Thursday, May 20, 2021

What's the Point of Ability Scores? (Part IV)

Lest it appear that my investigation into ability scores lies only in AD&D's treatment of them, I wanted to devote a single post to some oddities in OD&D in all its forms. Let's start with the 1975 version, which includes two easily forgotten and misunderstood rules. The first concerns changing character class. According to Volume 1 of OD&D, a character can only change his class if

they … have a score of 16 or better in the prime requisite … of the class they wish to change to, and this score must be unmodified. A Cleric with a "strength" of 15, for example, could not become a Fighting-Man. 

There are echoes of this in AD&D's rules for dual-class characters, though they are much more stringent and restrictive in the latter game.

More interesting – and baffling – are the following lines from further down the same page as the above quote. Under the heading "explanation of abilities," we read:

Strength is the prime requisite for fighters. Clerics can use strength on a 3 for 1 basis in their prime requisite area (wisdom), for purposes of gaining experience only …

Intelligence is the prime requisite for magical types. Both fighters and Clerics can use it in their prime requisite areas (strength and wisdom respectively) on a 2 for 1 basis …

Wisdom is the prime requisite for Clerics. It may be used on a 3 for 1 basis by fighters, and on a 2 for 1 basis by Magic-Users, in their respective prime requisite area …

With the addition of thieves as a class in Supplement I (1975), we also get the following:

Thieves use dexterity in the pursuit of their chosen profession. They may use 2 points of intelligence and 1 point of wisdom to increase their raw dexterity score so long as they do not thereby bring intelligence and wisdom scores below average.

What to make of all this? It seems pretty clear that, at least in the case of Volume 1 of OD&D, the rules state that a player may, depending on their character's class, lower certain ability scores to increase said character's prime requisite. Suppose, for example, that Boro the fighting man has a Strength of 14 and an Intelligence of 10. According to the guidelines above, Boro's player can lower his Intelligence from 10 to 8 in order to raise his Strength to 15, thereby increasing Boro's experience bonus from 5% to 10%. 

Where things get complicated is the phrase "for purposes of gaining experience only." My sense of it is that the rules suggest that the ability score is not actually increased but is simply treated as if it were higher for the purposes of determining a character's experience bonus. Strength and Wisdom have no clearly enumerated mechanical benefits beyond providing an experience bonus. Intelligence, on the other hand, provides additional languages. On my reading of Volume 1, a magic-user who lowered his Wisdom to increase his Intelligence would not gain any further languages from the increase. 

Is this the correct reading of the rules? I'm not certain, especially since the discussion of thieves and dexterity in Greyhawk makes no mention of the experience bonus. Likewise, the formulation in Supplement I contains the stipulation that the player of a thief may not raise Dexterity by lowering another ability score "below average," which presumably means "below 9." Whatever its intended meaning, what is clear is that, even in OD&D, the rules provide players with a limited faculty to ameliorate the results of the 3d6-in-order rolls.

The Holmes edition (1977) has a similar system, though some of the details are different (e.g. there's seemingly no restriction on how low a thief's player can lower other scores to increase Dexterity). With Moldvay (1981), there are a number of changes to this system. Firstly, the ratio of lowered score to raised score is universalized to 2 for 1, as is the restriction that no score may be lowered below 9. Only prime requisites may be raised, as in OD&D and Holmes, but, because all ability scores now provide clear mechanical benefits, this has additional effects beyond increasing the rate of experience gain.

There are two more points worth mentioning. Holmes introduces, and Moldvay retains, the "hopeless characters" guideline, which suggests that a character whose ability scores are all below average is not fit for play and may be discarded. Holmes expands upon this by adding that "more than one very low (3–6) ability score" is sufficient cause for a character to be discarded. It should be noted, though, that both Holmes and Moldvay leave this decision to the Dungeon Master, not the player. 

