Over the course of several months I have come across several observations on the comprehensive rules and Oracle texts. I post them here to make them more public and to encourage discussion on what course of action the rules manager should take on any or all of them. The observations follow.
[i]Conditional actions in shortcut proposals[/i]: Jul. 13, 2018: Moot; see comment 29.
_Two-Headed Giant and certain can't-attack-a-player effects_: Now moot as of [i]Dominaria[/i].
_Grafdigger's Cage and the_ Amonkhet _rules update_: Moot because the rulings that led to this issue were removed.
_Triggered abilities during the cleanup step_: Not actually an issue, so stricken out.
_Illegal targets and their active involvement_: Moot because I couldn't find an actual Magic card that illustrated this issue.
_For each ... A unless B_ (e.g., Oath of Lim-Dûl, Read the Runes): To better define how effects of this form are processed, the rules manager should include in the comprehensive rules a rule that applies something similar to the APNAP order rules (C.R. 101.4) to cases where only one player has to make multiple choices at the same time, such as those called for by most cards of the form "For each ..., A unless B".
_C.R. 903.9a: The "appropriate zone"_: I suggest that the rules manager should replace "the appropriate zone" (a particularly poor word choice) in C.R. 903.9a with "the zone where it would otherwise go".
_C.R. 607.2m and name changes_: I suggest that the rules manager should replace "cards named [this object's name]" with, say, just "cards named [name]" (see also C.R. 202.4b), so that the rule unquestionably applies no matter what names the object in question has.
_C.R. 608.2b _: I suggest that the rules manager should replace "parts of a resolving spell's effect" with, say, "parts of a resolving spell's or resolving ability's effect".
"Ninjutsu and C.R. 702.48c": I suggest that the rules manager should edit the text of C.R. 702.48a and C.R. 702.48d to say, in part, "... tapped and attacking the same player or planeswalker as the creature returned this way". (Compare with Kaalia of the Vast or the myriad ability [C.R. 702.115].) This would make C.R. 702.48c a reminder rule, and bring ninjutsu in line with the practice found in other keyword abilities such as bestow and soulbond.
"Leovold's Operative": This merely seeks to clarify an existing card to bring it in line with rulings. According to the rules manager, "[c]ards that have during-draft effects, such as Leovold's Operative, don't follow most of the normal rules for priority, sequentiality, or anything else in Magic: the Gathering" (see also C.R. 905.2a). Therefore, in my opinion, the rules manager should give Leovold's Operative text like "... pass the next booster pack you would draft without drafting a card from it"; the current text is open to more than one interpretation, in my opinion.
"Costs to attack with insufficient mana": I suggest that the rules manager should delete "Once the [active] player has enough mana in his or her mana pool" in C.R. 508.1i (see also C.R. 601.2h, where such a clause is not present).
This is a record of questions yet to be answered by the rules manager:
Does an ability that triggers when a permanent "dies" trigger even if that permanent both dies and leaves the game at the same time? (See also this thread.) Post.
[s]Cryptoplasm still says "...you may have Cryptoplasm become a copy .... If you do, Cryptoplasm gains this ability." Is the ability "gain[ed]" this way (still) copiable by other copy effects? Post.[/s]
If something makes Growth-Chamber Guardian enter the battlefield with a +1/+1 counter on it, does Growth-Chamber Guardian get that counter [i]before[/i] it enters, [i]while[/i] it does so, or [i]after[/i] it does so? Post.
Does Panharmonicon, Naban, or Torpor Orb affect Animation Module's first ability if an appropriate permanent enters the battlefield with a +1/+1 counter? (See also this thread.) Post.
For the purposes of Guardian Beast, does control of a permanent reverting to another player because a control-change effect ends (e.g., "gain control until end of turn") count as that player "gain[ing] control" of that permanent? (See also this thread.) Post.
I think C.R. 704.5 is ambiguous in part; e.g., "devotion to red and blue" may be interpreted to mean "all mana symbols in mana costs of red and/or blue permanents" rather than the intended "all red and/or blue mana symbols in mana costs of permanents". (See also this thread.) Post.
The updated Oracle text for Once More with Feeling includes: "A deck can have only one card..." Presumably "...no more than one card..." is meant? Post.
If Regal Force enters the battlefield, and before its ability resolves, Mind Bend changes "green" in its text to "black", is a card drawn for each black or for each green creature "you control" when the ability resolves? (See also this thread.) Post.
Before a player shuffles, if he or she has drafted Caller of the Untamed and Volatile Chimera, can that player reveal both cards and exile the same creature card (among others) for both? (See also this thread.) Post.
At least nine cards include the formulation "You may cast/play ... for as long as it remains exiled", namely:
The intent of all these cards is to generate a continuous effect that lets a player cast or play a particular card from exile. But arguably, it may not work as expected under the current rules. Namely, since the card is moved out of exile to the stack as the first step of casting it (C.R. 601.2a), the duration "for as long as it remains exiled" ends, and therefore, so does the effect that allows the player to cast or play the card (C.R. 611.2a). Therefore, after the player finishes proposing the spell under C.R. 601.2a-d, the card won't be in exile anymore, but the stack, so that the player arguably isn't allowed to cast that card anymore (C.R. 601.2e). And C.R. 601.3 says, "If that player is no longer allowed to cast that spell after completing its proposal, the casting of the spell is illegal" (see also C.R. 601.2e), so the entire action is reversed to the point before the spell's casting was proposed (and so rendering the cast of the spell not possible) (C.R. 601.2, 720.1). As a consequence of this, the spell never counts as being on the stack if the cast was reversed this way, so the card "remains exiled".
This leads to the absurd result that players can't cast that card at all, which is obviously not what is intended. And there appears to be no basis in the comprehensive rules that allows these effects to work as intended.
Note that none of this applies to playing a land, which works as intended because playing a land is usually a single step: the player simply "puts that land onto the battlefield from the zone it was in", here, the exile zone (C.R. 115.2a).
Does the ante zone always exist even if no card in the format in question can put cards there? That question has no clear answer under the comprehensive rules, in particular C.R. 400.1.
More generally, does a zone exist even if it contains no objects? Even that question is not explicitly answered by the comprehensive rules, even though it's ruled that it does exist in that case.
Under C.R. 704.3, if any state-based actions apply, they are "perform[ed] simultaneously as a single event". One case where the interpretation of this rule matters is if a player would lose the game for having 0 life while controlling and owning Lich's Mirror and owning a Grasp of Fate that exiled an Erebos, God of the Dead owned by an opponent. In this case, Lich's Mirror and Grasp of Fate would be shuffled into their owner's library, then Erebos would return to the battlefield "immediately after" that event and before seven cards are drawn (C.R. 610.3), so that Erebos would be on the battlefield in time to keep Lich's Mirror's controller from gaining life (which normally happens when a player goes from 0 life to 20 [C.R. 118.5]) (C.R. 611.3b, 611.3, 112.6, 101.3).
However, as I originally interpreted the text "simultaneously as a single event", the entire state-based action, including all the actions Lich's Mirror's replacement effect entails, would be condensed into one moment in time -- such that there would be no "time" (as far as the game is concerned) between the start of the state-based actions and their end -- so that Erebos wouldn't return to the battlefield until "immediately after" the entire state-based action is processed (that is, "immediately after" Lich's Mirror's controller goes to 20 life in this case). However, as I confirmed with "nastaboi" on the Ask a Magic Judge chat (#mtgrules), I was mistaken.
This suggests that the wording "... then performs all applicable state-based actions simultaneously as a single event" in C.R. 704.3 can be very counterintuitive, because as it turns out, it's there merely to indicate that state-based actions [i]begin[/i] simultaneously, even if some of those actions are replaced with multi-step events, as is the case with Lich's Mirror's replacement effect. If that is the intent, then such a possible rewording to C.R. 704.3 as "... then all applicable state-based actions begin simultaneously" would resolve my concerns, for it would leave no question that state-based actions [i]begin[/i] at the same time, but the actions of each one still occur sequentially.
[s]_Delayed triggered abilities: One token turns into several_: Whether, for the two tokens created due to Tatsumasa, the Dragon Fang's last ability and Doubling Season, one delayed triggered ability is created for each, or one for both, is not unambiguously answered by the comprehensive rules. This issue can be resolved, among other possibilities, by adding to the comprehensive rules a general rule that if a delayed triggered ability refers to one or more tokens and more tokens than usual are created, then a separate delayed triggered ability is created for each one.[/s]
[i]Conditional actions in shortcut proposals[/i]
C.R. 720.2a contains the following sentence dealing with legal shortcut proposals:
"It can't include conditional actions, where the outcome of a game event determines the next action a player takes."
It seems to me that the intent of this sentence is to disallow "four horsemen" or "Emrakul" shortcuts that involve a player shuffling the library and then involve something that cares about the order of cards in that library. In my opinion, though, these shortcuts are already prohibited by the requirement, in C.R. 720.2a, that shortcuts have "predictable results". Such purported shortcuts as those that--
depend on the outcome of a coin flip, or
depend on the order of cards in a library after it's shuffled,
as well as others that rely on randomness, do not have predictable results, for no matter how many coins are flipped or how many times the library is shuffled, a player is not guaranteed to win a coin flip or have a particular card appear on the top of the library, for example.
Moreover, that sentence appears to rule out such shortcuts as the one sanctioned by the annotated Magic Tournament Rules 2.4 ( http://blogs.magicjudges.org/rules/mtr4-2/ );
"A player may interrupt a tournament shortcut by explaining how he or she is deviating from it or at which point in the middle he or she wishes to take an action. A player may interrupt his or her own shortcut in this manner."
"[[ For example, Player A controls Basalt Monolith and Mesmeric Orb and says, 'I'd like to tap and untap this Monolith 30 times or until I reveal a land.' This suggests a shortcut and a stopping point for interrupting the shortcut.]]"
Because the outcome of a game event (here, putting a land card into the graveyard or some other card) determines the next action a player takes (namely, untapping Basalt Monolith), this shortcut arguably contains a conditional action and is arguably disallowed under the current version of C.R. 720.2a. Unfortunately, it isn't allowed under C.R. 720.2b either, since under that rule, only another player is allowed to "shorten" a shortcut, not the player who proposed the shortcut. (Though the example mentions "interrupting the shortcut", I believe this shortcut would be allowable in the absence of the sentence at issue here.)
Both these cases indicate that the sentence in question ought to be eliminated from C.R. 720.2a, in my opinion -- on the one hand, because it's redundant, and on the other hand, because it appears to disallow too much (and thus removing that sentence would bring C.R. 720.2a more in line with the Magic Tournament Rules).
(I am aware that the Magic Tournament Rules override the comprehensive rules in a sanctioned tournament [M.T.R., "Introduction"], but I feel this change is appropriate not just for sanctioned games, but in more casual games as well -- as is the case, by the way and in my opinion, for comprehensive rules changes that allow players to reveal cards from their hand at any time, that list certain tournament shortcuts, and the like.)
A related question: Is there a example of a shortcut disallowed by the sentence "It can't contain conditional actions...", but which doesn't involve randomness and is not within the scope of the Magic Tournament Rules excerpt I cited?
[i]Two-Headed Giant and certain can't-attack-a-player effects[/i]
EDIT (Apr. 29, 2018): Now moot as of [i]Dominaria[/i].
C.R. 810.7a says that in Two-Headed Giant, "Each team's creatures attack the other team as a group", so they don't attack individual players.
C.R. 810.7c, another Two-Headed Giant rule, says, in relevant part, "If an effect of an object controlled by a defending player prohibits a creature from attacking him or her, that creature can't attack the defending team."
C.R. 810.7c works quite well for nearly all cards that include effects in which creatures "can't attack" particular players, since in these cases, the effect comes from a static ability of a permanent and
are phrased "can't attack you" (since they apply only to the permanent's controller [C.R. 109.5] and end when the permanent leaves the battlefield [C.R. 112.6, 611.3b]). However, I see a few exceptions:
Arboria creates an effect in which creatures can't attack a player that's not necessarily Arboria's controller.
Because of the way this part of C.R. 810.7c is worded, it becomes unclear whether creatures can attack a defending player's team even if--
the effect came from a resolving spell, here either Chronomantic Escape or Illusionist's Gambit. Here, once the effect is created, the spell goes to the graveyard, so arguably isn't controlled by any player anymore.
the effect came from a resolving ability (from Orzhov Advokist or Web of Inertia), but its source is no longer controlled by the defending player (for example, "creatures can't attack" one player, but Orzhov Advokist leaves the battlefield or another player controls it).
the creature can't attack a player who doesn't control Arboria, but can attack Arboria's controller.
Because of C.R. 810.7a, it would be incorrect to clarify the cases mentioned above by saying that the creature can attack the other player on the defending team, since under C.R. 810.7a, creatures attack only teams, not individual players. And it would be similarly incorrect to say the creature can nevertheless attack the defending team, since the defending player in question is part of that team and it would likely come in conflict with the general rule that effects that provide that something "can't" be done prevail (C.R. 101.2).
This shows that this part of C.R. 810.7c is in conflict with C.R. 810.7a and doesn't cover all the cases it's intended to cover.
[i]Grafdigger's Cage after the Amonkhet rules update[/i]
EDIT (Oct. 15): Now moot, since the rulings in question were removed from Gatherer.
[s]
The rulings for Grafdigger's Cage, newly added in [i]Amonkhet[/i], don't appear to correspond with that card's Oracle text, which still says, in part: "Creature [i]cards[/i] can't enter the battlefield from graveyards or libraries" (which would mean that the ability checks the characteristics of the card in the graveyard or library, regardless of what it would be on the battlefield [C.R. 109.2]). However, the rulings (which appear to have been added in response to newly added C.R. 614.16d) suggest that the text was changed to "[i]Creatures[/i] can't enter the battlefield" (which unfortunately isn't the case; compare with Worms of the Earth, whose second ability, "Lands can't enter the battlefield", is governed by new C.R. 614.16d). Moreover, if the rulings stay, that would now unfortunately make Grafdigger's Cage's first ability quite unintuitive, and could even make it ambiguous, especially since "creature card" can now (under a given interpretation of new C.R. 108.2) mean either a creature card in the graveyard or library, on the one hand, or a creature card on the battlefield, on the other (see also C.R. 109.2a, 110.1).[/s]
[i]Illegal targets and their active involvement[/i]
[EDIT (Aug. 21): Soul's Fire turned out to be a poor example; edited it with a hypothetical card similar to Soul's Fire. (I have edited my answer to the thread that led to this issue with a more solid rule justification.)]
EDIT (Oct. 15): I will declare this moot for now since I can't find an actual Magic card where this matters.
[s]
Before [i]Magic Origins[/i], C.R. 608.2b read, in relevant part:
"[I]f any of [a spell or ability's] targets are illegal, the part of the spell or ability's effect for which it is an illegal target can't ... make that target perform any actions."
Since [i]Magic Origins[/i], that part of the rule now reads, in relevant part:
"Illegal targets, if any, won't be affected by parts of a resolving spell's effect for which they're illegal."
(Note that it should probably say "resolving spell's or resolving ability's effect", but that's a separate issue.)
Arguably, this appears to suggest that illegal targets are only kept from being [i]passively[/i] affected by the parts of the effect in question, and that such targets can still be [i]actively[/i] involved in that effect. An example will illustrate this.
Suppose a player casts a hypothetical spell similar to Soul's Fire (but reads instead "Target creature on the battlefield deals 2 damage to target creature or player") targeting a creature and a player, such that the creature will deal damage to the player. Then, before this hypothetical spell resolves, the creature gains shroud. Then, when the spell resolves, the creature will be an illegal target (C.R. 608.2b, 702.18a). In this circumstance, it's a stretch to imagine that being "unaffected" by the damage effect means that the creature won't deal damage, especially since dealing damage is [i]active[/i], whereas receiving damage is [i]passive[/i]. The rules before [i]Magic Origins[/i] made it clear that this hypothetical spell wouldn't make the illegal target deal any damage; that isn't clear anymore in the rules since [i]Magic Origins[/i]. (Contrast this with the case where the player gains shroud rather than the creature. In both versions of the rules, it would be clear that damage wouldn't be dealt to the player -- here, the illegal target, the player, is not "affected" by the hypothetical spell.)
[i]Triggered abilities during the cleanup step[/i]
EDIT (Oct. 15): Now moot as of [i]Ixalan[/i], if not before.
[s]In [i]Kaladesh[/i], the rules were changed to streamline how triggered abilities are handled. But under the current wording of C.R. 514.3a, triggered abilities "waiting to be put on the stack" would be put there below abilities that trigger during the cleanup step's state-based actions (which would go on the stack as the active player gets priority [C.R. 116.5]), a situation that the rule update seeks to avoid. Moreover, the current wording of C.R. 514.3a, as I see it, is defective in that it only says "those triggered abilities are put on the stack", that is, those triggered abilities that "are waiting to be put onto the stack" at the beginning of the process, and not, say, "triggered abilities that are waiting to be put onto the stack are put there", which also covers those abilities that triggered as a result of the state-based actions mentioned in that rule.[/s]
[i]Who decides for a player who has left the game? (RESOLVED)[/i]
Resolved with [i]Archenemy: Nicol Bolas[/i].
[s]As this thread shows, the comprehensive rules don't regulate which player makes a decision if it's not an object, but a game rule, that leads the active player to make a choice, but the active player isn't in the game anymore.[/s]
The intent of all these cards is to generate a continuous effect that lets a player cast or play a particular card from exile. But arguably, it may not work as expected under the current rules. Namely, since the card is moved out of exile to the stack as the first step of casting it (C.R. 601.2a), the duration "for as long as it remains exiled" ends, and therefore, so does the effect that allows the player to cast or play the card (C.R. 611.2a). Therefore, after the player finishes proposing the spell under C.R. 601.2a-d, the card won't be in exile anymore, but the stack, so that the player arguably isn't allowed to cast that card anymore (C.R. 601.2e). And C.R. 601.3 says, "If that player is no longer allowed to cast that spell after completing its proposal, the casting of the spell is illegal" (see also C.R. 601.2e), so the entire action is reversed to the point before the spell's casting was proposed (and so rendering the cast of the spell not possible) (C.R. 601.2, 720.1). As a consequence of this, the spell never counts as being on the stack if the cast was reversed this way, so the card "remains exiled".
This leads to the absurd result that players can't cast that card at all, which is obviously not what is intended. And there appears to be no basis in the comprehensive rules that allows these effects to work as intended.
Note that none of this applies to playing a land, which works as intended because playing a land is usually a single step: the player simply "puts that land onto the battlefield from the zone it was in", here, the exile zone (C.R. 115.2a).
The rule in question doesn't check about legality based on where the card is, because at that point the card is on the stack and you can't cast cards from the stack. If it does check location, it must be checking based on if the spell with those characteristics were in the the zone it was being cast from, which in this case would be a legal spell.
With respect to costs to attack (such as from Propaganda or Ghostly Prison), C.R. 508.1i says "Once the [active] player has enough mana in his or her mana pool, he or she pays all costs in any order." Arguably this can be read as requiring the player to pay costs only if he or she "has enough mana in his or her mana pool", that is, the step is skipped if he or she has only C, but the total cost to attack is 2.
I don't even understand the objection here. Indeed, you don't pay the costs if you can't. How is this unintended behavior?
The rulings for Grafdigger's Cage, newly added in [i]Amonkhet[/i], don't appear to correspond with that card's Oracle text, which still says, in part: "Creature [i]cards[/i] can't enter the battlefield from graveyards or libraries" (which would mean that the ability checks the characteristics of the card in the graveyard or library, regardless of what it would be on the battlefield [C.R. 109.2]). However, the rulings (which appear to have been added in response to newly added C.R. 614.16d) suggest that the text was changed to "[i]Creatures[/i] can't enter the battlefield" (which unfortunately isn't the case; compare with Worms of the Earth, whose second ability, "Lands can't enter the battlefield", is governed by new C.R. 614.16d). Moreover, if the rulings stay, that would now unfortunately make Grafdigger's Cage's first ability quite unintuitive, and could even make it ambiguous, especially since "creature card" can now (under a given interpretation of new C.R. 108.2) mean either a creature card in the graveyard or library, on the one hand, or a creature card on the battlefield, on the other (see also C.R. 109.2a, 110.1).
This is a fair point. They may have forgotten to update the Oracle wording.
As this thread shows, the comprehensive rules don't regulate which player makes a decision if it's not an object, but a game rule, that leads the active player to make a choice, but the active player isn't in the game anymore.
Clearly we need to switch to an object-oriented rules system.
C.R. 810.7a says that in Two Headed Giant, "Each team's creatures attack the other team as a group", so they don't attack individual players.
C.R. 810.7c, another Two-Headed Giant rule, says, in relevant part, "If an effect of an object controlled by a defending player prohibits a creature from attacking him or her, that creature can't attack the defending team."
C.R. 810.7c works quite well for nearly all cards that include effects in which creatures "can't attack" particular players, since in these cases, the effect comes from a static ability of a permanent and
are phrased "can't attack you" (since they apply only to the permanent's controller [C.R. 109.5] and end when the permanent leaves the battlefield [C.R. 112.6, 611.3b]). However, I see a few exceptions:
Arboria creates an effect in which creatures can't attack a player that's not necessarily Arboria's controller.
Because of the way this part of C.R. 810.7c is worded, it becomes unclear whether creatures can attack a defending player's team even if--
the effect came from a resolving spell, here either Chronomantic Escape or Illusionist's Gambit. Here, once the effect is created, the spell goes to the graveyard, so arguably isn't controlled by any player anymore.
the effect came from a resolving ability (from Orzhov Advokist or Web of Inertia), but its source is no longer controlled by the defending player (for example, "creatures can't attack" one player, but Orzhov Advokist leaves the battlefield or another player controls it).
the creature can't attack a player who doesn't control Arboria, but can attack Arboria's controller.
Because of C.R. 810.7a, it would be incorrect to clarify the cases mentioned above by saying that the creature can attack the other player on the defending team, since under C.R. 810.7a, creatures attack only teams, not individual players. And it would be similarly incorrect to say the creature can nevertheless attack the defending team, since the defending player in question is part of that team and it would likely come in conflict with the general rule that effects that provide that something "can't" be done prevail (C.R. 101.2).
This shows that this part of C.R. 810.7c is in conflict with C.R. 810.7a and doesn't cover all the cases it's intended to cover.
It's worth noting that spells or abilities are objects, and the effects created when they resolve should be covered by 870.1c. As for the controller, LKI should be able to see who controlled the spell or ability at the time the effect was created. Arboria does seem to throw a wrench in this, though.
Under C.R. 704.3, if any state-based actions apply, they are "perform[ed] simultaneously as a single event". One case where the interpretation of this rule matters is if a player would lose the game for having 0 life while controlling and owning Lich's Mirror and owning a Grasp of Fate that exiled an Erebos, God of the Dead owned by an opponent. In this case, Lich's Mirror and Grasp of Fate would be shuffled into their owner's library, then Erebos would return to the battlefield "immediately after" that event and before seven cards are drawn (C.R. 610.3), so that Erebos would be on the battlefield in time to keep Lich's Mirror's controller from gaining life (which normally happens when a player goes from 0 life to 20 [C.R. 118.5]) (C.R. 611.3b, 611.3, 112.6, 101.3).
However, as I originally interpreted the text "simultaneously as a single event", the entire state-based action, including all the actions Lich's Mirror's replacement effect entails, would be condensed into one moment in time -- such that there would be no "time" (as far as the game is concerned) between the start of the state-based actions and their end -- so that Erebos wouldn't return to the battlefield until "immediately after" the entire state-based action is processed (that is, "immediately after" Lich's Mirror's controller goes to 20 life in this case). However, as I confirmed with "nastaboi" on the Ask a Magic Judge chat (#mtgrules), I was mistaken.
This suggests that the wording "... then performs all applicable state-based actions simultaneously as a single event" in C.R. 704.3 can be very counterintuitive, because as it turns out, it's there merely to indicate that state-based actions [i]begin[/i] simultaneously, even if some of those actions are replaced with multi-step events, as is the case with Lich's Mirror's replacement effect. If that is the intent, then such a possible rewording to C.R. 704.3 as "... then all applicable state-based actions begin simultaneously" would resolve my concerns, for it would leave no question that state-based actions [i]begin[/i] at the same time, but the actions of each one still occur sequentially.
What if there are multiple state-based actions that have multi-step events? I'll bet someone here can come up with a scenario for that one.
Before [i]Magic Origins[/i], C.R. 608.2b read, in relevant part:
"[I]f any of [a spell or ability's] targets are illegal, the part of the spell or ability's effect for which it is an illegal target can't ... make that target perform any actions."
Since [i]Magic Origins[/i], that part of the rule now reads, in relevant part:
"Illegal targets, if any, won't be affected by parts of a resolving spell's effect for which they're illegal."
(Note that it should probably say "resolving spell's or resolving ability's effect", but that's a separate issue.)
Arguably, this appears to suggest that illegal targets are only kept from being [i]passively[/i] affected by the parts of the effect in question, and that such targets can still be [i]actively[/i] involved in that effect. An example will illustrate this.
Suppose a player casts Soul's Fire targeting a creature and a player, such that the creature will deal damage to the player. Then, before Soul's Fire resolves, the creature gains shroud. Then, when Soul's Fire resolves, the creature will be an illegal target (C.R. 608.2b, 702.18a). In this circumstance, it's a stretch to imagine that being "unaffected" by Soul's Fire's effect means that the creature won't deal damage, especially since dealing damage is [i]active[/i], whereas receiving damage is [i]passive[/i]. The rules before [i]Magic Origins[/i] made it clear that Soul's Fire wouldn't make the illegal target deal any damage; that isn't clear anymore in the rules since [i]Magic Origins[/i]. (Contrast this with the case where the player gains shroud rather than the creature. In both versions of the rules, it would be clear that damage wouldn't be dealt to the player -- here, the illegal target, the player, is not "affected" by Soul's Fire.)
Your argument is spurious. Being forced to deal damage to something is clearly being affected. There's no active/passive distinction. (But you're right about it should say spell or ability.)
Does the ante zone always exist even if no card in the format in question can put cards there? That question has no clear answer under the comprehensive rules, in particular C.R. 400.1.
More generally, does a zone exist even if it contains no objects? Even that question is not explicitly answered by the comprehensive rules, even though it's ruled that it does exist in that case.
I cannot imagine a scenario in which the answer to these questions matters.
C.R. 719.2a contains the following sentence dealing with legal shortcut proposals:
"It can't include conditional actions, where the outcome of a game event determines the next action a player takes."
It seems to me that the intent of this sentence is to disallow "four horsemen" or "Emrakul" shortcuts that involve a player shuffling the library and then involve something that cares about the order of cards in that library. In my opinion, though, these shortcuts are already prohibited by the requirement, in C.R. 719.2a, that shortcuts have "predictable results". Such purported shortcuts as those that--
depend on the outcome of a coin flip, or
depend on the order of cards in a library after it's shuffled,
as well as others that rely on randomness, do not have predictable results, for no matter how many coins are flipped or how many times the library is shuffled, a player is not guaranteed to win a coin flip or have a particular card appear on the top of the library, for example.
Moreover, that sentence appears to rule out such shortcuts as the one sanctioned by the annotated Magic Tournament Rules 2.4 ( http://blogs.magicjudges.org/rules/mtr4-2/ );
"A player may interrupt a tournament shortcut by explaining how he or she is deviating from it or at which point in the middle he or she wishes to take an action. A player may interrupt his or her own shortcut in this manner."
"[[ For example, Player A controls Basalt Monolith and Mesmeric Orb and says, 'I'd like to tap and untap this Monolith 30 times or until I reveal a land.' This suggests a shortcut and a stopping point for interrupting the shortcut.]]"
Because the outcome of a game event (here, revealing a land card or not) determines the next action a player takes (namely, untapping Basalt Monolith), this shortcut arguably contains a conditional action and is arguably disallowed under the current version of C.R. 719.2a. Unfortunately, it isn't allowed under C.R. 719.2b either, since under that rule, only another player is allowed to "shorten" a shortcut, not the player who proposed the shortcut. (Though the example mentions "interrupting the shortcut", I believe this shortcut would be allowable in the absence of the sentence at issue here.)
Both these cases indicate that the sentence in question ought to be eliminated from C.R. 719.2a, in my opinion -- on the one hand, because it's redundant, and on the other hand, because it appears to disallow too much (and thus removing that sentence would bring C.R. 719.2a more in line with the Magic Tournament Rules).
(I am aware that the Magic Tournament Rules override the comprehensive rules in a sanctioned tournament [M.T.R., "Introduction"], but I feel this change is appropriate not just for sanctioned games, but in more casual games as well -- as is the case, by the way and in my opinion, for comprehensive rules changes that allow players to reveal cards from their hand at any time, that list certain tournament shortcuts, and the like.)
A related question: Is there a example of shortcuts disallowed by the sentence "It can't contain conditional actions...", but which doesn't involve randomness and is not within the scope of the Magic Tournament Rules excerpt I cited?
Yeah, that's actually pretty ambiguous. I think the idea they're going for is that it depends on the outcome of a game event not described by the deterministic loop. For example, Emrakul's reshuffle is not a game action described by the loop, but the identity of the card milled is.
In [i]Kaladesh[/i], the rules were changed to streamline how triggered abilities are handled. But under the current wording of C.R. 514.3a, triggered abilities "waiting to be put on the stack" would be put there below abilities that trigger during the cleanup step's state-based actions (which would go on the stack as the active player gets priority [C.R. 116.5]), a situation that the rule update seeks to avoid. Moreover, the current wording of C.R. 514.3a, as I see it, is defective in that it only says "those triggered abilities are put on the stack", that is, those triggered abilities that "are waiting to be put onto the stack" at the beginning of the process, and not, say, "triggered abilities that are waiting to be put onto the stack are put there", which also covers those abilities that triggered as a result of the state-based actions mentioned in that rule.
Technically correct and should probably be fixed, but wow is that pedantic. (Yes, yes, I know that when it comes to rules pedantry is important.)
The ninjutsu ability says the card with the ability enters the battlefield "tapped and attacking" (C.R. 702.48a). Under C.R. 508.4, in this case, its controller chooses a defending player or planeswalker for that creature. But under C.R. 702.48c, the creature with ninjutsu "will be attacking the same player or planeswalker as the creature" returned to hand, which contradicts C.R. 702.48a. According to the rules manager, however, C.R. 702.48c "is not a reminder of other behavior but a descriptor for behavior that would otherwise not exist". In that case, why doesn't the ninjutsu ability say, for example, "... tapped and attacking the same player or planeswalker as the creature returned this way"? (Compare with Kaalia of the Vast or the myriad ability [C.R. 702.115].) As it stands, the text given in C.R. 702.48a is inconsistent with C.R. 702.48c because of C.R. 508.4.
Strictly speaking there's no contradiction. 702.48c restricts the choices you're allowed to make under 508.4. However, I guess that brings up the question of whether 702 keyword rules are supposed to have all of the actual rules content in 702.XXa and the rest of 702.XXx is just clarifications, or if the other 702.XXx are also meaningful.
This is not so much an issue with the comprehensive rules as it is with one card's allegedly vague Oracle text. Under a ruling, using multiple Leovold's Operative for the same booster pack has the player pass that many booster packs (and not merely the "next" booster pack) without drafting from it. But this differs from such effects as "doesn't untap during its controller's next untap step" or "beginning of the next end step". According to the rules manager, however, "[c]ards that have during-draft effects, such as Leovold's Operative, don't follow most of the normal rules for priority, sequentiality, or anything else in Magic: the Gathering" (see also C.R. 905.2a). In my opinion, the intent of Leovold's Operative would be clearer with text like "... pass the next booster pack [i]you would draft[/i] without drafting a card from it".
This all seems correct. "would" is the better phrasing.
[Reply to "Grafdigger's Cage after the Amonkhet rules update"]
This is a fair point. They may have forgotten to update the Oracle wording.
To be fair, Grafdigger's Cage's Oracle text was updated in Amonkhet, but only in its second ability.
[Reply to "Simultaneity of state-based actions"]
What if there are multiple state-based actions that have multi-step events? I'll bet someone here can come up with a scenario for that one.
Consider a scenario where, in addition to the objects mentioned in the Lich's Mirror scenario, a creature under a regeneration effect has been dealt lethal combat damage and, at the same time, the player controlling Lich's Mirror has been dealt enough combat damage to bring his or her life total to 0 or less. Like in Lich's Mirror, regeneration is a multi-step replacement effect: removing damage from the creature, tapping it, and removing that creature from combat (C.R. 701.13a-b).
[Reply to "The existence of zones in the game"]
I cannot imagine a scenario in which the answer to these questions matters.
If anything, a rule that clarifies that each zone always exists would clarify (more strongly than the rules already do), for example, that there is no time when players "have no hand", or that players don't "start a new stack".
[Reply to "Ninjutsu and C.R. 702.48c."]
Strictly speaking there's no contradiction. 702.48c restricts the choices you're allowed to make under 508.4. However, I guess that brings up the question of whether 702 keyword rules are supposed to have all of the actual rules content in 702.XXa and the rest of 702.XXx is just clarifications, or if the other 702.XXx are also meaningful.
In my opinion, C.R. 702.48c should at least be a reminder rule to C.R. 702.48a, which should be changed as I proposed. Notable examples of this policy include the bestow and soulbond keywords (C.R. 702.102, 702.94a).
Consider a scenario where, in addition to the objects mentioned in the Lich's Mirror scenario, a creature with a regeneration effect has been dealt lethal combat damage and, at the same time, the player controlling Lich's Mirror has been dealt enough combat damage to bring his or her life total to 0 or less. Like in Lich's Mirror, regeneration is a multi-step redirection effect: removing combat damage from the creature, tapping it, and removing that creature from combat (C.R. 701.13a-b).
Regeneration doesn't really feel multi-step because none of those are contingent on each other. I'm thinking something more like an Angel of Sanctions also dies to lethal damage as part of the event and it has a Humility under it. Assuming Erebos' controller has enough devotion for it to be a creature, we end up with a sort of race condition on which combination of permanents is actually on the battlefield when the mirror's replacement effect tries to make its controller gain 20 life.
So wait, in response to the "state based actions happen all at once"
What happens if a creature with undying and a +1/+1 counter on it gets a lethal amount of -1/-1 counters put on it? Does the creature die with or without a +1/+1 counter?
So wait, in response to the "state based actions happen all at once"
What happens if a creature with undying and a +1/+1 counter on it gets a lethal amount of -1/-1 counters put on it? Does the creature die with or without a +1/+1 counter?
You should ask such questions in the rules forum. This thread was created in the general forum specifically to discuss issues with the CR, not to answer rules questions.
Anyway, dies triggers look at the game state right before the creature died in order to determine if they trigger. And right before the creature died (so right before the check of state based actions) it had a +1/+1 counter, so undying doesn't trigger.
With Archenemy: Nicol Bolas, a new rule was added after C.R. 800.4f to the comprehensive rules. C.R. 800.4g now reads, "If a rule requires a player who has left the game to make a choice, the next player in turn order makes that choice." Therefore, the issue "Who decides for a player who has left the game?" has been resolved.
The Hour of Devastation set gave me a new issue on the comprehensive rules.
A clarification to C.R. 107.1b is in order
In C.R. 107.1b, I suggest that the rules manager should change "a calculation that would determine the result of an effect" to "a calculation that would determine the meaning of an effect" to help make the intended effect of the change to C.R. 107.1b in Hour of Devastation (to keep the X in such effects as "gets +X/+X ... where X is ..." from being negative) more consistent with the three examples given in that rule.
Paliano Vanguard, Aether Searcher, etc., and C.R. 905.2b
Paliano Vanguard, Noble Banneret, Smuggler Captain, and Aether Searcher include instructions to note information during a draft. However, C.R. 905.2b, which governs such noted information, doesn't appear to apply to any of those cards, since the information isn't noted as Paliano Vanguard itself, etc., is drafted.
The last ability of Chandra, Pyromaster says, in relevant part, "You may cast the copies without paying their mana costs". A ruling on this card says, in relevant part: "While resolving the third ability, ... [y]ou can cast zero, one, two, or all three copies", and this appears to allow the player to cast at least one copy this way even if an Extirpate card (with split second) is copied this way (because only one copy can be cast which renders casting further copies impossible while that copy is on the stack [C.R. 702.60a; see also this thread]), in which case there appears to be no basis for this ruling in the comprehensive rules (according to C.R. 608.2d, the option to "cast the copies", arguably, can't be chosen if not all the copies can be cast; compare Chandra, Pyromaster with Chandra Ablaze, which says "[c]ast any number" of cards meeting certain criteria).
To regard the agent of an occurrence as truly in a separate class from the patient is a language-relative thing. In logical terms, both the dealer and the damaged undergo a dealing relation which Soul's Fire imposes on them (and the damage is involved, too).
I do think that the language for casting legality is sort of clumsy. It is clear that it is trying to accommodate all the fancy ways that permissions are granted to spells - with keywords, with "as-though"s - but it has lost uniformity. Improvement is imaginable.
Private Mod Note
():
Rollback Post to RevisionRollBack
Epic banner by Erasmus of æтђєг.
Awesome avatar provided by Krashbot @ [Epic Graphics].
Trample and assigning less than lethal combat damage
In general, an attacking creature "can't assign combat damage to a creature that's blocking it unless, when combat damage assignments are complete, each creature that precedes that blocking creature in its order is assigned lethal damage" (C.R. 510.1c). However, under C.R. 702.19b, if the attacking creature has trample, its controller "need not assign lethal damage" to those blockers (but "in that case[,] ... can't assign any damage" to defending player or planeswalker) (C.R. 702.19b), so that that player can arguably assign less than lethal combat damage to the blockers (compare with C.R. 510.1c, the general rule). For example, under that interpretation, the following combat damage assignments are possible for a 6/6 attacking creature with trample (assuming the damage assignment order is 4/4 creature then 2/3 creature, and neither creature has damage marked on it):
- 0 damage to the 4/4 creature and 6 damage to the 2/3 creature.
- 0 damage to the 4/4 creature and 3 damage to the 2/3 creature.
- 3 damage to the 4/4 creature and 3 damage to the 2/3 creature.
- 3 damage to the 4/4 creature and 3 damage to the 2/3 creature.
- 4 damage to the 4/4 creature and 1 damage to the 2/3 creature.
- 5 damage to the 4/4 creature.
However, the sentence where the cited part of C.R. 702.19b appears is arguably not meant to imply that the damage assignment order can be ignored, but rather, as I was told, to "clarif[y] that [the attacking player] do[es]n't have to assign lethal damage to each blocker even if [he or she] could". In that case, only the following damage assignments would be possible in the same scenario given earlier:
- 5 damage to the 4/4 creature and 1 damage to the 2/3 creature.
- 6 damage to the 4/4 creature.
- 4 damage to the 4/4 creature and 2 damage to the 2/3 creature.
If that is true, then the statement "need not assign lethal damage" is quite ambiguous.
Before [i]Magic Origins[/i], C.R. 608.2b read, in relevant part:
"[I]f any of [a spell or ability's] targets are illegal, the part of the spell or ability's effect for which it is an illegal target can't ... make that target perform any actions."
Since [i]Magic Origins[/i], that part of the rule now reads, in relevant part:
"Illegal targets, if any, won't be affected by parts of a resolving spell's effect for which they're illegal."
(Note that it should probably say "resolving spell's or resolving ability's effect", but that's a separate issue.)
Responding only to that bold bit.
"Instant
Target Elf creature gets +1/+1 until end of turn.
Target green creature gains trample until end of turn."
This spell can target the same green elf twice. If that green elf becomes nongreen, it will still get +1/+1.
Before [i]Magic Origins[/i], C.R. 608.2b read, in relevant part:
"[I]f any of [a spell or ability's] targets are illegal, the part of the spell or ability's effect for which it is an illegal target can't ... make that target perform any actions."
Since [i]Magic Origins[/i], that part of the rule now reads, in relevant part:
"Illegal targets, if any, won't be affected by parts of a resolving spell's effect for which they're illegal."
(Note that it should probably say "resolving spell's or resolving ability's effect", but that's a separate issue.)
Responding only to that bold bit.
"Instant
Target Elf creature gets +1/+1 until end of turn.
Target green creature gains trample until end of turn."
This spell can target the same green elf twice. If that green elf becomes nongreen, it will still get +1/+1.
The issue you highlighted has to do, rather, with the fact that the sentence, as written, applies only to spells and not also to abilities.
When hideaway's triggered ability resolves, its controller looks at the top four cards of his or her library, then exiles one of them, then puts the rest on the bottom of that library in any order (C.R. 702.73a). However, the comprehensive rules (including the rules about the library [C.R. 401] and the rules about revealing or looking at cards [C.R. 701.14]) have no clear answer on whether it is known to other players which card among those four is exiled this way (even though the identity of that card is not revealed), namely, whether that card was the first, second, third, or fourth card from the top of the library. (Note that exiling one card, then putting the rest on the bottom, are two sequential actions.) See also this thread.
The issues "Grafdigger's Cage after the Amonkhet rules update", "Triggered abilities during the cleanup step", and "Illegal targets and their active involvement" are now moot, except for the issue that "Illegal targets, if any, won't be affected by parts of a resolving spell's effect for which they're illegal." should probably say "... resolving spell's or resolving ability's effect ...".
Rhetorical question: Which half of a fuse card has fuse? In which text box does the ability exist? If it's in both, does it have two instances? If it's in neither, why does CR 708.4c allow it to have fuse?
708.4c A split card has each card type specified on either of its halves and each ability in the text box of each half.
This subrule elaborates upon 708.4 "In every zone except the stack, the characteristics of a split card are those of its two halves combined. This is a change from previous rules."
The rules don't let us parse the card-anatomy of split cards with text box strangeness like a fuse card.
Fuse may be text outside of either text box. Cards may have such things.
702.94 Soulbond doesn't actually specify that a creature paired with another creature implies that the second creature is paired with the first, i.e. that paired is a commutative relation. Its definition only gives that the soulbond creature becomes paired with another when it enters, or perhaps another creature becomes paired with it when that one enters.
Private Mod Note
():
Rollback Post to RevisionRollBack
Epic banner by Erasmus of æтђєг.
Awesome avatar provided by Krashbot @ [Epic Graphics].
Have a third one. Are abilities truly independent of sources now?
702.1b An effect that grants an object a keyword ability may define a variable in that ability based on characteristics of that object or other information about the game state. For these abilities, the value of that variable is constantly reevaluated. Example: Volcano Hellion has the ability “Volcano Hellion has echo {X}, where X is your life total.” If your life total is 10 when Volcano Hellion’s echo ability triggers but 5 when it resolves, the echo cost to pay is {5}. Example: Fire//Ice is a split card whose halves have the associated mana costs {1}{R} and {1}{U}. Past in Flames reads “Each instant and sorcery card in your graveyard gains flashback until end of turn. The flashback cost is equal to its mana cost.” Fire//Ice has “Flashback {2}{U}{R}” while it is in your graveyard, but if you choose to cast Fire, the resulting spell has “Flashback {1}{R}.”
603.3. Once an ability has triggered, its controller puts it on the stack as an object that’s not a card the next time a player would receive priority. See rule 116, “Timing and Priority.” The ability becomes the topmost object on the stack. It has the text of the ability that created it, and no other characteristics. It remains on the stack until it’s countered, it resolves, a rule causes it to be removed from the stack, or an effect moves it elsewhere.
112.7a Once activated or triggered, an ability exists on the stack independently of its source. Destruction or removal of the source after that time won’t affect the ability. Note that some abilities cause a source to do something (for example, “Prodigal Pyromancer deals 1 damage to target creature or player”) rather than the ability doing anything directly. In these cases, any activated or triggered ability that references information about the source because the effect needs to be divided checks that information when the ability is put onto the stack. Otherwise, it will check that information when it resolves. In both instances, if the source is no longer in the zone it’s expected to be in at that time, its last known information is used. The source can still perform the action even though it no longer exists.
The identity of an ability -does- in same way depend on its textual instance at the source, as the echo cost of Volcano Hellion is re-evaluated at the time the trigger resolves. It does somehow detect a partial identity with a copy of itself or other instance of itself, through Rings of Brighthearth or through multiple activations of these Lorwyn Elementals.
If the cost referred to by an ability on the stack can change based on the application of a separate static ability of a certain permanent, in a continuous manner, then there is no independence. There is only a limited respect of persistence, as clarified precisely by the second sentence of CR112.7a. The first sentence is now spurious and misleading. As well, the structure defined between the other two rules and the behaviours of Rings of Brighthearth might reasonably be given more prominence and centrality. For instance, the language "It has the text of the ability that created it" might migrate into section 112. Abilities, and indicate there that this is a continuous redefinition (as in a .NET data binding).
The ability exists independently of its source, but is strongly linked to its textual counterpart.
112.7a Once activated or triggered, an ability exists on the stack independently of its source. Destruction or removal of the source after that time won’t affect the ability. Note that some abilities cause a source to do something (for example, “Prodigal Pyromancer deals 1 damage to target creature or player”) rather than the ability doing anything directly. In these cases, any activated or triggered ability that references information about the source because the effect needs to be divided checks that information when the ability is put onto the stack. Otherwise, it will check that information when it resolves. In both instances, if the source is no longer in the zone it’s expected to be in at that time, its last known information is used. The source can still perform the action even though it no longer exists.
If the cost referred to by an ability on the stack can change based on the application of a separate static ability of a certain permanent, in a continuous manner, then there is no independence.
The ability exists on the stack independent of it's source. That's the only independence implied or stated. Nothing you do to the source can remove the ability from the stack and vice versa. The rules never say that they're completely independent - in fact, such a claim is ludicrous because abilities are explicitly tied to their source for color and ownership/controller reasons.
702.94 Soulbond doesn't actually specify that a creature paired with another creature implies that the second creature is paired with the first, i.e. that paired is a commutative relation. Its definition only gives that the soulbond creature becomes paired with another when it enters, or perhaps another creature becomes paired with it when that one enters.
702.94b A creature becomes “paired” with another as the result of a soulbond ability. Abilities may refer to a paired creature, the creature another creature is paired with, or whether a creature is paired. An “unpaired” creature is one that is not paired.
I'm confused as to what you're referring to. The rule explicitly says that the two creatures are paired together.
Rhetorical question: Which half of a fuse card has fuse? In which text box does the ability exist? If it's in both, does it have two instances? If it's in neither, why does CR 708.4c allow it to have fuse?
708.4c A split card has each card type specified on either of its halves and each ability in the text box of each half.
This subrule elaborates upon 708.4 "In every zone except the stack, the characteristics of a split card are those of its two halves combined. This is a change from previous rules."
The rules don't let us parse the card-anatomy of split cards with text box strangeness like a fuse card.
702.101a Fuse is a static ability found on some split cards (see rule 708, “Split Cards”) that applies while the card with fuse is in a player’s hand. If a player casts a split card with fuse from his or her hand, the player may choose to cast both halves of that split card rather than choose one half. This choice is made before putting the split card with fuse onto the stack. The resulting spell is a fused split spell.
It's a static ability of the card, not in either text box.
Private Mod Note
():
Rollback Post to RevisionRollBack
To post a comment, please login or register a new account.
[i]Conditional actions in shortcut proposals[/i]: Jul. 13, 2018: Moot; see comment 29.
_Two-Headed Giant and certain can't-attack-a-player effects_: Now moot as of [i]Dominaria[/i].
_Grafdigger's Cage and the_ Amonkhet _rules update_: Moot because the rulings that led to this issue were removed.
_Triggered abilities during the cleanup step_: Not actually an issue, so stricken out.
_Illegal targets and their active involvement_: Moot because I couldn't find an actual Magic card that illustrated this issue.
_For each ... A unless B_ (e.g., Oath of Lim-Dûl, Read the Runes): To better define how effects of this form are processed, the rules manager should include in the comprehensive rules a rule that applies something similar to the APNAP order rules (C.R. 101.4) to cases where only one player has to make multiple choices at the same time, such as those called for by most cards of the form "For each ..., A unless B".
_C.R. 903.9a: The "appropriate zone"_: I suggest that the rules manager should replace "the appropriate zone" (a particularly poor word choice) in C.R. 903.9a with "the zone where it would otherwise go".
_C.R. 607.2m and name changes_: I suggest that the rules manager should replace "cards named [this object's name]" with, say, just "cards named [name]" (see also C.R. 202.4b), so that the rule unquestionably applies no matter what names the object in question has.
_C.R. 608.2b _: I suggest that the rules manager should replace "parts of a resolving spell's effect" with, say, "parts of a resolving spell's or resolving ability's effect".
"Ninjutsu and C.R. 702.48c": I suggest that the rules manager should edit the text of C.R. 702.48a and C.R. 702.48d to say, in part, "... tapped and attacking the same player or planeswalker as the creature returned this way". (Compare with Kaalia of the Vast or the myriad ability [C.R. 702.115].) This would make C.R. 702.48c a reminder rule, and bring ninjutsu in line with the practice found in other keyword abilities such as bestow and soulbond.
"Leovold's Operative": This merely seeks to clarify an existing card to bring it in line with rulings. According to the rules manager, "[c]ards that have during-draft effects, such as Leovold's Operative, don't follow most of the normal rules for priority, sequentiality, or anything else in Magic: the Gathering" (see also C.R. 905.2a). Therefore, in my opinion, the rules manager should give Leovold's Operative text like "... pass the next booster pack you would draft without drafting a card from it"; the current text is open to more than one interpretation, in my opinion.
"Costs to attack with insufficient mana": I suggest that the rules manager should delete "Once the [active] player has enough mana in his or her mana pool" in C.R. 508.1i (see also C.R. 601.2h, where such a clause is not present).
This is a record of questions yet to be answered by the rules manager:
At least nine cards include the formulation "You may cast/play ... for as long as it remains exiled", namely:
The intent of all these cards is to generate a continuous effect that lets a player cast or play a particular card from exile. But arguably, it may not work as expected under the current rules. Namely, since the card is moved out of exile to the stack as the first step of casting it (C.R. 601.2a), the duration "for as long as it remains exiled" ends, and therefore, so does the effect that allows the player to cast or play the card (C.R. 611.2a). Therefore, after the player finishes proposing the spell under C.R. 601.2a-d, the card won't be in exile anymore, but the stack, so that the player arguably isn't allowed to cast that card anymore (C.R. 601.2e). And C.R. 601.3 says, "If that player is no longer allowed to cast that spell after completing its proposal, the casting of the spell is illegal" (see also C.R. 601.2e), so the entire action is reversed to the point before the spell's casting was proposed (and so rendering the cast of the spell not possible) (C.R. 601.2, 720.1). As a consequence of this, the spell never counts as being on the stack if the cast was reversed this way, so the card "remains exiled".
This leads to the absurd result that players can't cast that card at all, which is obviously not what is intended. And there appears to be no basis in the comprehensive rules that allows these effects to work as intended.
Note that none of this applies to playing a land, which works as intended because playing a land is usually a single step: the player simply "puts that land onto the battlefield from the zone it was in", here, the exile zone (C.R. 115.2a).
Does the ante zone always exist even if no card in the format in question can put cards there? That question has no clear answer under the comprehensive rules, in particular C.R. 400.1.
More generally, does a zone exist even if it contains no objects? Even that question is not explicitly answered by the comprehensive rules, even though it's ruled that it does exist in that case.
Under C.R. 704.3, if any state-based actions apply, they are "perform[ed] simultaneously as a single event". One case where the interpretation of this rule matters is if a player would lose the game for having 0 life while controlling and owning Lich's Mirror and owning a Grasp of Fate that exiled an Erebos, God of the Dead owned by an opponent. In this case, Lich's Mirror and Grasp of Fate would be shuffled into their owner's library, then Erebos would return to the battlefield "immediately after" that event and before seven cards are drawn (C.R. 610.3), so that Erebos would be on the battlefield in time to keep Lich's Mirror's controller from gaining life (which normally happens when a player goes from 0 life to 20 [C.R. 118.5]) (C.R. 611.3b, 611.3, 112.6, 101.3).
However, as I originally interpreted the text "simultaneously as a single event", the entire state-based action, including all the actions Lich's Mirror's replacement effect entails, would be condensed into one moment in time -- such that there would be no "time" (as far as the game is concerned) between the start of the state-based actions and their end -- so that Erebos wouldn't return to the battlefield until "immediately after" the entire state-based action is processed (that is, "immediately after" Lich's Mirror's controller goes to 20 life in this case). However, as I confirmed with "nastaboi" on the Ask a Magic Judge chat (#mtgrules), I was mistaken.
This suggests that the wording "... then performs all applicable state-based actions simultaneously as a single event" in C.R. 704.3 can be very counterintuitive, because as it turns out, it's there merely to indicate that state-based actions [i]begin[/i] simultaneously, even if some of those actions are replaced with multi-step events, as is the case with Lich's Mirror's replacement effect. If that is the intent, then such a possible rewording to C.R. 704.3 as "... then all applicable state-based actions begin simultaneously" would resolve my concerns, for it would leave no question that state-based actions [i]begin[/i] at the same time, but the actions of each one still occur sequentially.
[s]_Delayed triggered abilities: One token turns into several_: Whether, for the two tokens created due to Tatsumasa, the Dragon Fang's last ability and Doubling Season, one delayed triggered ability is created for each, or one for both, is not unambiguously answered by the comprehensive rules. This issue can be resolved, among other possibilities, by adding to the comprehensive rules a general rule that if a delayed triggered ability refers to one or more tokens and more tokens than usual are created, then a separate delayed triggered ability is created for each one.[/s]
[i]Conditional actions in shortcut proposals[/i]
C.R. 720.2a contains the following sentence dealing with legal shortcut proposals:
"It can't include conditional actions, where the outcome of a game event determines the next action a player takes."
It seems to me that the intent of this sentence is to disallow "four horsemen" or "Emrakul" shortcuts that involve a player shuffling the library and then involve something that cares about the order of cards in that library. In my opinion, though, these shortcuts are already prohibited by the requirement, in C.R. 720.2a, that shortcuts have "predictable results". Such purported shortcuts as those that--
Moreover, that sentence appears to rule out such shortcuts as the one sanctioned by the annotated Magic Tournament Rules 2.4 ( http://blogs.magicjudges.org/rules/mtr4-2/ );
"A player may interrupt a tournament shortcut by explaining how he or she is deviating from it or at which point in the middle he or she wishes to take an action. A player may interrupt his or her own shortcut in this manner."
"[[ For example, Player A controls Basalt Monolith and Mesmeric Orb and says, 'I'd like to tap and untap this Monolith 30 times or until I reveal a land.' This suggests a shortcut and a stopping point for interrupting the shortcut.]]"
Because the outcome of a game event (here, putting a land card into the graveyard or some other card) determines the next action a player takes (namely, untapping Basalt Monolith), this shortcut arguably contains a conditional action and is arguably disallowed under the current version of C.R. 720.2a. Unfortunately, it isn't allowed under C.R. 720.2b either, since under that rule, only another player is allowed to "shorten" a shortcut, not the player who proposed the shortcut. (Though the example mentions "interrupting the shortcut", I believe this shortcut would be allowable in the absence of the sentence at issue here.)
Both these cases indicate that the sentence in question ought to be eliminated from C.R. 720.2a, in my opinion -- on the one hand, because it's redundant, and on the other hand, because it appears to disallow too much (and thus removing that sentence would bring C.R. 720.2a more in line with the Magic Tournament Rules).
(I am aware that the Magic Tournament Rules override the comprehensive rules in a sanctioned tournament [M.T.R., "Introduction"], but I feel this change is appropriate not just for sanctioned games, but in more casual games as well -- as is the case, by the way and in my opinion, for comprehensive rules changes that allow players to reveal cards from their hand at any time, that list certain tournament shortcuts, and the like.)
A related question: Is there a example of a shortcut disallowed by the sentence "It can't contain conditional actions...", but which doesn't involve randomness and is not within the scope of the Magic Tournament Rules excerpt I cited?
[i]Two-Headed Giant and certain can't-attack-a-player effects[/i]
EDIT (Apr. 29, 2018): Now moot as of [i]Dominaria[/i].
C.R. 810.7a says that in Two-Headed Giant, "Each team's creatures attack the other team as a group", so they don't attack individual players.
C.R. 810.7c, another Two-Headed Giant rule, says, in relevant part, "If an effect of an object controlled by a defending player prohibits a creature from attacking him or her, that creature can't attack the defending team."
C.R. 810.7c works quite well for nearly all cards that include effects in which creatures "can't attack" particular players, since in these cases, the effect comes from a static ability of a permanent and
are phrased "can't attack you" (since they apply only to the permanent's controller [C.R. 109.5] and end when the permanent leaves the battlefield [C.R. 112.6, 611.3b]). However, I see a few exceptions:
Chronomantic Escape and Illusionist's Gambit generate the effect from a resolving spell.
Orzhov Advokist and Web of Inertia generate the effect from a resolving ability.
Arboria creates an effect in which creatures can't attack a player that's not necessarily Arboria's controller.
Because of the way this part of C.R. 810.7c is worded, it becomes unclear whether creatures can attack a defending player's team even if--
Because of C.R. 810.7a, it would be incorrect to clarify the cases mentioned above by saying that the creature can attack the other player on the defending team, since under C.R. 810.7a, creatures attack only teams, not individual players. And it would be similarly incorrect to say the creature can nevertheless attack the defending team, since the defending player in question is part of that team and it would likely come in conflict with the general rule that effects that provide that something "can't" be done prevail (C.R. 101.2).
This shows that this part of C.R. 810.7c is in conflict with C.R. 810.7a and doesn't cover all the cases it's intended to cover.
[i]Grafdigger's Cage after the Amonkhet rules update[/i]
EDIT (Oct. 15): Now moot, since the rulings in question were removed from Gatherer.
[s]
The rulings for Grafdigger's Cage, newly added in [i]Amonkhet[/i], don't appear to correspond with that card's Oracle text, which still says, in part: "Creature [i]cards[/i] can't enter the battlefield from graveyards or libraries" (which would mean that the ability checks the characteristics of the card in the graveyard or library, regardless of what it would be on the battlefield [C.R. 109.2]). However, the rulings (which appear to have been added in response to newly added C.R. 614.16d) suggest that the text was changed to "[i]Creatures[/i] can't enter the battlefield" (which unfortunately isn't the case; compare with Worms of the Earth, whose second ability, "Lands can't enter the battlefield", is governed by new C.R. 614.16d). Moreover, if the rulings stay, that would now unfortunately make Grafdigger's Cage's first ability quite unintuitive, and could even make it ambiguous, especially since "creature card" can now (under a given interpretation of new C.R. 108.2) mean either a creature card in the graveyard or library, on the one hand, or a creature card on the battlefield, on the other (see also C.R. 109.2a, 110.1).[/s]
[i]Illegal targets and their active involvement[/i]
[EDIT (Aug. 21): Soul's Fire turned out to be a poor example; edited it with a hypothetical card similar to Soul's Fire. (I have edited my answer to the thread that led to this issue with a more solid rule justification.)]
EDIT (Oct. 15): I will declare this moot for now since I can't find an actual Magic card where this matters.
[s]
Before [i]Magic Origins[/i], C.R. 608.2b read, in relevant part:
"[I]f any of [a spell or ability's] targets are illegal, the part of the spell or ability's effect for which it is an illegal target can't ... make that target perform any actions."
Since [i]Magic Origins[/i], that part of the rule now reads, in relevant part:
"Illegal targets, if any, won't be affected by parts of a resolving spell's effect for which they're illegal."
(Note that it should probably say "resolving spell's or resolving ability's effect", but that's a separate issue.)
Arguably, this appears to suggest that illegal targets are only kept from being [i]passively[/i] affected by the parts of the effect in question, and that such targets can still be [i]actively[/i] involved in that effect. An example will illustrate this.
Suppose a player casts a hypothetical spell similar to Soul's Fire (but reads instead "Target creature on the battlefield deals 2 damage to target creature or player") targeting a creature and a player, such that the creature will deal damage to the player. Then, before this hypothetical spell resolves, the creature gains shroud. Then, when the spell resolves, the creature will be an illegal target (C.R. 608.2b, 702.18a). In this circumstance, it's a stretch to imagine that being "unaffected" by the damage effect means that the creature won't deal damage, especially since dealing damage is [i]active[/i], whereas receiving damage is [i]passive[/i]. The rules before [i]Magic Origins[/i] made it clear that this hypothetical spell wouldn't make the illegal target deal any damage; that isn't clear anymore in the rules since [i]Magic Origins[/i]. (Contrast this with the case where the player gains shroud rather than the creature. In both versions of the rules, it would be clear that damage wouldn't be dealt to the player -- here, the illegal target, the player, is not "affected" by the hypothetical spell.)
[i]Triggered abilities during the cleanup step[/i]
EDIT (Oct. 15): Now moot as of [i]Ixalan[/i], if not before.
[s]In [i]Kaladesh[/i], the rules were changed to streamline how triggered abilities are handled. But under the current wording of C.R. 514.3a, triggered abilities "waiting to be put on the stack" would be put there below abilities that trigger during the cleanup step's state-based actions (which would go on the stack as the active player gets priority [C.R. 116.5]), a situation that the rule update seeks to avoid. Moreover, the current wording of C.R. 514.3a, as I see it, is defective in that it only says "those triggered abilities are put on the stack", that is, those triggered abilities that "are waiting to be put onto the stack" at the beginning of the process, and not, say, "triggered abilities that are waiting to be put onto the stack are put there", which also covers those abilities that triggered as a result of the state-based actions mentioned in that rule.[/s]
[i]Who decides for a player who has left the game? (RESOLVED)[/i]
Resolved with [i]Archenemy: Nicol Bolas[/i].
[s]As this thread shows, the comprehensive rules don't regulate which player makes a decision if it's not an object, but a game rule, that leads the active player to make a choice, but the active player isn't in the game anymore.[/s]
The rule in question doesn't check about legality based on where the card is, because at that point the card is on the stack and you can't cast cards from the stack. If it does check location, it must be checking based on if the spell with those characteristics were in the the zone it was being cast from, which in this case would be a legal spell.
I don't even understand the objection here. Indeed, you don't pay the costs if you can't. How is this unintended behavior?
This is a fair point. They may have forgotten to update the Oracle wording.
Clearly we need to switch to an object-oriented rules system.
It's worth noting that spells or abilities are objects, and the effects created when they resolve should be covered by 870.1c. As for the controller, LKI should be able to see who controlled the spell or ability at the time the effect was created. Arboria does seem to throw a wrench in this, though.
What if there are multiple state-based actions that have multi-step events? I'll bet someone here can come up with a scenario for that one.
Your argument is spurious. Being forced to deal damage to something is clearly being affected. There's no active/passive distinction. (But you're right about it should say spell or ability.)
I cannot imagine a scenario in which the answer to these questions matters.
Yeah, that's actually pretty ambiguous. I think the idea they're going for is that it depends on the outcome of a game event not described by the deterministic loop. For example, Emrakul's reshuffle is not a game action described by the loop, but the identity of the card milled is.
Technically correct and should probably be fixed, but wow is that pedantic. (Yes, yes, I know that when it comes to rules pedantry is important.)
Strictly speaking there's no contradiction. 702.48c restricts the choices you're allowed to make under 508.4. However, I guess that brings up the question of whether 702 keyword rules are supposed to have all of the actual rules content in 702.XXa and the rest of 702.XXx is just clarifications, or if the other 702.XXx are also meaningful.
This all seems correct. "would" is the better phrasing.
To be fair, Grafdigger's Cage's Oracle text was updated in Amonkhet, but only in its second ability.
Consider a scenario where, in addition to the objects mentioned in the Lich's Mirror scenario, a creature under a regeneration effect has been dealt lethal combat damage and, at the same time, the player controlling Lich's Mirror has been dealt enough combat damage to bring his or her life total to 0 or less. Like in Lich's Mirror, regeneration is a multi-step replacement effect: removing damage from the creature, tapping it, and removing that creature from combat (C.R. 701.13a-b).
If anything, a rule that clarifies that each zone always exists would clarify (more strongly than the rules already do), for example, that there is no time when players "have no hand", or that players don't "start a new stack".
In my opinion, C.R. 702.48c should at least be a reminder rule to C.R. 702.48a, which should be changed as I proposed. Notable examples of this policy include the bestow and soulbond keywords (C.R. 702.102, 702.94a).
Regeneration doesn't really feel multi-step because none of those are contingent on each other. I'm thinking something more like an Angel of Sanctions also dies to lethal damage as part of the event and it has a Humility under it. Assuming Erebos' controller has enough devotion for it to be a creature, we end up with a sort of race condition on which combination of permanents is actually on the battlefield when the mirror's replacement effect tries to make its controller gain 20 life.
Also someone should submit this one to http://magicjudge.tumblr.com. I'm sure she'd love it.
What happens if a creature with undying and a +1/+1 counter on it gets a lethal amount of -1/-1 counters put on it? Does the creature die with or without a +1/+1 counter?
My CubeCobra (draft 20 card packs, 2 packs.)
430, Peasant, Very Unpowered
Why you should take your hybrids out of your gold section
Manamath Article
You should ask such questions in the rules forum. This thread was created in the general forum specifically to discuss issues with the CR, not to answer rules questions.
Anyway, dies triggers look at the game state right before the creature died in order to determine if they trigger. And right before the creature died (so right before the check of state based actions) it had a +1/+1 counter, so undying doesn't trigger.
Former Rules Advisor
"Everything's better with pirates." - Lodge
(The Gamers: Dorkness Rising)
"Any sufficiently analyzed magic is indistinguishable from science."
(Girl Genius - Fairy Tale Theater Break - Cinderella, end of volume 8)
Delayed triggered abilities: One token turns into several -- see comment 1.
A clarification to C.R. 107.1b is in order
In C.R. 107.1b, I suggest that the rules manager should change "a calculation that would determine the result of an effect" to "a calculation that would determine the meaning of an effect" to help make the intended effect of the change to C.R. 107.1b in Hour of Devastation (to keep the X in such effects as "gets +X/+X ... where X is ..." from being negative) more consistent with the three examples given in that rule.
Paliano Vanguard, Noble Banneret, Smuggler Captain, and Aether Searcher include instructions to note information during a draft. However, C.R. 905.2b, which governs such noted information, doesn't appear to apply to any of those cards, since the information isn't noted as Paliano Vanguard itself, etc., is drafted.
C.R. 607.2m and name changes -- see comment 1.
The last ability of Chandra, Pyromaster says, in relevant part, "You may cast the copies without paying their mana costs". A ruling on this card says, in relevant part: "While resolving the third ability, ... [y]ou can cast zero, one, two, or all three copies", and this appears to allow the player to cast at least one copy this way even if an Extirpate card (with split second) is copied this way (because only one copy can be cast which renders casting further copies impossible while that copy is on the stack [C.R. 702.60a; see also this thread]), in which case there appears to be no basis for this ruling in the comprehensive rules (according to C.R. 608.2d, the option to "cast the copies", arguably, can't be chosen if not all the copies can be cast; compare Chandra, Pyromaster with Chandra Ablaze, which says "[c]ast any number" of cards meeting certain criteria).
I do think that the language for casting legality is sort of clumsy. It is clear that it is trying to accommodate all the fancy ways that permissions are granted to spells - with keywords, with "as-though"s - but it has lost uniformity. Improvement is imaginable.
Awesome avatar provided by Krashbot @ [Epic Graphics].
In general, an attacking creature "can't assign combat damage to a creature that's blocking it unless, when combat damage assignments are complete, each creature that precedes that blocking creature in its order is assigned lethal damage" (C.R. 510.1c). However, under C.R. 702.19b, if the attacking creature has trample, its controller "need not assign lethal damage" to those blockers (but "in that case[,] ... can't assign any damage" to defending player or planeswalker) (C.R. 702.19b), so that that player can arguably assign less than lethal combat damage to the blockers (compare with C.R. 510.1c, the general rule). For example, under that interpretation, the following combat damage assignments are possible for a 6/6 attacking creature with trample (assuming the damage assignment order is 4/4 creature then 2/3 creature, and neither creature has damage marked on it):
- 0 damage to the 4/4 creature and 6 damage to the 2/3 creature.
- 0 damage to the 4/4 creature and 3 damage to the 2/3 creature.
- 3 damage to the 4/4 creature and 3 damage to the 2/3 creature.
- 3 damage to the 4/4 creature and 3 damage to the 2/3 creature.
- 4 damage to the 4/4 creature and 1 damage to the 2/3 creature.
- 5 damage to the 4/4 creature.
However, the sentence where the cited part of C.R. 702.19b appears is arguably not meant to imply that the damage assignment order can be ignored, but rather, as I was told, to "clarif[y] that [the attacking player] do[es]n't have to assign lethal damage to each blocker even if [he or she] could". In that case, only the following damage assignments would be possible in the same scenario given earlier:
- 5 damage to the 4/4 creature and 1 damage to the 2/3 creature.
- 6 damage to the 4/4 creature.
- 4 damage to the 4/4 creature and 2 damage to the 2/3 creature.
If that is true, then the statement "need not assign lethal damage" is quite ambiguous.
Responding only to that bold bit.
"Instant
Target Elf creature gets +1/+1 until end of turn.
Target green creature gains trample until end of turn."
This spell can target the same green elf twice. If that green elf becomes nongreen, it will still get +1/+1.
Older Magic as a Board Game: Panglacial Wurm , Mill
When hideaway's triggered ability resolves, its controller looks at the top four cards of his or her library, then exiles one of them, then puts the rest on the bottom of that library in any order (C.R. 702.73a). However, the comprehensive rules (including the rules about the library [C.R. 401] and the rules about revealing or looking at cards [C.R. 701.14]) have no clear answer on whether it is known to other players which card among those four is exiled this way (even though the identity of that card is not revealed), namely, whether that card was the first, second, third, or fourth card from the top of the library. (Note that exiling one card, then putting the rest on the bottom, are two sequential actions.) See also this thread.
This subrule elaborates upon 708.4 "In every zone except the stack, the characteristics of a split card are those of its two halves combined. This is a change from previous rules."
The rules don't let us parse the card-anatomy of split cards with text box strangeness like a fuse card.Fuse may be text outside of either text box. Cards may have such things.
Awesome avatar provided by Krashbot @ [Epic Graphics].
Awesome avatar provided by Krashbot @ [Epic Graphics].
Along with the set of cards in this Gatherer query, the first part of CR112.7a has been seriously undermined by new innovations in the rules:http://gatherer.wizards.com/Pages/Search/Default.aspx?action=advanced&format=+[%22Lorwyn-Shadowmoor%20Block%22]&text=+[resolved]+[this]
The identity of an ability -does- in same way depend on its textual instance at the source, as the echo cost of Volcano Hellion is re-evaluated at the time the trigger resolves. It does somehow detect a partial identity with a copy of itself or other instance of itself, through Rings of Brighthearth or through multiple activations of these Lorwyn Elementals.
If the cost referred to by an ability on the stack can change based on the application of a separate static ability of a certain permanent, in a continuous manner, then there is no independence. There is only a limited respect of persistence, as clarified precisely by the second sentence of CR112.7a. The first sentence is now spurious and misleading. As well, the structure defined between the other two rules and the behaviours of Rings of Brighthearth might reasonably be given more prominence and centrality. For instance, the language "It has the text of the ability that created it" might migrate into section 112. Abilities, and indicate there that this is a continuous redefinition (as in a .NET data binding).The ability exists independently of its source, but is strongly linked to its textual counterpart.
Awesome avatar provided by Krashbot @ [Epic Graphics].
The ability exists on the stack independent of it's source. That's the only independence implied or stated. Nothing you do to the source can remove the ability from the stack and vice versa. The rules never say that they're completely independent - in fact, such a claim is ludicrous because abilities are explicitly tied to their source for color and ownership/controller reasons.
I'm confused as to what you're referring to. The rule explicitly says that the two creatures are paired together.
It's a static ability of the card, not in either text box.