A Demon Haunted World

[Mord]

Priests of Nerull are Necromancers, but they use exactly the same Necromancer class as the Necromancers that aren't Priests of Nerull. Any more meaning you ascribe to being a Priest or a Priest of Nerull specifically just makes the game worse. It makes there be more game mechanical choices, but most of those choices are fake. And it turns what should be a strictly flavor consideration of what god your character likes the most and turns it into a min/max choice that has right and wrong answers. That was a step back in 1980, and the only way to salvage it is to erase it.

Yeah, but it's cool for a player who runs a Necromancer of Nerull to be in some meaningful way distinct from a free agent Necromancer or a Necromancer of Evening Glory. It's just as much of a fake choice as the option of being a Half-Orc Wizard instead of a Gray Elf Wizard.

I don't want to throw the baby out with the bathwater. I think it's legitimately cool that someone could say, "I'm a Fighter for Pelor" and have that mean something mechanically different than someone else who is a Fighter for Hextor. If there's no mechanical difference, then it's going to end up being about as cool and memorable as a character's eye color or whatever.

That said, it's easy to say "let's have functional mechanical differences between deities" and it's hard to actually come up with an implementation that doesn't run into the synergy problems you described or turn into bland samey nonsense across deific portfolios.

[hyzmarca]

You could also tie clerics even more closely to their gods, and make them direct agents who must do the will of their patron.

Been there, done that. The result seems to be that no one wants to play the cleric. The possible risks of weaselling out of jobs are obvious enough that people don't try consciously, but they may forget (or 'forget') until it's too late.

That's why you make everyone clerics of Friend Computer, and have the missions from god be the entire point rather than just an annoying way for the GM to screw you.

[Koumei]

If there's no mechanical difference, then it's going to end up being about as cool and memorable as a character's eye color or whatever.

Does your eye colour ever inform the decisions you make? Like, do you go "I think we should tackle this quest because I have blue eyes" or "As someone with heterochromia, obviously I'm going to take the word of the noble here and not the serf"? If you worship one deity over others, that's likely to influence the things you do in the game.

[hyzmarca]

If there's no mechanical difference, then it's going to end up being about as cool and memorable as a character's eye color or whatever.

Does your eye colour ever inform the decisions you make? Like, do you go "I think we should tackle this quest because I have blue eyes" or "As someone with heterochromia, obviously I'm going to take the word of the noble here and not the serf"? If you worship one deity over others, that's likely to influence the things you do in the game.

Well, your eye color usually proves that you're the biological heir of the old king, who everyone thought had been killed by revolutionaries during the coup and thus drives your decisions toward retaking the throne that is rightfully yours, and also killingyour optometrist before he can tell anyone that you wear colored contacts.

[Grek]

Yeah, but it's cool for a player who runs a Necromancer of Nerull to be in some meaningful way distinct from a free agent Necromancer or a Necromancer of Evening Glory. It's just as much of a fake choice as the option of being a Half-Orc Wizard instead of a Gray Elf Wizard.

Weirdly enough, I think 5e has the right idea here with Backgrounds. The difference between a Priest of Nerull and a Secular Necromancer is that one has the Acolyte Background and is allowed to crash at Nerullite temples and meet with the High Priest and gets a free holy symbol at the start of the campaign, while the Secular Necromancer took the Criminal Background and gets a bunch of black market contacts and proficiency with lockpicks.

[Username17]

Only that last one really requires something we'd recognize as a religion. For the others, they can have patron saints and generic non-denominational holy men wandering the countryside or being anchorites without needing any theology at all.

Exactly. There's a tactical need for Githyanki Strike Forces and Drow Raiding Parties to have armorered warriors who provide defense bonuses and healing. So the Paladin class is a good idea. But there's no reason that those armored warriors should be the religious leaders of Githyaki or Drow cities. The priestesses of those civilizations shoot purple energy out of their hands and dress in skin tight strappy outfits.

There's a need for a Paladin class. And there's a need for there to be priests of various civilizations. But there's no need for the priests of a lot of civilizations to be recruited from the ranks of the paladins and there's equally no need for most paladins to serve as religious authorities.

Paladin stays. Cleric goes.

Weirdly enough, I think 5e has the right idea here with Backgrounds.

I agree. In fact, I liked the 5e Backgrounds so much that when K and I wrote them into D&D we called them "Backgrounds" and wrote them up in 2006.

