“Defining the OSR”

I’ve seen a few people get quite territorial over the OSR acronym — mostly as a reaction to including modern games that emulate OSR-style play such as Dungeon World and Torchbearer.

There are arguments for inclusion and exclusion, but it’s worth noting that OSR stands for Old School Renaissance and a renaissance is above all a movement. This means that even with the roots of the OSR firmly planted in 70s D&D, this movement can embrace new and innovative ways to play with “old school feel” and up-to-date rules — following a different evolutionary path to D&D 3rd edition and later.1

One problem that’s compounded the OSR is its image as The OSR. Wikipedia presents the OSR as exclusively D&D retroclones, to the exclusion of modern pretenders and also all the other games we played in the 70s and 80s. I’ve found this definition helpful: 

OSR : renewed interest in playing and emulating games from the early days of the hobby

The OSR : a marketing umbrella term for clones of Original D&D.

(this definition has been borrowed and adapted from this fellow. Take it with a pinch of salt, he is a pirate.)

Now, this may sound like a complaint — that The OSR is somehow getting in the way of many other worthy candidates for their own old-school revival. Nothing gets nearly as much attention as D&D and its clones, despite clones of contemporaries in the works (OpenQuest, Zweihander, probably clones of Gamma World and similar games, too). But I don’t have much of a problem with that; I’m sure these games will continue to have their fans and quietly do their thing — and maybe they’re better off differentiating themselves in a market saturated with D&D me-too games.

And most importantly, without The OSR there would probably be no OSR. Yes, people like me would continue to blog about games from their youth, but it would be pure nostalgia, not progress. I happen to think that a lot of the things that the OSR delivers is progressive, even when looking back to games as old as I am.

One subtle property of the OSR is avoiding reinventing procedure, analysing the end product and matching it to player expectations and overall game feel. I have enough “new” games to last my lifetime, but there’s always room for discussion on how to revise and improve the existing product. I’m thinking about LotFP‘s encumbrance rules, Dungeon Crawl Classics character funnel and spell fumbles2, and similar innovations.

Some of the other qualities the OSR brings to light include:

  • mission focused games
  • location based games
    • with incumbent resource management (dungeon exploration, etc)
    • the infamous “sandbox”
  • player-GM adversity (I don’t mean outright GM-trying-to-kill-the-party adversity, but rather a strong focus on challenge and consequence of failure)
  • a strong focus on what the game provides based on the mechanics it provides


p>None of these are new ideas — but I think a lot of flabby, incoherent games could benefit from some introspection based on the above and pruning out the bits that don’t really serve a purpose.3

The OSR is not (just) a nostalgia trip. It’s not a bunch of reactionaries refusing to embrace the modern on principle. It’s not hipsters gaming ironically. It’s not just D&D, and it’s not just marketing. It is an opportunity for debate on what really works in games and delivering a lean, coherent product. And on that note I think Ron Edwards actually has a point when he describes OSR as “the Forge with vision”.

  1. I have never owned any version of AD&D, and only played a bit of 1e AD&D, so I’m not sure what the main evolution is. I can hazard a guess given what I know of D&D4e — feats, healing surges, attempts to emulate World of Warcraft. That’s about all I know.

  2. I also refrained so far from actually buying DCC, so my knowledge is secondhand. I’m not sure if it harks back to D&D 3e or an earlier edition, but from the size of the book I guess it’s a medium weight system. It looks gorgeous, though.

3. I guess the generic and all-encompassing toolkit has its place for a lot of people, but if I want a properly generic system I’d pick something minimalist.