Thursday, July 13, 2017

[D&D] Being cavalier about it

Seemingly everybody but me hates the AD&D cavalier class, introduced by Gary Gygax in Dragon #72 (April 1983) and later canonized in Unearthed Arcana. Some of it might be subconscious revulsion at Eric, the smug-but-cowardly cavalier character from the D&D cartoon, or resentment at a class that's basically built around the idea of getting all kinds of extra, unfair advantages by virtue of being born rich. There are also some valid complaints that the class as-described with its strict behavioral code doesn't fit in within "traditional" D&D paradigm where a bunch of rootless "murder hobo" adventurers crawl through dungeons and the wilderness employing stealth, ambushes, trickery, and bribery in order to garner loot (and do a lot of running away from danger to save their skins).

That's true, but it was also deliberate. Gary was intending with the cavalier (and similarly with the acrobat, and the other classes he mentioned in Dragon magazine but was never able to detail, like the mountebank and jester) to expand the scope of the game beyond the dungeon- and wilderness-crawling of OD&D, to include more in-town activities and interaction with the civilized world. Most D&D characters exist apart from, or at best on the fringes of, "normal" society, but cavaliers are required to be a part of it. They're not independent free agents the way other character are.

Some people don't like that, and want their characters to be free agents, so it's natural that this class, with its explicit social status and obligations, won't be appealing to them. That means that adding cavaliers (and the other "town-oriented" classes) makes it more incumbent that everyone be on the same page up-front regarding what the campaign is going to be about so that there aren't incompatible characters or resentful players who expected one thing and got something else. But that's part of the larger trend in how the game was growing and changing in the 80s (and the unfortunate coincidence that the growth was piecemeal and not explained (and ultimately incomplete) so that a lot of people didn't understand that it was happening and spent a lot of time trying to shove square pegs into round holes) of which the cavalier and acrobat are a symptom, not the cause.

I mostly do like the cavalier class and think that it fills an archetypal niche that is distinct from the soldierly "fighting man" just like the barbarian does, and - when the role is understood - that it works fine in the AD&D game, especially in a campaign where the players have a stable of several characters that they rotate between depending on the nature of the adventure.

The various advantages given to the class, including things that seem like they should (or at least could) also have been given to other classes like special parrying and ability score training and multiple weapon specializations and the ability to keep functioning with negative hit points, make sense, given the core conceit that unlike common men-at-arms (i.e. fighter characters), cavaliers have undergone intense, specialized, high-quality (and very expensive!) training and conditioning since birth. This is perhaps made more clear in the Dragon version, which includes some text that was rephrased or edited out of UA:
[T]he cavalier character must be of the correct social class, i.e. gentle or noble birth, or of the accepted aristocracy for candidacy to knighthood. This requirement usually means that the character must be of a knightly, noble, or royal family which has suitable financial means to support the training necessary for entrance to the class of cavalier.
Landless aristocrats (knights or nobles) are typically precluded from having a child immediately enter the cavalier class at 1st level, since they are unable to afford the training and equipment needed. Such families (as well as lesser families being particularly honored) might, however, be allowed to have a child candidate enter the cavalier class as a 0 level horseman retainer of a knight. 
Cavaliers in AD&D are, by definition, either rich kids or kids who've been adopted and sponsored by rich patrons. They are explicitly the privileged "1%" of AD&D characters, and that gives them advantages that other characters of humbler means simply didn't have access to. "It's not fair that cavalier characters get to improve their ability scores through training and other characters don't." Exactly - it's not fair at all. That's the point. The whole concept of the cavalier class is completely unfair and non-egalitarian, by deliberate design.

And, at least in a properly-managed campaign, those benefits don't come without a steep price. As noted above, cavaliers - at least at low to mid levels - are not free-agent adventurers in the way other characters are. They have strict social obligations that must be obeyed and maintained. Again, this is made clearer in Dragon #72 than in UA:
As stated above, service is the paramount requirement for assumption of cavalier status. This service can be to a deity, state, order, or any master, particularly one of high station. After attaining knighthood [ed. note: at 4th level - this isn't explicitly stated in the text of either Dragon #72 or UA, but it's strongly implied by the level titles and supported in context], the cavalier can renounce former service, of course. At such point, the cavalier then champions a creed or cause, or is simply a rogue. In all cases, social status is likewise of paramount importance, and this must be maintained.
Note also/especially that "if the order or liege lord of the cavalier demands it" the character can be forced out of the class, becoming a fighter and losing all cavalier benefits (except weapon of choice). That's something that I suspect gets overlooked and ignored a lot, but it seems very important to me as it's unique among all character classes - paladins, rangers, and monks lose class abilities if they violate their alignment, and clerics are subject to judgment by their deities, but the cavalier alone has their class abilities subject to the whims of an earthly patron (typically an NPC, but at least theoretically another PC) - if you don't do what your boss says to their satisfaction, at their discretion, you lose your class abilities.

