Gamasutra: The Art & Business of Making Gamesspacer
View All     RSS
October 24, 2014
arrowPress Releases
October 24, 2014
PR Newswire
View All
View All     Submit Event

If you enjoy reading this site, you might also want to check out these UBM Tech sites:

Against the Death Penalty
by Craig Wilson on 03/01/11 05:39:00 am   Featured Blogs

The following blog post, unless otherwise noted, was written by a member of Gamasutra’s community.
The thoughts and opinions expressed are those of the writer and not Gamasutra or its parent company.



It’s been many years since the first videogame hero’s spaceship exploded into tiny pixels and while much has changed visually, few have figured out what should happen next.

How do you deal with the hero dying? Can the story go on? How do the larger games incorporate player death into their narratives?

The birth of the death screenATDP-Insert-Coin

It wasn’t difficult for more cynical games to become exploitative back in what the histroy books refer to as the Arcade Glory Days. These games were designed to suck up kid's money and so player death was a common and accepted occurence.

Games which foster learning through such trial and error necessitate death to be a quick and painless flicker on the screen.

The stories told by these games were simple by design. Of course it is possible to read more into the likes of Pac-Man, Space Invaders or Donkey Kong and dress it up a bit but ultimately you’d just be waxing lyrical about a yellow circle, white splodge and a gorilla construction worker. The implications of dying bore no impact on these stories.

And on the third play he spawned again


The same just can’t be said about the games being made today. From Half Life, through Call of Duty, Bioshock, Crysis and beyond, games are boasting experiences driven by larger and more detailed stories. But with the old archetype requiring death to denote failure still prevalent in the design of games today, death is often an unwanted intrusion on the so-called rich narrative experience.

Most games still think it’s acceptable to simply respawn a dead player back into the world. Even with all the myriad sources of storytelling in the post-apocalyptic Fallout 3 – the current affairs of irradiated survivors, sounds of political rhetoric heard on the FM waves, the warnings painted on walls and old entries read on dimly lit computer screens as well as the implicit tales read from the abandoned homes and damaged Washington monuments – should the player’s health fall to zero, they are respawned at the last point they left a building with no explanation as to what occurred between their fall in battle and this new lease of life.

I understand the practical reasons of why it’s done. Entering and leaving a building triggers a quicksave and is therefore an easy checkpoint at which to restart the player and the action. But there's no attempt to resolve the implications of this system. They'd rather you ignore the part where you died and remember only the non-lethal ventures. Whether it’s checkpoints initiated by the player or those registered by the game what this says is that they’d rather discard completely what happens after such a failed state is reached.

Regardless of how it's implemented a save and load mechanism cuts into any narrative being told. It’s like flicking back a few pages in a choose-your-own adventure book and pretending nothing happened. The problem is that the more a section of say Call of Duty is played and replayed the less of a narrative experience it becomes and more it appears as the gussied-up automated shooting gallery that it is.

I’m not quite dead yet


Some games at least address the respawn mechanic and attempt to resolve it with the world. This gives rise to BioShock’s Vita-Chambers, large resurrection tubes in which the player revives.

There’s no backpedalling required with the narrative as in this soft fail state the player’s actions (damage dealt to enemies and items foraged) persist across a potentially unlimited number of revivals. The narrative structure remains intact.

In terms of game play it works but incorporating the devices into the world required a feat of narrative gymnastics. To explain why only the player revives in a Vita-Chamber we are told that the machines only work on a very specific genetic code even though they are nestled in practically every room throughout Rapture.

There’s also very little on how the player goes from bludgeoned corpse on the floor to stepping out of this wonder device- I never knew murderous splicers were prone to random acts of kindness. At other certain key moments in the story they are turned off to prevent further plot holes from emerging.

A much better example can be found in the Grand Theft Auto series. When the player dies, they emerge several hours later from a hospital and when they are caught by the 5-0 they pay bail, lose their weapons and step out of the police station. This seamless integration with the story has the right level of penalty however I know that as soon as the mission looks like its going to fail I open up the menu and load from my last save.

A similar case can be made for respawning fallen survivors behind barricaded closets in Left4Dead that require a fellow survivor to open. It satisfies both the demands of the gamplay and the story.

