Determination of Abilities

[aher]

"Prior to the character selection by players it is necessary for the referee to roll three six-sided dice in order to rate each as to various abilities, and thus aid them in selecting a role" (Men and Magic, 1974, page 10).

I see two obvious but disparate ways to parse this sentence:

1. Popular reading: "in order" (without "to") is adjectival. It modifies "dice". Most gamers seem to interpret the sentence this way -- roll the dice and record the results in order: STR first, INT second, etc. Neither the referee nor the player has any say how the scores are arranged. As the good book says, "We may throw the dice, but the Lord determines how they fall" (Proverbs 16:33).



Figure 1. Popular reading.
First of two possible Reed-Kellogg diagrams for sentence on page 10 of Men and Magic.

2. Alternative reading: "in order to" is a subordinating conjunction that introduces a series of infinitival clauses (without subjects). These two purpose clauses answer the question: For what purpose does the referee roll the dice? The infinitival clauses provide the goals that answer this question: "rate ... and aid ..." This interpretation seems best (to me, at least). Nowhere does it say to assign the first roll to STR, etc.; the referee could assign the rolls to abilities in any way he desires. There isn't even any requirement to roll six times; the referee could roll six times, thirty-six times (as in Method III in AD&D), or seventy-two times (as in Method IV in AD&D). The text nowhere mentions exactly six rolls -- the "each" being rated by the rolls are clearly the players, not their characters' abilities.



Figure 2. Alternative reading.
Second of two possible Reed-Kellogg diagrams for sentence on page 10 of Men and Magic.

If the popular reading were true, why would it matter who rolled the dice? But if the alternate reading were true, and the referee rolls and arranges the scores as he desires, then he can "aid them in selecting a role", i.e., push them in a certain direction, towards a class that best fits his campaign.

You could argue that the Xylarthen example contradicts the second (alternative) interpretation. "This supposed player would have progressed faster as a Cleric..." but presumably couldn't because of the order in which the dice were rolled. Not so! This only implies the player didn't arrange the scores himself. But remember, the referee rolled the dice. And the referee could have freely chosen to assign the results to the six abilities this way, favoring WIS, perhaps urging the player to become a cleric, maybe because the party lacked one. But the player ignored this hint. Regardless, the referee could have assigned a higher or lower score to INT. There's no implication he assigned scores to Xyarlathen's abilities strictly in the order they were rolled.

Do subsequent editions of the game provide any additional clues as to which reading is correct?

On page 11 of the AD&D DMG, Gary recalls with apparent regret OD&D's generation of ability scores, which some AD&D players refer to as Method 0, "While it is possible to generate some fairly playable characters by rolling 3d6,..." Nowhere does he say "in order" in reference to Method 0.

Before that, Gary wrote, "it is important to allow participants to generate a viable character of the race and profession which he or she desires." He achieves these goals of viability and of satisfying the player's appetite for a certain race/class by adopting four new generation methods --

Method I: "All scores are recorded and arranged in the order the player desires." This doesn't explicitly say who rolls: Referee or player? But it does explicitly say the player arranges the scores. A 4d6k3 roll-and-keep machanic is used. The mean ability score is 12.24, and the chance that the player gets an "18" in one ability he pre-selects is 9.3% (the same as the chance he gets an 18, since he can arrange scores as desired).

Method II: This uses 3d6 twelve times. The player retains six scores and arranges them as desired. The mean ability score is 12.74 (higher than Method I) but the chance of getting an 18 in a chosen ability is only 5.4%.

Methods III is rolled "in order" STR, INT, etc. But each ability gets 6 rolls and the maximum is kept. This maximizes viability (the mean ability score is 14.23), but at the cost of frustrating the player's preferences for race/class (the chance of getting an 18 in one preselected ability is only 2.7%).

Method IV also rolls 3d6 "in order." But twelve characters are generated and the player picks the one he wants. If the player picks the one that maximizes his pre-selected prime requisite, the mean ability score is 11.47, and the chance of an 18 in the prime requisite is 5.4%.

(BTW, the means and probabilities are from a 2008 post on RPG.net called AD&D Analysis of character generation methods. In November 2010, I reproduced these results using R.)

