Rules, Referees and Reality


Is it desirable, or undersirable, to have referee moderation of (and intervention in) an imaginary or virtual game world? Is it the sign of a healthy game, or a sick one? Is there such a thing as a perfect set of rules that would never require referee moderation? Are computer games superior to referee-moderated games, due to their enforced consistency and lack of bias? The Lurking Gherkin ponders these questions....

The Referee Rules:

An argument that all too frequently arises between players and referees is that of the sanctity of the rules of the game versus the referee's own judgement.

Sometimes, it's a matter of interpretation – I've never seen a pen'n'paper RPG system that didn't involve rule ambiguity somewhere in its pages, and the subjective nature of words and the multiple meanings we give them makes this inevitable. RPG systems are not written in a mathematically formalised way but instead use plain (or sometimes not-so-plain) English.

Other times, it can be a case of the referee 'over-ruling the rules'. He or she decides that the rules of the game as written by its authors do not adequately deal with the situation that has arisen, and so they will make a judgement call. If this happens frequently in their game this can give birth to a 'house rule'.

Not all players can happily stomach this.

Not all players can happily stomach this. In particular, people who tend to play competitively want a stable, explicitly known and conservative (i.e. unchanging) rule framework that they can exploit and seek to maximise their advantage within. Particularly galling for them is where a referee's judgements are clearly attempts to maintain 'game balance'. They feel that they have worked hard to exploit the system to maximum advantage, and now the referee is robbing them of the fruits of their labours. (If you think about this a little you may see a parallel of this situation in the political socialism-vs-capitalism debate:.)

From the referee's viewpoint, however, their attempts to restore balance to the game may simply be compensating for rules that they feel to be an inadequate description of the reality of their campaign world. Now, if the referee goes so far as to punish the powergaming player for their behaviour, well, that is a different matter – and not a desirable state of affairs in my book! Personally, I feel that people who have worked hard to develop their character should see a proportionate and realistic return for their efforts - that is to say, realistic within the context of the referee's conceived campaign reality.

Now, I have mentioned here several times that there may be a disconnect between the rules of the system, and the referee's conception of reality. But is this always necessarily the case? Surely the ideal would be to find a consistent set of rules that would always give results that the referee could agree with and the players could master and exploit without fear of the referee pulling the rug from under their feet with an ad-hoc judgement. In such an 'ideal' world, the referee's imaginative input to the game would simply consist of setting the scene for the players; a simple algorithm that encapsulates the rules of the game could thenceforth adjudicate the outcome of any given situation. The algorithm could be the referee applying the rules (a 'liveware' or 'meatware' implementation) or it could be a computer program ('software' and/or 'hardware'), but in either case it would be based purely on application of strict rules of logic with no creative re-interpretation or over-ruling of the rules.

What hope is there that such a perfect rule system might exist?

Ye Cannae Change The Laws Of Physics!

To attempt to answer this question, I want to start by looking at another set of rules to describe a particular reality; the 'Laws' of Physics.

Now, the idea of physics as providing a set of 'laws' that nature is obliged to follow is so rooted in popular culture that few people ever question this idea. As Scotty frequently quips, "Ye cannae change the laws of Physics!"

But in fact, physical theories are simply our so-far imperfect attempt to provide a descriptive model of physical reality. To my knowledge, pretty much no scientist since the eighteenth century has been so presumptuous as to refer to his or her theory as a 'law'. Newton's Laws of Motion, to take one example, are just plain wrong – which is why Einstein had to develop the Theory of Relativity. But Einstein's theories do not provide a totally accurate and comprehensive picture either – they have not so far successfully been reconciled with the other great pillar of 20th century science, Quantum Mechanics. The 'laws' of physics can be described at best as a work in progress. To suggest that nature obeys the laws we have written is simply nonsense. They are attempts to predict what nature will do under particular circumstances, but they do not oblige nature to behave in a certain way! Furthermore, there is reason to believe that we will never have a complete set of laws that answer all our questions about the nature of reality – our descriptions will become better and better but there will always be room for improvement.

So, the 'laws' of physics don't cover every situation, and are simply an approximation. But reality still goes on regardless of the imperfection of our descriptions of it, and sometimes proves our description false, in just the same way that the competent referee's campaign does not grind to a halt when the rules prove inadequate for a particular situation, and just as – sometimes – the referee overrules the rules. (Whether this means that our physical reality also has a 'referee' arbitrating the outcomes of every interaction is, of course, a philosophical or theological question beyond the scope of this discussion!)

...what hope do we have to create such a complete set of rules for an imaginary reality?

If we can't even find a complete set of rules describing our own physical reality, what hope do we have to create such a complete set of rules for an imaginary reality?

Well, that rather depends on the scope and level of detail that we require this imaginary reality to have. To invoke our 'laws of physics' example again, Newtonian Mechanics is a 'good enough' description of most everyday physical interactions between objects such as cars or billiard balls or even swords and shields. It is not very good at describing the physics of very massive or fast-moving objects – for this we need to invoke Relativity which is broader in scope – and is not very good at describing the physics of interactions occurring on a very small scale– for this we need Quantum Mechanics which deals with a finer level of detail.

So, provided we don't want to ask awkward questions, Newtonian Mechanics is fine. And in a similar way, a game system that is a gross over-simplification of reality works OK as long as we don't take it places where it wasn't designed to go, and are willing to content ourselves with its limitations.

So, given that, ultimately, all rule systems are to some degree an over-simplification of the reality we are seeking to describe, how do we cope with situations that the rules were not designed for? The thing is, game systems don't even have the scope or detail level of Newtonian Mechanics – and they don't embody scientific theories or principles such as chemical interactions or optics or thermodynamics. They operate at a much more macroscopic level of detail, and deal with the interactions of a fairly limited range of entities – so it doesn't take long before imaginative people start trying to step over the boundaries of the system's scope or want more detail than the system provides.

Pole Position

Furthermore, such systems can sometimes exhibit 'pole behaviour'. This is a term I am borrowing from analogue circuit theory – electronic circuits produce a particular output when subjected to a particular input, and a well-behaved circuit should aim not to have any 'poles' in the range of outputs that are produced by the expected range of inputs to which they will be subjected. A 'pole' is a particular finite input that will result in an infinite output (in practice physical limitations would prevent a truly infinite output but the output voltage could be sufficiently high as to damage another circuit further down the line).

Now most role playing or simulation game systems are, despite their imperfections from a scientific viewpoint, nevertheless so complex and so varied in their possible inputs from players and referees that it is simply not practical for their designers to analytically identify all the poles of their system. This is one of the reasons that games need to be playtested before releasing them. Of course, a small ruleset, such as that used in a scenario-specific simulation boardgame, is much easier to troubleshoot, both through analysis and playtesting, than a multi-volume RPG.

One way in which many game systems deal with 'pole behaviour' is to have enforced limits or 'caps' on game metrics such as character attributes or skill/class levels or armour class / defense / attack bonus / etc. An ideal system wouldn't need these artificial restraints, and the frequency with which these restraints appear is one measure (in my opinion) of how 'good' a system is. First edition D&D rules regarding character attribute scores that were introduced in Deities and Demigods are a fine example of this sort of thing – the maximum score in any attribute was set at 25; even the gods did not go beyond this limit! This was to prevent characters from increasing their attributes ad infinitum, of course. But if the system handled attribute scores in a sensible way and if the magic system (always the wildcard in any fantasy game system!) were more carefully written, those caps wouldn't need to be there.

'Not Fair!' the powergamer in the corner grumbles, missing the point:

Where the rules don't have such limits applied to truncate an undesirable instance of 'pole behaviour', the referee needs to apply his or her judgement to fix the problem. ('Not Fair!' the powergamer in the corner grumbles, missing the point:). Another thing that a referee can do, where appropriate, is to employ alternative sets of rules to resolve results depending on the game situation – although commonplace game activities (e.g. combat in most RPGs) tend not to lend themselves well to such treatment.

What about software / hardware rules implementations? Well, these tend to be even more in need of these limiting 'rules patches' than a human referee-moderated system. What if a player in an MMOG discovers a game 'pole' which enables them to level up at ridiculous speed? If they're a sharing sort of player this information will be all over the 'net in a day and everyone will have 1,000th level alts stomping around before the game admins even realise what's happened. Capping at a certain level prevents this sort of thing from occurring.

