REPOST: The Articles of Dragon: "How to Finish Fights Faster"

Along with falling damage, psionics, and alignment, articles about unarmed combat were a commonplace in the pages of Dragon during the years when I subscribed to that venerable gaming magazine. There's probably a reason for that: unarmed combat in AD&D was, in my experience, pretty much universally admitted to be unusable as written, a fact even Gary Gygax acknowledged on more than one occasion. Despite that, no single alternative system ever really took root, with most referees employing a welter of different approaches, some based on the official system, some based on earlier articles from Dragon, and some created whole cloth. That's what playing D&D was like during my formative years in the hobby -- a crazy mix of stuff all drawing inspiration from the same base and then running off in whatever direction one deemed most fun. Consequently, I can't help but chuckle at all the folks decrying the existence of "so many retro-clones," since, to my way of thinking, what we have now is pretty much what we've always had. The only difference is that, nowadays, it's easy to print up, prettify, and sell your interpretation of D&D to others, whereas, in the past, each referee had a photocopies and stapled collection of house rules he shared with anyone willing to listen.

Perhaps because no single alternative to AD&D's execrable rules emerged, it was inevitable that the redoubtable Roger E. Moore would eventually offer his own unarmed combat system. His article, "How to Finish Fights Faster," appeared in issue #83 (March 1984) and takes up only four pages, one of them being a humorous illustration of four rotund halflings attempting to bring down an eyepatch-wearing humanoid, who looks more annoyed than inconvenienced by his diminutive opponents. Moore divides unarmed combat up into three modes: pummeling, kicking, and grappling. Pummeling is straight up fisticuffs, with or without the use of aids, like dagger pommels or metal gauntlets. Kicking is, well, kicking and grappling is attacking to subdue. All three modes are fairly simple to use, working more or less like the normal AD&D combat system but with certain modifiers and special cases unique to them. This is particularly true of grappling, which has a number of different moves detailed, each of which has further modifiers and effects.

I never used Moore's system, so I can't comment on how well it plays in practice. I suspect it probably works better than AD&D's official system, but not as well as others. I say that, because it includes a lot of specificity in certain areas (grappling, for example) that necessitates either a good memory or referring to the article to adjudicate. That's not a bad thing in itself; there are lots of rules in D&D that require reference to a rulebook to handle. However, I'll admit that I find it baffling that unarmed combat rules so often wind up being much more complicated than armed combat. Why is it that we can accept that all it takes to adjudicate an armored fighting man's attack against an opponent is a 1D20 roll compared to a chart, followed by a damage roll if successful but we demand saving throws and percentage chances and so forth if he wants to throw a punch or wrestle someone to the ground?

Rules, Rules, and More Rules

I'm currently refereeing three different campaigns at the moment: House of Worms, using Empire of the Petal Throne; Barrett's Raiders, using the Free League edition of Twilight: 2000; and Dolmenwood, using the rules of the same name. Of the three, only two – EPT and Dolmenwood – can be called "old school" in the usual sense of the term, though T2K has a lot in common with many old school games, specifically its focus on hexcrawling and resource management. That said, I wouldn't really call Twilight: 2000 "old school" without some big caveats. That's no knock against it, since my players and I have been enjoying ourselves with it for more than three years now, but I think it's important to note these things, particularly in light of the topic of this post.

Empire of the Petal Throne is a very early RPG. Released in 1975, it's a close cousin to OD&D in terms of rules, meaning that it's not very mechanically complex. Dolmenwood is a little bit heftier, being largely derived from Moldvay/Cook Dungeons & Dragons (1981), itself a clarification and expansion of OD&D. Twilight: 2000 (2021) uses a variation of Free League's "Year Zero" rules, versions of which can be found in most of the company's games, like Forbidden Lands or Vaesen. The T2K variant is a bit more complex than the others, owing to its inclusion of modern firearms and vehicles.

In each campaign, I rarely use the game's rules as written. I don't mean that I've introduced lots of house rules (though I have in a few cases). I mean that I often ignore the rules. When playing, I often don't want to slow down the flow of the session by having to refer to a rulebook or a chart. Instead, I prefer to rely on my memory and that of the players, which means that we're more likely to strictly apply those that we remember than those we don't. I call these kinds of rules "sticky" rules, because they stick in your memory. 

One of the reasons I prefer old school RPGs like D&D is that I find their rules much stickier than those of newer games. To some extent, that's simply a function of familiarity. I've been playing D&D and Traveller for more than four decades; I know them almost like the back of my hand. I lack this familiarity with games I learned more recently. On the other hand, there's no question that most older roleplaying games are much more mechanically simple than those that came later. Again, this is a generalization and there are plenty of counterexamples. My point is that, as both a referee and a player, I'm much more comfortable with fewer and simpler rules, since I'm much more likely to remember and, therefore, use them.

But, as I already noted, even in games like EPT or Dolmenwood, I regularly handwave or outright ignore rules in the heat of play. For example, Empire of the Petal Throne includes spell failure rules. Depending on a character's level, psychic ability, and the type of spell, there's a chance a spell might not function. At mid to higher levels, this chance is minute, but there's still a chance of failure. Despite this, I don't always make the players roll, since there are many occasions when I feel it unnecessary or disruptive to the flow of the action. I defer to my own judgment here rather than the rules and the players have never complained. Were they to do so, I wouldn't hesitate to use the rules as written, but I like to think that, after a decade of play, we've built up enough trust that that no players worries much about how I'd adjudicate in-game situations.

I think about this question a lot, because many aspects of the new Twilight: 2000 rules, chiefly the combat system, are more complex than I like. There's nothing wrong with them and, by many measures, they're much simpler than the original GDW T2K combat rules. However, I'm not fond of them and I frequently dispense with many persnickety aspects of them in the interests of speed and simplicity. Again, the players rare complain about this and accept my judgments. Had I the ability to start this campaign over again, I might have opted for simpler, more straightforward rules, but, after more than three years, it's too late for that, so we muddle through. 

That's more or less where I am with rules these days: when give the option, I prefer simple, even simplistic, rules over more elaborate and complex ones. I'm not opposed to trying to model complicated situations and activities mechanically and, under the right circumstances, could even find that enjoyable. However, as a referee running a weekly game over the course of many years, I have come to find that rules I can't keep in my head without recourse to a book or a chart or a table don't hold a lot of appeal for me anymore. Consequently, my latest drafts of the rules for Secrets of sha-Arthan are decidedly much simpler than earlier ones. It's yet another way that my experiences as a referee have colored my own design work – and for the better, I hope.

