17 RPG Lessons from the Raiders of the Lost Ark Bar Fight (Part 1)

GM attempts in TTRPGs to create a “set piece” imitating film crumple when confronted with the uncertainty of player action and the randomness of the dice. Games are a different medium than film, and any adaptation of film technique needs to draw lessons and learn from underlying ideas, rather than merely imitate the action on screen.

A classic example that I return to again and again is the bar fight early in Raiders of the Lost Ark. In short, adventurer Indiana Jones visits a bar in Nepal operated by Marion Ravenwood, who owns a medallion he needs. The scene begins after Jones has visited and left, and Nazi officer Arnold Toht (who also pursues the MacGuffin) has arrived with hired thugs to take it.

How many lessons can we learn from this five-minute scene? Enough that this will be a two-part post!

The entire scene is here. The bolded text below each heading lead directly to the corresponding timestamp in the video. 

Shout character motivations from the rooftops

Shortly after her bar closes, Marion contemplates the medallion that Jones wants, considering her options. 

RPG NPCs don’t need much to drive them. Around the internet, I see complex generators and intricate prep tools designed to create page-long NPCs, and I find them to be overly busy. An NPC doesn’t need much more than a clear driving desire or goal that the players can learn and understand. Three bullet points is usually enough for almost any NPC. Everything else emerges naturally through play.

As an audience of the film, we get to see Marion, alone, contemplating the medallion. Without any dialogue, the audience understands that this character is going to be important to the rest of the movie. In a tabletop game, we’re going to have to be a lot more obvious in broadcasting Marion’s (temporary) refusal of the call to adventure. Figuring out what characters want usually isn’t the interesting part in an RPG. Shout the motivation at the PCs, or they may miss it.

Put the MacGuffin itself in peril

Marion hangs the medallion on a candleholder, where it will soon be at the center of the bar fight. 

Think about a typical RPG fight for an object that both sides want. A PC grabs it. An antagonist wins a grapple check and steals it away. It goes back and forth like this until the strongest, fastest, or most numerous prevail. Not too exciting.

What if instead, on each action, it was highly likely that the object would be dropped, knocked aside, thrown, rolled, ignited, bobbled, immersed, or obscured? The success-with-a-cost thinking of PbtA games’ 7-9 results is instructive here, versus the (superficially) pass-fail nature of D&D.

The Indiana Jones movies do this again and again. The grail is falling into the chasm. The stones are burning through the bag and dropping into the river. The treasure you are seeking is there, but your friend who is in danger is over here. Escape is this way, but the treasure is that way. We can do this in games too.

The antagonists should be visually distinct (...sometimes)

Toht and his three goons enter the bar. 

Note how the four men who enter the bar are all immediately different from each other in clothing, height, and facial hair. A little bit of description can be an anchor in the ensuing combat (“you want to hit the one with the scar?” “no, the one with the fancy boots”) Simple portraits for a rival adventuring party can go a long way from turning faceless bags of hit points into compelling antagonists.

That said, this is a device that can be used sparingly, because in tabletop versus film, so much more of the information is auditory rather than visual. In many cases, five orcs can simply be five orcs, and any additional detail is unnecessary unless the players ask for it. 

Positioning can help set a scene

As Toht and Marion converse, Toht’s henchmen spread out through the room in a threatening manner. 

As Toht moves from a purchase offer to a threat of violence, note how much of the drama of the scene is conveyed by the positioning of his minions, particularly behind Marion. This doesn’t have to be – and probably shouldn’t be – miniatures on a grid, at a point when the scene is still more conversation than conflict. 

Again, there are limits in how much film language is going to translate to an RPG. But think about how easy it would be to run this scenario in an RPG and essentially have the four men standing clumped together in the entrance, and how much better it feels when they move around while talking.

Narrate what initiative means 

Some of the antagonists move quickly to escalate the threat against Marion, while others move more slowly.

A low initiative roll for an antagonist can feel like a screw-up, in a way that undercuts the palpable menace they present. But the GM has the power to frame this action in different ways, presenting them as patient, or condescendingly deliberate, or waiting to base their action on what their ally does first. All could apply to the bar scene.

Narrating initiative effectively makes it feel as if this largely random ordering was intended all along.

Make improvised weapons make sense whenever possible

Toht uses a red hot poker to intimidate Marion. 

Improvised weapons are cool. They help “tell the story” of the location, encourage improvisation, and make for memorable fights. The battle where a PC used a hot poker; or a bottle of liquor; or a squealing farm animal as a weapon is going to be a lot more memorable than the 139th time Lothgar hit someone with his longsword.

If the rules of the game don’t already endorse improvised weapons, think about ways to make them more appealing. And have NPCs show off their effectiveness. Players who see NPCs do something cool will follow suit quickly enough.

Put conversations on the clock

Toht declines Marion’s attempt to make a Persuasion check.

