Slaying the First Colossus
Wednesday, 25 August 2010
When you sink your sword into the skull of the first colossus and end its life, do you feel a tinge of regret? Do you feel sadness, or perhaps empathy, for the great beast you have just felled? And whether or not you do, would you not agree that what you feel for overcoming this foe is intensified by the struggle you faced to climb it, that your satisfaction in attaining victory is enhanced by the trials that preceded its defeat? And if so, you must concede that the triumph you experience is not merely that which you can assign to the task you mastered, but also to the way it was represented.
In the Game Design as Make Believe serial, I discussed Professor Kendall Walton's concept of quasi emotions –states of genuine emotional arousal that are altered in their cognitive nature by the fact that what we are feeling is fictional in origin. For instance, when my wife plays Resident Evil, she reacts with fear to the zombie dogs attacking her – her amygdala is activated, adrenalin flows in her veins – but she does not flee the room or take any other action we would associate with being genuinely afraid. Walton would say she experiences quasi-fear; that it is fictional that she is afraid. This does not mean she is pretending to feel fear, per se – what she experiences is a genuine feeling. But it is fictional that this feeling is fear, since none of my wife's behaviour is consistent with the idea that she is truly terrified of the zombie dog.
In the same post, I also raised the discussion of quasi-emotions in videogames in the context of boss fights. Emotion researcher Paul Ekman refers to fiero as the emotion we feel when we succeed in overcoming adversity (the feeling which causes us to punch the air in victory), but we can use triumph as a more recognisable synonym for this emotion. The question I asked was whether or not we should talk of quasi-triumph when one defeats a boss foe. There is no doubt that the victory occurring is a fictional win – the monster you slay is purely an entity in a fictional videogame world – and that suggests it is fictional triumph (i.e. quasi-triumph) one experiences. A counter argument is that you really did have to master certain motor reflexes and responses in order to achieve victory, and thus it is genuine triumph, with no significant fictional element at the level of the emotion felt.
But what you experience when you slay the first colossus is not merely the triumph of beating the literal motor-skill task involved. If it were, you would feel the same way upon pushing the button atop an equally hard-to-scale giant top that caused it to stop spinning, or if the game simply presented you with a series of reaction and spatial tests that were equivalent to what you had to complete to overcome the monster. Embodying the foe as a beast, a mighty colossus does affect the emotional experience of victory against it. Not only may it elicit quasi-sadness (which the spinning top or the sequence of tests would not), but I will claim that the quasi-triumph is more intense than the triumph associated in mastering the strict task in and of itself. The representational elements of the game directly affect your emotional responses to it, and enhance the feeling of victory. Thus it is quasi-triumph you experience upon slaying the first colossus (accompanied, for many of us, with quasi-sadness), and we must conclude that representation directly affects the gameplay experience.
We hear all too often the phrase "gameplay is everything" but this is misleading. The genuine emotional experiences of play are intensified as a result of the fictional elements of the game – the quasi-emotions of play are enhanced by how the game is represented, and by this I do not simply mean the quality of the graphics. A text adventure uses no graphics, but it still represents its fictional world in a particular way. The experience of the Hitchhikers Guide to the Galaxy adventure game would be palpably different were it not based on Douglas Adams books, even if the puzzles were functionally identical. Magnetic Scrolls' Guild of Thieves would have been less enjoyable for me had it been called Kleptomaniac Garden Gnome because the fantasy of being an apprentice thief holds more appeal in my case than that of a compulsive lawn ornament. Gameplay is not everything. Representation is as vital as gameplay.
In the case of Shadow of the Colossus, it is absolutely crucial to the experience of this game that you are a lone swordsman slaying monstrous beasts that dwarf you, and which fight viciously to defend themselves against you. Identical play elements in a different representation would radically alter the way we would relate to the game. Those of us who experience regret upon slaying the first colossus have this specific emotional response because of the representational elements of the game that make it fictional that it is a living beast we have slain, not just an animated mesh of polygons, and this is an intimate part of the play of this game which cannot simply be set aside as gloss or irrelevance. Our empathy is aroused by representational elements of the game, and this directly affects the play of the game.
Yet the more colossi we slay the less we feel the quasi-sadness and the more we feel just the quasi-triumph instead. Perhaps the representation of Shadow of the Colossus gives us a glimpse into what it is like to be a hunter, a soldier or someone else who kills. The more lives you take, the more routine it becomes. And in this respect, as Miguel Sicart has asserted, the play of Shadow of the Colossus is a genuine moral experience, and of a kind quite distinct from other forms of art since you take an ownership over the choices of play (even down to the decision to keep playing) that you cannot take when you watch a movie character perform similar acts. We do not say "he slew the first colossus", we say "I slew the first colossus". True, it is merely fictional that we did so. But the emotions evoked are no less genuine for being quasi-emotions.
