A common assumption about adventure games is the inherent superiority of the Lucasarts design method to the Sierra approach. In a Lucasarts game, you can never die (with a few exceptions) and can always backtrack to get whatever item you need. In a Sierra game, bizarre crap kills you all the time, and sometimes you can’t backtrack to get a key item. This means starting over or going back to an old save. The assumption is that backtracking automatically makes the Sierra approach inferior. This argument is a bit unfair because it presumes that all adventure games are meant to do is tell a coherent story. Ron Gilbert (Lead Designer for Monkey Island 1 & 2) explains in a blog post why death and getting stuck are terrible for story oriented games, “As the story builds, we are pulled into the game and leave the real world behind. As designers, our job is to keep people in this state for as long as possible. Every time the player has to restore a saved game, or pound his head on the desk in frustration, the suspension of disbelief is gone. At this time he is most likely to shut off the computer and go watch TV, at which point we all have lost” (“Why Adventure Games Suck”, Grumpy Gamer, 22 May 2004).
The issue is that Sierra games aren’t always concerned with telling a complex story, sometimes their games are just a collection of puzzles scattered around a fairy tale landscape. Most of the plots in the early titles are explained from the very beginning. In King’s Quest IV, the fairy queen explains that the magic fruit that you need is located to the East. In Space Quest 2, you’re there to stop Vohaul from unleashing his Insurance Agent Robots. In Leisure Suit Larry, you’re told to find love or the closest thing to it. You find a lot of the closest thing to it. You solve a lot of puzzles along the way, but that’s pretty much how each game ends. There is no character development or even significant supporting characters. These games are just a laundry list of side characters and references to various myths, television shows, movies, or other original material. From this perspective, death is not really an impediment to story because such stories are just archetypal. The puzzles are the heart of the experience.
This assertion is mainly true for the early games. As each series progressed, later games would focus on story more than others and shifted their designs accordingly. Police Quest, for example, was the first one to really have much of a narrative or character arc. The Gabriel Knight games are organized tightly around plot. These games in turn are adjusted so that you can rarely die. Space Quest V either limited death to certain clearly dangerous zones or made it into a funny gag. Leisure Suit Larry abandoned death entirely by the fifth game, and it was a rarity in the second and third. On the other hand, the King’s Quest games remained fairly violent for their entire run.
The question is: what benefit is there to having terminal consequences in an adventure game? In terms of narrative, death keeps a decent amount of dramatic tension alive when the villain or monster can actually impose real consequences on the player. The Minotaur in the labyrinth really is about to kill you in King’s Quest VI, the Aluminum Mallard actually will explode in Space Quest III, and you’ll lose all your progress up to that point if you don’t act cautiously. It’s no accident that most Lucasarts games are comedies: if a character can’t do anything bad to you, then there’s no real weight to their threats. You never take anyone seriously, even a freaky zombie LeChuck, because you know you can’t die. The more serious games like Loom end up with awkward dramatic tension like the exchange with the giant dragon who won’t kill you. As a consequence, their games are limited in their scope. The only Lucasarts games that I’d say were outside the realms of comedy (the Indiana Jones games and The Dig) both include a few spots in which you can die.
Death also adds a certain tension to the game space itself because, when you’re entering a dangerous region, the consequence of losing progress adds real consequences to the game. You save your game before entering the desert in King’s Quest V because you’re warned that any number of disasters can strike. The same goes for beaming down to the dangerous planet before dueling with the Terminator droid in Space Quest V. Moving into new and narratively dangerous areas is something that the game expects you to take seriously, and it enforces this with the save system. To this day, I still play most games with a rotation of three saves, using the oldest one to re-write, because it always gives you room to back-up. Today it’s mostly for poor stat boosting decisions, but in a Sierra game, it was to make sure that you could always extricate yourself from dangerous situations. Saving was a part of how you played.
It also makes the space of the game more dynamic because the player has to pay attention to where their avatar is going and where they are clicking. They have to observe the space, examine things, and make conclusions about what something is or risk the death screen. King Graham will drown in any river he falls in throughout King’s Quest V, so you have to make sure he never walks near one. Lucasarts games with their no death policy are a lot more passive. You click on a spot and patiently wait for the character to walk over there. You can click around without any concern for the details in a space because they can never impose any risk. The player consequently only examines things that serve their own needs. You can rightfully argue that occasionally Sierra games will allow elements in the the landscape to kill you in ridiculous circumstances, such as the zombies do in Space Quest IV or those walking puzzles in the EGA era, and this tendency to fear the environment even made for a great joke in Monkey Island. Yet even with the occasional tedium, the possibility of the environment killing you could make a space more engaging.
The other question that you have to ask is how much does taking out death really fix the problem that Gilbert asserts. Death and puzzles that you can get stuck in because you missed an item breaks the suspension of disbelief because the flow of the story halts. In the audio commentary to the Monkey Island 2 remake there’s a point on Phatt island where Tim Shaeffer is commenting how they designed the puzzles. A key item for each puzzle on one island is located on another, meaning that the player was always travelling back and forth to find a key item. You also see the same issue crop up in Fate of Atlantis, backtracking all over the sunken city becomes necessary because you forgot an item. Ron Gilbert notes in the Monkey Island 2 audio commentary that, during both the jail cell and acid pit scene, puzzles where he locks the player in a single room were always too easy. The player knew that whatever they needed would be in the room, and they could just click on everything. The issue this raises is how the puzzles involving the entire game space seem to just be that same concept except on a much grander scale. In a Lucasarts game, I know that the item is somewhere accessible. Depending on how clueless you are about the puzzle, this can boil down to essentially trying to find a needle in a haystack. That strikes me as a pretty big suspension of disbelief tiger punch. Compared to a Sierra game where you had to backtrack to a save and just check back over your progress, the Lucasarts solution is at the very least equally unpleasant.
Adventure games always succumb to the issue of having an interface where you can theoretically do anything and having it confined to a space where you can only do what the designer allows. The player can always think of a logically valid solution that the designer does not allow. As Kieron Gillen points out on his playthrough of Monkey Island 2, “If you’re presenting a world which should have an obvious solution with an in-world element, the game’s fiction breaks. In other genres, it’s the atmosphere breaking of only being able to jump over certain walls (ala APB) or having unbreakable doors in a world where everyone has bazookas (ala most things). The difference is that in this classic adventure set up every single interaction in the world is based on this nonsense” (“An Hour With Monkey Island 2: Special Edition”, RockPaperShotgun, 12 July 2010). It’s a perplexing enough design problem that I am reminded of the audio commentary in Monkey Island 2 when you find the hint line on Dinky Island. While discussing the gag, Dave Gilbert ponders the fact that hint lines were one of the most profitable departments in the company before gamefaqs came along.