Spellcraft cannot be simply reading the V and S components of a spell

[Boci]

So a while ago I made a thread about problems with assuming that spellcraft involved reading verbal and somatic components to identify the spells. I had a problem with this interpretation, explaining that it affectively created a universal dual language of magic (words and handsigns) that everybody who cats magic, human and lithid, divine and arcane, needs to speak pretty precisely to cast spells. To put it mildly I found this unrealistic. Not impossible, but it would need to be a major feature of a setting and would need to be addressed, not implied through another rule and never spoken of directly.

I also provided an alternative fluff for how it worked: metasense. Sense the, for want of a better word, “energy markers” inherent in the nature of magic. Thus the skill becomes a 6th sense that deals with magical energy, which are made clearest when the spell is cast and communicate enough information that the caster can know the exact attributes of the spell.

In the thread, I outlined some problems I believed the UDLoM. There was some discussion and then I had the bright idea of actually reading the full of the spellcraft skill. I came up with these:

1. Identify a spell that’s already in place and in effect. You must be able to see or detect the effects of the spell. No action required. No retry.

2. Identify materials created or shaped by magic, such as noting that an iron wall is the result of a wall of iron spell. No action required. No retry.

3. After rolling a saving throw against a spell targeted on you, determine what that spell was. No action required. No retry.

Which cannot be explained under the UDLoM, since neither language is present to be witnessed for the above uses of the skill. Unfortunately no one ever responded to that post and the thread died.

So this is just me checking: have I missed something, or is that a major problem with the original theory? What do you think of my alternate theory?

[ericgrau]

Your theory works for 1-3 because that is exactly what is going on: identifying tell tale signs of magic. Like a gun specialist looking at some powder burn scorch marks and knowing which guns make marks in that shape.

V and S components, and spell preparation still work by the "UDLoM". Yes, that means spellcraft is for more than only UDLoM. On top of that I did notice some confusion caused by the term "UDLoM" recently though. You couldn't talk about the weather with it, because its "words" (actually symbols) only represent casting actions including gestures and sounds.

Like in the Naruto series to cast jutsu they form hand signs in series while channeling energy, for example fireball is "Serpent, Sheep, Monkey, Boar, Horse, Tiger". A spellbook might simply contain a series of symbols that mirror the gestures and words involved not only in casting but also in preparation. An experienced spellcaster could look at the way someone else writes their symbols and eventually figure out how they correspond. Then he writes them down in his own way. That's why other spellbooks aren't directly readable like a language, they must be deciphered or divined.

[Boci]

V and S components, and spell preparation work by the "UDLoM".

But why not make everything rely off metasense? Why even use the UDLoM when there are so many problems with it? (spell casters without spellcraft, druids with natural spell, the fact that in order to recognize a spell based on its components you need to have knowledge of every spell you can potentially identify, even if you cannot always do so in the heat of the moment).

Why not just use metasense for every use of spellcraft?

[Scow2]

But why not make everything rely off metasense? Why even use the UDLoM when there are so many problems with it? (spell casters without spellcraft, druids with natural spell, the fact that in order to recognize a spell based on its components you need to have knowledge of every spell you can potentially identify, even if you cannot always do so in the heat of the moment).

Why not just use metasense for every use of spellcraft?

Because Spellcraft isn't Detect Magic.

[Lord_Gareth]

Because Spellcraft isn't Detect Magic.

Pretty much this. Justifying it through metasense treads on DMagic's toes.

Besides, don't you need to interact with a spell to identify it?

[Boci]

Because Spellcraft isn't Detect Magic.

I know, and it will stay that way whether metasense or the UDLoM and metasense. This is a fluff issue, not a mechanical one.


Pretty much this. Justifying it through metasense treads on DMagic's toes.

What do you mean by that?


PBesides, don't you need to interact with a spell to identify it?

Pretty sure that's to get a save vs. an illusion.

