Once upon a time I wrote about railroading: https://thanuir.wordpress.com/2007/12/02/defining-railroading/ and https://thanuir.wordpress.com/2007/12/05/24/ and even https://thanuir.wordpress.com/2007/12/07/to-not-railroad/ . Jonne Arjoranta responded http://users.jyu.fi/~joolarjo/forge/role-playing/theory/by-train-for-a-change/ , but I learned of it only recently. Jonne did not link to any of my posts and so there was no pingback or other notification.
Jonne states the definition of railroading as someone restricting the diegetic (fictional) contributions of others, and concludes that this is too broad, since it is necessary for choices to have consequences, or for any roleplaying to coherence. Jonne would rather use the language of Markus Montola http://ropecon.fi/brap/ch14.pdf (in Beyond role play http://ropecon.fi/brap/practice.html ), involving integrative and dissipative actions in play.
My definition of railroading had three conditions: That someone restricted choices of others, that the others assumed they could make the removed choices, and that the action had no fictional (diegetic) reason. As such, the criticism of Jonne does not apply to the definition I stated, since my definition is much more specific than the one Jonne criticises, and the criticism is based on the large scope of the definition.
I should mention that my definition is meaningful when one consider making decisions and observing the consequences to be the central to play. If one think of something else as central to play – for example, causing a particular experience to the players as in Fastaval games http://alexandria.dk/english I read in Unelma keltaisesta kuninkaasta http://pelilauta.fi/index.php/topic,2184.0.html (forum thread in Finnish), then this definition might not be as meaningful.
This is also the reason why Will Hindmarch’s recent texts have not had much of an effect on me ( https://medium.com/gameplaywright-presents/sword-fighting-on-a-roller-coaster-railroading-for-the-best-in-rpg-play-547333c80359 , https://medium.com/gameplaywright-presents/the-illusionists-lament-dramaturgy-and-illusion-for-the-best-in-rpg-play-97d348bcb16 ). Based on a quick read (please correct me if I have interpreted them inaccurately) Will consider the game master as an artist or entertainer more than the facilitator of play who lets other participants make decisions and enforces their consequences.
njharman asked how to avoid railroading. For the purpose of this post, I assume reader does not want to railroad and has a good reason for it, such as not enjoying railroading or wanting to try new things. (Bad reason would be because some internet person told that railroading is inherently evil.) Any examples will be drawn from D&D because njharman used DM as a phrase, which kinda implies D&D or similar. I further a prep-heavy game (that is, not Wushu or similar).
Railroading can be avoided, and can happen, on several levels. The easiest and lest painful change is to change the macro level. The simplest method is to always prepare at least two adventures and let players essentially select which they pick. After one has been finished, the other should be altered as appropriate due to PCs ignoring it. The undead gains more minions or the orcish horde sacks more towns or the rakshasa infiltrates a position of power. Even if the adventures are on rails, players still have some choices: To engage this adventure or the other one, with the neglected situation often growing worse. Or maybe some other adventurers solve the other case. Something concrete that the players will notice.
Slightly less simple method is to ask players what their characters will next pursue and to build the next adventure around this. Pacing is important: The direction that PCs take should be clear at the start of each adventure, otherwise boring play and all symptons of that may start manifesting. Big twists and reveals should happen near the end of the session and the next adventure at the very end, preferably roleplayed to not make it boring, if people are in the mood for that. Email or simple face-to-face meetings between games may be an option based on the social environment of participants.
The key in both of the above methods is to clearly communicate with players that their choice matters. This may be out-of-game or in-game, whichever works more smoothly and reliable. Another important factor is to always start each game on the run. Either the plot hooks hit them on the head (more or less literally) by someone asking for help, some PC hearing rumours, the party being attacked, et cetera or the preselected adventure gets rolling right away, e.g. “You are in the sewers and have discovered the body. Who has the light source?”.
One way to get rid of the rails is sandbox gaming. It takes a heavy up-fron investment, though, and is not trivial to get correct. Do this only if you like building or memorising settings and have the time for it.
Not surprisingly, the method is to create a dynamic setting and then let players create characters and do as they will within the confines of the setting. The tricky part is “dynamic”. The setting must have interesting things going on on the scope that players can concretely affect. If they start with power and prestige, politics and wars are good default solutions. If they start as random farmboys, which is probably the better way to introduce elaborate settings for long-term play, slowly giving them power and prestige opens up many possibilities. The “slowly” part is to avoid player freesing at the terror of having the negotiate politics or an unknown setting with a fairly unfamiliar character.
There are some common pitfalls hidden sandboxes. First is players who don’t get interested. It is probably a good idea to start with a bang to avoid this. Burning farms is a long-time tradition among orcs and evil empires. Second is falling in love with the setting. Good gaming is the point. Setting is at best secondary. Players probably are not that interested in random setting bits (but if someone is, let him build parts of it or get involved in other ways). Also: Player characters are the most unreliable portion of the setting. They will blow it to pieces and reassemble them. Or not. Allow this to happen. Maybe they dethroned the emperor. Good for them. Play on. The setting is not sacred. Third is to make a totally impenetrable or alien setting. Resist the temptation. Players will start interacting with the setting only after they understand how it works, generally speaking. As clash bowley did when designing Book of Jalan, steal liberally from real world cultures, but feel free to mix and match cultural tropes, religion, environment, etc. Monotheistic roman dwarves living in jungle or something. It will look and feel exotic but also familiar, which is exactly the point.