Basic D&D went on a different trajectory...

Holmes: "Each player starts by rolling three 6-sided dice for each characteristic" (page 5).

Moldvay: "Roll 3d6 (for a result of 3-18) for each ability and put the result in pencil next to the name of the ability" (Page B5).

Holmes and Moldvay leave wriggle room for the alternative reading. Mentzer does not:

"Write the Scores down as you roll them, next to the names of the Abilities" (page 48). The accompanying picture shows STR 14, INT 9, WIS -blank-, and a hand rolling 3d6. No room for argument here, this is clearly "3d6 in order," but these rules were fairly late on the scene (1983), and I never played by these rules, despite owning the whole BECMI shebang.

So even if "3d6 in order" was intended all along, ever since Men and Magic, we don't get a crystal clear statement of it in an official rule book from 1974 for an entire decade until 1983 with Mentzer's red box.

But could that merely have been Mentzer's personal preference? My understanding is that he taught himself D&D, rather than learning it directly from Gary or Dave.

"Every link in the genealogy of dharma transmission occurs in documented historical circumstances: a specific place and time, identifiable individuals, and specific words and actions. At the same time, though, Zen texts also assert that true transmission consists of no transmission. In other words, it occurs only mind-to-mind" (William Bodiford, Dharma Transmission in Theory and Practice, 2008).

Allston (Rules Cyclopedia, 1991, page 6) keeps Mentzer's wording (above).

And finally the last word on the matter: "Take out ...paper... Write 'Strength' near the left-hand margin. Roll 3d6 and record the result next to Strength. Below Strength, write 'Intelligence,' then roll 3d6 and record the result next to Intelligence. Now do the same thing Wisdom, Dexterity, Constitution, and Charisma" (Doug Stewart, The Classic Dungeons and Dragons Game, 1994, page 12). Not only does this say to roll 3d6 "in order," it even bosses us and tells us exactly where on the sheet of paper to write everything down -- the left-hand margin, in case the anal-retentive detail of the prose caused you to stop paying attention. If that makes "3d6 in order" an official rule, then shouldn't Strength in the left-hand margin of the page also be an official rule???

In sum, the 1974 rules are ambiguous about rolling "in order." The AD&D rules present 4 new conflicting methods: half let player "arrange as desired"; half make him roll "in order," but even one of these lets the player choose one from twelve random PCs, allowing him to maximize a desired ability and thus negate the whole effect of "in order" rolling. Basic D&D initially retains OD&D's ambiguity, then eventually ordains the "3d6 in order" as official rule a decade later; and finally 2 decades later, classic D&D gets downright draconian (pun intended), telling the player not only how to roll, but also where on the page to write down the score. Personally, I see nothing definitive in all this.

"When an action is neither legally prohibited nor legally permitted there is a legal gap" (Joseph Raz, The Authority of Law, page 58).

[Finarvyn]

I'll confess that I don't understand Reed-Kellogg diagrams and they simply make your post more confusing to me. What I glean from your post is two interpretations for "in order"
(1) "in order" as in "in this sequence"
(2) "in order" as in "to determine"

The advantage of the first interpretation is that it forces character generation to be more unpredictable in that you don't know what kind of character you'll run until you see the numbers. The advantage of the second interpretation is that it offers more player choice.

Our group interpreted it in the first way back in 1975, and used the Xylarthen example as our basis. Later on, however, we abandoned this interpretation (without actually realizing we had adopted the second) because we wanted the option to choose which kind of character we could play. Honestly, I don't remember ever reading it the second way until your post. Now I do it different ways in different campaigns, depending upon what I want from my game.

I knew that different rules sets offer different choices, but never really considered it might trace back to two interpretations of that sentense. Interesting.

[Sean Michael Kelly]

Whoa! I hadn't done any diagramming in 20 years, and that was in Greek!

[Vile Traveller]

Hey, nice diagrams! Never seen that kind of thing before but I think it makes sense. I note there is nothing about order in Holmes, either.

[paulg]

Exalt, aher!