The cavalier's required body of retainers also worth paying attention to. Starting at 4th level the cavalier must acquire at least one lower-level cavalier retainer, and must increase the size of their retinue as they increase in level, to a total of 6 (3 lower-level trainee cavaliers and 3 0-level servants) by 8th level. The cavalier "is responsible for the actions of his or her followers and retainers, and is required to insure that others of the cavalier class live up to the standards of the class," and is also required to always travel with this substantial retinue until they achieve 9th level, unless their master orders them to travel solo. At first glance this may look like a benefit, but it's actually a substantial burden that characters of other classes don't face. Cavalier characters not only always have to answer to a boss above, they also are responsible for the well-being and behavior of a group of dependents, with no choice in the matter.

Many players will look at all the benefits and special abilities of the cavalier class and be envious of them, but it's a good bet that many of those same players (especially those accustomed to playing "traditional" free-agent adventurer characters) are also very turned off by all of the obligations and would be unwilling to abide by them. And that, of course, is exactly the point. You don't get one without the other. A cavalier without all those burdensome obligations, who doesn't have to answer to an NPC boss and doesn't have to shepherd around a group of dependents and is free to go out adventuring and behave in whatever manner he or she chooses is called, you guessed it, a fighter.

All of that said, while I mostly like the class, I acknowledge that there are some issues with it, just like there are with the barbarian, including some details from the Dragon version (which I think is worth reading - the UA version reorganizes and streamlines the text in a way that is more efficient but loses some of the flavor; plus I like Keith Parkinson's illustrations better than Jim Roslof's) that were left out or changed in UA - presumably by Jeff Grubb and/or Frank Mentzer - that I think shouldn't have been. Therefore, I recommend the following changes to the UA text:
  • Elf and half-elf cavaliers have the same level limits as fighters of the same race [per Dragon #96, seemingly overlooked in UA - and really it only makes sense]
  • The activity of the cavalier is such that it precludes any other profession - there can be no multi-classed cavalier or dual-classed cavalier. [per Dragon #72]
  • Add battle axe, bec de corbin, pole axe, falchion sword, and two-handed sword to the list of allowed/preferred weapons [all of these were on the list in Dragon #72 and were used by knights historically, so their deletion in UA seems to have a case of overzealous game-balancing overtaking source-fidelity - I smell the meddling hands of Grubb or Mentzer!] 
  • Add falchion sword as one of the options for the second weapon of choice [also per Dragon #72]
  • Add composite short bow as an option for the second or third weapon of choice for elf and half-elf cavaliers. Such characters have an increased rate of fire (3/1 at levels 6-10, 4/1 at levels 11+ (should they manage to attain such)) and are able to employ their full rate of fire even while mounted [per Dragon #72]
  • Weapon of choice "to hit" bonus is capped at 17th level (i.e. +3 for the third weapon) [implicit per Dragon #72]
  • The Protection from Fear aura of good-aligned cavaliers becomes a +2 bonus to saving throws against Fear-based effects, not a blanket immunity [per me - this effect is just too good as blanket immunity]
  • Delete the 90% resistance to mind-affecting magic and +2 saving throw vs. illusions [per me - legendary knights (from Arthuriana, Orlando furioso, Don Quixote, etc.) regularly succumbed to such effects and were constantly being charmed, beguiled, and fooled by illusions. If anything, it feels more appropriate and truer to the source material that cavaliers should have a penalty against such effects than a bonus!].
With these modifications, and with a firm understanding by both the player and especially the DM - if the DM doesn't enforce the restrictions and obligations and just allows the cavalier character to be "a fighter, only better" of course the other players will resent it - of how such a character fits into the game and setting (i.e. that they're only suitable for a game where what the characters do in-town actually matters and can't just be plugged interchangeably into a traditional group of vagabond murder-hobo adventurers) I think the cavalier can be a solid addition to and expansion of the scope of the AD&D game and bring in more of a romantic fantasy element alongside the grubby swords & sorcery flavor of the original/baseline game.