The method of gamemastering in a sandbox is to have the aforementioned dynamic forces, player characters who are involved with them, and the rest of the setting for context and ideas. When preparing, think about the motivation and goals of the dynamic parts. What will they do to accomplish them? What do they want player characters to do? What will they do to PCs? How will they be involved? There’s the adventure. Some fairly static but interesting locations and events are good to keep the game changing. A dragon sleeping in the cavern, some random ruins there, an enchanted island here. Just don’t assume that players will go after them. Restricting PC travel is useful, because one can always spring random encounters (that hopefully do have a point or serve a purpose besides depleting hit points) on hapless travelers.
More on the next post. These are the easiest to adopt from railroading background, I’d say. Also: remember to inform the players that you won’t railroad as much anymore. You can’t change a group’s style by yourself.
In my previous post, I defined railroading. Before that, I listed some heuristics for choosing to use or not use the game’s resolution mechanics. Adaen of Bridgewater asked if there is a connection between these two. I think there is.
Most railroading happens because the GM has something in mind and an action initiatiated by players is threatening it. Maybe it is an entire script. Maybe a single scene, like the death of a villain, has been planned. Maybe the player characters should simply be somewhere or meet someone. Maybe the pitiful goblins should really not have been a threat, only a simple diversion.
Generally speaking, players can ruin something in three ways:
- Ridiculing it out-of-character. “Wouldn’t it be a huge surprise if the tavern-keeper died right after we left the place?” If the plan was that someone assassinates the tavern-keeper, they have guessed it. Chances are that players know of the railroading if this happens with any frequency. They may or may not be okay with it. One should talk to the players in this situation.
- Making the “wrong” decision. Having discovered the big bad’s true identity, they choose to ally with him instead of the assumed blood spilling. Or maybe the clues are interpreted in a totally wrong way and party investigates the wrong rumour. This can be a sign of rebellion, not caring about the existence of prepared material, or simple mistake.
- Failing or succeeding where they should not have. This is a matter of rolling the wrong result or not figuring out the riddle or whatever.
Dice, in games where railroading can happen, generally are tied to the third option. So, that first.
The maxim “roll dice only when both failure and success are acceptable” is a relevant one. It deals with many unwanted effects of the third kind. If PCs need to find something (a clue, a secret door, whatever) simply letting them do so without rolling is a solution that makes any rails explicit, and, generally speaking, accepted. It also makes all relevant skills useless or at the very least less useful. Some other benefit should be given for them. For example, the better the skill check, the faster the research. The maxim essentially makes railroading easier and makes it more explicit.
One trigger for rolling that absolutely kills railroading is to roll the dice whenever two participants disagree about what should happen next, assuming the roll’s result is lived by. Including this statement in game text is a very strong anti-railroading message.
Rolling only when nobody particularly cares about the result, which is somewhat common among the people who think that it is a virtue to not use the rules of the game, makes railroading pretty easy if the default is that GM decides what happens otherwise. The opposite, rolling when the situation is dramatic, will encourage fudging when combined with preparation that tends to cause railroading, otherwise it is not special in any way.
Rolling when something might go wrong is pretty standard. Just thinking hard about what will go wrong before rolling allows on to avert many an undramatic death or failure and put something entertaining as the consequences. Failing the jump roll doesn’t mean a 50 metre fall into rapids for the character, but all his foodstuff does go down and he is now clinging to a steep cliff. Situation potentially becomes more dramatic in the short term (gotta climb up; especially cool if there are enemies there) and long term (gotta find food; especially effective if in hurry) and the game can go on.
I’ll be writing a bit more about railroading sometime soon. Specifically on how to avoid it, if one is willing.
Disclaimer: I think railroading is distasteful. My definition portrays it as negative.
Game master’s decisions are railroading when
- Players assume they can have an effect on a particular aspect of the game.
- There is no credible diegetic (fictional, in-game) reason for players to not be able to affect that aspect.
- GM’s decision negates or marginalises player input with regards to that aspect.
I think the first part makes railroading inherently bad and feel railroady to the players.
- A band of merry PCs must decide to go either south or west because they are at crossroads. Wherever they go, they shall encounter the same group of bandits. They have no reason to assume that there would or would not be bandits at either direction. Not railroading.
- As above, but reliable sources tell that there is only one group of bandits and they prey on the south road. The group of PCs take pains to make it seem like they are going south, but actually go west. They spread rumours, check out that they are not bieng spied, etc. Still they face the bandits. This I would call a minor instance of railroading, unless the bandits had a good reason to go west and it is revealed to the players.