I really enjoyed Dragon Age: Inquisition, but it was a bit of a mess. That mess was, however, not specific to that game: it’s entirely typical for a AAA game to throw in a kitchen sink of gameplay mechanisms.

Some AAA games manage to escape that pitfall: Ico and Shadow of the Colossus are particularly good examples. Or at least the versions of those games in my memories are; but I hadn’t played them in close to a decade! So: time to go through them again. And, having gone through the first one: Ico is very far indeed from a kitchen sink.


On an emotional level, I see Ico having two basic Strong Centers that permeate the game. I’ll label one as “separation”, with connection, loneliness, and fear as Echoes. And I’ll label the other as “mystery”, with exploration, wonder, understanding, and, again, fear as Echoes. And on a mechanical level, I see three basic types of actions: traversing the castle, solving puzzles in a specific portion of the castle, and combat with the monsters that appear.

Three basic types of mechanics sounds pretty small, but actually, that size isn’t unusual. Comparing this to the last two AAA games I played: you could say that Tomb Raider has the exact same basic triad of mechanics as Ico does, and you could say that Dragon Age: Inquisition has traversal, combat, and conversation as its three core mechanics. I’m not sure that’s entirely right: those two games both have collections of actions around capability increases (leveling up, inventory) that might be strong enough to call out as a fourth mechanic; in the Dragon Age case, you could even say that leveling up and crafting/inventory are two different top-level mechanical concepts. Still, at a top-level classification of mechanics, Ico‘s conceptual space seems pretty normal.

What is much less normal, however, is how little the tree expands when you drill down into these mechanics, both in terms of breadth and depth. If we consider terrain traversal in Dragon Age: there’s traversing the abstract map, there’s wandering around a region, there are distinct physical areas (towns, distinct chunks of terrain) within those regions, there are smaller portions of those areas (individual buildings within towns, hills within outdoor sections, etc.), there are dungeons to explore, there are landmarks, ocularia, etc. to discover and check off your list, there are herbs and ores to gather every few steps; in Ico, however, you’re moving from room to room and you’re moving around within rooms. (And I suppose the rooms themselves are grouped into sections, so let’s say that there are there levels of traversal.) For combat, the difference is even more stark: the list of layers of combat choices in Dragon Age would be even longer than the layers of terrain traversal, whereas it’s only a slight exaggeration to say that combat in Ico consists of repeatedly whacking your enemies with a stick.

Some of this is a difference in Levels of Scale that is proportional to the difference in length and scope of the different games. Dragon Age takes ten times as long to finish as Ico, and that difference of magnitude in duration is justified by a similar difference of magnitude in the scope of the plot; given that, for the games to feel balanced, Dragon Age needs to have more levels of a given mechanic than Ico does. But some of that difference in magnitude is because Dragon Age tries to appeal to a wide range of tastes: if you want to focus on the broad plot, you’ll traverse the game (including its physical spaces) in one way, if you want to focus on people and relationships, you’ll traverse the game in a different way, and similarly if your focus is on combat or on collecting. Or, for that matter, if your focus is on checking off boxes that the game sticks in your face, or if your focus is on the experiencing the terrain on its own terms instead of in instrumental terms. These aren’t different levels of scale, these are mechanics on the same level of scale that I read as an active choice to appeal weakly to a wide range of possible players instead of appealing strongly to a narrower range.


So: we have a vertical difference explained by having Layers of Scale appropriately applied in games of different scope. And we have a horizontal difference explained by Ico being more prescriptive in how it expects you to enjoy the game than Dragon Age. But that’s not all that’s going on here: the scale of combat is a vertical difference that is not explained by the difference in scope, combat in Ico is remarkably undeveloped even given the short length of the game.

Unlike so many other games, Ico isn’t treating combat as an unquestioned inclusion, as a means to its own end. If we analyze Ico as a game about separation and about mystery, then yes, combat does fit into the game: both of those centers link to fear, and enemies reify that fear. But Ico makes the decision to link the enemies most strongly with separation: the danger that the enemies bring (excluding the very end of the game) isn’t that they’ll kill you, it’s that they’ll separate you from Yorda. A hypothetical rich combat system that the player could learn to master would work at narrative cross-purposes to that reification of fear of separation: it would turn the focus onto the player and mastery and away from the focus on fear of separation. Contrast this with combat in Tomb Raider and in Dragon Age: Inquisition: the former game is a game about Lara coming into her own, and the latter game is about your inquisitor turning from a nobody into the leader of an army that can defeat demons and a world-threatening enemy; while I think combat in both games was a little bit overemphasized, it’s absolutely the case that development of the player’s combat skills fits into those games’ narrative themes in a way that it doesn’t in Ico.

I am on an anti-combat kick these days, however, and for the first part of this playthrough, I thought that the combat was a little long and tedious, that the game would have worked as well with much less of it. Now, though, I’m not nearly as sure of that: I acquired the mace in this playthrough, and once I had the mace, the battles were so fast that they had very little impact. So, based on that, I underestimated the power that was contained in the length of the initial battles: I’m not sure that the length of the battles was judged quite right, but maybe it was, and at the very least the length was a lot closer to being right than my original hypothesis was.


The other two mechanics, terrain traversal and puzzle solving, are much more developed than combat. Terrain traversal serves as a means to display the mystery of the castle: mystery turns to exploration turns to wonder. (And yes, turns to fear as enemies appear.) You and Yorda have to traverse the castle together, lest separation becomes a danger; and part of the mystery of the castle is its age, its beginning to fall to ruin: that highlights another form of separation, between yourself and the unknown people who built the castle. Returning to you and Yorda traversing the castle together: that connection is reified by your holding hands; but that in turn is animated in an asymmetric way, leading me to read the player at times as an excited child wanting to drag somebody older around, somebody who would be just as happy to take things a little bit more slowly. And that shows yet another, subtler, form of separation, namely one of age and knowledge: Yorda is older than you, knows more about the context of the castle than you do, knows more about the context of the struggles than you do, and will be affected differently by failure than you will.

The terrain traversal has Layers of Scale: the castle is divided into large sections, those sections are divided into rooms, and within each room, there’s a fair bit to traverse and examine. That examination turns into puzzles, which gives the game a form of Alternating Repetition (move, puzzle, move, puzzle); the puzzles reify one aspect of the mystery of the castle, and are designed with separation/reconnection as a key mechanic.

Those puzzles, from a traversal point of view, do provide a hint of The Void: they jerk you out of motion and force you to stop, look, and think. (Though puzzles frequently do fit into traversal: Non-Separateness, you could say, or Deep Interlock and Ambiguity.) But the puzzles are really more of a change in focus than they are a manifestation of The Void: your mind is active while grappling with them, your fingers are frequently active, and at times you’ll be frustrated, none of which I associate with The Void.

A better example of The Void, and an example which is one of my favorite mechanics in any game, are the couches. On a mechanical level, the couches are simply save spots, but they’re so much more. They divide sections of the games from each other, with each section being named in the save file. They highlight the importance of connection, because you can’t even use them unless Yorda is right there with you. They give down time and a space for contemplation for you as the character: you can sit there with Yorda, maybe just relaxing, maybe looking around at the castle, maybe thinking, maybe enjoying sitting there with somebody next to you. And they give down time and a space for contemplation for you as the player: as a boundary between sections, you’re told that you’ve accomplished something and should get ready to gather up your strength for what comes next, and as a save spot, you’re given explicit permission to take all the rest you need before continuing. I can’t think of another piece of furniture in any game that manages to accomplish quite so much.


Which makes it all the more powerful when the couches go away. For the last quarter of the game, you’re alone: and the game has reinforced the importance of connection between you and Yorda enough that her absence alone has a real impact, on the narrative of the game, the emotions of the game, and the mechanics of the game. (No more puzzles about helping her traverse a large gap, for example.) That would have been enough on its own, but the linkage of “no Yorda means no couches means no save spots” is what gave that part of the game such a visceral impact for me: I kept on wondering how much longer I’d be groping through the environments alone, how much I’d have to bear up, whether I’d have to stop and replay whole sections because I would have to do something outside of game that would prevent me from finishing that part of the game.

Normally, when games don’t let you save, I get quite annoyed at them: I see that as an active lack of respect for my time by the game. But I don’t feel that way about Ico at all: the game only does that once, it does a good job of signaling that something unusual has happened and places a save spot right before that, it does that for a clear narrative goal, there aren’t long-term failure states within that section, that narrative goal has a real emotional impact, and, ultimately, it only takes an hour and a half, maybe two hours to make it through that section of the game. (So it doesn’t require you to devote more unbroken time to the game than a movie would, it fits within the context of normal human lives.) Symmetry within games, in the form of Alternating Repetition, is all well and good, but you need to break that symmetry occasionally, to throw in a bit of Roughness: that’s what we have here.


Quite a game: I had a lot of respect for Ico going into this playthrough, and I have significantly more respect for it at the end.

Post Revisions:

This post has not been revised since publication.