I think the setting can make a difference in a game, but I think it's dangerous to set this as an absolute. Looking purely at game, is Snood significantly different than Puzzle Bobble (aka Bust-A-Move) because the former features creatures instead of bubble? Perhaps, but I don't think it's as big of a change as slaughtering the colossi in Shadow of the Colossus.
Likewise, personal interpretation is a vital part of the key here. I recently posted about Games as a mirror (http://psychochild.org/?p=991) where I argued that a lot of what we get out of games depends on our perceptions. To take a notorious example I mentioned: paying a hooker for healing in GTA. While this has the outward appearance of something controversial, I can believe some people saw this as purely a game mechanic; similar to paying for healing spells in a fantasy RPG. Of course, the trappings will be
meaningful to some people, and will arouse controversy. But, to say every player sees paying the hooker for sex as a titillating experience doesn't seem accurate to me.
That said, a designer absolutely has to make sure they pick a good setting for a game. Picking something that doesn't resonate can distract from an otherwise fine game. But, I'm not sure that a great setting can necessarily make up for the sin of having poor gameplay whereas great gameplay can occasionally shine through a mediocre setting. This is why people say that gameplay is the most important part.
Posted by: Brian 'Psychochild' Green | Friday, 27 August 2010 at 09:57
Brian: thanks for your comment, which raises some interesting points.
"Looking purely at game, is Snood significantly different than Puzzle Bobble (aka Bust-A-Move) because the former features creatures instead of bubble? Perhaps, but I don't think it's as big of a change as slaughtering the colossi in Shadow of the Colossus."
I agree in broad strokes - the more narrative the representation, the more important the nature of the representation. But even in the abstract games, the representation is important. You would have a very different audience for a Puzzle Bobble game in which you fire balls of poop, for instance. Part of the appeal of Bejewelled really is that you are playing with sparkling gems.
"But, to say every player sees paying the hooker for sex as a titillating experience doesn't seem accurate to me."
I have to take task here - because paying the hooker for health is a hugely inefficient way of healing, and if you are looking at it functionally I think you'd find very few players who do this. What made the hooker such a story in connection with the GTA games was the representation and where it lead - not that you can pay a hooker for sex, but that you can pay her for sex *and then run her over and get your money back*. It was this that made the story. And this is *not* equivalent to any healing spell in any game of any kind. The representation in GTA really did bring some new things to the table that cannot be fully understood if you only look at it functionally.
"But, I'm not sure that a great setting can necessarily make up for the sin of having poor gameplay whereas great gameplay can occasionally shine through a mediocre setting. This is why people say that gameplay is the most important part."
This statement has some validity, but it raises questions concerning what "poor gameplay" actually means, and what "mediocre setting" means. There are, for instance, fantasy RPGs with worlds that are not very interesting but for which the gameplay is sufficient. But even a mediocre fantasy setting still delivers representational elements key to that fantasy RPG player's enjoyment of the game. The same gameplay represented differently could have radically less appeal, say, as a wife-beating RPG. That a given setting could have been better isn't necessarily the relevant point here in judging the contribution of the representation.
There were many people who found Endless Ocean absolutely boring, and would accuse it of having poor gameplay. If this accusation is upheld, it's hard to explain why so many people (myself included!) absolutely loved this game. The game design was quite weak in many respects... but the representation was excellently done, and in this case *the representation was the most important part of the game*, because what the game is selling is the experience of being a marine biologist diving on a tropical island.
It's very common for game designers to insist - as you do - that great gameplay is "the most important", but this is a deeply suspicious claim. Most important for what or for whom?
If you look in market terms, there is little doubt that representation is king. Simpsons Hit and Run was very weak on the design side - it still sold 3 million units. Its representation was more important to its sales than its design. There are many other examples in the field of licensed games, because *the right license directly translates into sales* in a way that good game design simply does not.
Now great design can certainly help sales, but getting the right representation makes the difference between success and failure in the market place. The Jet Set Radio games were brilliantly designed - they flopped. The representation wasn't popular enough.
But you might want to object that looking in market terms is unfair to games, that we should adopt some kind of critical attitude (for which the games industry is still lacking foundations) in which we assess the game as an artwork and not as a commercial object. But then, what are you going to cite to me as a stellar game artwork which *doesn't* get its representation right? Complaints levelled at games will fall into both the functional (game design) and representational (art and game design) camps...
It's not so easy to declare gameplay the most important part of a game. Even the most stellar gameplay can be undermined by substituting poor representation. Dwarf Fortress would not have the same cult following if it were Smurf Commune, even if all of the mechanics were identical.