Modifications to Traveller (Part II)

Wiseman continues: 

The most common change will probably be in the available weapons. Any additions to the weapons should be especially well thought out and rationally based. Science fiction literature contains many more weapons than could be described in the basic rules. Many of these have only the flimsiest of scientific justifications, if they are justified at all. In television and movies, weapons are often created for the visual effect they have and are not usually very well thought.

A couple of points. First, we again see a reference to "science fiction literature" as the first source of inspiration for Traveller, which shouldn't be surprising. Anyone with even a passing knowledge of E.C. Tubb and H. Beam Piper – the two most influential authors on Traveller – can easily see where the game came from. That said, sci-fi TV and movies are mentioned, albeit unflatteringly. Second, it's very clear that Wiseman at least considered Traveller to be an attempt at a "serious" game grounded in real world science. He doesn't see it as a game of swashbuckling space opera. I'm OK with that, though I know plenty of people aren't.

In addition, all ramifications of the weapon must be considered. For example, any major change in lethality of hand-carried weapons is likely to have profound effects on military tactics, at least eventually (the military is sometimes a little slow to catch on).

This is a fair point. For the most part, Traveller's weapons – at least those readily available to player characters – are all on a fairly similar plane in terms of lethality. There are differences, obviously, but, aside from some of the stuff introduced in Mercenary, they're all of a piece. 

A few question to ask might be: is this weapon really required by my universe? What is the principle of its operation? (If the weapon is taken from literature, the principle of operation may be described or well-established.) Is the principle of operation a reasonable one? What are the advantages and disadvantages of this particular weapon, and why would it be used in place of more conventional weapon systems?

All reasonable questions, so I don't have much to add. I will note, however, that a common complaint about Traveller is that it includes too many "archaic" weapons among its firearms, like shotguns and submachine guns. This has never much bothered me, perhaps because of a comment in Book 1 of Traveller that seemingly addresses this point: "the surest way to injure or kill an adversary will be to subject him to a large dose of kinetic energy, and a simple way, easy way to administer that energy is by bullet impact."

Adaptations to the jump and maneuver drive systems and to our concept of jump space should be made only with care and consideration. The effects of modifications to these areas are wide-ranging and touch almost every rule in one form or another. In designing our interstellar drive system, we tried to come up with one which fitted in with the other aspects of our proposed mythos and which was reasonably close to the literature we drew it from. It was assumed, however, that some referees would not be satisfied, whatever system we devised. It was further assumed that any referee who felt strongly enough about the matter would change what displeased him.

This section might be one of the more controversial ones, because I know that many people both dislike Traveller's jump drive and believe that the game is perfectly playable without it. Wiseman acknowledges this above. For myself, I'm a big fan of the way jump drive works, because, when coupled with the lack of any form of FTL communications, it creates a setting that's somewhat akin to the Age of Sail, at least as far as the dissemination of information. That opens up a lot of possibilities for adventures both large and small. 

We do not recommend major revisions of the trade and commerce section without considerable thought. Minor deviations from the trade and speculation table, or reclassification of certain world types are acceptable; indeed, in most cases, some deviation from the letter of these rules is desirable, but the spirit (i.e. the general system) should be preserved.

As to how integral the speculative trade system is to Traveller, I'm not sure. I'll only say that I very much love the system, not because it's a good simulator of interstellar commerce – it's not – but because it provides a terrific campaign framework and a reason for characters to travel from world to world. I'll talk more about this in a future post.

Referees are cautioned against making psionic powers too common. Players will often urge that psionics become more widespread, but this is often only because they want to have every possible advantage without drawbacks. Powerful psionic abilities would soon dominate all aspects of play and most of the excitement of Traveller (the advanced technologies) will lapse into disuse, or at least be relegated to a secondary role. 

I can't really disagree with this section, but I've also never been a huge booster of psionics in Traveller (aside from when I've used the Zhodani), so my opinion is perspective is perhaps skewed.

The cautions which apply to major revisions apply to a lesser extent to expansions of existing rules. Many referees find certain rules do not go into enough detail to suit them. Examine the manner in which Book 4, Mercenary, expands upon the army and marines, or Book 5, High Guard, expands upon the navy for general guidelines on technique. You need not (and probably should not) duplicate the systems in these two books exactly, but they will prove a source of inspiration for your own efforts.

Even as someone with a decidedly mixed opinion about Traveller Books after 3, I agree with Wiseman here. Most of the later books do provide good models for those who wish to add more detail to their adventures and campaigns, even if I've rarely felt the need to do so myself. I prefer a simpler, less baroque version of Traveller, but I understand not everyone feels this way, especially in a game that's much more laser focused on a particular style of play.

Wiseman offers up a lot of food for thought in this section of Book 0. In future posts, I'll be returning to a few of them. Stay tuned.

Modifications to Traveller (Part I)

Deluxe Traveller, a boxed version of the game released by GDW in 1981, included Book 0: An Introduction to Traveller written primarily by Loren Wiseman. Book 0 was intended to be "a guide to the world of Traveller, written especially with the beginner in mind." It's written in a straightforward, conversational way, eschewing the dry, technical manual style that's common in RPG rulebooks, in order to reach an audience who might have no experience with this kind of game. Consequently, it provides some insight into how GDW saw Traveller as a game and what they felt important to convey to newcomers to it.

There's a section in Book 0 titled "Modifications" that focuses, more or less, on house rules. Though it's not quite two pages long, it nevertheless contains material worth pondering. The section begins:

In the process of playing scenarios, or while preparing to adapt a specific science fiction literary mythos to Traveller, many referees will wish to change the rules to a greater or lesser degree, either to expand some aspect which is not adequately covered in Traveller, or to modify some section which does not fit in with the referee's universe.

There's a couple of things here worthy of note. To start, there's the reference to "a specific science fiction literary mythos." I've been saying for years that Traveller is almost entirely inspired by SF books, particularly those written between 1945 and 1975, and that televisual and cinematic sci-fi had minimal impact on its design. It's always good to see further confirmation of this fact. That said, Wiseman couches the idea of modifying the rules first in the context of adapting the game to prior art and then in the context of a homebrew universe. 

Referees should feel free to modify any rule to whatever extent they see fit, providing they bear in mind that:

— The rules are interlinked to a great extent. If you change one section, you must also be willing to change all other sections, which are then rendered inconsistent. Naturally, radical departures from the rules will have greater and more far-reaching effects than minor changes. Referees who modify the rules without regard to the repercussions are doing themselves and their players a grave disservice.