PCs love to endlessly banter, query, and wheedle, even with antagonists on the precipice of violence. Since talking is (or at least feels) “free,” PCs will use it as much as possible. This is fine when it’s a genuine attempt to engage with situations beyond combat. But it can turn into an indefinite stall on interesting conflict.

It’s helpful to remind players that they can’t freeze the action, and the easiest way to do this is to say “the NPC answers you, but they’re continuing with their aggressive action at the same time they talk.” When Toht says “that time has passed,” it’s driving the situation towards interesting decision points for the PCs. 

Accidents will happen

Jones disarms Toht, accidentally lighting a curtain on fire in the process.

This feels like an interesting way to frame “partial success.” Jones’ player gets what they want, but there is spillover. Echoing “put the MacGuffin itself in peril” from above, this is something that happens a lot in the Indiana Jones movies.

In a game that doesn’t specifically instruct the GM to resolve partial success in this manner, we should constantly be looking for opportunities to interpret action in this way, because of how well it drives the action in interesting directions.

Next Week: 17 RPG Lessons from the Raiders of the Lost Ark Bar Fight (Part 2)

Alignment Languages Were Good, Actually

Previously: Putting the “Align” Back in Alignment 

AD&D 1E DMG, page 24:

“Each alignment language is constructed to allow recognition of like-aligned creatures and to discuss the precepts of the alignment in detail. Otherwise, the tongue will permit only the most rudimentary communication with a vocabulary limited to a few score words. The speaker could inquire of the listener's state of health, ask about hunger, thirst, or degree of tiredness. A few other basic conditions and opinions could be expressed, but no more. The specialty tongues of Druidic and the Thieves' Cant are designed to handle conversations pertaining to things druidical on the one hand and thievery, robbery and the disposal of stolen goods on the other. Druids could discuss at length and in detail the state of the crops, weather, animal husbandry and foresting; but warfare, politics, adventuring, and like matter would be impossible to detail with the language.”

Yeah, it raises a lot of questions. How early in life does one gain an alignment language? Is this learned by imitation or taught formally? Does it vary regionally like most languages do, or is it moral Esperanto?

Druidic and thieves’ cant are easier to place in the world. We can intuitively put together how characters in these classes would have learned these languages. They’re both evocative ideas that don’t require such mental gymnastics.

So why don't we just do that for all the alignment languages? Once the focus on profession and practical faction is clear, it’s a lot easier for players to grok. But the above paragraph only identifies two languages. How would we fill out the rest of the nine-grid chart?

• Lawful Good. The pure language from before The Fall. 
• Lawful Neutral. Mathematical speech, the aspiration of Babel.
• Lawful Evil. The language of command, ala They Live or the Manchurian Candidate. The language of the tyrant.
• Neutral Good. A polyglot creole of languages old and new, spread through trade, treaty, and camaraderie around a shared fire. 
• Neutral. Druidic. The secret tongue of plants and beasts.
• Neutral Evil. A base tongue of unrestrained id and unfettered ambition. A grim handshake between two sociopaths, a gesture worth a thousand words.
• Chaotic Good. The language of revolution and rebellion. Graffiti and samizdat. 
• Chaotic Neutral. Thieves cant, the language of rogues, spies, smugglers, and anarchists. 
• Chaotic Evil. The whispers of the Beast. The voice speaking directly from the dark place of the soul.

Sure, some of these are closer to the grounded fiction of the game than others. Some would require a bit more work to place in the world. But it’s not so hard to imagine tying them to factions and tangible hooks in the game world, in the same way druidic and thieves cant already work.

Putting the “Align” Back in Alignment

In the past year, I’ve enjoyed listening to several efforts to read the AD&D 1E DMG aloud and untangle its famously opaque contents. Listening raises a number of interesting questions and writing ideas. One that struck me was how nine-grid alignment was born, and then how it has deviated from its intended use.

Alignment, as Gary Gygax presented it in the 1E DMG, was both a guide to behavior (a roleplaying tool) and a chess board for faction play. The former application led to countless arguments and fruitless battles, and D&D-derived games have either mostly dropped alignment (as in D&D 5E, where it is mechanically almost vestigial) or have altered it significantly (for example, Dungeon World).

The latter use – for driving faction conflict – is as useful as ever, and probably underappreciated.

Law and chaos have as much oomph as they ever did, whether used in the style of the Elric books that inspired them, or simply to distinguish  group (lawful) and individual (chaotic) priorities. The latter tension still drives loads of modern media; it’s the crux of most of the conflict in the Marvel Cinematic Universe. So law and chaos can stay.

But how about replacing good and evil? The concepts of good and evil may exist in the world, but they’re abstract and philosophical, as they are in the real world. They’re not part of our new grid that we invite players to opt into when they create characters. What should we use instead to create a nine-grid alignment chart? 

We could certainly simply rename the good-evil axis. Celestial and terrestrial? Earthly? Telluric? Rational? Pragmatic? Different adjectives put different spins on the spectrum we will cross with law-chaos.

What if we move further afield? I’ve long been a fan of the Jungian treatment at From the Sorcerers’ Skull. You could build an entire RPG around the adventuring implications of this chart.

What about using social status as an axis? For a dungeon project I’ve been working on, I created the following nine-grid “alignment” chart for character background plus starting "special" equipment.

	Noble (1-2)	Servant (3-4)	Imposter (5-6)
Lawful (1-2)	Vicar (Holy Water)	Hunter (Bear Trap)	Spy (Forgery Kit)
Neutral (3-4)	Scholar (Tome of Lore)	Bricklayer (Trowel and Fast-Dry Mortar)	Thief (Lockpicks)
Chaotic (5-6)	Black Sheep (Gunpowder Bomb)	Fired Yesterday (Hidden Knives)	Goblin (Human Disguise)

Think about all the implicit world building work done by presenting a law-chaos/noble-imposter grid. 

For the same project, I used Patrician, Proletarian, Pastoral as categories. Again, these adjectives can be crossed with the law-chaos spectrum to create very interesting combinations. Just the intersection of those two ideas is enough to spark interesting characters and situations.

Next: Alignment Languages Were Good, Actually

Dungeon 23: The Grind

I have enough writing experience in different contexts – from professional settings to blogging to tabletop games – that I’m confident saying that editing is the most criminally underrated part of writing.

I understand why. Writing itself is enjoyable. Turning ideas into words. That’s the fun part. Editing is work.

But it’s valuable work. Removing a single bad idea is as useful as adding 10 good ones. Excising an unnecessary sentence is as helpful as writing a brand new paragraph. Scrapping an entire section that wasn’t pulling its weight could save an entire project. 

So, Dungeon 23.

Cut from the Bottom

Spending five minutes a day on Dungeon 23 produces some fast, zany, madcap results. They look pretty good fresh out of the oven. But in the cold light of February, some of those January dungeon rooms looked pretty bad.

So, we edit. At the end of every month, find the least-interesting ideas and cut them. For example…

Move the robots off the surface. Several of our rooms suggested that the reptile ancien regime fled from a surface society inhabited by robots. But these robots were incongruous; they didn’t really make sense. And they strictly limited how the dungeon could be incorporated in a game. Let’s move the robots deeper into the dungeon, rather than implying a surface setting we’re not interested in building. Our surface world can now be any normal, nominally human-settled civilization in a fantasy world.

Scrub out fixed scenarios. One of our rooms described a painted person examining the decapitated head of a robot from a nearby mechanical corpse. It was less a room, and more of a frozen tableau, a flash-frozen scene ready to play out when the players walk in on it. This kind of event fits better on a random encounter table than as a dungeon room.

Room to Breathe

I crafted my dungeon day-to-day, drawing heavily on the generator. Drawing on a 5.5’ by 3.5’ notebook, the rooms inevitably were tightly packed together. 

The reptilian ancien regime, the petrogeists, the surface-dwelling robots humans, the painted people and their summoned angel of light, and the sundry unaligned monsters were squeezed into relatively tight proximity. This is a known issue with otherwise excellent dungeons that try to compress an intriguing scenario into a too-small space of tightly keyed rooms.

Fortunately – because this particular Dungeon 23 project is merely a very rough draft – I can go back through these rooms and stretch out the proceedings, adding longer hallways, additional rooms, and obstructions and blockages to suggest a more believable shared dungeonscape between the factions.

The First Culling of Many 

To return to the subject of writing versus editing… how many editing passes does it take to turn draft writing into a finished product? 

“It depends,” but something like…

• A single editing pass for a text message or a DM to a friend.
• Two or three passes for a Discord post or a tweet.
• Four to six thorough reviews for a blog post (I will concede I sometimes fall short, and regret it when I re-read something later and see something I need to fix).
• Ten? More? Could it ever be too many? For something you will put your name on and publish for strangers to read.

In a way, a published module (adventure, whatever you like to call it) is the hardest thing to sufficiently edit. A TTRPG product is attempting to do double duty as not just an interesting fictional scenario, but also a practical tool for use at the table. Imagine combining the writing tasks of “short story” and “furniture assembly instruction manual” into one job.

I suspect a lot of creative people hate editing. Writing itself is the fun part, right? Encountering the ideas themselves in the wild and shaping them into tangible things others can enjoy.

Maybe you hate editing. That’s fine if you are a successful writer with a dedicated editor who can take on that work for you. But most people don’t have that luxury. Not bloggers, not independent publishers, certainly not homebrewing DMs. You are your own best editor. So embrace the value of editing. Even if it feels like work. Especially if it feels like work. Because it will pay off.