Sunday, May 1, 2011

Objections to Thieves

So, I've been pondering Thieves lately, and it seems like others have been too. It's been a while since there was a lot of talk about thieves in the OSR blogosphere, so I'm glad that it's coming up again. I like revisiting some of the same conversations over again because I think we get a chance to learn something new and develop existing ideas when we do.

I was just starting to lurk when they was a lot of discussion about them in the OSR, but discussion seemed to peter out when James released LotFP:WFRP and it included the Specialist class. Other bloggers that created "Old School" thief classes were James Maliszewski, Dyson and Telecanter.

I've been using a mash-up of Telecanter's Rogue and James' Specialist in my game, since Telecanter's Rogue doesn't include a system for adjudicating most of his skills, but it hasn't worked out as well as I would have liked. In retrospect, it was a good start, but it was pretty crudely hacked together, and it's time to try again.

So I've got a question for everyone. I've identified four Old School objections to thieves so far. They are:
  1. Thieves don't have an established archetype like the other classes do. All the characters, whatever their class, are thieves, or at least "rogues."
  2. Thieves make it so that other classes can't do what they do. Before Thieves came along, all the characters had to pull their weight doing thiefly things like disarming traps and picking locks and pockets.
  3. Thieves encourage skill systems and skill systems are (or at least can easily be) bad.
  4. Thieves are more likely to fail at their special skills than other character classes.
I'd like to be sure that my Thief class answers all, or at least most, of the OSR's objections so that it can be useful to others as well. Are there any other objections that I'm missing?

[And, yeah, go figure that I post something like this right after I announce a hiatus. I really will have limited time this next week or two, so I'll just be reading any conversation you all decide to contribute in the comments, and probably asking a few questions. I won't have any more posts about this until after the semester's coming to a close, most likely.]


  1. I've got an objection I've mentioned in a couple places that is related to #1 and #3: "Fighter" and "Magic-User" are archetypes, but "Thief" is a profession. In other words, the first two tell you what you typically do, while the latter tells you what you are able to do.

    I express archetypes as "I solve problems by ____". You can theoretically be any profession, but still be recognizable as a Fighter (solve problems by fighting back) or a Magic-User (solve problems by using magic.) I've toyed with archetypes that fill the thief role (solve problems with subtlety, for example,) but the thief as written is too focused on minutiae.

  2. Possible objections:

    Thieves discourage the creation of alternative fighter builds, eg fighters with high INT or DEX.

    Thieves encourage intra-party conflict, with tempting abilities to pickpocket friends etc.

    Personally I think the thief IS a mythic archetype in the Jungian sense, with figures like Renard the Fox, Raven, Prometheus... trickster figures in other words. And I like low-armour, high-dex swashbuckling heroes... the fighter tends to be punished for going that route.

    Hmm maybe 'trickster' would be a good alternative to a thief class? Seems like INT would be their prime stat though.

  3. Trickster is the name I use for my archetypal replacement. It does change the way thieves work, though.

  4. Thanks for the feedback! It's good food for thought.

    @Talysman: Have you written up a Trickster class? If so, I'd be interested in seeing it!

    @Dan: So far as inter-party conflict goes, that's something I've heard stories about, but never actually seen. I don't really understand why that kind of behavior seemed to be tolerated in those situations. Is there something I'm missing as far as why the rest of the PCs, in game, don't seem to kill the thieves when they catch them robbing the rest of the party?

    As for the different Fighter builds, that's a really interesting objection that I'd never heard before, and I think it is probably one of the best objections I've heard as well. I'll have to do some thinking about that. To help me think about it, do you have any examples of how you would run a high INT or DEX Fighter in a game without Thieves?

    On a slightly different note, one of the reasons I really like Telecanter's Rogue class is that different people can approach it with different archetypes, or facets of the Thief archetype in mind and come up with totally different "builds." If there isn't just one "Thief" archetype, or if there are multiple facets of the archetype, a class that accommodates those differences is attractive to me. I'd like to bundle traits of assassins, bards and Beyond the Black Gate's Megadungeon Explorer into the options for my Thief class. I can see all of them representing different facets of someone who solves problems with subtlety or bending/breaking the rules.


  5. Also, Talysman, thanks for giving me this way of thinking about archetypes. It makes sense to me. It's a bit more rigorous than just thinking up examples from inspirational material, but I think it has a good effect on the mechanics of Thieves (or whatever) if it's used to write them. I like the idea of Thieves using subtlety or bending/breaking rules or cheating, or some similar idea to solve problems, and I'll be applying that to whatever I come up with.

  6. Back when I wrote E&E, I addressed the issue pretty much the same way that the LotFP guy eventually did: remodel thieves as an "expert" class, drop the d% skills, and integrate a d6 skill system that experts are better at. It works like a charm.

    Lately, though, I've been thinking about how one would go about using thieves in a game without a skill system. (It's not my cup of tea, but apparently you have to do it at least once to count as a grognard[ling].) And I've been forced to arrive at the same conclusion as many: the thief d% rolls should count as an "extra" chance, over and above whatever a non-thief could attempt.