Traveller's rules are very elegant in my opinion. Marc Miller put much more thought into their design than did Arneson and Gygax when creating Dungeons & Dragons. Consequently, Traveller is not as easily modified or house ruled. It's not impossible by any means, but, as Wiseman notes here, doing so without reflection can have ripple effects.

— The balance of play should not be destroyed. A common change many newer players make (particularly those with backgrounds in fantasy role-playing) is to increase the occurrence of psionic talents, and permit training on a much higher level than the basic rules. This may seem like fun at first, but when any character can kill with a single burst of mental energy, where is the challenge? Psionic talents are best used sparingly.

Wiseman's reference to "newer players" and "those with backgrounds in fantasy role-playing" are intriguing. I wonder if GDW was seeing an expansion in Traveller's player base by 1981. If so, that would certainly explain some of the reason behind the publication of Book 0.

— All of the changes should be rational, logical, and scientifically sound (after all, Traveller is a science fiction role-playing game). A typical example is a suggestion we receive about three times a year for some form of anti-matter small arm (usually a pistol or rifle, but once a hand grenade). These suggestions always seriously underestimate the amount of energy necessary to maintain a magnetic bottle around the anti-matter for any length of time, and almost always have a maximum range of considerably less than the burst radius of the projectile. 

I love hearing little anecdotes like this one! When I spent time with Loren Wiseman at GenCon back in 2001, he shared many stories about dealing with fans of Traveller over the years. They weren't all like this one, but many were. 

— The speed of communication should never be allowed to exceed the speed of travel. This is a basic tenet of Traveller, and its violation will irrevocably alter the balance of the rules.

This one surprised me, not because I disagree with it – quite the opposite – but because, up until now, Wiseman had implied that any aspect of the game could be changed, provided the referee is willing to deal with the consequences. Here, though, he is emphatic: the speed of communication should never be allowed to exceed the speed of travel. This is a topic to which I'll return at greater length in a future post (or, more likely, posts), but I'm glad it was mentioned here.

— Do not expect other Traveller materials to match your universe if you engage in large-scale modifications.

This is, of course, the eternal danger of house rules. On the other hand, referees who make extensive use of house rules generally aren't using many official supplements and rules expansions, at least in my experience, so it's not a huge problem. 

(To be continued ...)

Rules of the House

I want to begin this post with a brief anecdote. 

The other day, a friend of mine with whom I've gamed for many years, confessed to something he considered "embarrassing." What was the source of his shame? He recently tallied up the pages in the house rules document for a game he was refereeing and discovered, to his surprise, that it exceeded sixty pages. The horror! I was quick to reassure him that this was no cause for unease. Indeed, quite the opposite, since these rules were, in part, the accumulation of years of thought and play – tweaks to make the game "as [he] would like it to be," in the phrase from the afterword to OD&D. 

My friend's reluctance to admit that he very thoroughly house ruled the game he was refereeing is not unique. Over the years, I've encountered quite a few people who have felt that the addition of house rules was somehow "wrong" and that the only "correct" way to play a RPG was exactly as its rules were written. I'm not quite sure of the origin of this mentality. If I had to guess, I'd lay the blame on the third edition of Dungeons & Dragons, part of whose culture is obsessed with "RAW" – "rules as written" – a term that I don't believe I ever heard prior to 3e. I don't want to lay the blame entirely on this edition; Gygax complained about rules lawyers all the way back in 1974, so the mentality has been a constant in the hobby, even if its prevalence seems to have increased in recent years.

I don't think I've ever played a single roleplaying game exactly as it was written. In some cases, this was due to ignorance or misreading of the rules. In other cases, it's because of the existence of a body of folklore that has supplanted the actual rules to the point that "everybody knows" this rule or that should be understood in a particular way (often contrary to the designer's intent). Yet, there are also examples of rules changes that are deliberate and the result of careful thought and consideration, added or subtracted to make the game run more like what the referee and players desire. 

Like rules lawyers, this is a venerable tradition, predating the existence of OD&D itself and arising out of the traditions of miniatures wargaming. Arguably, the entire history of RPGs is, to borrow from Benjamin Jowett, a series of house rules to D&D, sometimes explicitly so. Neither Gary Gygax nor Dave Arneson played OD&D precisely as written. From a certain perspective, AD&D might even be called an elaborate set of OD&D house rules, with the game expanded and honed into something Gary found more congenial to the style of play he preferred. There's nothing wrong with house rules and I must confess to be baffled that anyone might seriously think there were.

That said, I do feel that, before one introduces house rules into one's campaign, it's important that one reads, understands, and tries to use the rules as they are written. "Ivory tower" house ruling, which is to say, pre-emptive house ruling before play has actual begun no longer sits well with me, even though (or perhaps because I have engaged in it in the past). I am now much more firmly of the opinion that house rules should arise from play and reflect dissatisfaction with how the actual rules work in a given campaign. In this context, I find house rules not merely laudable but in fact inevitable. Any referee running a long campaign will, I think, introduce house rules into play simply because no RPG ever written is so perfectly written that its rules work for every possible configuration of players.

Again, I'd like to stress that I generally don't think house rules should occur in a vacuum. They should be reflective of play, not to mention reflective of thoughtful engagement with the rules they're supposed to modify or replace. Taking the time to read and understand the rules of the game you're playing seems to me to be a prerequisite to the creation of any house rule. Perhaps I am odd but that strikes me as commonsensical, after the fashion of G.K. Chesterton's metaphor of the fence: "don't tear it down until you know why it was put up in the first place." I think that's sound advice and have tried to put into practice in my own campaigns.

There is no shame in house rules, because they are almost unavoidable if you play a game long enough. That's exactly as it should be, which is why OD&D ends with an exhortation to make the game as you would like it to be. Fight on!

Free Parking

The other day, I was having a conversation about board games and the question arose, "Does anyone actually like playing Monopoly?" It's an engaging question, because, despite the widespread belief that Monopoly is a poorly designed and, more importantly, unenjoyable game, it's also, I believe, the best selling board game of all time. Ask almost anyone, even confirmed haters, if they have a copy of it and there's a very good chance that they do. And, with the proliferation of so many downright bizarre themed versions of the game, it's not at all unlikely that many people have multiple copies of the game. I think I own three copies myself, though, in my defense, all of them have been exiled to the garage, alongside old paint brushes and the bottle of coyote urine I use to keep raccoons out of my recycling bin.

