From a post on the rq-rules message group about character generation, written by Peter Maranci, who kindly gave me the permission to post it here too:
...
However, my very favorite method came from throwing out the rulebook altogether. We only used this for "serious" games, ones with an intense focus on roleplaying. The results were almost always spectacular.
The GM created each character by having a series of private discussions with each player over the course of several weeks. The conversations happened in person and on the phone (the net wasn't available in our dorms back then...it was the mid-1980s). We wouldn't discuss numbers. We'd discuss the personality and history of the character, along with the underlying concepts. And we'd do it in the terms that you'd use to describe a real person, i.e. NOT "he's got an 18 strength" but instead something like "he's apprenticed to the town blacksmith, and he doesn't know anyone in the village who is stronger than he is - although there are a couple of burly farmers who are probably about as strong".
Inevitably the character concepts evolved over the weeks of creation. The GM took sole responsibility for actually designing the characters, working out statistics by fiat. I don't know how other GMs who used this system did it, but *I* didn't make a fetish out of balancing the stats: instead, my goal was to make sure that each character had one or more interesting and useful abilities which were not duplicated elsewhere in the party. In other words, my goal was to balance playing time, plot involvement, and fun for the *players*, rather than balancing the stats of the *characters*. It helped that I knew all the players well, and could judge their strengths and weaknesses.
In the process, extensive histories for the characters naturally evolved. What's more, as things went forward the players would talk with each other and work out any shared history that their characters might have. We even tried playing out pre-campaign scenarios, talking in character to each other as our "young" characters in their earlier years. This built a sense of party unity and history that was unusually deep.
Typically, a game that used this method of character creation was run on a "sheetless" basis, i.e. the players never saw a character sheet for their character, never saw their own ability scores or stats. Instead, they knew as much about their characters as the character knew about him- or herself. Players would often keep extensive notes, of course, but they did not include numerical data.
The GM did all the system record-keeping, while the players rolled dice and employed strategy and tactics using a real-world approach. RuneQuest was an ideal system for this sort of sheetless campaign. Since it models reality well without being overcomplicated, the GM could easily accommodate the actions of the PCs without being TOO overburdened with bookkeeping.
The one flaw was that some people tended to lose track of the dividing line between reality and fantasy in roleplaying games. Sheetless gaming really seems to bring that out in some players. We all got obsessed - it's almost frightening how much more *involving* a sheetless game can be - but some people just lost it altogether.
But with good players, it can be a really mind-blowing experience.
It's not necessary to have a sheetless game in order to use the discussion method of character design, of course. The only problem is that argumentative players may object to the GM's actualization of their character, quibbling over stats or skills.
...
What do you all think? I found the idea of sheetless play quite interesting. Is there anyone else who have tried gaming like this?
Sverre.