-Username17

[Thaluikhain]

But what's the typical holy folk of Lathander, Pelor or Yondalla even are?

Paladin would be a fine fit if you're keeping that class around. One of the side bennies of getting rid of the all-in-one cleric is that Paladin concept of being a heavily armored energy channeler could actually be halfway distinctive for a change. Even as a young man I thought it was sort of dumb how you can't really tell apart a Lawful Good Cleric from a Paladin or Fighter-Cleric multiclass without knowing the math or spell lists like the back of your hand.

Out of interest, do you see the same problem with rangers and fighter-druids? Or, for that matter, perhaps even a fighter-cleric or paladin of a nature based deity?

[Username17]

Out of interest, do you see the same problem with rangers and fighter-druids? Or, for that matter, perhaps even a fighter-cleric or paladin of a nature based deity?

I don't see Rangers as inherently similar to Druids as Paladins are to Clerics of Healing and/or War are. But the big issue you're poking at the edges of is that 3rd edition's multiclassing system wasn't actually a good idea. In previous editions, Fighter/Druids simply didn't exist at all. And I'm unconvinced that adding Fighter/Druids to the game was a good idea.

So first of all, I can't tell the difference between someone who is 60% Scout 40% Swashbuckler and someone who is 70% Scout 30% Swashbuckler. Those things sound essentially identical from a narrative standpoint, so the extra granularity offered by 3rd edition's open multiclassing isn't actually good for anything. But the second issue where multiclassing was opened up to include character classes that were already secondary and tertiary character concepts isn't actually helpful. Why are we multiclassing Scout and Swashbuckler together at all? Such a thing wasn't possible in AD&D, and I haven't seen a great deal of evidence that it's a good idea now.

Now a caveat to all this is of course that you're going to want to have some sort of secondary shtick support that is more open ended than what AD&D offered. Because you want to be able to play things like Frost Giant Shamans without requiring that Dwarven Shamans get 9th level Dwarf powers. So the Frost Giant sacrifices his secondary shtick option to get level appropriate Giant abilities, while the Dwarven Shaman doesn't and can have some other secondary shtick. And since it would of course be incredibly stupid to tell us that Frost Giants or Fairies can't be Druids, you're obviously going to want to have Druids be capable of splitting off their secondary shtick - which means that it should be possible to be a Druid with Berserker or Hero or something as a subclass. But that still seems easy enough to distinguish from Ranger.

-Username17

[CapnTthePirateG]

Now a caveat to all this is of course that you're going to want to have some sort of secondary shtick support that is more open ended than what AD&D offered. Because you want to be able to play things like Frost Giant Shamans without requiring that Dwarven Shamans get 9th level Dwarf powers. So the Frost Giant sacrifices his secondary shtick option to get level appropriate Giant abilities, while the Dwarven Shaman doesn't and can have some other secondary shtick. And since it would of course be incredibly stupid to tell us that Frost Giants or Fairies can't be Druids, you're obviously going to want to have Druids be capable of splitting off their secondary shtick - which means that it should be possible to be a Druid with Berserker or Hero or something as a subclass. But that still seems easy enough to distinguish from Ranger.

-Username17

This seems like it would be just as open to abuse as much as having Pelorian clerics get special powers. You are definitely gonna have combos where Necromancer + Bard synergizes better than Necromancer + Archer.

[Username17]

This seems like it would be just as open to abuse as much as having Pelorian clerics get special powers. You are definitely gonna have combos where Necromancer + Bard synergizes better than Necromancer + Archer.

Well yes. Multiclassing rules are tremendously open to abuse and also create a lot of fail states. While it's trivially easy to avoid many of the specific pitfalls of multiclassing rules of specific previous editions, it's definitely true that you're going to have some kind of synergies and anti-synergies. Like, the thing in AD&D where a Fighter Wizard is exactly the same as a Wizard Fighter except that he has 21 extra hit points and a higher strength modifier is super bullshit, but it's difficult for me to imagine a scenario in which Berserkers get exactly equal benefits from splashing Sorcerer as they do from splashing Ranger.

As such, I think it's virtually inevitable that some multiclass combinations are going to be deprecated and others are going to be popular and while you should take active steps to minimize the real discrepancies as much as possible, it's absolutely ludicrous to expect that this won't be the case.