    So if a fighter has a 1-in-6 chance to find traps, the 1st level thief gets to roll 1d6 and d% and succeeds if either roll passes. And if a non-thief has no chance (e.g. climbing a sheer wall, hiding in naught but shadows), the d% roll is all you have, but passing that roll represents a truly extraordinary feat of agility.

    But leaving out a rogue-type class altogether? No way; they're too archetypical.

  7. I second the point that a Thief's abilities go beyond what a person is normally able to do. While anyone can sneak along, a Thief can Move Silently. A cat would not hear him. While anyone can climb a fieldstone wall or cliff or tree, a Thief can scale sheer surfaces like brick (or slimy whiteboard).

    Second, just bump up the percentage chances to at least 50% at first level and use a progression of +1-3% per level thereafter depending on skill.

    Third, Thief as a class is just a name. He's the guy who's skilled at dungeon delving and sneaking in general, at deciphering treasure maps and encoded messages, at slipping a knife into someone's back rather than a stand-up fight. I'd say that's as strong an archetype as "the soldier" or "the spellcaster."

    I don't know what to say about the skill system thing. Dwarves do it with underground feature detection and nobody cares.

  8. @Staples: I need to re-do the class system I've been working on, but here's the original version of the Trickster and here's a brief summary of the class which might be more useful.

  9. I've never understood the claim that thieves are not an archetype. Saying that all classes are thieves is like saying all classes are fighters -- true, but only so far. (Certainly from a Jungian perspective, the thief or trickster is as much an archetype as the wizard is.)

    Coming back to playing Basic D&D from 3.x/Pathfinder, the thief is the class that my players and I have found the most frustrating and least fulfilling. Skills are definitely the heart of the problem. I'm all eyes and ears for ways to improve -- better yet, fix -- the thief.

  10. @Theodoric:
    "Certainly from a Jungian perspective, the thief or trickster is as much an archetype as the wizard is."

    That's just it: the Trickster is an archetype. The thief isn't, because the class is presented as a profession: thieves know Thieves' Cant, belong to a guild, and have specific skills like Remove Traps and Pick Locks that should be just the result of training. In contrast, the M-U, at least in the early days, is just a character who casts spells, and the Fighter is just somebody who's good in a fight. Either can be reskinned easily without mechanical changes.

    A Trickster could be so much more than just a thief.

  11. Thanks to everyone who's commenting! This is all great food for thought for me. Unfortunately, I'm not going to have time to chew on it enough to contribute to the discussion for the next few days, or even longer. Please know that I am keeping track of the conversation, though.

    Carry on!

  12. Staples: regarding alternative fighter builds... I don't remember how viable these are and in any case it depends on exactly what system you are using; but they include:

    High DEX fighter can make an excellent archers. And proper archers, rather than the ranger image that all archers seem to be now. Also compare with the Scout class.
    High INT fighters may get a big advantage on initiative rolls - meaning they strike fast and first, as well as being better equiped to deal with tactics and non-combat problems. Shouldn't a general have high INT?

    Now depending on how the DM adjudicates things - whether they roll against attributes for improvised actions - a fighter with either of the above may find themselves doing all kind of thiefy things like climbing or tinkering or wearing less armour.

    On a slightly related note I think the whole modern "combat rogue" thing that evolved from backstab is just a different take on the fighter (separate from the actual thieving activities).

    Remember that Chainmail rules were originally designed to represent knights in plate armour - the system has sometimes struggled when asked to express fast moving swashbucklers; and especially when expressing both the above at the same time.

  13. Thanks, Dan, for clarifying that for me. I'm using Swords and Wizardry right now, so the particular high DEX and INT fighters you point out aren't things I've had to deal with. I do think rolling against attributes is very relevant to the way I ref, though, and can see your point there.

    I can also see your point with the "combat rogue." I've been thinking of, say, assassins as a specialization of a thief, who steals things: assassins steal lives instead of loot. I think that you can probably go overboard with letting rogue types fight well. I kind of think an assassin should be a decent fighter, on the level of, say, a cleric, unless he's got the surprise on his enemy. Of course, then there's the question of whether assassin abilities should be part of a rogue class at all… hmm, much to think about.

    Bottom line, though: a thief/rogue/trickster class should not infringe on or usurp the role of a fighting man.

  14. Talysman: In the long run, I think the argument of trickster vs. thief is a semantic one with no convincing resolution. If you look at the trickster tales, they abound with examples of thievery. What else does one find? Instances of deceit: which one could call robbing someone of their accurate perception of reality for one's own ends. So, a verbal distinction with no hard conceptual boundaries. Whether you call it a thief or a trickster, I think is just a preference of usage describing the same archetype with, perhaps, a little difference of emphasis. If I'm right, there's no substantial difference between us here. I'm happy to see Fr Dave tackle this from a Tolkenian point of view: