The general consensus that "rules-as-written" with respect to D&D 3.5 means

the core rulebooks, followed by their errata, followed by any later published source book and it's errata that does not contradict a previously published rulebook.

This comes from a very strict reading of the blurbs included at the beginning of each source books' errata, without taking into account the clear intent of the provided examples. This text has changed over time (possibly complicating the matter). This text in the Spell Compendium's errata

When the text within a product contradicts itself, our general policy is that the primary source (actual rules text) is correct and any secondary reference (such as a table or character's statistics block) is incorrect. Exceptions to the rule will be called out specifically.

By the understanding of this text espoused on this site, a later published source book cannot change any rules covered in a previous source book; thus allowed only to create something new or add to something (although I would argue; this is change and in many cases, contradictory). This is evident by the discussions on the Rules Compendium. The RC is not even an entirely original work, but one that ties in expansion rules from various source books into on volume. By stating that the RC is in violation; the consensus says the RC sources are too in violation (all the Complete series, et al).

It is my understanding (quite contrary to the "consensus") that this text was to prevent an error in a table or a secondary reference from being propagated when it contradicted the primary reference. This was not intended to prevent source books and errata from changing, expanding, formalizing or otherwise creating contradictions to the existing rules.

This site's consensus invalidates any answer that quotes from any source that "contradicts" a previous source and stifles understanding. Further, it invalidates any answer that quotes an faq or rules of the game article, if "rules-as-written" is in the question tags.

Possible solutions,

1) Convince people to abandon this approach that invalidates much of the published material.

2) Create a new tag, rules-strict and use this for this narrow reading of the errata directive.

3) To avoid confusion, answer any question where a rule is mentioned in more than one reference with a section of rules-blocks;

  • (Strict errata reading) information pertaining to original core books, DMG/PHB + it's own errata, disregard anything else

  • (Errata reading where source book can alter the rule) Expansion
    source book + it's errata [likely only need to look to RC for 99%]

  • (Additional material - technically non-RAW but may aid in
    understanding) Faq, web article, published adventure, et al

This seems a bit extreme, but would at least provide the OP with all the information and they can make an informed decision.

| |
  • \$\begingroup\$ Could you cite this usage, please? \$\endgroup\$ – Brian Ballsun-Stanton Nov 3 '14 at 22:46
  • \$\begingroup\$ Cite what usage? \$\endgroup\$ – Wyrmwood Nov 3 '14 at 23:39
  • \$\begingroup\$ You assert the general consensus. I'd very much like to see a) where this is shown, and b) your answers and the voting reaction that shows that there is downvoting to your viewpoint. \$\endgroup\$ – Brian Ballsun-Stanton Nov 3 '14 at 23:41
  • \$\begingroup\$ mxyzplk found the most glaring one pretty easily. Are you looking for citations for pedantic reasons, or do you disagree with my impression of the consensus? \$\endgroup\$ – Wyrmwood Nov 3 '14 at 23:43
  • 4
    \$\begingroup\$ Given that there are multiple good answers on that question, I disagree with your impression of the consensus. \$\endgroup\$ – Brian Ballsun-Stanton Nov 4 '14 at 0:24
  • 6
    \$\begingroup\$ The most upvoted answer on that question says "WotC says X in errata, but contradict that with Y in the RC" and dedicates plenty of text to the whys and hows of various interpretations and practical applications of the acknowledged mess. \$\endgroup\$ – BESW Nov 4 '14 at 0:43
  • \$\begingroup\$ I don't understand what you are asking. \$\endgroup\$ – Wyrmwood Nov 4 '14 at 2:02
  • \$\begingroup\$ Essentially, you're being asked to demonstrate that your interpretation is actually a consensus, given that there are multiple very good answers to a question on this subject which deals directly in this topic and contradicts your claim of consensus. \$\endgroup\$ – Lord_Gareth Nov 4 '14 at 19:13
  • \$\begingroup\$ My interpretation is NOT the consensus. The most up-voted answer in the link mxyzplk listed is. This comment thread makes no sense to me. Perhaps I should reword the question. \$\endgroup\$ – Wyrmwood Nov 4 '14 at 21:29
  • 1
    \$\begingroup\$ I believe there was a rules-lawyering tag at one point but it was merged into problem-players. Regarding your penultimate point, I feel your choice of wording is more confrontational than it needs to be, which probably isn't helping your cause \$\endgroup\$ – Wibbs Nov 4 '14 at 21:45
  • \$\begingroup\$ @Phil it was merged into [rules-as-written] actually. The meta Q is here. \$\endgroup\$ – doppelgreener Nov 4 '14 at 23:17
  • \$\begingroup\$ Ah my mistake - I knew something came up about it not that long ago \$\endgroup\$ – Wibbs Nov 4 '14 at 23:17
  • 2
    \$\begingroup\$ Re your last edit: this is proposing a strict protocol for how [rules-as-written][dnd-3.5e] questions are allowed to be answered. That's never going to fly. Certainly, a good question might take the form that you propose, but that's what your votes are for. Wanting to use site policy for this is like wanting to use a Howitzer to swat a mosquito. \$\endgroup\$ – SevenSidedDie Nov 5 '14 at 17:37