It almost goes without saying that cavaliers are especially suited to small, one or two-person, player groups and/or to younger players who might be more willing to accept being sent on missions by an NPC boss instead of being self-directed free agent treasure-seekers. Another interesting possibility in a large and long-running campaign would be to have cavaliers as "second generation" characters whose bosses are the earlier (now high-level, retired) adventurer PCs who carved the kingdom out of the wilderness and settled it. The first generation of PCs builds civilization out of the wilderness, then the second generation is charged with defending it.


  1. Just as "true trolls" and "ghouls" in CHAINMAIL (1969) are nailed down in D&D (1974) to be a specific thing the "knight" of CHAINMAIL takes the form of the cavalier:

    "Knights: Feudal Knights were ill-disciplined and generally refused to take
    orders from anyone — even their liege lord. However, they were exceptionally
    brave. Whenever Knights come within charging distance of an enemy they will
    charge regardless of any orders, unless a 6 is rolled on an "obedience die, " and
    regardless of any such roll if they can see other friendly troops moving towards
    the enemy, or attacking, they will charge or move towards the enemy if unable
    to charge. If more than one type of enemy troops are within charging distance
    the order of precedence that the Knights will follow is:
    1. Other Knights
    2. Any other mounted troops
    3. Baggage or missile troops firing upon them
    4. Armored Foot
    5. Missile-armed troops
    6. Artillery or siege equipment
    7. Heavy Foot
    8. Light Foot, Peasants, or Levies
    9. Pike armed troops"

    I verified this with Gary having had players obsessed with playing cavaliers. We had it down to a system of cavaliers did not function with chaotic characters so players would use two characters with the cavalier

    1. Yeah, I remember the "Aha!" feeling when I first came across that passage in Chainmail c. 1988 or so

  2. Interesting observations about the class, and I'm now thinking about the validity of a "civilized" game, with jesters and cavaliers and mountebanks and the like.

    On a different note, would you retroactively apply the cavalier class to the Arthurian entries in the D&DG, and in which way?

    1. I'd make pretty much all of them cavaliers. Galahad and Percival (and possibly Bors) are paladins.

      Actually, though, now that I think about it Arthur himself is probably a fighter rather than a cavalier. Both because he has a different background than the other knights (he wasn't raised/trained to be a knight) and because he is a Lord, while cavaliers are always servants - even their highest level is Cavalier Commander, basically a General, but not a Lord. Which is another important distinction I probably should've mentioned in the post, and another reason why some players will prefer the fighter class over the cavalier.

    2. Put another way, even the highest level cavaliers will always still have to answer to a Boss. High level fighters, by contrast, eventually get to BE the Boss.

    3. To further extend that line of thought, a name-level cavalier who wants to settle down and become a Baron and collect taxes from the peasants and let others do all the hard soldiering work could, presumably, voluntarily retire from their order and by doing so become a fighter, in the same way as if they had been involuntarily kicked out of the order. Of course whether the order will let their former brother go willingly and remain friendly or declare him or her an apostate and outlaw to be hunted down and slain will depend on the specific order.

  3. I had Knights of Quality as having all features of the cavalier class but Knights of Renown were as fighters with no restrictions. I also had the term "knight" as an add on to character classes for those dubbed a knight in service to Arthur. So a thief character was a "Knight Thief of the Round Table", etc.

  4. Gods I hate that term "murder hobo".


    1. Anything with "hobo" in sounds funny. There was a delicatessen called the "Hobo Deli" in upstate NY that used to make me and my cousins giggle when we were kids. Plus there was a "Ho-Bowl" with a "bum" on the sign that everytime I passed it someone had to say it out loud in a wry manner. Its just a catchy word. You can use it where it really belongs in the Lorraine Williams' era of TSR because DMs weren't giving out poop in those years to players and call them "story hobos". See? Its already catchy! ;)

    2. With you on this one, Allan. Every time I see the term "murder hobo" all I can think of is guys trying to feel cool and elitist by using the catchphrase du jour. "Fantasy heartbreaker" is another one of these that comes to mind.

