In 1981, the first issue of
Polyhedron, the RPGA's newsletter/magazine (edited by Frank Mentzer), included an interview with Gary Gygax where he shared his thoughts about some of the changes and additions that people were making to AD&D that he wasn't in favor of:
I don't mind creativity, I don't mind mutation, if it brings out better game play, and superior gaming in general. But from everything that I can see, all the changes that are made are usually foolish and meant to either baby players along or kill them off, one way or another. They're destructive. rather than creative.
Just think about some of the outstanding changes that were made in Dungeons & Dragons games, and Advanced Dungeons & Dragons games too, for that matter, and look at what their effects are. Consider the "double damage on a natural 20", which of course seldom went to the monsters, but only went to the players, therefore making it yet easier for the players to kill monsters.
Critical hits? Again, players never took critical hits, only monsters, for some reason, would take critical hits. The weapons expertise idea, that a player's chosen weapon ... he or she would do a lot better with it. And yet, monsters fighting with their natural abilities. fang and claw - who could be more expert than a tiger with its claws and teeth? -weren't getting any bonuses.
The spell point system, which allowed magic-users to become veritable machine guns of spells without ever having to seriously consider what they were going to take and just shoot everything down, made the magic-user the only character worth playing.
Some of the proposed classes, such as the barbarian I've heard of and the mighty knight, and one or two others that I've heard of, create super-powerful characters who just can ... again, it was the only one worth being. Then you just go through and beat up on everything The changes in the demi-human races create, again, super-powerful characters, so that everybody wants to be a dwarf, or an elf, or whatever it is, and nobody wants to be anything else, because it overbalances in favor. And generally these are done at the whim of a Dungeon Master, or from group pressure. to make a rather uninteresting campaign where everybody is one thing. These are usually the Monty Haul games.
On the other hand, you have the really silly monsters, or sure-death traps for the DM who seems to be rather sadistic and just wants to proceed to kill all of his players regularly, in capricious ways, without giving them any chance whatsoever. That's also guaranteed to spoil a game.
There's a lot to unpack here. In his infamous
Dragon #67 editorial that appeared in November 1982 ("Poker, Chess, and the AD&D Game") Gary drew a hard line and seemed to claim that anyone who added
any unofficial material to an AD&D game was no longer playing AD&D. Here he's taking a more moderate and sensible position - that adding to and modifying the game is okay in theory, he just thinks most of the changes being made aren't improvements and are making the game worse, not better. People familiar with
Unearthed Arcana, the Gygax-authored AD&D rules expansion released by TSR in 1985, likely feel some irony in this quote, since that book included many of the very things he's decrying above - weapon specialization, the barbarian and cavalier (knight) classes, and expansions to demi-humans - and critics of that book cite many of the same complaints about them that Gary does here (that they're too powerful and change the shape of the game too much). So what gives?
The answer is that this interview pre-dates the writing of the material in that book (which, as nearly as we can tell, took place sporadically between late 1981 and mid 1984), and a lot of what Gary was doing in that book was taking these concepts, which were obviously popular with players (hence their widespread inclusion as house-rules) and "doing them right," i.e. in a way that he felt fit within and added to the AD&D system. This becomes an interesting object-lesson in how Gary viewed the AD&D system - his sense of balance and priorities - and what he felt fit within it versus what he felt did not. Thus examining his additions sets an example that the rest of us can use for further expansion. It's illustrative to note, for example, that while Gary "rehabilitated" some of these ideas within the AD&D paradigm, there are others that he did not, presumably because he felt they were too incompatible with the game's core concepts - so there are no critical hit tables or spell-point system in
Unearthed Arcana.
I'd like to take a particular look at Gary's treatment of the barbarian class, both because it's always been one of the more controversial additions (even before the book was published - it was immediately controversial from its first draft appearance in
Dragon magazine) but is also one of my favorites.
The original edition of D&D included the "fighting man" class - a sort of professional soldier in the John Carter, Achilles, or d'Artagnan mold who could use any type of weapon or armor and proceeded over the course of play from being a mere "veteran" (just slightly better than an ordinary man-at-arms) to a "swordsman," "swashbuckler," "myrmidon," "champion," and ultimately a "lord" able to establish his own barony and collect taxes from the peasants. This is a very broad archetype, but it doesn't really capture the flavor of Conan (or his many, many literary imitations) - primitive barbarians who rely more on natural-born toughness and catlike reflexes than formal skill and training, who generally eschew heavy armor and fancy weaponry (preferring a trusty sword or axe), and have a special hatred for effete and corrupt wizards. I mean, you could play your D&D fighting man in that manner - declare that he will not wear armor and only use simple weapons and disdains magic - but doing so just creates an inferior character with a short lifespan. Underlining this disconnect, one of the illustrations in the original D&D set is of a long-haired, sword-wielding brute in a loincloth labeled "barbarian" - exactly the type of character the rules don't really support playing.
Given the popularity of these kinds of characters in fantasy literature, it's no surprise that players wanted characters like them, and nature abhors a vacuum so it wasn't long before fan-made barbarian classes began appearing. Two that I know of both appeared in print in 1977 - one version in issue #4 of the British D&D magazine
White Dwarf, another in David Hargrave's unauthorized collection of D&D house rules,
The Arduin Grimoire - and I'm sure there were many others besides these. These versions have some differences, and some elements in common - barbarians in both versions receive bonus hit points, have generally better saving throws (especially against Fear effects), have abilities to climb and hear noise like thieves, and can work themselves into a rage/frenzy to gain attack bonuses, in exchange for limitations of what kind of armor they'll wear and weapons they'll use, and how many languages they can learn. Interestingly, both versions require fewer XP to increase in level than standard fighters. The WD version has a couple of additional abilities - tracking like a ranger and catching missiles.
Another key development is that in
The Dragon #36 (April 1980) Gary Gygax, inspired by Lawrence Schick and Tom Moldvay's "Giants in the Earth" column of AD&D stats for literary characters, provided an extensive set of AD&D stats for Conan, including a large number of special Conan-only rules: Conan has an increased move rate, has a super-humanly high Dexterity score (and corresponding AC bonus), has a higher Charisma score that applies only to women, has way more proficient weapons than his fighter-level would allow, has higher number of attacks than his fighter-level would allow, is treated as having a magic sword (for purposes of hitting creatures only harmed by magic weapons) even though he "never willingly uses a magic sword," has thief levels as well as fighter levels, has a bonus to surprise and reduced chance to be surprised, heals from damage at an increased rate (even when not resting), has bonuses to his saving throws, has latent psionic abilities to detect magic and danger, and various other bonuses.
When Gary decided to do his own version of the barbarian class for AD&D, which appeared in
Dragon #63 (July 1982), he clearly looked primarily to his own treatment of Conan rather than to the existing "unofficial" barbarian classes that had been published. The class lines up pretty well with the stats he provided for Conan, with a few tweaks - instead of having superhuman Dex scores, they receive a higher-than-normal AC bonus for high Dex; instead of getting all thief abilities they get ad-hoc climbing and hiding abilities; instead of latent psionic abilities they get ad-hoc abilities to detect magic, illusions, and rear-attacks; their Charisma bonus applies to other barbarians rather than all females. They also get abilities to track like a ranger and to leap great distances. [The ability to hit creatures normally harmed only by magic weapons isn't included, but Gary added it as an addendum in
Dragon #65, claiming its omission was inadvertent. The same is done with the doubled healing rate in
Dragon #67.] The main drawbacks for the class are (1) they will never willingly/knowingly use any magic items,
ever; and (2) they have an extremely steep XP chart for advancement - 6,000 XP to 2nd level (compared to 2,000 for a standard fighter), and 500,000 XP per level above 8th (compared to 250,000 XP/level above 9th for a standard fighter).
Gary's version of the class was immediately controversial - garnering complaints from the
Dragon readership, mostly aimed at how seemingly-overpowered it was, despite the two heavy limitations. Gary devoted a significant chunk of the aforementioned "Poker, Chess, and the AD&D Game" ranticle in
Dragon #67 to responding to criticism of the barbarian (suggesting his frustration wasn't only with people making inferior additions to AD&D but also, apparently, to people refusing to accept
his additions - " All that is really being questioned is change, because this subclass
is different from others. Well, Gentle Players, that is what you’ve been asking for, and that is what I am here to do. Believe it or not, I actually know my game system and what or what will not work within its parameters!"). In addition to strongly arguing that the class is balanced (and, if anything, too weak - especially since long-term success depends on the character having multiple very high ability scores), he also added a couple of new details about the class: (1) that they can never be dual-classed - a character can neither switch to or from the barbarian to another class; and (2) barbarians do not require "training" in order to increase in level the way other AD&D characters do - rather, when they gain sufficient XP for the next level the increase is automatic.
After that, things were mostly quiet on the barbarian front for a while (though, notably, barbarian characters were included in both the
Dungeons & Dragons cartoon and LJN's line of
Advanced Dungeons & Dragons toys, both in 1983). However, with the publication of
Unearthed Arcana in 1985, the barbarian was back, and so was the controversy around them.
Apparently upon three years' further reflection Gary (likely in consultation with Frank Mentzer and/or Jeff Grubb, the two "developers" who were tasked with compiling
Unearthed Arcana from Gary's notes while he was busy trying to keep TSR afloat) decided that his original version of the barbarian was too weak after all, because the revised version doesn't tone down any of the things from the magazine version that readers had complained about, and instead gives it several new abilities. A couple of them are minor (staggering the saving throw bonus vs. spells from a flat +1 to starting at 0 and increasing up to +3 as the barbarian gains levels; giving high-level barbarians the ability to summon a "horde" of other barbarians who will follow them for a short period of time), and another one is a clever and fun roleplaying-oriented addition (giving barbarians who destroy magic items the XP award as if they'd kept it - so a barbarian who dumps out a
potion of growth earns 250 XP, one who snaps a
broom of flying receives 2,000 XP, one who smashes a
mirror of mental prowess gains 5,000 XP, etc.). However, appended to the end of the class description was a new table of abilities that has been the source of most of the vocal discontent with the class for the last 30+ years.
Right at the end of the class description is a table that enumerates the gradual restriction of both the prohibition on using magic items and the prohibition on associating with spellcasters (which in the magazine version had been a flat statement: "While a magic-user will be shunned by barbarians, clerical
spells are not regarded as magic (except for the more powerful
spells not typically usable by a tribal shaman or low-level cleric),
so barbarians will associate with clerics on occasion") as the barbarian increases in level - at 2nd level a barbarian will associate freely with clerics; at 3rd level they will use potions; at 4th level they will use magic weapons; at 6th level they will associate with magic-users "if necessary," etc. This was presumably added because Gary (and/or Frank and Jeff) felt the blanket prohibition was too strict and made the class too weak at higher levels, but I think it was a mistake and is problematic in several ways.
One is that it renders the ability to strike as a magical weapon - one of the class's most unique and flavorful abilities - completely redundant: barbarians gain the ability to strike as if they were a +1 weapon at 4th level, which is per the chart the same level where they gain the ability to use actual magical weapons. Yes, a barbarian isn't as dependent upon magic weapons and can still beat on a demon with his or her fists, but what feels like it should be one of the main distinguishing characteristics of the class has been reduced to a footnote ability that will rarely, if ever, come up in actual play.
Secondly, it undermines the forceful argument that Gary put forward in
Dragon #67 that the unique nature of the class - their huge number of special abilities - is justified in large part because, unlike other classes, the barbarian does not change to become more highly trained and sophisticated as they increase in level:
Playing
as a barbarian is offered to players as a
determined choice, not as one of several
possibilities — or a mere afterthought.
This is a part of the whole concept. Thus,
the level title for a barbarian never varies.
Such a character, properly role-played,
is bred, raised, grows, and dies a barbarian.
Barbarians do not need training to
go up levels, because they gain no sophistication.
They get tougher and more
wily.
Easing the restriction on use of magic items as they increase in levels runs contrary to this and weakens the concept of the class. As they gain levels they become less distinctive and archetypal, which is the opposite of how it's supposed to work.
Thirdly, quantifying exactly how and when the barbarian is willing to associate with spellcasters turns a roleplaying suggestion ("magic-users will be shunned") into a hard-and-fast rule - no association at all prior to 6th level, "if necessary" at 6th, "occasionally" at 8th level and higher - that creates problems at the table: unless the DM is willing to run two or more different groups of players, then a choice must be made - either the group can include a barbarian, or it can include magic-users, but not both. Anecdotal accounts suggest this led to both lots of unhappy players who weren't allowed by the rest of their group to play the kind of character they wanted, and also led to lots of players just ignoring this restriction or undermining it in a sort of nudge-nudge-wink-wink way (other characters distract the oafish barbarian while the elf fighter/magic-user casts his spells, etc.). In that same
Dragon #67 article, Gary had this to say about the restriction on association:
The barbarian
sees magic of two sorts — wizard magic
and god magic. The former is cast by
magic-users and their ilk — puling creatures
all. The latter sort of dweomer must
be tolerated, for who can argue with deities?
A brooch of shielding (hopefully a
rare find in any campaign) is so much
dross to a hard-nosed barbarian. He’ll
take the niggling damage from the
magic-user (that’s what his high hit
points are for) and then hew the cowardly
craven to pieces. Those magics which
allow saving throws are so much the better,
for the barbarian does have a better
chance to save against them. Those that
happen, happen. With everything that
the sub-class has, what real need is there
for magic items? Scarce and rare finds in
any well-run campaign, such wretched
stuff is not for true humans (barbarians)
in any event.