The thing is that while it's true and unfortunate that not all multiclass combinations are going to be well thought of, you aren't actually promising all the possible multiclass combinations when you make the book. You make a Druid class and a Necromancer class and you are promising that both of those are going to be playable. But you aren't specifically promising that a Druid/Necromancer or a Necromancer/Druid is playable. It would be nice if they both were, but you haven't made any specific promises along those lines. That's very different from a choice like "Worships Pelor" which you have committed to making a viable life choice by including it.

-Username17

[CapnTthePirateG]

The thing is that while it's true and unfortunate that not all multiclass combinations are going to be well thought of, you aren't actually promising all the possible multiclass combinations when you make the book. You make a Druid class and a Necromancer class and you are promising that both of those are going to be playable. But you aren't specifically promising that a Druid/Necromancer or a Necromancer/Druid is playable. It would be nice if they both were, but you haven't made any specific promises along those lines. That's very different from a choice like "Worships Pelor" which you have committed to making a viable life choice by including it.

-Username17

Sans outright calling this out in the book, I'm not sure where you're getting the idea that if you offer a necromancer/druid as an option people should secretly expect it to be ineffective. Generally in UI design people are taught that if it's an option they should be allowed to do it, if it's a waste of time why have it at all?

We've already established that part of the classplosion is going to be shit like swordmages, I don't see why you can't just put out other classes for the actually popular multiclass archetypes. The only thing you actually lose is Frost Giant Necromancers, and "playing a frost giant" is so far outside the realm of things people normally get to do that they'd be happy to just play a frost giant class.

[nockermensch]

The problem is that it's fucking 2018 and if you offer a D&D game "without multiclass" people will say you're not delivering the full experience.

Even if about 50% of the multiclass complaints seems to be actually about fighter/mages (and in this case you simply write a Duskblade, or Magus, or whatever), some people will come with character concepts like "it's a paladin, but knows some acrobactics and low level thievery stuff", or "she started as a sorcerer, but then joined a monastery".

And from a purely world building perspective, you also have to say what happens when characters who were already something change careers: The party captures a young thief after some shenanigans, and after asserting that the kid was just misguided and hungry, the magician decides to take him as an apprentice. Assuming that the kid has good Intelligence and etc, but is already "Rogue 1", what happens next?

Actually, most of the "reasonable" explanations people seem to come for multiclassing characters seem to be career changes right at the start. In other words, dips. In other words [2], a lot of multi-class character concepts seem to be really about people wanting to broaden the low level stuff their characters can do.

If this is correct, you don't need a truly open multiclass system. You need instead a way to make characters keep something from the levels 1-3 of each class if they decide to change careers. Before 3E and the mental gymnastics people got used to qualify for prestige classes, I never saw people legitimally wanting to take a level in Ranger or Necromancer once they were already competent at their choosen field.

[DrPraetor]

On the original topic, I'm still not entirely clear what the society looks like in a post-liberation demon haunted world.

Are people playing your hex combat army game at the same time? I can see how the setting would mesh with that, but it would require some work.

Is there a Church of Light Brotherhood of Steel International Guild of Murderhobos from which the players are expected to get missions? Is there more than one such organization and they get along well enough for players to share missions from different ones?

Suggestion: people may be members of thematically-distinct clans, but the Demons moved populations around as part of a strategy of control (and because they're dicks) so there are a lot of displaced people in any populated area, but they all joined the alliance. This also provides some clean historical precedent for how the fiendish empires were run: https://en.wikipedia.org/wiki/Populatio ... nt_Assyria
If you have an irredentist view that Elves should return to Murkwood and cleanse the sacred groves with the blood of the filthy goblins you are not really a member of the alliance, although such groups would make good evil cults and internal adversaries.

So even if Agrabah was a Scorpion Clan-Goblinoid monoculture in the before time, there are a bunch of Willow-elves and Oak-elves here now and they've A) sworn fealty to Sheik Gorbag who led the local struggle against the Fiends, and therefore B) have primary allegiance to mixed troubleshooter gangs that all support Agrabah FC even if they have some conflicted loyalties to the Willow and Oak councils back in the old country (who don't entirely get along.)

===
On multiclassing, for the umpteenth time. Okay, let me float a proposal.

Classes are 5 levels long and come in tiers. So "Fighter" (or whatever you rename it) is a 5 level class, and at level 6 you are a Cavalier or a Slayer or a Zen Archer or whatever.

