Noah Falstein’s “The 400 Project” is an “attempt to collect The 400 Rules of Game Design [from] the column “Better by Design” in Game Developer magazine.

It would be easy to mock, so I’ll start with a bit of mockery:

Rule 34. Emphasize Micromanagement for German Speakers. This rule is an instance of a more general rule to “consider national sensibilities”

I think you mistyped “national stereotypes” there, Noah.

Seriously, there’s plenty of sensible stuff in the list. But the “rules” approach bothers me. Every game, good or bad, breaks some of the rules. So how can you tell when it’s good to break a “rule” is good and when it’s bad? Let’s have a look at some of the rules.

Rule 11. Emphasize Exploration and Discovery

I don’t remember there being any emphasis on exploration in Tetris. Or Ikaruga, Wario Ware, or Frequency. But that’s a bit of a cheap shot. What this rule doesn’t tell you is that there are trade-offs to consider. Games based around exploration—“open world games” in this article by John Harris—can create a strong sense of immersion in a believable world. But they demand a lot of time and patience from the player, and suffer from longueurs and boredom: “what do I do next?” and “do I really have to walk round the whole world again looking for the secret I must have missed?” are questions that I’m sure most players ask themselves from time to time when playing games like Super Metroid and Castlevania: Symphony of the Night.

Rule 3. Maintain Level of Abstraction. Immersion is easily disturbed—don’t make the player re-calibrate his “suspension of disbelief” and lose touch with your game.

Some of the most effective moments in the first Metal Gear Solid are when it breaks the “fourth wall”. In one scene the villain points out that you haven’t saved your game for a long time and if you fail at the next challenge then your game will be over. This works because the threat (of having to replay the difficult section that led up to the current scene) is quite realistic, so it has the effect of increasing the tension.

Rule 12. Provide Parallel Challenges with Mutual Assistance. [i.e. construct the game so that the players faces several challenges at any given time and completing one challenge makes it easier to complete the others]

Hardly any games do this, so in what sense can it be a “rule”?

More to the point, this is something that can work in only some types of game, ones in which you have an inventory of items or a collection of powers, and in which the challenges are fairly generic, so that items or powers are generally useful, not specific to particular challenges. The canonical examples of such games are free-roaming RPGs like World of Warcraft—if stuck, you can go back and replay areas you’ve already visited in order to level up and improve your equipment—and certain strategy games like Pikmin where you keep your army from one battle to the next—if an enemy proves too tough you can go away and augment your army until you have enough units to defeat it.

These kind of games are subject to a well-known disadvantage: that “grinding” (increasing your powers within the game in a safe but repetitive and time-consuming way) is the dominant strategy. Even though it’s not particularly enjoyable, it’s less risky and much easier than the alternatives. Games have to go to some lengths to get around this. Pikmin imposes a global time limit, so there’s always pressure to tackle the harder challenges sooner; but this makes the game rather intimidating because you can make mistakes early on (wasting too much time) that make the game impossible to finish. (Becoming impossible to finish but not actually finishing seems like it ought to be something that there should be a rule against, but none appears in the list, at least not so far.) World of Warcraft does taper the rewards for fighting monsters, but in the end it is basically all about the grind.

Which leads us neatly to this:

Rule 93. Avoid Dominant Strategies that Trivialize Player Choice

Well, one simple way to achieve this is to break rule 12 and make each challenge more or less independent of the others.

Now for a long digression on rules.

During the neoclassical period in the 17th and 18th centuries in Europe, it was common for critics of drama to emphasize the importance of three rules, which they attributed to the ancient Greek and Roman dramatists they admired and wanted to emulate. These rules were the classical unities: of action (a play should treat a single subject, with no or few subplots); of place (the stage should represent a single place); and of time (there should be little or no compression of events; the action should represent little more time than it takes to perform).

John Dryden was the dominant figure of the neoclassical revolution in English literature. In his Essay of Dramatick Poesie (1668), he uses the unities to criticize Shakespeare’s history plays:

… they are rather so many Chronicles of Kings, or the business many times of thirty or forty years, crampt into a representation of two hours and a half, which is not to imitate or paint Nature, but rather to draw her in miniature, to take her in little; to look upon her through the wrong end of a Perspective, and receive her Images not onely much less, but infinitely more imperfect then the life: this instead of making a Play delightful, renders it ridiculous.

Not content with this critical blast, Dryden set out to show the superiority of the classical unities by writing a “regular” (that is, rule-following) version of the story of Anthony and Cleopatra. Shakespeare’s play starts soon after Antony’s meeting Cleopatra in 41 BC and follows the story chronologically to their suicide in 30 BC. The scene leaps from Alexandria to Rome, Athens, Actium, and Syria. The action treats the complexities of the intersection of politics and personal life: Antony’s wife Fulvia fights against Octavius but dies; Antony fights Sextus Pompeius and the Parthians; Antony marries Octavia; he breaks with the other triumvirs and allies with Cleopatra against them.

Dryden’s 1678 version of the story, All For Love, occupies only the last few hours before the deaths of the principals, and is confined to Cleopatra’s palace in Alexandria.

In his preface to the play Dryden explains the pains he has taken to respect the unities:

The fabric of the play is regular enough … and the unities of time, place, and action, more exactly observed, than perhaps the English theatre requires. Particularly, the action is so much one, that it is the only one of the kind without episode, or underplot; every scene in the tragedy conducing to the main design, and every act concluding with a turn of it.

But he also acknowledges that he had some difficulties in forcing the sprawling content of the story into the straitjacket of the classical tragedy:

That which is wanting to work up the pity to a greater height [that is, a tragic fate happening to a virtuous protagonist], was not afforded me by the story; for the crimes of love, which they both committed, were not occasioned by any necessity, or fatal ignorance, but were wholly voluntary; since our passions are, or ought to be, within our power. …

The greatest error in the contrivance seems to be in the person of Octavia; for, though I might use the privilege of a poet, to introduce her into Alexandria [the historical Antony had divorced her in 32 BC], yet I had not enough considered, that the compassion she moved to herself and children was destructive to that which I reserved for Antony and Cleopatra.

All For Love is by no means a failure; it remains one of the most performed of Dryden’s plays, though I suspect it does not have even one performance for every hundred of Shakespeare’s version of the story. But it is a clear indication of the difficulties that even a great writer can get into when he adopts rules without regard for whether the specific effect of following the rules suits the story he is telling.

So where did these rules come from and why was Dryden so keen to follow them? They derive ultimately from the discussion of tragedy in the Poetics of Aristotle, as filtered through the interpretations of classical scholars of the Renaissance. Classical learning was held in high regard in early modern Europe, and imitation of its forms was one way for dramatists like Dryden to claim a similarly high status for their work.

Aristotle is quite clear about the specific function of each of his recommendations. The goal, as he sees it, of tragedy being to “excite pity and fear”, each element must support this:

… the change of fortune presented must not be the spectacle of a virtuous man brought from prosperity to adversity: for this moves neither pity nor fear; it merely shocks us. Nor, again, that of a bad man passing from adversity to prosperity: for nothing can be more alien to the spirit of Tragedy; it possesses no single tragic quality; it neither satisfies the moral sense nor calls forth pity or fear. Nor, again, should the downfall of the utter villain be exhibited. A plot of this kind would, doubtless, satisfy the moral sense, but it would inspire neither pity nor fear; for pity is aroused by unmerited misfortune, fear by the misfortune of a man like ourselves. Such an event, therefore, will be neither pitiful nor terrible. There remains, then, the character between these two extremes—that of a man who is not eminently good and just—yet whose misfortune is brought about not by vice or depravity, but by some error or frailty. He must be one who is highly renowned and prosperous, a personage like Oedipus, Thyestes, or other illustrious men of such families. A well constructed plot should, therefore, be single in its issue, rather than double as some maintain. The change of fortune should be not from bad to good, but, reversely, from good to bad. It should come about as the result not of vice, but of some great error or frailty

But if you have some other goal for your drama than the excitement of pity and fear, or if your characters do not fit the mould of the tragic protagonist, Aristotle’s recommendations are not much help, and trying to follow them blindly can be positively harmful.

It seems to me that some writers on game design are in a similar situation to the neoclassical scholars, in that their description of their art is phrased in terms of rules, but they lack the analysis of how the rules work, and to what purpose, that would enable a practioner to select among them.

This missing analysis is exposed most starkly when rules directly contradict each other.

Rule 5. Make Subgames. Players want to participate in the course they take through your game—so give them plenty of opportunities to voluntarily take up ancillary challenges.

Rule 92. Trim the Fat. Ruthlessly trim away any parts of the game or story that do not directly contribute to the player’s enjoyment, and omit extraneous elements.

How should the game designer choose between these? (In most cases budgetary considerations will force the selection of the latter, but if you have the money, how do you decide how to allocate it between the main game and the subgames? Might it depend on what kind of game you are trying to make?)

Or what about this pair:

Rule 62. Make Challenges Require Skill

Rule 24. Don’t Penalize the Player

If a challenge requires skill then it must be possible to fail at it. And failure must come with a penalty, or in what sense is it failure? (Even if the only consequence of failure is that you have to try the challenge again, that’s still a penalty, of your time and effort.)

I don’t want to give the impression of picking on Noah Falstein. As I say, his list is generally pretty sensible. There’s worse out there, examples of more or less complete failure to understand how the elements of a game go together to makes it good.

For example, designer Ernest Adams, in his column “Bad Game Designer, No Twinkie! II”, discusses how unrealistic physics can spoil the player’s suspension of disbelief, and then comments,

the physics in video games is mostly ludicrous—the first time I saw Sonic the Hedgehog change direction in midair I heard a strange whizzing noise: Isaac Newton spinning in his grave

Adams is not a fan of platform games. If he had ever had to design one, he would know that Sonic, Mario and other jumping characters behave like this for very good reasons: (1) lining up a jump is hard and it’s easy to make mistakes; mid-air control means the player doesn’t have to be accurate in prediction; (2) the player continues to have an active part in the game during the jump instead of waiting as a passive spectactor, like a golfer watching the flight of the ball; (3) the player can react in real time to fast-paced events such as the changing paths of enemies, essential for the “hop and bop” gameplay (jumping on the heads off enemies to kill them). It’s instructive to contrast the platform games of the Super Mario Bros. generation with the previous generation of games like Donkey Kong and Manic Miner. The later games are much more fun to play, and mid-air control is a big part of it. That the physics is unrealistic is a comparatively minor aspect.