In general there has been a shift away from finite health bars to regenerating health states. Like in Call of Duty and Gears of War, a few moments in cover after taking fire provides a breather between skirmishes while the red splashes of damage across the screen slowly dissipate.

What remains however is that major failure, often manifested by death, is often treated as more of a temporary inconvenience rather than a real consequence to the player’s actions.

The Corpse Run


The video game equivalent of penance is the corpse run. I don’t know how prevalent it is nowadays but the corpse run was a feature of early MMOs that soured me on the genre. So excuse the lack of enthusiasm.

Should your level 10 Dumbass fall in pit and die during battle, the player’s ghost is resurrected a few miles back at some spiritual site. You now get to run as fast as possible back to your corpse in order to resurrect there and rejoin your level 10 friends who may or may not have waited on you.

This is a high penalty for death with loss of valuable experience points and weapon and item degradation all on the line. All this effort to save a +3 Axe of Awesome is utterly uninteresting to me but, from the story perspective, at least they tied it into the narrative. There are no instant do-overs. If a 40-man raid fails then all 40 of those fine men get the opportunity to attempt the raid later.

In Prey, when Tommy Tawodi dies he is sent to the spirit-realm. Here after collecting enough energy from fallen souls Tommy reawakens in his body and the lacklustre adventure continues.  Playing the spiritual Duck Hunt mini-game in Prey was the modern day equivalent to the corpse run: good for story, not so good for fun.

Death reincarnate


When death is treated as merely an unwanted occurrence along a blessed path then one option is to lift the penalty entirely. Prince of Persia addressed the literal pitfalls of the platformer genre by pairing the player with an acrobatic partner Elika. Rather than fall to their death the nimble Elika dives in and swiftly tosses the Prince back on to solid land. Elika also bails out the Prince from potentially fatal attacks.

This elegant design was met with much derision from portions of the player base. It was seen as dumbing down the experience rather than facilitating it. This reaction makes sense considering how prevalent the old ways of dealing with death are still today by breaking the fourth-wall with checkpoints or a respawn.

During Fable 2’s development, Peter Moore similarly removed the death penalty against the player. Instead the player’s character would become more scarred and disfigured the more damage they endured. This produced two curious responses. In the first camp, players willingly let their character suffer in order to produce the most deformed and mutilated badass possible. The second group simply reset their console: they forced their own death penalty. Old habits die never, apparently.

In Fallout 3 certain characters integral to the main story thread cannot be killed only knocked unconscious. Why the game doesn't afford the player this courtesy is a real missed opportunity. The possibilities of regaining consciousness in an unfamiliar location, like one of the many prisoner caravans, or having been looted and taken hostage by super mutants would have presented even more opportunites for interesting and dynamic storytelling.

Final Eulogy

So the sweet spot for death needs to have an associated weighted risk which isn’t simply tied to difficulty. It also needs to make sense within the world.

Much of the dissonance created between death and story lies in the range of stories currently on offer. In a medium which boasts astonishing variety in genres of game (strategy, action, puzzle, sport, simulation, ranging widely in depth and complexity) by and large the most common story being told is the hero’s journey.

There needs to be a large-scale break away from stories of heroes raging against a seemingly insurmountable enemy or at the very least they need to allow the risks of failure to be explored in as much fidelity as with the rewards of success. There needs to be the lows of failure in the tale of the hero as well as the triumphant highs.

Ultimately to make death work in video games in more interesting ways will only happen once the stories being told become increasingly diverse.

For more death, killing and misery come along to

For more death, killing and misery come along to

Related Jobs

Next Games
Next Games — Helsinki, Finland

Senior Level Designer
Magic Leap, Inc.
Magic Leap, Inc. — Wellington, New Zealand

Level Designer
DeNA Studios Canada
DeNA Studios Canada — Vancouver, British Columbia, Canada

Analytical Game Designer
University of Texas at Dallas
University of Texas at Dallas — Richardson, Texas, United States

Assistant/Associate Prof of Game Studies


Owain abArawn
profile image
I have been thinking that the world portrayed by J.R.R Tolkein in the Silmarillion would make for an interesting MMO.

For those who are familiar with Tolkein's Lord of the Rings universe, the events that take place in the Silmarillion also take place (at least in part) in Middle Earth, but thousands of years before the War of the Rings. Here, Sauron is not the chief antagonist. Sauron is the chief lieutenant of an older, more powerful entity originally named Melkor, but was named by the Elves Morgoth Bauglir, which translates from elvish as The Black Enemy, the Constrainer.

The war between the Elves and Morgoth took place in Berleriand, a land located far to the west of the territory familiar to fans of the Lord of the Rings. Berleriand was utterly destroyed at the end of the First Age of Middle Earth when the Valar, the gods of Arda (Earth) arose from their dwelling in Valinor, west of the Sea and came to the aid of Elves and Men. In that battle, Morgoth was defeated, but Berleriand was destroyed, sinking beneath the waves.

As an MMO, the world of the Simarillion would have features very recognisable to people familiar with the Lord of the Rings. There are elves, men, dwarves, orcs, trolls, balrogs, dragons, and so forth. The events related in the Lord of the Rings take place in a period of time that covers little more than a year. The war against Morgoth in the First Age lasted thousands of years.

So, how could you handle death in a Silmarillion based MMO. Simply put, if you die, you die. If you are an elf, your essence would go to the Halls of Mandos, and after a period of waiting, you would be reborn. As a human, your soul would leave circles of the world, as this is the gift of Illuvatar (the creator of the universe) to men. Either way, in the game, you would be reborn as a new character as either a man, or an elf, or for that matter, a dwarf, or an orc.

Since the war lasts for thousands of years under conditions that didn't change very much (other than continually getting worse for Elves and Men), from a story line point of view, your individual death doesn't much matter. There are always battles to be fought, and challenges to be met.

This kind of game would best be implemented as a sandbox game rather than a quest driven game. That is not to say that there wouldn't be quest worthy activities to undertake. Launch an attack against an Orc stronghold, and now that fortification is controlled by the alliance of Elves and Men. If players could play as either men, elves, dwarves, or orcs (all roughly equivalent in capabilities), there would be a continual struggle to control territory, resources, and strategic points. Men can be allied either with Elves or with orcs under Morgoth. Dwarves during the first were mostly neutral. Sometimes they fought with elves and men, sometimes they fought against them, mostly over gold and treasure. I don't think dwarves ever fought with orcs, or for Morgoth.

Getting the rights to the Silmarillion would be expensive, but a fantasy MMO could be developed using a different, yet similar environment. It would be quite a bit different from any existing MMO, and the idea of permadeath would take some getting used to. Perhaps the best way to get around that is to make the game an MMO shooter in a way. There are no classes. There are no skills, other than mouse and keyboard skills. There is gear, but for the most part, no epic gear. You can get a crappy sword to start with, a good sword sooner or later, and if you are lucky, maybe a very good sword, but there are no OMFGLOLBBQ weapons. For the most part, there is little if any magic, although elves are almost supernaturally stealthy.

This is a game that would probably appeal strongly to old school Ultima Online players, but maybe not so much to gamers who were introduced to MMOs with WoW. As such, this might be seen as a very long shot and a high development risk for a developer.

It certainly would be different, and I think MMOs have to move beyond WoW and it's clones.

Steven An
profile image
Demon's Souls.

All I wanted to say is this: In DS, death mattered. It wasn't a mere inconvenience - you couldn't just reload a save and "try again." The consequences of death were real, often brutal, and that made death in the game meaningful. When you explored a new area further and further, you were scared to your core because you might just die.

Now, DS was designed with this brutal mechanic in mind, which is why I thought it was such a success. Even then, tons of people just don't like the game. I for one loved it.

Cary Chichester
profile image
I would argue that death in DS was not a significant penalty simply because of how often you died. It's normal to spend far more time in the game dead than alive.

Craig Wilson
profile image
Dead Rising was a similar roguelike game which also split critics mainly because of its weighty death penalty. Do I save another survivor or do I avoid any risk of dying and save myself? I for one grew to eventually love Dead Rising for this very reason (see: Stockholm syndrome).

Sting Newman
profile image
Arcades forced developers to develop play mechanics that got people paying money to play there game. The idea that death is 'antiquated' design is a misnomer. It's whether death is skillfully dealt with, with a game design that makes it meaningful or interesting. Even most games today still have the threat of death. If you take the threat of death away many games do not have the same impact. Consider Prince of persia 2008, you always knew that when you fell your female sidekick would save you.

Craig Wilson
profile image
You see for me, death would still have held very little threat in Prince of Persia. Say there was no sidekick, then the following occurs:

1) I miss a jump and fall

2) The game fades to black and the Prince respawns at some point before the ledge (I do nothing)

3) I try the jump again

Compared with their solution:

1) I miss a jump and fall

2) Elika throws me back to the ledge (I do nothing)

3) I try the jump again

In both methods I fail (I miss the jump) which is the only impact I really care about in a platformer. However the second method works around the need to explain the Prince plummeting to his death in the story and more importantly had me back playing the game quicker rather than a fade-to-black respawn that goes unmentioned.

I agree with your point that death shouldn't be viewed as an antiquated notion- I'd rather fail, with or without consequence, than 'die' unless that is the end of that character's tale.

Hakim Boukellif
profile image

Actually, that's not entirely true. Usually when you die in a game, you respawn at some kind checkpoint, such as the entrance of the room you're in or a more arbitrary mid-way point, not right in front of the thing that killed you (and although there are people who disagree with me, I personally believe there to be merit in this). But if there's an in-game existence that keeps saving you when you're about to die, then it would be silly to not send you back to the closest possible safe area. "Gee, thanks for saving me, but did you have to take me all the way back here?"

There's also the issue of believability. Maybe it's just me, but I find the narrative

"...and then the prince fell to his death. In another parallel universe, where hopefully the prince is a bit more careful..."

much easier to swallow than

"...and then the prince was about to fall to his death, but was saved by Elika right in the nick of time. Like she always does when he's about to die. Every. Single. Time. And she never fails, too!"

Trying to make the narrative seamless is a lofty goal, but adding artificial constructs that stretch the player's suspension of disbelief just for that is too much of a sacrifice to make, I think.

Ronildson Palermo
profile image
My main quarrel with "game over" screens is that they remind you that you're playing a game and, depending on your objective, focus and aim with your game experience, that's not really suitable or desirable. Therefore, it should be extinct... But then, how do we introduce challenge? Upon very long introspection ventures and analysis of games I found myself finding the answer in play-time.

No matter how cool a game segment is, people want to see what's next. So, why not make then stay? I think that's the core of the idea: Make the player stay.

If a player screws up a battle plan his unit is ambushed and forced to retreat, making it necessary for the player to devise a new plan and re-execute it, taking more time for him to see what's the next development in the game's storyline.

Should the player be having an horror experience, if he fails to keep hidden from the monster then he needs to run, if he yet again fails to evade the attacks, he's forced to fight in order to make an opening for a new attempt at running away. Bottom line is: the player only needs to believe he'll die or think he will lose in case he fails that particular segment. And one might even argue that such system would numb the player into accepting each failure will only lead to a new task, which can again be taken lightly, but no.

The trick to this is rewarding effort. The ultimate penalty, DEATH, should only happen to those who do not try to win, do not move when faced with danger or don't fight when attacked. If a player is told: "RUN!" and he stands still, then he dies, he loses, he sees a game over screen. But if the player starts moving, even if lightly, the player understands that he is trying and rewards that by, if the player fails that challenge, he is only set back in time. Having to complete more puzzles, cover more ground or search harder for an answer or solution. For as long as the player is invested in the game, he's believing in it, involved.

Again, one might say that when the player figures this out he'll start making minimal effort in order to proceed in the game. And this particular problem can only be solved by developing a system which creates a baseline of the player's skill level and compares it to that, crafting an unique experience to each player playing the game at the time.

And this whole comment is just a long version of this: If the player fails, make him go back and do something to make up for what he failed to accomplish. Try not to repeat yourself in the objectives, but if you have to, at least throw in something new. To keep the game from becoming down right BORING. And if the player shows no interest or investiment in the game's objective or focus, kill him.

So, as long as the player is TRYING, the game simply sets him back with each failure. This still requires him to beat some sort of challenge, but doesn't break the experience in case he fails, simply asks of him to try something different or the same. Giving him a chance, but punishing him with increased play-time, avoiding a "game over" screen.