You also get Kits, which give you a list of level-scaling abilities and may very well be subclasses like Petty Sorcerer or Thief. Meaningfully-non-human races count as at-least-one of these (so a Druid/Dwarf would have Dwarf as a subclass, we'll get to Giants in a bit); you select two kits at first level, and another kit at levels 3, 6,8, 11,13, 16,18.

Elite races - Gith, Thri-Kreen, stuff like that - could count as 2 kits.

Advanced races - where every monster in the village is opposition for a medium level party - wake up with five monster hit dice and anywhere from 3 to 0 kits. So a basic centaur, gargoyle, hill giant, troll, medusa (sorry Prak) etc. all show up with 5HD and if you want hill giants to be bigger and meaner than trolls you stat up a "hill giant militia" as the default opposition and give them a class level or two.

You can play as advanced races if you are level 6+: you go straight into a 2nd tier class and you give up some or all of your kits.

There are also Lord-tier Races where everyone has 10+ HD, and Demigod-tier where everyone has 15+ HD.

[Username17]

On the original topic, I'm still not entirely clear what the society looks like in a post-liberation demon haunted world.

It has approximately 20% of the population working the farms with the typical city worker getting about a pound of silver (50 sp) per month and paying 3 copper per day for regular meals (9 sp). It has a rough power-rule of income where 1/10th of the population makes four times as much as the remaining 9/10ths. So in a village of 100, you expect there to be 90 people making it on about 5gp per month, 9 people on 20 gp per month, and one asshole making 80 gp per month, which converges to a GDP of about 90 gp per person per year.

More culturally speaking, it looks like a Dungeons & Dragons society with a lot of ruins near populated areas, a bunch of evil cults that are surprisingly well connected, a tradition of sending small armed strike forces to deal with monster problems and a lot of monster problems considering the populations involved. It also has a bunch of cultural artifacts left over from fiendish occupation. So Halflings have been forced to make factories that produce steel goods and the Human empress has bat wings.

Are people playing your hex combat army game at the same time? I can see how the setting would mesh with that, but it would require some work.

The intention would be for the players to graduate to hex combat army games later in the campaign. So the first thing they do is have a couple of Keep on the Borderlands or Descent Into the Depths of the Earth style adventures where the players go explore an area and find areas that are still under the control of Fiend Lords or Fiend Lord sympathizers and then they fight it out with the leaders of small forts or villages. And then when they have conquered a bit of area they can start doing kingdom management shit. Benighted Troglodyte warrens probably have a GDP of like 30 gp per person with more than half the population doing fishing and mushroom gathering and shit. So there's a lot of room for improvement there.

In any case, I'm looking at an Adventurer -> Conqueror -> King progression, but hopefully without the OD&D throwback bullshit.

Suggestion: people may be members of thematically-distinct clans, but the Demons moved populations around as part of a strategy of control (and because they're dicks) so there are a lot of displaced people in any populated area, but they all joined the alliance.

That's a good idea, yeah. Rather than have universities or even guilds that could train people in skills, the Fiend Lords probably just bought and sold groups of people from other Fiend Lords who were expected to have the needed skills. "We need some miners, let's buy some Kobolds from Tiamat."

-Username17

[Koumei]

So out of curiosity, given the discussions on classes and prestiging and multiclassing and so on, how much of this is "Frank is running basically a D&D game, using Tome stuff presumably then adding bits of other stuff later like the Hex based things, to entertain the other doctors when they're all on break hahaha just kidding, it's England, there are no breaks for doctors" and how much is "Frank is assembling a solid setting to go with the new rules he's putting together to lead the world into a golden new age of Post-Tome D&D-ish stuff"?

[Username17]

So out of curiosity, given the discussions on classes and prestiging and multiclassing and so on, how much of this is "Frank is running basically a D&D game, using Tome stuff presumably then adding bits of other stuff later like the Hex based things, to entertain the other doctors when they're all on break hahaha just kidding, it's England, there are no breaks for doctors" and how much is "Frank is assembling a solid setting to go with the new rules he's putting together to lead the world into a golden new age of Post-Tome D&D-ish stuff"?

A little more on the latter side of things, to be honest. I've long felt that to actual move forward with D&D rather than just throw ever more epicycles onto 3rd edition that you'd have to redesign the core challenges. That is, you have to make a new monster manual if you want to determine what kinds of spells that Sorcerers need, but you also need that monster manual in place to determine when character concepts like "Ranger" stop being salvageable.