I am sure that on-line multiplayer games nevertheless require human intervention from time to time to 'fix', even with safety net features such as this.

From Hero to Zero

As well as 'poles', analogue circuits exhibit things called 'zeroes'. This corresponds to a null response to a given input. The corresponding phenomenon in game terms would be where the rules simply don't cover a particular activity. In a game, this would manifest itself as a place you simply can't get to or an activity you can't perform, and this tends to be the biggest drawback of computer games. Can your avatar spit in an enemy's face? Can they father a child or become pregnant? No? Then those are system zeroes. Now this, even more than fixing undesirable 'poles', is where a human referee comes into their own in comparison with a software / hardware rules implementation. A human referee can take their players into those places that there are no rules for and moderate what happens using their imagination and real-world experience.

It may be pointed out that in a computer game there is nothing to stop you from imagining that you have done the above things. After all, in a pen'n'paper RPG, all you are doing is imagining. Why should a computer game that gives you a visual representation of your character prevent you from imagining that your character is doing things that the interface makes no allowance for? Could two players not pretend that their avatars have engaged in a sexual union and conceived a child?

This is a valid point to be making. However, the factor that I think tends to select against bilateral screen-plus-mind's-eye play of this nature is that the game reality is then spanned across multiple storage media – the on-line environment, and the player's own imagination and memory. This is kind of an awkward representation of game reality to maintain. If your character is visually represented on-screen in front of you, it's not easy I think to keep a modified image in your mind of how they 'really' look and act in their imagined pregnant state (for instance). It's much easier if the whole thing is either in your mind, or else entirely on the screen, but not half-and-half. I think also that complex audio-visual input such as that provided by a computer game tends to de-activate the imaginative faculties of the human mind. And difficult as it may be to picture your digital avatar as pregnant, what happens when they give birth? Can you create a baby avatar?

I am perfectly aware that most people don't log onto MMOGs to make babies, and so this is not perceived as a limitation of the system by the majority of players. I merely cite it as an example of a system zero in most computer games that can easily be handled by a human referee in a pen'n'paper RPG.

The Computer: Friend or Fiend?

So am I saying that referee-moderated games are better than computerised ones? Well, that all depends on what you want from the game, of course! At present, computers are dumb beasts compared to the human brain with its 100 billion neurons and their 100 trillion connections. And they cannot, as yet, do something as clever as decide to use different sets of rules based on an intuitive grasp of what best suits the current game situation. However they are very good at performing specialised operations at astonishing speed – for example, handling a melee combat between avatars in real-time!

In comparison a referee and group of players rolling dice and looking things up in tables and books are like a pack of tortoises on tranquilisers. The other day I ran a combat that lasted 8 melee rounds in game time which took around four hours of real time (admittedly this was a fairly complex engagement involving multiple melee combats, psionic combats, missile fire and magic use, taking place in a rather unusual environment). The question is, would it have been more enjoyable if it had only taken eight minutes of real time? With this slow-motion blow by blow expos̩ of the engagement everyone got to share in the drama of each other's struggle, and that in itself was part of the fun I think. It's simply a different sort of experience Рmore of a shared group experience and less of a personal experience. 'Me, me, me!' players who are impatient to get on with their own character's actions and who don't care about the detail of what the other characters are going through will probably gravitate towards the faster-paced computer format. Provided the inherent zeroes of the computerised system do not appear in the output response generated by their range of likely inputs, or they do not mind restricting their inputs to those that will not generate zeroes, this format will be more satisfying to them Рlack of flexibility is a worthwhile price to pay for the ability to do more of what they enjoy in less time than it takes in a pen'n'paper system. Although I think it will tend to cause their unsused imaginative faculty to atrophy somewhat Рcaged birds aren't the strongest of flyers.

Though I may seem to be making the case for Luddism, I am not opposed to the principle of computerised game worlds

Though I may seem to be making the case for Luddism, I am not opposed to the principle of computerised game worlds – although I think that, like the Emperor dressed in his new clothes, many people have blinded themselves to their present limitations, and do not acknowledge that the main reason they keep coming back for more is dependency and operant conditioning (similar to the psychological tricks employed by casino owners in Vegas). And why stop doing something if it feels good? Any unrepentant heroin addict will agree with this sentiment, having forgotten what life was like before they took that first shot in the arm.

OK, I am perhaps being a little extreme here! I don't actually think that a bit of MMOGing now and then is really so bad, but the more activities you find yourself giving up to make more time for online gaming in its present limited format, the more worried I think you should be, if you care about the wellbeing of your psyche.

Provided human civilisation doesn't receive a knock back in the impending future, computerised games will of course continue to grow and develop, and the range of what can be done in the online environment will gradually improve over time. There will eventually be much more diversity, especially when 'MMOG World Builder' software becomes available that enables any Tom, Dick and Harriet to create their own online game world with its own rules (and then just watch the intellectual property lawsuits fly!). Genetic evolutionary algorithms could develop game software automatically to meet a given set of requirements without human intervention. Human-computer interfaces will improve as our understanding of neurobiology increases. Artificial Intelligence will blur the distinction I have made here between human vs computer moderated games – a sufficiently savvy AI could tackle those poles and zeroes I mentioned, writing new campaign material and game rules on the fly. However, we will probably be approaching, or even have passed, a technological singularity by the time these later things happen which will cause such a cultural upheaval that it is impossible at this time to predict exactly how things will pan out. Also the existence of AI will raise all kinds of ethical concerns and there will be those opposed to even creating such a thing, much less employ it as a plaything in an online game world.

Reality you can rely on

Well, I started this rather lengthy ramble by talking about the nature of rules and campaign reality, and the referee's role as the final arbiter of that reality. I guess I should do some kind of summing up:

  • Referees should not be afraid to make their own rules up or make a judgement call that disagrees with the 'official' rules of a game, if they are doing so because they feel those rules are not a good depiction of reality in their game world. And players need to accept that this is part and parcel of the game (although if they feel the referee is being in some way inconsistent or biased, they should feel free to politely point this out!). It is not even 'wrong' for a referee to choose to switch between rule systems in order to find the most appropriate solution to a given game situation.
  • No-one has ever discovered a 'perfect rules system', and in fact there may be no such thing at all. Even the 'laws' of Physics aren't a perfect set of rules; reality is whatever happens, rules simply try to describe this in a formalised way, but what actually happens takes precedence over what the rules say should happen. And the final arbiter of what happens in a campaign world is always the referee. (In real life – who knows, for sure?). The key point is – Reality takes precedence over Rules, and it is healthier to regard game rules as an attempted description of, rather than an ultimate definition of, the game reality.
  • Computer game worlds are not in any sense 'more perfect' than human-moderated ones; they simply have different advantages and disadvantages. Key limitations are that in such worlds the rules DO define rather than describe reality, and lacking a referee these worlds offer no faculty to provide ongoing creative input in reaction to unusual statements of player intent – although the removal of a human referee may be seen as an advantage for those who regard a human referee as a biasing factor or unpredictable factor that reduces their ability to compete on a 'level playing field' in the limited range of activities that the game specialises in.
  • The main advantage of computer games is fast handling of a small range of specialised activities – although this may, by some, also be regarded as a disadvantage, as it reduces the level of detailed shared experience.
  • The present distinction between computer-moderated and human-moderated games will gradually blur, but the barrier will not truly collapse between them until certain new revolutionary technologies come into force. But by the time that happens, all bets are off anyhow!
  • Excellent article LG!

    Just a few comments before I rush off to work:

    "The other day I ran a combat that lasted 8 melee rounds in game time which took around four hours of real time (admittedly this was a fairly complex engagement involving multiple melee combats, psionic combats, missile fire and magic use, taking place in a rather unusual environment). The question is, would it have been more enjoyable if it had only taken eight minutes of real time?"

    No. It takes longer to describe a fight than to participate in one. However, you go on to describe a number of things that slowed your combat down that are not description/narration. You experienced frustration with managing different actions, speeds, and events. The mechanics of your game are slowing you down unnecessarily -- players taking turns breaks up the flow of time, cross-referencing tables makes the results harder to access, and the results of your dice rolls need to be translated back to real world events. Many people just accept this as a limitation of all tabletop RPG's. I don't. Here is a shameless plug: .

    "They feel that they have worked hard to exploit the system to maximum advantage, and now the referee is robbing them of the fruits of their labours."
    Excellent point -- speaks to implied objective. I was planning on writing an article on the topic of various game objectives. Defeating the rule-system, or using it to maximum advantage, was going to be one of my threads.

    "One way in which many game systems deal with 'pole behaviour' is to have enforced limits or 'caps' on game metrics such as character attributes or skill/class levels or armour class / defense / attack bonus / etc. An ideal system wouldn't need these artificial restraints, and the frequency with which these restraints appear is one measure (in my opinion) of how 'good' a system is."

    Bravo! (A tumultuous clapping and banging ensues) Skill progressions should not be capped, but normalized. The system can be open by identifying correctly the components of any measurement. Running fast is contingent on the mechanics of your gait (training, the muscle power that you have, and the length of your limbs. (There is a formulae for calculating the speed of extinct dinosaurs that compares the physical limb length to the body structure -- it has quite broad applicability -- especially for us gaming nerds). Bottom line is that two of the three components pull the characters skill back to the middle. With the other two factors the same your size proportion and speed proportion are identical. Game balance on advancement is created by identify the correct scale of each measurement.

    Lots of good stuff to comment on here. Must go now though. Thanks LG.

    A really interesting article, LG.
    However, you reminded me of my studies with the Zeros and Poles, so that's minus one point for you (it's a good thing you didn't bring up higher-order poles or....damn!) :-)

    In any case, I pondered if a combination refree/computer environment would be better at most things that either solution alone, but haven't managed to get around to checking this assumption (using NWN). Any ideas?

    - reading a signature is silly -

    Another plus for computers is that they are impartial and objective. Too often when i have a real life game master, deus ex machina occurs. In fact that's one of the reasons we have rules is keep each player and even the game master accountable.

    For example one starwars game my party got into a fight with some aliens and killed them. One of the players decided to be lil' psycho and take the head of an alien as a hostage. My game master considered this distasteful and arbitrarily ruled that the blood of the alien caused a skin disease in the psycho player. Did my gamemaster decide ahead of time that the blood reacts badly to human skin? Does it say that anywhere in the rules?

    What... did I have a point? Anyways great article.

    -Mr. Rogers is pissed-

    Computers usually are unbiased referees. I guess the day they get even an Artificial Intelligence, they will be as prone to biases as any other sentient being. Until then, GMs make better referees just because they can improvise. There are several ways to speed the game up, my own is simplifying the rules, especilally combat and magic rules. You can still make anything you wish and your choice matters. Throwing dices is kept at the minimum, however. The rules are just an approximation of reality anyway.

    "[Rules] keep each player and even the game master accountable."

    Accountable to what?

    Impartial referees work well in a game where there are two sides. They balance the rules and make an even playing field.

    What is the goal of a role-playing game? What is the victory condition? If the goal of an adventure is to have the characters succeed then why is the GM involved? The GM has nothing at stake in that victory condition. You object to overt manipulations -- probably with good cause.
    Perhaps you need to recognize that your GM doesn't want to play in a game where the characters are macabre and psychotic. Instead of talking about the goals of your adventures, it sounds like he lashed out in frustration.

    "[Rules] keep each player and even the game master accountable."

    What I mean is that the rules keep the game just a little bit realistic. The referee might think that his super powered squirrel of death might be able to beat my fighter. However, when the dice are thrown and the numbers are crunched that may not be the case. What if a munchkin player thinks that his power ranger, stealthy mysterious sniper, half dragon assassin can kill anything and anyone? If there weren't any rules he could just stroll through any encounter and emerge unscathed.

    In both of these examples we see imaginations run rampant. Both the player and the referee cannot be right. Either the squirrel of death wins or the assassin wins the encounter. And, that's where the rules come in. They set a standard baseline that all must be held accountable to. Numbers and dice can't lie. They can't fudge. They can't take sides.

    -Mr. Rogers is pissed-

    LG: Thought provoking article! Nice work.

    zipdrive: you might want to check out this:

    My attention was drawn to this recently by one of my players, which reminded me of what I'd said about 'system poles' in this article way back....