The second point is Moldvay's innovation of what came to be known as "ability checks." Moldvay indicates that, because "there's always a chance," the DM may allow a player to roll 1d20 against his character's relevant ability score to determine if the character is successful in some difficult task (with modifiers to the roll as deemed appropriate). If the roll is lower than the relevant ability score, the character succeeds. While both OD&D and Holmes imply that abilities might be used by the referee to determine success in circumstances not otherwise covered by the rules, the 1981 edition of D&D is the first to explicitly mechanize it in this fashion. 

Part I | Interlude | Part II | Part III


  1. The way I'd interpret it would be like this:

    Say you have a cleric with 18 strength, intelligence and wisdom (obv very unlikely, but it divides well).

    He adds 6 to his Wisdom due to his strength, 9 due to his intelligence, and so ends up with an effective wisdom of 33 for the purposes of gaining experience.

    I strongly suspect the table doesn't go that high, though. But if it does, I bet that's what they meant.

  2. what an incredibly fiddly arrangement.

  3. Your interpretation as the other ability points increasing how the Prime Requisite is treated as though higher, rather than actually being increased, is exactly how I interpreted the rules.

    It was interesting to see how that evolved into the 2 for 1 tradeoff in actual scores in Moldvay (which is what I used for a long time, even in AD&D).

  4. You're a little off, James.

    Bottom of page 11 (OD&D Book 1):
    "Note: Average scores are 9-12. Units above may be used to increase prime requisite total insofar as this DOES NOT bring that category below average, i.e. below a score of 9."

    Boros could not thus raise his strength to 15 by lowering intelligence to 8.

    The table on the same page clearly lists the benefits (and penalties) to experience for a high (or low) prime requisite. Page 8 of Greyhawk (Supplement I) under DEXTERITY states explicitly that

    " is also the prime requisite for thieves."

    Which explains why that class gains the ability to make adjustments to dexterity: it will give the character a better x.p. bonus.

    While I think your interpretation of the text "for purpose of gaining experience only" is fine and valid, I've always inferred that to mean that the only reason fighters get to raise strength is because it is their prime requisite (and thus impacts their ability to earn experience). In other words, it is the fighter's concentration (just as MUs have intelligence, etc.). Classes can only raise the ability in which they focus and no other (you can't adjust your fighter's CON or DEX, much as you might want to). I just feel the text has some clunky it does in other parts, too.

    Moldvay's inclusion of the "ability check" is the absolute worst part of his has led to everything from non-weapon proficiencies to universal D20 skill checks. I'm wondering if there was a pre-1981 (TSR) adventure module that featured some sort of "roll under" save to avoid a weird trap that generated this idea, or if Moldvay is the first presentation of this mechanic.

    1. In regards to Moldvay's "ability check" what happens when a character wants to do something other than combat or spellcasting but there is a chance for failure? What if that chance of failure is reasonably affected by how strong the character is, how smart, or how dexterous?

      For example I am in a swamp and just took out a lizardman's outpost and found their treasure horde. Unfortunately I don't have enough room to carry it out but noticing the nearby reeds, I attempt to weave a basket to carry it all in. Except part way in I have to rush it because a incoming party of lizardman to large for the party to handle is spotted and will be at the site in 10 minutes.

      Will the basket always hold the treasure while the party flees?
      There is an arbitrary odd assigned?
      Will it always fail to hold the treasure?
      Does it matter if a character has a 16 dexterity and tries to do this?
      Does it matter if a character has a 16 intelligence?

      Yes weaving a basket sound trivial and lets to the question of "Why bother?" But there things a character can do outside of combat and spellcasting is near infinite.

      And of that the number of times the outcome of attempting those other things is uncertain is substantial.

      And of those uncertain outcome some will be logically affected by one of the character's six attributes.

      And back to my example, each of the possible responses to the questions similar to "Does the rushed reed basket hold the treasure during pursuit?" has a different consequence to how campaigns for that referee plays out.

    2. "I've always inferred that to mean that the only reason fighters get to raise strength is because it is their prime requisite"

      Agree. Which means the OSR darling of "3d6 in order, without point swaps" is based on... a complete misunderstanding?

      Jaquays' Cavern of Thracia published in 1979 had Ability Checks. So did 1979's B1 - In Search of the Unknown. It's been in the hobby from the very beginning. Moldvay was wise to codify it. I'll never understand the dogmatic stigma against it.

    3. Do you recall where ability checks are mentioned in B1? I ask because B1 is a strange bird in terms of its rules additions to Holmes Basic and I'd be curious to see how Carr describes the mechanic.

    4. @James:


      B1, p.11, Room 8. To identify mysterious ingredients in a wizard's workroom, you "multiply a character's wisdom times 5 to give the percentage chance of positive identification."

      Then again on p.20, Room 36, PCs must save vs CON to dive underwater and recover their belongings: "...due to the extremely cold temperature of the water, characters will depend upon their constitution rating to see if they can stand the water enough to dive for things on the bottom. One check can be made for each character, with a 5% chance per point of constitution that they will be able to take the cold water (for example, a character with a constitution rating of 11 would have a 55% chance of being able to take the cold water and dive effectively)."

      Of course in both cases, if you roll a d20 instead, no multiplication is necessary. :)

      In Caverns of Thracia, it's p.23, room 1, you roll vs DEX to avoid slipping in bat guano. Also in rooms 2 and 14.

      Encounter 19 is a rope bridge haunted by giant bats. "If the bats score a hit against a character on the rope bridge, that character must roll less than or equal to his dexterity +2 on a d20..." or fall from the bridge.

      1979. Jaquays. Carr. Moldvay. What kind of credentials does a rule need to be taken seriously around here? Embrace it already. There's no shame in it. :)

      Btw, thanks for this blog. I read it all the time.

    5. Actually. From my understanding of the research, Arneson's game essentially started as nothing but ability checks, tge characters being entirely ability score based. Even combat was run via ability scores. Characters had a larger number and much wider range of scores, as well, than the six that Gary eventually settled on.

    6. Here are two articles about Arnesonian Ability Scores:

      So really, the "roll under" Ability Score system is the original default, "re-invented" by Carr, Moldvay, etc.

    7. @Mishler: Older than old school! Thanks.

    8. @Mishler: Wow. So, did Gygax keep Ability Checks out of AD&D just because of his lengthy campaign (pun intended) to keep royalties away from Arneson?

    9. Thanks for this. It's interesting that Carr employs a percentile roll rather than a d20 one.

    10. Agree. I wonder if it's because d20 rolls in D&D were all about rolling high. An Ability Check is an anti-Saving Throw; you actually want to roll low on a d20.

      Carr's "the percentage chance" of something solves the communication around this non-intuitive Save. Unfortunately, it adds the extra, unnecessary step of multiplying by 5.

      All that said, it didn't stop Jaquays from going right to a d20 roll vs an ability.

      And neither of them name this roll.

    11. Another early instance of an ability check:

      AD&D Players Handbook (1978), p. 76, "Dig." "Any creature at the edge (1') of such a pit uses its dexterity score as a saving throw to avoid falling into the hole, with a score equal to or less than the dexterity meaning that a fall was avoided."


    12. I'm not sure of where I first heard of d20 <= attribute checks, but I'm pretty sure I was exposed to them by 1978. CSIO does have roll d100 <= attribute. d20 may have come by word of mouth. There could be something else buried in an early JG product (CSIO, First Fantasy Campaign, Thieves Fortress Badabaskor, Tegel Manor are what I'm sure I saw that early, I don't remember when I saw Citadel of Fire).

  5. OH...and regarding players "ameliorating 3D6 in order rolls." Remember that the text in OD&D states it is the DM that rolls the PC's ability scores. If you're handed a character with a low strength and WANT to play a fighter (because there are no ability-based restrictions on class in OD&D...and not even a penalty to attack rolls!), you can use the rule here to help make your character become a faster-leveling fighter, regardless of the hand dealt by your DM.