The claim advanced by gamers is that "gameplay is everything" and this is thoroughly misleading. It is a claim that functional elements are the sufficient condition of a "good game". But this is an error. Effective functional elements are a *necessary* condition of a great game, but so also are effective representational elements a necessary condition of a great game. Not just any representation will do. It must be suitable for the functional aspects of the play, and it must offer a fiction that players want to engage in. And these elements are not mere gloss - they are central to what makes a great game, whether you are talking about sales or artistic credentials.
I welcome challenges to this claim! :)
Posted by: Chris | Friday, 27 August 2010 at 12:04
Gah, I made this comment right before going on a cross-country trip, so I forgot about it. Pardon the "comment necromancy", if you would.
Note that I agree that the setting has a large impact on a game, but I'm questioning if it's all pervasive as you assert in the post. Unfortunately, this is not something we can resolve since theme is one of a large number of elements that make up a game. Did a game with a great setting sell because of that setting, or did it sell because of some other element? For example, I've heard people assert that playing with gems is one of the appeals of Bejeweled, but I suspect there are a lot of other reasons why that game succeeded so brilliantly beyond what icons they chose.
To that end, a few direct comments:
"And [using a GTA hooker then running her over to get the cash back] is *not* equivalent to any healing spell in any game of any kind."
I'd argue that cheating the blind reagent merchant in Ultima 4 is similar. This does make the game harder to complete if you want to gain avatarhood, but it's a tactic you can use if you're in a tough spot in finances.
But, the question is: did you avoid cheating the blind merchant because you felt bad about taking advantage of someone (influenced by the setting), or because the game punished you for it (influenced by the gameplay)? Personally, I robbed the blind merchants... er, blinder? Until I found out it was hurting my overall progress and then I stopped. It's not that I personally advocate taking advantage of those less fortunate, but I saw it as a game situation and I optimized based on the information I had.
I have to admit I've not played the GTA games, so I'm speaking from second-hand experiences here. But, I think your point also supports my assertion: you can effectively get "free sex" and the appropriate titillation. And, while this got a lot of press coverage, I assumed that anyone quickly ignored that aspect of the setting and got on with the gameplay.
At any rate, it's an interesting discussion. Thanks for your insightful reply!
Posted by: Brian 'Psychochild' Green | Thursday, 23 September 2010 at 06:15
My posts are *always* open for comment, even years after the fact, so "comment necromancy" is no cause for apology. In this case, it's far from a case of necrosis anyway, since it's only been a few weeks. :)
You are of course right to raise the issue that we *can't* seperate the role of representation from the role of gameplay in determining the success of a game. My goal in this piece is to demonstrate that the oft-attempted attempt to do so *fails* by assuming that gameplay can be given the absolute position. Whatever the relative contributions of representation and gameplay, it clearly cannot be claimed that one or the other has sole relevance.
"It's not that I personally advocate taking advantage of those less fortunate, but I saw it as a game situation and I optimized based on the information I had."
Sicart deals with this nicely in his book on The Ethics of Computer Games: in the instance you cite, there were clearly game-mechanical advantages to be had from behaving in a certain way and so you *cannot* draw usual moral judgements from the representation in this instance. But in the case of the hooker in GTA, I believe the issue is more complicated because of the inefficiency of the activity, which renders its game mechanical aspects (nearly? wholly?) tangential to the representational interpretation.
"But, I think your point also supports my assertion: you can effectively get 'free sex' and the appropriate titillation. And, while this got a lot of press coverage, I assumed that anyone quickly ignored that aspect of the setting and got on with the gameplay."
This was certainly a small aspect of the play of the game, but it was a *big* aspect of the story of the game in the world *outside* of the game, and thus of the identity of the game. And precisely my claim would be that the very identity of the GTA games have emerged from their use of representation (or, if you prefer, from the integration between representation and gameplay). But this is, perhaps, tangential to your point.
Thanks for returning to the discussion!
Posted by: Chris | Thursday, 30 September 2010 at 08:54
Thanks for the reminder, Chris. I am certainly one of those who leans more towards the design purity over theme side of things. But I can't deny your argument. Looking at myself, I see that I've passed over the Super Mario Galaxy games even though they are highly recommended. Yet if the theme was spelunking or ruins exploration, or platforming on a scrap metal planet with a less cartoony avatar I would certainly play it.
Posted by: Josh Foreman | Thursday, 07 October 2010 at 20:50
Josh: thanks for your comment; to my knowledge, no-one has done any research on the role of representation, despite its relevance to people's enjoyment of games - and to subsequent sales. In this respect, the research community (including myself!) has somewhat dropped the ball.
Posted by: Chris | Tuesday, 12 October 2010 at 07:16