In 1986, Doug Sharp released The King of Chicago, a profoundly ugly game for the Macintosh; not just in its characters looking like bizarre puppet things, but also in its overall tone. "There's no redemption to be found amongst King of Chicago's many possible story arcs, just crime and bloody murder and revenge," notes Jimmy Maher in his overview of the game. The possibilities in the story are numerous, but the actual space of the story is narrow and restricted, by design. In 1987 Doug and a team at Cinemaware cleaned the game up, added far more attractive art, and released it on the Amiga, bestowing onto the world a strikingly original approach to narrative storytelling that remains to this day damn near unique to the game on which it debuted.
A more thorough overview can be found in Jimmy's article linked above, but in short: The King of Chicago places you in the role of Pinky Callahan, a gangster with ambitions of retiring his boss one way or the other and, after taking his place, making all of Chicago the North Side's turf. The mechanics of the game are extremely simple, involving mostly watching a scene featuring dialog between two characters, one of them usually Pinky, and at one or more junctures selecting a decision via 2-3 floating thought bubbles above Pinky's head. A complete game of The King of Chicago consists of a few minutes to a little over one hour of a sequence of these scenes. The decisions the player makes throughout rarely have a strict, hard-branching effect, though; this is not a Choose-Your-Own-Adventure. The decisions will more often nudge a hidden number here or there, and the number would represent something like the Gang's attitude toward Pinky, or the attitude of Pinky's girlfriend, Lola. Every few scenes you also have the opportunity to adjust the ledger dictating how much you're trying to get out of your various enterprises, and how much you're paying everyone; also in this particular scene you're able to plan your next move in carving out another piece of Chicago for the North Side. Take all four pieces and the game is won.
All of these factors are taken into account behind the scenes by the engine in determining what scene to serve up next. The scenes themselves are written so that they can be plugged into an ongoing story without being incoherent, and a scene or sequence of scenes is tagged with criteria that describe numerically its general premise. "The gang likes Pinky but we don't have much territory", or "Lola likes Pinky but the gang is losing faith in him", or "Ben (Pinky's right-hand man) isn't getting paid enough", and so on. These are all premises that can be numerically boiled down to "High gang morale and territory less than 2", or "high Lola attitude but low gang morale", or "low Ben salary". There are occasions where the next scene needs to be restricted to be one of a particular type, but largely the selection is a weighted random draw from a pool of valid (sensical) scenes. With this setup we cross the threshold from CYOA to game mechanic-driven dynamic narrative.
With a system like this, one need only ensure that there is at least one start-to-finish path through the game, and then from there add more and more appropriately-tagged scenes; the game's engine will figure it out from there. In this way, The King of Chicago approaches the naive ideal of "Interactive Movie" more than any other game before it, or after it. Any scenes given before Pinky gets rid of the Old Man becomes Act 1, the offing or forced retirement of him becoming the inciting incident for Act 2. In this Act 1 there's a very singular focus, and one can even conclude it in the first scene if the game gives the player the opportunity to kill the Old Man in his car, but one can also take a winding path through consulting the rest of the gang about how they feel about the boss, and feeling out perpetual right-hand man Ben's opinions on him. Things like the gang's and Ben's opinion on the Old Man, as well as most other numeric variables, are randomized at game start, so any number of outcomes can occur. The player is eventually funneled into a confrontation, but there's a remarkable amount of choice just in this opening bit of the game.
Act 2 can be a straight shot to the top, taking over each piece of Chicago one-by-one, or it can be a subplot-packed exploration of life as a gangster. Here one is given more or less free reign to decide what to do next in a broad sense, but there is a hard time limit that when reached the gang gets fed up and makes you leave so that the engine doesn't start scraping the bottom of the barrel and serving up wildly inappropriate scenes. Act 3, if there is one, consists in one of several ending sequences of scenes, each with their own possible twists, again wholly dependent on the decisions you've made in a micro sense. The small things you do nudge hidden numbers this way and that, in a way that you can't have any real concrete knowledge of, and so the flow of the narrative and its outcome usually feels very natural and intentionally constructed.
The natural-ness of the flow of the story can be attributed partially to the scenes themselves being carefully written to be slotted in like this, but foremost working to the game's advantage is the amount of restrictions involved. In Act 1, there's no way around it: you're going to have a confrontation with the Old Man. There's variation within the establishing of Pinky's character as someone with unwavering ambition to dethrone him and the ultimate confrontation, but those two events will occur. In Act 2, subplots will come and go, get introduced and conclude, but ultimately you'll reach one of a handful of good or bad endings within a couple dozen cycles of play. There aren't that many endings, either; the actual genuine branching is relatively small. There is restriction, too, in just how the player can provide input. There are some quite bad arcade sequences sprinkled in throughout, but the player mostly selects one of 2-3 options at particular junctures and balances a ledger every cycle and that's the entirety of what the player can do. The majority of the scenes involve Pinky speaking for himself, and the player never really feels themselves embody the protagonist like they might for one more directly controlled.