But doing that it occurred to me that the most important question is not actually "What does a CR 11 Giant look like?" but rather "What does an 11th level campaign look like, and are you fighting Giants in a meaningful way?" That while it's possible to present the 4th edition "Always Fighting Orcs" model with a serious of palette swapped bruiser monsters with increasingly large numbers, but that actually isn't interesting or meaningful. 4th Edition's presentation of what high level adventuring meant was so terrible that it made me physically ill to read about it. While there should probably be some 11th level monsters that you are expected to face at 11th level, the question of what you're doing and why you might be fighting such a thing is far more important.

Questions of campaign progression also go to questions of acquisition of material wealth. Once characters could stop fighting mummies in sewers and start pursuing other avenues of wealth acquisition, it becomes important to understand what would happen if they did that. Thus, the question of "What is the price of tea in Kara-Tur?" is actually more important than the question of how the numbers are different on a Fire Giant and a Cloud Giant. It's the work of moments to figure out the street value of the raw materials presented after you turn a cow into salt or create a Wall of Iron from the aether - but the broader reality is that whatever the economics or the abilities in the setting do, that players are going to be able to get the equivalent of many pounds of gold at some point if that is what they want to do.


And that's assuming that the enemies don't simply have giant piles of gold in the first place, which is a pretty dumb assumption in my opinion.

So it means that one of the main issues that is facing medium and high level play is actually "What do high level characters do with gold?" and thus the economics of raising and feeding armies and building castles and shit is actually a significant and central point if you want to design a new edition. The reason that 3rd edition's forays into Stronghold Building and Heroes of Battle and shit were so half-assed is that the finances of such weren't hammered in to the core rules at all. Instead we were told that a level appropriate sword for a 14th level character cost 50,000 gp, which meant that getting chainmail onto a few hundred Hobgoblins was just never going to happen and it would ballbust the expected balance points of the game for the players to even get remotely close to controlling enough wealth to seriously try.

So here I am, working on the back end of how the kingdoms work and how it would work for the player characters to raise an army and build a fortress because all of those questions have to have solid answers before it makes sense to even ask what abilities high level characters can, can't, and must have.

So yes. I'm working out how much money you could make by buying Drow-made lingerie in the Underdark and selling it in Human or Elf majority cities on the surface. And then I'm working out how many Orcish medium infantry you could hire for the proceeds. Because while I don't expect every, or even most players to spend time playing Starflight 2 and arbitraging their way to wealth and power, the numbers have to work out if that is a thing you do. And further, that whatever you do to accumulate wealth and power, whether it be founding a merchant clan or taxing a developing agricultural region, or looting a dragon's cave, the end result should be that you build a fortress and raise an army and not that you spend your every last copper piece on a shinier sword to go murder hoboing with.

Regardless, I'm aware that I'm going to need actual other authors when it comes to crunch time of filling in all the monsters and class features and shit. But for now, I'm content to sounding board broader ideas while I crunch the math on how many Goblins it takes farming rice to keep a Hobgoblin army in the field.

-Username17

[angelfromanotherpin]

Frank, have you looked at Riggsby's various low-tech economy articles for GURPS? He's a pro scholar writing for a gaming perspective.

[CapnTthePirateG]

The problem is that it's fucking 2018 and if you offer a D&D game "without multiclass" people will say you're not delivering the full experience.
.

You probably could get away with slaughtering that particular sacred cow, 5e killed vancian casting and literally no one cares. As long as it's not locked to more godawful design like 4e/5e people will be fine with it. No one actually likes multiclassing as far as I can tell, it just produces overpowered or underpowered characters and is just despised by everyone. As you mention career changes can just be handled by background and some mumbling about how you've forgotten all your old necromancer spells because you never use them anymore.

[Trill]

You probably could get away with slaughtering that particular sacred cow, 5e killed vancian casting and literally no one cares.

I'm guessing it's less because it's not a sacred cow anymore, and more because it's 5e and nobody cares about 5e

[Username17]

The problem is that it's fucking 2018 and if you offer a D&D game "without multiclass" people will say you're not delivering the full experience.
.

You probably could get away with slaughtering that particular sacred cow, 5e killed vancian casting and literally no one cares. As long as it's not locked to more godawful design like 4e/5e people will be fine with it. No one actually likes multiclassing as far as I can tell, it just produces overpowered or underpowered characters and is just despised by everyone. As you mention career changes can just be handled by background and some mumbling about how you've forgotten all your old necromancer spells because you never use them anymore.

