"But we cannot relegate computer games to the cesspit of pop culture solely on the evidence of the current crop of games. The industry is too young and the situation too dynamic to dismiss computer games so easily. We must consider the potential, not the actuality. We must address fundamental aspects of computer games." --Chris Crawford, The Art of Computer Game Design
It's been a while since we've done a book review. And yes, this is yet another game on videogame design. And it's yet another argument that games could and should be considered art. But this one is a little different from the others, for one simple reason: it was written in 1984.
Comical opening aside, Chris Crawford is something of a well respected legend in game studies. In part, that's for the games he's designed. While I haven't actually played any of them (the majority of them being made before I was in kindergarten), they did generally enjoy a rather good reputation at the time, especially his magnum opus, Balance of Power, which still is touted by some as a quality game. In part, his reputation comes from his role in hosting the first Game Developers Conference in his living room, or as it was known then, the Computer Game Developers' Conference. And he also founded The Journal of Computer Design (although he was largely the only person writing in it, from what I understand, which diminishes the accomplishment a little). But a lot of his impact in game design theory comes simply from the fact that he was the first. The first to publish, the first to articulate his ideas, the first to set on paper an argument that games could be more than they were. And today, we're going to talk about that book, The Art of Computer Game Design.
The Art, as I am cleverly abbreviating it, consists of seven chapters and about 120 pages. Unlike many game studies books today, it doesn't really have an argument, beyond the one that there is something here worth arguing about. His preface sets the terms of the discussion: games are art, or, at least, they have the potential to be art. Their level of audience participation allows them to offer something more engaging than art as well. And while the current level of games (circa early 1980s) is the equivalent of junk food, time will develop a larger palate. That brings him to his first chapter, wherein he establishes a definition of games and offers reasons why people play them. The basic definition: a game is a formal closed system that subjectively represents a subset of reality. (Personally, I object to the closed part, but the rest of it is a good start.) In particular, they offer interaction (there's that word, already vague by the early 1980s), conflict (which he distinguishes from violence), and safety, in that you can experience this conflict with fewer consequences than their real life equivalents (I haven't gone through the real life equivalent of hunting super mutants in the apocalyptic ruins of DC, but I admit that it wouldn't be very safe). As for reasons to play, he includes learning safely, expressing behavior that society may frown on without consequence (shooting and killing and so forth), indulgence in fantasy, and, on the social level, games act as social lubricant and allow players to show off their talents and skills.
The second chapter is Crawford's taxonomy of games, which is based purely on what was available at the time. The two main categories are skill- and reflex-based and cognitive based, with subdivisions including things like Paddle games and sports games for the former, and adventure games and war games for the latter. I should note that many of these games have large, full-colore screenshots, which is rare for game studies (although I imagine copyright for games circa 1984 isn't what it is now), and appreciated. Chapter 3 considers the computer (and as always, circa 1984. You know what--I'm going to stop saying that. Just remember that everything is the state of the industry as it was a year before Golden Girls was a thing.) as a platform for game technology. It's dynamic and responsive, but limited by its I/O. He generates seven principles to allow design to go smoother. These include designing with the I/O specifically in mind (game port designers take note), taking advantage of the computer's better points and focusing on dynamic programming over static information sets, and branch softly, which basically means presenting options rather than restricting paths. It's Crawford's version of the debate we see today (or maybe yesterday) between emergent and directed gameplay. I'm more interested in the second point up there, that computer games are better equipped to handle dynamic processes than static blocks of information. Games such as Mass Effect and Elder Scrolls have a huge amount of text-based backstory, which accumulate in either the codex or found books, respectively. The problem is that they can be rather tedious to read, and presented (in Mass Effect's case) in rather tiny font. A decade or so earlier, you would have found that information in the manual. I do think it's easier for the player to process that sort of information in that form, if only because you can reference it while still actually playing the game. Instead, due to costs in printing, we get more and more shoehorned into the game itself. It works well for tutorials, because they generally do take advantage of the game's dynamic nature, but for other bits of information, it feels a bit more forced.
But I digress. To continue the discussion at hand, Chapter 4 is Crawford's version of the design sequence. It's a rather familiar pattern: determine idea, research, design. Then move into programming, evaluation, playtesting, and finally polishing. The interesting parts are in the details. The idea, for Crawford, is the heart and soul of the game, and consists of two parts: the goal (that is The Meaning of the game) and the topic (how it accomplishes getting its meaning across). If the goal and topic don't speak to you, he maintains, then you're doing it wrong. He's also extremely opposed to bringing in players rather than designers for gametesting, as he basically doesn't trust your average gamer to know what they want: “I will concede that such methods can prove to be a useful way to guide the mass production of cheap games by designers of limited talents, but this book is not written for them.” Of course, the game tester of the time wasn't as sophisticated as today's-- OMG, is that Batman? Made out of lego? Out of the way, reasoning adult self! There's nostalgia to be had! Ahem. Crawford also advises would-be designers to write their own manual, to appreciate the work writers do and to get a better understanding of their own game. As a scholar studying game manuals and a writer, I approve.
Chapter 5 goes over some basic tips Crawford has for AI management. First, you can just give the AI better resources, but that's lazy. He prefers improving the routines, through an application of point systems, coordination, and careful transitions algorithms. And in terms of player balance, he brings up asymmetry and intermediaries--you can tell he cut his teeth on wargames. IMHO, the most interesting game he describes here is Russian Civil War, a game so obscure that even Mobygames turns up a blank. The idea is that you have more than two players, and each player gets some Red armies and some White armies. And each round, you can use one of your armies to attack someone else's army of the opposite color. The game ends when one color is eliminated, and the player with the largest number of remaining forces wins. What I like about the game is that it makes you think about more than the battle at hand, and sometimes question whether a particular battle is worth winning. It also captures that sense of a civil war, in that you're on both sides, so to speak.
Chapter 6 is Crawford's account of the development of his game Excalibur. It's basically an instant of the method developed in chapter 7. It's an interesting anecdote of how games (or at least this game) were developed back then. The final chapter is on the future of video games. Crawford warns that games can't go up in sales forever. (A prescient statement, given that the Atari collapse happens a little while later.) And while technology improvements will mean better looking games, unless the artistic side improves, there may not be better games. The car and television, he argues, went through four phases: a pioneering phase, a conquest expansion phase, a general transformation of society by the new technology, and a transformation of the technology by society. (And it's very North American to use a pioneer colonization metaphor to describe technology changes.) Games will be changed by society in that there will be an expansion, an expansion beyond just games for teenage males.
So after that exhaustive investigation, what can be said? Is the book still relevant today? Yes and no. Or, the better question is perhaps, who is the book still relevant for? (or for whom, if you're the pedantic grammar type) The taxonomy part is the most obviously dated, in that those divisions don't work as well for games any more because they've moved on. And the theory regarding games themselves is rather unsophisticated for a modern game scholar, what with our aesthetic theories and our ludologies. The game design stuff ages a little better, but with one very, very significant caveat. The big difference between the game developers of 1983 and 2012 is scale. A single person in a development team numbering hundreds simply doesn't have the luxuries Crawford describes. You can't start over, you can't choose to work on something that appeals to you artistically, and you can't avoid having long game testing sessions, because none of that is your call. The designing steps work best, in fact, for the auteur designer, the indie gamer who is looking to make a statement and an identify for themselves.
The book's other audience is, well, me, or rather, those interested in it in terms of its value as a historical artifact. Crawford isn't really interested in player response to games, as long as the game sells well; that much you could guess from the dismissing of game testers. But as a starting step towards such things as game balance and the definition of a game and game taxonomies, a study like this has value. Even the chapter on Excalibur is rich in detail in terms of how designers saw themselves, and how their process worked in action. Crawford personally wrote the novella that came with the manual; can you imagine how long it's been since the programmer and the story teller for a blockbuster game were the same person? (And not, say, dozens of people?) It's a smorasbord of "the way things were." Granted, you could get a lot of the same information by simply sitting down and talking to 80s designers, but... well, we aren't doing that. Not enough, at least. And as a regular journal keeper can tell you, a written record preserves and abides the way a spoken anecdote doesn't.
And what about Crawford's predictions of videogame future? Unfortunately, it seems his worst fears have come to pass; the majority of videogames are blockbuster shleck, for lack of a better term. And the genres he hoped to see spring into being--soap opera games, country music games, goth romance games--haven't appeared. (Well, maybe Vampire: The Masquerade for the goth thing.) On the other hand, we have seen the reins slip from the teenage boys, with the development of the so-called casual gamer, and the simple aging of videogames' demographics. It has led to an expansion of what games are, but also a proliferation of clones and shoddy games. And this is from someone who likes videogames. Ah well. Dealing with a certain amount of shleck is the burden the pop culture sholar must bear. My favorite Crawford prediction of things to come, however, appears on the final page: “By 1985, software stores will be as common as record stores; by 1990, they will be as common as bookstores.” It's rather sad that, in 2012, this has become true by virtue of the record and book stores dwindling away.
At any rate, the one sentence summary: Crawford's book does stand the test of time, but its greatest value is to either the indie designer or game historian.