If Ghostwalk (Monte Cook and Sean K Reynolds) had been released by TSR in 1995, it would probably be considered one of the classic rpg settings. It's got that exact pitch of mid-90s weirdness - a bold high concept that nonetheless remains wedded to vanilla D&D orthodoxy, even when doing so is dull or counterproductive (the last 50 pages of this 220 page book are a series of site-based adventures which only intermittently take advantage the unique aspects of the setting and they probably took me twice as long to read as the rest of the book combined).
Unfortunately, Ghostwalk was released in 2003, which means it never really got a chance to benefit from a dozen different authors churning out quick, disposable supplements that gradually distilled its best elements into an immortal meme setting that never quite existed on the page (I'm looking at you, Planescape). This book needed to hit the ground running, and it couldn't quite pull it off.
The setting's basic pitch is great - in this world, the land of the dead is a physical location that exists underground. Whenever someone dies, they become a ghost, and in order to arrive at the True Afterlife, they must venture to the Ghostwalk, the holy pilgrimage where their spirit descends into the depths of the earth through a series of sacred caves. And the land closest to the Ghostwalk has the miraculous property of giving ghosts a solid physical form allowing them to interact normally with the living. So the area around the entrance to the underworld is a thriving city called Manifest, inhabited by both the living and the dead.
But, all of that was on the first page of the book. The subsequent 219 pages are a study in how to take a great idea and gradually wear it down with persistent D&D-isms. With the brief exception of the monster chapter, this book gets less and less itself as time goes on. The first page of the first full chapter talks about how every damned creature type (literally, listed alphabetically) except humanoids can't actually become a ghost, and is exempt from the most distinctive aspect of the setting metaphysics. The description of the city of Manifest is largely done through a keyed map, where half the locations don't have anything at all to do with ghosts and could easily fit into any fantasy setting. Then, you get to the DM's chapter and it reveals the true nature of the afterlife (with a keyed map!) and it's so aggressively dull that it will need its own paragraph. And you wind up with an adventure chapter that feels like it was written by someone who has not had a chance to read the completed book, with example after example of map-based storytelling that somehow manages to tell me nothing about the world of Ghostwalk (seriously, one of the adventures is just fighting a random manticore in his lair and I guess it's saying that manticores are . . . bad?).
Don't let my griping fool you, though. I'm only disappointed because I was enchanted by this book's potential. There's so much it could have done, if it was willing to be just a bit funnier or spookier or more philosophical.
Resurrection magic is easier and less traumatizing in the city of Manifest, so adventurers go there to bring back their comrades. And many of the city's ghostly residents work for the living in hopes of saving up enough to raise themselves from the dead. Which makes me think that Ghostwalk could have been the D&D setting that really went all out with the game's stereotypical revolving door afterlife. People could be completely casual about dying and jaded to the miracle of being brought back. You could take a cue from the temple that's trying to raise money for charity by selling its resurrection services and put the power of life and death into the hands of otherwise ordinary characters, who use it for distressingly mundane purposes. But that's essentially a comedy premise and Ghostwalk takes itself too seriously for that.
Alternately, Ghostwalk could be a go-to setting for D&D Halloween one-shots. You actually get rules for playing as a ghost, including powers like telekinesis, possession, and the manipulation of ectoplasm! Super neat . . . except that they decided to accomplish this by relying on multiclassing as the system's load-bearing mechanic. When you advance in level, you can take the eidolon or eidoloncer class in lieu of a regular character class . . . if you're dead . . . and if you aren't exceeding your living glass levels (lest you be drawn to the True Afterlife by the Calling). Mechanically, this doesn't work at all, because levels are scarce. You have to decide whether you want to sacrifice skill at your day job and if you don't, you're going to be a crummy ghost (this was late enough in 3rd edition's life cycle that they knew this about spellcasters, which is why the eidoloncer gives you 20 levels of spellcasting, but apparently still too early for them to realize that the only reason other classes were less affected is because they had less cool stuff to lose). And if you do decide to invest in being a ghost, you're not going to be able to use any of the powers you paid for if someone brings you back to life (a thing most ghosts want to happen).
The solution the book comes up with is to allow ghost PCs to trade in their ghost levels for regular class levels whenever they're brought back, but doing so involves going through the level advancement math in reverse and leaves you less cool on those occasions when you do have an opportunity to use your ghostly abilities. It's a dilemma that runs straight into the biggest mathematical weakness of 3rd edition as a whole, and it's one you're going to have to navigate if you want to engage with the book's biggest selling point.
