Videogames and Rules

Apr 30, 2011

Rules in Computer Games Compared to Rules in Traditional Games is now available as an updated and improved version of this essay, presented in video form.

This is a refinement and continuation of last month’s entry, Games are artificial. Videogames are not. Games have rules. Videogames do not. Explanatory footnotes, additional examples, and sources are at the bottom, reachable by clicking the linked numbers inside braces [#].

First, to clear up a few points:

  • Though I am using very specific meanings of words, it is not my intent to argue in favor of specific meanings, nor to suggest that any words ought to be used differently. My purpose is not to say something about words, but to highlight a distinction between videogames and non-digital games that exists apart from the terms used here to call attention to it.
  • I am not making this point for its own sake, but because I believe it can help expose more aspects of videogames for consideration in formal design and criticism.
  • I have a personal interest in fairly broad uses of videogames, including non-entertainment, notgames, and the fascinating/worthwhile though not ‘fun’ Chris Crawford projects. Although this distinction is not important to the argument – what is said here certainly applies to Super Mario World and NBA Jam – my interest in advancing fringe uses of videogames is partly the source of my desire to better understand properties of videogame-like gameplay beyond what can be uncovered through cloning and incremental innovations in entertainment uses.

I think that when understood, the observation here is actually not so controversial as misunderstandings of it seem. If you are among the readers that already grokked the previous entry, the material below extends and explores the point – this is not simply a second attempt to introduce the same idea.

Bernard Suits, in The Grasshopper, defines a game by four aspects:


Prelusory goal – an “achievable state of affairs”, such as getting a golfball in a hole.

Lusory means – “means which are permitted (are legal or legitimate) in the attempt to achieve prelusory goals.” [1]

Constitutive rules – “rules which prohibit the most efficient means for reaching a prelusory goal”, for example disallowing golfballs being carried and dropped into the hole instead of hit.

Lusory attitude – voluntary acceptance of the inefficient means.

He offers a single definition to tie these four terms together: “To play a game is to attempt to achieve a specific state of affairs [prelusory goal], using only means permitted by rules [lusory means], where the rules prohibit use of more efficient in favor of less efficient means [constitutive rules], and where the rules are accepted just because they make possible such activity [lusory attitude].”

Reoccurring in how Suits defines constitutive rules is the notion that, “the permitted means for moving… are of less scope than the possible means for moving… [2]”

In a videogame, however, the permitted means are of the same scope as the possible means. No means are possible and prohibited, except perhaps for hacking [3]. In videogame chess, unlike non-digital chess, we are truly unable place a piece where the rules do not allow for its movement; in videogame golf, unlike non-digital golf, we actually cannot pick up the ball and carry it to the hole.

This deemphasizes lusory attitude, because the player does not need to accept inefficient means in order to play.

Because all possible means are permitted means in videogames, players do not need to learn the rules to play. Karen Sideman noted that we can instead learn to play by playing, removing the necessity for study and/or peer explanation which occurs prior to first playing most non-digital games [4]. Additionally, Chris Crawford observed [5] that thanks to the exactness and speed of rule evaluation in videogames, there is no room for disagreement over the rules, and the rules can exceed the complexity of what humans are capable of learning and enforcing.

One common type of disagreement related to rules occurs over judgments made in rule enforcement for a particular case: did the ball bounce on the boundary line, was the stopwatch started too early, were referees not paying attention when a player knowingly employed means that are possible though not permitted? In videogames these measurements are exact; rather than needing to measure something that exists physically, independent of measures, the coordinate itself that the representation exposes is already accessible as a measurement infinitely precise by its definition (the ball is exactly where we render it, because we render the ball exactly where it is).

Crawford’s second observation, that computers offer much greater speed and precision, enables real-time videogames. Though many real-time non-digital games exist, they all assume natural laws and human perception as foundations. In videogames the physics, representation and input are open to a broad range of meaningful reimagining [6]. Developing a tacit understanding of how these elements are constructed in each project is usually central to gameplay in real-time videogames.

To be human enforceable, non-digital games emphasize rules as discrete events that humans can easily distinguish. Listing rules for Super Mario Bros, we might include, “Holding the jump button makes Mario jump higher. Holding the run button makes Mario run faster.” In sports, humans jumping and running make sense, however Mario is not a real human, and what Mario does is neither running nor jumping, but instead these are abstract invented actions unlike anything else, which only superficially and conceptually resemble running and jumping [7].