Leaving aside the more specific question of why nearly everyone owns at least one copy of the game, I sometimes advance a theory about the broader question of why most people seem to hate it – two theories, actually. The first and perhaps less contentious theory is that Monopoly is no different than Candyland or Mouse Trap – to cite just two more terribly unenjoyable games – in that it was designed for children. Children's board games have different design goals than games geared at an adult audience. To expect that a childish entertainment should be able to hold the attention of adults is both absurd and unfair (and likely explains why so much of contemporary pop culture is awful, but that's a topic for another day). On this theory, most of the people who vociferously hate Monopoly are adults or older children, who are not the intended audience of the game.

My second theory is a bit more contentious and even I will admit that it's not without its flaws. I contend that the reason Monopoly is unfun is because almost no one plays it correctly. Most of us never bother to read the rules of the game, because "everybody knows" how to play it. You need only do a very quick search online to find many articles pointing out which aspects of the game's actual rules people either misunderstand or don't use at all. All these various house rules and folkloric interpretations contribute to making the game a tedious, tiresome slog that no one in his right mind would call entertaining. It's my feeling that, if the game were played as intended, it wouldn't be nearly so abhorrent. To be fair, I've never tested this theory – and, honestly, don't have much enthusiasm for doing so – so I could well be completely mistaken. However, I don't think it's unreasonable to suggest that, if game is widely played in ways that are contrary to its actual rules, one cannot then blame those rules for being terrible or poorly designed. That's not to say that Monopoly is, in fact, a good game, only that, to make that claim, would have to play the game as written and vanishingly few people do.

Naturally, all of the foregoing is a prelude to several issues near and dear to my heart, namely, the nexus of game design, house rules, and "everybody knows" interpretations of the rules as written. Squabbling about these matters is a time honored tradition of the hobby, one that's been going on since its inception (and, arguably, contributed to its birth in the first place). I don't see these squabbles as fruitless and in fact believe they're a good source of insight into both game design and what we, as individuals, want out of a game. 

They're also an entry into explorations of the "culture" that inevitably grows up around any communal human activity. Over the years, I've played RPGs with many people, some of whose early experiences are similar to my own and others whose experiences are not. For instance, some, like me, were inducted into the hobby by a mentor, while others taught themselves to play simply by reading a rulebook. That seemingly small difference has far-reaching consequences and I find those fascinating. Then there are regional differences in style and content, as well as temporal ones and probably many more besides, all of which explain why it is that gamers don't always understand one another, despite ostensibly playing the same games (I find myself reminded of the apocryphal Shaw quote about Americans and Englishmen being divided by a common language).

Venomous Thoughts

While working on translating my Dwimmermount notes into something others can read and understand, I was reminded some weirdness regarding poison in OD&D. In Volume 1 of the LBBs, in the section describing how saving throws work, it's stated that

Scoring the total indicated above (or scoring higher) means the weapon has no effect (death ray, polymorph, paralization, stone, or spell) or one-half effect (poison scoring one-half of the total possible hit damage and dragon breath scoring one-half its full damage).

It's a peculiar section, because it implies that poison deals damage rather than simply being save or die, as longstanding tradition would have it. Likewise, there's the 4th-level clerical spell neutralize poison, whose description in the same volume states that it is

A spell to counter the harmful effects of poison. Note that it will not aid a character killed by poison, however. It will affect only one object. Duration: 1 turn.

That, too, is pretty peculiar, because, once again, it implies that poison deals damage rather than instant death. However, no poisonous creature in the LBBs seem to deal damage with their venom that I can find and even the fact that they deal death is more a matter of interpretation.

In AD&D, a 2nd-level clerical spell is introduced, slow poison. Its description notes that it can restore "even .. a supposedly dead individual" back to life, if they died from poisoning within a number of turns less than or equal to the level of the cleric who cast the spell. That description suggests that "real" death from poison occurs not immediately upon failing a saving throw but some time afterward. AD&D's description of neutralize poison is roughly consonant with its OD&D predecessor, in that it doesn't seem to affect someone who's been killed by poison (though, presumably, it could be used in conjunction with slow poison), but, rather, could be used to make poisonous food edible or even de-venom poisonous creatures. Interestingly, the 1981 Expert Rules seem to combine the effects of AD&D's slow poison with OD&D's neutralize poison.

This spell will cancel the effects of poison and revive a poisoned character if cast within ten rounds. It can also be cast on a poison or poisoned item to make it harmless. It acts only on poison present at the time it is cast.

It's a bit of a muddle, if you try to go with just the bare text, unless, as is quite likely, I'm missing some crucial passage somewhere in the LBBs that clarifies it all. That's why, back when I was preparing to start the Dwimmermount campaign, I briefly contemplated having poison deal damage rather than death. The problem I had was trying to figure out just how much damage it ought to deal and whether that damage ought to be in addition to the damage dealt by the attack that delivers it. That is, if a giant spider deals 2-8 points of damage, does that include the poison damage or is it in addition to it? I never came to a satisfactory conclusion and, the more I thought about it, the more questions it raised, so I simply decided to ignore the implications of those passages in the LBBs and go with the "traditional" understanding that all poison is deadly poison.

I'm still not sure that was the "right" decision, but it's a decision that worked -- a few dead henchmen and hirelings to the contrary.

The Articles of Dragon: "How to Finish Fights Faster"

Along with falling damage, psionics, and alignment, articles about unarmed combat were a commonplace in the pages of Dragon during the years when I subscribed to that venerable gaming magazine. There's probably a reason for that: unarmed combat in AD&D was, in my experience, pretty much universally admitted to be unusable as written, a fact even Gary Gygax acknowledged on more than one occasion. Despite that, no single alternative system ever really took root, with most referees employing a welter of different approaches, some based on the official system, some based on earlier articles from Dragon, and some created whole cloth. That's what playing D&D was like during my formative years in the hobby -- a crazy mix of stuff all drawing inspiration from the same base and then running off in whatever direction one deemed most fun. Consequently, I can't help but chuckle at all the folks decrying the existence of "so many retro-clones," since, to my way of thinking, what we have now is pretty much what we've always had. The only difference is that, nowadays, it's easy to print up, prettify, and sell your interpretation of D&D to others, whereas, in the past, each referee had a photocopies and stapled collection of house rules he shared with anyone willing to listen.

Perhaps because no single alternative to AD&D's execrable rules emerged, it was inevitable that the redoubtable Roger E. Moore would eventually offer his own unarmed combat system. His article, "How to Finish Fights Faster," appeared in issue #83 (March 1984) and takes up only four pages, one of them being a humorous illustration of four rotund halflings attempting to bring down an eyepatch-wearing humanoid, who looks more annoyed than inconvenienced by his diminutive opponents. Moore divides unarmed combat up into three modes: pummeling, kicking, and grappling. Pummeling is straight up fisticuffs, with or without the use of aids, like dagger pommels or metal gauntlets. Kicking is, well, kicking and grappling is attacking to subdue. All three modes are fairly simple to use, working more or less like the normal AD&D combat system but with certain modifiers and special cases unique to them. This is particularly true of grappling, which has a number of different moves detailed, each of which has further modifiers and effects.