But shaky mechanics are only half of it. Ghostwalk sends very mixed messages about whether or not it understands its own inherent goth appeal. The first chapter talks about how ghost characters can have grisly appearances based on the circumstances of their death, but the city of Manifest doesn't have a single gloomy underground bar with a funereal atmosphere where the patrons take ghastly delight in freaking out the living by comparing their various wounds. There's no romance of the shadows. No tenuous line between ghosts that are basically just dead people and ghosts who have embraced the power of death (even though the setting makes a distinction between regular ghosts and "undead spirits.") It just continues to be D&D, but doesn't try to get any mileage out of that either.
Then again, you could say that not capitalizing on good ideas is kind of Ghostwalk's whole deal. Nowhere is that more evident than with the True Afterlife. I can't disagree with the DM chapter - players are going to want to go there. It's a place that's relatively easy to find (if you discount the subterranean perils of the ghostwalk itself) and it's a natural culmination to the campaign's exploration of the line between life and death. After spending time a ghost and then returning to life, possibly several times, you're going to want to know what the big deal is about the afterlife itself.
And when you get there, you're going to discover that it was all hype. There is no big deal, nothing at stake, and practically nothing to learn. The True Afterlife is just a series of islands where a bunch of humanoid ghosts hang out. They only have bodies if someone brings their remains down through the ghostwalk, but even when reunited with their bodies, they all just sort of find a place to be and then do the same sort of stuff they did when they were alive. And if you think that I'm being frustratingly vague about all this, then we're on exactly the same page. You get it.
It would be bad enough if the True Afterlife were just a regular campaign world, sans growing old, having to eat and drink, and without character progression. You get isekaied to the exact same situation you just left, and maybe that's a bit ridiculous, but it could potentially be something. But the True Afterlife is actually a much duller and less diverse world than the world of the living. Remember all those "Aberrations, constructs, dragons, elementals, fey, giants, magical beasts, monstrous humanoids, oozes, outsiders, plants, shapechangers, undead, and vermin" who "simply pass on to the True Afterlife when they die?" Well, it turns out, when they get there, they "[arrive] in the True Afterlife as little more than an intangible shade (even within the Manifest Ward), incapable of affecting the world around [them]. These poor creatures are called the bodiless dead [and] can make no attacks, can cast no spells or use any extraordinary, spell-like, or supernatural abilities . . . cannot speak above a whisper. . . and cannot influence or manipulate solid, incorporeal, or ghost touch objects . . . The bodiless dead normally flit about for a few days before abandoning consciousness, which in effect is a permanent form of self-dispersal. Some of them join the ranks of servants of the other dead . . .'
And so on. You may recognize this as being total bullshit.
Which isn't necessarily an automatic strike against a setting, but it is a strike against a book when it fails to realize that a bullshit situation is one that inevitably invites a reckoning (i.e. an exciting story for a D&D campaign). The bodiless dead can get bodies in the afterlife, if their remains go through the ghostwalk and arrive in the True Afterlife, but that doesn't often happen because humanoids (and humanoid ghosts) built a fucking city in the way.
And that's a campaign. The whole book uses Yuan-ti as antagonists because, as monstrous humanoids, they can't become ghosts, and so they take out their envy on humanoid civilization, but it never realizes that the Yuan-ti kind of have a point. Not that this injustice necessarily excuses their violence and cruelty, but I can see where they're coming from. There's this place that should belong to the entire world, but some people built a city on top of it and they use that to act as gatekeepers to the afterlife, and the whole thing is just the biggest, most monumental political conflict it's possible to imagine.
So, maybe you could tinker with the metaphysics a bit to better make it work. Maybe it's the case that only humanoids become ghosts because they control Manifest? Maybe the shipment of bodies through the ghostwalk is a big fucking deal, governed by treaties, orchestrated by priests and arcanists who explicitly seek to control the afterlife. Maybe we should be left asking if Manifest even deserves to exist?
To sum up my feelings - this is a book that feels like it could be torn apart to serve as the basis of a much better book. But also, maybe the time to do that was late 2002, before the final draft was sent to the printer's. Like, maybe the math issues can be forgiven, with the relative lack of d20 sophistication that existed back then, but the thematic, structural, and aesthetic shortcomings should have been obvious, especially since I know both of these authors are capable of better work. I still really like the game's elevator pitch, but I come away from Ghostwalk feeling like it exists purely because of the elevator pitch, and that during the writing and development, the pitch was not allowed to evolve. Overall, I'd say this book has 20 or 30 great pages that I'd love to see get turned into a campaign setting one of these days.
Ukss Contribution: My favorite part of the book was the microfiction at the very beginning, about a young man who goes on a pilgrimage to Manifest, in order to carry his grandfather's bones to the ghostwalk. I really liked that vibe - carrying a corpse to a particular location is a sacred duty. Some group of people in Ukss will feel the same way.