The announcement of Dark Sun as the next 4th edition campaign lead me to go take a look at the Dark Sun boxed set, a gift I received a couple years ago but never actually dug into much.

One of the more unexpected surprised of Dark Sun was the introduction of character trees. In a nutshell, in a campaign using this concept as written, each player would roll up four different characters. They could choose which character to play during each session, and whenever the character they were playing gained a level they could also have one of the inactive characters (of a lower level) also gain a level.

The idea was two-fold; first, it would allow the players to have a stable of characters so that they could draw on the one best suited for a particular adventure or even their mood. Second, in the harsh world of Dark Sun, starting out at lower levels is especially bad (or so they make it seem), and this gives the players a bit of a buffer for when they lose their best character.

It’s an interesting concept, and one that I think could be successfully applied to modern games. So let’s look at some possible implementations.

Classic Character Trees

Exactly as described above, or similar at least. Each player gets a stable of characters, and gets to choose which to play during any given adventure. Whenever the active character gains a level, one of the inactive ones does as well.

Pros
This style would work for any system where it’s not uncommon or particularly undesirable for PCs to have differing levels of experience. Point-based systems in particular, such as Savage Worlds, White Wolf, or Shadowrun would work well here, although you’d have to adjust it a bit. For example, if a player earns 20 experience points, they add the full 20 to the active character and also get to distribute 20 points among the inactive characters however they want.

This approach is also nice because, simply by virtue of having several characters who each necessarily have “off-screen” things to occupy their time when they’re not adventuring, it ties the players more strongly to the game and to the setting.

Cons
In games like Third and Fourth edition D&D, a character who’s not close to the average party level (and, by extension, the level of the encounters they’ll be facing) is mathematically moot. At best they’re ineffectual and boring, and at worst they’re a liability to themselves in others. Simply put, this approach will not work in such a system (but continue reading…).

Fully-balanced Character Trees

As above, but all characters in the tree are always kept at the same power level, either by splitting up exp evenly or just advancing all inactive characters at the same time rather than one at a time.

Pros
This is a necessary fix to the concept for certain systems, such as the aforementioned D&D third and fourth editions. It also prevents an unplayed character from falling too far behind and becoming “unplayable” (or, at least, unpalatable).

Cons
The rapid gain of levels and powers among four separate characters may prove difficult for some players to come to terms with, particularly in the case of new players or even experienced players who are playing complicated classes or classes which they don’t have much experience with.

UA Gestalt

In the Third edition Urban Arcana book of optional rules, they presented a variant called gestalt in which players pick two classes and smoosh them together, gaining all the benefits of each class and, in cases like saving throws and such, taking the better between the two classes.

Pros
It’s fairly simple and organized. You only need to keep track of one character. The player always has access to all of their toys. It makes some character options much more feasible (e.g. fighter/mages in D&D third edition are usually fairly crippled unless you’re playing a class specifically designed as such. A gestalt fighter/mage, though, would be quite capable in both respects).

Cons
It absolutely requires a class-based system. I can think of no way to adapt this as-is to a point-based system (but, again, continue reading…). It represents a definite power creep, though not quite as much as you might think. Gestalt characters are just plain more powerful than non-gestalt characters, especially if they min-max their choices to minimize weaknesses and maximize strengths.

Final Fantasy Gestalt

I’m not sure if this is a concept or not unique to the Final Fantasy franchise, but it’s the only example I can think of off the top of my head. In essence, your character can have multiple different classes… but only one at a time, and they’re all kept separate and distinct. So for example you could be a level 5 fighter/6 wizard, and at the beginning of each adventure you’d choose whether you wanted to be in “fighter mode” or “wizard mode.”

Pros
This adopts many of the pros from the character trees above, since that’s basically what you’re doing. You have multiple different character sheets, but rather than each representing a different character they all represent the same character wearing a different hat.

It does have the obvious advantage over the character tree method in that it’s a bit simpler to maintain party cohesion if, story-wise, it’s always the same character.

Cons
That’s a double-edged sword, however. If you’re into verisimilitude, then this might not be the best approach as you have to figure out why your character was swashing buckles with the best of ‘em yesterday but today he can’t carry a sword to save his life.

Character Pool

The final option to consider is a pool of characters which don’t belong to particular players. Instead, each player chooses a different character each game. You could even take it a step further and have some extra characters, so that if you’ve got, say, five players you have seven characters and thus there’s a bit more choice.

Pros
It strikes me that this could be great for groups which have trouble with consistent turn outs. Bob’s not here this week? No biggie, Dan’ll play the cleric. If you go with the extra-characters-in-a-pool idea, this could also be a great way to gracefully “absorb” temporary additions, such as potential new players or friends visiting from out of town.

You also only have to worry about one character per player (or a bit more if you’ve got a pool), which is always nice.

Cons
This could make it difficult for players to become attached to characters and, by extension, the game itself. It also might end up a bit of a self-fulfilling prophecy, as it removes a bit of the social obligation to show up when you know the party won’t be completely screwed without you there.