I never used Moore's system, so I can't comment on how well it plays in practice. I suspect it probably works better than AD&D's official system, but not as well as others. I say that, because it includes a lot of specificity in certain areas (grappling, for example) that necessitates either a good memory or referring to the article to adjudicate. That's not a bad thing in itself; there are lots of rules in D&D that require reference to a rulebook to handle. However, I'll admit that I find it baffling that unarmed combat rules so often wind up being much more complicated than armed combat. Why is it that we can accept that all it takes to adjudicate an armored fighting man's attack against an opponent is a 1D20 roll compared to a chart, followed by a damage roll if successful but we demand saving throws and percentage chances and so forth if he wants to throw a punch or wrestle someone to the ground?

Purely Off the Top of My Head ...

In reading all these comments about the relative ineffectiveness of shields in D&D, I was reminded of the fact that, when Dexterity adjustments to armor class were first introduced in Greyhawk, they were limited to fighting men. I've long liked that limitation, since it provides a small but useful boost to fighters. So, I thought: what if, instead of Dexterity, characters could add their Strength bonus to their AC when using a shield?

Obviously, in such a case, I'd eliminate the Dex-based bonus to AC entirely and replace it with a Strength bonus when a shield is used. Is this idea at all helpful, even as an idea seed, or is it just completely without merit? Feel free to say so if it is, because I only just came up with it after a few seconds thoughts, so odds are that it is useless.

House Rules

I've been refereeing the Dwimmermount campaign for a little over two years now, but planning for it began several months beforehand. Part of my planning included giving some thought about what house rules I wanted to make to the LBBs, the results of which can be seen here. Looking back on that post, I feel a small amount of embarrassment, because I can see that I fell into a trap that I've often cautioned against in other posts of mine, namely, treating OD&D's rules as something distinct from their actually being used in play. Of the rules modifications I listed in that entry, most of them have never been used or, if they were, they were quickly abandoned. I use the alignment system (though it has minimal mechanical effects), the hit dice re-rolling, liquid courage was used briefly at 1st and 2nd levels, and scrolls (also less used now than they once were). The others have all come up in play from time to time, but, more often than not, we simply forget them.

What this suggests to me is that, like so much else, house rules are usually better if they're a consequence of having to deal with a genuine problem that crops up in play. I've noticed that lots of gamers will claim that rules X or feature Y of OD&D "doesn't work" or is "broken" and sometimes they're right. But "brokenness" is situational and what one groups finds problematic, others will not. Likewise, many referees -- I know this, because I am one of them -- become obsessed with "fixing" this or that or introducing some mechanic that they think is neat, only to find that the fix is intrusive or that his players couldn't care less about the mechanic (such as my early attempt to introduce a weapon vs. AC modifier table à la Supplement I).

My point is that, unless one has extensively played a particular system with the same group of players over time, creating house rules in advance of actual play is jumping the gun. The most durable -- and useful -- house rules I've created for the Dwimmermount campaign are those that were created, often on the spot, to address a problem that arose at the game table. The most forgettable -- and useless -- "house rules" I've created are those I came up to scratch some intellectual itch of my own. Consequently, I've largely stopped tinkering with OD&D's rules in the abstract, which is why I so rarely make game mechanical posts to the blog anymore. The ones you do see are things I either used in actual play or that I created in response to an immediate (or at least imminent) need in the campaign. I won't say this is the only approach, but it's definitely my preferred one these days.

Spell Complexity

As most of you almost certainly know already, spells in Chainmail are rated according to their complexity. There are six levels of complexity, which not at all coincidentally maps on to the six levels of magic-user spells in LBB-only OD&D. An optional rule pertaining to complexity compares the complexity rating to to power of the magic-user casting it (there are, in descending order of power, wizards, sorcerers, warlocks, magicians, and seers). The more powerful the magic-user, the more likely his spells will take immediate effect, while less potent casters have a greater chance for their spells to be delayed until the next turn or to be negated entirely.

This is a rule I find intriguing and one I've often contemplated adding into my OD&D game, but I've never done so, both because I'm not sure how I'd implement it and I'm uncertain the effect it'd have on gameplay. I'm familiar with the original Chainmail rules, as well as adaptations of the spell complexity system in Spellcraft & Swordplay and Brendan Falconer's article from issue 2 of Knockspell, so I do have resources to draw upon should I ever go ahead with the idea. However, I do worry about the impact it'd have on spellcasting in an OD&D campaign. Clearly, it'd make things more unpredictable, which I like, but it might also make magic-users more ineffective as well, unless one adopts the interpretation that only negated spells vanish from the magic-user's memory. In that case, it might be a fair trade-off, I don't know.

Does anyone have experience with using spell complexity on a long-term basis?

"We Always Ignored That Rule"

It will come as a surprise to no one that I have some pet peeves when it comes to discussing the history of RPG rules, particularly the rules of Dungeons & Dragons. One of the biggest ones usually manifests itself like this: I'll start talking about a certain rule, say weapon vs. armor class adjustment, and I'll express some bafflement as to how the rule was supposed to work or why it was presented in the fashion it was. In reply, my interlocutor will say simply, "We always ignored that rule," as if that answers my questions.

Now, I expect that, in any given RPG, there are large numbers of rules that get ignored, for a wide variety of reasons, from ignorance to laziness to outright disagreement about the utility of the rule in question. I don't have any problem with this fact, speaking as someone who's ignored more than his fair share of RPG rules over the years. My beef is more with the notion that, because a given rule is easily ignorable and likely was ignored by a lot of gamers that it's not worth discussing or trying to figure out why it was included in the game and what the designer hoped to achieve by doing so.

A related peeve of mine pertains to the so-called "Rule Zero." I have no clue where this term first originated, but I suspect it's a fairly recent coinage. According to popular usage, it refers to the supreme authority of the referee to change anything he wishes in his campaign, including the game rules. Now, I don't have any particular problem with Rule Zero as such; I question why it's necessary to codify it at all, let alone in such a goofy way (I am unhappily reminded of Asimov's "Zeroth" Law of Robotics). What bugs me about it is the way that it too is often used to short circuit discussions of odd rules in RPGs. Say I want to talk about the rather low demihuman level limits in OD&D and what its implications for the implied setting of the game and my interlocutor replies, "Well, the LBBs were only intended as a framework to be changed as the referee desires, so I don't think you can draw any conclusions from the way the rules were written." I think it's possible to admit the first clause of that reply without conceding to the second; otherwise, one wonders why any rules were needed at all.

I enjoy delving into the whys and wherefores of game rules. I hold the possibly ridiculous notion that rules aren't included in a game "just because." They're all there for a reason, even -- perhaps especially -- bad rules. Understanding why they're there is not only interesting in its own right, but it also provides firmer ground on which to modify or reject existing rules. That's why I can get worked up when I hear, "We always ignored that rule" or "The referee can change it, so why worry about it?" Call it a weakness of mine, but there it is.

Open Friday: Psionics Use

As some correctly guessed, I wanted to ask about psionics this Friday.

This is a question for people who actually liked and used psionics in their D&D campaigns, not those who didn't use them, didn't like them, or thought they had no place in a fantasy setting: did you use the psionics rules as written (either in Supplement III or the PHB) or did you modify/house rule them or use a variant of them from some other source?

The psionics rules are notoriously difficult to get a handle on, after all, and, when I used them back in the day, I know I wasn't using them "as intended." I'm wondering how common my experience was and it's that that I'd like to hear about today.

Making Babies

As most of you probably know, dwarves in my Dwimmermount campaign are a male-only race. Even if they were immortal, which they're not, accidents and violence would ensure that their numbers would inevitably dwindle if there were not some way to create more dwarves. That way is through the carving of a "son" from living rock, with embellishments and adornments of precious metals and gems. With enough attention and craftsmanship -- outsiders say obsession -- a weird magic takes hold of the carving and imbues it with the spark of life, becoming a new dwarf.

In my campaign, Vladimir is in the process of creating a son for himself and I decided we needed rules to cover how this would work. Basically, a dwarf gets one chance per level to imbue his son with life. His player rolls 1D20 and adds the character's current level as a bonus. If the final result is 20 or more, the carving comes to life. This roll requires a minimum 10,000 gp investment in materials, however, with an additional +1 bonus to the carving roll for each additional 10,000 gp invested. Consequently, most low-level dwarves will not be able to make a carving roll, unless they're very lucky and/or diligent when it comes to seeking out loot. The investment bonus carries over from level to level, so a dwarf who invests 10,000 gp at level 5 and then invests 20,000 gp at level 6 gets a +3 bonus the next time he makes a carving roll (in addition to the +6 he gets from being 6th level).

There are some wrinkles to this system, of course. A roll of 1 is not just a failure but a critical one, resulting in something unusual occurring, according to the following table:

1D6 Roll	Result
1-2	Gnome
3-4	Inert
5-6	Knocker

Gnomes are an offshoot race of dwarves, created when something goes awry in the process of carving a dwarf son. Gnomes and dwarves have a complicated relationship, something I'll discuss in a couple of days in a separate entry. A knocker is a twisted dwarf, one so obsessed with gold and gems that it retreats into mines, dungeons, and other such subterranean locales in order to sate its lust for riches. Knockers are selfish, malevolent creatures with a penchant for trap-making and cruel practical jokes on other beings, most especially dwarves. A result of "inert" means that the carving is "stillborn" and unusable. The dwarf must start work over again on a new son. Inert dwarf sons are never discarded or destroyed. Instead, they are given a place of honor somewhere safe, often within a dwarf hold, which typically has a large "cemetery" filled with sons whose carvings failed to bring them to life.

A newly-carved dwarf is level 1 and has the same ability scores as his father. There is a 10% chance per ability score that it might be higher (or lower) than that of his father, owing to some quirk in the process of his carving. Roll 1D6, with even numbers conferring a +1 bonus to the ability score and odd numbers a -1 penalty. The new dwarf also owes his father an amount equal to the cost spent in carving him. Each month, there is a 75% chance the father will receive 1d12 x 100 gp from his son in partial repayment of this debt. The other 25% of the time, he receives nothing. This money cannot be converted into experience points, while the cost of carving a son can. (Remember that, in my campaign, only gold spent earns XP)

A dwarf PC, by the way, begins play in debt as well, to the tune of 1D10 x 10,000 gp. He is not under any obligation (other than a social one -- dwarves who don't attempt to repay their fathers are ostracized by others of their kind) to make good on this debt, although any money spent in doing so earns XP at the usual rate. Paying off a debt, however, does have one mechanical benefit. Usually, dwarves are limited to level 8 in advancement. For every 25% of their debt they pay off, they may advance an additional level, up to a maximum of 12.

Retrospective: Claw Law & Arms Law

Along with last week's retrospective, another product -- well, set of products -- I remember seeing relentlessly advertised in the pages of Dragon was Claw Law & Arms Law, one of several rulebooks that would, in time, join together to create the game system known as Rolemaster (as, as my friends and I called it "Rulemaster"). Back in 1982, though, (when the boxed set pictured here was released) products like Claw Law & Arms Law were initially sold as add-ons to other RPGs. Certainly, one could buy all the various Law books (such as Spell Law and Character Law) and combine them to play a wholly different game (or simply buy the boxed Rolemaster set that included them all), but there was no way I was going to do that. My friends and I were still deeply devoted to D&D and there was no way we were going to forsake it for another fantasy RPG. We'd house rule the heck out of D&D, of course, but, in our minds, that was somehow different and so it was that we decided to take the plunge and add Claw Law & Arms Law into our campaigns.

You must remember that 1982 was the tail end of the Golden Age. The fantastic realism that carried the day in the early part of the Silver Age was becoming an undeniable force in gaming culture, with lots of our older contemporaries dabbling in games they'd earlier told us were "for weirdos," like RuneQuest and Chivalry & Sorcery. Why? Partly because those other games had more "realistic" combat systems. That's where Claw Law & Arms Law came in: they were designed to make your RPG's combat system more realistic -- and deadly. That greatly appealed to us, as we'd already had solid experience with critical hit tables and, while their results were decidedly mixed, we nevertheless continued to see the appeal in making D&D's combats bloodier.

The big problem was that, for all their advertisements as add-ons to other games, they worked very poorly with AD&D. Like Rolemaster itself, Claw Law & Arms Law used percentiles for its combat system. To use it with Dungeons & Dragons, you weren't merely modifying the existing combat rules; you were completely replacing them. That caught us off-guard and probably ensured that we'd never adopt the rules on a permanent basis. To our way of thinking, it was perfectly fine to add or subtract to the existing combat rules, but to replace them entirely was a different thing altogether. At the same time, we wanted to see what all he fuss was about and so we decided to test out Claw Law & Arms Law.

The result not pretty, not because the new combat system was difficult to use; it wasn't. Indeed, despite our own mocking of Rolemaster by calling it Rulemaster, the game isn't particularly rule-heavy but it is chart-heavy. The new combat rules slowed down play, because we had to keep consulting charts. I am sure that this was because of our inexperience with the system. Indeed, I know it was, as I'll explain shortly. Charts catch a lot of undeserved flak in certain quarters, but my experience is that they're often better at presenting complex rules than are formulae, unless the formulae are very simple. Once one becomes familiar with which charts are needed and when, speed is increased considerably.

But the simple truth of the matter is that, for D&D, the addition of Claw Law & Arms Law just never felt right. We were far too accustomed to the existing combat rules (including weapons vs. AC modifiers, but not speed factors) to change midstream. The additional detail these add-ons provided simply didn't justify the cerebral rewiring necessary to make it all run smoothly. Yet, there was still something about these rules that we did like, which is why, when Middle Earth Roleplaying was released in 1984, we happily played it. MERP felt nothing like Tolkien's novels in my opinion, but the rules presentation was cleaner and more compact than in Rolemaster. We enjoyed playing the game as an alternative to D&D rather than as a replacement for it. That seemed to be the best way to use Rolemaster and its derivatives and we had a lot of fun doing so.

Magic-Users of Dwimmermount

This morning, I was inspired and finally put the finishing touches on the character class write-ups for Dwimmermount. Here's the chart for magic-users, to give you an idea of how I'm presenting them.

As I always intended, I've included level titles, both because I think they're something that's sadly absent from all the retro-clones to date (though I understand why) and because I think their presence can, if done properly, add a lot of flavor to a campaign setting without the need for adding any mechanics. In this case, the titles are Thulian in origin (or, rather, Termaxian-influenced Thulian) and correspond to degrees within the state-sponsored schools of magic. So influential were these schools that, even after the fall of the Thulians and their Termaxian masters, the degrees remain common terms by which to refer to magic-users of varying power.

MAGIC-USER LEVEL PROGRESSION
Experience	Level	Title	Hit Dice	Combat Bonus
0	1	Neophyte	1	+0
2,501	2	Zealot	1+1	+0
5,001	3	Beholder	2	+0
10,001	4	Practitioner	2+1	+1
20,001	5	Philosopher	3	+1
40,001	6	Minor Adept	3+1	+1
80,001	7	Major Adept	4	+1
160,001	8	Exalted Adept	5	+2
310,001	9	Magister	6+1	+2
460,001	10	Mage	7	+2
610,001	11	Archmage	8+1	+3
760,001	12		8+2	+3
910,001	13		8+3	+4
1,060,001	14		8+4	+5
1,210,001	15		9+1	+5
1,360,001	16		9+2	+6
1,510,001	17		9+3	+6
1,660,001	18		10+1	+6
1,810,001	19		10+2	+7
1,960,001	20		10+3	+7

Chartless Combat

A frequent complaint about old school Dungeons & Dragons is the need for a matrix to tell you what your character needs to hit an opponent in combat. This fact is also often used as a justification for the shift to an ascending armor class system, like that introduced in D&D III, which obviates the necessity of such a chart. While it's true that old school D&D has traditionally employed a chart to present "to hit" numbers, that's not the only way the information could be presented.

In my own Dwimmermount campaign, I don't use a combat chart at all when refereeing. Instead, when a monster attacks, I roll a D20, and add the monster's Hit Dice and the target's (descending) armor class to the result of the dice roll. If the sum is 20 or more, the attack is successful. This system is simple and quick and I don't need to consult any charts.

The players write down their attack numbers on the character sheets, so they don't regularly use a chart either. Of course, they could use a similar system to what I use for monsters if they wanted. They'd just have to change the addition of Hit Dice to the addition of a class-based "combat bonus" that boiled the charts down to a single +x to hit. For example:

Fighting Man
Level	Combat Bonus
1-2	+0
3	+1
4	+2
5	+3
6	+4
7-8	+5
9	+6
10-11	+7
12+	+8

Example: A 4th-level fighting man wearing plate mail (AC 3) is facing off against an orc champion wearing chain mail and carrying a shield (AC 4). His player rolls 1D20 and scores a "2." To this the player adds his character's combat bonus (+2) and the orc's armor class (4), resulting in a total of 8. Since that doesn't equal 20 or more, the fighting man misses. The orc (who has 2 Hit Dice) attempts to attack in return. The referee rolls "15." To this he adds the orc's Hit Dice and the fighting man's armor class (3). The result is exactly 20, meaning that the orc's attack succeeds.

Going Lightly Armored

Supplement I introduces the idea of giving Fighting Men a bonus to their Armor Class based on their Dexterity. While I appreciate Gygax's feeling that he needed to "beef up" Fighters, I have been considering ways to make going lightly armored more attractive to all classes in OD&D. One option I am considering is allowing a Dexterity bonus to Armor Class only to characters wearing leather or cloth armor. Another option is allowing Fighting Men so armored -- and only Fighting Men -- the chance to deal double damage from behind if they surprise their opponent. This would give a tangible benefit to a "sneaky" Fighter but at the cost of heavier armor. I have discovered that, in OD&D, being able to wear chain and, especially, plate armor is a huge boon to survivability, so foregoing that in exchange for the chance to deal more damage under the right circumstances seems a fair trade.

REVIEW: Original Edition Delta: Book of Spells

One of the most enjoyable aspects of the old school renaissance is seeing the large number of house rules and variant interpretations of OD&D that are used in individual campaigns. From my perspective, each campaign should be as unique as the referee who runs it and the players who participate in it. Some of the best old school blogs include not only listings of house rules but also discussions of the rationales behind them. I find material of this sort endlessly fascinating, since it's a latter day example of the kind of idiosyncratic creativity that was commonplace in the early days of the hobby. It also helps put paid to the notion that the old school movement supposedly marches in lockstep according to one rose-colored tune.

One of my favorite blogs is Delta's D&D Hotspot, written by Daniel Collins. Dan's been posting his thoughts about D&D online for a long time -- far longer than most -- and he's been playing OD&D even longer. I consider him one of the more thoughtful and experienced commentators on OD&D out there and, even when I disagree with his opinions, such as giving saving throws for any spell that has potentially negative consequences, I'm always interested in his explanation of why he holds that opinion. His musings on his blog and elsewhere have often caused me to re-evaluate my own thoughts, which is a remarkable feat, given how hidebound I can be. In short, Daniel Collins is someone worth listening to.

On his blog, Dan's been discussing his version of OD&D, which he calls "Original Edition Delta" or "OED" for short. OED nicely boils OD&D down to its essentials, clarifying some rules without eliminating the need for referee judgment calls and rationalizing other rules without eliminating the charming quirkiness of the original game. Again, I don't agree with all the choices he's made, but OED is nevertheless a version of OD&D I'd enjoy playing and I've pilfered a few ideas from it in my own Dwimmermount game.

So, when Dan released Original Edition Delta: Book of Spells, I was intrigued. What did he have in store? As it turns out, Book of Spells is an 18-page product (available in both print and PDF formats) that presents, as fully open content, all the magic-user spells of 6th level and below presented in the little brown books, with select additions from Supplement I. Dan did this for a couple of reasons. First, it fills the void left by the removal of legal OD&D PDFs by Wizards of the Coast earlier this year. Second, it gives players of magic-users (and referees) a "spellbook" they can easily consult during play rather than having to flip through an entire rulebook for just a few specific pages.

What's most remarkable about the spell descriptions is that, with very exceptions, they're no more than three lines of text long, including information on range and duration. Most of the descriptions eliminate "negative" statements and limitations, on the notion that OD&D magic should be rare and potent and that the primary determinants of what it can and cannot do are player ingenuity and referee adjudication. This results in a very bare bones presentation, but one that is strangely inspiring to those of us who see OD&D not so much as a complete game in itself as an invitation to create one's own game from its piecemeal rules.

Book of Spells is an admittedly specific book and I can't say it's a "must-have," particularly for those who already own the LBBs + Supplements. However, it's nevertheless a very fascinating take on OD&D magic-user spells, one that strips away the limitations of later editions, concessions to convention play, and other needless worries that have emasculated D&D magic over the last 30+ years. Reading Book of Spells is like reading a recently-unearthed ancient text, one free from later glosses so that it can be read with new eyes.

Original Edition Delta: Book of Spells consists almost entirely of dense text, with only four illustrations (counting the cover) to break it up. It is, however, very readable and well edited. It's also a bit expensive considering its length but that's hardly a damning critique of what is a useful and inspiring product. I don't expect it to set the old school gaming world on fire, but I do think it could occasion quite a lot of fruitful discussion about the power and utility of magic in OD&D, even among those who are already playing the game. That's a pretty remarkable feat in my opinion.

Presentation: 6 out of 10
Creativity: 7 out of 10
Utility: 6 out of 10

Buy This If: You're looking for a concise presentation of OD&D magic-user spells free from later accretions.
Don't Buy This If: You prefer more expansive (and restrictive) spell descriptions.

Level Limits

In pre-Greyhawk OD&D, dwarves were limited to 6th level as fighting men, halflings were limited to 4th as the same, and elves could progress no higher than 4th level as fighters and 8th level as magic-users. Supplement I loosened things up a bit if a demihuman character had exceptional ability scores. Dwarves with 18 Strength, for example, could reach 8th level as fighters, while Elves with 18 Intelligence could reach 9th level as magic-users. In general, though, the trend was to limit demihumans to 9th level and below (with thieves being a notable exception to this rule).

It's my belief that level limits were intended to serve two purposes. First, since demihumans got a number of abilities that humans lacked (infravision, improved saving throws, etc.), level limits were an early attempt at "balancing" their advantages against humans' lack of same. Now, as I think I've made clear over many months, I'm not a big fan of the balance über Alles school of game design and find it almost always makes a RPG less fun rather than more fun. Moreover, OD&D already possesses a means of balancing the varying abilities of characters -- variable XP charts. If one believes that the abilities of elves, dwarves, and halflings are good enough that they ought to exact some kind of "penalty," why not use XP as a means of representing this?

This is the approach that Moldvay Basic opted for in its particular interpretation of race-as-class. Elves and dwarves -- though not halflings -- require more XP to gain a particular level than do their human counterparts. It's a decent solution, I think, although the treatment of halflings undermines the logic behind it somewhat. Halflings get a number of advantages, most notably excellent saves, and yet their XP charts are identical to those of human fighters. Their main drawback is that their advancement is capped at 8th level, the lowest level limit of all the demihuman races. The halfling thus throws a wrench in the notion that there was any kind of consistent philosophy behind the demihuman XP charts.

The second purpose behind level limits, I believe, was genre emulation. Most of the pulp fantasies that influenced Gygax and Arneson didn't include lots -- or any -- non-humans as major protagonists. Non-humans were rare, exotic beings, far more likely to be talked about but never seen. The idea of a largely demihuman adventuring party (Tolkien's works to the contrary) was likely seem as peculiar. By making demihumans more limited in their progression, I suspect it was hoped that they'd prove less attractive and thus less numerous. The limit of halflings to 4th level, for example, has always struck me as the authors' way of saying, "Well, sure, you could play one of those hobbit guys, if you really want, but they're not really cut out for adventuring."

As I've played OD&D more extensively, I've come to some conclusions about all of this. Firstly, I like race-as-class a great deal and see it as a good way not only to keep non-humans strongly archetypal but also as a way to more implement the variable XP idea that Moldvay didn't seem to have followed through with. Secondly, I don't think level limits are a good way to discourage people from playing elves or dwarves or whatever. If the referee wants to limit the number of demihumans in his campaign, then he should simply do that: limit them. That's what I've done in my home Dwimmermount campaign. Dordagdonar, for example, is the only elf the characters have ever encountered. Even in a large city like Adamas, elves are exceedingly rare and he often raises eyebrows when he meets people who've heard of the existence of elves but never had the occasion to actually meet one.

Of course, Dordagdonar's uniqueness cuts both ways. Elves in my game are generally reclusive and stay out of worldly affairs. For whatever reason, Dordagdonar is different; he breaks a lot of the stereotypes about elven behavior, spending all his time with "ephemerals," as he does. Consequently, I don't plan to limit his level. Like his human companions, he can continue to progress indefinitely, although I do use a different XP chart to represent his abilities. Mind you, in OD&D, level advancement is slow and I doubt the characters in the campaign will ever see much beyond 9th or 10th level anyway, so it's largely a moot point.

In the end, I can't say level limits do a lot for me. I don't hate them or think them an abomination against "good" game design, but I also don't see their elimination as necessarily problematic. Far worse in my opinion is the ubiquity of demihumans, which destroys a lot of the flavor I associate with OD&D's literary influences. Having a party consisting largely of, say, elves bugs me more than the possibility that the single elf allowed in a campaign might one day reach higher than 8th level. But then I'm weird that way.