Why I Dislike the term Railroading

[Doug McCrae]

First is the insistence that this is the One True Campaign by the rules. That other styles of Campaigns are wrong as they don't follow The Rules set out by the creators of the game. As pointed out by another poster, the original rules were not all inclusive and therefore other emergent styles of Campaign play are no more wrong than developing rules for how a character can jump a gorge. Those who insist or imply that the type of Campaign you describe is the One True Way are the first cause of friction and ill feelings related to this subject among those in our greater gaming community.

I was wondering about this. In several places, 1974 OD&D says the GM must have a mega-dungeon ready before play can begin. Wilderness and town adventures are also discussed but they are more of an adjunct to the centrepiece of play, the dungeon, which is vast.

There's also a fair amount of material on building a stronghold and fighting involving armies. In some respects, OD&D is halfway between rpg and wargame. It even describes itself as a "fantastic-medieval wargame".

Otoh it says that play does not have to be medieval, that it can "stretch from the prehistoric to the imagined future, but such expansion is recommended only at such time as the possibilities in the medieval aspect have been thoroughly explored."

The most extensive requirement is time. The campaign referee will have to have sufficient time to meet the demands of his players, he will have to devote a number of hours to laying out the maps of his "dungeons" and upper terrain before the affair begins.

First, the referee must draw out a minimum of half a dozen maps of the levels of his "underworld", people them with monsters of various horrid aspect, distribute treasures accordingly, and note the location of the latter two on keys, each corresponding to the appropriate level.

Before it is possible to conduct a campaign of adventures in the mazey dungeons, it is necessary for the referee to sit down with pencil in hand and draw these labyrinths on graph paper. Unquestionably this will require a great deal of time and effort and imagination.

The so-called Wilderness really consists of unexplored land, cities and castles, not to mention the area immediately surrounding the castle (ruined or otherwise) which housed the dungeons. The referee must do several things in order to conduct wilderness adventure games. First, he must have a ground level map of his dungeons, a map of the terrain immediately surrounding this, and finally a map of the town or village closest to the dungeons (where adventruers will be most likely to base themselves).

"Blackmoor" is a village of small size (a one-horse town), while "Grayhawk" is a large city. Both have maps with streets and buildings indicated, and players can have town adventures roaming around the bazaars, inns, taverns, shops, temples, and so on. Venture into the Thieves' Quarter only at your own risk!

Whether this is how the game had to continue is another question. AD&D 1e puts far less emphasis on mega-dungeons than OD&D. Ariosto's preferred playstyle is a very OD&D-influenced brand of AD&D, imho.

 

[awesomeocalypse]

whatever you want to call the playstyle, and I'm absolutely fine with calling it a "railroad" even if others are bothered by the stigma, adventures with some sort of largely preplanned plot are just about the only types of game I've ever enjoyed.

Above all, I play d&d to feel like the hero of a fantasy story--not the "protagonist", or the "main character", and certainly not just "some dude who lives in a fantasy world", but the hero (or one of them, at least).

Far too many "sandbox" campaign I've played have utterly failed to deliver on this point--we've walked around doing what "we want", in the sense of pursuing individual goals that happen to interest us, and we've come into conflict with NPCs who stand at odds with those goals. But I've never played in a sandbox which organically produced the "the world needs saving, and you're the only ones to do it" plot structure that I look for.

In other words, I don't want to be some guy strolling around the shire or skirmishing with orcs on the plains of Rohan. I want to be frikkin Aragorn or Gandalf or Frodo, the "chosen one", "big damn hero", "here to save the world" guy whom the entire world essentially revolves around in a metaplot sense. Maybe thats narcissistic, but its what I look for, and if I don't get it then I generally don't have fun.

In my experience even an average DM if given time to prepare and lay out a campaign can reliably produce this sort of experience. Producing it organically out of pure sandbox play is much, much harder--I'm not saying its impossible, but I've never played in a game in which it happened.

 

[Vyvyan Basterd]

I think that the contents of the booklets are plain for anyone to see. I know what I, and everyone I know of, produced by following those instructions.

I think you are reading too much into this.

I may be, but being told that the instructions of the game tell me "This Is The Way It Is" kind of lead me into that line of thinking.

I'm not saying everyone who speaks on this subject believes this. I'm not even saying that Ariosto believes this. But my perception is that I'm being told that I'm doing it wrong because I'm not following the instructions.

 

[Doug McCrae]

Fascinating look at D&D stuff from before my time, thanks!

It's a bit before my time too, I started in 1982. By then, mega-dungeons were dead. We learned to play from Moldvay red box D&D, AD&D, modules, White Dwarf magazine, and a wide variety of other rpgs, such as Call of Cthulhu, Traveller and RuneQuest. Play was very module-influenced, the GM would have one adventure prepared and that was that. I've actually never experienced mega-dungeon play ever.

