed that the player was just not getting it. These would range from enigmatic ("The armor it wears seems brittle...") to patronizing ("Grab on to the colossus's tail..."), but, much as the opening exposition seemed a concession to help ground players in a story that left them little to latch onto, these messages were basically an admission that the puzzles were not as elegantly constructed as they ideally might have been for an organic experience.
The puzzle-battles could be frustrating in that way, but the game was more downright infuriating in others. Shadow of the Colossus had many technical glitches that were forgivable because it clearly pushed the PS2 hardware harder than any other game, but it also suffered from more serious shortcomings.
As great an enemy as any of the colossi, the camera would almost always be inadequate to contain all necessary information. Because many of the colossi were so huge that, at close range, they took up most of the frame, it was very difficult to maneuver around them to an advantageous position on the ground. It was simply impossible for the camera to capture both what you were running from and where you were running to, so quite often you would be working with humongous blind spots. A pulled-back camera would have been easier to use, but Ueda and his team must have chosen the closer angle for dramatic effect. All things considered, that probably was the right decision for the sensations they wanted to excite, but sometimes the camera would also just wander for no reason during platforming sections. The recurring example would have me hanging from a ledge, needing to jump off and grab onto an opposing ledge, but, because Shadow of the Colossus employed camera-relative controls, as opposed to an absolute character-relative scheme a la Resident Evil, an unexpected change in camera angle would suddenly alter my jumping angle and cause me to miss my target.
I was also foiled often by the loose controls and physics, which were carried over almost exactly from ICO. Like Ico, Wander moved with very jerky motions, operated with delayed response to button inputs, and could only swim very slowly. But the most annoying thing was that, if he was knocked to the ground, it would take him forever to get back up. And instead of trying to make off with Yorda, the enemies in this game would hang around to target you some more. There was one colossus in particular that was not shy about exploiting this prolonged rising period, and it would repeatedly attack me just as I was getting up, leaving me no chance to do anything about it, until finally I ran out of health and had to start over. That was the sort of cheap that would break me out of the experience and leave me seriously wondering if whatever I was fighting for was worth it.
I suppose I should also discuss the horse mechanics. In Shadow of the Colossus, when you mounted Agro, you did not actually directly control the horse, but rather you continued to control Wander, who attempted to control Agro. There were buttons for acceleration and steering, but these were more like suggestions to the AI Agro, who, while never disobedient, could come to a screeching halt if, for example, you leaned too heavily on the stick to make a turn at high speed. The horse was not useful for much other than getting across the map, and its idiosyncrasies were more bothersome than endearing. It was surprisingly effective in select battles, however, where you could largely let it steer itself while you focused on aiming your bow and arrow.
Back to the game's virtues, the same camera that was so problematic when Wander was on the ground became central to the experience once you managed to get on top of the colossus. As it writhed about to shake you off, and you gripped the R1 button to hang on, just as Wander clutched for dear life to the hairs on the beast's back, the wild swings of the camera to track the violent motions of the living creature that had become the stage itself, along with music that transitioned deftly from ominous to triumphant, and even the game's use of the controller's vibration feature, combined to produce a thrill beyond any before felt in gaming. There was a rarely achieved sense of weight to the action, as the world shook and Wander dangled from one arm, and you realized that this Goliath was resisting just as hard as you had fought to get on it, and whoever survived now depended on who was more determined to defy the other. Then, in that moment when it could shake no more, and you could feel its life as though it were a tangible object in your hands, the feeling of victory was almost godlike. If a player drew nothing else from the experience, this sensation alone made it worth playing Shadow of the Colossus.
ICO was a flawed masterpiece. Shadow of the Colossus, in addition to inheriting most of its predecessor's blemishes, also introduced a large number of new and more serious problems. But there are those exemplary titles that approach perfection in their command of established mechanics, and then there are those that shatter expectations by creating entirely new experiences, and those latter are no less masterpieces despite the roughness that comes with the uncharted. It did not matter ultimately that Shadow of the Colossus did not succeed in all technical aspects, because it showed me things I'd never seen before. The case could certainly be made for it being the PS2's best game.