    3. Aww, I think it's funny, plus it pretty accurately describes a pretty common approach to D&D, the way we used to mostly play as kids, where the PCs just go from one adventure to another killing the bad guys and taking their stuff, with no home, no family or background, and no friends except for each other (and even there there was usually so coveting and scheming to get ahold of particularly desirable magic items). We obviously never thought about it that way at the time (beyond perhaps comparing ourselves to Clint Eastwood's "Man With No Name"), but looking back, yeah, it really is like our PCs were a band of drifters who just went from town to town killing things.

    4. I don't recall anything like murder hobo play although the opposite, super controlled penniless voids with characters on a train track arriving at each story station, were rampant in the early 80s (my friend was embedded in one of these from 79 to 83) but not on the pop level. Those became popular after the rot set in post-1985. I had actively avoided 83 Basic (Mentzer edit) groups so maybe that's the root of hobo world. The real early play that fizzled was monsters as allies. The more fantasy weak and rule excessive the gameplay became the less likely you were going to run into characters with monsters in their party.

    5. It might be [that the murder-hobo thing arose out of the Mentzer-edit Basic]: the introductory story-game thing in that set that teaches the rules certainly does feature a bunch of characters hanging out in a dungeon that exists for no reason (it's just a set of caves outside of town) and fighting and killing each other for no purpose other than that "it's what adventurers do," without questioning any of it. Later in the set it talks about coming up with reasons for adventures, but by that point the players have already been indoctrinated into the idea that you're just supposed to wander into the caves outside of town and look for stuff (including "bad" people) to kill.

    6. My brother ran a Mentzer-edit Basic campaign in Middle School from 84 to 86 but I never saw any of his players because I was in the High School and those kids weren't driving like all the ghouls that I had for players. I used to say "Mentzer" like it was a verb sometimes! "Don't be Mentzer-ing everything! etc. Thats how I would argue with him about bringing BECMI material (In fact, I didn't even call it BECMI [Did I coin this term?], I called it "Mentzer") over to my games. Some "good" old days! Its funny because, when I met Frank Mentzer, I really felt bad especially in retrospect with all the nostalgia questioning that I was barraging Gary with. Perhaps, I should have said,"YOU RUINED EVERYTHING BETWEEN ME AND MY BROTHER, YOU FUCK!" ;) Seriously, Franks a good guy.

    7. Gary clearly liked Frank (and vice versa) and Frank's assistance definitely helped when Gary was busy with other stuff - without Frank we definitely never would have seen T1-4, and probably not UA, and maybe not even the MM2. That said, Frank has a very literal approach - that the rules are the physics of the game world - that I don't really like and think undermines the sense of wonder and leads to boring adventures (like the later sections of T1-4). So I've got mixed feelings about his stuff. I think he was probably better suited as an editor/developer than as a writer.

  5. While this wouldn't be absolutely required to run a campaign including cavaliers, I think one with battles and mass combat - human nations conducting politics by other means between each other - is one way to use the restriction of being tied to a lord. Lords are more likely to care about their temporal power and increasing their status. A significant amount of a cavalier's time should be devoted to their lord's military designs.

    1. It might be worth coming up with a table that the cavalier player has to roll on on a weekly or monthly basis to see what duty is required of them - training, escort, patrol, siege, battle, ceremony, etc. with only a small chance that the lord has no pressing duty and the cavalier is free to go wandering/adventuring at will - though some of the other activities could also become adventures, just not self-directed ones.

      The same idea could be applied to any character who is a member of a guild or organization - with the difference being that other types of characters can choose whether or not they want to join or stay in the group (and get benefits of association) whereas for a cavalier it's mandatory.