Nonetheless, for all its restrictions, it's a story that feels alive in a way that a linear narrative game can't, alive in a way that a CYOA book or your usual work of choice-based interactive fiction can't. And it is precisely because of its restrictions that it can provide such a living narrative. No, you can't just decide to follow the Old Man and serve under his leadership. No, you can't just give up the life and go be a grocery store clerk. You can't do literally anything. You get fairly significant input in how the story plays out, but there is a story to tell, here. You have a setting, and you have to work within it. You have a role to play, and you have to play it. Games are, at bottom, about setting restrictions in which the player can act; in other words, they're about rules. And The King of Chicago, like any game, has rules.
But when we have rules in place, a set plan for how the story will go and what the player can do, we still have lots and lots of freedom in adding the events that occur between points A and B. More importantly, we have lots of workable freedom in this. The combinatorial explosion that occurs when one tries to account for literally every possibility at every story juncture becomes an unreasonable logistical quagmire very quickly; either one makes a game with unimaginable breadth but lasts half an hour or, more likely, gives up trying. Absolute freedom in a story world is a great thing in theory, but more often than not it's 100% equally as useful to provide instead the illusion of absolute freedom, for 1% of the work. Funneling a player into a predetermined outcome without them noticing is an art. It's part of the illusion that a convincing game world creates.
Systems that present scenarios to the player via selection based on a game state affected by minute player decisions lived on in various Japanese dating sims and virtual novels, most notably Konami's Tokimeki Memorial series. Tokimeki Memorial 2 in particular goes kind of crazy with story granularity, with its data spanning 5 CDs and its average playthrough length of around 8 hours. It's a fantastic game, well worth playing if you know enough Japanese, and the first game's Super Famicom port has an English fan translation as well. This particular mode of storytelling seems to have become predominantly lodged in this niche however, and even within it the feel is less "interactive movie" and more "JRPG"; you go running to increase your Athleticism stat, you meet the girl who likes running, you pursue her romantically, you're treated to several scenes featuring the running girl, and so on. There's a lot of random battle-style repetition, a lot of the same canned conversation couched around a few lines of unique dialog and a choice for the player, running throughout. The numbers are usually far more overt than anything in The King of Chicago, and are usually very clear in their effect. The flow of the game resembles more a series of episodes than a coherent narrative. What I really want to say is that dating sims are, well, simulations, and Chicago is very much not a simulation of being a gangster in the 1930s; it's very clearly a focused story. It's not that I'm even saying Chicago is a better game than Tokimemo 2, the latter is much tighter mechanically and has a far more engrossing world overall; it's just, why has no one else even tried to do this concept as organically as The King of Chicago does?
Like, to be clear, I'm having to dig into a very specific genre that's fairly niche to the western world, that generally encompasses only one broad subject, to find anything that even resembles it in terms of storytelling mechanics. Most of what arrived to us as "interactive movies" in the days where that phrase wasn't a cause for much scornful laughter were, at best, graphical adventure games with lots of video clips. Point and click adventures died an ignominious death at the hands of developers who couldn't understand how to properly place the paradigm of "just walking around, looking at stuff, and talking to people" into a context that doesn't involve puzzle solving. Any of the genre's advantages evaporated as action adventure games and RPGs assumed and exceeded their audiovisual luster. But is that really all there is to it? The fate of the AAA mainstream adventure game was to be subsumed into other genres and there wasn't any alternative? I play a game like The King of Chicago, a game that, again, was made in 1986, and begin to doubt such a proposition.
Is it just that the game offers little side paths and story tributaries in the course of its main story? That's not that special, obviously. See basically any large sidequest-packed game made in the past two or three decades for countless examples of that. No, it's that the forward thrust of the plot changes based on what you do. The world is reactive in a way that reaches beyond game mechanics merely interacting with each other, beyond "player doesn't have x so they can't do y." World reactivity is something we've never really gotten a hold of in a consistently satisfying manner, and Chicago has it in spades; it's the whole premise. This isn't to say that adventure games should have become this kind of low-interactivity decision making game that serves up canned scenes in succession for maybe the length of a feature film. What I am saying is that there were other places to go in adventure gaming that didn't involve picking up objects and using them with other objects, and didn't necessarily require being merged with action or rote RPG combat mechanics. There was a path we, as human beings, as artists, could have tread. The path was in fact there from the very beginning. We just branched almost immediately.
When Will Crowther wrote Adventure in 1975, it was his interest in the then-novel tabletop role playing game Dungeons & Dragons that informed its interface. You're presented a situation describing yourself, the player, in a given setting, inhabiting it, in the second person. What do you do? Well, a lot of what you did involved picking up objects and using them or placing them elsewhere, and most of the content of the game lies in navigating the world's geography. But it was 1975 and such basic mechanics were all that could be achieved on equally basic computer hardware; it was the fact that the game initiated a player-to-GM-like dialog at all that made it revolutionary. What a strong illusion this must have produced for its players in its time! The key thing to understand here is that the game is a treasure hunt as a result of its technical limitations and limited cave-obsessed setting, but the primary inspiration for the form that it takes is that of the rules-based collaborative storytelling game that is D&D.
The understandable infatuation with the game in itself though, led to an equally understandable desire to make it again, but bigger. Hence, Zork, another treasure hunt through an even more massive winding geography. By the time Infocom came into being and created its own Zork series for disk-based home computers, and by the time Scott Adams began creating diminutive and hilariously terse text adventures for lesser-equipped tape-based home computers, a genre had ossified, wholly different in character from that primordial urge to "have D&D on the computer." All of a sudden, you had Infocom in the mid-80s trying to figure out how to make a fulfilling adventure game that wasn't a puzzle-solving treasure hunt, as if it were some fundamentally impossible task. We completely lost the plot; we forgot what Adventure was even trying to do in the first place.
Parallel to these developments we have the various dungeons and Multi-User Dungeons, leading into the first genuine CRPGs. These too focused on the exploratory aspect of D&D, but instead of attempting to imitate the player-to-GM dialog they instead implemented as many D&D-like stats as possible, for use in combat encounters. Again, it would have been difficult to implement anything richer than that for computers at the time; even having NPCs that spoke a single line of canned dialog constituted luxurious world-building, so anything like a conversation system was out of the question. A lot of mechanical-narrative scaffolding would have had to been constructed for things like Persuasion checks to be viable outside of the player-GM dialog of D&D, and certainly no one at the time, impressed just to see a game occurring on their computer screen at all, was up to the task. "You hit the Orc; Orc's hitpoints reduced by 2" asks far less of the player's imagination than something as abstract as "You persuade the shopkeeper; Shopkeeper's willingness to haggle increased by 2." (How did I persuade him, what did I say, etc.) Once more these games became popular not for their imitation of a collaborative storytelling experience, but for the elements that were imitated in themselves, and so these too would grow up and ossify into a form not wholly unlike the mainstream videogame RPGs we have today, largely based around exploration and combat couched in a mostly non-reactive story, with perhaps some bones thrown toward skill checks in conversation systems here and there.
From a mutual point of origin we built two prisons, only then to occasionally wonder how we could escape them, as if they were foregone conclusions, never to wonder what it was that got us there in the first place. Ultima is one of few series consistently having something close to the correct vision, at least beginning with Quest of the Avatar, but these games too were more like the linear stories of any given CRPG, just couched in a broader and more sophisticated simulation than random combat. No one wants to bake bread or live a life as a baker of bread, in Ultima VII. It's neat that it can happen, but a compelling narrative it ain't. Worse off is something like The Elder Scrolls IV: Oblivion, the most prominent and popular example of Bethesda's obsessive over-simulation of everything, as if uncanny automatons running around in a too-big-for-its-britches, partially procedurally generated world can accidentally clutter up into a cohesive, much less interesting, story for the player. Simulation is a blind alley for game narrative; the cracks will always show, if there's not a person at the other end of it all telling the stories. The computerized GM is not an AI, it's reams of human-authored material woven skillfully into a system.
Special mention should be made of Quest for Glory, a series by seasoned tabletop players Corey and Lori Ann Cole, and an outlier in Sierra's lineup of flashy but negligently cruel adventures in that its games followed the general mold of a graphical adventure, but incorporated RPG stats and dice rolls in conjunction with multiple solutions to problems. These are not only delightful games well worth anyone's attention, but also tantalizingly close to the ideal. "Here's a situation." "Okay, let me try to do [...]" "Alright, I'll allow it. Roll for [...]". This is the basic gameplay of a standard tabletop RPG, and this is the basic gameplay of Quest for Glory. The computer provides the world and the situation and rolls the dice for you, and does all the other bookkeeping; the player just has to provide the input. And therein lies the rub.
How do you account for the possible usage of every combination of items on every thing on every screen beforehand? A GM can do this handily; even if they disallow a mere attempt at a particular course of action, justification can be provided. You typically won't get a blank-faced "You can't do that right now" from a human being running a game. D&D requires an ongoing dialog between the player and the GM, notably a fellow human being, who has emotional intelligence and can read intentions and ask for clarification. This is all horribly messy business to deal with in a game run by a computer, and until we have Star Trek-style androids available in the household we're not going to get anything but a human to run a game with the flexibility of a tabletop RPG while still prompting the player for arbitrary atomically constructed inputs. So, again, how does the GM of a computer game, that is, the author, account for everything the player can try in advance? It's totally infeasible, isn't it?
For me, Disco Elysium is a double-edged sword. "Finally," I thought, after completing the game for the first time some months ago, "someone understands how to make a computer role-playing game." There are "problems, not puzzles," as the Quest for Glory team would stress, and there are multiple solutions to them. While significant actions consume in-game time, everything is effectively done on a turn-based scale and not a modicum of reaction time is ever necessary. All of the skill exhibited in the game is that of your character's, based on dice rolls weighted against your stats. Strategy isn't required, just decision-making. And to drive it home, all decisions the player can make are provided in advance. Yes, it's the choice-based format of Disco Elysium that makes it such a pure shining example of how to do this genre right: in the absence of a genuine GM, how do we account for all the things the player could try to do? We simply restrict the things the player can try to do from the outset, down to a list manageable enough to write around yet healthy enough to make the player feel as though they're never being deprived of an obvious course of action.
All of the time spent outside of menus is spent walking (more likely, running) to an object or person and interacting with it, after which the player is presented a usually very robust menu of choices. These choices can sometimes be lawnmowered through one by one, but not always. Some of the choices have skill checks attached, and some of these skill checks can't be repeated. The others can, but only after leveling up the stat in question. Experience for leveling up these stats is gained through making particular choices. Some of the actions you can take have visual consequences, but everything is always described in text by the narrator. To sum it all up, here: basically everything that isn't navigating through rooms and initiating an interaction is in text. All the action happens in the ribbon that appears when you interact with something. From a purely mechanical standpoint, the rest of the game could be replaced by a yet another menu of things to interact with. To be sure, we would be missing out on the beautifully illustrated world of Revachol, and the sporadic but effective character animations that accompany the action, and so it would certainly not be desirable to replace the entirety of navigation with a menu, but it could be done. This is the other edge of the sword.
It's not to Disco Elysium's detriment that all of its mechanical richness lies in choicemaking and selecting stats to improve, that is in fact what makes it brilliant, and it's not that the game has a relatively linear story expressed as a turn-limited journey through a handful of days, as there are loads of little world-reactive moments ranging from small asides from the characters you interact with to absolutely thunderous payoffs to plot setups you didn't even know you were setting up. In actuality, it's difficult for me to find any fault at all with the game, and this is why I find myself unable to write a proper review. Never have I played something that achieves the old Infocom ideal of making the player feel like they're "waking up inside of a story," until I played Disco Elysium. But in all of what it does, none of it is revolutionary, technologically. We could have had something like this game, beautiful (albeit lower res) art and voice acting included, in the early ages of the CD-ROM. And thus, the inevitable question is:
Why didn't we?
For all that I love this game, I also hate it for how it's affected my view of all adventure games up to this point; nay, all of long-form narrative games in general. Games like this should have been the norm 30 years ago. In the strive for living stories so few developers ever thought to work holistically within a set of limitations and instead chose, time and time again, to continue beating one of a stableful of dead horses. Parsers are a dead end. Puzzle solving and combat as a means to "break up the story" is a dead end. In interactive fiction of all stripes, the story is the game. If a game isn't designed from the ground up to deliver a story, and instead consists of game and story in discrete parts, then it is failing the medium of video games utterly.
To pull away from these hysterics momentarily, let me be clear that I've loved, and continue to love plenty of these "failures of the medium." Hell, Dragon Quest XI is one of my favorite games, and structurally speaking you can't get more bog-standard classic linear JRPG than that. Great aesthetics and a general friendly attitude toward the player go a long, long way. But occasionally a game like Chicago or Elysium crosses my path and casts a dark shadow over otherwise beloved games, for allowing me a glimpse at the truth that we've just been running in circles for decades now. On the one end, we have The King of Chicago's strikingly original and innovative system for story delivery, basically never to be picked up again; a game that I can only wish I could say was "before its time." On the other, we have Disco Elysium, a game that offers nothing particularly innovative whatever, and so that it is so notably great is a tragedy.
If I may now be (uncharacteristically?) direct, here are some key takeaways from all of this:
I'll be the first to admit that my knowledge of RPGs and adventure games is hardly comprehensive. If there's some really cool game along these lines that I've failed to mention here, please let me know in the discussion thread below. Now it is my intention to sit down and play video games for several hours. See ya!