    ALSO (sorry, last note): since pre-Greyhawk OD&D had no advantages for Strength, one might consider that the line "for experience purposes only" is a reference to the (later discussed) text about changing classes, i.e. "this score MUST be unmodified."

    Without this stipulation (noted twice), it would be quite possible for an OD&D player to receive a PC from her DM with a WIS of 15, choose "fighter" as a class, raise WIS to 16, and then multi-class (i.e. switch classes) after a few level to gain both the benefits of fighters AND clerics. The text puts the kybosh on that in two places:

    - you may only raise the prime requisite of you character class for experience purposes (the only benefit of PR ability scores besides languages)
    - you may not switch classes unless the original prime requisite was un-modified

    Okay, that's it.
    ; )

  6. I think Gygax and/or Arneson had two conflicting ideas here:

    (1) Although each class has a prime ability, other abilities should improve performance in a class (smart and strong fighters should advance a little faster than those who are just strong.) This explains the different point ratios and why M-Us can't use Strength to improve earned experience.
    (2) An actual point swap, to reduce the risk of bad rolls.

    The problem is, they should have gone with #1 without #2, or #2 without #1. Trying to hang on to both leads to confusing language about lowering other ability scores, but only "for purposes of earned experience". Are we lowering the scores, or not?

    Holmes and the Basic/Classic line went with a literal point-swap. Some retroclones went with no point swap, but giving at least one ability (usually Wisdom) an experience bonus effect for all classes. AD&D kept experience bonuses for the prime requisite, but ditched the "secondary and tertiary requisites". WotC ditched experience bonuses completely.

    Can't say I'm entirely satisfied with any of those answers.

    1. RE #1 above:

      Huh! I never thought of that! This actually makes perfect sense (even the 3 for 1 or 2 for 1 of wisdom for fighters and magic-user respectively)

      But, yeah, in execution it's not clear (because it's actually reducing ability scores and once PR abilities start having meaning besides x.p. adjustments...hoo-boy!).

      I wonder if there's a way to work #1 without the need for some fairly complex tables? Probably not, unfortunately.

    2. Oh, wait. Now I get it. Yep. it.
      : )

    3. Posted:

    4. That was a good post. Makes sense of why each class makes use of a different set of attributes. If you could trade points, why wouldn’t it be more flexible?

      The bonus to XP rather than combat or spellcasting or other abilities goes to show the relatively greater importance of the long campaign in OD&D.

  7. Gygax needed an editor to help him with clarity in several places. Putting together the text cited, it is clear he meant the virtual increase of stats for the purpose of increased experience gain from a prime requisite for the classes. I put the following text in my OD&D clone:

    Under Strength:
    "For each 3 full points of strength above 9, clerics may consider their prime requisite (Wisdom) 1 higher for the purpose of gaining experience."

    Under Intelligence:
    "For each 2 full points of intelligence above 9, fighters, clerics, and thieves may consider their prime requisite (strength, wisdom, and dexterity, respectively) 1 higher for the purpose of gaining experience."

    Under Wisdom:
    "For each 3 full points of wisdom above 9 for fighters, 2 full points for magic-users, or 1 point for thieves, they may consider their respective prime requisites 1 higher for the purpose of gaining experience."

  8. Wow. So for 40+ years the rule has been misinterpreted, and the misinterpretation extended in Moldvay's Basic book?

    Very interesting.

  9. Thanks for an interesting post. My two takes:
    1)given Gary's ability to articulate the logical operators of rule mechanics, it is a miracle anyone ever figured out how this game can be played in an approximately rational and self-consistent way.
    2) thank the lord above for Moldvay's revival of the attribute as a meaningful part of the game. I think the failure to develop the d20 roll-under-an-attribute approach as a core mechanic of play was one of the worst design decisions in the history of roleplaying. Out of that choice came decades of baroque complexity in modifiers and class powers, and clashing approaches to different sorts of situations (to-hit rolls; listen rolls; etc.). I love pre-1980 D&D, but think it absolutely demands a deep set of house rules developed in this direction.