    2. A "required service" table is a great idea. I like this. A lot.

  6. Its funny that players have such a thing about Cavaliers but not Paladins. Both serve a superior that's essentially a NPC under DM control. I always liked the Cavalier myself (thus the "Knight" class in C&C) but I do agree its one of those classes that require work from the DM to make sure the player is fulfilling his/her obligations...or else. ;)


  7. I just wanted to make a quick comment about how I recall the time that cavaliers came out (because my brother was instantly obsessed with them). D&D people, that I knew, waited for the cartoon (1983-1985) with bated breath. At that time, I was playing AD&D minimum at least once a day (Gamma World once a week. Boardgames in general everyday.) I believe there was more correlation with then current unrelated dragon culture out there than what was on TV and video games. "Pussy dragons" were the rage since "Puff the Magic Dragon" (posters, Pern, etc.) but the scary ones, like in D&D, were correlated with full plated dragon slayers and there was a cultural niche of, now hard to find, imagery (usually St. George, etc.). By 83, the cavalier class was starting to make full plated miniatures be associated with it by players and the various companies. Priorly, or as it came out in the Dragon article, the name "cavalier" had been reserved for plated fighter miniatures with fancy hats but by 1986 or so it seemed like if it was heavily armored it was no longer associated with fighting men and clerics but stuck with "cavalier","paladin" or "knight". To this day when I grab a warrior figure with chain I don't think "cavalier" for a second.

  8. My problem with the cavalier's ability increase doesn't have anything to do with whether it's fair. It has to do with whether or not it makes sense.

    "Yes, Paul. Your character can exercise and thereby increase his physical stats."

    "No, Luke. While your character can certainly exercise just as much as Paul's, your character cannot increase his physical stats."

    The argument against this isn't "poor Luke." I could care less about poor Luke's feelings. The argument against this is that it makes no sense.

    If it's reasonable to say that people should be able to get stronger by lifting weights, or increase their constitution through endurance training, then put a system into the rule set that applies to every character capable of lifting weights or going through endurance training.

    I've always thought a lot of the problems with the cavalier class stem from two main sources:

    1. fan based player feedback was making it clear that masturbatory power creep sells, and

    2. the core of the system wasn't designed to allow for easy expansion into other classes. How, for instance, could they have another fighter class that didn't outshine the fighter? Because the fighter never had anything past a bare-bones package.

    1. But doesn't the "why can't everybody do it?" logic apply to a lot of other class abilities too - why can't I learn to pick locks or disarm traps? Why can't I study spell books and cast spells? Etc. Sure you could make a system that itemized out how much training and practice time was needed to gain all of those abilities, but doing so would be a different game than AD&D. In AD&D you've got classes that are based on archetypes and each one has a set of abilities, including some stuff that theoretically could also be done by other people but isn't because that's how archetypes work.

    2. No, the logic doesn't apply because this is a faulty comparison. Herein we are trying to compare an everyday act (exercising) to a trained skill (disarming traps, casting spells, etc.) These are two different things and while the comparison seems to hold water on its face, it's fallacious.

      It's like trying to compare digging a ditch to getting a doctorate in quantum physics. They are only the same insofar as they are both things human beings can do. Beyond that, there are clear delineations.

      Yes, a character could learn to cast spells or disarm traps, but the character would have to take the class in question to learn the appropriate skill. This is reasonable, and I put forward that most reasonable folks would agree.

      But anyone can exercise. Anyone can build muscle. In fact, everyone builds muscle and endurance as they grow older. Suggesting that someone needs a degree in exercising to get stronger by lifting weights is just goofy.

      It's already a failing of the system that characters cannot increase their strength through common sense methods. This failing is compounded (imo) when one class is suddenly afforded the verisimilitude of this ability while other classes remain sheets of paper. It becomes a slap in the face, so to speak. The gains from exercise are acknowledged by the game system, but ultimately denied to the bulk of characters in the game. And for what? Because someone got a little too carried away in their admiration of "the bestest trained!" class. The bias, it shows. And that makes for poor design. In my opinion.

    3. For me it's like working with a professional trainer to formulate an efficient, targeted exercise regimen, and then sticking with it consistently. That's part of why cavaliers are required to spend a portion of every single day doing training exercises. Sure, anyone else could theoretically do the same thing, if they had both the initial training and the discipline to stick with it, but the archetypal nature of AD&D characters means that as a general rule they don't - sure it could be arranged as a special case exception between a player and DM, but the non-cavalier archeypes don't include that sort of intensive, ongoing training. Cavaliers are the pro-athletes of D&D - they hit the gym every single day. Fighters, barbarians, and rangers aren't that single-minded - the ranger is busy cross-training other skills, the fighter and barbarian are out partying.

      Yeah, it's easy to punch holes in that logic, but it works well enough for me that I'm not bothered by the idea, and think it's a nice flavor differentiation between the classes. YMMV.