You are giving an edition specific example, Rules-as-written is system agnostic tag and should not be changed because one edition of D&D handles their rules canon in one particular way.

| |
  • 3
    \$\begingroup\$ When the RAW tag is being used in the context of D&D 3.5, then the Spell Compendium errata he's quoting would naturally be reasonable to cite in an answer as it's part of the rules of that system. \$\endgroup\$ – BESW Nov 3 '14 at 22:04


  1. We don't have the power to dictate how individuals or whole communities interpret a game, which is what you're asking us to do.

  2. Even if we did have the power, using it in this matter contradicts our tags policy: tagging reflects how the wider RPG community uses a term, not how we at RPG.SE might like the term to be used.

To change the meaning of RAW in regards to D&D 3.5e, you have to go out there and change the existing community consensus until the majority of respected voices use it in the way you'd prefer. How to do that is left as an exercise for the reader…

| |
  • \$\begingroup\$ ... and is something I've personally tried and don't recommend, because getting someone to reevaluate their axiom that RAW is a closed and complete system is inherently hard. \$\endgroup\$ – SevenSidedDie Nov 4 '14 at 19:48

I'm going to add an answer here which I think needs stating:

If someone asks for rules-as-written information, going so far as to tag their question with the rules-as-written tag, we can assume that they want answers based on the rules-as-written.

If they want answers that are reasonable when the rules-as-written aren't, which, let's face it, is a lot of the time, they can simply not ask for rules-as-written in the first place.

Your suggestion that the meaning of the rules-as-written tag be extended to include rules that, rules-as-written, aren't valid, is changing the meaning of the tag to be contrary to its name.

If you want a tag that includes all rules regardless of whether they are valid under the rules-as-written, I would absolutely agree that there could be space for it. (Not sure what I'd call it, but that's a different problem.)

| |
  • \$\begingroup\$ This fails to address the problem in the question, because it's doing what the question is objecting to: taking a particular position on how to interpret RAW and speaking of it as if it is the only possible correct position to take. This comment isn't the place to argue for or against the current community consensus about how to interpret RAW in D&D 3.5e... but neither are these answers the place for it. \$\endgroup\$ – SevenSidedDie Nov 5 '14 at 2:39
  • 2
    \$\begingroup\$ What is more reasonable, and what is in fact our standard of best practices, is when someone tags with [rules-as-written] but their question could be answered differently depending on what sources are acceptable, we ask that they clarify what sources are acceptable. And then we answer that. Not some idealised personal conception of what RAW means, not a carefully-negotiated definition, not what we wish RAW meant. We answer their actual question. \$\endgroup\$ – SevenSidedDie Nov 5 '14 at 2:40
  • \$\begingroup\$ @SevenSidedDie Since there tend to only be at most three reasonably-possible sources (the primary source/errata, per the strict definition, supplements asserting primacy, per their text, and the FAQ, per... I don't know, wishful thinking?), I have found it both more useful and more efficient to simply discuss the discrepancy itself and what each has to say. \$\endgroup\$ – KRyan Nov 5 '14 at 13:24
  • \$\begingroup\$ @Miniman We don’t have such a tag because that is the default here. \$\endgroup\$ – KRyan Nov 5 '14 at 13:26
  • \$\begingroup\$ @KRyan Yeah, I was imagining a "how do I…" question where clarifying sources mattered, but for a "how does X work…" question, covering the field like you say is the other best practice. :) \$\endgroup\$ – SevenSidedDie Nov 5 '14 at 15:59
  • \$\begingroup\$ This is a well thought out answer, but does not address the question (assumes the consensus and invalidates the question). \$\endgroup\$ – Wyrmwood Nov 5 '14 at 16:27

I think there's a misunderstanding here.

When the text within a product contradicts itself, our general policy is that the primary source (actual rules text) is correct and any secondary reference (such as a table or character's statistics block) is incorrect. Exceptions to the rule will be called out specifically.

And this:

It is my understanding (quite contrary to the "consensus") that this text was to prevent an error in a table or a secondary reference from being propagated when it contradicted the primary reference. This was not intended to prevent source books and errata from changing, expanding, formalizing or otherwise creating contradictions to the existing rules.

You're right. That is what that text means. An example case is one of the classes in Complete Divine, where the text says "each level, gain a spellcasting level", but the table only shows it every other level. The text you quoted is meant to fix that type of editing error.

What you're disputing is different text:

Another example of primary [versus] secondary sources involves book and topic precedence. The Player's Handbook, for example, gives all the rules for playing the game, for playing PC races, and for using base class descriptions. If you find something on one of those topics from the Dungeon Master's Guide or the Monster Manual that disagrees with the Player's Handbook, you should assume the Player's Handbook is the primary source. The Dungeon Master's Guide is the primary source for topics such as magic item descriptions, special material construction rules, and so on. The Monster Manual is the primary source for monster descriptions, templates, and supernatural, extraordinary, and spell-like abilities.

The whole mess is outlined effectively in this answer. But in a nutshell, what happened is the books and the errata say "the books and their errata are primary and overrule everything else". So if what the book says and what a web article or FAQ entry say, even if written by the same person who wrote the rules, even if the article/FAQ came out later, and even if it was specifically created to fix some confusing nonsense in the books, the books win. It reaches extreme nonsense levels when the Rules Compendium gets involved...

Don't Worry So Much About It!

The spirit of is that you should answer the question citing the rules. For that purpose, just ignore the FAQ/web articles and stick to books when dealing with D&D 3.5, and you will be doing the right thing 99% of the time (cases where a published book contradict another published book in a way that'd mess this up are rare enough that you shouldn't lose sleep over it).

If the RAW answer results in something that doesn't make sense, it's perfectly acceptable to add another section to your answer addressing that. Here's a simple example where I did that.

The first part of the answer addresses the question by RAW (no rule quotes are needed since the question already quotes the relevant rule). The second part is how I'd play it in a real game, because the RAW is kinda weird here.

It's in that second part where if you want to bring in secondary sources (or even homebrew, like KRyan's alternate flying rules), you should feel free to do so. Just make sure that you've already answered the RAW question with primary sources before you do that. I've used this approach successfully for quite a while

Yep, it can be a bit quirky. But quirky describes a lot of 3.5. ;)

Other Games

Other games don't handle this the same way. Pathfinder is the most obvious example, as one of the game designers making a ruling on the Paizo forum is considered valid RAW. 20pfsrd.org will often be updated with those rulings, and there's a FAQ list out there as well.

If you're not sure what counts as RAW for a given game, then you can list what source you're using, or ask in a comment on the question what sources are considered acceptable.

| |
  • \$\begingroup\$ Yeah, but I don't agree that's what point 2 means. RC is nothing more than a reprint, so this issue is in ALL the expansion material. I do not think the designers ever intended to prevent rules-creep (it's their livelihood...) What they are trying to prevent is an erroneous reference to a rule from being misconstrued as a rules change. Specifically called out rules changes (that is, exceptions) were never intended to be prevented. \$\endgroup\$ – Wyrmwood Nov 5 '14 at 15:57
  • \$\begingroup\$ @Wyrmwood I don't think you're wrong, and I find the whole RC issue to defy common sense, personally. In fact I use the RC in my [rules-as-written] answers. I make sure to say that's my source, and the question asker can decide if they consider that useful or not. Usually, they do, as the RC doesn't actually contradict other sources that often (it expands and clarifies far more often than it contradicts). In practice, I think it's less of a problem than it is in theory. :) Generally speaking if you use published books & the SRD for your answer, you will be fine 99% of the time. \$\endgroup\$ – Tridus Nov 5 '14 at 16:01
  • \$\begingroup\$ @Wyrmwood The Rules Compendium is not exclusively a compilation of rules from other sources in one volume. Often when the Rules Compendium differs from the core rules, the Rules Compendium is the first and only place those rules ever appear. \$\endgroup\$ – Hey I Can Chan Nov 5 '14 at 18:54

@JoshuaAslanSmith is correct in that the tag is of general application and so there's not really anything related to the tag we can/should do to mitigate this problem.

Now, in terms of your real problem (people not accepting later books in their understanding of RAW on 3.5e questions) - yes, you're right, it seems like the majority of vocal 3.5e folks on this site have a very... strict reading of the books and don't recognize newer books, the FAQ, the Rules Compendium, etc. Does the Rules Compendium overrule the core books? illustrates this. You could hold a referendum just on changing the site attitude here, but given the voting on that fairly recent question I suspect you won't make headway. I strongly agree with you that this stance places some strange form of barracks-lawyering well above the value of practical, working game fun, but that appears to be the predominant local 3.5e players' mode d'emploi, so you just have to accept that at some level.

I would recommend just posting your answers and sticking to your guns. I personally vote up an answer that has a reasonable basis in "The FAQ says" or "A later 3.5e book says." The answer to the RC question has nothing to do with the validity of an answer of that sort. It may pull downvotes from those who disagree, and that's the risk you take. Feel free and flag/point out anywhere that overzealous community members may be crossing the line from voting their conscience to trying to muffle reasonable viewpoints and the mods will take action. Of course, there are questions where someone's after super-strict RAW, and other kinds of answers might not necessarily be welcome there. But for those asking for "I don't understand, how's this supposed to work," answers based on those other sources are IMO super useful.

| |
  • 2
    \$\begingroup\$ Ah yes, the D&D rules lawyer equivalent of "is god powerful enough to create a rock that even he can't lift?" That question really showcases the absurdity of the whole thing nicely. \$\endgroup\$ – Tridus Nov 4 '14 at 15:35
  • 1
    \$\begingroup\$ Yeah, I have to say I don't think a lot of the argument that "since the first books don't say later books can supersede them they can't - even if the designers insist in the text of that book it does!" It's ridiculous, but if that's how that community rolls, then c'est la vie. \$\endgroup\$ – mxyzplk - SE stop being evil Nov 4 '14 at 19:02
  • 1
    \$\begingroup\$ It should be noted that in the specific case of 3.5, WotC went out of their way to define FAQ rulings as being non-official and non-binding, which is good because they're usually wrong. Also: what in all nine blasphemous circles of Hell is "barracks-lawyering"? \$\endgroup\$ – Lord_Gareth Nov 4 '14 at 19:19
  • 4
    \$\begingroup\$ I'll let you Google the definition on your own, it's easy. And sure, the FAQ may be "non-official and non-binding" - but what I think you miss too often is the context. If it's someone wanting to do some pure RAW rules lawyering, then sure, but sometimes people are trying to actually play the game and are confused and want to maybe know what a reasonable rule or intent was - in that case, "it says X in the D&D FAQ" is certainly a reasonable response, and more authoritative than "some random Internet dude says something different" IMO. Which is why I distinguish between the 2 cases above. \$\endgroup\$ – mxyzplk - SE stop being evil Nov 4 '14 at 19:27
  • 1
    \$\begingroup\$ mxyzplk, it's called answering the question asked. If the question-asker tagged their question rules-as-written or otherwise indicated that RAW was what they were after, it is not your place to judge them for wanting that information. Which is what you routinely do, such as in this answer for instance. You could challenge their question, but per your own meta thread on that subject you have to answer their question first and rudeness and attacks on people who play the game differently from you (or even don't, and want to know for other reasons!) is never appropriate. \$\endgroup\$ – KRyan Nov 4 '14 at 23:41
  • \$\begingroup\$ You all take your RAW-is-law philosophy to all 3.5e (and many other) questions whether it's RAW-seeking or not, and oh look, people end up asking questions on meta because of how hostile it is. You are completely right, I have no respect for it. This question isn't someone seeking RAW, if you haven't noticed, despite your crusade to spread your particular fundamentalist approach to gaming everywhere. I know you two don't understand that other people don't subscribe to that and that we should be shouted down whenever it occurs, but... Too bad. \$\endgroup\$ – mxyzplk - SE stop being evil Nov 5 '14 at 0:15
  • \$\begingroup\$ If "You all take your RAW-is-law philosophy to all 3.5e (and many other) questions whether it's RAW-seeking or not" is true, it is an important issue, but it's an entirely different issue to this one. This question isn't "people end up asking questions on meta because of how hostile it is", it's asking specifically about rules-as-written questions. \$\endgroup\$ – Miniman Nov 5 '14 at 0:41

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .