What conditions and effects does the spell freedom of movement [abjur] (PH 233) overcome?
I want a fair, reasonable, and consistent method to determine what spells and conditions are overcome by the spell freedom of movement. I'm not looking for a comprehensive list of spells and conditions that the spell freedom of movement overcomes, although examples are okay. I am unconcerned with freedom of movement's effects on grapples and underwater combat.
Results of My Research
This runs way long. You've been warned.
The FAQ
I read the June 30, 2008, 3.5 Main FAQ that contains the question, "Does the freedom of movement spell protect a character from being stunned? The argument is that 'stun' is a condition that hinders movement" (82), but the answer seems unaware that the spell hold person [ench] (PH 241) causes paralysis, which freedom of movement specifically negates. Further, the FAQ urges the DM to house rule effects as "mental impediments" and exclude those from freedom of movement's effects. Call me lazy if you like, but I'm unwilling to classify arbitrarily every effect in the game either a mental impediment or not a mental impediment because the FAQ suggests a house rule.
Message Boards
Then I read a dozen or so online discussions. To understand those results, here's the important part of the spell freedom of movement:
This spell enables you or a creature you touch to move and attack normally for the duration of the spell, even under the influence of magic that usually impedes movement, such as paralysis, solid fog, slow, and web.
I emphasize that word even because, during my research, discussions sometimes accidentally conflated the less powerful and limited travel Domain granted power--which negates only magical effects--with the more powerful and versatile freedom of movement spell--which negates both magical and mundane effects. Further, sometimes discussions touted that list as inclusive, which it isn't.
The online discussions--including Pathfinder ones as the D&D 3.5 version is largely unchanged in the Pathfinder version--usually reach a consensus (or, at least, the folks who type the most repeat loudest) that freedom of movement only cares about the affected creature's ability (or inability) to actually move (often ignoring the and attack phrase in the spell's description), with freedom of movement not caring that the affected creature can or can't take actions.
To summarize that argument, because paralysis says nothing about losing one's actions--a paralyzed creature can take all his actions... the paralyzed creature just can't move when he does--freedom of movement therefore can't negate effects that reference actions (e.g. dazed, stunned). A creature's ability to move actually isn't impeded by being unable to take actions; instead, because of his inability to take actions, the creature can't move. That the creature can't move because the creature can't take actions isn't the condition's fault but the creature's fault for not having some way to move that doesn't involve actions. In other words, something that limits the ability to move is different from something that limits the ability to take actions.
I can almost accept this argument, although it seems forced. I mean, darn few ways exist to "move and attack normally" when a creature "can't take actions," even though it's an interesting limit on the spell's scope. However, even accepting that argument doesn't address conditions that impose penalties; penalties, after all, do seem to prevent the creature from being able to "attack normally" without affecting the creature's ability to take actions. (It's not normal for a creature to suffer a -2 penalty on its attack roll unless the creature's always sickened or shaken, for example.) Furthermore, accepting that argument doesn't address the freedom of movement spell's ability to negate the spell slow [trans] (PH 280), which, in addition to not letting a creature move normally by reducing the creature's speed, specifically limits the affected creature to "only a single move action or standard action each turn." Any conclusion that says a creature affected by both the spells slow and freedom of movement should take less than his full allotment of actions yet suffer no penalties is, bluntly, just weird.
A Similar Spell
I can't help but look at the spell freedom [abjur] (PH 233); you know, right above freedom of movement. It reads
The subject is freed from spells and effects that restrict its movement, including binding, entangle, grappling, imprisonment, maze, paralysis, petrification, pinning, sleep, slow, stunning, temporal stasis, and web. To free a creature from imprisonment or maze, you must know its name and background, and you must cast this spell at the spot where it was entombed or banished into the maze.
The spell freedom is a 9th-level spell compared to the 4th-level freedom of movement, but freedom has a longer list of effects that it overcomes, and those effects are specifically those that "restrict [a creature's] movement." Using that much more expansive list (and noting that it, too, isn't inclusive) in addition to freedom of movement's list makes freedom of movement a very powerful effect. However, many high-level effects freedom of movement could suppress after the fact can't actually be affected by freedom of movement because the caster can't touch the target. Further, if the duration of the effect exceeds the duration of the freedom of movement, the effect resumes when freedom of movement's duration expires (e.g. a creature under the effects of the spell freedom of movement who fails a saving throw versus a medusa's gaze is petrified when the freedom of movement spell's duration expires). The spell freedom of movement, then, is a temporary barrier against such effects rather than a fix for such effects, and barriers are often lower-level than fixes.
Maybe if the game had emphasized death spells, magical death effects, energy draining, and negative energy effects like it emphasized conditions that prevent creatures from being able to "move and attack normally," this question would instead be about death ward [necro] (PH 217). I don't know.
Into the Past
I went to the 3.0 Player's Handbook to see if the spell freedom of movement had changed from 3.0 to 3.5, and it did--a little. The 3.0 spell freedom of movement [abjur] (PH 207) reads, in part,
This spell enables you or a creature you touch to move and attack normally for the duration of the spell, even under the influence of magic that usually impedes movement, such as hold person, paralysis, solid fog, slow, and web.
That italicized paralysis is in the original for no reason. It's clear that during the transition between 3.0 and 3.5 somebody took a look at freedom of movement, read the spell hold person [ench] (PH 214), and decided to consolidate paralysis and hold person into paralysis. Whoever did that probably also rewrote 3.0's hold person (which made no mention of paralysis) to 3.5's hold person [ench] (PH 241). So, yeah, that happened.
I also looked at my 1st edition Dungeon Master's Guide at the ring of free action (which--shock!--references hold, slow, and web spells and underwater movement--grappling wasn't really a thing) and the virtually identical item from volume 3 of the Encyclopedia Magica. Those were little help.
Conclusion
I'm going to speculate a bit about authorial intent here, so hang on: contemporary D&D 3.5 players understand the action economy and the importance of battlefield control, but the design team 14 years ago when freedom of movement was published didn't. Or, if they did, it was an example of Ivory Tower Game Design, a sort of backhanded congratulations for system mastery. Either way, immunity to all those effects just didn't seem that big of a deal then as it is now.
Possibly the spell's supposed to be a DM's call, a legacy of the earlier game wherein the DM's house rules developed as the game progressed, and the players accepted them, and if the fighter with his ring of free action showed up at someone else's table the ring might work completely differently. That was totally okay for the majority of RPG history and even when D&D 3.0 was released, but video games and the Internet have made it so folks want these effects nailed down. That's okay, too. Seriously, I prefer a consistent gaming environment over a transitory one. I'm not sure this effect even can be nailed down given that it's just 4 words ("move and attack normally") out of millions, but I'd like to hear fair ways it has been nailed down.