Oct 8, 2012 5:27:36 GMT -6 Finarvyn said:

What I glean from your post is two interpretations for "in order"
(1) "in order" as in "in this sequence"
(2) "in order" as in "to determine"

The advantage of the first interpretation is that it forces character generation to be more unpredictable in that you don't know what kind of character you'll run until you see the numbers. The advantage of the second interpretation is that it offers more player choice.

The idea for the second interpretation (if I understand aher correctly) is that it offers the players less choice but more guidance. The referee arranges the scores for the purpose of pushing the players to select classes most suitable for the campaign.

This interpretation is both plausible and interesting.

[geoffrey]

The opening post clearly shows that, linguistically, the sentence under discussion can be equally understood in either of two ways.

Now let me share why I think the context of page 10 of M&M indicates that ability scores are to be diced in a particular order (namely, S, I, W, C, D, Ch). Note that the example magic-user character (Xylarthen) has a higher wisdom score (13) than intelligence (11). The commentary even states, "The supposed player would have progressed faster as a Cleric, but because of a personal preference for magic opted for that class."

That sentence indicates that the ability scores are to be generated in a particular order. Otherwise, the player of Xylarthen would have assigned the "13" to his intelligence.

On top of that, the text continues, "With a strength of only 6 there was no real chance for him to become a fighter." This sentence has little meaning if the player could assign his rolled scores however he desired.

[Deleted]

I think this is a case of over-analyzing text and losing the intent. Perhaps it is because we are so far from the original players of the game, though there are several of us here (Marv, Vile, myself, and most notably, GronanofSimmerya). Or, perhaps it is because some players began modifying character generation right away.

I don't know and it doesn't matter anyway. No matter how the sentence can be parsed, the author's intent has been shown by how he played the game and taught it to others.

Two important points to keep in mind:

1) I mean no offense to the original poster, I'm merely trying to voice a dissenting opinion in what I hope is a respectful manner.

2) I'm not saying Aher's parsing of the sentence is incorrect.

If Aher or anyone else wishes to arrange scores to taste, I say all power to him. This is not a case of me telling someone else how to play the game. I'm merely trying to highlight the perils of exegesis regarding text written by (at that time) a non-professional writer.

[Deleted]

"I think this is a case of over-analyzing text and losing the intent. "

Ya THINK?

Let me settle this inexorably:

GARY AND DAVE BOTH PLAYED 3d6 IN ORDER.

Doesn't matter if the sentence is convoluted, unclear, or not. The INTENT, as ENACTED BY THE AUTHORS, is 3d6 in order.

[Deleted]

Thanks for the input and affirmation of my post, Mike.

edit: Have an exalt for being such an amazing historical resource.

[Finarvyn]

But it's also significant that some of us played "in the day" without the benefit of having Dave or Gary explaining how to do it, and our group (at least) interpreted "in order" to mean Str, then Int, then Wis, then Con, Dex, and finally Cha. The character example cited several times was our guide, since we didn't have anyone to teach us the game.

Nice to know that we did it "right" but I agree that if others like a different method that's okay, too. As I said before, I've done it more than one way depending upon my whim....

[Deleted]

Oct 8, 2012 19:35:54 GMT -6 Finarvyn said:

But it's also significant that some of us played "in the day" without the benefit of having Dave or Gary explaining how to do it, and our group (at least) interpreted "in order" to mean Str, then Int, then Wis, then Con, Dex, and finally Cha. The character example cited several times was our guide, since we didn't have anyone to teach us the game.

Nice to know that we did it "right" but I agree that if others like a different method that's okay, too. As I said before, I've done it more than one way depending upon my whim....

There is a difference between "this rule is unclear" and "if you spend a lot of time and energy you can make something reasonably straightforward seem dubious." The original post fails on grounds of, among other things, Occam's Razor. This is hardly adding new illumination to an old ambiguity in the rules. It is the verbal equivalent of what my statistics professor meant when he spoke about "if you crunch the numbers long enough, you can get anything to correlate."

[Aplus]

I've always interpreted the sentence the second way, but just assumed the dice scores were meant to be assigned in order due to the Xylarthen example and the stuff it says about trading scores for prime requisite ability on the bottom half of page 10. Until this post, I never even imagined that anyone would interpret the "in order" as being applied to the dice roll.

Honestly my favorite thing about the original books is the ambiguity. It makes being a referee more than just "guy who looks up the rules".

[Sean Michael Kelly]

Oct 9, 2012 10:35:18 GMT -6 Aplus said:

Honestly my favorite thing about the original books is the ambiguity. It makes being a referee more than just "guy who looks up the rules".

[raises a pint and tips elbow]

I'll drink to that!

[aher]

Gronan is a treasure trove of information. Learning the game based on what Gary did rather than what Gary wrote, through direct observation and personal interaction, for a year-and-a-half before the rules were published -- these are experiences I could only dream of as a kid.

And it would be interesting (to me at least) to know if Gronan rolled his own dice during character generation, or if Gary rolled, as per the written rules? Did Gary ever playtest Methods I-IV from AD&D in your Lake Geneva circle, before AD&D became a separate product from D&D? And if so, when did he begin experimenting with these alternative methods?

And most importantly, when did Gary dump the "3d6 in order" rule in his OD&D games? By 2005 Gary was running brown box D&D, with no supplements, and using "Method I" in favor of "3d6 in order." The rules he used are summarized all over the web, including here, here, and here:

No Supplements (Greyhawk, Blackmoor, etc.) are used, just the original three booklets. Ability scores rolled as best 3 out of 4d6 (arrange scores to taste).

Gary wasn't the only one who ditched "3d6 in order." Looking around the web it quickly becomes apparent that gamers developed a dizzying assortment of houserules for ability determination. For example, over at DF, where Gary was a frequent contributor, players invented such techniques as

• 3d6 in order, reroll the lowest score, then swap any two scores
  
• 3d6 in order, reroll until you get a character you are happy with
  
• 3d6 reroll 1s
  
• 3d6 in order. They can rearrange everything except CON and CHA
  
• 3d6, arrange as desired
  

One funny comment there reads "the mob that would come after me for 3D6 in order stat gen would burn down my game table!"

It's also crystal clear why all these houserules exist: to overcome the shortcomings implicit in "3d6 in order", shortcomings Gary himself made explicit in the AD&D DMG: (1) non-viable characters, and (2) frustrating player preferences for race/class (see the OP for precise quote and page number). Even the 3LBBs had the ability adjustment system, a hotfix for these two problems.

Now while many people apparently "paved over" the written rules with ad hoc houserules, others of us kept the rule "referee rolls 3d6 in order to aid ... and guide ...", literally, and took advantage of the linguistic ambiguity to let the referee arrange the scores to his taste.

The Xylarathen example doesn't contradict the referee arranging to taste, it merely contradicts the players arranging to taste, an invalid reading, and one I never argued for in the OP.

It's great that Gronan can tell us exactly what Gary did, but my interest in the OP was what Gary actually wrote, because for many of us, that's all we had.

It's easy for Gronan to dismiss the analysis in the OP as pilpul, since he never had to learn the game from a book. However, I just don't get the accusation of "multiplicans entitates praeter necessitatem." And even if it were true, what's the harm? Doesn't Volume I page 4 say "New details can be added and old 'laws' altered" and "If your referee has made changes in the rules and/or tables, simply note them in pencil"?

Nor do I find it self-evident that "in order" must be read as an adjective in this context. The sentence under analysis has a similar structure to the preamble to the US Constitution:

We the People of the United States, in Order to form a more perfect Union, establish Justice, insure domestic Tranquility, provide for the common defence, promote the general Welfare, and secure the Blessings of Liberty to ourselves and our Posterity, do ordain and establish this Constitution for the United States of America.

It would be crazy to read this amazing text as -- We (subject) do ordain and establish (verb) this Constitution (d.o.) in order (adjective): to (conjunction)

• (first) form
  
• (second) establish
  
• (third) insure
  
• (fourth) provide
  
• (fifth) promote and
  
• (sixth) secure
  

so that, for example, we must establish justice before we can provide for the common defence. The order isn't important here.

By claiming that the written rule is ambiguous is not in any way to disparage the game or Gary. As Aplus and SMKSensai wrote, it's one of the game's most endearing features. William Faulkner's novels are still considered great today precisely because of their ambiguity. Even the four evangalists didn't always see the same thing the same way, and this lack of consistency has much to do with what makes their books the cornerstone of our culture and literary canon.

Later editions of the game are marred by a kind of fundamentalist zeal to enumerate rules to cover every little detail and possible contingency and then to insist: You must do it this way! There's only one way to do it! My way or the highway! Even AD&D slouches towards this kind of totalitarianism.

Freedom to interpret the 3LBBs as a "framework around which you will build a game of simplicity or tremendous complexity" (Volume I, page 4), rather than as a complete and consistent set of rules carved in stone, drove me back to the game from the later editions. Even something as seemingly straightforward as elementary arithmetic cannot be both complete and consistent; insofar as the 3LBBs presuppose elementary arithmetic, they can't be complete and consistent either. And this lack leaves room for endless imagination and novelty.

[Deleted]

No Supplements (Greyhawk, Blackmoor, etc.) are used, just the original three booklets. Ability scores rolled as best 3 out of 4d6 (arrange scores to taste).

There is nothing wrong with this method, so please don't misread the following. It is my understanding (from conversation with Gary) that he used the above method because he ran a lot of "pick up" games for folks who showed up to play a game with one it's creators.

It is merely a point I thought might be of interest.

[kent]

Of the two readings it is difficult say which interpretation was intended, however, I would stress that even following the second reading it remains very likely that rolling dice for separate stats is implied in the absence of any conflicting method. Swapping the numbers about at will is a gross assumption from the text as presented at that time and a misperception influenced by reading later iterations of D&D.

For the record I am indifferent to the manner in which stats are initially generated in my own AD&D game.

>> " ... thus aid them in selecting a role."

Consider that if one is free to place the highest of six rolls where one wishes, and then the second highest and so on, that one has typically predetermined a 'role' for the character. In other words when free to sort the numbers the dice rolling comes after the selection of a role and so rarely 'aids ... in selecting a role'. On the other hand if we roll for each stat separately then the dice genuinely aid in selecting a role.

[murquhart72]

Oct 8, 2012 11:58:35 GMT -6 @gronanofsimmerya said:

GARY AND DAVE BOTH PLAYED 3d6 IN ORDER.

Doesn't matter if the sentence is convoluted, unclear, or not. The INTENT, as ENACTED BY THE AUTHORS, is 3d6 in order.

My question is did Gary/Dave roll these dice themselves or did the players roll their own abilities? The words written in Men & Magic tend to support the former

[Deleted]

At first he rolled them, later we rolled, especially if several people are starting at once.

[scottyg]

I asked Rob about the scores of Robilar and other Greyhawk characters. They are pretty good and I assumed they had been increased over time by different magical effects, especially wishes. Rob's response was that wasn't the case. Wishes were saved for emergencies: hopelessly lost, PC deaths, etc. He explained that for generating ability scores they did use 3d6 in order, but they rolled multiple sets and chose the one they wanted.

[sixdemonbag]

Jul 18, 2017 11:42:32 GMT -6 scottyg said:

I asked Rob about the scores of Robilar and other Greyhawk characters. They are pretty good and I assumed they had been increased over time by different magical effects, especially wishes. Rob's response was that wasn't the case. Wishes were saved for emergencies: hopelessly lost, PC deaths, etc. He explained that for generating ability scores they did use 3d6 in order, but they rolled multiple sets and chose the one they wanted.

Interesting, that's how most crpgs work. You keep pushing the "generate random stats" button until you get a set you like.

(Assuming you forego any point-buy options, which I always do...)

[Scott Anderson]

Pshaw. Iron man rules or bust!

I got a rare 18 strength on my last iron man guy. Therefore he was obviously a fighter.

Now I play a 18/40 fighter

With 2 hit points. Which suggests another question: how do we determine hit points in 0E?

[howandwhy99]

We roll Hit Points, but are balanced by Hit Dice.

Also, I believe the Referee might've rolled every roll in some early games. The reason being fairness, referees being a neutral party.

I see the same reasoning for Character Generation rolls. If players roll stats, you need to watch 'em like a hawk.

[murquhart72]

Jul 18, 2017 10:21:25 GMT -6 @gronanofsimmerya said:

At first he rolled them, later we rolled, especially if several people are starting at once.

Definitely a helpful memory, thank you for that!

[Scott Anderson]

Jul 18, 2017 20:02:02 GMT -6 howandwhy99 said:

We roll Hit Points, but are balanced by Hit Dice.

Also, I believe the Referee might've rolled every roll in some early games. The reason being fairness, referees being a neutral party.

I see the same reasoning for Character Generation rolls. If players roll stats, you need to watch 'em like a hawk.

I mean to say, when we roll hit points for games I referee, wr do it once, and take the result. Then when we level up, we roll all our hit dice again. We either take the new total or the old total plus one. That's not btb though.

How do you do it?

[hamurai]

That's exactly how I did it in most of my games, too. That's how I first read it in the XP table, actually. It doesn't say +1HD, it states all HD on any given level, so I thought rolling all HD again would be btb - back then my English wasn't good at all, though.

I was wondering just yesterday, which method would be best. I guess rolling all HD tends to produce more "average" results as the high rolls will often "cancel out" the low rolls.
Adding a single HD every level probably may produce more extreme HP results if players roll lucky or unlucky.

[magremore]

Oct 8, 2012 4:18:14 GMT -6 aher said:

And the referee could have freely chosen to assign the results to the six abilities this way, favoring WIS, perhaps urging the player to become a cleric, maybe because the party lacked one.

I’ve mentioned this elsewhere, but I’m no stranger to redundancy... The above is not even true. BTB, the Xylarthen character will progress equally as a magic-user or cleric due to the experience-only ability adjustments detailed lower on the on same page. “Use” 4 WIS for 2 INT and Xylarthen gets his 5% XP boost, ring-a-ding-ding.

Also, what the OP calls the alternative reading seems to be literally correct based on the evidence—Men & Magic is only 34 pages and “in order to” is used 13 other times as a subordinating conjunction, an obvious authorial tendency. But it hardly matters. Based on the rest of the section it’s apparent that “aid [the players] in selecting a role” has to do with how the abilities relate to the classes. And really the fairest reading of “rate each [dice roll] as to various abilities” is something like: determine Strength, roll; determine Intel, roll; determine Wisdom, roll, etc. The exact order has never really mattered, it just makes sense to roll up the abilities following the order they’re given in.

[howandwhy99]

I was thinking about the OP in my response. We roll 3d6 for each stat in preset order & 3d6x10 for gold. That's not optional, that's core design or the game is unbalanced at the start. (It's not about the end result)

Of course you can generate a character as often as you like. I don't have a problem with rolling up a sheet of pregen stats beforehand. (And really the only way to reach most subclasses means using the Min/Max rules)

[Scott Anderson]

Aside from charisma, there's little reason to worry about the several ability scores. They are about 80% qualitative. The meager bonuses you get are almost rounding errors.

But on that basis, it is boring to get all your scores clustered around the 9-12 range. It's nice to have one or more high or low scores to help you imagine what kind of fellow he might be.

[howandwhy99]

Each class is uniquely balanced. The 3 core classes are the 3 core stats: Strength, Intelligence, and Wisdom. High Constitution, Dexterity, and Charisma improve every class. However, thise 3 core classes *mostly* only benefit from their Prime Requisite scores being higher, They can "dump stat" the other two.

Look at the 2:1 vs 3:1 Min/Max trade offs to see Gary's assessment of the needs of each class. IIRC, only these 3 core stats can be raised and lowered, but I could see a 3:1 or even 2:1 trade between the bottom 3 scores to reach a subclass requirement.

And I disagree these mods are rounding errors. Look at To Hit and Saving Throws ranges, those are very small progressions over 1st to 10th, the whole game basically. A +1 or -1 really reveals itself over the long term. The supplements add more subsystems too making Abilities more important as a whole.