While it is factually true that you don't need people to be able to play Rogue/Wizards when they can jolly well play Assassins and have sneaky tricks and also some magic, that is not the only thing that Multiclassing does. Multiclassing has done different things in different editions, and filled various needs to greater or lesser extents.

The most pressing thing you need to do with your multiclassing rules is to allow for high powered monsters as characters. It's fine to have players play Orcs or Elves and have some sort of bullshit low level character package that they get for that, and even races that are modestly powerful like Wererats and Dryads can be handled with level minimums and blocked out magic item slots. But what about races where the race is genuinely powerful and capable of solving level-appropriate tasks by virtue of its providence? Sometimes players are going to want to play Succubi or Frost Giants or Air Elementals? Those races are damn near an Enchanter, Berserker, or Assassin just for getting up in the morning. And yet, people aren't just going to want to write "Giant" on their character sheet, they are going to want to play an actual character. A Giant Shaman or a Giant Ranger. A Succubus Paladin or a Succubus Druid. Your system wants to handle those things. It has to handle those things, because it's a design requirement from decades ago and failing to hit it with a new edition is fucking inexcusable.

But here's the thing: once you've created a system by which someone can play a cut-down version of a Druid and hybridize it with the Tempter Fiend Racial Class, people are going to ask why they can't play a cut-down version of a Druid hybridized with a cut-down version of the Assassin. And there's honestly no reason why they can't other than the fact that you obviously have not playtested all 136 two class combinations (272 if order matters) in the basic book and there's even less chance that you've playtested all possible multiclass combinations involving classplosion classes from expansion books.

-Username17

[hyzmarca]

You could also decide to just give up on game balance and give races the full benefit of their Monster Manual entry without jumping through hoops like a racial class.

And while I'm sure it seems incredibly regressive to just give up on the idea of game balance, a cooperative game actually can have a PC who is just better than everyone else, just so long as he's Superboy in the Legion of Superheroes better and not Superboy in the League of Assassins better.

[Foxwarrior]

By sheer virtue of the fact that there aren't extremely draconian rules about level design and what sort of campaigns you can run, the implications of making one new book a month, and the players getting character customization options at all, there's likely to be some PCs who are better than other ones. You don't need to actively give up on game balance for that to happen to some extent, but giving up does increase the odds of it getting too unbalanced to handle.

[DrPraetor]

In order to be better than D&D 3rd edition - and therefore worth writing - your heartbreaker doesn't need perfect balance (which would be nonsensical anyway, given different campaigns would have different foci), but you need enough balance that the major archetypes share the limelight against the appropriate challenges.

Pyromancers can go ahead and be somewhat better than Psychic Warriors but the Psychic Warrior had better contribute something meaningful to level-appropriate challenges.

For bonus points, your hearbreaker can also hold together better at higher levels, through some combination of a kingdom-management minigame and a thing-that-Exalted-pitches-to-you that doesn't blow choad.

[Judging__Eagle]

Frank, have you looked at Riggsby's various low-tech economy articles for GURPS? He's a pro scholar writing for a gaming perspective.

I've been taking a look at GURPS Low-Tech over at the rem.uz website, and it's an interesting read to be sure. I wish I had read it before the "Tech Levels" thread wherein people were trying to define the graduations between TL0 and TL1 in trying to discuss societies more futuristic than the 20th/21st century.

[Username17]

Pathfinder experimented with the "pay nothing" school of monsters as player characters. The end result is that Minotaurs and Drow Nobles and shit are so obviously more powerful than Dwarves and Grey Elves that Mister Caverns won't let you play them at all. It ends up beinf even more restrictive than 3rd edition's "pay too much" system. At least in 3rd edition you had the option of playing an exotic underpowered character. In Pathfinder it's just hard banned at every table.

Getting the powers of a Half Fiend or a Medusa or a Vampire has to have a cost because it obviously has a benefit. There's a lot you can do with level minimums and expended magic item slots, but a Half Fiend player probably wants to pick up new Fiend powers as they level up while still being a Paladin or Assassin or whatever, and what is that if not a multi-class progression?

Further, on a game design end you're already going to want to design a Giant progression that is less than the Berserker player character class because you have Giants as small-group thug monsters at almost every level. Once you've created what is essentially a character class that outputs something less than a full player character, why wouldn't you let people staple some additional class-themed abilities onto it and call it a player character progression?

-Username17