 

[Beginning of the End]

That's a classic! The "quantum documents" soon raise all sorts of questions. For instance, why am I rolling to find the documents? Isn't that a task? Don't I have a 'real concern' beyond that?

If the documents are in the safe, then one can get the documents by opening the safe. The basic idea is that one might like to make a role-playing game of finding out where the documents are, you see.

Dice are tools.

The basic misunderstanding here arises from failing to understand that task resolution and conflict resolution are different tools and they're used for different reasons.

You're basically mocking a screwdriver for not being a hammer.

Task resolution, as you note, is useful for players who are only interested in roleplaying (i.e., making decisions as their character). Task resolution allows you to explore the game world as if you were your character.

Conflict resolution, on the other hand, is useful for giving players a degree of narrative control.

In a task resolution system, the player is saying: "I would like to open the safe. Let's use the mechanics to see if I can do that." It's an exploration of the game world.

In a conflict resolution system, the player is saying: "I want the incriminating documents to be inside the safe. Let's use the mechanics to see if that is true." It's a definition of the game world.

This distinction is often muddled up by people who want to describe D&D combat as a conflict resolution system. It looks similar, but identifying it as such requires you to equate mechanical outcome ("reduce enemy hit points to 0") with the player's desired outcome by default. And if you do that, then the distinction between task resolution and conflict resolution becomes non-existent.

Take the "opening the safe" vs. "finding incriminating documents" demonstration, for example. The distinction between task resolution and conflict resolution disappears if you equate the mechanical outcome ("does the PC open the safe?") to the player's desired outcome ("I want to open the safe").

Conversely, let's look at D&D combat but change the player's desired outcome to "I want to make the bad guys run away" or "I want to impress the princess by beating my rival in single combat". This clearly demonstrates that treating D&D combat as a conflict resolution system is an illusion created by assuming that the player's goals will equate to the system's outcomes.

Now, if we changed the rules for D&D combat so that a player who reduces an enemy's hit points to 0 gets to dictate how the combat is resolved ("they die", "they fall unconscious", "they surrender", "they run away screaming in terror"), we would then have a conflict resolution system.

One of the reasons that the RAW version of 4th Edition skill challenges fail is because they're mechanically trying to wed a task resolution mechanic onto a conflict resolution mechanic, but can't quite figure out how to make that work. Which is why most people who having success in using skill challenge mechanics have consciously or unconsciously modified them to either become a fully dedicated conflict resolution system (in which the PCs are defining the path they want to take); turning them into a railroaded task resolution system (in which the GM predefines acceptable task-based paths to success); or simply discarding the conflict resolution portion of the mechanic when it conflicts with the task-resolution portion of the mechanic (or, less commonly, vice versa).

For instance, why am I rolling to find the documents? Isn't that a task? Don't I have a 'real concern' beyond that?

The problem with the "infinite regression" argument is that it can be applied with equal validity to task resolution.

For example, I want to search a door for traps. Or do I want to search the entire room for traps? Or do I want to search the entire dungeon for traps? Conversely, if I want to search the door for traps don't I have to search the handle, the keyhole, the lintel, the frame, and the hinges?

Do I make one check for the entire dungeon complex? Or do I make 5 separate checks for every door? There's a rather huge range of abstraction between those points. (Which aren't even the end points of those progressions.)

The decision of exactly how much task is going to be resolved by a single die roll is ultimately an arbitrary one guided by our sense of what will be fun. IME, it will frequently vary depending on circumstance.

The same applies to conflict resolution. Is it theoretically possible in a sufficiently flexible conflict-based system to make a single roll against "I want to catch the bad guy"? Sure.

Should the system be used that way? Maybe. But probably not.

Just as you probably don't require your players to make 5 different Search checks every time they check a door for traps.

 

[Desdichado]

This is why usurpation of player agency is important in my definition; players have the right to give away any agency they so desire. If I agree to play Savage Tide, for instance, I have no cause to go crying that the DM is forcing me to play an AP. If the DM creates unnatural limitations on how I can approach the material because the scenarios as written do not envision my going a certain way, or using a certain tactic, that would still be railroading, because agreeing to play ST does not mean that I agree my actions can be curtailed in this way.

This is exactly why the more I think about it, the more convinced I am that the only meaningful railroading can happen at the table. A module that's designed and written to be played as a railroad probably will be as often as not, without major rework from the GM. But if the players voluntarily surrender some of their control over their characters, if the GM doesn't disallow actions that they think are reasonable for arbitrary reasons, if they buy into it and do it together; the most structured adventure in the world might not really be a railroad.

The only really meaningful definition of a railroad is the describe the experience at the actual gaming table, with the correllary that by transference, you can also call an adventure that is structured so that a railroad experience is likely a railroad too.

Hmmm.

I think you are reading too much into this.

I think there's also quite a bit of transference of percieved accusations of blame here. By which I mean, although you (or Ariosto, or whomever) may not actually be saying something, it is still part of the conversation because of past experiences with people who have "bundled" certain opinions together frequently. I.e., some diehard sandboxers have a "bundle" of opinions that includes exalting sandbox as the ordained method of play, and all other methods of play are wrong. Even if they can accept them as, "hey, whatever floats your boat" there's a need to create denigrating labels for them, since they are no longer "not a game" or "not D&D" or whatever.

Again; I'm not specifically calling anyone out in this thread as saying that, but I think it's come up because it's something that's expressed frequently.

And to nip in the bud the call for "Cite?", that's absurd. I know this because I've had it expressed to me clearly. It's a personal experience. I don't cite personal conversations or experiences, and the notion (which was expressed frequently in the other thread a couple of months or so ago on this same topic) that anyone won't believe my personal experience without a citation so they can look at it for themselves is both insulting and cretinous.

 

[Beginning of the End]

Instead of a threshold, railroad is seen as one direction on a continuum, the other direction being referred to as sandbox. Railroad means less player freedom, sandbox means more. There is, probably, no such thing as a perfect railroad or a perfect sandbox in actual play. Neither railroading nor sandboxing is inherently good or bad, they are just tools to be used by individual gaming groups, and adjusted to taste.

As I mentioned a couple pages ago, I don't think this works well because the opposite of "railroad" isn't "sandbox" and the opposite of "sandbox" isn't "railroad".

Severity of railroad generally goes from macro- to micro-. If the GM insists that the player's take an adventure hook because it's the adventure hook he has prepared but the PCs are free to do whatever they want to accomplish the adventure's goal, most people would describe that as very light railroading. The severity of the railroad increases as the GM

Sandboxes, OTOH, are defined from macro- to micro-. If the PCs are free to go anywhere in the world and pick up any adventure thread they want, most people would define that as a sandbox even if some (or all) of the individual adventure hooks aren't perfectly non-linear.

Or, to put it another way: The term "railroad" generally applies to adventure design. The term "sandbox" generally applies to campaign design. Putting them on the same continuum just distorts the meaning of one or both of the terms.

 

[Desdichado]

Task resolution, as you note, is useful for players who are only interested in roleplaying (i.e., making decisions as their character). Task resolution allows you to explore the game world as if you were your character.

Conflict resolution, on the other hand, is useful for giving players a degree of narrative control.

In a task resolution system, the player is saying: "I would like to open the safe. Let's use the mechanics to see if I can do that." It's an exploration of the game world.

In a conflict resolution system, the player is saying: "I want the incriminating documents to be inside the safe. Let's use the mechanics to see if that is true." It's a definition of the game world.

I've never once heard those two terms used that way. That's an interesting concept, although I think coining that vocabulary amongst gamers as a whole would be tricky. I rather like the more straightforward description of conflict resolution systems as systems that exist to resolve conflict between characters.

Therefore combat is absolutely a conflict resolution systm, because two characters (presumably a PC vs. an NPC, but clearly not necessarily) are using the combat mechanics to resolve a conflict. For that matter, a skill system can also be conflict resolution, if, for example, a character is attempting use a Bluff skill vs. a Sense Motive or whatever. That's character conflict, and the rules exist to resolve them. Hence, conflict resolution.

I mean, I like your concept of narrative control mechanics as an interesting discussion item (personally I don't really like mechanics for players to determine whether or not their are papers in the safe, but that's neither here nor there), but I think your labels for them are all wrong.

 

[Doug McCrae]

Or, to put it another way: The term "railroad" generally applies to adventure design. The term "sandbox" generally applies to campaign design. Putting them on the same continuum just distorts the meaning of one or both of the terms.

Good point. I agree that that is how both terms, railroad and sandbox, are most frequently used.

 

[Desdichado]

Or, to put it another way: The term "railroad" generally applies to adventure design. The term "sandbox" generally applies to campaign design. Putting them on the same continuum just distorts the meaning of one or both of the terms.

I think that's a useful theoretical distinction. I mean, as a point of devil's advocacy, there isn't always such a sharp line between adventures and the campaign in many games, particularly those of a sandboxy variety, where there isn't necessarily going to be much in the way of adventure design in the traditional sense, but still. I think that's a good point.

However, I think that as playstyles they still stand out as antonyms. A sandbox playstyle is one that focuses on freedom of character action, and casts the GM in a reactive role. A railroad is one that focuses on setting action, and casts the players in a more reactive role.

A little context around how the words sandbox or railroad are being used in a given conversation is obviously pretty crucial to determining exactly in what way it's meant. The quick and dirty definitions given in the paragraph above are obviously as broad as possible, and can apply to design goals, experience at the table, or simply playstyle more generally.

And, as has been mentioned many times in this thread and others, I'd imagine that by far most gamers don't really like either endpoint on that particular spectrum, and prefer some hybrid that uses a few elements of both.