Non-digital rules are either situated in pure abstraction (chess, checkers, Go) so as to proceed only in relation to their internal rules, or situated in reality and all its physical/physiological complexities (basketball, golf, boxing). Although all videogames “proceed only in relation to their internal rules,” through thousands of internal ‘rule’ calculations per second many real-time videogames, particularly real-time or arcade-style, create the impression of taking place in the latter, physical domain. The difference, between real reality and this illusion of reality created entirely by internal rules, is that the particulars of spatial, temporal, and visual implementation are not necessarily any more based on reality than Mario’s “jumping” or “running.” For many real-time videogames the authoring of these decisions is every bit as significant and characteristic to the gameplay as the causal, discrete events one might document as rules [8]. Such matters of implementation are fundamental to differences in gameplay between videogames, likely to have a significant effect on player experience/enjoyment, and may even be construed as a source of authored meaning (especially if deviating noticeably from reality or convention).

This distinction of human enforceability and selectively authored causal, perceivable reality enables us to dig deeper into the earlier idea of disagreement over judgment in enforcing a rule. In reality, friction and gravity do not need any referee to function, and though “friction” and “gravity” in arcade videogames can be implemented in many different ways, these are in every case equally real phenomenon, in the sense that they are not imaginary and disregarding them (as we can disregard a rule of basketball by running without dribbling) will not prevent it from affecting player mobility (as in platform jumping in Super Mario World) and degree of control (as in Asteroids, Gravitar). Likewise, the overlap of two ‘physical’ objects produces distinctly different behaviors in many videogames – sometimes blocking movement, sometimes repelling backward, sometimes allowing passage with damage or ignoring collision altogether – though the spatial consequences of physical contact is not something which in reality we entrust to authored or subjectively enforced rules. The difference is not simply that a computer is referee, that enforces the rules instead of humans, but that videogames construct relationships and continuous behaviors in such a way that they do not require referee nor enforcement. To a videogame, it is equally unambiguous whether a player made the shot from in-bounds before the timer stopped as what precise slowdown should be exacted on the player’s avatar every logic cycle that the directional pad is not being held.

Acknowledging that difference in the precision of simulated rules, videogames can, of course, incorporate simulated rules to simulate games. Videogames can also simulate fallible referees with narrow attentional limitations, as is sometimes done in pro wrestling videogames to simulate illegal maneuvers that can be performed without penalty when the referee looks away [2b]. Along these lines, videogames can simulate aerodynamics in aircraft-based projects. However because there is such a broad range in the implementation and intended purpose of flying videogames, aerodynamics are handled in a variety of different ways, and surely it would be an inaccurate overstatement to say that (all?) videogames “have” aerodynamics. It is in connection to that simulated, ambiguous, inconsistent form of inclusion that I bring into question whether videogames “have” rules – some videogames simulate rules, but what that means varies just as dramatically case-by-case, and even then there are projects when the simulation of such rules can be secondary to the fundamental gameplay in comparison to the details that are not captured by what would be identified as analogous to non-digital rules [8].

The overlap between possible means and permitted means in videogames also affects cognitive load. The arbitrary spatial, temporal, and representational complexity possible in videogame systems can turn gameplay into the experimentation, execution, and continual refinement of heuristic algorithms [9]. The player, not responsible for ensuring that enemy ship pieces and shots move as they should, is free to dedicate all mental energy to finding a more successful response to the system, which may or may not involve an increasingly accurate understanding of how the underlying system functions. Remembering Crawford’s point about potential for computational complexity in videogames that exceeds human limitations [5], a completely accurate understanding of the system may even be impossible, especially to someone without access to the source code. Put another way, gameplay may not consist of wondering, “how is my understanding about the system incorrect,” but focusing instead on, “how do I need to change the pattern of my reactions to more consistently achieve the desired end result.”

Though the distinction is subtle, and can be applied to many non-digital games as well (chess, poker, and sports involve similar learning experiences), the effect of this detail is that a player can play a videogame – and even become skilled in it – without ever actually learning or understanding the rules by which it operates. In the case of non-digital games, the rules have to be learned in order to play, but to extend Sideman’s notion [4] that “…with digital games it is possible… to learn the rules of the game as it is being played,” I propose that with videogames it is possible to achieve mastery through play without ever learning most of the game’s ‘rules’.

For example: if a deathmatch videogame player learns that every time they get the flamethrower and run around firing wildly they win, they are unlikely to deduce through analytical experimentation the explicit functionality of the flamethrower in its relationship to range, rate of fire, fuel expenditure, and so on. A tacit feel for the overall effect will suffice, much as a football player does not need to understand the numerical physics of air resistance to throw a successful pass. Curiosity and analytical habits notwithstanding, our player may even ignore anything else about the videogame’s gameplay besides how to find and fire the flamethrower, until some circumstance pushes back and usually causes that strategy to fail. Even then, the player is likely to explore other easy options (“can I use the same strategy, but focus on the rocket launcher instead?”) before, if ever, resorting to more thorough analysis of flamethrower functionality in search of some way to succeed. This relationship is why balanced tuning in videogame design affects duration of entertainment value – it increases the circumstances which push back and motivate deeper exploration. It is also why a videogame attempting to communicate a systematic concept by simply facilitating interaction with a model of that system is unlikely to result in rigorous learning (for more online reading about everything in this closing point, see the links in [9]).

Sources and Footnotes

1– Suits uses the terms lusory means and constitutive rules in a closely related way, which he acknowledges: “We have defined lusory means as means which are permitted without examining the nature of that permission. This omissions will be repaired directly by taking up the question of rules.” Lusory means are the means permitted, constitutive rules specify which means are permitted, the terms are two sides of the same coin. Consequently it is difficult to say something about one without meaning something about both. To minimize redundancy and remain consistent, in discussing both I will primarily refer to constitutive rules, which directly affect lusory means in a way that I believe will be sufficiently clear from context.

2– (a.) Bernard Suits, The Grasshopper: Games, Life, and Utopia: “Since the knight, for example, is permitted to move in only a highly restricted manner, it is clear that the permitted means for moving the knight are of less scope than the possible means for moving him… the way in which those rules function is to prohibit the use of the most efficient means for achieving that state of affairs.” Emphasis added. He continues, “I think you will agree that the simplest, easiest, and most direct approach to achieving such a goal is always ruled out in favor of a more complex, more difficult, and more indirect approach.” (b.) Suits also acknowledges rules of skill, which are not rules in the same sense, but rather wise tips to keep in mind in order to play a game well, such as “keep your eye on the ball.” More relevantly, he then clarifies a third case, “There is a third kind of rule in some games which appears to be unlike either of these. It is the kind of rule whose violation results in a fixed penalty, so that violating the rule is neither to fail to play the game nor [necessarily] to fail to play the game well, since it is sometimes tactically correct to incur such a penalty [e.g. in hockey] for the sake of the advantage gained. But these rules and the lusory consequences of their violation are established by the constitutive rules and are simply extensions of them.” In the same way, videogames that simulate penalties for violating simulated rules, with or without simulating the attentional limitations of a referee, do not contain violations of the game’s defining rules, but rather afford suffering (or risking) that penalty as a valid option within playing the game.

3– (a.) Consider as a constitutive rule that videogame controllers and console switches are the only permitted means of interaction, so as to exclude file and memory hacking. In general, the subject of what is ‘permitted’ for a videogame, especially single player since social rules are then not at stake, is a complex question perhaps varying by authorial intent; is a classic controller with a turbo button, slo-mo switch, or programmable sequence a violation of rules? What about exploiting significant glitches? Surely there are videogame authors that welcome further enjoyment of their products through use of Game Genie, GameShark, or similar devices – does that make such hacking no longer cheating? It may be useful in discussing this aspect to think of the videogame as a product like a toy or car, rather than an abstract idealized experience, which the consumer is generally free to do anything legal with in the interest of satisfying their wants and curiosities. (b.) Though house rules, tournament rules, and user-determined meta-games (ex. agreeing to acknowledge a course though a level then racing camera-guided rockets through it in a deathmatch game) consist of breakable constitutive rules, these are not part of a videogame itself, but instead how the videogame fits into encapsulating non-digital games, such as round robin tournaments. In such cases the rules and victory conditions exist outside the videogame, and the videogame is utilized as a deciding piece in the game structure, like rolling very complicated dice.

4– (a.) Salen and Zimmerman, Rules of Play, summarizing Karen Sideman’s 2000 Game Design address: “In a typical board game it is necessary for at least one of the players to learn the rules and understand them fully before a game begins… with digital games it is possible… to learn the rules of the game as it is being played; to make the discovery of the way that the game operates part of the play of the game.” Note that similar study and peer explanation activities are possible for videogames, such as teaching friends hidden combo moves, or holding back on combat play while they figure out the basics – just not necessary as in the case of non-digital games. (b.) Primarily mechanical games such as Crossfire, Labyrinth, or Hungry Hungry Hippos, and especially enclosed examples such as pinball or crane claw (to prevent direct physical manipulation of the game parts without breaking the game), are in this regard more like videogames than they are like other non-digital board or skill games. However, as explained later above, all such physical games exist within the physical and representational system of reality, as opposed to the arbitrarily constructed conceptual approximation of physics, time, representation and input in videogames.

5– Chris Crawford, Art of Computer Game Design (free PDF): “Rule disputes and administrative foul-ups are part of the unavoidable dangers of boardgames. The computer… can administer the game, freeing the player to concentrate on playing it. This allows one other big advantage: the computer can implement complex arithmetic and logical rules. With other technologies, game rules must be overly simple because the humans implementing them cannot be trusted to perform simple numerical computations. The computer eliminates this restriction.”

6– Atari 2600 examples: the one-brick-per-top/bottom-trip in Breakout, the distance-from-center ball control in Pong, and the ability to curve “guided” tank shots in Combat. More complex examples: compare Street Fighter 2 to fighting, Super Mario Kart to kart driving, NBA Jam to basketball, or Quake III to gun battles.

7– (a.) Steve Swink in Game Feel, p. 201-228 goes into depth on the subtleties of Mario input and movement. 28 pages are devoted mostly to Mario’s movement – whatever Mario is doing, it clearly is not running and jumping, it is “running” and “jumping.” (b.) If we decrease abstraction to resolve this discrepancy, considering programmatic piecewise-functions as rules, we are then labeling as “rules” complex evaluations happening dozens of times per second which players will never explicitly learn, akin to calling air resistance on a passed ball a rule in a sport. Juul explores this distinction a bit more in Half-Real, with regard to FIFA 2002, though does not seem to treat as significant the degree of interpretive freedom in translating physical reality for sports into physical-seeming reality for videogame versions.

8– Imagine a 2D Sonic the Hedgehog videogame in which the character runs and jumps like Mario, instead of like Sonic, or if the cars in Grand Theft Auto San Andreas moved like karts in Super Mario Kart. The same rules might be documented about running and jumping, or accelerating and steering, though the gameplay would be dramatically different. Which is less like Super Mario Bros: a videogame in which Mario runs and jumps with the speed and air control of Sonic, or a videogame in which Mario’s controls/movements stay the same but the videogame is stripped of game elements such as score, level number, lives, or winning/losing, instead procedurally generating Mario level indefinitely as the player moves to scroll the screen, and restarting in a new random level after each death?

9– This is an idea I have been exploring and trying to articulate in several forms (see Are Complex Systems Learned by Interacting with Complex Systems? and the other entries it links to), but the gist is that because players cannot accidentally perform invalid interactions within videogames (again, on account of the permitted means being identical to the possible means), and because players are not burdened with manually handling the execution of videogame rules/logic, players can trial-and-error through situations or iteratively adapt/recondition habits to succeed without needing to understand the system they are up against. Despite playing Mike Tyson’s Punch-Out extensively and doing quite well in it, if it were up to me to identify which counter-punches deserve to be awarded with a Super-Uppercut Star (say, in a copy of the game with programming modified to award a new star only when a button on controller 2 is pressed, leaving me to demonstrate my understanding of the rule for earning a star), I would have only a very vague sense of when to do it, and by giving away more or fewer stars than the videogame is currently balanced to award I would be making the videogame less or more difficult than it was designed to be.

Learn and practice team game development with Gamkedo Club.
Membership worldwide. Professional support. Proven process.

Subscribe by e-mail to receive weekly updates with Gamkedo.Community interviews and YouTube training videos for game developers!

All contents Copyright ©2018 Chris DeLeon.

Site production by Ryan Burrell.