Alex Belzer
profile image
I couldn’t agree more; just because death in videogames is ubiquitous, that doesn’t mean it’s elegant.

Food for thought: Fathom (a Flixel based flash game) and Queens (another Flixel game). These short (and free!) indiegames treat the death of the player character as events that add meaning to the experience. I think as long as death has meaning then it is solid design to include. Therefore Demon’s Souls passes the test, with the death of the character worked into the story itself, and death used as a means of weighing risk versus reward. Without it, the game wouldn’t be nearly as exciting (or infuriating—but extreme highs must be met with extreme lows). Even in games like Donkey Kong Country or Mario, death has meaning; these games are games, about mastering the platforming mechanics of the game; and they have built in economies of harvesting extra lives. They are games about improving your skill, and death (well, failure) is necessary to accomplish that. Then there are the games that Game Over anytime the player leaves the bounds of the designer’s narrow intention. I mean if you’re designing a heavily story-based linear game then why let player’s make a death-inducing decision in the first place? (Uncharted, I’m looking at you). What about Heavy Rain, which limited player actions to only pushing the story forward; no game over screen, ever. For all its flaws, HR is revolutionary in many ways (and for that matter, and despite its flaws, so is 2008’s Prince of Persia).

Alex Belzer
profile image
PS. Actually, this really goes back to the games-as-games versus games-as-media dialogue from the other week


If it's a game-as-game, then failing at the game makes sense. If it's a game-as-media then death should impart some sort of meaning onto the experience.

Robert Bevill
profile image
I feel that attempts to work death into a game's narrative only work if the universe allows for it. Death can break a player's immersion, but there are not many alternatives to punishing the player for failure. In most cases, the player wants to get back into the action and try again, and any narrative explanation is only going to get into the way. Also, games are too different for there to be a clear answer to this problem. You would not take Demon's Souls's death penalty and apply it to Super Meat Boy.

That said, I do like how Assassin's Creed handles death. It reloads the animus, reminding the player that Ezio never actually died, and that they were simply viewing his memories wrong.

Jamie Mann
profile image
Excellent article :)

However, I'm on the fence when it comes to "death" - yes, it can be used inappropriately and it can be jarring within a story-based game, but it's also a clear, understandable and direct penalty which maps well onto the increasingly "hollywood" settings mainstream games use. After all, the miracles that actors pull off in movies are generally either unlikely or physically impossible - and we rarely get to see the three dozen outtakes which were required to produce the final cut [*]. And while (by definition) people may not like to be penalised, it's impossible to achieve success without the possibility of failure.

On a side note, while this article (and the comments!) does a good job of analysing the various approaches, I'm sometimes concerned that people are rejecting the "death-penalty" purely because it's an established mechanism - much as when there was a drive to remove all HUD elements from the screen for the sake of immersion. There's definite benefits from this approach, but it can also cause significant issues - for instance, it's very difficult to interact with Issac's inventory in Dead Space 2 when standing in a lift, because the close-up camera angle pushes his HUD-display offscreen. Back to the "anti-death" brigade, and Bioshock's Vita chambers could be abused to grind down enemies, rather than learning how to skillfully defeat them, resulting in an unsatisfying gaming experience.

[*] E.g. Scott Pilgrim; something as seemingly simple as Scott throwing a box over his shoulder and into a bin took several dozen reshoots...

james sadler
profile image
This is a good article and it chronicles a lot of the history of death in video games. I read an similar article here back in August or September which has made me really think about how death is presented in my games. Sometimes I don't worry about it, but other times I really try to add a value to it as it can help move the game forwards, make the player more connected, and also add an element to the story that was otherwise lacking.

Christopher Engler
profile image
Respawing is a device I harken to film's use of sound in space. In a vacuum, sound doesn't travel, but in most movies, director's bend this reality for the sake of the overall experience. Most people don't mind this, but the few who do admire when designers abandon the device to achieve a more believable story. If we play a game's character as a role, screwing up is like a rehearsal; whereas, when we get I right, it's like a performance. I think death/respawning might bother more people if games posted death counts and restarts next to their achievements on XBL or PSN.