Magic performed by clerics, particularly
clerics who serve the deities of the
barbarian and his or her tribe, is another
matter. That sort of thing must be abided.
Who in a barbarian tribe would stoop to
using even the dweomer of deities? Why,
that’s simple: Men and women too old to
fight, weaklings, and those odd individuals
“touched” by some super-being. In
a life-and-death situation, any self-respecting
barbarian would allow a proper
servant of a known deity to do whatever
the deity directs through that servant. If
it goes against the barbarian grain, then
the offending cleric can be thereafter
shunned — whether out of embarrassment,
dislike, or fear is entirely open to
question. If absolutely necessary, such
spells can be tolerated for short periods
of time, but by choice any barbarian
must seek more direct solutions with
arms. Obviously, faced with a situation
which required the barbarian to perform
a given plan, and that action was impossible
without magic — possibly even wizard
magic — the intelligent barbarian
would be forced to stoop to such low
means to reach the end. Shunning
doesn’t mean the same as never associating
with: Look the word up. Again, it
doesn’t assert that barbarians will slay all
magic-users just because they reek of
noisome magic, nor does it state that
clerics casting spells above 2nd or 3rd
level will be done to death by the outraged
barbarian. Low-level spells are
merely the power of a shaman/cleric
given by some deity — not even god
magic. Higher-level spells of a clerical
nature are disliked by barbarians, and
they will not voluntarily be around those
who make a practice of employing magic.
Circumstances, as usual, alter cases.
Remember the spirit of the rules, instead
of trying to find the letter by reading between
the lines.
This is a roleplaying guideline - the barbarian dislikes magic and spellcasting, and considers those who use it weaklings and cowards. A self-respecting barbarian doesn't need the aid of spells or enchanted items (due to their wide array of special abilities) and won't willingly employ them unless they have no other choice - and even then they will resist and resent it. They don't trust spellcasters, look down on them as weaklings and cowards, and will never seek their aid. That's pretty straightforward and easy to understand, and easy to portray at the table. There's no need for two separate parties - though if the DM is willing the barbarian is particularly well-suited to solo adventuring, and back in the 80s we had such a side-game that was a lot of fun - the barbarian just sneers and grumbles at spell-casting and acts superior, and as long as they can keep him or her from destroying the magical treasure the other players are probably fine with the barbarian's hard-line stance since it means more magical treasure and aid for them!
With all of these considerations, I really like Gary's version of the barbarian class, and encourage all AD&D DMs who don't allow it (and over the years I've encountered
way more DMs who don't allow the class than who do) to give it another look. I think it's really fun to play, works really well in the "traditional D&D paradigm" of dungeon and wilderness-crawling (unlike the cavalier and acrobat, which both work best in more urban and civilized environments), and is a great addition.
Looking at how the class fits into the game, how it relates to and was balanced against the other classes, shows how Gary's
situational, instinctive, and descriptive approach to rules-making worked in action. The barbarian stretches the rules-framework of AD&D in a way that created a lot of resistance among the more traditionally-minded , but when everything is taken together in a gestalt it
works - it fits right into the game and expands and enriches it without changing it. The class is undeniably very front-loaded, and at first level will likely overshadow most other characters, but the steep XP chart means that unless the barbarian is doing a lot of solo adventuring on the side they will quickly fall behind the other characters, especially once "name level" is achieved. This makes them the opposite of magic-users, who start out extremely weak but eventually become the most powerful class. I like that symmetry, and the way it causes players with different characters to approach the game differently and employ different strategies.
I do have a couple of house rules for barbarians in my own games: as mentioned above, I revert to the magazine version's blanket restriction on magic item use, I allow barbarians to gain XP from magic items
only from destroying them, not from selling them (closing a probably-unintentional loophole in the rules), and I also severely restrict their starting funds, to 5-30 (5d6) g.p., which helps balance them at first level (since they won't be able to afford armor or good weapons). Note that because barbarians don't have to pay for training they tend to acquire an enormous amount of gold during play, so I've also toyed with (but never formalized before now) some sort of ad-hoc "treasure attrition" rule for barbarians - that for each game-week of inactivity 0-90% of their cash-equivalent treasure simply, unaccountably,
disappears. Presumably it was spent on ale and wenches, swindled away by unscrupulous merchants, or otherwise lost. This models the way that almost every Conan story began with him broke, no matter how much treasure he acquired at the end of the last one, and also prevents the situation of the barbarian becoming a de facto banker to the other PCs, which is totally out of character for the class.