[Lord_Gareth]

Well, turns out it's another option (http://www.d20srd.org/srd/skills/spellcraft.htm), but more difficult.

Personally, I feel as though Spellcraft identifying verbal/somatic components is about recognizing things that are common to all magic, then identifying things from there. For example, in my settings any language can be used to cast spells (some mages superstitiously try to suit language to spell, like using Ignan for fire spells), but all verbal components have a sort of double-voice aspect, as though another voice is speaking just under or just over the mage in question. From there, the mage analyzes factors common to certain magical effects (fireball's voice is a low bass and is spoken with a rapid in-and-out breath control) to help identify the spell.

[Boci]

Well, turns out it's another option (http://www.d20srd.org/srd/skills/spellcraft.htm), but more difficult.

Still nothing about interaction, I just have to see the affect


Personally, I feel as though Spellcraft identifying verbal/somatic components is about recognizing things that are common to all magic, then identifying things from there. For example, in my settings any language can be used to cast spells (some mages superstitiously try to suit language to spell, like using Ignan for fire spells), but all verbal components have a sort of double-voice aspect, as though another voice is speaking just under or just over the mage in question. From there, the mage analyzes factors common to certain magical effects (fireball's voice is a low bass and is spoken with a rapid in-and-out breath control) to help identify the spell.

That's an interesting interpretation, but it still has its problems. How does this translate to sonomatic components, how does spellcraft allow you to identify a spell without ever seeing its components (+ the knowledge of all spells, but that's not neccissarily a problem).

[Lord_Gareth]

First, the 'interaction bit' is "25 + spell level - After rolling a saving throw against a spell targeted on you, determine what that spell was. No action required. No retry.".

Secondly, I try to emphasize themes in somatic components as well. Sight-based spells (Detect Evil, Darkvision, et cetera) involve touching the eyes, evocations make sculpting motions (shaping the energy you're creating), illusions have small, subtle gestures, etc. You can see the spell gather and form in or around the spellcaster, and failed spells leave a short-lived "spelltaint" that scatters like dust.

As far as identifying spells that you've never seen before, it's less about knowing "He's casting avascular mass!" and more about seeing and/or hearing the incantation and using one's grasp of spellcasting theory (Spellcraft ranks) to understand the effect they're creating. You might not know it's called Avascular Mass when you identify it, but you know that something horrible is about to happen to your veins.

[Boci]

First, the 'interaction bit' is "25 + spell level - After rolling a saving throw against a spell targeted on you, determine what that spell was. No action required. No retry."

Fair enough. But there is still the "Identify a spell that’s already in place and in effect". How would that work in your system?


Secondly, I try to emphasize themes in somatic components as well. Sight-based spells (Detect Evil, Darkvision, et cetera) involve touching the eyes,

How many spells are there that are sight based, and how many ways can you touch your eyes in a manner unique enough that others can identify the spell?


As far as identifying spells that you've never seen before, it's less about knowing "He's casting avascular mass!" and more about seeing and/or hearing the incantation and using one's grasp of spellcasting theory (Spellcraft ranks) to understand the effect they're creating. You might not know it's called Avascular Mass when you identify it, but you know that something horrible is about to happen to your veins.

You know a bit more than that. You know its duration, whether it offer's spell resistence, its range, who it targets. You know everything about the spell. This strains even metasense, but at least there its more plausable that you know so much.

You're idea sounds like it would work great if spellcraft worked like knowledge, and for every 5 points by which you beat the DC you get an extra piece of info about it. But it doesn't work that way. It's all or nothing.

[ericgrau]

You are basically identifying a spell by mundane means. When that spell hasn't even happened yet, or basing it on information unrelated to the spell's effect, at minimum you need at least some understanding of the words and gestures themselves to figure it out. Maybe it's indirect rather than direct, but it must still be there.

Claiming that you don't understand the gestures at all and yet figuring it out on the spot would be harder, not easier, than figuring it out based on the spell's effect. If not impossible. And yet the DC is lower not higher.

[Boci]

You are basically identifying a spell by mundane means. When that spell hasn't even happened yet, or basing it on information unrelated to the spell's effect, at minimum you need at least some understanding of the words and gestures themselves to figure it out. Maybe it's indirect rather than direct, but it must still be there.

Claiming that you don't understand the gestures at all and yet figuring it out on the spot would be harder, not easier, than figuring it out based on the spell's effect. If not impossible. And yet the DC is lower not higher.

Sorry, but I don't understand the point you are trying to make. Could you try rephrasing?

[Lord_Gareth]

Fair enough. But there is still the "Identify a spell that’s already in place and in effect". How would that work in your system?

Check it against your knowledge of spells in general first, then compare it to spells that look similar to see where the differences are. Of course, it's difficult to do if you can't actually see what's going on, but thankfully that's modeled pretty well by the rules (so I might not, for example, be able to define a Protection from Evil until or unless I see it repulsing an evil summoned creature).


How many spells are there that are sight based, and how many ways can you touch your eyes in a manner unique enough that others can identify the spell?

Detect Alignment - Smear eyes with aligned spelltaint, which evaporates.

Darkvision - Cover eyes briefly

True Seeing - Draw symbol around eye in evaporating spelltaint.

Etc, etc, etc.


You know a bit more than that. You know its duration, whether it offer's spell resistence, its range, who it targets. You know everything about the spell. This strains even metasense, but at least there its more plausable that you know so much.

How much can you learn about an object by observing its impact crater, even if the majority of the object is gone? How much can you learn about a person by observing the tracks they leave? Spellcraft is similar, and it's true that it does get harder and harder, requiring more ranks to produce a similar effect. Low level spells are easy, and more common knowledge because they're more commonly practiced. By the time you're identifying these traits about high level spells using inference and comparison you're an expert or even a leader in your field on the Prime Material Plane, writing books that the 2nd level PoS apprentices study to learn Spellcraft from.

Does it seem odd that you can identify a spell, thoroughly, that you've never encountered before? Sure. But that's evidently how defined spellcasting works in D&D, so the idea behind my interpretation is that these have cues (audible, somatic) that can help the experienced magical theorist identify them even if they don't cast magic themselves - which the synergies for Spellcraft seem to agree with.

[LTwerewolf]

I work spellcraft where you can only identify spells of the type you can cast. Wizards can only identify arcane, clerics divine, etc. At higher levels each can start identifying some of the trivial spells of the other. For example a lvl 12 wizard can pretty easily figure out a cleric casting CLW.

[Boci]

Check it against your knowledge of spells in general first

But you don't have to have knowledge of the spell?


Detect Alignment - Smear eyes with aligned spelltaint, which evaporates.

Darkvision - Cover eyes briefly

True Seeing - Draw symbol around eye in evaporating spelltaint.

Etc, etc, etc.

3 problems. 1: why is is spellcraft untrained? This is pretty elementary. 2: how can a spellcaster fail to identify a spell they cast the last round with these kind of tells? 3: swift action and quickened spells. How are you making these gestures in 0.5 seconds?


How much can you learn about an object by observing its impact crater, even if the majority of the object is gone? How much can you learn about a person by observing the tracks they leave? Spellcraft is similar,

Similar maybe, but not the same. How are you knowing the range of an unknown spell just to by seeing it being cast?


I work spellcraft where you can only identify spells of the type you can cast. Wizards can only identify arcane, clerics divine, etc. At higher levels each can start identifying some of the trivial spells of the other. For example a lvl 12 wizard can pretty easily figure out a cleric casting CLW.

Interesting idea. What are the exact mechanics for a wizard IDing divine spells? What about spell-like abilities?

[LTwerewolf]

I think I can explain why someone wouldn't be able to identify a spell they just cast. It's the same reason that you need a spellcraft check to read another wizard's spellbook. I see them all as doing something slightly different to get the same effect. There might be a different accent or even an entirely different gesture involved.


Interesting idea. What are the exact mechanics for a wizard IDing divine spells? What about spell-like abilities?

I haven't actually gotten around to telling the players yet (they know the general concept, but I keep a lot close to the chest, because their characters wouldn't know exactly either) but I do half spell level (rounded up) -1. So if you can cast 8's, you can identify 3's.

[Boci]

I think I can explain why someone wouldn't be able to identify a spell they just cast. It's the same reason that you need a spellcraft check to read another wizard's spellbook. I see them all as doing something slightly different to get the same effect. There might be a different accent or even an entirely different gesture involved.

But if its an entierly different gesture involved, how is it successfully ID? And if its only subtally different, well he's covering his eyes in a subtally different manner. I think I can still guess the spell he's casting.


I haven't actually gotten around to telling the players yet (they know the general concept, but I keep a lot close to the chest, because their characters wouldn't know exactly either) but I do half spell level (rounded up) -1. So if you can cast 8's, you can identify 3's.

Interesting. Tying it to the level of the spell you can castalso delays the sorceror, which is arguably apropriate given their lesser acedemin nature.

[LTwerewolf]

But if its an entierly different gesture involved, how is it successfully ID? And if its only subtally different, well he's covering his eyes in a subtally different manner. I think I can still guess the spell he's casting.

Successfully ID'd because you can kind of figure it out as it happens. Basically it's making an epic educated guess.



Interesting. Tying it to the level of the spell you can castalso delays the sorceror, which is arguably apropriate given their lesser acedemin nature.

Admittedly it's the first campaign we've been using it, however it's been pretty successful thus far. We haven't had a sorcerer in a campaign since we started using it, so I hadn't thought about that, but I still think it applies pretty well.

[Boci]

Successfully ID'd because you can kind of figure it out as it happens. Basically it's making an epic educated guess.

And epic educated guess, on a spell you've potentially never heard of before, that nets you school, range, target, duration, whether or not it allows SR or a save and affect?

[LTwerewolf]

And epic educated guess, on a spell you've potentially never heard of before, that nets you school, range, target, duration, whether or not it allows SR or a save and affect?

Kind of like words of power, there are some that just mean something. Like this word means fire, that word means storm, etc. Even on successful spellcrafts I don't always tell people every aspect of the spell, such as duration. We houserule a lot of things like that. For example in my game if you cast identify on everything willy-nilly you're bound to come across an artifact and have to start making saves vs confusion, blindness, etc. A knowledge check of 35834689348967324967438957259872354 still won't let you know this super obscure thing that was never written down 76,000 years ago by the gods.

[Boci]

Kind of like words of power, there are some that just mean something.

Spellcraft is too specific for that though. You seem to have taken steps against that, but as it you said, that's a houserule.

[LTwerewolf]

Spellcraft is too specific for that though. You seem to have taken steps against that, but as it you said, that's a houserule.

Spellcraft is just one of those that's just too broken (like knowledges, diplomacy, and several others) not to house rule.

[Qwertystop]

3 problems. 1: why is is spellcraft untrained? This is pretty elementary. 2: how can a spellcaster fail to identify a spell they cast the last round with these kind of tells? 3: swift action and quickened spells. How are you making these gestures in 0.5 seconds?

1: Well, firstly, it really isn't that elementary. Especially since there's a good case that these are simplified. For example, casting Blindness could also be covering the eyes, Darkvision could be shading them, some other spell could be shading them with little fingers curled in, etc.

2: Let's say you're not looking directly at their eyes when they do it, so you can't tell more than "eye-related... maybe". Or they're twiddling the fingers on their other hand. Or you blinked.

3: That's metamagic and specific spells made for short casting. Different, simpler gestures.

[Boci]

1: Well, firstly, it really isn't that elementary. Especially since there's a good case that these are simplified. For example, casting Blindness could also be covering the eyes, Darkvision could be shading them, some other spell could be shading them with little fingers curled in, etc.

This is covering one problem by pointing to another. There are a lot of spells, and only so many ways you can covered your eyes, only so many unique jestures you can make.


2: Let's say you're not looking directly at their eyes when they do it, so you can't tell more than "eye-related... maybe". Or they're twiddling the fingers on their other hand. Or you blinked.

And this happens every single time someone casts a spell?


3: That's metamagic and specific spells made for short casting. Different, simpler gestures.

Which are no harder to identify. Plus that means each spell that 1 full round or standard action casting time needs 2 sets of components, one normal and 1 for quicken.


Spellcraft is just one of those that's just too broken (like knowledges, diplomacy, and several others) not to house rule.

I've heard that said about diplomacy before, but never spellcraft or knowledges.

[Lord_Gareth]

Boci, at this point it seems like you're suffering from confirmation bias. The theory is sound - yes, there's only so many unique gestures that involve your eyes, but how much do you wanna bet that the number of spells involving sight is much smaller? As far as Spellcraft being untrained, it's not (http://www.d20srd.org/srd/skills/spellcraft.htm) (it's defined as Int-Based, Trained Only). Did you fail the check to identify a spell someone else cast? Maybe the differences in how they cast it (using a different language threw off how you interpreted the voice inflection, greater or lesser dexterity made the gesture seem subtly different, differing approaches in magical theory or alignment make the spelltaint look odd) are to blame. The point is that the interpretation is internally consistent, which means that you CAN use Spellcraft as-written without having to hack in some metasense.

As far as metamagic goes, that depends on how you want to interpret the metamagic itself. Maybe when you memorize a Quickened spell you're using complex magical theory to shorten the final gesture needed to fire it off. Metamagic makes the spell more complex (higher level), so interpretations of that sort tend to fly with me.

[Spuddles]

Universal magic languages are hinted at- it's why wizards get draconic for free. In the dragon books (rotd, draconomnomnomnomicon, dragon magic) it's made relatively clear that the hyper precision and complexity of the draconic language is due to it being fundamental to the underlying reality, blah etc blah.

I figure casting spells with waiving your hands around is the same as swinging a sword- it doesn't matter if you're illithid or human, the sword is still a sword and a swing is still a swing. Higher level spells are harder to identify, just like it's harder to catch what a master swordsman is doing while he's doing it.

If magic is a real, physical piece of reality, then it'll be just like math or physics. 2+2=4, rocks fall faster than feathers, etc.

That different spellcasters use different notation to record how to perform their judo chop spellcasting is reflected in the spellcraft checks required to decipher their writing.

It's interesting this thread showed up cause I have been thinking about this today. The reason you need to see/hear a spell being cast to identify it is because the spell doesn't exist yet. There is literally nothing to identify the spell by. For instance, fireball is a very obvious spell, but until you see the bead and the detonation, you won't know what it is.

[Boci]

Boci, at this point it seems like you're suffering from confirmation bias. The theory is sound - yes, there's only so many unique gestures that involve your eyes, but how much do you wanna bet that the number of spells involving sight is much smaller?

Given that each handsign needs:

To be unique to the spell
Distinct enough to make that spell recognizable
Complex enough to require special training to recognize
Performable in as little as 0.5 seconds

Yeah, I fell that's a pretty safe bet IMO.


As far as Spellcraft being untrained, it's not (http://www.d20srd.org/srd/skills/spellcraft.htm) (it's defined as Int-Based, Trained Only). Did you fail the check to identify a spell someone else cast?

What if you don't have spellcraft ranks? Then its not a matter that you failed, you can never succeed, its impossible, even with genuis levels of intelligence.


The point is that the interpretation is internally consistent, which means that you CAN use Spellcraft as-written without having to hack in some metasense.

How many problems have I brough up that would be solved with metasense?

I'm not seeing component reading spellcraft as "internally consistent".


As far as metamagic goes, that depends on how you want to interpret the metamagic itself. Maybe when you memorize a Quickened spell you're using complex magical theory to shorten the final gesture needed to fire it off.

But doesn't affect the spellcraft check to identify the spell in anyway. Plus this means you need multiple components for almost every spell. More when you consider the components of a wildshaped druid.

[Scow2]

I've heard that said about diplomacy before, but never spellcraft or knowledges.
Knowledges are broken because Bear Lore. And people not being able to identify their own species.

[Spuddles]

Knowledges are broken because Bear Lore. And people not being able to identify their own species.

Explains the demiraces....

[Bakkan]

I would just like to jump in here and confirm that the number of possible hand motions is far and away greater than the number of spells out there. Let's consider the "covering the eyes" option. A person puts their hand near their eyes. Their palm could be toward their face, away from their face, up, or down (there are lots of other options, but lets leave it at that). Each one of their five fingers could be straight, partially bent, or totally bent (again we're assuming very few distint positions are possible). During the casting of the spell, the hand is staying still or moving in one of the six cardinal directions (up, down, left, right, forward, back).

This gives us
4x3x3x3x3x3x6 = 5832. There aren't that many distinct spells in all of 3.5 if you assume that the somatic component is the same for all casting classes, and there are fewer than twice as many as that if you allow for the sorcerer to have a different movement than the cleric, for instance.

So, yes, it is completely plausible that each spell for each class has two versions (one for quickened, one for normal) distinct from each other and every other class-spell combination in the game.

[Boci]

I would just like to jump in here and confirm that the number of possible hand motions is far and away greater than the number of spells out there.

What about sentient creatures with less, more or no fingers? Can they not perform sonomatic components? Even if you rule they cannot by default, druids with natural spell can. There's also beholder mages. Probably some other aberrations.

[Bakkan]

Given any sort of body capable of manipluating its environment with more detail than just "launch your body into it" you could come up with practially unlimited different motions that could be used. Multiplication is just that powerful.

[Boci]

Given any sort of body capable of manipluating its environment with more detail than just "launch your body into it" you could come up with practially unlimited different motions that could be used. Multiplication is just that powerful.

But there is more than just that. All signs need to be unique, usuable by any creature who casts spells, distinct enough to be identified, but complex enough that it requires training to do, to the point where it is impossible without. I'm gona need a bit more than "Multiplication is just that powerful" to swallow that tibit of fluff.

[Bakkan]

The motions don't have to be the same for different creature or body types. Spellcraft is Trained Only preciesly because a beholder's eye-stalk movements can't replicate or be replicated by the movememnts of a human hand, or a bear's paw, or an Awakened squid's tentacle. Spellcraft training allows you to idenitfy the spell even when an unusual creature is casting it.

[Boci]

The motions don't have to be the same for different creature or body types. Spellcraft is Trained Only preciesly because a beholder's eye-stalk movements can't replicate or be replicated by the movememnts of a human hand, or a bear's paw, or an Awakened squid's tentacle. Spellcraft training allows you to idenitfy the spell even when an unusual creature is casting it.

But how? Remeber you can identify spells you never seen before (unless all casters know of every spell they can identify).

So how are you identifying the spell based on the movement of a beholder's eye stalk? You need an encyclopedia for the component reading fluff of spellcraft to work.

[Pickford]

What about sentient creatures with less, more or no fingers? Can they not perform sonomatic components? Even if you rule they cannot by default, druids with natural spell can. There's also beholder mages. Probably some other aberrations.

That's because the druid is using a body part to emulate the somatic component (drawing in dirt with a paw, gesturing with a talon, etc...)

So no, if you lack natural spell and lack arms you can't do somatic spells.

"You must have at least one hand free to provide a somatic compoenent."

[Lord_Gareth]

But how? Remeber you can identify spells you never seen before (unless all casters know of every spell they can identify).

So how are you identifying the spell based on the movement of a beholder's eye stalk? You need an encyclopedia for the component reading fluff of spellcraft to work.

Observing the shape and quality of the spelltaint. Noting where the waving of the eyestalks seems similar to the end motion of hands, etc, etc - or, you know, hearing the double-voice incantation.

[Pickford]

But how? Remeber you can identify spells you never seen before (unless all casters know of every spell they can identify).

So how are you identifying the spell based on the movement of a beholder's eye stalk? You need an encyclopedia for the component reading fluff of spellcraft to work.

You can only identify a spell as it is being cast (in order to counterspell it for example) if you are able to perceive the verbal 'or' somatic components.

PHB 180
"A spell-like ability has no verbal, somatic, or material component"

So no, you can't use spellcraft to identify a spell-like ability as it is being cast. Thus the question of beholders and the like is obviated.

[zlefin]

How does spellcraft interact with new spells? Seeing as the rules for independent research allow for the creation of new spells, or variations on existing spells; spellcraft needs to be able to accomodate that.

[Lord_Gareth]

How does spellcraft interact with new spells? Seeing as the rules for independent research allow for the creation of new spells, or variations on existing spells; spellcraft needs to be able to accomodate that.

In the interpretation I used, it works essentially the same way as identifying a spell you've never seen before; you look/listen to the elements of the spell, compare it to similar magics, and deduce what's going on from there.

[Kelb_Panthera]

It seems to me that the most likely explanation is that the "universal language of magic" is a mathematic language. It's not the particular gestures and words that are important but their underlying patterns.

Think about it. Complex mathematical equations can be calculated in several ways to come to the same conclusions.

Mathematical notation varies widely based on the base numeric system (base 8 V base 10), the symbols used (roman numerals V phonecian numerals V katakana numeric symbols) and the process symbols used (addition, subtraction, multiplication, etc can all be indicated by varying symbols). Even how the equation is laid out

1+1=2

Vs

1
+1
2

can be different.

Then consider that all these different equations, no matter how they appear in written form, can be used to describe the angle of an appendage's movement, the length of the arc that the phalanges move through, the cadence and rythm of the words spoken and other such details without any real concern for exactly what those digits look like or how many of them the caser has or what those words actually are.

It becomes even more apparent when you consider the existence of spells expressed as geometric figures or architectural structures.

Since everything has an underlying pattern that can be explained mathematically it just makes sense to me that spellcraft is the skill of recognizing and deciphering those patterns in whatever form they're expressed.

Wizards, Archivists, and Wu-Jen actually calculate the figures in their own ways. Prepared divine casters have the patterns inspired by their faith and spontaneous casters of all kinds are a sort of magical savant that simply intuits how the patterns work without really understanding them.

How does that explanation grab you?

[Duke of Urrel]

I think I can explain why someone wouldn't be able to identify a spell they just cast. It's the same reason that you need a spellcraft check to read another wizard's spellbook. I see them all as doing something slightly different to get the same effect. There might be a different accent or even an entirely different gesture involved.

While I don't share LTWerewolf's house rules, I do agree with LTWerewolf's idea that every individual spellcaster casts the same spell in a different way. I do believe in a single, universally intelligible Language of Magic, but one that is in some respects easier, in others harder to understand than any mundane language.

I'll discuss the traits that make the Language of Magic harder to understand here below, and the traits that make it easier to understand in my next posting.

"Magic words belong to an extraordinary language, the Language of Magic, which is unlike any mundane language. Magic words possess the power, when enunciated correctly and under the right conditions, to make magic happen." […]

"A personal magic word is both the true name of the thing it denotes and a true expression of the creature that speaks or writes it. For everything that exists, there are as many personal magic words as there are spellcasters, and although every personal magic word must sound somewhat like all the others that denote the same thing, it also must be as unique as the spellcaster who utters it. Therefore, every spellcaster finds every other spellcaster’s personal magic words difficult to understand as they are spoken and equally difficult to read as they are written.

Spellcasters of every class learn to pronounce personal magic words, so that they can cast spells and read magic scrolls aloud, but they do not learn how to interpret personal magic words as they are spoken by other spellcasters until they start to acquire Spellcraft skill.

Personal magic words are always written in magic runes; therefore, spellbooks and magic scrolls are written exclusively in magic runes. All characters who have at least one rank of Spellcraft or Use Magic Device skill know how to read some magic runes, and characters who are wizards or who have the Scribe Scroll feat learn how to write some magic runes as well. However, nobody ever masters the entire Language of Magic, either as it is spoken or as it is written. The Read Magic spell makes reading magic runes much easier, and even advanced spellcasters often use it, but no spell makes personal magic words easier to say, to write, or to interpret as they are spoken by anybody other than you.

The incantation that any spellcaster other than you utters to cast a spell always consists of one or more personal magic words that differ slightly from the ones that you would use to cast the same spell. This is why you must ready a mental standard action and use Spellcraft skill to identify a spell as it is being cast by another spellcaster, for example for the purpose of counterspelling." […]

"The same is true of an incantation that any spellcaster other than you reads aloud to discharge a spell from a magic scroll that you have not written. If the magic scroll that another spellcaster reads aloud to you is one that you have written, it is easier for you to identify the incantation as you hear it. You still must ready a mental standard action to do to, but your Spellcraft check automatically succeeds.

The magic runes that any spellcaster other than you has written differ slightly from those that you would write, even to describe exactly the same spell in a spellbook, or to capture exactly the same spell in a magic scroll. This is why it is difficult for you to read a magic scroll that anybody else has written, or a spelltext that any wizard has written, or to transcribe either kind text into your own spellbook if you are a wizard."

[Duke of Urrel]

Now, as promised, I'll talk about the ways in which interpreting the Language of Magic is actually easier in some cases than interpreting words spoken in any mundane language.

The reason I'm sharing all this with you is because I believe we can both argue persuasively that interpreting a spell as it is being cast is an extraordinary ability, if not a supernatural one. It isn't labeled as an extraordinary ability, and doesn't need to be, as far as I'm concerned. After all, we don't call Jump skill an "extraordinary ability" when a high-level rogue makes a Jump check that enables her to grab a ledge 16 feet above the ground (surely breaking all Olympic records, but requiring a Jump check score no higher than 32). But in all honesty, such an achievement using this "mundane" skill is extraordinary, and the use of Spellcraft skill to interpret spellcasting actions is no different.

Realistically, and according to the RAW as I understand them, if you're surrounded by combat noise, a distraction penalty of +5 should be added to the Listen DC of any spoken words, and you must add another +10 to understand the words as well as hear them. Then you must add a range penalty of +1 for every 10 feet that separate you from the speaker.

This raises the question: How many DMs do you suppose require spellcasters routinely to make Listen checks at DC 15 or higher before they make Spellcraft checks to counterspell an enemy's spell?

I'll confess here and now: I don't. If a spell is cast anywhere within counterspelling range – which I assume is the same as the spell's own range or the range of the Dispel Magic spell (depending on your counterspelling method) – I don't require you to make a Listen check before you make a Spellcraft check to counterspell. By waiving the usual Listen check rules, I'm effectively admitting that a spellcaster's power to hear and to interpret magic words is extraordinary.

(I apply the usual Listen check rules only to spells that are cast outside of counterspelling range, or to spells that are cast surreptitiously using Sleight of Hand skill. If an enemy spellcaster makes a Sleight of Hand check to conceal a spellcasting action, I let you make either a Spot check or a Listen check to notice it, whichever is more likely to succeed. If you beat the enemy spellcaster's Sleight of Hand check, you either see or hear the spell well enough to counterspell it.)

With anyone who applies the rules as I have just described them, we can both can argue convincingly that Spellcraft, at least when it is used to interpret spellcasting actions, is an extraordinary ability. I would not go so far as to call it magical or supernatural, however. Like Darkvision and numerous other extraordinary abilities, it violates our laws of physics, but does not for that reason have to be magical.